The Theology of robots

Edmund Furse
Department of Computer Studies

 

Research Interests My research is primarily in the area of computational models of human learning. I am interested in how people learn to perform a variety of different tasks, and what is the underlying cognitive architecture. Most of this research has been in subjects taught at school and university. I am also interested in Intelligent Tutoring Systems, Parallel Processing and the Philosophy of mind. Clearly much of human learning is in terms of what we already know, and I am thus mostly interested in the learning of knowledge rich tasks. However, in building computational models this creates a problem because one needs to build into the machine all the required prior knowledge. For this reason, much of my research has been in domains where little prior knowledge is used. Research is being conducted into learning in a number of domains including mathematics, board games, computer programming and foreign languages. In each case the research aims to produce a computational model of how people learn the subject area with as little built in knowledge as possible. In each case the task drives the research and leads to the discovery of appropriate learning mechanisms. The research into the learning of University level mathematics uses the the Mathematics Understander (MU) as the computational model of how students learn pure mathematics from textbooks. MU reads texts written in the Formal Expression Language (FEL), and its underlying cognitive architecture is the Contextual Memory System (CMS). MU is an existence proof that Penrose's views on mathematics are incorrect. This is argued in the slide show on the web, known as Mathematics is Scruffy, but Computable There is a popular account of my Theory of Learning and Memory. There is also a popular talk entitled A Theology of Robots On 25 October the Pope announced that he believed in Darwin's theory of evolution, but was doubtful about reductionist models of the human person. See my letter Towards the first Catholic robot? published in the Independent newspaper. Personal Interests I enjoy walking, particularly in the Brecon Beacons and live in a 200 year old cottage. I like cooking and entertaining, enjoy listening to classical music, particularly early church music. I am an active member of my local film society. I am a practicing Christian and organise the local Taize prayer. I am married with no children. Suggestions to efurse@glam.ac.uk Last updated 25 October 1996 Introduction This talk is about the theology of robots. We will be concerned with questions like whether it is possible for robots to sin, can robots pray, and whether robots will be ordained to the priesthood. But first it is necessary to give you some background about myself, then once I have explained some of the terms I can go on to give an outline of the talk. I am a practising Christian and this talk is from the perspective of Christian theology. I make no claims as to the views of Hindu, Jewish or Muslim theologians on robots. A previous version of this talk was published in the Dominican journal "New Blackfriars" and I contributed to a discussion in the letter pages of the Tablet at the end of 1993 following an article by Fraser Watts. I work in the fields of Cognitive Science and Artificial Intelligence. Cognitive Science is concerned with the development of computational models of human mental processes like learning, memory, problem solving and language. Artificial Intelligence is concerned with not only understanding human intelligence but the development of intelligent machines regardless of whether the way they work is the same as humans. To illustrate the difference consider the work in computer chess. Researchers like Herb Simon are interested in how people play chess and has developed a number of models of how people use patterns to choose moves. In contrast, a lot of AI work in chess uses techniques that are not used by humans, like very deep search. Incidentally Deep Blue which beat Kasparov for one chess game, was mostly based on brute force with little AI technology. My own research is in the area of modelling how people learn in fields such as mathematics, games, computer programming and foreign languages. Most of you know what a robot is from Science fiction, for example the robots of Isaac Asimov, or C3P0 in Star Wars, or Mr. Data in Star Trek. These robots are very different to robots we have today. There are a large number of robots used in car manufacturing, but these robots are not very intelligent, and certainly have little grasp of language. Thus this talk is about the intelligent robots of the future. That is to say a robot which is autonomous, thinks for itself, makes its own decisions, can deliberate about its own thoughts, can learn and adapt to new situations and can communicate with humans through the use of language. In general, these robots will be able to do any intelligent task currently done by people. This thesis that one day a computer will be able to do any intelligent task done by people is known as the "Strong AI Hypothesis." This talk has two aims. First, to convince you that one day there will be intelligent robots on earth. Second, to argue that the robots will also have a religious life just as people do. There are five parts to this talk. First, I want to give arguments for strong AI. Second, I want to rebut the arguments against strong AI. Third, I want to look at the sorts of lives that intelligent robots will lead. Fourth, I will consider the religious life of robots. Fifth and last, I will look at some of the moral and social issues surrounding intelligent robots. Arguments for Strong AI Introduction The first argument is the argument from scientific progress. The second argument is the one from technological progress The third argument views the human brain as a machine The fourth argument is from progress in Artificial Intelligence The fifth argument is based on the Church-Turing thesis The sixth argument for strong AI is based on the nature of learning Caveats Introduction I want to present six arguments for strong AI. Note that I am not saying that we will have intelligent robots in the next few years. I personally believe that we will have them within 30 years, but I am not going to put forward arguments for this belief today. Rather, I am putting forward an "in principle" argument. That is, that at some point in the future, perhaps 20 years, perhaps 300 years, but there will come a time when we have intelligent robots that can do any intelligent thing that we can do. Note the caveat "intelligent" since some of the things that humans get up to, like sleeping, eating and our sex lives, may not necessarily be part of the experience of robots. The first argument is the argument from scientific progress and has been argued by Fredkin from MIT. I believe it is essentially a scientific question to understand the nature of the human mind. Part of this understanding will come from neuroscience, and a great deal of progress has been made in the last 100 years in our understanding of the brain. Recent advances with the use of nuclear magnetic resonance scanners have enabled researchers to study small parts of the brain whilst subjects solve problems. There is no reason to believe that within time we should not have a complete map of the neuroanatomy of the brain. However, we must not underestimate the contribution of 200 million years of evolution in the development of the brain. It is possible that evolution has produced a brain that is so complex that it may take a very long time to understand its structure. But, given time it should be possible to understand the neuroanatomy of the brain. However, to understand the human mind it will not be sufficient to know the complete map of the brain wiring. Understanding the full circuit diagram of a microcomputer will not help you to understand much of how it runs an application program. But there has also been progress in Cognitive Science in building computational models of human tasks, and in time these models will cover a wider range of human experience. Furthermore, eventually the cognitive science models will relate human behaviour back to our experience and to appropriate circuits in the brain. Clearly, to understand the mind there will have to be progress in philosophy as well as other fields, but again there has been a lot of progress in the last few years, and increasing interest in the philosophy of mind. Once we understand the nature of the mind it should be possible to build artificial minds based on our understanding. The second argument is the one from technological progress. This argument was most cogently argued about 10 years ago by Clive Sinclair. In the 1960s the most complex computer used thousands of valves and occupied a large room. Over time the size of computers has decreased and the number of switching elements, transistors, has increased. Now it is possible to put nearly a million transistors on a single integrated circuit. Sincliar has pioneered a technology known as wafer scale integration which uses the whole five inch silicon wafer as a complete electronic component, rather than breaking it up into about a hundred chips. He believes that within about 20 years it will be possible to build a machine with 10 thousand million transistors in a box no larger than the human brain. Provided that such a machine can also have the very high interconnection required, it will be comparable in its complexity to the human brain, and the same size. But, of course, without corresponding scientific progress, we will not know how to program such a machine. Thus I argue that scientific progress will enable us to understand the mind, and technological progress will allow us to build a mind. The third argument views the human brain as a machine. albeit a very complex one, and thus able to be built in an artificial technology such as silicon. Few neuroscientists would doubt the role of neurons in human thought, and we can think of the brain as a very complex network of neurons. This is a simplification because other cells such as glial cells may play some important part, and we need to remember that protein structures play a role in human memory. Nevertheless, most scientists would be happy to view the brain as a vast but complex machine. As such it should then be possible to purely replicate the brain using artificial neurons. This has already been done for very simple life forms such as insects which only have a few thousand neurons in their brains. In principle, it would not be necessary to have a full scientific understanding of how the brain works. One would just build a copy of one using artificial materials and see how it behaves. The fourth argument is from progress in Artificial Intelligence. AI programs can do a wider range of intelligent tasks and increasingly complex ones. Programs can show understanding of natural human language, solve problems and learn. It used to be believed that a program can only do what it is programmed to do. But since we have developed programs which can learn, this is no longer the case. In the last five years there has been increasing interest in computational models of creativity and discovery, and whilst some people used to believe that computers could not be creative, there are now machines which discover mathematical hypotheses, paint pictures and compose poems. Attempts by Dreyfus and others to identify things that computers cannot do have only proved to be new challengers for researchers to achieve. The fifth argument is a technical one from Computer Science known as the Church-Turing thesis. They separately argued that given an algorithm running on one computer, it could always be rewritten and run on another computer. Thus, in some sense all computers have the same abilities. Now we can apply this argument to humans and existing computers. Given a problem that can be solved by a person, this problem solving can be thought of as an algorithm, and this algorithm can then be run on an ordinary digital computer. Of course the digital computer may run the algorithm much slower than the human brain, and it will need all the knowledge that the person had in executing the algorithm, but at some level of analysis, it is essentially the same algorithm. Incidentally this argument turns on its head an argument by Roger Penrose who argues the opposite position. The sixth argument for strong AI is based on the nature of learning. If we could understand the nature of human learning, then we could build a machine with the same learning mechanisms. Such a machine if brought up in a suitably friendly environment would acquire knowledge and experience much in the same way as a human infant. Daniel Dennett argues that there may be as many as forty different learning mechanisms in humans, but given time there is no reason to believe that we should not understand them. Since people do learn, and we can observe what they know before and after a learning task, and even their behaviour whilst learning, this gives us a handle to discover the nature of the learning mechanisms. Of course, there is more to being intelligent than learning, and it may take some time for a very smart learning machine to learn to understand language without already having some special hardware. Caveats Of course, there may be some insurmountable barrier to scientific progress. It may be that there is some aspect of human cognition that can never be understood and will always remain a mystery. Or there may be some part of brain function that of necessity needs to use neuroanatomical tissue, so called "wetware," and cannot be achieved with any artificial materials. But I believe we have no grounds for these doubts. Arguments Against Strong AI Introduction The first argument against strong AI is that it is impossible for robots to feel emotions The second argument against strong AI is that robots cannot experience consciousness. The third argument against strong AI is that machines never understand the meaning of their processing The fourth argument against strong AI is that machines cannot have free will. The fifth argument against strong AI is that God created humans as intelligent persons Introduction I now turn to part 2 of my talk, arguments against strong AI, and their rebuttal. No doubt some of you have already been thinking "Yes, but...", and I hope to address some of your misgivings. I cannot hope to cover all the arguments against strong AI, and no doubt some of these will come up in questions and discussion later. I want to consider five arguments against strong AI. The first argument against strong AI is that it is impossible for robots to feel emotions to feel pain, or to get excited about something. Some might argue that it will not be necessary for intelligent robots to feel emotions, for example in Star Trek, Vulcans like Mr. Spock survive OK being purely logical creatures. However, I believe it is likely that intelligent robots will need to feel emotions in order to function effectively. There has been quite a lot of research into emotion from an AI perspective, for example by Aaron Sloman's group at Birmingham. It is possible to build computational models of emotions, and I do not see any reason, in principle, why a machine should not feel emotions. The second argument against strong AI is that robots cannot experience consciousness. Consciousness has become a hot topic for research in the last five years, with scientists such as Francis Crick believing that it is the most important scientific question of our times. Whilst Roger Penrose believes that the brain needs to use special biological structures to support consciousness which exploit quantum mechanical effects, this is not a widespread view. Others such as Daniel Dennett in his book "Consciousness Explained" believe that we already have a good understanding of the nature of consciousness. I personally believe that there is still progress to be made in philosophy and cognitive science before we have a good understanding of consciousness, but I see no reason to believe that we will never understand it. Certainly it will be an important scientific milestone when we have a good theory of consciousness, but it will come. The third argument against strong AI is that machines never understand the meaning of their processing This is a view most strongly argued by John Searle, that machines do not understand semantics. However, this is a view widely contested in the literature, e.g. in the journal Brain and Behavioural Science. Furthermore, it shows a misunderstanding of the nature of semantics. In some sense even a simple computer understands semantics when it understands that the command "print" means that something is to be printed. Language understanding programs show much more complex levels of understanding of semantics. The fourth argument against strong AI is that machines cannot have free will. People argue that any man made mechanism is necessarily determined in its behaviour, and therefore it cannot make free choices. This view relies on a too simple view of what a mechanism is. We tend to think of things like clocks and engines as machines, but they are far simpler in their organisation than a computer. Even our experience of computers is too limited; the computers we encounter are fairly dumb and very rarely do they take the initiative when communicating with us. When was the last time your computer asked you about what you did last night, or suggested that it was time you did some work? But there is no limit to the complexity of computer programs, and Margaret Boden has convincingly argued that machines like humans will be able to deliberate about their choices. That a machine will be able to consider options A,B and C just as well as a human, and thus its choices are as free as human choice is. The fifth argument against strong AI is that God created humans as intelligent persons, and He made no other such beings. In this view, humans are unique among creation being made in the image and likeness of God, and therefore we will always be different to any artefacts made by us. There are a number of responses to this argument. First of all a distinction needs to be made between humans and persons. Humans are the species homo sapiens on the planet earth; we might also consider other species as humans, e.g. homo erectus. A person is an autonomous agent who is capable of intelligent communication, for example, a human, a Martian, an angel in heaven, and I would argue, an intelligent robot. A number of religious writers in science fiction, for example C.S. Lewis and James Bligh SJ, have suggested that there may be intelligent life forms on other planets - persons in my terminology - and further that these persons will have a religious dimension to their lives. Thus it can be argued that the gospel should be preached on Mars, and throughout the universe. Thus, by considering aliens we can escape the anthropocentric view of persons as only being homo sapiens. Just as when a child is brought up we want it to come to know God, the same argument applies to robots. Why should God not desire the salvation of robots? The Lives of Intelligent Robots Will robots be slaves? Will robots be all knowing? How will robots be created? Will there be male and female robots? Will robots have legal rights? Will there be varieties of robots? Will robots marry and have children? Will robots have different lives to humans? Will robots think better than humans? Will robots lie? Will robots die? I now come to the third part of the talk where I want to consider the lives of intelligent robots. What sorts of lives will intelligent robots live? First, I want to look at two popular misconceptions about intelligent robots. Will robots be slaves? Asimov's picture of robot life seems to view the robot as an intelligent slave. I think this is mistaken. Consider the following scenario. You come home from work, and find that your household robot sitting in your favourite armchair reading the paper. You are shocked, and ask why is it not doing the cleaning? The robot replies that it has finished all the cleaning, and wanted to see what was being discussed in the letters pages of the paper. Is this too fanciful? I think not. The trouble with Asimov's robots is that they are required to be autonomous and intelligent, but only within very circumscribed limits. If you want a robot to be curious and creative in its work, then it is likely to be in other aspects of its life. Will robots be all knowing? Other views of machine intelligence in science fiction see the machine as all knowing and purely logical. This again seems to me to be a mistaken view. Certainly there will be questions which we humans do not know the answer to, which an intelligent robot may be able to find the answer to very quickly since he will think much faster than we do. But it seems to me that there will always be things that the robot does not know. For example, it is not difficult to prove that the subject of pure mathematics is uncountably infinite, and therefore there are always new mathematical theorems to discover. How will robots be created? Science fiction seems to only deal with adult robots, manufactured with a vast amount of knowledge. This is certainly possible since once a robot has been brought up in a particular experience, its memory could be copied into other robots. However, each robot would have memory of its history, and thus each of these robots would still remember the same early memories. It seems more likely that robots will be brought up in a whole variety of experiences, and that robots will want to have some say in how they are brought up. Will there be male and female robots? It is not obvious to me that robots need necessarily have any gender at all. It seems perfectly possible to imagine a robot that is completely sexless. Ursula le Guin in her book "The Left Hand of Darkness" considers the possibility of a race with three genders. Of course, if one wanted male and female robots, then one needs to ask what this would actually mean. In Star Trek, Mr. Data is clearly a male robot, and indeed he is "perfectly equipped." But it is not clear that we would necessarily need to design robots with a specific gender. In this talk I have tended to talk about robots as "he," but they could equally well be "she" or "it." Will robots have legal rights? Once intelligent robots are recognised by humans as having legal rights, the robots will soon start to represent themselves in courts of law. The robots will thus be keen to protect their own interests, and they will have to be given equal rights to humans. Will there be varieties of robots? Some people think of robots as all being the same, but this is certainly very unlikely, and probably impossible. Given different experiences, robots will come to have different knowledge, and different approaches to questions. But it would also be possible to design robots in very different ways. Evolution has designed humans in a fairly haphazard way, for example, parts of our brains which were devoted to a sophisticated sense of smell are not used by the brain today. With the design of robot brains it would be possible to quickly improve on evolution, giving them more powerful memories, faster thinking speeds, and perhaps specialised perceptual abilities like infra-red cameras. Will robots marry and have children? It is not obvious that robots would necessarily either want to marry, or to have children. Some robots might prefer to live on their own, others might prefer to live in communities of robots, or in mixed communities with humans. Nevertheless, the possibility of a lifelong commitment of one robot for another is certainly a possibility. Likewise, mixed marriages between robots and humans are presumably possible, although here children would presumably be impossible, or at least difficult with today's science. If a robot was to have a child, it would not be necessary either for the robot to be female, or for the child to be born from within the robot's body. Nevertheless, robot parents might desire that some of their own design characteristics were incorporated within the new robot child. But it is not clear how a system as effective as genetic recombination could be effected in robots. Will robots have different lives to humans? The simple answer is yes. Robots will have some capabilities probably denied to humans, in particular until we develop matter transmission and Star Trek's transporter technology, only robots will be able to travel at the speed of light. It will be possible to transmit the complete design data of the current state of a robot from one place to another, and then implant this data in a new robot shell. This will enable robots to travel much further distances than humans. Will robots think better than humans? Robots are likely to think much faster than humans, and as a result assuming that they read they will also be better informed than humans. Robot-robot communication will also be better than human-human communication. As a consequence it is likely that many lines of human thought will have been rejected already as fruitless. However, robots could, in principle, have whole varieties of ideas, partly driven by their previous experience, and at the extreme might take views which others, even humans, might consider to be prejudiced. Will robots lie? I will argue in the next section that it will be possible for robots to sin, and therefore it will be possible for them to lie. Asimov thinks that robots will be incorruptible and therefore will make excellent politicians, but he may be mistaken in this view. However, there is the possibility of designing robots in such a manner that there thoughts are totally open, transparent and inspectable by other people. If this was done, then one would know if a robot had an ulterior motive in his actions, and thus lies would be detectable. However, robots might consider it to be an invasion of privacy to have one's every thought open to other peoples' inspection, and this view might prevail in the courts. Will robots die? This is a difficult question to answer. If a robot became ill, perhaps some circuit started to malfunction, then robot doctors would be morally obliged to cure the robot, perhaps by replacing the defective part. It is likely that in the fullness of time, such robot medicine would become relatively perfect, with very few robot illnesses not being treatable. As a result, one might consider that robots would live for ever. However, we need to look at how the knowledge the robot has develops over time. Given enough time, say a few hundred years, an old robot would inevitably be out of date compared to younger models. Certainly, he could have his processors updated, so that his thinking went faster, comparable to the current models. But, he would be stuck with his memories of the past. No robot would want his past memories destroyed, they make up part of his nature as a person. But the robot's previous knowledge inevitably shapes the way he understands the world. The first few years experience of the robot will probably shape the rest of his life. Thus, it is likely that as the robot gets older he will in time be more and more out of touch with his society. As a result of this changing nature of the robot as he gets older, he may increasingly think about deeper issues of existence, reflect on what good he has done in his life, and begin to look forward to death and the life hereafter. Thus there will probably be voluntary euthanasia for robots. Moral and Social Issues Should we create intelligent robots at all? Is the creation of an intelligent robot an act that only God should do? Will there need to be some regulation about the creation of robots? Will intelligent robots take away all forms of human employment? Where are humans to derive their meaning and purpose in life? Should we create intelligent robots at all? Would it not be better for society if we totally banned the design and manufacture of intelligent robots? We might go further and ban all research using intelligent robots. But would this be desirable? Is there not a certain amount of species pride here, that we homo sapiens think we have the right to dictate which sapient life forms should exist on our planet. Clearly, if intelligent robots already existed, and governments decided to wipe them all out, then this would amount to a serious crime on a par with crimes against humanity, or the eradication of nations. If robots were consulted, then surely they would want to live. Besides, we might benefit from another sapient life form on our planet. Is the creation of an intelligent robot an act that only God should do? In other words, in creating robots, we are taking on a role of God which we should not do. The simple response to this is that every time a couple decide to have children, they are also taking on this God like role in bringing forth a new person into the world. If it is OK to bring new humans into the world, why is it not OK to bring a new robot into the world? Of course, one could argue that the book of Genesis tells us that God created the world for humans to live in, and not necessarily for intelligent robots. The simple reply to this is how do we know that God does not desire the creation of robots just as much as people. Will there need to be some regulation about the creation of robots? Initially, the manufacture of robots might be expensive, but this would soon decrease with advancing technology. Furthermore, their maintenance costs will in time be lower than those of humans. Do we want the planet to be totally overcome by robots, with many more robots than humans. Probably not, at least initially. The crucial question is whether robots will be brought up by parents in families. Would a robot when it was young have an owner? Or is the whole notion of ownership inappropriate when considering intelligent robots. We would not really say that we own our children. Will intelligent robots take away all forms of human employment? We are already seeing machines taking up more and more aspects of employment, and this trend is likely to continue. Many human occupations are relatively simple, and can be performed by not very intelligent machines. At present this raises the possibility of humans doing the more skilled work, with the more mundane work being done by machines. But once intelligent robots become as intelligent as us, and even more intelligent, then employment patterns will begin to shift. Just as science fiction, in particular Star Trek, has shown that society gets used to working with people from many different alien species, so also we will get used to robots taking a greater part in the workplace. Where are humans to derive their meaning and purpose in life? We already have this problem in our society with over 10% of the population in unemployment. Of course, it is possible that society will adapt so that there are many more employment positions which would be specific to humans and difficult for robots to do. For example, if robots did not take much interest in food or drink, then there might be opportunities for professional food and drink tasters, and it is unlikely that robot cheffs would be as good as human ones, despite their ability to replicate existing recipes. Meaning and purpose have in human history often been related to the quest for survival, whether at the bare necessities of food, clothing and shelter, to the more sophisticated levels of today's society of the car, the television and the mortgage. But if we have a society in which not only do we already have most of our needs' catered for, but where robots can meet new needs much faster than humans can, this really does raise new questions for us. Meaning and purpose then will have to come from elsewhere than from our employment. I can see various potential approaches to providing meaning and purpose in life if all our basic needs are already met. Some people might derive meaning and purpose through a seeking after new experiences and in particular, after thrills and dangers. Thus mountain climbing, car racing and other sports come into this approach. But also, we have to consider the excitement of a life of crime, of taking drugs, and the seeking of emotional highs. It seems to me that there is a clear distinction between these two approaches to meaning and purpose based on how we feel after we have engaged in the activity. Thus the climbing of a mountain can lead to a positive change in someone's self image. But it is not clear how taking drugs achieves this. Essentially, it is a question of whether one can find a long term basis of meaning and purpose, that continues to provide one with sustenance in the future, rather than just a short term fix. But there are less ephemeral sources of meaning and purpose. First, and most obviously is prayer and adoration. If there is nothing for us to do, then we can always worship God. Indeed this is the picture we have of our future lives in heaven. But we might get bored with a 10 hour day of praising God. Second then, there would also be plenty of scope for personal development and education. Given a lifetime of 150 years according to the latest scientific research, this would give us plenty of time to visit all sorts of places, to study many cultures and subjects. Of course, this would mean real education, a love of learning for its own sake, rather than in order to achieve wealth or fame. Third, there would be plenty of scope to learn to be better at loving our neighbours, at caring for our families and friends. In a sense love does make the world go round, and it is no accident that stories of self sacrifice for others such as that of Schindler or Jesus himself move us greatly and form an important part of our culture. Jesus taught according to St.Paul that it is better to give than to receive, and loving other people can give a very deep meaning to one's life. And, of course, this would mean in this future society, loving both humans and robots. The Religious Life of Robots Introduction Can a robot sin? Can a robot believe in God? Should robots be baptised? Can a robot pray? Would robots go to Church? Would robots receive the Holy Eucharist? Could a robot be ordained to the priesthood? Could a robot go to Hell? Introduction Now we come to the fourth part of the talk, the religious life of robots. What will be the religious experience of robots? In answering this question I inevitably also address the question about the religious life of humans. If intelligent robots are curious about the world, surely they will be curious about religion. Just as a robot may wish to read Jane Austin, 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. Can a robot sin? For, if a robot can never sin, then he is in no need of the salvation of Christ's death and resurrection. Robots will have choices to make, and some of these choices are moral choices. Is it possible that a robot could a) know that a choice was wrong, and b) choose to take it? It seems to me that whilst it is possible that much of the time a robot may consciously choose to do what it believes to be good, it is impossible for it to have perfect knowledge and therefore will frequently sin due to its living in an imperfect world. Thus robots will become corrupted by humans. One problem for Cognitive Science is that the behaviour of an intelligent system is largely viewed in terms of the knowledge it has. This naturally leads one to a similar view to Plato's of the nature of evil; namely that a person sins because they do not know any better. But ultimately it is probably a question of loving oneself above one's love for other people; in some sense this is a sort of knowledge, but how it may be represented is beyond our current scientific understanding. Of course, it is possible that the robot does sin, does even know it has sinned, but it may not necessarily have any conviction of its sinning. In other words, the robot may not ascribe any great significance to the fact of its sinning. This is a common human experience: there are plenty of people who lead sinful lives, but do not examine this deeply, and make little attempt to change their lives. Others may lead fairly moral lives, and even engage in self analysis to improve their behaviour, but find no deep conviction of their sin. The problem is essentially a question of a knowledge of God. If God plays no part in a person's life, and God's existence is not even thought relevant to life, then one is unlikely to see sin as separating one from God. On the other hand, the person with a knowledge of God is likely to see how the experience of sin separates them from God. Can a robot believe in God? Clearly, a robot on reading the world's religious literature can come to believe that many humans believe in a divine being known as God. Humans believe that God is all knowing, that He created the universe, and that He loves humanity. Will the robot continue to sit on the agnostic fence talking about the God that people believe in, without attempting to communicate with God himself? Part of the answer to this question is whether the robot sees it necessary to believe in God in order to make sense of the world. Central to this question is the nature of miracles. The robot could take a number of views on the matter. First, that all miracles are fakes which did not really happen. Second, that the miracles did happen, but they can be explained by some brilliant thinking by the robot. Third, that the miracles happened, he does not know how, but believes there will be an explanation soon which does not require God. Fourth, that the miracles took place, and the best explanation is the one offered at the time, namely that they were due to divine intervention. Fifth, that he does not care whether the miracles took place, or not, let alone whether there is a God who caused them. Sixth, the robot may have had personal experience of a miracle, or have a robot friend who has, and this may prejudice him in favour of believing in miracles. Seventh, the robot may already be a Charismatic Christian who has caused through the power of prayer various miracles to already have happened! Let us assume that the robot does believe that God might exist. The next question is why should a robot WANT to believe in God? "See how they love one another" was how the early Christian community was seen by others. Perhaps if the robot had Christian friends, and he had personal experience of their life of love and care, he might want to have something of what they have. Certainly, if the robot had experienced his own failure in attempting to love others, then he might be more predisposed to want to experience this Christian life. But, of course, if all the robots' Christian friends were indistinguishable in their lives from non-Christians, this argument will not cut much ice. Should robots be baptised? Unless there is large-scale ecumenical progress there are likely to be two approaches to robot baptism: either to be baptised as robot children or as adults. In the former case, the parents would be practising Christians and would undertake to bring up the robot as a Christian. In the latter case, we are talking about the prior conversion of the robot to Christ, and its seeking to be baptised. If a robot asks to be baptised, why should we deny him the gift of God. Assuming that God desires the salvation of robots, then we can assume that in baptism they will receive the Holy Spirit. Can a robot pray? The Lord's prayer is the model of prayer given to Christians by Jesus, and seems perfectly acceptable to robots, although "give us this day our daily bread" might have to be replaced by "give us our regular electric feed." Essentially, a robot should be able to have a relationship with almighty God, to be dependant upon God, and to seek His will. Thus just as a robot can be in relationship with humans, I see no reason why a robot should not form a relationship with God. Indeed if the robot views humans as rather frail in comparison to himself, there may be great merit in the robot relating to a being superior to himself. Thus it should be possible for robots to meditate, to worship God, and to intercede for his needs, the needs of robots, and the needs of the whole world. Would robots go to Church? Although it is possible that the robot might choose to be an anonymous Christian, it is likely that he will want to celebrate together with other Christians. But robots might find human Church services rather slow and boring. A robot might not necessarily appreciate hymns, for example, and the human congregation would not appreciate the creed being said 1000 times faster by the robot. So, it is possible that there might be specialised churches for robots where together they can have their own services. On the other hand it is possible that robot services could be conducted over the Internet. Would robots receive the Holy Eucharist? The problem here is whether robots would necessarily have the apparatus for eating and drinking. It can be argued that eating and drinking is fundamental to the human condition, and therefore if a robot is to adequately relate to humans it will have to be designed to also eat and drink. If that is the case, then a practising Christian robot would naturally receive the Eucharist. However, if it could not eat, then other forms of the sacrament would have to be devised, perhaps an oiling of the head? Could a robot be ordained to the priesthood? From the foregoing you can see that I see no objection to the ordination of robots. There are a number of arguments that might be made against this. First, that Christ was a man, and therefore in so far as the robot represents Christ at the altar, this is inappropriate. A simple repost to this would be that Christ was also a Jew and a former carpenter and yet these are not deemed to be essential requirements for priesthood. It seems to me that Christ died for all persons, male, female, human and robot. A second argument might be that a robot is unlikely to be an icon of Christ at the altar, but I suppose that priestly robots could grow long hair and a beard if desired. Could a robot go to Hell? This is the ultimate question about the sinfulness of robots. Could a robot steadfastly set its face against the will of God. Could a robot continuously know what is the right thing to do, and yet choose to go against it. Could a robot ultimately choose to reject God and all goodness, and desire to be cut off from God and his grace for all eternity? Surely a robot being so knowledgeable would choose a path of goodness. But we have to allow for the possibility of free choice, and in allowing the robot this possibility, we also have to allow for it to ultimately to go to Hell.