Can Robots be People?
By Aubrey O'Fallon

What is a Person?

"To be a person is to have a special moral status, to have the right to live one's life without unwarranted interference from others," say the philosophy students at Illinois State University. "We are calling this the moral right of self-determination." Being a person in America means you have rights and the government is obligated to defend those rights. But with the rise of artificial intelligence, we are facing new questions about just what it takes to constitute a person. Sometime in the future computer programmers may be able to develop a robot that behaves just like a person. If it behaves just like a person, does that mean it is a person? If so, we should give it the moral right of self-determination, and not use it merely to serve mankind. The answer to this ethical dilemma turns on the definition of the word person. I will discuss several possible definitions one by one.

I will start to look for my definition in the most obvious place. The American Heritage Dictionary defines person as, "1. A living human being, especially as distinguished from an animal or thing. 2. The composite of characteristics that make up an individual personality." Okay, but we want to know exactly what set of characteristics it takes to distinguish a thing from a person, so we can test our robot by them. This definition gives us two specific characteristics - "human" and "living." A robot is obviously not a human, but our discussion is on whether we should give it rights anyway. Now we are left with "living."

A Person is a Living Being

Well, what does it mean to be alive? Biologists have been discussing this for years, and it isn't an easy question. They by no means pretend to have answered it, but they do have a sketchy list. This list I got from a Bio 1 lecture outline at The University of Missouri which the teacher posted in the Internet.

1. Complex, Organized, and made of organic molecules.

2. Acquire and process materials and energy.

3. Homeostasis. (Meaning: staying the same)

4. Growth.

5. Response to environmental stimuli.

6. Reproduction using deoxyribonucleic acids (DNA).

Robots, at least as far as I know, do meet a lot of these requirements. They are extremely complex and organized, they process energy, they stay the same, and they respond to their environment. But they are not made of organic molecules, they don't grow, and they don't reproduce using DNA. An organism has to meet all the above requirements to be alive, so robots can't be alive. Therefore they fail the dictionary definition of persons as living beings.

In The Positronic Man, a science fiction novel by Asimov and Silverberg, Andrew is a robot who desires to be a person. In older to reach this goal, he strives to make himself "alive" by gradually replacing his robot parts with more human-like parts. He has his brain placed into an android body made of organic materials, and he eventually invents things like digestion for this body, so he can actually eat food. In the end, Andrew still does not feel like a real person until he is able to die. So he allows his positronic brain to come into contact with the organic juices in his body, and they gradually break it down until it stops functioning.

In the real world, if we made robots that were so lifelike they were able to die, they wouldn't be as useful. One of the reasons for making robots is we want helpers who are more durable than humans. But I don't think a robot needs to be alive to be a person. There are other definitions to consider, which are more to the point.

You see, we don't give people rights because they are alive. If that were so, we would give trees and flowers rights. What is the fundamental difference between a flower and a person? A person can think and feel. He is aware of his existence and his experience. He is intelligent. In our experience so far things which are intelligent are always alive. Could it be possible to break this rule? Could a non-living thing, however well programmed, ever have the qualities of self-awareness, intelligence and consciousness? That is the real question.

A Person has Intelligence, Self-Awareness, and Consciousness

The students at Illinois State University, whom I started my essay with, have devoted a web page to discussing whether it's possible for robots to have these three characteristics. The list of characteristics come from the Star Trek: Next Generation episode called "Measure of a Man." In it, Commander Data, a very advanced android who is part of the crew on the Starship Enterprise, wishes not to be dismantled for experimental purposes. So he claims he is person who deserves the right to be protected from this. Starfleet sets up a trial to decide if Data is a person, and they pick intelligence, self-awareness and consciousness as their "measuring rod" of personhood.

Can a Robot be Intelligent?

The American Heritage Dictionary defines intelligence as, "the capacity to acquire and apply knowledge." A programmer programs knowledge into a robot and tells it what to do with that knowledge (how to apply it). But can robots acquire knowledge? Most people would say "No, all a computer can do is represent the knowledge acquired by the programmer. It can only do what a programmer tells it to do." In other words, it can't learn, therefore it isn't intelligent.

However, programmers can create computer programs with the framework they need to learn by trial and error. An example of this kind of program is "Larry Learner," a program that can play a pencil game called "Last One Loses." The game starts with 10 pencils and two players, who can pick up one, two or three pencils at a time. The player to pick up the last pencil loses the game. Larry plays this game, and each time he loses he records his losing move, and doesn't repeat it the next time. Eventually he gets to the point where there are only winning moves left in his database, so he wins the game every time. Larry has learned how to win "Last One Loses." He has acquired the knowledge of the winning moves. Does that mean he's intelligent?

John Searle, professor of philosophy at the University of California at Berkeley, would say no, because Larry does not understand what he is doing. All Larry is doing is manipulating symbols according the rules the programmer has put into him, but he does not know what the symbols mean. Searle calls this the difference between syntax (manipulation of symbols) and semantics (meaning of symbols). Searle is most famous for illustrating this difference in his compelling Chinese Room argument.

The argument asks you to imagine you're sitting in an enclosed room with an "input" slot on one side and an "output" slot on another. Chinese symbols are fed into you through the "input" slot, and you have a rule book in front of you that lists these symbols and their corresponding matches. Your job is locate the matching symbol, write it on a piece of paper, and place it into the "output" slot.

Imagine you don't know Chinese or even know you are working with Chinese symbols. Now imagine that the symbols being fed in are questions about a story, and the output you've been producing are insightful answers to these questions. The people outside of your room watching what was happening would think you must be extremely fluent in Chinese. But actually, you do not understand Chinese at all. All you know how to do is manipulate symbols. Searle says it is the same way with a computer or robot. It may appear to be very intelligent but it's really not.

There are many replies to Searle's argument from people who think someday robots will be actually intelligent (this belief is called "strong artificial intelligence," as opposed to "weak artificial intelligence.") The most compelling one I found is called "The Robot Reply." This reply argues that yes, a computer program does not understand what words mean. But what if we were to construct a robot with a television camera, one that could "see"? If this robot could see chairs, it would know what the word "chair" stands for, therefore, what it means. And if we could make a robot that could walk, or watch others walking it would know what "to walk" means, and so on for all the names of things and actions in our world.

This argument, combined with computer programs that can learn, makes it sound possible that if we could program an advanced enough robot, maybe it could be intelligent. In "Measure of Man," Professor Maddox decides that Data is intelligent because "it has the ability to learn and understand and to cope with new situations."

Can a Robot be Self-Aware?

Now we turn to the next characteristic -- self-awareness. Self-awareness is being aware one's thoughts and feelings, and able to critically analyze those thoughts and feelings. It is a person's ability to decide maybe he has an incorrect belief or a wrong attitude. In other words, people are self-aware because they can have thoughts about their thoughts. Can robots do this?

Douglas Hofstadter, who wrote an article for Scientific American, came up with a way we could program robots so they could do this. He says the main problem with machines is they are not creative, so they get stock in ruts. For example, robots like those on old science-fiction movies who miss the door and bang into the wall over and over again, completely unaware of the futility of their actions, are obviously not self-aware or creative.

But what if we programmed robots with higher-level "monitoring" programs to "watch over" their lower-order programs, and save them from ruts? These programs would say something like, "Check to see if A is working. If not, try B." It seems like a solution. But there is a problem with the higher-level programs. The robot would still get stuck in ruts, just bigger ruts. In order to avoid this, the monitoring programs would have to have programs to monitor them so they don't get stuck in a rut! And those programs would have to have programs to monitor them... and so and so forth, back into an infinite regress. I asked my fiancé Steve (who has a degree in computer science and took an artificial intelligence class in college) what he thought about this problem. He said it didn't have to be an infinite regress if the programmers wrote a very general process at the top of the hierarchy that has no definition for right or wrong behavior. Something along the lines of, "Try something different every once in awhile." This would throw some randomness into the equation.

Not being a programmer myself, I don't know if that solution would work. But whether it would or not, it seems if a robot did have a hierarchy of programs to monitor the productiveness of its approaches to problems, that would be roughly equivalent to humans' ability to have thoughts about their thoughts. If we accept this as a definition for "self-aware," maybe robots could be self-aware.

Can a Robot be Conscious?

The last quality, consciousness, is the hardest to describe. The American Heritage Dictionary defines it as, "Having an awareness of one's own existence, sensations, and thoughts, and one's own environment." It's impossible to tell if a robot could have this feeling. Does a robot know "what it feels like to be a robot"? Who knows, but I doubt it.

John Searle has an interesting theory about consciousness which might show robots couldn’t have it, but it might not. However it is worth considering. Searle talks about how "states of matter" (which he calls higher-order properties) are different from the states of the individual molecules composing the matter. For instance, a table is solid, but the atoms which make up the table are not. Searle says that consciousness is just a higher-order property of neurons firing. If that is so, then robots don't have neurons, so they can't have this special property of neurons.

However, someone could say, "There are many different kinds of molecules that make up solid substances, so maybe different molecules could make up conscious substances. Maybe consciousness could be a higher order property of silicon chips firing." However, if that were so, all the computers today would be conscious also, and I don't think there's anybody who thinks they are. But, someone could say in response to this, "Possibly it takes a special arrangement of silicon chips to achieve consciousness, like it takes special conditions for water to be solid."

I personally come out on this question in an entirely different place, because I think consciousness is more than a state of matter. I believe it is a spiritual thing, and robots could not possibly have it, because they are only material substances.

The students at Southern Illinois University actually address this very issue in their discussion of Searle's Chinese room. I found this information on the Internet in the lecture outline to their introduction to philosophy class. "Searle's analogy shows more than he thinks it does," they say. "It not only shows that computers cannot think. It also shows that brains cannot think. Thus, if Searle is correct, we must have souls in order to think." They then draw a chart comparing the different elements in the Chinese room to elements in a computer and elements in the human body. I will reproduce it for you here.

According to this chart, the neurons in a person's brain do not understand English just as the man in the Chinese room and the CPU in the computer do not understand English. So maybe we need an immaterial soul or mind to be conscious and intelligent.

Searle responds to this argument this way: "The answer is that the brain does not merely instantiate a formal pattern or program (it does that too), but it also causes mental events by virtue of specific neurobiological processes." But Searle doesn't show why a brain can cause mental events and a CPU cannot. I don't find his "states of matter" explanation sufficient for the same reason -- he does not show why silicon chips could not have the state of consciousness as well. I believe that it does come down to people needing souls to be a people, and I will come back to this in a later definition.

A Person is Responsible for His Choices

Now we have discussed intelligence, self-awareness, and consciousness, but we are not done yet. There is another aspect of personhood that we should talk about. My fiancé Steve, who I mentioned above, brought this to my attention. "Robots always are completely predictable," he said. "If you know their program, you know exactly what they will do next, every time." People, on the other hand, are not predictable. No matter how much psychologists study them they still can't know exactly what they will do in any given situation.

Why is this important? It is important because robots' predictability comes from them not having a free will. Robots don't make choices of their own. They do what they are determined to do. Many philosophers say humans do not have free will either, and maybe that is ultimately so. There is a theory called "functionalism" that says the human brain is a large computer program itself. If so, only a being that was above us could view this program, if we have a creator or "programmer," that being would be God. Since this creator Cod would know everything we were going to do next, we still face the issue of whether we are determined.

But even if we are predestined, it doesn't really make any difference to our life down here. We are still held responsible for the choices we make. Nobody can get out of this responsibility by saying they are just a program and they can't help what they do. But with a robot it is different, because we know for sure he is a program, and we know exactly what the program says. Nobody is going to hold the robot responsible for making a mistake if his program is wrong, or his wiring is wrong. That would be unfair. The programmer and the manufacturer are responsible for that.

Well, in our society, the right of self-determination, or being one's own master, goes along with responsibilities. If someone does not fulfill his responsibility to his fellow man and decides to murder him, we take away this person's right to freedom, and put him in jail. As children grow up, their responsibilities increase, and we give them freedom from their parent's authority accordingly. So, if we can't hold robots responsible for their actions, then they don't deserve the right to self-determination that accompanies this responsibility.

Can a Robot have Free Will?

However, this is not the end of the story. There is something new on the horizon that might eventually make even robots unpredictable. In the June '98 issue of Discover magazine there is an article entitled "Evolving a Conscious Machine." The subtitle of the article says, "Some computer scientists think that by letting chips build themselves, the chips will turn out to be stunningly efficient, complex, effective, and weird -- kind of like our brains." What are they talking about here?

When computer chips are manufactured they are usually hard-wired. All the paths that the bits of information travel down are set permanently in place. But at the University of Susa in England, computer scientists like Adrian Thompson and Unman Harvey have developed "programmable chips," whose wiring is able to be changed electronically in a billionth of a second by instruction programs sent to the chips. These instruction programs, like all other programs, really boil down to nothing more than a string of 0's and l's.

Thompson and Harvey are working on creating computer chips optimally designed to handle a specific task... something like distinguishing tones at two different pitches. But the way they are designing them to do this is revolutionary. They are using genetic algorithms to encode the bit strings which will wire the chips. The algorithm generates slight variations to the bit strings, and each variation is tested to see which is the best at coding the chips to recognize the different pitches. The "fittest" strings are then "mated" together in the same way chromosomes combine on genes in humans. All the offspring of this match are tested, and the "fittest" of them are picked and mated. The engineers continue this for thousands of generations, throwing in a few genetic mutations along the way for good luck. When they are done, they end up with a chip that is more efficient at distinguishing the two notes than a chip programmed by a human for the same task. This is exciting.

But what's even more exciting is Thompson says that he "has been unable to pin down how the chip was accomplishing the task" (78). They are doing it in very weird ways which no computer programmer would expect, and have become just as hard to analyze as the human mind. In some environments they work better than in others. And many different chips can be created that do the same task in different (mutually inexplicable) ways.

If we used chips like this to make up an entire robot brain, it is mind-boggling what we might come up with. I'm still not sure the result would be conscious, but it would most certainly be unpredictable! No one would be able to read its program or know what caused it to do what it did. Although it may still be determined, it wouldn't be any more determined than a human being is.

A Person Has Intrinsic Value

After all this, there is still a question I have not addressed.  In our society there are certain basic rights we give to every person regardless of whether they are capable of bearing responsibility. The right to self-determination is not one of them. But the rights to life and to not be physically abused are. We give people these rights because we believe in everyone’s fundamental worth as human beings.

So you see, in the end, after all I have said, the question really comes down to what makes all people valuable. We have put forward a number of guesses. Are people valuable because they are alive? No, that isn’t really it. Are they valuable because they are intelligent, self-aware, and conscious? Maybe. But if so, why? Is it because these qualities make them able to well-serve society? Is it because these qualities make them like us? I don’t really know.

In my psychology class I have heard the theory that we humans want to protect each other because we share the same genes. Our altruistic drive comes from evolution and the struggle for survival. We are just acting according to our genes, which happen to tell us to preserve our own species. These genes are like our program, planted within us by chance. If it is true that this is all we are, evolution does not leave us with much of a purpose. At most is random purpose, because the program we lived by could have evolved a myriad of different ways but this is just sort of the way it turned out. I would like to illustrate this point with a quote by the evolutionist Stephen Jay Gold, which first appeared in a Life Magazine article entitled "The Meaning of Life." (I found it in book by Ravi Zacharias called Can Man Live Without God?)

"The human species has inhabited this planet for only 250,000 years or so -- roughly .0015 percent of the history of life, the last inch of the cosmic mile. The world fared perfectly well without us for all but this last moment of earthly time -- and this fact makes our appearance look more like an accidental afterthought than the culmination of a prefigured plan.

Moreover, and more important, the pathways that have led to our evolution are quirky, improbable, unrepeatable and utterly unpredictable. Human evolution is not random; it makes sense and can be explained after the fact. But wind back life’s tape to the dawn of time and play it again -- and you will never get humans a second time...We cannot read the meaning of life passively in the facts of nature. We must construct these answers ourselves -- from our own wisdom and ethical sense. There is no other way." (55)

Along with leaving human life and ethics devoid of any intrinsic meaning, the evolutionary view also reduces us to organic machines, with no way to distinguish between ourselves and a robot.

However, there is a perspective that set us apart from machines and gives us intrinsic purpose and meaning. That is the perspective that there is a supernatural world which we cannot see directly, but have evidence of in supernatural events that have happened. There are many different views that ascribe to humans a spiritual nature. One such view is Christianity, which I would like to address since it is my personal belief.

A Person is Made in the Image of God

In his book What if Jesus Had Never Been Born?, D. James Kennedy says, "Prior to the coming of Christ, human life on this planet was exceedingly cheap" (9). Some of the practices common to the ancient world were: exposing unwanted children, treating women like slaves, killing off the elderly, slavery, killing other humans for sport, cannibalism, and suicide. "But Jesus Christ -- He who said, ‘Behold, I made all thing new’ (Revelation 21:5) -- gave mankind a new perspective on the value of human life" (9). Christians were some of the first to set up orphanages, elevate the worth of women, and were some of the main players in the abolition of slavery. The book also quotes a telling anecdote told by James C. Helfy about the end of cannibalism in one tribe because of Jesus Christ.

"During World War II, on a remote island in the Pacific, an American GI met a national who could speak English carrying a Bible. The solider pointed to the Bible and grinned knowingly. 'We educated people don’t put much faith in that book anymore,' he said. The islander grinned back. 'Well, it’s a good thing for you that we do," he said while patting his stomach, 'or else, you’d be in here by now.'" (24).

Why do Christians believe that human life is sacred? Because we believe that man is created in the image of God. This is what gives people their worth, and this is the reason we are obligated to preserve the rights of our brothers. Here are some examples of the Bible invoking this reason: "Whoever sheds the blood of man, by man shall his blood be shed; for in the image of God has God made man." Genesis 9:6. And, "With [our tongue] we bless our Lord and Father; and with it we curse men, who have been made in the likeness of God...My brethren, these things out not to be this way." James 3:9-10.

You see, since man is made in the image of God, we are disrespecting Him when we disrespect or devalue His creation. But robots are not created by God or made in His image. The closest they come is being made in the image of man. Of course we humans will be proud of our creation, but there is not the same moral impetus to value it simply because of what it is.

A Person Has a Soul

But wait. Someone could say, "Since man is made in the image of God, and robots are made in the image of man, maybe that is close enough. It’s just like making a photo copy of what God did. Maybe this ‘image of God’ stuff would come along with the package when we tried to create a being with a mind like ours."

Maybe this could be so, but I doubt it, because part of what it means to be made in the image of God is we have God’s spirit within us. Genesis 2:7 says, "Then the LORD God formed man of dust from the ground, and breathed into his nostrils the breath of life; and man became a living soul." Here we have the Biblical basis for the belief that man is made of matter and spirit. The Bible says it is God’s spirit within us that gives us understanding. "But it is the spirit in a man, the breath of the Almighty, that gives him understanding." Job 32:8. It also says that this spirit is what allows to be aware of our thoughts. "For who among men knows the thoughts of a man except the spirit of the man which is in him?" 1 Corinthians 2:11.

So, without God’s spirit within in them, robots would have something crucially missing. And God’s spirit is not the kind of thing a human can create or bestow upon a robot. God has given us great powers of creativity and intelligence, but there are limits to what we can do. Human beings may be able to create a brain but they cannot create a soul.

Conclusion

I have discussed a variety of definitions of the word "person," and as you can see, with each definition comes different answers to our question, "Is a robot a person?"  It is up to you to decide which definition or combination thereof sounds the most plausible to you, or maybe you have your own definition.  But hopefully I have caused you to consider some new ways of looking at a question that may become very important in our future.  I hope it will.  What an exciting thing it would be if robots someday become so advanced that they cause us to wonder.