B U L L E T I N
of the American Society for Information Scinece
Go to Bulletin Index
Volume 25, No. 2
December / January 1999
Human Work in a Computer AgeASIS Special Award Winner Herb Simon Addresses the Annual Meeting
by Steve Hardin
Herbert A. Simon, 1978 Nobel Prize winner, received the 1998 Special Award of ASIS inrecognition of his research in, and the magnitude of his contributions to, computer science, psychology, economics, philosophy and numerous other fields. Preceding the presentation of the award at the recent ASIS Annual Meeting, Simon spoke to the gathered information professionals.
Researcher Herbert A. Simon discussed some of the challenges facing humanity as it strives to deal with information, computers, nature and its own societal shortcomings. Simon, who won the 1978 Nobel Prize in Economic Sciences, talked about Human Work in a Computer Age at the opening plenary session of the Annual Meeting of the American Society for Information Science in Pittsburgh.
ASIS President Michael Buckland introduced the Carnegie Mellon University professor by quoting from one of Simon's works:
What information consumes is rather obvious: it consumes the attention of its recipients. Hence awealth of information creates a poverty of attention and a need to allocate that attention efficiently among the overabundance of information sources that might consume it.
Simon notes that the lot of forecasters is not easy. He points out how astrologers of old faced a deadly dilemma. If their forecasts were unfavorable, they were imprisoned, tortured or beheaded. If their forecasts were favorable, but inaccurate, they met the same fate. He says that people really do not want forecasts of the future; they want assurances of the future. They want to know everything will turn out all right. So, he notes wryly, "in inviting me here to talk about the future of our species and our work in the next millennium, you are giving me very little incentive to level with you!"
Simon notes another problem with predicting human affairs. In nature, solar and lunar eclipses take place in a particular place and at a particular time no matter what we do, but human affairs are our affairs, usually about what we do or don't do. So something is wrong with the concept of "prediction" of human affairs.
The objective really is not to make predictions but to propose futures. We want to propose futures that are desirable, certainly tolerable. We do not want to propose utopias; they cannot be realized. We cannot repeal the laws of nature. Our plans must account for the things we propose to do, the things we propose to change, the laws of nature that govern them and the laws of human nature that are so flexible they are hard to use for predictions.
We also need goals, Simon says, conditions for "satisficing" the needs of the human condition. These goals will also describe the negative things we wish to minimize. We should be careful not to include goals like perfection of human morality or intelligence or the body. Simon states there have been two great planning experiments in his lifetime. The first was the Soviet Union; the second was the People's Republic of China. Both experiments depended on and tried to bring about a reform in human character. But whatever they accomplished - good and bad - they did not change human character. Russians and Chinese still behave very much like humans in the rest of the world.
Simon turns next to information. He says he accepts without argument the premise that during the last half-century we have been passing through the early stages of a profound revolution. He also argues that there is vastly more to come. We humans spend the vast majority of our time processing information. It is as vital to our survival as processing food. Information processing begins with sensing; it ends with acting. And it involves everything that goes on inside us in between. It also involves not only our nerves, but also the "nerves" of our computers and of our communication nets.
This is the third information revolution, he says. The first was the slow one that turned apes into human beings and greatly enlarged the information processes we call thinking. It introduced language as a vital tool in thinking and communication. Maybe we should call it "evolution" instead of "revolution" - as a matter of fact, it usually is called just that!
The second great revolution was the invention of writing, the ability to make the storage of information independent of the properties of human memories. By allowing information to be recorded on durable materials, this revolution preserved it even beyond the human lifetime. If you borrowed money 7000 years ago, and your transaction was recorded, you would have to be careful that the record was still around. In the more modern era of high sulfide paper, you are safer borrowing millions, because in 50 years there will not be any evidence for the debt. And with electronic transactions, he joked, you can be safely assured that every evidence of your debt will disappear in five years!
What should be our concerns about the third information revolution? When artificial intelligence first appeared, questions arose about what we should do with human work when machines can do it better. The day when this will be true is coming. Simon says it would not surprise him if it happened in the next century. The answer is not that we will spend our time in leisure. Except for children, who must learn to become adults, and the small number of parasites, wealthy or needy, worthy or not, that societies commonly maintain, all of us have to work to survive. He noted he was not talking just about economic survival: studies of retirees show that our sense of value and worth often derive from work. Whatever early cave dwellers did, evolution saw to it that they learned to like it. He said he has not seen any studies of people's abilities to survive leisure, but he increasingly encounters - as the age of his friends increases - people who have a hard time with it.
Simon notes that the recluse scientist Henry Cavendish wrote a lot of papers. Many were not published, but Simon suspects they gave him a lot of satisfaction. Cavendish lived to be 79.
Not all of the unemployed can be Cavendishes engaged in research. What are the other outlets for human skills? Actually, Simon remarks, this problem came up first in the industrial revolution. When one tenth - to take a round number - of the population could produce the food and shelter that previously required everyone's efforts, new wants were created that the rest of the population could supply. The adjustment was not without its trials and tribulations, and we could do better, but roughly the same proportion of the population is employed now as was employed before the industrial revolution. Working hours are shorter, so we did substitute some leisure for work. . . or did we? Do we include the hours commuting by car, not just to work but to the nearest market or to ferry our children to their activities? Do we include as leisure the hours spent in home improvements or serious hobbies? People have filled their time to the point of constant complaint of too much to do. He asked how many people in the audience found themselves busier than they would like to be. Many hands were raised.
It is not fair to describe the new wants as simple "make work" artifacts. Even with all the problems of the health care system, people do live longer. People are better educated, too. Still, leisure appears to be a burden to some people, who use it in a way that some of us do not wholly approve. Forget alcohol, drugs and the fascination with weapons. They were probably around in pre-industrial times. What give us pause are the totally passive activities such as watching TV that are now acceptable. Simon says he includes most forms of surfing the Web in this totally passive category, too.
But before we start legislating for mankind, we need to determine whether work is better than these forms of leisure. The evidence that work is more pleasurable than leisure is negative. Work songs, Simon says, are all about the pleasures workers will have after their work is over. Is work more mind-stretching than leisure? The evidence is mixed. Farming was mind-stretching, particularly if you did it as an independent. But working on the assembly line was not usually intellectually taxing.
Nevertheless, when all is said and done, Simon says he personally takes a rather dim view of the couch potato. He says he would like to design a society in which people have options other than being couch potatoes. Can we create a world in which there are more person hours spent playing games and sports than watching them? Can we avoid creating illusory needs to keep people employed? Can we get people to feel good about their self-worth and to stop thinking that worth depends on keeping humankind unique and apart from the rest of nature? Instead of worrying about computers "taking over," we should be worrying about the human behavior that uses our tools for destructive ends and how to understand and improve it. Computers, by simulating human thinking, can help us in that too, he says.
The problems we are creating have little to do with computers and more to do with strengthening people's curiosity about their world and their sense of responsibility for their future, to fill the vacuum created by the loss of being needed for traditional work. Again, these are not novel issues; the leisured ruling class has always faced them. We would probably learn more about the problems of the future by reading the literature of their societies to see what they thought their problems were. Simon says his crystal ball does not portend either a grim or utopian future. The future is not to be predicted, but to be made by us. The first step is to understand what the issues are and what our resources are for dealing with them.
What about the struggle of the burgeoning human population to live in harmony with the rest of nature? The number of people who think the earth can support an unlimited population is diminishing. Most of us are convinced you cannot have exponential growth without something bad happening eventually. We will need to create societies that use resources at rates that do not have a negative impact on the earth. This issue, Simon says, is even more important than unemployment. The principal tool to handle this issue is knowledge, and computers and communication are becoming ever more essential components of this tool. In the past 10 years, computer and information scientists have made great strides in developing the tools they will need to handle these problems. Robots have come a long way in learning to deal with the complexities of the world. Until we build systems that can interact with the external environment, we are not addressing the problems we need to address. Robotics is a key field, exciting and rapidly moving. He notes parenthetically that Carnegie Mellon University has three of the world's champion robot soccer teams.
Another important area, Simon says, carries the name of data mining. We need a better name, he says, because data mining makes us think of horsepower and energy, not thinking. We need to apply intelligence to data mining.
Keyword searching is not the answer to information retrieval. He adds we cannot solve the problems by brute force. In the chess match between IBM's Deep Blue and Kasparov, the victory did not go to the brute force speed of the computer, but to the knowledge it had obtained from its human trainers that enabled it to search in a highly selective manner. Brute force is not the answer - intelligence is. That is what artificial intelligence is all about.
Artificial intelligence tries to study human intelligence and learn the sly tricks we have developed tocompensate for the fact that our brains are very slow computers. We will need to find out what we use to compensate and apply large doses of it if we are going to be successful in mining the information we want.
Simon notes that the Roman statesman Cato the Elder always ended his speeches with "Cartago delendum est" - "Carthage must be destroyed." He said he would end with a refrain, too, already suggested by Dr. Buckland in his introduction. We must keep in mind that information is not the limiting factor in what we can accomplish. Our problems will not be solved by processing more information but by processing the right information. The scarce resource is human attention, and increasingly, it will be computer attention, too. So getting the right information is essential. That principle should be at the center of all of our design of information processing, he concluded.
How to Order
@ 1999, American Society for Information ScienceLast Update: February 21, 1999