main | about us | download | faq | readings | blogs | support | tutorial | *

Using reasoning to solve our complex social problemsDonald Steward8/17/2011

I believe there is a clear but as yet unrecognized reason for the rising frustration and anger we are now seeing around the world. As the world has become more interconnected and complex, our ability to use reason to solve our problems has not risen commensurately. As a result we find ourselves in a sea of fantasies that no one is able to adequately dispute. This results in fruitless arguments and anger, but no solutions. And this lack of solutions eventually produces more anger, leading to rage and sometimes violence.

Why aren’t we able to solve these complex problems? Solving problems involves dealing with facts and relations, and the logic to put them together to draw their conclusions. My work appears to show that people are usually able to collaborate to collect the facts and relations. Where they are lacking is in their ability to use logic. Can a computer program help them apply the logic?

Generally it has been assumed that the computer cannot do the necessary logic to help us connect the dots. The logic that the computer would have to be able to do is not the familiar propositional logic, because that assumes the system is open and infinite. An unknowable cause outside the system can mysteriously cause some event within the system to be true. What is needed is a logical process to deal with systems that are closed and finite. It was also necessary to use graph theory in an unusual way to deal with circuits in the cause-and-effect statements. No other method known to me deals with circuits in the cause-and-effects. How to get the computer to do this is by no means obvious. It has required someone who is pathologically persistent and willing to spend almost two decades to finally work it out. But that has now been done. A program has been written to do solve these problems in cases not involving circuits, and is currently being extended to solve those problems that do involve circuits.

Now there is a chasm to be crossed. Since people cannot readily imagine how it might be done and generally do not understand logic, they assume that it cannot be done. (However, those who do have a deep understanding of logic have been able to see how and why it works.) And if no one is willing to look at it more carefully, it will not gain the faith they need to use it. So effectively it has not been done. When people are presented with examples of how it has been done, they don’t evaluate it based on whether the reasoning looks reasonable, but by whether it supports the fantasies they wish were true. “We love our fantasies. Don’t you dare take them away from us!” But this gets us nowhere!

So here we are with the abilities to unmask these fantasies and deal with reality, but an unwillingness to do so.

The scientific method deals with relations to find the best assumptions that would explain the behaviors we observe or the actions that would produce the behaviors we desire. I have found that the relations from which many problems arise can often be stated in the form of cause-and-effect statements. ‘Joblessness’ is caused by ‘Businesses unable to hire’ caused by ‘Businesses unable to sell their goods and services’ caused by ‘Joblessness’.

The computer can be used to help people collaborate to collect these cause-and-effect statements. Then the computer can use abductive logic to use the cause-and-effect statements to find all the sets of assumptions that would explain or produce specified behaviors. But because the behavior is usually not adequately described, the computer will usually come up with many extraneous behaviors that predict behaviors that do not occur. So the computer uses deduction to find all the behaviors that each explanation would predict. Then humans make observations and tests to determine whether these other behaviors actually do occur. Then they select those explanations that predict the behaviors to be explained but do not predict behaviors that do not occur. When a valid explanation is found, the computer produces a scenario that summarizes the reasoning used to find that explanation. This scenario can be used to detect flaws in the cause-and-effect statements that can be corrected in an iterative process to improve the cause-and-effect statements, or to convince people of the validity of the conclusions.

This method has been demonstrated to show how is can be used to solve such problems as finding the cause of the economic crisis and how it can be prevented from happening in the future. It can also be used by a panel of doctors who keep up with the latest medical literature to provide a diagnostic subscription service to practicing physicians who can improve their diagnoses while using fewer expensive tests to protect themselves from malpractice suits. It can be used to formalize and use the computer to help in applying the scientific method to solve a wide variety of problems.

This method could be used as a basis for a social network in which people discuss and help solve the complex problems of the day. People love to share in discussing these problems. But now that we have the means to solve these problems, we don’t have the will to see how they work and use them.

Dr. Donald V Steward
Creator of DSM & the Explainer
Emeritus Prof. - California State University, Sacramento  

Look for more blogs |