Deduction, induction and paradox

In examining puzzles, mystery, enigma or paradox, the first instinctive reaction is to apply logic. This is the science of reasoning that looks for defensible arguments based upon sound and provable tenets. As J. S. Mill defined logic : "Logic is not the science of belief - but, of proof or evidence".

Unfortunately, when dealing with fast changing environments like the Internet, there is very little in the way of evidence. The Internet is not only complex, it is without any sustainable precedence. There are no case histories or previous experience known today that will necessarily apply tomorrow, so, pure logic does not appear to be a suitable tool to use when contemplating suitable strategies for e-commerce.

Rhetoric and dialectic are logical processes of thought used by Hegel to merge contradictions into a higher form of truth. This is a special branch of logic which according to Aristotle was invented by Zeno of Elea and scientifically developed by Plato. This method relies on a continuous series of questions and answers to gradually bring opposing arguments together. It is the art of critical examination into the truth of an opinion through discussion.

However, as many flame wars in Internet discussion forums will testify, this well founded traditional approach to logically resolving opposing theories is easily disrupted by paradox. As complexity in general, and e-commerce in particular, is beset with countless examples of paradox, the rhetorical and dialectical approach to understanding would appear to be inappropriate.

It is worth looking closely at this phenomenon of the paradox. It will be encountered over and over again when dealing with complexity and e-commerce strategy. For our purposes we shall define a paradox as:

"A proposition or an observance which seems at first sight to be absurd or self contradictory, conflicting with common sense or preconceived notions of what is reasonable or possible - upon further reflection, or, with new evidence or explanation, the proposition may prove to be well founded and essentially true".

In 1685, Hobbes observed "The Bishop speakes often of paradoxes with such scorn or detestation that a simple reader would take a paradox either for a felony or some other heinous crime - whereas perhaps the more judicious reader knows that a paradox is an opinion not yet generally received".

Brooks, in describing the art of poetry in 1942, commented: "Few of us are prepared to accept the statement that the language of poetry is paradox - yet there is a sense in which paradox is the language appropriate and inevitable to poetry."

It might also be argued that paradox is the language of complex systems and, by inference, the language of e-commerce because, as we shall see, the route to understanding is littered with paradoxes of all kinds. Thus, if paradox is a hindrance to logical thought, how can we get to understand e-commerce?

Inference is a term that is often used in the context of solving problems where the rules of logic cannot be directly applied. This can be defined as the forming of a conclusion from data or premises, either by inductive or deductive methods. It implies reasoning from something known or assumed to something else that follows from it. Mill (1843) wrote: "Cases of inference, in the proper acceptance of the term, are those in which we set out from known truths, to arrive at others quite distinct from them. In any act of perceiving, observation and influence are intimately blended". It is interesting to note that some philosophers restrict this term to apply only to methods of induction.

Sherlock Holmes is noted for his skills of deduction, but, the latter day meaning of deduction is: "the process of inferring particular instances from general laws". Deduction, therefore, goes from the general to the particular. Applying deduction to complexity and to e-commerce would see us having to make decisions based upon established principles and proven techniques. This is the way of the traditional business world; it makes decisions based upon procedures and concepts that have been established through past experience. This is not appropriate for e-commerce, where there is precious little past experience. What there is cannot be reliably applied in such a continuously evolving environment as the Internet. It would seem then that the methods of deduction will be of little use to us. So, should we have to abandon the methods of Sherlock Holmes?

W. V. Quine, writing in 1940, made the observation: "The highly explicit way of presenting formal deductive systems which is customary nowadays dates back only to Hilbert (1922) or Post (1921)". As Connan Doyle wrote about Sherlock Holmes in the previous century, this suggests that Sherlock Holmes methods of explaining the unexplainable was not by the process of deduction as we understand it today. More likely his investigations proceeded by means of induction rather than deduction.

Induction is the reverse of deduction. It is the process of inferring a general law or principle from the observation of particular instances. It doesn't, like deduction, mean the process of looking at the general law, or, big picture, and from this working out the details. It means looking at the little things that happen and then trying to discover the wider, more universal influences causing them.

This is how Sherlock Holmes worked. He looked for little significant clues that suggested a pattern: a pattern of behaviour and circumstance which were part of the broad pattern of a crime. There is a story that Sir Francis Bacon gave birth to the principle of induction when he noticed the complementary similarities between the coasts of South America and those of Africa. They seemed to fit like pieces of a jig-saw puzzle. It cried out for a reason or a logical explanation. Three centuries later the explanation came, by way of the discovery of plate tectonics.

Confirmation, of the way in which Holmes reasoned was provided by Connan Doyle's biographers. They describe how Doyle once worked with and was a life long friend of an eminent surgeon of the time. There are several records of Doyle referring to and admiring this surgeon's powers of 'deduction' as he used small, sometimes seemingly insignificant, observations to diagnose a patient's condition. As this was before the time of the modern understanding of human biology, this surgeon must have had to use methods of induction rather than deduction.

In the research establishment, where I spent much of my time as a student, it was patently obvious that there the scientist were using methods of induction rather than deduction to expand the frontiers of new and emerging technologies. They were working at the leading edge of newly emerging technologies; there were no previous work or experimental evidence on which to base conclusions or give guidance as how to progress and move forward. The mind sets of those scientists, who used inductive methods of reasoning, were quite different from the deductive mind sets of the corporate management executives of the Industrial Age. The corporate decision makers have a wealth of procedures, techniques, statistics and previous knowledge to draw upon.

Perhaps then, it is going to be the method of reasoning known as induction that is going to prove the most fruitful approach to understanding and being successful in the enigmatic world of e-commerce. This will not rely on the process of deduction used regularly by large companies to create new strategies out of experience and established practice. It will involve a process that looks for clues in the here and now and then searches for explanations and causes.