Wednesday, February 01, 2006

Using Occams razor sparingly

Using Occam’s razor sparingly: Exercising judgment in thinking

  • You have been reading up on a topic, and you have to write an essay about it for homework.

  • You are reading about an unusual event and you want to know why it happened.

  • You are curious about why some people behave the way they do.

Deciding what to write about, what is important, and reaching meaningful conclusions about the causes of certain events, or coming to understand the human condition and what motivates people, are all difficult things to do.

There are so many theories, ideas, and slants.  There seems to be so much information on any given topic, that reaching the right conclusions rather than jumping to them is difficult, if not impossible.

There are ways to reach sensible conclusions, and realizing that there may not be any right answers to complex questions is one such way.  There are sensible, plausible answers, but most of the time, there is rarely just one answer that is absolutely correct.

The physical sciences seem to represent the closest we get to ‘right’ answers, but any scientists will tell you that what is considered correct only applies at a certain level of analysis.

In the mosaic of information in other subjects, much of the information actually conflicts, and we know that statistics can be manipulated to prove anything, don’t we?

A friend of mine was recently informed by a doctor that his apparent deafness was due to a spinal degeneration that was limiting the amount of blood getting to the ear.  Another doctor examined him and told him he had an ear infection from swimming in water that may not have been chlorinated enough to prevent contamination.  
He accepted the latter diagnosis.  The question is: Was he right to do that?  Was he right to accept the diagnosis that he most wanted to believe?

He accepted the simplest explanation.  He used the principle known as ‘Occam’s razor’ to decide what to believe. He had a vested interest in so doing, and indeed, who wouldn’t want to believe the second explanation rather than the first?

Occam’s razor states that:  ‘One should not increase, beyond what is necessary, the number of entities required to explain anything.’  It is similar to the principle of parsimony or the principle of simplicity, which is a criterion for deciding among scientific theories or explanations.

‘One should always choose the simplest explanation of a phenomenon, the one that requires the fewest leaps of logic.’
(Principia Cybernetica Web)

In plain language, the simplest explanation is most probably the correct one, given that all the facts have been dealt with, the logical principle put forward by William of Occam, a medieval English philosopher.

Here is a real life example of Occam's Razor in practice.

(1) Crop circles began to be reported in the 1970s. Two interpretations were made of the circles of matted grass. The first one was that UFOs had made the imprints. The second was that someone had used some sort of instruments to push down the grass. Occam's Razor would say that given the lack of evidence for UFOs and the complexity involved in UFOs arriving from distant galaxies, the second interpretation would be the simplest and therefore the one most likely to be correct.

Of course, both explanations could have been wrong, but again, the second was by far the simplest, and so, applying the principle of Occam’s razor, would be the one most likely to be correct.  

More evidence would be needed before the first one could be accepted.  In fact, two people later admitted that they hade made the circles, corroborating the second explanation.

For a given set of observations or data, there is always an infinite number of possible models explaining that same data. This is because a model normally represents an infinite number of possible cases.

Evidence is critical in judging competing explanations where otherwise inexplicable events are concerned.

We talk about reaching conclusions, and we talk about jumping to conclusions, and while it is true to say that the latter is usually applied to more day to day matters, it is also true to say that reaching conclusions based upon examination of sound evidence is always preferable to jumping to conclusions, or reaching a conclusion before any attempt at examining all the evidence has been undertaken.

Acceptance of a proposition based upon incomplete evidence is known as prejudice.   To be honest though, prejudicial behavior is usually displayed in connection with issues that concern people rather than, for example, with scientific phenomena.

Nevertheless, if we were to invariably accept the simplest explanation as being the nearest one to the truth, or to reality, then we would be guilty of habitually ignoring potentially important data just because it complicated the issue or because it was not presented initially.

A complete, in depth understanding of complex issues or equations demands that we examine all the relevant data before we pass judgment or define something, or decide upon something.

The principle known as Occam’s razor is most helpful in helping us to limit the amount of data that is relevant to our understanding of a particular topic.

It is the awareness of the existence of other variables that is necessary, along with deeming them relevant or otherwise.  

In this age of the information super-highway, the Internet, it is those who are able to select relevant data and use it sensibly who will be more successful, and a plethora of information increases the difficulty of being able to do that.
Robert L. Fielding


Post a Comment

Subscribe to Post Comments [Atom]

<< Home