Monday, January 14, 2013

Observation and Truth: A Philosophical Entropy

I've been reading and re-reading a lot of philosophy recently via the excellent medium of Modern Philosophy by Roger Scruton.

A large lump of philosophy is concerned with two aspects of truth; the difference between what we observe as being true and what is "really" true, out there, beyond our thoughts and senses.

This gap runs through all aspects philosophy. How do we know that a scientific theory is true for example? Only by application to the world. Application to one person's senses might not indicate truth, a theory that polar bears are blue might be true to a mad explorer who thinks polar bears are blue but that doesn't mean it's a valid scientific theory... does it?

Is there a difference between what is reasoned or calculated and what is observed? Reason or calculation needs data to work on and that comes from observation, or a previous calculation. Ultimately all reason or calculation must stem from a previous observation. The information from observation might be manipulated, or it might be lost or distorted, but the resulting reason or calculation can never create something "more true" that the observation. All calculation takes inputs, malforms them and produces an output, and here the input is that which is observed (or the result of something else which was ultimately observed).

This is a crucial point because it implies that if there is a "reality" beyond what we cab observe then it cannot ever be grasped, even theoretically, even by guesswork or insight. Nothing can be more true than observation, only as true or less true.

It may sound like an implication that only our personal view of the world is true. This is a lonely view, like that of Descartes. He is sadly no longer thinking, but when he was, he was it. Could truth have a social aspect?

This is a communication problem. Observation communicates data to us, it is manipulated and produces ideas, calculations, reasons etc. The data can be degraded in quality or lost but never made more pure. The original observation is as good as it will ever get. Information can also come from prior ideas, calculations etc. but when taking data again, some information may be lost or degraded.

Now, ideas or observations can also come from other people and these may be more true than your personal observation. Remember the mad explorer? One thousand other people may attest that polar bears are white rather than blue, if they observe them as white. If polar bears are truly white then then best that an observation can get is that people see them as white. Errors can creep in; someone with bad eyes might see them as pink, or yellow, or even blue. A distant memory or madness might degrade the information too, someone who originally thought one white might change his mind later. All of this shows that any initial truth can be lost but never gained.

What then if a person who saw the bear as yellow, later by chance recalled it as white? By chance they perhaps reached a truth, but their recollection of whiteness is no more valid than if they had recalled it as any colour because the change happened by chance. What if they saw more polar bears and later concluded that they were white after all, or if more people convinced them of the bear's whiteness? Then a truth had been reached, but it would only be as true as the truth of their other observations or the truth of the other people's observations, it would never be more true than an observed truth. Nothing closer to truth beyond a personal observation would have been created.

As data can be lost at observation and whenever it is communicated or used then the total pool of observations by all people who can communicate with each other defines the best truth that there can ever be. Any truth beyond what can be observed will remain forever out of reach. Any scientific theory that proposes to apply to the "real world" will apply only to the collective observations of those who observe and nothing beyond that.