August 20, 2017

Facts and hypotheses – a false dichotomy

Most people—including many scientists—assume that there is a difference of kind between a “fact” and a “hypothesis”.

A fact, it is usually considered, is something that has been established “empirically”, i.e. by observation; a “hypothesis” is just a hunch that remains to be verified empirically.

This is a false dichotomy, and is only made by those who are ignorant of the nature of perception or observation.

The latter is not simply a mechanical process like taking a photograph, as empiricists would have us believe. It is an organisational one. Data is detected, “transducted” or translated into the informational medium of the brain and organised in that pattern of information that the brain contains.

It is only when this has occurred that the data constitutes information; and this information is not a fact but rather a hypothesis based on the interpretation of the data in the light of our particular model of the system. Information is organised in our brain to form a model of our relationship with our environment, i.e. of the system of which the two are part.

However, each interpretation must vary in accordance with the model used and since, as a result of our different characteristics and different experience, we have all built up different models of our relationship with our environment, so our models and hence our interpretations will be different.

It is for this reason that people see, hear and smell different things and that perception is so subjective.

The information obtained by means of observation in terms of which we verify our hypotheses is itself but another hypothesis, from which it must follow that there can be no such things as facts as distinct from hypotheses.

The main reason why this is not more apparent is that for day-to-day requirements the subjective probability of the hypotheses postulated on the basis of our perception is so high that for all practical purposes they can be regarded as certain.

As a result, we have an innate tendency towards what might be called “perceptive realism”—a form of subjectivism in which we assume the reality of our perceptions, and which provides the psychological basis of the empiricist fallacy.

Thus if I see a dog sitting on the green and eating a bone, I feel I have established a fact. In reality I have postulated a hypothesis. It might for instance be a jackal, or a wolf, or a mechanical contrivance dressed up as a dog. Similarly, what I have interpreted as being a bone might be something quite different—a stone, for instance, or a porcelain figurine, or a piece of wood. However, I shall almost certainly treat these suggestions with the scorn they deserve, for the simple reason that my original hypothesis fits in extremely well with my model, and therefore has high subjective probability, whereas the alternative hypotheses simply do not.

The “hypothetical” nature of perception becomes much clearer in the case of models with a lower degree of probability. Thus, to return to our dog; once I am satisfied that he is eating a bone, I might postulate a number of further hypotheses. For instance, I might identify it as John Smith’s dachshund eating a bone. I might go further and make certain assumptions about the origin of the bone. Since I know that John Smith’s family is away on holiday and that the dog is being looked after by John Smith’s mother-in-law, who is notoriously mean, and even more notoriously indifferent to dogs, I would be pretty certain that the bone had been given to the dog by the local butcher. I could complicate my model still further by guessing what sort of bone it was, at what time in the morning the dog visited the butcher, the expression on the butcher’s face when he gave the bone to the dog, etc. Even the staunchest empiricist would admit that these final details were in the nature of hypotheses, and that they did not constitute “empirical knowledge”, as did the original facts alluded to. Empiricists would thus establish a sort of dualism between valid knowledge obtained by perception and not so valid hypotheses. If so, however, where then is the frontier to be drawn between these two categories of information? Thus, if it cannot be said that I saw John Smith’s dachshund eating a veal bone given to it by a smiling butcher at 11 a.m., since much of this information is assumed or deducted, can I say that I saw John Smith’s dachshund eating a bone given to it by a butcher? If not, can I say that I saw John Smith’s dachshund eating a bone? If I must limit myself to saying that I saw a dog eating a bone, why should the frontier be drawn at this point rather than any other? The answer is that I did not “see” any of these things, if “seeing” refers to an objective mechanical process such as detecting. What I saw was a mass of lights and shadows which I then proceeded to interpret, in terms of my systemic model, by postulating the most probable hypothesis, whose generalities I feel certain of, i.e. have very high subjective probability, and whose successive particularities I am increasingly less certain of, i.e. have increasingly lower subjective probability.

One is thereby forced to the conclusion that all knowledge simply consists of hypotheses with greater or lesser probability or what is the same thing. . . that there is no dichotomy between facts, however well they may be verified empirically, and mere hypotheses.


  • Twitter
  • Facebook
  • Digg
  • Reddit
  • StumbleUpon
  • Diaspora
  • email
  • Add to favorites
Back to top