Eric Stockton: Thoughts on Ideological Minimalism

Harold Walsby circa 1940s.

I enjoyed being one of George Walford’s readers and correspondents although I did not have the pleasure of meeting him. I have no idea what he might have thought of what follows but I like to imagine that he would have found it worth reading. I write as an amateur philosopher and long retired professional scientist. This essay is mainly about epistemology .

The impact of science is twofold: it generates ‘knowhow’ and it generates disciplined inquiry; these two functions are reciprocally motivating. Philosophical study is the prime example of the precept that “it is better to travel hopefully than to arrive.” The remit of philosophy is distinct from that of science but neither gets very far without the other being well done. As to certainty, more and more knowhow is certain – practically, we really can make things that work – but scientific knowledge is always uncertain (in that it is radically revisable) in principle.

Philosophers do not, typically, arrive at certainty – at least not at agreed certainties – but in their hopeful journeys of the mind they do demonstrate that, generally, easy answers should be treated with the disrespect that they deserve. We are evidently faced with a complex and often frightening world; practically, we have to assume that our sense experiences originate in realities outside of us and reach our brains in sufficiently dependable, usable, form that can be recalled and recorded. We do not merely face the perceived realities; within the scope they allow us, we have to acquire knowhow and to do so we must, practically, assume that nature behaves uniformly. We have also to assume that the undeniable empirical data can be ordered meaningfully by recourse to elementary logic.

Practically we have to make those assumptions but we must not pretend, deny or forget that assumptions is precisely what they are. Knowhow but, we must suppose, of a non-reflective kind, satisfies the survival needs of the many species that possess consciousness; birds, fishes, tigers have the knowhow to fly, to swim, to hunt and – all of them – the knowhow to heed, to feed and to breed.

But non-reflective knowhow alone does not satisfy us; we generally aspire to know that xyz is the case, often over and above any direct need, for any identifiable practical purpose. It is acting on this yen for ‘know that’ that enables us to progress – human history for the last ten thousand years is full of events, developments, successes and failures of our making. The history of the other animal species over that time, and indeed for far longer, is a rather simpler tale of heed feed and breed with varying success in face of varying opportunities – resulting in fluctuations in their respective populations. The histories of these other species are routine and quantitatively changeable; our history is far more than that; it is a tale of purposeful endeavour and of qualitative change. Our quest for knowthat is so relentless, so obsessional almost, that epistemology (the study of the general question on what grounds can it properly be claimed that xyz is in fact the case?) is perhaps the most important subject we need to reflect upon and one with which we take the most dangerous liberties. We take liberties with epistemology precisely because it is such a challenging subject, precisely because of the difficulty – perhaps an insurmountable difficulty – of identifying any certainties whatever and it is precisely certainties that we hanker after in a manifestly uncertain life in a baffling world.

The first liberty we take is recklessly to exaggerate what we can claim a priori. It may be that there are some things that we can know independently of observation but that is no excuse for the hasty recourse to “it stands to reason that xyz is the case” when, very often, the reality is “it sits on hunch that xyz is the case.” Even the respectably a priori assertion, that if a=b and b=c then a must equal c, must have had a long history of empirical acceptance long before people cerebrated about it and pronounced it to be true, a priori. (After all, how can countless generations of prehistoric hunters have survived except on the assumption that two sets of identical footprints – the ‘a’ and the ‘c’ terms – are identical because of being made by the same set of feet – the ‘b’ term?).

The notion of a priori truth may have its proper place but, as well as being a standing temptation to epistemological haste, it is also politically suspect. Articulate persons with power can often sustain that power by making a priori assertions which lesser mortals hesitate to challenge. If such an assertion were made on avowedly empirical grounds then lesser persons would soon look to see if it did actually match their experience. In a very general way, one might say that a priori is the favoured epistemological gambit of autocracy while overriding respect for empirical data is the proper stance of liberal democracy.

Two other liberties that are often taken are to endorse the ideas that “everybody knows that xyz is the case” and that “xyz cannot be the case because that possibility is distasteful to me or perhaps upsets my established habits of thought.” To be a little more precise, the conformity test and the congeniality test are not epistemologically valid – even if sometimes they work out right in the event. How then do we find the way to absolute truth? The answer, that is increasingly given by thinking persons, is that there is no such way to be found – except possibly in very simple and perhaps contrived cases. Unfalsifiable unverifiable a priori assertions carry less and less conviction – as many a religious apologist will tell you, privately.

My own view is that the notion of a conceptual model is the key to making the best use of the empirical data we can acquire, the best way in the quest both for knowthat and for knowhow. The conceptual model is an adult version of the childhood game “let us pretend that” . When faced with a body of empirical data, the conceptual modeller says, “Let us set the real world aside and construct a related but simplified imaginary world having certain recognisable features – viz – consistency internally and consistency with the data.” The model has to satisfy the tests of coherence and of correspondence.

Having conceived the model, its author and others try to falsify it, try to detect incoherence in it and to find data that cannot be fitted into it. In so far as the model fails those tests it has to be amended or even abandoned; if the model passes those tests then the hasty conclusion is that it is true. But the temptation to be hasty is to be resisted. It may be that a quite different model will serve just as well; it may be that the model that seems to be secure is destined to be undermined by subsequently observed data. (That, say, horses breed true is consistent with the Genesis model of the fixity of species; but other data lead one to adopt an evolutionary model. Then the horse species breeding true is an accurate observation, so far as it goes, simply because the observer’s time-scale is so very short).

It is in the natural sciences that conceptual models have proved to be indispensable and no reputable scientist would deny their value. The flat-earth model seemed reasonable but the globular model proved to be better (i.e. passed better the two tests mentioned). Pictures, taken from orbiting manned space vehicles, make the globular model scarcely a model at all but a simple fact. The geocentric model of the universe seemed to be satisfactory until Copernicus showed that the heliocentric model of the solar system was better and Galileo practically clinched it. The “big bang” model seems, to those competent to judge these things, to be a satisfactory model of the “origin of the universe” – whatever that phrase may be thought to mean.

In biology, evolution is a highly successful model both in respect of coherence and correspondence but, in this case, it was ridiculed by people who were wedded to conformity to the traditional creationist model but mere conformity has dropped out of the running in this case because many religious people are now evolutionists. The current critique of evolution is on account of its uncongeniality to a select band of people with contrary theological presuppositions. This has triggered, in them, a compulsive yen for other versions of coherence and correspondence in the minds of the creationist model makers. In science, conceptual models need not be held to be absolutely true in order for them to be useful as keys to ordering the data and to pointing us in the direction of new data (the Gaia model of the earth, for example, does not entail a belief that the planet is alive; it merely draws attention, on an “as if” basis, to an environmentally more positive approach than we might otherwise adopt). But, as in the case of the globular model of the earth – some models do become statements of truth accepted as such by all but a tiny little rearguard of eccentrics. The move away from literalism in religion is increasingly a matter of such thoughts as it is as if there is a loving, omniscient, omnipotent creator whose ways, admittedly, sometimes puzzle us but only because our understanding is necessarily not as full as His. It is a matter of observed fact that some people adopt just such a model – but it is a matter of extreme difficulty for those of us who listen sympathetically, yet critically, to such people to decide whether they actually believe in such a creator or whether they are simply content with a creator model that they can suppose is coherent and which they perceive to correspond with experience. What is clear is that the conceptual model idea is spilling over from science (where demonstrable fact and clear vision are sought) to religion (where a subjective feeling of being in touch with overriding, elusive, all-encompassing truth is the quarry). The notion of conceptual models is perfectly proper in many areas of thought but discipline has to be exercised in its use and here the principle of Occam’s Razor is the key. “Entities should not be multiplied beyond necessity” means that our models should be both as complex as necessary and as simple as possible, to accommodate the data coherently. Too simple an hypothesis entails ignoring some of the data; too complex a hypothesis entails fantasising beyond the reach of the discipline that the data should be allowed to exercise. Not that fantasy is necessarily unproductive – the case of Kekul dreaming about carbon atoms in a ring did, after all, lead to the solution of the puzzling problem of the structure of benzene. But it is absolutely inadmissible to make a hypothesis for the sole purpose of supporting a preceding, but inadequately founded hypothesis. (For example, if I make the unsupported claim that Mr XYZ is a homicidal maniac, it is inadmissible for me to try to support that by resorting to “but he hasn’t yet got around to it.”)

We get nowhere epistemologically by making unfalsifiable assertions to back up unwarranted ones – people often go far in propaganda and persuasion by committing just that offence. That offence we can term rationalisation. Ideological minimalism has as its basic tenets the adherence to Occam’s principle and the avoidance of runaway rationalisation. Our models must never desert the data, leave the data far behind. The seductiveness of coherence without the salutary discipline of correspondence is the engine of such desertion. We must not be so seduced. Ideology is often precisely this resort to runaway rationalisation – never mind the correspondence, feel the coherence! For this reason ideology can often be a vice. The sought after Theory of Everything could become a Pretext for Anything. Models can be possible truths, they can be satisfactory, but not necessarily true, ‘ways of looking at things.’ My model of ideological minimalism does not even pretend to be true but it says metaphorically what I could only say far less economically by direct description. I think it is as if we live in a large hollow sphere whose inner surface is a continuous concave mirror. Within this sphere we live out our shared lives in a world that includes causation – we have many instances of ‘C’ making’ E ‘ happen; there is also contingency – but for’ A,’ ‘B’ could not happen or could not exist; empirically, there is also chance and choice. Empirically, we are aware of these things and we have to make what we can of them. If we try our hand at what are grandly termed ‘ultimate questions’ we are, on this model, trying to look out through the concave mirror encircling us all. We all know what happens when we look into a concave mirror – depending on our distance from it, we see either an upside-down distorted image of ourselves or a blur or an upright distorted image of ourselves. Whatever else we see, we do not see through the mirror. And, what is more, the harder we stare at the enclosing mirror the less do we heed what is actually afoot inside the sphere, the less do we heed our fellows who are not trying to look through the wall.

On this model, we are cautioned to be wary of people who go on and on about “ultimate questions” – questions that, I suspect, are designed to be unanswerable and therefore not worth asking. I am tempted to propose two laws of the epistemologically reflective life: “Ultimate questions are silly questions.” This is Law 1. Law 2 is the Law of Ideologically Constant Volume; “The wider the scope of a belief system the shallower it is likely to be.” We and the other animals do well enough on knowhow; we should not expect too much of our limited powers to knowthat. Some people think that they really know, some people know that they only think… that xyz is the case. You can do business with the second kind but not with the first.

Given such a minimalist attitude we may yet come to know more than we currently suspect that we can know and we will certainly be more tolerant. The motto of the ideological minimalist was well expressed by a maths teacher of mine sixty years ago when setting Euclidean ‘riders’ – “Don’t jump at conclusions; you might scare them away.”

September 1995

George Walford International Essay Prize

Eric Stockton: Thoughts on Ideological Minimalism

Last posts