Thursday, 10 September 2009

Rational Incrementalism

In my last post I was considering scepticism and the certainty of knowledge. I signalled at end that I would discuss a term I had coined: Rational Incrementalism. Though what I propose here is nothing new, nor is it some new way of thinking about knowledge and certainty, it is rather, a organising structure - a kind of conceptual map. A map that we can compare theories and propositions to. I am, mostly, considering scientific theories here, however in a later post, I will consider its application outside of the sciences.

Modern science as opposed to theology or certain non empirical strands of philosophy acquires knowledge in piecemeal form. It builds up a picture of the world in fits and starts, gradually or - incrementally. Theology, makes grand claims to knowledge on the basis of a few dogmatically held positions; it can then happily and whimsically build elaborate and ornate systems of belief based on these suppositions; should though, one supposed foundation be found wrong or invalid - then the whole edifice comes crashing down. Some might argue that science is similar: “a ugly fact can kill a theory”, indeed, but science in making more piecemeal claims seeks to find a number of coherent, mutually supportable beliefs (or theories) with attendant facts and evidence. Though a fact can kill a theory, more often than not it is assimilated into the existing paradigm or theory, thereby enriching, expanding or modifying our understanding.

What I will outline is the conceptual terrain that a theory may pass through before we can say, with sufficient certainty, that it can count as knowledge, or, at the very least, having a high probabilty of being correct.To capture the progress in a metaphor - our understanding of the world according to this paradigm - progress like a person on a escalator. They start out at the bottom -yet gradually, incrementally, they get lifted higher and higher - they “see” more, they understand more, as their experience of the world increases. Furthermore, by proceeding cautiously and methodically they do not greatly risk falling over, or the ground giving way under them. They do not make one big improbable “leap” to knowledge - loudly proclaiming truth - no; whatever claims a scientist, generally makes, are the result of careful laborious research and painstaking years of study. The theory progress via six stages: 1. Idea. 2. Speculation. 3. Hypothesis. 4. Theory. 5. Error theory. 6. Scientific explanation or scientifically established theory.

1. Idea

Firstly, before there is any theory, before any hypothesis can be tested, there must first be ideas or possibilities. This is perhaps, where a creative aspect to science and philosophy comes in. I should define an idea as merely a possibility: that, which does not have any evidence for it either way. Ideas can be conceived in two main ways. Firstly, they can be conceived in what could be termed an “a priori” position. Thus, the thinker knows no important facts or circumstances upon which he is thinking about - he thinks up ideas and thereby seeks evidence to confirm or disconfirm his idea. The second kind of idea - one studies a body of evidence or set of circumstances - after which the thinker/scientist/philosopher attempts to conceive a theory which explains these facts - this, is post facto style of explanation; one needs to be careful however - that they do not yarn a JUST SO STORY. There first kind of idea is, what I shall term hypothetical ideas, the latter being explainer ideas.

2. Speculation.

Now we have our idea. We either have an idea that will attempt to explain what we have seen or experienced (explainer ideas) or we will attempt to make the evidence fit our burgeoning theory (hypothetical ideas). Either way, what we must do now is test our idea or concept. Except in some cases, perhaps in the nitty gritty sciences of molecular biology or astrophysics -we can do a lot of preliminary testing from our armchairs. We look to see if there are any reasons or evidence in support of our supposition, or indeed, reasons that would seriously count against it. If we can see something that could drastically kill the idea (falsify - make highly improbable, or indeed if it’s highly improbable in the first place -thus not worth pursuing) - then we are back to the drawing board, if not then we advance to step three

3. Hypothesis.

We now have tested our concept in a purely informal and largely unscientific way, indeed a lot of the work up until now concerns language and concepts and armchair reasoning. Questons like, what do we mean? What would prove or disprove it? What kinds of evidence am I looking for? Is what I am talking about coherent with what we already know? Is it internally consistent? Now however, it is time for philosophy to give way to science. We have hammered out a concept or proposition to test; the aim, being either to provide verification of our hypothesis, or falsification. If our experiment goes well, and it is repeated with success - peer reviewed satisfactorily and so on - then we are on to stage four.

4. Theory.

We now have a theory, perhaps a new one, or a competing one or new alternate theory which explains some phenomena. The important thing however to note, is that the new theory is only one among many; subsequently, it will have to compete against these other theories, and at the same time answer any potential criticisms before it can be respected as an established theory. The reason being, that, for any body of data we can have multiple interpretations or theories. Our new theory then, is just a new kid on the block: out to “prove itself” - why it’s a better theory than the others on the “free-market of ideas”. Though the new theory has some empirical support - it is far from conclusive; the next goal then, is to point out why the alternate explanations are wrong or that the objections are misguided or misinformed.

5. Error Theory.

Strictly speaking, a new theory does not always call for an error theory. It’s possible that something original has been discovered that does not appear to supersede or refute an existing theory. This is not always so however, as the case of Darwin and Einstein showed. What an error theory purports to show, is why we were wrong to think the former theory was correct. It many ways, this is a powerful and important philosophical tool; it, in effect, performs a Ju-Jitsu move on the opposition - using the force of the opponents argument against them. An Error theory, then, needs to explain coherently why the originally theory or belief was mistakenly held in the first place. I should note, though, that an error theory needs to be deployed after some positive evidence has been produced for a new theory which explains the facts; as anyone can do an error theory for anything established - but its useless if they do not provide in the old theory’s place an explanation of the phenomena.

6. Scientific theory or established theory.

Finally, the theory is ready to be considered a fully established, scientific theory. It has considerable explanatory power - better than any other theory. It demonstrates a high degree of consistency with the evidence; it leaves no ambiguities or unexplained problems; additionally, it coheres with what we already know to be true; furthermore, it is simple: it does not rely on obscure, largely unproven, or implausible assumptions; lastly, and perhaps, most importantly - it is testable - with a large body of experimental data to back up its claims.

Putting theory into practice - where the rubber meets the road.

Lets now compare this conceptual map to say - Darwin’s theory of evolution by means of natural selection. Darwin it should be noted, was not the first person to conceive of evolution: a number of Greek philosophers noted its possibility - they did not, however, test it or provide evidence for it - so it remained merely an idea. Darwin’s Grandfather, Erasmus, understood its possibility. There was, in addition, a small number of other thinkers, who, before Charles Darwin, explored the possibility - with some references to artificial selection . It could be said however, that their ideas were speculation - they had some evidential support, but it lacked scope and sufficient rigour.

Darwin provided the hypothesis and the theory - alongside Alfred Russell Wallace. Darwin, after years of thinking and amassing evidence provided testable hypotheses, and a mechanism or means by which evolution unfolds: natural selection. Darwin could point to the fossil record and argue that lower forms evolve into more complex forms (by way of transitional fossils) - over great expanses of time. Indeed, the artificial selection, and hence - evolution - of dogs, pigeons and livestock; which subsequently, though well know, provided additional and highly persuasive evidence for the burgeoning and highly controversial theory of evolution.

Darwin and Wallace then, developed the theory ( evolution by natural selection) which promised enormous explanatory power, it was in its day - highly plausible and consistent with the evidence that was available. However, it could be argued that it was not until relatively recent times - that the last two features of our schema were provided for: an error theory and that Darwin’s theory became a fully established, and secure, scientific theory.

Many thinkers have provided reasons as to why we did not conceive of evolution sooner and why many have trouble grasping it, or rather - accepting it. If we set aside religious propaganda, and examine the factors which gave rise to movements like creationism - we will find - intuitively attractive, or seductive “reasons”, though entirely unsound, - why some “think” design so convincing and evolution so absurd.

Firstly, evolution requires a vast amount of time to work - the earth being only a few thousand years old - hence not enough time for the theory to work. This objection, which is still made today, was solved in Darwin’s own time, where the age of the earth was shown to be very old indeed. Humans who only live, if at best, for a few decades, subsequently - many cannot grasp deep geological time.

The second reason we have trouble with evolution is that we are a tool making species. Darwin’s theory, as Daniel Dennett notes: “is a strange inversion of reason…. You never see a pot making a pot maker, never see a spear making a spear maker…. Never see a car making a carmaker.” Darwin’s theory when presented to our common-sense intuitions - flat out nonsensical - it has it the wrong way round. It is only complex things which can create or design less complex things; but Darwin and scientists ever since, have shown time and again why, this is wrong.

Since Darwin wrote Origin of Species, there has been a number of scientific breakthroughs that have supported and extended his theory. Firstly, there was the discovery of genes - this solved the problem that vexed Darwin as to how the information from parent to child was passed, thus ensuring the preservation of information which allowed superior organisms to survive and reproduce. Secondly, with the advent of computer technology and complex mathematical algorithms - the theory of natural selection as a mechanism - can be tested, abit theoretically. Furthermore, after more than a hundred years of investigations, in biology, in palaeontology, bacteriology, epidemiology - Darwin’s theory, is, as Richard Dawkins affirms: “the only game in town”.


To sum up then, we can use this conceptual map or schema to assess on what point of the scale a theory or hypothesis rests on. Furthermore, we can see what work lies in front of us if we wish to establish an idea or theory as knowledge. Finally, this paradigm allows us to independently assess what degree of support a theory or explanation has in the scientific community - we can seek out and examine studies, experiments, books and experts - in order to determine how well established and respected a theory is.

Saturday, 5 September 2009

On Scepticism

Recently, I have been considering the uses, values and applications of scepticism. I will try to distinguish firstly, what I mean by the term scepticism. We ought to bare in mind the sharp and frequently misunderstood distinction between academic, philosophical or theoretical scepticism or Scepticism, and “applied” or “practical” scepticism. I shall briefly sketch the differing forms of this position, Finally, I will recommend some practical recommendations when applying sceptical thought to real life problems.

Consider academic scepticism: this can be taken to be either one of two positions. The first being that all knowledge claims are of equal merit: we cannot rationally distinguish any claims to knowledge - we hold all claims, or pretensions to certainty, as being equally probable. The second, is a more moderate position: any inference between fact and theory - then there is always a underdetermination of evidence. This means for example, that when I throw an apple up in the air - the established and empirically backed theory of gravity explains - why the apple falls to earth. It is possible however, to construct a consistent (though implausible) theory of why the scientists are wrong - that it is, instead, a invisible demon pushing the apple down.

I shall now outline the problems with this form of Scepticism (this will apply both to the stronger and weaker forms.) Given however, the persuasiveness of arguments produced by philosophers such as Hume, Russell and Quine, it would be reasonable to say that we can never be absolutely sure of any claims to knowledge, even gravity or claims involving mathematics (Quine). Though I broadly accept this argument; while subsequently rejecting however, the idea that all knowledge claims are equally probable or likely; I find, in the end, that this sort of Scepticism is next to useless; moreover, it allows people to be lazy, or worse, insincere - “you believe what you believe and I believe what I believe - in the end its all just opinion.”

This kind of Scepticism is useless, for, like Pyrrho - you end up with your head stuck in the ground, not knowing whether or not it’s a good idea to try and remove it. Secondly, this form of scepticism is self-defeating. Like the Marxist argument concerning super-structure or false consciousness - it can be boomeranged back against the opponent and vice versa - without any address to the substance of each other’s argument. Each person take up a position of complete scepticism against the other, thus - no progress is made. Next, the claim that one knows nothing or cannot know anything - is contradictory - for one cannot claim to know even that. Finally, the claim, though sound - that we can never have absolute certainty, leads moreover, to a non-sequitur if we say that there does not exist probabilistic degrees of certainty or rational expectation - that any claim to knowledge has a 50/50 probability. Bertrand Russell closed the door to this fallacy forever when he came up with the argument from ignorance: the celestial teapot. We can never be sure that there is a teapot orbiting the sun, or that there are fairies in the bottom of the garden - though we cannot disprove such things, there is scarcely a reason to believe that a china white teapots currently orbits space or that there are pixies playing in the garden.

Turing now to the more profitable uses of scepticism then, - keep in mind how one could, potentially, be wrong. Seek out then, alternative explanations and other possibilities, indeed, invite criticism and debate - to ensure that you have not fallen for the first explanation you have been presented or came up with. Be careful when judging claims that are outside of your field of expertise; furthermore, be mindful of the emotional reactions when you comes across an argument that contrasts with your position - that “extraordinary claims require extraordinarily evidence”; finally, beware the claims that seek to bolster self-esteem or attempt to flatter one sense of self.

Here are a few useful rules of thumb when using positive or practical scepticism. I draw some of them from Bertrand Russell’s essay on the values of scepticism.

1. When experts are agreed, the opposite claim cannot be held to be certain. 2. When they (the experts) are not certain - no opinion cannot be held to be certain by a non-expert. 3. When they hold that no sufficient ground exist for a view - the non-expert suspends judgment.

In addition to these guidelines, I would add the metaphor of a juror considering evidence in a court case. Imagine what would, in the circumstances of the case, consist of reasonable doubt - has, then, those doubts been met? Consider what evidence or arguments would force you to revise or abandon your conclusion? Attempt then, to seek these alternate explanations, possibilities and doubts out - if you do not find them, or if they are not persuasive - then maintain your position. Finally, ask yourself, are you being objective and dispassionate in your analysis and evaluation, or, are you being driven by emotion and prejudice.

This short primer then, should help us keep in mind the important and often forgotten distinction between philosophical and practical scepticism. Furthermore, the maxims outlined are useful rules to keep in mind when encountering strange and extraordinary claims - especially if they lie outside our normal range of experience and expertise. In addition to this short missive on rational thought, in my next blog, I will consider a gradient or structure where we can track our ever increasing certainty when considering a theory - the process I call Rational Incrementalism.