When reading a text as you are now, what actually reach you are sense data: you then interpret the images as letters or directly as words. There then has to be some way that you judge whether what you are reading makes sense or not. Finally you assess the message's truth, its logical status, its comprehensiveness, refutability, reliability, ....

§6.1 Building Knowledge

If you thought that I was an authority dispensing true knowledge and wisdom then you might accept my arguments unthinkingly: those arguments would require no justification beyond the fact that it was I who presented them. This cannot be the case for any reader with a philosophical turn of mind. Assuming that you are reading critically and not simply skimming in quick overview then I think it most probable that what your critical faculties are busy doing, as you read and reflect, is to ask two sorts of questions:

1. Does each line of argument hold water in itself? This is mainly a matter of logic.
2. Do the premises and conclusions fit in with what you know? This is the coherentist criterion.

Given 1, that a new argument or piece of information is logically self-consistent, what else must you be able to say about it for it to qualify under 2 as well? Perhaps the first thing I had better do is to drop the "you": an essential feature of coherentism is that it is personal, subjective, idiosyncratic and a matter of self-responsibility. So I am qualified only to talk about my coherent body of knowledge.

Images of knowledge itself have been proposed in several different forms.

• Some favour an ever-expanding area of light in the all-encircling gloom of ignorance: as knowledge expands the perimeter gets bigger and bigger, letting us realise increasingly the extent of our ignorance.
• Perhaps slightly less fancifully, one group of philosophers adopts the building metaphor: a foundation of sure and certain truth is identified. Then that foundation is used:
  • to deduce all the facts that can be derived by logic.
  • as a base on which to build by experience and learning.
  Unfortunately there is no unanimity when it comes to criteria for foundational knowledge, nor as to what makes up such a foundation. We have already seen (Introduction and Synopsis) that Descartes' cogito ergo sum was in this category.

Now it seems to me that the facts of the real world in which we live simply don't fit any such images. We have got to take account of several circumstances that rule them out:

• It is not possible for any person to know all that is known jointly. Perhaps in ancient Greece an individual might reasonably hope to know everything there was to be known about mathematics, politics, science, religion and philosophy, though I rather doubt it personally. It does not need stating that nowadays no individual can hope to keep up to date with the detailed work in more that one or two specialisms within one or two disciplines: the "generalist" is necessarily a picker of plums, a skimmer of surfaces, valuable though subject integration can be.
• Lots of things are initially counter-intuitive. "Counter-intuitive" is, by the way, a euphemism for "incoherent", ie., for not fitting in with the things that had been understood up to the time of coming across them. The main purpose of the preceding Chapters has been to give examples of explorations of the world which seem to me to have inevitable but counter-intuitive conclusions. By way of reminder, here are the main ones:
  • Sensing mechanisms are unreliable.
  • The interpretation of accurate sense data is unreliable.
  • The cost of running a car depends on what you want the information for.
  • Dice can be found which form a circle of superiority rather than the expected slope from best to worst.
  • It is a simple matter to produce a piece of paper with but a single side and a single edge.
  • No self-referencing negative paradox can be significant. Actually, that conclusion is a hypothesis rather than a fact.
  • The cuckoo's route to Africa is coded in a digital pattern of four different chemical bases.
  • A particle can take two routes from A to B simultaneously.
  • The 'normal', logical, cause-and-effect linkage does not operate at the individual quantum level.
  Plenty of other examples could be added. Familiarity does tend to breed acceptance, if not contempt. What is totally incoherent to one generation may be normal to the next: one has only to imagine Mozart's reaction to the ideas underlying a Walkman or Shakespeare's to satellite television to make the point. Similarly, those bits of theoretical physics in Chapter Five which seemed to us, mere people-in-the-street, to be so incredible must be daily normality to specialists working on such things. Here are a few more cases which may or may not be familiar but which do seem odd when we - when I - think about them.
  • There is an absolute limit on how cold things can get.
  • There is an absolute limit on how fast things can go.
  • The whole can be greater than the sum of its parts. A termite colony for instance can build a very sophisticated nest complete with air conditioning (controlling both temperature and humidity), orientated in the best possible direction relative to the sun which they can't see, on a plan which no termite brain could conceivably conceive and which no foreman insect could possibly co-ordinate in practice.
• The same is true at the conceptually much more modest level of the conjuring trick. Since I am not a member of the Magic Circle I am forced by the nature of the case to accept some sort of incoherent "fact" - some piece of knowledge that does not cohere with the other pieces of knowledge that I have already built into my personal edifice of knowledge. There are two possibilities:
  • My eyes were playing me tricks, ie., some significant error took place somewhere along the chain discussed in Chapter One.
  • The lady was indeed sawn in half but is not the kind of lady that I have previously known about since she could be halved without pain and rejoined without scar.
  Of the two, the lesser damage is done to my coherent edifice if I plump for the first, assuming that the "error" was in fact engineered by the magician.

So the problem of reliable knowledge is not a black-and-white one concerning some absolute truth. It is not even "Do I accept this new proposition as part of my knowledge?" or even "Does this new proposition cohere with my present edifice of knowledge?". Instead we have got to concern ourselves with probabilities: we are here, inevitably, faced with guessing, or plumping as in the conjuring trick example. What an abhorrent admission for any self-respecting philosopher to make! But how else might it be expressed?

§6.2 Criteria For New Knowledge

We need an example to get our teeth into. Let's think about the case of the poor unfortunate philosophers who were active during the period of the great debates about the earth or the sun as the centre of the universe. We need not list all the arguments, but the following will give the flavour:

• GEOCENTRIC THEORY
  It is intuitively obvious that the ground is stationary. I learn from the leaders of my religion that man is the most important thing in God's creation and therefore that man's home is the centre of it.
• HELIOCENTRIC THEORY
  Science has shown that planets have moons, so it is not the case that all heavenly bodies are fixed to crystal spheres centred on the earth as previously taught. The mathematics of the motions of the earth, moons, planets and sun is very much simpler if the sun is taken as the fixed point.

There could at that time be no independent logical proof of either theory. Most thinkers had been brought up on the geocentric theory, and had incorporated it into their own private coherent bodies of knowledge. But the newfangled replacement blatantly conflicted with it so it wasn't possible to hold that both could be right: a commitment was necessary, earth-centred or a moving earth and stationary sun. To accept ANY of the heliocentric arguments necessarily involved dismantling some of the existing edifice. To change metaphor, some significant chunks of the resident database had to be edited out.

And please don't argue that that sort of thing happened only at crucial points in history: contrariwise, it is happening every day. In the normal process of education - formal schooling or simply finding out about society - crude versions are learned first and are then repeatedly refined to varying degrees.

• In early schooling atoms are Dalton's little bits of matter: they later change to electrons and nucleons with post-1935 terminology. Components are later discussed in terms of more fundamental particles, which may in turn be displaced by the mathematical probability-energy fields of our Chapter Five.
• To a child the populace can be divided into goodies and baddies: we later learn to make more sophisticated distinctions. And we have already had cause to refer to the demise of Father Christmas.
• The young thinker may well consider that every meaningful question must have an answer: the philosopher has learned that this is not the case with any of the most significant questions.

So where does that leave us? Given a statement of the form "Floggits are hickly phanderoles" how do we decide - or to return to the personal case, how do I decide - whether or not to accept that as a piece of knowledge?

• If it slots neatly into what I already know about floggits and hicklyness and phanderoles then it is unlikely that any difficulty will arise. Indeed, it may well follow logically from what I have previously understood.
• If floggits are a kind of phanderole new to me then I may well not feel it necessary further to query their hicklyness. Similarly with the other possible combinations: given that two elements are already held in my edifice of knowledge, the new proposition could perhaps cohere without difficulty.
• But if the new proposition is contrary to something I already know - to something in my edifice of knowledge - or seems to be inconsistent or illogical then I have got to decide between:
  • rejecting the new proposition altogether.
  • putting the new proposition on "hold", awaiting further evidence or consideration. It may perhaps be pigeon-holed alongside some other related fact with which it does cohere.
  • accepting the new proposition and dismissing its converse. This is certainly likely to be the case with increasingly refined propositions about the same thing, when learning successively about atoms, electrons and quanta for example.

The time must come, perhaps when a new proposition is being pigeon-holed or possibly in the still of the night, when the contents of the pigeon-holes must come up for sorting and review. A more coherent and beautiful edifice may result if an existing section is sliced away and replaced with some fresh formulation based on new elements from one or a number of pigeon-holes. It is worth bearing in mind that this is what underlies all cases of scientific and inventive genius. Genius hardly ever discovers new objective matters of fact: that is the function of the slogger.

What genius does is to form or discover new relationships - new edifices. There may for instance be several propositions each of which is, on its own, plainly wrong, impossible, pointless, counter-intuitive, incoherent, or whatever. But put a number of them together and they support each other in the way that the poles of a wigwam do. As an example we might again think of the impossibility of there being any such thing as the square root of a negative number and yet finding that i, the square root of minus one, was so very useful. It provided a route to several insights in quantum physics in Chapter Five.

Again, it would be too simplistic to assume that once a proposition has been incorporated into the edifice there can be no room for another parallel one. For example, you may be curious about those orange-yellow street lights.

• A perfectly full and sufficient explanation is available in terms of causality - in terms of electrical excitation of sodium atoms, electron jumps and so on.
• But an equally full and sufficient justification can be given teleologically, as a matter of purpose - as the cheapest way of providing the light levels that people and safety demand.

That doesn't really signify anything very serious, but there are parallels in philosophical subjects too. This may not be the best place to discuss it further, but cause and purpose can easily be confused. The distinction took our attention in §4.2 in the context of a living process being caused to happen and/or achieving some purpose.

§6.3 Coherentism Grows

Deserting any ivory towers, the real world seems to be full of people. And it seems to me that the most coherent explanation of that "seeming" must be in terms of those other people-like images or whatever actually being people. And furthermore I'm prepared to accept that some of them are pretty competent scientists, philosophers, airline pilots, musicians, ....

Again, I personally am prepared to accept that -

• what you and I take to be books really ARE books.
• that they were written by real people.
• that in common with video recordings, photographs, databases and so on, some of them embody propositions which I ought sensibly to consider for inclusion in my coherent edifice of knowledge.

In short, it seems to me to be a coherent proposition that I should accept the testimony of experts. That's a relief: I shan't feel I have to prove quantum mechanics from first principles after all.

On the other hand, I can't accept all evidence from all authorities since experts differ. So I am faced by a second version of the original problem, namely how to distinguish acceptable expert evidence from unacceptable expert evidence. This has to be a matter of probabilities. I am more likely to respect the statements of a university teacher than those of a salesman. Extensions to orthodox thinking are easier to take on board than are eccentric novelties: but if the person putting forward the novelty can produce what seems to be sensible and logical justification then that's fine too.

So I give a high rating to propositions which come into any of several categories:

• They fit straight into or onto my existing knowledge structure.
• They branch out logically in a new direction and I don't already know anything that conflicts with them.
• Those that carry some sufficient form of logical authority so that a conflict with some existing part of my own database forces me to re-assess what I thought I already knew.
• Propositions which add new bits or re-structure what is already there and so improve the beauty, simplicity or coherence of my body of knowledge.

(Incidentally, these are pretty much the same criteria as are conventionally discussed in the matter of competing or successive scientific theories.)

It would be nice if there were a word to encapsulate all those four, but I certainly don't know of one. I suppose I could try and devise some imposing Latin- or Greek-based term but I should be likely to show my ignorance in the process. It is a very personal matter, and so the easiest way for me to rate propositions is on the basis of whether or not they are acceptably "hoskentic".

In the end, then, I can take as knowledge (for the time being at least) all those propositions which are sufficiently hoskentic.

• Some propositions will arise from sense data. But sense data will normally be accepted at first sight, at initial-interpretation value, only when the final picture is coherent: conjuring tricks are by definition non-hoskentic because I'm not one of the Magic Circle initiates.
• Some propositions will arise from testimony. In fact a very high proportion of information and knowledge comes in an indirect form through the speech of others, or even more remotely from its origin via text, diagrams and pictures on paper or TV/VDU. The degree of hoskence is compounded of the perceived reliability of the source as well as the coherence of the material.

§6.4 Knowledge / Belief

You may be feeling that my so-called coherent edifice of knowledge is nothing more than a pattern of beliefs. I am happy to accept that "coherent edifice" can equate with "pattern" though I think my term is a shade more appropriate, perhaps even insightful. We at least agree that the various components relate to each other in some structured way, whatever metaphor we choose. The more serious matter is whether or not "knowledge" equates with "belief". Philosophers tend to make a significant distinction between the two. Even the man in the street uses them differently, usually implying full agreement with "knowledge" but possibly having some degree of reservation with "belief": the latter smacks too much of mere "opinion".

To the philosopher there is a spectrum of certainty-words:

• OPINION is what someone thinks about something, with reasons which seem sufficient to him or her.
• BELIEF is a strongly held opinion founded on reasonable but not totally conclusive evidence. It is perfectly possible to be mistaken: one can believe something to be the case which turns out not to be so.
  • FAITH is a belief which is not unreasonable but its main justification is something other than pure reason.
• KNOWLEDGE is stronger and carries an implication that the proposition of which one has knowledge must be true: one cannot, by definition (or by the way it is used in the language), KNOW something that turns out not to be the case. We shall not attempt any fuller analysis. Relationships between them depend on definitions and definitions are (by definition) matters of opinion.

I am not unique in all this. Assuming that people in general are happy to make do with a large proportion of second-hand testimony it is possible for them - for us - to draw a parallel with accepted bodies of knowledge, such as perhaps the whole of mainstream natural science. By what criteria can science itself expand and grow? Only by internal self-coherence. C.A. Coulson made it clear like this -

"Scientific truth means coherence in a pattern which is recognised as meaningful and sensible. It is acceptable only so long as it does 'hold together', without internal contradiction, and is able to grow, either by the prediction of new phenomena or the absorption of old ones."

But if you want totally reliable, justifiable, irrefutable science that comes with a copper-bottomed, irrevocable guarantee then you're destined to be out of luck. Some creationists for example, in denying biological evolution "know" that when God created the world He included fossils here and there as an intellectual trap and test of faith for mankind. Just to pile on the agony and to illustrate the ultimate, there is in fact no way of proving that the whole universe was not created just a moment ago, complete with books and records, fossils and memories, and light created at the same time as if part way from distant stars. On that basis, then, no proposition can ever constitute knowledge for anyone: the best that can be hoped for is true belief.

So must we accept that coherentism is concerned (at best) with a pattern of true beliefs? It would seem so. But how do we know they are true? Answer: we don't - we can't. In fact it is the very impossibility of certainty about beliefs or knowledge that makes the whole discussion necessary. So if the fundamental question is about true knowledge it must be re-phrased: it is pointless asking for total certainty. The meaningful question is simply this:

In the absence of total certainty
what is the next best state of affairs
so far as knowledge / belief is concerned?

It seems as if my knowledge - we'll stick with that term I think - that my knowledge is simply that body of propositions which I have accepted (for the time being at least) as coherent - as holding together in a suitably hoskentic way.

I have not offered an actual definition of "coherent" for use in this context. And indeed even that is a matter of opinion, of idiosyncratic preference. Those suitably privileged assign a supreme probability value amounting to total certainty to propositions coming to them by what they see as divine revelation - a subject to which we return in a later Chapter. Some rate logic very highly while others feel obliged to pay greater heed to gut feeling, to intuition. For yet others, harmony and fittingness may be valued though less easily pinned down, let alone defined. If you feel that everything should be verified, that you need to arrive at the same proposition from two different directions, then that's your privilege: I may not feel the need so strongly.

If you don't like such a woolly conclusion, if you feel that it's all much too confused and subjective, well hard luck. If absolutely certain knowledge were easy to come by there would be no need for philosophy. It's too late to run back to Mummy when philosophy shows that you have got to be responsible for your own knowledge.

So is that it? Nothing about reality? Hardly a mention of truth? Verifiability? Reliability? No: they're all up to you and your version of what qualifies for inclusion in your body of knowledge.

So yes, that IS it.

< < < < < < < < < <+> > > > > > > > > >