Essay Four

I only encountered Guy's work in 2005, but I soon saw that he had anticipated several of my own ideas -- except he manages to express in two paragraphs what it takes me several pages to say! Indeed, in this paper Guy manages to convey in nine thousand words what took me countless thousands more (in Essay Twelve Part One and Essay Thirteen Part Two -- as yet unpublished).

On Misunderstanding Science

Guy Robinson

"You've seen to an almost unprecedented extent what I've been up to. I couldn't have identified my position so clearly at the time I wrote Structure."

Letter from Thomas Kuhn

I am going to lay my cards on the table and say that I don't think there is any room in philosophy for theories and theses. So I get nervous and suspicious when the 'isms' come marching by. One that makes me particularly nervous is "Scientific Realism". The reason for that is that I think that historical facts extraneous to both philosophy and to the sciences have been a major subterranean motivation for the belief in what might be called an 'ultimate reality' as a goal of, or limit on scientific work. If we describe that 'ultimate reality' as a goal in the sense of something worked toward, and as a limit in the sense of what it is that gives us something to measure our theories against and test their adequacy, then we can see that it is a very appealing notion, one that seems to solve a lot of problems at once.

Unfortunately, I don't think it is as simple as that and we can see why if we consider why we should be suspicious of the notion of theories and theses in philosophy. Theories, theses, beliefs and opinions are all propositional in character. That is, they say how things are and how they might not be. That is what gives them their true/false status. In relation to that we need to ask whether the world could change in any way so that Scientific Realism would become true (having been false) or so that it became false (having been true). Since pretty obviously it could not, we have to say that Scientific Realism is not a theory or thesis and that something else is going on here, something that is being obscured and misrepresented.¹

To get at what is going on and the nature and source of the obscurities I think we are going to have to go a long way round, to look at an alternative way of analysing scientific development and progress and to consider what kind of work the notion of reality does both in the sciences and outside of them.

I want to start by looking at the radical shift in viewpoint, method, style and conception of what it is to understand the sciences and scientific development proposed by T.S. Kuhn in the 1960s. I want to try to bring out just how radical a challenge he was issuing to a conception of the sciences themselves, what it was to account for their development and what progress in them consisted in. The conception which he was challenging can really be said to have dominated our era from the time of Descartes.

Thirty years ago in his Structure of Scientific Revolutions, Kuhn made a plea for a new conception of what it was to understand science and scientific development and progress, one to replace the conception of the task as that of supplying a logical model of that development and progress. That 'logical model' conception had dominated the field and defined its aim from the beginning -- and still largely does. And we have to ask why? And we have to ask why the historical conception that Kuhn proposed was largely misunderstood and then buried and tiled over with 'isms,' so that Kuhn is now often seen through the wrong end of a telescope, a reversed telescope, moreover, that is focussed on a distorting mirror. That is, his careful argument and analysis, his setting out of the concrete detail of the circumstances and the arguments surrounding several crucial moments of scientific development and his demonstration of the inadequacy of the usual 'logical models' to make sense of them, have all been boiled down and evaporated and reduced to some kind of 'ism' -- in this case 'irrationalism' or something of that kind.

All reductions of that sort, the reduction of a complex argument to an 'ism,' involve turning the telescope around and looking through the wrong end. One can see the advantage of this kind of procedure in that it allows you to avoid dealing with the detailed arguments of the position that has been so reduced and abstracted, but that single advantage doesn't seem to be enough to account for the current dominance of a style of philosophic argumentation that can be described as 'ism shuffling'. Perhaps an historical account is needed here too.

But of course, the notion of an historical understanding is one that is profoundly alien to our era. For that era, scientific understanding has been the dominant mode, to the point that other forms of understanding are thought to need legitimating by a reduction to the scientific. We at least think we know what scientific understanding is -- even as we struggle to give an account of it with no evident success. But the notion of historical understanding is quite beyond us and our impulse is to try to reduce it (in the manner, perhaps, of Carl Hempel) to a poor sort of scientific understanding that appeals implicitly, and perhaps illicitly, to hidden laws and generalisations that would 'explain' things if only they could be identified and formulated. Hempel simply assumed that there was only the one kind of explanation (derivation or deduction from a generalization or law) and took his task to be that of showing how history and historical explanation could be reduced to that model. If you have ever seen history written in that style you will know what dreadful stuff it is, cluttered up with gaseous generalisations that are utterly unconvincing and easy enough to shoot down.

If we could get over the initial block, perhaps we could have a historical explanation of why it is that the notion of a historical explanation is so difficult for us. I think one is available and I have hinted at one in outline before now, but that is rather a long story that involves some alien notions and I wouldn't expect it to be taken in in one go.

The reaction to Kuhn's work and the particular kind of misunderstanding of it that was exhibited are a very nice illustration of a process that Kuhn was himself describing and documenting in the sciences. It was the dominance of the very model of understanding that Kuhn wanted to challenge and replace that made it impossible for people to take in his message and to understand his proposal.

The assumption has been that there is only one kind of understanding, one that finds the reason for the thing in some larger or more general principle. Understanding something is to have a theory of it that derives or deduces it, sees it as an example of, the more abstract and universal truth. Simply because Kuhn purported to be offering us a way of understanding the sciences and their development, it was assumed that he was offering a theory of scientific development, and for that reason some 'ism' had to be found that would characterise his message. "Subjectivism," "irrationalism," a "sociology of science" and even a "psychological theory" of scientific development were among those attributed to him -- apart from that old favourite "relativism."

All these bogeymen have been conjured up to threaten scientific rationality itself and placed between us and Kuhn. Or rather, Kuhn has been dressed up in these bogeyman clothes, given these masks to wear and made to threaten that scientific rationality with its promise of objectivity and truth. And if the sciences can't give us the truth, what can? It is all too terrible to contemplate. We have a lot invested here. It is the secular equivalent of the denial of God.

To understand the forces making for misunderstanding and rejection one has to get a measure of the full depth and nature of Kuhn's challenge to the reigning paradigm and conception of what it was to understand the sciences, the conception that gave rise to the project of logical modelling. For Kuhn's challenge was completely subversive of that paradigm, and the changes of aim, outlook and practice that he was calling for were total, a reversal of the previous account.

Since Descartes, a project has been laid before us that has almost defined that branch of philosophy, epistemology, that came into being then, and since then has shaped and directed our search for an account and understanding of the sciences, a theory of their development and progress. The sciences have been the focus because they have been for us a paradigm and the chief hope for obtaining that certain and objective knowledge sought in the post-feudal era, a vision of knowledge which was central to the enlightenment dream of universal human emancipation.

That project was exemplified in Descartes' Meditations, and it laid two demands on any account of knowledge and the means to knowledge, demands that set the standard and defined the adequacy of any account. There had been urgent reasons for making those demands but the reasons were historical rather than philosophical and came from the individualistic model of humanity that played such a pivotal role in the era's project of eliminating feudalism's remnants in thought and social institutions, and the project of justifying the conceptions and arrangements that were replacing them. That story needs to be elaborated, and will get some elaboration in the next chapter. What is important here is that those demands have been accepted since without serious critique or examination of alternatives.

The first of the demands, describable as a "democratic" or "individualistic' one, was that a method be found that was available to each separated individual to apply privately and severally in the search for knowledge. The second, relating to the knowledge thus found, was that the method would lead all who conscientiously applied it to the same, objective and timeless true view of things.

The historical task here taken on was that of finding a secular alternative to Christendom as means for bringing unity and a common focus to an increasingly fragmented world of separated individuals. This task was an urgent one in the post Reformation world that faced the individuating and secularising force of the market without a single, unifying Christendom and Church to counterpose it. But that story, as I have said, needs elaboration.

For the moment I would simply like to note how these two demands, the "democratic" or "individualistic" and the "unifying" are implicit in the aims declared by the many books in this century with the titles The Foundations or The Principles of this or of that: The Foundations of Logic and Mathematics (Carnap), The Foundations of Science (Campbell), The Foundations of Empirical Knowledge (Ayer) etc., etc. The point of foundations is precisely to be fixed, immovable, and as far as possible, eternal and the same for all. The image of "Principles" is procedural rather than architectural, but the thrust is the same. The object of setting them out in a book is to enable the individual reader who grasps them to proceed by their means to a timeless and objective true view of the world, a view that, because 'objective,' will be the same for all.

Now we have to look at the way in which Kuhn's analysis and account of the development of the sciences demands a complete rejection of both of these demands -- not only setting them at nought, but turning them on their heads. Kuhn asks us to see the sciences not as abstract structures of principles and deductions, of observations and postulates that each individual could in principle discover individually, but rather as a human activity and moreover as a communal one. He goes on to spell this out by adding that this human activity is not, as the reigning paradigm demands, the slow and painful uncovering of the divine order (as it were), the objective, eternal laws, order and truth, that transcends the secular world of growth, decay and change that we inhabit.

This human activity is, on the contrary, a historically conditioned one, one that is carried out against a particular concrete background of existing theories and models, vocabulary and equipment, hopes and expectations that may come from adjacent fields, prospects and projects of consolidation as well as being shaped by the standards of accuracy and comprehensiveness reigning in the field at the time.

All of these are the historical legacy of the immediate past, its successes and its failures, its projects and its hopes. They form the shaping environment in which any particular scientific activity is carried out. The experimenting and the theorizing do not stand in any strict logical relation to that environment, but they are shaped by it nevertheless in ways that are hardly irrational (so that the charge of "irrationalism" against Kuhn can itself only be described as "obfuscating and unhelpful").

Perhaps the best model of what is going on is the way in which the hand is guided by the model or the scene in making a drawing. There is no mechanical or strict determination going on. (Though one can describe some draughtsman's work as "mechanical," this is an "as it were" meant to indicate that habit and technique have come to displace sensitivity and freshness of vision.) But the hand that is being guided is the hand of an artist with perhaps years of looking and drawing entering into the way in which the eye sees and the hand moves in tune with it. (Setting this down now, I am reminded that many years ago, Kuhn told me that he wanted to turn his attention to art as a way of illuminating the sciences. Perhaps he had just that point in mind. At the time I didn't see what he was after.²) The image seems to me to be a useful and illuminating one even though it will not give us the basis of a theory or theoretical account and so will not satisfy those whose vision is limited by the dominant paradigm of understanding as theorizing.

Apart from anything else, Kuhn's radical and subversive suggestion and way of looking has the effect of dethroning the scientists as priests and theologians of this transcendental order (whether ordained of God, or just there) whose job is the uncovering and interpreting that order to us. That itself would account for some unpopularity and resistance.

To be fair, those pretensions to priesthood are not found much among the general population of practising scientists, though there is sometimes a natural increased concentration among cosmologists and fundamental particle physicists. The very conception of their task as searching for ultimates and origins makes them prone to those pretensions. And there are, of course, the 'Sunday scientists' whose stock-in-trade and delight is making just the sort of grand metaphysical interpretations that will impress a gullible lay audience.

But what if Kuhn has not been trying to give us a theory of anything? And of course he is not. His message is that theorizing is the wrong way to go about trying to understand the sciences and scientific development. The various theories of scientific development, attempts to discover universal rules, principles or tests that are supposed to define scientific rationality and account for the development of the sciences, haven't worked and can't work as a description of the historical developments we have before our eyes. His message is the Wittgensteinian one: "Look and see."

But in an era dominated by the scientific model of understanding, "looking and seeing" is regarded as only a preliminary to theoretical analysis, its role limited to providing the data that a theory will have to account for. Without that theory, the dominant view is that there is no understanding.

Kuhn makes an analysis of various crucial historical developments in the sciences all right, but it is not an analysis that issues in a theory that gives us what has been so long sought -- a logical model of scientific development and progress, an account of scientific rationality that fits our dominant model of rationality as logical and deductive. On the contrary, the analysis that he makes shows that such a logical model of scientific development and scientific rationality is not to be had. Of course scientists use logical and deductive methods -- who doesn't? But Kuhn's analysis shows that there is a characteristic pattern of development in the sciences which at certain historical moments involves critical readjustments of viewpoint and the introduction of new procedures and styles of analysis, readjustments and innovations for which these logical methods and arguments are not determining.

Kuhn's suggested way of looking at the moments of radical change in the sciences brought to the centre of the stage the notion of a paradigm, the notion of a crisis for a given paradigm and, most importantly the notion of a paradigm shift. There has been much huffing and puffing about the notion of a paradigm and at one time acres of journal-space were devoted to it. For some it was the key to the universe, for others a major piece of obfuscation -- mainly because of Kuhn's use of it to show that the really crucial developments in the sciences are not available to logical modelling.

Where the Cartesian project sought a method that would lead separated individuals to a common view and to community, Kuhn says that a community of view and of practice is a condition and a requirement of progress in the sciences. The notion of paradigm is involved here because the acceptance of a common paradigm and way of seeing is what makes communication possible amongst a group of scientists, that acceptance is the source³ of common practices and vocabulary which make them into a community and therefore makes progress possible through mutual criticism and discussion, through the repetition of results and experiments, through the adoption of procedures and practices that prove successful -- or, just as revealingly, run up against limits and encounter failures.

The notion of a paradigm is simple enough, however, and, so long as your project is one of clarification and not theory building, it will not run you into complexities and difficulties of definition and so forth. It won't bear the whole weight of a theory, and Kuhn doesn't mean it to. (Before long we are going to have to face and examine that conception of understanding as limited to giving a theoretical account.) The reasons for the unavailability of logical determinations in "paradigm shifts" are themselves logical. A key is perhaps given in Wittgenstein's discussion in Philosophical Investigations II, IV of the possibility of regarding someone as an automaton: "My attitude to him is an attitude to a soul. I am not of the opinion that he has a soul." My attitude is a matter of my whole demeanour toward him involving the rights and duties and customs of my culture. Not something that is simply true or false like an opinion, a matter for argument and proof. Paradigms, attitudes and ways of looking are not matters of proof and disproof.

To get at the notion of paradigm, take the example that Newton gives to illustrate his Third Law of Motion: that every action is accompanied by an equal and opposite reaction, or, as he himself puts it: "To every action there is always opposed and equal reaction: or, the mutual actions of two bodies upon each other are always equal, and directed to contrary parts." As the sense of this is not obvious to someone coming fresh to it, as were Newton's contemporaries, he offers an illustration. That illustration is, in effect a paradigmatic example meant to give the sense of the law, to show how it is to be applied:

"If a horse draws a stone tied to a rope, the horse (if I may so say) will be equally drawn back to the stone; for the distended rope by the same endeavour to relax or unbend itself, will draw the horse as much toward the stone as it does the stone toward the horse...".

I have italicised the phrase "if I may so say" because this signals the status of what, because of its misleading propositional appearance, might easily be mistaken for a proposition or claim with a truth value. But Newton is not offering us a claim or a theory that is testable and has a truth value. The phrase "if I may so say" indicates that he is clear that he is offering us a way of looking at the situation, a method of analysing it, rather than a testable, falsifiable claim about it.

And here we have to say that Newton was a lot clearer about the status of what he called "axioms" and "laws" of motion than were later generations who looked on them as universal, and perhaps providential, truths about the cosmos. It took Henri Poincaré's hard work and careful analysis to bring out the fact that what was perhaps the most promising candidate of the three laws for empirical status and testable content, the Second Law -- nowadays rendered as "Force equals mass times acceleration," -- was not in fact a testable, falsifiable claim about the cosmos or the things in it. Poincaré showed that there was no way of measuring each of the three components, the force, the mass and the acceleration independently in any concrete situation and that therefore no experiment could bring the law to the test. And so too for the other two of Newton's three "laws" of dynamics.

Poincaré concluded that what looked like a law or theory, and had the outward form of a testable proposition was in fact a disguised definition. Naturally he got boxed into an 'ism' for his pains -- I have forgotten which -- and is these days only referred to as the representative of some 'ism' or other. I recommend forgetting the 'ism' and going back to the actual argument beautifully set out in the chapter called "The Classical Mechanics" in Science and Hypothesis.⁴

As will be suspected from what I have been saying, I don't myself think that "definition" is the most perspicuous way of describing the Newtonian "laws" of motion but would suggest "way of looking" and "method of analysis" would give us a clearer picture of what is going on. What is common to both descriptions, however, is that they bring out the fact that those "laws" of motion are not propositional or theoretical in character. They are not true or false, are not candidates for testing on that true/false axis. They are guides for practice that stand to it much as the model or the scene stand to the drawing. (And my remarks themselves are not theoretical or propositional in character. They are not up for testing and refutation. Though you may persuade yourself and come to the conclusion that they are, or else are not, a valuable way of looking at things.)

Methods and ways of looking are not propositional in character. They are not true or false and we create great confusion for ourselves if we try to assess them in that way or mistake a method of analysis for a theory, a proposition or anything else two-valued and testable. We do assess methods and ways of looking, but in ways that are generally scalar rather than bivalent. The Pragmatists merely confused matters by trying to bundle everything up into one package and making no distinction between propositions, methods and ways of looking. Apart from anything else, the Pragmatists' "usefulness", being scalar, is for that reason not appropriate for giving an account of propositions, which are manifestly bipolar. Usefulness may be an appropriate way of assessing methods, since generally a single a goal or thrust is defined for them. For ways of looking, by contrast, there are many other scales on which they may be assessed. Ways of looking may be deeper, more sensitive, funnier, enlightening, unifying, more human, more rigid, more puritanical, more self-serving, sexist, etc., etc., etc.

Now, let's get back to Kuhn's "paradigms" and what is meant by calling them "incommensurable" by way of signalling that logical canons are not available which will force a decision in favour of one of two or more competing paradigms. Of course, one can use logical canons to make a decision between two paradigms as one might use anything else, tossing a coin, for example. But that way of making the decision is not the way the scientists work and it would not advance the sciences any.

The notion of incommensurability seems now to be even less well understood than it was in the 1960s and 1970s when Kuhn's work provoked a great deal of journal activity. Hilary Putnam,⁵ for example reduces it (or, if you like, expands it gaseously) to a claim about the untranslatability of the theoretical vocabulary of different scientific "cultures". He then produces a 'knock-down' argument that reminds one somewhat of Anselm's way of dealing with "the Fool who in his heart hath said 'there is no God.'" That is, Putnam fastens on what it is that is supposed to be getting denied to show that the very notion of it subverts the denial. The argument is that we can't assert that there is a gap or disparity between the two vocabularies if we can't say something about the gap, and that is to say something simultaneously about both sides of it, and to do that requires a vocabulary which bridges it. -- Q.E.D.

A snazzy little argument. Unfortunately, like most snazzy little arguments it doesn't reveal much to us. Snazzy arguments have too short a way with things to be much use in philosophic clarification. More important here is that it doesn't really touch Kuhn's notion of incommensurability because that incommensurability lies in the first instance between paradigms. Since these are ways of looking at things and related practices and methods of working, two different paradigms don't lie in the same logical space. They are not true or false any more than building in steel is as against building in wood, stone, concrete or reed. One or other may be a more appropriate building technique given the available skills or materials, the use of the building or appropriateness to its surroundings, or the exigencies of wind and weather, and so forth. And here again we have a range of scales for assessing the activity, project or technique.

Any incommensurability of vocabulary between different scientific practices derives from the incommensurability of the paradigms that guide those practices in which the vocabulary is embedded and from which it draws its sense. The disparities in vocabulary are not the cause, basis or reason of that paradigmatic incommensurability. Things run the other way round and the incommensurabilities of vocabulary derive from that incommensurability in paradigms and ways of looking.

The piece of incommensurable vocabulary that Putnam takes as an example and then proceeds to hammer is that of "temperature" as used by the Seventeenth Century scientists with only proto-thermometers to measure it by and the word's use and meaning in our own time in which there a range of methods to measure it. This turns out to be rather a soft target that Kuhn offers up by certain things that he incautiously says about changes in the notion of mass from Newton (for whom it was conserved) and Einstein (for whom it was not.) What Kuhn says can be said to be "incautious" only in that it offers up this soft target to an unsympathetic interpreter, as Putnam is here. Putnam would not have had such an easy time bridging the incommensurabilities if he had been considering differences in vocabulary, viewpoint and practice embedded in, say, the homeopathic model of disease and treatment and those of what is called "the medical model" by the people who view it from amongst the alternative strategies.

It might have been better still to have considered the vocabulary gaps revealed in the story of the Indian of Northern Canada who had no difficulty believing his informant who told him that as a result of the Apollo Mission men had been to the Moon. He was in no way surprised or particularly impressed. "My uncle, the Shaman," he said, "has been to the Moon many times."

Finding a bridge between the vocabularies and the life and whole worldview connected with those two ways of talking would not have appeared such a simple business. One can, of course, take the easy way of the hard-nosed and say that the Indian and his people inhabited a delusional fantasy world in accepting the notions of shamanistic travel to the underworld or to the heavens.

That is, of course, an option. What it does not do, however, is to find a bridge between two incommensurable outlooks and ways of talking. It simply denies sense to one of them. And, given that that way of talking of the ascents of the shaman or his descents into the underworld is rooted in a whole way of life and set of practices, that denial of sense to it is not very persuasive or convincing.

The task of bridge building between incommensurable outlooks and associated vocabularies is in no way made easier by the disparities of viewpoint and ways of looking being contained within a single individual. A nice case here is that of the young Trobrian Islander who went to Port Moresby to the University there. When he was in Port Moresby engaged in his studies and the university world, he said, he knew that ghosts didn't exist. When he was back home in the Trobrian Islands, he knew that they did. They were part of the world-view and the practices of the society that he had re-entered. To ask which of these seeming incompatibles he really believed is perhaps not helpful here. Though it certainly is possible to recognise when someone has rejected a point of view and practices from which a certain vocabulary draws its sense and its strength leaving that person mouthing formulaic words that no longer have life or force.

With our Trobrian Islander it could go either way. He might become so identified with the Port Moresby world that he found himself unable to re-enter his native world to take a genuine part in its rituals and practices and ended up merely going through the motions, like an unbeliever in church for whom the invocations, the prayers and the practices have no meaning and are only a dumb-show. On the other hand, he might be so pulled by the tribal identity and solidarity that the urban, essentially secular viewpoint of Port Moresby lost its hold. I have no information about the subsequent history of the actual individual interviewed. He might even have continued to move between these two irreconcilable points of view.

Each of us employs a variety of vocabularies with disparities of that sort across which bridges cannot be built -- political, moral, medical, aesthetic, religious, scientific vocabularies, all associated with and taking their meaning from different areas of our life and practices that have quite different places there. On the whole these different practices and ways of viewing the world are not in competition with one another, except that they may compete for our time and attention. On the whole they have not got vocabularies that have the deceptive appearance of overlapping -- though there are certain notorious exceptions -- words from different practices that have the same form and enough analogy between their functions in those practices that they get taken to have exactly the same function and meaning. Much confusion results from that identification and literal transfer, confusion which generally needs careful philosophic unpicking.

Transfers from one such sphere to another can be made deliberately for poetical effect, or to make a joke. Transfers that are made without knowing or noticing are a source of confusion rich enough to supply philosophy with its raison d’être. An example is the confusion endemic in the sort of AI talk we looked at in Chapter Five and Chapter Eight, talk which involves the unexamined transfer of the notion of purpose from the human to the mechanical world and the description of certain machines as though purposes can be attributed to machines in the same way that purposes can be attributed to people. But the purpose that a machine has is given to it by its human designers or its users, and may be withdrawn or changed if the use of the machine is forgotten or a new use is found and preferred.

The reduced and derivative nature of that transferred attribution of purpose to the machine infects the attribution of all the other 'human' attributes up to and including 'intelligent' and 'creative' and shows them all to be transferred and metaphorical uses, whether poetry, joke or mere confusion. The 'creative' machine or program is not creative on its own behalf and according to its own lights, but only according to our aims, purposes and standards. That fact is overlooked in the eagerness to transfer the human epithets to the machine. Only confusion results from this even though all the participants in the 'great debate' about AI know at some level that they are only kidding one another. The underlying form of the debate is: "Prove to me that I'm kidding you." What I mean by that is that no one of the participants is willing to "put his money where his mouth is" and to show that he really believes in the 'sensitivity,' 'intelligence,' or whatever of the machine by behaving in a way appropriate to that belief. If he were to, he would be carted away.

It is Descartes who has licensed this kind of stuff with his method of "hyperbolical" doubt that asks us to "employ" (whatever that may mean) doubts that no one feels or could act on without being rounded up and put into an institution. I regard Descartes as having done immense damage to the practice of philosophy thereby, and with his "method" having created problems wholesale that his method was incapable of solving, and, more seriously, having infected philosophy with a kind of insincerity that has turned it all too often into a dumb-show, a stage-battle between puppets, abstractions and 'isms'.

But here we are getting away from the main line of our problem which was to elucidate the notion of paradigm as a way of looking and as a guide to practice, pointing out how paradigms are necessarily incommensurable in the way practices are. They may be competing all right, but in that competition they will get assessed on a whole range of scales and not simply as true/false, in the manner of beliefs, opinions, assertions, assumptions or anything propositional in character. Practices and ways of looking can't be measured on that single true/false scale and as a result logic and observation will not settle the choice and force the agreement amongst scientists. One can recommend a way of looking or a practice and point to a whole range of different and completely disparate things may make it attractive (or not.)

Perhaps now, after some of the things that have been said, it is possible to have a clearer grasp of what Kuhn had in mind in saying that community and agreement among scientists, a common vocabulary, practices and standards are a condition of progress and fruitful work and not a consequence of it.

But one has to be careful here, because the relations are not ones of strict logical priority but the characteristically zig-zag dialectical relations of a concrete historical development. That is, at one stage the rough agreements in practice, vocabulary and standards and conception of the problems, that have come from the allegiance to a single paradigm, make possible fruitful work and successes (or significant failures) that in their turn lead to finer and closer agreements amongst practitioners, and so forth. If, by contrast, we pursue the 'foundational' approach, looking for strict and absolute priorities, we are only heading for confusion. Actual historical development is not like that and cannot be made to fit that requirement.

But now we have to face the problem of scientific progress, a problem which is thought to arise particularly acutely for Kuhn because of his presentation of the sciences as a historical human activity and his emphasis on internally set standards and norms, his emphasis, reminiscent of Wittgenstein, on agreement in paradigms, practices, vocabulary and even judgments. These may be what it is that welds the scientists into a community, and enables them to communicate with one another, but doesn't that leave their activity without any external check and therefore any external reference? They may be able to compare results because of this ability to communicate, but if this ability requires a common way of looking at things and this common vocabulary requires "agreement in judgments" aren't we left with an internally organised activity which has no more external reference than dancing has? Wittgenstein, for his part tells us only that this requirement of "agreement in judgments" for the possibility of communicating "seems to abolish logic, but does not really do so." (PI §242) He leaves us there.

What is worse, Kuhn even suggests that at certain historical junctures, the development of some particular science might have gone another way. Does that view of the sciences not remove from them any claim to an external, objective reference? I don't think so, even though a first view may give the impression that the sciences are a cosy old-boy network admitting or rejecting whom they want and settling things among themselves. We will have to come back to that. But first we will have to say something about the problem of scientific progress itself, because the notion of progress itself has its own entanglements that make it pretty well intractable for almost any view of things.

The first cause of trouble we have to look at is a conception of progress that is powerful and has exercised its fascination since Socrates used it as an argument for the existence of the Forms. This conception of progress sees it as necessarily involving a final goal and the nearer approach to that goal. That conception underlies and gives urgency and a particular cast to the whole debate on "Scientific Realism." It requires us to think that there is something that one might call "ultimate reality" which in some way or other the sciences are approaching.

There are two things to be said about that. One is that though progress is sometimes like that, defined and assessed in terms of a goal aimed at, it is probably less often like that than it is a matter of an improvement over a previous condition. If I have progressed in drawing or playing the piano or running or doing philosophy there is no implication of an ultimate skill I am approaching. If I am making a table, that is a different matter. There is a point at which it is done, and my progress in the task of making it is measured by that: "It's only half-done," "Nearly done," "A couple more hours"...etc.

The other thing that must be said about that conception of scientific progress as an approach to some ultimate true picture (or whatever 'ultimate' you want) is that if it is taken seriously and not just as a metaphorical, allusive, poetical, or (dare I say it?) religious way of speaking, then it seems that coherent sense cannot be made of it. For one thing, we have to notice that the would-be 'goal' is a transcendental one and beyond reach. There could be no point at which the sciences or the scientists could come actually to have achieved it, come to have grasped that 'ultimate reality' in all its fullness. That notion can't be given sense. But then, if the end or goal is not achievable, what sense can we give to the notion of approaching it that was meant to be involved in our idea of scientific progress?

In mathematics we can, of course, talk about an asymptotic approach to something, one that never reaches the thing however long continued. We can talk about it there because the limiting value, the asymptote, is defined and therefore the notion of approach is defined; we can talk about getting "as close as we please" and so forth. But in this would-be notion of the progress of the sciences toward an unreachable goal we have not got, as we have in mathematical case, an independent means of defining, describing or characterising the goal in a way that will allow us to give any sense to the notion of approach. With the 'approach' notion of progress, scientific progress becomes an intractable problem for everyone.

Popper had a sense of this difficulty, which arose for him because in his attempt at logical modelling he had only the two categories in which to put theories: "falsified" and "not yet falsified."  This simple division was incapable of giving him any measure of progress. To get round this he tried to introduce a scalar notion of verisimilitude which he wanted to define in logical and mathematical terms. Roughly, verisimilitude was to be measured by taking all the deductions from the theory that turned out to be true and all the deductions that turned out to be false and noting the excess of the one over the other.

As a measure it ended up in that never-never land to which the logical modellers often retire. The notion of all the true deductions or of all the false deductions are never-never land notions. The operation could never be carried out because the would-be classes are infinite and, in addition, they are not infinite in a way that would allow comparisons between those classes on the model, say, of Cantor's 'Diagonal Argument.' That is, the notion of verisimilitude that Popper offered turned out to be not only unworkable but meaningless. When I pointed that out in 1971 in Analysis I got a chilly reception from the Popperians, who, after much argument among themselves came in the end to the same conclusion.

But in concentrating on the impossibilities and inappropriateness of the mechanical test I see now that I missed something deeper and more important about Popper's notion of verisimilitude, something for which he should be given much credit. It lay in the aim of that notion which was to give us a test of improvement so that we didn't have to say that one theory was "better than another" because it was "closer to" some mythical "ultimate or final picture." Though Popper did not provide us with a useful answer, he did something deeper and more important. He provided us with a more useful question or project in directing us away from the 'approach' notion of progress toward the 'improvement' model.

Because of the impossibilities in the 'approach' notion we have to scrap it and to try to get a better grip on the 'improvement' conception. This seeks to define progress in relation to what has gone before instead of looking vainly to an ideal something lying in a future that we know will never arrive.

But that change of direction removes one of the supposed functions of the notion of reality and we need to look hard at the work that the notions of the real or reality are actually given in the daily working of the sciences.

The answer is "not much if any." It is not in itself a substantial term. Its role, as Austin pointed out in Sense and Sensibilia is pretty well confined to negating other terms with real content such as: toy, stuffed, simulated, imitation, fake, artificial, makeshift or poor. Even those old favourites of epistemology: dream, mirage, hallucination, illusion, are all specific in themselves and are not to be bundled up into one bag as they used to be by the epistemologists, and defined as perceptual lapses from the real. A dream is not an illusion, and a mirage is not a hallucination. They all have their separate senses and work to do whereas the notion of real has only the job of denying them and gets what sense it has from what is denied. The gun that is real as against imitation is one that can be fired. The scotch that is real wasn't made in Japan. The leather that is real was not produced from petroleum.

In the actual working sciences the job of the notion of the real is chiefly to contrast with the suggestion that some result was 'an artefact' of the experimental situation. For a time chemists pursued something that came to be called "polywater" -- a form of superfluid water that appeared in minute quantities in certain experiments. In the end they decided that it "didn't really exist," that it was an artefact of the experimental situation. Though you can say that polywater turned out 'not to be real,' that is a less helpful and clear way of expressing things than it would be to say what was thought to be a new form of water turned out to be an artefact of the experimental situation. Much the same goes for the recent furore over 'cold fusion,' which was itself interesting in that it tended to divide the physicists from the chemists, who were more prepared to consider the possibility. The ungenerous have suggested that this was because the physicists had invested huge amounts of time and oceans of money in the plasma approach to fusion. They too had had that same artefacts problem that had led to premature claims of having achieved the production of energy by controlled fusion. For the moment, the consensus on cold fusion is 'artefact.' However my grapevine sources tell me that the Japanese have moved in to bankroll Martin Fleishmann's work, so that maybe the tallies are not all in yet.

I think I'll stick my neck out here and say that the working scientists could manage to get on with their work quite happily without the word 'real.' You could put it mischievously and say that there is no real use for it in the sciences. It is only the ideologues of science, the Sunday scientists who can't seem to get on without it.

My own preferred term for the task of trying to capture and represent the 'externality' of the sciences and scientific activity is 'objectivity,' which for several reasons I think does a much better job of bringing out the nature of the checks and limits that give scientific activity its external reference than does a notion of the real or ultimate reality which lies in splendid isolation in the misty distance, never to be really grasped. Part of the problem with the notions of the real or reality is that they carry with them the inescapable picture of being 'pre-formed,' so that to grasp the real properly is to grasp it only in one way.⁶ This leaves us wandering always in darkness in the sciences, because we can never actually reach that "one way" and, as we have  made out above, we can't even give sense to talk about "approaching it."

The objective on the other hand, does not lie in the unreachable distance. Quite the opposite: it is what is 'thrown down' before us, what we trip over and knock against, on the one hand, or pick up and use, on the other. That is one side of it. The other side is its relation to our projects. What we trip over is a matter of where we are going and what we pick up a matter of what we are up to.

This introduces something that the notion of reality and cognates miss entirely, namely, the historical dimension of the objective. What I have in mind here is that the projects that are possible to us develop over time with our past achievements and successes, both as individuals and as cultures. Nuclear fusion was not a possible project for the Greeks nor bridging the Rhine for the Neanderthals. Playing Scarlatti or climbing Everest are not projects open to me at my age, even though they are projects that exist and are recognised in my culture. We will have to come back and make use of that notion of a project being established or existing in a given culture.

Since the job description for the notion of reality was to be eternal and without history as well as single, presenting the same face to everyone, we have really two tasks here: one is to try to make out whether anything could carry out that role for the sciences and the secular world without muddling matters and ending by giving the sciences a quasi-religious role; the second is: whether the job needs to be done at all and whether, without it, we can put together a view of the external sources of both limit and possibility. These are not small tasks. For the notion of the objective has generally been given that very same old timeless and abstract role as the real, and it is hard for us to think of anything as objective that isn't the same for everyone and the same at all times. I think there is no hope of detaching the notion of reality from that job and those characteristics. It seems to have a natural affinity for the word "ultimate" and a hankering for theology.

On the other hand, I think there is some hope for the notion of the objective -- so long as we start from the conception of it as "what we encounter" in our projects and activities and work out from there. But we will have to keep in mind constantly that what we encounter in our activities and our projects is from one side determined by them, by the practices that are embedded in those activities and the way of life and the way of looking in which they in turn are involved. What we encounter must not be treated, as reality has, as a notion that pretends to an abstract sense that is independent of and detached from those human contexts of practices and way of life. It is precisely that pretence that removes the notion of reality from any role, place or sense within the secular world and projects onto it a quasi-theological role.

These tasks are, however, implicit in Kuhn's project and vision, and we have been circling around them all along. The unwillingness to take them on board has been the reason for the distortions and misunderstandings of his work.

Notes

1. And it is no good shifting the ground and saying that the philosophic 'theories' and 'theses' are not really about the world but about language as a reflection of the world. Any facts about language -- those things that are but might be otherwise -- would not be a matter for philosophy but for linguists, the historians of particular languages, morphologists, etc. Philosophers will be interested in the aspects of language that are variable and not necessary, but in a negative way, as Michelangelo had to interest himself in those parts of the block of marble that were obscuring the statue imprisoned within. What is hidden behind the changeable in language could be called, in Wittgenstein's term, the 'grammar' of the notion of language itself.

2. Kuhn later told me (1994) that he doubted that he was making that point at the time of our meeting sometime in the mid '70s, but that he agreed fully with it and thought it important.

3. It is just here that we have to resist the pull to specify the relation between paradigm and practice in logical or causal terms. It is just here that the image of model and drawing is perhaps most useful. That image helps to emphasise the two-sided nature of the relation.

The model does not determine the drawing or cause the movement of the artist's hand and the relation of the lines. But those movements are not unconnected with the model either. The connection involves the aim and the skill of the artist. The skill and vision that goes into making that connection has itself a history of development that may be long and complex. And anyone who looked for a causal connection between that history and that skill would convict himself of fatuousness.

The story is told that when someone asked scornfully of one of Matisse's simplest and most pared-down charcoal drawings, "And how long did that one take to produce, M. Matisse?" Matisse answered: "Fifty years."

4. (New York: Dover, 1952). Readers of this edition should be warned that there is a transposition of "start" and "end" that makes nonsense of the final paragraph of that chapter and does not reproduce the French. That transposition may itself be evidence of the strength of the older model and its influence on the translator.

5. Reason, Truth and History (Cambridge, Cambridge University Press, 1981), Ch. 5 "Two Conceptions of Rationality."

6. Partisans of The Real have not, to my knowledge, settled accounts with Relativity Theory. Perhaps they would say that Relativity Theory removes temporal relations of priority etc. from the ambit of The Real (on analogy with the theological treatments of the problem of the conflict between human freedom and God's foreknowledge.) Would that leave us unable to say that the Battle of Waterloo really came after the French Revolution?

 

© Guy Robinson 2010

 

Hits Since 17/02/10:

 

Office Depot Coupon Codes