The Social Affairs Unit

Print Version • Website Home • Weblog Home

Use the buttons below to change the style and font size of our site.
Screen version     Print version:   
September 28, 2007

Paul Boghossian claims to have blown both relativism and constructivism out of the water - in fact, argues David Wootton, he has missed both targets: Fear of Knowledge: Against Relativism and Constructivism - Paul Boghossian

Posted by David Wootton

Fear of Knowledge: Against Relativism and Constructivism
by Paul Boghossian
Oxford: Clarendon Press, 2006
Hardback, £15.99; Paperback, £9.99

For thirty years Richard Rorty (who died on 8th June 2007) campaigned against conventional accounts of knowledge which claim that we know the truth when our theories correspond with how the world really is. He insisted that there is no way of knowing how the world really is. The idea of an unconditional truth is "unhealthy". Rorty wanted to "eradicate" the idea of objectivity. In place of a conventional epistemology he offered what we might call a political philosophy of knowledge. Knowledge is what we agree it to be, and agreement should be established by discussion, debate, and the building of consensus. Since agreement takes place within particular communities, knowledge is relative to the communities that reach agreement.

Paul Boghossian's little book is a sustained attack on Rorty's relativism. The basic argument is simple. Rorty's claim is that when we say "dogs are mammals" we really mean "according to biologists, dogs are mammals". But is this to be taken as a true statement? Surely it must be a statement true for a particular community. So we need to unpack it as "according to schoolteachers, according to biologists, dogs are mammals".

But the same move can be made over and over again in an infinite regress, and so Rorty's relativism is incoherent. There would seem to be a reply to this. The idea of a community is not a logical construct but a sociological fact, and so there is only a finite set of communities to whom one might appeal in deciding how to interpret the word "mammal".

Boghossian has another target, aside from Rorty, and that is the constructivism of the Science Studies school, whose best-known representative is Bruno Latour. Latour has rejected the claim that the pharaoh Ramses II died (1213 B.C.) of tuberculosis.

How could he pass away due to a bacillus discovered by Robert Koch in 1882?
he asks, arguing that this is an anachronism comparable to saying that he died of machine gun fire.
Before Koch, the bacillus had no real existence
- this despite the fact that scientists have found the bacillus in the mouth of Ramses II and extracted its DNA from another mummy.

For Boghossian what is astonishing about this line of argument is that it implies that there were no facts at all before there were human beings - a view that would scarcely faze Latour. What ought to astonish Boghossian is the comparison between the tuberculosis bacillus and a machine gun. When Gatling invented the machine gun in 1861 he took out a patent because it was something that had never existed before. When Koch discovered the tuberculosis bacillus he claimed that it had always been the cause of the disease previously known as phthisis: the discovery announced itself as a reinterpretation of the past. Gatling and Koch were clear about the difference between discovery and invention, and so were the people who wrote the law of patents. It seems very odd that Latour should blur this distinction, which we all recognize in everyday life. One might just as well claim that the sun went around the earth until Copernicus proposed otherwise.

Most people will have little difficulty with Boghossian's conclusion, that (p. 131)

there is a way things are that is independent of human opinion, and that we are capable of arriving at belief about how things are that is objectively reasonable, binding on anyone capable of appreciating the relevant evidence regardless of their social or cultural perspective.
A follower of Rorty, however, would baulk at the way that "independent", "objectively", and "binding" are used in this sentence, and would insist that all evidence is theory-dependent. How do you choose between Rorty's relativism and Boghossian's realism? The sensible answer is that you don't. You find a way between the two: the model to follow is Ian Hacking's The Social Construcion of What? (1999)

Boghossian's book, unlike Hacking's, proceeds at a very high level of abstraction. It is not clear that we are going to learn much about knowledge by worrying about whether dinosaurs existed before human beings discovered them; it might be better to look carefully at the sorts of knowledge-claims Gatling and Koch made.

Indeed for the last forty years there has been a symbiotic relationship between history and philosophy of science. When Thomas Kuhn came up for tenure at Berkeley (where he had a joint appointment in history and philosophy) as the author of The Copernican Revolution (1957), the philosophy department voted he should be tenured, but tenured in history. Since he was on the point of producing The Structure of Scientific Revolutions (1962), a work which is the most important contribution to philosophy of science of the last century, we can safely say that they were being obtuse.

Boghossian, unlike Kuhn, seeks to work out what knowledge is without doing case studies. This is not an impossible task - providing you smuggle into the argument things we all know (or think we know) about the world in which we live - that discoveries are different from inventions, for example. But problems begin as soon as one starts asking exactly what is being smuggled in, and how far one is entitled to take it for granted.

Boghossian's lack of interest in case studies means that when he does introduce them he makes a mess of them. Cardinal Bellarmine was central to the condemnation of Copernicanism in 1615: "the infamous Cardinal Bellarmine" as Boghossian calls him, claiming he is "reputed" to have refused to look through Galileo's telescope (p. 60). Actually Bellarmine looked through the telescope in April 1611; it is the philosopher Cremonini who is usually reputed to have refused to look through Galileo's telescope, but even he must have done so because he protested that it gave him a headache.

Bellarmine did not claim, as Boghossian implies, that Scripture was the only reliable source of knowledge about the heavens. He already held a view of the heavens which was unusually compatible with Galileo's discoveries, for he held that there were no crystalline spheres, as Ptolemaic astronomers had traditionally claimed (Galileo argued that the moons of Jupiter would have crashed into them), and that the planets swam through space like fish through the sea. His position on Copernicanism was a very sophisticated one: since movement is relative, all the planetary movements posited by Copernicanism could be equally successfully modelled in a system where the earth was stationary and the sun moved. Indeed there was in existence an excellent system of this sort - the system of Tycho Brahe.

It was a feature of Galileo's argument that he claimed that for almost all purposes the movement of the earth was imperceptible: objects, for example, dropped from a tower fall at the foot of the tower, not to the East. Galileo thought there was one physical process on earth that could be used to demonstrate that the earth moved, and that was the flow of the tides: Bellarmine was quite right to find this argument unconvincing, since it is the gravitational attraction of the moon, not the movement of the earth, which causes the tides. Galileo, who worked with pendula, unfortunately did not realise that the movement of a pendulum could be used to demonstrate the movement of the earth (as in the Foucault pendulum, 1851).

Boghossian sets up Galileo v. Bellarmine as a simple contest. Galileo believes in objective reality; Bellarmine believes in words in a book. But this isn't the case at all. Both Bellarmine and Galileo believe in objective reality; both believe (or at least claim to believe) that the Bible is divinely inspired; both believe that astronomers construct theoretical models that bear a complex and often uncertain relationship to reality. They differ on fine details of how to select between possible models of planetary movement while retaining these three commitments; and when they differ Galileo is not always right and Bellarmine always wrong.

What Boghossian has done is take the debate between Galileo and Bellarmine and hopelessly simplify it so that it corresponds to a debate between a crude realist and a crude constructivist, and then use that simplification to argue that crude realism is to be preferred to crude constructivism, and to imply that as a consequence crude realism is adequate - the earth moves, and that's a fact. Boghossian claims to be following Rorty in his account of Galileo v. Bellarmine, but Rorty never made such crass mistakes. He never said that Bellarmine refused to look through the telescope (at a guess Boghossian gets this from the legal philosopher Richard Posner), and, unlike Boghossian, he acknowledged that Bellarmine recognized Copernicus to be both ingenious and useful.

Boghossian only discusses one other case study in any detail. Like many in the debates over realism and relativism, it comes from the work of the anthropologist E. E. Evans Pritchard. EP (as he was known in the trade) claimed that the Azande believe that if a man is a witch, then all his patrilineal descendents are witches.

To our minds it appears evident that if a man is proven a witch the whole of his clan are ipso facto witches, since the Zande clan is a group of persons related biologically through the male line. Azande see the sense of this argument but they do not accept its conclusions…
Instead they take the view that only the close patrilineal kinsmen of a witch can be known to be witches. Philosophers debate whether the Azande have a different notion of logic from us - while the real question is whether philosophers have as good a grasp of the facts of life as the Azande do. The identity of a child's mother is usually well-established; the identity of a child's biological father is a very different matter. EP is wrong when he says
the Zande clan is a group of persons related biologically through the male line.
It is, instead, a group of persons who believe themselves to be related biologically through the male line; and the Azande are right to recognize that the uncertainty of whether two people are biologically related through the male line grows as one moves outward to more and more distant relationships. Boghossian's difficulty in understanding Azande logic seems to derive entirely from his failure to bring into consideration some obvious facts about ordinary life.

There's a further problem with Boghossian's book: it is preoccupied with words, and ignores deeds. Most forms of knowledge have uses for us; some forms of knowledge are useful because they enable us to navigate social worlds, but many forms of knowledge are useful because they enable us to act effectively on the natural world: they, like the Gatling gun, are technologies. In most cases technologies either work or they don't.

There are problem cases of fantasy technologies (Hippocratic medicine, astrology, alchemy) that have gained wide acceptance, but technologies that don't work are usually self-defeating. One can see the experimental method as a way of making sure that all scientific knowledge is applied knowledge: experiments are local technologies for performing specific, artificial tasks. Pre-Galilean astronomy was primarily a technology for predicting the movements of the planets in order to make possible astrological forecasts, and for that purpose it was entirely reliable. Had it not been it would have been quickly discarded.

Boghossian acknowledges that we only acquire knowledge on subjects in which we have an interest, but he doesn't ask himself what knowledge is good for. He only asks if it is true, not if it is functional. One of the points at which this gap in Boghossian's argument becomes apparent is in his discussion of what he calls "blind reasoning". We all start thinking, not from scratch, but from the resources made available to us within our culture; we cannot start anywhere else. Consequently we cannot reasonably claim that our starting point is the right one: it is simply the one we have. But what happens when two different cultures meet, and compare their differing ideas of what constitutes knowledge? Boghossian writes (pp. 100-101):

It is not entirely out of the question that we should come across an alternative epistemic system that made us doubt the correctness of our own epistemic principles. How would we imagine this? Well, we would imagine encountering a different community with what are clearly much more advanced scientific and technological abilities. And yet they deny fundamental aspects of our epistemic system and employ alternative principles.

For this encounter to have the desired effect, this alternative epistemic system would clearly have to be a real-life epistemic system, with a proven track record, not just some theoretical possibility. Its actual achievements would have to be impressive enough to make us legitimately doubt the correctness of our own system.

Such encounters are the stuff of science fiction: in the Star Trek series societies without warp drive are sharply demarcated from societies with warp drive. But they are also the stuff of real life - and there are plenty of case studies that consider them. The Spanish conquest of the Incas and of the Aztecs involved a society with gunpowder weaponry encountering ones without; throughout the Third World in the period after the Second World War western-trained doctors armed with penicillin were in competition with traditional healers with no knowledge of antibiotics.

It should be obvious that the test of an alternative epistemic system is not whether it is true; it is whether it enables you to do things you already want to do (kill at a distance; cure diseases). And it follows from these examples that radically different intellectual systems may address sufficiently similar practical problems for their adequacy to be directly compared (a possibility that Kuhn denied), and the test of knowledge is often not just true belief (as Boghossian thinks), or agreement (as Rorty claimed), but the ability to do things in the world.

Let me give an example. In 1668, Francesco Redi conducted experiments to show that small creatures such as flies do not generate spontaneously, as biologists had always believed. When microscopic creatures were discovered a decade later scientists disagreed about whether or not they were spontaneously generated, and for more than two centuries, until John Tyndall's Floating Matter in the Air (1882), experiments sometimes proved that there was spontaneous generation of microorganisms, and sometimes proved, equally convincingly, that there wasn't.

There is a conventional way of telling this story which says that the whole question was resolved in 1864, when the French Academy of Sciences declared that Pasteur had disproved spontaneous generation. Yet in 1868 the leading British medical microscopist, J. Hughes Bennett, was convinced he had new experimental evidence proving that Pasteur was wrong.

What ensured Pasteur's victory was not any incontrovertible experimental evidence - the experimental evidence was ambiguous and confusing until 1882. His triumph was ensured by Joseph Lister, who in 1867 had found a practical application for Pasteur's germ theory in medicine, with the invention of antiseptic surgery. Bennett was backing the wrong horse. Pasteur's triumph depended on the fact that it had proved possible to turn his argument about spontaneous generation into a form of applied knowledge. Why would one argue with something that worked in everyday life? By the time Tyndall tidied up the experimental evidence the battle was already over.

Does introducing the principle of applied knowledge make a difference to the debate between realists and relativists? It depends on what sort of applied knowledge you have in mind. Almost all societies (perhaps of necessity all societies) fight wars and attempt to cure diseases. So gunpowder and penicillin must make some sort of sense in any society, even if the way in which they work is incomprehensible.

But some forms of knowledge appear extremely useful to one society, and useless to another: the Azande claimed to be able to identify a witch by post-mortem dissection. For them this was a valuable form of knowledge, but not for us. Relativists and constructivists assume that a lot of knowledge is of this form. There was recently a television programme about a Buddhist monk said to be the happiest man on earth. How do I choose between his system of knowledge, which has made him happy, and mine, which has produced the hydrogen bomb? Can his system meet Boghossian's test of having actual and impressive achievements?

A good deal thus depends on whether the utility of a form of knowledge is socially specific or cross-culturally universal. It is precisely here that the dispute between realists and relativists should be most interesting. Cross-culturally universal technologies are narrowly constrained. There are only a limited number of ways of killing at a distance in societies with pre-nuclear technology: throwing devices (slings, trebuchets); stored tensile or torsional energy (bows, crossbows, onagers); controlled explosions (firearms, rockets). Socially specific technologies are infinitely variable: the morse code, for example, is just one of innumerable possible ways of conveying coded information.

Realists need to show that knowledge is valid cross-culturally; relativists need to show that knowledge cannot survive translation from one culture to another. Here again, the best way of thinking about the problem is to look at case studies (at the rejection of firearms in Tokugawa Japan, for example, or the rejection of inoculation against smallpox in eighteenth-century France), and the interesting cases will all be cases of applied knowledge, for those are the cases where one would expect translation from one culture to another to occur.

This issue of cross-cultural validity brings us to a further problem. Kuhn argued that Bellarmine and Galileo lived in two different worlds. Bellarmine lived in a world in which, if he was sitting in a chair, he was stationary in space; Galileo lived in a world in which, if he was setting in a chair, he was simultaneously spinning around the centre of the earth and circling around the sun. In Kuhn's view because they lived in different worlds they could not communicate with each other, and indeed there could be no objective test of which of them lived in the "real" world - this is the problem of incommensurability.

Boghossian, of necessity, dislikes this line of argument. If Bellarmine and Galileo meet in Bellarmine's room, then they are sitting in the same room, talking about the same telescope. This is true and not true: they can both agree about where they are, and how to get from where they are to somewhere else; but one believes the room is whirling through space, and the other thinks it is stationary.

Both Boghossian and Kuhn seem wrong here: in lots of ways, Galileo's world maps perfectly onto Bellarmine's world, so there are plenty of topics on which they can communicate straightforwardly with each other, and, where their two worlds don't map perfectly onto each other, there are ways in which they can explain to each other the relative advantages (as each sees them) of their differing viewpoints - they can translate from one world into the other. Kuhn overstates the problem of incommensurability.

On the other hand the problem cannot simply be wished away, as Boghossian implies. If Galileo tosses an apple to Bellarmine, Bellarmine will probably succeed in catching it. But he would be incapable of drawing its trajectory through the air (for, as a good Aristotelian, he believes that it first travels, steadily losing speed, in a straight line, and then falls), a task that Galileo (who knows the apple is travelling at a more or less constant speed in a straight line, while simultaneously falling at an accelerating speed) would find straightforward.

And consequently if we asked Galileo and Bellarmine what they had seen as they watched the apple fly through the air, they would be unable to agree. And if we took them to a gunnery range they would disagree how to aim a gun in order to hit a distant target - though they would be perfectly able to reach agreement about which of them had succeeded in hitting a target and which had failed.

Boghossian claims to have blown relativism and constructivism out of the water. In fact he has missed his chosen target. To hit this particular target you need to look at real cases, and you need to look at applied knowledge.

Reading this book is a bit like watching Cardinal Bellarmine on a gunnery range: he will always aim in the right direction, but the projectile will always land short of the target, and this will keep on happening until he starts learning from experience. Interestingly, as early experiments showed, he will still keep missing if he simply replaces his Aristotelian physics with Galilean physics - a fact that was rather embarrassing for Galileo's disciples. This is because the gunner not only has to take account of air resistance, but also of the fact that cannon balls and round shot spin in the air.

In the debate between objectivists and relativists we are about as far along as Bellarmine and Galileo were in their analysis of the trajectories of cannon balls. We are still missing the target.

David Wootton is Anniversary Professor of History, University of York. He is the author of Bad Medicine: Doctors Doing Harm Since Hippocrates.

Comments Notice
This comments facility is the property of the Social Affairs Unit.
We reserve the right to edit, amend or remove comments for legal reasons, policy reasons or any other reasons we judge fit.

By posting comments here you accept and acknowledge the Social Affairs Unit's absolute and unfettered right to edit your comments as set out above.

Maybe I missed something here (not having read the book), but why doesn't this author deal with falsifiability and predictive value? To my mind, Karl Popper cut these pomo clowns off at the knee long before this author took them on. In other words, I believe, based on experience and the current state of available knowledge, that gravitational effects will cause unpleasant results if I were to step off a 15-story roof; those who believe that gravity is a social construct are invited to act upon their beliefs.

Posted by: Mitch at September 30, 2007 02:16 AM

As any relativist would say, upon hearing a criticism of relativism:

"But that's just what you think"

Posted by: Benjamin Bilski at October 2, 2007 03:29 PM

First I would like to say concerning the use of the physical Theory of Relativity to justify moral relativism – dung can at least be used as a fertilizer. My reading of the Alexandria Quartet is confined to summaries, which seemed pornographic enough. All they would have achieved would have been to convince both the Egyptians and the British that the other’s country was a den of iniquity. However, in the spirit of Oklahoma:

I’d like to say a word for the cowboy
The road he treads is difficult and stony
He rides for days on end with jist a pony for a friend

I’d like to put in a word for Rorty. I’d much rather take “2+2=5” from Radiohead than from O’Brien. And when it comes to moral or spiritual truth, the use of force belies the truth or reinforces the untruth of what is forced. But as to the lengths to which he took this, and his Trotskyist double thinking (as bizarre as that of Emily Kingham’s young friends, and which in the case of Frida Kahlo delivered a message which spoiled an otherwise excellent medium), they both prompt me to finish:

I shore am feelin' sorry for the pony.
Posted by: Robert H. Olley at October 2, 2007 10:30 PM
Post a comment

Anti-spambot Turing code

Creative Commons License
Except where otherwise noted, this site is licensed under a Creative Commons License.

The Social Affairs Unit's weblog Privacy Statement