Stove's Discovery of the Worst Argument in the World


James Franklin

Philosophy 77 (2002): 615-24.

(pdf version)


In 1985, the year Sydney University threatened him with disciplinary action over his complaints about `Jobs for the Girls', David Stove ran a Competition to Find the Worst Argument in the World. In his marking scheme, half the marks went to the degree of badness of the argument, half to the degree of its endorsement by philosophers. Thus an argument was sought that was both very bad, and very prevalent.

He awarded the prize to himself, for the following argument (Stove, 1995)


We can know things only



we cannot know things as they are in themselves.


Perhaps that argument does not look familiar at first glance. It will be argued that it is extraordinarily common, and that it has underpinned many irrationalist programs in the history of thought, from classical idealism to recent relativisms in the philosophy of language, the philosophy of science, ethics and elsewhere.

But to start closer to home. Two short passages from Stove's later book, The Plato Cult, deal with people everyone has actually met. Speaking of the typical products of a modern high school, he writes:

Their intellectual temper is (as everyone remarks) the reverse of dogmatic, in fact pleasingly modest. They are quick to acknowledge that their own opinion, on any matter whatsoever, is only their opinion; and they will candidly tell you, too, the reason why it is only their opinion. This reason is, that it is their opinion. (Stove, 1991, 168).

And who can fail to recognise Stove's picture of another group of players in the intellectual world:

The cultural-relativist, for example, inveighs bitterly against our science-based, white-male cultural perspective. She says that it is not only injurious but cognitively limiting. Injurious it may be; or again it may not. But why does she believe that it is cognitively limiting? Why, for no other reason in the world, except this one: that it is ours. Everyone really understands, too, that this is the only reason. But since this reason is also generally accepted as a sufficient one, no other is felt to be needed. (Stove, 1991, 167)

These arguments – or, less euphemistically, dogmas – are versions of Stove's `Worst Argument' because all there is to them as arguments is: our conceptual schemes are our conceptual schemes, so, we cannot get out of them (to know things as they are in themselves). In Alan Olding's telling caricature, `We have eyes, therefore we cannot see.’ (Olding, 1998; further in Olding, unpublished)


Stove himself was most concerned with this argument as it occurred in classical idealism. Berkeley argued `the mind … is deluded to think it can and does conceive of bodies existing unthought of, or without the mind, though at the same time they are apprehended by, or exist in, itself.’ (Berkeley, par 23). That is, `you cannot have trees-without-the mind in mind, without having them in mind. Therefore, you cannot have trees-without-the-mind in mind.’ (Stove, 1991, 139) This argument, which Stove called `the Gem’, is a version of the `Worst Argument’ because it argues from the fact that we can know physical things only under our own mental forms to the impossibility of knowing physical things at all. Stove finds this argument in many later idealists. Fascinating as High Victorian idealism is, its hold over modern thought is not what it was, so let us leave that topic aside — except to mention Stove’s complaints about the extra pomposity added to the argument as each successive stage: `Thus you never say, for example, "things as they are," and still less, "things". You say "things as they are in themselves," or better still, "things and their properties as they exist both in and for themselves."’ Then you can construct a seriously heavyweight argument, like:

We can eat oysters only insofar as they are brought under the physiological and chemical conditions which are the presuppositions of the possibility of being eaten.


We cannot eat oysters as they are in themselves. (Stove, 1991, 151, 161)


Let us not attempt to say what deconstruction or postmodernism are, to express their essence or true nature. How crude and unsympathetic that would be. Still, as in negative theology, it is possible to say what they are not, or at least to make a few playful gestures in that direction. There are obvious difficulties with presenting the arguments in the original works of Derrida, or Lacan, or Baudrillard. They do not write in any natural language, they do not put the premises before the conclusion, the conclusion is distributed over the text rather than appearing in any one sentence, positions are assumed to have been established outside the texts one is actually reading, in previous texts, or perhaps future ones, and so on. But a broadly accurate summary of one of their basic arguments is given by an author who writes in a tradition in which conciseness of argument is not unknown. He writes

Based on the Saussurean principle of the sign, which is that the relationship between the signifier … and the sign … is arbitrary, the structure of language for Lacan is such that `language’ is already cut off from `reality’. What is taken as the meaning … of any word, for example, is always going to be the result of that word’s difference from all other words within a particular language. Meaning, then is a result of difference, and difference is a result of language as a system … Consequently the Saussurean-based theory of language … is radical because it erases `reality’ from the system: reality is never present `in’or `to’ the system of language … The gap between word and thing … is a necessary one inasmuch as language can never be identical with what it names, for example, and vice versa … From this it follows that presence (truth, reality, self-identity) is an effect of a system that is constituted by absence and separation. The very lack within language and the very gap between word and thing is what makes reality possible, making it seem present. (Lucy, 1997, 23; cf. 42; similar in Novitz, 1986, 49)

This is a `Worst Argument’, undoubtedly, close to a Gem. It is just a linguistic version: we cannot speak about things except through the forms of language, therefore we cannot speak about things as they are in themselves. The apparent preceding reasoning from Saussure’s view of linguistic structure is no more than a softening-up operation: while you’re cowering in your foxhole disoriented from hearing that `cat’ gets its meaning merely by contrast with `dog’ and not from any connection with your experience of cats, the real Gem is coming across the wire at you. The Gem stands by itself as an argument in the passage quoted, and if there is any relevance in the argument from Saussure, that argument must be circular, since it already assumes the disconnection of `cat’ from experience of cats.

It is not necessary to move in the French orbit to be imposed on by the same argument for linguistic idealism. Immersion in the semi-idealist tradition long noted in America is sufficient. (Partridge, 1936) Putnam adduces the Löwenheim-Skolem theorem in symbolic logic in support of that position. The theorem says that a theory, conceived as a set of uninterpreted symbol strings, necessarily has unintended interpretations (at least, if the theory is not excessively simple). Putnam suggests there are three basic positions in the philosophy of language on how words (so to speak) go out and attach to things: either there is a mysterious Platonic faculty of `understanding’ that does it; or there is a moderate realist `causal reference’ position that tells a natural causal story about how a certain word becomes attached to a certain thing; or words don’t attach to things. The moderate realist position that we would naively like to believe, says Putnam, is ruled out by the Löwenheim-Skolem theorem. So why exactly cannot the causes – any causes – distinguish between the interpretation of words we want and unintended ones? Because `adding … a body of theory titled "causal theory of reference" is just adding more theory’ – and hence itself, by the theorem, could have unintended referents. (Putnam, 1980, p. 477; discussion in Lewis, 1984, Taylor, 1991) In other words, `We can talk about things only via words, so, we cannot talk about things as they are in themselves.’ The Löwenheim-Skolem theorem acts merely as softening-up, like Saussure’s theory earlier: the theorem does not apply to sufficiently simple theories, but it does not appear that Putnam proposes to draw a different conclusion for those theories.

Here is another example from American anti-realism, perhaps an even purer one:

The difficulty is that whatever we observe, or, more generously, whatever we interact with, is certainly not independent of us. This is the problem of reciprocity. Moreover, whatever information we retrieve from such interaction is … information about interacted-with-things. This is the problem of contamination. How then, faced with reciprocity and contamination, can one get entities both independent and objective? Clearly, the realist has no direct access to his World. (Fine, 1986, 151, quoted in Musgrave, 1999)

In quoting this passage in his article, `Conceptual idealism and Stove’s Gem’, Alan Musgrave rightly calls attention to the role of the hyphens. Hyphenated entities like `interacted-with-things’ are a reliable sign of a Gem in play.


Let us consider another Gem-laden lode, post-Kuhnian philosophy of science. The replacement of the logic and philosophy of science by its history certainly raises the suspicion that there is a `Worst argument’ at the bottom of it. Stove writes:

The Kuhnian is scandalized if you call a current scientific paradigm `true’ or an earlier one `false,’ or if you say that the later one is `probably nearer the truth’ than the earlier. Paradigms are incommensurable, he tells you, and no special authority attaches to one which governs a field of science now. And why must we accept this astounding and sordid democracy of paradigms? Why, just because, in any field, even the best scientific knowledge which is current now, or at any time, is always rigidly constrained within the limits imposed, by the paradigm prevailing at the time, on scientific knowledge. (Stove, 1991, 168)

It is not clear how accurately this represents Kuhn himself. Partly, this is because he just said, `Let’s do history, as it is so much more exciting than boring old logic.’ He does, it is true, state conclusions that seem to require such an argument, such as `There is, I think, no theory-independent way to reconstruct phrases like "really there"; the notion of a match between the ontology of a theory and its "real" counterpart in nature now strikes me as illusive in principle. Besides, as a historian, I am impressed with the implausibility of the view.’ (Kuhn, 1970, 206-7; discussion in McGrew, 1994). But no argument is included. His followers have made up the slack, especially those in the `Strong Program in the Sociology of Knowledge’ or social constructivism, like Bloor. They propose to replace all considerations of logic, of what scientific theories are reasonable, with considerations of sociology, that is, of what interests theories serve. The real reason for their views is their conviction that since science is done by people, its explanation should be in the realm of causes acting on people, not the realm of abstract reasons. People, they think, can be acted on by their interests, or patronage, or the social milieu, but abstract facts like 2 + 2 = 4 do not act. So explanations of how people, including scientists, think ought to be sociological. This argument appears in various forms, mostly not very explicit ones. Thus, Bloor argues that observation `underdetermines’ theory – that is, that several theories are logically compatible with any given body of observations – and concludes immediately that it must be social factors that determine which theory is chosen. (Bloor, 1976; Bloor, 1991, 171-2) He says that the `existence of nature’ does not account for (scientific) theories and that simple `attention to nature’ will not adjudicate the merits of our theories. (Barnes, Bloor and Henry, 1996, 48) He reserves particular anger for the opinion that belief in rea onable theories is at least in part explained by their being reasonable, while mistakes require causal explanations; Bloor says sarcastically that this is an attempt to render science `safe from the indignity of empirical explanation.’ (Bloor, 1976, 7, 5) It must be emphasised that Bloor does not admit any possibility of co-operation between causes and reasons: explanation in terms of causes is quite different to that in terms of reasons, he says; if one is right, the other is wrong. (Bloor, 1976, 9)

This argument, the central plank of the social constructivist position, is a version of Stove’s `Worst Argument’ because it says: `We can know things only via causal (social) processes acting on the brains of real scientists, therefore, the content of our theories is explained without remainder by the social factors causing them; that is, we cannot know things as they are in themselves.’ This is why no amount of raging about relativism, scepticism and truth is found to make any impact on constructivists. They have a last line of defence in the argument: `Those entities in Platonic worlds, like truths and theories, cannot cause belief in themselves. Scientists are people, after all, and as such are responsive only to social or similar causes.’

Like all such arguments, Bloor’s says, in effect, that the mere fact that a theory is accepted is a reason for not accepting it.


So far it has been taken for granted that the invalidity of the `Worst Argument’ is obvious. That is because its invalidity is obvious: its conclusion, `we cannot know things as they are in themselves’ just does not follow from the premises about how we can know things only as they are related to us. Nevertheless, what is wrong with the argument can be seen more clearly through a parallel case. Take an electronic calculator. Why does the calculator show 4 when someone punches in 2+2? On the one hand, there is a causal story about the wiring inside, which explains why 4 is displayed. But the explanation cannot avoid mention of the fact that 2+2 is 4. On the contrary, the wiring is set up exactly to implement the laws of arithmetic, which are true in the abstract. The causal apparatus is designed specifically to be in tune with or track the world of abstract truths. If it succeeds, the causal and abstract stories co-operate, and the explanation of the outcome requires both. If it fails, the mistake is explained by some purely causal story. It is the same with brains that do science. For whatever reason, brains have the ability to gather reliable basic sensory information about the world. Information gathered by observation and experiment supports some scientific theories better than others, and a brain that draws the correct conclusions is performing well, while a brain that does not do well needs its mistakes explained. (basic argument in Sokal & Bricmont, 1998, ch. 4)


It is no surprise that relativist ethics is also a field strewn with Gems. A version particularly untroubled by philosophical sophistication appears at the beginning of E.O. Wilson’s Sociobiology:

… self-knowledge is constrained and shaped by the emotional and control centers in the hypothalamus and limbic system of the brain. These centers flood our consciousness with all the emotions — hate, love, guilt, fear, and others — that are consulted by ethical philosophers who wish to intuit the standards of good and evil. What, we are then compelled to ask, made the hypothalamus and limbic system? They evolved by natural selection. That simple biological statement must be pursued to explain ethics and ethical philosophers, if not epistemology and epistemologists, at all depths. (Wilson, 1975, 3)

The argument is: `We cannot know ethical truths (if there are any) except through the urgings of our back-of-brain plumbing, therefore, we cannot know ethical truths at all.’ Attempts to make this argument more philosophically svelte are unlikely to change its basic logical form.


So, are all the philosophical arguments that offend common decencies and corrupt the youth reducible to Stove’s `Worst Argument’? No. That is not true even all those of an idealist tendency. Talk of `forms of perception’, and `things in themselves’ may suggest Kant, but it is not clear that Kant was imposed on by a `Worst Argument’. Stove does pin a few small Gems on him (Stove, 1991, 160), but they are not central to his argument. Well before that stage in his reasoning, Kant relied on arguments from antinomies, transcendental arguments and considerations about constructions in geometry and the activity needed in counting, none of which are Gems. The arguments of classical scepticism, too, although generally along the lines of asking `how can we get out of the prison of our personal experience?’, have a further serious argument at their core: the symmetry argument that says `there is no mark to distinguish the true and the false.’ (Franklin, 1991) Arguing that there is no reason in experience to prefer the realist hypothesis to demon scepticism has no resemblance whatever to the `Worst Argument’.


It is natural to think: `All right, the argument as it stands is wrong, but surely there is something in it. If we see things only from our perspective, through the filter of our cultural experience, and so on, is there not some problem about how to get out (and see the world as it really is)?’ That may be so. But what made the `Worst argument’ worst was not that it raised a question about how to get out, but that it claimed immediately that there was no way to get out.

How-to-get-out plans come in three kinds. A very brief survey of them will reveal the scope and limits of `Worst Argument’ diagnoses.

The first plan, common to such diverse philosophers as Wittgenstein and Stove, denies that we are ever `in’, and hence believes there is no problem about how to get out. Such philosophers think that as soon as one admits the force of metaphors like `in the mind’, and starts taking them literally, one is stuck inside forever. But, they say, there is no need to do so. We are in direct contact with the world, like thermostats are, and no more need be said, except scientifically. Naturalised epistemology is in the same line of business, by suggesting that once one has done the science, there are no epistemological questions left over.

The difficulty with this plan, as is well known, is that it is hard to reconcile with the scientific picture. It could have been that human knowledge worked the way Aquinas says it does for the separated soul after death – by going out and so to speak digesting the object of knowledge. But human perception does not work like that. Perceptual experience supervenes on what is inside the head, and hence there is a how-to-get-out problem, whether the head is human or robotic, and whether what is inside it is material, mental, electronic or astral. Calling its relation to the outside world `direct’ does not help.

The second kind of escape plan, at the opposite end of the spectrum, takes the problem very seriously – so seriously, that it believes divine help is necessary, or something close to it. Descartes invoked God’s help directly, while the scholastics relied on a divinely-implanted `active intellect’ with amazing powers of insight into reality. There are not many modern adherents of this plan, but there is an interesting argument in Richard Taylor’s Metaphysics that exhibits its virtues very quickly (Taylor, 1963, 96-101):

You are travelling on a train. Looking out the window at a hillside, you see some stones, which appear to you to form a pattern reading, THE BRITISH RAILWAYS WELCOMES YOU TO WALES. You have two choices: either you think someone put them there to make a message, or you think they came there by some natural process. Either could be a good theory, but only if you believe the first theory can you take the stones to be a reason for thinking that you are entering Wales. If you believe they came there by a natural process, you cannot at the same time take them as constituting a message about Wales. Similarly with your sense perception. Either someone created it to form a message about something outside itself, or it came to exist by some natural evolutionary process. If the latter, you cannot take it to be a message about something outside itself. (Obviously, to invoke evolutionary theory as itself a reason to believe in sense perception is circular, since the only reasons to believe in evolution rely on sensory data.)

The problem with this plan is that few at this late historical stage want to believe the conclusion. But also, since you just open your eyes and see, it looks like it must be easy — needing help from ganglia, perhaps, but not gods.

The third and more moderate how-to-get-out plan relies on inference to the best explanation. It admits the internalist point of view, but suggests we can get beyond it by the kind of argument that leads from experimental evidence such as Brownian motion to the existence of atoms. The existence of the external world is said to be much the best explanation of our experience. (Devitt, esp. 108; Sokal & Bricmont, 1998, ch. 4)

The downside of that plan is that inference to the best explanation is so poorly understood. Even on matters of the most general principle, it is not well enough understood to permit a convincing account of why exactly Berkeley’s explanation of sense experience is a less good explanation than the realist one with physical objects. The situation is even worse if details are asked for about how to measure the goodness of an explanation, or the probability of an inference to the best explanation, or an account of how the coherence or simplicity of a theory affects its goodness as an explanation. Considering the importance of such arguments, and the meagre results of a decade of work trying to understand them (e.g. Josephson, 1994; Magnani, 2001), our ignorance of them is simply embarrassing.


There is therefore a genuine problem about how we can see, given we have eyes. That gives no support to fallacious arguments of the form `We have eyes, therefore we can’t see.’




B. Barnes, D. Bloor, & J. Henry, Scientific Knowledge: A Sociological Analysis (Chicago: Chicago University Press, 1996).

G. Berkeley, A Treatise Concerning the Principles of Human Knowledge, ed. J. Dancy (Oxford: Oxford University Press, 1998).

D. Bloor, Knowledge and Social Imagery (London: Routledge; 2nd ed, London: University of Chicago Press, 1976).

M. Devitt, Realism and Truth (2nd ed., Oxford: Blackwell, 1991)

A. Fine, `Unnatural attitude: realist and instrumentalist attachments to science’, Mind 95 (1986), 149-79. <

J. Franklin, `Healthy scepticism’, Philosophy 66 (1991), 305-24.

T. Kuhn, The Structure of Scientific Revolutions 2nd ed (Chicago: Chicago University Press, 1970).

J.R. Josephson & S.G. Josephson, Abductive Inference: Computation, Philosophy, Technology (Cambridge: Cambridge University Press, 1994).

L. Magnani, Abduction, Reason and Science: Processes of Discovery and Explanation (Dordrecht: Kluwer, 2001).

D. Lewis, `Putnam’s paradox’, Australasian Journal of Philosophy 62 (1984), 221-36

T. McGrew, `Scientific progress, relativism and self-refutation’, Electronic Journal of Analytic Philosophy 2 (1994).

A. Musgrave, `Conceptual idealism and Stove’s Gem’, in M.L. Dalla Chiara, ed, Philosophy of Science in Florence (Dordrecht: Kluwer, 1999), repr. in A. Musgrave, Essays on Realism and Rationalism (Amsterdam: Rodopi, 1999), pp. 177-84, and as `Idealism and antirealism', in R. Klee, ed, Scientific Inquiry: Readings in the Philosophy of Science (New York: Oxford University Press, 1999), 344-52.

D. Novitz, `The rage for deconstruction’, Monist 69 (1986), 39-55.

A. Olding, `Religion as smorgasbord’, Quadrant, 42 (5) (May, 1998), 73-5.

A. Olding, unpublished, `Common sense and uncommon nonsense’.

P.H. Partridge, ‘The social theory of truth’, Australasian Journal of Psychology and Philosophy 14 (1936), 161-75.

H. Putnam, `Models and reality’, Journal of Symbolic Logic 45 (1980), 464-82.

A. Sokal & J. Bricmont, Intellectual Impostures: Postmodern Philosophers’ Abuse of Science (London: Profile Books, 1998).

D.C. Stove, The Plato Cult and Other Philosophical Follies (Oxford: Blackwell, 1991).

D.C. Stove, `Judge’s report on the competition to find the worst argument in the world’, in D.C. Stove, Cricket Versus Republicanism (Sydney: Quakers Hill Press, 1995), 66-7.

D.C. Stove, Anything Goes: Origins of the Cult of Scientific Irrationalism (Sydney, Macleay Press, 1998).

R. Taylor, Metaphysics (Englewood Cliffs: Prentice-Hall, 1963).

B. Taylor, `Just more theory’, Australasian Journal of Philosophy 69 (1991), 152-66.

E.O. Wilson, Sociobiology: The New Synthesis (Cambridge, Mass: Belknap Press, 1975).