Motte and Bailey Doctrines
One of the difficulties of getting people to behave better epistemically is that, whilst intellectual dishonesty is wrong, it is difficult to convict people of intellectual wrongs. As David Stove showed in his wonderful paper ‘What is Wrong with Our Thoughts?’ (The Plato Cult and Other Philosophical Follies Chapter 7 ), there are indefinitely many ways of cheating intellectually and for most there is no simple way to put one’s finger on how the cheat is effected. There is just the hard work of describing the species in detail.
Some time ago I wrote a paper entitled The Vacuity of Postmodernist Methodology (here or here or here ) in which I described and named a number of such cheats that I detected in postmodernism. One of these I named the Motte and Bailey Doctrine. There has recently been a flurry of use of this concept to analyse ethical, political and religious positions (e.g. here, here) so I am taking the opportunity to have a look at it again.
A Motte and Bailey castle is a medieval system of defence in which a stone tower on a mound (the Motte) is surrounded by an area of pleasantly habitable land (the Bailey), which in turn is encompassed by some sort of a barrier, such as a ditch. Being dark and dank, the Motte is not a habitation of choice. The only reason for its existence is the desirability of the Bailey, which the combination of the Motte and ditch makes relatively easy to retain despite attack by marauders. When only lightly pressed, the ditch makes small numbers of attackers easy to defeat as they struggle across it: when heavily pressed the ditch is not defensible, and so neither is the Bailey. Rather, one retreats to the insalubrious but defensible, perhaps impregnable, Motte. Eventually the marauders give up, when one is well placed to reoccupy desirable land.
For my original purposes the desirable but only lightly defensible territory of the Motte and Bailey castle, that is to say, the Bailey, represents philosophical propositions with similar properties: desirable to their proponents but only lightly defensible. The Motte represents the defensible but undesired propositions to which one retreats when hard pressed.
Diagnosis of a philosophical doctrine as being a Motte and Bailey Doctrine is invariably fatal. Once made it is relatively obvious to those familiar with the doctrine that the doctrine’s survival required a systematic vacillation between exploiting the desired territory and retreating to the Motte when pressed. Clearly, the diagnosis is not confined to philosophical doctrines: others may suffer the same malady.
I have been impressed by how well the recent users of this concept have deployed it as a tool of analysis and don’t myself want to add to those analyses here. What I do want to do is clarify something about its nature. Some people have spoken of a Motte and Bailey Doctrine as being a fallacy and others of it being a matter of strategic equivocation. Strictly speaking, neither is correct. A fallacy is an argument that is invalid and equivocation is giving different meanings to the same terms when you should be keeping their meaning invariant. A doctrine, however, is a body of propositions, not an argument. So a Motte and Bailey Doctrine cannot be a fallacy and shifting from asserting the Bailey propositions to the Motte propositions is not in general effected by giving different meanings to the same words or statements.
It is possible that I am partly to blame for that misunderstanding since in the original paper I mentioned that a Motte and Bailey Doctrine could be constructed from a set of what I called Troll’s Truisms. Troll’s Truisms are indeed based in equivocation A Troll’s Truism is an ambiguous statement by which an exciting falsehood may trade on a trivial truth. For example, ‘morality is socially constructed’ sounds like a radical assertion of cultural relativism until we are told that by ‘morality’ the speaker means not morality itself but just our beliefs about right and wrong. Of course, these beliefs are, in some sense, socially constructed, if only because our acquisition of many beliefs is mediated by language and beliefs about right and wrong are certainly among those acquired in that way. Hence in this sense of ‘morality’ the statement is true and trivially so. (At the time I named Troll’s Truisms I did not know of the phenomenon of internet Trolling: now that I do it seems to me a serendipitously fortuitous naming. Daniel Dennett has recently taken to calling Troll’s Truisms ‘Deepities’.)
So a Motte and Bailey Doctrine may be asserted by the use of Troll’s Truisms, using the trivial truths for the Motte propositions and the exciting falsehoods for the Bailey propositions. In such a case it might be correct to speak of the use of strategic equivocation. However, in general they are not asserted in that way and there are in fact many ways they can be advanced. In the paper itself I analyse in detail the ways in which Michel Foucault, David Bloor, Jean-François Lyotard and Richard Rorty set up Motte and Bailey Doctrines by Humpty Dumptying, deploying what I call ‘Equivocal Fulcra’ and dancing the Postmodernist Fox Trot before they finally fall into the black hole of absolute irrationalism.
All that being said, once a Motte and Bailey Doctrine is in place, it offers extensive opportunities for deceitful trickery in argument. The basic fallacy that is available is offering the arguments for doctrines in the Motte as if they were arguments for the doctrines in the Bailey. The crudest such fallacy would have the form ‘Motte, therefore Bailey’, and certainly, despite its crudity, there is no shortage of such argumentation. Something similar is going on in the strategy of advancing the Bailey and then retreating to the Motte when criticised. On such foundations, a myriad of persuasive fallacies may be built, and indeed, that myriad is the very point of the Motte in a Motte and Bailey Doctrine: without it the Bailey is lost.
So it is, perhaps, noting the common deployment of such rhetorical trickeries that has led many people using the concept to speak of it in terms of a Motte and Bailey fallacy. Nevertheless, I think it is clearly worth distinguishing the Motte and Bailey Doctrine from a particular fallacious exploitation of it. For example, in some discussions using this concept for analysis a defence has been offered that since different people advance the Motte and the Bailey it is unfair to accuse them of a Motte and Bailey fallacy, or of Motte and Baileying. That would be true if the concept was a concept of a fallacy, because a single argument needs to be before us for such a criticism to be made. Different things said by different people are not fairly described as constituting a fallacy. However, when we get clear that we are speaking of a doctrine, different people who declare their adherence to that doctrine can be criticised in this way. Hence we need to distinguish the doctrine from fallacies exploiting it to expose the strategy of true believers advancing the Bailey under the cover provided by others who defend the Motte.
For a final worrying lesson, here is someone who has grasped the concept but who entirely mistakes its point, apparently because he takes the Sophist’s perspective that goodness in epistemic behaviour is answerable only to the standards of effective persuasion.
Update: Scott Alexander has added some further very interesting analyses using this concept in his new post here and his commenters add an impressive range of examples in the comments below.
Update 2: I’d like to add a clarification to an example given in the original paper in response to challenges from a number of different people (who I thank for their interest) in a number of different locations. In that example I showed how Foucault sets up a Motte and Bailey doctrine about truth using Humpty-Dumptying (the arbitrary redefinition of a word). The bailey in this case contains radical philosophical propositions about truth (identifying truth and power) which are what his statements express if we take his use of ‘truth’ to mean truth. The motte contains the trivial propositions that his statements express under his redefinition of ‘truth’ to mean something other than truth.
Defending him on the grounds of his redefinition is missing the essential dishonesty of what he is doing. He purports to be offering a theory about truth, not something else, but the trickery of the redefinition gives him a way out when it is proved (as I do in the paper) that his theory of truth is false. Then it is claimed that he wasn’t ever talking about truth but always talking of something else. But that makes no sense of him using the word ‘truth’. He intends to be understood to be offering a theory of truth, a radical theory of truth that undermines our complacent beliefs about it. The function of the redefinition is precisely to play the role of the impregnable role motte when placed under philosophical scrutiny. Under that redefinition, his statements are true, but they’re not about truth. Once the philosophical scrutiny is relieved, he and his adherents return to what they really wanted, the radical (but false) propositions about truth.
Sometimes evading this last point is attempted by claiming that the refutation is based on speaking of logical or scientific truth, which is not the truth he is talking about. This is then defended by mentioning the apparent possibility of a proposition lacking scientific truth but possessing some other kind of truth. This, however, is confused. There are not separate properties, logical truth, scientific truth, every day truth, to be spoken of. There is just truth. Speaking of logical truth, scientific truth, every day truth and so on, are just rather loose ways of speaking of logical truths, scientific truths, every day truths and so on. Logical truths are truths of logic, scientific truths are truths of science, every day truths are truths about every day matters. Logical truths are true and in being so are possessed of exactly the same property of truth as are possessed by scientific truths and every day truths. So the adjectives ‘logical’, ‘scientific’ etc. characterise the subject matter of the propositions, not the semantic property of truth had by those propositions.
Finally, a great deal of philosophical mischief and nonsense in postmodernism is recognisably traceable to Humpty-Dumptying. For example, the foolishness of linguistic ontological relativism is exposed by the old riddle of how many legs does a dog have if you call a tail ‘a leg’. The answer is four: it doesn’t matter what you call a tail, that doesn’t make it a leg.