Methods of the Metaphysicians

 There are a few general principles of reasoning which are so useful or important that they have been given names of their own and are frequently alluded to in philosophical discussions of consciousness. A few of these are considered here.

Leibniz's Law home

Leibniz's Law states the identity of indiscernibles. What this means, roughly, is that if there is no conceivable way of telling the difference between A and B, then A is B. More precisely, A is B if, and only if, all the attributes of A are also attributes of B (and vice versa). To put it yet another way, A is B if everything which is true of A is also true of B (and vice versa again).

Informal arguments along these general lines have been used to support the case that computers could be conscious - "if it looks like a duck, walks like a duck, and quacks like a duck, then it is a duck". If a robot is indistinguishable from a person, then, it is argued, it is a person. The trouble with Leibniz's Law, though, is that in order to work it has to apply rigorously: all the attributes of A and B have to coincide. It's not enough for them to look or seem the same from a certain point of view, they have to have the same position in space, previous history, and so on. It's really a definition of identity rather than a principle which allows the discovery of new truths: only in the abstract world of formal logic does it become useful.

A looser variety of this kind of argument says that A is B if it has all the relevant properties of B. This involves us in new difficulties, though, since now we have to which properties are relevant. If we can find properties that define B, or are (somehow) essential to B, we'll be OK - but if we could do that for consciousness, the main problem would be solved. Otherwise, we end up with a mere rule of thumb - a set of properties which, as it happens, will pick out Bs quite accurately in our current circumstances, but don't really catch the essence of B-hood, and will be unreliable when something new comes along. We can use "featherless biped" as a definition of a conscious entity quite successfully until someone plucks a chicken.


Blandula I'd just like to point out that what definitely won't do as a substitute for Leibniz's Law is the view that if A can be mistaken for B in certain circumstances, then A= B, though this is what some readings of the  Turing Test seem to claim. A waxwork, or a pile of clothes in a darkened room, can be mistaken for a person in some circumstances, but that doesn't make them conscious.


Occam's Razor home

William of Occam (or Ockham) laid down the principle that 'entia non sunt multiplicanda' - entities are not to be multiplied. Loosely, this means that the simplest possible explanation - the one which doesn't require you to imagine or believe in a lot of additional mechanisms or entities - is the one you should always adopt. This is an extremely useful principle, often invoked. There are always in principle lots of alternative explanations for anything (in fact an infinite number, since we can always add another irrelevant element to any theory), so we need some way of choosing between them.

But there are some problems. It's not always easy to tell which hypothesis is the simpler one or the one with fewest entities. It could well be argued that the simplest explanation of everything is that God makes it the way it is by direct action of His will. If we adopt that hypothesis, we can do without all the constants, laws of nature, forces, fields, and all the rest of the apparatus required by physics (and hey, what are these 'laws of nature', anyway? What legislature passed them and which policeman enforces them?). Most philosophers, I think, would regard God as the redundant entity, amounting in himself to a much more complex and ontologically burdensome commitment than any amount of science. But it's not always obvious which way the sharp edge of Occam's razor is pointing.

Secondly, Occam may predispose us towards inadequate explanations. After all, it's not enough to be simple: an explanation must really explain things, too. If we work too hard at applying the razor we might find ourselves slicing out things we actually need ('I don't see the need for this Gravity business. Things just fall down, that's all. It's true by definition - 'down' just means the direction things fall in - obviously').

In the third place, what justification is there for assuming that the simplest explanation is the true one? The answer, I'm afraid, is none, really. The simplest explanation sometimes turns out to be wrong. Some would argue that within the infinite sea of possible explanations, the true ones are inevitably near the simpler end, so as a general rule, the simpler the better. But even if you accept that argument, it only means the simpler hypothesis is slightly more likely to be true. The truth is, I think, that as a matter of practicality we have to pick out one of the possible explanations in any given case, and simplicity is the only criterion which can be relied on, even in principle, as an objective way of singling out just one option. What else could we do? Choosing the most complex hypothesis won't work, because, there is no limit to the complexity we can add. The most beautiful or elegant explanation (unless these terms are used as synonyms for 'simplest') is a matter of opinion.


Morgan's Canon


Occam's Razor is a principle of parsimony; and there is another principle of parsimony which is of special relevance to consciousness. This is Morgan's Canon, put forward by Conwy Lloyd Morgan. It runs as follows.

'In no case may we interpret an action as the outcome of the exercise of a higher mental faculty, if it can be interpreted as the exercise of one which stands lower in the psychological scale'

Morgan, (1852-1936), was one of the founders of modern animal psychology: his work was characterised by scrupulous, painstaking observation and careful attention to methodology,and his canon was intended as a corrective to the anecdotal and anthropomorphic accounts of animal behaviour given by others. It is generally interpreted broadly, as meaning that consciousness is only to be invoked as an explanation where nothing else will do. This reading was taken up enthusiastically by behaviourists (who obviously thought consciousness need never be invoked), but Morgan himself was far from being a behaviorist, and actually had in mind a somewhat narrower point: don't attribute human-level mental processes to animals if simple instinct or training provide a good enough explanation. Some see the canon as a special case of Occam's razor, others see potential conflicts between the two - Steve Grand, for example, in his book 'Growing up with Lucy'


Reductio ad absurdum home

Reductio ad absurdum is a logical tactic which consists of assuming the negation of the thing you want to prove, and then showing that that assumption leads to a direct contradiction. Since a self-contradictory view cannot be true, the opposite one must be. This is another technique which is perhaps most at home in formal logic, where it is sometimes known as 'brute force', because if you can't find a short and pretty proof of your premise in propositional calculus, reductio will always give you one which works, though it may be relatively laborious and ugly.

You do occasionally see a more general argument carried on by working back from the absurdity of a conclusion to the falsity of the premises, but I'm afraid reductio most often appears in a rhetorical form, as a vehicle for ridiculing someone else's views - 'the fact that this fellow ends up believing in the consciousness of thermostats simply provides a reductio of his whole theory...'