To: Tom M.
From: Geoffrey Klempner
Subject: Zombies and philosophy
Date: 18 March 2002 15:50

Dear Tom,

Thank you for your e-mail of 17 February, with your first essay for the Philosophy of Mind program, 'What is the philosophical significance of the idea of a zombie?'

This is a clearly laid out essay, which follows the main lines explored in the program. You show a good, clear grasp of what it would mean to talk of a zombie in the philosophical sense.

One question which you explore is the thought experiment of 'becoming a zombie'. As this appears to be central to the case for the coherence of the idea of a zombie, it is worth looking in detail at the criticisms you make of it. You say:

(a) 'A zombie is empty inside, without thoughts or fears. If an experience as described happened to us [viz. the experience of alternating 'blind' left and right sides] we would be cut off from our bodies, receiving no input from our senses and exerting no control over our actions. However, there would still be emotions and thoughts in our inner world and thus by definition we would not be zombies.'

(b) 'It also appears implausible that you could have a detailed awareness of what you could not sense.'

As the worry in (b) is one which I voice in the unit, citing the example of drawing a flower from life, I will concentrate on objection (a).

The logical structure of the argument for the possibility of experiences which would give support to the hypothesis that one had become a zombie for a specified period is similar to an argument which has been given (by Richard Swinburne in his book 'Space and Time') for the coherence of the hypothesis of a world in which every process stopped for a specified period.

Obviously, there would be no-one around to witness that everything had stopped (and we are not relying on the idea of a God-like observer outside of all temporal processes). However, if everything on the one 'side' of the universe was observed to stop every two years with lawlike regularity and everything on the other 'side' of the universe was observed to stop every three years, that would constitute acceptable empirical evidence that everything in the universe - the two sides simultaneously - stops every six years.

It is also worth noting that John Searle offers an argument where he describes the experience of someone who finds that parts of their consciousness are disappearing, bit by bit. However, this does seem vulnerable to the objection that you give, namely, that the process can only go so far.

One question is whether zombiehood can be conceived of as something that could happen a bit at a time, as Searle thinks, or as I describe in the thought experiment of the bus, while the non-zombie part of you 'looked on'. Suppose that we did accept that such an idea is coherent, i.e. that this is a possible experience. In that case, you need to supply a reason why we could not find ourselves in the position of making the same kind of prediction as the inhabitants of the universe where the two alternate regions freeze at regular intervals.

My own view, however, is that Swinburne and the philosopher who puts forward the 'zombie on a bus' argument have made a similar mistake. They imagine that they have explained what it *means* for there to be a zombie, or what it means for all events in the universe to stop, by explaining what it would be to *discover* or find reliable evidence that this was so. My question would be, Why make this concession to verificationism? ('verificationism' = the idea that the concept of factual truth is tied to the possibility of empirical verification).

In other words, why can't the dualist say, 'I have arguments for my view which do *not* depend on the idea that there could be evidence that a zombie existed. It is a consequence of my dualist theory, however, that it is logically possible for there to be a zombie, although evidence for the existence of a zombie could never, in principle, be found.'

In saying this, the dualist is making a similar claim to one who, in the parallel case of the possibility of a time when nothing changes, argues that this idea makes sense even if nothing could count as sufficient evidence that a period of time without change had elapsed.

(As you will discover in due course, I think that the zombie hypothesis is incoherent, but not for the reasons discussed here.)

All the best,

Geoffrey