C O G I T O VOL . I NO .
2
Can There Be a Test for Consciousness?
Mark Peters
HE desirability of a test for consciousness derives from its potential to resolve conflict in several debates on interesting subjects of serious ethical importance where there is a potential to minimise suffering, such as foetal consciousness, euthanasia, animal consciousness and (perhaps soon) machine consciousness. Such a test, if formulated, might also establish the usefulness, one way or another, of debate on extra-terrestrial, plant, sub-atomic and cosmic consciousness, all of which have their ardent adherents. Finally, a test might enable us to detect consciousness where it has hitherto been unsuspected. To be useful, then, the test must be universal. If it is applicable only to communicating humans, it begs the question what consciousness is, in the worst possible way. To be universally applicable it must recognise the generalised characteristics of consciousness, not just the form of it with which humans are familiar. As Lucy Suchman puts it: Mind is best viewed as . . . an abstractable structure implementable in any number of possible physical substrates. . . . This view decouples reasoning and intelligence from things uniquely human.1 To be valid, any test must correlate objective physical events with individual phenomenal events. Secondly, any test which purportedly resolves a question must contain a certain definition of the question, expressed in terms of the results of the test and the probability of its accuracy. In other words, the statement “consciousness in a subject under investigation will produce result R1 with probability P1 , and absence of consciousness will produce result R2 with probability P2 ” amounts to one kind of definition of consciousness. Though this may seem circular, it isn’t. It is simply a way of showing that some definition of consciousness is necessary and integral to the test. The test must be built on the ground plan of an agreed physical...