US philosopher whose background in engineering provided a model for his carefully crafted theories of epistemology
The American philosopher Fred Dretske, who has died aged 80, worked mainly in epistemology – the study of knowledge – and the philosophy of mind. His first degree was in electrical engineering: in his subsequent work, he liked to use examples from engineering, and constructed theories with many well-designed parts carefully fitted together to form functioning wholes.
He belonged to the naturalist tradition, discounting explanations that extend beyond the laws of nature to the supernatural or spiritual. Although he did not suppose that philosophy and science were exactly the same enterprise, he did think that philosophical theories should be scientifically respectable. And much of his work sought to show how elements of the mind are natural phenomena that can be understood in scientifically acceptable terms.
In the 1960s, epistemology was dominated by the idea that knowledge requires justification, and that one can be justified in believing a false proposition. It was also widely held that one could have a belief that is both justified and true, and yet not have knowledge. For example, I may believe that there is petrol in my tank because the petrol gauge tells me it is full. If in fact there is petrol in the tank, but the gauge is broken and just by coincidence reads "full", then my belief is true and justified. But because the justifying belief – the belief that the gauge reads "full" – does not provide conclusive evidence for the belief that the tank is full, and could equally well have justified a false belief to the same effect, my belief that the tank is full does not amount to knowledge. Most epistemologists reacted to this sort of example by arguing that knowledge is belief that is true and justified and meets some further condition which would rule out the broken-gauge kind of example.
Dretske, on the other hand, argued that justificatory beliefs have to provide conclusive reasons for the beliefs they justify. But whether something is a conclusive reason depends on the circumstances. If a petrol gauge is working properly, then its readings can provide conclusive reasons for beliefs about the tank. If it is not, then they cannot. In this way, Dretske could account for animals and infants having knowledge based on perception. A dog could know that there was a bone buried in the ground on the basis of its scent, even if the reason for the scent justifying the dog's belief are far beyond the dog's ken.
For Dretske, beliefs themselves were rather like petrol gauges: their function (determined by evolution or learning history) is to carry information about other things. If belief-forming mechanisms in the subject are working properly and are in an amenable environment, then they will reliably indicate external states of affairs. In such cases, beliefs are justified and provide knowledge.
In his philosophy of mind, Dretske's central idea was that the representational content of mental states such as beliefs could be understood in terms of their indicator function, and he constructed an elegant, deep and detailed theory based on this idea. Just as a sound petrol gauge co-varies with the amount of fuel in the tank, so beliefs are physical states that have the function of indicating, or co-varying with, states of the world. Dretske's basic idea was that what makes a state of an organism – say, a belief that the Democrats will win the next US election – is explained in terms of its having the function of indicating just such a state of affairs.
Dretske was born to Frederick and Hattie Dretske in Waukegan, Illinois, where he grew up. On the way to obtaining his electrical engineering degree from Purdue University, Indiana (1954), Dretske took a course in philosophy, and quickly decided that this was the subject for him. So he changed tack for his PhD from the University of Minnesota (1960).
For the bulk of his career he taught at the University of Wisconsin-Madison. His next chair was at Stanford (1988-99), and for the remainder of his life he was senior research scholar at Duke University, North Carolina. He was awarded the Jean Nicod prize (1994) and the Humboldt prize (2008). Some of his numerous articles are collected in the fifth of his books, Perception, Knowledge and Belief (2000).
A gentle, modest, deeply honest but very strong-minded individual, Dretske was very well-liked by his colleagues and students. He was a keen bridge player and an aficionado of very dry gin martini cocktails, which he took without ice, but with one or two olives. He is survived by his wife, Judith; daughter, Kathleen; son, Ray; stepson, Ryan; and three grandchildren.
• Frederick Irwin Dretske, philosopher, born 9 December 1932; died 24 July 2013