Response from Peter Lipton on November 6, 2005

Lexicography and conceptual analysis must anyway be different, since a good dictionary has lots of good definitions, while philosophers are hard pressed to give a correct conceptual analysis of any of the concepts that centrally concern them. One reason for this may be that conceptual analysis tries to go deeper than lexicography, to give the underlying contours of a concept. Take the concept of knowledge. My dictionary defines knowledge as 'the facts, feelings or experiences known by a person or group of people'. Not very deep, because it uses 'known' in the definition. Philosophers try to avoid that sort of circle. The traditional philosophical definition of knowledge is 'justified true belief', which probably avoids a circle; but it is also wrong. Like pretty much every philosophical analysis, it is both too narrow and too broad. Thus you can know without having a justification, say if you just see something, and you can have a justified true belief without knowing, because it may still be just a matter of luck that your belief is true. The attempt to improve on the traditional definition has been a philosophical cottage industry for the last forty years, and we have not reached any closure on the issue. It is striking that we can be entirely competent at using a concept, yet so poor at giving its analysis.