Concepts embody our knowledge of the kinds of things there are in the world. Tying our past experiences to our present interactions with the environment, they enable us to recognize and understand new objects and events. Concepts are also relevant to understanding domains such as social situations, personality types, and even artistic styles. Yet like other phenomenologically simple cognitive processes such as walking or understanding speech, concept formation and use are maddeningly complex.Research since the 1970s and the decline of the "classical view" of concepts have greatly illuminated the psychology of concepts. But persistent theoretical disputes have sometimes obscured this progress. The Big Book of Concepts goes beyond those disputes to reveal the advances that have been made, focusing on the major empirical discoveries. By reviewing and evaluating research on diverse topics such as category learning, word meaning, conceptual development in infants and children, and the basic level of categorization, the book develops a much broader range of criteria than is usual for evaluating theories of concepts.
We all know what ‘concepts’ are, and yet few of us are able to explain them.
The ‘classical’ view has lasted for about 2000 years, up to Eleanor Rosch’s work in the 1970s. According to that view concepts are definitions. We classify things (like the furry thing in the garden) under concepts (like dog) because the things meet a set of necessary and sufficient conditions in order to be categorised in that way. So our concepts are definitions constituted by those necessary and sufficient conditions.
As an account of concepts this has a simplicity and elegance which makes it attractive. It also fits nicely into the black and white world of logic where things are either p or not-p.
The problem with the classical view arises when we ask whether this is supposed to be a description of how humans do concepts, or is it an idealised model of perfect conceptualising which humans should aim for.
If it is meant to be a description, then it starts to look implausible because psychology experiments have shown that human practice is rather different. One of the key problems is that no one seems to be able to give the definitions which constitute concepts. And so there are disagreements about which concepts apply. For example, a salmon is a fish, but is an eel? If there really were a clear set of necessary and sufficient conditions, then we could answer this question easily by reference to the list. But we can’t.
And some concepts seem to be fuzzy, with ‘family resemblances’ rather than specific conditions of membership. Wittgenstein famously gave the example of the word ‘game.’
By chapter 3 the book takes it that the classical view is dead. And then begins the exploration of alternatives. We hear of prototype theory and the way it involves mental constructions of concepts from elements which are more and less familiar amongst the members of the concepts. And there is exemplarism which focuses upon specific members, as formative of concepts.
The book ends inconclusively by suggesting that all the models have important elements, and there is more work to be done before a successful theory of concepts can be formulated. This is frustrating, but an honest admission of the complexity of a problem which philosophers are still working on.
Stylistically, the book combines technical philosophical language with informal asides. For example the author tells us at one point ‘…I have waffled on the question of whether I am talking about concepts or words…’ (p385). It is a demanding read, but is accessible to general readers who are prepared to pause and think about passages, as necessary.
It was a comprehensive review, albeit a bit outdated. The author had strong preferences -- not a bad thing, but take it with a grain of salt. It would have been great if the presentation of empirical findings could be better balanced with computational models. To be honest, I don't have high hopes that another book like this would come out any day soon. The field has moved away from these fundamental, and possibly idealistic questions around categories since the 2000s. Now it's all about social cognition.
I read this for work. It is a very clear, carefully thought out book about what human experiments have taught us about the nature of concepts in the human mind-- how they are acquired, how similar concepts affect one another, whether they are formed by exemplars of prototypes, their relationship with categories, their structure, and so forth. The discussion of basic level concepts, and their relation to superordinate and subordinate concepts, was particularly interesting (since I was less familiar with the material). I would have liked more speculation about what kind of model could give rise to such results, but Murphy is too careful an author to indulge. The language is precise, while still being readable.