Knowledge is costly, and hence it is only rational to pursue it till the point when a marginal increment in the amount of knowledge you get is less valuable than the marginal increment in the resources you should devote to getting it. As a matter of fact, this entails that most people will not pursue knowledge very far, except, given our division-of-labour society, knowledge about those things one is earning one's life with, i.e. knowledge which happens to be very valuable for him or her. An expressed goal of Russell Hardin's book How do you Know? The Economics of Ordinary Knowledge (actually a collection of previously published, independent papers) is to extract some interesting, and in some cases amazing, consequences of this fact. The first chapter is devoted to explain the notion of ‘ordinary knowledge’, and to contrast the ‘economic’ analysis of everyday knowledge with the ‘epistemological’ study of scientific knowledge of philosophy. Chapter 2 explains why it is reasonable that ordinary people have a poor acquaintance and understanding of scientific knowledge. Chapter 3 analyses the effects of rational ignorance on democratic participation. Chapter 4 is a short comment on Hayek's view of society and distributed knowledge. Chapters 5 to 7 are devoted to moral, institutional and religious ‘knowledge’. Chapter 8 discusses the notion of culture, and Chapter 9 applies some of the ideas of the book to the problem of fundamentalism and extremism. In the Preface (p. xi) Hardin describes his goal as follows:

The main problem with How do you know? is that it fulfils this promise in a rather limited way, mainly because the ‘looseness’ Hardin announces is too big: the proposed ‘theory’ is only economic in the sense that it recognizes that an ordinary person's acquisition of knowledge is most of the times not an economic activity at all, but mere happenstance, just a by-product of other activities (it is more or less as if one derived a ‘biological’ theory of rocks from the fact that rocks are not alive), but also partly because Hardin's ‘economic theory’ can only very loosely be qualified as a theory: it can be more appropriately understood as a collection of reflections on some factors that unintendedly promote ignorance, and on the consequences of this ignorance on some collective choice situations that have happened to be interesting for Russel Hardin in his previous works. The book becomes more suggestive, perhaps, if one just delves into it concentrating on the questions each chapter independently discusses, but without the misguiding assumption that one is to discover within those discussions anything remotely similar to an ‘economic’ unified and systematic approach to the distribution of knowledge, beliefs and ignorance amongst the crowd of ordinary people. Unfortunately (at least for the author of this review), it is easy to misunderstand the real goals of Hardin's new book.

Another concern about How do you Know? is that it leaves the reader without clear responses about the situations analysed in it. Of course, one might say that the aim of Hardin's theory is merely descriptive, or explanatory, about ‘what it is’, and has not the goal of suggesting how the world ‘should be’. At times, the book gives even the impression that Hardin defends a kind of fatalist approach to the question of ordinary ignorance: ‘there is nothing to do about it, full stop!’, but there is hardly any explicit defence of such a fatalism in the book, besides the general recognition that, being knowledge costly, it is not reasonable to expect that people get much more of it than what they've got. Instead, one would have expected, from a real ‘systematic theory’ about the causes of the distribution of knowledge in society, some suggestions about how to improve this distribution, either in general, or regarding the specific problem situations the author discusses. For example, it is disappointing that almost nothing is said about the organization of primary and secondary learning, besides some in-passing comments while talking about the disputes on intelligent design theory (chapter 2) and on the schooling of Amish children (chapter 9), and even in these two cases, Hardin hardly reveals in an explicit way any suggestion about whether the laws should be changed or maintained, or perhaps he does it in such a subtle way that I just missed it. Something similar happens in the discussion of democratic participation: are there some alternative ways of organizing democratic societies that supply ‘better’ outcomes, or at least that promote in some way a better knowledge and a better use of this knowledge by the common citizens? Is the curse of ignorance about political issues going to corrupt western democracies, or is it harmless? Again, no answer. Yet one small piece of advice is offered in passing while talking about religious fundamentalism (p. 158):

I cannot but applaud this wise recommendation, but one has to recognize that, as the only practical, economic or political conclusion from a thoughtful social science ‘theory’ about the causes and effects of religious beliefs, it is a rather meagre reward.

From a theoretical point of view, I have some other complaints about the vision of knowledge offered in How do you Know? In the first place, I find particularly disappointing the fact that Harding chooses to demarcate in an absolute way ‘ordinary knowledge’ from its ‘expert’ or ‘scientific’ counterparts, and, particularly, his claim that ‘epistemological’ considerations are relevant only for the latter, and not for the former (p. 24). It is dubious that epistemology as regularly practiced by philosophers has really a big relevance for understanding scientific knowledge; as a matter of fact, most of the people that would now qualify themselves as ‘epistemologists’ (i.e. those speculating about the ‘nature’ or the ‘essence’ of knowledge, above all within the analytic tradition) have an almost complete disregard for scientific examples of knowledge and by the intricacies of the scientific method: they are, rather, experts on what we might call ‘ordinary knowledge’, in a analogous fashion as to what philosophers of ordinary language did in order to replace the speculations of logical positivists on ‘perfect’ formal languages. Scholars outside the analytic tradition, who are still more reluctant to treat science as something like a rational standard, would consider even less justified this distinction between ordinary and scientific knowledge. Epistemology is not so much about the canons of scientific research, but about the meaning of concepts like knowledge and belief, and about the norms that regulate the use of those terms in everyday life. I do not see any reason why Hardin's project could not have been relevant for such an endeavour, nor vice-versa. But perhaps Hardin was not actually thinking in this passage about epistemology, but about a (sadly) different branch of philosophical studies: philosophy or methodology of science, which (at least sometimes) concentrates on the procedures that may warrant the cognitive quality of scientific knowledge items. Nevertheless, contemporary philosophy of science shows that there is not an absolute difference between scientific knowledge and other types of knowledge, it is more a matter of degree (often scientific hypotheses are also provisional, tacit, based on authority, not perfectly articulated from a logical point of view, and so on). What is more important, there are plenty of circumstances in which science and the rest of society interact, and some ‘hybrid’ forms of ideas, claims, beliefs, etc., emerge from this interaction and play a fundamental role in political and everyday discussions in the parliaments, the media, the companies, or on the street. It would be extremely interesting to have an economic theory of how this interaction and this ‘hybridization’ takes place, as well as on how some limits are conventionally settled at some levels of the continuum between everyday knowledge and scientific theories.

Second, I profoundly disagree with Hardin's strategy of mixing together knowledge with belief, and factual knowledge with moral knowledge. Regarding the former distinction, it may be true that, from the point of view of studying the influence that beliefs have on behaviour, it is not very relevant if an agent really knows what he thinks he knows, or just believes it. As a matter of fact, economic theory, or the rational choice theory the former is grounded on, does not usually refer at all to ‘knowledge’, but just to subjective probabilities as a mathematical representation of beliefs. In this sense, an ‘economics of ordinary knowledge’ might function in a similar way, and just consider the psychological, subjective, cognitive states of individuals as causally relevant (not the epistemological status of such states). But if one refers to his theory as an economics of knowledge, and not merely of opinions, one seems to be assuming that there is something that allows us to distinguish knowledge (either ‘ordinary’ or ‘expert’) from mere opinion, or from mere belief. As a matter of fact, it is simply not true that ordinary people do not make this distinction, for, though they usually take their actual opinions as true knowledge, they are also very conscious of the fact that not all opinions (e.g. many of the opinions of other people) deserve to count as knowledge, and each individual also knows that, about many things, he or she does not have a sure opinion, but just a less than full belief. One would have expected, from an ‘economics of ordinary knowledge’, some reflections about how people establish criteria that help determine what opinions deserve to be taken as knowledge, what to do in case of disagreement, and what to do in case of uncertainty. These criteria, of course, will not necessarily be the same ones as those of scientists or epistemologists, but it would be interesting to know what are exactly the differences between them, how do they evolve, what influence can the collision of ordinary and expert criteria have, and things like that.

Regarding the distinction between factual and moral knowledge, again I admit that from the point of view of the ordinary person, in practice one knows in almost the same way that Barcelona is farther from Madrid than Valencia, that omelettes are made so and so, and that it is wrong to steal, and the mechanisms propagating factual beliefs can be very similar to those allowing to propagate moral beliefs (though an assumption is not the same as a proof, of course: it would have been nice if some analytical discussion of the identity or the differences between these mechanisms had been offered). Nevertheless, it is difficult to swallow that ordinary people lack in an absolute way the capacity of distinguishing between ‘what is’ and ‘what must or should be’ (i.e. their ideas about the former and their ideas about the latter), and that this distinction is not manifest in some procedures that help people take decisions about what to do, at least when their moral or social norms do not imply it in a deterministic way. A study of these procedures would have been interesting in an ‘economics of ordinary knowledge’. But even admitting that factual and moral beliefs are not essentially different in the case of ordinary people, I think that How do you Know? is again a missed opportunity to analyse the best models of ‘psychological states’ that might be used in the social sciences; for standard rational choice theory does not only take for granted that moral beliefs are different from factual beliefs; it does not even consider moral beliefs as a kind of beliefs at all: beliefs are represented by the subjective probability function, whereas moral considerations affect exclusively the subjects’ utility functions, making some actions or consequences more or less morally desirable. Instead, Hardin's suggestion that moral and factual ideas are of the same kind could have made us reflect on the possibility of building a different model of behaviour, one which allowed to make sense of Hardin's suggestion that ‘we might in some contexts be able to argue convincingly that some beliefs virtually entail actions, so that believing is tantamount to acting’ (p. 196). I think that, for example, a model of belief and action like Robert Brandom's inferentialism (Reference Brandom1994), in which the notion of commitment is central, might play this role, but How do you Know? offers few hints about what we might find along this unexplored path.