I've enjoyed many of Robert Goodin's books in political theory, but the arguments in this one don't compare favorably to his previous work. The chapters have been put together from research papers. As usual the separate origin of each chapter is reflected in the argument, which becomes excessively detailed in some places and is insufficiently integrated on the whole. I was particularly annoyed by a section on the Condorcet jury theorem and Bayesian probability. These topics are clearly peripheral to the main argument. After 50 pages of elaborate but completely pointless formalism the author finally comes to the obvious conclusion: political decisions don't have an objectively correct yes/no answer.

As for the main argument, "democratic deliberation within", I have to say that I found this idea unworthy of publication. The point is simply that democratic decisions would be improved if most voters would imaginatively put themselves in other shoes before voting. The interests of disadvantaged and disenfranchised people, and even those of future generations and the natural world, could supposedly be democratically protected in this way. The author does not say much about how this paradisiacal ideal could be reached in practice, but apparently it would be helpful if people read more novels and broadened their spheres of social intercourse. Frankly, a professional political theorist should come up with better ideas than this. I would direct prospective readers towards Goodin's other books, such as Reasons for Welfare, before this one.