Sometimes we speak of an agent's action being justified on the basis of particular normative practical reasons. At other times we speak of an agent's action being required on the basis of particular normative practical reasons. It is perspicuous that these claims do not have the same meaning. One may be justified in acting thus and so, yet not be required to so act. Nevertheless, we think that in order to be required to act, one would have to be justified in so acting. It is this asymmetry involving the different ways in which normative reasons can function in practical deliberation that grounds Gert's view that normative reasons play two logically distinct roles: considerations that require action and considerations that justify action. Normative reasons that function in a requiring role are considerations that make otherwise irrational actions rationally required, while normative reasons that function in a justifying role are considerations that make otherwise irrational action rationally permissible.

Traditional single-value accounts of normative reasons make the rational status of action merely a matter of weighing up all the relevant considerations on the balance of reasons that count in favour of or against acting in a particular circumstance, and acting in accordance with the option favoured by the strongest reason(s). As a result, such accounts maintain that the strength of sufficient justifying normative reasons will always be determinative in establishing a normative requirement to act. Gert seeks to challenge such traditional views as being too simplistic and in need of further subtlety in order to provide an adequate view of the terrain of normativity. On his view, in addition to different normative roles for practical reasons, there can be different justifying and requiring strengths for normative reasons, such that it is possible for the strongest reason on the balance of reasons rationally not to require action. On Gert's view, ‘no matter how strong the justification for some action becomes, it never follows, simply in virtue of the strength of such a justification, that one is required to do the action’ (p. 21).

While philosophers such as Bernard Gert (Joshua's father), Joseph Raz and Jonathan Dancy have also discussed the possibility of normative reasons having these two distinct roles of justifying and requiring, none has gone as far as Joshua Gert to identify and develop systematically an account of practical rationality and normative reasons that centrally incorporates this fundamental idea. Of particular interest for many will be an account of how the different roles that normative reasons play can elucidate why some reasons may not be relevant to the rational status of one's action and why some reasons possess normative force and authority independent of what one happens to care about. In addition to pressing this important distinction within the philosophy of practical reason, Gert uses this distinction to argue that there exists a class of rationally permissible actions that includes not only morally required action, but also the possibility of self-interested and immoral action. The upshot is that we can provide a robust explanation, for instance in the case of altruistic and moral action, for why there may be strong normative reasons that rationally justify acting in an altruistic or moral fashion, but that do not rationally require such action.

Concomitant to Gert's account of normative reasons is a theory of practical rationality that distinguishes between objective and subjective senses that has similarities to (and important differences from) Bernard Gert's theory of practical rationality. According to Joshua Gert, ‘an action is objectively irrational iff virtually everyone would regard the action as irrational, if they were fully informed about all nontrivial consequences of the action’ (p. 140), while ‘an action is subjectively irrational iff it proceeds from a state of the agent that (a) normally puts an agent at increased risk of performing objectively irrational actions, and (b) has its adverse effect by influencing the formation of intentions in the light of sensory evidence and beliefs’ (p. 160). Aspects of Gert's account, however, remain unclear in places where we are not provided with an explicit account of what would constitute full information, how awareness of the consequences of the action in coming to regard it as irrational grounds its objectivity, or how an internalist account of normative reasons avoids the problem of failing to have a reason to become more rational in the face of having no desire to be so (which one would have were one fully rational).

The vast majority of the chapters are previously published essays (some of which have been somewhat revised). One unfortunate result, which only mildly detracts from Gert's excellent work, is that his central distinction between requiring and justifying does tend to become overly repetitive at times and leads to uneven transitions between chapters.

There are also a few philosophical interpretations in the book that may strike some as idiosyncratic. For instance, at one point Gert describes J. S. Mill as ‘a psychological egoist, and not a rational egoist’ (p. 13), although Mill's theory of sanctions, as part of his account of why individuals would be motivated to act morally, clearly diverges from Bentham's psychological egoism. He also, as a result of his construal of his distinction between reasons internalism and externalism, labels dyed-in-the-wool externalists such as David Brink and Derek Parfit as ‘Kantian internalists’. Gert does so because he thinks that in relation to the question ‘Can we have a [normative] reason to act in some way, and fail to be motivated by it, even if we are fully rational?’ externalists answer ‘Yes’ and Kantian internalists answer ‘No’ (p. 172). This is, however, a somewhat odd classificatory principle, since a negative answer to this question need not be motivated by a distinctively Kantian view (especially a commitment to a view about an internal source of normative reasons, or tying the concept of a normative reason to the concept of rationality and motivation). Be that as it may, such idiosyncrasies do not detract from the overall thought-provoking arguments advanced by Gert.

On the whole, this is an impressive book replete with a number of valuable and intriguing arguments. The rigour and perspicacity of Gert's approach to the often tangled and bewildering literature concerning normativity reveals what seems to be a distinctive and profitable addition to the literature regarding the nature of practical rationality and normative reasons for action.