Despite its broad title, this book explores how far various strands of Joseph Raz's legal theory are internally consistent and present a coherent picture of law in modern societies. Given that the author is reviewing a series of works written over nearly 40 years, in one sense it is not surprising if the various ideas developed over that long time are not fully consistent. But it matters whether there is a coherent legacy from all his writing. Martin offers a valuable challenge to the interest of that legacy, but is less effective in challenging its internal coherence.

In terms of internal coherence, a central part of the author's critique is that what Raz has written on the subject of adjudication undermines the claims of the sources thesis. She argues “At the very centre of his account of the nature of law is an understanding of the role of judges” (p. 3). Raz's sources thesis argues that it is possible to identify what is law and what is not without recourse to moral judgments. Martin argues that, in central ways, this thesis does not work. In particular, the identification of legal rules involves not only a factual enquiry about which textual provisions were enacted by a legislator or decided by a judge in a previous case, but requires interpretation. That exercise of interpretation involves the making of evaluative judgments – which is the best way to achieve the purpose of the provisions? In Martin's view, common law reasoning is particularly difficult to fit with the sources thesis in that there is no canonical factual source. Common law reasoning is not even a matter of identifying a rule and grafting on to it an exception, but, following Simmonds, it is typically a matter of casting the rule in a new light (see Chapter 2). Treating this feature as a peculiarity of common law reasoning is not helpful to her argument. Anyone familiar with codified legal systems realises that there is no greater difficulty in finding the English rules on trusts than the current norms associated with particular articles of a civil code (e.g. article 1384 of the French Civil Code). Martin would have been better stressing the distinction between identifying a legal text (code, statute, or precedent ruling) and identifying a legal norm arising from any other those sources (what must I do from the legal point of view?). As Simmonds explains, casting in a new light is a general feature of legal reasoning. This leads her in Chapter 3 to explain Raz's “morally robust theory of adjudication” in which the legal reasoning associated with deciding cases in which there are difficult questions of law is seen as a subspecies of moral reasoning. Chapter 5 again returns to the issue with a discussion of Postema's critique of Raz (which also shapes the concluding Chapter 7). Postema's argument that interpretation often involves judgments about the “reasonableness” of understanding a previous case or statute (p. 110) makes it hard, she says, to characterise the citizen or judge as engaging in purely factual identification of legal rules. She picks up Raz's statement in his debate with Postema that “much of legal reasoning is interpretative reasoning, and interpretative reasoning is not, in general, autonomous” (cited e.g. p. 159, also p. 109). She concludes that this undermines the sources thesis because it brings in moral argumentation into the process of deciding what the legal rules are. But, in her conclusion on p. 159, she accepts that reporting the assumptions and evaluative views of legislators is not moral argumentation. So, when judges say that they are interpreting the intention of Parliament, they are in fact undertaking Simmonds's casting the words of the statute in a new light (cf. Lord Reid, Black-Clawson [1975] A.C. at 613), rather than deciding on the basis of their personal moral code. So not all forms of interpretation are moral reasoning of a kind that undermines the sources thesis.

I do not agree that the theory of adjudication is a central concern of Raz. It is a matter he addresses on numerous occasions in response to others. But it is not his principal starting point. Martin recognises the seminal character of Practical Reason and Norms (“PRN”, OUP 1999, originally 1975) in Raz's thinking (see pp. 12ff.). That work made central use of the idea that the law works by giving citizens exclusionary reasons – reasons that pre-empt the normal process of moral reasoning and encourage the citizen, in Bankowski's words, “don't think about it, do it”. The classic instance of exclusionary reasons is the soldier driving a tank who is expected simply to obey the orders of his commander, not to engage in moral reasoning (except in the most extreme cases) (cf. PRN, p. 38). Although Martin recognises that PRN focuses on the reasoning of citizens and not on adjudication, the main thrust of her own work is on adjudication.

It seems to me that Martin does not give adequate place in relation to the sources thesis to the importance in PRN of reasoning from the legal point of view. There is a difference between the citizen, who is a person whose status within the legal and political system requires loyalty to the law and who is expected to act from the legal point of view, and the ethical human being who is expected to take a holistic viewpoint about the rights and wrongs of particular courses of action. An obvious example of the difference is the leading French Conseil d'Etat judge who was a Communist, yet was recognised as one of the best exponents of French administrative law of his generation and who wrote one of the leading textbooks on the subject. Raz's idea of the legal point of view makes sense of that man's life, which involved arguing normatively in the afternoon about what French administrative law required of state officials and citizens, and then talking normatively over dinner in the evening about the need to bring down the French state. Of course, as Martin points out forcefully and Raz clearly acknowledges, there is not a bright line between acting from the legal point of view and making personal moral judgments. First, you have to make an existential commitment to act from the legal point of view. That requires a holistic judgment, an act of faith for which one can give justifications. Secondly, there are issues of interpretation. Now, here, there is a difference between doing one's best to continue to reason from the legal point of view in a situation of uncertainty and simply closing the law books and saying that you will decide according to your personal moral code, like Azdak in Brecht's Caucasion Chalk Circle. In the former, the citizen or judge is taking personal responsibility for an interpretation of the legal point of view, but can legitimately say “left to my own personal convictions, I would have acted differently”. Martin is right that Raz tends to blur these two approaches. But the solution is not to abandon the sources thesis as incoherent, but to revisit what is meant by the legal point of view.

Martin does actually suggest a way of understanding the legal point of view, particularly in Chapter 6, which sets out her own point of view and arguably a more attractive agenda for jurisprudence than the analytical approach of legal positivism. She comments that “Raz's conception of law as a set of content-independent reasons for action tells us little about how law actually works in the world” (p. 133). For her, we should focus on how a citizen reasons in a normal legal system – how does one give an account of the law as part of their reasons for action. People want to do the right thing. But law is technical and requires much study. As Martin suggests, ordinary people are not typically guided by the pure text of the law, but by the way its requirements are transmitted through various intermediaries, lawyers, agencies, friends, family, etc. The interesting question is how far legal requirements fit into people's other values. What is the normal relationship between what the legislator enacts and what people accept as right? The imperfect formulation of law and the need for change makes law in the real world always in a process of becoming, and that positivism's abstract and stylised account of law fails to capture.

It is a pleasure to read a work of jurisprudence that is accessibly written. Given the complexity of the subject matter, the author has done well to present it in a way which can be understood easily. There is some significant overlap in the main ideas discussed in Chapters 2–5 and 7. In particular, Chapter 7 is less of a conclusion than a kind of appendix to the main thesis, which more naturally ends with Chapter 6. But it is often difficult to restructure a thesis into a book and the author has basically performed that task well. Her serious and thoughtful engagement with Raz's output over 40 years merits being read widely.