The essence of Shapiro's view of law appears on the second last page (p. 399): “Although it is not a particularly inspiring or romantic description, the law is, in the end, an instrument”. Law, for Shapiro, is an instrument designed by a designer or group of designers in order to solve certain problems. It is, further, a necessarily efficient instrument, for its solutions obviate the need for moral or political deliberation, thereby reducing deliberation costs. What problems does it solve? It solves problems that arise for large groups of people with diverse value commitments, and thus problems that require solutions that are “complex, contentious or arbitrary”. How does it solve them? It solves them by “providing a highly nimble and durable method of social planning” (p. 213). Who solves these problems? The problems are solved in the first instance by designers of the ‘master plan’, and thereafter by officials who are appropriately self-disciplined, i.e. those who “think inside the box” (original emphasis, p. 398), suspending their own moral and political judgements, and sticking to a process of interpretation of the master plan and its sub-plans that does not unsettle the very problems that the law aims to settle.

All this Shapiro elaborates on, and argues for, with great clarity and gusto. His writing is patient; one hears at times the voice of a teacher (especially in his summaries of Austin, Hart and Dworkin); and sometimes the voice of a father (“try to explain law to a five-year old”, he implores; see p. 51). He has plenty of thought-experiments, including both imaginary genealogical stories (e.g. the creation of an Island community, and its eventual need for law) and everyday anecdote-like scenes or sequences (e.g. the description of the Cooking Club). And even when discussing very complex and abstract concepts, Shapiro always tries (on the whole, very successfully) to be simple and clear. As a result, the book is long (400 pages), but all the better for that (a rare achievement).

Shapiro is a battle-hardened positivist, and some parts of the book read like classroom-rehearsed defences to now very familiar complaints against positivism. Indeed, the second half of the book – chapters 8 to 13 – are essentially 1) a summary of Dworkin's attack on positivism; and 2) a defence of positivism against Dworkin (the essence of which is that we should be associating law with ways of obviating, not requiring, moral and political deliberation). The last chapter, 14, is a defence against certain traditional natural law objections (principally as articulated by Finnis and Fuller). As has become customary since Hart, Austin is sacrificed on the positivistic altar (in chapter 3, which largely follows Hart's criticisms of Austin), and, as is less customary, the discussion of Hart (in chapter 4) is restrained. Indeed, although many aspects of Hart's account are praised (including later in the book, when discussing legal reasoning), Shapiro announces his departure from Hart, arguing (most importantly) that Hart made a “category mistake” in describing the rule of recognition as a practice. Rules, says Shapiro, are “abstract objects … they are objects of thought, not entities that exist within space and time”. Practices, however, are “concrete events. They take place within the natural world and causally interact with other physical events” (p. 103). Something – in this case, the rule of recognition – cannot both be a rule (an abstract object) and a practice (a concrete event).

Shapiro's improvement on Hart, his original contribution to the philosophy of law, and also the key to his defence of positivism, rests on the idea of plans, including the activity of planning and interpreting plans. Here, Shapiro draws heavily on the work of Michael Bratman, according to whom ‘human beings have a special kind of psychology: we not only have desires to achieve complex goals, but we also have the capacity to settle on such goals and to organise our behaviour over time and between persons to attain them’ (as summarised by Shapiro, p. 119). Shapiro's self-styled positivism, then, is planning positivism, or, as he calls it, the Planning Theory of Law.

There are a number of reasons why Shapiro thinks plans are an improvement on, for instance, rules or practices. The most important feature of plans is that they do the thinking for us – and yet, they do so in ways that are arguably more flexible than rules. Plans, says Shapiro, are “partial, composite and nested”; simplifying this a little, plans can be designed in more or less detail from the beginning, but even when designed with some detail, they can be – and typically are – filled in as we go along. In this respect, Shapiro attempts – though he is not always as explicit about this as he could be – a balancing act: on the one hand, plans must be sufficiently detailed in order to do the thinking for us, obviating the need for deliberation on their merits, but on the other hand, they must not be so rigid as to not enable us to fill them in.

Shapiro's image of plans seems to be one where they stabilise both ends and means (those means rationally connected to the ends), where any subsequent filling in is but a rational way of further stretching the means, but never, it seems, a way of challenging either the means or (especially) the ends. For example, I can form the intention to cook dinner this evening, and the means that is connected to that end might involve, for instance, going to the supermarket. When I arrive at the supermarket I can then decide to fill in the details as to what I will buy (see p. 123). Fair enough, but at no point does Shapiro consider the possibility that my seeing what is available (or better, what is not available) at the supermarket throws up a challenge to my original intention/end (e.g. there is no meat, and I cannot imagine cooking dinner without meat). This point might seem trivial, but it is not: it indicates a certain difference in attitude between plans and rules; I may be less likely to reconsider the applicability of a rule (e.g. adopting a rule that I ought to cook tonight) than I would a plan (in other words, I might change my preference for having dinner with meat; more generally, then, one might argue that whereas rules express commitments, plans express preferences). Capturing this difference in normative attitudes might be key to a sufficiently psychologically rich theory of law (more on this below).

Plans are different to practices because they are “abstract propositional entities that require, permit, or authorise agents to act, or not act, in certain ways under certain conditions” (p. 126). This, of course, renders them very much like rules, but the implication here is that, even where law looks like it is full of rules, what it is really full of are plans – legal rules are plans (after all, as noted above, them being plans makes them, in Shapiro's eyes, more flexible than rules). Here, we might wonder whether Shapiro might have considered that plans are expressed differently than, or take on different forms to, rules. Imagine we adopt Shapiro's view of law as made up of plans: might there be a specific art to drafting plans, as distinct from rules? If plans are to be abstract, propositional entities then, presumably, the form they will take will differ from rules (e.g. perhaps they will be propositions more tentatively formed, or ones that easily enable filling in).

Plans, says Shapiro, are also norms – or at least the plans that characterise the law are norm-like. They are guides for conduct and standards for evaluation. These norm-like plans are also “positive entities” (which is important, for also renders them different to other kinds of arguably non-positive norms, such as the norms of logic). Given that they are positive, their status as norms depends on them being ‘created by adoption and sustained through acceptance’ (p. 128). Shapiro has some discussion of what it means to “adopt” a plan, but less so about what it means to “accept” one. This is a pity, for one of the key issues here, if we are go along with Shapiro's characterisation of plans as norms or norm-like, is whether there is a distinctive kind of planning normativity – and thus a distinctive kind of matching normative attitudes associated with plans. On one view of it, rules, norms, conventions and plans are all ways in which we might go about reducing future deliberation costs – but are they, in that sense, nothing more than rough markers for picking out some of the attitudes that we might have to the future? Is not what is key here less the abstract object (the plan or the rule etc), and more so the kind of attitude behind it, the point being that there is no neat correlation between kinds of abstract objects and normative attitudes, and that there are many more of the latter than the former? If so, one might have wanted more detail on what Shapiro means by “acceptance”.

Perhaps the key to the advantage that plans have for Shapiro (as indeed they do for Bratman) is that they can be shared, or better, that they enable a specific kind of joint intentionality. Indeed, what is crucial about the applicability of plans and planning to law is their power to co-ordinate human behaviour and their ability to do so in “complex, contentious and arbitrary environments” (p. 135) where there are doubts and disagreements about how to solve particular problems (Shapiro characterises these contexts as “circumstances of legality”, it being further law's “moral aim” to solve the “moral deficiencies” of those circumstances – though why this aim and these deficiencies are moral is not entirely clear). Plans are so useful in this context because they enable further planning: we can specify, in a plan, who, in the future, is to plan and how. Because there are so many of us, and we are all busy with our lives and often alienated from governance, we may all agree – adopt and accept – a “master plan” that enables further planning. Further, because any such group may have differing attitudes of trust to those who it delegates planning to, so we may have different methodologies for interpreting plans in the future (chapters 12 and 13, which make this argument, are for this reader the standout chapters in this book).

Shapiro's book is unabashedly universalistic in scope – he sets out to offer a theory of the nature of law, answering what he asserts is a vital, fundamental question: what is law? And yet, his answer – that law is a form of social planning – is remarkably parochial. Discussions of a “master plan” (a term which re-appears throughout but is never quite defined) cannot help but remind the reader of the Constitution of the United States. The discussion of interpretative methodologies – for instance the debate between textualism and purposivism (see chapter 13) – also feel very US-based, where so much is at stake in interpretations made by the Supreme Court. There are moments when Shapiro qualifies his view by reference to the common law, though these are brief and not particularly convincing (can the common law really be described as an instance of incremental, bottom-up planning?). But the great majority of the discussions – including the few historical examples (e.g. the reconstruction of the intellectual backdrop to the Constitutional Convention of 1787 in chapter 11) – are very much US-based.

But in evaluating Shapiro we ultimately have to go back to the beginning: is law really best thought of as an instrument, designed by a designer or group of designers? Shapiro's view tries to make some room for non-designed elements (non-planned elements, to use his terminology), but only where the master plan permits it and validates it. He also tries to define social planning broadly enough to include decentralised mechanisms and bottom-up practical reasoning. But time and again he assets that “legal institutions are institutions of social planning and their fundamental aim is to compensate for the deficiencies of alternative forms of planning in the circumstances of legality” (p. 171; these alternative forms including, for instance, customary law). One cannot help but feel that Shapiro has in mind a particular ideal: a political order designed by enlightened founders, supported by a foundational text (the master plan), with institutions staffed by self-disciplined officials, trying their best to honour the wisdom of those original designers. This is, perhaps, at best, a form that law can take (or serve?), but surely not the nature of law?

Shapiro is one of our generation's best legal theorists. This book is a very considerable achievement, and there is much to learn from it, both for new and seasoned students of jurisprudence. The association, for one, of law with the trust-sensitive management of deliberation costs is a fascinating one to explore further (the figure in the background here, and one which one would have expected, in this respect, to play a more explicit part in the book, is Raz). The one hope this reader has is that, in future writing, Shapiro will be willing to challenge his analytical talents by the riches of legal history and comparative law outside the US context.