In two recent contributions to this blog, Professors Goldsworthy and Oliver have put forward two quite different bases for Parliament having sovereign power (and for courts not having the power to strike down legislation). For Goldsworthy, Parliamentary sovereignty is a fact about the constitution: it reflects the beliefs of most officials about the power of Parliament vis-a-vis courts. For Oliver, by contrast, Parliamentary sovereignty is a pragmatic necessity based on the practical need for mutual respect or comity between Parliament and courts
Which, if either, of these accounts is correct? In this post, I want to say something about how to answer that question. The topic, we might say, is constitutional methodology. My aims are twofold. First, I want to introduce this topic and some of the problems it throws up. Secondly, I want to offer an outline defence of one particular method of doing constitutional theory. I shall suggest after Ronald Dworkin that the theories of Goldsworthy and Oliver (along with every other theory of the constitution) must be understood as rival interpretations of British constitutional history and practice. A theory is correct, I shall say, if it provides the most morally appealing interpretation of the constitution.
Constitutional Methodology: Some Issues and Problems
Public lawyers disagree about many aspects of the constitution. They disagree, for instance, about whether, or in what sense, parliament is sovereign. They disagree about whether judges have the power to invalidate statutes. They disagree about the nature of the judicial role in judicial review proceedings. And they disagree about whether individuals have rights that cannot be removed by parliament. What is sometimes less obvious is that public lawyers also disagree about how to answer those types of questions. They disagree, that is, about what makes it the case that parliament has absolute legislative powers, or that parliament’s powers are limited by judicially enforceable legal rights and principles, or that some other distribution of rights, duties and powers obtains in the constitution.
Here are just three of the different constitutional methods that one can detect in the public law literature, beginning with an elaboration on the work of Goldsworthy and Oliver (above).
1. Goldsworthy implies in his work that constitutional theory is descriptive and morally-neutral in character. Parliament is sovereign, he says, because this is what most officials accept (i.e. say, believe, or otherwise practice). It is the Hartian ‘rule of recognition’ in the constitution. Whether Parliamentary sovereignty is morally acceptable is irrelevant to its existence. Those theorists who argue for a judicial strike-down power, he says, are simply making a factual mistake about the practice of officials.
2. Oliver argues in her work that constitutional theory involves a pragmatic (non-principled)judgement about what works in the constitution. The reason that Parliament is sovereign, and that judges respect that sovereignty, she says, is that a range of undesirable consequences would flow from any attempt by judges to upset that arrangement (e.g. damage to the reputation of the courts, institutional gridlock). A distinction has to be made, Oliver contends, between the theoretical/principled case for judges having a strike-down power and the practical workability of them having such a power.
3. For theorists such as Allan, Jowell, Craig and Laws, constitutional theory is thoroughly normative in character. The legislative powers of Parliament, and the adjudicative powers of judges, they argue, depend on the moral principles that justify those powers. It follows that if Parliament possesses absolute legislative power, there must be a compelling moral basis for such power. Allan, in particular, contends that there is no plausible justification for Parliamentary sovereignty. A more convincing moral reading of the constitution, he says, points to a theory of liberal constitutionalism and ‘strong’ judicial review.
Even taking this very small sample of different approaches to constitutional theory, it is not difficult to see the potential for confusion. If different theorists are using different methods to identify the powers of Parliament and courts, then their disagreements suddenly look rather futile. Indeed, it might be said that they are not disagreeing at all: that they are talking past each other or pursuing different projects. We can reinforce that gloomy diagnosis by imagining the following dialogue between Goldsworthy and Allan:
G – Parliament is sovereign because this is what most officials accept.
A – It doesn’t matter what most officials accept. My constitutional method does not depend on acceptance by officials, but on the normative justification for the legal powers of institutions. Parliamentary sovereignty cannot be justified. The principles of the rule of law and the separation of powers demand that judges have the power to invalidate legislation.
G – You are making a factual mistake. For centuries, judges and theorists have denied that judges can strike down legislation.
A – That may be so, but what judges and theorists accept, or have historically accepted, may diverge from the true normative powers of institutions.
G and A – (Audible sigh)
There is clearly something deeply unsatisfactory about this imaginary exchange. Each theorist begins their analysis in a different place: one begins their analysis with acceptance; the other begins with justification (incidentally, this is precisely the way that I read the actual exchanges between Goldsworthy and Allan). Is there any hope of resolving such a fundamental difference in approach? Before I attempt to provide a positive response to that question, let me first consider a possible negative response.
It may be that some constitutional theorists, at least, are indeed pursuing different projects. Goldsworthy and Allan (along with most other constitutional theorists, we can suppose) seem to be interested in identifying the true rights, duties and powers of institutions and individuals in the British constitution. In other words, they are trying to offer a theory of the normative effects of British constitutional practice. Other theorists seem to be less interested in, or less sympathetic to, that project. For political constitutionalists such as Bellamy, constitutional theory involves constructing an ideal model of the relationship between Parliament and courts. The theory of Parliamentary sovereignty, he implies, may or may not reflect the prevailing balance of power in the constitution, but it is the theory that approximates most closely to the ideals of political constitutionalism. In a different vein, Griffith can be read as bringing the perspective of a political scientist to constitution analysis. His interest is in the realpolitik of the constitution (i.e. where the sources of brute power can be found). Thus, it is the Government rather than Parliament, he says, that possesses sovereign power (perhaps it would be Rupert Murdoch or Tesco today…). There are hints of Griffith’s approach in the work of Oliver (above). By her claim that there is a gap between the reality of institutional powers in the constitution and the (normative) theory of those powers, Oliver is perhaps similarly attempting to offer the descriptive view of a political scientist rather than the normative view of a constitutional theorist. I shall resist that reading of Oliver in the section that follows.
Interpreting the British Constitution
I now want to offer a positive response to the question that I posed above (viz. is there any hope of resolving the fundamental difference in approach taken, for instance, by Allan and Goldsworthy?) Let me first try to sharpen the inquiry a bit. Our interest is in whether theorists who seem to employ very different methods of constitutional analysis (as illustrated in 1-3 above) can intelligibly disagree with each other. At the same time, we want to be in a position to say that one theory of the constitution provides the correct theory, or a better theory than some other theory. The object of the inquiry, then, is to arrive at some common framework, or some common measure of success for all theories of the constitution. It is the interpretive approach mentioned at the start of this blog item that, I think, holds the key to this ambition.
Rather than attempt a long and detailed exposition of Dworkin’s interpretive approach, let me spell out as plainly as possible a) how I think this approach makes disagreement between Goldsworthy, Oliver and Allan possible; and b) how, according to this approach, one of these theories can potentially provide the correct understanding of the constitution.
In answer to a), the level at which the disagreement between Goldsworthy, Oliver and Allan can take place must be more abstract than we have so far considered. In the imaginary dialogue above, both Goldsworthy and Allan make arguments that are internal to their preferred constitutional method: Goldsworthy argues on an empirical level about official acceptance, while Allan argues on a normative level about how to justify institutional powers. The real disagreement between Goldsworthy and Allan, I suggest, concerns the deeper question of why the powers of Parliament and courts should depend either on empirical argument or normative justification (or, to include Oliver’s approach, pragmatic judgment). These are rival approaches to constitutional analysis which each require a positive argument in their favour. Goldsworthy must explain why acceptance by officials is the decisive factor; Oliver must explain why pragmatic judgment is the decisive factor; and Allan must explain why moral justification is the decisive factor. By extension, each theorist must try to show why the factors identified by other theorists are erroneous.
This brings us to b). It will be tempting for some theorists to claim that their favoured method of constitutional analysis is descriptively correct: that their method is not so much a theory as a factual statement of the way the constitution works. This recalls the claims made by Griffith (and perhaps Oliver) above. It also recalls the way that Herbert Hart sought to characterise his approach to legal theory (it is not altogether clear to me whether Goldsworthy endorses that characterisation). Unfortunately, such a descriptive claim is doomed to fail. Every theorist presumably believes that his or her method provides the correct understanding of the constitution, and that other methods provide an inferior understanding. Every method identifies particular facts, features or standards of constitutional practice as representing the reality of the constitution. There is no way that any theorist can stand above or outside these mainstream debates. They must defend their method with arguments like every other theorist. That defence, I suggest, can only be an interpretive (or justificatory) one. Each theorist must attempt to show that their preferred method provides the most morally appealing interpretation of British constitutional history and practice.
These responses to a) and b) now require some unpacking. What exactly does an interpretive defence of any particular constitutional method involve? Goldsworthy, Oliver and Allan (along with all other constitutional theorists), I suggest, must show two things at once. They must first show that their preferred method has independent moral appeal as an account of law, government and the state. If a theorist relies on (or presupposes) a defective general legal and political theory, then we can almost certainly discount their theory as a viable account of the constitution. Secondly, assuming that their underlying legal and political theories are sound, a theorist must then show that their preferred method is capable of making sense of the salient facts and features of British constitutional practice. In other words, they must show that they are advancing, not a Utopian theory, but a theory of the particular constitutional practices and traditions in Britain.
What might an interpretive defence of the methods of Goldsworthy, Oliver and Allan look like?
Goldsworthy’s argument might (and, in part, does) go something like this. That the powers of Parliament and courts depend on a consensus of acceptance among officials can be justified by such values as certainty, clarity and constitutional stability. If members of each branch of government must agree to the powers of Parliament and courts, then the risk of a constitutional crisis is minimal. Neither Parliament nor courts can unilaterally unsettle the system. A theory based on official consensus also fits the day-to-day practice of judges and officials. Judges give effect only to the clear meaning of statutory text or to clear rules contained in previous judgments. There is no sense in which judges make moral judgments either about the proper powers of institutions or about the substantive content of the law.
Oliver’s argument might go something like this. That the powers of Parliament and courts depend on pragmatic, all things considered judgments can be justified by the values of comity and mutual respect. If courts and Parliament do not trample on each other’s territory, then things will work out better in the long run. Courts will maintain their authority, and there is less scope for a clash between branches of government. A theory based on comity and respect fits the day-to-day practice of courts and officials. Judges frequently defer to the opinion of officials, and they suppress any thought of striking down legislation. Likewise, Parliament and the Government generally refrain from criticising judges or judgments.
Allan’s argument goes something like this. That the powers of Parliament and courts depend on moral principles can be justified by values such as integrity, equality and a range of basic liberties. The very point of law and government is to ensure that people are treated in accordance with these fundamental principles. It follows that the state does not have the power to act in a way that contravenes such principles. A theory based on moral values and principles fits the democratic structures of government. It also fits the way in which judges bring arguments of constitutional principle to their adjudicative task.
These are merely sketches of what an interpretive defence of these different constitutional methods might look like. The crucial point to note for our purposes is how each theorist must go about defending their constitutional method and the theory they derive from it, and how they must go about attacking other methods and theories. Constitutional theory, I have suggested (at greater length than I initially hoped), must be interpretive (rather than descriptive, conceptual, logical or whatever).
Stuart Lakin is a Lecturer in Law at the University of Reading