L ego, puto, scribo
In much legal scholarship, methodology is an instance of a ‘known unknown’. Academic lawyers know that their university colleagues in any number of fields of inquiry use, debate, and study a range of different approaches to answering questions within their respective disciplines. Well aware of the fact of methodological debates in other disciplines, many academic lawyers fail to see the reason why legal scholarship should burden itself with such distractions. Indeed, in witnessing some of the methodological debates animating colleagues in private law theory and general jurisprudence, some academic lawyers may see reason to avoid the burdens of methodology altogether. They will believe that these debates about how to approach research answers to research questions illustrate the quip about university politics often attributed to Henry Kissinger: the debates are so vicious because the stakes are so small.
Although it will be acknowledged that the ‘law and … history, sociology, anthropology, economics, etc’ scholars among us enrich the scholarly field, it will be said that they do so by bringing the methodological commitments of other disciplines to the law, rather than by exposing the methodological presuppositions of legal scholarship itself. Legal anthropology assists the researcher in understanding the internal point of view; socio-legal studies draw attention to feature of the law-in-action; and so on. However, to continue the caricature of academic lawyers, let us say that if legal scholarship was forced to adopt a methodological motto, it might content itself with: I read, I think, I write.
In his MLR contribution, Chris Thornhill argues that the ongoing theoretical debates between proponents of legal constitutionalism and proponents of political constitutionalism suffer from a methodological flaw. That flaw, for Thornhill, is this: the ‘debates have, to date, been conducted in something close to a sociological vacuum, showing only limited consciousness of the material forces driving the rise of the legal or judicial constitution across different societies’ (212). As a result of this inattention to the sociology of constitutions (their real world formation and evolution), Thornhill argues that the ‘theories expressing a preference for one or other model of constitutionalism do so without appreciation of the reasons why certain patterns of constitutional normativity have evolved, and without recognising how particular constitutional norms are correlated with broader social processes’ (212). Thornhill’s charge against the legal and political constitutionalists and their debates is formulated in an unforgiving way: ‘expressions of theoretical preference … tend to be reductive, socially abstracted, and evidentially questionable’ (212). In other words, the inattention to methodology has lead one of the dominant constitutional debates astray.
Thornhill claims that greater attention to the methodology informing the political vs legal constitutionalism debates would allow ‘legal and political theorists to test their analyses through a material reconstruction of constitutional law and the social settings in which it evolves’ (213). Thornhill’s invitation to ‘think sociologically’ about constitutions is part of his more general scholarly commitment to ‘sociological constitutionalism’ (see here and, more generally, here). In the specific context of his MLR contribution, however, his target is the legal vs political constitution debate and, especially, the proponents of political constitutionalism. He never identifies one or more specific proponents as especially guilty of the charge of ‘sociological under-reflection’ (213). There are footnote references to JAG Griffith, Richard Bellamy, Graham Gee, Adam Tomkins, and myself, among others (including some, like Martin Loughlin, who would resist the political constitutionalist label), but no one proponent of political constitutionalism is awarded pride of place in conversation with Thornhill. Given that Thornhill’s argument alleges insufficient attention to detail by others, it should set for itself a rather high standard for detail. Against that standard, it is liable to the same order of criticism it levels against the political constitutionalists. For example, a more searching review of the scholarship would complicate Thornhill’s claim that ‘the defence of political constitutionalism is badly let down by its rather conventionalised, even simplified, use of its core concept: politics’ (213). A more detailed review of the scholarship would need to confront Marco Goldoni’s ‘internal critique’ of political constitutionalism’s focus on ‘ordinary politics’ to the exclusion of ‘constitutional politics’, a theme that Goldoni has explored more than once, including with Chris McCorkindale. Neither Goldoni nor McCorkindale are referenced in Thornhill’s argument.
There are other aspects of Thornhill’s framing of the legal vs political constitutionalism debate that give pause, including his emphasis on comparisons with political systems that differ fundamentally from the Westminster system that animates much political constitutionalist scholarship. My focus, however, is on the basic motivation for his argument; namely, that the theoretical debates, and especially the theoretical commitments of political constitutionalists, fail by reason of their sociological under-reflection. I argue that the charge is misdirected. To see why, consider how Thornhill deploys his sociological method.
A Sociological Invitation
Thornhill argues for a ‘more sociologically reflected understanding of politics, and of the political content of a society more widely’ by using ‘a macro-sociological perspective to observe ways in which, in different societies, certain interactions have been constructed as generically and categorically political’ (213). His argument runs, first, by outlining different aspects of constitutional development. The claims here are made in very broad strokes (‘most contemporary societies’, 208; ‘In most societies’, 217; ‘Almost uniformly’, 217) and without much by way of supporting evidence. Second, Thornhill’s argument runs through three case studies: Russia (227-233), Kenya (233-240), and Bolivia (240-246). The conclusion of his argument is stark: ‘before the intersection between national laws and international human rights laws, few societies had been able enduringly to abstract and to preserve a clearly and distinctively political structure’ (246). That conclusion is supplemented by the equally strong claim that, ‘[i]n most societies, it is only by virtue of the fact that they are accompanied, insulated, and internally checked by, international human rights norms that conventional patterns of political-systemic formation have become sustainable’ (246). On the specific relationship between his conclusions and the legal vs political constitutionalism debate, Thornhill concludes his MLR contribution with this thought: ‘the conventional distinction [between] political and legal or judicial constitutionalism appears sociologically ill-constructed’ (247).
The claims supporting Thornhill’s conclusions are all claims of fact, of the way the world is. They are based on the sociological method ‘to observe’. As claims purporting to describe a state of affairs, they stand or fall on their truth as descriptions. It either is the case that international human rights norms have allowed ‘conventional patterns of political-systemic formation to become sustainable’, at least for ‘most societies’, or it is not the case. It either is the case that only a ‘few societies’ were able ‘to preserve a political structure prior to the intersection of international human rights law and national laws’, or it is not the case. I have my doubts about the veracity of these conclusions, but to argue against Thornhill’s conclusions, I would need to look out into the world and observe the history of ‘most societies’ to see if his claims hold for all but a ‘few societies’.
To do so in a manner that engages with Thornhill’s specific project, I would need to share his framework, including his posited (but not argued for) account of politics (213-214):
Politics is understood here, sociologically, as a structure of inclusion, in which societies, over time, acquire the capacity to set some norms and some collective functions apart from everyday processes of social contestation, and in which decisions are produced which can be plausibly explained and justified, and so have a reasonable chance of gaining compliance, across the divergent sectoral, functional, and geographical fields that a society contains.
Even accepting this account of politics, my attempt to measure the accuracy of Thornhill’s conclusions would be complicated by his penchant for labels. He uses (I think interchangeably) the terms ‘legal constitutionalism’, ‘judicial constitutionalism’, ‘transnational judicial constitutionalism’, and ‘judicial-constitutional model’; in addition, he deploys (I think not always interchangeably) the labels ‘political constitutionalism’, ‘classical or political constitutionalism’, ‘corporatist constitutionalism’, and ‘privatilistic constitutionalism’. Any attempt to evaluate the truth of Thornhill’s claims against the evidence will be complicated by his frame of analysis, which at times does not assist in the clarity of his exposition. Nonetheless, those seeking to measure the truth of Thornhill’s claims would do so by reference to the states of affairs under study.
Criticism of Thornhill’s argument would miss its mark if it sought to argue that his conclusions are wrong or false on their moral merits; that, although factually true, the states of affairs he describes are not as they ought to be. Along similar lines, criticism of Thornhill’s argument would be misdirected if it claimed that his account of the rise and content of legal constitutionalism does not align with TRS Allan’s theory of legal constitutionalism or, again, that what is being described falls prey to the evaluative criticisms of Bellamy or Tomkins regarding the failings of legal constitutionalism. These criticisms would all fall flat because, as I have argued, it is possible to describe and explain a state of affairs without evaluating its moral merits, but instead by attending to the evaluations of the persons whose states of affairs are under study. In other words, it is possible to ‘observe’ without ‘endorsing’. There is no requirement that when I describe, I need also to endorse the content of what is being described. So, when Thornhill reports that the source of legitimacy has shifted from a powerplay of forces within the political state to international law, I read him as employing the sociological account of legitimacy (reporting what the relevant persons under study consider to be legitimate). I do not read him as employing a normative account of legitimacy (signalling the justification of this shift as good and right) that would signal his endorsement.
Thornhill’s sociological description of major shifts in legitimation and constitutional ordering is thus immune, as a description, from theories arguing in favour of the moral merits of alternative constitutional arrangements and from theories arguing for different major shifts. Thornhill’s argument is vulnerable only to competing and better descriptions of the same subject matters under study. It is this separateness of the descriptive and evaluative projects that is at the heart of the original misstep in Thornhill’s challenge to political constitutionalists.
A Theoretical Rejoinder
As Graham Gee and I have argued, political constitutionalist scholarship took a distinctively ‘normative turn’ in the post-Griffith era. Some, like Tomkins, have offered normative re-interpretations of aspects of the British constitution, breaking from received understandings, and thus injecting a normative quality to much of the day-to-day workings of the constitution. Others, like Bellamy, have developed more explicitly theoretical accounts, which are of immediate relevance to our understanding of the British constitution, but without being descriptive of its day-to-day workings. These and other self-consciously theoretical projects by political constitutionalists stand or fall on the evaluative merits of their propositions: the moral merits of their conceptions of political equality, political activity, citizenship, reasonable disagreement, majority voting, and the like.
Political constitutionalist scholarship is sometimes accused of taking an idealised view of politics and political activity that abstracts away from the horse-trading, the passions, the crowd mentality, and the excessive partisanship of politics in Westminster or Washington or elsewhere. The accusation is important. It warrants careful attention and either rebuttal or response. It is an accusation that holds, however, only insofar as the political constitutionalist advocates reform of politics in Westminster or Washington or elsewhere along the lines of his or her scholarly theory. As a theory, however, the argument that the theory is idealised is neither here nor there. The accuracy of the theory’s premises or conclusions when measured against the real world is not a condition of the theory’s success as a theory. The theory’s success rather turns on the reasons favouring it: are they sound and coherent or not? Neither Hobbes’ state of nature nor Rawls’ veil of ignorance is liable to the charge that they are ‘idealised’ because there is no historical record of either. They are liable only to charges that their accounts are unreasonable on account of their conception of the human condition. (For an argument along these lines, see Michael Sandel’s criticism of Rawls’ Theory of Justice.) A charge that their theories are ‘idealised’ can hold only if it is a charge that challenges the soundness of the reasons favouring their theories, perhaps because one or another theory is inattentive to the human condition and so cannot adequately construct a theory of political activity. The merits of a theory rest not on whether it ‘fits the facts’ of this or that state of affairs, but whether it ‘fits reason’.
This distinction between theory and fact appears to be alive in Thornhill’s account of political constitutionalism’s theoretical focus. He notes that ‘[t]he contemporary rise of judicial constitutionalism, and its contrast with political constitutionalism, thus gives acute expression to a body of long-standing, diverse, and cross-national theoretical controversies’ (211, emphasis added). The ‘eclectic front of constitutional theorists’ who defend political constitutionalism do so, highlights Thornhill, by ‘invoking theories of political conflict and contested will formation as the foundation for their rejection of judicial power’ (211, emphasis added). Thornhill’s focus on theory as the main register of recent political constitutionalist scholarship is accurate.
And yet, his accurate report makes all the more surprising his pivot to criticise that scholarship on sociological grounds. His claim that ‘theories expressing a preference for one or other model of constitutionalism do so without appreciation of the reasons why certain patterns of constitutional normativity have evolved’ (212) misses its mark. Thornhill’s criticism is in the factual register; the theories he criticises are in the normative register. As theories, they are not subject to accusations of ‘sociological deficit’ (212) or to claims that they are inattentive to ‘sociological motivations and societal pressures which, in different concrete socio-historical contexts, have shaped the growing judicial emphasis of constitutional law’ (212). The political constitutionalists’ ‘expressions of theoretical preference’, as Thornhill puts it, do not ‘tend to be … evidentially questionable’ (212) for the simple reason that they do not purport to describe a state of affairs in Russia, Kenya, Bolivia, or elsewhere, be it today or at some other historical moment in time.
Thornhill’s proposed ‘sociological approach to current patterns of constitution making’ may, as he conceives of it, be ‘intended to add a dimension of empirical sociology to debate about legal and political constitutionalism’ (211-212), but that is not what it succeeds in doing, if it succeeds at all. It does not add a new dimension to existing theoretical debates about legal and political constitutionalism. Rather, given his sociological method, Thornhill is best understood to be attempting to add a new debate on the relationship between legal and political constitutionalism, one that runs in parallel to, but never crosses into, the theoretical debates. Insofar as his argument succeeds, Thornhill has shed light on important but neglected dimensions of the constitutional terrain. But, in drawing attention to the sociological dimensions of that terrain, we must recall that his contribution is not in the same register of the political constitutionalists he purports to engage with.
Keeping the factual and normative spheres separate helps clarify the different ways in which the appeal to legitimacy should be understood. When reporting the views of political constitutionalists, he writes that political legitimacy ‘cannot be occluded – say, by judiciaries, international bodies, or formal bills of rights – against the particular dynamics of social contestation’; it must instead ‘necessarily reflect, and mediate between, diverse positions in the spectrum of social conflict, dissent or deliberation’ (212). Political constitutionalists, writes Thornhill, perceive that the ‘immediate democratic provenance of law is a vital indicator of its legitimacy’ (212). These claims about legitimacy by proponents of political constitutionalism are not sociological; they do not purport to report the views of citizens of this or that state, at this or that time. They are normative, saying that the justified way in which a community should order its decision-making procedures is this rather than that way. That normative claim survives even if a great majority of citizens of this or that state, at this or that time, think that decisions on fundamentals should be made by the flip of a coin, or by the Supreme Leader, or by international economic institutions. Against the standard of legitimacy (normatively understood), any citizen who thinks in this way is wrong and holds an illegitimate (unjustified) view.
In turn, when Thornhill reports on his sociological study to affirm that ‘the primary mechanism through which states seek to obtain legitimacy’ (223) is via international human rights norms and their judicial enforcement, he is reporting on a sociological fact (if fact it is). He is not making a claim that states are legitimately (as in justifiably) acting when they do so. He is reporting that they do think this. The only one who can be liable to the charge of being wrong here is Thornhill: he is either correct in claiming that this is what states perceive to be legitimate or he is incorrect. He presents insufficient evidence in support of his claim for his reader to draw independent conclusions. Nor does he engage with the very real possibility that the ‘states’ he refers to act through officials who are unlikely to be unanimous in their evaluations of the merits of international human rights norms and their judicial enforcement.
How should we approach answers to questions? That is methodology’s most welcome inquiry. The caricature of legal scholarship with which I began uncharitably suggested wilful ignorance of method by a great many academic lawyers. However, every able lawyer knows of the basic distinction between descriptive and normative undertakings. The able lawyer knows, too, how common law reasoning invites some normative evaluation in some of its descriptions, when weaving together disparate precedents into a coherent whole and extracting one or more principles to guide the future development of the law. The distinction, though not always easy to draw, is always present. Without bearing it in mind, one risks asking and answering the wrong question.