Explaining and justifying political and legal accountability
Elizabeth Fisher has argued that accountability is ‘closely related to issues of democracy'4 and ‘has a profoundly normative aspect'.5 It is ‘not an autonomous principle' or a neutral concept,6 and an ‘argument for more accountability is really an argument about wanting to align governance regimes to a particular normative vision.
The process of holding a decision-maker to account is a process of debating what the standards should be'.7 To quite some extent, this position must be correct: the term ‘accountability' assumes a situation or process which is normatively desirable, a point underscored by Carol Harlow's and Richard Rawlings' suggestion that accountability is one of a trilogy of ‘administrative law values' (the others being transparency and participation) around which a ‘growing consensus' of support has ‘crystallized' as something which it is thought appropriate for the law to promote.8 Debate concerning the exact meaning of accountability and the content of the best or least bad accountability regime tends to be normative in character, even if only because the theorists concerned are assuming at a very general level that some value attaches to the idea of accountability: for otherwise, there would be little or no point in debating the best or least bad interpretation. This point can be made even before consideration is given to the well known assertion of Harlow and Rawlings that ‘Behind every theory of administrative law there lies a theory of‘ “Never Allow a Crisis Go To Waste”: The Wright Committee Reforms to Strength the House of Commons’ (2011) 64 Parliamentary Affairs 612.
4 E Fisher, ‘The European Union in the Age of Accountability’ (2004) 24 Oxford Journal of Legal Studies 495,508.
5 Fisher, 2004,509.
6 Fisher, 2004, 508 (compare with 496).
7 Fisher, 2004,513.
8 C Harlow and R Rawlings, Law and Adm-inisiriui-on (Cambridge: Cambridge University Press, 3rd edn, 2009),46. the state'[868]—‘theory' meaning normative theory—although the two clearly reinforce one another.
It is thus unsurprising that leading normative accounts of the proper roles of courts, legislatures and executive bodies—more specifically, the accounts known as legal and political constitutionalism[869] —entail their own assumptions about or views concerning accountability. As is well known, Harlow and Rawlings talk of ‘red light' and ‘green light' theorists: loosely-defined camps, some of the distinctions between which broadly echo the divisions between legal and political constitutionalism.[870] Adam Tomkins characterizes ‘red light' theorists as believing:
(1) that law is autonomous to and superior over politics; (2) that the administrative state is something which needs to be kept in check by the law; (3) that the preferred way of doing this is through rule-based adjudication in courts.[871]
By contrast, ‘green light' theorists believe:
that law is nothing more than a sophisticated (or elitist) discourse of politics... [and] that the objective of administrative law and regulation is not merely to stop bad administrative practices, but to encourage and facilitate good administrative practices... and... that the best institutions to achieve these aims will not necessarily be courts.[872]
Harlow and Rawlings, while acknowledging that ‘accountability' is nowadays championed as a value by both red and green light theorists,[873] are nonetheless clear that on an historical and conceptual basis, ‘Red light theory prioritizes courts; green light theory prefers democratic or political forms of accountability'.[874] An analogous point is made by Tomkins, using the broader language of political and legal constitutionalism: ‘A political constitution is one in which those who exercise political power (let us say the government) are held to constitutional account through political means, and through political institutions (for example, Parliament).
Thus, government ministers and senior civil servants might be subjected to regular scrutiny in Parliament... A legal constitution, on the other hand, is one which imagines that the principal means, and the principal institution, through which the government is held to account is the law and the court-room'.[875]As Harlow makes clear, ‘important choices' therefore have to be made about ‘the proper balance between different forms of accountability and the extent to which, in any given situation, legal, political, or regulatory accountability is preferable'.[876] Such choices are to a significant extent normative, focusing on the values promoted by the form of accountability in issue—and, in turn, on one's general view concerning the proper role of the institutions in play (including, but not only, the legislature). For example, Tomkins suggests that the weight one places on values such as openness, transparency, participation, representativeness, deliberation, and the safeguarding of minorities—all normative concerns, and sometimes in competition—may affect one's general support for political or legal constitutionalism and associated accountability mechanisms.[877] A not-dissimilar point can be made about the accountability of legislators to the electorate (an issue to which considerable attention has recently been paid in the UK): the extent to which one believes that a member of the legislature should explain decisions to the electorate, and/or take account of the electors' will but exercise independent judgment, and/or be mandated, will have very direct implications for the accountability relationships which should exist— for example, whether there should be some form of direct popular accountability in addition to the political checks and balances which exist within the legislative process—as well as resting on a normative view of the proper role of legislators and the legislature in a democracy.
Nonetheless, to fully understand the distinction between political and legal accountability, it remains necessary to pay attention to c ertain descriptive or definitional issues (hence the slightly qualified endorsement given to Fisher's argument above[878] ).
For it is plainly the case that, whatever the normative aspects of arguments concerning accountability, such arguments can also have linguistic or purely analytical components. This point is captured in Dawn Oliver's understanding of accountability arrangements as things which, while promoting the public interest (conceived of in normative terms as an ideal which legitimates government), can also be divided analytically into four categories. First, political accountability is owed to politicians, encompassing ministerial accountability to Parliament (in the UK) or local authority accountability to central government and Parliament. Within this category, accountable actors and bodies are exposed to possible political censure and electoral risk, with political costs sometimes being exacted at a personal level (for example, via the forced resignation of a minister) if performance falls below the expected standard.[879] Secondly, public accountability is owed to the general public or interested sections of it. Most obviously, elected national and local politicians are politically obliged to explain and justify their actions to electors, with political penalties to be paid at the ballot box if an adequate account is not offered..[880] Thirdly, legal accountability is owed to the courts, most obviously in the UK via judicial review of executive action, something which hopefully encourages transparency in decision-making.[881] Fourthly, administrative accountability entails the duty to account to non-political regulatory bodies such as the Parliamentary Commissioner for Administration.[882] Analytical components can also be found in Carol Harlow's treatment of accountability in the European Union. Thus, Harlow associates legal accountability as a category with proportionality review, in the sense that administrative measures need to be appropriate as well as necessary to achieve their desired objectives, and with the requirements of due process.[883] She also argues that: ‘[T]here is always an adversarial element in accountability, the purpose of which is to encourage admissions of responsibility and to hold people to account for their actions'.[884] In addition, she suggests that effective accountability may, as a mechanism, entail a measure of externality and autonomy given that it becomes hard for those who have taken part in a decision or in policy-making to censure or blame the decision-maker.[885]The arguments just discussed go to the structure of accountability viewed, analytically, as a distinct concept: this is why they can be described as descriptive or definitional. A further approach operating beyond the purely normative level is set out by Tomkins.
He has suggested that appropriate roles for political and legal constitutionalism can be evaluated not only by reference to the values promoted by each—a clearly normative criterion—but also by assessing their efficacy as tools for promoting the values concerned.[886] This additional criterion is clearly normatively-tinged, it being hard to detach one's assessment of the effectiveness of a constitutional tool from any analysis at all of the desirability of its underlying values. Nonetheless, Tomkins' explicit separation of the two evaluative methods would, in logic, be difficult to explain without accepting to some extent that it is possible to assess a constitutional tool by reference to its success ‘by its own lights': that is, by reference to how effective it is (including by comparison with other possible tools) in promoting its own demonstrable goals. Since many such goals are normative, it would plainly be artificial to claim that an efficacy-focused assessment of this type could plausibly be seen as purely empirical (at least where normative goals are in play). Nonetheless, it is clearly different from an evaluation which asks only whether the tool concerned promotes normatively desirable goals, and judges the tool on that basis. For an assessment of efficacy ‘by its own lights' does not entail a prior evaluation of the (un)desirability of the goals in play. The goals are instead presupposed and the focus of assessment is on how well the tool serves in assisting with their delivery. It should thus be possible to assess constitutional rules from the standpoint of their efficacy in promoting a given type of accountability, and for this exercise—while quite possibly normatively-tinged— not to be an out-and-out argument ‘for a particular normative vision' in the sense envisaged by Fisher.For the sake of completeness, two further points need to be stressed. First, while it seems obvious that judges' interpretations of the workings of individual constitutional mechanisms, including those with implications for accountability, will be affected by the properties of the broader constitutional setting (or judicial understandings of such properties) within which the mechanisms concerned operate,[887] constitutional theorists' analyses—of the general idea of accountability and the distinction between political and legal accountability as well as of individual mechanisms— may well be similarly affected, albeit at varying levels.
For example, a theorist whose primary focus is the constitutional law of the UK may well be content to equate legal accountability (that is, to courts) with statutory interpretation and (so far as it is distinct) judicial review of executive action[888]; a United States constitutional theorist, by contrast, is more likely to see the main tool of legal accountability as judicial review of legislation, with statutory interpretation and review of executive action being placed slightly lower down the list of significant accountability devices. In consequence, it may be a matter of degree how far these two theorists can share a workable understanding of legal accountability. Despite this, however, it is also clear that some general ideas can be and are shared. An obvious example is the notion that certain privileges and immunities attach to those engaged in the business of the legislature, a protection that seems to apply—albeit with different boundaries from case to case—in jurisdictions where legislation is open to judicial review by courts as well as in the small number where it is not.[889] Furthermore, the two efficacy criteria for evaluating accountability arguments (normative desirability and ‘by its own lights') would seem, in logic, to be applicable regardless of localized constitutional characteristics.Secondly, normative theories about desirable accountability arrangements sometimes rest on ‘best case' assumptions about the performance or characteristics of the institutions under scrutiny. Sandra Fredman, for example, suggests by reference to UK and South African legislative debates about voting rights for prisoners that key political constitutionalist Jeremy Waldron ‘idealizes legislative decision-making', from which ‘much relevant evidence and opinion is frequently omitted' in practice.[890] By analogy, leading legal constitutionalist Ronald Dworkin conceded that the US Supreme Court had come in the late 2000s to be dominated, in implicit contrast to his ideal of integrity-driven decision making, by a ‘right-wing phalanx' which seemed ‘guided by no judicial or political principle at all, but only by partisan, cultural, and perhaps religious allegiance'.[891] These examples may help reinforce the suggestion that in practice, constitutional accountability is likely to be most effectively promoted not through ‘best case' idealizations of the role to be played by a single institution, but instead through a combination of political and legal mechanisms.
A more principled version of this suggestion has been articulated, originally by reference to the Canadian Charter of Rights and Freedoms, by Janet Hiebert. Hiebert puts forward a ‘relational approach' whereby Parliament and the courts can both offer valid insights into the best interpretation of legislative objectives. From this perspective:
The normative goal is not that Parliament aspire to ensure that legislation addresses all judicial concerns or, alternatively, that the judiciary defer to Parliament's judgment. Rather, it is that each body satisfy itself that its judgment reflects [constitutional] values, particularly when faced with the other's contrary judgment. Thus, a relational approach is informed by the assumption that parliamentary and judicial judgments be guided by a degree of modesty about the superiority of their conclusions and by respect for the other's contrary interpretation. Respect is measured by a sincere effort to understand the reasons and motivations that led to a contrary assessment, even if each ultimately disagrees with the other.[892]
On this view, the legislature and judiciary should each satisfy themselves that their assessment is justified. When differences arise, the two institutions should ‘reflect on the opinion of those situated differently, relative to the.... conflict' and ‘[i]n Parliament's case, a court's contrary judgment should be considered and its different concerns addressed, even if ultimately disagreed with'.[893] This forms part of a ‘call for shared responsibility', whereby the legislature is encouraged to ‘exercise principled judgment about the merits of legislative decisions’ involving careful and deliberate evaluation of the goals of legislative proposals relative to individual rights, even if its judgment ultimately differs from that of a court.[894]
In turn, Tom Hickman categorizes approaches of this type as involving a ‘constitutional dialogue between courts, legislatures and executives’.[895] In the UK, such approaches are often associated with the Human Rights Act 1998, although there have been important disagreements between ‘dialogue’ theorists about their exact operation: for example, about whether the idea of ‘dialogue’ entails the adoption of divergent judgments by courts and the legislature, whether courts are essentially engaged as political actors in the process of persuading Parliament to reach a final judgment, or whether the nature of the interaction between the two institutions depends upon the contestability of the issue in play.[896] Questions have also been raised about the point at which the process of ‘dialogue’ can be seen as beginning and ending, and about the quality of the deliberation required on each side.[897] It is perhaps unsurprising that Alison Young, a supporter of ‘dialogue’ theory, acknowledges its ‘malleability’.[898]
Debate about the best approach to ‘dialogue’ theory lies beyond the scope of this essay. Instead, the aim of the next two sections is to tie discussion of the respective roles of the legislature and courts to an analysis of accountability issues, ultimately through the lens of ‘dialogue’ arguments. It will be suggested that an effective analysis of the accountability arrangements prevailing in relation to the Westminster Parliament—that is, arrangements surrounding ministerial responsibility and/or accountability, Parliamentary questions and Parliamentary debate, as well as judicial scrutiny of executive power—must be understood in terms of legal and political components viewed in combination. More particularly, the combination of judicial and political scrutiny in the anti-terrorism context provides a key—and centrally focused—illustration of the correlation between the promotion of accountability and the existence of judicial and political methods working in tandem.
B.