Margit Cohn: Response to Comments on a Theory of the Executive Branch: Tension and Legality

I am honoured to be discussed in this venue by top public and administrative law scholars, who invested their time in reading A Theory of the Executive Branch and offered insightful comments. In this response I set aside the praises (although they were, of course, a pleasure to read). The book was written in the hope that it would contribute to the still-limited comparative and theoretical study of the executive as a leading institution in the modern democratic state; the comments offered by the esteemed participants in this symposium show that the book may have achieved this initial aim. The challenges and critical comments offered in this blog have led me to further consider some of the book’s main arguments; although I stand firm behind the exercise that is my book, they constitute the type of discussion that I hoped would follow the publication of the book; I therefore thank Professors Cane, Jhaveri, Loughlin, MacDonnell and Rundle.

          Rather than focus on each of the comments separately, my response is ordered by topic; I apologize if some of the comments were lost in this ordering exercise. I open with what I conceive to be one of the strongest critiques. Several commentators, namely Cane, Loughlin and Jhaveri, find my focus on the central government and its head (prime minister or president) to be misdirected, if not incorrect. The basis for this critical comment is the observation of the structural diversification of executive bodies in the modern state. Discussed by many as one of the features of the so-called “administrative state”, Cane raises the question of whether “the massive growth of the non-political, non-elected bureaucracy” still enables my focus on the central government;[1] in a similar vein, Loughlin notes the centrality of bureaucracy and the emergence and rise of “public bodies that carry out executive functions,” operating “at a considerable distance from the main institution of democratic legitimation.”[2] Jhaveri further considers the fact that some of these bodies’ decisions “are made outside the value system typically associated with the state (for example, commercially motivated profit-based activities versus welfare-oriented decision-making).”[3] The two observational elements in play here—dispersal and institutional independence—are addressed below. First, however, I move to Loughlin’s discussion of the new foundational basis for administrative law that is to replace the understanding of law as command.

          In Loughlin’s view, expressed also in his important book on the nature of public law, eighteenth century constitutional ideas are outdated because they are based on the recognition that “government acquires it authority from popular sanction, and… to ensure it exercises its powers to the public interest, government must be divided into three branches” that “keep within their constitutionally conferred powers.”[4] This rejected vision of state powers is tied with the work of earlier continental theorists; Léon Duguit is granted a central role in forming a new theory of state power under which government bodies act to advance individual needs and social solidarity, instead of relying on principles that support a top-down version of governance.

          I do not aim, within the confines of this blog, to address the link made there between such nascent and current visions of the administrative state and the rejection of the principle of the separation of powers and other such constitutional principles. In my view, the brave new world that is the administrative state is, certainly for Duguit, a rejection of sovereign power that is “unified and indivisible” which commands action; as such, it should not necessarily be read as a direct challenge to the principles of liberal democracy.[5] In other words, the repudiation of visions of absolute power does not necessarily lead to a rejection of the recognition that the executive branch, as one of the three separate and interacting branches of government, is ideally limited to the execution of law. Nor does the focus on the dispersion of power in the administrative state lead to the rejection of such ideals. Indeed, different conclusions on the nature of the modern theory of the state are possible, but I find that this renewed vision of the state can remain linked with theories that award the executive, as such, powers to act within the tri-branch formulation. 

          I was of course aware of the burgeoning literature on the administrative state. Part of Chapter 1 of my book addresses this issue. In this context, I challenged the arguments that identify the burgeoning of administrative agencies as a weakening of central government power due to the so-called independence of agencies. In Section E.2 of Chapter 1, I argue that degrees of institutional independence of bureaucratic bodies vary; appointment and removal processes introduce points of intervention by the central government or other political players; de facto oversight of agencies in the US is ensured by sets of executive orders, and the distance of British bodies from government departments is lesser; furthermore, appropriation acts provide additional points of entry, inter alia by the insertion of riders in annual appropriations acts.[6] Finally, it is impossible to ignore the central role played by heads of governments, who operate as the politically dominant decision-maker in democracies, even in those turned “administrative”.

          As for Jhaveri’s correct observation regarding the transfer of functions from the public to the private sphere, which involves the infusion of so-called market and profit values into decision-making, two distinctions are useful here. The distinction between outsourcing services that were previously meted out by government and the privatization of regulation is useful here. The question of the volume and content of state-organized direct supply of services does not necessarily imply the transformation of the core functions of government, here presented as regulatory and supervision of markets. Bodies that operate as supervisors are, in fact, generally closer to the scrutiny of central government, despite their so-called independence, and private bodies functioning as regulators, when they exist, are, or should be, further supervised by the state.[7]

          The next comment is also addressed in Chapter 1 of the book. Cane’s contribution to this blog contains an overview of the differences between not only the constitutional structures of the UK and the US, but their distinctive legal cultures. Echoing his important book on the matter,[8] Cane finds the readiness to compare these two systems to be based on unconvincing arguments. Under his account, “at the apex of the UK system we find not a legal document but a political institution with inherent power—the Queen in Parliament”; under this structure, the principle of parliamentary sovereignty is essentially political. Conversely, in the U.S., under the legal document that is the Constitution, all branches of government are “understood equally to exercise not original power but rather power delegated to them by the people. The Constitution is supreme law, and the whole government is understood to be directly or indirectly derived from and subject to that supreme law.”[9] For Cane, then, all aspects of governance must be assessed through the lens of this fundamental difference, which carries a direct effect on all government branches, including the judiciary.

          I set aside the claim that the study of merely two legal systems falls short from representing a “general theory”. This is of course true, and the best that I could offer here was a tentative claim, backed by political facts evident in two systems. In my view, the dominant political status of the head of the executive is clearly part of the life of both states. While Cane’s account is valuable, and points at substantive differences in the framing and operation of the respective constitutional frameworks, I argue, in Section E.1 of Chapter 1, that the retention of such distinctions carries lesser force when the analysis embraces political facts beyond legal structure. A distinction between presidential and parliamentary regimes is of course important and suitable in many analyses of the nature of the state and the legal culture underlying it. However, reliance on it when discussing the interaction between legislatures and executives and the relative power of the executive may overlook similarities, just as much as it can smooth over the complexity of the permutations of relative executive power. Using veto-point theory, I find a spectrum of relative powers, under which the strongest politically would be a prime minister who enjoys a stable and strong majority in parliament, as she enjoys direct access to parliament, unlike a president who likewise operates relatively freely, enjoying the support of both houses of Congress. The weakest executive under this spectrum of structural interaction is likely to be a president who has no such support. The zigzag-type placing that oscillates between parliamentary and presidential regimes shows that a simple distinction between the two systems, based only of the type of legitimacy enjoyed by a directly-elected president in comparison to a prime minister’s indirect support is untenable.[10]  

          Cane’s and Loughlin’s arguments concerned with the validity of my distinction between “acceptable” and “excessive” fuzziness are valid, and require further attention. In Chapter 9, I argue that reliance on fuzziness is problematic in any system that professes a commitment to governance under law, but also serves the interest of decision-makers in the public sphere: hence, fuzziness is an inevitable feature of the law that instructs the action of executive branches. Cane and Loughlin seem to argue that my recognition of some forms of “acceptable” fuzziness reflects a flaw in the model itself. This may be so; I’d need to reconsider this aspect.

          Offering a different version of this type of argument, several commentators beyond this blog symposium have asked whether it would be possible to draw a framework that distinguishes between the extent of fuzziness by identifying fields of action that are better served by fuzziness. I agree that these aspects could be further developed, but all I could do in this context was to point at fuzziness. Beyond this, I am sceptic about the utility of a search for classification by fields of action. The world does not fold out into neat divisions and distinctions; the interests and difficulties that generate fuzzy law, its operation, and the proposal of ways to constrain fuzziness, are all ordered in this fuzzy way. All that can be done is to identify the forms of law and to propose tools for some sort of ordering via accountability processes, which will never be sorted into neat and clear operable doctrines. Hence, I doubt whether a list of fields or contexts can be devised, as politics lean towards fuzziness. The most I could do was contained in Chapter 9.

          As for Rundle’s link of my account of the nature of law to the Hart-Fuller debate, I profess that this was not part of my exercise; mine was more of a blend of law and politics. Clearly, though, open-ended and fuzzy forms of law should be addressed by legal philosophers, a task relegated to others. My angle is more limited, but I can consider myself to be, at least in this context, a ‘dismayed positivist’. More accurately, although fuzzy law is likely to pass the structural requirements that qualify a set of rules as binding (in this aspect, under a positivist account), its substance leads, as Rundle comments, to the questioning of the propriety of relying on a positivist account. There is much more to be said about the link of a ‘fuzzy law’ and theories of the nature of law. This book merely offers a glimpse into the discussion, intentionally limited contrasting core positivism with competing content/substance-based theories of the nature of law, exemplified in Fuller’s account of ‘proper law’. Clearly, more has to be done to develop this nascent analysis to a full-fledged discussion of the nature of fuzzy law.    

          The inclusion of the final chapter, concerned with the role of the judiciary, was motivated by this dismay: while writing, the risk of readers’ tying-in my book with Posner and Vermeule’s somewhat celebratory acceptance of presidential dominance (theirs is a book on the US, even if professed as having relevance to other systems) became pressing enough to justify the inclusion of the chapter. Part of Posner and Vermeule’s argument was that the judiciary should not be an active checking force, due to the power of politics that, in their view, provides sufficient constraints.[11]  Since, like them, I have identified the political reality of grey holes and fuzzy law and argue that they are inevitable, I did not want to be understood as a follower of their normative stance.

          Participants’ comments to this chapter range from arguments that I have given excessive power to courts, implied in Jhaveri’s comment, through Cane’s question regarding the ability of the judiciary itself to create novel nodes of fuzziness, to MacDonnell’s note that judicial review is often not realistically available.[12] All that can be done here is to note, in a bullet-point fashion, that I consider the judiciary as one of the checking mechanisms, and reject arguments about judicial dominance, once again in favour of non-hierarchic accounts of the nature of the decision-making process; that although courts may in theory create new fuzzy forms, they function more as a body that has the power to grant approval to such forms generated by other players; and that belief in the nature of courts as the final saviours of all that is improper in the public sphere, open to all and readily available, would be unrealistic.

          Finally, two comments on the role of other key players as promoters of fuzziness are due. MacDonnell’s reframing of the question concerning the notion of a-legality, and her focus on the executive control over the content of law[13] seems useful, although this angle could be tidily inserted into the book’s framework, as executive-generated forms on fuzziness. A similar response can be made to Cane’s argument that at least in a system that operates under a written constitution, the legislature could be considered merely as an exploiter of the openings provided in the constitution. While this, too, can be accepted as a valid argument, the workings of government have sometimes extended beyond recognized rules of constitutional status, be they written or unwritten. Unilateral executive action, analyzed in detail in Chapters 4 and 5 is considered to be a “constitution-generated” form of fuzziness, but my concept of the constitution is much broader than the text of a central document, and moves to include judge-made and politically-supported doctrines that operate both in the U.K. and the U.S.           In conclusion, I have much enjoyed reading the comments provided in this blog and considering possible responses. The best type of support may sometimes be expressed in criticism: it reflects the readiness of readers to seriously engage with a text, in this case one that is spread over 322 pages. Being taken seriously is, in my mind, the strongest form of recognition. Of course, praise is never rejected, but I thank all blog participants for taking their precious time to assess the value of the book. I hope that the book further inspires similar dialogues and conversations: the executive branch deserves further attention. 

Margit Cohn is Professor and Henry J. and Fannie Harkavy Chair in Comparative Law at the Faculty of Law, Hebrew University of Jerusalem.


[1]     Peter Cane, blog, p. 2.

[2]     See also Jhaveri, blog, p. 3.

[3]     Jhaveri, blog, p. 3.

[4]     Loughlin, blog, p. 1. See also Martin Loughlin, Foundations of Public Law (Oxford: Oxford University Press, 2010), Chapter 15, esp. 448–452.

[5]     Léon Duguit, Law in the Modern State, Frida and Harold Laski trans. (London: Allen & Unwin, 1921), 118.

[6]     Margit Cohn, A Theory of the Executive Branch: Tension and Legality (Oxford: Oxford University Press, 2021), 31–39.

[7]     See Yael Kariv-Teitelbaum, “Privatization of Regulation: Promises and Pitfalls”, in Avihay Dorfman and Alon Harel (eds.), The Cambridge Handbook of Privatizaton (Cambridge: Cambridge University Press, 2021) 263. Of course, many issues can be raised here. Is there indeed a defined core of function that is, or should be, left in the hands of central government? Can a clean distinction between grant of services or goods and supervisory function can be maintained? These, and linked questions, cannot be discussed in detail here.

[8]     Peter Cane, Controlling Administrative Power: An Historical Comparison (Cambridge: Cambridge University Press, 2016).

[9]     Cane, blog, pp. 2–3.

[10]   Cohn 2021, 27–31.

[11]   Eric A. Posner and Adrian Vermeule, The Executive Unbound: After the Madisonian Republic (Oxford: Oxford University Press, 2010).

[12]   Jhaveri, blog, p. 3; Cane, blog, pp. 2–3; MacDonnell, blog, p. 2.

[13]   MacDonnell, blog, pp. 1–2.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s