¹ John Locke identified an additional branch, which he called the federative. It is associated with international affairs and it has not figured substantially in modern constitutional theory.

² Large-scale failures of governance indicate that radical changes might be necessary – or, in present terms, that the existing system should be destabilized by the adoption of large-scale constitutional changes.

³ Carl Schmitt's idea of the commissarial dictatorship has some resonances with this thought, though Schmitt's perception that governance structures had to be able to survive extreme shocks led to his well-known views about the impossibility of defining the contours of states of exception by binding law.

⁴ Among the special considerations are problems of legislative gridlock and ossification, terms familiar in the literature of the modern U.S. administrative state. A well-designed constitution's provision for amendment might be sufficient to deal with these problems – or, put another way, their persistence might be an indication that the amendment mechanism is not well-designed.

⁵ Constitution of Massachusetts, art XXX (“In the government of this commonwealth, the legislative department shall never exercise the executive and judicial powers, or either of them: the executive shall never exercise the legislative and judicial powers, or either of them: the judicial shall never exercise the legislative and executive powers, or either of them: to the end it may be a government of laws and not of men.”)

⁶ James Madison, Federalist No 48 (The Independent Journal 1788).

⁷ James Madison, Federalist No 51 (The Independent Journal 1788).

⁸ Levinson, Daryl J. & Pildes, Richard H., ‘Separation of Parties, Not Powers’ (2006) 119 Harvard Law Review 2311.Google Scholar

⁹ For a powerful discussion of the phenomenon described in the text, see Lars Vinx (ed & tr), The Guardian of the Constitution: Hans Kelsen and Carl Schmitt on the Limits of Constitutional Law (Cambridge University Press 2015) 125–150 (reprinting an essay by Carl Schmitt discussing “the development of parliament into the arena of a pluralistic system”).

¹⁰ Tucker, Paul, Unelected Power: The Quest for Legitimacy in Central Banking and the Regulatory State (Princeton University Press 2018)Google Scholar, frames his inquiry into the proper domain of bodies (mostly for Tucker administrative agencies) independent of the traditional three branches almost entirely in terms of the limits on the ability of actors in those branches to give credible commitments to maintaining a stable course with respect to specific policies, of which Kelsen would have said preserving the constitution was one.

¹¹ I should note here that my exposition from this point on is not an effort at the exegesis of Kelsen's specific writings. It is instead my effort to construct an account of the fourth branch that is roughly consistent with his ideas but might be inconsistent with some of his specific points. Put another way, it is my effort to construct a constitutional theory inspired by but not bound to what Kelsen wrote.

¹² If Kelsen was correct in identifying a “new” function that would ensure regime stability, then that function should exist in any governance order that claims to settle things (for more than a short period). So, for example, we should be able to identify a “guardian of the theocratic constitution” or a “guardian of the one-party constitution” in nations with such constitution. And indeed we do: the Guardian Council in Iran, the Central Committee of the Communist Party in the People's Republic of China. In the remainder of this Article, I consider only fourth branch institutions in regimes roughly qualifying as constitutional democracies, though with a rather expansive definition of that category. For a review of modern uses of the phrase “guardian of the constitution” in written constitutions and court decisions, see Jones, Brian Christopher, Constitutional Idolatry and Democracy: Challenging the Infatuation with Writtenness (Edward Elgar 2020) ch 7CrossRefGoogle Scholar.

¹³ It is a standard view that high-level corruption is a threat to the constitutional order because of its effects on public support for that order. See, eg, Huq, Aziz, ‘Legal or Political Checks on Apex Criminality: An Essay on Constitutional Design’ (2018) 65 UCLA Law Review 1506.Google Scholar

¹⁴ EE Schattschneider, Politics, Pressures, and the Tariff (New York: Prentice-Hall Inc 1935) 288.

¹⁵ Some platforms might aim at long-term entrenchment, supported by political actors who concede that the platforms cannot be enacted in the short run (and advocacy thereof might contribute to short-run defeats). Here we should think of “fringe” or seemingly “minor” parties that aim for long-term influence.

¹⁶ The Workers Party was able to enact the program because Brazil benefited from high prices for commodities such as oil that fueled the Brazilian economy. When commodity prices fell the family-allowance program became a drag on the economy.

¹⁷ The statement was made by economic adviser Herbert Stein in congressional testimony in 1976.

¹⁸ Hirschl, Ran, ‘The Judicialization of Mega-Poiltics and the Rise of Political Courts’ (2008) 11 Annual Review of Political Science 93CrossRefGoogle Scholar. Hirschl uses the term to encompass judicial consideration of questions going to fundamental matters of national identity as well.

¹⁹ Compare here the suggestion that the constitutional courts in Israel and Germany have been careful to exclude from the ballot only parties with quite small followings, and that the German constitutional court's most recent decisions on these questions explicitly attend to the political significance of excluding a party that has significant support.

²⁰ See Fuller, Lon L, ‘The Forms and Limits of Adjudication’ (1978) 92 Harvard Law Review 353CrossRefGoogle Scholar; Chayes, Abram, ‘The Role of the Judge in Public Law Litigation’ (1976) 89 Harvard Law Review 1281CrossRefGoogle Scholar.

²¹ Fuller argued that disputes that he called ”polycentric” were generally unsuitable for ordinary adjudication, and that were courts to attempt to dispose of them they would, as he put it, depart from ”the judicial proprieties.” Writing in the 1950s, Fuller had what we now can see as an essentialist view of courts as institutions, but we can also use his terminology to describe the conclusions reached in the text. IPDs are institutions that constitution-designers use to off-load some polycentric problems away from the courts, and when courts attempt to deal with such problems they experience stresses because of features that make the problems polycentric.

²² The details of the Indian and Korean bodies’ design are not important for the current inquiry. India's Constitution specifies that there be a single election commissioner with tenure similar to that of supreme court justices. Parliament can and has added additional members to the body. The most influential election commissioner, TN Seshan, was a career civil servant during the period of Congress party dominance. The Korean EMB has nine members, each serving a six year term. The President, the legislature, and the supreme court each choose three members, all but one of whom typically hold other positions as well. One Commissioner works full-time in that capacity, and the chair is by custom a supreme court justice.

²³ Katju, Manjari, ‘Institutional Initiatives Towards Expanding Democracy: The Election Commission of India and Electoral Mobilisation’ 29(2) Contemporary South Asia 147, 154–157Google Scholar (details the difficulties associated with compiling accurate electoral rolls in India).

²⁴ The report is cited in McMillan, Alistair, ‘The Election Commission’ in Jayal, Niraja Gopal & Mehta, Pratap Bhanu (eds), The Oxford Companion to Politics in India (Oxford University Press 2010) 113Google Scholar.

²⁵ For the Code's origins, see ibid 109.

²⁶ Gilmartin, David, ‘One Day's Sultan: T.N. Seshan and Indian Democracy’ (2009) 43 Contributions to Indian Sociology 247, 257, 261, 267CrossRefGoogle Scholar; Jaffrelot, Christophe, ‘T.N. Seshan and the Election Commission’ in Qurashi, SY (ed), The Great March of Democracy: Seven Decades of India's Elections (Penguin Viking 2019) 109Google Scholar.

²⁷ ibid 268, 274–275.

²⁸ ibid 277, quoting Swapan Dasgupta.

²⁹ Sridharan, Eswaran & Vaisnaw, Milan, ‘Election Commission of India’ in Kapur, Devesh, Mehta, Pratap Bhanu & Vaishnav, Milan (eds) Rethinking Public Institutions in India (Oxford University Press 2017) 441, 442Google Scholar; Gilmartin (n 26) 281, connects Seshan's approach to “long-standing assumptions among literate Indian elites that they had a special duty… to tame the unruliness of India's everyday life” through advancing “transcendent principles of law.”

³⁰ A brief period between April 1960 and May 1961 was an exception.

³¹ Lee, Yoonkyung, “Political Parties” in Moon, Chung-In & Moon, M Jae (eds), Routledge Handbook of Korean Politics and Public Administration (Routledge 2020) 83Google Scholar.

³² Prohibition of Distribution of UCC (User-Created Content) in Prior-Electioneering, 21-2(a) KCCR 311, 2007Hun-Ma718, July 30, 2009. Five justices out of nine would have held the prohibition unconstitutional, but the Korean Constitution requires a six-judge majority to invalidate the statute. The Court later held that the statute would be unconstitutional were it interpreted to prohibit political expression and election campaigning on the internet, arguing that the possibility that defamation and false information might be disseminated could not justify a complete ban of online campaigning even during a limited period. Prohibition of Internet Use for Political Expression and Election Campaign 23-2(B) KCCR 739, 2007Hun-Ma1001, 2010Hun-Ba88, 2010Hun-Ma173⋅191(consolidated), December 29, 2011.

³³ For full details, see Lee, Youngjae, ‘Law, Politics, and Impeachment: The Impeachment of Roh Moo-Hyun from a Comparative Constitutional Perspective’ (2005) 53 American Journal of Comparative Law 403CrossRefGoogle Scholar.

³⁴ Impeachment of the President (Roh Moo-Hyun) Case, 16-1 KCCR 609, 2004Hun-Na1, May 14, 2004.

³⁵ Petition to Invalidate the Notice of Compliance Request for President's Duty of Impartiality toward Election, 20-2(a) KCCR 139, 2007Hun-Ma700, January 17, 2008.

³⁶ The official translation of the opinion into English is more stilted than others I have quoted, and I have freely adapted the language to make it more readable.

³⁷ The majority ended by finding that Roh's statements violated the duty of impartiality: They were made within six months of the scheduled presidential election, at a time when the potential candidates and “their general policies were already known to the public,” they were made at relatively large public gatherings (rather than in small private settings), and they occurred during the ordinary hours of work.

³⁸ Here too we see the constitutive role of the Election Commission and the Constitutional Court, as described by Stacey & Miyandazi of this volume.