James Harrington insists on the possibility of an empire of laws. The claim that institutional arrangements can force self-interested behavior to serve the common interest underlies the system of separated powers and checks and balances central to the American constitutional order. But is he right? Can this theory of “dividing-and-choosing” give us confidence?

“Two [girls] have a cake yet undivided which was given between them that each of them therefore may have that which is due. ‘Divide,’ says one to the other, ‘and I will choose, or let me divide, and you shall choose.’ If this be but once agreed upon, it is enough, for the divider dividing unequally loses in regard that the other takes the better half. Wherefore, she divides equally and so both have right.”

—James Harrington, The Commonwealth of Oceana, 1656

Lex Luthor delivers a wonderful line in Superman (1978): “Some people can read War and Peace and come away thinking it’s a simple adventure story. Others can read the ingredients on a chewing gum wrapper and unlock the secrets of the universe.” English political theorist James Harrington would have made Luthor proud: he found the whole secret of the rule of law in a commonwealth in the method two “silly girls” might use to split a cake.

Harrington, a seventeenth-century republican, is best known for his theory of the economic basis of political power and forms, outlined in The Commonwealth of Oceana and other works. As John Toland wrote in an account of Harrington’s life introducing Oceana, “empire follows the balance of property.” Roughly, Harrington’s idea is that regime stability depends on correspondence with the distribution of property in a society. A monarchy is the form of government that corresponds to a distribution in which a single person holds the bulk of property or all of it; an aristocracy to a society where property is more broadly distributed, but still concentrated in the hands of a nobility; and a “commonwealth” or government that includes participation by a larger portion of the people—which he favored as the best fit for England in his time—to a scenario where property is much more widely distributed. Regimes that impose a political form not corresponding to the distribution of property rely on force and thus are tyrannical and inherently unstable.

In his study on Harrington’s political thought An Immortal Commonwealth (1960), Charles Blitzer describes how Harrington sought to explain not only the determinants of political stability, but the means of linking the power to rule, or “empire,” with “authority,” or legitimate rule. Every regime that endures has power, but not every regime exercises genuine authority. In a preliminary section of The Commonwealth of Oceana outlining his fundamental principles of government, Harrington identifies his theory with what he calls “antient prudence,” in contrast to “modern prudence”:

Government (to define it de jure, or according to antient prudence) is an art wherby a civil society of men is instituted and preserv’d upon the foundation of common right or interest; or (to follow Aristotle and Livy) it is the empire of laws, and not of men.

And government (to define it de facto, or according to modern prudence) is an art wherby som man, or som few men, subject a city or a nation, and rule it according to his or their privat interest: which, because the laws in such cases are made according to the interest of a man, or of som few familys, may be said to be the empire of men, and not of laws.

In Blitzer’s words, one of Harrington’s major questions is: “Granting men’s selfishness, is it possible to create a system of government which will lead men to act in accord with the common interest?”[1] Harrington’s theory heavily relies on institutions, or rules for collective decision-making, to link “empire” with “authority.” Harrington employs the cake-cutting example to demonstrate the possibility of the rule of law. It’s a concrete example of how an institution can channel private self-interest to serve the common interest.

The institutional arrangement Harrington identifies and the model of a commonwealth he sketches show the theoretical possibility of government by divided power and escape from the arbitrary rule of a single person or entity. “Rule of law” is possible. But, as Harrington and other proponents of what I’ve described as “republican institutionalism,” recognized, institutions function within a context of power relationships and social conditions related to civic virtue.[2] Neither institutions nor virtue alone guarantee liberty and justice for all.


Putting ourselves in the cake scenario helps us think about how institutions work. The problem is we both want to eat the most cake we can. Neither of us can be impartial, so neither can be trusted to make the decision on our own. But, if we adopt the right rule or institution, the interaction between our self-interested desires and actions can produce a just outcome. The fair outcome of an equally divided cake results from a simple process of what is called “backward induction” in game theory—as the divider, I cut the cake fairly in half because I know you, the chooser, will take the largest available piece for yourself.[3] The whole point is that there is no third party.

Political theorist Frank Lovett treats securing the rule of law as the central element of Harrington’s constitutional and political theory.[4] Harrington claims the principle of dividing and choosing makes an “empire of laws, and not of men” possible, even though men compose a commonwealth. According to Harrington, the divide-and-choose method is the only way to “give the upper hand in all cases to the common right or interest, notwithstanding the nearness of that which sticks to every man in private.” Thankfully, it’s not too difficult; two girls dividing a cake can manage it! “That which great philosophers are disputing upon in vain is brought unto light by two silly girls, even the whole mystery of a commonwealth, which lies only in dividing and choosing.”

The idea that well-designed institutions can incentivize self-interested actors to behave in publically beneficial ways is neither obvious nor universally accepted. Thomas Hobbes, who published Leviathan in 1651, four years before Oceana, flatly denied the possibility of an empire of laws as opposed to men. Ultimately, Hobbes said, rules are just words; the will of the powerful by force is the only kind of empire. There must be at least one unlimited wielder of power in a polity; if an agent’s power is limited, then who limits the limiter? The final limiter has personal, unconstrained, and arbitrary power. Dr. Lovett describes a similar argument from Dante’s De Monarchia to the effect that, logically, some person or group of persons has final say on the content or applicability of a rule; appeals to third parties to adjudicate disputes cannot go on ad infinitum. The rule of law ultimately collapses into the rule of men.

Harrington insists on the possibility of an empire of laws. The claim that institutional arrangements can force self-interested behavior to serve the common interest underlies the system of separated powers and checks and balances central to the American constitutional order, at least as envisioned by theorists like Publius and John Adams. Article XXX of the Massachusetts Constitution, of which Adams wrote the original draft, provides for separating the legislative, executive, and judicial powers “to the end it may be a government of laws and not of men.”[5]

The central feature of Harrington’s model is a bicameral legislature: “In a commonwealth consisting of a single council, there is no other to choose than that which divided.” The Senate is the divider, and the popular assembly the chooser:

Dividing and choosing in the language of a commonwealth is debating and resolving, and whatsoever upon debate of the senate is proposed unto the people and resolved by them is enacted Auctoritate Patrum Et Jussu Populi, by the authority of the fathers and the power of the people, which concurring make a law.

The senate and only the senate can propose legislation. Likewise, only the representative assembly has the power to adopt legislation, and that is its only power. The main point is the powers of proposal and adoption are split among different councils, or bodies of state. Separating those powers allows the magic, or better the technology, of backward induction to take effect, incentivizing lawmakers to promote the common interest as they pursue private interests.

Harrington’s institutional theory is elegant, but there are several problems. In an extended discussion of Harrington’s cake example, legal scholar Richard Epstein points out that the chooser in the example has the upper hand because of the possibility of cake-cutting error.[6] He also raises several dissimilarities between the cake example and actual politics, especially situations involving more than one participant.

There is not a one-to-one correspondence between dividing-choosing and proposing-adopting. Making law and policy isn’t always like cutting a cake. The power to propose policy, the “agenda-setting” power, is a major one. Even if I’m allowed to veto a proposal, that’s not the same as getting something closer to my preferred policy onto the table for consideration.

As Dr. Epstein notes, the senate-assembly division raises questions about internal decision rules for each body and about which body may have a favorable bargaining position in different circumstances. One of them relates to the agenda-setting power—the cost of holding out for a better proposal may weaken the assembly relative to the senate. Unlike in the cake example, debating and adopting take place against the backdrop of a given status quo. All the proposers have to do is propose something a tad better than the status quo, leaving the adopters with the option of the status quo or a paltry proposal. In other words, very often, the true option is not “I’ll divide and you choose,” but “I’ll propose, and you take it or leave it.”

Finally, one wonders how the institution gets set up in the first place. Interestingly, according to Mr. Blitzer, Harrington’s notion seems to be that a single ruler, say a Lycurgus or an Oliver Cromwell type, has to wind up the clock and set in place, getting the institutional structure right at a single point in time. Luckily, it’s probably an oversimplified view of how institutions come into being.

Harrington oversells the power of the cake example, but it is nevertheless suggestive of how institutions can structure interactions in situations of conflicting private interests such that they serve common interests. Dr. Epstein concedes that Harrington’s scheme shows the possibility of a workable, if not perfectly reliable, division of power that avoids the need for arbitrary, personal rule. Dr. Lovett agrees, in the formal sense:

If we consider the institutional framework Harrington has outlined, and imagine that everyone does indeed adhere to the specified rules, we would have a very particular sort of political system: namely, one in which there is no single omni-competent political agent (whether a natural person, or a corporate actor) . . . This would indeed constitute an empire of laws and not of men in the required sense.

But, as Dr. Lovett notices, there is a deeper problem. What if powerful actors aren’t satisfied by the outcomes of the institution? What if they find themselves denied the political or policy outcomes they want? Well, such actors may find workarounds, or simply ignore the formal system of checks and balances entirely. Harrington’s critic Matthew Wren raised this point, forcing him to connect his institutional model with his theory of the economic basis of political stability and forms in later writings.


Harrington’s model of the legislature operates within a broader context of a particular distribution of power, specifically economic power, which makes it workable. The Senate represents an aristocratic class, and the assembly the mass of the people. Harrington is describing institutions for an “equal commonwealth,” where the balance of property is held by the people and not concentrated in the aristocratic class. He summarizes his model as follows:

An equal commonwealth . . . is a government establish’d upon an equal Agrarian, arising into the superstructures or three orders, the senat debating and proposing, the people resolving, and the magistracy executing by an equal rotation thro the suffrage of the people given by the ballot.

The divide-and-choose system does not operate in a vacuum, but within the context of a distribution of property. Critically, an agrarian law preserves that distribution by preventing accumulation of property beyond a certain degree.

Modern political scientists have also argued that there must be certain power relationships in place for constitutions, especially those prescribing that major offices are determined by elections with broad or universal suffrage, to really be effective. Political scientist Adam Przeworski has argued: “Constitutions are neither sufficient nor necessary for democracy to survive.”[7] The idea is that an initial agreement on constitutional rules does not guarantee sustained adherence to the rules. In Przeworski’s words, “To survive, democracy must be an equilibrium at least for those political forces which can overthrow it: given that others respect democracy, each must prefer it over the feasible alternatives.” As an aside, an implied irony is that constitutional adherence may be most likely where explicit constitutional rules are least necessary because there is an existing balance of power.


Dr. Epstein points out that Harrington’s scheme acknowledged the importance of virtue, without relying on it. Dr. Epstein says

Individuals must be raised in a civil society where they are taught to understand that their own individual wills and desires must be subordinated to those of the larger society. If many individuals learn to behave in this way, then they can develop into a sufficient critical mass, which will form an effective counterweight to the arbitrary power of wayward government officials.

Harrington supposes that broad participation in elections and office holding may contribute to the development of civic virtue.

Publius’s theory, also incorporating elective systems as the primary, though not exclusive, bulwark against tyranny, echoes this notion. In Federalist #57, Publius writes

The aim of every political constitution is, or ought to be, first to obtain for rulers men who possess most wisdom to discern, and most virtue to pursue, the common good of the society; and in the next place, to take the most effectual precautions for keeping them virtuous whilst they continue to hold their public trust.

Along with its commitment to the rule of law, the Massachusetts Constitution of 1780 acknowledged the importance of civic virtue, “As the happiness of a people and the good order and preservation of civil government essentially depend upon piety, religion, and morality.” Likewise the argument that “Wisdom and knowledge, as well as virtue, diffused generally among the body of the people, being necessary for the preservation of their rights and liberties” justified public provision for education.

Some thinkers oppose public virtue to institutions as the more necessary and reliable bulwark of liberty and justice. But, as Dr. Epstein points out, both Harrington and many of America’s founding constitutional theorists thought both were important. Epstein summarizes Madison’s balanced view: “Civic virtue is nice when you can get it, and in the long run a nation that places a succession of fools and knaves in high places will not be able to survive unscathed regardless of its constitutional structure.” But relying on enlightened statesmen and paying no attention to institutional constraints is equally foolish.

Cakes and Constitutions

So is Harrington right? Should the cake example give us confidence in the magic of backward induction, of dividing and choosing? Yes and no. Logically, institutions dividing power can generate socially beneficial outcomes, and there is some empirical evidence they have.[8] However, as Harrington, Publius, Adams, and later Tocqueville understood, institutions operate in a broader context of power relationships and social norms.

Dr. Epstein concludes that institutional features designed to approximate Harrington’s cake example are necessary but not sufficient to achieve a “just social state”:

Once the question becomes, is this the best we can do, then the rule of law is only a necessary and not a sufficient condition for the just social state. So long as that one conclusion is kept firmly in mind, we can recognize the achievements of any political system in which the rule of law protects us from the ravages of arbitrary power, without forgetting the enormous amount of substantive work that has to be done even after the rule of law is firmly entrenched and universally accepted.

Even if Harrington’s Cake does not fully unlock the secrets of commonwealths, it suggests institutions can contribute to the establishment of rule of law, which also depends on economic and social conditions. The goal of linking power to justice involves institutions, virtue, and a heavy dose of fortune, or perhaps Providence.

The Imaginative Conservative applies the principle of appreciation to the discussion of culture and politics—we approach dialogue with magnanimity rather than with mere civility. Will you help us remain a refreshing oasis in the increasingly contentious arena of modern discourse? Please consider donating now.


[1] Charles Blitzer, An Immortal Commonwealth: The Political Thought of James Harrington (New Haven: Yale University Press, 1960), 146.

[2] Ben Peterson, “Republican Institutionalism for a ‘Government of Laws’: The Polybian Political Science of John Adams,” American Political Thought 8, no. 3 (Summer 2019).

[3] William Spaniel, “Game Theory 101: Backward Induction,” (YouTube, Game Theory 101).

[4] Frank Lovett, “Harrington’s Empire of Law,” Political Studies 60, no. 1 (2012): 59-75.

[5] The Massachusetts Constitution can be accessed here.

[6] Richard A. Epstein, “Beyond the Rule of Law: Civic Virtue and Constitutional Structure,” George Washington Law Review 56, no. 1 (November 1987): 149-171.

[7] Adam Przeworski, “Self-enforcing Democracy,” Department of Politics at New York University, June 28, 2005.

[8] Stephen Weymouth, “Political Institutions and Property Rights: Veto Players and Foreign Exchange Commitments in 127 Countries,” Comparative Political Studies 44, no. 2 (December 8, 2010): 211-240; Susan Alberts, “How Constitutions Constrain,” Comparative Political Studies 41, no. 2 (January 2009): 127-143.

The featured image is “The Reluctant Bride” and is in the public domain, courtesy of Wikimedia Commons.

All comments are moderated and must be civil, concise, and constructive to the conversation. Comments that are critical of an essay may be approved, but comments containing ad hominem criticism of the author will not be published. Also, comments containing web links or block quotations are unlikely to be approved. Keep in mind that essays represent the opinions of the authors and do not necessarily reflect the views of The Imaginative Conservative or its editor or publisher.

Leave a Comment
Print Friendly, PDF & Email