Arbitrary Power: Caricature and Concept

Abstract

Arbitrary power is often understood as bearing some kind of relation to tyrannical rule, a relation that is thought to explain why arbitrary power is objectionable. But what is tyrannical rule? What precisely is the relationship between arbitrary power and tyranny? Why (if at all) is arbitrary power objectionable? Arbitrary power, this paper argues, is best understood through the figure of the tyrant. The figure of the tyrant is a caricature with stock character traits, usually of a disliked other, who is, or is thought to be, powerful. Arbitrary power, the paper argues, is power that is constituted to express, or enable the expression of, caricatured tyrannical traits. Arbitrary power is bad or objectionable (if and when it is) for mimetic reasons. Understanding arbitrary power in this way allows us to see problems with how it is deployed, understood and communicated and has advantages over standard accounts.

Introduction

The rule of law is often understood in opposition to arbitrary power, an opposition that is often thought to explain why the rule of law is valuable. Arbitrary power is often understood as bearing some kind of relation to tyrannical rule, a relation that is thought to explain why arbitrary power is objectionable.

But what is tyrannical rule? What precisely is the relationship between arbitrary power and tyranny? Why (if at all) is arbitrary power objectionable?

Arbitrary power, this paper argues, is best understood through the figure of the tyrant. The figure of the tyrant is a caricature with stock character traits, usually of a disliked other, who is, or is thought to be, powerful (Section II).

Noone argues that only tyrants possess or exercise arbitrary power. Rather, wielders of arbitrary power are thought to be like tyrants. Since rule by tyrants is objectionable, rule by those with tyrannical powers is also objectionable (continues the thought). But in what respect are wielders of arbitrary power like tyrants?

Arbitrary power, the paper argues, is power that is constituted to express, or enable the expression of, caricatured tyrannical traits. Arbitrary power is bad or objectionable (if and when it is) for mimetic reasons (Section III). Understanding arbitrary power in this way allows us to see problems with how it is (and has been) deployed, understood and communicated (Section IV) and has advantages over standard accounts (Section V). While drawing on genealogical and historical sources, the account offered here is interpretive – it aims to makes sense of how arbitrariness is used in discussions about the rule of law, including in legal doctrine.¹

What is a Tyrant?

Amongst contemporary accounts of what makes power arbitrary, in discussions of the rule of law, Gerald Postema’s stands out for its force and richness:

Arbitrary power is capricious and arrogant. “I will do this, because I can,” says the wielder of arbitrary power. To be subject to the arbitrary power of another is to be subject to the whim, the mere pleasure, of the wielder. The wielders’ will is their only law; they are, we say, laws unto themselves. A will that is law unto itself is a feral will. …Nothing outside themselves constrained them, nothing but their absolute choice.

The wielder’s action says, “My will alone matters; yours is irrelevant. Do this for no reason you need to bother about but only because I say so.”²

One of the virtues of Postema’s personification of arbitrary power is that it compels us to reckon with the figure of the tyrant that looms, often unacknowledged, over contemporary discussions of the rule of law. Reckoning with this figure requires acknowledging the continuity between our contemporary understanding of arbitrary power and historical thought on the tyrant as the wielder of arbitrary power. This reckoning is essential to arrive at a good account of arbitrary power with the potential to illuminate the rule of law. Foregrounding the figure of the tyrant is also essential to explore deep-seated problems with how the concepts of arbitrary power and the rule of law are deployed.

Anglophone neo-Roman thinkers inherited and contributed to a tradition in Western thought which employed a caricature of a disliked other to represent a tyrant, and to describe arbitrary power. For the neo-Roman thinkers, this other was the Ottoman Sultan, often described as a despot.³ Further back, for classical thinkers,⁴ the subject of the tyrannical caricature was the Persian emperor, with caricatures of the emperor Xerxes being particularly well-known.⁵

We can identify stock traits typically associated with tyrants in Western political thought, particularly by neo-Roman Anglophone writers (thought to be particularly influential over our contemporary ideas of the rule of law and arbitrary power).⁶

• 1. The tyrant is greedy, and particularly power-hungry⁷ with a tendency to “overreach”.⁸ Power is important to the tyrant who is often characterized as seeking further power.⁹ The tyrant’s devotion to power is such that he ‘thrive[s] when there are [wide] discretionary powers’.¹⁰

• 2. The tyrant does not recognize any moral constraints to act only according to the law,¹¹ natural law or religion.¹² Many of these traits reflect the tyrant’s unwillingness to recognize any moral constraints on how his will is exercised.¹³

• 3. The tyrant does not recognize moral constraints on his will based on even "his subjects’ most sacred" rights,¹⁴ including property rights.¹⁵ Even if the tyrant in fact respects lives, property and persons, he does not regard these as constituting a moral limit to his exercise of power.

• 4. The tyrant is haughty.¹⁶ The tyrant does not recognize moral constraints on their will based on the views of those who are subject to their powers. Those subject to their power are “as flies” to a divinity, “not worth bothering about” and “at the[ir] mercy”.¹⁷ It follows that the tyrant offers their subjects no space to be heard or to contest exercises of their power;¹⁸ subjects are, in Postema’s words, “utterly without standing to question or utter a vain protest”.¹⁹

• 5. The tyrant is indifferent to the public interest. The tyrant does not recognize moral constraints on his will based on the interest(s) of the public. The idea that the exercise of public power implies moral constraints on those who exercise it²⁰ is not one that a tyrant accepts.²¹ There is a long history of contrasting the tyrant (who serves his own interest) with a ruler who serves the public interest.²²

  The public interest might be understood in different ways. It might encompass the interests that members of the public have as a collective e.g. in flourishing public culture, or in strong civic institutions. It might encompass universal (or at least very common) individual interests of members of the public e.g. in having adequate shelter or healthcare.²³ However the public interest is understood, the tyrant is indifferent to it.

• 6. The tyrant is “capricious”, “a law unto themselves” and subject to his whims;²⁴ in Postema’s powerful description, he possesses a “feral will”.²⁵

• 7. The use of coercive force is a tool in the tyrant’s toolbox²⁶ and casts a shadow over his fearful subjects, even if it is never used.²⁷ The tyrant may also use deception and manipulation.²⁸

• 8. The tyrant can be generous and cruel, rewarding those he favours and torturing those he disfavours.²⁹

• 9. The tyrant’s traits have a deleterious impact on his subjects who become cowed, flatterers and “small-souled”.³⁰

While the traits above could belong to anyone, they were represented as having an elective affinity³¹ – a tendency to attach more readily – to the disliked other (e.g. for classical writers, the Persian emperors, and for neo-Roman writers, the Ottoman rulers). Many influential works on tyranny were also written in a context where there was an incentive to characterize certain others as natural tyrants.

For instance, early descriptions of the Ottoman Sultan’s rule were not offered as descriptive political science to satisfy the intellectual curiosity of European readers. Instead, the Sultan offered a cautionary tale, a foil, or a comparison to better illuminate existent or ideal governance in Christian Europe represented by a rule-bound monarchy or (later) a republic.³² Influential writing on the Sultan-as-tyrant was sometimes motivated by belligerence, given the conflicts between the Ottoman and European regimes.³³ Some writing was motivated by a desire to prevent emigration from Christian Europe to the Ottoman Empire due to the perceived superiority of living conditions in parts of the Ottoman Empire.³⁴ The Sultan was, unsurprisingly then, caricatured as exemplifying the tyrannical traits listed earlier, enabled by a complete lack of legal constraints and by the absence of private property rights or a nobility who could temper his power. The Ottoman people were also caricatured as a compliant and slave-like populace.

The influence of the caricature of the sultan-as-tyrant on foundational concepts of Western political thought – as well as constitutional debate – has long been recognized.³⁵ Alexander Hamilton criticized the invocation of precisely this caricature by those concerned about the magnitude of power concentrated in the US President:

“Here the writers against the Constitution seem to have taken pains to signalize their talent of misrepresentation. [The President] has been shown to us with the diadem sparkling on his brow and the imperial purple flowing in his train. He has been seated on a throne surrounded with minions and mistresses, giving audience to the envoys of foreign potentates, in all the supercilious pomp of majesty. The images of Asiatic despotism and voluptuousness have scarcely been wanting to crown the exaggerated scene. We have been taught to tremble at the terrific visages of murdering janizaries, and to blush at the unveiled mysteries of a future seraglio ”.³⁶

Recent scholarship argues that how influential neo-Roman thinkers like John Locke and John Harrington conceived of arbitrariness influences our concepts of arbitrariness and the rule of law today.³⁷ Their caricature of the tyrant was at the heart of their accounts,³⁸ and it haunts our understanding of arbitrary power as well. Arbitrary power is often understood as bearing some kind of relation to rule by such a tyrant, a relation that is thought to explain why arbitrary power is objectionable.

The next section attempts to offer a more precise account of how tyrannical rule – and particularly the figure of the tyrant – relate to the concept of arbitrary power. Doing so helps us understand why arbitrary power is considered objectionable. It also exposes, section IV will show, problems with how the concept of arbitrary power is deployed, understood and communicated.

Tyrannocentric Arbitrariness

Tyrants walk among us in the form of the authoritarian populist leaders who have seen recent success, in the form of a certain type of petty bureaucrat that most of us have had the ill-fortune to encounter, and in the form of family members who abuse their power and authority. Though arbitrary power was associated with tyrants, on the contemporary understanding of arbitrary power in rule of law discussions, you do not need to be a tyrant to have arbitrary power.

Rather, the wielder of arbitrary power is like the tyrant. But in what respect? Arbitrary power, this Section argues, is power that is constituted to express, or enable the expression of, caricatured tyrannical traits.

Power is arbitrary to the extent to which it is constituted to express, or enable the expression of, caricatured tyrannical traits.

An exercise of power is arbitrary to the extent it expresses caricatured tyrannical traits

‘Expression’ is used here in Anderson and Pildes’ sense of a manifestation of an actor’s state of mind, their apparent state of mind, character traits or apparent character traits. (‘Expression’ is a term of art that I use reluctantly for continuity with the literature.) Expression requires neither that the actor intended to manifest, nor even possessed, the state of mind or trait expressed. A furtive shoplifter may express her nervousness in the way she talks to the cashier at the counter, even though she wants to hide her nervousness.³⁹ When someone fails to shake hands out of ignorance of social conventions (and not dislike or disrespect), she nevertheless expresses – manifests – dislike or disrespect.⁴⁰ Importantly for us, a ruler might express haughtiness, greediness and other caricatured tyrannical traits without possessing them, and without intending to express them. Section IV will suggest that this unsettles some commonly-held views about the morality of arbitrary power.

A statement that an actor has expressed a state of mind or trait involves an inference to best explanation.⁴¹ It is a statement of what, if true, would best explain their action.⁴² If a person walks down the streets of Berlin wearing a T-shirt with a swastika on it, in the absence of other known facts, neo-Nazi views best explain that action.⁴³

Understanding arbitrary power in terms of the expression of tyrannical traits helps us make sense of how it is used, particularly in the law. It allows us to distinguish between arbitrary and mistaken decisions, and explains why arbitrary power is taken to be objectionable.

Administrative law perhaps provides the best illustration of how arbitrariness is understood in an area of law that directly serves the rule of law. Common law grounds of review of administrative action are often rationalised as responses to arbitrariness. The ultra vires doctrine (which holds administrators to the limits of their powers), natural justice requirements, standards of reasonableness and requirements of reason-giving, and the expectation that power be exercised for public (not private) purposes, speak directly to caricatured tyrannical traits of power-hunger, haughtiness, caprice and self-interestedness. These grounds of review appear in some form in practically all common law jurisdictions; the fact that they reflect tyrannical traits so closely is telling.

It might also be useful to home in on a jurisdiction where government arbitrariness is constitutionally proscribed per se and has required judicial attention. (Many other jurisdictions prohibit arbitrary takings, arbitrary punishments etc.)⁴⁴ The Indian Constitution prohibits all arbitrary state action.⁴⁵ But while state action is routinely challenged and struck down for arbitrariness, the nature of arbitrariness has never been made clear. Courts largely proceed on the basis that they know it when they see it. The account offered here can make sense of these cases.

Leading Supreme Court judgments clearly associate arbitrariness with the figure of the tyrant, with “the whim and caprice of an absolute monarch”⁴⁶ acting “capriciously or at pleasure”⁴⁷ unconstrained by principle or reason.⁴⁸

But the figure of the tyrant features in subtler ways as well. Shayara Bano v Union of India featured a challenge to divorce practices associated with India’s personal law system, which applies state interpretations of religious law to religious groups in family matters. Influential judgments in the case declared the legal recognition of instantaneous and unilateral divorce by a  husband to be unconstitutional because it was “manifestly arbitrary in the sense that the marital tie can be broken capriciously and whimsically.⁴⁹ A husband’s action in divorcing his wife instantaneously and unilaterally expresses his capriciousness, haughtiness and failure to recognize moral constraints from his spouse’s rights or interests.

In the influential case of Ajay Hasia v Khalid Mujib Sehravardi,⁵⁰ the Indian Supreme Court found the admissions process to a university course to be arbitrary because it accorded significant weight (33.3%) to the applicants’ performance in the oral interview. It also held that the particular admissions decision challenged was arbitrary once it found that the interview only lasted a few minutes and involved irrelevant questions.

The court’s reasoning has been criticized as offering insufficient guidance on the nature of arbitrariness, and how state officials can avoid it.⁵¹ But a careful reading of the case reveals that the court assessed the arbitrariness of the decision by drawing inferences about the traits of the decisionmaker as immoral, “nepotistic” and “corrupt”.⁵² It assessed the decision-making power as arbitrary because it enabled such decision-making. In other words, it understood arbitrariness as the expression of caricatured tyrannical traits of indifference to the public interest and a lack of recognition for moral constraints.

In State of Mysore v. S.R. Jayaram,⁵³ a case involving recruitment into government jobs, the court condemned rules that allowed the government wide discretion in appointing officials, as arbitrary. This discretion was described as “vest[ing] in the Government an arbitrary power of patronage”,⁵⁴ expressing the traits of a tyrannical ruler who distributes fear and favour over a cowed population of flatterers.

Courts in these cases appeal – albeit implicitly – to the figure of the tyrant to distinguish arbitrary action and the power to act arbitrarily from mistaken action and the power to make mistakes. A husband could make an incorrect assessment of his chances of remaining happy while married to his wife, and initiate a divorce on that mistaken basis. University admission committees might incorrectly assess the nervousness of a candidate being interviewed as a lack of enthusiasm or potential and fail to give them a place. An appointment committee might give a seasoned bureaucrat a plum role over a promising up-and-comer, on the mistaken assumption that her experience will translate to better policy. The courts distinguish these and the hundreds of other mistaken decisions that we can imagine people making from arbitrary decisions, i.e. decisions that express caricatured tyrannical traits.

The account of arbitrary power defended here is also supported by the wealth of Anglo-American jurisprudence in which arbitrary power is explicitly or euphemistically represented by figures or devices associated with the sultan-as-tyrant. One figure is “the great caliph [who] would sit on his cushion and decide each case intuitively, without regard to precedents or reasoned elaboration of law."⁵⁵ This figure is closely identifiable with that of the tyrant: ‘caliph’ is a title claimed by the sultan-as-tyrant. The qadi (sometimes spelt kadi or cadi), an Ottoman judicial official closely associated with the sultan-as-tyrant, also makes frequent appearance in Anglo-American jurisprudence as a caricature of the arbitrary judge or decision-maker.⁵⁶ The qadi is often contrasted with the rule-bound, principled, appropriately deferential judicial decision-making.⁵⁷ Perhaps best-known in US jurisprudence are Justice Frankfurter’s remarks: “This is a court of review, not a tribunal unbounded by rules. We do not sit like a kadi under a tree dispensing justice according to considerations of individual expediency”.⁵⁸ More euphemistically, common law judges have long been fond of referring to ‘palm tree justice’, by which they mean ‘a form of justice dispensed by a cadi sitting under a palm tree without the advantage of books or precedents’⁵⁹, ‘without regard for legal principle or precedent⁶⁰ and ‘according to his personal whim and pleasure’⁶¹

In short, this account of arbitrary power helps make sense of how arbitrariness and cognate terms or devices are used in legal doctrine relating to the rule of law. But the account has more to recommend it. A good account of arbitrary power should help us understand why it is considered wrong or regrettable. This is relatively straightforward if arbitrary power is understood as expressing caricatured tyrannical traits. We object to being ruled by a tyrant. We object also to being ruled tyrannically – i.e. by someone expressing tyrannical traits – even if the ruler is not a full-blooded tyrant and their tyrannical actions are limited, occasional or unthinking. In the neo-Roman tradition, rule by a tyrant implies or creates a relationship which is traditionally described as slavery⁶² – the tyrant rules over cowed subjects forced to use flattering arts.⁶³ In the Indian case law discussed earlier, the relationships between the husband and wife, the admissions committee and the potential students, and the government and the job applicants might be described in similar terms, even if the subordination is attenuated compared to rule by a full-blooded tyrant. The power of the husband and the committees is constituted to enable capricious, haughty decision-making that is unresponsive to the public interest and fundamental rights.⁶⁴ This situation would not pass Philip Pettit’s eyeball test.⁶⁵ Those subject to this kind of power cannot look power-wielders full in the eye, needing neither to flatter or fear them.⁶⁶

In saying this, it is important to emphasise that it does not follow from this account that arbitrary power is necessarily immoral or unjustified (in fact, as will be discussed later, the account pushes back against exaggerated assumptions about the immorality of arbitrary power). What is offered here is an account of arbitrary power which seeks to make sense of how it is used in rule of law discussions, including in legal doctrine. In those discussions, arbitrary power is taken to be immoral or unjustified. As with other work of this type,⁶⁷ this account makes sense of these views, while remaining open to the possibility that they are incorrect or outweighed by other considerations.

So this paper does not aim to defend a full-blooded normative position on arbitrary power. But the account of arbitrary power offered here provokes ethical and epistemic questions about how the idea of arbitrary power is deployed, understood, and communicated. The next section will argue that the tyrannocentric account does more than make sense of how the concept of arbitrary power is used where the rule of law is implicated. By attending to the caricatured nature of the tyrant, this account reveals that we ought to be alert to the possibility that our understanding of arbitrariness and the rule of law is inflected with troubling presuppositions, including about the disliked others who are the subjects of caricature.

The Trouble with the Caricatured Tyrant

The caricature of the tyrant lies at the heart of the concept of arbitrary power, the previous sections argued. Anglophone neo-Roman thinkers inherited and contributed to a tradition in Western thought in which caricatured disliked others – Persian and Ottoman rulers – represent the tyrant, and arbitrary power. While there are strong continuities of thought about the tyrant across time and place, the caricatures were culturally and historically particular. Stock traits – such as the ones discussed in section II – were played up or played down to better fit the perceived characteristics of the disliked other.⁶⁸ Accounts of the characteristics of the disliked other were adjusted to better fit established stock traits.⁶⁹

This tradition of caricaturing disliked others to represent a tyrant, it has been plausibly argued, influences our concept of arbitrariness and the rule of law today.⁷⁰ If this is so, we ought to be alert to the possibility that (what we would today recognize as) the ethnocentricism of their notion of tyrant-centred arbitrary power inflects how we use the concept today.

This section suggests that there is reason to fear that the intellectual tradition of caricature of tyranny predisposes us to deploy the charge of arbitrary power inaccurately, misapprehend the features and conditions of arbitrary power, narrow our imaginative options in how to respond to arbitrary power, and exaggerate and universalise the wrongfulness of arbitrary power. Contemporary examples are offered to illustrate these dangers.

First and most obviously, the caricature predisposes us to seeing arbitrary power in some places rather than others, and amongst some peoples rather than others. Many prominent contributions to our understanding of tyranny and arbitrary power presumed that it was a natural state for particular climates,⁷¹ particular latitudes,⁷² particular ethnicities (‘natural slaves’),⁷³ or adherents of particular religions (particularly Islam).⁷⁴ James Harrington – thought to have had a strong influence in contemporary notions of the rule of law⁷⁵ – described arbitrary rule as flourishing under the ‘slavery, beggary and turkism’ associated with despotism.⁷⁶ Henry Parker argued that the King must not have arbitrary powers lest it “make[] us as servile as the turks”.⁷⁷

Justifications for colonialism were constructed on the foundations of this ethnocentric understanding of arbitrariness. American imperialism in the Philippines was explicitly defended by President McKinley as “substituting the mild sway of justice and right for arbitrary rule”.⁷⁸ The British Empire in India was justified by comparison with its arbitrary despotic past:

The view that India had long been enslaved by the tyranny of Oriental despotism made law a critical instrument by which Britons simultaneously established their authority and differentiated colonial law and order from the anarchy of previous regimes. In contrast to the personal and arbitrary rule of the Oriental despot, Britons saw their empire as an empire of law and liberty – not an empire of men, and certainly not an empire of violent men…⁷⁹

Elisabeth Kolsky documents the quotidian nature of private racial violence against Indians by Britons during British rule. Signs on walls of colonial hotels read: “Gentlemen are earnestly requested not to strike the servants.”⁸⁰ This kind of violence, she concludes, was condoned or tolerated in part because, while the British could see tyranny easily in the Mughal former rulers of India, they could not or would not see it in their own empire.⁸¹ Even where the arbitrariness of British rule was recognized, for instance by the impeachment of Warren Hastings,⁸² it was explained as an outcome of corruption by the East India Company’s ‘accommodation and imitation of “Asian despotism”’.⁸³ Even British arbitrariness was thought to be caused by Oriental tyrants.

The idea that arbitrariness is somehow inherent in native governance persists. In Ajay Hasia, the significant weighting (33.3%) accorded to the oral interview impugned the entire admissions process because of “the conditions prevailing in the country, particularly when there is deterioration in moral values and corruption and nepotism are very much on the increase”.⁸⁴ While the value of interviews was acknowledged, the possibility that an Indian admissions committee could run an interview process with integrity was never seriously contemplated.

The specific impact of representing arbitrary power though caricatured Mughal rulers also continues to be felt today, though it takes a different form in postcolonial India. Successive Hindu ethno-nationalist governments in India have rewritten school history textbooks to downplay the role of the (Muslim) Mughals in India’s history, and to erase Mughal exemplars of religious tolerance such as the emperor Akbar.

While downplaying examples of Mughal tolerance and syncretism, Hindu nationalist discourse also reinscribes the caricature of the Mughal as tyrant, exemplifying arbitrary power. One influential Hindu nationalist media outlet decries the “whimsical, arbitrary and unscrupulous” nature of rule by an emperor who was so capricious that approaching him could prove either “fatal or beneficial” “depending on one’s luck or clout”, and who promoted a “a climate of servility, which fundamentally debased the human character by smoking out self-respect” and requiring flattery.⁸⁵ (It is hoped that this description sounds familiar from section II.) In the current moment, Hindu nationalists, rather than British imperialists, present themselves as the alternative to a tyrannical past.

Hindu nationalists also make a connection between the arbitrariness of the historical Mughals and the dangerous character and aspirations of contemporary minority Indian Muslim citizens. As one nationalist media source argues: “The Hindu of today who is still unfamiliar with [Mughal tyranny] … stands to suffer the same fate” – “mass enslavement” by Muslims – as those subjected to Mughal rule.⁸⁶

This linking of Indian Muslims to tyranny is in line with significant legal discussions of arbitrariness. The Shayara Bano case discussed earlier involved Muslim men invoking a customary form of unilateral instantaneous divorce by pronouncement, known colloquially as triple talaq. Prominent judgments in the case that held that legal recognition of the practice was arbitrary were subsequently used by the government to justify its criminalization.⁸⁷ Muslim men accused of having pronounced divorce can now be imprisoned for up to three years.⁸⁸ Family law scholars and activists have noted the disparate treatment of Muslim men compared with other Indian men who might desert their families or spouses, and the disproportionate punishment the law prescribes.⁸⁹ The law is widely viewed as an attempt by the government to criminalise and control a disliked minority.⁹⁰ Ironically, caricatures of powerful Mughals and Sultans as exemplifying arbitrary power have contributed to the subordination of a relatively powerless class of citizens. Mughal tyranny, and hence arbitrariness, has an elective affinity with this class.

The dangers that attend our intellectual tradition of arbitrariness are also well illustrated by returning to the figure of the qadi. The Anglo-American judicial caricature of the qadi discussed earlier has likely informed legal and political opposition to qadis (or qazis) dispensing justice in South Asia. This opposition is sometimes explicitly framed in rule of law terms.⁹¹ While there is no doubt that actual (as opposed to caricatured) qadi justice has serious shortcomings, ethnographic studies show that in at least some contexts, contemporary qadis operate as accessible, inexpensive, familiar officials who serve the interests of women and marginalised communities.⁹² It is troubling then that the caricature may present an obstacle to accessible justice for these communities.

Beyond India, in recent times in the West, the Oriental exemplar for tyranny also seems to have played a role in media discourse supporting regime change in the Middle East. A study of reporting on Iraq in a national Australian newspaper finds support for the ‘Coalition of the Willing’ to be based on “Orientalist tropes such as that of “Oriental despotism” (and by extension arbitrary power) and a “narrative in which the Western world is the legitimate legatee of democracy and therefore reserves the right to democratise – under fire if necessary - the despotic Middle Eastern “other.””⁹³ The impacts of the caricature of the tyrant – particularly how that caricature shapes who is thought of as naturally arbitrary – have therefore been significant in contemporary law-making, policymaking and public discourse.

A second danger of our intellectual tradition of tyranny is that having the Ottoman Sultan as the caricatured tyrant may mislead us about the features that make power non-arbitrary: particularly, property rights and subjection to legal authority. The development of the European idea of the ‘absolute monarch’, who was not subject to any other human authority, raised the question of whether the Sultan was really so different from absolute monarchs in Europe.⁹⁴ He was different, it was argued, because the subjects of absolute monarchs enjoy private property, by contrast (it was wrongly thought) with the subjects of the Sultan.⁹⁵ The idea that private property precluded or protected against tyranny was important to anyone committed to preserving the superiority of European over Ottoman governance.⁹⁶ Against this history, we ought to be alert to the possibility that contemporary defences of property rights as a bulwark against arbitrary power⁹⁷ are exaggerated.⁹⁸ We ought to be doubly alert because colonial history also demonstrates that rule of law measures to secure private property often widened colonial power.⁹⁹

European monarchs, by contrast to the Sultan, were also thought to be constrained by law – divine and human.¹⁰⁰ (Other European writers at the time were aware of the independent authority of muftis and religious scholars and legalists in the Ottoman empire, but this contrast continued to be made.¹⁰¹) The fact that law was used to mark out tyrants, as part of an ethnocentric project, should make us question whether a ruler being constrained by law really is a distinguishing feature of non-arbitrary power.¹⁰² A failure to question the Sultan-caricature may have partially obscured our contemporary vision to the phenomenon of ‘autocratic legalists’ – who have all the other marks of a tyrant but rule through and within the law.¹⁰³ It may have allowed authoritarian, ethno-nationalist leaders – including Narendra Modi and Viktor Orban – to hide behind their legality.¹⁰⁴

Third, our inherited ideas about arbitrary power might also constrain how we see the options available to respond to it. The answer to the problem of arbitrary power is, for many, law, or at least rules. And one can understand why. We noted earlier how administrative law, as the branch of law concerned with executive/administrative power, perhaps provides the best illustration of how the law can be shaped to address arbitrariness. Of course the law protects fundamental rights through constitutional bills of rights, legislation and common law protections. These kinds of protections were a priority for neo-Roman thinkers concerned with tyranny, so that “in the case of our most important rights, such acts of arbitrary interference should be rendered legally impossible.”¹⁰⁵

Postema’s account of the rule of law is rightly modest: “the rule of law promises protection and recourse against the arbitrary exercise of power using the distinctive tools of the law.”¹⁰⁶ He acknowledges that a “number of other devices or strategies might be employed against the arbitrary exercise of power”.¹⁰⁷ But the intellectual history of the caricatured idea of arbitrariness might be obscuring other, rather obvious ways, of tempering arbitrary power. One of the most objectionable (to European writers of the past) features of despotism – exemplified by the Ottoman rule – was the absence of a role for a hereditary nobility in governance. The nobility was thought by some to offer a counterbalance of power of the European monarch.¹⁰⁸ By contrast, there was an understanding amongst Western writers that the role that the nobility played in Europe was performed in Ottoman governance by officials chosen for their ability and virtue.¹⁰⁹

We should consider whether this traditional suspicion towards selecting administrators for their virtue, and the relative emphasis on the law, might be colouring our assessment of how best to constrain arbitrary power. There is no denying the role that the law can play in reducing arbitrary power. But, as section V will discuss, dominant contemporary accounts reduce arbitrary power to the absence or inadequacy of constraints or accountability. The standard accounts emphasise means of checking power that are external to the power-wielder (particularly the law) and overlook internal means.

But inculcating and encouraging anti-tyrannical character and dispositions – humility, respect, concern for others, public-spiritedness – would reduce arbitrary power everywhere.¹¹⁰ Placing power in the hands of such an anti-tyrant (as opposed to someone who lacked these dispositions) would be an effective internal means of constraining the expression of tyrannical traits. This suggests that those concerned with arbitrary power should be concerned with civic education and the training and selection processes of public officials, and not just external checks.

Finally, the traditional characterization of subjects of particular tyrants as slaves¹¹¹ should lead us to question whether we exaggerate and misunderstand the moral significance of arbitrary power. It is unsurprising that a concept based on a caricature exaggerates particular features of the concept. It is the nature of a caricature to exaggerate. The intellectual history of tyranny may have influenced the exaggeration in particular directions. Noel Malcolm, in his sweeping account of the function of the Ottoman empire in Western political thought, demonstrates (1) the ubiquity of the characterization of the Sultan’s subjects as his slaves, and (2) the fictional nature of this characterization, probably based on a misreading of Machiavelli’s description of the Sultan’s administrators as his slaves.¹¹²

Contemporary republican theorists treat the power of a master over those that they enslave as a paradigm case of arbitrary power. The slavemaster’s power over those that they enslave is arbitrary. The problem with making that power the paradigm case of arbitrary power is that there so much else terribly wrong with the institution of slavery.¹¹³ Many instances of arbitrary power are not anything like the power of a master over those that they enslave. The admissions committee in the Ajay Hasia case¹¹⁴ discussed earlier perhaps better represents the typical holder of arbitrary power.

If we are able to examine and overcome any tendency to exaggerate the evil of arbitrary power, due to its equation with slavery in our intellectual tradition, we might be more attentive to other concerns about power and how it is exercised. Power might not be arbitrary but may be exercised mistakenly. A department of over-worked and under-trained officials overseeing welfare payments, who are by no means holders of arbitrary power, should perhaps attract some of the concern that arbitrariness receives. There is potential for them to make mistaken but non-arbitrary decisions that cause enormous harm. If we think of arbitrariness as akin to slavery, we might understandably discount these other harms.

One way to deflate our traditional exaggeration of the evil of arbitrary power is to emphasise that arbitrariness is sometimes a matter of appearance.¹¹⁵ Arbitrariness is immoral (when it is) because it is like (or more precisely, expresses the traits associated with) rule by a caricatured tyrant. But as section III noted, someone can express tyrannical traits unintentionally and even without possessing those traits. In other words, power or action may be arbitrary – for mimetic reasons – because it appears to be tyrannical, and not because it reflects the true traits of the ruler.

Lawyers are familiar with the idea that actions may be objectionable because of how they appear, from the injunction that justice not merely be done but also be seen to be done, and in the scrupulous avoidance of the appearance of bias in judicial and administrative proceedings.¹¹⁶ While the legal system may be attentive to potential mimetic immorality, it is worth remembering that appearing bad or immoral is not generally immoral in itself. Even the most innocuous action can be misunderstood or misconstrued depending on the onlooker and context such that it expresses tyrannical traits, and it cannot always be the actor’s responsibility to guard against such appearances.

Marcia Baron provides helpful examples. Assuming a context where a consenting sexual relationship between a graduate student and their thesis advisor is objectionable, should an advisor reduce the frequency of their public interactions with a student because false but very believable rumours about their relationship (based, say, on other students’ jealousy of the graduate student’s success) are floating around? Assume that even thoughtful reasonable people are influenced by the rumours, say because of the ubiquity of such relationships in the institution and the ability of such rumours to create loops of confirmatory feedback. Assume in other words that a sexual relationship is the best explanation for the parties’ behavior, given the limited information available to onlookers. The advisor’s actions could then be characterized as expressing various tyrannical traits. But here acting to preserve or create the right appearance – by meeting the student in secret, by meeting less or by appearing notably cold to the student – is not justified.

Some of the usual reasons why people care about appearances – not setting a bad example, preserving norms and institutions whether legal or academic – might apply in this case. But it seems too much to ask the advisor to take these considerations into account, let alone to avoid the appearance of wrongdoing.

No doubt for some officials – such as judges – the consequences of appearing to act immorally can cause immense damage. But there is a huge diversity of ‘rulers’ who are held to the non-arbitrariness standard: national presidents, minor bureaucrats, administrators of public universities. They might unintentionally and unknowingly express haughtiness, indifference to individual rights or the public interest or other tyrannical traits.

So the condemnation of arbitrary power is not always justified. Actions, arrangements or character traits are not objectionable just because they resemble something objectionable. Perhaps even more importantly, genealogies of tyranny suggests that we might be predisposed to exaggerate or even manufacture moral concerns with caricatured tyrannical traits. Some of these traits may not be as bad as we are intellectually predisposed to think they are; and insofar as arbitrary power is objectionable because it expresses such traits, it is not as bad as we think it is.

When we deflate the exaggerated assumption that arbitrary power is always immoral, we can better evaluate its moral significance across a broader range of cases.

In summary, by centering the tyrant as a caricature, the account of arbitrary power defended here helps us to:

• a. deploy the charge of arbitrary power more accurately, avoiding ethnocentricism and stereotypes;

• b. avoid misleading assumptions about how far law and property rights protect against arbitrariness;

• c. open up the options available to respond to the problem of arbitrary power; and

• d. moderate overblown and universalised claims about the wrongfulness of arbitrary power.

But how does it fare against competing accounts?

Competing Accounts of Arbitrary Power

Painting with very broad brush strokes, arbitrary power is generally thought of – in philosophical, legal and lay discussions – in one or more of the following ways.¹¹⁷ This section suggests why the account defended here has advantages over these accounts, at least insofar as our interest is in the rule of law.

Unconstrained power

Power is arbitrary, on one view, when it is not “reliably controlled by effective rules, procedures, or goals”.¹¹⁸ This idea positions arbitrariness in opposition to the formal conception of the rule of law, which emphasizes governance by stable public rules.

Unpredictable power

Constrained power might only be constrained in some respects, such that its exercise is unpredictable. This is true of the power that constitutes the offices of a number of contemporary authoritarian leaders.¹¹⁹ But to take a benign example, imagine a case where an official has power to make decisions ‘to further the public interest in health’,¹²⁰ and the official is reliably constrained to make only such decisions. The official might still exercise the power unpredictably because there are a number of ways to further this public interest e.g. by building hospitals, increasing air quality, planting trees, investing in primary care etc. This kind of power would be arbitrary on a different account that conceives of arbitrary power as the capacity to exercise power in ways that are unpredictable by those subject to the power.¹²¹

On first blush, these two accounts of arbitrary power as unconstrained or unpredictable power are attractive to anyone seeking to understand the rule of law; arbitrary power on these accounts bears an oppositional relation to the rule of law. The trouble is that the opposition tells us nothing further about the rule of law. Arbitrary power, on these accounts, just is the negation of the rule of law.¹²² Power is arbitrary on these accounts when it does not have to be exercised through stable, predictable, public rules. This is simply a restatement of the formal conception of the rule of law. If this is all there is to arbitrariness, talk of ‘arbitrary power’ does not really advance our understanding of the rule of law. If the idea of arbitrary power is to help rule of law theorists, it must be related to the rule of law, but must do more than mirror it.

But there is a more fundamental concern with these accounts. Predictable power that is controlled by effective rules can still, others argue, be arbitrary.¹²³ Paradigm cases of arbitrary power – that of a master over those he enslaves – are not rendered non-arbitrary by the existence of some reliable legal constraints (e.g. against maiming or torture).¹²⁴ Nor is the master’s power over the enslaved non-arbitrary because the master exercises power in a predictable fashion.

Unaccountable power

Power, according to a third view, is arbitrary unless its exercise is directly or indirectly responsive (accountable) to those subject to the power. On Philip Pettit’s influential view, to avoid arbitrariness, power must be constrained to track the interests of those under its sway, as seen by their own lights.¹²⁵

This is an attractive moral ideal of how power ought to be constituted. It is however better characterised as a democratic ideal¹²⁶ rather than as an account of arbitrariness. This democratic ideal is much more demanding than the idea of arbitrariness, as understood in rule-of-law discussions, in most legal doctrine and even in everyday conversation. Indeed, Pettit’s later work does not describe the ideal in terms of arbitrariness¹²⁷ and his account of the rule of law does not require non-arbitrariness in this democratic sense.¹²⁸

Umbrella accounts

Some scholars bring these kinds of power – unconstrained, unaccountable and unpredictable – together under ‘arbitrariness’ as an umbrella term for (some) morally objectionable kinds of power.¹²⁹ But what unites these kinds of power and distinguishes them from other morally objectionable kinds of power? ¹³⁰ In other words, is the label ‘arbitrary’ adding anything meaningful to discussions of the rule of law?¹³¹ Would anything (apart from brevity) be lost if we simply spoke of unconstrained, unpredictable and unaccountable powers?

There are expansive answers out there for what unites these kind of powers: e.g. that wielders of such powers can ‘choose not to “give[] due weight to ...the genuinely respect-worthy interests, expectations, and rights of all relevant persons.”’¹³² I struggle to think of many examples of powers which are not arbitrary on this account. Where you have a modicum of discretion and any room for the slightest mistake¹³³ about the weight due to all the respect-worthy interests, expectations, and rights of all relevant persons – you have arbitrary power.

A good account of arbitrary power should distinguish it from adjacent or related concepts. Since everyone agrees that non-arbitrary power is possible, an account of arbitrary power should not collapse the distinction between arbitrary powers and non-arbitrary powers (typically powers exercisable in a mistaken or imperfect way). The umbrella accounts of arbitrary power fail this criterion.

Contemporary thinking about arbitrary power disassociates it with any particular kind of person. It is after all potentially exercisable by a range of people: state officials, business tycoons, social media influencers and others. By contrast, this paper has argued that focusing on the tyrant as a caricature – informed by our intellectual history - reveals an understanding of arbitrariness that has the potential to illuminate, rather than merely mirror, the rule of law.

But this paper does not argue that the standard accounts of arbitrariness discussed above are dramatically flawed. While often presented as competing accounts of arbitrariness, they all reflect a concern with different tyrannical traits. If the ruler is accountable in the exercise of their powers, the expression of haughtiness is less likely. To the extent that a ruler’s power is constrained, the expression of power-hunger becomes less likely. In other words, constraining power, as well as making its exercise predictable and accountable, are all good strategies to check arbitrary power. But to understand the concept of arbitrary power is to appreciate that these strategies are so important because the central concern is with rule that expresses caricatured tyrannical traits.

Conclusion

This account of arbitrary power (centred on the tyrant as a caricature) makes sense of how it is used in discussions about the rule of law, including in legal doctrine. While this paper is not a moral evaluation of arbitrary power, it aims to help us see problems with how the concept is deployed, understood and communicated. It suggests that arbitrariness may not always be objectionable. The account offered here also opens up some important further questions that this paper did not explore.

For example, those who are committed to identifying the kind of power which is always objectionable in the hands of a ruler might ask: if arbitrariness does not always track objectionable power, should we turn the spotlight away from the tyrant and the power-wielder, and on to those subject to power? The intellectual traditions that influence our current understanding of arbitrariness and the rule of law are attentive to the impact of tyranny on its subject. Those subject to tyranny are flatterers, distrustful of each other, lacking in forthrightness and “small souled”.¹³⁴ Someone committed to identifying the kind of power which is always objectionable might take the account offered here as encouragement to press further on the question: what kind of power produces this kind of subject?¹³⁵