Moderation as Government: Montesquieu and the Divisibility of Power

ABSTRACT

The principle of moderation can be regarded as an ethical principle of virtue or as a principle of government. On the basis of the former, moderation has a personal, ethical sense—not to go towards extremes. The latter model is more generalized and impersonal: moderation as the limitation of power by power. Both conceptions actually meet, though with the latter model more salient, in the work of Montesquieu. This article outlines Montesquieu’s view of moderation emphasizing the extent to which this view cannot be understood apart from his concept of despotism. It shows that Montesquieu’s understanding of moderation entails the balancing and interaction of different powers in a State. As such, his view is more “dynamic” and relational than “substantive.” By exploring the interaction between the models of moderation as virtue and moderation as government in Montesquieu’s work, I then develop the notion of “ruling fear” in considering the continuing relevance of Montesquieu’s perspective for current understandings of moderation and political power.

KEYWORDS:

• moderation
• despotism
• Montesquieu
• ruling fear

Introduction

The concept of moderation is ripe for renewal in social and political theory. Of course, the concept has a lengthy, variegated and complex history, but two main emphases can be distinguished. On the one hand there is an ethical or personal model: moderation understood as a kind of golden mean, a middle position, away from the extremes. This was largely the classical conception of moderation. Crucial here was the affirmation of balanced conduct, justice as a virtue. In On Duties, Cicero invokes moderation as “doing the right thing at the right time” or “right-ordering.”¹ More generally this type of moderation is associated with following a “middle path” between extremes. In Greek and Roman ethics the good citizen was one who embraced the mean in this sense, who was moderate in all things.

On the other hand, there is a more modern conception of moderation. This does not entail finding a middle path on an ethical level so much as balancing different impersonal types of power. Moderation, here, means a mutual tempering of powers; it is not so much an ethical or personalized idea as, at its root, a political one whereby different forces moderate each other in the sense of balancing each other out.

This latter conception can be associated, in particular, with Montesquieu.² Although as we will see later on, he did not repudiate the importance of ethical moderation—taking the middle way—as a political virtue, it can nevertheless be said that Montesquieu pushed the emphasis more towards de-personalized moderation; that he refashioned the idea of moderation for circumstances that were “after virtue.” To do so he proposed a very basic dualistic topology of power, with moderation understood in opposition to despotism.

Usually Montesquieu’s notion of power is held to be tripartite. If one reads through The Spirit of the Laws, published in 1748, one will indeed be struck from the very beginning by the author’s quasi-Aristotelian threefold typification of possible regimes.³ These are “republics” (including democracies and aristocracies), “monarchies” and “despotism” (10). However, as Montesquieu’s argument progresses and deepens, he rather abandons or at least downplays the trichotomy of regimes for a more basic dichotomy: that between despotism generally and “moderate government” of whatever sort.⁴ Moderate governments are nowhere precisely defined by Montesquieu, although there are exemplifications—arguably, the famous treatment of English government in Book 11, chapter 6 being the paradigm. Here Montesquieu paints the famous picture of what later became familiar as the “separation of powers.” But what seems best to characterize these sorts of regime is not so much an essence that can be defined as their distance from indivisible power—that is, from despotism.

This means that it is impossible to understand the meaning of moderation for Montesquieu if we do not first understand its antithesis.

Despotism as Indivisible Power

Despotism is much more than tyranny for Montesquieu. It is more than just a political regime; or rather it is a regime where the politics spreads down into non-political life, civil society, the personal sphere, the family. It is in that sense an entire state of society. In a despotic regime, everything is despotic (60). Here the universal spring or basic principle is fear, and fear is both self-perpetuating and all-consuming (28). In despotism there is no law, or at least despotism is without independent law; rather the law is reducible to the whim of the despot himself—despot’s law. With despotism, then, there is monolithic, indivisible power. Everything comes through the despot, and even the functionaries of the system are themselves despots; “in despotic government, one alone, without law and without rule, draws everything along by his will and his caprices” (10; emphasis in the original). There is, then, no divisibility of power. Despotism is one-dimensional domination; in fact if we are to follow the spirit rather than the letter of Montesquieu we might say that it does not need a despot as such, but simply occurs wherever there is no distinction and discrimination of powers. This is why Tocqueville was writing in a spirit derived from Montesquieu when he wrote of the despotic tendencies of majoritarian opinion in a democracy.⁵

Yet there is a sense in which despotism has rather a deliberately “staged” character in Montesquieu’s work. It has something of an abstract, deliberately hyperbolic status. It is, so to speak, the zero degree of all power.⁶ As Louis Althusser wrote, it is “the idea of absolute evil, the idea of the very limits of politics as such.”⁷ It is not just another “type” of regime but simply the summum malum of all political life, the worst that can happen. Despotism, at the extreme, is so pernicious that everyone, even the despot, is captured by it: “Despotism is so terrible that it even turns against those who exercise it” (197).

Reconstructed in this way, perhaps despotism seems like an overblown concept, and no doubt a somewhat “Orientalist” one (283).⁸ Are not just about all of the instances that Montesquieu invokes—China, Japan, Russia, Turkey—non-Western or at least non-European? Yet although Montesquieu’s examples were indeed largely—though not quite entirely—of this sort, it is also the case that he regarded despotism as a universal condition, latent to all systems of power. And if, as such, his image of despotism seems inordinately bleak then it is best to see it as a sort of “pure type” rather than as a straightforwardly descriptive category. Indeed, as a descriptive empirical category perhaps no actual society fits it completely. In that sense it is indeed a kind of exaggeration; and Max Weber was purportedly influenced by Montesquieu’s conception in his invention of the “ideal-type” method whereby one uses abstract categories to “measure” reality on the ground.⁹ No actual society completely fitted Weber’s ideal-type of bureaucracy; like Montesquieu’s despotism, it is as much an abstract—and so a critical and analytical—category as an empirical one. It is a tool of critical measurement; a measure against reality rather than being, exactly, reality itself. Certainly, with his central category of despotism Montesquieu supplies us with a rather extreme vision of the very worst. Hence writers such as Voltaire, in his “ABC, or Dialogues between A, B and C,” were impressed but sceptical, seeing the concept as something of a satire on power.¹⁰

The point about despotism, then, is that it is best seen as a conceptual yardstick rather than straightforwardly as an empirical category. By depicting it as the worst, one can gauge political systems in terms of their relative distance from the extreme. In Montesquieu there is no substantive summum bonum at all. There is no vision of a perfected society. His influence even on the French Revolution was marked, but always vitiated by what was perceived as his formalism or relativism, his refusal to indicate the best way.¹¹ From Montesquieu’s kind of perspective, those who think they have found positive answers almost certainly have not, and can themselves be a source of monolithic power; and in any case, when it comes to politics all answers are only provisional.

We do know that Montesquieu admired the English constitution—hardly the ethical values of the English themselves—as an instance of moderate government; but not as an ideal, still less a utopia. England seems to have been an example more of the least-worse than the absolutely good. Montesquieu gives us, then, something like a negative kind of vision; he delineates the worst and invokes the good only in negative relation to that. This is not, however, out of absolute despair but from what was a principled scepticism concerning the more utopian, and hence dangerously unrealistic, tendencies of power and politics. Indeed, it is more a question of realism than outright cynicism or despair; or rather Montesquieu, in despairing of the good, makes way for a kind of complex, provisional hope in the possibility of the least-worse.

So if there is no absolute good, what is the least worse? The chapter on England, Book 11, chapter 6, of The Spirit of the Laws is, as already noted, the obvious source (156–66). It is in some ways an odd chapter. It was written some time before the main body of The Spirit of the Laws (that is, by around 1734) and in some ways—for instance in its rather formalistic tone (the chapter is “about” England but England itself is not specifically mentioned)—it is anomalous.¹² What was it about England? It is certainly not that Montesquieu regarded the English themselves as particularly virtuous types; if they are moderate it is not by way of personal morality or virtue. In The Persian Letters (1721) the English are described (by the fictional protagonist Uzbek) as having “a restive disposition. … Meekness and compliance are the virtues on which they pride themselves least.”¹³ In his personal reflections, Montesquieu notes of the English: “The nation, insolent; the individuals, modest.”¹⁴ The merits of English government were historical and constitutional rather than having anything to do with personal virtue; the English could get rid of kings if they felt they were illegitimate, as they had done in 1688. And they had produced—perhaps more by accident than intent—a “system of political liberty” that consisted of a separation or rather balancing of powers.

Divisible Powers: Moderation as Balance

As for the so-called separation of powers, it is important to observe that it is the concept that features rather than the actual words. Montesquieu does not invoke the separation of powers specifically as a principle; but, certainly, in his description of the English way of government he invokes something like the balancing of powers and counter-powers as key to the spirit of moderate government.¹⁵ Again, this is not so much a political formula in a positive sense as a register of the degree of distance that might plausibly be attained from the rule of despotism, understood as the rule of one person that pervades the entire social body indivisibly and without remainder. After all, if for Montesquieu despotism was a kind of latent state of all political conditions, the key was, if possible, to escape it even if such an escape could not be perpetually prolonged or even be absolute or final. Where despotism is about the indivisibility of power, moderate government is about its divisibility and what the English had discovered—at least to an extent—was not how to be morally good or virtuous so much as how to divide up power (66).

The English constitution, then, balanced powers, or so Montesquieu thought. In his “Notes sur L’Angleterre,” Montesquieu argued that England is the most free country in the world in that the monarch has limited power, but also because the House of Commons was limited by the separation of legislative and executive power. King and Commons limit each other.¹⁶ It is not necessarily outright separation, but nor is it exactly balance in a positive sense; it is more like balance as a kind of mutual—negative—restriction. Montesquieu seems to have taken the legislative-executive distinction from Locke’s Second Treatise, but to have adapted the meanings of the terms. In fact the exact contours and limits of these terms themselves are unclear and they seem to evolve as the chapter progresses.¹⁷ But the basic principle is there: that in England power is not indivisible or hollowed out, that it has levels and dimensions that interact with each other. Rather, it is mixed, graduated power at some remove from the undifferentiated, molar power of despotism.

In Book 11, chapter 6 of The Spirit of the Laws Montesquieu asserts the general principle that the various kinds of power in a State should not be held by the same person or the same body. Above all, there should be gradations of power. Montesquieu’s actual terminology is usually, for instance in Book 2 of The Spirit of the Laws, to pouvoirs intermédiaries, intermediate powers (17–19). But “intermediate” here should not perhaps be thought of in terms of a coming between two or more kinds of power (for instance in terms of voluntary associations in civil society that are “between” state and private life), but as the mutual “mediation,” the mutual inter-relation and balancing of powers. Actually the notion of balance here is fairly complex: “In order to form a moderate government, one must combine powers, regulate them, temper them, make them act; one must give power a ballast, so to speak, to put it in a position to resist another; this is a masterpiece of legislation that chance rarely produces and prudence is rarely allowed to produce” (63).¹⁸

It was Carl Schmitt, citing Montesquieu’s dictum of le pouvoir arrête le pouvoir (power checks power) who seems to have been first to argue that it is better to translate the basic thought here in terms of the balancing not the separation of powers. “The theory of the so-called separation of powers becomes incomprehensible for someone who places too much emphasis on separation and division rather than on the notion of balance … arrêter, enchainer, lier, empêcher [to stop, put in chains, tie, hinder] are the key words of the famous chapter 6.”¹⁹

The idea of balancing powers was not in itself new. In Aristotle there is the idea of “mixed government” (not so much between powers as between classes); Polybius has the notion in Book 6 of his Histories of a separation of powers; Machiavelli invoked the conflict of classes at imperial Rome; and something like the principle of moderating powers appears in the writings of John Calvin, which then influenced Locke. Montesquieu, though himself certainly influenced by Locke, quite probably took the idea from Bolingbroke’s Craftsman.²⁰ But as Schmitt and others have pointed out, the immediate context of the argument in Montesquieu’s thought was actually located nearer home in the famous thèse nobiliare; it was Louis XIV’s establishment—as it were, over the heads of the nobility—of the Intendant system (something that was also very much to exercise Tocqueville later on in his analysis of the French Revolution). Whereas previously the nobility had acted as a kind of buffer to monarchical power, the Intendants were direct expressions of it.²¹ The state was being hollowed out towards what Montesquieu and his predecessors in the nobiliare tendency regarded as a quasi-despotism. In this sense the concept is far from being Orientalist in its conception; on the contrary, the whole idea is to direct it towards the West, towards—in Montesquieu’s terms—us.²²

Does the idea of the balance of power not add up to a normative interpretation of State power? If so, it is only minimally; it is something of a distortion to think that Montesquieu is contributing to constitutional theory as conceived by later generations. Subsequent writings from within liberal and constitutional types of theory have tended to treat him as offering a fixed theory of the State, involving the executive, legislative and judiciary, and then finding this theory wanting because it is neither entirely coherent nor empirically viable.²³ Indeed it is not coherent nor always empirically viable; but then it is not such a theory. In fact it is quite striking that Montesquieu has no theory of the State, as such, at all; and although he was concerned with the laws and with legal theory, this is not restricted to constitutional arrangements alone. For him, the balance of power is certainly no single, substantial kind of “thing,” a balance between legislative, executive and judiciary. Rather, it is a dynamic principle: a generalised means of restricting power by power.

The appropriate image of the moderated balance of power is of divisible power, a series of overlapping institutions, not—as with despotism—a single, vast one. Now, liberal theory has tended to be either obsessed with the State (disliking it – Robert Nozick) or more or less ignoring it—although certainly assuming it (John Rawls). Montesquieu’s perspective offers us a way out of this false alternative. One has to look at modes of government rather than just the State; and the best State, or the least-worst, the kind that is farthest from the despotic kind, is obviously that which is most free from fear. Here Montesquieu does perhaps connect up to liberal theory but to a sort of negative version of it.²⁴ The point is that Montesquieu regards government as being integrally bound up with the exercise of power and never shies away from that. Good government is not, as liberal theorists often seem to think, about minimizing power away. For Montesquieu, power is there—it is not intrinsically bad (little can be done without it) but is always dangerous: “Government, though a necessity, is a dangerous necessity.”²⁵

If the concept of moderation, understood in these terms, is essentially negative—as a mutual restrictiveness of powers—it should be emphasized, nonetheless, that this can be differentiated from conceptions of negative as opposed to positive freedom. The contrast is useful however. Isaiah Berlin famously distinguished between negative and positive liberty.²⁶ For Berlin, negative liberty was the absence of restrictions on doing or believing something, whereas positive liberty was the freedom to do or believe something. Berlin held, quite rightly, that both conceptions could be distorted towards malign ends; negative liberty for instance was compatible with tyranny, and positive liberty was liable to capture by those who, in the wake of Rousseau and Marx, would force us to be free. Certainly, Berlin devoted more pages to the problems of positive liberty, but that did not mean that he did not recognize the problems and limits of each conception. But since his suspicion was, in effect, with power itself, he seemed to prefer, and for that reason, the negative version as being the least concerned with multiplying power.

Montesquieu’s version of “negative” moderation has a different emphasis. It is not about minimizing power overall. If anything it demands more power; or at least it demands power to countermand power. It is about the proliferation of powers as an antidote to the concentration of power, whether that concentration is itself minimal or all-enveloping; and it is about the separating and countermanding of powers by powers. So Montesquieu’s version of moderation is not a despair of or scepticism towards power per se so much as a methodological and political wariness of the uses to which kinds of power can be put. Hence the alternative to despotism is not the absence of power; it is more to do with the multiplication of powers. But even then there is no final “end” to be either hoped for or expected. There is no teleology in Montesquieu. There must only be a constant vigilance towards power, since any action against such more concentrated and monolithic kinds of power is always provisional and never definitive.

Institutions are important for Montesquieu in this endeavour. Berlin’s notion of negative freedom would not, in contrast, particularly have to do with institutions; indeed, institutions tend to get in the way of negative freedom because Berlin thinks of institutions themselves negatively, as restrictions upon freedom.²⁷ In contrast, Montesquieu’s own “negative” or at least critical conception of power places a great emphasis on institutions rather than persons; not just because institutions balance each other, but because institutions can be used as counter-powers to persons themselves—that is, to capricious power and individual whim. Persons can, so to speak, be locked into virtue by institutions, or at least the right kinds of institutions. Whether people are “bad” or “good,” institutions should be moderating forces on their passions.²⁸

Moderating Institutions

In fact Montesquieu’s conception of moderation invites us to think about institutions differently from what has often been the norm in liberal theory. For Montesquieu, institutions are themselves powers that balance other powers (institutions). It is not about being outside power, but about moderating power. That is what institutions do. Outside despotism where all institutions are merely emanations of the despot, all institutions are really counter-powers; any institution is countering another and so diffusing, differentiating—but not necessarily diminishing—the sum of power.

But what, really, are institutions? Contemporary sociologists make a distinction that might be broadly useful here—one between institutions themselves and organizations.²⁹ Both are in Montesquieu’s sense powers, but organizations are, as it were, concrete powers (an assembly, a parliament, a court, even the State), whereas institutions are practices (a certain type of family structure, the market economy). Obviously this distinction is somewhat provisional to say the least and of course the two types of power certainly overlap (the family can be an organization and a type of practice), but at least the distinction helps us recognize the range of what counts as a form of power in Montesquieu’s style of thinking. More specifically, it helps us see that powers are not just a matter of constitutional arrangements in a narrow or fixed sense, for instance in that of political philosophy’s simplistic triumvirate of executive-legislative-judiciary. Certainly the constitution will consist of an array of organizations linked together in certain ways. But on a more general institutional level, other forms of moderating power are important—the powers of moeurs and mores, for instance, institutionalized orientations based on social class, or for that matter the great theme in The Spirit of the Laws (Part 4) of commerce as a means of softening the barbs of power, or even religion as moderating in some contexts (460). In sum, what Montesquieu gives us is a much wider remit than most political philosophy or political or constitutional theory would normally allow in terms of institutions, their nature, their scope and their relative capacities for moderation.

Emphasizing this feature of Montesquieu—a style of thought that makes him in today’s parlance a “sociologist” of the political as much as a “political theorist” or a “political philosopher”—does not mean we have to agree with his substantive analyses. For instance, several questions beg themselves about Montesquieu’s picture of the English constitution, not least—is it accurate? As Robert Shackleton points out, it is misleading to see England, as Montesquieu to many readers appears to do, in terms of any outright “separation” of powers since from Walpole’s time there was something more like a fusion of executive and legislative powers, and then the cabinet system, which was not featured in any formal description of the English constitution.³⁰ David Hume, publishing his first version of his Essays in 1741, noted that the House of Commons in effect had supreme power in the English constitution.³¹ The three powers—legislative, executive and judiciary—clearly had overlaps at the very least. In Federalist 47 James Madison doled out the by-then familiar praise for Montesquieu, then pointed out that he had got the English constitution wrong. Executive and legislative overlapped, as did executive and judiciary (since the executive appointed the judiciary), as did legislative and judiciary (the Lords).³²

On the other hand, there was also evidence if not of tyranny or despotism then certainly of a narrowing of powers. In this context the most telling criticism came over a century later (in the 1860s) from Walter Bagehot. Without referring to Montesquieu by name, Bagehot scoffed at the idea of a separation of powers as being the basis of the English constitution. Such ideas were merely the result of reflection “on paper” not in practice, whereas the “efficient secret” of the constitution was exactly the opposite of the paper version: “the nearly complete fusion of the executive and legislative powers.”³³ Any reader of Montesquieu might pause to reflect on Bagehot’s vocabulary. Was there by the mid-nineteenth century indeed a kind of despotism in terms of cabinet government? Was the prime minister in effect a despot?

A great deal had changed between Montesquieu and Bagehot. Yet no doubt Montesquieu misunderstood much of the English constitution. More embarrassingly perhaps, his own portrait of it did not obviously fit very well into any of his formal categories of government as outlined in The Spirit of the Laws. It was not technically a republic and obviously not a despotism—but was it really simply a monarchy? Hume was to describe it as an evolving hybrid, between monarchy and a republic, if perhaps tending more and more towards monarchy.³⁴ Perhaps that was the point. It was a hybrid; and the more hybrid a form of government presumably the more “moderated”—balanced and counter-balanced—and so, at least from Montesquieu’s point of view, un-despotic. Of course this was all provisional and in no way did English government amount to some superior “stage of development” for Montesquieu. Notwithstanding the fact that it is, in any case, misleading to see Montesquieu as offering anything amounting to a “philosophy of history,” the separation of powers is not a kind of culmination point; it is a process, and, as I have indicated, it is not really a question of outright “separation” at all. At no point does Montesquieu actually suggest that the powers in the English constitution were completely separate powers. If they are separate at all it is in so far as all or several are not combined in one person or one body. But they are not isolated completely from one another; rather they are entangled, they overlap and balance each other. They restrict—they moderate—each other. Montesquieu’s own perceptions are messier—more complex, more sceptical—than much of the political science literature that has come after him. In being so, they are also assuredly more productive.

Depersonalizing Government

At this point we might revisit a little more critically the rather basic distinction that I have been making between personal moderation (as a matter of personal ethics or virtue) and the more impersonal principle of moderation, as outlined here, in terms of institutions of government. It is certainly the case, as interpreters of Montesquieu such as Judith Shklar have emphasized so well, that Montesquieu’s own position tended to be oriented to moderation as government rather than to issues of personal virtue.³⁵ Shklar insists that it was crucial to his project to minimize the role of personal virtue in political affairs, which was one reason for his undoubted scepticism towards exclusively republican institutions. Shklar characterizes Montesquieu’s entire political science as “a highly depersonalized account of social phenomena.”³⁶ Montesquieu was too sceptical to trust to what Shklar labels “good characters” in government. Rather government had to be set up so that even “bad characters” could engage in it without too much diminution of legitimate functioning.³⁷ In other words, it is best to base political trust so far as possible on institutions rather than persons.

Certainly, there is a sense in which any kind of personalized power is dangerous for Montesquieu. Unmoderated personal power can become unpredictable and coercive. Any kind of centralizing power might approach despotism if based merely on caprice—a favorite term of Montesquieu’s. And although Montesquieu was at heart a monarchist, what he feared most even about monarchy was the fact that, in extreme cases, a single person could decide on war or capricious behavior and attempt to gain glory on that basis, so sacrificing many others to the pursuit of their own personal interest. Moreover, in this emphasis Montesquieu was in line with the “scientific” character of his intellectual times. Montesquieu was in many ways a Cartesian and it is assuredly true that this pro-science perspective informed his scepticism towards moderation understood in terms exclusively of morality or virtue.³⁸

On the other hand, Montesquieu’s emphasis on moderation as depersonalized power did have implications for the personal model and he certainly did not forget it altogether. He was not, in that sense, entirely “after virtue” and Shklar’s type of emphasis on Montesquieu’s conception of depersonalized power can be overstated. Depersonalized institutions are precisely what are likely to keep “bad characters” more or less true; people were more likely to behave well if not with outright virtue then as basic, more or less “good” characters, if powers were balanced against each other, thus holding them to account in their own institutional spheres. What people do or do not do and how they relate personally to institutions certainly matter; and of course institutions themselves have to be designed by persons.

So personal conduct still matters even if Montesquieu’s analytical emphasis is on, as it were, the impersonal regulation of personal power. The principle of moderation invoked here is a mosaic of the surveillance of power by power, each power “moderating” others; in other words, a perpetual scrutiny of some parts of government (and society) by others, and mutually, a scrutiny that would at least disincentivize bad conduct at a personal level. As such, power needs, so far as possible, to be made public. It needs to go beyond the personal. Powers need transparency.

This theme was to have a long pedigree after Montesquieu; it is the theme of the distrust in power that is so central to any principle of moderation; the constructive distrust in politicians, scepticism about power which is, somewhat paradoxically, integral to the legitimacy of moderate political systems.³⁹ “Is it objected against the regime of publicity,” Bentham wrote in his Political Tactics, “that it is a system of distrust? This is true; and every good political institution is founded on this base.”⁴⁰ This principle of what could be termed moderation by mutual surveillance had to do with the oversight of government business: transcripts of debates, records of votes, public admission to the assembly and so on.

Montesquieu, in his own way, also emphasized the principle of publicity in this sense; in fact it could be argued that he pioneered the approach (Book 6, chapter 8).⁴¹ In his notes on England, Montesquieu invokes the committee system of the English parliament in terms of a moderation of the deliberations of the House of Commons itself: “after the matter has been adequately discussed [in committee] the Speaker resumes his place, and they deliberate in the House.”⁴² The details are less important than the principle; that people are kept “good,” certainly, but that this is achieved not simply by their intrinsic, personal sense of virtue or honor but by external oversight of their activities.

Ruling Fear

In other words, as Montesquieu saw it, the moderation of virtue and the moderation of government effectively presupposed each other, or at least the latter might be a spur to the former. But there is also a sense—and this is a theme very much pertinent today—in which moderate styles of government impact on personal power and virtue in a more immediate way still. This indeed has very much to do with the personal sensibilities of those in power themselves. To see this, we need to recall that when it comes to his considerations on his summum malum, despotism, the basic principle for Montesquieu has to do with an exorbitant type of fear. In fact, Montesquieu introduces this theme in Book 3, chapter 9 of The Spirit of the Laws not with the term “peur” (fear) but with “crainte” (something more akin to extreme fear or terror).⁴³ This is the sort of fear that “must beat down everyone’s courage and extinguish even the slightest feeling of ambition” (28).

Now, this theme can be regarded largely in terms of the fear of the populace subjugated to the powers of the despot, and indeed that is how Montesquieu often writes of it (28–29). In despotism all those who are subject to rule are cripplingly fearful. But there is also a fear that is perhaps more dangerous than the fear that the populace has of the despot, and that is the fear experienced in such regimes by those in power themselves—above all, the despot’s own fear. In a despotism it is the despot’s own terror—and its consequences—that is to be most feared.

For Montesquieu, despots and tyrants are as much forced into bad behavior as they choose it. Despotism does not allow despots to be honorable (27–28). This has to do with the very logic of the regime. If everything is based on terrorizing fear then this has to be perpetuated, kept at a constantly intense level. This brings about the curious paradox that despotic regimes can long look to be remarkably stable, only to fall with incredible speed once they do start to crumble. Once the spell of fear has been broken, the guiding principle of the regime begins to collapse and, since despotisms are systems of indivisible power, once that concentration of power is breached then the rest follows swiftly. Despotism, in modern parlance, has a single point of failure: the despot himself, the personalized epitome of unmoderated rule. The despot, then, has an interest in generating fear not just because he happens to be a bad character or that he has innately cruel dispositions but because he is fearful himself. This issue—not so much of the fear of power on the part of the subordinates of power, but the fear experienced by those in power themselves—was actually more developed by Montesquieu in The Persian Letters than in The Spirit of the Laws. But the two analyses are not inconsistent.

It could be called ruling fear. Montesquieu’s great predecessor, Montaigne, perhaps put it best: “What is it that makes tyrants lust for blood? It is their worries about their own safety and that when they fear a scratch their cowardly minds can furnish them with no other means of security save exterminating all those who simply have the means of hurting them, women included.”⁴⁴ Montesquieu’s Persian Letters explores the logic of ruling fear—paranoia might be as adequate a term—in a satirical way; for Montesquieu the seraglio was a metaphor for despotic rule generally.⁴⁵ Uzbek, the book’s protagonist, leaves his seraglio to go and tour around Europe. Without his constant presence instilling fear in his wives, chaos and rebellion break out amongst them—the entire system of power is then in peril. Uzbek is portrayed by Montesquieu as by nature a reasonable, gentle individual, but is by now in abject fear of losing his authority, the entire basis of his rule. Only now does he realize that he must act properly like the despot he actually is, whether he likes it or not. “This letter gives you unlimited powers over the entire seraglio,” he writes to his chief eunuch, “Your commands have as much authority as my own. Let fear and terror be your companions; go with all speed to punish and chastise in room after room; everyone must live in dread, everyone must weep before you. Interrogate the whole seraglio, beginning with the slaves.”⁴⁶

The lesson is clear. After all, if those in power are fearful they are likely to be even more fearsome themselves. There is an escalating spiral here: fear breeds fear breeds fear. And the more he is alone in power the more the despot, so to speak, becomes an individual, one wholly “unmoderated” by powers other than his own capriciousness. Thereby, the more personal virtue would count as an asset, the less likely it is actually to be so. Individuals in personalized rule tend, so far as Montesquieu is concerned, to be erratic and mercurial, just as are popular passions when left ungoverned; and as with all passions, though not necessarily bad in themselves, they need constant, ongoing moderation.

Now, of course the whole “classical” literature on virtue was designed just for this purpose: to limit the passions so as to restrict the individual, capricious vicissitudes of power. When Cicero wrote of the duties of those who govern, the abiding concern was to limit the capriciousness of rule.⁴⁷ Virtuous persons were required to moderate themselves. This idea of the personal virtue of the ruler was the prevalent model of moderation, at least until challenged by writers such as Machiavelli; indeed The Prince has at times been regarded as a kind of “answer” to, or even a satire of, Cicero, even though there are as many points of comparison as there are contrasts between the two thinkers.⁴⁸ At any event Machiavelli was important to Montesquieu precisely because he held up a mirror to the real as opposed to the idealized propensities of princes; reading Machiavelli should, as Montesquieu put it, cure us of our Machiavellianism (389). Montesquieu’s own basic response was, do not merely accentuate the positive in terms of personal character—virtue—but provide “depersonalized” institutions and mechanisms that restrict the bad propensities of rulers, that will restrict ruling fear itself.⁴⁹ These mechanisms—the balancing of powers each against the others—are moderate not in the “substantive” sense of being intrinsically moderate but in a dynamic sense; moderation here is a process rather than a substance—it is to do with powers moderating each other rather than simply being, in essence, “moderate.”

Montesquieu, Realism and Modernity

Perhaps there is something deeply “modern”—as opposed to classical, Ciceronian—about this conception.⁵⁰ Or rather I may put this a little more paradoxically; it is that Montesquieu’s conception of moderation resonates with the modern without being substantively modern in itself. Perhaps the best way to see this is in terms of a certain kind of realism that Montesquieu pioneered.⁵¹ As noted earlier, he is not really “against power”; he treats it, rather, with a perpetual suspicion—especially with regards to its possibilities for concentration. What is important is not to get rid of power but to subject it to balance, to divide it up—indeed to “moderate” it. Of course, Montesquieu’s is a practical, pragmatic, sceptical realism not a version of remorseless, ideological Realpolitik. It is a realism of moderation not of outright cynicism. In other words, this sort of realism is basically sceptical of power; not either addicted to it (Realpolitik) or allergic to it (idealism). Montesquieu recognizes that power is a basic and necessary feature of human life, but the solution provided by the principle of moderation is not to rid life of power but to use power against power: to make power divisible.

Now, Montesquieu’s own substantive concerns here were particularly with his own class—invoking the term broadly—the aristocracy. Indeed, in monarchies he associated moderation above all with the presence of aristocracy itself (51). But there are lessons for modern post-aristocratic, more egalitarian politics here even though—and perhaps even because—Montesquieu was not himself a “modern.” Given the thèse nobiliare, perhaps Voltaire was right in seeing Montesquieu’s treatment of despotism as in fact a kind of “satire” of his own times.⁵² So this is not moderation in the modern, liberal sense; or at least it is at best an “aristocratic liberalism” where the role of the aristocracy in the parlements is to balance monarchical power.⁵³ Aristocracies moderate monarchies. But in doing so aristocracies also serve to moderate the fear or the solitary vulnerability of the monarch himself. Monarchs left to their own devices are dangerous; in moderating them, other powers also moderate their exposure, their vulnerability, their own fear.

Where this aristocratic or rather nobiliste conception nonetheless points, so to speak, towards the modern is that Montesquieu has actually hit here upon a general problem for anyone reflecting on the nature of power—how those in power who “abuse” that power perceive themselves, the extent of their ruling fear. And indeed there is a general principle here. It is in the interests of everyone in a State—including a democratic State—that those who have power do not have personal kinds of fear—ruling fear—for themselves, that they do not aim to hold onto power at all costs. Power on this sort of conception has to be understood in terms of “office”; those who govern “hold” office and are not personally identified with their position such that they need to hold onto it at all costs, for that would be, in Montesquieu’s terms, something like despotism.⁵⁴

Under moderate styles of government, then, person and office have to be held separate. Institutional arrangements, in any kind of State, that restrict the varied kinds of ruling fear can actually serve also to moderate propensities to cruelty and injustice. This is not necessarily about such devices as granting legal kinds of immunity to those in power—devices which can serve, on the contrary, to exacerbate sorts of tendencies that might be associated with “bad characters”—but devices which restrict power, that allow for the relinquishing of power in a peaceful, stable manner. One obvious example from liberal democracies is the expectation for periodic elections. As Karl Popper liked to claim, the function of democratic elections is not so much to establish governments as to get rid of them: “For it does not matter who rules so long as it is possible to get rid of the government without bloodshed.”⁵⁵ Here the electorate is seen less as a source of “positive” sovereignty than as a “negative” or critical moderating power. And for the rulers themselves, it can also be preferable to be moderated in this way. It is at any event better to lose one’s office than one’s head. As Popper insisted, elections get rid of politicians without violence; they lower the political stakes, and in that sense they are an important—perhaps the most important—device of moderation in modern States, however inadequate those States might be. Authoritarians, autocrats, populists of various kinds—all these tend to begin by making war on the principle of elections, either by rigging them or by prolonging legitimate terms of office such that the role becomes personalized to them.⁵⁶

In democratic contexts, once those who rule feel that they have some kind of immunity from other powers, some kind of “right” to rule beyond the electorate, we approach the circumstances of unmoderated, indivisible power—indeed we approach, at whatever the distance, Montesquieu’s conception of despotism. In claiming that there is something “modern” about this conception, the point is not to sign up Montesquieu to the likes of latter-day political or other kind of ideology. Montesquieu was not a “modern” not just because he was ideologically and personally not so, but also in that his writing predates the era of industrial capitalism and in that he could have had no advance conception of mass democracy or even ideological liberalism. But he saw something very basic about the connections between fear and power, and not just with regard to those who are ruled, but with regard to those who rule themselves—ruling fear.

A further aspect of the realism invoked here is simply that this view of moderation has no utopian or idealized features whatsoever. Much of today’s liberalism for instance—particularly in the guise of what has become known as neoliberalism—is compromised by a certain kind of fundamentalism, or the sense that the “free market” is the only principle that is required. But the single unmoderated principle of the free market can become in effect a despotic principle. There is little scope for moderation here in Montesquieu’s sense. For neoliberalism the market serves as the basic principle of all political life. Montesquieu himself saw markets very differently. For him the world of commerce was itself a moderating power (337–97). Commerce moderated, for instance, the tendencies of Princes to be capricious about power. Commerce for Montesquieu was not just a question of interests versus passions; it could itself be a passion capable of moderating some of the more malign passions of political life.⁵⁷ For one thing, it could be a harmless distraction, typically employing peaceful rather than violent methods; for another—the famous doux commerce thesis—it could soften manners generally; and finally it could serve as an incentive to Princes to moderation, since rulers of necessity had to accommodate themselves to economic demands that were wider than their own private interests (342, 348–49, 389).

For Montesquieu nothing is really of political value in itself but for its effect on the whole system of power. As already noted, there is no sense that there could ever be a telos to history, some kind of final state of affairs to which everything tends. There is certainly no telos of the moderate State. Despotism of whatever sort and to whatever extent is more likely; and the emergence of moderate government is often, so Montesquieu insists, just due to chance rather than art (63). Montesquieu’s realist conception of moderation, then, works with political circumstances as they are. And his conception of depersonalizing power that is so important a component of his notion of moderation is not in any way a conception of “rationalizing” politics towards some ultimate enlightened State. Certainly, the less government is down to individual whim or caprice the better; but this does not mean that it should necessarily be “legal-rational” or merely rule-based, only that the rule of law balances other powers (78). The rule of law is one part of moderation, but the law too needs to be moderated. Montesquieu’s kind of depersonalization is not, then, Max Weber’s disenchantment or rationalization any more than it is Francis Fukuyama’s neoliberal end of history. Power is moderated away from individual caprice as much by individuals—and collectivities—as by impersonal forces, rules, or laws. But this is always provisional. A continuous alertness is necessary. Moderation is not a state but a process.

Conclusion

The leading image of moderation outlined here, then, is of one power or powers “moderating” another power or powers. It is, as it were, impersonal or institutional moderation as opposed to the model of moderation as virtue.

But, as I hope to have made clear, Montesquieu’s conception is not one that would be, as Weber put it, “without regard for persons” at all. It is not that persons should not govern—his is not a recipe for universal bureaucracy—but that their passions should be moderated by other powers when they do. So it would be an error to overstate my initial contrast between moderation as virtue and moderation as government. For Montesquieu it is more as if the former can only be approached so long as consideration is given to the latter. As Shklar insists, he does emphasize the depersonalized conditions of government, but this does not mean that he places no stock on personal moderation as well. Rather, a moderate polity will lead to moderation itself as a political virtue: “A good legislator takes a middle way … ” (93).

Acknowledgement

I wish to thank the three editors of this special issue and two anonymous referees for their scholarly and professional criticism, help and encouragement in improving this article. I also owe a huge debt to Gabriel Osborne for his expertise in the history of political thought and insight into Montesquieu’s place in the history of moderation.

Disclosure statement

No potential conflict of interest was reported by the author.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

The early part of this work was supported by the Leverhulme Trust under grant MRF-2015–18.

Notes on contributors

Thomas Osborne

Thomas Osborne is Professor of Social and Political Theory at the University of Bristol. He is the author of Aspects of Enlightenment: Social Theory and the Ethics of Truth (University College London Press, 1998), and The Structure of Modern Cultural Theory (Manchester University Press, 2008), and of many articles on social and political subjects in academic journals. His most recent research has been directed at questions in political ethics, liberalism and the analysis of power for which he was awarded a Major Research Fellowship by the Leverhulme Trust, UK.

Notes

^1. Cicero, On Duties, 142.

^2. Carrese, Democracy in Moderation, 22–49. This is an excellent text that puts Montesquieu in the tradition of French political liberalism. The best modern text on moderation and Montesquieu, however, remains chapter 2 of Craiutu, A Virtue for Courageous Minds. The current article—influenced by the work of Judith Shklar—stresses the difference between moderation as virtue and as government (though this is also a guiding theme of Craiutu’s Faces of Moderation), entailing emphasis on what is here called the “negative” or critical character of Montesquieu’s notion of moderation, and emphasis on Montesquieu’s particular brand of political “realism.”

^3. Montesquieu, Spirit of the Laws. Hereafter parenthetical page numbers cited in the text refer to this work.

^4. Carrese, Democracy in Moderation, 47.

^5. Tocqueville, Democracy in America, 816–21; Osborne, “Civil Society.”

^6. Osborne, “Power Degree Zero.”

^7. Althusser, Politics and History, 75. Althusser’s text, in spite of its ideological leanings, remains one of the best—and most evocative—texts on Montesquieu’s view of power and government.

^8. Malcolm, Useful Enemies; Osterhammel, Unfabling the East. Classic texts on the intriguing—and complex—history of the concept of despotism include Koebner, “Despot and Despotism”; Richter, “Despotism”; and Venturi, “Oriental Despotism.” See also Krause, “Despotism,” and Osborne, “Power Degree Zero,” and for a contemporary perspective, Keane, The New Despotism.

^9. Hennis, Politics as a Practical Science, 211.

^10. Voltaire, Political Writings, 87–100.

^11. Groethuysen, Philosophie de la Révolution, 126–27. Groethuysen’s short text on Montesquieu remains a classic of interpretation.

^12. Shackleton, Montesquieu, 238.

^13. Montesquieu, Persian Letters, 190.

^14. Montesquieu, My Thoughts, 437.

^15. Schmitt, Dictatorship, 84.

^16. Montesquieu, Oeuvres Completes, 875–74.

^17. Shackleton, Montesquieu, 300–301.

^18. See also Montesquieu, My Thoughts, 257–58.

^19. Schmitt, Dictatorship, 84.

^20. Shackleton, Essays on Montesquieu, 11.

^21. Schmitt, Dictatorship, 81–83.

^22. In this sense Montesquieu’s perspective could be understood within the terms of Wolin’s idea of “inverted totalitarianism” in Democracy Incorporated. See also Paul Rahe, Soft Despotism, Democracy’s Drift; and Osborne, “Power Degree Zero.”

^23. See on this Jeremy Waldron’s valuable Political Political Theory, chap. 3.

^24. Walzer, “On Negative Politics.”

^25. Shackleton, Essays on Montesquieu, 5 note 12 for this quotation. Shackleton quotes Voltaire’s Zadig as voicing a similar view: “Tout est dangereux ici-bas et tout est nécessaire.” Later, another thinker in the French liberal tradition will claim that the issue is not that everything is bad but precisely that “everything is dangerous”—Michel Foucault.

^26. Berlin, Liberty, 166–217.

^27. Shklar, Ordinary Vices, 10.

^28. Ibid., 226–49.

^29. Elster, Making Sense of Social Behavior, 427.

^30. Shackleton, Montesquieu, 300–301.

^31. See Hume, “Of the Independency of Parliament,” 26.

^32. Hamilton, Madison, and Jay, The Federalist, 235.

^33. Bagehot, English Constitution, 9.

^34. Hume, “Whether the British government,” 31.

^35. Shklar, Montesquieu.

^36. Shklar, Ordinary Vices, 214, 34–35.

^37. Ibid., 226–49.

^38. I am grateful to an anonymous referee from this journal for making this point.

^39. Hardin, “Liberal Distrust”; Shklar, Ordinary Vices, 238.

^40. Quoted in Elster, Securities Against Misrule, 166.

^41. See on this the comments of Stephen Holmes in Rosenblum, ed., Liberalism and the Moral Life, 241.

^42. Montesquieu, My Thoughts, 480.

^43. Jaucourt, “Crainte,” 428; Shklar, Montesquieu, 84–85.

^44. Montaigne, Complete Essays, 792; but cf. Carrese, Democracy in Moderation, 35.

^45. On this, see Silverman’s “Refléxions sur une métaphore.”

^46. Montesquieu, Persian Letters, 271.

^47. Cicero, On Duties.

^48. Giorgini, “Cicero and Machiavelli.” Of course, Machiavelli can be appropriated—even for something like the liberal tradition—in different ways; see, for example, Osborne “Machiavelli.” For Machiavelli himself as a thinker of moderation, see now the illuminating discussion in Mithen, “Politics as Moderation.”

^49. Of course, much might be said, in this connection, also of Montesquieu’s near-contemporary, David Hume and the notion of politics understood as a “science”—but this would require another treatment.

^50. As discussed in various contributions to Carrithers, Mosher, and Rahe, eds., Montesquieu’s Science of Politics.

^51. Closely linked with what Carrese in his discussion in Democracy in Moderation characterizes as Montesquieu’s complexity, 23, 29, 47.

^52. Voltaire, Political Essays.

^53. Manent, An Intellectual History.

^54. See on this Hennis, “The Idea of Office and the Concept of Democracy.”

^55. Popper, “On the Theory of Democracy,” 94.

^56. Cheeseman and Klaas, How to Rig an Election.

^57. See Hirschman, Passions and the Interests.