The Arbitrary

Norbert Campagna


1Whereas the substantif arbiter has no negative connotation, the adjective arbitrary is often equivalent to condemnation of the decision to which it is applied. We find that negative connotation in Montesquieu, who opposes the arbitrary to justice, reason or yet to nature.

2It is the medieval theologians, like William of Ockham, who engage Western thought on the path of voluntarism. They conceive God as able at any time to change the world order, the divine will not being bound by any sort of normativity that might have existed before it. With the sacralization of political power in the seventeenth century, this theological conception is propagated into politics, as was recently again shown by Arlette Jouanna. Montesquieu takes clear exception to Ockham’s theological voluntarism. Thus he asserts from the very first chapter of The Spirit of Laws that “creation, which seems to be an arbitrary act, supposes rules as invariable as the fatality of the atheists” (“la création, qui paraît être un acte arbitraire, suppose des règles aussi invariables que la fatalité des athées”, EL, I, 1). If God wants there to be a world, he must create and govern it according to those rules. The laws of Newtonian physics are not arbitrary creations of God, but they are constitutive of all the possible worlds that God could have created. In this framework, there is no place for miraculous interventions that bracket these laws, for, to take the example of the sun which is supposed to have stopped to permit Joshua to carry the victory, “if the sun really stops […] there is no more movement, no vortex, no sun, no Earth, no men, no Jews, no Joshua” (Pensées, no. 22, prior to 1731). However, Montesquieu lets it be understood that God – “heaven” – has an extraordinary power, but does not abuse it and “acts with as much order and rule as if its power were dependent” (Pensées, no. 640, prior to 1734).

3What goes for the physical laws goes also for the moral laws, which relate back to “relations of equity prior to the human law that establishes them” (ibid.) – human or divine law, little does it matter. Contrary to Ockham’s God who can overturn all values and make blasphemy into a moral duty, Montesquieu’s God is not the master of the moral world. But if God himself cannot govern the world by arbitrary decisions, a fortiori human governments will not be able to govern their state by such decisions. But arbitrariness is a constant phenomenon in the political world.

4In a passage of the Pensées with contractualist overtones, Montesquieu evoked the birth of the first governments: “Chance and the mentality of those who came together established as many forms of government as there were peoples, all good since they were the will of the contracting parties.” (Pensées, no. 1267, prior to 1739). Contrary to other thinkers who judge that some forms of government are more in conformity with nature than others – monarchy being often considered as the most natural and thus also the best, since it reflects God’s government of the universe – Montesquieu assumes that all forms of government are of equal validity provided they are founded on the consent of those who establish them. In that sense, the initial establishment of governments was arbitrary. But “what was arbitrary has become necessity […]” (ibid.). Once a form of government is established, it imposes itself on the “mentality” of those subjected to it, and these therefore cannot change it arbitrarily. In this context, Montesquieu rejects the hypothesis of unanimous consent of all individuals to pass from one form existing of government to another. He does not, however, absolutely exclude all change of government, but he inscribes it in a legal frame: “It was necessary that all the changes that took place in the established laws be an effect of those established laws […] and the people itself could take back its authority only when that was allowed by civil or natural law” (“Il a fallu que tous les changements arrivés dans les lois établies fussent un effet de ces lois établies […] et le peuple même n’a pu reprendre son autorité que lorsque cela lui a été permis par la loi civile ou naturelle”, ibid.). If there exists a “right to revolution”, in other words to change the form of government, its exercise cannot be founded on the fact that any form of government is natural and that men are thus free to set up the government they wish. The very existence of a form of government creates a sort of normative fact that puts an end to the initial arbitrariness.

5International relations are a domain in which it seems the arbitrary ought to occupy a position of choice, for they still seem to embody the state of nature, that is, a state with no laws because it has no sovereign. In The Spirit of Law, Montesquieu condemns certain jurisconsults who “have plunged into the arbitrariness” and affirmed that a conqueror could do what he wanted with the conquered people, notably exterminate it (EL, X, 3). There will be “streams of blood that will flood the earth” if in the framework of international relations, especially of the right of war, we base ourselves “on arbitrary principles of glory, propriety, utility” (EL, X, 12). To avoid these streams of blood, Montesquieu judges that where war is concerned we must hold to “rigid justice” (ibid.). As far as ius ad bellum is concerned, that means that a war should be undertaken only it if is the sole means of self-preservation. Just as individuals, states also have a natural right to preserve themselves in the face of those who wish to destroy or conquer them. That natural right still does not imply an unlimited right. As for what is ius in bello and ius post bellum indeed, it is not permitted to kill someone who no longer represents any danger. Where war is concerned, the arbitrary is limited by four types of law: the law of nature, the law of natural insight, the law that forms political societies, and the law derived from the thing itself (ibid.). Not to respect these laws is to open the door to a continual war between states and, in the long run, to the destruction of the political world. International law (droit des gens), which includes the right of war, “is naturally founded on the principle that the various nations must do each other, in peace, the most good and the least harm possible, without prejudice to their true interests” (“est naturellement fondé sur ce principe, que les diverses nations doivent se faire dans la paix le plus de bien, et dans la guerre le moins de mal qu’il est possible, sans nuire à leurs véritables intérêts”, EL, I, 3). Naturally and not arbitrarily. Even the fact of considering one’s own true interests ahead of another’s interests does not imply an arbitrary will to power, but the natural right to self-preservation.

6The arbitrary is also found within states, and notably at the level of the legislative power. Montesquieu mentions “states where the laws are nothing, or are but a capricious and transitory will of the sovereign” (“des États où les lois ne sont rien, ou ne sont qu’une volonté capricieuse et transitoire du souverain”, EL, XXVI, 2). In such states law is “a pure act of power” (EL, XIX, 14), the sovereign’s simple will being a sufficient reason for the law: Sic volo, sic jubeo, stat pro ratione voluntas. There you have the essence of legislative arbitrariness. The legislator’s will need not be submitted to any kind of “prior relations of equity” or, more generally, to a normativity that extends beyond it. The legislative act creates the just, even the equitable. But according to Montesquieu, “a thing is not just because it is the law; but it ought to be law because it is just” (Pensées, no. 460). This represents a critique of Hobbes, who asserted in Leviathan that justice does not exist independently of the sovereign’s laws. But Hobbes nevertheless allowed the existence of an equity independent of those laws.

7Legislative arbitrariness can manifest itself in several domains. As for penal laws, Montesquieu observes: “It is the triumph of liberty when criminal laws draw every penalty from the particular nature of the crime. All arbitrariness ceases; the penalty does not come down from the legislator’s whim, but from the nature of the thing; and it is not man who does violence to man” (“C’est le triomphe de la liberté, lorsque les lois criminelles tirent chaque peine de la nature particulière du crime. Tout l’arbitraire cesse ; la peine ne descend point du caprice du législateur, mais de la nature de la chose ; et ce n’est point l’homme qui fait violence à l’homme”, EL, XII, 4). In this way Montesquieu justifies the lex talionis, which would have “an intelligent being who has harmed another intelligent being should deserve to receive the same harm” (EL, I, 1). Penal arbitrariness is placed in relation to the domination of man over man and violence, whereas the refusal of penal arbitrariness is placed in relation to freedom. Man is not free if he is subject to another’s will, but he is when he is subject to natural relations.

8Still on the subject of penal law, Montesquieu mentions the crime of lèse-majesté, a “vague” (EL, XII, 7), that is to say ill-defined crime, “arbitrary” (Romans, xv, OC, t. II, p. 199), and even “more arbitrary […] when it is a matter of indiscrete words (EL, XII, 12). If political freedom is defined as the opinion each person has of his security, not knowing whether what one says constitutes a crime of lèse-majesté or not saps that freedom. The penal laws must be as precise as possible and must enumerate the cases that fall under them, while avoiding vague expressions that cause them to “revert to arbitrariness” (EL, XXIX, 16).

9Just as he must know the precise nature of the crime, the accused must also know precisely the penalty he incurs if he commits the crime. When the laws abandon the fixation of penalties to the will of the judges, the accused fear, like the subjects of feudal lords who could often freely set the fine, “the events of the trials” (EL, XXVIII, 35), and they lose any sense of their own security.

10Fiscal law is another field where arbitrariness can be exercised. After noting that “nature is just toward men” insofar as it rewards men in proportion to their efforts, Montesquieu accuses political power of raising disproportionate taxes: “But if an arbitrary power takes away nature’s rewards, one reverts to distaste for labor, and inaction seems to be the only good” (“Mais si un pouvoir arbitraire ôte les récompenses de la nature, on reprend le dégoût pour le travail, et l’inaction paraît être le seul bien”, EL, XIII, 2). Besides being unjust, exorbitant taxes destroy the economy by discouraging labor, the source of individual and collective wealth. Among the numerous measures which Servius Tullius, sixth king of Rome, took in favor of the people, Montesquieu mentions that he “removed arbitrariness in taxation” (Pensées, no. 1479, prior to 1740). Opposed to this fiscal arbitrariness we find the idea of necessity. Montesquieu mentions a supposed charter of Louis the Headstrong in which the king “renounces imposition of any taille or aide without evident necessity or evident utility. But would that not have been vain if he had been the sole judge of both?” (“renonce à imposer aucune taille ou aide sans une évidente nécessité ou une évidente utilité. Or cela ne serait-il pas vain s’il avait été seul juge de l’un et de l’autre?” Pensées, no. 1184). Evidence and necessity must be recognized by the estates, which represent those who must be subject to the taille or aide. Royal arbitrariness is thus counterbalanced by the consent of the realm.

11Still on the subject of fiscal law, Montesquieu condemns taxes on “various clauses in civil contracts” (EL, XIII, 9). Laws or regulations introducing such taxes are subject to interpretation, and the person of whom they are required does not have the necessary knowledge to defend himself against the allegations of the officer who deducts the tax. In so doing, the officer, who interprets the law or regulation, “exercises an arbitrary power over people’s fortunes” (ibid.). It is he and not the text that decides whether someone must pay, and how much.

12To avoid total arbitrariness, a state needs set laws. The plebeians called for them in Rome, “so that judgments would not be the effect of a capricious will or an arbitrary power” (EL, XI, 15). Wherever the legislator does not limit power in proclaiming laws, people must consider themselves fortunate if there are sacred books, as in the despotic states of the East: “The religious code supplements the civil code, and fixes the arbitrary” (EL, XII, 29). That does not mean that arbitrariness cannot also find its place in these codes, for beside holy precepts “entirely based on an eternal reason”, there are those that “are purely arbitrary and are rather a sign of religion than the religion itself, and these are the ceremonial ones” (Pensées, no. 205). In Montesquieu’s view, it is “necessary to society that there be something set” (EL, XXVI, 2). The despot’s arbitrariness must be tamed; he must be forced to “follow some order, and suffer some rule”, or else he would lose his power (EL, VIII, 10). But at the same time, a great empire, such as a universal monarchy in Europe, requires that “the law be on a single head, in other words endlessly changing, like the accidents that always proliferate in a state in proportion to its size” (“la loi soit dans une seule tête, c’est-à-dire, changeante sans cesse, comme les accidents qui se multiplient toujours dans l’État à proportion de sa grandeur”, Reflections on universal monarchy in Europe, viii, OC, t. II, p. 347, repeated in The Spirit of Law, VIII, 19). If God in his wisdom can govern the greatest of all empires solely by fixed laws, such is not the same for men.

13Laws alone do not suffice, but require a repository, an authority that reminds him who governs of them. If in the French monarchy this role is played by the parlements, it is the clergy that assumes it in monarchies that neighbor on despotism: “Where would Spain and Portugal be since the loss of their laws without this power [the clergy] which alone limits arbitrary power?” (“Où en seraient l’Espagne et le Portugal depuis la perte de leurs lois, sans ce pouvoir qui arrête seul la puissance arbitraire?” EL, II, 4). Montesquieu seems nevertheless to suggest that the despot cannot simply escape certain facts, “circumstances derived from climate, religion, the situation, or the genius of the people” (EL, VIII, 10).

14The existence of fixed laws – divine or human – and of a repository for these laws allows the elimination of the arbitrary only from a purely formal point of view. That suffices to guarantee society’s existence, but not necessarily anything more. To the non-arbitrary form must be added a non-arbitrary substance, and this presupposes that the laws that fix the arbitrary take into account the relations of equity that predate them.

15Just as legislative decisions, governmental decisions may be arbitrary. Thus Montesquieu sees the loss of monarchy “when a prince thinks he shows his power more by changing the order of things than in following it, when he takes natural functions from some in order to give them arbitrarily to others, and when he is more in love with his fantasies than with his desires” (“lorsqu’un prince croit qu’il montre plus sa puissance en changeant l’ordre des choses qu’en le suivant, lorsqu’il ôte les fonctions naturelles des uns pour les donner arbitrairement à d’autres, et lorsqu’il est plus amoureux de ses fantaisies que de ses volontés”, EL, VIII, 6). Some persons are by nature more apt than others to exercise certain responsibilities, and the prince must abstain from distributing responsibilities according to his whims, on pain of seeing the monarchy become corrupt.

16The arbitrary is also to be found on the administrative level, and notably in the application of the government’s decisions and the laws. While it is characteristic of despotic states that the law is nothing other than the despot’s will, Montesquieu judges that that will, “once known, must have its effect just as infallibly as a ball cast against another” (EL, III, 10). The idea is also found in the Pensées: “Arbitrary orders ought not to be carried out arbitrarily, and it is in the interest of an unjust prince that he who carries out his desires, even the most tyrannical of them, observe in the manner of carrying them out the most exact rules of justice” (“Les ordres arbitraires ne doivent point être exécutés arbitrairement, et il est de l’intérêt d’un prince injuste que celui qui exécute ses volontés, même les plus tyranniques, observe dans la manière de les exécuter les règles de la justice la plus exacte”, Pensées, no. 1898, prior to 1751). Despotism depends on the blind obedience of those to whom the despot has delegated the daily exercise of power, and leaving them the slightest right to judge if and how they are to apply the despot’s decision would be equivalent to sapping the bases of the government and more precisely its popular footing. That is why Montesquieu writes: “The people must be judged by the laws, and the grandees by the prince’s fantasy; the head of the least subject must be secure, and that of the bachas always exposed” (“Il faut que le peuple soit jugé par les lois, et les grands par la fantaisie du prince ; que la tête du dernier sujet soit en sûreté, et celle des bachas toujours exposée”, EL, III, 9). Only the grandees should feel the arbitrary power, as at the end of Louis XI’s reign: “Arbitrary power arises and takes shape in an instant. At the end of this last reign, there was not a single lord who could be sure of not being assassinated” (“Le pouvoir arbitraire s’élève et se forme dans un instant. À la fin de ce dernier règne, il n’y avait pas un seigneur qui pût être assuré de n’être pas assassiné”, Pensées, no. 1302).

17The administration of justice is a domain lending itself to arbitrariness, and Montesquieu devotes many pages to the question. It is not new, and the debate can be traced back to Aristotle. In his Rhetoric he contrasts recourse to arbitrage and recourse to lawsuits (1374b 20). Whereas the jurors in a trial take the law as their guide, the task of arbiters is to make honesty prevail, which is defined as “the justice that is situated beyond the written law” (1374a 25) and for that reason is akin to equity (epieikeia). For certain acts, either the written law does not exist, or it exists but because of its consubstantial generality neglects certain particularities, and its strict application therefore can lead to verdicts which, though indeed in conformity with legal justice, are not commensurate with natural justice. The arbiters’ decision thus does not go against justice, but it seeks on the contrary to achieve justice more perfectly than the law.

18Montesquieu several times mentions arbitrage. In the compilation of Roman law known under the name of Collectio juris, he notes: “He is deemed arbiter who does in a sense the duty of judge seeking by his sentence to end the discussions of the parties, not the one who intervenes to try to accommodate them” (“Celui-là est censé arbitre qui fait en quelque façon le devoir du juge voulant par sa sentence finir les discussions des parties non celui qui intervient pour tâcher de les accommoder”, OC, t. XI, p. 59). The judge and the arbiter have in common to settle the case, which distinguishes them from simple mediators. But whereas the judge decides in function of the letter of the law, the arbitor frees himself from that letter or even decides in the absence of any letter of the law. Basing himself on the Digest, Montesquieu affirms that the parties must submit to the arbiter’s decision, “whether it is just or unjust” (ibid., p. 60), unless the arbiter makes a decision explicitly against the law (ibid., p. 61). And he says of this sentence of the arbiter that it must be “fixed and certain” (ibid., p. 60), two characteristics which, in Montesquieu, are often attributed to the laws.

19Arbiters (in whom Montesquieu takes a particular interest in 1745, when he draws an extract from Cardin Le Bret’s Ordo perantiquus judiciorum civilium) are also mentioned several times in The Spirit of Law. There he takes up bishops who were “arbiters of the judgments” in times of the “first race” (EL, XVIII, 31), a sign of the clergy’s authority in those distant times. A few centuries later, the parents of a dead person who had made no will had to obtain “from the bishop that he name, along with them, arbiters to decide what the deceased should have given had he made a will” (“de l’évêque qu’il nommât, concurremment avec eux, des arbitres, pour fixer ce que le défunt aurait dû donner, en cas qu’il eût fait un testament”, EL, XXVIII, 41). As this last example shows, the arbiter is the one who decides in the absence of a precise written text or of any text at all.

20If in a well-constituted republican government the laws are precise and cover every possible case, avoiding recourse to arbitrage, that does not apply to monarchies, where the law, given the complexity of the social structure is not always precise: “[…] the judges there assume the manner of arbiters; they deliberate together, share their thoughts, come to agreement; one man’s opinion is modified to make it conform to another’s […]” (“[…] les juges [y] prennent la manière des arbitres ; ils délibèrent ensemble, ils se communiquent leurs pensées, ils se concilient ; on modifie son avis pour le rendre conforme à celui d’un autre […]”, EL, VI, 4). For these judges, it is not enough just to have eyes to read the letter of the law, but they must discern the spirit of the law, which must refer to the principle of monarchical government and its fundamental laws. The judges’ decision is therefore not taken in a complete vacuum, but with a background of data tied to the very essence of the government and society. The judges’ arbitration in a monarchical state is imposed on the parties, for the judges are not simple arbiters, they are content to adopt their “manner”.

21Nor are arbiters foreign to aristocracy, where “the nobles’ disputes must be decided promptly; otherwise contestations between persons become contestations between families. Arbiters can end the lawsuits, or prevent them from arising” (“les différends des nobles doivent être promptement décidés ; sans cela, les contestations entre les personnes deviennent des contestations entre les familles. Des arbitres peuvent terminer les procès, ou les empêcher de naître”, EL, V, 8). In this type of situation, the arbiters play a pacifying role and their aim seems to be less to decide in function of a natural justice superior to legal justice as to see that a dispute finds an acceptable solution, or simply comes to an end. Moreover, arbiters make up for the slowness of justice and allow the parties to find a compromise which, if it is not ideal, at least spares them the drawbacks of justice.

22It is in the framework of penal justice in a moderate government that arbitrariness is most firmly condemned by Montesquieu. Judgments ought to be only “a precise text of the law” (EL, XI, 6), and the judges “the mouth that pronounces the words of the law” (ibid.). The judges in question here are the jurors, who judge only the fact. The law names the fact as criminal, and the jurors must observe whether the accused has committed a crime or not. It is not for them to presume anything at all, for “when the judge presumes, the judgments become arbitrary; when the law presumes, it gives the judge a fixed rule” (“[l]orsque le juge présume, les jugements deviennent arbitraires ; lorsque la loi présume, elle donne au juge une règle fixe”, EL, XXIX, 16). In this sense, the law must clearly define crimes of lèse-majesté so as to avoid the judge’s having to evaluate by himself whether an act constitutes such a crime. If it were up to him to evaluate, he would be transformed into (co)legislator. But the power to judge must be disassociated from that of legislating – outside of the three famous exceptions (EL, XI, 6): “If it were combined with the legislative power, power over the life and liberty of the citizens would be arbitrary; for the judge would be a legislator” (“Si elle était jointe à la puissance législative, le pouvoir sur la vie et la liberté des citoyens serait arbitraire ; car le juge serait législateur”, EL, XI, 6). As here it is solely a question of the judging of fact, Montesquieu does not have to ask himself what a judge should do if he is confronted with an arbitrary law, in other words a law that violates the relations of equity antecedent to human laws. His role is not to state the right (droit), but to determine the fact.

23This question arises nevertheless for the judge of rights (droit). It retained Montesquieu’s attention at the time when he was compiling the Institutions of Justinian and other works of jurisprudence. Thus he translated a passage of the Digest (i, 3, 18): “The laws must always be interpreted favorably” (Collectio juris, OC, t. XI, p. 1). But elsewhere he copies textually an extract from the addition to the Gloss Cautum sit: “The judge must not distance himself from the written right (droit) in favor of unwritten equity unless a specific principle orders him to do so” (Latin original: judex non debet recedere a jure scripto propter aequitatem non scriptam nisi speciali principio mandato; Collectio juris, OC, t. XII, p. 580). Thus the judge must stick to the written law, even if he thinks that in the case he has before him the application of that law leads to an inequitable verdict. Only if he receives an order to judge in terms of equity can he do so. He returns to this question when mentioning rescripts: “For as the judges by Roman law could not judge by what seemed to them equitable or just but by the letter of the laws, in the cases where they thought they should reject that maxim they would consult the prince […]” (“Car comme les juges par le droit romain ne pouvaient point juger selon ce qui leur paraissait équitable ou juste mais selon la lettre des lois dans les cas qu’ils croyaient devoir écarter cette maxime ils consultaient le prince […]”, ibid., p. 835). This opens a small space for the judges to be arbitrary, for they will have to decide when they address the prince. It is up to them to appreciate whether the strict respect of the letter of the law could lead to an inequitable verdict. In the Pensées Montesquieu evokes the case of excessively severe penalties and rejects the idea of a radical change. One must proceed “imperceptibly”, notably by the “diminution of penalties in the most favorable cases, leaving that to the arbitrage of the judges” (Pensées, no. 1897, a passage which occasioned an addition to the posthumous edition of The Spirit of Law, VI, 13, §7).

24A place of choice for arbitrariness is judgments dealing with magistrates. Montesquieu mentions on this subject the ephors, who judged “arbitrarily, there being no laws to direct them” (EL, VI, 3). Five ephors, elected annually, exercised a control over politics, justice, finances and administration, and it fell to them to judge whether the magistrates in charge of those domains acquitted themselves well of their task. They could, should the case arise, charge the king a fine. They themselves had no one to answer to. Montesquieu condemns this type of institution, but he approves of the idea of a control of the executive power by the legislative power (EL, XI, 6). He nevertheless excludes allowing the king to be the direct object of such a control, for “his person ought to be sacred, as being necessary to the state, so that the legislative body will not become tyrannical” (“sa personne doit être sacrée, parce qu’étant nécessaire à l’État pour que le corps législatif n’y devienne pas tyrannique”, ibid.). It is to the ministers that poor execution of the laws will be imputed, and “they can be sought out and punished” (ibid.). By whom? For what acts, exactly? By virtue of what laws? So many questions that Montesquieu leaves open, and with which our contemporary democracies are still confronted.


Extract of Ordo perantiquus judiciorum civilium by Cardin Le Bret, ed. Christian Cheminade, OC, t. XVII, 2014.

Aristotle, Rhetoric, ed. Pierre Chiron, Paris: Garnier-Flammarion, 2007.

Jacques Ellul, Histoire des institutions, XVIe-XVIIIe siècle, Paris: Presses universitaires de France, [1956]1999.

Jean-Marie Carbasse, Introduction historique au droit, Paris: Presses universitaires de France, 1999 (2nd edition).

Arlette Jouanna, Le Pouvoir absolu: naissance de l’imaginaire politique de la royauté, Paris: Gallimard, 2013.