Introduction

In colloquial language, the terms authority and power are often used interchangeably. However, they are two distinctive, though related, concepts in political philosophy. The difference between the two terms is more pronounced in natural law theory than in legal positivism. Authority is the right to issue commands and merits compliance by moral obligation or dutyFootnote 1. Power is the extent to which the command can be enforced, typically by physical coercion. To elucidate the subtle difference between authority and power and underscore the conflicts between authority and autonomy, consider the following example. When compelled to exit my vehicle because a robber is holding me at gunpoint, I concede that the robber has power over me because my life takes priority over the value of my car. However, I acknowledge no legitimate authority from the robber. In this case, my autonomy is being violated, for if I had the freedom of choice, I would not surrender my car to the robber, but insofar as the robber has power over me, I have no choice but to comply. In contrast, when a police officer commands me to step out of my vehicle because I was speeding, the officer has power over me since he or she carries a gun, but the officer also, at least putatively, has legitimate authority over me. However, the question is, since I am an autonomous individual, even if I have transgressed a law to which I didn’t personally consent, does the police officer truly have legitimate authority over me? I may as well reason that, if I had the choice of keeping on driving without any punitive consequence, I would have kept on driving given that I have no moral obligation to obey. If the officer does not have legitimate authority over me, whether his or her command is morally different from that of a robber is a salient concern. For a political law to preserve one’s autonomy, the foundation of its authority must not be derived from the mere fear of reprisal, but from one’s moral duty.

Fig. 1
figure 1

Illustrated relationship between HI and CI and first-order desire and second-order desire/volition (Created with Miro Online Whiteboard (Version 3.11.2).

RealtimeBoard Inc. https://miro.com/app/dashboard/).

Full size image

The conflict between autonomy and authority is illustrated by the idea of public reason, which can trace its roots to Immanuel Kant’s essay on What is Enlightenment, although it has a different meaning than what is popularized by John Rawls in contemporary political philosophy. The Kantian definition of public reason can be interpreted as the right to express one’s opinion to the entire public through readership viz., the right to the freedom of the press (Braeckman, 2008: pp. 285–306). This definition is to be distinguished from the definition adopted in this manuscript. Here, public reason is defined as the requirement for regulative principles (e.g., political laws) of society to possess at least some justification over whom they hold authority (Quong, 2018). The traditional view of the philosophy of law posits that political laws can be justified by natural law theory or legal positivism, although the third theory of law has also been proposed by Ronald Dworkin (Mackie, 1977, pp. 3–4). A detailed account of both justifications is far beyond the scope of this paper; for the present purpose, it is sufficient to consider the former justifies the political laws and obliges citizens by moral duty, while the latter does so by legal duty. I define legal duty as the duty of citizens, or more broadly, of residents, in a sovereign state, to obey the laws of the state through explicit (such as the oath taken during the naturalization process) or implicit consent (being a resident). If the duty is compromised, viz., one’s action violated the laws of the state, the doer will be subjected to fines, imprisonment, or punitive consequences enforced by the state. Moral duty is to be distinguished from legal duty, which is defined as: under the Kantian assumptions of rational beings, obeying moral laws derived from practical pure reasons. When moral duty is compromised, insofar as it is not reflected as a violation of political laws, one does not receive any materialistic punishment; however, they are stripped of autonomy and dignity fundamental to a rational being. The two forms of duty oblige citizens to obey political laws through different avenues, each with unique strengths and weaknesses philosophically. I aim to synthesize the two forms of duties, namely, natural law theory and legal positivism, to provide justifications for political laws.

In this essay, with references to H.L.A. Hart’s contemporary legal positivism theory, I will consider the legitimacy of political laws grounded on moral justification, namely, on the natural law theory, and how it can command subjects with legitimate authority while preserving their autonomies, mainly from the perspective of Kant in his Groundwork of the Metaphysics of Morals, as well Jean-Jacques Rousseau’s On the Social Contract in which he tackled the critical question, “how can we find a form of commonwealth, which will defend and protect with the whole common force and the person and goods of each associate, and in which each, while united himself with all, may still obey himself alone, and remain as free as before (Rousseau, 1999, pp. 54–55, translation modified)?” I will explore (1) the Kantian formulas and how the state has legitimate authority over the violation of Categorical Imperatives (CI) but debatable authority over disputes of Hypothetical Imperatives (HI), (2) unanimous direct democracy as the only theoretical solution to the conflict between authority and autonomy regarding political laws derived from HI, (3) the General Will (GW) as sharing elements from both legal positivism and natural law theory, (4) The emergence of the General Will (GW) from majoritarian direct democracy and the problem of the tyranny of the majority, and (5) the GW as a synthetic a priori concept analogous to the CI and how it can be used to solve the conflict between authority and autonomy practically if we accept a problematic assumption. In short, I will defend my thesis that the Kantian CI successfully justified a limited number of political laws, although the deontological system is not without defects. Rousseau offers an innovative solution to the conflict between autonomy and authority with the GW as the source of the legitimate authority of the state which is comparable to the CIs as the supreme moral laws that oblige every individual, but he falls short of solving it completely due to the key assumption being empirically unverifiable constructs. Nonetheless, by substituting the current legislative procedure with Rousseau’s GW, the current socio-political system of the U.S. may be improved.

Political laws based on CIs are the absolute authority over human actions

In the anarchistic argument, all states on earth, either currently or historically, are de facto, which have power (descriptive), but not legitimate authority (normative) over us (Casey, 2012). The de jure state should be normative (i.e., command by authority) in addition to being descriptive (i.e., command by power). The maintenance of autonomy is required for the authority to be legitimate. That is, a de jure state should have both power and legitimate authority by virtue of individuals autonomously consenting to the state’s laws, thereby creating a moral obligation to comply. As human beings living in a society, we are governed by moral laws on the one hand and political laws on the other. It may be helpful, for now, to consider the moral laws as the necessary natural laws, and the political laws as the contingent human laws, as differentiated in Thomas Aquinas’s Summa Theologica (Aquinas, 2005: pp. 463–472). Natural laws emanate from moral reasoning, while political laws aim toward common interests (Locke, 1980). However, a stringent sense of natural law necessites complete agreement of political laws with moral laws. In reality, the agreement between political laws and moral laws varies. The ones that have no moral foundation are not necessarily unjust. A relevant example lies in the directionality of driving, such as driving on the right side in the U.S. or the left side in the U.K. The directionality is a social convention that the citizens who consented to the social contract should obey for a more efficient society. These kinds of laws, though lacking moral foundation, are justified by the social contract and legal duty of citizens for a more efficient social order. The state exceeds the limit of authority when political laws contravene natural laws, as exemplified by Nazi Germany where political laws opposes natural laws but were enforced by the power of the state. We do not need to invoke the knowledge of political laws, which is limited for most people anyway, to perform every action while being law-abiding citizens. We know what is generally the right thing to do thanks to our moral compass, as Kant wrote in the Groundwork of the Metaphysics of Morals,

I do not need any wide-ranging acuteness to see what I have to do for my willing to be morally good. Inexperienced with regard to the course of the world, I just ask myself: can you also will that your maxim becomes a universal law? If not, then it must be rejected, not because of some disadvantage to you or to others that might result, but because it cannot fit as a principle into universal legislation [based on CIs] … It would be easy to show, with this [moral] compass in hand, [how human reason] is well informed in all cases that occur, to distinguish what is good, what is evil, and what conforms with duty. (Kant, 2012, p. 19, emphasis added)

Kant asserts boldly that our faculty for moral judgment is inherent, insofar as we do not value our moral duty to be less than immediate inclinations. Consequently, political laws emerge as the last line of defense for moral laws when the balance of judgment tilts in favor of inclination over moral duty. When the moral compass fails to guide our actions, the fear of legal repercussions acts as a deterrent to rightfully prevent us from committing atrocities. However, Kant emphasizes that our refrainment from actions such as murder should not stem from fear of punitive consequences but from the recognition of its inherent immorality and contradiction of our moral duties.

I would imagine that most people would be reluctant to murder a stranger for fun even if no punishment would ensue because it is contrary to our CIs. The CIs are derived from pure practical reason, independent of any inclination. The CI can therefore be a practical law that restrains a person morally. Within the societal context, CI based on moral duty is principally transformed into political laws that restrain us by legal duty. This transformation, however, is nuanced. Firstly, the reflection of CI in political laws is not universal. Laws prohibiting murder are directly aligned with CI, but laws enforcing speed limits do not involve CI. Secondly, moral duty can be further differentiated into perfect and imperfect duty (Trafimow and Ishikawa, 2012, pp. 51–60). Most political laws, even if based on CI, are grounded on perfect duty commanding one ought not to do somethingFootnote 2. On the other hand, political laws based on imperfect duties commanding one ought to do something are comparatively rare and often contingent on special circumstances. For example, the Emergency Medical Treatment and Labor Act (EMTALA) mandates healthcare professionals to perform lifesaving procedures on patients regardless of their ability to pay, whereas there exists no legal obligation for a layperson to perform CPR on a stranger in a medical emergency. Thirdly, Kant argues that an action has moral worth only when it is done from the goodwill itself, not merely doing what is “supposed to be done” while being motivated by other motives. This requirement is almost impossible to be enforced by political laws in practical settings. Despite these challenges, a limited subset of political laws grounded on perfect duty can arguably be justified by CI.

Another major distinction between moral and political laws is that legality does not necessarily imply morality. The historical legality of slavery underscores this disparity. Legal positivism, in the strict sense, might justify slavery given the socio-historical context, as it purportedly served the common interest of “citizen proper” rather than the common interests of all citizens because slaves were perceived as “sub-humans”. Nonetheless, from the perspective of natural law theory which held that political law is only justified when rooted in necessary and universal moral truth, such as the Kantian humanity formula, slavery would unequivocally be unjustified. Indeed, slavery as an illegitimate political law was eventually abolished for want of a necessary and eternal moral foundation. On the other hand, actions may be illegal but morally justified. For example, civil disobedience against racial discrimination was deemed illegal at that time, but it was the morally right thing to do. It is morally justified because it is grounded on the universalizability and humanity formulas that all people ought to be treated equally. From the standpoint of natural law theory, political law is authoritative only if it reflects “objective morals law knowable through reason” (Wolfe, 2003: p. 38). In other words, the authority of political laws is not inherent but originates from the moral laws embedded within. The schism between natural law theory and legal positivism, however, is not as drastic as one would presume. Some moral justification is necessary for the state to be instituted according to the social contract in the first place. All political laws must to some extent appeal to the natural law of peace and order among the subjects, as echoed by Thomas Hobbes (Hobbes, 2005: pp. 593–594). Even the proponents of legal positivism such as Hart granted that a legal system must have a minimum content of moral facts, viz., “a core of indisputable truth in the doctrine of natural law” (William, 1984: pp. 683–684). When the moral laws run afoul of the political laws, we must realize that the ultimate authority remains in the moral laws, although the power of the state can coerce us to obey the political laws. In this case, however, the state is no longer legitimate. This is why Aquinas and many other political philosophers have proposed that the citizens should have the right to petition against unjust laws that violate the fundamental natural laws, or when peaceful means fail, to overthrow an illegitimate government through revolution.

While the preceding argument highlights the susceptibility of political laws to criticism, the authority of the moral laws is above reproach. However, to fully examine the justification for moral laws and understand their role in preserving autonomy, it is essential to delve into the underlying principles. Kant provides possible answers through a rather counterintuitive claim. He wrote,

The ground of all practical legislation lies objectively in the rule and the form of universality, which [according to the universalizability formula], makes the rule capable of being a law [of nature]. Subjectively, however, it lies, according to the [humanity formula], in the subject of all ends, which is every rational being as an end in itself. From this follows the autonomy formula, the idea of the will of every rational being as a universally legislating will. (Kant, 2012, p. 43, translation modified)

The moral laws are justified, objectively, by the pure nature of the CI as the supreme principle of practical reason, creating a moral obligation for all rational beings (Kant, 2004). For the Kantian autonomy formula to apply, the laws that one wills must be universal without exception, as Kant wrote, “the principle of every human will as a will universally legislating through all its maxims… would be well fitted to being a CI [founded on no interest which the will obeys unconditionally]” (Kant, 2012, p. 44). In contrast to the contingent nature of political laws, which may be subjected to the dynamics of socio-historical perspectives, moral laws are universal and enduring, providing a stable foundation for them to be the governing principle of society. Their universalizability minimizes favoritism and contextual bias. The humanity formula reinforces this, necessitating the moral laws to uphold the intrinsic worth of every human being equally. Most importantly, moral principles stem from the self-legislation. As both the legislator and subjects of moral laws, individuals contribute to and obey their own creation of political laws. Thus, the self-imposed submission to moral authority through practical reason and moral duty preserves our autonomy.

Obedience to the CI is the sine qua non for an autonomous, moral, and rational agent. The individual has a moral duty to respect the universal law which is justified solely by goodwill as an end per se, irrespective of its consequence. Actions done from moral duty do not pursue specific ends since they are ends in themselves. The refusal to violate CI originates not from fear of punishment, but from the recognition that such violation contradicts the will of a rational being. In instances where CI is violated, autonomy is compromised. Following the autonomy formula, the conflict between the authority and CI can be resolved, for whatever actions that violate the CI morally justify the authority and power of the state to restrain that individual who failed to achieve his or her duty. The violation of moral duty signals the renouncement of autonomy, justifying the need for external restraints to maintain moral principles imperative to societal orderFootnote 3.

Political laws based on HIs have questionable authority over human actions

The difficulty concerns the justification of political laws instituted from HIs that originate from means-end reasoning. Kant wrote, “the HI represents the practical necessity of a possible action as a means for something else [desired end] such as happiness” (Ibid, pp. 28–29). Actions done from HI are not done from duty as an end in itself, but from an anticipated end, derived from desire and context rather than pure reason, which requires one to use his or her practical reasons to conceive the means to attain it. When the end fails to be achieved, the HI, therefore, is not justified. Most of our actions are done from the HI. However, the HI is inherently unpredictable and conditional in that it is justified by the purported end, unlike the CIs which are absolute and unconditional moral commands that are ends in themselves. Kant wrote,

The concept of happiness is a concept so indeterminate that even though everyone wishes to attain it, he can never definitely understand what it is that he truly wills. The concept of happiness is unexceptionally empirical. It is impossible for finite beings able to determine with complete certainty, what will make them truly happy because omniscience would be required for this. (Ibid, 31)

The contamination of the HI by a posteriori conditions divorces it from the infallible nature of a priori morality. This epistemological argument will be important in the latter part of the essay. The HI cannot serve as the foundation for practical laws, as Kant wrote, “the CI alone expresses to be a practical law, while all the others (HIs) may be called the principles of the will but not laws” (Kant, 2012: p. 33). In the strict sense, the state is not justified to enforce political laws upon citizens unless the laws are based on CI; however, the difficulty is that most political laws are based on HIs as most human actions are. Kant wrote, “Heteronomy admits only of HIs: I ought to do something because I want something else” (Ibid, p. 52). According to Kant, the political laws (principles of will) based on HIs cannot impose legitimate authority over the citizens because however they may appear to be justifiedFootnote 4, they create no duty, and hence no moral obligation for the subjects to obey. If one accepts the laws based on HIs, their autonomy is transformed into heteronomy, in which the individual wills their action from the direction of external objects (i.e., the state). I must admit that, if the laws are not exceedingly atrocious like slavery, perhaps it would just be begging the question merely for the sake of philosophical argument to examine whether a law like speed limits can withstand moral interrogation especially when considering the practical function of these laws within the broader societal framework. For the pragmatic legal positivists, “[if] it (political laws) leads to the realization of the anticipated end for the most part”, is sufficient to create a legal obligation for the citizens to obey, even if it is not completely fair and just, since “laws are morally relevant, but not morally conclusive” (William, 1984: p. 689). Therefore, even if a law is found on HI, insofar as part of the HI is relevant to promoting the common interest of the citizens, it would have been partially justified. John Rawls even goes so far as to argue that “the injustice of law is not, in general, a sufficient reason for not adhering to it anymore than legal validity legislation is a sufficient reason for going along with it… unjust laws are binding provided that they do not exceed certain limits of injustice” (Rawls, 1971: pp. 350–351). This pragmatic approach is reflective of the fact that not all laws necessitate moral scrutiny, and their justification may lie more in societal conventions, safety considerations, or other practical concerns.

Certainly, while it may not be a conventional philosophical inquiry to scrutinize laws like speed limits, it can be insightful to use these laws as examples and consider why they may not entirely align with moral duties to demonstrate the insufficiency of HI in justifying political laws. In light of Kantian moral reasoning, driving over the speed limit is a means to get to the destination faster (anticipated end). If it is guaranteed a posteriori that no accident will ever occur, then everyone should be allowed to drive above the speed limit. The end anticipated by the HI may not be achieved, however, since accidents may slow down or prevent the drivers from getting to their destination. Hence, we see that the HIs fail to justify speeding over the limit, but it also does not justify the speed limit as a political law. Even with the speed limit, there are accidents, although with a lower probability than if no speed limit were set. The HI of the state in enforcing the speed limit is to reduce the likelihood of accidents. Indeed, we may concede that the speed limit seems like a CI, but Kant cautions us that “all imperatives that appear categorical may yet be covertly hypothetical (Kant, 2012: p. 32). Some of the German autobahnen have no speed limits. Can we say that the HI of the German government is to maximize the accident rate? This is of course absurd; the HI of the German government should be the same one mentioned above. However, we see that there are two means of realizing the same end because of the conditional nature of the HI. How can we then assert which one is more justifiable than the other? All HIs are subordinate to the CI and are equal in their authority over all rational beings, i.e., there is no moral superiority in HI. If I insist that my HI would arrive at the destination faster, the HI of the state does not create a moral obligation for me to obey in that all HIs are equally justified.

Consider the following example. Emergency vehicles are legally permitted to drive over the speed limit with the anticipated end of saving other people. A husband may be rushing his laboring wife to the hospital but was punished with a speeding ticket regardless of his anticipated end, which was similar to that of the emergency vehicle. What then justifies the exception of emergency vehicles for a reason other than that it is a state-run service? The emergency vehicle may also cause accidents since it cannot predict what will happen. Therefore, the traffic laws based on a particular HI create no moral duty for the citizens to comply in that the anticipated end may fail, and different HIs can come into conflict with no verdict on which one has moral superiority.

All human beings have the right to be the legislators of practical laws according to the humanity formula; the HI of the state is no more justified than that of the individual citizen. The husband may argue that he has the same end as the emergency vehicle. In this case, the authority of the state is no longer legitimate because it enforces heteronomy. Therefore, for the HI to have moral authority, it should admit no exceptions. That is, it must become quasi-categorical in observing the universalizability principle. For example, all cars should be manufactured in a way that allows no speeding over the limit, or the speed limit should be abolished altogether so everyone is at liberty to drive as he wishes. The former solution defeats the purpose of the proposition because people are no longer given the freedom of choice to control their speed, which indeed deprives them of autonomy. The latter solution is paradoxical, for to preserve the autonomy of all citizens, it eliminates the authority of the state. In short, while speed limits have practical purposes in traffic regulation and safety, they do not directly appeal to our intrinsic sense of moral duty, highlighting the nuanced relationship between legal obligation and moral duties. It seems now that if the law is based on a HI, then either autonomy must be transformed into heteronomy, or the authority must become anarchism. In the next section, I will discuss potential avenues for overcoming the conflict between authority and autonomy.

Unanimous direct democracy is the only theoretical justification of political laws

Indeed, there are exceptions to the presumed mutual exclusiveness of authority and autonomy concerning laws based on HI. However, it is not so much that the state is justified morally as there are no longer opposing opinions. There cannot be heteronomy if the state has the same will as all individuals. That is, if all citizens unanimously agree on the same law (i.e., their HI converges on the same one), then authority and autonomy find harmony. Therefore, when we live in association with other people, the most conspicuous solution to the conflict between authority and autonomy is legislation by unanimous direct democracy.

This legislative procedure circumvents the problem with laws that are not based on the CIs, for if everyone has the same HI, it nonetheless becomes universal, which eliminates heteronomy and anarchism. If everyone agrees on the same HI, for example, to maximize the safety of everyone, everyone should obey the speed limit. It follows that the authority of the state is the same as the autonomy of the people. Expressed mathematically, the arithmetic mean of all opinions should be a positive or negative integer. That is, the sum of opinions divided by the number of people should become +1 or −1. If a society has 10 individuals, and each of them agrees that driving over the speed limit is wrong (+1), then the sum of their opinions would be +10 divided by 10 (number of people), yielding +1. Considering it in terms of percentage, it should be 100%. Since the opinion of the association is identical to that of the individual, the authority of the society is legitimate. In this scenario, the subjects who are also the legislators preserved their autonomy while obeying the authority. We would imagine that it would be much easier if we were just dealing with a limited number of people, but the difficulty increases when we live in a society, and the more populated the society is, the more difficult it becomes because of the diversity of opinions and stronger influence of social context and the power of majority that prone autonomy toward heteronomy. Furthermore, unanimity of opinion may be accidental and does not guarantee that what everyone agrees on would maximize the benefits for everyone practically. Unanimous direct democracy provides a theoretical solution to the conflict of authority and autonomy, but it falls short in practical settings. To continue the quest of solving the conflict between autonomy and authority, we must proceed to Rousseau’s idea of the General Will (GW).

General Will shares elements from legal positivism and natural law theory

The GW is a complicated concept that cannot be dichotomously categorized as belonging to either legal positivism or natural law theory (Kain, 1990: pp. 315–334; Hiley, 1990: pp. 159–178). My task is to delve into the justifications of political laws legislated by the GW as natural laws in addition to being human laws. While political laws grounded on GW align with human laws owing to their democratic origins, I will demonstrate that GW can institute natural laws that oblige all citizens morally, which grants the legitimacy of the state. While Dorwkin’s third theory of law differs from the Rousseauvian republic in that the former argues that nonelected judges should hold judicial review powers and the latter contends that the citizens are the absolute sovereign, there are some similarities between the GW and the third theory of law (Nordahl, 2013: pp. 317–346). The third theory of the law occupies an intermediary position between the realm of legal positivism and natural law theory. It agrees that all laws are human laws to some extent and unique to each legal system and sovereign power, but hints at an “objectively correct answer” reminiscent of natural law theory. (Himma, 2003: pp. 345–377). Interestingly, the GW can also be interpreted as the common ground between legal positivism and natural law theory (Noone, 1972: pp. 23–42; Melzer, 1983: pp. 633–651). For the remaining part of the manuscript, I will demonstrate how GW contains characteristics from both legal positivism and natural law theory.

General Will as the ultimate, primary, and secondary rule

The GW is significant in justifying the state in addition to the CIs because individual reasoning may fail as we are not perfect beings (Kaufman, 1997: pp. 25–52). Kant was criticized precisely for his overly optimistic assumption of our reasoning capacity. Being social animals, however, our faculty of communication enabled us to verify our moral reasoning with other people. The GW is the manifestation of this interdependent collating process. Specifically, citizens do not simply vote on the laws that they prefer but have deliberative discourse among each other to vote for the law that they think is “right” in that it promotes the common interest of all. In light of Hart’s legal positivism, the GW can be thought of as a secondary rule that is used to create or alter primary rule which is duty-imposing. In other words, the GW is a power-conferring rule, viz., a legislative procedure in addition to being the governing rule itself (William, 1984: pp. 676–677). The GW is special because it justifies political laws via a democratic process, but bases its authority on natural laws, hence creating a moral duty for the citizens to comply and the legal duty imposed by a democratic republic.

In other words, GW is not only a legislative procedure, but also an ultimate rule of recognition, which according to Hart, is a supreme rule that harmonizes the primary and secondary rules and according to which the validity of laws is assessed, i.e., it is how a law can be justified (William, 1984: pp. 678–680). If natural law theorist adopts this concept, it would be equivalent to the moral scrutiny of the laws. This makes sense because the GW is the supreme source of authority according to Rousseau and represents the citizens who obey the political laws that they have prescribed for themselves. To satisfy the former criterion, the GW must be the ultimate rule of recognition so political laws can be validated, to satisfy the latter in light of the autonomy formula, it must also be a secondary rule of legislative procedure. In the following section, I will first examine the role of GW as a secondary rule (legislative procedure).

General Will emerges from majoritarian direct democracy

The GW is to be found by counting votes from the entire sovereign, viz., it must be found by a majoritarian direct democracy. The state has legitimate authority over the citizens insofar as it is directed by the GW which is determined by the people and aims for the common interest. Rousseau wrote, “Only the GW can direct the power of the state according to the purpose for which it was instituted, which is the common good. The sovereignty, being only the exercise of the GW can never be transferred [i.e., it is inalienable] and cannot be represented except by itself [i.e., the entire populace]” (Rousseau, 1999: p. 63). This seemingly simple requirement is not so easily satisfied. A representative democracy like the U.S., or more precisely, all modern democratic states, can be said to have no legitimate authority based on the Rousseauvian argument because GW can be found only by majoritarian direct democracy. Rousseau further argues that,

The sovereign cannot be represented for the same reason that it cannot be alienated; it consists essentially in the GW which cannot be represented; it is either itself or it is something else. The GW does not allow it to be represented. The English people are free only during the election of the members of Parliament. Once they are elected, the populace is enslaved. (Ibid, 127, translation modified)

Following Rousseau’s argument, the sovereign is alienated in all modern states for its master is not the people that constitute it but elected officials that putatively represent the will of the people. Hence, sovereignty is indivisible and unrepresentable for the same reason that it is inalienable. The will is either general or it is not. It is either the people as a whole or only a part (Ibid, pp. 64 and 127). The former is the GW by which the sovereign has authority, the latter is the private will by which the elected legislator forces others’ obedience by power. It seems clear that Rousseau argues that legislation must follow the GW determined by direct democracy because any other legislative procedure will involve heteronomy. He wrote,

The sovereign, having no other force than the legislative power, acts only through the laws determined by the GW… There should be mandatory, periodic assemblies that convene the populace to vote. (Ibid, pp. 122–123, translation modified)

Heavily influenced by Rousseau, Kant puts forward a similar argument in his less-read work Doctrine of Rights from The Metaphysics of Morals in which he wrote,

Every state contains three authorities within it… the sovereign [legislative] authority in the person of the legislator; the executive authority in the person of the ruler [government]; and the judicial authority…The legislative authority can belong only to the united will of the people… When someone makes arrangements about another [i.e., enforces heteronomy], it is always possible for him to do the other wrong; but he can never do wrong in what he decides upon with regard to himself [i.e., autonomy]. Therefore, only the … united will of all, insofar as each decides the same thing for all and all for each, and so only the general united will of the people, can be legislative. (Kant, 1996: pp. 90–91).

If we divide political power into three branches (executive, legislative, and judicial), the first sentence of Rousseau’s quote seems to contradict my previous statement on classifying the GW as an ultimate rule of recognition that belongs to judicial power. Rousseau does not write extensively about judicial power, but it is reasonable to infer from this statement “Drawing lot suits those in which common sense, equity, and integrity are sufficient, as with judicial appointments, because in a well-constituted state such quality is common to all citizens” (Rousseau, 1999: p. 140) that Rousseau would support the jury system in the U.S. in which law professionals will be the mediator between the parties while the jury selected by lot. Kant also wrote that “neither the head of the state nor its ruler can judge, but only appoint judges as magistrates [mediators]… hence only the people can give judgment upon one of its members, although indirectly by means of the jury” (Kant, 1996: p. 94). However, the problem with the U.S.’s legal system is that the ultimate rule of the recognition is not the GW of the people through jury, but the will of the Supreme Court judges. So far different from the Supreme Court in the U.S. in which the judges are appointed by the government, Rousseau would argue that the GW of the people should also act as arbiter of the Supreme Court. If the validity of the law is called into question, the Supreme Court judges should not be the arbiter for the same reason that Congress should not be the legislators. Therefore, GW should be the legislator of the law and the ultimate rule of recognition concerning the validity of the law.

A potential objection to my argument may be that the jury with a limited number of people drawn by lot cannot represent the GW of the people, to which I answer that for the same reason that direct democracy can be used to legislate laws thanks to modern technology, so can a direct democracy be used to validate a law insofar it is a law that concerns all citizens (hence motivate all citizens to participate). It is imperative to note that in a society like the U.S., although there is a separation of power, the body politics is governed solely by the government, which has an internal separation of power but encompasses all three political powers. For Rousseau’s republic, the body of politics has only the sovereign and the government, with the former entitled to both legislative and judicial power and the latter only to executive power. The Supreme Court judges the validity of the law aristocratically because they receive scarce input from the people, while the Rousseauvian court does so in a democratic fashion. The recent overturning of Roe v. Wade by the U.S. Supreme Court revoked the federal right to abortion in opposition to the majority opinion (DeAngelis, 2023): p. 80. While the morality of abortion is beyond the scope of this paper, this example illustrates how GW which delegates the judicial power to the democratic process may present an alternative to the current Supreme Court system of the U.S.

Most modern and historical political systems have some components of paternalism (Le Grand and New, 2015). For this reason, with the premise that the sovereign is the supreme authority and the common qualities of citizens, perhaps Rousseau thought it would be redundant to separate judicial power from the legislative power if both ultimately rely on the GW of the people which can make laws (legislative) and amend them (judicial) without appealing to a higher authority. Narrowing the paternalism of the state to private individuals, one would envision that it is indeed cumbersome if we have to ask our parents first before we decide to eat an extra scoop of ice cream, given that we are already independent adults. I have established the downfalls of representative democracy and the current Supreme Court system. In the next section, I will discuss the practical feasibility of direct democracy.

On the possibility of direct democracy

For the rest of this section, I will analyze GW from the dimension of being a “direct” and “majoritarian” democracy in the section that follows. Rousseau proposed that everyone’s vote must be counted for the GW to emerge; however, he also criticized that,

Taking the term in the strict sense, a true democracy has never existed and never will. It is contrary to the natural order that the majority governs, and the minority is governed. It is unimaginable that the people would remain constantly assembled to handle public affairs. Were there a people of gods, they would govern themselves democratically, so perfect government is not for men. (Rousseau, 1999: pp. 101–102)

The strict sense of true democracy here refers to a unanimous direct democracy. Of course, Rousseau acknowledges the difficulty of achieving a unanimous decision, so he proposed that “[based on the importance of the deliberation], the more important the deliberation are, the closer the majority opinion should be to unanimity… And the more urgent the matter, the smaller the difference in the division of opinion should be” (Ibid, 136–139, translation modified). What Rousseau proposes here is a direct majoritarian democracy (Ebisch, 1977: pp. 14–20) with dynamic requirements on what counts as the majority.

One may criticize that direct democracy is nearly impossible, especially in a society with a large population such as the U.S. However, what was impossible in Rousseau’s time is not impossible anymore given the development of communication technology in the modern world. The advance of information and communication technology (ICT) enabled E-democracy, facilitating the democratic process and enhancing voter engagement (Congge et al., 2023; Kneuer, 2016; Lorenz-Spreen et al., 2023). Given the popularity of smartphones, it is possible to create an App where citizens can browse proposed laws, comment on these laws, exchange opinions with other citizens, and cast votes after identity verification. Of course, the citizens would still have the choice to vote through ballots, but advancement in information technology has already largely eliminated the difficulty of direct democracy enabled by ICT. The problem, therefore, is not so much the difficulty of direct democracy as to why majoritarianism is a morally justified secondary rule.

On the justification of majoritarian democracy

Now, let us consider the “majoritarian” part. Following the mathematical expression outline in the unanimous direct democracy section, let’s imagine that now one person disagrees (−1). then the equation would yield 9/10 = 0.9. The sum of all opinions is no longer equivalent to all individual voices since there is one dissenter. The association can choose to ostracize that dissenter, or the dissenter can choose to secede from the association. Either way 9/9 becomes +1, which restores the harmony of authority and autonomy. We must note, however, that the dissenter now lives in anarchism. If the dissenter stayed in the association and drove over the limit, the remaining 8 people would overpower and punish them accordingly, but they would not have the legitimate authority to do so in that the majority opinion also stems from HI which is no more justified than the HI of the dissenter. The dissenter can submit to the majority’s decision and choose to be peaceful, but their decision is forced by the tyranny of the majority which overpowers them. Therefore, the dissenter loses autonomy in obeying the heteronomous authority which now becomes illegitimate since it creates no moral obligation, but violent coercion for the dissenter. The authority is therefore reduced to power via the tyranny of the majority. The concept of tyranny of the majority is illustrated by John S. Mill’s On Liberty in which he wrote, “If all mankind minus one were of one opinion, mankind would not be more justified in silencing that one person than he if he had the power, would be justified in silencing mankind” (Mill, 1859: p. 18). Authority and autonomy therefore cannot be equated with the calculus of power. It follows that we must vouch that the dissenter was not willing to autonomy, i.e., violated the CI, for the authority to be legitimate. Of course, we cannot empirically verify the will of the individual due to the intrinsic limitation of deontology. How can we know that the will of the dissenter is false while the will of the majority is genuine? The agnosticism of the will cannot justify its authority. The solution to this problem is grounded on the epistemological account of the GW as the synthetic a priori truth to which we have access through the result of voting directed at the common interest. GW is not simply a majoritarian democracy in light of legal positivism. Its component rooted in natural law theory justifies the majoritarian decision while preserving the autonomy of the minority. In the next section, I will explore how this is possible according to Rousseau.

Epistemic justification of the GW as the supreme source of authority

We will now explore how the GW resolves the conflict between autonomy and authority from an epistemological perspective. An indispensable assumption of Rousseau’s argument is that we must assume that the GW cannot err. That is, we must take the GW to be omniscient. The GW is the legitimate authority since it is analogous to the CI. Although a vote is required for the GW to emerge, Rousseau assumes that the GW would always lead to the anticipated end of promoting the general interest of all citizens independent of empirical context. Rousseau hypothesized that “the GW is always right and always tends to the public welfare, but it does not follow that the decisions made by the people have the same rectitude. We always want what is good for us, but we do not always see what it is” (Rousseau, 1999: p. 66). The argument can be understood in this manner. When your opinion differs from that of the GW, you are voting for your private interest that springs from your inclination (HI), rather than the common interest that requires deliberation with autonomy and rationality (CI). In this sense, the authority of the GW is epistemologically justified. Therefore, when one votes for his or her private will, he or she does not know for certain whether it will lead to happiness, but if he or she votes for the GW, he or she can be certain that it will lead to happiness because it is infallible (the GW is always right).

I will use an example adapted from Robert Wolff’s Defense of Anarchism to further elucidate this argument (Wolff, 1970: p. 51). Suppose that you are in a hurry and need to go to Boston. You ask the conductor for directions, and he or she points to train A, but you are in a rush, so you dash to train B, which is headed to Chicago. Now the conductor has two choices, he or she can either let you ride train B, which putatively preserves your autonomy, or he or she can force you into train A since what you actually wanted is train A. If you consider this scenario, which should be the moral thing for the conductor to do? Would you be freer if the conductor forced you to train A, which headed to the correct destination, or would you be free if the conductor let you ride train B, which headed to the wrong destination? In this example, you are the citizen, the conductor represents the state directed by the GW, train A represents the GW which is your original intent willed by autonomy and pure reason, and train B represents your immediate but wrong decision that is influenced by impulses and context (reminiscent of akrasia in Aristotelian Ethics). Sentimentally, the answer to the question above depends on the preference of the individual, but following Kantian philosophy, one would be freer if the conductor forced him or her to train A. In this scenario, what the conductor directs leads to moral liberty while the passenger’s personal decision leads to natural liberty. According to the interpretation of Prof. Neuhouser, the GW secures the “[moral] freedom of individual citizens … by functioning as the embodiment as well as a precondition of such freedom” (Neuhouser, 1993: p. 363). Being a citizen of the state embodies moral freedom, which is defined as “obedience to laws one has prescribed oneself”. At the same time, such freedom requires one to act within the sphere of activities that do not violate the common interest of the community (Neuhouser, 1993: p. 366).

To give a perspicuous explication of this difficult concept, we can further distinguish between moral liberty and natural liberty. The obedience of the GW grants one moral liberty in exchange for natural liberty. Moral liberty makes a person truly the master of himself or herself. To be driven by appetite alone is slavery but the obedience to the law one has prescribed for oneself is liberty. For example, a father earns $100 per day. He has willed that this money will be used to feed his family. The grocery store is right next to a liquor store. Instead of purchasing groceries for his family, he was compelled by inclination to purchase alcohol, leaving his family in starvation. The father has the natural liberty to purchase anything as he pleases, but he preserves his moral liberty only if he rejects the inclination and chooses the original decision of buying groceries for his family. A more detailed discussion on desires versus freedom of will can be found in endnote 1.

A similar account can be found in Isaiah Berlin’s two concepts of liberty in which he defined negative liberty (natural liberty) as the freedom to do whatever one pleases without external impediments and positive liberty can be interpreted as living the life of a rational and moral person (Berlin, 1958). Negative liberty concerns the individual alone while positive liberty (moral liberty) concerns the individual as a member of a collective society. Rousseau wrote,

The passage from the state of nature to the civil state allows the voice of duty to replace physical impulse and justice to replace appetite. Only does man, who had until now taken only himself into account, find himself forced to act upon principle and to consult his reason before listening to his inclination. Although he deprives himself of several advantages in the civic state, he regains greater ones. (Rousseau, 1999: pp. 59–60, translation modified)

Rousseau’s assertion that “when one refuses to obey the GW, he will be compelled to do so by the whole body [politics]; while means nothing else than he will be forced to be free” (Ibid, 58), can therefore be interpreted as such: when one enters the social contract, he or she is forced to renounce his or her negative liberty in exchange for positive liberty, which can only be obtained in a civic society directed by the general will. The recent turmoil surrounding vaccine and mask mandates serves as great examples in illustrating how obeying one’s private will can be less free than obeying the GW. In short, when the sovereign forces you to obey the GW, it forces you to obey your autonomy and rationality, rather than your immediate inclination and desires. In this way, natural liberty is exchanged for moral liberty. However, GW still involves the risk of paternalism. How do we know that the decision of the GW is better than ours? I will explore the answer to this question in the following sections.

Transcendental conception of the GW parallels the Synthetic A Priori nature of CIs

The GW, when placed into the Kantian framework, can be presented as such: what one wills to the common interest is also what is willed by all other rational beingsFootnote 5. When the sovereign legislates for the political laws that carry moral authority, everyone must, through voting, present their willed law. With the Condorcet assumption which we will discuss later, the majority opinion of the voting result would be guaranteed to be a political law that carries the same supreme moral authority as the CI.

Before we proceed, we must emphasize that democracy is necessary but not sufficient for the GW. The citizens must vote for the general interest of society rather than their private interest. The former manifests the GW that unconditionally obliges all citizens morally even if one’s opinion is different from that of the GW, while the latter only manifests the will of all that has no moral authority over those who hold opinions contrary to the will of all. The will of all can be thought of as the sum of all HIs that aim for selfish ends with exceptions, while the GW is the manifestation of the CIs that allows no exception and equally obliges all citizens morally. According to Prof. Weinstock, “a Kantian legislator cannot … legislate so as to attempt to satisfy the preference of citizens” (Weinstock; 1996: p. 402). Even in a representative democracy, the legislator cannot base their decision solely upon the preference of the constituents but consider additionally what laws can maximize common interest. This is a critical distinction that ought to be taken into account carefully when examining the possibility of GW. There are numerous philosophical inquiries on what is representative of the common interest and how to derive them (Ginsberg, 1963: pp. 99–116; Rawls, 1958: pp. 164–194). One of the most famous accounts is John Rawls’ idea of the Original Position that if everyone possesses a thin theory of the good and will which laws society should adopt behind the veil of ignorance, then they will necessarily arrive at the same set of laws that applies justly to everyone. The laws conceived must also follow the equal liberty, equal opportunity, and difference (max-min) principles. It follows then that “members of [society] are autonomous and the obligations they recognize self-imposed” (Rawls, 1971: pp. 11–22; 136–142). The validity of Rawsl’s argument remains to be examined critically; however, I will not digress too much from the central thesis. Interested readers as encouraged to read A Theory of Justice in detail (Rawls, 1971).

For the will of all to be justified, it would require unanimous consent of all citizens associated with impracticality which we have discussed, while GW is sufficiently justified with only a majoritarian vote, provided that every citizen truly votes for the common interest they have willed to be. This conception of the GW is known as the transcendental conception, which is different from the democratic conception. Rousseau holds that the GW is not a simple outcome of majoritarian democracy, but a metaphysical manifestation of citizens’ common interest that exists in abstraction from their immediate interest (Bertram, 2012). Rousseau wrote,

It is through GW that the people are citizens and have freedom. When a law is proposed in the people’s assembly, what is asked of them is not precisely whether they approve or reject the proposal, but whether or not it conforms to the GW that is theirs. Everyone, by voting, states his or her opinion on the matter, and the manifestation of the GW is drawn from the counting of votes (synthetic). When, therefore, the opinion contrary to mine prevails, this proves merely that I was in error (the GW is true a priori), and that what I took to be a GW was not so. (Rousseau, 1999: p. 138, translation modified)

The GW must therefore be viewed not simply as a democratic, but as a transcendental concept that justifies the state and its political laws. The democratic conception of the GW views it as a posteriori like the HI which involves means devoted toward an anticipated end whose realization can be verified only empirically. The democratic conception of the GW as the will of all does not justify it as the source of legitimate authority.

The transcendental conception posits that the GW is a synthetic a priori concept like the CI (Rustighi, 2022; p. 134). I have hitherto paralleled CI with GW, but they are not precisely identical. To reiterate the Kantian jargon, the GW is synthetic in that it is manifested by the majoritarian vote of all citizens through a democratic process, that is, it relates to the wills of all citizens, it is a priori because each citizen voted for what he or she believed to be the common interest which is an objective concept presupposed to exist as the guiding principle for the commonwealth. The GW is different from the will of all as the former is a priori and the latter is a posteriori. GW is not justified by experience, viz., whether the anticipated end is achieved, but by the transcendental concept that it is directed purely toward the common interest, giving it the characteristics of apriority: necessary, unconditional, and universal. Since the GW is a priori not precisely by pure reason but by the transcendental conception that it is the genuine representation of the common interest collectively willed by all citizens, it is justifiable even if the end fails and allows for exceptions. For example, if everyone believes that the speed limit would maximize safety which is a common interest of society, then the political law would hold authority over the citizens even if the end may fail. Similarly, if everyone believes that emergency vehicles should be allowed to speed above the limit, for it accords with a common interest, then such a law is justified. The GW is less rigid than the CI, yet the proposition that the GW manifested by the majoritarian democracy is the representation of common interest requires further justification, viz., why can’t the minority opinion represent the common interest? If we assume mathematics to be a priori, Condorcet’s theorem supports the idea that the common interest can be found in majoritarian direct democracy. Based on Condorcet’s theorem, each citizen must have a greater than 50% probability (p) of getting the truth. If p > 50%, then it is mathematically guaranteed that the majority would arrive at the truth (Ladha, 1992: pp. 617–634). In other words, “if the likelihood of a correct judgment … of the representative is greater than that of an incorrect one, the probability that the majority vote is correct increases as the likelihood of a correct decision by the … legislator increases (Rawls, 1971: p. 358). With the assumption that p is greater than 50%, it is guaranteed that the majority will have arrived at the truth. Furthermore, it justifies direct democracy over representative democracy. The number of legislators is negatively correlated with the obligatory proximity of the voting result to unanimity. It can be interpreted in this way. Assume there is a biased coin with a 60% p of getting the head (truth) and 40% p of getting the tail (falsity). If you toss the coin 10 times, you may get 6 heads, but you are also likely to get 5 heads or even 4 heads and lower due to the limited number of trials. However, as the number of tosses increases infinitely, the result would approximate 60% head more closely. If the number of trials is limited, then p must be very high to assume that the majority has arrived at the truth. This is part of the underlying implication in Rousseaus’s previous statement that “the more important the deliberations are, the closer the majority opinion should be to unanimity” (Rousseau, 1999: pp. 136–139). From this reasoning, representative democracy cannot be justified by Condorcet’s theorem because there are a limited number of legislators. In a direct democracy which is what Rousseau’s GW requires, all citizens will participate in the voting, which increases the likelihood that the majority would arrive at the truth. It follows a smaller population requires that the majority approximate unanimity as much as possible. The requirement for p to be greater than 50% decreases as the population increases. However, an undeniable objection to Condorcet’s jury theorem is that it must assume p > 50% even with an infinite population size. Without this Condorcet assumption, the GW would always arrive at the wrong decision if p < 50% or sometimes right and sometimes wrong if p = 50%. While GW can be justified a priori if we make the Condorcet assumption, the problem is that p can only be verified a posteriori for we cannot derive p a priori. That is, we cannot know whether the majority decision leads to common interest until the law is implemented. If p is always greater than 50%, then the GW is morally justified to represent the legitimate authority, but if p is equal to or smaller than 50%, then the GW cannot be a legitimate authority. Indeed, Rawls pointed out the epistemic insufficiency of majority rule. He wrote, “Yet majorities are bound to make mistakes, if not from a lack of knowledge and judgment, then as a result of partial and self-interested views” (Rawls, 1971: p. 354). In sum, we have no way of ensuring that people will genuinely vote for their common interest, and if we assume that they do, we must make another far-fetched assumption that p is greater than 50% so the majority is always right. Therefore, Rousseau’s idea of the GW also fails to justify the state ultimately unless we make risky assumptions.

There are still caveats associated with the arguments for GW. Firstly, one may not be able to derive what tends toward the common interest due to the limit of human reasons. Rawlsian theory of justice provides helpful heuristics, but it has many additional criticisms such as detachment from reality and insufficiency of the difference principle (Choptiany, 1973: pp. 146–150). To know what is right is in itself a philosophical conundrum. Secondly, even if we suppose that we know what is objectively right, this calls into question why a democratic process is required for the GW to emerge. If we assume GW is a priori, then we are also implying the objective validity of GW can be recognized individually without prior deliberative discussion, contradicting the requirement for the democratic process. The state can just impose laws that maximize the common interest without the democratic process, which is more efficient. Thirdly, philosophers like Robert Nozick and Adam Smith laid out the argument that the collection of individual pursuits of private interest is what maximizes the common interest through the “invisible hand” process in the absence of government interference (Nozick, 2013; Teitelman, 1977: pp. 495–509; Hutchison, 1976: pp. 507–528). Lastly, there are additional challenges to the implementation of GW in a large and diverse society. One cannot deny that Rousseau may have been affected by the socio-historical and geopolitical context of the Geneva Republic (Mason, 1993: pp. 547–572) being a small country with a relatively homogenous population. GW cannot completely justify political laws, as with any other philosophical attempt. Despite all the limitations, by replacing representative democracy with direct democracy, the GW still presents an alternative to the current political system and may better ensure the justice of policies and the moral foundation of laws based on Condorcet’s assumptions.

Conclusion

In this essay, I initiated our discussion by differentiating authority and power, moral duty, and legal duty. I proceeded to discuss how the CI can justify the legitimacy of authority while preserving autonomy on the limited number of laws. Following that, I demonstrated how political laws based on HI are not justified and proposed GW as a potential solution. The ensuing discussion unveiled the GW’s nuanced attributes, having elements from both legal positivism and natural law theory. Transitioning from this, I demonstrated how GW emerged from majoritarian direct democracy. Furthermore, I examined how the GW is justified epistemologically if we assume it is a synthetic a priori concept based on Condorcet’s theorem. However, we grapple with the quest for finding a legislative procedure that justifies the authority of the state while preserving the autonomy of all citizens except in the case of violating CIs.

Kant defended that if laws such as “thou shalt not murder” are a CI, it does not give occasion to the conflict between authority and autonomy in that one cannot autonomously will anything contrary to reason. If autonomy is not involved in the reasoning process, then the authority will be justified in punishing him or her. This proposition, however, has limited application because of the rigidity of the CI and the fact that most political laws are based on HI, which introduces the mutual exclusiveness of autonomy and authority. Rousseau puts forward GW as the solution to the conflict between autonomy and authority arising from disputes over political laws. Nonetheless, there are limitations due to the difficulties associated with the emergence and application of GW. Even if we ignore all the technical difficulties associated with direct democracy, we cannot know whether the p would be greater than 50%. If it is equal to or smaller than 50%, then the assumption that the GW cannot err and the entire of Rousseau’s argument will fall apart, not to mention the disastrous consequence if a society is under the direction of such GW. Furthermore, there is also an equally important shortcoming that legislators may just vote for their partial interest rather than the common interest of all people, i.e., how do we know that the voting result is the GW and not the will of all? Because of the agnostic nature of p as well as the will of the legislator, direct majoritarian democracy still falls short of perfectly justifying political laws.

Further objections to GW can be proposed. For example, I selectively skipped the problem of indoctrination, censorship, and civil religion which are familiar criticisms of GW so that the essay does not digress from the central thesis too much. I want to caution the readers lest I am misinterpreted as arguing for anarchism because no perfect justification for the legitimacy of authority can be found. I believe that a minimal state that conforms to the fundamental principles of justice is necessary for the normal functioning of a society. In essence, my central thesis is simply that, considering current socio-political systems based on majoritarian representative democracy, the adoption of the GW holds the potential to achieve direct democracy: a stride toward legislating just laws and rectifying the socio-political system of the U.S.