Hostname: page-component-7b7567c8c-2n9c9 Total loading time: 0 Render date: 2026-11-13T19:36:20.288Z Has data issue: false hasContentIssue false
  • English
  • Français

Ranking the Regimes in Aristotle's Politics: The Four-Principles Approach

Published online by Cambridge University Press:  25 November 2020

Ross Mittiga
Rights & Permissions [Opens in a new window]

Abstract

There is a long-standing debate over which constitution Aristotle regards as best in the Politics. I attempt to clarify his view by reconstructing four principles he uses to assess constitutions, in both ideal and more ordinary circumstances: (i) the supremacy-of-virtue principle, (ii) the more-virtuous-citizens-are-better-than-fewer principle, (iii) the equality principle, and (iv) the stability principle. I apply these principles to defend a rank ordering of constitutions, which situates the ideal aristocracy of books 7 and 8 at the top, and tyranny, along with unmixed forms of democracy and oligarchy, at the bottom.

Information

Type
Research Article
Information
The Review of Politics , Volume 83 , Issue 1 , Winter 2021, pp. 1 - 20
Copyright
Copyright © The Author(s), 2020. Published by Cambridge University Press on behalf of University of Notre Dame.

There is a long-standing debate over which constitution Aristotle regards as best in the Politics. The disagreement pertains not only to Aristotle's view in the Politics as a whole, but to his view in particular books and chapters of the work. Hence Richard Kraut reads book 3 (especially chapters 17 and 18) as an endorsement of kingship and/or aristocracy as the best constitutional form, whereas Daniel Devereux interprets chapters 9 through 18 of book 3 as an endorsement of the regime Aristotle ambiguously refers to as “polity” (politea).Footnote 1 Josiah Ober and Jill Frank, who focus on the “polis of our prayers” described in books 7 and 8, argue for something in-between: an “aristocratic democracy” or “democratic aristocracy.”Footnote 2

I share the view that the “unqualifiedly” best constitution is aptly described as an inclusive (if not democratic) aristocracy. But while Ober reaches this conclusion by elaborating Aristotle's natural and historical accounts of political development, and Frank by showing how the cultivation of virtue (through the bonds of civic friendship) is possible within real democracies like Athens, I identify four principles Aristotle uses to evaluate the manifold constitutional arrangements he identifies in Politics. After properly ordering these principles, I apply them to establish a rank order of regimes—both in an absolute sense and relative to particular circumstances. I call this the “four-principles approach.”

The principles are (i) the supremacy-of-virtue principle, (ii) the more-virtuous-citizens-are-better-than-fewer principle, (iii) the equality principle, and (iv) the stability principle. I consider (i) and (iii) to be principles of justice, and (ii) and (iv) to be principles of advantage or benefit.Footnote 3 As Aristotle maintains throughout the Politics, “correct” constitutions are both just and beneficial (e.g., 3.17.1287b37–40; 5.8.1307a11–14).Footnote 4

Certain other maxims can be deduced from combinations of these four: for instance, taking (ii) and (iii) together seems to entail that citizens should (know how to) both rule and be ruled.Footnote 5 Similarly, combining (iii) and (iv) suggests the need to balance arithmetic and proportional distributive principles (5.1.1301b25–1302a15). Finally, all of these principles, taken together, promote civic friendship and the eudaimonia of the city-state—at both the collective and individual levels (3.9.1280b35–40; 7.2.1324a23–25).Footnote 6

My argument unfolds in four parts. I begin by exploring the textual grounds for each principle and some of the correlated maxims that can be drawn from them. In part 2, I defend a priority ordering of the principles. In part 3, I demonstrate how different constitutions can be considered best qualifiedly (that is, with respect to particular principles or sets of principles), while only one constitution (that described in books 7 and 8) is unqualifiedly best—namely, best in relation to all four principles.Footnote 7 Part 4 concludes with a more complete ranking of the main constitutional forms Aristotle discusses in the Politics.

The four-principles approach responds to an important and long-standing objection: that Aristotle is ultimately not interested in settling the question of which constitution is absolutely best, but rather is concerned with the more pragmatic questions of which constitution is “realistically” best, which is most stable, which is most just, and so forth, which leads him to label different constitutions “best” in relation to different circumstances.Footnote 8 While I agree that the Politics pursues multiple, distinct inquiries along these lines, this does not rule out the possibility that Aristotle regards one constitution as best overall. I show this by applying the four principles to generate a single, concise ranking of the primary regimes discussed in the Politics. While some of the conclusions I reach here echo those of scholars such as Ober and Frank, my approach offers a new and distinctive explanation of Aristotle's various and seemingly conflicting remarks about which constitution is best.Footnote 9

1. Defining the Four Principles

(i) Supremacy of Virtue

The first principle stipulates that any person, singular or corporate,Footnote 10 who has more virtue and power than all the other parts of a society ought to rule that society, as a matter of justiceFootnote 11 (3.13.1284a3–11, 3.9.1281a1–5, 3.11, 7.14.1332b13–22, 7.3.1325b10–11). The following passage offers a clear expression of this principle:

But if there is one person or more than one (though not enough to make up a complete city-state) who is so outstanding by reason of his superior virtue that neither the virtue nor the political power of all the others is commensurable with his (if there is only one) or theirs (if there are a number of them), then such men . . . would be treated unjustly if they were thought to merit equal shares, when they are so unequal in virtue and political power. (3.13.1284a3–10)

Aristotle later concludes that it seems “natural” for “everyone to obey such a person gladly, so that those like him will be permanent kings in their city-states” (3.13.1284b32–34).

Principle (i) does not pertain only to kingships, however. In fact, we find Aristotle applying it in his assessment of the governing capacity of the multitude. Hence, he claims “the majority too have a just claim [to participate in politics] against the minority, since they are stronger, richer, and better, when taken as the majority in relation to the minority” (1282a37–40). The fact that (at least in some city-states) the multitude, taken collectively, possesses greater virtue than the other citizens or groups in the same society entitles them to take part in governing in the ways a multitude can—chiefly, through deliberative and judicial bodies (3.11.1281a40–b21, 1281b22–35).

Yet, while virtue alone qualifies a person to rule, Aristotle stipulates in the passage above (3.13.1284a3–10) that, to be effective in ruling, it is also necessary to possess a similar plenitude of political power (politikē dunamis).Footnote 12 This idea emerges even more explicitly later at 7.3.1325b10–12.Footnote 13 Importantly, the possession of political power does not itself comprise a legitimate claim to rule (3.13.1283b23–27); rather, Aristotle appears to regard such power simply as a necessary condition of ruling effectively.

Both in expressing the idea that the most virtuous person(s) ought to rule, and that such virtue must be wed with sufficient power if the rule is to succeed, Aristotle is echoing and extending arguments made by Plato. In the Laws, for instance, Plato argues that “where supreme power in a man joins hands with wise judgment and self-restraint, there you have the birth of the best political system.”Footnote 14 This same idea stands behind the paradox of the philosopher-ruler in the Republic: “Until philosophers rule as kings or those who are now called kings and leading men genuinely and adequately philosophize, that is, until political power and philosophy entirely coincide . . . cities will have no rest from evils.”Footnote 15 Aristotle, like Plato, doubts that such a “blessed” union of supreme virtue and political power is likely ever to be found in a single person.Footnote 16 Aristotle departs from Plato, however, in applying this principle to collective persons (e.g., a wise multitude) as well as individuals.Footnote 17

(ii) More Virtuous Citizens Are Better than Fewer

The second principle holds that it is better for a city-state to have more citizens with virtue rather than fewer.Footnote 18 I take virtue here to encompass all of the moral virtues plus practical wisdom.Footnote 19 Aristotle concedes that it is usually “difficult”Footnote 20 for more than a few citizens of any given polity to possess virtue in this sense (3.7.1279a39–b1).Footnote 21 Nevertheless, a polity with more citizens of this type is, on his view, clearly preferable to one with fewer.

Aristotle provides three essentially pragmatic reasons for endorsing (ii), which together make it a principle of advantage. To begin with, he claims that having more excellent citizens is advantageous insofar as this allows for a broader distribution of the manifold political duties associated with running a city-state well (3.16.1287b7–15). Second, having more persons who are both “good people” and “good citizens” taking part in governing reduces the likelihood of corruption. This is because it is easier for the “judgment of an individual” to become “corrupted when he is overcome by anger or some other passion of this sort, whereas . . . it is hard to get all the citizens to become angry and make mistakes at the same time”; thus, the larger a group is, the “more incorruptible” it will be (3.15.1286a32–35, 1286b3–5). Third, and perhaps most importantly, having more virtuous citizens serves to make the city-state more excellent, which is desirable because excellence is integral to the highest goal: eudaimonia (3.16.1287b10–15, 7.13.1332a33–5, 3.15.1286a37–1287b8). Aristotle actually identifies two ways in which a citizenry might be considered excellent: (1) collectively—that is, when the people, taken together as a whole, are virtuous even if few or no constituent individuals are;Footnote 22 or (2) individually, together—that is, when each individual is virtuous such that the collective is, too, by extension.Footnote 23 But note that while both (1) and (2) operate on the same basic assumption (that virtue can compound in groups), Aristotle believes the latter, in which there are proportionally more citizens with virtue, is “more choiceworthy,” given that virtue exists at both levels.

(iii) Equality

The equality principle mandates equal treatment for equals and unequal treatment for unequals (7.3.1325b7–10; cf. 3.13.1284a7–10). More specifically, it stipulates that political offices, honors, and responsibilities ought to be distributed equitably, such that similar or equal shares are allocated to citizens who are similar or equal in politically relevant respects, while dissimilar shares are allocated to citizens who are dissimilar or unequal in those respects.Footnote 24 “Politically relevant respects” refers to the three primary grounds on which the distribution of political power might be legitimately based: wealth, freedom, and virtue (3.13; 4.8.1294a14–25).Footnote 25

The equality principle is foremost a principle of justice: “the political good is justice and . . . what is just is some sort of equality” (3.12.1282b16–18; cf. 5.1.1301a25–30, 3.13.1283b38–40). Indeed, some commentators have referred to principle (iii) as the principle of “formal justice.”Footnote 26 Yet, as Aristotle notes, while everyone agrees that “what is unqualifiedly just is what is according to merit,” there remains widespread disagreement over what constitutes merit (5.1.1301b35–39; cf. 3.17.1288a20–23). Consequently, there exists no single or universally agreed-upon notion of justice (see, e.g., 3.12–13, 3.9, 5.1). Instead, there are three ideals of distributive justice, corresponding to the three claims mentioned above: one based on freedom, which is democratic justice (e.g., 6.2.1317a40–b16); one on wealth, which is oligarchic justice (e.g., 3.5.1278a19–20, 4.7.1293b10); and one on virtue, which is aristocratic justice (e.g., 3.9.1281a1–7).Footnote 27

Aristotle further distinguishes between “numerical” equality and what sometimes gets translated as “proportional” equality or (somewhat misleadingly) “equality according to merit” (5.1.1301b25–1302a15).Footnote 28 Numerical equality pertains to what is equal “in number or magnitude,” whereas proportional equality concerns what is “the same and equal in ratio” (5.1.1301b30–34).Footnote 29 Aristotle considers democratic equality a form of numerical equality, because the shares of political power will be equal to, or evenly distributed among, the number of free citizens dwelling in the city-state (6.2.1317b1–10; 6.6.1321a1–5). Oligarchic and aristocratic equality, on the other hand, are both forms of proportional equality, since political power in these regimes is distributed in proportion to the relative wealth or virtue, respectively, of each citizen.

Aristotle seems to regard all three distributive ideals as incomplete and so imperfect (3.9.1280a9–11, 1281a8–10). In other words, unmixed forms of democracy, oligarchy, and aristocracy all “possess justice of a sort” but, “unqualifiedly speaking, they are mistaken” about the absolute primacy of their notion of it (5.1.1301a36–38). Hence, when two or more distinct groups—e.g., the rich, the virtuous, the poor, or some amalgam group such as the well born—are present in a city-state, that city-state should attempt to accommodate each of the corresponding notions of equality by mixing them or balancing them against each other in its constitution (3.13.1283b1–1284a17; 5.1.1302a2–15).Footnote 30 Should the constitution fail to balance these disparate ideals, the city–state risks extreme forms of injustice and/or instability; in Aristotle's words, “faction is everywhere due to inequality” (5.1–2, esp. 1301b26–7; 3.13.1283b14–28).

In this sense, the equality principle is a principle of justice and benefit (5.8.1308a10–12). It is a principle of justice insofar as justice consists of equality. And it is a principle of advantage insofar as achieving justice (by correctlyFootnote 31 balancing different ideals of equality in the constitutional order) tends to promote the eudaimonia and stability of the city-state (5.8.1308a9–18; 7.14.1332b25–29; 3.9; 5.1). Thus, we can say that a constitution that justly and equitably distributes political power among qualified citizens is, ipso facto, just and choiceworthy.

(iv) Stability

Although on Aristotle's view stability largely hinges on the presence of equality and justice, it nonetheless represents an independent good that can be used to evaluate constitutions.Footnote 32 The stability principle maintains that, all else being equal, constitutions that are more stable and easily preserved are, for that reason, better and more choiceworthy. This principle is especially salient in books 4–6, which examine various mechanisms, reforms, institutions, and so forth that are aimed at achieving greater stability (or reduced instability) and thus preserving constitutions.Footnote 33 Two important maxims result from his inquiry, which serve as extensions of the stability principle:

  1. (a) All else being equal, it is better for a constitution to be supported (or at least not opposed) by the greatest number of citizens possible.Footnote 34

  2. (b) All else being equal, it is better for a constitution to include as many qualified groups of citizens as possible, and to exclude any individuals who are, or are likely to become, grossly superior (and so dissimilar and unequal) in wealth, influence, or any other politically relevant good except virtue.Footnote 35

Aristotle provides two variants of principle (iv)(a).Footnote 36 The strong variant stipulates that “if a constitution is to survive, every part of the city-state must want it to exist and to remain as it is” (2.9.1270b20–23; see also 4.9.1294b34–40 and 6.5.1320a14–17). More plainly, regime stability depends on no part of society wishing for a different arrangement and all parts wanting the existing order to persist. This unanimity requirement makes the strong variant of (iv)(a) particularly stringent.

Perhaps recognizing this, Aristotle presents a weaker formulation in other passages, which requires only that the majority in a city-state wish the current constitutional order to persist: “it is essential that that part of the state which desires the permanence of the constitution should be stronger than that which does not” (4.12.1296b14–16; see also 4.13.1297b4–6 and 6.5.1320b26–28). Aristotle refers to this as “the most important principle”: “ensur[ing] that the number of those who wish the constitution to be maintained is greater than that of those who do not” (5.9.1309b16–18). As with the strong variant, the animating idea here is that by securing the broad support of citizens, a constitutional order will be (more) stable and so more easily preserved.

While both the strong and weak variants of principle (iv)(a) could be described as principles of consent, principle (iv)(b) would be more aptly characterized as a principle of inclusion. Again, however, on Aristotle's view, both consent and inclusion are potentially relevant aspects of, and ways of effecting, constitutional stability and preservation. The argument he gives in support of (iv)(b) is subtle and complex, but can be stated in outline as follows: by including certain groups of citizens (in certain cases), greater equality is achieved; and since inequality is a key source of faction and instability, greater equality will reduce faction and so promote stability; thus, a constitution that includes more (groups of) citizens in equitable ways is, all else being equal, more stable and likely to be preserved.Footnote 37

This line of reasoning explains why many of the reforms outlined in books 4–6 call for greater inclusion: of the poor/many in oligarchies, of the rich in democracies, and so forth (see, e.g., 5.9.1310a2–11). It also explains why in many of the same passages Aristotle warns of the “dangers” of exclusion (3.11.1281b25–35). It is perhaps unsurprising, then, that Aristotle considers polity to be both one of the most inclusive and one of the most stable constitutional forms (4.12.1296b38–1297a3; 4.11.1296b34–a12).

Three important qualifiers must be noted with respect to (iv)(b). First, for Aristotle, inclusion must be delimited by higher-order concerns with good decision-making, justice, and virtue (more generally).Footnote 38 Thus, only those citizens who are free and have the capacity for virtue and the opportunity to achieve it (by being unburdened by “necessary” tasks) should participate in the constitution. This explains why he advises against including “vulgar craftsmen,” “natural slaves,” and certain other groups.Footnote 39 However, these same higher-order considerations also form the basis of his support for including the middle class into virtually any political order in which it exists (4.12.1296b34–40).Footnote 40 Second, while Aristotle clearly believes that inclusion of key groups is critical for achieving and sustaining constitutional stability, he also argues that, in certain cases, the exclusion (i.e., ostracism) of particular individuals—namely, those who are grossly superior in certain respects—is necessary for the same reason.Footnote 41 By ostracizing overly powerful (and thus dissimilar and unequal) individuals, a city-state is able to preserve equality and, thereby, stability. Notably, Aristotle thinks this practice is compatible with justice and so may be implemented even in correct constitutions (3.13.1284b3–5, b15–17). Yet he also claims that the need for ostracism could (and, ideally, should) be obviated through constitutional design (3.13.1284b18–19); this suggests that the unqualifiedly best constitution would not need the practice. Third and finally, it is important that the stability principle not violate any of the principles that precede it. Thus, if including or ostracizing certain (groups of) citizens is unjust, this should not be done, regardless of whether that inclusion or exclusion would promote stability.

Of course, Aristotle considers other aspects of regime stability: for example, military virtue, personnel, and fortifications (2.7.1267a19–21; 7.4; 7.11), geopolitical position (e.g., 2.10.1272b15–17), and public education (5.9.1310a12–20). Nevertheless, consent and inclusion are, for him, the most important facets of constitutional stability and preservation.

2. Ordering

There are strong textual reasons for adopting the ordering of these four principles as I have. The priority of (i), the supremacy-of-virtue principle, is established in several passages. For instance, in discussing the different claims to political power (wealth, freedom, and virtue) Aristotle asserts that if a godlike person were to have more virtue and power than all the rest (combined), the “only option” would be “for such a person to be obeyed and to be in control not by turns but unqualifiedly” (3.17.1288a28–30; cf. 3.13.1284a3–11).Footnote 42 A superabundance of virtue constitutes an “authoritative” claim to ruleFootnote 43—one that outweighs any competing claim, be it democratic, oligarchic, or otherwise. We can call this the priority of supreme virtue. As commentators have noted, however, such supreme virtue is likely to be exceedingly rare,Footnote 44 and may only exist in archaic polities.Footnote 45

In the many cases in which a superabundance of virtue is not concentrated within a singularly powerful person, the claim of virtue will lack absolute priority: namely, it will not entirely outweigh other competing claims. As Devereux explains, the claim of virtue (outside the godlike case) does not justify a “monopoly” on political power, since other (e.g., democratic and oligarchic) claims to power are at least partially just and correct, and so should be recognized and accommodated by all correct constitutions in which the claims exist.Footnote 46 While this is certainly correct, it is nonetheless clear that Aristotle regards virtue as preeminent and thus orders it prior to other claims (like those of wealth and freedom). This is suggested by the myriad passages in which Aristotle asserts that a central purpose of a city-state is to promote the virtue of each and all of its parts (3.9.1280b5–8, 1281a1–7). It follows from this that “those who contribute most” to this end ought to “have a larger [though not absolute or exclusive] share in the city-state than those who are equal or superior in freedom or family but inferior in political virtue, and those who surpass in wealth but are surpassed in virtue” (3.9.1281a1–7). In short, as the true function of a city-state is to promote virtue and the doing of noble deeds, the strongest and most authoritative claim to rule is always that of virtue. If virtue is present in superabundant and concentrated form, that claim will be absolute; otherwise, it will have only a relative priority to other claims that must be balanced in accordance with principles (iii) and (iv).

The position of (ii), the more-virtuous-citizens-are-better principle, is dictated by the principles around it. Principle (ii) presupposes (i), since we must first accept that good people comprise a good and beneficial part of a city-state before arguing that it is desirable to have as many of them in the city-state as possible. The priority of (ii) over (iii), the equality principle, is suggested in the discussion of ostracism above: if equality were to constitute a higher-order concern than having a greater number of virtuous persons, then the practice of ostracism would apply to persons who are grossly superior in terms of virtue. Yet Aristotle repeatedly argues against ostracizing the virtuous (3.17.1288a24–29). It follows from this that (iii) must be subordinate to principles (i) and (ii). Moreover, the realization of principle (ii) provides one set of conditions that makes (iii) necessary. For if it is just that the virtuous rule, then only in instances in which multiple citizens of the same society are simultaneously virtuous will the equitable distribution of power among them become a concern to be addressed by principle (iii). In these cases, (ii) is logically prior to (iii) (3.17.1287b40–a1).

Of course, equality, and so principle (iii), will become a concern any time there is more than one person (or part of a city-state) fit to hold political power who (or which) possesses a legitimate claim by dint of being similar or equal (in point of virtue, wealth, and/or freedom) to others also claiming power. In fact, there is only “one set of circumstances” in which principle (iii) is not applicable, namely, when a single person of society has more virtue and power than all the other people of that society combined (3.17.1287b37–1288a5; 3.13.1284a3–11). In the absence of a divine or quasi-divine monarch,Footnote 47 however, principle (iii) applies as a matter of basic justice; for, as Aristotle claims, “where people are similar and equal, it is neither beneficial nor just for one person to control everything” (3.17.1287b40–1288a1; see also 2.2.1261a30–b5).

There are three reasons for situating (iv), the stability principle, after (iii). First, and most generally, stability is a lower-order good than equality because stability concerns what is advantageous or beneficial for a constitution, whereas equality pertains to what is just. Second, much of what fuels concern over stability stems from issues pertaining to (in)equality. Thus, properly attending to principle (iii) should obviate most concerns at the level of principle (iv). Finally, stability exists in deviant constitutions—at least if we construe stability according to the weak variant of (iv)(a), which is to say, as a situation in which a simple majority of the population wishes the current constitutional order to persist.Footnote 48 This suggests that principle (iv) does not give us as meaningful an insight into the excellence of a given constitution as do the other principles.Footnote 49

Within the stability principle, (a) appears to have priority over (b) for at least two reasons. First, greater acceptance of a constitutional order may be achieved without resorting to greater inclusion, making it a more expedient option. Hence Aristotle's claims that the demos may accept (or at least not oppose) nondemocratic constitutional arrangements so long as the multitude is materially secure and not suspicious that public offices are being exploited for private profit (4.13.1297b1–12; 6.4.1318b14–16; 5.8.1308b34–1309a5).Footnote 50 Second, while both acceptance and inclusion are pursued for the sake of stability, acceptance leads directly to that end, while inclusion does so only indirectly. That is, greater inclusion is pursued, in the first instance, for the sake of greater acceptance, which, in turn, promotes stability. Thus, (b) is valuable, at least in certain cases, only insofar as it realizes (a), which is tied directly to stability. This caveat is important: in cases in which the demand for inclusion is based on a legitimate claim to political power, greater inclusivity ought to be pursued for reasons of justice rather than just as a way to promote stability (3.10.1281a28–32 and 3.11).

3. The Best Constitution—Qualifiedly and Unqualifiedly

Before applying the principles to create a rank ordering of constitutions, it is necessary to emphasize that the comparative excellence of a constitution cannot be determined algorithmically. The ideal is less about maximizing value with respect to one or more of the principles and more about achieving balance, proportion, and harmony among them. Thus, although the principles have been ordered in terms of their relative importance (or logical priority), the unqualifiedly best constitution will be such that it need not sacrifice stability for equality, or equality for virtue. Rather, it will reflect all four principles and the values they represent. It is better and more choiceworthy for a constitution to be excellent with respect to each (and every) principle, which entails being excellent overall, than for a constitution to do well overall without doing well with respect to each.

Thus, an extremely equitable but unstable constitution may be worse than one that is moderately equal and moderately stable; similarly, a constitution in which a greater amount of virtue is extremely concentrated in one or a few persons may be worse than a constitution in which slightly less virtue (overall) is present in a greater number of persons. In this sense, we might say that each of the four principles is valuable for its own sake, as well as for the sake of eudaimonia—the highest end of political society (1.2.1252b30; 3.9.1281a2–8).

This helps to explain why Aristotle makes several ostensibly conflicting textual claims concerning “the best constitution” throughout the Politics.Footnote 51 Each claim is, in a sense, correct, but (with one exception) qualifiedly. In other words, all save one of the constitutions that Aristotle describes as best are so only relative to specific principles. The single exception is best unqualifiedly—that is, with respect to all of the principles. I show this in what follows by applying each of the four principles in turn and discussing the results.

Based on principle (i) alone,Footnote 52 it appears that kingship and aristocracy are equally apt candidates for the best constitution.Footnote 53 This is because both represent rule of the virtuous, and principle (i) stipulates that the virtuous ought to have political power. Principle (i), alone, is insufficient for establishing a rank order between kingship and aristocracy; however, a qualified ordering appears possible in reference to particular circumstances. Hence, Aristotle claims that absolute kingship is justified only in “one set of circumstances”: when a single person has more virtue than all other persons and parts of society (3.17.1288a5). In such a state, kingship would be best;Footnote 54 under different circumstances, however, principle (i) provides no clear reason to favor kingship over aristocracy. Rather, both regimes are equally situated, since each constitutes a form of virtuous rule that aims at the common good. This conclusion corresponds to Aristotle's claims in 3.18 and 4.2.1289a26–33, which suggest that kingship and aristocracy could both be considered best.

But what of polities in which the multitude, collectively, is the most virtuous element of society? We find this in two constitutions: well-mixed aristocracy and well-mixed polity. The former is represented here as one possible form of aristocracy; the latter, however, is not a candidate, based on (i) alone. The reason for omitting polity is that, while aristocracy directly and necessarily involves rule of the virtuous (since offices are distributed directly in accordance with that factor), polity only incidentally does so (see 4.7.1293b1–21). Moreover, it is not clear that all polities possess the favorable conditions that allow them to approximate rule of the virtuous: it is not clear that all polities, qua polities, possess a collectively virtuous multitude. This is suggested by the fact that Aristotle delimits his remarks concerning the “wisdom of the multitude” such that they pertain only to certain peoples (3.11.1281b19–20). That said, if there is strong textual evidence showing that a polity always and necessarily has a virtuous multitude—or even that it always and necessarily includes a middle class, which Aristotle associates with virtue in the collective sense (4.11)—then it would be appropriate to include polity among the candidates for best constitution according to principle (i) alone.Footnote 55

When we also take principle (ii) into account, the situation changes. That is, based on principles (i) and (ii), it seems that (some form of) aristocracy is more choiceworthy than kingship. This is because aristocracies have more persons with virtue rather than fewer, while still being characterized by rule of the virtuous. Aristotle makes this point explicitly in 3.15: “So then, if the rule of a number of people, all of whom are good men, is to be considered an aristocracy, and the rule of a single person a kingship, aristocracy would be more choiceworthy for city-states than kingship (whether the rule is supported by force or not), provided that it is possible to find a number of people who are similar” (1286b3–7).Footnote 56 Aristocracy is “more choiceworthy” than kingship because it fulfills principle (ii), which is a principle of advantage. That said, aristocracy may only be equally just as kingship, since both fulfill principle (i). Thus, assuming roughly equal justness between aristocracy and kingship, principles (i) and (ii), taken together, favor aristocracy as the best constitution. This ranking receives textual support at 4.7.1293b1–7 and 1293b18–21 (cf. 4.3.1289b40–1290a5).

If we consider the constitutions in light of principles (i) through (iii), it seems again that aristocracy, in one of two forms, prevails. The first form is represented by the ideal constitution described in 7 and 8, in which citizens are all similar and equal in the relevant respects and are therefore equal participants in the constitution (7.14.1332b25-28).Footnote 57 This is an unmixed form of aristocracy that distributes political power solely to those with virtue. It might therefore be asked why this constitution is considered equitable and just, since, as noted earlier in reference to 3.10, constitutions that deny legitimate claims (e.g., freedom and wealth) commit a kind of injustice. Why is the constitution of 7 and 8 not guilty of this?

In 3.13, Aristotle suggests that competing legitimate claims to political power exist only when the corresponding elements of a society, representing those claims, are present. As the ideal constitution in 7 and 8 does not have these multiple, conflicting elements—because its citizens are all similar and equal in the relevant respects—it is not unjust for it to award power solely on the basis of virtue, since this is the only legitimate (and differentiating) claim that exists in that society. Thus, the ideal constitution is able to accommodate all of the legitimate claims to rule (virtue, wealth, and freedom), not by balancing—again, the citizenry's homogeneity in these regards obviates the need for deliberate constitutional balancing—but, in a sense, automatically. Since all are free-born and similarly well-off (euporos), there exists no part of society (no faction) that uses freedom or wealth as a claim to rule. Virtue remains the sole differentiating factor, as it is unequally distributed among the citizenry. But since the citizens are all equal in their capacity for virtue, and in the education they receive toward realizing it, this claim is accommodated simply by allocating political power as a function of individual age—which makes sense, given that the acquisition of phronēsis requires time and experience (7.14.1332b31–1333a3).

The second form is well-mixed aristocracy, in which citizens are not all similar and equal in point of wealth, freedom, and virtue. In this constitution, it is necessary that shares of political power be distributed according to principles that take these differences into account (4.7.1293b1–21). The supremacy of aristocracy in a general sense (covering both of these forms) is supported by many of the passages cited above. But in addition to these are others in which Aristotle identifies the well-mixed “middle constitution”—in particular, well-mixed aristocracy—as best (see 4.11 and 5.7). “Middle constitution” may be taken to refer to either well-mixed aristocracy or polity (see 5.7.1307a5–10; cf. 5.1.1301b40–1302a–15). There are also several places (especially in book 4) in which Aristotle appears to conflate these two constitutions for the purposes of his analysis.Footnote 58 That said, the fact that a well-mixed aristocracy deliberately balances all three claims (virtue, wealth, and freedom) suggests that it is superior to polity, which only balances two (wealth and freedom). Aristotle makes this point explicitly at 4.7.1293b1–21. Thus, as noted earlier, polity can be said to constitute rule of the virtuous only in a qualified sense, since it depends on the happy accident of a city-state's having a predominant middle class, rather than on the implementation of a constitutional structure that directly allocates power on the basis of virtue.

Evaluating the constitutions in light of principles (i) through (iv), it initially appears that the ideal constitution of 7 and 8 and well-mixed aristocracy are, again, equal claimants to best, since both (i) award power on the basis of virtue, (ii) have more virtuous persons rather than fewer, (iii) distribute political power justly, which is to say, in accordance with the relevant and legitimate claims to rule, and (iv) are, for these reasons, stable. On closer inspection, however, it seems that the ideal constitution outperforms well-mixed aristocracy on principles (ii) and (iv). Regarding principle (ii), the ideal constitution is realizable, it seems, only in city-states that are fortunate enough to have a citizenry that is similar or equal in all of the relevant respects,Footnote 59 which means that all “complete”Footnote 60 citizens will have the four main virtues. Consequently, the ideal constitution performs better with respect to principle (iv), since it is able to include all (complete) citizens as equal participants in the constitution (7.13.1332a32–35). It is therefore likely to have a very wide (if not universal) basis of acceptance (7.1–3). Moreover, because of its extensive inclusivityFootnote 61 and the opportunities for ethical development through political participation this entails,Footnote 62 in this regime the virtue of a good (complete) citizen equals the virtue of a good person, unqualifiedlyFootnote 63—a point that corresponds to Aristotle's cryptic claims about the unqualifiedly best constitution in 3.4.Footnote 64

Well-mixed aristocracy, by comparison, may require the various parts of society to rule in different ways (e.g., members of the demos may be restricted from holding high office), or perhaps even to be ruled permanently (without ruling in turn).Footnote 65 Thus, in a well-mixed aristocracy, the virtue of the good citizen will equal that of a good man only in a qualified sense—that is, only for certain citizens (namely, rulers).Footnote 66 In light of all four principles, then, it seems the ideal constitution of 7 and 8 is best unqualifiedly, since it is most just and most choiceworthy overall. If the very fortunate circumstance of having a citizenry that is completely similar and equal does not obtain, however, then well-mixed aristocracy, of the sort that takes into account all the relevant claims to power, would be best (i.e., most just and, relatively speaking, most choiceworthy).

4. A More Complete Ranking

We can now consider a more complete rank ordering of the many constitutional forms Aristotle discusses in the Politics. The constitutions, including subforms, are listed in a strict and descending order of excellence. While not exhaustive, my hope is that the argument has been methodologically transparent enough that omitted regimes (e.g., certain variants of oligarchy and democracy) could be placed with little difficulty.

I have purposefully omitted godlike kingship from the ranking, however, for two reasons. The first is epistemological: I am narrowly interested in political orders that do not contain godlike person(s) with more virtue and power than all the other parts and persons of their society combined. I take it that this condition is sufficiently rare, even on Aristotle's view, to validate this assumption.Footnote 67 The second reason is conceptual: even if a godlike leader were to rule, it is not clear whether the resultant order would, properly speaking, amount to a constitutional order, or if such rule could accurately be described as political.Footnote 68 Consider, for instance, Aristotle's claim that, for godlike rulers, “there is no law, since they themselves are law” (3.13.1284a12–15; 3.17.1288a1–5). If we accept that the presence and sovereign power of a system of laws is an important (and desirable) element of constitutions, then it would appear that this form of kingship exceeds a normal conception of constitutional order (3.16, 1287a19–21). Moreover, Aristotle repeatedly insists that distinctively political rule—as opposed to the rule of a master or household manager—involves “ruling and being ruled in turn” (e.g., 3.4.1277a26–30, 1277b8–15; 3.13.1283b40–1284a5; 3.17.1287b36–1288a12; 3.5.1278b4). He also states plainly that it would not be “proper to claim” that a divinely virtuous and powerful king “deserves to be ruled in turn” (3.17.1288a24–26); indeed, such “a god among human beings” should not even “be regarded as a part of the city-state” (3.13.1284a6–12), nor can his rule—which is characterized by the predominance of a part over the whole—be understood as “natural” (3.17.1288a26–28).

The Correct Constitutions

  1. 1. The ideal aristocracy described in books 7 and 8Footnote 69

  2. 2. Well-mixed aristocracyFootnote 70

    1. a. Well-mixed aristocracy that balances virtue, wealth, and freedomFootnote 71

    2. b. Well-mixed aristocracy that balances virtue and freedomFootnote 72

    3. c. Well-mixed aristocracy that balances virtue and wealth

  3. 3. Well-mixed polityFootnote 73

    1. a. Aristocratic polity, which privileges wealth over freedomFootnote 74

    2. b. Democratic polity, which privileges freedom over wealthFootnote 75

  4. 4. Unmixed aristocracyFootnote 76

  5. 5. (Archaic) kingshipFootnote 77

The Deviant Constitutions

  1. 6. Moderate, partially mixed forms of:

    1. a. Democracy

    2. b. OligarchyFootnote 78

  2. 7. Unmixed forms of:

    1. c. Democracy

    2. d. OligarchyFootnote 79

  3. 8. Tyranny and extreme forms of democracy and oligarchy

Insofar as this ranking comports with those devised by other scholars, my hope is that the four-principles approach serves as a useful supplement to their arguments. Where this ranking conflicts with those of others, however, the burden is on them to show that their views either more accurately comport with the principles I have described here, or else take into account some other principle(s) I have failed to identify. To the extent either of these cannot be shown, there are strong grounds for accepting the ranking presented above.

Footnotes

An earlier version of this article was presented at the 2016 annual meeting of the Southern Political Science Association. I thank the attendees, copanelists, and especially the discussant, Stu Gray, for their feedback on that occasion. I also thank George Klosko, Doug Reed, the journal's editor Ruth Abbey, and two anonymous reviewers for their helpful comments. My deepest gratitude goes to Dan Devereux, without whom this paper would never have existed.

1 Kraut, Richard, Aristotle: Political Philosophy (Oxford: Oxford University Press, 2002), esp. 231n71, 399–402; Daniel Devereux, “Aristotle on the Just Distribution of Political Power: Politics III.9–18” (unpublished manuscript).

2 Ober, Josiah, “Nature, History, and Aristotle's Best Possible Regime,” in Aristotle's “Politics”: A Critical Guide, ed. Lockwood, T. and Samaras, T. (Cambridge: Cambridge University Press, 2015), 224–43; Frank, Jill, A Democracy of Distinction: Aristotle and the Work of Politics (Chicago: University of Chicago Press, 2005).

3 Aristotle distinguishes between “justice” and “advantage” (or “benefit”) throughout the Politics, but without explicitly specifying what constitutes the latter. I assume that whatever conduces to virtue is advantageous, even if the good in question is not (a) strictly speaking, required by justice or otherwise necessary for the formation of virtue, or (b) exclusively a good for virtuous regimes or people (as opposed to vicious regimes or people) (3.7.1279a25–33). Advantage, then, is not necessarily an ethical concept. Moreover, views about what counts as advantageous may differ between those ruling a polis and those subject to that rule. From the perspective of the rulers in a “deviant” polity, for instance, stability is likely to be considered advantageous, while subjects would likely regard stability as (personally and collectively) disadvantageous, insofar as it implies a continuation of oppressive, unjust, or vicious conditions.

4 While this paper largely follows C. D. C. Reeve's translation of the Politics (Indianapolis, IN: Hackett, 1998), every quoted or discussed passage has been evaluated against Carnes Lord's (Chicago: University of Chicago Press, 2013) and Trevor J. Saunders's (London: Penguin Books, 1981) renderings of the text. Potentially significant points of disagreement are indicated in footnotes.

5 See, e.g., 3.4; 3.13.1283b40–1284a1.

6 I share the common view that Aristotle's primary aim in the Politics is to identify regimes in which it is possible to live well—i.e., in which the full extent of human flourishing can be realized and sustained (1.2.1252b27).

7 The unqualifiedly best constitution is, as is well known, very closely related to Aristotle's conceptions of well-mixed aristocracy and aristocratic polity, especially as these are described in book 4 (see 1293b1–20, 1273a2–5, 1294b10–11; cf. Kraut, Aristotle, 231n72).

8 This view has its roots in Werner Jaeger's distinction between Aristotle's practically and ideally best constitutions (Jaeger, Aristotle: Fundamentals of the History of His Development, 2nd ed. [London: Oxford University Press, 1948]), and is supported by Aristotle's own description of his investigation in 4.1 (esp. 1288b20–30). For recent attempts to settle the question of the best constitution by distinguishing between multiple lines of inquiry (and thus multiple “best” constitutions), see Coby, Patrick, “Aristotle's Four Conceptions of Politics,” Western Political Quarterly 39, no. 3 (1986): 480–503; Mulgan, Richard G., Aristotle's Political Theory (Oxford: Oxford University Press, 1977), 102; and Alexander, Liz Anne, “The Best Regimes of Aristotle's Politics,” History of Political Thought 21, no. 2 (2000): 189–216. For arguments against Jaeger's distinction, see Kraut, Aristotle, esp. 428n2; Lord, Carnes, “The Character and Composition of Aristotle's Politics,” Political Theory 9, no. 4 (1981): 459–78; and Frank, Democracy of Distinction, 164.

9 Ober also suggests why there are multiple answers to this question: different regime types may be best at different points in the historical sequence (“Nature, History, and Aristotle's Best Possible Regime,” 229). While I agree, it is still helpful to distinguish between constitutions that are “qualifiedly” best versus the one that is “unqualifiedly” best—i.e., the “polis of our prayers.”

10 See 5.8.1308a9–18 on the idea of a collective person, or “people,” representing a single “part” of society.

11 Note that Aristotle also thinks rule of the virtuous will be beneficial, since the virtuous contribute “the most” to a true polis (3.9.1281a1–6).

12 In these passages and elsewhere (e.g., 5.9.1309a33–39), Lord translates politikē dunamis as “capacity” or “the capacity to achieve the best things.” Saunders similarly translates this concept as “the capacity for statecraft,” which he describes further as “the ability to contribute to the life of the state by proper discharge of the functions of a citizen in his capacity as a statesman” (Politics, 213n7).

13 On this point, see also Garver, Eugene, Aristotle's Politics: Living Well and Living Together (Chicago: University of Chicago Press, 2011), 98.

14 Plato, Laws, trans. Trevor J. Saunders (London: Penguin Classics, 2004), 712a. Note that in the passage cited, Plato is discussing the best constitutional form when founding a state, not necessarily the best constitution in an absolute sense.

15 Plato, Republic, ed. C. D. C. Reeve, trans. G. M. A. Grube (Indianapolis, IN: Hackett, 1992), 473c–d.

16 Plato, Laws 680e–681d; cf. 709e–710a, 711d–712a, 875a, 691c–d. Aristotle, Politics 3.15.1286b7–13, 1284a3–b35; 7.14.1332b21–23; 3.15.1286b3–21.

17 But should there be multiple simultaneous philosopher-rulers in the Republic, then Aristotle's view tracks Plato's that much more. For an interesting defense of the claim that Aristotle's best regime is actually one ruled by philosopher-kings, see Alexander, “The Best Regimes of Aristotle's Politics.”

18 See 3.16.1287b7–15, 1286a38–b5, 1332a33–39.

19 Aristotle associates this with hoi spoudaioi (Pol. 1281b1–17).

20 Though seemingly not impossible—hence, the ideal constitution in books 7 and 8.

21 But cf. Aristotle, Nicomachean Ethics 1.9.1099b18–20, which suggests that this difficulty may be more attributable to educational shortcomings than ordinary human defect.

22 This is the essential idea behind Aristotle's famous “wisdom of the multitude” argument in 3.11. It appears again in 3.15.1286a22–35 and 7.13.1332a35–39.

23 Aristotle makes this argument explicitly in 3.15.1286a36–1287b9 and somewhat more obliquely in 3.5.1278a6–11 and 3.7.1279a39–b4.

24 See 7.3.1325b7–10; 7.14.1332b25–29; 2.2.1261a22–b6; 3.12–13; 3.17.1287b40–a1; 3.6.1279a8–15; 4.9; 3.9; 3.6.1279a8–15. See also 5.8.1308a9–18; 3.16.1287b25–34; 3.13.1284b4–15.

25 Aristotle sometimes discusses other bases—e.g., “good birth” (e.g., 3.13.1283a1–b20). He usually reduces these to the primary three, however—for instance, he describes being “well born” as a function of being born free to a virtuous and/or wealthy family (3.13.1283a32–37; 4.8.1294a20–22).

26 See, e.g., Lyons, David, “The Weakness of Formal Equality,” Ethics 76, no. 2 (1966): 146–48; cf. Georg Spielthenner, “The Role of Formal Justice in Ethical Reasoning,” Res Publica 21, no. 1 (2015): 78–81.

27 See also 3.9 and 5.1.1301b29–30. Note that “aristocratic justice” is related to the supremacy-of-virtue principle, in that it stipulates that aristocratic justice takes priority over the competing distributive ideals of oligarchic and democratic justice whenever there is conflict among them.

28 Cf. Nic. Eth. 1131a14–b23.

29 By making this distinction, and by attributing one form to democracy and another to aristocracy and oligarchy, Aristotle's discussion again mirrors that in Plato's Laws (see especially 757a–758a, 744b–e).

30 Note, however, that the absolute best constitution described in 7 and 8 is not a mixed constitution, as it lacks various factions and so rival ideals of distributive justice. This is discussed further below.

31 Aristotle explains: “what is correct must be taken to mean what is equitable; and what is equitable in relation to the benefit of the entire city-state, and the common benefit of the citizens” (1283b38–40).

32 Garver also argues that stability constitutes an additional criterion for assessing constitutions (Aristotle's Politics, 83).

33 But, as Ober points out, considerations of stability also factor into Aristotle's assessments of “correct” constitutions like kingship or aristocracy (“Nature, History, and Aristotle's Best Possible Regime,” 230, 235).

34 See 2.9.1270b20–23; 4.9.1294b34–40; 4.12.1296b14–16; 4.13.1297b1–12; 5.8.1308b34ff; 6.4.1318b14–28; 5.9.1309b16–18; 6.5.1320a14–17.

35 See 3.11, 1281b25–35; IV.12, 1296b34–a12; 3.16, 1287b25–34.

36 This distinction is suggested by Reeve in an editorial footnote (Politics, 52n109).

37 Ober seems to recognize this feature of stability when he claims that the best constitution includes all “would-be citizens”—i.e., all those with the “natural capacity to exercise citizenship” and thus the “cultural expectation” of exercising it (“Nature, History, and Aristotle's Best Possible Regime,” 224–26).

38 7.9. On this see Reeve, Politics, 206n43.

39 For instance, women, children, resident aliens, and foreigners. See 1.13; 3.5.1278a7–12; 7.9.1329a17–26; 3.4.1277b1–10; 7.4.1326a17–20; 6.4. See also Kraut, Aristotle, 214–20, 277–305, 464–65.

40 For further discussion of Aristotle's views on the middle class, see Thomas L. Pangle, Aristotle's Teaching in the “Politics” (Chicago: University of Chicago Press, 2013), 186–94.

41 Specifically, when an individual is too influential in a democracy or too rich in an oligarchy; not, however, when an individual is greatly superior in virtue (3.13.1284a17–b25; 3.17.1288a24–29).

42 For discussion of this form of government, see Newell, W. R., “Superlative Virtue: The Problem of Monarchy in Aristotle's ‘Politics,’” Western Political Quarterly 40, no. 1 (1987): 159–78; Lindsay, Thomas K. “The ‘God-Like Man’ versus the ‘Best Laws’: Politics and Religion in Aristotle's Politics,” Review of Politics 53, no. 3 (1991): 488–509.

43 Albeit not an exclusive or unqualified claim to rule. For city-states with multiple parts—e.g., a rich, a poor, and a virtuous part—all of the claims to political power must be balanced for the sake of justice (and stability). On this point, see 3.9.1281a8–10; 3.13.1283a26–29, 1283b19–30; 3.10.1281a28–34. See also Coby, “Aristotle's Four Conceptions of Politics,” 490–92; Devereux, “Aristotle on the Just Distribution.”

44 Newell, “Superlative Virtue,” 174.

45 Ober, “Nature, History, and Aristotle's Best Possible Regime,” 228–30.

46 Devereux, “Aristotle on the Just Distribution,” 5. See also Pol. 3.13.1283b27–30, 1283a26–29.

47 See 7.14.1332b12–27. Pangle stresses the same point, arguing that the divine monarch represents the “most superior” regime, while the regime described in books 7 and 8 is the “best republican regime,” which is to say, “the best political community for a multitude of citizens of ‘similar’ civic capacity” (Aristotle's Teaching, chap. 5, esp. p. 225). In section 4, I reject the claim that divine monarchy is the best regime, if it even counts as a “regime” at all.

48 If we construe stability according to the strong variant of (iv)(a), however, this may not hold. See 4.9.1294b35–40, where Aristotle seems to suggest that this stronger notion of stability is the only one unqualifiedly compatible with correct regimes.

49 Equality may also exist (to some extent) in deviant constitutions; those constitutions fall short, however, of being completely or unqualifiedly equitable (5.1.1301a36–38).

50 These points echo Plato's justifications for excluding the producer class in Republic.

51 Unless a finer distinction is made, I use “best” throughout this section to indicate both what is most just and most choiceworthy.

52 Note that principle (i) can only settle what is best in the sense of being most just; it does not itself bear (directly at least) on the question of what is most choiceworthy.

53 This is Kraut's view of book 3 (Aristotle, 225n63, 359, 359n4, 416–18). Others, however, interpret book 3 as a defense of democracy or polity: e.g., Lindsay, Thomas K., “Aristotle's Qualified Defense of Democracy through ‘Political Mixing,’” Journal of Politics 54, no. 1 (1992): 101–19; Lindsay, , “Liberty, Equality, Power: Aristotle's Critique of the Democratic ‘Presupposition,’” American Journal of Political Science 36, no. 3 (1992): 743–61; Nichols, Mary P., Citizens and Statesmen: A Study of Aristotle's “Politics” (Lanham, MD: Roman and Littlefield, 1991); Salkever, Stephen, Finding the Mean: Theory and Practice in Aristotelian Political Philosophy (Princeton: Princeton University Press, 1990); and Bates, Clifford Angell, Aristotle's Best Regime: Kingship, Democracy and the Rule of Law (Baton Rouge: Louisiana State University Press, 2003).

54 This explains the various instances in which Aristotle appears to regard kingship as the best constitution (e.g., 4.2.1289a39–b5; cf. Nic. Eth. 8.10.1160a31–36).

55 For an argument that aims to demonstrate precisely this, see Devereux, “Aristotle on the Just Distribution.” For other attempts to show that Aristotle considers “aristocratic democracy” or “democratic aristocracy” best overall, see Frank, Democracy of Distinction; and Ober, Josiah, “Aristotle's Natural Democracy,” in Aristotle's “Politics”: Critical Essays, ed. Kraut, Richard and Skultety, Steven (Lanham, MD: Rowman and Littlefield, 2005).

56 This qualifying clause is significant in light of 3.13.1283b19–23.

57 Although it is generally uncontroversial to regard the ideal constitution in 7 and 8 as an aristocracy, note that Ernest Barker argues against this (Barker, The Politics of Aristotle [London: Oxford University Press, 1958], 153).

58 On this point, see Hansen, Mogens, Reflections on Aristotle's “Politics” (Copenhagen: Museum Tusculanum Press, 2013), 96–97.

59 This condition is part of what makes the ideal constitution “blessed” (7.13.1332a20–30; 7.2.1324a23–25). Note also that the similarity and equality found in the ideal constitution suggest that it would possess an extreme amount of civic friendship (3.16.1287b32–34).

60 This excludes children, who are yet to possess full practical wisdom (3.1.1275a13–18; 8.5.1339a29–32; 7.14.1332b31–1333a3).

61 It includes everyone capable of acquiring the principal virtues. Of course, other constitutions may be as inclusive but without awarding each citizen equal political power; e.g., a polity may include all citizens capable of acquiring virtue, but restrict some from holding certain offices. Moreover, some constitutions may be even more inclusive, say, by allowing even those citizens who are incapable of achieving virtue in the strict sense (e.g., hoi banausoi) to participate in the constitution, as is the case in unmixed democracies.

62 Coby draws attention to this point (“Aristotle's Four Conceptions of Politics,” 491).

63 4.7.1293b5–6; 7.13.1332a30–35; 7.14.1333a10–15; 5.12.1316b9–10.

64 See esp. 1288a37–39. Kraut also links Aristotle's argument in 3.4 to his endorsement of aristocracy, though he does not develop the claim in the way I do here (Aristotle, 359, 366–68, 428n1).

65 See 3.10; 3.17.1288a9–12; 3.5.1278b4. Kraut draws attention to this possibility (Aristotle, 359, 359n4, 366–68), but associates it with kingship and aristocracy simpliciter; he also argues that permanent rule of the few over the many is characteristic of the best constitution—a point I contest here and below.

66 3.4, esp. 1277a20–24. See also Pangle, Aristotle's Teaching, 111–21.

67 This is suggested at 7.14.1332b21–23; 3.15.1286b3–21. See also Newell, “Superlative Virtue,” 174; Pangle, Aristotle's Teaching, 156.

68 Garver also raises doubts about whether divine monarchy is a form of “political rule,” or more akin to household management (Aristotle's Politics, 100–102).

69 (1) is ranked highest not only because it is completely just and beneficial, but because it is the most “blessed” and most compatible with civic friendship (7.1–2; 7.13; 3.16.1287b32–34).

70 The ordering of the subforms of well-mixed aristocracy is based on 4.7.1293b1–21.

71 (2a) is ranked highest because in balancing all three claims, it is most completely just and, for this reason, presumably most stable.

72 (2b) is ranked higher than (2c) because Aristotle considers democracies more stable than oligarchies, making an aristocracy that privileges the claim of the people more stable and so more choiceworthy than one that privileges the claims of the rich.

73 By not intentionally distributing political power in accordance with virtue, polity does not adhere to principle (i) in a strict or deliberate sense, making it less excellent than well-mixed aristocracy. That said, polity still outranks unmixed aristocracy, reflecting Aristotle's claim that certain multitudes, when acting collectively, can “be better than the few best people” (3.11.1281a40–b1).

74 (3a) is ranked above (3b) because polities that privilege wealth are more aristocratic in nature, since wealth tends to secure education and the material preconditions of virtue; a properly calibrated property assessment ensures that as many individuals so educated and equipped are able to take part in the constitution as is possible (see, e.g., 4.8).

75 Even though (3b) may be more stable than (3a) by dint of being more democratic, it is ranked lower than (3a) for being less just (which is a higher-order good than stability).

76 (4) emphasizes justice perhaps at the cost of stability, though it fulfills principle (i) and, at least in archaic times, may have been relatively stable.

77 (5) is ranked below (4) because kingship has fewer virtuous people.

78 (6a) and (6b) outperform (7) and (8) both in justness and stability. (6a) is ranked higher than (6b) because, while both are only partially just, democracy is more stable (and more conducive to civic friendship). On these points, see Kraut, Aristotle, 432, 444–51; Ober, “Nature, History, and Aristotle's Best Possible Regime”; Frank, Democracy of Distinction, 167–78.

79 (7a) and (7b) perform poorly on principles (iii) and (iv). (7a) is ranked higher than (7b) because democracy is more stable.