Toward the end of Politics V. 12, Aristotle criticizes Plato’s discussion of political change in Republic VIII-IX. Scholars often reject Aristotle’s criticism, especially because it portrays Plato’s discussion, allegedly unfairly, as developing a historically testable theory. I argue that Aristotle’s criticism is adequate, and that the seriousness with which he considers Plato’s account of political change as an alternative to his own is both warranted and instructive. First, apart from criticizing Plato’s account for its historical inaccuracies, Aristotle also exposes theoretical insufficiencies and internal inconsistencies within it. Second, Aristotle’s criticisms of historical inaccuracies in Plato’s discussion of political change are not misdirected, since there are reasons to think that Plato does intend that discussion to accord with the historical facts.
Toward the end of Politics V. 12, Aristotle criticizes Plato’s discussion of political change in Republic VIII-IX. This criticism is often taken to be unjustified or misdirected on several points. In particular, it has been suggested that Aristotle unfairly portrays Plato’s discussion as developing a substantive and historically testable theory of political change, similarly to Aristotle’s (but inferior to it), whereas Plato’s purposes are in fact entirely different.1 Indeed, as I shall show in section 2 below, Plato makes it clear that his account of political change is intended to establish a fact not so much about politics at all, but rather about human psychology – it is intended to show that cultivating justice in one’s soul is preferable for the individual over the alternative. Political history seems to be beside the point. However, Aristotle’s criticisms of Plato’s account of political change should not be so readily dismissed, or so I shall argue (in sections 3-6 below). First, these criticisms do not merely challenge the historical accuracy of Plato’s account. Aristotle argues, further, that Plato’s account in Rep. VIII-IX suffers from both theoretical insufficiencies and internal inconsistencies. Second, Aristotle’s charge of Plato’s account of political change as being historically inaccurate should be taken seriously as well. Even though it is not Plato’s only or even primary concern to present a correct or complete account of political change, there are reasons to think that he would nevertheless wish for his account to be historically accurate, and that failing to do so would compromise his overall project in various ways. I survey Aristotle’s different types of criticism of Plato’s Republic VIII-IX, viz. charges of theoretical insufficiencies, detection of internal inconsistencies, and criticisms concerning historical inaccuracies. I show that these criticisms are adequate, and that, furthermore, the seriousness with which Aristotle considers Plato’s account of political change as an alternative to his own is both warranted and instructive.
2 The Plan of Plato’s Republic VIII-IX
At the end of Republic IV (444e-445e), Socrates returns to the challenge put forth by Glaucon in Book II (357a – ff.), and which prompted the discussion in the ensuing parts of the dialogue – he sets out to establish that justice is preferable over injustice in every way. Glaucon retorts that this task has already been accomplished. At this point in the dialogue, Socrates has already delineated the nature of justice. He has also shown that justice is analogous to health, which grounds, to his interlocutor’s satisfaction, the status of justice as a good of the highest kind, which is also preferable over its contrary in every conceivable way. Socrates, however, insists on continuing the discussion, so as to establish its conclusions as clearly as possible (ὅσον οἷόν τε σαφέστατα) (445b5-7).
In particular, Socrates sets out to distinguish between five different types of constitution (corresponding to five different types of human soul [445c-d]) – one of which (viz. the kallipolis) is ‘good’ and ‘correct’ while the others are ‘bad’ and ‘misguided’ (Rep. V, 449a) – and to explain the way in which these change from (and into) one another (449a-b). The purpose of Books VIII-IX is to supply such an account, after the topic has been diverted at the beginning of Book V by Adeimantus and Polemarchus (449b). Socrates is clear on the point that accounting for the good and just constitution (and soul-type) and comparing it to bad and deviant ones is intended to settle the issue concerning the preferability of justice over injustice, so that ‘either, being persuaded by Thrasymachus, we shall pursue injustice or, [being persuaded] by the argument which is now forthcoming, [we shall pursue] justice’ (VIII, 545a-b).
To the above end, Republic VIII-IX describe in detail the way in which the ‘good’ and ‘correct’ constitution mentioned in Book IV (viz. the kallipolis) deteriorates into progressively worse and more misguided ones (first into a timocracy, then an oligarchy, a democracy and, finally, a tyranny). They also describe, in detail, a parallel process of deterioration in the types of soul corresponding to those constitutions. Indeed, establishing the comparison between these different types of soul seems to be the primary concern of the discussion, since it is the main purpose of Books VIII-IX (as it is, in a sense, of the Republic as a whole) to determine whether an individual person should pursue justice over injustice, and why. The primary reason why the discussion of soul-types is accompanied by a discussion of types of constitution is that, as we are already in a position to know based on the ‘letters analogy’ of Book II, the (moral) character of cities is both more clearly given than, and once given is then indicative of, that of the human soul (545b-c).
As to be expected, the discussion in Rep. VIII-IX reveals that the most just constitution and soul-type are also the ones in the most desirable state. Of the remaining constitutions and soul-types, the more a constitution or soul is unjust the less desirable its state, culminating in tyranny and the tyrannical man, who is a miserable person (e.g. VIII, 567e-568a). By the end of Book IX it is concluded, in large part on the basis of the process of political change and the corresponding process of deterioration in the soul, that ‘indeed in every way he who praises just things speaks the truth, whereas he who [praises] unjust things speaks falsely’ (589b8-c1).
Plato lays such a noticeable emphasis on the inquiry into justice in the individual human soul, in fact, that it is tempting to conclude that the entire discussion of political change in Republic VIII-IX functions merely as a means toward that inquiry, rather than a serious attempt at political history or theory. On such an interpretation, potential difficulties with the account of political change given in that text, such as historical inaccuracies, would not compromise Plato’s project. Most importantly for our purposes, as Skultety puts it, on such a view ‘regardless of how many discrepancies Aristotle would like to point out between actual history and Plato’s theory, he will be missing the point’, in so far as Plato’s proposed theory of political change ‘is little more than a literary device’.2 Indeed, Skultety agrees that ‘[i]f Aristotle’s attacks were limited to the empirical criticisms of Pol. V. 12, such responses would certainly dull the edge of his critique’.3 He therefore recommends looking for adequate criticisms of Plato’s theory of political change, not in Pol. V. 12, but rather in V. 1-4.4 By contrast, I shall argue that a careful examination shows that, even though Plato was mainly interested in accounting for individual justice, there is still room for all of Aristotle’s criticisms of Republic VIII-IX presented in Pol. V. 12.
3 Theoretical Insufficiencies
The first set of arguments to be considered are Aristotle’s criticisms of certain gaps in Plato’s account of political change. These have mainly to do with Plato’s accounts of the causes of change in certain types of polis, the mistaken assumptions underlying them, and the various points that they either leave unexplained or simply leave out. First, Plato claims that the cause of change in the kallipolis is unique to it. But he does not explain why this cause (viz. the fallibility of human beings qua percipient beings) could not cause change in other types of polis, and indeed in non-political aggregations (even those of non-human animals). Second, Plato assumes that this cause of change in the kallipolis would not hinder the initial successful construction of the kallipolis, but this too requires an explanation. Third, Plato thinks that a polis with a certain type of constitution (say, an oligarchy) can come about without adopting the model of justice distinctive to such a constitution (say, wealth-based justice). Aristotle shows that this idea is mistaken, and that it rests on the dubious presupposition that it is the psychological proclivity of the leaders in a polis that determines the identity of its constitution. Fourth, Plato thinks that there can only be one possible cause of change in an oligarchy. Aristotle challenges this idea by enumerating many possible events which could bring about such a change (as many, in fact, as would trigger the change in the model of justice adopted by the city in question).
Aristotle’s first criticism of Plato’s Socrates’ account of political change in Republic VIII has to do with the first change mentioned in that text, viz. the change of the kallipolis into a Sparta-like timocracy. There (546a-547a), Socrates (reporting the words of ‘the Muses’) says that, since everything coming into being must also dissolve, so must the kallipolis. Specifically, the kallipolis will necessarily have rulers who, despite being as knowledgeable as humanly possible, are nevertheless only human and are therefore liable to mistakes. Thus, these rulers, who, as Socrates has made clear in Book V, are in charge of procreation through fixing marriages among the guardians of the kallipolis, would inevitably at some point miscalculate the ‘fertility and barrenness of human beings’, viz. make mistakes in applying the ‘human number’ (more on this below). This would ultimately result in the deterioration of the population and mixture of classes within the city, preparing the ground for the dissolution of the city’s constitution.
Aristotle’s criticism of this account begins with his contention that Socrates does not attribute to his kallipolis a change specific to it (1316a3-4: οὐ λέγει τὴν µεταβολὴν ἰδίως). Aristotle explains this statement by analyzing Socrates’ view into two main claims, which he then goes on to criticize (1316a4-14):
For [Socrates] says that:
(1) The cause (αἴτιον) [of the change] is that nothing remains but rather [everything] changes in some cycle (ἔν τινι περιόδῳ).
(2) And the source (ἀρχὴν) [of the change] lies in those [numbers] ‘of which the bases giving four thirds, coupled with five, give two harmonies’ [cf. Rep. VIII, 546c2-3], whenever the number of the diagram, he states, becomes cubed, meaning that (ὡς):
(2b) nature sometimes generates [people who are] simple-minded (φαῦλοι) and too bad for education. In fact, in saying this he is perhaps not [speaking] badly, for it is possible that it would be impossible for some men to be educated and to become excellent.
(3) But why would this change be peculiar to what he calls the best constitution, more than to all the others and indeed all things coming into being?
Proposition (1) requires clarification right at the outset. Specifically, though it may seem that Aristotle is here attributing to Plato a view of cyclical political change, that is not the case. Plato does introduce a pattern of political change, which he argues political constitutions, if and when they change, must follow, and Aristotle will criticize him later on for holding that view (see section 4). However, as we shall see, and as Aristotle himself recognizes, Plato is not explicitly committed to the view that his proposed pattern of political change repeats itself cyclically anywhere in Republic VIII.
The cycles discussed in Plato’s text appear first when Socrates speaks of plants and animals as having their period of fertility and barrenness determined by the time ‘when the circuits conjoin the circumferences of the cycles (κύκλων) of each’ (546a4-6). The ‘circuits’ determining the periods of fertility and barrenness of animals and plants, Socrates says, are ‘of a short route for the short lived, and the opposite for the opposite’ (546a6-7). These suppositions are crucial for Plato’s account of political change, since they explain how even an ideal polis, once established, will necessarily change, in particular deteriorate until it is destroyed. This is so because maintaining that polis depends on the recognition by its rulers of the aforementioned cycles determining the periods during which human beings should procreate in order to generate optimal offspring, and the application of that information in fixing marriages.
The regular and circular change Plato mentions in the text, then, is of the human beings within the polis, rather than a change of constitutions. Plato’s view that everything exhibits circular change does not of course entail that every change must be circular. On that view, poleis exhibit circular change, e.g. in so far as their populations are subject to periodic fertility and barrenness, and perhaps in some other ways as well. But poleis can still undergo further, non-cyclical, processes of change, and one such process may well be what we have termed ‘political change’, namely the substitution of one constitution for another. Thus, the cyclical change mentioned in Plato’s text is only relevant to political change in so far as the generation of inferior human beings in the polis, due to procreation at the wrong period, would result in the deterioration of the polis (in this discussion, specifically the kallipolis). This is why Aristotle, in formulating proposition (1), says that the fact that nothing remains but rather everything changes in some cycle is the cause of political change. In saying this, Socrates does not ‘rely … on the simple claim that “nothing remains”, that is, that over time all things change (metaballein), even forms of rules’.5 Rather, Socrates identifies specific cyclical changes relevant to determining the nature and direction of (noncyclical) political change.
Next in Plato’s text, περίοδος (the word Aristotle actually uses in formulating proposition (1)), is mentioned at 546b4, in Socrates’ prelude to his discussion of the ‘human’ or ‘nuptial’ number. This is the number which reveals or determines the periods of fertility and barrenness in human beings as mentioned above, and which the rulers of the kallipolis ought to use, and would necessarily occasionally misuse, in order to fix marriages among the guardians. At 546b4, then, Socrates says that there is a perfect number encompassing the cycle (περίοδος) which determines divine births. Here, again, περίοδος refers to that cycle which determines the period in the life of certain beings (here, gods)6 relevant to calculating the best time for them to procreate (the discussion is then followed by the exposition of the number specific to human births; see below). Again, the reference is not to any cyclical view of political change. This point will be especially important to keep in mind once we get to Aristotle’s evaluation of Plato’s view of the political change in tyranny. For Aristotle asks there whether tyranny changes in Plato’s view, and if so, whether it necessarily changes back into the kallipolis. The answer to the second question is no, precisely because Plato does not commit himself to a cyclical view of political change.
Proposition (2) corresponds to Plato’s discussion of the ‘human number’. The details of that discussion are notoriously obscure, and need not concern us here.7 The point relevant for our purposes is that Socrates mentions this number as the one determining the appropriate time at which people (specifically, the guardians of the kallipolis) ought to procreate. It is this fact about the number which makes it (or the numbers leading up to it) constitutive of the ‘principle’ or ‘source’ (ἀρχή) of the change of the kallipolis mentioned above. For, it is by miscalculating this number or misapplying it to practical cases that the rulers of the kallipolis bring about the deterioration and eventual destruction of their city.
Aristotle goes on to explain proposition (2) by saying (and in fact agreeing with Socrates on the fact) that nature occasionally produces simple-minded and uneducable people (2b). In saying this, Aristotle does not distort Socrates’ view, although he does portray it in a way which is conducive to his own critical purposes. In a sense, it is Socrates’ view that nature is responsible for the deterioration of the kallipolis by producing uneducable people, since it is the nature of the rulers in the kallipois, viz. human nature relying on sense perception in determining marriages,8 which makes them liable to error (Rep. VIII, 546a7-b4). But by representing Socrates’ view in this light Aristotle sets the ground for his upcoming criticism of it.
According to the criticism in question (proposition (3) above), it is not clear why the change attributed to the kallipolis should be peculiar to that constitution, without applying either to constitutions generally speaking or even to all things coming into being. The relevance to this criticism of the previous representation of Plato’s text as conveying the idea that nature occasionally produces inept people is clear. For if human nature in general is capable of leading rulers to mistakes ending in such results, then that should be the case in constitutions other than Plato’s ideal one as well. Indeed, Aristotle seems to think that similar errors could cause political change in other constitutions. Ober argues (citing Pol. III. 15) that for Aristotle ‘[t]he regime change from kingship to “politeia” does not … happen (as in Plato’s Republic) because of an error on the part of the rulers’,9 and that, again unlike Plato, Aristotle believes the ‘native-born males’ of the best polis will not be ‘inherently … uneducable’, which explains why there is, as Ober puts it, ‘no hint of any’ provisions relevant to dealing with such occurrences in Aristotle’s Politics.10 But Aristotle does also dedicate an entire chapter of his Politics (i.e. VII. 16) to discussing the ways in which legislators must supervise and optimize marriages and childbearing in his own ideal polis (which of course differs in constitution from the kallipolis), apparently echoing Plato’s concern with regard to the eventual deterioration of the kallipolis. In any event, Aristotle’s point that the errors Plato ascribes to the rulers of the kallipolis, if they are to apply to any constitution at all, should not apply exclusively to that best city, already refutes the idea, to which Aristotle thinks Plato is committed, that the change of the kallipolis is unique to it.
But Aristotle’s criticism need not (and does not) stop there. For, if human nature is liable to error even when granted the highest form of philosophical apprehension, as in the case of the philosopher kings, one would assume that under normal conditions the nature of non-rational living things would sometimes lead them to produce inferior offspring by procreating at the wrong time as well (though, in their case, of course, this would not result from an error in judgement or calculation of the number determining good births). Presumably, similarly to the human case, such results would bring forth the deterioration of the group (e.g. the herd, flock, etc.) to which these specimens belong, and ultimately possibly its demise. Thus, Aristotle shows, based on a calculated, though not misguided, reformulation of Plato’s own point about the process of change in the kallipolis, that such a process need not be peculiar either to that polis or even to political regimes in general.11
Aristotle continues his discussion of Plato’s account of political change in the kallipolis with a second criticism (1316a14-17). With everything changing through time as Socrates says, would things which do not begin to come about together nevertheless change together? He exemplifies as follows: If a thing A comes to be after a thing B by one day of the cycle (τῆς τροπῆς), would A and B change simultaneously? This example is curious, since in it Aristotle seems to attribute to Plato a cyclical view of time and change (presumably including political change). Since this view is foreign to the text of the Republic, it has been suggested that Aristotle is here responding to a discussion in another Platonic dialogue, i.e. the Statesman.12 If we are to retain the example, then the discussion of 1316a14-17 as a whole should probably be taken to be a marginal remark concerning Plato’s view outside of the Republic.
However, it should be mentioned that the example is in fact omitted in at least one of an authoritative family of manuscripts (viz. cod. Ambrosianus 126). But if this bit of text, i.e. οἷον… µεταβάλλει at 1316a16-17, is to be excised, then we can make sense of the question preceding it as making a point about the relevant discussion in Republic VIII. In this case, Aristotle could be taken to pose a further challenge to Socrates’ idea that the kallipolis changes as a result of the change in its population, in particular the generation in it of uneducable people. When Aristotle asks whether, on Socrates’ proposed view, things which do not begin to come about together would nevertheless change together, he might have in mind as such things specifically the kallipolis and the nature shared by all human beings, including the philosophers establishing and ruling over that city. The kallipolis is supposed to come into being through the separation into classes by the philosopher kings, who would then control births in order to retain that social structure. But, if change in the kallipolis begins with the misjudgments and miscalculations of those very rulers about marriages, which they are sure to commit due to their human nature, then why should we assume that similar misjudgments would not occur already at the earliest stage of establishing the kallipolis (viz. the process Socrates describes as the philosophers painting the constitution on a clean slate; cf. Rep. VI, 500d-501c)? In order to assume that they do not, and in order for the kallipolis to come into existence in the first place (let alone survive), we would need to assume that human nature, viz. the nature of the philosophers, changes at some point, in particular, that it becomes liable to error (after a period of functioning optimally). That point would mark the beginning of the change process of the kallipolis, viz. its deterioration. But there is no reason to assume that human nature changes in such a way. So, a fortiori, there is no reason to assume that it changes along with the kallipolis.
This point leaves the Platonic position with two options. The first is that the kallipolis potentially stays in existence forever (contra Socrates himself; cf. Rep. VIII, 546a1-3), since human nature is not fallible enough to cause the philosopher kings necessarily to be incapable either of establishing the polis or of retaining it. As Garver suggests, according to the model of political change which Plato adopts, and which Aristotle rejects, constitutions alter only as a result of internal defects or imperfections, whence it should follow that a city with a perfect constitution would remain fixed.13 It is at the risk of denying absolute perfection to the kallipolis, then, that Plato explicitly attributes to it an internal cause of change, namely the inevitable errors on the part of its rulers. Admitting such an imperfection, however, leaves the Platonic position with only one option: the kallipolis could not even come into existence, since the fallibility of human nature precluding philosophers from retaining the polis would also render them incapable of constructing it (except, perhaps, by some lucky coincidence, which is surely not the way Plato would think this polis should come about).
We find criticisms of a similar kind in Aristotle’s discussion of Plato’s view of the political change into and from an oligarchy. First, Aristotle states that it is absurd (ἄτοπον) to think that constitutions change into oligarchies due to the fact that those in political office are ‘money lovers and money-getters’ (φιλοχρήµατοι καὶ χρηµατισταὶ) (Pol. V. 12, 1316a39-b1), as Plato does (cf. Rep. VIII. 550c-551b). Rather, Aristotle thinks, the relevant fact should be that the wealthy think it is unjust (οὐ δίκαιον) for the less-than-wealthy to enjoy equal political rights and power with them (1316b1-3). This remark echoes his idea that types of constitution are determined on the basis of the specific models of justice operative in them. Whereas constitutions generally adopt a model of distributive justice, determining the citizens’ share in the constitution based on merit (ἀξία), there are of course various ways of determining the merit of citizens (Pol. III. 9; V. 1; cf. NE V. 3).14 Thus, in democracies, for example, the merit of citizens is determined by their freedom (and is hence equal), and in oligarchies it is determined by their wealth. Oligarchy, in particular, came into being ‘from people being unequal with respect to one thing taking themselves to be entirely unequal (for, being unequal with regard to wealth they take [themselves] to be unequal without qualification)’ (Pol. V. 1, 1301a31-3).
Aristotle takes Plato’s account of the political change into an oligarchy to be mistaken, because that account makes it possible for an oligarchy to come about without the oligarchic model of distributive justice coming into effect. In Plato’s account, an oligarchy comes into being as a result of those in office being themselves preoccupied with money (Rep. VIII, 550c-551b). But, as Aristotle says, it is perfectly possible for people with such a preoccupation to run a democracy, and moreover to wish for their constitution to remain a democracy, and to do so consistently. Indeed, in some democracies, those in political office are permitted to, and in fact do, make money (Pol. V. 12, 1316b5-6). So, at least in certain cases, officials who are ‘money lovers and money getters’ would not turn their constitution into an oligarchy (and would not and should not aspire to do so). And, by the same token, in many oligarchies those in political office are legally prevented from making money (1316b3-5), so that it is perfectly possible for officials who are not ‘money lovers and money getters’ to run an oligarchy, while again remaining consistent with their proclivities. By contrast, Aristotle would argue that it is impossible for a city either to remain an oligarchy while abandoning its model of wealth-based justice or to remain a democracy while abandoning its model of freedom-based justice. An oligarchy, for Aristotle and contra Plato, is not that type of regime in which the rulers are motivated by wealth. Rather, it is that type of regime in which people ‘rule because of [having] wealth’ (διὰ πλοῦτον: Pol. III. 8, 1279b39-1280a2).
The difference between Aristotle and Plato on the issue of the causes of political change and faction has been understood in several different (sometimes incompatible) ways. Saxonhouse, for example, argues that Aristotle deviates from Plato’s account in the Republic in so far as he presents a unified psychological cause of stasis, essentially and invariably including ‘jealousy and resentment’, whereas for Socrates such causes differ from one regime to the next (and in the case of the kallipolis are not specified at all).15 Skultety, by contrast, argues that it is precisely for positing a single root cause for faction, in particular appetitive desire, that Aristotle criticizes Plato, and that ‘if we had to pick just one element to play the leading role in the Aristotelian drama of stasis, it would surely be the content of the agents’ rational beliefs concerning what is just, appropriate, and beneficial’.16 Surely, Aristotle does think there are many causes of political change and stasis.17 He enumerates and discusses such causes in Pol. V. 1-4, followed by a discussion of further causes of stasis in specific kinds of constitution in V. 5-7. But, rather than simply criticizing Plato for stating either too many or too few causes of political change, or for focusing on either desires or beliefs as such causes, the point of Aristotle’s criticism in 1316a39-b3 is that Plato completely assimilates political change and its causes to psychological change in individual citizens and its causes.
For Plato, psychological change in individual members of a polis is necessary and sufficient for explaining the change in the constitution of their polis. By contrast, in Aristotle’s theory, such psychological causes are only one side of the coin. Aristotle views the polis as a hylomorphic compound, whose form and matter are, respectively, its constitution and citizenry (Pol. III. 3, 1276b1-13). Thus, as Polansky points out, Aristotle should explain political change as he explains change in general (in Physics I. 6-7), i.e. as the taking on of a new form (here, a new constitution) by an underlying matter (here, a population).18 On that view, psychological features of individual citizens, which constitute part of the matter of the polis, may only bring forth political change by triggering a change in the form of the polis, i.e. in its constitution. In particular, as we have seen, in order to cause political change, such features of individual psychology (or whichever other relevant factors or events there may be) would have to enact and render operative a new model of justice in the polis. For instance, to use a case described by Polansky, political change may take place when the beliefs by individuals concerning the type of constitution appropriate for their city (a material fact) clashes with the type of constitution currently instituted in that city (a formal fact).19 Such an explanation would advance over Plato’s model, which reduces all causes of political change to psychological facts.20 Focusing on the case of oligarchy, but extending to the general account of political change in the Republic, Aristotle’s criticism in 1316a39-b3 charges Plato with just such a reductionist approach.
Aristotle seals the discussion of Plato’s account of oligarchy with a criticism of his view of the change from oligarchies into another type of constitution (in Plato’s view, specifically into democracy). As Aristotle notes, Socrates identifies only one cause for such a change in Republic VIII: the citizens becoming poor as a result of mishandling their finances (1316b14-17; cf. Rep. VIII, 555b-e). Aristotle agrees that such an occurrence can (under some circumstances, and by no means always) lead to a change in an oligarchy (1316b18-20). But he thinks Plato is wrong in focusing on it as the only cause of the change in question. As Aristotle states even prior to mentioning Plato’s explicit view on this point, oligarchies may change into democracies ‘without anyone becoming poorer’, e.g. by the poor becoming a majority (the reverse type of change may come about by the rich gaining control) (1316b10-14).21 Oligarchies may also change due to people causing stasis in reaction to being mistreated or dishonored, even if they do not lose their money or possessions in the process (1316b21-5; cf. V. 2-4, 6). In fact, in Politics V. 6 Aristotle describes a whole host of further causes of faction in an oligarchy, which are not necessarily tied up with financial losses either, from internal conflicts between oligarchs to the oppression of the rich who are not in office by the oligarchs.
Garver argues that it is by shifting from speaking of political change as caused exclusively by internal imperfections to speaking of factions as external causes of such a change that Aristotle does away with Plato’s view that each constitution has ‘its own way of passing away’ (and that each constitution can only change into one particular type of constitution).22 It is of course true that Aristotle does not think it is only internal defects in the constitution that lead to political change. But even if he did, he still would have been able to criticize Plato for positing only one kind of internal cause of change per constitution type, since he lists numerous and various internal causes of change in a single type of constitution, e.g. in the case of oligarchy, as we have just seen (and similarly for the direction of change).
It is important to note that the aforementioned theoretical gaps which Aristotle exposes in Plato’s account of political change do not assume that this account is intended to be historically correct. Aristotle, as we shall see, does provide criticisms that do assume such an intention. The criticisms discussed so far, however, focus on the adequacy of the argumentation and explanations used in Plato’s account, regardless of their correspondence with the historical records. It is true that Aristotle appeals to examples from political history in some of these criticisms. For instance, he appeals to the fact that in some democracies those in power are permitted to make money in arguing against Plato’s point that a city counts as an oligarchy due to the preoccupation of its rulers with money (cf. 1316b5-6). But here the historical data are used in order to expose an inadequacy in Plato’s very conception of an oligarchy. Plato is not here criticized for getting the history wrong – he is criticized for misunderstanding what an oligarchy is.
4 Internal Inconsistencies
The next set of criticisms, similarly to the first, is not concerned with the historical testability of Plato’s account of political change. Rather, by presenting this set of criticisms Aristotle is interested in exposing the following contradictions or inconsistencies latent in Plato’s account: First, although Plato does not mean to present a cyclical pattern of political change, his principles commit him to upholding just such a pattern. Second, although Plato insists on the point that the kallipolis (alone) is unified and indivisible, his theory nevertheless commits him to the view that the kallipolis is divisible into ‘two poleis’, similarly to (his conception of) poleis with other types of constitution.
The first criticism occurs in the context of Aristotle’s discussion of Plato’s account of change in a tyranny. Aristotle points out the omission in Rep. VIII of a discussion of whether tyrannies change, and, if so, how and for what reason (1316a25-6). Indeed, as Aristotle goes on to say, this would have been a difficult issue for Socrates to address (1316a27-9):
For it [sc. the issue of whether or not tyrannies change] is indeterminate, since according to him [sc. Socrates] [tyrannies] should [change] into the first and best [constitution, viz. the kallipolis], for thus it would become continuous and cyclical.
ἀόριστον γάρ, ἐπεὶ κατ’ ἐκεῖνον δεῖ εἰς τὴν πρώτην καὶ τὴν ἀρίστην· οὕτω γὰρ ἂν ἐγίγνετο συνεχὲς καὶ κύκλος.
Aristotle’s point here has been taken to be that Plato intends for his pattern of political change to be cyclical.23 But, as we have seen, Plato’s text indicates no such thing.24 Alternatively, we could take ‘οὕτω γὰρ ἂν ἐγίγνετο … κύκλος’ to be explanatory, not only of ‘κατ’ ἐκεῖνον δεῖ εἰς τὴν πρώτην καὶ τὴν ἀρίστην’, but rather of ‘ἀόριστον γάρ … τὴν ἀρίστην’. On that reading, Aristotle would be taken to say, more plausibly, that in introducing his pattern of political change Plato did not mean for it to turn out to be cyclical. However, his principles incidentally led him to such a conclusion, possibly unbeknownst to himself.
Plato’s envisaged pattern of political change in Republic VIII, as we have seen, is a process of deterioration, from the kallipolis to a timocracy, an oligarchy, a democracy, and, finally, a tyranny. Since Plato’s pattern excludes the change of one type of constitution from or to more than one other type, we would expect tyranny to change into the only constitution left for it to change into, and that is the kallipolis. The resulting cyclical view of political change, Aristotle charges, at least requires some defense or explanation, as it is ‘indeterminate’ (ἀόριστον).25 It is also at odds with what seems to be Plato’s own intention. Indeed, if Aristotle thought that Plato’s Socrates did originally say that the pattern of political change is cyclical, then he could not have seriously reported that Socrates does not say whether tyranny changes or not (οὐ λέγει: 1316a25-6), as a conclusion would have been directly implied by the very assertion regarding the cyclical pattern.
There is, of course, one further alternative available to Plato, which could save him from endorsing the circular view of political change (which he does not seem to be interested in defending), and hence from inconsistency. The alternative is that tyrannies do not change into any type of constitution. Aristotle recognizes that this may well be Plato’s considered view, and he responds to it directly, as we shall see in the next section. For now, all we need to point out is that Aristotle manages to show that, given Plato’s theoretical principles, he (Plato) cannot consistently attribute political change to tyrannies within a non-cyclical paradigm of political change.
The second charge of an internal inconsistency occurs in Aristotle’s discussion of Plato’s explicit view in Republic VIII that an oligarchy is ‘two poleis’, viz. a city of the rich and a city of the poor (1316b6-7; cf. Rep. VIII. 551d). Aristotle suggests that there is no good reason why oligarchies alone should be divisible in this way. If an oligarchy can be said to be composed of two distinct cities by virtue of having two separate classes of citizens differentiated on the basis of wealth, then a timocracy, for instance, can be equally said to be divided into two cities by virtue of having two separate classes of citizens differentiated on the basis of honor, and the same would be true for any constitution whatsoever (1316b8-9: ὁποιασοῦν ἄλλης). Interestingly, the next example that Aristotle goes on to give is of a city which would be distinguished into two poleis by virtue of having two classes differentiated on the basis of whether or not the citizens in them are ‘good men’ (1316b9-10: ἀγαθοὶ ἄνδρες). The reference here is to Plato’s kallipolis, in which guardians by far supersede in virtue and goodness members of the class of producers.
According to Aristotle’s criticism here, then, the same considerations that have led Socrates to maintain that an oligarchy is divisible into two cities necessarily lead to the conclusion that the kallipolis is divisible into two poleis as well. But Plato is of course committed to the view that the kallipolis, unlike non-ideal constitutions, is an indivisible unity (Rep. IV. 422e-423b). This kind of unity, which Plato takes to be unique to the kallipolis, is the main target of Aristotle’s criticism of the Republic in Politics II. 1-5. There, Aristotle’s main point is that the unity Plato envisages for his kallipolis is exaggerated, as it befits a household or even a human being rather than a polis (Pol. II. 2, 1261a15-22). Here (V. 12, 1316b6-10), Aristotle complements that criticism by adding the further point that the kallipolis could not be uniquely unified in the way that Socrates proposes since it is at least as divisible into classes as non-ideal poleis are.26
This further criticism focuses on obliterating the fundamental distinction that Plato draws between his ideal city and all others, and it therefore would have been helpful for Aristotle’s argumentative purposes to stress Plato’s explicit view of the divisibility of all non-ideal poleis. The fact that Aristotle focuses on the case of oligarchy, which he then goes on to contrast first with timocracy and then with the kallipolis and with ‘any [polis] whatsoever’, should not lead us to conclude that his criticism ‘implies that Plato thinks that only oligarchic cities split into opposing sections’.27 Rather, the criticism merely implies that Plato only elaborates on the reason why, or the way in which, an oligarchy is divisible into two distinct poleis. As Keyt mentions, Plato is ‘especially emphatic in Book VIII about the bifurcation of oligarchy’.28 Aristotle focuses on this particular account, because it shows that the general distinction between the unified kallipolis and the divisible non-ideal poleis which Plato commits himself to (e.g. at Rep. IV. 422e) is faulty. The division of oligarchy into a ‘city of the x [=rich]’ and a ‘city of the y [=poor]’ shows us why both timocracy and the kallipolis must be divisible in the same way, and for the same reason, even though this reason is only fully developed in Plato’s account of oligarchy. This special role of oligarchy in establishing Aristotle’s general criticism of Plato’s view of the relationship between non-ideal and ideal political constitutions also explains why that criticism appears here, in the context of Aristotle’s particular discussion of Plato’s view of oligarchy (and not e.g. in the context of his criticism of Plato’s account of political change in the kallipolis).
5 Historical Inaccuracies
Having dealt with the two kinds of criticism intended to reveal theoretical inadequacies in Plato’s account of political change, I now move on to discuss those criticisms which focus on the fact that various features in Plato’s account clash with the historical evidence. Criticisms of this kind are particularly difficult to defend on Aristotle’s behalf. For, though it is unlikely to be contested that Plato’s account of political change should be gapless and internally consistent, it is often claimed that this account is not intended to be historically accurate. Nevertheless, I shall attempt to show that Aristotle is quite justified in mounting these criticisms (even though Plato could quite possibly conjure up responses to some of them, as we shall see below). The criticisms are roughly as follows. First, though Plato presents a fixed pattern of political change in one direction (viz. kallipolis → timocracy → oligarchy → democracy → tyranny), Aristotle presents historical cases which clearly contradict it. Second, though Plato may think that tyrannies do not change at all (or else change exclusively into the best city), Aristotle presents historical cases to the contrary. Third, although Plato thinks there is only one kind of each type of constitution (e.g. only one kind of oligarchy or democracy), and that therefore no constitution can change into another constitution of the same type, Aristotle thinks that goes against the facts, as he recognizes various subkinds of the same type of constitution and even observes political change from one subkind into another within the same constitution-type.
First, then, Aristotle (1316a17-24) criticizes Plato’s fixed pattern of political change as a whole. The pattern, as Aristotle describes it following Republic VIII, is as follows:
According to Aristotle, there are reasons to reject this pattern. By contrast to the gradual change presented in it, constitutions often change into their contraries, rather than to the types of constitution close to them (1316a18-20), and, even within the gradual scheme, we often see constitutions changing in the opposite direction (e.g. from democracy into oligarchy; indeed, Aristotle says that this happens more often than the change of democracies into a monarchy) (1316a23-4). Aristotle also gives a list of examples of e.g. oligarchies changing into tyrannies (1316a34-9).29 He later on states that there is no reason to assume that an oligarchy would change specifically into a democracy, as Socrates supposes in Republic VIII (1316b20-1).
Keyt (speaking specifically of 1316a17-39) suggests that such criticisms are ‘misguided’, since ‘clearly Plato was not writing as an historian when he composed Book VIII of the Republic’.30 He establishes his point by stressing the speculative nature of the first item in Plato’s proposed chain of political change, viz. the kallipolis:31
[Plato’s] description of the deterioration of the first and best constitution is speculative and a priori, and cannot be intended as constitutional history (…) But if the first term of the sequence is not historical, the historicity of the entire sequence is undermined. The sequence is clearly determined by psychology rather than history. In constructing it Plato had his eye, not on the constitutional history of historical cities, but on various configurations of the parts of the soul.
The contrast between the seriousness with which Aristotle takes Plato’s account of political change as historically testable and the original intent of that account, at least with regard to the kallipolis, is still widely mentioned. Thus, Kalimtzis states that Plato’s ‘story’ concerning the deterioration of the best regime, though ‘taken seriously’ e.g. by Aristotle, has also ‘been interpreted as a heuristic tale’,32 and according to Saxonhouse, even though Aristotle acknowledges that ‘the problems with the [nuptial] number (…) may have some validity’,33 Socrates’ identification of this number as the cause of change in the best city may be either meant ‘seriously or not’.34 We have also already mentioned (in section 2) the discussion in Skultety of the view that by criticizing Plato’s general account of political change for historical discrepancies Aristotle is ‘missing the point’, since that account is intended as ‘little more than a literary device’.35
There is no doubt that Aristotle responds to Plato’s account of political change as if it were intended to be historically correct. His arguments center on actual cases disproving the pattern suggested by Plato. There is also no doubt that Books VIII-IX of the Republic aim not only at political but also at psychological insight, as they are concerned with delineating the nature of justice and defending it against injustice in both poleis and individual souls. As we have seen (in section 2), the main argument of these books continues the discussion of types of constitution interrupted in Book V (449b), which is in turn presented in order to fully answer Glaucon’s challenge in Book II of showing that justice not only is good for its own sake as well as for the sake of its consequences, but is also preferable over injustice in every way (357a-ff). In order to fully meet the challenge, Socrat Cambridge: Cambridge University es in Books VIII-IX applies the ‘city-soul’ analogy, also introduced back in Book II, and presents a pattern of gradual deterioration of types of constitution along with a similar pattern of gradual deterioration of the types of soul corresponding to them. Throughout the discussion it becomes increasingly clear that tyranny and the tyrannical soul, which are the least just, are in a much less desirable state than the kallipolis and the type of soul corresponding to it, which are of course the most just. But none of this, as far as I can see, entails that Plato intends his analysis of political change in Republic VIII to be purely speculative or, as Keyt puts it, ‘just another Platonic myth’.36
Furthermore, even if we grant that Plato introduces his pattern of political change solely for the purpose of supporting his psychological theory, it is entirely possible that he would have wanted that pattern to be historically accurate.37 Indeed, precisely because Plato’s guiding principle throughout the Republic is the ‘city-soul analogy’, he might think that a correct analysis of political change is (methodologically) necessary in order to arrive at a correct account of psychological change, in particular the process of moral degeneration ultimately resulting, if carried on to an extreme, in the ‘tyrannical soul’. Less directly (and somewhat rhetorically), Plato might have thought that giving a correct historical analysis of political change would lend some support to his psychological conclusions merely by way of bestowing credibility on the discussion as a whole. But even without assuming any relation of support between his account of political change and his account of justice in the individual soul, Plato might have simply been interested in giving a correct historical account of political change for its own sake (doing so would certainly not be incompatible with a focus on psychology).
The claim that Plato does not mean for his account of political change to be historically true, and that Aristotle must therefore be wrong in focusing his criticism on the historical inaccuracy of that account, lacks the proper support. In particular, and as noted above, Keyt bases his version of this claim on the non-historicity of the kallipolis, which he takes to imply that Plato’s view of the transition from the kallipolis into a timocracy (and from timocracy into an oligarchy, etc.) is meant to be non-historical as well. Granted, the kallipolis, if it is ever established, is certainly not instantiated frequently enough for it to figure in any reliable historical analysis of political change. But it is precisely this fact that makes it plausible that Plato would have meant his own analysis of political change to be historically correct. For Plato might have aspired to establish an adequate view of change within his best polis based on what actually happens in cities.38 On this interpretation, the process of deterioration beginning from the second item in Plato’s pattern (i.e. from timocracy downward) would serve as evidence for constructing a view of the kallipolis as it would function and change, not merely as a model (Rep. IX, 592b1-3), but under actual circumstances (such a project would not have been foreign to Aristotle himself, of course, as it closely resembles that of Politics VII-VIII).
A further historical inaccuracy which Aristotle attributes to Plato has to do with the place of tyranny in the pattern of political change presented in Rep. VIII. Though Socrates is silent on the issue of whether tyranny changes (and if so, then into what), Aristotle considers the implications of the positions available to him. As we have seen in the previous section, holding that tyranny does change (i.e. into the kallipolis) would saddle Plato with an internally inconsistent account of political change. The alternative, which we shall presently discuss, is that tyrannies do not change into any type of constitution. This alternative is recognized by Aristotle in his very first complaint about Plato’s view of tyranny, viz. his remark that Socrates does not state whether or not there would be a change of tyranny (1316a25-6). The complaint is of course legitimate, since even if it is Plato’s view that tyrannies do not change we would still expect him to state it explicitly in his discussion of tyranny in Republic VIII-IX, and to give his reasons for holding it. But Plato might have nevertheless seriously regarded tyrannies as unchangeable. As Keyt puts it:39
Since it is also part of Plato’s theory that constitutional decline is always caused by personal decline among the rulers, what cause could there be for tyranny to change, given that tyrants have reached the limit of personal decline?
Though Plato does not explicitly answer or even raise such a question in Rep. VIII (which is partly what Aristotle complains about), I believe Keyt is wrong in suggesting that ‘we should consider the possibility that Plato never intended these questions to be answered’, and that this is ‘one more indication that [Plato’s] sequence of constitutional change is just another Platonic myth’.40 In Republic X, for example, he speaks of tyrants as ‘incurably bad’ (615e2-3), and of tyranny as being responsible for ‘incurable bad things’ (619a4-5). Plato may well have regarded the worst possible type of constitution, instantiated by irreparably bad people and thereafter leading to irreparable damage, as irreparable itself, in which case it would also necessarily be unchangeable (since, on Plato’s own classification of constitutions, there are only better types of constitution for it to change into). Aristotle, for his part, concedes that this may well be Plato’s view.41
Now, Aristotle’s closing remarks about the nature of change in a tyranny in fact count against both of Plato’s available alternative solutions to the problem of change in tyrannies. For there (1316a29-34), Aristotle provides examples which show that tyranny need neither change exclusively into the best type of constitution nor remain unchanged. The examples are of tyrannies which have been known to change into other tyrannies, as well as into oligarchies, democracies, and aristocracies.42 Here, again, Aristotle treats Plato’s account of political change as a serious proposal in political science, testable by reference to the historical records. I have argued already that he cannot be shown to be mistaken in thinking so. However, at this point a further objection to this claim could be raised. If all that Aristotle needs to do in order to show that change in a tyranny is possible is to appeal to well-known and readily available cases of tyrannies which have undergone such a change, how could Plato possibly seriously think that such a change is either impossible or restricted to the transformation into one particular type of constitution? And, with the mistake being so obvious, how could we not interpret Plato’s account as being deliberately non-historical?43
One obvious possibility is that Plato would have disagreed with the very classification of the poleis in Aristotle’s example as tyrannies (or as turning into oligarchies, democracies or aristocracies). For instance, Plato might retort that the examples given by Aristotle do not fit neatly into his (Plato’s) classification of constitutions. Rather, they could be e.g. what Plato terms ‘intermediate’ constitutions, in which case his account of political change based on his hierarchy of distinct types of constitution might not apply to them (or at least not straightforwardly) (Rep. VIII, 544d). Indeed, discussions of possible disagreements and misunderstandings concerning the classification of constitutions are not unprecedented in Aristotle. He thinks that the correct classification of constitutions depends on an adequate understanding of the nature of the polis and its identity throughout time (Pol. III. 3, 1276b9-13), which naturally leaves room for mistakes in applying the name of a type of constitution to particular cases. In fact, Aristotle goes as far as saying that ‘in most poleis the type of constitution is badly called’ (IV. 8, 1294a15-16).44
It is quite possible, then, that Plato would have had examples of his own to defend his account of political change against Aristotle’s criticism, and that each would have disagreed on the import of the evidence used by the other.45 But such disagreements would only strengthen the legitimacy of Aristotle’s interpretation of Plato’s account as being meant to be taken seriously and literally as a testable hypothesis.
Implicit in Aristotle’s criticism of Plato’s downward progression of political change in the Republic, especially if he takes Plato’s intended view to be that tyrannies remain unchanged, is the rejection of the general idea that political improvement is impossible. As Kraut argues, ‘Plato holds that ordinary cities leave no room for improvement’, which is why he ‘advocates withdrawal’ from political life (unless one happens to be a member of the kallipolis).46 ‘By contrast’, Kraut continues, ‘Aristotle’s Politics is founded on the assumption that Plato’s pessimism is unfounded’.47 By listing in V. 12 examples of constitutions (e.g. tyrannies) changing into superior types Aristotle makes the point that, contra Plato, political improvement is not only possible but indeed observable.
Aristotle’s final remark in V. 12 also challenges Plato’s analysis of historical data. According to Aristotle, Socrates speaks of the changes in oligarchies and democracies as if there were only one kind of each, whereas in fact there are many subkinds of oligarchy and democracy (1316b25-7). Aristotle has here in mind his own classification of subkinds of democracy and oligarchy, as it appears in Politics IV. 4-6. In fact, this classification is the topic of the discussion immediately following Aristotle’s treatment of Plato’s Republic in Pol. V. 12. In VI. 1, Aristotle differentiates and ranks the different subkinds of democracy and oligarchy based on the population dominant in them and the different combinations of parts within them (1317a22-33). As Aristotle goes on to say, knowledge of these different subkinds of oligarchy and democracy and the ways in which they come to be (and are destroyed) is useful for the purpose of initiating and controlling political change both from and into such constitutions (1317a33-5).48 Plato, according to Aristotle’s last bit of criticism in V. 12, fails to make use of such distinctions, and errs in thinking that any political change into and from oligarchy and democracy is of the same kind.49 This is false, as Aristotle explains, not only because oligarchies and democracies do not change exclusively into democracies and tyrannies (respectively), but because oligarchies and democracies themselves are divided into subkinds, having their own specific sets of characteristics and processes of political change. To repeat one obvious example, Aristotle thinks that a specific type of constitution can change into a further subkind of the same type of constitution. He gives the example of a tyranny changing into another tyranny in V. 12 (1316a29-31), and the same is true also in the case of the different subkinds of oligarchy and democracy which he goes on to discuss in Book VI.50
The fact that Aristotle contrasts his own classification of kinds and subkinds of constitution, as well as his own view of the kinds and directions of political change between these different kinds and subkinds, with Plato’s discussion in Republic VIII, reinforces the seriousness with which we have so far seen him take Plato’s political theory as an alternative to his own throughout the last part of Pol. V. 12.
Aristotle is quite justified in regarding Plato’s view of political change as a serious alternative to his own view, and his responses to this alternative are both careful and adequate. Aristotle criticizes different features of Plato’s account for being either theoretically insufficient, internally inconsistent or historically inaccurate. Aristotle’s own political theory, surely, is meant to overcome these shortcomings. As we have seen, there could be room for further debate. Plato’s position could have conceivably been defended against at least some of Aristotle’s objections, for instance by reevaluating the historical data (e.g. the examples used by Aristotle in order to criticize Plato’s view of political change into and from tyrannies). Other objections seem decisive, as they reveal, for instance, basic tensions within Plato’s own theoretical framework (perhaps most clearly shown by Aristotle’s criticism of Plato’s opposition between the ‘unified’ kallipolis and the divisible non-ideal poleis). But whether or not Aristotle’s criticisms are ultimately successful, what is more important for our purposes is that they cannot be said to be fundamentally misguided, e.g. by misrepresenting Plato’s views and the motivations underlying them. Or so I have argued. If I am correct, then at least Aristotle’s criticisms of Plato’s theory of political change in Politics V. 12 are not susceptible to the claims so often raised against his criticisms of Plato, and indeed against his criticisms of his predecessors in general.51
W. L. Newman (comm.), The Politics of Aristotle (Oxford: Clarendon Press, 1887-1902), vol. 1, p. 519: ‘The Republic, we feel, has a great practical end in view – to recall the State and the individual to a right view of the importance and nature of Justice – and we can forgive it, if in its language on the subject of constitutional change it to a certain extent sacrifices historical accuracy. Aristotle, however, who is often a somewhat unsympathetic critic, loses sight of this, and bluntly enumerates the points in which Plato’s account of the subject falls short’; D. Keyt (trans. and comm.), Aristotle Politics Books V and VI (Oxford: Clarendon Press, 1999), pp. 186-7: ‘This criticism [here, specifically Aristotle’s criticism in Pol. V. 12, 1316a17-39 of Plato’s proposed pattern of political change in Rep. VIII] seems misguided. For clearly Plato was not writing as an historian when he composed Book VIII of the Republic … In constructing it [sc. his sequence of political change] Plato had his eye, not on the constitutional history of historical cities, but on various configurations of the parts of the soul’; R. Polansky, ‘Aristotle on Political Change’, in D. Keyt and F. D. Miller (eds.), A Companion to Aristotle’s Politics (Oxford: Blackwell, 1991), pp. 323-345, p. 343: ‘The attack on Plato is really a defense of the scientific standing of Aristotle’s work. Aristotle merely uses Plato’s text as a foil (even straw man) for defending his own approach. He knows Plato’s intention in the Republic was not a comprehensive analysis of political change, but avails himself of this entertaining way to secure his own position’. See also J. Adam, The Republic of Plato (Cambridge: Cambridge University Press, 1902), ad VIII. 543a; R. L. Nettleship, Lectures on the Republic of Plato (London: Macmillan and Co., 1922), pp. 295-6; S. C. Skultety, ‘Delimiting Aristotle’s Conception of Stasis in the Politics’, Phronesis, 54 (2009), pp. 346-70, p. 363 (see below). P. L. P. Simpson adds that Aristotle could have himself recognized and appreciated Plato’s (psychological, or at any rate non-historical) purpose of proposing an account of political change, and that his criticism could be addressing ‘certain students of the Republic’ who adopt the problematic ‘empirical way of reading the Republic, whether it was the way Plato intended or not’ (P. L. P. Simpson, A Philosophical Commentary on the Politics of Aristotle [Chapel Hill: University of North Carolina Press, 1998], pp. 417-18).
Skultety, ‘Conception of Stasis’, p. 363.
Skultety, ‘Conception of Stasis’, p. 363.
Skultety, ‘Conception of Stasis’, p. 363. But notice that Skultety thinks at least one criticism mounted in V. 12, 1316a39-b3, modeled on the discussion in V. 1-4, is also relevant (Skultety, ‘Conception of Stasis’, p. 367). More on this below.
A. W. Saxonhouse, ‘Aristotle on the Corruption of Regimes: Resentment and Justice’, in T. Lockwood and T. Samaras (eds.) Aristotle’s Politics: A Critical Guide (Cambridge: Cambridge University Press, 2015), pp. 184-203, 185; cf. Keyt, Politics Books V and VI, p. 185 (discussed below).
Or, according to some scholars, the cosmos as a whole. See R. Jenks, ‘The Machinery of the Collapse: On Republic VIII’, History of Political Thought, 23.1 (2002), pp. 21-9, p. 23 and n. 5, who also offers a third suggestion of his own.
For further discussion see Jenks, ‘Machinery of the Collapse’, pp. 21-9; Adam, Republic of Plato, ad 546a-ff.
For further discussion of the role of perception in the miscalculation of the nuptial number by the philosopher kings see C. D. C. Reeve, Philosopher-Kings: The Argument of Plato’s Republic (Princeton: Princeton University Press, 1988), pp. 113-15 and Nettleship, Lectures, pp. 302-3. As has been helpfully pointed out to me by an anonymous referee, it has also been argued that a certain obstacle to reason, rather than (or in addition to) the reliance on perception, can function as the cause of the decline of the kallipolis (see Jenks, ‘Machinery of the Collapse’, pp. 23-6, who also finds an initial formulation of this interpretation in E. Erhardt, ‘The Word of the Muses (Plato, Rep. 8.546)’, Classical Quarterly, 36.2 , pp. 407-20). Related to this point is Jenks’ discussion of Plato’s Critias, describing a city nearly equivalent to the kallipolis (i.e. Atlantis) which collapses as a result of there being ‘in principle no rational way’ to make a certain kind of political decision concerning the defense of the city (Jenks, ‘Machinery of the Collapse’, p. 25). Thanks to the referee above for a useful comment on this issue.
J. Ober, ‘Nature, History and Aristotle’s Best Regime’, in T. Lockwood and T. Samaras (eds.) Aristotle’s Politics: A Critical Guide (Cambridge: Cambridge University Press, 2015), pp. 224-43, p. 229.
Ober, ‘Aristotle’s Best Regime’, p. 240. Ober acknowledges that at 1316a8-11 Aristotle seems to agree with Plato’s view that such uneducable people might be born (even in the best regime). However, Ober argues that Aristotle may be simply conceding that point for the sake of argument (Ober, ‘Aristotle’s Best Regime’, p. 243 n. 22).
A further controversy about the text needs to be mentioned at this point. When Aristotle, in item (3) above (viz. 1316a11-14), criticizes Plato’s view for failing to ascribe to the best constitution a change that would be peculiar to it (or even peculiar to constitutions), I have so far taken it for granted that he means to be saying that (i) the change in the kallipolis as Plato describes it is no different from the change of other constitutions (and other things) in that the cause of the change in the kallipolis could equally cause change in other constitutions and things as well (as noted e.g. by Newman, Politics of Aristotle, vol. 4, p. 481), and that (ii) the cause of the change in question is what Plato describes in Republic VIII as the ‘nuptial number’ (as noted e.g. by Polansky, ‘Political Change’, pp. 343-4) or ‘nature’ (as noted e.g. by Simpson, Commentary on the Politics, p. 418) (in my reading, specifically the application of the nuptial number by the nature e.g. of human beings). Keyt, rejecting this reading, takes Aristotle’s criticism to focus solely on the type of change in the kallipolis versus other constitutions and things (Keyt, Politics Books V and VI, pp. 185-6). This seems misguided, given the focus on causes of change in the exposition of Plato’s view immediately preceding the criticism (viz. items (1)-(2b) above; 1316a4-11). Furthermore, as Keyt himself recognizes, ‘Aristotle has not explained why the best constitution should have a distinctive mode of change’, so that the question that he poses to Plato on Keyt’s reading would in fact be expressing ‘an undefended and questionable idea’ (Keyt, Politics Books V and VI, p. 186). Keyt’s motivation in proposing his reading is his assumption that the cause of change Aristotle ascribes to Plato is ‘the fact that nothing endures but everything changes in a certain cycle’, which admittedly would not have yielded a satisfactory criticism, as Plato does give us a cause of change in the kallipolis which is more specific than that (Keyt, Politics Books V and VI, p. 185; cf. Saxonhouse, ‘Aristotle on the Corruption of Regims’, p. 185, discussed above). But, as we have seen, there is a real sense in which the specific cause which Plato provides could apply to changes in both non-ideal constitutions and things other than constitutions.
Keyt, Politics Books V and VI, ad loc; J. Adam, ‘The Myth in Plato’s Politicus’, The Classical Review, 5 (1891), pp. 445-6, p. 446.
E. Garver, Aristotle’s Politics: Living Well and Living Together (Chicago: University of Chicago Press, 2011), p. 139. Garver links this Platonic view to the Socratic idea that fully virtuous agents do not suffer (Garver, Aristotle’s Politics, p. 139; cf. Apology 41d1-2).
For a discussion of distributive justice in those texts see D. Keyt, ‘Supplementary Essay’ in R. Robinson (trans. and comm.), Aristotle Politics III-IV (Oxford: Clarendon Press, 1995), pp. 125-48; Keyt, Politics Books V and VI, ad. V. 1 esp. pp. 56-62. On the connection between those texts and Pol. V.12, 1316b1-3, see Keyt, Politics Books V and VI, p. 189; A. Lintott, Aristotle’s Political Philosophy in its Historical Context (Abingdon/New York: Routledge, 2018), p. 155.
Saxonhouse, ‘Corruption of Regimes’, p. 189.
Skultety, ‘Conception of Stasis’, p. 366 and n. 30. Skultety argues against the interpretation offered in K. Kalimtzis, Aristotle on Political Enmity and Disease (Albany: SUNY Press, 2000), pp. 118-23. I also agree with Skultety that the criticism in 1316a39-b3 has its roots in the earlier parts of Politics V (see above), though I of course disagree that this is the only adequate criticism of Plato to be found in V. 12, as Skultety seems to suggest.
See F. D. Miller, Nature, Justice and Rights in Aristotle’s Politics (Oxford: Clarendon Press, 1995), p. 295 n. 615; A. Lintott, Aristotle’s Political Philosophy, pp. 27-30; E. Cohen de Lara, ‘Aristotle’s Politics: Ethical Politics or Political Realism?’, in E. Cohen de Lara and R. Brouwer (eds.), Aristotle’s Practical Philosophy (Cham: Springer, 2018), pp. 13-33, pp. 16-19.
Polansky, ‘Political Change’, p. 326. Applying the four Aristotelian causes model to Aristotle’s account of stasis has been objected to on various grounds; see Skultety, ‘Conception of Stasis’, p. 349 n. 5 and A. Hatzistavrou, ‘Faction’, in M. Deslauriers and P. Destrée (eds.), The Cambridge Companion to Aristotle’s Politics (Cambridge: Cambridge University Press, 2013), pp. 275-300, p. 298 n. 15. I agree that one should not seek an Aristotelian account of stasis in terms of (e.g.) its form, as Keyt seems to do when he speaks of the formal cause of stasis as ‘the dispositions of the members of a faction’ (Keyt, Politics Books V and VI, p. 78). As Skultety notes: ‘stasis does not exist in the way that a chair exists or a specific living substance exists; rather, stasis designates a group of people who are acting in a certain way toward the constitution of their city’ (Skultety, ‘Conception of Stasis’, pp. 348-9). However, as Polansky shows, Aristotle’s account of political change does seem to be plausibly based on the distinction between the form and the matter of the entity in which such a change takes place, viz. the polis.
Polansky, ‘Political Change’, p. 327.
Polansky recognizes that Aristotle takes issue with the ‘reductionistic’ nature of Plato’s account of political change (1316a39-ff), and for making the ‘logical error’ … ‘that matter [i.e. the desires of the community members] determines form [i.e. the type of constitution changed into]’ (1316a17-39) (Polansky, ‘Political Change’, pp. 343-4). However, as mentioned above, Polansky also thinks that ‘Aristotle merely uses Plato’s text as a foil (even straw man)’ (Polansky, ‘Political Change’, p. 343). Similarly, Hatzistavrou argues that Aristotle, rather than offering ‘a “psychologically reductionist” account of the causes of political factions’, ‘explains faction from two different perspectives, a sociological and a psychological perspective’ (Hatzistavrou, ‘Faction’, p. 293). But note that by Aristotle’s ‘sociological’ account of faction Hatzistavrou has in mind the explanation of faction in terms of political injustice causing mental states in individual citizens being wronged, which in turn leads them to initiate political change (Hatzistavrou, ‘Faction’, pp. 294-5). Aristotle does not seem to confine himself to such occurrences. Rather, he seems to count as a cause of political change any event resulting in a relevant change in the constitution of the polis, as explicated above.
Newman notes that Susemihl ‘transposes οὐδενός, [1316b]10 – νοῦν, 14, to after πολιτείαν, 21’ (Newman, Politics of Aristotle, vol. 4, p. 487), which is in line with my reading of the passage.
Garver, Aristotle’s Politics, pp. 147-8.
See Keyt, Politics Books V and VI, p. 187.
See also the comment on 1316a27-9 in Lintott, Aristotle’s Political Philosophy, p. 154.
Apart from the difficulty in maintaining and defending a cyclical view of political change as such, there is a further reason to think that Plato might want to avoid the particular cyclical pattern that view would commit him to (viz. kallipolis → timocracy → oligarchy → democracy → tyranny → kallipolis …). As Simpson recognizes, such a view would be problematic for Socrates to accept since his ‘account is all about how regimes get progressively worse, and contains nothing about how the worst of them could suddenly return to the best’ (Simpson, Commentary on the Politics, p. 419). See also R. Kraut, Aristotle: Political Philosophy (Oxford: Oxford University Press, 2002), p. 173 (discussed in section 5 below).
Indeed, as Newman points out, this further criticism too makes an appearance in Politics II. 5, at 1264a11-26 (Newman, Politics of Aristotle, vol. 4, p. 487). There, Aristotle says that if the producers in Plato’s kallipolis are to have private property (as they should, if they are to differ from the guardians), then the case of the kallipolis would turn out to be one in which ‘there necessarily are two poleis out of one, and these set against each other’ (1264a24-6).
Keyt, Politics Books V and VI, p. 189.
Keyt, Politics Books V and VI, p. 189.
The examples, which are then supplemented with additional arguments against Plato’s position, are as follows. Most oligarchies in Sicily changed into tyrannies: in Leontini, the oligarchy changed into the tyranny of Panaetius; in Gela the oligarchy changed into the tyranny of Cleander; in Rhegium the oligarchy changed into the tyranny of Anaxilaus.
Keyt, Politics Books V and VI, p. 186.
Keyt, Politics Books V and VI, p. 187.
Kalimtzis, Aristotle on Political Enmity, p. 122 and p. 198 n. 24. Kalimtzis does not commit himself to this interpretation, but rather cites the comments on Rep. VIII, 546b-c in G.M.A. Grube (trans.), Plato’s Republic (Indianapolis: Hackett, 1977), 197 n. 6.
Saxonhouse, ‘Corruption of Regimes’, pp. 184-5.
Saxonhouse, ‘Corruption of Regimes’, p. 203.
Cf. Skultety, ‘Conception of Stasis’, p. 363.
Keyt, Politics Books V and VI, p. 187.
In modern times, the interpretation of Republic VIII-IX as presenting a historical account of political change has been most famously defended by Karl Popper. Popper contends that ‘we should look upon the Books VIII and IX of the Republic, in view of their close parallelism with the Third Book of the Laws, as a simplified historical sketch of the actual decline of the ideal city of the past, and as an explanation of the origin of the existing states’ (K. Popper, The Open Society and its Enemies [Princeton: Princeton University Press, 1966], vol. 1, p. 223 n. 27 on chapter 4).
Popper argues further that Plato’s vision and description of the kallipolis in general are due to his historical analysis of the actual poleis of his day (especially Crete and Sparta) and the change within them. In particular, Popper says, ‘[Plato’s] best state was … to be reconstructed in such a way as to eliminate all the germs and elements of disunion and decay as radically as this could be done; that is to say, it was to be constructed out of the Spartan state with an eye to the conditions necessary for the unbroken unity of the master class, guaranteed by its economic abstinence, its breeding, and its training’ (Popper, Open Society, vol. 1, p. 55; but see p. 45, where Popper adds that Plato’s kallipolis also has some ‘intentionally unhistorical elements’).
Keyt, Politics Books V and VI, p. 187.
Keyt, Politics Books V and VI, p. 187 (emphasis mine).
But note that there is also some evidence for attributing to Plato the view that tyranny does (or could) change (viz. improve), though this evidence is mainly external to the Republic. Popper reads the Laws (709e-714a) and the Statesman as supplying such evidence, which he connects to Plato’s own project of ‘reforming the younger Dionysus’ tyranny over Syracuse’ (Popper, Open Society, vol. 1, p. 44 and p. 222 n. 24 on chapter 4).
The examples given are as follows. In Sycion, the tyranny of Myron changed into that of Cleisthenes; in Chalcis, the tyranny of Antileon changed into an oligarchy; in Syracuse, the tyranny of Gelon changed into a democracy; in Lacedaemon, the tyranny of Charilaus changed into an aristocracy, as happened also in Carthage.
Thus, e.g. Adam, Republic of Plato, ad Rep. VIII, 543a: ‘Plato himself must of course have known as well as Aristotle that the historical development of Greek constitutions did not by any means always correspond with his scheme.’
Reading 1294a15 with ‘κακῶς’, as added by W.D. Ross in the OCT.
Plato may have similarly contested the examples Aristotle uses in the criticisms mentioned previously in this section, for instance Aristotle’s examples of the change into tyranny from oligarchies (cf. 1316a34-9).
Kraut, Aristotle: Political Philosophy, p. 173. Kraut also qualifies this statement by saying that ordinary cities ‘are incapable of significant improvement so long as they maintain their current constitutions’ (Kraut, Aristotle: Political Philosophy, p. 173). But, if this refers to political change into better constitution types, then this option seems to be excluded by the account given in Rep. Cf. Simpson, Commentary on the Politics, p. 419 (discussed above).
Kraut, Aristotle: Political Philosophy, p. 173.
Aristotle then suggests that a person founding a constitution should not attempt to have all of its features conform to its basic principle (1317a35-38), presumably because different subkinds of one type of constitution would require different configurations of the features generally characteristic of that type. As E. Cohen de Lara notes, citing this text, inter alia, Aristotle thinks that legislators should understand ‘which regime-specific elements preserve it and which destroy it’ (Cohen de Lara, ‘Aristotle’s Politics’, p. 22).
On the criticism in Pol. V. 12, 1316b25-6 and its relation to Aristotle’s view that political experts must grasp the various types and subkinds of constitutions, as well as the causes of their destruction and preservation, see Cohen de Lara, ‘Aristotle’s Politics’, p. 28.
Newman notes that Aristotle, despite his criticism of Plato here, does not develop his own account of political change in Book V with regard to the subkinds of oligarchy and democracy either (Newman, Politics of Aristotle, vol. 4, pp. 488-9). But note that as early as Politics IV. 6, 1293a12-34 Aristotle does describe the processes through which one subkind of oligarchy may change into other subkinds.
Some of the ideas appearing in this paper were presented by the author in his seminar on Aristotle’s Politics at the University of South Florida during the Spring semester of 2015. Thanks are due to the participants in that seminar. Thanks also to Ronald Polansky and anonymous referees for their helpful feedback on previous versions of this paper.