The beginning of Politics 3 addresses the issue which both Politics 1 and 2 attempted to deal with, the city or the polis.(1) Thus, Book 3 is a return to the beginning of an inquiry into politics. Concerning this inquiry, Aristotle in Book 3, chapter 1 states what is clearly the object of his investigation:

For one investigating the regime ... virtually the first investigation concerns the polis, to see what the polis actually is (3.1.1274b32-33).(2)

Thus, the first question of Book 3 which Aristotle requires us to address is "what is the polis?" or, as Carnes Lord translates it, the city.(3) Yet if we look for this investigation in the following paragraph of Politics 3.1, we encounter no clear discussion of the city.

The sentence following the first sentence of Politics 3.1 argues not about what the city is, but rather addresses the debate about who is responsible for certain political acts, the city or the rulers of the city (3.1.1274b34-35). Although the third sentence claims that the "entire activity of statesmen (politikos) and legislators is concerned with the polis," the sentence then goes on to discuss the regime as "a certain arrangement of those who inhabit the polis" (3.1.1274b35-38). Thus we get no discussion of what the first sentence says is the proper question: What is the polis?

Given that Aristotle suggests it is required to address the question--what is the polis--yet in the immediate context he fails to do so. It is thus fitting, before we go on to the other issues raised by Politics 3, to address this question. Is the polis merely the Greek city of the 4th and 5th century BC? Or is it the proto-organization of the modern Nation-state?

The Polis as State?

Leo Strauss and Harry Jaffa have addressed at length this question and argue that both state and city-state are inadequate and unhelpful translations of polis, so it is only necessary to refer the reader to them.(4) Also, the state is a product of modern political philosophy. The term "the state" is a creation of Machiavelli.(5) The modern state--and its contemporary embodiment, the modern liberal state--as we know it, although conceived by Machiavelli, is the child of Thomas Hobbes. Although Hobbes does not explicitly define the state, it is nevertheless a conceptual product of this understanding of political community.(6)

The term Hobbes uses to specify the political community is the commonwealth. For Hobbes the commonwealth is the political entity within which human political behavior will be actualized. The real force of Hobbes' commonwealth must be understood in the expression of the sovereign power toward the subjects of the commonwealth. For Machiavelli the state is the articulated will of the prince--an actual person. Hobbes takes this concept of the state and argues that it is not to be specifically the will of a prince or actual sovereign.

What is the sovereign power in Hobbes political thought? It is the articulated will of the social compact which gives legitimacy to the commonwealth. The reason the sovereign power arises out of a social contract is that, for Hobbes, a political community is not by nature, but is rather a humanly made construct. Thus, for Hobbes, political community is an artifact and, as such its political expression is the abstracted will of that which forms the political community.(7) Yet the term political community is no longer appropriate for the political entity which is being constructed; rather, sucha a community is referred to as "the body politic."

The body politic is a term or, more correctly a metaphor from Medieval political thought, which attempts to explain the relationship between a king and the realm he rules.(8) Hobbes' use of the term sovereign has long led readers to think he is speaking as if the sovereign will were a single human ruler, a king or a monarch. Hobbes is not referring to an actual human sovereign, but using the term as a metaphor to describe not a person but the embodied will of that which authorizes the body politic. The sovereign is thus no longer the body of the sovereign, i.e., the king or prince, but the abstracted will of the whole body politic.(9)

The notion of the state is further developed along Hobbesian lines by both Rousseau, who shows how not only the state is a product of human construction but also of human rationality,(10) and Kant, who 1) shows that all moral action is an act of the will--e.g., the categorical imperative, rather than an outcome of natural predispositions and 2) makes explicit that the state is a disembodied will.(11) Thus, the modern state is no longer understood as the articulated will of any specific ruler.

The concept of the modern state reaches its intellectual peak in Hegel's articulation of it.(12) The history of the concept of the state entails a rejection of Aristotle's understanding of political community as a natural condition (this is to say environment or habitat) for human beings. Rather, "the state" offers a human construct that is nothing but the disembodied will of the body politic.(13) Thus, translators of and commentators on Aristotle's works, who insist on construing the polis as the state, fail to see how that construction does not fit at all with what Aristotle means when he talks about the polis. First, the concept of the state is one which was not developed until the Renaissance.(14) The state is an artificial construct, which represents the disembodied will of the body politic, whereas the polis is a natural form of human association. Therefore, when scholars construe polis as for Aristotle, the state, they commit an egregious anachronism that obscures, as well, what "the state" implies conceptually.

What is The Polis?

So what do we call the polis? Carnes Lord, who is a student of Strauss, in his translation of Aristotle's Politics translates polis, as we have mentioned earlier, as the city.(15) Yet is it fitting or correct to translate polis as the city? The answer to this question is that there is a problem with translating polis into English as city, just as much as it is problematic to translate it as city-state or as state.

In English there are two ways to understand city: 1) an urban area with a large aggregation of peoples from a multitude of diverse backgrounds and 2) a political community, entity or unit. The double meaning of the English word city leads us to a skewed understanding of the polis. Calling it the city, one tends to conceptualize it as a commercial, cosmopolitan and urban, rather than rural and agrarian, entity.(16) Yet, as long as we realize that Aristotle is only referring to the polis as a political unity and is not describing it either in sociological terms or as an urban unit, one could provisionally translate polis as city--since it is the best of the given translations. Yet is there a better translation?

The best translation of polis ends up being the term which is used to define it. The definition of polis turns out to be what Aristotle uses to clarify the term polis at 1.1.1252a5-6; he says that the polis can be called the political association, or koinonia.(17) Or another way to say the political association would be to render it simply as the political community.(18) To call the polis the political community, rather than the city, is closer to what Aristotle intends to describe--a unity of political organization.(19)

The Polis

Now let us look closely at Politics 1.1 regarding the question of the polis. Politics 1.1 starts off with a conceptualization of what is the subject matter of politics: the polis. This arrangement is seen by Aristotle's original definition of the polis.

Since we see that every polis is some sort of association (koinonia), that every association is constituted for the sake of some good (for everyone does everything for the sake of what is held to be good), it is clear that all associations aim at some good, and that the association that is most authoritative of all and embraces all the others does so particularly, and aims at the most authoritative good of all (1.1.1252a1-6).

Thus, the polis is the association or community that embraces all other associations or communities.(20) It is also the community which aims at the most authoritative good. He then goes on to distinguish the type of rule which arises out of the polis, political rule. He argues that political rule is different in kind from the rule of the household and the rule of slaves (1.1.1252a7-9).(21) Aristotle goes on to argue that ruling a polis is not the same as ruling over a large household, that household management and politics are different in kind and not in terms of the size of the association (1.1.1252a9-13).(22)

In Politics 1.2, Aristotle then speaks of how the polis comes into being from the household (oikos)--which is formed from the associations dedicated to 1) sexual reproduction and 2) the rule of the foresighted over those lacking foresight--and later the village--an extension of the household by comprising several households (1.2.1252b10-30). Yet in some ways the polis is more than merely a sum of the numbers of villages or households within it.(23) The central role of the polis in human life is even more emphasized by the claim that the polis is by nature (1.2.1252b60-31).

A complete understanding of the polis as the political community necessitates a turn to Politics 3. As I have argued earlier, it is not only the third beginning of Aristotle's inquiry into politics, but it is the one which frames the rest--as well as the earlier specific discussions--of his Politics. Now, as I have argued, Politics 3.1 does not at any length discuss the polis as political community. In fact, the next place in Book 3 that discusses the polis is Politics 3.3, in the context of a debate of who is responsible for an action done in the name of the polis: the rulers of the polis or the polis itself (3.3.1276a6-9). This question is raised following the discussion in Politics 3.2, where Aristotle deals with the possible extension of citizenship following a revolution.(24) Thus, the issue of who is responsible is directly related to another question, whether it is just or unjust to make people who were not citizens citizens.(25) But the initial issue addressed in chapter 3 is not the justice or injustice of the act but who is responsible and whether the polis is the same polis or a different one depending upon who rules it.

To repeat: the question turns out to be "who is responsible when a revolution occurs?" Is the old ruler responsible for the acts of the polis, after they are disposed with a new ruling body? Also, Aristotle asks, given that a revolution has occurred, should the polis be spoken of as being the same as it was or is it somehow different (3.3.1276a17-19)? One must realize that Aristotle has yet to introduce the concept of the regime and, because of this, the discussion of revolution will be brought up elsewhere (i.e., Politics 5). These questions pose the framework in which the polis is brought up in Politics 3.3.

What is The Polis? Revisited

Concerning the polis and the way it is brought up in Politics 3.3, Aristotle suggests a way to understand the term:

Now the most superficial way of examining this question [about whether the polis remains the same after a revolution] concerns the location and the human beings [constituting a polis]; for the locations and the human beings can be disjoined, with some inhabiting one location and others another, and it will still be a polis (3.3.1276a19-22).

Aristotle argues that a superficial understanding of the polis merely deals with its location and its human inhabitants. It is superficial to understand the polis in these two senses, because, as he says, locations can be different and the human inhabitants could be "disjoined" and it could nevertheless remain a polis. This problem is clear from his comment that to understand the polis only in those terms is mistaken. He says of location,

The question in this form is to be regarded as a slight one, for the fact that the polis is spoken of in several senses makes the examination of such cases easy (3.3.1276a22-24).

He next speaks of the question of the inhabitants. He says of this issue, "And similarly in the case of human beings inhabiting the same location, [if one asks] when the polis should be considered [as] one [entity]" (3.3.1276a24-26). The question here is "what makes a polis a polis?"

In addressing what makes a polis a polis, Aristotle first rejects the view that a wall around an area makes it a polis (3.3.1276a26). He says that if walls make a polis then "it would be possible to build a single wall around the Peloponnese" (3.3.1276a26-27). Aristotle then suggests that "Babylon is perhaps

a polis of this sort, or any that has the dimensions of a nation (ethnos) rather than a polis" (3.3.1276a27-29).(26)

The fact that it is said to have taken three days to conquer Babylon seems to indicate the difference between what Babylon is and what a polis is (3.3.1276a29-30). Aristotle then drops the issue of size, saying that "this question will be useful on another occasion." Yet he goes on to say,

For the size of the polis--as regards both quantity and whether it is advantageous to have one or several [locations] --should not be overlooked by the statesman (politikos) (3.3.1276a31-33).

Aristotle then addresses the issue of how the character of the inhabitants affects one's understanding of the polis.

How Does The Polis Change

In addressing the issue of how differences or changes in the inhabitants affect the polis, Aristotle raises a few questions.

But where the same persons inhabit the same location, must it be asserted that the polis is the same as long as the stock of inhabitants remains the same, even though some are always passing away and some being born (as we are accustomed to speaking of rivers and springs as the same even though more water is always coming and flowing away)? Or must it be asserted that the human beings are the same for this sort of reason, but that the polis differs? (3.3.1276b33-40).

Aristotle seems to answer the first question by his analogy of rivers. Although the water may be constantly different, one nevertheless understands the river being referred to as the same river previously understood. So what is important is not the particular matter (or form) in question, be it water particles or persons, but the realization that the object in question is made up of water particles or persons.(27)

The second question is of more interest. It is of more interest because it addresses the issue of how polises differ from each other. Although the human character of the inhabitants remains, can it be that a change in the character of the inhabitants will cause a change in the polis?(28) Aristotle answers this concern as follows:

For if the polis is a type of community (koinonia), and if it is a community (koinonia) of citizens in a regime, if the regime becomes and remains different in kind, it might be held that the polis as well is necessarily not the same (3.3.1276a40-b3).

Although he has yet to explain fully the concept of the regime, he argues that a change in the character of the persons in the polis will cause a change in the regime. When the polis's regime changes it is generally understood that in some manner the polis itself also changes. This line of argument follows from two premises: 1) a polis is a type of koinonia and 2) the polis is a koinonia of citizens, which forms a regime. Thus, the logic of this argument is that if the character of the citizens change, the regime will change and if the regime changes so, too, will the polis.

From the above explanation--how a change in the make up (or character) of a thing can, in fact, lead to a change in a thing--one can see how one thing could change within a polis and its change could in turn change the character of the polis to such a degree that it is no longer what it was. In demonstrating how a change in a polis may alter it to such a degree that it is no longer what it was before, Aristotle gives an example of a chorus:

At any rate, just as we assert that a chorus which is at one time comic and at another tragic is different even though the human beings in it are often the same, it is similar with any other association (koinonia) and any other compound, when the compound takes a different form--for example, we would say that the mode is different even when the notes are the same, if it is at one time Dorian and at another Phrygian (3.3.1276b3-9).

Although the chorus may consist of the same members, it is different in character by the type of tune it sings.(29)

But here what is really different about the chorus is the song or type of song they sing. Therefore, the chorus is defined by the form of song it performs. So too with the city. A city is defined by the form it takes to reach its preferred end, or telos, and this is understood to be its regime.

By pointing out how the various differences in form affect our understanding of a thing, we come to understand the polis. We must come to grasp its form, i.e., its regime. Aristotle shows this by stating:

If this is indeed the case, it is evident that it is looking to the regime above all that the polis must be said to be the same; the term one calls it can be different or the same no matter whether the same human beings inhabit it or altogether different ones (3.3.1276b9-13).

Thus to understand the polis, and not only how it changes, but also its very character, we need to understand the regime, that which gives form to the polis. Yet this is not what is discussed in the rest of Politics 3.3. In fact, the regime will not be fully discussed until chapter 6.

The last sentence of chapter 3 suggests the question of whether it is just or unjust to fulfill agreements of the previous ruling body after a revolution--a form of the question which opens this chapter--is to be dropped. It is dropped because, as he says, "that it is another argument" (3.3.1276b113-15). Aristotle claims that the question of the justice of complying with agreements is in fact another question. That question is one of the issues of what is just and what is not just--the issue of justice itself--as we will see later, that will vary depending upon the nature of the regime in question. Another reason why the question must be dropped is that there is still not enough information regarding the parties and what was exactly done and why to make a correct answer to this question. Thus, Aristotle drops this question and moves on to another issue.

Thus, Aristotle ends Politics 3.3 by suggesting that the polis points to the important role the regime plays within his politics. Yet, one must remember that this discussion of the polis is in the middle of a discussion of the citizen, which will also point to the importance of the regime for Aristotle. Because the regime will be brought up later in Politics 3, a fuller and complete examination of the polis will have to wait until we come to understand the regime. Also because we must wait until the regime is addressed, other issues of the polis--such as the claim that it is by nature and human beings are political animals--must also be addressed later.


1. Some commentators on the Politics have said that the aim of Aristotle's political science--the polis--is not fundamentally different from the modern concept of the state. Some also speak as though the polis and the state are identical. The only difference, these scholars argue, between the polis and the state is size. This is why they tend to refer to it as the 'city-state.' See Barnes 1982 and 1987, 249-63; Mulgan 1977, 13-35 and 140; Irwin 1988, 407-22 and 1987, 73-98; Bradley 1991, 13-56; and Johnson 1990.

Whereas both Strauss 1978, 30-31 and 45-46 and Jaffa 1975 9-13 have given extremely plausible argument against, if not refuted, those who call the polis either a state or a city-state.

2. The tranlation used for this and all other passages from the Politics are from Lord 1984, although I felt free to alter his translations where I thought his word usage altered the meaning of the Greek.

3. Lord 1984 is not alone in this usage of calling the polis, city. See Strauss 1978, 30 and Nichols 1991.

4. See Strauss 1977, 30-33 and 1989, 37-50; and Jaffa 1975, 9-13.

5. See Strauss 1936, xv; de Alvarez 1989, xii-xvii and xxxii-xxxiii; Mansfield 1983, 849-57; and Hexter 1956, 113-138. Also see Strauss 1989, 39-55.

6. Strauss 1936, xv.

7. See Strauss 1953 and Manent 1994a and 1994b..

8. See de Piza 1994 and Kantorowicz 1957.

9. In this sense Foucault (1979, 3-31) is the radical outcome of Hobbes' line of argument concerning the abstraction of will and its relation to the body politic. Also Foucault's comments about the body of the sovereign indicates how liberal politics fail to understand how the underpinning of liberal politics depends upon a certain understanding of how to perceive the body politic.

10. See Rousseau, Second Treatise and The Treatise on Language.

11. See Kant 1991. It would be useful if one wished to understand the history of modern liberalism to turn to Manent 1994a. It is the best concise account of the origins and growth of liberal political thought.

12. See Hegel, The Philosophy of Right.

13. The above brief history of the modern state is indebted to Pierre Manent. See Manent 1994.

14. Viroli 1992, Mansfield 1983 and de Alvarez 1989, xiii-xviii.

15. The problem with translating polis as city is that that city is generally perceived to be an urban area. See Strauss 1978, 30. Compare to Hansen 1995.

16. See Hansen 1995. In this outstanding book, Hansen undermines the traditional understanding of the Greek Polis as an urban and commercial center, in favor of its being a rural and agrarian center. He also sees Aristotle defending the agrarian polis and its fundamentally democratic character from excess urbanization, which results in the forgetting of the polis's rural foundations.

17. Lord insists on translating koinonia as "partnership" (Lord 1984, 278). This is a very problematic and misleading translation of koinonia because partnership invokes being involved in not only some form of an association but more specifically in Anglo-American legal parlance, a pact or contract. Common usage speaks of members of a compact or contract as partners. Thus the usage of partnership for koinonia could lead one to understand Aristotle's use of koinonia in terms of it being some form of contract or compact. This is explicitly what Aristotle does not want to do. Later Aristotle will reject the view that polis is some form of contract (Politics 3.9). Given this one would think that Lord would find it necessary to somewhere defend his translation anywhere within his translation, however, he somehow does not find it necessary to so.

The reason, I believe, he does not feel it is necessary to defend his translation of koinonia is that he agrees with Ambler 1985. Ambler 1985 argues that the polis for Aristotle is not really natural. Rather, he holds the polis to be wholly conventional in character. This is because the regime is conventional in character and there can be no polis without a regime. However, I believe Ambler is utterly wrong in his reading of the naturalness of the city and would point the reader to Miller 1989 and 1995, 3-26 and 336-45; Arnhart 1990; 1994a; and Kullmann 1991.

Because of this, I am extremely unsympathetic to Lord's rendering of koinonia. Rather I prefer either the traditional rendering "association" or "community." In some ways community is a more preferable translation in that koinonia is related to koinos, common. Several Aristotle scholars also perfers to translate koinonia as community (Jaffa 1975, 12-13; Miller 1995, 54-55; and Voegelin 1957, 315-16).

On the other hand, Saxonhouse 1992, 189 translates it as "sharing." She does this to stress the sense of commonality. Although Saxonhouse's translation is more akward than community, nevertheless it is far more preferable to Lord's.

18. When we speak of community there is a tendency to speak of it as an abstraction. Real community is not an abstraction but a concrete environment of local assocations which if not form or shape the character of its inhabitants, it does indeed retrict and control their behavior. Concerning what community really is it would be useful to turn to Wendell Berry who says that "if the the community is to mean or amount to anything, it must refer to a place (in its natural integrity) and its people. It must refer to a placed people" (1993, 168).

19. Although I believe that political community best translates Aristotle's use of the word polis, in this work I will either leave it in the transliterated Greek or occasionally fall into calling it the political community, with the provision that it is not to be strictly understood as we understand the city today--as mass urban metropolis. This is contra Strauss 1978, 30.

20. See Wilson 1980, 80-96.

21. See Arnhart 1990, 477-92.

22. The scholars who use Aristotle's notion of political animals in order to attempt to defend the natural sociability of human beings against Hobbes' and Rousseau's rejection (e.g., Master 1989; Arnhart 1994a, 1988a, 1990; and Wilson 1993a and 1993b), nevertheless distort Aristotle by their undermining of Aristotle's distinction of the household and the political community.

23. To understand the relation between the parts to the whole in Aristotle's Politics 3 see Quinn 1986 and Saxonhouse 1993, 215-24.

24. The use of revolution here is ambiguous, in that one is not sure if a regime change has occured or only a change of the rulers has occured? It is my hunch that Aristotle's later account of revolution, found in Politics 5, is not implied in this passage.

25. In Politics 3.1-.5, the polis is said to be made of citizen parts, whereas in Politics 1.2, the parts of the polis were said to be the household and the village. Later in Politics 3 and 4-6, Aristotle will finally argue that the polis is composed of different classes of persons, each of which possess and live within villages and households. Regardless which view what exactly are the parts of the polis, the parts are always a composite of the differing parts. Again, for an explanation of the particular relationship of the parts to the whole in Politics 3, see Quinn 1986.

26. To translate ethnos as "nation" then to add to the identification of polis as a urban center rather than a political unit. The modern nation and the ethnos is merely roughly similar, in that it controls a vast area, but it differs in that the ethnos was more tribal and racial or ethnically unified than the modern nation state. Whereas the modern nation state, centers upon an abstraction of a given geographic location as sharing the same way of life and the same principles. Thus the concept of the modern nation more close resembles the polis as political entity than does the modern nation resemble the ethnos.

Aristotle has a interesting discussion of the difference between the polis, the ethnos and the empire in Politics 7.7.1327b19-28a15, where he equates the principle of the ethnos as freedom at the exclusion of civilization, empire as the principle of civilization at the exclusion of freedom, and the polis as a balancing of both these principle for the well being humans. Strauss says that the modern nation state is likewise a balancing of freedom and civilization and thus it can be seen as competing to see which one fulfill human happiness best (Strauss 1978, 30-31).

27. See Saxonhouse 1993, 218. She argues that Aristotle is addressing in a political context the metaphysical question posed by Heraclitus, about whether a river is still the same river even after all the water has changed. Heraclitus claims that all existence is flux and therefore naming a thing is a arbitrary act of the will, whereas Parmenedies claims it is pure stillness or status. Aristotle, on the other hand, argues that existence is a condition between pure flux and pure stillness, that it is understood to be a state of revealing or becoming.

28. This treatment of change (metabole) prepairs the reader for a more thorough discussion at Politics 5.

29. See Saxonhouse 1993, 218-219. The relation she points to is the tension between unity and disunion and the nature of the chorus is a unity that arises out of differences of the parts. Yet, the unity that is arrived at is a in between a pure unity and a pure flux of difference.

[an error occurred while processing this directive]