without any reference to the number of inhabitants (SL 157, 5.14, Ar. 141; see Parens 1995, 62–67). Averroes’s approach to Plato’s Republic is even more explicit. One passage in the dialogue appears to limit the city to a thousand guardians (Republic 423a9). Averroes praises the view that a community should have a limited size, but adds that its proper bounds vary considerably across space and time, and can be fixed in each case only by political deliberation (Averroes 1974, 46.1–5). Plato’s choice of only one thousand guardians was appropriate for Greece, but in Averroes’s own time the required number would be somewhat larger (46.5–15).
Nevertheless, Mahdi takes a statement that occurs in the Virtuous City to imply a certain preference for the city. “The city whose association intends cooperation concerning the things through which true happiness is attained, is the virtuous city …. The Umma, all of whose cities cooperate to attain happiness, is the virtuous Umma” (VC 230.9–10). In Mahdi’s view, the second phrase indicates that “The virtuous nation [Umma] … presupposes virtuous cities” (Mahdi 2001, 143). This reading would be convincing if the cities of the virtuous Umma were themselves qualified as virtuous. But since they are not so qualified, it remains unclear whether cooperation entails virtue filtering up from the cities to the Umma or down from the Umma to the cities.42 Perhaps both are possible, in which case the view of the equal receptivity of cities and Ummas to virtue and happiness still holds. The obscurity of these passages does not end here. Galston has observed that while mere “happiness” is attained in the Umma, “true happiness” is attained only in the city (Galston 1990, 152; VC 230.7–9). It is hard to grasp the meaning of this distinction. Alfarabi sometimes sets “true happiness” in opposition to “what is supposed to be happiness without being such” (BR 101.20–22, Ar. 52.10–12), but does not do so here. As tempting as it may be to regard the mere “happiness” of the Umma as inferior to the “true happiness” of the city, we lack any clear point of reference for interpreting this distinction.43
The ambiguity of these passages remains frustrating. As Galston concludes, “The relationship between cities and nations in Alfarabi’s political thought is ambiguous and needs to be studied further” (Galston 1990, 153). We will have many occasions to study this issue further throughout the rest of this book. Even at this stage of the discussion, it already seems clear that Alfarabi’s ambiguity about the superiority of the city to other political forms contrasts starkly with Aristotle’s forthright declaration of it.
It must be emphasized, however, that Aristotle’s focus on the polis and its uniqueness does not betray any innocence about the significance of the ethnos. It is discussed at some length in the opening chapter of the Politics. According to Aristotle, both the city and the ethnos emerge gradually from the household. But only the former is based on partnership, first of households into villages, then of villages into cities (Politics 1252b15–31). In both cases, the partnership implies a degree of political cooperation that transforms the character of the original community, permitting it to live a full and independent political life for the sake of living well (1252b28–31, 1280b33–35). This seems to involve, among other things, ruling and being ruled in turn (1277a30–33, b8–16). But with regard to the ethnos, Aristotle says nothing about partnership or shared rulership. Instead, the ethnē are ruled by kings, just as households are ruled by patriarchs. The original community of the household becomes bigger in size, first to a village and then to an ethnos, but does not change its patriarchal structure (1252b16–20). There is no clear limit to how large this community can become: kings may equally rule a polis, an ethnos, or many ethnē (1285b33–34). The politeia, however, can flourish only in a small community such as the polis, where everybody knows one another and the assembly or army is not so large as to be unable to hear the voice of orators (1326b2–7). For Alfarabi, political cooperation in its highest sense is possible in a community of any size (VC 230.7–11), while for Aristotle political partnership is limited to the polis. It is important to note the difference in meaning between the Greek koinōneia and its imperfect Arabic translation ta‘āwan: while the former implies a measure of equality and political participation, the latter, usually rendered into English as “cooperation,” could easily describe a strictly hierarchical society where everybody performs his particular function well. This is indeed how Alfarabi proceeds to describe the virtuous city (230.12 ff.).
Despite these differences between Aristotle and Alfarabi on the relative stature of nations and cities, both emphasize the actual prevalence of nations. Aristotle implies that the ethnos ruled by a king, far from being an outlier, is the kind of government and society that initially prevails everywhere. He observes that all humans speak of the gods as being ruled by a king, thus assimilating the gods to themselves (Politics 1252b24–27). “All humans” surely includes the Greeks, since they too were ruled by kings when they first fancied Zeus to be king of the gods. By quoting Homer in this context (1252b22–23), Aristotle indicates that the polis as he knew it was yet to exist in Homeric times. Even among the Greeks of Aristotle’s own time, the polis had not established itself universally. Aristotle speaks explicitly of Greek ethnē, such as the Arcadians, who have never managed to organize themselves into a polis (1261a29–31). He later suggests that the Greeks are divided into ethnē of various qualities, some as wild as the people of Europe and others as docile as the people of Asia (1327b33–36). Among the non-Greeks, the polis hardly seems to exist at all: even nominal cities such as Babylon are large enough to be called ethnē (1276a28–30).
The foregoing analysis suggests that in Aristotle’s view the ethnē comprise the vast majority of humankind, and even a significant portion of the Greeks. The bulk of Aristotle’s work on politics treats a political form that by his own acknowledgment is quite rare. Aristotle evidently thinks that there is something unique and noble about the vibrant, self-sufficient political life of the polis. The ethnos does have a distinct meaning, but a rather negative one: it is a loosely defined community that cannot be governed in the free, political manner of a city (1326b5). The Greek-barbarian distinction becomes less important than the distinction between potentially free cities and unfree ethnē.44 Aristotle’s exclusion of the ethnē from the blessings of political life attained only in a few cities could even be said to foreshadow a certain Judaeo-Christian usage of the term, according to which a few faithful believers live surrounded by many, almost anonymous, gentiles.
The elevation of the city vis-à-vis the nation constitutes a central theme of Aristotle’s Politics, but it is absent in Alfarabi. And why would Alfarabi have wanted to elevate the city? While the city was rare enough in Aristotle’s time, and limited mainly to Greece, in Alfarabi’s Babylon and neighboring Byzantium its last vestiges had long been engulfed by a series of vast empires and sweeping claims to revelation. The only kind of government known to Alfarabi and his contemporaries was kingly and imperial: this remained true in the Islamic world well into the twentieth century. Praising or even discussing a political form that no longer existed might have appeared hopelessly anachronistic. The absence of independent cities or popular governments in the thought and practice of the medieval Islamic world might be another major reason why the Politics never gained much currency within it. Unfamiliarity with the contents of the Politics would have discouraged translations, and even if translations were made, philosophers might have been leery of publicizing something that would have seemed so preposterous to much of their audience. The political situation in thirteenth-century Europe, in which the Politics publicly resurfaced, was already quite different: republics had emerged in northern Italy, so that thinkers like Marsilius and even Thomas Aquinas felt free to ask the Aristotelian question of whether kings, aristocracies, or popular assemblies should rule.45 The rediscovery of the Politics helped stimulate a centuries-long European debate over which class of people should govern: no comparable debate occurred in the Islamic world until the twentieth century.46 I propose this as a plausible answer to the query with which Brague concludes his article: how do we explain the relative lack of interest in the Politics in the Hellenistic, Roman, Byzantine, and Muslim worlds (Brague 1993, 432)?
Our analysis of Plato and Aristotle has shown that their treatment of nations does not display simple indifference or contempt. On