Language develops further through an important intermediary, namely, chance agreement (ittifāq).8 Alfarabi argues that particular words are fixed by chance agreement, as humans, making use of the sounds that their tongues have formed, begin to speak to one another in a fairly haphazard manner. Whenever two people happen to agree (ittafaq) to use a word in order to designate a certain meaning, they establish a linguistic convention (iṣṭilāḥ), which eventually spreads across a larger community (BL 137.17–21, 138.3–4, #120). Although the words established through these agreements eventually spread far beyond their originators, Alfarabi does not believe that even a long series of such agreements would be sufficient to create all the necessary words, so that a language-giver must arise at some point (138.4–8, #120). It is this language-giver who endows language with its conventional character (cf. PR 61.65, Ar. 70.6).9 Convention creates and standardizes what fails to come into being instinctively through nature or haphazardly through chance. Language could never emerge as an effective tool of communication among large groups, even for necessary things (BL 138.7–8, #120), until it becomes standardized and therefore conventional.10
Alfarabi explains the conventionality of language even more bluntly in his Commentary and Short Treatise on Aristotle’s “De interpretatione.” The primary sensibles and intelligibles are identical for all humans, experienced in the same way by Arabs and Indians. However, the words used to signify these terms, in writing and in speech, are already entirely conventional. The manifest proof of their conventionality is that they vary from Umma to Umma. Alfarabi goes so far as to compare the giver of utterances (wāḍi‘ al-alfāẓ) to the giver of laws (wāḍi‘ ash-sharī‘a): both are governors who establish pure conventions (CA 12.11–20, Ar. 27.9–18). To return from convention to nature seems quite impossible, at least with regard to language. If the simplest nouns are already conventional, then the thorough development of convention is the only way for language to evolve. Without it all humans would be equally natural, but equally ignorant: our natural faculties for observing the sensible and grasping the intelligible things (BL 135.6–14, #115) would simply remain uncultivated.
The Book of Letters confirms, and develops, this general point about the inherent conventionality of language. As the Umma and its language continue to evolve, they move farther and farther away from their natural origins. The more complex the language becomes, the more its speakers entrench themselves in the linguistic customs that have been standardized by earlier generations and passed down to their descendants (BL 141.16–142.4, #128). The gifts of beauty and refinement that are conferred upon the language by a class of language experts and storytellers serve to enhance its customary character (143.1 ff., 144.6–11, #130). The culmination of the process occurs with the development of linguistic science. Alfarabi explains how the dialect on which this science rests should be taken from the most remote and savage members of the Umma, whose life in the wilderness has protected them from contact with foreigners and left them least inclined to speak or understand utterances different from those to which they have been long accustomed (145.8–146.20, #143–44). Linguistic science, the last of the syllogistic arts developed by the Umma and peculiar to it (cf. 148.14–20, #138),11 is the most inextricably rooted in the arbitrary customs of the Umma to which it belongs. Every Umma is formed by nature, but consummated only by custom, so that by the end of its development the natural elements seem almost to have disappeared. Although the triumph of custom is not the end of this ongoing story, it is certainly an important part of it.
Alfarabi’s Definition of the Umma
I have shown that in Alfarabi’s view the Umma and its language emerge out of nature, but are perfected only through custom and convention. The link between convention and the Umma is not unique to Alfarabi, but implicit in the traditional Arabic meaning of the term. One of the rarer Qur’ānic meanings of Umma is “custom” or “way of life” (Qur’ān 43.22). Alfarabi echoes this Qur’ānic meaning in the Philosophy of Plato, by associating the Umma with a certain way of life (PP 22.18).12 Yet in the absence of any description of the conventions and ways of life characteristic of the Umma, our discussion is bound to remain vague. What kind of customary institution is the Umma, and how does it come to embody a specific way of life?
We may begin with a process of elimination. Many political matters that are elsewhere of the deepest significance to Alfarabi are not even mentioned in the long account of the evolution of the Umma in the Book of Letters, or in the shorter account of its physical traits in the Political Regime.13 The words for “regime” (siyāsa), “city” (madīna), “ruler” (ra’īs), “king” (malik), and “association” (ijtimā‘), which occur frequently in Alfarabi’s political works, do not appear at all.14 The only discernible allusion to political power comes in the sections describing the establishment of the language, which cannot take place without some kind of binding authority (BL 138.4–8, #120), and the influence of the Umma’s linguistic elite (143.5, #130). In these passages the word “governor” (mudabbar) and its cognates are indeed used. But Alfarabi restricts the governor’s authority to linguistic matters (143.5–6, #130). It seems unlikely that purely linguistic authority could be upheld without the help of some kind of external force, such as a king or a warlord. By saying nothing about political authority, Alfarabi may imply that it does not help or hinder the linguistic development of the Umma as much as one might think. The Greek Umma flourished under a large variety of mixed regimes, mostly in cities, while the Arab Umma flourished under tribalism followed by multinational empire.
While there are some vague references to politics in Alfarabi’s account of the origin of languages and Ummas, there are no references to religion at all. Alfarabi invokes the Arabs once, as a model of the proper method of collecting the words and phrases on which linguistic science is based (BL 145–47), but says nothing about the more famous exploits of their prophet Muhammad and his followers. In Alfarabi’s two thematic accounts of the Umma in the Book of Letters and the Political Regime, no connection between religion and Umma is established. This omission may have helped convince some scholars that Alfarabi ignores the religious Umma (cf. Vajda, 250), an impression that can and will be corrected by considering other passages.
The Umma is therefore reduced to two things: natural, physical traits caused by climate and nutrition, and the mainly conventional trait of language. Of the two sets of qualities, those arising from language are in every way the more dynamic. Although language is passed down from generation to generation, it is not merely a static, inherited trait. In an earlier passage in the Book of Letters, Alfarabi quotes the anonymous opinions of people who designate an Umma according to natural qualities inherited from parents, and moral qualities inculcated by them. These opinions employ the Arabic terms sha‘b and qabīla, frequently translated as “tribe” or “clan,” as synonyms of Umma (BL 98.11–12). The implication is that these people fail to properly distinguish between clan and Umma. By occupying themselves merely with ancestry and the moral upbringing afforded by parents, they are left unable to properly define the Umma (98.9–14, 100.5–7).15 Alfarabi points out an error that must have been endemic in his genealogy-obsessed age.16 He reminds his contemporaries that even if parents and ancestry determine the material constituents of the child, they do not fix its form, or moral and intellectual character, any more than strong wood fixes the shape of a bed (99.21–100.5). While education within the home has an undeniable effect on character, it is wrong to assume17 that all education takes place there (98.17–99.4). Language is learned from elders, but not solely from parents (142.1–2, #128, 143.8–9, #130). The broad, public dissemination of language lends the evolution of the Umma a vitality that could not be sustained within individual families or even small tribes. It allows some members of the Umma to enjoy a much richer education than could ever be provided by their immediate clans.18
The significance of the link between the Umma and language