FAÚRAÚBÈ, ABUÚ NAS®R, Muslim philosopher of the 4th/10th century.


The sources for the life of Fa@ra@b^ are such as to make the reconstruction of his biography beyond a mere outline nearly impossible. The earliest and more reliable sources, i. e., those composed before the 6th/12th century, that are extant today are so few as to indicate that no one among Fa@ra@b^'s successors and their followers, or even unrelated scholars, undertook to write his full biography, a neglect that has to be taken into consideration in assessing his immediate impact. His fame, however, began to grow, apparently in association with and as a result of the renown of Avicenna (q.v.) who, through his explicit recommendation and endorsement of Fa@ra@b^ in his writings, presented himself as Fa@ra@b^'s successor in philosophy. When major Arabic biographers came to write comprehensive entries on Fa@ra@b^ in the 6th-7th/12th-13th centuries, the period of the greatest expansion of philosophical studies in Islamic lands, there was very little specific information on hand; this allowed for their acceptance of invented stories about his life which range from benign extrapolation on the basis of some known details to tendentious reconstructions and legends. Most modern biographies of the philosopher present various combinations of elements drawn at will from this concocted material.

The extant original sources are of two basic kinds, documentary and narrative. The documentary sources consist of notations and other incidental information in manuscripts of Fa@ra@b^'s works relating to his biography; few though these may be, they are our most reliable sources. The narrative sources also fall into two major categories, those written before the 6th/12th century and those after. The sources prior to the 6th/12th century consist of: (1) an autobiographical passage by Fa@ra@b^, preserved by Ebn Ab^ Osáaybe¿a (II, pp. 134-35), tracing the transmission of the instruction of logic and philosophy from antiquity to his days. Fa@ra@b^ situates himself also in this process at the very end of it, which would appear to be the purpose behind his otherwise quite tendentious account (Gutas, 1999), but there is little reason to doubt the specific statements about his immediate predecessors and himself. (2) Reports by Mas¿u@d^ (Tanb^h, p. 122), Ebn al-Nad^m (ed. Flügel, p. 263, ll. 8-14), and Ebn H®awqal (pp. 510-11), all younger contemporaries of Fa@ra@b^, as well as by S®a@¿ed Andalos^ (d. 462/1070), the first biographer to have devoted an entry to him (pp. 53-54). These reports are implicitly reliable though of limited extent. This is significant in the case of the Andalusian tradition represented by S®@a@¿ed, for it can be assumed that it would contain all accessible information on Fa@ra@b^ because of its almost exclusive adherence to his philosophy. S®@a@¿ed, however, though he summarizes a number of Fa@ra@b^'s books, knows nothing more about his life than the preceding sources other than the mere fact of his association with the Hamdanid Sayf-al-Dawla (r. 333-56/944-67).

The sources from the 6th/12th century and later consist essentially of three biographical entries, all other extant reports on Fa@ra@b^ being either dependent on them or even later fabrications: (1) the Syrian tradition or collection of biographical narratives on Fa@ra@b^ represented by the entry by Ebn Ab^ Osáaybe¿a (II, pp. 134-40), and to a lesser extent by Ebn al-QeftÂ^ (pp. 277-80); (2) the pro-Turkish tradition, compiled and composed as a continuous narrative by Ebn K¨alleka@n with the purpose of documenting a Turkish ethnic origin for Fa@ra@b^ (ed. ¿Abba@s, V, pp. 153-57; tr. de Slane, III, pp. 307-11); and (c) the scanty and legendary Eastern tradition, represented by Z®ah^r-al-D^n Bayhaq^ (pp. 16-20, no. 17). Of these, the Eastern tradition of Bayhaq^ (q.v.; d. 565/1169) can be discounted: the few accurate data derive from the earlier sources, whereas the added material is obviously fabricated. Ebn al-QeftÂ^ (or the extant epitome of Zawzan^, compiled in 647/1249) actually offers a combination of the Andalusian and Syrian traditions, for he copies S®a@¿ed for the most part and has additional material only on Fa@ra@b^'s association with Sayf-al-Dawla. This leaves the Syrian and pro-Turkish traditions of the biographical entries in Ebn Ab^ Osáaybe¿a's ¿Oyu@n (final recension completed in 667/1268) and in Ebn K¨alleka@n's Wafaya@t (completed in 669/1271) respectively. These present themselves as our most extensive and detailed sources though they date a good three centuries after Fa@ra@b^'s death. Ebn Ab^ Osáaybe¿a's entry, which is the earlier one, consists of a collection and patching together of all the diverse pieces of information that were available to him in Syria at that time. It includes much legendary material, but Ebn Ab^ Osáaybe¿a also quotes Fa@ra@b^ where he can. Ebn K¨alleka@n's entry, by contrast, is a response to that of Ebn Ab^ Osáaybe¿a: the latter had mentioned at the beginning of his entry, and for the first time by any extant biographer, that Fa@ra@b^'s father was of Persian descent; Ebn K¨alleka@n's entry is completely animated by the effort to prove that Fa@ra@b^ was ethnically Turkish. To this end, Ebn K¨alleka@n first gave Fa@ra@b^ an additional nesba, one he never had, al-Tork^. Abu'l-Feda@÷, who copied Ebn K¨alleka@n, corrected this, and changed the word, al-Tork^ "the Turk," which reads like a nesba, to the descriptive statement, wa-ka@na rajolan tork^yan "he was a Turkish man" (Mokòtasáar II, p. 104). Second, at the end of his entry, Ebn K¨alleka@n spent considerable time giving the correct spelling and vocalization of all the names which he says are Turkish and are associated with Fa@ra@b^: the names of his alleged grand- and great-grandfather, T®arkòa@n and Awzalag@ (adding explicitly, wa-homa@ men asma@÷ al-tork, "these are Turkish names"), and the toponymics of his origins, Fa@ra@b, OtÂra@r, Bala@sag@u@n, and Ka@œg@ar (the information on the toponymics is derived from Sam¿a@n^, under the nesba al-Fa@ra@b^, though Sam¿a@n^ does not refer to the philosopher). In between, Ebn K¨alleka@n offers a continuous narrative of Fa@ra@b^'s life as reconstructed by him.


Since almost every detail of Fa@ra@b^'s life found in one source is contradicted by that in another, it will be helpful to list first those items from the documentary and earlier narrative sources which are certain and to present the dubious and legendary material on the later sources in the next section. His name was Abu@ Nasár Mohá®ammad b. Moháammad Fa@ra@b^, as all sources, and especially the earliest and most reliable, Mas¿u@d^, agree. In the famous passage about the appearance of philosophy preserved and reported by Ebn Ab^ Osáaybe¿a (II, p. 135 ll. 20-21), he is quoted as having said that he had studied logic with Yu@háanna@ b. H®ayla@n up to and including Aristotle's Posterior Analytics, i.e., according to the order of the books studied in the curriculum, Fa@ra@b^ said that he studied Porphyry's Eisagoge and Aristotle's Categories, De Interpretatione, Prior and Posterior Analytics. His teacher, Yu@háanna@ b. H®ayla@n, was a Christian cleric who abandoned lay interests and engaged in his ecclesiastical duties, as Fa@ra@b^ reports. His studies of Aristotelian logic with Yu@háanna@ in all probability took place in Baghdad, where Mas¿u@d^ tells us Yu@háanna@ died during the caliphate of al-Moqtader (295-320/908-32). This is further indicated by the entire approach and contents of his logical work, which is imbued with the thought world of Alexandrian Aristotelianism as resuscitated in Baghdad by Abu@ Beœr Matta@ and his teachers (Zimmermann, pp. lxviii-cxxxix; see also below, section on Fa@ra@b^ and Greek philosophy). Fa@ra@b^ apparently stayed on and worked in Baghdad. Yaháya@ b. ¿Ad^, a resident of Baghdad according to Ebn al-QeftÂ^ (p. 361 ll. 9-10), was among his students, and he composed at least two of his works for Baghdad personalities: a treatise on the validity of astrology for the Christian scholar and translator Abu@ Esháa@q Ebra@h^m b. ¿Abd-Alla@h Bag@da@d^ (Mahdi, 1975-76, p. 265) and his great book on music for the vizier of the caliph al-Ra@zµ^, Abu@ Ja¿far Moháammad b. Qa@sem Karkò^ (Keta@b al-mu@s^q^ al-kab^r, pp. 30 and 35, n.1). We know that he was definitely in Baghdad until the end of the year 330/September 942. As we learn from notes in some manuscripts of his Maba@de÷ a@ra@÷ ahl al-mad^na al-fa@zµela, he had started to compose the book in Baghdad at that time and then left and went to Syria. He took the book with him, and he finished it in Damascus the following year (331), i.e., by September 943 (cited in Fa@ra@b^'s Keta@b al-mella, p. 79 and by Ebn Ab^ Osáaybe¿a II, pp. 138-39; tr. in Mahdi 1990, pp. 721-22). In Syria Fa@ra@b^ also lived and taught for some time in Aleppo; Ebn al-QeftÂ^ mentions that he went to Aleppo to Sayf-al-Dawla, a report that is corroborated by another manuscript note, copied by Ebn Ab^ Osáaybe¿a (II, p. 139 ll. 19-20), which says that he had dictated a commentary on the Posterior Analytics to Ebra@h^m b. ¿Ad^, a student of his in Aleppo (and Yaháya@'s brother). Later on Fa@ra@b^ visited Egypt; the note in the manuscripts of the Maba@de÷ also informs us that he wrote the six sections (fosáu@l) summarizing the book in Egypt in 337/July 948-June 949. He must have returned to Syria, however, soon thereafter, for we are certain of his place and date of death: Mas¿u@d^, writing barely five years after the fact (955-6, the date of the composition of the Tanb^h), says that he died in Damascus in Rajab 339 (between 14 December 950 and 12 January 951). His stay in Syria was somehow associated with Sayf-al-Dawla, though we do not know precisely how, how long, and in what capacity. S@a@¿ed al-Andalos^, the first to report this connection, simply says (p. 54 l. 19) that Fa@ra@b^ "died in Damascus in 339 under the protection (f^ kanaf)" of Sayf-al-Dawla. Later biographers greatly embellish this association.


The above is all that can be said with certainty about Fa@ra@b^'s biography. The remaining reports in the later sources are dubious at best and legendary at worst, beginning with his pedigree and origins. There is confusion and uncertainty, first of all, about the names of his grandfather and great-grandfather, which are given variously by the sources. The consensus in secondary literature is to list T®arkòa@n as the grandfather's name, but this is not supported by the sources, some of which do not have it at all (most of the earliest sources), while others have it as the name of the great-grandfather (Fehrest; Ebn Ab^ Osáaybe¿a), and Ebn K¨alleka@n has it as the name of the father (a discrepancy which was remarked upon by the careful S®afad^, I, p. 106).

Actually, it would seem none of them is right. In some manuscripts of Fa@ra@b^'s works, which must reflect the reading of their ultimate archetypes from his time, his full name appears as Abu@ Nasár Moháammad b. Moháammad al-T®arkòa@n^, i.e., the element T®arkòa@n appears in a nesba (Fa@ra@b^, Keta@b al-mu@s^q^ p. 35, note 1; Aháka@m al-noju@m, p. 46). This indicates that T®arkòa@n was not necessarily the name of Fa@ra@b^'s grandfather but rather that of a more distant relative from whom his family claimed descent (cf. Sam¿a@n^, ed. Yama@n^, IX, p. 63, s.v. the nesba al-T®arkòa@n^). Moreover, if the name of Fa@ra@b^'s grandfather was not known among his contemporaries and immediately succeeding generations, it is all the more surprising to see in the later sources the appearance of yet another name from his pedigree, Awzalag@. This appears as the name of the grandfather in Ebn Ab^ Osáaybe¿a and of the great-grandfather in Ebn K¨alleka@n. Ebn Ab^ Osáaybe¿a is the first source to list this name which, as Ebn K¨alleka@n explicitly specifies later, is so to be pronounced. In modern Turkish scholarship the pronunciation is given as Uzlug@ (ËA V, p. 451), without any explanation. The first appearance of this distinctly Turkish sounding name in the later sources in the context of attempts to claim a Turkish ethnic background for Fa@ra@b^ is accordingly questionable.

The nesba, universally given as al-Fa@ra@b^, would indicate a place of ultimate origin in the district of Fa@ra@b (the older Persian form Pa@ra@b is given in Háodu@d al-¿a@lam) on the middle Syr Darya (Jaxartes). This is corroborated by the geographer Ebn H®awqal, a younger contemporary of Fa@ra@b^ who was also somehow associated, like Fa@ra@b^ later in his life, with the Hamdanid Sayf-al-Dawla, since the first edition of his famous S®u@rat al-arzµ was dedicated to that prince. Ebn H®awqal notes from his travels in Transoxania that Fa@ra@b^ was "from" (men) the town of Vas^j in Fa@ra@b (EstÂak@òr^ does not mention Fa@ra@b^ in association with Vas^j). This has been taken to mean that Fa@ra@b^ himself was born there, but this need not be necessarily the case. Ebn H®awqal is contradicted by no less an authority than Ebn al-Nad^m, who was also a younger contemporary of Fa@ra@b^ and had close personal contacts with Yaháya@ b. ¿Ad^, Fa@ra@b^'s most successful student, from whom he received a significant amount of his information about philosophical studies for his Fehrest. Ebn al-Nad^m states (ed. Flügel p. 263 l. 9) that Fa@ra@b^'s origins (asáloho) lie in Fa@rya@b in Khorasan (men al-Fa@rya@b men arzµ Kòora@sa@n), that is, the town half way down the road from Marv-al-ru@dò to Balkò. Bayhaq^ in his Tatemmat Sáewa@n al-háekma (p. 16.7) conflates the two traditions and says that Fa@ra@b^ was "from Fa@rya@b in Turkestan."

These variants in the basic facts about Fa@ra@b^'s origins and pedigree indicate that they were not recorded during his lifetime or soon thereafter by anyone with concrete information, but were rather based on hearsay or probable guesses. When in the 7th/13th century Fa@ra@b^'s ethnic origin was made into an issue by the biographers, dogmatic statements without acknowledgment of source begin to appear. We thus hear for the first time, from Ebn Ab^ Osáaybe¿a, that Fa@ra@b^'s father was a commander of the army and of Persian (fa@res^) descent, to which Ebn K¨alleka@n responded as described above. Ultimately pointless as the quest for Fa@ra@b^'s ethnic origins might be, the fact remains that we do not have sufficient evidence to decide the matter.

There was also no information about Fa@ra@b^'s early life and studies, other than the mere fact that he had studied logic with Yu@háanna@ b. H®ayla@n. The fabricated stories accordingly proliferated. In this case, as perhaps in others, these stories may have started as plausible extensions of the little that was known about him or as well-intentioned elaborations of elements in his works. Since, as his nesba indicated, he was supposed to be from the East, he was said to have come to Baghdad as a young man and to have studied there. Almost every source mentions Yu@háanna@ b. H®ayla@n as his teacher, except that Ebn K¨alleka@n adds Abu@ Beœr Matta@, plausibly because by Ebn K¨alleka@n's time Abu@ Beœr had gained the reputation of having been the master logician before Fa@ra@b^ himself, and Fa@ra@b^ could not but have studied with the best. Ebn Ab^ Osáaybe¿a (II, p. 136 ll. 23-24) also mentions, citing an otherwise unidentified History (Ta÷r^kò), that Fa@ra@b^ used to take grammar lessons from the famous Baghdadi grammarian Ebn al-Sarra@j (d. 316/928) in exchange for logic lessons, a report which would appear to be based on the numerous references to grammar in Fa@ra@b^'s works as well as on Ebn al-Sarra@j's reputation of having introduced logic in his grammatical analyses (cf. the discussion and references in Zimmermann, pp. cxviii-cxxii).

A similar basis in Fa@ra@b^'s works would seem to lie also behind the legends of Fa@ra@b^'s talent for languages. As is well known, in his incessant efforts to differentiate between universal logical structures of thought and particular grammatical structures, Fa@ra@b^ has in a number of his works references and glosses in Persian, Sogdian, and Greek (but no Turkish; cf. Walzer, 1985, p. 3), and this apparently gave rise to his reputation as a polyglot (Zimmermann, p. lxxvii n. 2). There are variations on this theme, depending on the purposes of the author using it. Ebn K¨alleka@n, in line with his pro-Turkish bias, makes the outlandish claim that Fa@ra@b^ knew no Arabic when he came to Baghdad but only "Turkish and numerous other languages," and that he mastered Arabic only afterwards. Later on, in his story about Fa@ra@b^ at the court of Sayf-al-Dawla (see further below), he has Fa@ra@b^ say that he knew more than seventy languages. Related to this would appear to be an interpolated statement in a manuscript of Fa@ra@b^'s account of the appearance of philosophy preserved in Kabul (Library of the Ministry of Information, Arabic MS 217, fol. 154r; cf. S. De Laugier de Beaurecueil, Manuscrits d'Afghanistan, Cairo, 1964, p. 293, no. 40). The Arabic text from this manuscript remains unpublished, but according to Mahdi (1971, pp. 523-24), after the report that Fa@ra@b^ studied with Yu@háanna@ b. H®ayla@n (i.e., after what corresponds to the end of the passage quoted by Ebn Ab^ Osáaybe¿a, II, p. 135 l. 24), the Kabul manuscript adds, "After this, he traveled to the land of the Greeks and stayed in their land for eight years until he completed [the study of the] science[s] and learned the entire philosophic syllabus." If this is accurate, it would be another instance of the legend of the polyglot Fa@ra@b^, this time also based on his reputation as expert of the Greek syllabus. This is further corroborated by a statement reported by S®afad^ (I, p. 106) which presents the gist of the legend of the polyglot Fa@ra@b^: "It is said that he learned philosophy from the Greek language because he knew it as well as other languages." Qazv^n^ (d. 682/1283), a later representative of the Eastern tradition, invents yet another accomplishment; he says that Fa@ra@b^ translated Aristotle into Arabic from Greek (AÚt¯a@r, p. 548).

From the Syrian tradition, Ebn Ab^ Osáaybe¿a reports two romantic stories about Fa@ra@b^'s youth. The first, reported personally to Ebn Ab^ Osáaybe¿a (háaddat¯an^) by the famous theologian AÚmed^ (d. 631/1233), says that Fa@ra@b^ was a warden in a Damascus orchard in his early life and that he used to read philosophy at night by the light of the watchman's lamp. The other story, apparently a variant of the first and reported anonymously by Ebn Ab^ Osáaybe¿a ("from the manuscript notes of some shaikh"), has it that Fa@ra@b^ was a judge (qa@zµ^) in early life but gave it all up for philosophy, which he would study by the light of watchmen's lamps. The same anonymous source also adds that Fa@ra@b^ took up philosophy when he accidentally read some books by Aristotle left with him for safekeeping. As is Ebn Ab^ Osáaybe¿a's wont, he does not try to harmonize these stories with the others about his education in Baghdad.

With regard to Fa@ra@b^'s education and early life, Ebn K¨alleka@n makes up yet another story on the basis of what he read in the available literature. In the parallel account by Mas¿u@d^ of the transmission of philosophical instruction from Alexandria to Baghdad (Tanb^h, pp. 121-22; cf. Stern; Gutas, 1999), Mas¿u@d^ says that the teaching of philosophy was transferred "in the days of al-Motawakkel to H®arra@n and then in the days of al-Mo¿tazµed it eventually got to Qowayra@ and Yu@háanna@ b. H®ayla@n, who died in Baghdad during the reign of al-Moqtader. . ." From the context, Ebn K¨alleka@n apparently inferred that Yu@háanna@ was still in H®arra@n when he taught Fa@ra@b^—since Mas¿u@d^ does not explicitly mention that the teaching was transferred to Baghdad but only that Yu@háanna@ had died there—and thus has Fa@ra@b^ take a side trip from Baghdad to H®arra@n to study with Yu@háanna@ and then return to Baghdad.

Fa@ra@b^'s association with Sayf-al-Dawla, understandably, figures prominently in the Syrian tradition. Ebn al-QeftÂ^ first mentions that Sayf-al-Dawla honored Fa@ra@b^ and offered him protection. The anonymous source (œaykò) of Ebn Ab^ Osáaybe¿a adds that he would accept as stipend no more than four silver dirhams per day—a paltry sum, in contrast with the notorious liberality of Sayf-al-Dawla—and that upon his death funeral services were led by Sayf-al-Dawla himself. Ebn K¨alleka@n repeats these stories and adds that he was buried outside the Ba@b al-S®ag@^r in the south of Damascus. Ebn K¨alleka@n is also the only source to mention that Fa@ra@b^ was eighty years old when he died. Bayhaq^ finally, has the romantic story of Fa@ra@b^ being killed by highway robbers on his way from Damascus to Ascalon. As A. Adnan Adévar suggests (ËA V, 453a), this would appear to be in imitation of the story of the death of the poet Motanabb^.

The personal qualities which these stories in the later sources are designed to convey are single-minded devotion to study accompanied by frugality bordering on asceticism, or what was then regarded as the ideal philosophical way of life. He is said to have had no care for possessions or gainful employment and to have subsisted on little more than soup and wine. Ebn al-QeftÂ^ adds that he would go around in "Sufi garb" (be-zeyy ahl al-tasáawwof). Understandably, as already discussed, this detail in Ebn K¨alleka@n becomes "Turkish garb" (be-zeyy al-atra@k). Ebn K¨alleka@n also adds that Fa@ra@b^ liked solitude, and that he would compose his works and meet with students in parks by the pond. As for his accomplishments, other than his alleged knowledge of languages mentioned above, and universal praise as philosopher, mention is also made of his musical talent, obviously deriving from his massive work on musical theory and practice. This occurs most prominently in the story told by Ebn K¨alleka@n about how Fa@ra@b^ entered the court of Sayf-al-Dawla unannounced and unrecognized and, after exhibiting an astounding knowledge of languages, made people laugh, reduced them to tears, and put them to sleep by playing on his lute that he would tune differently for each purpose. Interestingly, a similar story about Fa@ra@b^ is reported by Bayhaq^ from an unidentified work "Morals of the Philosophers" (Akòla@q al-háokama@÷), only this time it occurs in the court of the S®a@háeb b. ¿Abba@d (q.v.) in Ray. Bayhaq^ appears to be oblivious to the anachronism: the S®a@háeb was about twelve when Fa@ra@b^ died and, as far as we know, Fa@ra@b^ never went to Ray. Qazv^n^ repeats the story about Fa@ra@b^ and the S®a@háeb, and adds another one of his own: Fa@ra@b^ would go from city to city in disguise because the kings were looking for him, and in each city that he liked he would buy a house and orchards and slaves and slavegirls. When he tired of them, he would marry off the girls to the slaves, give them the real estate as a present, and leave and never come back. One can see how with each retelling, here as in the previous stories, the legendary material snowballs with more colorful additions.

As for the real Fa@ra@b^, beyond the idealized image of abstemious philosopher of the later biographers and of the Turkish philosopher of Ebn K¨alleka@n, we have no direct means of knowing. Indirectly, it can be observed that Fa@ra@b^ went through life unnoticed (epitomizing the Epicurean precept la‚the bio@sas), which would explain the lack of general information about him. Even his senior colleague in Baghdad, Abu@ Beœr Matta@, engaged in social activity and debated the grammarian S^ra@f^, but there are no such reports about Fa@ra@b^, not even by Tawhá^d^, the great gossip of philosophical gatherings. After all those years of teaching and writing, we hear only of two disciples, the brothers Ebn ¿Ad^, Yaháya@ and Ebra@h^m. Yaháya@ eventually headed the Aristotelian school in Baghdad, and Ebra@h^m apparently accompanied his teacher to Aleppo. These, and the few patrons for whom he wrote some works, are the only traces of Fa@ra@b^'s social life. Indirectly also, we are left with his works, but they reveal very little, if anything, of his personal circumstances. There have been attempts to read into his philosophical works his life and religious beliefs—like Walzer's suggestion (1975, p. 5) that he was an Imami Shi¿i on the basis of his statements in the Maba@de÷, or Mahdi's conjecture (1971, p. 524a) that he studied in Constantinople (!) in order to explain his allegedly particular brand of Plat onism—but these are speculative in the extreme and are best avoided. We must be content with Fa@ra@b^'s philosophical legacy in his works.


Fa@ra@b^'s works have been listed in a conventional and uncritical way a number of times (see the Bibliography). There is no critical inventory on the basis of the manuscripts that will identify double listings under different titles and eliminate misattributions, much less are there historical and philological studies of the transmission of individual works. These are tasks for the future. A first attempt at analyzing the transmission of the Maba@de' was made by Mahdi 1990.

The narrative biographies present some interesting information in this regard that should be kept in mind. Ebn al-Nad^m, first of all, knows surprisingly few titles of Fa@ra@b^'s works—essentially, some commentaries and paraphrases of Aristotle. This fact needs to be explained in studying the transmission of his works. The report by Bayhaq^, whose credulity for legends is compensated for by his bibliographic acumen, complements that of Ebn al-Nad^m. He says that in his time, i.e., the middle of the 6th/12th century, few of Fa@ra@b^'s books were available in Khorasan; most of them were to be found in Syria. In addition, he says that he saw in the library of the chief Shi¿ite leader (naq^b al-noqaba@÷) in Ray autograph manuscripts by Fa@ra@b^ as well as works of his copied by his student Yaháya@ b. ¿Ad^ (who was, incidentally, a scribe by profession). The availability of Fa@ra@b^'s books in Syria is to be expected; some of these would be the redactions, or clean copies, prepared by the other brother, Ebra@h^m b. ¿Ad^ who was, according to Bayhaq^ again, Fa@ra@b^'s redactor or editor (modawwen). On the other hand, the copies in Ray were by the hand of the other brother, Ya@háya@, and these would have come there from Baghdad. The detail about the Shi¿ite leader's library also needs to be taken into consideration in studying the dissemination of Fa@ra@b^'s works among Shi¿ites. Furthermore, this report indicates two additional things: first, that there was no interest in his works in Khorasan, and perhaps in the East generally, right after his death and until the appearance of Avicenna; and second, that when Avicenna's work made philosophy a popular subject in the East, Fa@ra@b^'s works were overshadowed by those of Avicenna, at least until Bayhaq^'s time. This is to be contrasted with Ebn K¨alleka@n who, writing in Syria and Egypt about a century after Bayhaq^ says that most of Fa@ra@b^'s works were transmitted in the form of disconnected paragraphs and notes, and that some of them were found incomplete and dispersed because he used to compose on loose pieces of paper and not in (bound) quires. Though the reason adduced by Ebn K¨alleka@n is apparently fictitious (it proves Fa@ra@b^'s frugality and lack of interest in worldly matters), the state of preservation of Fa@ra@b^'s works that he describes for his time and place has to be verified against extant manuscripts.

Bibliography (for cited works not given in detail, see "Short References"): Sources (texts of all the Arabic sources reporting on the life and activities of Fa@ra@b^ and known to the editor have ben collected in one volume by H®. ¿A. Maháfu@zá, Al-Fa@ra@b^ f^ mara@je¿ al-¿arab^ya, Baghdad, 1975). Abu@'l-Feda@÷, Al-Mokòtasáar f^ ta÷r^kò al-baœar, Istanbul, 1286/1869-70. S®a@¿ed Andalos^, Keta@b tÂabaqa@t al-omam, ed. L. Cheikho, Beirut, 1912. Z@áah^r-al-D^n Bayhaq^, Tatemmat Sáewa@n al-háekma, ed. M. ˆaf^¿, Lahore 1351/1932. Ebn Ab^ Osáaybe¿a, ¿Oyu@n al-anba@÷ f^ táabaqa@t al-atáebba@÷, ed. A. Müller, Cairo, 1299/1882. Ebn al-QeftÂ^, Ta÷r^kò al-háokama@÷, ed. J. Lippert, Leipzig, 1903. Abu@ Nasár Fa@ra@b^, Aháka@m al-noju@m in J. AÚl Ya@s^n, ed., al-Fa@ra@b^: Resalata@n^ falsaf^yata@n^, Beirut, 1407/1987. Idem, Keta@b al-Mella wa nosáu@sá okòra@, ed. M. Mahdi, Beirut, 1968. Idem, Keta@b al-mu@s^q^ al-kab^r, ed. GÚ. Kòaœaba and M. Háefn^, Cairo, 1967. Zakar^ya@÷ b. Moháammad Qazv^n^, AÚt¯a@r al-bela@d wa-akòba@r al-¿eba@d, ed. Beirut, 1380/1960. K¨al^l b. Aybak S®afad^, Al-Wa@f^ be'l-wafaya@t I, ed. H. Ritter, Istanbul and Leipzig, 1931.

Secondary literature. A. Adnan [Adévar], "Fa@ra@b^," ËA V, pp. 451-69. D. Gutas, "The Alexandria to Baghdad Complex of Narratives," Documenti e Studi sulla Tradizione Filosofica Medievale 10 (forthcoming). I. Madkour, La place d'al-Fa@ra@b^ dans l'e‚cole philosophique musulmane, Paris 1934. M. Mahdi, "Al-Fa@ra@b^," in C. C. Gillispie, ed., Dictionary of Scientific Biography IV, New York, 1971, pp. 523-26. Idem, "al-Ta¿a@l^m wa'l-tajreba fi'l-tanj^m wa'l-mu@s^qa@" in al-Fa@ra@b^ wa'l-háazµa@ra al-ensa@n^ya (Mehraja@n al-Fa@ra@b^, 1975), Baghdad, 1975-76, pp. 260-80. Idem, "Al-Fa@ra@b^'s Imperfect State," JAOS 110, 1990, pp. 691-726. M. Steinschneider, al-Farabi (Alpharabius) des arabischen Philosophen Leben und Schriften, St. Petersberg, 1869, pp. 1-11 (the first critical discussion of Fa@ra@b^'s biography; although Steinschneider had only the later sources at his disposal, his analysis is still valuable). S. M. Stern, "Al-Mas¿u@d^ and the Philosopher al-Fa@ra@b^," in S. M. Ahmad and A. Rahman, eds., Al-Mas¿u@d^ Millenary Commemoration Volume, Aligarh, 1960, pp. 28-41. R. Walzer, Al-Farabi on the Perfect State, Oxford, 1985. Idem, "Al-Fa@ra@b^," in EI2 II, pp. 778-81. F. W. Zimmermann, Al-Farabi's Commentary and Short Treatise on Aristotle's "De Interpretatione," London, 1981.

Fa@ra@b^'s works (bibliographies listed in chronological sequence). A. Ate¶, "Fa@ra@b^'nin eserlerinin bibliyografyasi," Belleten 15, 1951, pp. 175-92 (apparently a [revised?] reprint of the author's bibliography which first appeared in Fa@ra@b^ tetkikleri, Istanbul, 1950, pp. 111-26). K. ¿Awwa@d and M. ¿Awwa@d, "Mo÷allafa@t al-Fa@ra@b^," al-Mawred 4/3, 1975, pp. 223-68. M. Cunbur, Ë. Binark, and N. Seferciog¡lu, Fa@ra@b^ Bibliyografyasé, Ankara, 1973. H®. ¿A. Maháfu@zá and J. AÚl Ya@s^n, Mo÷allafa@t al-Fa@ra@b^, Baghdad, 1395/1975. J. Ja@wuœ^, "AÚt¯a@r Abu@ [sic] Nasár al-Fa@ra@b^," in Abu@ Nasár al-Fa@ra@b^: faylasu@f al-Esla@m wa'l-mo¿allem al-t¯a@n^, Damascus, 1409/1989, pp. 117-55. N. Rescher, Al-Fa@ra@b^: An Annotated Bibliography, Pittsburgh, 1962. In addition to these bibliographies which deal with all the works of Fa@ra@b^, inventories of his logical work alone were drawn in two other publications by N. Rescher, al-Fa@ra@b^'s Short Commentary on Aristotle's "Prior Analytics," Pittsburgh 1963, pp. 12-17, and The Development of Arabic Logic, Pittsburgh, 1964, pp. 122-28, no. 26. None of these bibliographical surveys is completely reliable, and each should be used with caution.

(Dimitri Gutas)


Most of Fa@ra@b^'s extant writings deal with logic and the philosophy of language. Many of these writings take the form of commentaries on, or summaries of, the Aristotelian Organon, which, following the tradition of the Alexandrian commentators of late antiquity, included Porphyry's Isagoge as well as Aristotle's Rhetoric and Poetics. Fa@ra@b^ also produced a number of independent treatises in the fields of logic and linguistic philosophy, such as the Keta@b al-háoru@f, Keta@b al-alfa@zá al-mosta¿mala fi'l-mantáeq, and parts of Keta@b al-tanb^h ¿ala@ sab^l al-sa¿a@da and Ehás®a@÷ al-¿olu@m. Throughout these works, Fa@ra@b^'s perspective on the nature of logic remains constant: logic is defined principally as an instrumental science that safeguards the mind from error. It includes both syllogistic and non-syllogistic branches, and it culminates in the study of demonstrative science, which Fa@ra@b^ often identifies as the method of philosophy itself.

The most important general theme in Fa@ra@b^'s writings on logic may be the relationship between logic and grammar. Fa@ra@b^ upholds the concept that logic is a form of universal grammar which provides the human mind with rules that it must follow in order to reason correctly in any language, whereas grammar, according to Fa@ra@b^, provides only those rules that have been established by convention for speakers of some particular natural language such as Arabic or Greek. A passage in his Ehásáa@÷ al-¿olu@m describes this succinctly: "This art [of logic] is analogous to the art of grammar, in that the relation of the art of logic to the intellect and the intelligibles is like the relation of the art of grammar to language and expressions. That is, to every rule for expressions which the science of grammar provides us, there is a corresponding [rule] for intelligibles which the science of logic provides us" (ed. Am^n, p. 68).

In the Keta@b al-h®oru@f, Fa@ra@b^ attempts to flesh out in more detail the exact nature of the relationship between philosophical logic and the ordinary grammar of Arabic. The work opens with an extended classification of Arabic particles and explores how their popular meanings in everyday usage are transformed into technical philosophical terms expressing ideas related to the ten Aristotelian predicaments or categories (ed. Mahdi, pp. 61-130). The second, better known, part of the text discusses the origins of language, the history of philosophy, and the relationship between philosophy and religion. It offers a broad picture of how human language in general originates and how it develops technical terms from popular usage (ibid., pp. 131-61). In the final part of the Keta@b al-h®oru@f, Fa@ra@b^ returns to his consideration of the technical uses of philosophical terminology, this time through a classification of interrogative particles based upon their uses in different types of philosophical inquiry and their relationship to the four causes of Aristotelian philosophy (ibid., pp. 162-226).

Fa@ra@b^'s commentaries on the Organon itself are not simple summaries of the original texts, nor do any but the long commentary (œarhá) on the De interpretatione attempt an exhaustive line-by-line exegesis of Aristotle's work. Instead, Aristotle's writings are used as a focus around which Fa@ra@b^ can offer his own interpretations of Aristotelian logic and the school tradition that had grown about it, sometimes incorporating non-Aristotelian elements such as Stoic logic. In keeping with his interest in the relations between logic and grammar, Fa@ra@b^ is conscious of the need to adapt Aristotelian logic to an Arabic context, a need he openly acknowledges in his discussion of the Prior Analytics, where he promises to "strive to express [Aristotelian syllogistic], as much as possible, by means of words familiar to people who use the Arabic language" (Keta@b al-q^a@s al-sáag@^r, tr. Rescher, p. 49).

In the area of syllogistic theory, Fa@ra@b^ upholds a broadly hierarchical conception of the syllogistic arts, which for him include rhetoric and poetics. He considers the purpose of the syllogism to be fulfilled primarily by the method of demonstration, as articulated by Aristotle in the Posterior Analytics. As he states in Ehásáa@÷ al-¿olu@m, "logic seeks its principal intention only in this part, and the remainder of its parts have been invented only for its sake" (ed. Am^n, pp. 87-89). The non-demonstrative syllogistic arts are relegated, in this broad scheme, to the role of supplementary skills with the primary purpose of preventing philosophers from falling into error by using insecure methods.

Despite this broadly hierarchical view of the syllogistic arts, expressed in works like Ehásáa@÷ al-¿olu@m, Fa@ra@b^ elsewhere shows a greater appreciation for other logical methods used in the pursuit of philosophy. A fine example of this recognition is found in his presentation of the role of dialectic in the opening of his epitome of Aristotle's Topics, the Keta@b al-jadal, where Fa@ra@b^ identifies five important roles for dialectic within philosophy: (1) offering training in argumentative skills, (2) providing the student of philosophy with a first glimpse of the principles of the special demonstrative sciences, (3) alerting the philosopher to the innate self-evident principles of demonstration, (4) helping to develop the ability to communicate philosophical conclusions to non-philosophers, and (5) cultivating the tools by which sophistry can be combated (in ¿Ajam and Fakòr^, III, pp. 29-38).

Fa@ra@b^'s interpretation of the Aristotelian theory of demonstration is most completely articulated in Keta@b al-borha@n, the part of his epitome of the Organon devoted to the Posterior Analytics. A parallel consideration of the epistemological foundations of the theory of demonstration is also provided in an independent treatise on the nature of certitude, the Keta@b œara@÷et al-yaq^n. Both works offer extensive analyses of the logical and epistemic conditions that must be satisfied in order to attain true science or knowledge (¿elm = Greek episte@me@). Fa@ra@b^ begins the Keta@b al-borha@n by distinguishing between the two basic cognitive acts that characterize human thought, namely, conceptualization (tasáawwor) and assent (tasád^q). Conceptualization applies to any act of knowing by which we apprehend simple, discrete concepts. Its ultimate aim, when complete, is to allow us to grasp the essence of the object conceived so as to formulate its proper definition. Assent, by contrast, names an act of knowing which is by its very nature complex and involves a judgment of truth or falsehood. Perfect assent is the act that yields certitude of knowledge, and hence it is the aim of a demonstrative syllogism. Since Aristotle's Posterior Analytics considers both the proper way of discovering definitions and the construction of demonstrative syllogisms, the couplet of perfect conceptualization and perfect assent becomes the organizing theme for Fa@ra@b^'s interpretation of the Aristotelian theory of science (Keta@b al-borha@n, in ¿Ajam and Fakòr^, IV, pp. 19-22, 45).

In developing his views on the perfect act of assent that produces demonstrative certitude, Fa@ra@b^ offers a detailed analysis cast in modal terms. Most significant is Fa@ra@b^'s explicit articulation of the view that absolute certitude must be understood as a form of what contemporary philosophers call second-order knowledge. According to Fa@ra@b^, then, it is not sufficient for certain knowledge that the knower (the person possessing that knowledge) believe that the true statement to which assent is given cannot be otherwise: "And certitude is for us to believe, concerning the truth to which we have assented, that it is not possible at all for what we believe about this matter to be different from what we believe; and that, in addition to this we believe, concerning our belief, that another belief is not possible—to the extent that whenever some belief about the first belief is formed, it is impossible for it to be otherwise, and so on ad infinitum" (Keta@b al-borha@n, in ¿Ajam and Fakòr^, IV, p. 20). Thus, certitude for Fa@ra@b^ requires both that we know some proposition to be true and that we know that we know it. Having defined certitude by these criteria, the way is open to Fa@ra@b^ to develop a distinction between necessary certitude, which applies only to beliefs that can never be otherwise, and non-necessary certitude, which is certitude that holds "only at a particular time." Only necessary certitude requires an object of belief that is itself necessary and unchangeable: "Necessary certitude and necessary existence are convertible in entailment, for what is verified as necessarily certain is necessarily existent" (Keta@b al-borha@n, in ¿Ajam and Fakòr^, IV, p. 22; cf. Keta@b œara@÷et al-yaq^n, in ibid., pp. 97-104). Necessary certitude, then, is the goal of scientific demonstration in its strictest sense; and in keeping with the Aristotelian view of strict demonstration, it can be applied only to a limited range of subjects which are in their very natures eternal and unchanging such as separate, immaterial substances and universals. But non-necessary certitude can apply to a much broader range of subjects and propositions, thereby allowing Fa@ra@b^ to take into account the possibility of attaining some form of certitude about variable and contingent phenomena as well.

The overall spirit and inspiration of Fa@ra@b^'s logical teachings thus remains thoroughly Aristotelian, but the details of his own presentations of basic Aristotelian logical themes introduce nuances and distinctions which are not explicit in Aristotle himself. In particular, Fa@ra@b^'s desire to make Aristotelian logic compatible with expression in the Arabic language forced him to develop Aristotle's rather brief and scattered teachings on the relationship between logic and language in new directions that would address the challenge created by the introduction of Aristotelian logical theory into a very different linguistic and cultural milieu from the one in which it originated.

Bibliography (for cited works not given in detail, see "Short References"): Works by Abu@ Nasár Fa@ra@b^. Ehásáa@÷ al-¿olu@m, ed. ¿O. Am^n, 3rd ed., Cairo, 1968. Keta@b al-alfa@zá al-mosta¿mala fi'l-mantáeq, ed. M. Mahdi as Al-Fa@ra@b^'s Utterances Employed in Logic, Beirut, 1968. Keta@b al-borha@n, ed. in ¿Ajam and Fakòr^, IV, pp. 19-96. Keta@b al-eba@ra, ed. in ¿Ajam and Fakòr^, I, pp. 133-63; tr. in Zimmermann. Keta@b al-háoru@f, ed. M. Mahdi as Alfarabi's Book of Letters, Beirut, 1969. Keta@b al-jadal, ed. in ¿Ajam and Fakòr^, III, pp. 13-107. Keta@b al-q^a@s al-sáag@^r, part. ed. in ¿Ajam and Fakòr^, II, pp. 65-93; tr. N. Rescher as Al-Farabi's Short Commentary on Aristotle's "Prior Analytics," Pittsburgh, 1963. Keta@b œara@÷et al-yaq^n, ed. in ¿Ajam and Fakòr^, IV, pp. 97-104. Keta@b al-tanb^h ¿ala@ sab^l al-sa¿a@da, ed. J. AÚl Yas^n, Beirut, 1985. ˆarhá al-Fa@ra@b^ le-Keta@b Arestáu@táa@l^s fi'l-¿eba@ra, ed. W. Kutsch and S. Marrow, Beirut, 1960; tr. in Zimmermann.

Studies. S. B. Abed, Aristotelian Logic and the Arabic Language in Alfarabi, Albany, N.Y., 1991. R. ¿Ajam and M. Fakòr^, eds., al-Mantáeq ¿end al-Fa@ra@b^, 4 vols., Beirut, 1986-87 (editions of several texts by Fa@ra@b^). R. Arnaldez, "Pense‚e et langage dans la philosophie de Farabi (aà propos du Kitab al-huruf)," Stud. Isl. 45, 1977, pp. 57-65. Th.-A. Druart, "Substance in Arabic Philosophy: al-Farabi's Discussion," Proceedings of the American Catholic Philosophical Association 61, 1987, pp. 88-97. D. M. Dunlop, "Al-Farabi's Introductory Sections on Logic," Islamic Quarterly 2, 1955, pp. 264-82. Idem, "Al-Farabi's Eisagoge," Islamic Quarterly 3, 1956, pp. 117-38. Idem, "Al-Farabi's Introductory Risa@lah on Logic," Islamic Quarterly 3, 1957, pp. 224-35. Idem, "Al-Farabi's Paraphrase of the Categories of Aristotle," Islamic Quarterly 4, 1958, pp. 168-97; 5, 1959, pp. 21-54. A. Elamrani-Jamal, Logique aristote‚licienne et grammaire arabe (e‚tude et documents), Paris, 1983. P. E. Eskenasy, "Al-Fa@ra@b^'s Classification of the Parts of Speech," Jerusalem Studies in Arabic and Islam 11, 1988, pp. 55-82. M. Galston, "Al-Fa@ra@b^ on Aristotle's Theory of Demonstration," in P. Morewedge, ed., Islamic Philosophy and Mysticism, Delmar, N.Y., 1981, pp. 23-34. Idem, "Al-Farabi et la logique aristote‚licienne dans la philosophie islamique," in M. A. Sinaceur, ed., Aristote aujourd'hui, Toulouse, 1988, pp. 192-217. Idem, Politics and Excellence: The Political Philosophy of Alfarabi, Princeton, 1990. H. Gätje, "Die Gliederung der sprachlichen Zeichen nach al-Fa@ra@b^," Der Islam 47, 1971, pp. 1-24. A. Hasnawi, "Fa@ra@b^ et la pratique de l'exe‚geàse philosophique (remarques sur son Commentaire au De interpretatione d'Aristote)," Revue de Syntheàse, 3rd ser. 117, 1985, pp. 27-59. J. Lameer, Al-Fa@ra@b^ and Aristotelian Syllogistics: Greek Theory and Islamic Practice, Leiden, l994. J. Langhade, "Grammaire, logique, e‚tudes linguistiques chez al-Fa@ra@b^," Historiographia Linguistica 8, 1981, pp. 365-77. G. Vajda, "Langage, philosophie, politique et religion, d'apreàs un traite‚ d'al-Farabi," JA 258, 1970, pp. 247-60. F. W. Zimmermann, Al-Farabi's Commentary and Short Treatise on Aristotle's "De Interpretatione," Oxford, 1981 (contains translations of Keta@b al-¿eba@ra and ˆarhá al-Fa@ra@b^ le-keta@b ArestÂu@tÂa@l^s fi'l-¿eba@ra).

(Deborah L. Black)


The question of determining the subject matter of metaphysics has always been a matter of dispute. Aristotle's Metaphysics, with which metaphysics as a discipline originates, exhibits ambiguities and oscillates between two main projects: (1) a study of what is common to all beings, i.e., being as such and other universal notions such as oneness, and (2) a study of the ultimate causes, i.e., God and other immaterial beings. The two projects later became known as general metaphysics and special metaphysics respectively (Frede, pp. 81-95).

This ambiguity led to the well-known different positions taken by Avicenna and Averroes on this issue. Avicenna gives primacy to general metaphysics, whereas Averroes assigns priority to the study of God. Recently, Dimitri Gutas has shown (Gutas, pp. 238-54) that Avicenna's position stems from his reading of a brief treatise by Fa@ra@b^ on the purposes of metaphysics (Ag@ra@zµ ma@ ba¿d al-tÂab^¿a, ed. F. Dieterici in Alfa@ra@b^'s philosophische Abhandlungen, Leiden, 1890, pp. 34-38 [repr. Osnabrück, 1982]; part. tr. by D. Gutas in Gutas, 1988, pp. 240-42; tr. Th.-A. Druart in Druart, 1982, pp. 38-43). This text presents a rather complicated introduction in which Fa@ra@b^ explains the purpose of Aristotle's Metaphysics (tr. Gutas, pp. 240-42), followed by a presentation of the purpose of each book (ed. Dieterici, pp. 34-38; tr. Druart, pp. 40-43). In this introduction Fa@ra@b^ carefully distinguishes metaphysics from kala@m, denying that the purpose of the Metaphysics is to examine God, the Intellect, and the Soul. The real concerns of the Metaphysics are with common notions, i.e., being as such, oneness, their "species and consequent properties, things which are not specific accidents of each individual object studied by the particular sciences (like priority, posteriority, potentiality, actuality, perfection, imperfection, and similar things), and the common first principle of all beings, which [alone] ought to be called by the name of God" (tr. Gutas., p. 241). God, then, far from being the subject matter of the Metaphysics, is only part of it and this only in an indirect manner as the common first principle of all beings. General metaphysics is so central that special metaphysics becomes included in and subordinated to it. "Theology ought to belong" to metaphysics simply "because God is a principle of absolute being, not of one being to the exclusion of another" (ibid.). This constitutes Fa@ra@b^'s understanding of Aristotle's own project and purpose, thus distinguishing it from the enterprise and purpose of kala@m and from the conception of many others who remain unspecified.

Here Fa@ra@b^ follows Aristotle rather strictly, but he is well aware that the Metaphysics leaves many questions unanswered. In working out his own conception of metaphysics, the "second master" (al-mo¿allem al-t¯a@n^) hints that Aristotle made short shrift of special metaphysics and neglected relevant aspects developed by the Neoplatonists. This may explain why in his Falsafat Arastáu@táa@l^s (ed. M. Mahdi, Beirut, 1961; tr. M. Mahdi as "The Philosophy of Aristotle" in Mahdi, 1962, pt. 3) Fa@ra@b^ says little about the Metaphysics. On the one hand, he limits himself to stating that full understanding of natural and human science cannot be achieved "without completing the inquiry into, and investigation of, the beings that are above things natural in their rank of being" and that Aristotle in a book called Metaphysics proceeded "to inquire into, and to investigate, the beings in a manner different than natural inquiry" (tr. Mahdi, p. 130). On the other hand, a few lines further down, he does not hesitate to claim that "we do not possess metaphysical science". This last claim cannot mean that Fa@ra@b^ does not have Aristotle's text, but rather that in his opinion this text remains unsatisfactory; while developing a general metaphysics, its presentation of special metaphysics remains far too narrow, confined as it is to book lambda (la@m). It is a simple assent to a principle of motion rather than of being and does not include a descent grounded in emanationism (Druart, 1987).

In order to understand better Fa@ra@b^'s own approach to metaphysics, we need to examine other texts purporting to offer his own views:

1. Keta@b tah®sá^l al-sa¿a@da (ed. J. AÚl Ya@s^n, Beirut, 1981; tr. M. Mahdi as "The Attainment of Happiness" in Mahdi, 1962, pt. 1). This text is fundamental for understanding Fa@ra@b^'s philosophy. Its presentation of theoretical virtue, i.e., the sciences, explains what Fa@ra@b^ expects from metaphysics and provides it with a program, albeit a rather sketchy one. The text reveals that Fa@ra@b^'s own conception of metaphysics includes much more than the content of the Aristotelian text. As Fa@ra@b^ makes clear, his aim is to search for the principles of being. The physical inquiry into the heavenly bodies shows that they require principles that are neither natures or natural things nor bodies or something in bodies (tr. Mahdi, p. 21). In contrast, metaphysics will provide these principles, since it is "the science of what is beyond natural things in the order of investigation and instruction and above them in the order of being" (ibid., p. 22). In the same way, the inquiry into the rational animal, its soul and intellect, leads to a discernment of those principles for the sake of which such entities are made. These principles are metaphysical beings which possess no matter at all; hence investigating them will bring us to the discovery of a being which "is itself the first principle of all [these] beings," since it is their efficient, formal, and final cause (ibid., p. 24) whereas Aristotle's unmoved mover was only a final cause of motion. Once the ascent to a first principle of being has been achieved, one must then set out on a descent which investigates "what properties the other beings possess as a consequence of their having this being as their principle and the cause (sabab) of their being" (ibid., p. 24). The descent should begin "with the being whose rank is higher than the rest (that is, the one furthest from the first principle). One will thus come to know the ultimate (aqsáa@) causes of the beings. This is the divine inquiry (al-nazáar al-ela@h^) into them. For the first principle is the divinity, and the principles that come after it—and are not bodies or in bodies—are the divine principles" (ibid., p. 24).

The Tahásá^l al-sa¿a@da provides only a partial view of Fa@ra@b^'s overall metaphysics, neglecting the perspective on general metaphysics to focus almost entirely on the ascent to the ultimate principle and the subsequent descent from it. Fa@ra@b^ may have thought that Aristotle had given us the essentials of general metaphysics, while curtailing any full account of complete ascent to an ultimate principle of being and completely neglecting any account of the ensuing descent. Beings are hierarchically organized, and this requires a proper explanation. An emanationist descent would offer such an explanation, and this is the theme explored by Fa@ra@b^ in the first part of both Maba@d^÷ a@ra@÷ ahl al-mad^na al-fa@zµela (Opinions of the people of the virtuous city) and al-S^a@sa al-madan^ya (Political regime).

2. In his own metaphysical works, Keta@b al-wa@háed wa'l-waháda (ed. M. Mahdi as Alfarabi's On One and Unity, Casablanca, 1989) and the Keta@b al-háoru@f (ed. M. Mahdi as Alfarabi's Book of Letters, Beirut, 1990), Fa@ra@b^ does not merely offer refinements on Aristotle's general metaphysics but also develops a full blown special metaphysics, thereby completing Aristotle's project. Yet scholars still encounter significant difficulties in their attempts to comprehend the details of al-Fa@ra@b^'s metaphysical views, since they are scattered in various texts whose chronology, intended audience, and purpose remain obscure. The emanationist texts seem to be among the last of his writings.

The Keta@b al-wa@háed wa'l-waháda is firmly rooted in the tradition of general metaphysics, with emphasis on a common notion, oneness. It illustrates a point Fa@ra@b^ himself makes in his introduction to Ag@ra@zµ ma@ ba¿d al-tÂab^¿a: "The primary object of this science is absolute being and what is equivalent to it in universality [¿omu@m], namely, the one. But since the knowledge of contrary correlatives [motaqa@bela@t] is one, theoretical inquiry into privation and multiplicity is also included in this science" (tr. Gutas, p. 242). So the text begins and ends with some elaboration of the many ways a thing can be said to be one, as in Aristotle's Metaphysics 5.6, 10.1. The central part, on the other hand, completes Aristotle's own contribution in dealing with the many ways in which things are said to be many, "the elaborate account of the various types of opposition between 'one' and 'many,' the 'many' generated from each one of the classes of 'one,' and the various ways in which 'one' and 'many' are related to one another" (Keta@b al-wa@háed, tr. Mahdi, preface). Mahdi remarks that the elaborate discussion of the generation of 'many' from 'one' does not refer to the origination or emanation of many things from the First One, as is the case in the emanationist texts (ibid.). This observation is perfectly true but not surprising, since the origination or derivation in question is of the logical order and not of the ontological order. The Keta@b al-wa@háed wa'l-waháda limits itself to general metaphysics.

The Keta@b al-háoru@f is a difficult and unusual book which gives us further clues about Fa@ra@b^'s conception of metaphysics. The Arabic title is rather ambiguous. It may mean "the book of letters," i.e., a traditional name in Arabic for Aristotle's Metaphysics (since each of its books is designated by a letter of the alphabet). Mahdi adopts this translation and even adds a subtitle, "Commentary on Aristotle's Metaphysics" (Keta@b al-háoru@f, ed. Mahdi, English title page). However, the title may also mean "the book of particles" referring to the parts of speech called particles (háoru@f) in classical Arabic grammar (prepositions, conjunctions, etc.), since several sections of the book focus on interrogative particles. The latter title reflects more accurately the content of the book, which contains far more explicit references to Aristotle's logical works than to the Metaphysics.

The main focus of the Keta@b al-háoru@f seems to be a reflection on the categories (maqu@la@t) and their relation to the various disciplines, including rhetoric and poetry as well as the demonstrative sciences. The categories themselves appear to be derived from questions about the various ways something can be said to be. This explains the importance given to interrogative particles (háoru@f al-so÷a@l) and to the ways of answering them. The categories state in how many ways sensible material things can be said to be. After distinguishing the primary intelligibles (ma¿qu@la@t), which refer to things outside the mind, from secondary intelligibles (ma¿qu@la@t t¯awa@n^) which refer to things inside the mind (such as our cognition of the primary intelligibles) Fa@ra@b^ then applies this distinction to separate the categories as objects of logic from the categories as objects of metaphysics. The logical categories refer to the "utterances qua utterances" of the primary intelligibles, e.g., genera and species. In contrast, the properly philosophical categories deal with the intelligibility of things existing outside the mind. This interpretation explains why Aristotle treats the categories both in logic and in the Metaphysics (5.7).

The relation to the categories underpins the examination of the section on the primary subjects of the arts and sciences (Keta@b al-háoru@f, pp. 66-70). Physics deals with natural things inside the categories; yet in studying their causes it discovers an ultimate efficient cause of being as well as an ultimate final cause, both of which are outside the categories. Metaphysics deals with things that lie outside the categories; it is this science which "examines such things, tries to know them, and examines what is comprised by the categories in as much as those things are their causes, including what mathematics and politics comprise as well as the practical arts encompassed by politics" (ibid, p. 69). This brief description of metaphysics reveals a lively interest in "metaphysical beings" and in special metaphysics generally. Then from such beings considered as causes it derives what concerns the categories, mathematics, and even politics broadly constructed and so provides principles for the sciences, be they theoretical or practical. This derivation is linked to the descent accomplished in the more formal and detailed first part of the emanationist texts.

While the beginning of Part I, the Keta@b al-háoru@f highlights issues in special metaphysics and the descent from the ultimate principles, subsequent sections develop aspects of general metaphysics by studying universal notions that cut across the categories, such as "being" (al-mawju@d; ibid., pp. 110-28) "itself" (al-dòa@t; ibid., pp. 106-10), and "thing" (al-œay÷; ibid., pp. 128-29). Interrogative particles, such as "wherefore" (alladò^ men ¿ajlehe; pp. 129-30) and "wherefrom" (¿an; p. 130), are also considered, leading to the affirmation of the existence of beings outside the categories. The section on being, for instance, explains why the Arabic term mawju@d was chosen to reflect the Greek "on," lists its various philosophical meanings (as in Metaphysics 5.7), and states the possibility of the existence of a being outside the categories. On the other hand, the chapter on substance (jawhar; Keta@b al-háoru@f, pp. 97-105; Druart, 1987, pp. 88-97) inquires whether one can properly speak of a substance existing outside the categories, all the while recognizing that such a substance would be more substance than any other. This last question remains unresolved, although in the emanationist texts Fa@ra@b^ applies the term substance to the first cause. Part II leaves aside general notions in order to focus on the origins of words, of philosophy, and of religion, across various cultures and languages (tr. Berman, pp. 171-78). Part III returns to an examination of interrogative particles. The whole text exhibits striking attempts at adapting general metaphysics to the Arabic language, and at developing such metaphysics in an original fashion.

3. The emanationist texts: Maba@de÷ @a@ra@÷ ahl al-mad^na al-fa@záela (ed. and tr. R. Walzer as Al-Fa@ra@b^ on the Perfect State: Abu@ Nasár al-Fa@ra@b^'s Maba@di÷ a@ra@÷ ahl al-mad^na al-fa@dáila, Oxford, 1985) and the Keta@b al-s^a@sa al-madan^ya (ed. F. Najja@r as Keta@b al s^a@sa al-madan^ya al-molaqqab be-Maba@de÷ al-mawju@da@t, Beirut, 1964). In their first half, both texts follow the program laid down for metaphysics in the Keta@b al-h®oru@f. They study the metaphysical beings which are beyond the categories and are neither bodies nor in bodies (the First Cause and the Ten Intelligences) as well as the principles of natural beings, thereby providing principles for physics. The latter principles are not bodies but are in bodies (nafs, "soul"; sáu@ra, "form"; and ma@dda, "matter"). They also offer a metaphysical descent from such principles to all beings. The second half of each book presents a political science derived from the study of these principles. The focus here will be on the Keta@b al-s^a@sa al-madan^ya, which is metaphysically more sophisticated.

The Keta@b al-s^a@sa al-madan^ya begins abruptly by stating that there are six principles for all beings. These are listed in descending order: the First Cause, the Secondary Cause (the first nine intelligences), the Agent Intellect or Tenth Intelligence (al-¿aql al-fa¿¿a@l), soul, form, and matter. The secondary principles emanate from the First Cause either directly or indirectly—directly in the case of the First Intelligence, indirectly for all the others. Each of the secondary causes emanates directly from the one above it or from the First Cause, and through intelligizing the First Cause, originates another Intelligence and through intelligizing itself originates an ensouled celestial sphere. The number of intelligences and celestial spheres are in accordance with Ptolemy's astronomical system. As for the last Intelligence, or Agent Intellect, it only originates the sphere of the moon and the intelligibles for human intellects, thereby actualizing this lower rung of intellect.

The celestial spheres are a kind of intermediary between the metaphysical beings, which exist beyond the categories, and the realm of nature, which is within the categories. Although these spheres have neither form nor matter in the proper sense and, therefore, are not subject to the hylomorphic composition, they nevertheless have both a substrate and a form-like soul. Consequently, "only the most excellent categories belong to them" (al-S^a@sa al-madan^ya, p. 54). From the motions common to all celestial spheres emanates matter, and from the motions proper to each and their combination emanate the various forms. The hylomorphic or possible sublunary beings emanate from forms (including souls) and matter.

This constitutes special metaphysics and includes a descent that derives the various principles from one another as well as from the various beings. The First Cause is the ultimate efficient and final cause of all. Yet the perspective of general metaphysics is not entirely neglected. The examination of the First Cause or God reveals that the universal notions, being and oneness, which cut across the categories, belong primarily and perfectly to the First Cause, since all other beings receive from the First Cause their own derivative and limited being and oneness. The focus is no longer on the ways things can be said to be, but on the ways they are and why they are.

Unfortunately, we do not possess a complete presentation of Fa@ra@b^'s metaphysics in which the relations between general metaphysics, special metaphysics, and the descent would be carefully explored and developed.

Bibliography (works by Abu@ Nasár Fa@ra@b^ are given in the article): R. Arnaldez, "L'âme et le monde dans le systeàme philosophique de Fa@ra@b^," Stud. Isl. 43, 1976, pp. 53-63. L. V. Berman, "Maimonides, the disciple of Alfarabi," Israel Oriental Studies 4, 1974, pp. 154-78 (contains a partial translation of Keta@b al-háoru@f). C. D'Ancona Costa, La casa della sapienza, Milan, 1996, pp. 122-32. Th.-A. Druart, "Le traite‚ d'al-Fa@ra@b^ sur les buts de la Me‚taphysique d'Aristote," Bulletin de Philosophie me‚die‚vale 24, 1982, pp. 38-43. Idem, "Al-Fa@ra@b^ and Emanationism," in J. Wippel. ed., Studies in Medieval Philosophy, Washington, D.C., 1987, pp. 23-43. Idem, "Substance in Arabic Philosophy: Al-Fa@ra@b^'s Discussion," in J. Wippel, ed., The Metaphysics of Substance, Washington, D.C., 1988, pp. 88-97. Idem, "Al-Fa@ra@b^, Ethics, and First Intelligibles," Documenti e Studi sulla Tradizione Filosofica Medievale 8, 1997, pp. 403-423. M. Frede, "The Unity of General and Special Metaphysics: Aristotle's Conception of Metaphysics," in his Essays in Ancient Philosophy, Minneapolis, Minn., 1987, pp. 81-95. D. Gutas, Avicenna and the Aristotelian Tradition, Leiden, l988. M. Mahdi, Alfarabi's Philosophy of Plato and Aristotle, Ithaca, N.Y., 1962. S. Pines, "Les limits de la me‚taphysique selon al-Fa@ra@b^, Ibn Ba@jja et Maïmonide," in J. P. Beckmann et al., eds., Sprache und Erkenntnis im Mittelalter I, Berlin and New York, 1981, pp. 211-25. A. Rachid, "Dieu et l'être selon Al-Fa@ra@b^: le chapitre de l'être' dans le Livre des Lettres," in Dieu et l'être, Paris, 1978, pp. 179-90. R. Ramo‚n Guerrero, "Al-Fa@ra@b^ y la Metafísica de Aristo‚teles," Ciudad de Dios 196, 1983, pp. 211-40.

(The‚reàse-Anne Druart)


Fa@ra@b^'s philosophical moorings and direct affiliation lie in the Greek neo-Aristotelian school of Ammonius in Alexandria, in the form in which it survived and was revived after the Islamic conquest among Syriac Christian clerics and intellectuals in the centers of Eastern Christianity in the Fertile Crescent. This school, traditionally but inappropriately called Neoplatonic, was essentially Aristotelian in its basic orientation, structure, and contents. It had been transformed, however, in a number of ways along Neoplatonist lines, most notably in its acceptance of Plotinian emanationism (for various aspects of this transformation see Sorabji, passim). By Fa@ra@b^'s own account (cf i. biography), the Nestorians Yu@háanna@ b. H®ayla@n, his immediate teacher, and Matta@ b. Yu@nus, his older contemporary and colleague in Baghdad, were direct descendants in this tradition. To Matta@ apparently belongs the credit for reviving Aristotelian studies in Baghdad and establishing both a curriculum of school texts and a method for their study (Endress, 1989, pp. 844-45). Fa@ra@b^'s claims about his philosophical pedigree are independently substantiated by the demonstrably numerous points of actual doctrinal congruity with the Alexandrian tradition (Mahdi, 1967, pp. 233-37; Gutas, 1983, p. 255); the scholastic tradition of both philological and philosophical commentary on the core texts of Aristotle survives in Fa@ra@b^'s commentaries and in the Paris MS. (Arab. 2346) of Aristotle's Organon (Zimmermann, 1972 and 1981; Hugonnard-Roche, 1990 and 1993).

Equally as important as Fa@ra@b^'s educational background for an assessment of his relation to Greek philosophy is the intellectual climate during his time in Baghdad. When Fa@ra@b^ was studying in the ¿Abbasid capital at the end of the 9th century, the translation and study of the Greek sciences were almost as old as the city itself (founded in 145/762). The translation movement, which was generated, funded, and promoted by almost all the upper strata of early ¿Abbasid society, had provided Arabic versions of the majority of Greek philosophical and scientific works, and had reached such a state of maturity where the translations of certain fundamental works (e.g., Aristotle's Physics, Euclid's Elements, or Ptolemy's Almagest) were repeatedly done anew or revised (Gutas 1998, ch. 6). The intellectuals who sponsored these translations championed the authors, methods, and ideas of their liking, and in some instances had created schools of thought. In philosophy, particularly notable was the circle of philosophers and scholars around al-Kend^ (d. 870s) and his successors, as well as the sui generis work of the autodidact Abu@ Bakr Ra@z^ (d. 925 or 935), Fa@ra@b^'s older contemporary. Fa@ra@b^ thus experienced in Baghdad a fertile environment of pluralist philosophical thinking at least a century old and at an advanced stage of sophistication.

The prevailing atmosphere of intellectual fermentation and experimentation was unchecked by political authority, particularly after the abortive attempt by al-Ma÷mu@n and his two successors to impose a centralized ideology (the mehána, 833-49). As a result, the center stage for ideological supremacy was contested by many groups in competition with each other, for the different trends of thought had not yet coalesced into established doctrines with sufficient social backing to impose themselves upon the rest. The period from 850 to 950, covering Fa@ra@b^'s entire life, saw attempts to bring order to the intellectual disarray, consolidate and systematize the various disciplines and establish methodological and doctrinal uniformity, e.g., Ebn Moja@hed's (d. 936) reforms of Kokanic readings (qera@÷a@t), Ebn Ab^ H®a@tem Ra@z^'s (d. 938) formulation of the principles of Hadith criticism (al-jarhá wa'l-ta¿d^l), etc. (Gutas 1998, pp. 101-104). The revival of Aristotelian studies represented by the translations and commentaries of Matta@ and brought to fruition by the entire oeuvre of Fa@ra@b^ must be seen in this light as performing a similar function in the field of philosophy.

Fa@ra@b^ sought to present philosophy as a coherent system, promulgate neo-Aristotelianism as the one true philosophical doctrine, and rationalize its practice and show its validity for contemporary ¿Abbasid society. It seems also that he wished to emancipate philosophy from subordination to medicine, a position it occupied in early ¿Abbasid times for historical reasons, and establish it as an independent, and possibly even leading, intellectual discipline (Zimmermann, 1976, pp. 407-8, 412-14). To begin with, his conception of true philosophy was that it "was handed down to us by the Greeks from Plato and Aristotle only . . . . In what they presented, their purpose is the same, and . . . they intended to offer one and the same philosophy" (The Attainment of Happiness, in Mahdi, 1969, pp. 49-50). However, his understanding of the identity of the philosophies of Plato and Aristotle is not doctrinaire, as in Porphyry, but nuanced in favor of Aristotle, as espoused in the school of Ammonius (Endress, 1991, pp. 242-43). Fa@ra@b^'s Concordance of the Opinions of Plato and Aristotle (Jam¿ ra÷yay al-háak^mayn) presents Plato as a respected precursor of Aristotle whose main achievement was to provide moral exhortation for the purposes of social education. He presents Aristotle as the philosopher who began where Plato left off, perfected philosophy, and put it on a demonstrative basis (Endress, 1991, pp. 249-51).

Given this understanding of true philosophy, Fa@ra@b^'s teachings and writings fall into different categories (cf. Druart). First, at an elementary level and apparently for a wide audience, he edited and adapted certain existing translations of Greek texts; (he did not know Greek and had no access to Greek works other than those translated into Arabic.) He worked by slightly rewording the argument, adding explanatory details, and clarifying syntactical and semantic obscurities of the Arabic of the translation (Gutas, 1983, p. 252 and note 51, p. 256 and note 61). Such editorial reworkings by Fa@ra@b^ include his recasting of (1) the late Alexandrian introductions to the study of Aristotle in his Prolegomena to the Study of Aristotle's Philosophy (Ma@ yanbag@^ an yoqaddama qabla ta¿allom falsafat ArestÂu@; Gutas, 1985, pp. 115-16); (2) similar prolegomena by Elias(-David) in the logic part of his Enumeration of the Sciences (Ehásáa@÷ al-¿olu@m; Gutas, 1983, pp. 231-38, 255 ), and (3) Galen's Synopsis of Plato's Laws in his Precise Exposition of the Synopsis of Plato's Laws (Talkò^sá jawa@me¿ nawa@m^s Afla@tÂu@n; Gutas l997).

Second, Fa@ra@b^ wrote school commentaries on the works of Aristotle, as well as paraphrases and other popularizations (Gutas, 1993, pp. 47-50). Third, and most importantly, he created a philosophical system of his own on the basis of principles and orientations which he inherited from neo-Aristotelianism and by using as his material the entire array of Greek philosophical thought that was available to him in translation. These crucial principles and orientations were: (1) a procedural tool of analysis by division, (2) the "tree" of Porphyry, (3) the classification of the sciences, and (4) the theory of language. The neo-Aristotelian school of Ammonius had developed these procedures and positions for the purposes of philosophical pedagogy and in the process of exegesis of Aristotle's works, in particular the initial books of the Organon: Porphyry's Eisagoge and Aristotle's Categories and De Interpretatione. Fa@ra@b^ fully followed these practices, but he used them in a way which resulted in a philosophical system that went far beyond the scholastic interests of his Greek and Syriac predecessors.

1. Analysis by division (diairesis, qesma). Analysis by division was introduced by Plato into the discussions of scientific method (Phaedrus 265, Politicus 263, Sophist 221-231) and firmly grounded as a procedure in definitions by Aristotle (Topics and Categories, Metaphysics Z and Posterior Analytics B13, 96b25-97b6, and Parts of Animals i). For an analysis, particularly relevant for Fa@ra@b^'s understanding of the theory and practice of scientific method in Aristotle, the articles by Balme, Lennox, Bolton, and Gotthelf (Gotthelf and Lennox, pp. 65-198) should be consulted. Division became the standard method of analysis through its detailed exposition by Porphyry in his commentary on Plato's Sophist, and from him it passed into the works of all subsequent philosophers writing in Greek and Syriac (Hein, pp. 131-45). From the works in particular of David, Elias, Ps.-Elias, and their reproduction by Paul the Persian, analysis by division appears in the earliest extant Arabic work on logic, Ebn Behr^z's H®odu@d al-mantÂeq, and constitutes one of the fundamental analytical tools of Fa@ra@b^.

2. Hierarchical classification. Closely related to, and ultimately derived from the above, is Porphyry's elevation of the division of genera and species into a hierarchical classification of being that ranges from the highest genera (the ten categories), which are not species to any superior genus, to the lowest species, that are not genera to any lower species. Every level between these extremes constitutes a genus to its inferior and a species to its superior levels (Eisagoge, ed. Busse, pp. 4ff.; MantÂeq ArestÂu@, ed. Badaw^, pp. 1027ff.). Fa@ra@b^'s systematization of all reality, both ontological and social, into a hierarchical whole in which each level is governed by the one above it, and governs the one beneath it, follows precisely such a classification.

3. Classification of the sciences. In the two centuries before the advent of Islam, Alexandrian scholars in the school of Ammonius erected, by means of a rigorous application of analysis by division, an elaborate schema of classifying Aristotle's works, in which each individual treatise came to denote a single field of study. The result of this approach was that the classification of Aristotle's works became, in effect, a classification of all the sciences, and hence of all human knowledge. The function of this classification was initially descriptive and pedagogical, used in the introduction to the philosophical curriculum to elucidate the definitions and divisions of philosophy or all knowledge. Eventually, however, it also acquired two further interrelated values: a normative value, in that the precise classification of Aristotle's works and hence of all knowledge was assumed to reflect ontological reality, the way things are; and a historical value, in that the classification of the sciences also purported to reflect historical reality or the chronological sequence of the development of the sciences in human history. Both the classificatory schema and all of its functions were adopted by Fa@ra@b^, mainly through the Arabic translation of Paul the Persian's treatise on the subject, and made into the cornerstone of his philosophical system (Gutas, 1983, pp. 256-60).

4. Theory of language. Alexandrian neo-Aristotelianism engaged in an intensive cultivation of the preliminaries to the study of Aristotle's Organon. Both Porphyry's Eisagoge and other related introductory material formed the focus of much philosophical study. The issues treated were predominantly related to the philosophy of language and meaning, if only because Aristotle's Categories and De Interpretatione deal with these subjects. This heightened preoccupation with semantics, syntax, and semiotics, and in particular, with concepts such as homonymy, synonymy, and paronymy at the beginning of the Categories, and name, verb, and sentence at the beginning of De Interpretatione, put linguistic analysis at the center of philosophical practice. Fa@ra@b^ exhibits a similar preoccupation both because of his philosophical training in the neo-Aristotelian tradition which cultivated these studies and because of the central position of linguistic studies in contemporary Baghdadi intellectual life (cf. Abed). Two of his works are entirely devoted to the subject: al-Alfa@zµ al-mosta¿mala f^ l-mantÂeq (Vocables Employed in Logic) and Keta@b al-h®oru@f (Book of Particles); and even his essay on the intellect (Resa@la f^'l-¿aql) is concerned with differentiating the various meanings of the homonymous term "intellect" (¿aql).

On the basis of these principles, and with a wide variety of Greek philosophical and scientific texts in Arabic translation at his disposal, Fa@ra@b^ created an original and compelling philosophical system. A pre‚cis of that system is offered in his Maba@de÷ a@ra@÷ ahl al-mad^na al-fa@zµela (The Principles of the Opinions of the People of the Excellent City). At the heart of the system lies the theory of the intellect, or noetics, which animates and lends coherence to Fa@ra@b^'s entire philosophy. This is the natural extension of late Greek neo-Aristotelianism, which combined on the one hand a long tradition of commenting on and extrapolating from the few and cryptic statements by Aristotle on the nature of the intellect (both the unmoved mover and that of humans), and on the other an ontological enhancement of the status of the intellect that was developed in particular in the Neoplatonic school of Athens (Walzer, 1957, pp. 229-30, 201-2; Walzer, 1974; Finnegan; Jolivet).

Following standard neo-Aristotelian doctrine, Fa@ra@b^ considered the "noblest" part of logic to be apodeictic demonstration, the primary function of the intellect. Accordingly, Aristotle's Posterior Analysis forms the centerpiece and culmination of the entire Organon, while the four preceding books (Porphyry's Eisagoge, Aristotle's Categories, De Interpretione, and Prior Analytics) are said to introduce demonstration and the final four to "protect" it, by disclosing the ways in which apodeictic certainty can be derailed by arguments that are dialectical (Topics), sophistic (Sophistici Elenchi), rhetorical (Rhetoric), or poetic (Poetics). Fa@ra@b^ accepts this fivefold division of arguments or propositions (i.e., demonstrative, dialectical, sophistical, rhetorical, and poetical) not only on the level of description or analysis, but also grants it ontological status by claiming that the human mind can think only in these five ways (Gutas, 1983, pp. 256-57, 265-66). Thus in the final analysis even logic, originally a methodological discipline, is made subservient to, and dependent on, ontological noetics.

In cosmology Fa@ra@b^ accepted Plotinian emanationism, which he combined with Ptolemy's planetary system from the Hypotheses (Walzer, 1985, pp. 363-65) to create the rigidly hierarchical "tree of Porphyry" system of the supralunar cosmos. Noetics is at the center of this system insofar as the emanation of the spheres, with their intellects and souls, depends on the creative act of intellection of the superior sphere. The same hierarchical structure of constituent elements is reproduced in the world of nature on earth (humans, animals, plants, minerals), within the human body (limbs and organs), within the human soul and its ranked faculties, and within the rational part of the soul (the intellect itself). Contact between the supralunar and the sublunar realms is effected through the emanation of the intelligibles from the active intellect (the intellect of the tenth sphere, that of the moon) to the human intellect, which is then actualized, and on occasion to the imaginative or representative faculty, bringing about prophecy and divination.

In his philosophy of the individual and society, noetics is again central to Fa@ra@b^'s hierarchical system. He sees human happiness in the life of the intellect, or in the actualization of the human intellect, which is to be understood as the life of philosophical reflection. As he states in the closing paragraph of his exposition on Aristotle's philosophy, "Investigation and theoretical inquiry into the intelligibles which are of no benefit for the soundness of the body and the soundness of the senses are necessary for the perfection of man," which he goes on to define as the actualization of man's intellect. Of the various sciences studied in philosophy, Fa@ra@b^ continues, metaphysics is "necessary for the development, as a city-dweller, of his intellect, for the sake of which man is made, while all other sciences are investigated . . . in order for the human intellect, for the sake of which man is made, to be made perfect" (translated according to the sounder text of Fa@ra@b^'s Falsafat ArestÂu@ quoted in ¿Abd-al-LatÂ^f Bag@da@d^'s Keta@b al-nasá^háatayn, MS. Bursa, Hüseyin Çelebi 823, f. 87r; cf. Mahdi, 1969, p. 130). The human intellect, for the sake of which man is made, is specified in this passage as belonging to a "city-dweller" (madan^) because man needs the social organization of a city in order to meet his physical needs and to attain perfection (Maba@de÷ in Walzer, 1985, p. 229). In the Maba@de÷ Fa@ra@b^ describes the man who has reached perfection as he who "has become actually intellect and actually being thought" (Walzer, 1985, p. 241).

This is in accordance with Aristotle's description of human happiness as the contemplative life (Nicomachean Ethics X) or of the divine as intellect engaged in perpetual self-intellection (Metaphysics XII) but it is particularly close (especially with its emphasis on the lack of practical benefit in philosophical reflection) to Aristotle's theory of leisure as the end of both individual and social activity (Politics VII.15, 1334a11-b28). Within that context, it is also close to his statement that intellect is the end of nature: "In men, reason (logos) and mind (or intellect: nous) are the end towards which nature strives, so that the birth and training in custom of the citizens ought to be ordered with a view to them" (Politics 1334b15-17, in Jowett's translation in Barnes, p. 2117). Consequently, human communities, which Fa@ra@b^ classifies hierarchically in accordance with the principles of the "tree of Porphyry" listed above (Maro‚th 1978; in greater detail, Maro‚th 1994, pp. 215-22; cf. Pines, pp. 156-60; Rosenthal), attain or fall short of perfection to the extent to which their rulers and inhabitants are able to actualize their intellects, receive the intelligibles, and hold the correct opinions as required by the philosophical sciences and revelation (Maba@de÷, chaps. 15-19 in Walzer, 1985, pp. 229-329, and esp. pp. 277-85).

Fa@ra@b^'s discussion of human communities is subordinate to his noetics and intended to complete it by working out, in the interests of a classificatory comprehensiveness also exhibited in other parts of the Maba@de÷, all the details of a highly hierarchical system; there is no question here of an independent political philosophy. The always perspicacious Ebn K¨aldu@n (q.v.), as the conscious originator of political philosophy in Islam, recognized that Fa@ra@b^'s discussion was based on noetics; he says that by s^a@sa madan^ya (the title of one of Fa@ra@b^'s works) "the philosophers mean the disposition of soul and character which each member of a social organization must have, if, eventually, people are completely to" have no need of "rulers." They call the social organization that fulfills these requirements the 'ideal city' (al-mad^na al fa@zµela, part of the title of Fa@ra@b^'s Maba@de÷). The norms observed in this connection are called 'political utopias'"(s^a@sa madan^ya). Accordingly Ebn K¨aldu@n dismissed the theory as irrelevant to real life and as something "rare and remote" (Moqaddema II, tr. F. Rosenthal, p. 138).

The absence of an independent political philosophy in Fa@ra@b^'s work is to be expected. In neither Athenian Neoplatonism, nor Alexandrian neo-Aristotelianism, the twin sources of Fa@ra@b^'s philosophy, was Plato's and Aristotle's political philosophy an integral part (Walzer, 1985, pp. 425-26), while the actual political ideas in the Byzantine world from Justinian to Fa@ra@b^'s time (Dvornik, chaps. 11, 12) are completely alien to Fa@ra@b^'s analysis. Besides, the focus on noetics is also reflected in the title of the Maba@de÷: it is not, as it is often elliptically but misleadingly referred to, the "excellent city," but "the principles of the opinions of the people of the excellent city." Subsequent philosophical tradition in Islam well understood the noetic basis of the work; philosophers who adopted Fa@ra@b^'s system dispensed with his classification of human communities as an inconsequential appendage. It was only in the tradition of the akòla@q (q.v.) books written in Persian and Turkish, which followed the Aristotelian tripartition of practical philosophy into ethics, oeconomics, and politics, that Fa@ra@b^'s classification of human communities found a place to fill in the section on politics. This tradition was inaugurated by Nasá^r-al-D^n T®u@s^'s Akòla@q-e na@sáer^ (q.v.), which was followed in both Persian and Turkish by similar works of Dava@n^ (q.v.), Wa@¿ezá Ka@œef^, Qena@l^za@da ¿Al^, etc. This apart, Fa@ra@b^'s classification of communities had no influence either in Islam or the West (Gutas, 1990, pp. 357, 354).

The various discrete Greek sources of Fa@ra@b^'s philosophical system and of the Maba@de÷ in particular have been identified and analyzed by Walzer (1971 and 1985), but this work as a whole and especially its conception must be regarded as original to Fa@ra@b^ himself (cf. Mahdi, 1990). Within Islam, Fa@ra@b^'s system was taken up by Avicenna, who further developed and refined it to create a philosophy that was to remain dominant in the East. In the West, it influenced both Arabic Andalusian (Jewish and Muslim) and Latin European philosophy. Fa@ra@b^'s achievement is that he was the first philosopher who succeeded to internationalize Greek philosophy by creating in a language other than Greek a complex and sophisticated system far surpassing the elementary efforts of both the early medieval Latins and his Syriac predecessors. As such, he stands at the head of all subsequent philosophers who made Greek philosophy Western philosophy.

Bibliography (Apart from references cited in the article, the bibliography also lists works which deal with Greek antecedents of Fa@ra@b^'s philosophy. For cited works not given in detail, see "Short References"): S. B. Abed, Aristotelian Logic and the Arabic Language in Alfa@ra@b^, Albany, N.Y., 1991. ¿A. Badaw^, ed., MantÂeq ArestÂu@ III, Cairo, 1952. J. Barnes, The Complete Works of Aristotle, Princeton, 1984. C. S. F. Burnett, ed., Glosses and Commentaries on Aristotelian Logical Texts, London, 1993. Th.-A. Druart, "Al-Farabi and Emanationism," in J. F. Wippel, ed., Studies in Medieval Philosophy, Washington, D.C., 1987, pp. 23-43. F. Dvornik, Early Christian and Byzantine Political Philosophy, Washington, D.C., 1966. Ebn K¨aldu@n, The Muqaddimah, tr. F. Rosenthal, Princeton, 1967. G. Endress, "Matta@ b. Yu@nus," in EI2 VI, pp. 844-46. Idem, "«La Concordance entre Platon et Aristote», l'Aristote arabe et l'e‚mancipation de la philosophie en Islam me‚die‚val," in B. Mojsisch and O. Pluta, eds., Historia Philosophiae Medii Aevi, Amsterdam and Philadelphia, 1991, pp. 237-57. J. Finnegan, "Al-Fa@ra@b^ et le Peri Nou d'Alexandre d'Aphrodise," in Me‚langes Louis Massignon II, Damascus, 1957, pp. 133-52.

A. Gotthelf and J. G. Lennox, eds., Philosophical Issues in Aristotle's Biology, pt. II on 'Definition and Demonstration: Theory and Practice,' Cambridge, 1987, pp. 65-198. D. Gutas, "Paul the Persian on the Classification of the Parts of Aristotle's Philosophy: A Milestone between Alexandria and Bagµda@d," Der Islam 60, 1983, pp. 231-67. Idem, "The Starting Point of Philosophical Studies in Alexandrian and Arabic Aristotelianism," in W. W. Fortenbaugh, ed., Theophrastus of Eresus: On His Life and Work, New Brunswick, N. J. and Oxford, 1985, pp. 115-23. Idem, "Ethische Schriften im Islam," in W. Heinrichs, ed., Neues Handbuch der Literaturwissenschaft V: Orientalisches Mittelalter, Wiesbaden, 1990, pp. 346-65. Idem, "Aspects of Literary Form and Genre in Arabic Logical Works," in C. S. F. Burnett, ed., Glosses and Commentaries on Aristotelian Logical Texts, London, 1993, pp. 29-76. Idem, "Galen's Synopsis of Plato's Laws and Fa@ra@b^'s Talª^sá," in R. Kruk and G. Endress, eds., The Ancient Tradition in Christian and Islamic Hellenism, Leiden, l997, pp. 101-19. Idem, Greek Thought, Arabic Culture, London, l998. C. Hein, Definition und Einteilung der Philosophie, Frankfurt, Bern, and New York, 1985.

H. Hugonnard-Roche, "Les traductions du grec au syriaque et du syriaque aà l'arabe," in Rencontres de cultures dans la philosophie me‚die‚vale: Traductions et traducteurs de l'antiquite‚ tardive au XIVe sieàcle, Louvain-la-Neuve and Cassino, 1990, pp. 131-47. Idem, "Remarques sur la tradition arabe de l'Organon d'apreàs le manuscrit Paris, Bibliotheàque nationale, Ar. 2346," in Burnett, 1993, pp. 19-28. J. Jolivet, "L'intellect selon al-Fa@ra@b^: quelques remarques," Bulletin d'Études Orientales 29, 1977, pp. 251-59. M. Mahdi, "Alfarabi against Philoponus," JNES 26, 1967, pp. 233-60. Idem, Alfarabi's Philosophy of Plato and Aristotle, Ithaca, N. Y., 1969. Idem, "Al-Fa@ra@b^'s Imperfect State," JAOS 110, 1990, pp. 691-726. M. Maro‚th, "Griechische Theorie und orientalische Praxis in der Staatskunst von al-Fa@ra@b^," Acta Antiqua 26, 1978, pp. 465-69. Idem, Die Araber und die antike Wissenschaftstheorie, Budapest and Leiden, l994. S. Pines, "Aristotle's Politics in Arabic Philosophy," Israel Oriental Studies 5, 1975, pp. 150-60. E. I. J. Rosenthal, Political Thought in Medieval Islam, Cambridge, 1968, pp. 122-42. R. Sorabji, ed., Aristotle Transformed, London, 1990.

R. Walzer, "New Studies on al-Kind^," Oriens 10, 1957, pp. 203-32, repr. in idem, Greek into Arabic, pp. 175-205. Idem, "Al-Fa@ra@b^'s Theory of Prophecy and Divination," Journal of Hellenic Studies 77, 1957, pp. 142-48 (repr. in idem, Greek into Arabic, pp. 206-19). Idem, Greek into Arabic, Oxford, 1962. Idem, "Al-Fa@ra@b^," in EI2 II, pp. 778-81. Idem, "L'e‚veil de la philosophie musulmane," REI 38, 1970, pp. 7-42, 207-42. Idem, "Early Islamic Philosophy," in A. H. Armstrong, ed., The Cambridge History of Later Greek and Early Medieval Philosophy, Cambridge, 1970, pp. 652-66. Idem, "Lost Neoplatonic Thought in the Arabic Tradition," in Le Ne‚oplatonisme, Paris, 1971, pp. 319-28. Idem, "Aristotle's Active Intellect, nous poie@tikos, in Greek and Early Islamic Philosophy," in Plotino e il neoplatonismo in Oriente e in Occidente, Rome, 1974, pp. 423-36. Idem, ed. and tr., Al-Farabi on the Perfect State: Abu@ Nasár al-Fa@ra@b^'s Maba@di÷ a@ra@÷ ahl al-mad^na al-fa@dáila, Oxford, 1985. F. W. Zimmermann, "Some Observations on al-Farabi and Logical Tradition," in S. M. Stern, A. Hourani, V. Brown, eds., Islamic Philosophy and the Classical Tradition [Festschrift Richard Walzer], Oxford, 1972, pp. 517-46. Idem, "Al-Farabi und die philosophische Kritik an Galen von Alexander zu Averroes," Akten des VII Kongresses für Arabistik und Islamwissenschaft, Göttingen, 1976, pp. 401-14. Idem, Al-Farabi's Commentary and Short Treatise on Aristotle's "De Interpretatione", London, 1981.

(Dimitri Gutas)


In the history of Middle Eastern music Fa@ra@b^ remains unequalled as a theorist, but this aspect of his manifold achievements has been obscured by his more widely known writings on philosophy. Scholars of medieval European music are seldom aware of Fa@ra@b^'s importance for music theory or the significance of his commentaries on the works of the ancient Greek music theorists. At the same time, his contributions to musical theory are often entangled in legendary accounts which bestow a supernatural dimension on his powers as a musician, relating that he could make his audience laugh, cry, and sleep against their will (Bayhaq^, pp. 30-35; Ebn K¨alleka@n, tr. de Slane, III, p. 309).

Of the approximately one hundred and sixty works attributed to Fa@r@a@b^, eight are on music, but only four have survived (Sawa, 1983-84, p. 3). The first and least significant is the very short chapter on music in Keta@b ehásáa@÷ al-¿olu@m. Paradoxically, it was this sketchy work alone that was available in medieval Europe through several Latin translations (Farmer, 1934). Fa@ra@b^'s masterly and comprehensive Keta@b al-mu@s^q^ al-kab^r, as well as his Keta@b al-^qa@¿a@t and Keta@b ehásáa@÷ al-^qa@¿a@t (on rhythms) remained unknown.

Fa@ra@b^ wrote Keta@b al-mu@s^q^ al-kab^r for Abu@ Ja¿far Muháammad b. al-Qa@sem Karkòò^, the vizier of the caliph al-Ra@zµ^ (d. 329/940), who wished to learn about the science of music according to the ancient Greek theorists. Fa@ra@b^ agreed to the request because he had found serious shortcomings in the Greek works available to him in Arabic translation (through Syriac). He blamed the shortcomings on the poor quality of the works chosen for translation or that of the translations themselves. He also found shortcomings in the Arabic writings of his predecessors (al-Mu@s^q^, ed. K¨aœaba, pp. 35-37; Sawa, 1989, p. 14), including the philosopher Kend^ (d. after 256/870) and the singer, lutenist, composer and theorist Esháa@q Mawsáel^ (q.v.; d. 235/850). As a philosopher, rather than a practicing musician, Kend^ uncritically followed Greek musical theories (in Arabic translation) that were unrelated to Middle Eastern practice. On the other hand, Esháa@q Mawsel^ lacked the necessary philosophical training to apply a rigorous, logical approach to his writings on the science of music. In contrast, Fa@ra@b^'s training as a logician and a practicing musician meant that his theory reflected practice in a clear discourse. He did indeed expound Greek music theory but he also made his readers aware of those aspects not applicable to music in the Middle East. In addition to the Greek theories he also described the musical practices of his own time and provenance, i.e. the early ¿Abbasid era from Iraq, Persia, and Transoxiana, as well as the reported practices of the Omayyads and early Islamic era in Mecca, Medina, and Damascus (Sawa, 1989, pp. 14-17). As the vizier was not knowledgeable in the art and theory of music, Fa@ra@b^ explained music by borrowing terms, concepts and paradigms from the Greek sciences and contemporary disciplines such as arithmetics, Euclidean geometry, Aristotelian logic, architecture and textile, civil and mechanical engineering, politics, Arabic grammar, phonology, prosody, poetics, rhetoric, and koranic sciences (Madian; Sawa, 1981, p. 80; idem, 1990).

The Keta@b al-mu@s^q^ al-kab^r was supposedly composed of two books. The second, which was intended as a commentary on the works of previous writers, is lost and perhaps was never finished. The first book is in two parts. Part one is an introduction (madkòal) to the art of music, consisting of two discourses, one dealing with the philosophy of music and the other with elementary acoustics. Part two deals with the craft (sáena@¿a) of music, arranged according to three arts (fann). The first art comprises basic theoretical elements such as acoustics, music intervals, and melodic and rhythmic modes. Fa@ra@b^ reports that the Greeks as well as the early writers in the Middle East confined their investigation to this art alone (al-Mu@s^q^, ed. K¨aœaba, pp. 38-39; Sawa, 1989, p. 15). In contrast, the discussion of the second skill is, in Fa@ra@b^'s words, his own innovation (al-Mu@s^q^, ed. K¨aœaba, pp. 38-39; Sawa, 1989, p. 15). It deals at length with the description of common musical instruments and how the tone systems discussed in theory could be obtained on them. These instruments include the ¿u@d (lute), the Baghdadi as well as the Khorasani tÂonbu@r (pandore), the mezma@r (flute or reed pipe), the sorna@y (oboe), the raba@ba (rebec), the me¿zafa (lyre) and the sáanj (harp). The discussion of the third art deals with musical composition, consonances and dissonances, melodic movements and rhythmic modes used in practice, and specific details regarding vocal and instrumental performance practice, the relation between language and music, classification of voice types, and the purpose of music. Thanks to Fa@ra@b^'s lucid description of musical practices, we can see how much has survived to our own time (Sawa, 1981). The treatise exists in a number of manuscripts (Farmer, 1965, pp. 27-28; Sawa, 1989, pp. 18-20; Shiloah, 1979, pp. 104-07), as well as in an Arabic edition (K¨aœaba) and a French translation (D'Erlanger). However, more work is needed to elucidate the ambiguous and obscure passages in the book. Part of the problem is inherent in the subject itself: music being a non-verbal and non-visual art, Fa@ra@b^ had to resort to many disciplines to express his thoughts. The modern reader must therefore be familiar with these disciplines in their 10th century context and must be equally aware that at times Fa@ra@b^ freely borrows technical terms from a discipline and applies them to music, distorting their original meaning in the process.

One of the difficult areas in the Keta@b al-mu@s@^q^ al-kab^r is the treatment of rhythm. The outstanding German scholar Eckhard Neubauer rightly considers the two chapters on rhythm as an impenetrable thicket and thinks that Fa@ra@b^ himself realized, or perhaps was persuaded by friends, that a revision was needed (Neubauer, 1968-69, pp. 196-97). This was done in two subsequent treatises, the Keta@b al-^qa@¿a@t and Keta@b ehásáa@' al-^qa@¿a@t (Sawa, 1989, pp. 20, 36-37). The latter was discovered only recently (in 1951) by Ahmet Ate¶ in Manisa, Turkey (Sawa, 1983-84, p. 4; 1989, p. 20). In these two treatises Fa@ra@b^ perfected his rhythmic theory and rhythmic notation system. He developed general formulae, which he named the basics, and codified sixteen contemporary ornamental techniques which altered and beautified rhythms and allowed for an infinite number of rhythmic variations. Of immense value are his notated examples of rhythms in their basic and ornamented forms, as well as meticulous captions supplied under the notated examples explaining the arrangement of the basic rhythmic attacks and ornamental additions. With this kind of accuracy the problem of understanding, deciphering, and transcribing the medieval rhythmic modes was at last solved (Sawa, 1983-84; 1989, pp. 35-71). Additionally, the Keta@b ehásáa@'al-^qa@¿a@t is of great importance for the early history of rhythms: in a section of the treatise Fa@ra@b^ quotes and comments on the writings of Esháa@q Mawsáel^ and Kend^ (fol. 79b-81b, 88a-89b; Sawa, 1989, pp. 235-36), thus preserving for us unique documents from the 3rd/9th century.

As mentioned above Fa@ra@b^ twice revised his theory of rhythm and rhythmic modes after he wrote the Keta@b al-mu@s^q^ al-kab^r. Unfortunately no such revision was made in the case of his work on the melodic modes, which to this day have resisted decipherment. For this reason, the focus here will be on his rhythmic theory. He defined rhythm as "the motion through the notes within durations well defined as to their length and proportions" (al-Mu@s^q^, ed. K¨aœaba, pp. 435-36). He explained that musical sounds are produced by the action of an attack (naqra) as in the striking of the membrane of a drum, the plucking of a string, the impact of the air pushed out of a wind instrument or a singer's throat. The attack is further defined as being timeless, that is, it carries no time in itself, and it occurs in the present (a@n) which separates the past from the future. He imagined the attack as the striking of a solid body by a very thin body, the thinner the better, so that the contact is imagined as a point. This led Fa@ra@b^ to borrow Euclid's postulate in geometry: as a point has no length, as the line has no surface, as the surface has no volume, the attack has no duration; furthermore as a point is separated from another by a line, as a line is separated from another by a surface, as a surface is separated from another by a volume, an attack is separated from another by a duration. Then he defines the durations starting with the shortest perceivable time as the standard of measurement, a concept he borrowed from Greek music (chronos protos). He refers to this shortest time by the symbol ta, and double the shortest time by the symbol tan, and double that by the symbol

tann. By naming and defining these durations and their proportions he created a very precise tool to describe as well as define the rhythmic modes which were current in his time and before his time. Side by side with measured music built on rhythmic modes, which is comparable to poetry, there existed unmeasured types of music comparable to speech. In the latter music the proportions of the durations are not integer numbers. It is significant to note that these genres of music still exist to this day in the Middle East: the unmeasured is comparable to the Persian a@va@z (q.v.); the Arabic mawwa@l, laya@l^ and taqs^m; and the Turkish gazel and taksim; whereas the measured is comparable to the Persian tasán^f, the Arabic mowaœœah and the Turkish pe¶rev and semaî.

Fa@ra@b^'s work is of interest to musicologists doing research on Ancient Greece or medieval Europe, but for the history of music in the Middle East it is absolutely crucial. His terminology, concepts, and methodological approach have had a lasting impact on later music theorists, be they Arab, Persian, or Turkish. Furthermore, because of strong continuity in the musical traditions of the Middle East from the medieval to the modern era, Fa@ra@b^'s writings continue to offer useful models for music analysis in the region today.

Bibliography (for cited works not given in detail, see "Short References"): Works by Fa@ra@b^. Keta@b ehásáa@÷ al-^qa@¿a@t, in MS Manisa, Turkey, Genel Kütüphanesi, no. 1705, fol. 59-90a. Keta@b al-^qa@¿a@t, in MS Istanbul, Topkapé Sarayé Kütüphanesi, Ahmet III, no. 1878, fol. 160b-167a. Keta@b al-mu@s^q^ al-kab^r, MSS Istanbul, Süleymaniye Kütüphanesi, Ragép Pa¶a, no. 876; Köprülü, no. 953; MS Leiden, Universitätsbibliothek, Or. 651; MS Madrid, Bibliotheca Nacional, Res. 241; MS Milan, Biblioteca Ambrosiana, no. 289; MS Princeton, N.J., University Library, Garrett 1984; ed. GÚatÂtÂa@s ¿Abd-al-Malek K¨aœaba (with revisions and introduction by M. A. Hefn^), Cairo, 1967; tr. R. d'Erlanger as Grand traite‚ de la musique: Kita@bu' l-Mu@s^q^ al-Kab^r, La Musique arabe, 2 vols., Paris, 1930-35.

Other sources and studies. Abu'l-H®asan ¿Al^ b. Zayd Bayhaq^, Tatemmat S®ewa@n al-háekma, ed. M. Kord ¿Ali as Ta÷r^kò háokama@÷ al-Esla@m, Damascus 1946; repr. Damascus, 1976. M.-T. Da@neœpa‘u@h, Moda@wamat dar osáu@l-e mu@s^q^-e Èra@n, Tehran, 2535=1355 ˆ./1976, pp. 54-62 (with further bibliographical information). H. G. Farmer, ed., Al-Fa@ra@b^'s Arabic-Latin Writings on Music, Glasgow, 1934. Idem, The Sources of Arabian Music, Leiden, 1965. ¿A. ¿A.-H®. Madian, "Language-Music Relationships in Al-Fa@ra@b^'s Grand Book of Music," Ph.D. diss., Cornell University, 1992. E. Neubauer, "Die Theorie von Èqa@¿ I: Übersetzung des Kita@b al-Èqa@¿a@t von Abu@ Nasár al-Fa@ra@b^," Oriens 21-22, 1968-69, pp. 196-232. Idem, "Die Theorie von Èqa@¿ II: Übersetzung des Kita@b Ihásáa@' al-Èqa@¿a@t von Abu@ Nasár al-Fa@ra@b^," Oriens 34, 1994, pp. 103-73. G. D. Sawa, "Al-Fa@ra@b^'s Theory of the Èqa@¿: An Empirically Derived Medieval Model of Rhythmic Analysis," Progress Reports in Ethnomusicology, 11/9, 1983-84, pp. 1-32. Idem, Music Performance Practice in the Early ¿Abba@sid Era 132 A.H./750 A.D.-320 A.H./932 A.D., Toronto, 1989. Idem, "Paradigms in al-Fa@ra@b^'s Musical Writings," in N. van Deusen and A. E. Ford, eds., Paradigms in Medieval Thought: Applications in Medieval Disciplines, Lewiston, N. Y., 1990, pp. 81-92. Idem, "The Survival of Some Aspects of Medieval Arabic Performance Practice," Ethnomusicology, 25/1, 1981, pp. 73-86. A. Shiloah, The Theory of Music in Arabic Writings (c. 900-1900): Descriptive Catalogue of Manuscripts in Libraries of Europe and the U.S.A., Munich, 1979.

(George Sawa)


The central theme of Fa@ra@b^'s political writings is the virtuous regime, the political order whose guiding principle is the realization of human excellence by virtue. Fa@ra@b^ conceives of human or political science as the inquiry into man insofar as he is distinguished from other natural beings and from divine beings, seeking to understand his specific nature, what constitutes his perfection, and the way through which he can attain it. Unlike other animals, man is not rendered perfect merely through the natural principles present in him, and unlike divine beings he is not eternally perfect but needs to achieve his perfection through the activity proceeding from rational understanding, deliberation, and choosing among the various alternatives suggested to him by reason. The initial presence of the power of rational knowledge and of the choice connected with it, is man's first or natural perfection, the perfection he is born with and does not choose. Beyond this, reason and choice are present in a human being to use for realizing his end or the ultimate perfection possible for his nature. This ultimate perfection is identical with the supreme happiness available to him. "Happiness is the good desired for itself, it is never desired to achieve by it something else, and there is nothing greater beyond it that a human being can achieve" (Maba@de÷, ed. Dieterici, p. 46: cf. Keta@b al-s^a@sa, pp. 72-75, 78).

Yet, happiness cannot be achieved without first being known, and without performing certain orderly (bodily and intellectual) activities useful or leading to the achievement of perfection. These are the noble activities. The distinction between noble and base activities is thus guided by the distinction between what is useful for, and what obstructs, perfection and happiness. To perform an activity well, with ease, and in an orderly fashion requires the formation of character and the development of habits that make such activities possible. "The forms and states of character from which these [noble] activities emanate are the virtues; they are not goods for their own sake but goods only for the sake of happiness" (Maba@de÷, ed. Dieterici, p. 46; cf. Keta@b al-s^a@sa, pp. 43-44). The distinction between virtue and vice presupposes knowledge of what human perfection or happiness is as well as the distinction between noble and base activities.

The virtuous regime can be defined as the regime in which human beings come together and cooperate with the aim of becoming virtuous, performing noble activities, and attaining happiness. It is distinguished by the presence in it of knowledge of man's ultimate perfection; the distinction between the noble and the base and between the virtues and the vices; and the concerted efforts of the rules and the citizens to each and learn these things, and to develop the virtuous forms or states of character from which emerge the noble activities useful for achieving happiness.

The attainment of happiness means the perfection of that power of the human soul that is specific to man, of his reason. This in turn requires disciplining the lower desires to cooperate with and aid reason to perform its proper activity and also acquiring the highest arts and sciences. Such discipline and learning can be accomplished only by the rare few who possess the best natural endowments and who are also fortunate to live under conditions in which the requisite virtues can be developed and noble activities performed. The rest of men can only attain some degree of this perfection; and the extent to which they can attain that degree of perfection of which they are capable is decisively influenced by the kind of political regime in which they live and the education they receive. Nevertheless, all the citizens of the virtuous regime must have some common notions about the world, man, and political life. But they will differ with regard to the character of this knowledge, and hence, with regard to their share of perfection or happiness. They can be divided broadly into the following three classes: (1) the wise or the philosophers who know the nature of things by means of demonstrative proofs and by their own insights; (2) the followers of these who know the nature of things by means of the demonstrations presented by the philosophers, and who trust the insight and accept the judgment of the philosophers; (3) the rest of the citizens, the many, who know things by means of similitudes, some more and others less adequate, depending on their rank as citizens. These classes or ranks must be ordered by the ruler who should also organize the education of the citizens, assign to them their specialized duties, give them their laws, and command them in war. He is to seek, by persuasion and compulsion, to develop in each the virtues of which he is capable and to order the citizens hierarchically so that each class can attain the perfection of which it is capable and yet serve the class above it. It is in this manner that the city becomes a whole similar to the cosmos, and its members cooperate toward attaining happiness.

The virtuous regime is a non-hereditary monarchical or an aristocratic regime in which the best rule, with the rest of the citizens divided into groups that (depending on their rank) are ruled and in turn rule—until one arrives at the lowest group that is ruled only. The sole criterion for the rank of a citizen is the character of the virtue of which he is capable and that he is able to develop through his participation in the regime and obedience to its laws. Like the regime itself, its citizens are virtuous, first, because they possess, or follow those who possess, correct similitudes of the knowledge of divine and natural beings, human perfection or happiness, and the principles of the regime designed to help human beings attain this happiness; and, second, because they act in accordance with this knowledge in that their character is formed with a view to performing the activities conducive to happiness.

Once the main features of the virtuous regime are clarified, the understanding of the main features and the classification of all other regimes become relatively simple. Fa@ra@b^ divides them into three broad types: (1) The regimes whose citizens have had no occasion to acquire any knowledge at all about divine and natural beings or about perfection and happiness. These are the "ignorant" regimes. Their citizens pursue lower ends, good or bad, in complete oblivion of true happiness. (2) The regimes whose citizens possess the knowledge of these things but do not act according to their requirements. These are the "wicked" or "immoral" regimes. Their citizens have the same views as those of the virtuous regime; yet, their desires do not serve the rational part in them but turn them away to pursue the lower ends pursued in ignorant regimes. (3) The regimes whose citizens have acquired certain opinions about these things, but false or corrupt opinions, that is, opinions that claim to be about divine and natural beings and about true happiness, while in fact they are not. The similitudes presented to such citizens are, consequently, false and corrupt, and so also are the activities prescribed for them. These are the regimes that have been led astray or the erring regimes. The citizens of such regimes do not possess true knowledge or correct similitudes, and they, too, pursue the lower ends of the ignorant regimes. The regimes in error may have been founded as such. This is the case with the regimes "whose supreme ruler was one who was under an illusion that he was receiving revelations without having done so, and with regard to which he had employed misrepresentations, deceptions, and delusions" (Maba@de÷, ed. Dieterici, p. 63; cf. Keta@b al-s^a@sa, pp. 103-4). But they may also have been originally virtuous regimes that had been changed through the introduction of false or corrupt views and practices.

All these types of regimes are opposed to the virtuous regimes because they lack its guiding principle, which is true knowledge and virtue or the formation of character leading to activities conducive to true happiness. Instead, the character of the citizens is formed with a view to attaining one or more of the lower ends. These ends are given by Fa@ra@b^ as six, and each of the general types mentioned above can be subdivided according to the end that dominates in it: (1) the regime of necessity (or the indispensable regime) in which the aim of the citizens is confined to the bare necessities of life; (2) the vile regime (oligarchy) in which the ultimate aim of the citizens is wealth and prosperity for their own sakes; (3) the base regime is the purpose of whose citizens is the enjoyment of the sensory or imaginary pleasures; (4) the regime of honor (timocracy) whose citizens aim at being honored, praised, and glorified by others; (5) the regime of domination (tyranny) whose citizens aim at overpowering and subjecting others; (6) the regime of corporate association (democracy) the main purpose of whose citizens is being free to do what they wish.


To combine divine and political science is to emphasize the political importance of sound beliefs about divine beings and the principles of the world. Both Islam and classical philosophy are in agreement concerning this issue. Muslims believed that the primary justification of their existence as a distinct community was the revelation of the truth about divine things to Moháammad, and that, had he not come to them with his message, they would have continued to live in misery and uncertainty about their well-being in this world and the next. It was also because of such considerations that Plato thought that kings must become philosophers or philosopher-kings. Once the quest for the best regime arrives at the necessity of combining divine and political science, it becomes necessary that the ruler should combine the craft of ruling with that of prophecy or philosophy. The ruler-prophet or the ruler-philosopher is the human being who offers the solution to the question of the realization of the best regime, and the functions of the ruler-prophet and of the ruler-philosopher appear in this respect to be identical.

Fa@ra@b^ begins his discussion of the supreme ruler with the emphasis on the common function of the ruler-philosopher and the ruler-prophet as rulers who are the link between the divine beings above and the citizens who do not have direct access to knowledge of these beings. He is the teacher and guide "who makes known" to the citizens what happiness is, who "arouses in them the determination" to do things necessary for attaining it, and "who does not need to be ruled by a human in anything at all" (Keta@b al-s^a@sa, pp. 78-79). He must possess knowledge; not need any other human to guide him; have excellent comprehension of everything that must be done; be excellent in guiding all others in what he knows; have the ability to make others perform the functions for which they are fit; and have the ability to determine and define the work to be done by others and to direct such work toward happiness. These qualities evidently require the best endowments, but also the fullest development of the rational faculty. (According to Aristotelian psychology as Fa@ra@b^ presents it in his political works the perfection of the rational faculty consists of its correspondence to, or "contact" with, the Active Intellect; see ¿AQL.) The supreme ruler must be a human being who actualizes his rational faculty or who is in contact with the Active Intellect.

This supreme ruler is the source of all power and knowledge in the regime, and it is through him that the citizens learn what they ought to know and to do. As God or the First Cause of the world directs everything else, and as everything else is directed toward Him, "the case ought to be the same in the virtuous city; in an orderly fashion, all of its parts ought to follow in their activities in the footsteps of the purpose of its supreme ruler (Maba@de÷, ed. Dieterici, pp. 56-57; cf. Keta@b al-s^a@sa, pp. 83-84). He possesses unlimited powers and cannot be subjected to any human being or political regime or laws. He has the power to confirm or abrogate previous divine laws, to enact new ones, and "to change a law he had legislated at one time for another if he deems it better to do so" (Keta@b al-s^a@sa, pp. 80-81). He alone has the power to order the classes of people in the regime and assign in them their ranks. And it is he who offers them what they need to know.

For most people, this knowledge has to take the form of an imaginative representation of the truth rather than a rational conception of it. This is because most people are not endowed, or cannot be trained to know divine things in themselves, but can only understand their imitations, which should be made to fit their power of understanding and their special conditions and experience as members of a particular regime. Religion contains such a set of imaginative representations. The divine law is legislated for a particular group of human beings. It is necessitated by the incapacity of most human beings to conceive things, especially the highest or divine things, rationally. Still, they need to know these things in some fashion. They need to believe in the imitations of divine beings, and of happiness and perfection, as presented to them by the founder of their regime. The founder must then not only present a rational or conceptual account of happiness and the divine principles to the few, but also adequately represent or imitate these same things for the many. All the citizens are to accept that with which he entrusts them: "the ones who follow after happiness as they cognize it and accept the [divine] principles as they cognize them are the wise human beings; and the ones in whose souls these things are found in the form of images, and who accept them and follow them as such, are the believers" (Keta@b al-s^a@sa, p. 86).

Thus far, Fa@ra@b^ identifies the ruler-prophet and the ruler-philosopher. They are both supreme rulers absolutely, and both have absolute authority with regard to legislating beliefs and actions. Both acquire this authority in virtue of the perfection of their rational faculty, and both receive revelation from God through the agency of the active intellect. Wherein, then, does the ruler-prophet differ from the ruler-philosopher?

The first and primary qualification that the ruler of the virtuous regime must possess is a special kind of knowledge of divine and human beings. Now, a human being possess three faculties for knowledge: sensation, imagination, and reason (both theoretical and practical), and these develop in him in that order. Imagination has three functions: (1) It acts as a reservoir of sensible impressions after the disappearance of the objects of sensation. (2) It combines sensible impressions to form a complex sensible image. (3) It produces imitations. It has the capacity to imitate all sensible things (human desires, temperament, passions) through sensible impressions or certain combinations of them. When later the rational faculty develops, and a human being begins to grasp the character, essence, or form of natural and divine beings, the faculty of imagination receives and imitates these rational forms also, that is, it represents them in the form of sensible impressions. In this respect, imagination is subordinate to the rational faculty and depends on it for the "originals" that it imitates; it has no direct access to the essence of natural and divine beings. Further, the imitations that it fabricates are not all good copies; some may be more true and nearer to the originals, others defective in some respects, and still others extremely false or misleading copies. Finally, only the rational faculty that grasps the originals themselves can judge the degree of the truth of these copies and of their likenesses to the originals. The rational faculty is the only faculty that has access to the knowledge of divine or spiritual beings, and it must exercise strict control to insure that the copies offered by the imaginative faculty are good or fair imitations. It may happen in rare cases that this imaginative faculty is so powerful and perfect that it overwhelms all the other faculties, and proceeds directly to receive or form images of divine beings. This rare case is the case of prophecy (Maba@de÷, ed. Dieterici, p. 52).

The description of the nature of prophecy leads to the distinction between the faculty of imagination and the rational faculty. It explains the possibility of prophecy as the perfection of the faculty of imagination, and that imagination can almost dispense with the rational faculty and receive the images of divine beings directly and without the latter's mediation. There are two powers by means of which a human being can communicate with the Active Intellect: his imagination and his rational faculty or his intellect. When he communicates with it by means of his imagination, he is "a prophet who warns about what will happen and who informs about what is taking place now"; while when he communicates with it by means of his rational faculty he is "a wise human being, a philosopher, and has complete intelligence" (Maba@de÷, ed. Dieterici, pp. 58-59).


Wisdom or philosophy is an indispensable condition for the founding and survival of the virtuous city. Prophecy, on the other hand, is indispensable for founding a virtuous city, but not for its survival. In enumerating the qualities of the supreme ruler or the founder of the virtuous city, Fa@ra@b^ stipulates the coincidence of excellent rational and prophetic faculties. This requirement is imposed by composition of the virtuous city as a political community, that is, the fact that it must be made up of two broad groups: (1) the few who are philosophers or can be addressed through philosophy, and who can be taught the theoretical sciences and hence the true character of divine and natural beings as they are; (2) the many who (because they lack the necessary natural endowments or have no time for sufficient training) are not philosophers, who live by opinion and persuasion, and for whom the ruler must imitate these beings by means of similitudes or symbols.)

While the few can be made to grasp rationally the meaning of human happiness and perfection and the rational basis for justification of the virtuous activities that lead to a human being's ultimate end, the many are incapable of such understanding and have to be taught to perform these activities by persuasion and compulsion, that is, by explanations that could be understood by all the citizens regardless of their rational capacity, and by prescribed rewards and punishments of an immediate tangible kind. The supreme ruler teaches the few in his capacity as a philosopher, and he presents similitudes and prescribes rewards and punishments for the many in his capacity as prophet. To be believed and practiced by the many, these similitudes and prescriptions should be formulated by the prophet, and accepted by the citizens, as true, fixed, and permanent; that is, the citizens should expect definite rewards and punishments for belief and unbelief, and for obedience and disobedience. The prophetic faculty culminates, then, in laying down laws concerning both the beliefs and the practices of the many, and the prophet who assumes this function becomes a prophet-legislator. The rational faculty, on contrast, culminates in teaching the theoretical sciences to the few. In his summary of Plato's Laws, Fa@ra@b^ also understood Plato to say that these virtuous few "have no need for fixed practices and laws at all; nevertheless they are very happy. Laws and fixed practices are needed only for those who are morally crooked" (Talkò^sá, p. 41).

It is only as viewed by the subjects that laws are fixed and are of unquestionable divine authority. The supreme rule of the virtuous regime is the master and not the servant of the law. Not only is he not ruled by any human being, he is also not ruled by the law. He is the cause of the law, he creates it, and he abrogates and changes it as he sees fit. He possesses this authority because of his wisdom and his capacity to decide what is best for the common good under given conditions, and conditions can arise under which the changing of the laws is not only salutary but indispensable for the survival of the virtuous regime. In so doing he must be extremely cautious not to disturb the faith of the citizens in their laws, and should consider the adverse effect that change has on attachment to the law. He must make a careful appraisal of the advantage of changing the law as against the disadvantage of change as such. Thus he must possess, not only the authority to change the laws whenever necessary, but also the craft of minimizing the danger of change to the well-being of the regime. But once he sees that changing the law is necessary and takes the proper precautions, there is no question as to his authority to change the law. Therefore, so long as he lives, the rational faculty rules supreme and laws are preserved or changed in the light of his judgment as philosopher.

It is this coincidence of philosophy and prophecy in the person of the ruler, or at least the coincidence of philosophy and rulership, that insures the survival of the virtuous regime. As long as rulers who possess such qualities succeed one another without interruptions, the same situation obtains (Keta@b al-mella, pp. 49-50).

The coincidence of philosophy and prophecy is extremely rare, and chance may not even favor the virtuous regime with the availability of a human being who possesses all the necessary natural endowments and whose training as philosopher proves successful. Thus the question arises as to whether the virtuous city can survive in the absence of a human being with all the qualifications required of the prophet-philosopher-ruler or of the philosopher-ruler. Granting that the best possible arrangement demands the existence of such qualifications in one person who must rule, can the regime originated by the prophet-philosopher-ruler survive at all in his absence and in the absence of a philosopher-ruler as his successor? Fa@ra@b^ is willing in the Maba@de÷ to consider the possibility that this city can survive in the absence of both such rulers, but only if provisions are made for the presence of proper substitutes for prophetic legislation. These substitutes consist of (1) the body of laws and customs established by the "true princes," and (2) a combination of new qualities in the ruler that make him proficient in the "art of jurisprudence," that is, knowledge of the laws and customs of his predecessors, willingness on his part to follow these laws and customs rather than change them, the capacity to apply them to new conditions by the deductions of new decisions from, or the discovery of new applications for, established laws and customs, and the capacity to meet every new situation (for which no specific decisions are available) through understanding the intention of previous legislators rather than by the legislation of new laws or by any formal change of old ones. So far as the law is concerned, this new ruler is a jurist-legislator rather than a prophet-legislator. He must, however, possess all other qualities, including wisdom, that enable him to discern and promote the common good of his regime at the particular period during which he rules.

In the event that no single human being should exist who possesses all these qualifications, then Fa@ra@b^ suggests a third possibility: a wise man and one other human being (who possesses the rest of the qualities, except wisdom) should rule jointly. Were even this to prove unobtainable, he suggests finally a joint rule of a number of human beings possessing these qualifications severally. This joint rule does not, however, affect the presence of the required qualifications but only their presence in the same human being. Thus, the only qualification whose very presence may be dispensed with is prophecy. The substitutes for prophecy are the preservation of old laws and the capacity to discover new applications for old laws. To promote the common good and preserve the regime under new conditions as these emerge, neither the coincidence of philosophy and prophecy in the same human being, nor the coincidence of wisdom and jurisprudence, proves to be an indispensable condition. It is sufficient to have wisdom in the person of a philosopher who rules jointly with another human being or a group of human beings who possess, among other things, the capacity to put old laws to new uses. Unlike prophecy, wisdom cannot be dispense with, and nothing can take its place. Unlike the presence of prophecy, the absence of wisdom is fatal to the existence of the virtuous regime. There is no substitute for living wisdom.

Bibliography: Works by Fa@ra@b^. Keta@b al-mella, ed. M. Mahdi as Alfarabi's Book of Religion and Related Texts (Keta@b al-mella wa nosáu@sá okòra@), Beirut, 1968. Keta@b al-s^a@sa al-madan^ya, ed. F. Najjar as Al-Farabi's The Political Regime (al-Siya@sa al-Madaniya Also Known as the Treatise on the Principles of Beings), Beirut, 1964; part. tr. by F. Najjar in R. Lerner and M. Mahdi, Medieval Political Philosophy: A Sourcebook, New York, 1963. Maba@de÷ a@ra@÷ ahl al-mad^na al-fa@zµela, ed. F. Dieterici, Leiden, 1895; ed. and tr. R. Walzer as Al-Farabi on the Perfect State, Oxford, 1985. Talkò^sá nawa@m^s Afla@tÂon, ed. F. Gabrieli as Alfarabius Compendium Legum Platonis, London, 1952.

For further discussion and extensive bibliography, see M. Galston, Politics and Excellence: The Political Philosophy of Alfarabi, Princeton, 1990.

(Muhsin Mahdi)