Corps de l’article

When we turn from late Antiquity to Islamic and Jewish philosophy in the early Middle Ages, we can no longer focus exclusively on commentaries as a technique for doing philosophy. Of the three philosophers to be examined in this paper, Averroes wrote almost only commentaries on Aristotle, whereas Maimonides did not write a single work that falls into a traditional philosophical genre. And yet, they share in many respects the same philosophical-religious outlook. The way these philosophers approach philosophy and exegesis is determined by a twofold task : to take up and continue the project of ancient philosophy and to define its place in a society in which the authority of the divine Law is undisputed. In this paper I will attempt to clarify how the roles of philosophy and exegesis were shaped by the strategies employed to carry out this twofold task.

Averroes (d. 1198) and Maimonides (d. 1204) are the last two important representatives of the falsafa tradition in Muslim Spain, an intellectual tradition that originated in the Muslim East with al-Fârâbî (d. 950).[1] Al-Fârâbî is, of course, not the first Arabic philosopher, but I take it to be uncontroversial that the identity of falsafa as a self-conscious philosophical movement was largely shaped by him. Maimonides who is the most important Jewish representative of this school has been called “the disciple of Alfarabi” by Lawrence Berman, because, according to Berman, he took al-Fârâbî’s general theory of the relationship between philosophy, religion, theology, and jurisprudence and applied it to Judaism.[2] The same can, I think, be said about most of the falâsifa. It certainly can be said for Averroes’ interpretation of Islam.[3] A comparison between Averroes and Maimonides appears to be promising for several reasons. Both were born in Córdoba and received their scientific education in the context of Andalusian Aristotelianism. Both were not only philosophers, but also prominent doctors and experts in Islamic respectively Jewish law.[4] Both dealt with the same philosophical problems of which the most important were al-Ghazâlî’s criticism of the falâsifa, the incompatibility of Ptolemaic astronomy and Aristotelian celestial physics and the questions arising from the sceptical epistemology that al-Fârâbî is reported to have set forth in his lost Commentary on the Nicomachean Ethics.[5] Finally, Maimonides read Averroes’ commentaries on Aristotle and recommended them to his students.[6] If, then, it is true that Averroes and Maimonides were disciples of al-Fârâbî, and reshaped his legacy under a comparable set of circumstances, one would expect a comparison between them to yield many similarities. But a look at the corpus of their writings shows that the opposite is the case.[7] The great majority of Averroes’ works are commentaries on Aristotle. They come in different formats and serve different purposes. Sometimes they take on the form of a systematic treatise on a philosophical topic, e.g. philosophical psychology or celestial physics. But all of them reflect what is undoubtedly Averroes’ central intellectual concern : understanding Aristotle. Averroes also wrote a number of medical treatises and treatises on Islamic law. But in relation to his work on Aristotle these are clearly marginal. Finally, he also wrote three works that pertain to what can be described as philosophical theology. In these works, however, he states expressly that he would not have written them at all if al-Ghazâlî had not questioned the legitimacy of falsafa from the standpoint of Islam.[8] For Averroes, then, philosophy and exegesis come together first and foremost in his impressive series of commentaries on Aristotle.

Maimonides, by contrast, did not compose a single work that strictly speaking can be described as philosophical. The authenticity of the one that comes closest is disputed : his Maqâla fî ṣinâ‛at al-manṭiq [Treatise on the Art of Logic] that he is said to have written in his youth.[9] But even if Maimonides did, in fact, compose it, it falls into the genre of “introductions to philosophy” that goes back to Porphyry’s Eisagôgê. It leads to the threshold of philosophy, as it were, without being philosophy itself.[10] Maimonides is, on the other hand, a prominent commentator as well. Only that his commentaries do not explicate Aristotle, but the Law of Moses. His first important work is the Sirâj [Light], his Arabic commentary on the Mishna. His last important work — usually referred to as his chief philosophical work — is the Dalâlat al-ḥâ’irîn [The Guide of the Perplexed], which is at least presented as a book of Biblical exegesis. According to the introduction, Maimonides’ goal is to explain to perplexed Jewish intellectuals “the meanings of certain terms,” as well as “very obscure parables occurring in the books of the prophets” in order to show to them that no real conflict exists between the teachings of the prophets and the teachings of the philosophers.[11] In addition to explaining the Law of Moses as a commentator, Maimonides also puts order into it as a jurisprudent, most importantly in the Mishneh Torah, his 14-volume code of Jewish law. But if for Maimonides exegesis consists first and foremost in explaining and systematizing the Law of Moses, this does not mean that he is not in important ways concerned with philosophy in general and with Aristotle in particular. In fact, Maimonides’ most orthodox expression of Aristotelianism occurs in the opening four chapters of the Sefer ha-madda‘ [Book of Knowledge], the first book of the Mishneh Torah that begins by stating the Hilkhot yesode ha-torah [Laws Concerning the Foundations of the Torah]. As the “foundations of the Torah,” it turns out, Maimonides legislates a set of opinions that amount to a summary of Aristotle’s metaphysics and physics : from the existence of God, inferred from the eternal motion of the celestial spheres all the way down to the four elements of the sublunar world.[12] But to find Aristotle here of all places clearly does not mitigate the sense that Averroes and Maimonides have left markedly different legacies behind.

The main question that I would like to address, therefore, is whether this strikingly different relation to philosophy and exegesis in Averroes and Maimonides can be explained as two ways of implementing a conceptual framework established by al-Fârâbî. In other words : are the projects of the commentator on Aristotle and of the commentator on Moses projects of two disciples of al-Fârâbî ? The thesis for which I will argue is that this is on the whole the case for Averroes, but only in a qualified way for Maimonides. Maimonides’ philosophical-exegetical project, I will try to show, is in part based on al-Fârâbî, in part creatively transforms al-Fârâbî and in part relies on premises that can clearly not be derived from al-Fârâbî. Although in parts of this paper I will build on much important earlier scholarship on the three authors to be examined, the specific question that I laid out above has to the best of my knowledge not yet been satisfactorily answered.

I. Al-Fârâbî

Let me begin by outlining my understanding of al-Fârâbî’s project. Al-Fârâbî sees himself as inheriting and continuing the project of ancient philosophy of which in turn he has a distinctive conception : in the Kitâb al-ḥurûf [Book of Letters] the second-highest and the highest stage in his curious account of the development of human knowledge are associated with Plato and Aristotle. At “the time of Plato,” knowledge has reached the epistemic status that can be achieved by dialectical means. But only in “Aristotle’s time,” al-Fârâbî writes, “scientific speculation [al-naẓar al-‛ilmî] is completed, the mathematical methods are all distinguished, theoretical philosophy and universal practical philosophy are perfected, and they cease to contain any object of examination. It becomes an art that is only learned and taught.”[13] Elsewhere he writes that true philosophy “was handed down to us by the Greeks from Plato and Aristotle alone.”[14] But note that the account in the Ḥurûf is clearly not meant to be primarily historical. It is a theoretical model of how human knowledge evolves assuming an absolute starting point that al-Fârâbî illustrates by what he takes to be the development of Greek philosophy.[15] In fact, in Taḥṣîl he traces the origin of philosophy back to Ancient Mesopotamia, from where it was transmitted to “the people of Egypt, from there to the Greeks, where it remained until it was transmitted to the Syrians and then to the Arabs.”[16] Since al-Fârâbî wrote most of his works in Baghdad, the political and intellectual capital of the Abbasid caliphate, this passage seems to suggest that by taking up the project of ancient philosophy, he is merely returning philosophy to where it originally came from.[17] In yet another text that is important for this discussion — Fî ẓuhûr al-falsafa [On the Appearance of Philosophy] — al-Fârâbî draws a line that through a long chain of intermediaries leads from Aristotle to himself.[18] He suggests that after the pagan period Christian authorities prohibited the teaching of those parts of the philosophical curriculum that were perceived as a threat to Christian doctrines, and then presents himself as the first to restore philosophy to its full scope in the context of Islam.[19] As historical documents these texts are, of course, of little interest. But they allow us to see how al-Fârâbî situates himself in relation to ancient philosophy and how he conceives his own role in the process of philosophy’s transmission to the Muslim world. His self-presentation clearly corresponds to how he was perceived by later falâsifa who take him to be the foremost philosophical authority after Aristotle.[20]

Much of al-Fârâbî’s work can be understood in light of the task he set himself of taking up the project of ancient philosophy, culminating in the writings of Plato and Aristotle. This task, I suggest, he conceived as twofold. For one thing the writings of the Greek philosophers that had recently become available in Arabic translation needed to be explained.[21] For having a translation and understanding it are clearly two different things. They needed to be explained for two related reasons : firstly, because for al-Fârâbî, as we will see shortly, these writings provide the key to attaining happiness and human perfection and secondly, to ensure the continuation of the transmission of philosophy, which al-Fârâbî had traced from Aristotle to himself. The second task consists in clarifying the relationship of philosophy to the divine Law. This was an urgent task for a number of reasons. Al-Kindî (d. ca. 870), for instance, had made unacceptable philosophical compromises in al-Fârâbî’s view in order to integrate philosophy into an Islamic framework — so much so that he is passed over in silence in al-Fârâbî’s account of the “appearance of philosophy” in Islam.[22] We saw, moreover, that Christian authorities, according to that same account, had prohibited teaching parts of the philosophical curriculum for religious reasons. On the opposite side of the intellectual spectrum Muslim freethinkers, such as Abû Bakr al-Râzî, called into question that prophetic guidance was required at all given the guidance of the intellect that God had bestowed on all human beings. The plurality of prophetic religions, moreover, and their claim to exclusive truth had only given rise to destructive religious strife in Râzî’s view.[23] We know that al-Fârâbî wrote a refutation of one of Râzî’s works, the Kitâb al-‛ilm al-ilâhî [Book of Divine Science].[24] Unfortunately neither Râzî’s work nor al-Fârâbî’s response have been preserved, but quotations from Râzî by later authors suggest that the book contained aggressive anti-prophetic statements.[25] Thus a number of questions arise. Must the philosophy of Aristotle be modified in order to fit into an Islamic framework ? Must it be partially prohibited ? And conversely, does the divine Law become redundant if we can rely on the guidance of the intellect ?

Al-Fârâbî’s twofold task, as I suggested describing it, reflects challenges specific to introducing philosophy to the Muslim world. While historians of Islamic philosophy broadly agree that al-Fârâbî’s intellectual outlook was significantly shaped by the late Alexandrian tradition of Neoplatonic commentators on Aristotle, the continuity between the philosophical curriculum in late ancient Alexandria and early medieval Baghdad accounts only partly for al-Fârâbî’s project.[26]

Let me start, then, with the first task and offer a hypothetical reconstruction of how al-Fârâbî’s writings can be related to it. I do, of course, not claim that it corresponds to the chronological order in which al-Fârâbî wrote his works or to a preconceived plan that he systematically executed. The reconstruction is meant as no more than an interpretation of how several parts of al-Fârâbî’s corpus fit together. To begin with the project of ancient philosophy itself needs to be clarified. The general outline that al-Fârâbî provides in Taḥṣîl al-sa‛âda shows that he takes its central concern to be an inquiry into the constituents of “sa‛âda [happiness]” and into how to attain and to disseminate it. In Falsafat Aflâṭûn [The Philosophy of Plato] and Falsafat Arisṭûṭâlîs [The Philosophy of Aristotle] al-Fârâbî goes on to explain how this project informs the corpus of Plato’s and Aristotle’s writings.[27] Of course the question arises whether Plato and Aristotle can be taken to pursue the same project at all. Although we saw that for al-Fârâbî Aristotle is superior to Plato, he stresses that the “purpose” of their respective philosophy is the same.[28] Hence he writes a treatise that aims to harmonize their views on issues on which they appear to disagree.[29] Next the systematic structure of the scientific curriculum needs to be established which al-Fârâbî does most prominently in his Iḥṣâ’ al-‛ulûm [Enumeration of the Sciences].[30] A number of introductory works exhort to or prepare for the study of philosophy.[31] After all these preliminary steps have been completed the strictly philosophical work can begin : the explanation of Aristotle, mainly in form of commentaries, but also in form of treatises that elucidate a topic in a systematic manner. Since Aristotle is taken to be in agreement with Plato and at the same time superior to him, no explanation of Plato is required. Al-Fârâbî stresses that the Greek philosophers did not only leave an outline of philosophy itself, but also of “the ways to it and of the ways to re-establish it when it becomes confused or extinct.”[32] By “the ways” he certainly means Aristotle’s logic, and the Organon clearly stood at the center of his interest. He commented on all parts of the Organon as it had been established in the late ancient Alexandrian tradition — from Porphyry’s Eisagôgê to the Rhetoric, and the Poetics.[33] This should not surprise us. If al-Fârâbî’s goal is to establish philosophy in the Islamic world, providing the “instrument” for establishing it must be a central concern to him. Much of al-Fârâbî’s authority for philosophers like Averroes and Maimonides rests, in fact, on his outstanding contribution to logic.[34] But al-Fârâbî also wrote commentaries and treatises on the philosophical sciences properly speaking, both the theoretical sciences — mathematics, physics, psychology, metaphysics — and the practical sciences — ethics and politics.[35]

I turn now to the second part of al-Fârâbî’s task : determining the relationship between philosophy and the divine Law. This task is carried out in the framework of al-Fârâbî’s political philosophy which in turn was mainly shaped by Plato. Scholars have long wondered why Aristotle’s Politics was not translated into Arabic and why al-Fârâbî — and medieval Muslim and Jewish philosophers in general — turned to Plato rather than Aristotle for their political philosophy despite considering Aristotle superior to Plato.[36] But recall that Aristotle’s superiority, according to al-Fârâbî, stems from the fact that his scientific method is superior to Plato’s. Whereas in Plato’s time knowledge was still pursued by means of dialectics, in Aristotle’s time “scientific speculation” was “completed” i.e. reached the stage of “burhân [demonstration],” the method set forth in the Posterior Analytics. Aristotle himself, however, stresses that “political philosophy” is not an exact science. Its premises and conclusions are only “generally [epi to polu]” and not necessarily true.[37] For medieval Muslim and Jewish philosophers this means that political philosophy by its nature cannot generate more than dialectical propositions.[38] Hence even on al-Fârâbî’s conception of Greek intellectual history Plato would seem to be as good as Aristotle with regard to political philosophy. Plato, of course, had given more thought to political issues than Aristotle. And the way he proposes to conceive the relationship between philosophy and political power in a virtuous polis and the role of philosophy in the pedagogical-political guidance of the citizens to virtue provided a much better conceptual framework for interpreting Islam and Judaism as philosophical religions than Aristotle’s Politics. It is also not true that Aristotle plays no role in al-Fârâbî’s political thought. First and foremost the Organon, but also the De anima and the Nicomachean Ethics provide crucial concepts that are integrated into the overall Platonic framework.[39] We will see an example of this below. Let me finally add that recent scholarship has on the whole convincingly argued that also in his political thought al-Fârâbî was significantly influenced by Neoplatonic traditions and is best understood in the context of the Neoplatonic reception of Plato’s political philosophy.[40] How, then, did Plato’s political philosophy prove useful to al-Fârâbî for determining the relationship between philosophy and the divine Law ?[41]

Throughout Plato’s dialogues, the goal of his philosophical-political project remains the same : to make the citizens better by leading them to virtue.[42] In the Socratic dialogues the key to virtue is knowledge which the Socrates of the Apology thinks is to some extent accessible to all citizens.[43] Knowing what is good for Socrates is a necessary and sufficient condition for doing what is good.[44] Political success thus consists in directing the citizens through philosophical inquiry to knowledge.[45] But from the middle dialogues onwards Plato no longer considered knowledge as sufficient for the task. When he came to see the human soul as having both rational and irrational parts, and most souls as dominated by one of the latter, the problem of non-philosophers — i.e. of those who cannot be led to virtue through philosophy — became a central concern for him.[46] My understanding of Plato’s solution is this. Although philosophy remains a necessary condition for becoming perfectly virtuous, all human beings can be led to a lower level of virtue through a pedagogical-political program designed by philosophers for non-philosophers. This explains the elaborate discussion that Plato devotes to non-philosophical devices — most strikingly in the Laws. The program’s main components are religious stories, laws, persuasive speeches, and religious practices. The crucial point for my purpose is that Plato abandons the Socratic attempt, portrayed in the Apology, to lead all citizens to virtue through philosophical inquiry. In the Republic, Plato, in fact, explicitly criticizes the use of the elenchos for testing the beliefs of non-philosophers, since it will cause them to loose the traditional beliefs in which they were brought up and they lack the ability to “discover the true ones.” After having been “law-abiding,” therefore, they “become lawless.”[47]

Al-Fârâbî adopts Plato’s fundamental premise that human beings are unequal by nature and divided into a minority of philosophers and a majority of non-philosophers.[48] In Falsafat Aflâṭûn, moreover, he clearly distinguishes between the Socratic method and the method advocated in Plato’s later dialogues. “Socrates,” he claims, was only able “to conduct a scientific investigation of justice and the virtues […] but did not possess the ability to form the character of the youth and the multitude [al-aḥdâth wa-al-jumhûr].” The “philosopher, the king, and the lawgiver,” by contrast, must be able to do both : to instruct “the elect [al-khawâṣṣ]” by means of “the Socratic method” and to form the character of “the youth and the multitude” by means of a pedagogical-political program.[49] From al-Fârâbî’s Jawâmi‛ kitâb al-nawâmîs li-Aflâṭûn [Epitome of Plato’s Laws],[50] we know that at a minimum, he was familiar with the main traits of the pedagogical-political program that Plato had worked out in the Laws.[51]

We saw that in the Ḥurûf, al-Fârâbî describes the process by which the theoretical and practical sciences reach perfection. This is followed by an outline of the two methods used for disseminating the results of this process to the political community. They correspond precisely to the two methods that we just saw : the “instruction” of philosophers “proceeds by demonstrative methods,” whereas the instruction of non-philosophers, “which is public, proceeds by dialectical, rhetorical, or poetical methods.” This second kind of instruction in turn constitutes “religion [milla]” which al-Fârâbî takes to be an “imitation of philosophy [muḥâkîyya li-l-falsafa].”[52] Religion thus conceived fulfills precisely the role of Plato’s pedagogical-political program : “through religion, the multitude is taught, educated, and given all that is needed to attain happiness.”[53] Its purpose is to convey “theoretical and practical matters that have been inferred in philosophy, in such a way as to enable the multitude to understand them by persuasion or imaginative representation, or both.”[54] Religion thus serves as the “tool” of philosophy which makes philosophical contents accessible to non-philosophers.[55] God’s description as a king in Scripture, for instance, is seen as a pedagogically useful metaphorical imitation of the philosophical doctrine of God occupying the first rank in the hierarchy of existents. The notion of the king conveys an approximate idea of God’s rank to non-philosophers who cannot understand the ontological order, but who do understand the political order.[56]

Plato’s philosopher-king who has the task of guiding both philosophers and non-philosophers to the perfection and happiness possible to them is replaced by al-Fârâbî through the prophet.[57] The virtuous political community is, therefore, by the same token a virtuous religious community. The difference between the philosopher and the prophet is explained in terms of Aristotle’s psychology : the prophet has not only perfected his intellect like the philosopher, but he also has a perfect imagination. And one of the imagination’s functions, according to al-Fârâbî, is precisely “to imitate” things.[58] In other words, the prophet is not only a philosopher, but a poet and orator as well which allows him to guide both the philosophers and the non-philosophers in his community. It should be clear by now how the late ancient version of the Organon is integrated into this Platonic framework : the prophet instructs philosophers by presenting them propositions about things as they truly are and then leads them to assent by providing a demonstration. To non-philosophers, on the other hand, he presents propositions that for the most part poetically imitate reality and then leads them to assent through rhetorical or dialectical arguments.[59] Thus religious texts for al-Fârâbî consist mainly of metaphors, parables, and rhetorical and dialectical arguments.

One implication of conceiving religion as an “imitation” of philosophy is that much of its content will be false if understood literally. God, for example, according to the philosopher, is not truly a king as he is poetically represented in Scripture. But the representation is true if it is understood as a metaphor for God’s ontological rank. A religious text is thus true when considered in terms of its allegorical content, i.e. when the imitated doctrine is made visible behind the imitation. That al-Fârâbî is aware of these implications is clear from his considerations of what occurs when a community adopts from another community both a virtuous religion and the philosophical sciences which that religion imitates. In this case neither the philosophers nor the non-philosophers know that their newly acquired religion is an imitation of their newly acquired philosophy. As a consequence, they will perceive the two as being in conflict. For the philosophers, al-Fârâbî claims, this conflict will be resolved once they learn that the seemingly false content of the religion consists in parables representing true doctrines. The philosophers are then called to “oppose” the “assumption” of non-philosophers “that religion is in conflict with philosophy.” They “strive to rid [the non-philosophers] of this assumption by seeking to make them understand that what their religion contains are parables.”[60] But if true religion coincides with true philosophy on the allegorical level, then the difference between the two — which seemed to be implied in the notion of “imitation” — disappears. Only taken literally, religion is an imitation of philosophy. Taken allegorically religion and philosophy are the same.

Al-Fârâbî is now in a position to respond to al-Kindî and the Christian authorities that there is no need to either compromise philosophy or prohibit part of its teachings. For once understood correctly, a virtuous religion is in complete agreement with philosophy. He can also respond to Abu Bakr al-Râzî that the literal content of the prophetic teachings is not redundant because God did not bestow equal intellectual abilities on all human beings.[61] As a consequence, the intellect is not sufficient to guide all of them to perfection and happiness.

Two final points must be added to this outline in order to understand how it informs the relationship between philosophy and exegesis in Averroes and Maimonides. Firstly, al-Fârâbî subdivides the group of non-philosophers into two sub-groups : the majority which is guided by poetic, dialectical, and rhetorical means and a minority which knows things “according to their true nature through the insights of the philosophers, following them, assenting to their views and trusting them.”[62] In other words, these non-philosophers apprehend things not in form of poetic imitations, but as they truly are. They do not, however, assent to them on account of demonstrations, but on account of the authority of the philosophers. Secondly, the group of philosophers and non-philosophers are not entirely static for al-Fârâbî. In the Mabâdi’, for instance, he describes the different stages in the long process leading to intellectual perfection — from the inborn natural disposition to receive intelligibles to the final state in which it can be said that “the active intellect has descended” on a person.[63] The implication is that nobody is born a philosopher. At the beginning of their intellectual careers the potential philosophers, like the non-philosophers, will not be able to grasp the true nature of things by means of demonstration. Like everybody else their understanding will be shaped by poetical, rhetorical, and dialectical devices.[64] In contrast to non-philosophers, however, the philosophers will reject the parables imitating philosophical doctrines as inadequate and false, once they advance in their education and apprehend things according to their true nature. This I take to be the group that al-Fârâbî in the Mabâdi’ calls the “seekers of the right path [al-mustarashidûn] :”

It is not impossible that among those who know these things through imitating parables, there is someone who puts his finger on the grounds for objection to those parables and holds that they are inadequate and false. There are different kinds of these people : first those who seek the right path. When one of them rejects anything as false, he will be elevated [rufi‛a] towards a better parable which is nearer to the truth and is not open to that objection ; and if he is satisfied with it, he will be left where he is. When that better parable is also rejected by him as false, he will be elevated to another rank, and if he is then satisfied with it, he will be left where he is. Whenever a parable of a given standard is rejected by him as false, he will be elevated to a higher rank, but when he rejects all the parables as false and has the strength to understand the truth, he will be made to know the truth and will be placed into the class of those who take the philosophers as their authorities. If he is not yet satisfied with that and desires philosophical wisdom [tashawwaqa ilâ al-ḥikma] and has the strength for it, he will be made to know it.[65]

It now turns out that the philosopher-prophet who rules the virtuous community has one additional pedagogical task. Besides instructing philosophers by means of demonstrations and making the true nature of things as much as possible accessible to non-philosophers by means of imitations, he also gradually “elevates” potential philosophers, who are advancing on the “right path,” from the imitations back to the true nature of things. The “seekers of the right path” react, in fact, in a way similar to the philosophers living in a community which has adopted its religion and its philosophy from another community and who fail to realize that the former is an imitation of the latter. In both cases the imitations are rejected as false, because they are understood literally. Once the imitation is recognized as imitation and its allegorical content as true there is no longer reason for rejecting it. Thus “elevating” potential philosophers to understanding the true nature of things means both, elevating them to philosophy and to the allegorical content of religion.

II. Averroes and Maimonides : Disciples of al-Fârâbî ?

It is, of course, not difficult to identify Averroes’ commentaries on Aristotle as continuing what I characterized as the first aspect of al-Fârâbî’s project. Averroes shares al-Fârâbî’s high opinion of Aristotle. In his Long Commentary on the De anima he writes, for example : “I think that this man [Aristotle] was a model in nature and the instantiation that nature found for showing the highest human perfection.”[66] The twin reasons that I suggested motivated al-Fârâbî’s project as a commentator on Aristotle apply to Averroes too : attaining happiness and human perfection and ensuring the continuity of the transmission of philosophy. But Averroes, in addition to taking up the project of al-Fârâbî, also builds on al-Fârâbî’s contributions to it. Thus we learn from Ibn Ṭumlûs, a student of Averroes, that al-Fârâbî’s works on logic were considered the most accurate introduction to the subject.[67] Averroes’ explanation of the legitimacy of his commentaries from the point of view of Islam follows from al-Fârâbî’s premises, but is not made explicit by al-Fârâbî. To understand it we must first examine how Averroes conceives the relationship between philosophy and Islam.

In general, Averroes adopts the Platonic interpretation of religion that we have seen in al-Fârâbî. This becomes particularly clear in his commentary on Plato’s Republic in which he frequently quotes important passages from al-Fârâbî’s works.[68] But for the application of this conceptual framework to the interpretation of Islam as a philosophical religion we have to turn to Averroes’ philosophical-theological treatises of which the most important is the Faṣl al-maqâl wa-taqrîr ma bayn al-sharî‛a wa-al-ḥikma min al-ittiṣâl [Decisive Treatise and Determination of the Relationship between the Divine Law and Philosophy]. Averroes’ fundamental assumption concerning this relationship is the following :

Since this Law [sharî‛a] is true and calls to the reflection leading to cognition of the truth, we, the Muslim community, know firmly that demonstrative reflection cannot lead to something differing with what is set down in the Law. For the truth does not contradict the truth [al-ḥaqq lâ yuḍḍâd al-ḥaqq] ; rather, it agrees with and bears witness to it.[69]

Averroes, of course, knows that this cannot be the case if the sharî‛a is understood literally. For then it contains much that is at odds with what philosophy demonstrates. The reason for this is that for Averroes, like for Plato and al-Fârâbî, there is an important “‘difference in human nature [ikhtilâf fiṭra al-nâs],” namely that between philosophers and non-philosophers, and that the divine Law is addressed to all Muslims, and not only to the philosophers among them.[70] To achieve this, the prophet proceeds as follows : for one thing he calls the philosophers to pursue true knowledge on the basis of demonstrations. In addition he translates this knowledge by means of rhetorical and dialectical arguments, as well as poetic representations into a language accessible to non-philosophers. As a consequence, contradictions arise between the literal sense of the divine Law and the doctrines demonstrated by the philosophers. These contradictions can be solved, according to Averroes, through “exegesis [ta’wîl]” which discloses the “allegorical sense [bâṭin]” of the divine Law.[71] Up to this point, Averroes follows al-Fârâbî fairly closely. He goes one step beyond al-Fârâbî, however by expressly confining allegorical exegesis to the philosophers. As far as I can see, nothing in Averroes corresponds to the educated audience to whom al-Fârâbî thought philosophy could be disclosed in a non-demonstrative manner.[72] The difference between philosophers and non-philosophers with respect to the truth is thus twofold : Only the philosophers have access to the truth through scientific demonstrations and only the philosophers have access to the “allegorical sense” of the divine Law. For Averroes pointing out in public that the literal sense of the divine Law is false and disclosing its allegorical sense would precisely undermine the intention of the prophet who concealed the allegorical sense because of the “difference in human nature.” Averroes explains this by drawing an analogy between the role of the medical doctor and the role of the lawgiver in which he opposes the lawgiver to the person who intends to disclose the allegorical content of the divine Law :

Here is a parable of these people’s intention as contrasted to the intention of the Lawgiver [al-shâri‛] : Someone intends [to go] to a skilled physician who intends to preserve the health of all of the people and to remove sickness from them by setting down for them prescriptions to which there is common assent [mushtarakat al-taṣdîq] about the obligation of practicing the things that preserve their health and remove their sickness, as well as of avoiding the contrary things. He is not able to make them all become physicians, because the physician is the one who knows by demonstrative methods [bi-al-ṭuruq al-burhânîyya] the things that preserve health and remove sickness. Then this one [the allegorical exegete] goes out to the people and says to them : “These methods this physician has set down for you are not true.” And he sets about rejecting them until they have rejected them. Or he says : “They have interpretations.” Yet they do not understand [these interpretations] and thus come to no assent as to what to do because of them.[73]

To the “health” in the parable corresponds the perfection and happiness to which the prophet and lawgiver intends to lead all human beings to the extent that they can attain it. To the “prescriptions” corresponds the divine Law. What Averroes means is that if the beliefs based on the literal sense of the divine Law are taken away from non-philosophers who do not understand the allegorical sense, because they lack the required intellectual abilities for understanding it, than these non-philosophers will fall into nihilism. For they will not follow the guidance of the lawgiver on account of its literal sense which has lost its authority for them, nor will they follow it on account of the allegorical sense, because they do not understand it. They loose, for instance, their belief in God as a king who rewards the obedient and punishes the disobedient. At the same time they are unable to understand the notion of a first cause and how it relates to a virtuous life. Hence they loose both their belief in God and their belief in the value of a virtuous life. It may be noted that Averroes’ critique of publicly interpreting the divine Law has the same rationale as Plato’s critique of subjecting non-philosophers to elenchoi. Again and again Averroes stresses that the allegorical sense of the divine Law is not to be made public. His sharp criticism of Muslim theologians who “strayed and led astray” is motivated above all by the fact that they “revealed their allegorical interpretation to the multitude [ṣaraḥû bi-ta’wîlihim li-l-jumhûr],” i.e. did not respect the divisions due to the “difference in human nature.”[74]

Let us now return to Averroes’ commentaries on Aristotle. To the extent that he took them to have established true philosophical doctrines by means of demonstrations he must have considered them as coinciding with the allegorical content of the divine Law. But here a question arises : does not setting these doctrines forth in books that in principle can be read by non-philosophers transgress the strict prohibition of disclosing the divine Law’s allegorical content ? To this Averroes responds that books that “use demonstrations are accessible only to those who understand demonstrations.”[75] In other words, Averroes takes his commentaries on Aristotle to be protected from non-philosophers on account of their difficulty. This is an original point. Both al-Fârâbî and Maimonides distinguish between two kinds of esoteric writing : the use of “parables and riddles” on the one hand and the use of “obscurity and brevity” on the other.[76] To this Averroes adds a third kind : difficulty.

Unlike al-Fârâbî, Averroes conceives the relationship between philosophers and non-philosophers as static. As far as I can see, Averroes does not consider the case of the “seekers of the right path” who must be successively elevated to levels of understanding of the divine Law closer to the truth in correspondence with their intellectual progress.

Precisely al-Fârâbî’s concept of the “seekers of the right path” provides, on the other hand, a key to understanding Maimonides’ project. As I stressed in the introduction : Maimonides’ work could hardly be more different than that of Averroes. He is, above all, a commentator on the Law of Moses and not a commentator on Aristotle. He does, on the other hand, share all the premises informing al-Fârâbî’s and Averroes’ Platonic interpretation of religion and uses them to present Judaism as a philosophical religion : the assumption that human beings are by nature unequal and hence subdivided into philosophers and non-philosophers, the notion of the prophet-philosopher who has a perfect intellect as well as a perfect imagination and who teaches philosophers by means of demonstrations and non-philosophers by means of rhetorical, dialectical, and poetical devices, the importance of concealing the allegorical content of the divine Law for the protection of non-philosophers — all these elements can be found in Maimonides as well.[77] This notwithstanding, the Guide is presented as a book of Biblical exegesis. To understand why, it is important to clarify who the “perplexed” are whom Maimonides is addressing. In the introduction to the Guide, Maimonides characterizes the “perplexed” person as a Jewish intellectual who has “studied the sciences of the philosophers and come to know what they signify :”

The human intellect having drawn him on and let him to dwell within its province, he must have felt distressed by the literal meanings of the Law [ẓawâhîr al-sharî‛a] […] as he continued to understand them by himself or was made to understand them by others. Hence he would remain in a state of perplexity and confusion as to whether he should follow his intellect, renounce what he knew concerning the terms in question, and consequently consider that he has renounced the foundations of the Law. Or he should hold fast to his understanding […] of these terms and not let himself be drawn on together with his intellect, rather turning his back on it and moving away from it, while at the same time perceiving that he had brought loss to himself and harm to his religion. He would be left with those imaginary beliefs [al-i‛tiqadât al-khayâliyya] to which he owes his fear and difficulty and would not cease from heartache and great perplexity.[78]

In al-Fârâbî’s words from the characterization of the “seekers of the right path” one could say that the Jewish intellectual has “put his finger on the grounds for objection” to many terms and parables in the divine Law because he recognized them as “inadequate and false,” i.e. because they do not correspond to the true nature of things. And the philosophical-exegetical program of the Guide is precisely a response to this situation. As I mentioned above, Maimonides’ purpose is “to give indications [tanbîh]” by explaining “the meaning of certain terms” and “very obscure parables occurring in the books of the prophets.” In a sense, therefore, it seems that the project of the Guide can be characterized as “elevating” Jewish intellectuals from the literal sense of the divine Law, designed according to pedagogical and political considerations for non-philosophers, to the allegorical sense of the divine Law, corresponding to the “truth as it is” and accessible only to philosophers. Maimonides, in fact, describes the process of “elevating” potential philosophers who are on the way to becoming actual philosophers in detail in Guide I, 33. And in the “Epistle Dedicatory,” which prefaces the Guide, he presents himself precisely in the role of the teacher when portraying his relationship to Joseph ben Judah, his beloved student, whose departure prompted, according to the Epistle, the composition of the Guide. But while Joseph fits neatly into al-Fârâbî’s account of the “seekers of the right path,” the term can only in a qualified way be applied to the perplexed Jewish intellectuals described in the Guide’s introduction who already are philosophers but fail to understand the relationship between philosophy and religion. As a consequence they resemble more al-Fârâbî’s philosophers living in a community which has adopted its religion and philosophy from another community without at first recognizing that the former imitates the latter. For although Maimonides claims that in the remote past philosophy was part of Judaism, he also claims that this ancient Jewish wisdom was lost in the course of the Diaspora history of the Jewish people.[79] Jewish intellectuals of later times must, therefore, learn philosophy from non-Jewish authors which in Maimonides’ case means from Greco-Arabic sources.[80] As a consequence they do not recognize the literal content of their religion to be an imitation of philosophy. If, therefore, we are to characterize Maimonides’ purpose in the Guide within the conceptual framework of al-Fârâbî, my first suggestion would be that he carries out one aspect of the pedagogical task which al-Fârâbî attributes to the prophet-philosopher : he shows Jewish philosophers who take the literal sense of the divine Law to be “inadequate and false” how to coordinate the knowledge acquired from non-Jewish sources with the divine Law by “elevating” them from its literal to its allegorical content.[81] But like Averroes, Maimonides holds that publicly teaching philosophy and disclosing the allegorical content of the divine Law will not only cause “confusion” in the beliefs of non-philosophers but will ultimately lead to their “absolute negation [ta‛ṭîl maḥḍ].”[82] For this reason one would expect the Guide to refrain from straightforward philosophical exegesis. Maimonides indeed claims to have concealed its content by means of esoteric devices — not through the difficulty of Averroes’ “books of demonstration,” nor through the “parables and riddles” used by the prophets, but through obscuring his argumentation by means of deliberate disorder and contradictions.[83]

But why did Maimonides not write commentaries on Aristotle ? After all, since the alleged ancient Jewish wisdom was lost, studying Greco-Arabic philosophy is the only way to gain access to the allegorical content of the divine Law. Moreover, in a letter to Samuel ibn Tibbon, who translated the Guide from Arabic to Hebrew, Maimonides highly praises Aristotle and the Aristotelian tradition, in particular Alexander, Themistius, al-Fârâbî, Ibn Bâjja, and Averroes. And like al-Fârâbî and Averroes, Maimonides takes Aristotle to have “reached the highest degree of human knowledge.”[84] In a sense one could argue that Maimonides’ conception of Jewish wisdom precludes commentaries on Aristotle. For had he justified composing commentaries on Aristotle because their difficulty prevented them from being studied by non-philosophers, he would have had to explain why no ancient Jewish philosophical texts are extant. Maimonides, in fact, insists that in Antiquity philosophy was transmitted only orally from one generation of Jewish sages to the next until historical circumstances interrupted the chain of transmission.[85]

In light of Maimonides’ insistence on concealment, however, it should be all the more surprising that much of the Guide — and most strikingly the first part — is devoted to a straightforward explanation of the allegorical meaning of terms and parables occurring in the divine Law. These explanations are, after all, accessible to anyone who can read. Maimonides, it seems, does precisely what Averroes (and he himself !) stressed must be avoided. The reason for this, I contend, is that much of the Guide is based on a creative adaptation of the concept of “elevating” the “seekers of the right path” to which there is no counterpart in either al-Fârâbî or Averroes. For Maimonides not only Jewish philosophers need to be “elevated” from the literal to the allegorical content of the divine Law, but to some extent non-philosophers as well. Non-philosophers, of course, cannot acquire knowledge by means of demonstrations. But they can, according to Maimonides, be habituated to opinions that correspond to the true nature of things. Maimonides thinks it possible for non-philosophers to acquire opinions through habituation that coincide with the knowledge acquired by philosophers through demonstration. The most important example for this in the Guide is the doctrine of God’s incorporeality.[86] This doctrine “ought to be inculcated in virtue of traditional authority [taqlîd] upon children, women, stupid ones, and those of a defective natural disposition, just as they adopt the notion that God is one, that He is eternal, and that none but He should be worshipped. For there is no profession of unity [tawḥîd] unless the doctrine of God’s corporeality is denied.” But this doctrine is obviously at odds with much of what the Bible has to say about God :

When people have received this doctrine, are habituated to and educated […] in it, and subsequently become perplexed over the texts of the books of the prophets, the meaning of these books should be explained to them. They should be elevated to the knowledge of the interpretation of these texts [unhiḍû’ li-ta’wîlihâ], and their attention should be drawn to the equivocality and allegorical sense of the various terms — the exposition of which is contained in this Treatise — so that the correctness of their belief regarding the oneness of God and the affirmation of the truth of the books of the prophets should be safe.[87]

On the basis of the expanded version of Aristotle’s Organon, al-Fârâbî and Averroes distinguished between the demonstrative, dialectical, rhetorical, and poetical method of disseminating knowledge. To these four methods Maimonides adds a fifth that is not derived from the same conceptual framework : “inculcation in virtue of traditional authority.” Here al-Fârâbî meets the Almohads, i.e. the “professors of God’s unity [muwaḥḥidûn]” who made the strict understanding of tawḥîd — God’s unity as entailing God’s incorporeality — into the official doctrine of the Almohad kingdom that all subjects, independently of their intellectual abilities, were forced to profess.[88] It seems that Maimonides supported this aspect of the Almohad project which he thought would lead — for example with regard to the doctrine of God’s incorporeality — to the habituation of non-philosophers to an opinion that is not an imitation of the true nature of God but an accurate description of it. The implicit assumption seems to be that in this manner the truth can be progressively disclosed to non-philosophers. Moses habituated them to the belief in God’s numerical unity thus turning them away from pagan polytheism. But since, according to Maimonides, “a sudden transition from one opposite to another is impossible […] man, according to his nature, is not capable of abandoning suddenly all to which he was accustomed.”[89] Hence Moses did not impose the belief in God’s incorporeality as well. But in Maimonides’ time the commitment to God’s numerical unity could be taken for granted. Thus he can introduce a further aspect of the true notion of God by means of habituation. Plato and Averroes expected non-philosophers to fall into nihilism if their traditional beliefs were called into doubt because of their inability to replace them through true ones. Maimonides, by contrast, appears to think that this nihilism can be avoided if the true beliefs are gradually imposed through habituation. As a consequence of this habituation non-philosophers too will experience perplexity over the literal meaning of terms and parables contained in the divine Law which represent God in an anthropomorphic way. Maimonides response to this problem is again based on al-Fârâbî : non-philosophers “should be elevated to the interpretation of these texts,” which is precisely what he claims to have accomplished “in this Treatise.” It turns out, therefore, that much of the exegetical program of the Guide aims at resolving the perplexity of non-philosophers !

In the fist four chapters of the Mishneh Torah’s Commandments concerning the Foundations of the Law, Maimonides seems to extend the Almohad project of legislating true opinions. Here again Maimonides does not teach philosophy by means of demonstration nor by means of rhetorical, dialectical, or poetic devices. Instead he imposes a succinct summary of Aristotelian metaphysics and physics through the authority of the law. In light of the passage just examined I would suggest that his long term goal was to habituate non-philosophers to the basic concepts of a sound scientific worldview. In his optimistic moments Maimonides may have thought it possible to “elevate” all non-philosophers through habituation to the intermediary class of citizens that we saw in al-Fârâbî : those who know the objects of physics and metaphysics “according to their true nature through the insight of the philosophers, following them, assenting to their views and trusting them.” One version of the conception of progressively disclosing the divine Law’s allegorical content was adopted by Samuel ibn Tibbon, the founder of medieval Maimonideanism. It contributed considerably to establishing philosophical commentaries on the Bible as one of the main genres of medieval Jewish philosophy.[90]