Against Ptolemy? Cosmography in Early Kalam.

• Author: Anchassi, Omar

In 219/834 Ahmad b. Hanbal (d. 241/855) was hauled before the ealiphal confidante Ishaq b. Ibrahim (d. 235/850) for interrogation (i.e., mihna). (1) Shown "an image of the heavens and the earth and other such things," Ibn Hanbal responded with confusion. "I know nothing about it," he protested, "I do not know what this is (ma adri ma hadha)." (2) It is not clear why cosmography--ideas concerning the physical structure of the cosmos (3)--featured in this reported encounter between theological "rationalism" and "traditionalism." (4) We know that Ibn Hanbal pitted himself against a range of adversaries, among whom the Baghdad Mu'tazilis. (5) We also know that this period marked the apogee of the Translation Movement. which brought the corpus of late ancient learning, including astronomy, into Arabic. (6) One wonders which vision of the cosmos Ibn Hanbal's lvlVtazili persecutors subscribed to, traditionalist or Aristotelian-Ptolemaic. (7) Later sources suggest that the rationalist mutakallimun (exponents of kalam, dialectic theology) were typically hostile to the Aristotelian-Ptolemaic "science of the stars" ('ilm al-mtjum), an amalgam of astrology and astronomy for much of this period, (8) and that this system had long been established as the regnant "scientific" paradigm in the Near East. (9) The overall significance of the episode, nevertheless, remains unclear.

While Ibn Hanbal's confusion sheds little light on the faultlines between theological rationalism and traditionalism in the third/ninth century, it does raise interesting questions. We are relatively well informed about the traditionalist cosmography and its sources, but know far less about the relevant views of the mutakallimun. (10) A number of important studies have addressed the physical theory of kalam, but these are focused on "kalam atomism [...] the one notable exception [...] ends in the eleventh century and does not explore later developments." (11) In this article I aim to remedy the deficit, focusing not on atomism or "cosmology"--an extremely broad category encompassing a multitude of topics--but on the physical structure of the heavens and the earth as discussed by rationalist theologians (mostly Basran and Baghdad Mu'tazilis. Ash'aris, and Maturidis). (12) Josef van Ess collected some of the relevant material in his monumental Theologie und Gesellschaft (1991-95), but his attention is directed toward the second and third Islamic centuries; in addition, many important sources have since been edited. My focus is on the period from the second/eighth to the early sixth/twelfth centuries, viz., from the dissemination of the Aristotelian-Ptolemaic paradigm in Arabic to al-Ghazali's (d. 505/1111) generation.

For the purposes of this article, the Aristotelian-Ptolemaic cosmography is the view that the earth is an immobile sphere, nestled in the center of a series of concentric celestial orbs or shells. The precise number of orbs or shells, and the exact nature of their interaction, along with much else, was the subject of considerable disagreement among Aristotle and Ptolemy's interpreters and interlocutors in later centuries, Muslim or otherwise. (13) Notwithstanding their divergences, however, they "were in agreement with regard to the sphericity of the Earth and its position at the center of the universe, as well as the sphericity and the circular motion of the heavens." (14) These fundamental features of the Aristotelian-Ptolemaic paradigm remained stable down to modernity, (15) and for this reason I choose to present them as a single model or paradigm.

This cosmography espoused by Muslim advocates of 'ilm al-nujum in the period surveyed is the alternative against which traditionalists defined themselves, as shall become clear. The traditionalist cosmography, to which I devote a separate section below, consists of belief in a flat earth, beneath which are a further six earths. Instead of the celestial spheres of the Aristotelian-Ptolemaic cosmography, there is a series of seven heavens (either domed or vaulted), above which is God's throne ('arsh). What lies beneath the seventh and lowermost earth was, in the traditionalist corpus, subject to considerable disagreement. Nonetheless, the basic features of the traditionalist cosmography permit one to refer to it as a single model, or paradigm, since it remains remarkably stable.

I make three key interventions. First, I demonstrate that, other things being equal, the disagreement between mutakallimun on cosmographical questions broadly resembles the late ancient Christian debate on the merits of the Greek and Hebraic models of the cosmos. Then, more significantly, I establish that there is no correspondence between (rationalist) theological affiliation and cosmographical opinion. (16) That is to say, one can find Basran Mu'tazilis who are basically traditionalist in their cosmographical outlook alongside other Basran Mu'tazilis who reject the plain-sense meaning (zahir) of the Quran's cosmographical passages. This general rule of noncorrespondence between theological school and cosmography requires qualification: (1), the Ash'aris are consistent in their occasionalism, which shapes their views on, e.g., the cause of the immobility (wuquf) of the earth, and (2), the Maturidis are not "genetically" or geographically linked to the other three major schools, and are therefore typically less interested in exploring questions of cosmography in a kalam context. (17) With these qualifications, the rule of noncorrespondence holds. My third intervention relates to questions of diachronic development. Up to the early sixth/twelfth century there is no evidence of a trend of increasing accommodation of Greek cosmography by theologians of the four main rationalist schools. The question of the shape of the earth, for example, continued to be contested. This remained true even as the distinction between astronomy ('ilm al-hay'a) and astrology (ahkam al-nujum) emerged with greater clarity, and the former was (increasingly) shorn of metaphysical assumptions--developments that would clearly facilitate such an accommodation. (18) To make my case, I draw on the published corpus of kalam treatises of the period, along with Quran commentaries authored by theologians, and, occasionally, works from other genres. My evidentiary basis for these conclusions is comprehensive, though by no means exhaustive, and will admittedly have to be revised as new sources come to light.

I begin with a brief survey of quranic cosmography. I then proceed to explore traditionalist doctrine, and whether and to what extent it represents a departure from the vision of the cosmos presented in the Quran. I also attempt a characterization of traditionalist attitudes toward rival nontraditionalist viewpoints. These are necessary contexts for the study of the cosmography of the mutakallimim, representing the standard against which they reacted. Their discussions show evidence of extensive engagement with cosmographical traditionalism, sometimes by way of critique and at other times of adherence, in preference to the Greek alternatives. I then address the cosmographical views of individual rationalist theologians whose works are extant or whose opinions are preserved in later sources in more or less chronological order. This discussion forms the bulk of the paper.

I have made a number of choices about terminology. I write of traditionalist "cosmography" rather than the more typical "cosmology," the term Anton Heinen used in his seminal study and translation of al-Hay'a al-saniyya of al-SuyutT (d. 911/1505). (19) While "cosmology" is broader than "cosmography" and suggests the "divinely governed order of the universe and the place of humans within it," (20) and "cosmography" has no clear parallel in the traditionalist lexicon, the most proximate equivalent being hay'a, though this is not used systematically in the formative period, I regard it, following David King, as appropriate. (21) I include under this rubric a number of related discussions, such as those on the finitude of the cosmos, the causes of the immobility of the earth, and, to a lesser extent, void. Because the traditionalist cosmography includes material about the structure and composition of the heavens and the earth, there is a certain overlap with the discourse of the astronomer-astrologers (munajjimun) of the period. I have already alluded to the problem of disaggregating astrology and astronomy, which needs still to be treated systematically in its Islamicate context. (22) In addressing the formative period of Islamic thought, in particular, one cannot avoid the problem of the dearth of extant sources. Unfortunately, these problems continue beyond the second/eighth century. Large multivolume works such as the Quran commentary (tafsir) of Abu 'A1T al-Jubba'i (d. 303/915f.) are lost to us, surviving only in quotation (especially by later Twelver authors, because of their Mu'tazili commitments). (23) Where possible I have sought to corroborate quotations of lost works by drawing on multiple sources, including the five-volume series Mawsii'at tqfasir al-mu'tazila and the recently published encyclopaedic al-Tahdhib ft al-tafsir of al-Hakim al-JishumT (d. 494/1101), both published in Beirut. (24) As a general rule, there is a high degree of consistency in the cosmographical views attributed to early theologians by later authors. I have not noticed any notable discrepancies between different genres in this regard. Any divergences in the views attributed to particular figures will be noted where relevant.

1. THE QURANIC AND TRADITIONALIST COSMOGRAPHIES

   The provenance of quranic cosmography has long been of interest to Western scholarship, with theories that the seven heavens of Babylonian mythology were the (ultimate) source of Muslim belief on this point, (25) that one cannot speak of "scientific" interest in the cosmos among Muslims until more than a century after the...