User Tools

Site Tools


The Role of Astronomy in the Dialogue Between the Arab Islamic World and Europe

George Saliba


From its very beginnings, Arabic/Islamic astronomy originated, took root, and flourished in an intercultural environment. By the mere geographical proximity, and later overlapping, with two major earlier empires to the east and to the west, much of what was known in early Islamic times was inspired one way or another by the richness of those earlier traditions. On the Eastern Sasanian domain, where the emerging Islamic civilization came in close contact with the Persian/Indian astronomical tradition during the first two centuries of its own existence, Arabic/Islamic astronomy relied heavily on the legacies of older texts, observations, records, and methods of computations that were prevalent in the lands that became later the eastern part of the Islamic world. A similar and slightly more complex engagement happened on the western side as well.

The astronomical literature that was encountered in the east had in all likelihood survived on the folkloric cultural level, and at best survived as astrological applications rather than theoretical astronomy. A trend was then set to utilize the folkloric expression of those sciences and try when possible to understand the theoretical foundations upon which those applications were once based. As a result there developed a double tier conversation encountered mainly in the early Islamic astrological sources where one would find references to the ancient Persian Zīj-iShahriyār orZīj-iShāhī, on one level, and in turn encounter the more ancient Indian and Greek sources which inspired this text. As we shall soon see, this search for theoretical origins began to take a more formal shape during the caliphate of the second Abbasid Caliph al-Manṣūr (754-775 AD), when historical sources preserve accounts of translations of the various Sanskrit Siddhantas, and more particularly the Surya Siddhanta (known in the Arabic sources as the Sindhind).1

This attempt at incorporating the Indo/Persian astronomical tradition was not very long lived. For soon after that, this eastern tradition found its competitive match in the equally rich Greek tradition coming from the west which could easily outstrip the eastern tradition in terms of supplying equally practical answers for astrological practice as well as respond to the more theoretical needs better than the other eastern tradition could.

Nevertheless, in the beginning these multifaceted eastward and westward conversations took place across many platforms and through the initiatives of various individuals who were either in the employ of governmental patrons or in the employ of aspirants to such governmental positions. As a result, the conversations sound to modern historians as having been somehow chaotic, unorganized, and over ambitious. Many individuals attempted to appropriate the same texts, each in his own way, or attempted to double check the validity of earlier scientific results, each in his own way as well. And as far as we can tell, there never arose a centralized place to localize all those cultural activities, notwithstanding all what we hear about such semi-legendary institutions like the House of Wisdom (Baital-Ḥikma) in ninth century Baghdad, which were supposed to fulfill such functions, but were not as active in such cultural appropriation enterprises as is commonly believed. At times these multifarious activities were repeated over and over again, as if to express, in each case, the difficulty that is usually encountered when texts, ideas, and even crafts and cultural objects cross linguistic and cultural barriers.

This phenomenon of repeated attempts at appropriating texts (engaging in serious multilayered dialogues) is nowhere more apparent than in the attempts to acquire, for example, the works of Euclid, Ptolemy and Dioscorides, to name only three of the classical Greek scientific giants. The works of each and every one of them, the Elements, the Almagest, and the MateriaMedica, were in fact ‘translated’ into Arabic more than once, and in some instances, as in the case of the Almagest and the MateriaMedica, up to four and even five times. And yet in each and every one of those ‘translations’ we not only usually find a considerably different text from the original Greek, but we also find considerable differences among the various translations of the same texts, as if to betray and give new meaning to the concept of translation itself or expose the difficulty of the process involved. This kind of activity must mean that the receiving culture, in this instance the early Islamic culture, by initiating dialogues with the earlier cultures through the translation of those cultures’ scientific texts, was not simply engaged in preserving the original Greek, Indian or Persian texts as sacrosanct. It was not faithfully translating similes of, but that it was intentionally appropriating the productions of those earlier cultures in order to forge a vision of its own. That vision was informed by the same earlier cultures but definitely not restricted to inhabiting a world circumscribed by the confines and contours of those earlier cultures.

A similar phenomenon seems to have taken place, although it was slightly differently nuanced, when these variegated ‘translated’ Greek, Indian and Persian texts were once more ‘re-translated’ from Arabic into Latin. In this latter translation an extra layer of a translation was added on top of another translation, and at the same time transforming and transmuting the essence of those texts, as was seen fit and as was required in the new Latin environment. The peculiarities of this Arabo-Latin cultural conduit can be easily detected when we usually find the earlier ‘translations’ from Arabic, during the early Middle Ages, almost always consciously attempting to keep the spirit and phraseology of the Arabic texts, as much as that could be done. While at the later period, during the Renaissance, ‘translators’ took greater and greater liberties with the original Arabic, and at times even discarded the whole process of translation and went straight to the contents of the texts in order to utilize those contents. From that vantage point, scientist working during the European Renaissance seem to have been willfully appropriating the earlier Arabic scientific texts into their own Latin scientific works, without even pretending to be preserving the texts of the earlier Arabic/Islamic tradition..

This inter and multicultural dialogue is best illustrated in the discipline of astronomy, which itself highlights the main features of this dialogue that spanned several centuries and ‘successfully’ crossed several cultural barriers, and seems to have found those barriers easily penetrable. By moving from direct translations during the middle ages, where we still have the original Arabic texts that were rendered into Latin, to a more mature appropriation during the Renaissance, where we find ideas first found in earlier Arabic texts freely used in Renaissance scientific texts without necessarily going through the actual process of translation. In this shift the very concept of translation should then gain a much deeper meaning.

That was the state of affairs across the border between the European and the Islamic world. In what follows, the simpler translation movement of the middle ages, with its recognizable texts and translators will be intentionally sidestepped in favor of exploring the much less known aspect of that border dialogue during the Renaissance which gave rise to the subtle integration of particular Arabic astronomical ideas and observational results into the conceptual and structural schemes of Renaissance astronomy, and the general science of the Renaissance at large.

The Islamic Conversation with Earlier Cultures

As was quickly mentioned before two particular texts illustrate rather well this part of the dialogue: the Almagestof Ptolemy, and theMateriaMedica of Dioscorides. These texts demonstrate how endeavors of ‘translation’ began to gain a life of their own just as the process of translation from Greek into Arabic was taking place. They demonstrate how that independent life within the Arabic Islamic culture eventually set the resulting ‘translations’ apart from their Greek originals. And as they could also be conceived as part of a lively dialogue, these two texts also spawned other side conversations in the form of other subsidiary scientific interests within the receptive Islamic culture. Very often, it was in these subsidiary disciplines that one would witness the emergence of real breakthroughs within the Islamic culture. And it is those breakthroughs that begin to tell the story of the full impact of those texts and the emergence of the defining character of an Islamic scientific tradition.

As examples of the transmission of scientific texts those two texts demonstrate as well how eventually there came to be many Arabic Almagests, and equally multiple renditions of the MateriaMedica. In the case of the Indian astronomical texts, a similar phenomenon also materialized in which we find the astronomical work of one Muḥammad b. Mūsā al-Khwārizmī (fl. 850), which was originally written under the inspiration of the Sanskrit siddhantas, gaining a life of its own to be transmitted to the Latin west without the Sanskrit original, and to survive there for centuries on its own. In that migration the independent al-Khwārizmī text seems to disappear in the original Arabic, except in the few citations and fragments that survived in other Arabic astronomical, historical, and geographical sources.

The Emergence of the new Arabic Astronomical Tradition

Once naturalized within the Arabic Islamic tradition, texts like the Almagestand the Sanskrit Siddhantas gave rise to new opportunities to examine them very closely, to correct their mistakes, and to modify them dramatically so that they would fit the new Islamic cultural requirements. At the same time they initiated a series of ancillary sciences that tried to determine the methodological and constitutional reasons that allowed those mistakes to appear in the first place.

As a result, the new Arabic astronomical texts that were initially inspired by the Almagestand the Siddhantas, quickly began to form their own tradition, and to refocus the material that seemed dispersed in several Greek and Sanskrit texts, or that seemed not to have been fully articulated in those same texts, and to use it as a basis for new epistemological schemes that would shape the future of Islamic astronomy. In a relatively short period, Arabic astronomy managed to develop a new conceptual framework, and within that framework quickly began to produce comprehensive astronomical texts of its own, mostly referred to as hayatexts. Those newly formulated Islamic texts did not simply restrict themselves to a phenomenological description of the universe, in order to distance themselves from astrological texts that sought to know the influence of the celestial objects on earth bound individuals as they did in both the Greek and the Sanskrit traditions, but also contained the preliminary mathematical, philosophical, and physical foundations of a new astronomy, theoretically different from the inherited astronomies. In these same new texts there came to emerge a new discussion of planetary theories proper, and a rather innovative description of the errant behavior of the planets. At the same time attempts were made to outline the implication of those celestial phenomena to the observers who were bound to inhabit the various regions of the earth, and as a byproduct these attempts produced a veritable discipline of mathematical geography. Almost all hayatexts always included a whole section (usually one out of four) devoted to the earthly effects of celestial phenomena, or the manner in which celestial phenomena appeared to earth bound observers or conditioned the observers’ perception of those phenomena. Finally those hayatexts eventually tried to determine the actual dimensions of the universe we inhabit.

Naturally these genres of texts crossed various disciplinary lines and encompassed in their discussions physical, philosophical, mathematical, and of course genuine cosmological problems. And although these texts were rather comprehensive in nature, they were rarely translated into Latin, if at all, probably because they came into their prime when the massive medieval translations from Arabic into Latin were waning at the time. Furthermore, the very form of Arabo-Latin transmission had already morphed by then, in the late medieval and renaissance times, to become more of a direct appropriation of ideas, as we shall soon see, rather than a translation of texts as was done before. As a result several important hayatexts never crossed into the Latin domain. Instead in this later Latin period, especially during the sixteenth and seventeenth centuries, we begin to note the emergence of a European class of astronomers and scientists who could either read the original Arabic competently to benefit from their findings, or who could have access to collaborators and assistants who could unlock for them the language of those Arabic texts to make their novelties ripe for the picking as we shall see below.

The New Astronomical Texts and their European Reception.

In their rudimentary phase, Arabic astronomical texts were first at the stage of bringing the astronomy of the earlier Greek and Indian cultures within focus, as was done in such texts as the Jawāmi ‘‘ilm al-nujūm (compendium of the science of the stars) of al-Farghānī (Latin Alfraganus c. 860)2, the just mentioned al-Khwarizmi’s hybrid Indian Arabic astronomical handbook (zīj)3, and al-Battānī’s (Latin Alpategnius) astronomical handbook.4 When these texts were first translated into Latin during the Middle Ages, and all of them were, they apparently had such a wide impact in the Latin west on texts that ranged all the way from pure astronomy, where the Arabic sources were simply redressed in Latin garb, to literary texts, where they were used by such people as the humanist Dante in his various writings.5 In both cases, the contents of the earlier Arabic texts, which were now made available in Latin, were freely harvested. Much is known about this dialogue during the Middle Ages.

But the more intriguing dialogue, which is still relatively less known, was the one that took place during the Renaissance, when, as we just said, the very concept of translation and transmission had been radically transformed in order to make use of the contents of the Arabic astronomical texts at a much more sophisticated level. To illustrate this stage I take two examples where we find Renaissance astronomers and scientists of the caliber of Copernicus (d. 1543) and Galileo (d. 1642) using scientific ideas that were first developed in the Islamic world some two to three centuries before their time in the very epistemological constructions of their own works. And although we can be pretty certain that neither Copernicus nor Galileo would have read those astronomical and philosophical texts in their Arabic originals, yet we find their free use of those very ideas, which had first appeared in Arabic, problematic enough to warrant a new reconsideration of the genealogy of the ideas of those Renaissance scientists.

To make this point clear, let’s consider the situation when, on the one hand, we find Copernicus freely using Arabic models of mathematical astronomy that he could easily translate to heliocentrism, while everything else in the mathematical models remaining the same, and on the other hand, in two particular instances, we find him even making mistakes in his description of those models which were supposed to be of his own creation, when in fact they were correctly described in the earlier Arabic sources. Such grafting of ideas, rather than translations of the same, and his obvious misunderstanding of their full mathematical import, in all likelihood must mean that Copernicus must have at least visited the contents of those earlier Arabic texts, in some form or other, and must have had a good idea of those contents in most cases, and some poorer ones in others.

In the first instance we take the curious parallelism between the works of Copernicus (d. 1543) and those of Ibn al-Shāṭir of Damascus (d. 1375), where in the case of the purely geocentric moon, for example, we find the two astronomers using identical models to describe the movements of this planet (Figure 1).

Fig. 1a Fig. 1b

Fig.1 The Mathematical models for the motion of the moon as proposed by Ibn al-Shāṭir (1a) and then some two centuries later by Copernicus (1b). Both astronomers depart critically from the Ptolemaic model, and use double epicycles, two spheres, represented here as two smaller circles, one of them riding on the circumference of the other (and both riding on the circumference of the larger circle/sphere when conceived as circles), or within the thickness of the other, when conceived as spheres.

The most distinguishing features of this common model are that it was at once capable of accounting for the observational results of Ptolemy (d. c. 170), and yet did not distort the size of the apparent diameter of the moon as was done by Ptolemy’s model. Both astronomers were therefore adopting a model that was not only radically different from that of Ptolemy, but were also correcting the mistakes of Ptolemy. As far as we can tell, Ibn al-Shāṭir was the first to simply state that the moon was never seen to be twice as large when it was one week old as was required by Ptolemy’s model. He simply said: It was never seen as such (lamyurākadhālika). Seeing that Ibn al-Shāṭir and Copernicus were both resolving the Ptolemaic observational problems, by resorting to the very same techniques and mathematical constructions, and knowing that Ibn al-Shāṭir constructed his model at least a century and a half before Copernicus, it becomes incumbent upon historians of astronomy to explain the manner in which such identical solutions to the same problem could have come about, if one did not wish to subscribe to the idea of intercultural borrowings and influences. In light of what we shall see below, about the other similarities between other features of Copernican astronomy and the earlier Islamic astronomical sources, we shall see that such intercultural borrowings could not be easily ruled out.

Another instance of perplexing parallelisms is the occurrence of a mathematical theorem that was first tentatively formulated by the astronomer Naṣīr al-Dīn al-Ṭūsī (d. 1274), in the year 1247, in the context of his objection to the Ptolemaic description of the latitudinal motion of the planets (Figure 2). This early tentative attempt (2a) was later developed into a full fledged mathematical theorem (2b), complete with a rigorous mathematical proof, in the year 1260.

Fig 2a Fig 2b

Figure 2. The first formulation of Ṭūsī’s theorem, now known as the Ṭūsī Couple (2a) in the year 1247 [Courtesy of the British Library India Office] and its more mature reformulation, with proof in 1260 (2b) [Courtesy of the Vatican Library].

The particular theorem was originally formulated to answer a curious astronomical question that had bedeviled the Greek astronomer Ptolemy. While describing the latitudinal motion of the planets, especially Mercury and Venus, Ptolemy stipulated the existence of two cosmologically impossible spheres that performed all sorts of motions that could not be physically realized. In addition those spheres produced wobbling motions instead of the regular uniform circular motion they were supposed to perform. In this particular instance, Ptolemy knew that he was asking too much, and begged the reader to look the other way for he says “how can we (humans) imitate the behavior of the gods?” Here he meant that the mathematical contraptions that he had proposed in order to describe the latitudinal motion of the lower planets (the divine bodies), were physically impossible to achieve (by humans) in reality.

At this point, and while trying to expound the complicated Ptolemaic astronomy, probably for school purposes, Ṭūsī had to scream, “this kind of speech is outside the craft of astronomy,” (hādhākalāmunkhārijunanal-ṣināa) meaning of course that it was not permissible to make such statements in a mathematical discipline as theoretical astronomy was then perceived. And it was then that he proposed to create a new mathematical device (now known as the Ṭūsī Couple) that could generate linear motion out of two circular motions, thus achieving the oscillations required by Ptolemy’s observations without forcing the spheres to move in a wobbling motion.

In essence, the Ṭūsī Couple, stipulated the real existence of two physical non-imaginary spheres, both moving in place around axis that passed through their centers, and yet their combined motions produced a linear motion. This linear motion was exhibited thus: In Figure 3 if the larger sphere of the Ṭūsī Couple were to move counterclockwise around its center D, and the smaller sphere, which was half the size of the larger one, moved in the opposite direction at twice that speed, around the axis that passed through its own center Z, then the combined motions would force the point H which was originally the point of inner tangency of the two spheres, to move linearly along the diameter AB of the larger sphere.

Figure 3. The Ṭūsī Couple, where a large sphere AGB is supposed to move counterclockwise around its center D. The same sphere carried inside it another smaller sphere GHD, half its size, which moved in the opposite direction at twice the speed. The combined motion of the two spheres, would make point H, which was originally the point of inner tangency of the two spheres, move linearly along the diameter of the large sphere AB.

Even before getting to this point, Ptolemy had once before also stipulated the existence of another sphere that moved uniformly in place around an axis that did not pass through its center, which was plainly impossible, not to say physically absurd. Ṭūsī, like all other serious astronomers working in the Islamic tradition, knew very well that the existence of such a sphere, moving uniformly, in place, around an axis that did not pass through its center, was tantamount to assuming the existence of basketball that could be spun on the top of a finger that did not point towards the center of the ball. Once this cosmological impossibility hit this dead end, Ptolemy did not utter any word as to how it could be resolved, and did not resort to the same technique he used while excusing the failure of human imagination as he later did in the case of the latitudinal motion of the planets.

Now armed with his successful and unexpectedly fecund new theorem, Ṭūsī was in a position to solve this other Ptolemaic problem as well, which was by then called the problem of the equant that was embedded in all of the Ptolemaic models. Ṭūsī’s theorem worked well for the motion of the upper planets too, but Ṭūsī failed to deploy it for the motion of the planet Mercury, and frankly confessed that. In his work of 1260, he stated that if he ever managed to solve the intricate problems of the planet Mercury he would insert the answer at that point of his book.

This solution together with others was left for Ibn al-Shāṭir to resuscitate about a century later. Instead of restricting the application of the Ṭūsī Couple to the latitudinal motions of the lower planets Venus and Mercury, or to the solution of the equant problem of the upper planets, as was done by Ṭūsī, Ibn al-Shāṭir now applied the same theorem to change the size of what would now be called the orbit of Mercury, then known as Mercury’s epicycle, thus allowing it to shrink in size and expand, during its apparent motion from the earth. At the same time, his mathematical contraption had to successfully account for the Ptolemaic observations which up till then were thought to be superior to all others..

Figure 4. The model for the motion of the planet Mercury as conceived first by Ibn al-Shāṭir and taken up later on by Copernicus, demonstrating the source of Copernicus’s mistake when he stated that the orbit of Mercury appeared largest at 90° away from the apogee. He seemed to have confused the absolute size of the object with its apparent size. The dotted angle marked as “maximum elongation”, which indicates the maximum size of the planet Mercury’s orbit as seen by an observer on earth at O, clearly demonstrates that Mercury’s orbit at ± 120°, encompassed by the dotted lines, is clearly larger than the angle with the continuous lines that encompasses the orbit at 90°.

It was in this new reconfiguration of the motions of the planet Mercury that both Ibn al-Shāṭir and Copernicus had to resort to the same technique again and at two different junctures. First they both knew that with the appropriate application of the Ṭūsī Couple, Mercury’s motion could be described in such a way that it accounted for the Ptolemaic observations, and still avoid the cosmological absurdities embedded in the Ptolemaic model. And so once more, Copernicus deployed the same model that was used by Ibn al-Shāṭir for the solution of the motions of the planet Mercury by keeping the whole model intact and only shifting the center of the universe from the earth to the sun. All other components of the motions of the planet Mercury essentially kept the very same elements of Ibn al-Shāṭir’s model and now only introducing the sun as the reference point instead of the earth, which is mathematically relatively trivial. But the revealing indebtedness came when Copernicus attempted to describe the actual motions of the planet Mercury for an observer on earth. At that point he mistakenly made the claim that the planet Mercury would appear to reach its maximum orbit when it was 90° away from it apogee, instead of ± 120° as was stipulated by both Ptolemy, who backed his claim with observations, and Ibn al-Shāṭir who followed suit. In the language of Noel Swerdlow, the editor and commentator on Copernicus’s early astronomical work, the Commentariolus: “This misunderstanding must mean that Copernicus did not know the relation of the model to Mercury’s apparent motion. Thus it could hardly be his own invention for, if it were, he would certainly have described its fundamental purpose rather than write the absurd statement that Mercury “appears” to move in a larger orbit when the Earth is 90° from the apsidal line. The only alternative, therefore, is that he copied it without fully understanding what it was really about. Since it is Ibn al-Shāṭir’s model, this is further evidence, and perhaps the best evidence, that Copernicus was in fact copying without full understanding from some other source, and this source would be an as yet unknown transmission to the west of Ibn al-Shāṭir’s planetary theory”.6

That was not all, since both Ibn al-Shāṭir and Copernicus, after him, knew that they had to use the Ṭūsī Couple to resolve the Mercury motion, Ibn al-Shāṭir went ahead and assumed his reader knew all about the work of his predecessor Ṭūsī, whom he repeatedly mentioned in his own work. But Copernicus who was addressing a completely different audience (a Latin reader this time) felt obliged to demonstrate the workings of the Ṭūsī Couple in a separate chapter of his DeRevolutionibus III,10, before he would go ahead and use it in his description of the Mercury model. And here again he seems to have followed his predecessor too closely, and ended up with a misreading. In Figure 5, which compares the proof of the Ṭūsī Couple both in the works of Copernicus, on the right, and Ṭūsī, on the left, the late Willy Hartner had noted that Copernicus used the Latin phonetic equivalents of the same alphabetic letters that were used by Ṭūsī before to designate the same geometric points that were designated by Ṭūsī.

Figure 5. Proof of the Ṭūsī Couple by Ṭūsī, on the left, and Copernicus, on the right, and where the Arabic letter “alif” is rendered with Latin A, “bā’” with B, etc, except for the sole difference in rendering the Arabic letter “zain” with the Latin F. As the Arabic letters zain and fā’ are orthographically very close, this demonstrates a possible misreading of zain for a fā’, on the part of Copernicus, or whoever was assisting him in deciphering the Arabic figure.

This is not the whole story of the impact of the Ṭūsī Couple on Renaissance thinkers. In the Islamic world, the commentators on Ṭūsī’s work had already noted that this production of linear motion as a result of two circular motions had other implications as well.7 On the one hand, it played havoc with the neat Aristotelian division between the celestial region of the cosmos, where all motions were naturally circular, and the sublunar region where the motions were naturally linear. On its own, linear motion had complications as well when it came to two motions that were linear and opposite in direction. Two such opposite motions were supposed to annul each other out, thus causing generation and corruption, as Aristotle would say. But such opposing motions also generated a second, much more serious problem, which in antiquity had already pitted Aristotle against Plato, as Aristotle was prone to do.

In this instance Plato had held, in the Phaedo, that there was no moment of rest between two opposing motions as was the case with the subterranean waters that were constantly in motion back and forth without rest.8 To illustrate that position later philosophers, like Richard of Middleton, for example9 among others, would use the argument that it would be inconceivable for a small pebble tossed against a large falling stone to bring the large stone’s motion to a moment of rest, and thus have both the pebble and the large stone stand still before the pebble reversed its direction to that of the falling large stone. Nevertheless Aristotle insisted that such a moment of rest had to occur between the pebble’s two opposing motions.

Later commentators on the Greek philosophers kept this debate alive throughout antiquity, and when it reached the Islamic world many people, particularly philosophers such as Ibn Sina (Latin Avicenna d. c. 1037) and Abū al-Barakāt al-Baghdādī (fl.1100) after him participated in the discussion. Abū al-Barakāt even went a step further to side with Plato in denying a moment of rest between two contrary motions by proposing the example of a ruler, with a hole in its midst, and a plumb line passing through that hole. He then said that as one passed his hand which held the other side of the plumb line from one end of the ruler to the other and moved his hand in a steady continuous motion till the other end, the plumb line would descend and ascend without reaching a moment of rest between the two motions, as both motions were caused by a continuous motion of the hand in one direction.

Commentators on Ṭūsī’s work saw the same potential with Ṭūsī’s Couple, namely, that it also demonstrated very clearly the possibility of two contrary motions caused by continuous circular motions and thus did not pause for a moment of rest. If the cause of the motion did not rest, they argued, the result of the motions would not rest either.

This discussion was not exploited by Copernicus as far as we know, nor was the Ṭūsī Couple exploited in this sense by Ṭūsī himself, despite the fact that his commentators in the Islamic world did so. In the Latin west, it was apparently Galileo who found the discussion particularly productive for his physical studies of motion. In his early works, especiallyDemotuantiquiora, when he tried to refute the Aristotelian claims on the nature of motion and more specifically the Aristotelian claim that every two contrary motions must have a moment of rest separating them, Galileo went ahead and used the Ṭūsī Couple in the same manner it was used by Ṭūsī’s commentators. For Galileo too, Ṭūsī’s theorem clearly demonstrated the possibility of a continuous circular motion producing continuous linear contrary motions without any moment of rest between them. This particular feature of the theorem must have moved Galileo to inform us, in explicit terms, that the mathematical construction he had encountered in the works of Copernicus was very useful for the discussion, and thus could demonstrate that there was no necessary moment of rest between two contrary motions. The “two circles” Galileo spoke about in hisDemotuantiquora, were none other than the ones which were used by Copernicus before him in his own attempts to produce linear motion from continuous circular motion particularly for the motion of the planet Mercury as we just saw. The fact that Copernicus was in that instance mimicking the diagram in the Arabic text of Ṭūsī should be clear to us by now.

So when Galileo refers to this very theorem in the Copernican work to buttress his own argument that one could find contrary motions that were not separated by a moment of rest, Galileo was in essence inheriting the same Ṭūsī couple through Copernicus, and was drawing from it the obvious philosophical implications for Aristotelian physics that the commentators on Ṭūsī’s work had already drawn before. Furthermore, Galileo was also inheriting a much longer Arabic tradition in which such Aristotelian problems were widely circulated and discussed in the Islamic culture, and in all likelihood were themselves the fertile grounds where one could profitably seek the genesis of the Ṭūsī Couple itself. As we just mentioned, anti-Aristotelian attacks were apparently known for centuries before the time of Ṭūsī, and in particular by some who obviously lived before the time of the famous philosopher Abū al-Barakāt al-Baghdādī (fl.1100). For in Abū al-Barakāt’s famous philosophical, and relatively anti-Aristotelian, work al-Mutabar, there is a reference to someone or ones (baal-fuḍalā) who were cited by al-Baghdādī to have even produced a mechanical example which illustrated the continuity between contrary motions.

As we also stated, by Ṭūsī’s time almost all commentators on his major astronomical work in turn raised their own concern with the philosophical implications of the Ṭūsī Couple. And like al- Baghdādī they too would at times even supply examples of their own, as was done by Ṭūsī’s student Quṭb al-Dīn al-Shīrāzī (d. 1311). In order to demonstrate the continuity of the linear opposing motions, Shirāzī suggested the example of a bowl whose edge was of uneven heights and had a string passing through a hole at its bottom. The end of that string was supposed to have a plumb weight attached to it. Shīrāzī then stated that when one moved the other end of the string along the edge of the bowl the motion will produce an oscillating up and down linear contrary motions of the plumb weight as a result of a continuous single circular motion, without any moment of rest to separate the contrary motions.

We do not know if these Arabic anti-Aristotelian debates were known in the Latin west in their full richness. Nor do we know the routes they would have followed when they reappeared in their Latin garb. But the very existence of such anti-Aristotelian arguments within the various epistemological disciplines of the Islamic culture means that the Arabic philosophical tradition seems to have easily overlapped with the astronomical tradition, and must have shared some common ground with the mathematical domain in the form of theorems describing physical phenomena, as seems to have happened with the Ṭūsī Couple. These theorems in their turn, added a mathematical rigor to the import of those debates by supplying formal mathematical proofs as was done with the Ṭūsī Couple. It is probably that very feature that made them attractive to people like Copernicus, and through him to Galileo, to incorporate them in their own arguments.

What we note through the close investigation of texts of all cultures concerned, whether Sanskrit, Greek, Arabic or Latin, is that disciplines were allowed to overlap, languages and the concepts that were embedded in them were forced to merge, ideas that were the subject of a vast appropriation project were forced to float from one culture to the other, whether from Greek into Arabic, or from Arabic into Latin. All of those activities demonstrate the very fluid and dynamic dialogue that was truly universal in nature. If Islamic civilization did not do anything except create the right environment for such a universal dialogue to take place, by asking the kind of scientific and philosophical questions that were truly culture free, despite their being anchored in particular cultural moments, then it would have indeed laid the foundations for what we can now confidently call a universal modern science.

Further Appropriations of Texts and Instruments

This tradition of incorporating earlier results from the Islamic world into Latin works of the Renaissance seems to have continued well into the seventeenth and eighteenth centuries. The field of instruments seems to have experienced a similar fate.

In the case of texts the most obvious examples come from two Arabic astronomical works that are still preserved at European libraries, and which were once owned by the Renaissance scientist and man of letters, a one time royal professor of mathematics and oriental languages at the Institut Royal that was to become later on the Collège de France. The man in question is the famous Guillaume Postel (1510-1581), who at one point had in his possession two astronomical manuscripts, now dispersed, and presently kept at the Vatican Library (Arabo 319), and the Bibliothèque Nationale de France (arabe 2499), respectively.

In the first instance, the Vatican manuscript clearly illustrates the care with which Postel must have read the contents of the text as we can still see his hand-written Latin annotations on the margins of select pages of that manuscript (see, e.g. Figure 6).10

Figure 6. A marginal note in the hand of Guillaume Postel, in the lower corner of folio 14v, of the Vatican ms Arabo 319, is placed next to the section of the text which discusses the concepts of apogee and perigee. Annotating such elementary concepts, and transcribing, rather than translating, the Arabic word for perigee “Hadhidh” (sic), as is clearly seen, simply means that Postel did not yet have full command of the Greek or Latin terms for perigee, and was probably still learning both Arabic and astronomy through such Arabic texts. Courtesy of the Varican Library

Furthermore, the manuscript itself is a copy of the famous work of Naṣīr al-Dīn al-Ṭūsī already mentioned above, and of course contains the statement of his most famous theorem, the Ṭūsī Couple, which, as we have seen, was of multifaceted fecundity in both astronomy and philosophy, in addition to its own field of mathematics. But on the general level, when one sees a learned contemporary of Copernicus such as Postel, and a professor at the Collège de France no less, still struggling with the Arabic scientific texts that he had bought on his several trips to the Orient; this can simply reflect the mood and the social setting of Renaissance science, vis a vis its main interlocutors from the Islamic lands.

Other manuscripts, including those that were also owned by Postel, contain much more extensive marginal annotations, all indicating the diligence with which this dialogue between the world of Islam and Renaissance Europe was kept up at various occasions and by a variety of scholars. And at every juncture, when those annotations are studied carefully, they always reveal remarks that look like they have been made by a person who was in the process of closely studying the basic contents of the texts. Technical terms, such as the words for trigonometric concepts [Figure 7], when annotated so diligently, must mean that Postel was apparently either studying trigonometry through such texts or preparing a list of concepts to share with his students.

Figure 7. Fol. 3v, of another manuscript that was owned by Guillaume Postel, now kept at the Bibliothèque Nationale de France, arabe 2499, where the interlinear annotations reveal a close reding of such elementary trigonometric terms: “versus” above the word al-ma‛kūs, the last word of the seventh line from the bottom where the phrase al-jaib al-ma‛kūs (versed sine) is mentioned. Similarly he had the term “recto” over al-mustawī (straight), “subtedens” over the term watar ḍu‛f al-qaws (chord of twice the arc), and “absol” [short for absolute] over the term al-jaib al-muṭlaq (absolute sine) for the full diameter of the circle. Courtesy of BNF.

Other pages of the same BNF manuscript (arabe 2449), which is a text on planetary astronomy, are much more extensively annotated, and in one instance a mistake in the original Arabic text is even corrected, all to denote a clear understanding of and an engagement with the text. Figure 8 reveals the extent of those annotations for those who care to see.

Figure 8. Three pages of the Arabic manuscript BNF arabe 2499, exhibiting the extensive annotations on folios 112v-113r, on the left, and the interference with the text by Postel to correct the mistake on folio 124r shown on the right. Courtesy of BNF.

Almost half a century after Postel’s death, this practice of annotating Arabic manuscripts for pedagogical reasons and also for the purpose of studying the science that they contained seems to have spread into England, where the same phenomena can be witnessed on at least one of the Arabic manuscripts that are still kept at Oxford’s Bodleian Library. This particularly important and unique Arabic manuscript, Arch. Seld A 11, is of great scientific value for the early history of Islamic astronomy, and was apparently first owned by John Greaves (1602-1652), the Gresham College professor of Geometry at London from 1631 to 1637, and later the Savilian Professor of Astronomy at Oxford from 1643 to 1649. After Greaves’ untimely death at age 50, in 1652 the manuscript passed on to the possession of John Selden, the great scholar, politician, and supporter of scholars and libraries, and thus became part of the Selden collection of manuscripts now housed at the Bodleian Library.11

The manuscript is a treasure trove of astronomical material containing four important texts, by different authors, from various chronological periods, all bound together in a single volume. It contains the work of the famous ninth century astronomer al-Farghānī, already mentioned before, the work of his contemporary Qusṭā b. Lūqa (820-912) on what could be legitimately claimed as the first extensive cosmological treatise (haya) of Islamic astronomy, an astrological work by the polymath Abū al-Raiḥān al-Bīrūnī (d.c. 1048), and finally the elaborate astronomical work of the early Islamic astronomer Alī b. Sulaimān al-Hāshimī (lived before 929), on the causes of the variations in the early Islamic astronomical handbooks.

It is not surprising that John Greaves, the astronomer/mathematician, would take an interest in this text and that he would study it very closely, leaving his annotations on several of its pages and publishing selections from it, or using it for the purposes of studying other astronomical works.12 In my opinion he probably used this very manuscript to study Arabic astronomy, by using al-Farghānī’s text as an introduction, and later publishing selections from chapter 16 [Figure 9] of this same treatise, in 1652.

Figure 9. fol. 18r of al-Farghānī’s treatise where Greaves annotated such elementary terms as “Zod[iac],” “Aux” for apogee, “epicycli” and even superlinear translations that are hard to see but slightly magnified in the detail from the same page. The last double-pages give Greaves’s publication of selections from al-Farghānī’s text, here showing the beginning of chapter 16.

In the same publication he included excerpts from another elementary astronomical work by Alī Qushjī (1403-1474), but the lion’s share in the publication went to another astronomical work that he owned in a different manuscript attributed to a Shah Cholgi (rl. Malwā 1435-1469), whose name was used in the title of the publication.13

In the Arch. Seld. A 11 manuscript, Greaves made no elaborate annotations on the cosmological treatise of Qusṭā, as he did on al-Farghānī’s and al-Hāshimī’s works, nor did he seem to care much for Bīrūnī’s astrological discussion to warrant his comments. As a practicing astronomer, and more importantly an astronomy teacher, he seems to have devoted much attention to the first and last treatises, namely al-Farghānī’s and al-Hāshimī’s, where in the first he seems to have annotated the basic astronomical terms, much in the tradition of Postel, and in the second made more substantial comments upon its contents [Figure 10]. After all, as we just said, al-Farghānī’s text constituted a good introduction to someone who wished to study Arabic astronomy, while in Hāshimī’s text one could find an almost comprehensive overview and critique of earlier Arabic astronomical works that were the subject of Hāshimī’s commentary. In addition Hāshimī’s treatise contained the much more useful astronomical parameters that would allow one to acquaint himself with a wide range of Arabic astronomical literature. He apparently treated Ibn al-Shāṭir’s astronomical handbook (al-zīj al-jadīd) in the same fashion and annotated it as well.14

Figure 10. Pages 94v-95r, the first two pages of Hāshimī’s text with extensive annotations both in Arabic and English scripts, and a detailed comment on the precession motion of the star Regulus on fol. 95v.

Other texts that were published by Greaves were similar to the text he published under the title of Astronomica Quaedam ex traditione Shah Cholgi, and almost always included explanations of Persian and Arabic astronomical terms, thus giving the impression that he intended to use those texts for educational purposes with his students [Figure 11].

Figure 11. Greaves publication of an anonymous Persian text in which elementary Arabic astronomical terms are explained in Persian, and translated into Latin by Greaves.

If one were to focus on the purely astronomical works of Greaves, forgetting the other works he devoted to antiquarian subjects and pyramidology, one can see a continuous trend, very similar to that of Postel, in that Greaves seemed to be learning astronomy from Arabic and Persian sources and wishing to disseminate this learning to his students at Oxford and elsewhere. He was certainly interested in whatever astronomical parameters he could cull out of those Arabic and Persian sources, either to use them for his own training as a practicing astronomer, or to incorporate their contents in his observational programs that took him to Rhodes, Constantinople, Alexandria and Cairo. When we realize that this activity was taking place in the midst of the seventeenth century, almost a full century after the death of Copernicus in 1543, and at the eve of Newton’s publication of his Principia, in 1687, one cannot contain his amazement at the diligence with which seventeenth century astronomers were still seeking interlocutors in the medieval Islamic astronomical texts.

Greaves was not alone, there were certainly many others, among whom people like Edward Bernard (1638-1696), John Flamsteed (1646-1719) the Astronomer Royal, and later on John Hevelius (1611-1687) and Thomas Hyde (1636-1703) who were also seeking similar information to incorporate in their own works. But with this group of people we begin to witness a merging between the Arabic/Persian astronomical texts and the Latin texts that were being produced at the time. When Bernard sends, for example, a list of astronomical observations that he had culled from various Arabic sources, like the coordinates of a group of fixed stars, to a fellow Arabist such as Robert Huntington, who in turn communicated the information contained therein to the Royal Society, or when he sent the various observational measurements of the ecliptic inclination, to John Flamsteed, the Astronomer Royal himself, who also turned the information over to be Royal Society, the information thus entered the public domain when it was published in the Philosophical Transactions of the society.15 The table of the fixed stars was also reproduced by the Astronomer Royal Flamsteed in the introduction to his ownHistoriaCoelestisBritanica,16 British star catalogue, thus becoming part and parcel of the technical astronomical literature of the seventeenth century.

Some, like Thomas Hyde, even went further and published a full edition of a star catalogue [Figure 12] which he extracted from Ulugh Beg’s astronomical handbook, thus also incorporating the information from this fifteenth century Persian text into the “modern” astronomical literature. The same catalogue was further updated and printed as late as 1917, thus bringing it well within the purview of our own times.

Figure 12. Cover page of Thomas Hyde’s edition of Ulugh Beg’s star catalogue, and pages 98-99 published in 1665, and the cover page of the more recent publication of the same (with updates) by Knobel at the Carnegie Institution in Washington, DC, in 1917. All in Public Domain

On the whole then, there was an active interest in seventeenth century England to incorporate astronomical results that were already obtained from the Islamic world. But as we have seen in the cases of Greaves and Hyde there was apparently also a concerned effort to make Islamic astronomical texts accessible to students of Islamic languages and Islamic sciences alike. The fact that this intercultural scientific dialogue was still taking place at such a late date, and as was said before, in between the Copernican and the Newtonian revolutions is all the more surprising, and a clear testament to the vitality of the impression the Islamic scientific tradition must have made on the intellectual life of seventeenth century England.

In Italy the situation was slightly different, in that Italian cities had had a much longer engagement with the Islamic world, and kept their intercultural dialogue going on well into the seventeenth century as we have seen with the case of Galileo. Galileo’s friend and fellow academician at the Academia de Lincei, Giambattista della Porta (1535-1615), should also be mentioned on account of his direct engagement in the dialogue with the Islamic world.. Della Porta wrote several books, and used Islamic sources quite freely. But in one of his publications, on meteorological and sundry matters, which he called De Aeris Transmutationibus17 [Figure 13] he even went as far as asking one of his friends, a certain Marci Dobeli (or Marco Dobel [Murqus Du’aybili]), a professor of Arabic at a gymnasium in Rome, to compose a poem, in Arabic, chanting the praises of his book. This poem was presumably supposed to give the book a better publicity by associating it with things Arabic. All della Porta’s other books, just like this one, are filled with references to Arabic sources, and direct citations from Arabic sources in Latin translations, all revealing an intimate knowledge of that legacy on the part of this early seventeenth century Neopolitan scientist and academician.

Figure 13. The title page of della Porta’s book Aries Transmutationibus, and Dobeli’s Arabic poem affixed to its front matter.

On the level of astronomical instruments, Italy offers us as well a brilliant example of a dialogue between Europe and the Islamic world that crossed both centuries of time and cultural linguistic barriers and remained artistically and scientifically scintillating. Among the papers of the Florentine architect Antonio de Sangallo the Younger (1484-1546), one of the architects who participated in the building of St. Peter’s Cathedral in Rome, that are now kept at the Uffizi Gallery in Florence, there is a two page drawing of an astrolabe that illustrates this dialogue in quite unexpected ways. One side of de Sangallo’s paper, the more interesting side, carries a drawing [Figure 14], of two copies of the face and back of an astrolabe.18 And because de Sangallo was apparently such a fine draftsman, and wanted to leave nothing unrepresented on the astrolabe he must have had in front of him, he did not only meticulously draw all the curves and Arabic markings that he saw on the astrolabe, but even went further than that to include on the drawing of the back of the astrolabe, the left part of Figure 14, the name of the original astrolabe maker, meticulously copied in what looks like the original kufic script, along the edge of the upper right hand corner. The signature simply said: ṣanaahuKhafīfghulāmAlīb. ‛Isā, i.e. “made by Khafīf the student (ghulām) of ‛Alī b. ‛Isā,” the usual phrase astrolabe makers used to sign their works. In later times, such signatures were either placed in a separate cartouche slightly below the center on the back, or engraved along the lower edge of the astrolabe, again on the back.

This astrolabe is not only unusual in this respect, but is also of such antiquity that it could be used to date the surviving copies of it, now at science Museums, such as the Oxford’s Ashmolean Science Museum. The rich Arabic tradition in biographical dictionaries luckily preserved a record of this astrolabist, Khafīf, and that of his teacher ‛Alī b. ‛Isā, and places them in Baghdad, around the middle of the ninth century.19 The interesting question is why was a renaissance architect of the caliber of de Sangallo copying the astrolabe so meticulously, to include the astronomically uninteresting name of the astrolabe maker, if he did not think that anything that was Islamic and scientific was worthy of drafting with extreme care. The second question is what did de Sangallo intend to do with this drawing? Could he have been thinking that he could take the drawing to an artisan and ask him to execute a brass astrolabe copy for his own personal use? Unfortunately there remains no text or indication to help answer such a question. All we can tell is that a sixteenth century Italian architect was apparently enamored by Islamic material scientific instruments.

Figure 14. A drawing from the archives of de Sangallo the Younger at the Uffizi galleries in Florence, Italy, depicting the front edge and matter of an astrolabe, to the right, and the back, to the left. On the upper right hand corner on the back there is scribbled along the edge of the astrolabe the name of the astrolabe maker, in the phrase, ṣana‛ahu Khafīf ghulām ‛Alī b. ‛Isā (made by Khafīf student (ghulām) of ‛Alī b. ‛Isā), who lived in Baghdad towards the middle of the ninth century. (Courtesy of the Uffizi Galleries, Florence, Italy).

In Germany, or in what was then the Holy Roman Empire, the astronomer and celestial atlas maker, Peter Apianus (1495-1552) had a slightly different, but still very interesting dialogue with the earlier Arabic astronomical sources. In his most intriguing book, the Astronomicum Caesareum,20 [Figure 15] which could double as a scientific instrument on account of the volvelles (the revolving cardboard disks) that were produced with the book for calculating planetary positions, he not only produced a text describing the constellations but included drawings of those constellations.

Figure 15. From left to right: The title page of Aprianus’s Astronomicum Caesareum, the volvelles for the planet Venus included inside the book, the atlas that includes the collective constellation iconography, and a text of the constellation Draco in which Apianus mentions Azophi’s description directly.

The earlier models of constellations that Apianus seems to have been familiar with go back to the tenth century Buwayhid astronomer ‘Abd al-Raḥmān al-Ṣūfī (903-986), mostly known in the west as Azophi, or Azophi Arabus, as he was called by Albrecht Dürer (d. 1528) [Figure 16].

Figure 16. Dürer’s Celestial Atlas, and detail thereof of the lower right hand corner where he has Azophi Arabus, turban and all, in the company of such august astronomers like Aratus and Ptolemy on the upper corners of the Atlas.

This Ṣūfī produced his own book on the constellations which he called Ṣuwar al-kawākib al-thābita (the Images of the Fixed Stars).21 In it he undertook to harmonize two different systems of constellation representations. On the one hand he needed to update and record the Greek tradition of the fixed stars, whose iconographic representations were mostly reflections of Greek mythology. In this tradition one saw the constructed imagery of Hercules, Andromeda, Bootes, Orion, etc, all characters well recognized in the Greek mythological pantheon, and all imagined to inhabit the starry dome overhead. On the other hand Ṣūfī was very well acquainted with a completely different constellation tradition, namely, that of the Bedouin Arabs who had their own imaginary representations that enlivened their own view of the starry night sky. They had their own stories of the daughters of Na‛sh, walking in a curved row behind their father’s beer, thus the constellation called Banāt (daughrters) Na‛sh known from the Greek tradition as the Big Bear (Ursa Major). They also had starry clusters in that same constellation that were not even recognized in the Greek tradition like the three pairs of stars on the feet of Ursa Major, [Figure 17] commonly designated on modern day Atlases with their Arabic names still as Alula (for al-Ūla = first), Tania (for al-Thāniya = second) and Talitha (for al-Thālitha = third) which to the Bedouin Arabs represented the first, second and third, pairs of marks left in the sky by skipping gazelles running in fright from the nearby Lion.

Figure 17. The hybrid constellation of Ursa Major, combining the Greek iconographic tradition of the whole constellation of the Bear, and the three pairs of small stars (Borealis and Australis) on the feet of the Bear, designated from Bedouin iconography with Alula, Tania, and Talitha. (Courtesy Stellarium and Bodleian Marsh 144.)

Ṣūfī wished to preserve both traditions in his book, thus making it one of the most detailed records of a dialogue between the ancient Greek and Islamic civilizations. And to preserve this dialogue, he would describe each constellation with the Greek designations and coordinates first, and then follow that with the tales the Arabs, as he would call them, had spun around the asterisms that were within that specific Greek constellation or in its vicinity.

Apianus did more or less the same thing, almost slavishly imitating Ṣūfī’s style, and he too would give a summary of the Greek description of the constellation and then follow that with what Ṣūfī had to say, without alerting the reader that what he attributed to Ṣūfī were in fact the tales of the Arabs attached to the particular constellation. For example, in the constellation of Draco, [Figure 18], Ṣūfī relates that the five stars on the head of Draco were according to the Arabs representations of four she camels, surrounding a small faint star in their midst, unnoticed by the great Greek astronomer Ptolemy, representing a young camel calf, with the mother camels encircling it in order to protect it from a possible attack by a pack of two wolves and a male hyena represented by the stars on the last curve of the tail. Two of the stars in between, numbered 20 and 21, are referred to by the Arabs as the claws of the wolves.

Figure 18, Ṣūfī’s depiction of the constellation of Draco from a late Persian translation now kept at New York Public Library, and the text of Apianus mentioning “Azophi Arabs (sic for Arabus)” with his five camels and two wolves. Although the usual depictions of Draco in the Arabic manuscripts does not include the small faint star on the head, here marked in red in this Persian version, Apianus seems to know about it and to count it with the “five” camels instead of the clearly larger four.

This reference of Apianus to the work of Ṣūfī does not only reveal an intimate knowledge of that tradition, but also a deeper knowledge of the more ancient layer of the star names attributed to the ancient Bedouins by Ṣūfī. What Apianus decided to highlight from Ṣūfī’s work was that Bedouin layer, and thus many of the star names preserved in his Atlas, and in the text of hisAstronomicumCaesareum, come from that layer rather than from the translation of the Greek tradition. And as Apianus’s Atlas became a source of inspiration for later European astronomers, the Bedouin star names contained therein were incorporated in the later star maps, and thus managed with time to make it into our modern skies, all the way to modern star maps produced in modern day European and American circles. The latest version of this transmission appears in the art design of the current software called Stellarium from which Figure 17 was taken with all the star names marked therein coming from the Bedouin tradition. In this sense then, the star names continue to represent one of the longest lasting dialogues the Islamic world has ever had with the European civilization.

By the time Johan Hevelius (1617-1687) came to draw his own Atlas of the fixed stars, and his depiction of the lunar surface, he was simply an heir to this long tradition of dialogue that had already preceded him with the Islamic world, and he felt that he was particularly indebted to two major astronomers from that tradition, namely, Muḥammad b. Jābir al-Battānī (d. 929) who was also well known to Copernicus and other Europeans under the Latin transliterated name of Alpategnius, and the famous astronomer and central Asian potentate, Ulugh Beg (d. 1449) who built the equally famous Samarqand Observatory, and whose astronomical work was the subject of intense study in seventeenth century Oxford as we have seen. Being enamored by those two astronomers Hevelius decided to include them in the frontispiece to his [Figure 19] FirmamentumSobiescanum (published in 1690). In that frontispiece we see Hevelius depicted with bent knees, presenting his star Atlas to the famous astronomers who had preceded him and including such august figures as Ptolemy, Hipparchus, Copernicus and Tycho Brahe, but also including Ulugh Beg, standing next to Tycho (in the middle of the left row) and Albategnius standing right next to Ptolemy (second from the left in the right row). Urania and her planetary children naturally occupy the center stage.

(a) (b) ©

Figure 19. Frontispieces for Hevelius’s works all depicting his fascination with things Arabic/Islamic: (a) Frontispiece of his star atlas, Firmamentum Sobiescianum, depicting Hevelius with bent knee in the center offering his work to Urania and the august astronomers surrounding her, standing on both sides of her, five on each side, with both Albategnius, second to her left, and Ulugh Beg in the center of the astronomers assembled on her right. (b) In his frontispiece for Prodromus Astronomiae he has Ulugh Beg further elevated to be seated directly to the right of Urania. © The frontispiece for his Selanographia has the banner bearing the title page of his book carried by Ibn al-Haytham (Alhazen) on the left, and Galileo on the right, but both, especially Galileo, dressed in Arab garb

This dialogue with the Islamic world which was apparently pervasive during the European renaissance, had other long lasting dimensions as well. Towards the end of the sixteenth century, when Pope Gregory XIII (1502-1585) wished to reform the ecclesiastical calendar, he resorted to a committee of specialists who knew much about astronomical matters, including the most important parameters of the lunar month and the solar year. The most up to date values this committee had to resort to were the values that its members could cull from Arabic/Islamic sources that by then had refined those parameters to a great degree of perfection. As it also happened, there came to Rome around the year 1577 a Jacobite Antiochian Patriarch of the Syriac church, by the name of Ni‛matallah (d. 1590, known as Nehemias in the Latin sources), who had also brought with him several major scientific Arabic works that were of great use to that committee. This Patriarch was officially appointed to the calendar reformation committee by Pope Gregory XIII, and participated actively in its deliberations. Even when the committee did not fully agree with his opinions on matters of sequencing the Christian feast of Easter, and the Jewish Passover, with which it is historically intertwined, they still sought from him the latest values for the astronomical parameters that he could find in the Islamic sources. The fact that he was using those sources for such purposes is amply demonstrated in a treatise that he penned at the occasion, a copy of which, written in Karshūnī (Arabic written in Syriac script), is till preserved at the Laurentiana library in Florence (Or. 301). Ni‛matallh’s intervention is still with us in the form of the Gregorian calendar which is now almost universally employed.

A good number of the other books that Ni‛matallh had in his possession ended up being used by Ferdinand de Medici, a cardinal and later Duke of Tuscany, in his enterprise of printing Arabic books in Europe through the Medici Oriental Press [Figure 20].

Figure 20. samples of Arabic books published by the Medici Oriental Press. On the left is a page of a reworking of Euclid’s Elements published in 1594 and on the right is the title page of Avicenna’s medical work the Canon of Medicine, published in 1593.

Lest we think that this dialogue was a one way traffic, from east to west, we also have evidence that by the sixteenth century the traffic began to flow eastward from Europe to the Islamic world. The evidence suggested by one particular astronomer, Taqī al-Dīn Ibn Ma’rūf (d. 1586), confirms this trend. This Taqī al-Dīn had ingratiated himself at one point to the Ottoman sultan so that he was given the position of court astronomer/astrologer, and much funds were expended on his behalf for the building of an observatory in Istanbul that he claimed he needed for his work. Much of the personal dealings of this astronomer, his eventual demise and the destruction of his observatory in 1586 is well documented. But what is not well known is his exposure to European culture, as well as the possibility of his being taken captive to Rome for a few years, and such turns in his life which are still subjects of study and much speculation.22 What is certain though, is that this astronomer was apparently a very educated man and must have had a huge library at one time. In this writer’s limited experience with Arabic astronomical manuscripts he has already seen his signature in at least a dozen manuscripts now scattered at libraries all over the world. One of those manuscripts which was similarly signed by him on the cover page contains the famous Arabic translation of Ptolemy’s Almagest. It is now part of the possessions of Tunisia’s National Library (BN 7116 ) [Figure 21].

Figure 21. Title page of a copy of the Arabic translation of Ptolemy’s Almagest, marked with Taqī al-Dīn’s signature on the last line of the diagonal note in the lower half of the page (see detail). This note by Taqī al-Dīn mentions that he had consulted the work of the Italian lexicographer Ambrogio Calepino (c.1450 - c.1510). This single title, with its representation of a Greek text, having been translated into Arabic, and also recording acquaintance with Latin dictionaries represents a multilayered dialogue across three languages and cultures. (courtesy of Bibliothèque Nationale Tunis).

In the note carrying his signature Taqī al-Dīn had this to say in regard to the author of the book in his possession, Claudius Ptolemy, and its title, the Almagest:

“What one finds in all of the Greek copies is Kilawdi, with a ‘K followed by the vowel i’ and a ‘D also followed by the vowel i’ which is a name attribution in accordance with their custom. As for Fulūdhi, with the letters ‘F followed by the vowel u’ and L and ending with the letter Dhal’ that is an attribution to the name of a city [usually confused in the Arabic sources with the city of Pelusium] in which he was supposed to have been born as is said in the Geography. He migrated afterward to Alexandria where he learned philosophy (?)and where he observed, and was thus sometimes related to it by being called al-iskandarī, meaning al-iskandarānī (the Alexandrian). As for the Almagest, it means the greatest in their language (meaning the Greek language), thus I read it in the book of Ambrogio Calepino. Abū al-Raiḥān [al-Bīrūnī] said in his al-qānūn al-mas‛ūdī, that the Almagest is the Syntaxis, and the name Syntaxis is said to mean the well arranged introductions. That is all that I was lucky enough to find about this book. Written by the poor Taqī al-Dīn Ibn Ma‛rūf (in signature form) the observer at the victorious city of Constantinople.”

This signed note does not only reveal the wide range of sources that Taqī al-Dīn was familiar with, but also reveals the various cultural borders of Greek, Italian/Latin, and Arabic that this learned man managed to cross.

In another work, this time of his own composition, that has survived in several copies in European libraries as in Paris BNF and the Bodleian, for example, and was published by Sevim Tekeli in an edition of the original Arabic with a Turkish and English translations,23 he deals with the construction of clocks. Clocks obviously served as necessary astronomical instruments to measure time, and have multiple other societal applications. In this work as well, we are also treated to the wide ranging knowledge of Taqī al-Dīn, and the horizons he was acquainted with and the borders he had crossed, thus representing the multiple dialogues he had carried out. In a note on the quality of clocks in his Islamic lands he says:

“These instruments are difficult to construct and required skilled masters of lowly crafts. In addition, early Islamic society did not exhibit great interest in them, much less in the simpler instruments required for prayer times. And then there began to come to these lands [meaning Islamic lands] some of those instruments, especially those that were made by the people of Lān [normally referring to the mountains of the Caucasus and possibly southern Russia, Tekeli translates as Holland], Hungary, France, and Germany, which were of the utmost precise construction and execution, and of superb beauty and form, together with all that was used by way of gilding of their parts for which much gold was expended. And yet they were made available at low prices. Imitating such instruments required the expenditure of much energy.”24

There is no doubt that Taqī al-Dīn was not only witnessing European artifacts inundating the Ottoman market, especially at the circles of the high court, but was himself apparently fascinated by these new crafts, and took it upon himself to produce a text that would guide the craftsmen of his country to adopt such instruments, and to transform them so that they would become useful for the service of their own Islamic religion in determining the times of prayers and the like. When he addressed in his introduction the purpose of his writing the book, he says: “As for its definition [meaning the definition of the science of clock making] it is a science through which one knows the manner in which instruments for telling time are constructed. Its subject is that of particular motions, in particular objects that last for particular distances, knowing that perpetual motion is impossible in this world. Its purpose is to tell the time of prayers and other such matters, without having recourse to the planetary motions, nor to measuring their heights with instruments, and to tell the night time for the performance of prayers or for the contemplation of matters of government.” 25

In Taqī al-Dīn’s mind these new gadgets that the Europeans were exporting to the Ottoman world, although they were not intended to answer to religious needs, they nevertheless could be deployed for those purposes. He had no doubt that mechanical instruments, such as clocks, could easily cross cultural and religious barriers and be made useful in both cultures. But more importantly, these remarks of Taqī al-Dīn demonstrate very clearly the extent of the dialogue that was taking place across the cultural border between Europe and the world of Islam, this time being sensitive to the filter of the eastward traffic. From his reading the Italian/Latin dictionaries of Calepino in order to verify the name of the Almagest, to his surveying of Greek manuscripts of the Almagestthat he could get his hands on in order to verify the exact spelling of Ptolemy’s name, to his monitoring of foreign artifacts coming into the Ottoman empire in order to modify their usage to fit the local cultural needs, all of these activities speak of the role he was facilitating in the unfolding of this dialogue.


With this quick and illustrated survey of the various shapes and forms through which the dialogue between the Islamic and the European worlds had taken place over many centuries, but most intensely during the time of the European renaissance when Europe was engaged in defining its own cultural identity, one could easily see how porous were the cultural barriers between those two cultures, and how easily very sophisticated and technical astronomical ideas could pass through those barriers. More significantly, one could also hopefully see how difficult it has become for the modern historian of science to separate the scientific legacy of one culture from the other. And yet what is sadly remarkable in this whole story is the ease with which many people nowadays dismiss this intimate interlocking dialogue in preference for seeking the differences that separate those two cultures. Many of those differences do indeed exist, but when weighed on one side of the scale with the multiple points of interdependency and similarities on the other side, the scale is always seen to tip in the direction of more intimate dialogue, and more dependency, so much so that one could see both of those cultures standing along a continuum line with blurred borders in between and at either end.

Notes and References

1 An example of such texts is the Sūrya Siddhānta: A Testbook of Hindu Astronomy, translated by Ebenezer Burgess, American Oriental Society, New Haven, CT, 1860.

2 The Latin translation of this text was published during the latter part of the fifteenth and first half of the sixteenth centuries, Gregg De Young, Farghānī, in The Biographical Encyclopedia of Astronomers, ed. Thomas Hockey et al. Springer, 2007, p. 357, and was also published together with the edition of the Arabic text by Jacobo Golius, Amsterdam, 1669.

3 See Heinrich Suter, Die Astronomischen Tafeln des Muhammed ibn Mūsā al- Khwārizmī, København, A.F. Høst & søn, 1914.

4 See Carlo Nallino, al-Zīj al-Ṣābi’, Al-Battānī sive Albatenii opus astronomicum, Milan, 1899.

5 See for example, Paget Toynbee, “Dante’s Obligations to Elementa Astronomica of Alfraganius,” originally in Romania (Paris), 24 (1895), reprinted in 1974, pp. 413-432 and in Dante Studies, London, 1902, pp. 55-77,.

6 Noel Swerdlow, “The Derivation and First Draft of Copernicus’s Planetary Theory: A Translation of the Commentariolus with Commentary.” Proceedings of the American Philosophical Society 117, no. 6 (1973): 423–512, esp. p. 504.

7 George Saliba, “Aristotelian Cosmology and Arabic Astronomy,” in De Zénon d’Élée à Poincaré, ed, Régis Morelon et Ahmad Hasnawi, Peeters, Louvain, 2004, pp. 251-268.

8 For a succinct statement of this pendulum action described in Plato’s Phaedo, and vehemently opposed by Aristotle, both in the meteorologica and more generally in the Physics, see the superb article of Bert Hall, “The Scholastic Pendulum”, Annals of Science, 35 (1978) 441-462.

9 As quoted by Edward Grant, God and Reason in the Middle Ages, Cambridge University Press, 2004, p. 226.

10 For a detailed study of Postel’s annotations on the two said manuscripts, see George Saliba, “Arabic Science in Sixteenth Century Europe: Guillaume Postel (1510-1581) and Arabic Astronomy,” Suhayl, 7 (2007) 115-164.

11 For more information on Greaves and Selden and their interest in Arabic studies and collection of Arabic manuscripts, see Gerald Toomer, Eastern Wisdome and Learning: the Study of Arabic in Seventeenth Century England, Oxford, Clarendon Press, 1996, pp. 64ff, but more pertinently pp. 177ff. Also see the very detailed study of the astronomical Arabic manuscripts belonging to these Orientalists by Raymond Mercier, “English Orientalists and Mathematical Astronomy,” in G. A. Russell (ed), The ‘Arabick’ Interest of the Natural Philosophers in Seventeenth‐Century England, Leiden: E. J. Brill, 1994, pp. 158–214

12 For more information see Mercier, “English Orientalists,” esp. pp. 162ff.

13 John Greaves, Astronomica quaedam ex traditione Shah Cholgii Persae: una cum hypothesibus planetarum: studio et opera Johannis Gravii nunc primum publicata, Londonii, Typis Jacobi Flesher, MDCLII (1652). Farghānī’s selections are on pp. 90-93 of this publication.

14 Mercier, “English Orientalists,” p. 164. For more on Greaves see also Gregg De Young, “John Greaves Astronomica Quaedam: Orientalism and Ptolemaic Cosmography in Seventeenth Century England,” Indian Journal of History of Science, 39.4 (2004) pp. 467-510.

17 Io Baptistae [Giambattista] Portae Lyncei Neopolitani, De Aeris Transmutationibus Libri IIII, Roma 1610, and republished in 1614.

18 The present author has published a full description, with pictures, of the front and back of de Sangallo’s paper in George Saliba, “A Sixteenth-Century Drawing of an Astrolabe Made by Khafīf Ghulām ‘Alī b. ‘Isā (c.850 A.D.),” Nuncius, Annali di Storia della Scienza, 6 (1991) 109-119.

19 See for example, al-Nadīm, al-Fihrist, ed. R. Tajaddud, Teheran, 1971, pp. 342-343.

20 See for example, Peter Apian, Astronomicum Caesareum, Ingolstadt, 1540.

21 ‘Abd al-Raḥmān al-Ṣūfī, Ṣuwar al-kawākib, Hyderabad, 1953.

22 The latest of these tales, anecdotes, and speculations about this astronomer’s dealings is now recorded extensively in Avner ben Zaken, Cross Cultural Scientific Exchanges in the Eastern Mediterranean, 1560-1660, Baltimore, John Hopkins Press, 2010, passim.

23 Sevim Tekeli, 16'ıncı asırda Osmanlılarda saat ve Takiyüddin'in Mekanik saat konstrüksüyonuna dair en parlak yıldızlar adlı eseri: The Clocks in Ottoman empire in 16th Century and Taqī al-Dīn‘s ―The Brightest Stars for the Construction of the Mechanical Clock, Ankara, Ankara University Basimevi, 1966.

24 My translation, based on the Arabic text in Tekeli, p. 216.

25 Ibid. My translation, p. 218

Case studies

Abū al-Raiḥān Muḥammad b. Aḥmad al-Bīrūnī

(Sep. 4, 973 – c. 1052 or shortly before).

This Central Asian polymath, is one of the very few astronomers of the Islamic world who consciously sought out other astronomical traditions and spent a life time integrating what he learned and reporting about it. In his own right he was also the most original encyclopedists of the Islamic civilization, with a primary interest in astronomy and mathematics. His ethnographic and anthropological works on Indian culture, and on various other cultures that were either extinct or poorly documented make him the best example of a cultural interlocutor with a scope covering several cultures at the same time.

His most important works that stand to epitomize the general character of Islamic civilization as a culture of inter-dialogue are his book on India taḥqīqli-l-hindminmaqūlamaqbūlaal-‛aqlawmardhūla (literally,Verifying All What the Indians Say: the reasonable as well as the unreasonable), and his even wider encompassing book al-Āthāral-Bāqiyaanal-qurūnal-khāliya (Chronology of Ancient Nations, that covers a multitude of other cultures, their feasts, their calendars, their folklore, their astronomical traditions, as well as their myths and legends [see figure depicting the origins of Persian kings]. Both works were edited and translated by Edward Sachau during the 19th century. For relatively complete bio-bibliographies of this astronomer one should consult the entry devoted to him in Encyclopedia of Islam, 2nd edition, 1960, and the more extensive one in Dictionary of Scientific Biography, 1970.

Illustration from Bīrūnī’s al-Āthār al-Bāqiya, depicting the legendary origins of Persian kings, almost parallel to the Biblical creation story. Edinburgh manuscript Or. 161, detail of fol. 48v, photograph by G. Saliba. (courtesy of the Edinburgh University Library)

His encyclopedic astronomical work the inimitable Mas‛udic Canon (al-Qānūnal-Masūdī , was published in Hyderabad, only in Arabic, in 3 volumes. In it he did not only give a systematic overview of what was inherited from the Greek astronomical tradition, but supplemented that with all the innovations and research that was carried out during Islamic times, of course peppered with what he himself had learned from the Indian tradition, to make his work an exhaustive reference for all things astronomical up to his time, and an all encompassing dialogue across the borders of at least three different cultures.

Bīrūnī’s other works are immense. All together he produced some 150 titles, each averaging about 90 folios, that is close to 200 printed pages, and almost half of them were on astronomical and mathematical subjects. Only a miniscule number of his output, almost one seventh 22 / 146, has survived, and even only about half of that has been published.

Notable among these works are: Elements of Astrology (al-tafhīmli-awāilṣināatal-tanjīm) which gives the most comprehensive treatment of the topic, although his personal opinion of it was “as weak as that of the least of its adherents,” as he would put it. This although the majority of people of his time believed that astrology was “the fruit of the mathematical sciences.”

The Determination of the Coordinates of Places for the Correction of Distances between Cities (taḥdīdnihāyātal-amākinli-taṣḥīḥmasāfātal-masākin) is Bīrūnī’s masterpiece in mathematical geography. In it he does not only lay down the observational principles for the determination of longitudes and latitudes of places, but caps that discussion with the vexing spherical trigonometric problem of determining the religiously mandated direction of Mecca for the performance of the five daily prayers, from his own local horizon at Ghazni, near modern Kabul. He uses that occasion to defend the role of the mathematical sciences against the attacks of the religious scholars who could not understand their utility.

His relatively minor works are only minor in their size but equally important, if not more so, in their sophistication. Bīrūnī’s Keys to Astronomy (maqālīdilmal-haya), Pharmacology (kitābal-ṣaydana), Gems (al-jamāhirmarifatal-jawāhir), and a treatise On Shadows (ifrādal-maqālamral-ẓilāl), all deal with particular subjects, but in a perfect Bīrūnī manner, each of them contains extremely original comments on various scientific subjects. In the introduction to his treatise on Gems Bīrūnī gives an elaborate description of man’s place in nature and society, and the social need for economic systems, thereby explaining the role of silver an gold in such economy.

In the introduction to his book on pharmacology he raises the issue of the relative worth of languages. He then posed as an outsider to both Arabic and Persian as he evaluated the scientific utility of those two languages. And in that discourse he enunciated his now famous personal preference to be criticized in Arabic than be praised in Persian.

His Exhaustive Book on Astrolabes (istīābal-wujūhal-mumkinaṣanatal-asṭurlāb), raises another important issue, namely, the possibility of the earth’s motion, as a consequence of a particular case of one astrolabe projection. But he quickly dismissed the whole issue as philosophical speculation, not to preoccupy someone like him whose feet were always on the ground, observing and recording results. The rest of the book is a mathematical tour de force which details all the various projections of astrolabe parts, mainly retes, that were known to Bīrūnī or could be imagine by him.

But on the subject of astronomical cosmology, a subject usually broached by authors of a genre of astronomical literature called, hayatexts, much in the tradition of Urḍī (d. 1266), Ṭūsī (d. 1274), and Ibn al-Shāṭir (d. 1375), to name only a few, Bīrūnī did not seem to have any interest in this kind of theorizing. Only one hint, in a seemingly now lost book, hisIbṭālal-buhtānbi-īrādal-burhān(DisqualifyingfalsehoodbyProducingProof), suggests that he may have approached such speculative cosmological questions. But even then the context in which this hint is mentioned also suggests that he apparently restricted himself to discussing the particular problem of latitude theory in Ptolemaic astronomy. He may not have carried the discussion any further than he did in the case of the cosmologically interesting possibility of the motion of the earth.

Abū al-Ḥusayn ‘Abd al-Raḥmān b. ‛Umar al-Ṣūfī


The great and well deserved reputation of this astronomer rests primarily on his work on constellations, and the frequency with which his work was illustrated in the various extant copies still preserved in libraries around the world. Together with the several Arabic renditions of Dioscorides’s pharmacology that are also usually illustrated those two scientific works have attracted the attention of art historians who studied them for their aesthetic qualities rather than the contents of their texts. [see figure of Orion from two Ṣūfī manuscripts now kept at the Library of Congress and the BNF].

Illustrations of the constellation Orion from Ṣūfī’s work Book on the Constellations of the Fixed Stars (Kitāb ṣuwar al-kawākib al-thābita), here presented together as testimony to their artistic qualities. On the left, from the Libray of Congress Manuscript, SM 16 DLC, p. 134, photograph, G. Saliba, and on the right from BNF, arabe, 5036, 194r (Courtesy of the Library of Congress and BNF).

But the contents of Ṣūfī’s work are far more important for the dialogue they initiated within and across cultural borders. By the time Ṣūfī decided to study systematically the figures of the constellations and to observe the longitudes, latitudes and magnitudes of their individual stars there were two different traditions of star lore that were already well known to his contemporaries: one was the lore of the ancient Arabian Bedouins of pre-Islamic times, and the other was inherited from several Greek sources, mostly via the Almagest of Ptolemy which was rendered into Arabic several times as early as the ninth century, that is, about two centuries before Ṣūfī’s time. Since the iconography of the two traditions were remarkably different, one spoke of desert animals like deer, lions, camels, wolfs and the like, and the other spoke of Greek mythological figures drawn mainly from the Greek pantheon, Ṣūfī decided consciously to bring these two traditions into conversation with one another by identifying the asterisms once in the Greek style and then identifying them in the Bedouin style, thus painting a much richer tapestry in the celestial sphere. And when the Renaissance astronomer Peter Apianus decided to render the contents of Ṣūfī’s work into Latin, he in effect was widening the dialogue that was already started by Ṣūfī and forcing it this time to cross yet another cultural and linguistic barrier. It was across this last barrier that several of the Bedouin names of particular asterisms were adopted by Apianus and from him passed on to modern celestial atlases, giving us our modern Arabic names of the stars that are still in use in modern atlases.

But Ṣūfī was not only interested in harmonizing the two traditions, he was also interested in updating them, in critiquing them, and in setting them back on firm scientific foundations by bringing his own observations and methods of inquiry to bear on the subject. When he found misrepresentations, he would say so. If Ptolemy missed a specific star, Ṣūfī would certainly correct the mistake. In almost every constellation he would have occasion to say this star was not recorded by Ptolemy or the coordinates of this star or its magnitude were so and so in Ptolemy’s work but should be otherwise as per his own observations. In that context he also found occasion to mention the new asterisms that he himself had observed that were either neglected by Ptolemy or were unknown to him. One memorable observation was Ṣūfī’s acute notice of the andromeda galaxy, which is now referred to as galaxy M 31, whose discovery is duly credited to Ṣūfī. He referred several times to this cluster of stars with the term patch (latkha), to mean a cluster of stars, while describing the stars of Andromeda, but not before noting that the same constellation of Andromeda was known in the ancient Bedouin tradition as the fish (al-ḥūt) and that the patch was located on the mouth of the fish [see Figure of Andromeda].

Similarly, when he found the authors of books following the traditional Bedouin asterisms making gross errors in their description, he did not refrain from critiquing them as well, as he did with Abū Ḥanīfa al-Dīnawarī for example. His sole criteria of judgment was always observation, and would blame both the professional astronomers, who followed the tradition of Ptolemy, or the more traditional men of letters who followed the ancient Bedouin tradition, if they did not verify what they wrote with their own observations of the specific star

Figure depicting the constellation of Andromeda with the fish across its body and the cluster of stars, the new galaxy, represented by clustered dots on the lower fish’s mouth [Courtesy Bodleian Library Marsh 144,]