Pathfinders

by Jim Al-Khalili

  So why is this incredible Graeco-Arabic movement not a well-known chapter in world cultural history, alongside other similarly seismic events? Baghdad between the eighth and tenth centuries should be spoken of in the same breath as the golden age of Athens during the time of Pericles in the fifth century BCE, or Alexandria of the Ptolemies a few hundred years later, or Renaissance Florence of the Medici family in the fifteenth century. Even if the translation movement were all we have to thank the Abbāsids for, this would and should still be regarded as a major epoch in history. But it was not all we have to thank them for; it marked just the beginning of the golden age. So before we explore exactly what was translated and by whom we have to look carefully at why it took place at all.

  As with the reasons for the eventual decline of the golden age, it turns out that it owes its origins to a variety of factors, not all of them obvious. Many historians are now making compelling arguments for these and are belatedly overturning years of an over-simplistic historiographical view.

  Let us first consider the three standard reasons usually held up as having enabled and brought about the translation movement.

  The first is that it began on the whim of one or two enlightened caliphs, as might be deduced from al-Ma’mūn’s famous dream about Aristotle, which instantly ignited within him a lifelong obsession with Greek scholarship. However, the translation movement began much earlier than the reign of al-Ma’mūn – under his great-grandfather and the founder of Baghdad, al-Mansūr – and was in full swing by the time al-Ma’mūn had his dream. In fact, if the dream story is true, it would have been entirely in keeping with the cultural atmosphere he was an integral part of. So one can more correctly say that the dream was the result of the translation movement and the intellectual climate that brought it about, rather than the other way round.

  The funding of the movement came from right across Baghdad’s society. Along with the caliphate, it included their courtiers, military leaders, officials of state, administrators, even the leading scholars who had become wealthy as they rose through the ranks as translators themselves. During al-Ma’mūn’s reign the most famous of these scholars – men such as Hunayn ibn Ishāq – did not work in isolation, but employed teams of students, translators and scribes.

  Without the patronage and encouragement of the caliphs themselves, the translation movement that blossomed and flourished in Baghdad simply would not have taken place on anything like the same scale. But the enthusiasm and commitment to scholarship of the early caliphs was just part of this wider intellectual movement.

  A second reason for the start of the translation movement that is often heard is the spread of Islam itself; that since it is the religious duty of all Muslims to search for knowledge and enlightenment, this inevitably led them to seek out the secular Greek texts on science and philosophy and have them translated into Arabic. While it is certainly true that early Islam encouraged a general spirit of enquiry and a curiosity about the world that was less evident in Christianity and Judaism, we are still left with the issue of why the translation movement started when it did and not earlier, during the Umayyads.

  Moreover, the translation movement transcended religious boundaries. A large number of the translators were Christians, and they would not have played such an important role if the main motive behind the movement had been religious, following the teachings of the Qur’an or the advice of the Prophet (the Hadīth). The all-important patronage of the movement also cut across society to include non-Muslims. The fact that the teachings of the Qur’an and Hadīth encouraged the seeking of knowledge was certainly a necessary factor in the development of original schools of thought in theology, philosophy and even the exact sciences. But these started somewhat later than the translation movement.

  Which leads us to the third widely perceived origin for the movement; that it was Greek-speaking Christians living in the lands previously part of the Byzantine Empire and well versed in Greek science, who should be thanked for transmitting this knowledge on to the Abbāsids. The reality is somewhat different. The study of the work of Greek philosophers like Aristotle and Plato, as well as some medical and astronomical texts, was, it is true, going on within Byzantine centres such as Antioch and Edessa in northern Syria, where a modest Graeco-Syriac (a dialect of Aramaic, the ancient Semitic language that evolved into Arabic and Hebrew) translation movement was taking place. But it could be argued that these translations tended to be of poorer quality than what was to come later and were not subject to the rigorous demands of accuracy nor the depths of intellectual understanding.

  With the spread of Islam throughout the region, political and religious barriers that had existed between the various sects and denominations became less important. Thus, while religious tensions continued, such as Christian Iconoclasm, Karaite versus Talmudic Judaism and sectarian arguments within Islam itself, Christian and Jewish scholars were now freer to share their knowledge in a spirit of more open collaboration. But this openness of early Islam to other faiths does not entirely explain why relatively few translations took place during the hundred years of Umayyad rule, yet increased dramatically in number as soon as the Abbāsids took over, when many of the most influential and skilled Christian and Jewish translators travelled to Baghdad seeking fame and fortune.

  So, if it was not the spread of Islam, enlightened caliphs or Christian scholars carrying the torch of ancient Greek science and philosophy in the Islamic world, what was it? How, for instance, did al-Ma’mūn know of Aristotle in the first place? More generally, why should the Arabs, those uncultured nomads of the desert, suddenly become interested in Greek philosophy? The answer is that they did not – not until after the start of the translation movement, anyway. It was only then that we begin to see translations of the medical texts of Galen, the philosophy of Aristotle, the geometry of Euclid and the astronomy of Ptolemy. The first translation of a Greek text would often be into Syriac, then from Syriac into Arabic. Subsequent more careful and accurate translations were made directly from Greek to Arabic only as understanding of both the Greek language and the scientific content of the original texts improved.

  What then were the true reasons for the translation movement? Before the arrival of the Abbāsids, we witness what historians refer to as smaller-scale ‘translation activities’ rather than a full-blown movement. These involved the translation of astronomical and medical texts from Indian to Pahlavi in the Sasanian Empire and from Greek to Syriac in the Byzantine, Sasanian and Umayyad empires. Then, around the time of the Caliph al-Mansūr in 754, a quite sudden and dramatic change took place. I believe three important factors helped trigger the translation movement. Not all three happened at once and no one factor was sufficient to explain what happened. But together, they offer a compelling argument.

  Unlike the Umayyad dynasty, whose capital, Damascus, had been part of the Greek-speaking Byzantine Empire, the Abbāsids had moved the whole operation further east into the heart of what had been part of the Persian Empire of the Sasanians. This was no accident. Powerful Persian clans such as the Barmakis and the Nawbakhts had helped them to power and continued to maintain a strong influence in government for many generations. The Abbāsids, in turn, needed the support of this Persian nobility and encouraged the interweaving of Arab and Persian cultures and identities.

  But Arabic was now the official language of the empire, and there was immediately seen to be a need for translation of Pahlavi texts into Arabic, and full caliphal support was offered to the project. Some of these texts were Persian in origin; others, such as many medical, mathematical and astronomical works, had been originally translated into Pahlavi from Greek and Indian and were in use in cities such as Gondēshāpūr (called Jundaysābūr in Arabic).2 So the first and most important factor in bringing about the translation movement was this Abbāsid obsession with Persian culture. This was typified by one translator who, when asked why he searched for Persian books to translate into Arabic, is supposed to have replied: ‘we [the Arab
s] have all the words, but they [the Persians] have all the ideas.’3 This need for translation was at the outset almost entirely for practical purposes; it had to be seen as useful and necessary.

  This is where the second factor comes in: an obsession with astrology. Sasanian ideology, based on Zoroastrian myth, was very appealing to the Caliph al-Mansūr, who developed a deep personal interest in astrology. He also knew he needed the support of the influential Persian aristocracy, most of whom were still Zoroastrian and had not converted to Islam. His interest in astrology, therefore, while wholehearted and genuine, can also be seen to have a shrewd political motive. Astrology, as distinct from astronomy, was embedded in Persian culture and played a fundamental role in Persian daily life, in stark contrast to the Arabs, who saw it as being against the teachings of Islam with its links to fortune-telling and divination. But such was the influence of Sasanian culture on the Abbāsids that astrology underwent a revival in the second half of the eighth century. Astrologers were employed in the caliph’s court to cast horoscopes, offer advice and glorify his achievements. We have already seen how al-Mansūr employed his three top astrologers to advise him on the right day to start building his new capital.

  It is not surprising therefore that the first ‘scientific’ discipline to be systematically translated from Pahlavi into Arabic was astrology. One of the earliest texts was the hugely influential five-part astrological work of the prophet Zoroaster, The Book of Nativities, which was first translated into Arabic between 747 and 754.4 Astrology, the art of plotting the positions of the stars for horoscopes, had become a perfectly acceptable branch of knowledge. Known as ’ilm al-nujūm (the science of the stars), it was indistinguishable from mathematics and astronomy (’ilm al-falak) and those interested in astrological texts were keen to get hold of star charts and mathematical tables. So this early Abbāsid interest in astrology naturally led to the search for astronomical texts already available in Pahlavi or still in Sanskrit, the language of the Indian mathematicians and astronomers.

  Al-Fazāri, one of the astrologers who advised al-Mansūr, and the man credited with building the first astrolabe in the Muslim world, is also associated with translating several astronomical texts from Sanskrit into Arabic. It has even been claimed that he was the first to translate the Siddhanta, written by the greatest Indian mathematician and astronomer, Brahmagupta (598–668), into Arabic. This was arguably the very first time that the Abbāsids encountered Hindu astronomy, but the date and authorship of the translation are uncertain owing to confusion surrounding several contemporary scholars, all called al-Fazāri.5

  The word Siddhanta is a Sanskrit term meaning ‘The Doctrine’ or ‘The Tradition’. It was originally written, as was the tradition among Indian mathematicians, completely in verse. But frustratingly, Brahmagupta provided no proofs of the many mathematical theorems it contains. The Arabic translation of this text is known as the Sindhind and, along with Ptolemy’s Almagest and Euclid’s Elements, it was destined to have a huge influence on the scholars of Baghdad. It appears likely that the Siddhanta had originally been translated into Pahlavi, possibly in the Persian city of Gondēshāpūr, which had been a great centre of Sasanid scholarship. The book contained not only tables and star charts, but mathematics and crude trigonometry. However, it was notoriously obscure and difficult to follow.

  There is a dubious apocryphal story which, if true, not only dates the translation of the Sindhind to the time of al-Mansūr but explains the reason for its lack of clarity. The story describes how Arabs first conquered then settled in the land of Sindh (today one of the provinces of Pakistan) in the early days of Islam. When the Abbāsids came to power these settlers seized the opportunity to declare themselves independent. But al-Mansūr would not tolerate this and sent an army to quell the uprising. After his victory, a delegation from the defeated Sindh came to Baghdad. In the party was an Indian sage named Kankah, who spoke no Arabic or Persian, and his speech, describing the wonders of Indian astronomy and mathematics had to be translated first into Persian by an interpreter, and then into Arabic by a second interpreter, a process which rendered the final form of his instruction very involved and abstruse. What he had been describing was Brahmagupta’s Siddhanta.

  Later Islamic scholars such as the polymath al-Bīrūni in the eleventh century dismiss this story as highly unlikely, claiming that a more likely scenario was that the Sindhind was a translation of a Persian version already in use in Gondēshāpūr. The only likely truth in the story is therefore that the Siddhanta did indeed pass through two translations on its way to the Arabs.

  It is not until well into the ninth century that we see among Islamic scientists and philosophers an emerging confidence in a new rational and scientific world-view that led many to criticize astrology as not having a place alongside true sciences like mathematics and astronomy. Some, however, continued to dabble in it, including the mathematician al-Khwārizmi. Others, even centuries later, would recognize its importance in convincing their less-enlightened rulers to continue funding their astronomical projects. One such scholar was the Persian al-Tūsi, who had to feign an interest in astrology to persuade the Mongol ruler Hūlāgū Khān to fund his new observatory in Marāgha in north-west Persia in the mid-thirteenth century.

  But back in the eighth century it was this early widespread obsession with astrology that helped ignite interest in translating the predominantly Greek works in the other sciences.

  The third factor that played a role in establishing and accelerating the translation movement was the serendipity of emerging technologies. A knowledge of subjects like geometry was required for engineering projects6 such as arched stone bridges, waterwheels and canals; accurate astronomical data were needed to predict the phases of the moon for timekeeping; and arithmetic was vital for accountancy. All these played an important role; but all would have been equally important to earlier civilizations and so do not explain the sudden acceleration in the volume of translations. However, it was the arrival of one technology in particular that made all the difference in the world.

  The first paper mill in the Abbāsid Empire was built in the city of Samarkand in Central Asia on the silk route between China and the West. The city was already one of the greatest in the Persian Empire many centuries before the Islamic conquest and was to continue as a centre of learning and scholarship well into the Middle Ages. The Muslim army defeated the Chinese in 751 on the banks of the river Talas several hundred miles north-west of Samarkand in modern Kyrgyzstan. This Abbāsid victory marked not only the furthest expansion of the Chinese Tang dynasty westwards but the furthest east into Asia that the Islamic Empire would venture. Pertinent to our story is that among the Chinese prisoners of war were those who had knowledge of papermaking – an invention of the Chinese in the second century CE – who were taken back to Samarkand. There, their knowledge was crucial in the building of the first paper mill, helped by the abundance of raw materials like flax and hemp crops in the region. The first paper mills in Baghdad began to appear in the last decade of the eighth century.

  Parallel with this was a rise in technologies associated with the production of books: the development of dyes, inks, glues, leather and book-binding techniques,7 all of which exploded on the scene within a very short time. Paper quickly became far cheaper as a writing material than papyrus and parchment, and multiple copies of texts would often be produced by a whole team of scribes working side by side.

  Prior to this, the codex (in which sheets are bound together as books between covers, usually wooden) had replaced the scroll much earlier than the invention of paper itself. First used by the Romans and Hellenistic Greeks, codices were originally made of papyrus and parchment. In fact, it is claimed that during the Prophet Muhammad’s lifetime, the pages of the Qur’an were kept as a codex between wooden boards.

  So the translation movement owes its beginnings to the appeal of Persian culture, and astrology in particular, to the Abbāsids, along with the development of paper-makin
g technology they had learned from the Chinese. But once it began, this obsession with translating ancient texts sparked the beginning of a golden age of scientific progress.

  A sharp increase in the number of translations took place during the reign (786–809) of Harūn al-Rashīd. Medical, astronomical and mathematical texts began to be translated from Greek, Syriac, Persian and Indian. But at this relatively early stage, scholars were careful in their choice of manuscripts to translate. The importance of any scholarly scientific work is said to be measured by the extent to which it renders all earlier ones on the same subject superfluous. It was quickly realized that many Persian works on science were themselves translations from the original Greek. Soon those Greek originals were being sought. By this time, Islamic scholars and patrons of the translation movement had moved on from an interest in purely practical subjects like astrology, medicine and agriculture to mathematics and astronomy. What is missing from this early list of disciplines is philosophy, as exemplified by the work of the two Greek giants: Plato and Aristotle. The later translation movement from Greek and Arabic into Latin began with philosophy because of a desire to understand these great works. But interest in this field among Islamic scholars came relatively late in Baghdad’s translation movement and finally took off for a quite touching reason: Muslim scholars felt somewhat lacking in their reasoning and debating skills on matters of theology alongside their Christian and Jewish counterparts already familiar with Aristotle and Plato and therefore more experienced in such logical disputations.