Ibn Butlan’s Risalat da`wat al-atibba

Detail from a miniature from Ibn Butlan’s Risalat da`wat al-atibba.

L.A. Mayer Museum for Islamic Art, Jerusalem

Al-Kimiya: Notes on Arabic Alchemy

Historians have uncovered evidence of the immense influence of Arabic alchemy—a largely unexplored piece of the alchemical puzzle.

Note: Arabic words in this article are given in a simplified transliteration system: no graphical distinction is made among long and short vowels and emphatic and non-emphatic consonants. The expression “Arabic alchemy” refers to the vast literature on alchemy written in the Arabic language. Among those defined as “Arabic alchemists” we therefore find scholars of different ethnic origins—many from Persia—who produced their works in the Arabic language.

According to the 10th-century scholar Ibn Al-Nadim, the philosopher Muhammad ibn Zakariya Al-Razi (9th century) claimed that “the study of philosophy could not be considered complete, and a learned man could not be called a philosopher, until he has succeeded in producing the alchemical transmutation.” For many years Western scholars ignored Al-Razi’s praise for alchemy, seeing alchemy instead as a pseudoscience, false in its purposes and fundamentally wrong in its methods, closer to magic and superstition than to the “enlightened” sciences. Only in recent years have pioneering studies conducted by historians of science, philologists, and historians of the book demonstrated the importance of alchemical practices and discoveries in creating the foundations of modern chemistry.

A new generation of scholarship is revealing not only the extent to which early modern chemistry was based on alchemical practice but also the depth to which European alchemists relied on Arabic sources. Yet scholars are only beginning to scratch the surface of Arabic alchemy: a general history based on direct sources still has to be written, and an enormous number of Arabic alchemical manuscripts remain unread and unedited—sometimes not even cataloged—in Middle Eastern and European libraries. This brief survey is offered in hopes of giving Chemical Heritage’s readers a glimpse into this fascinating yet largely unexplored world.

The Origins of Arabic Alchemy

In the 7th century the Arabs started a process of territorial expansion that quickly brought them empire and influence ranging from India to Andalusia. Fruitful contacts with ancient cultural traditions were a natural consequence of this territorial expansion, and Arabic culture proved ready to absorb and reinterpret much of the technical and theoretical innovations of previous civilizations. This was certainly the case with respect to alchemy, which had been practiced and studied in ancient Greece and Hellenistic Egypt. The Arabs arrived in Egypt to find a substantial alchemical tradition; early written documents testify that Egyptian alchemists had developed advanced practical knowledge in the fields of pharmacology and metal, stone, and glass working. The first translations of alchemical treatises from Greek and Coptic sources into Arabic were reportedly commissioned by Khalid ibn Yazid, who died around the beginning of the 8th century. By the second part of that century Arabic knowledge of alchemy was already far enough advanced to produce the Corpus Jabirianum—an impressively large body of alchemical works attributed to Jabir ibn Hayyan. The Corpus, together with the alchemical works of Al-Razi, marks the creative peak of Arabic alchemy.

As is typical in the chain of transmission of ancient knowledge, the origins of alchemy are steeped in legend, and the links of this chain are either mythical or real authorities in the fields of ancient science and philosophy. The doctrines on which Arabic alchemy relied derived from the multicultural milieu of Hellenistic Egypt and included a mixture of local, Hebrew, Christian, Gnostic, ancient Greek, Indian, and Mesopotamian influences.

The presence of the Arabic definite article al in alchemy is a clear indication of the Arabic roots of the word. Hypotheses about the etymology of the Arabic term al-kimiya hint at the possible sources for early alchemical knowledge in the Arab world. One of the most plausible hypotheses traces the origin of the word back to the Egyptian word kam-it or kem-it, which indicated the color black and, by extension, the land of Egypt, known as the Black Land. Another hypothesis links kimiya to a Syriac transliteration of the Greek word khumeia or khemeia, meaning the art of melting metals and of producing alloys.

A third interesting but far-fetched etymology suggests that the word al-kimiya derives from the Hebrew kim Yah, meaning “divine science.” The idea of a connection between the origins of alchemical knowledge and the Jews was widespread among medieval Arabic alchemists, who saw in this etymology a possible confirmation of their belief. These alchemists tended to attribute the mythical origins of alchemy alternately to the angels who rose against God, to the patriarch Enoch, to King Solomon, or to other biblical characters who taught humankind the secrets of minerals and metals. This interpretive strategy dignified the origins of alchemy and attributed alchemical books pseudepigraphically to authorities of the past, providing a safe mechanism for spreading alchemical knowledge, which could otherwise be persecuted for its proximity to magic.

In contrast with the modern term alchemy, the word al-kimiya lacks abstract meaning. Rather than designating the complex of practical and theoretical knowledge we now refer to as alchemy, it was used to describe the substance through which base metals could be transmuted into noble ones. In Arabic alchemical books al-kimiya tended to be a synonym of al-iksir (elixir) and was frequently used with the more general meaning of a “medium for obtaining something.” Expressions like kimiya al-sa‘ada (the way of obtaining happiness), kimiya al-ghana (the way of obtaining richness), and kimiya alqulub (the way of touching hearts) testify to the broad meaning of this word. What we now call alchemy was called by other words: san‘at al-kimiya or san‘at al-iksir (the art or production of the elixir), ‘ilm al-sina‘a (the knowledge of the art or production), al-hikma (the wisdom), al-‘amal al-a‘zam (the great work), or simply al-sana‘a. Arabic alchemists called themselves kimawi, kimi, kimiya’i, san‘awi, or iksiri.

The contribution of Arabic alchemists to the history of alchemy is profound. They excelled in the field of practical laboratory experience and offered the first descriptions of some of the substances still used in modern chemistry. Muriatic (hydrochloric) acid, sulfuric acid, and nitric acid are discoveries of Arabic alchemists, as are soda (al-natrun) and potassium (al-qali). The words used in Arabic alchemical books have left a deep mark on the language of chemistry: besides the word alchemy itself, we see Arabic influence in alcohol (al-kohl), elixir (al-iksir), and alembic (al-inbiq). Moreover, Arabic alchemists perfected the process of distillation, equipping their distilling apparatuses with thermometers in order to better regulate the heating during alchemical operations. Finally, the discovery of the solvent later known as aqua regia—a mixture of nitric and muriatic acids—is reported to be one of their most important contributions to later alchemy and chemistry.

Arabic books on alchemy stimulated theoretical reflections on the power and the limits of humans to change matter. Moreover, we have the Arabic alchemical tradition to thank for transmitting the legacy of the ancient and Hellenistic worlds to the Latin West.

Theoretical Assumptions

The alchemical authorities most often quoted as sources in Arabic alchemical texts were Greek philosophers, such as Pythagoras, Archelaus, Socrates, and Plato. During the Middle Ages, Aristotle himself was considered the authentic author of the fourth book of Meteorologica, which deals extensively with the physical interactions of earthly phenomena, and of one letter on alchemy addressed to his pupil Alexander the Great. Arabic language sources also quoted Hermes, the supposed repository of the knowledge God gave to man before the Deluge and to whom legend attributes the famous Tabula smaragdina (Emerald Tablet); Agathodaimon; Ostanes, the Persian magician; Mary the Jewess (probably 3rd century), for whom the bain marie (akin to a double boiler) is named; and Zosimus of Panopolis (3rd–4th centuries), believed to be the author of an alchemical encyclopedia in 28 books. Indeed, Zosimus is said to have introduced religious and mystical elements into the alchemical discourse: his books meld Egyptian magic, Greek philosophy, Neoplatonism, Babylonian astrology, Christian theology, pagan mythology, and doctrines of Hebrew origin in a highly symbolic writing full of allusions to the interior transformations of the alchemist’s soul.

Arabic alchemists largely worked from an Aristotelian theory of the formation of matter in which the four elementary qualities (heat, coldness, dryness, and moisture) generate first-degree compounds (hot, cold, dry, and moist), which, in turn, combine in pairs, acquire matter, and generate the four elements: hot + dry + matter = fire; hot + moist + matter = air; cold + moist + matter = water; cold + dry + matter = earth. Everything on earth consists of varying proportions of these four elements. A particularly clear explanation of how alchemists made sense of Aristotelian theory can be found in the pseudo-Avicennian treatise De Anima in arte alchimiae (Basel, 1572), an alchemical work probably of Arabic origins that survives only in Latin translation. According to this treatise, every existing body is a compound of the four elements: if a body is defined as cold and dry, this means that the qualities of coldness and dryness predominate, while heat and moisture occur in minor proportions and thus remain concealed. An external cause—either natural or artificial—could generate a change in the structure of the body, rebalancing the natural proportion of its external and internal qualities, thereby changing its appearance. The alchemist in his laboratory seeks to artificially overturn the balance of qualities in the body he is trying to transmute by adding or removing heat, coldness, dryness, or moisture.

Arabic natural philosophy similarly accepted the classical theory of the formation of minerals in mines. This explanation held that two different movements take place in the depths of caves as the caverns are heated by the sun: particles of water (cold and moist) rise to the surface and generate vapors (bukhar) when they make contact with air (hot and moist); particles of earth (cold and dry), however, rise to the surface and generate fumes (duhan). The meeting of vapors and fumes creates quicksilver, if the vapors predominate, or sulfur, if the fumes predominate. Gold is generated when quicksilver and sulfur are pure and in a balanced proportion, and the soil and astral conditions are positive. Imperfections in any of these conditions create metals of progressively lesser value. An impressive description of the formation of metals in caves can be read in Epistle 19, on mineralogy, of the Rasa’il Ikhwan al-safa’ (Epistles of the Brethrens of Purity), a 10th-century encyclopedia of science, religion, and ethics attributed to a group of philosophers influenced by Neoplatonism and Pythagorism.

The alchemist’s goal, to be achieved through study and practical expertise in the laboratory, was to reproduce these natural processes in a shorter period or to interfere somehow with the natural processes to produce “natural accidents.” The alchemist’s knowledge was, therefore, often compared to the creative power of God (for instance, in the 10th-century treatise Rutbat al-hakim, by Al-Majriti) and represented the highest level of knowledge attainable by humans. Yet Arabic alchemists were, for the most part, able to harmonize alchemical doctrines with Islam. The belief in a pure and absolute version of monotheism led Islamic theology to assume the existence of a single creator: according to classical Islamic philosophy, God is the creator of everything that exists and is the direct cause of every action that takes place in the sublunary world. Since only God can create a change—a fasl (differentia specifica, substantial difference)—alchemy, with its aim of changing the internal nature of metals and stones, could have been considered religiously unacceptable. In the 12th century, however, the alchemist Al-Tughra’i proposed an intriguing solution: since nothing can be created unless God wants it to be so, the alchemist simply prepares matter to receive the fasl God will bestow.

Perhaps because of alchemy’s association with divine knowledge, Arabic alchemical treatises persistently appeal to secrecy: alchemists should avoid the transmission of recipes to greedy people whose main aim is to obtain riches rather than wisdom. As would their European followers several centuries later, Arabic alchemists used rhetorical tricks to conceal the secrets of the art from the uninitiated. In the introductory essay to his translation of the first 10 books of Jabir ibn Hayyan’s Kitab al-sab‘in (The Book of the Seventy), Pierre Lory underlines the author’s habit of “scattering knowledge” (tabdid al-‘ilm) by intentionally presenting alchemical procedures out of order so that only the initiated could understand how to read the text. Alchemical authors used a highly enigmatic language, marked by abundant metaphors and technical and allusive terminology, to describe their processes and ingredients. Like the Hellenistic alchemists before them, the Arabic alchemists referred to a metal by the name of the planet that was thought to exert influence over it, so that recipes included Moon for silver, Mercury for quicksilver, Venus for copper, Sun for gold, Mars for iron, Jupiter for tin, and Saturn for lead. Modern readers must bear in mind that even when the names of the alchemical ingredients appear identical to those used in modern chemistry, they rarely designate the same substance.

Arabic Alchemists

Our knowledge of Arabic alchemists has been largely mediated through the voices of their Latin translators, whose works are more likely to have survived to the present day. Scholarly research in this field is still in the preliminary stages, and every new discovery, every new edition of a manuscript, can lead to substantial changes in our perception of the history of Arabic alchemy. Even so, two philosophers have emerged as leading figures.

Jabir ibn Hayyan was born in Tus (in present-day Iran) in 721/2. Besides his Islamic studies, he was well educated in mathematics and science. After settling in the city of Kufa, he became the court alchemist of the Abbasid caliph Harun Al-Rashid (786–809) and was reportedly a close friend of the sixth imam, Ja‘far AlSadiq. He probably died in 803. Given the enormous number of alchemical books that have been attributed to him (more than 300) and the fact that the word jabir can mean “the one who rectifies things,” some scholars have suggested that the Corpus Jabirianum should be seen as the work of a group of anonymous alchemists. Some of the most famous books traditionally attributed to Jabir include Mi’a wa-ithna ‘ashara kitaban (The One Hundred and Twelve Books), which explains how to produce the elixir from vegetables and animals and was supposedly based on Ja‘far Al-Sadiq’s teachings; Kitab al-sab’in (The Book of the Seventy), a rich source for studying the operations and the equipment of medieval Arabic alchemy; Kutub al-tashih (The Books on Rectification), a survey of the progress of earlier alchemists; and Kitab al-mizan (The Book of the Balance), in which Jabir clearly outlines the double aim of his alchemical practice as both the transmutation of bodies in the laboratory and the transformation of his own soul. Jabir’s importance is not limited to the history of Arabic alchemy: numerous translations of his works appeared in Latin, and an abundant pseudo-Jabirian literature was transmitted under the name of Geber.

Muhammad ibn Zakariya Al-Razi was born around 864 in the city of Rayy (in present-day Iran). A versatile mind, he was well learned in mathematics, astronomy, astrology, music, and medicine. In this last field Latin translations of his works— together with Avicenna’s Canon—became the basis of the cursus studiorum for European students of medicine. Tradition holds that he lost his sight as a consequence of one of his alchemical experiments, but in spite of his blindness he was appointed head of the Baghdad hospital, where he remained in charge until his death in 925. His most important and influential alchemical book is the Sirr al-Asrar (the Latin Secretum secretorum, Secret of Secrets), in which he explains alchemical operations in detail and describes the equipment and ingredients needed in a medieval alchemical laboratory in a plain and clear style.

Historians of science would do well to look to the works of Al-Razi rather than Jabir’s highly complex and symbolic Corpus for evidence on how to reconstruct a medieval alchemical laboratory. Al-Razi mentions two groups of instruments: those used for melting metals and those used for preparing other substances. In the first group he lists the furnace (kur), bellows (minfakh), crucible (bawtaqa), double crucible (but bar but, known as botus barbatus to Latin alchemists), spoon (mighrafa), tweezers (masik), scissors (miqta‘), hammer (mukassir), and file (mibrad). In the second group we find the cucurbit (qar’), alembic with evacuation tube (anbiq dhu khatm), receiving matrass (qabila), blind alembic (al-anibiq al-a‘ma), vessel for liquids (qadah), cauldron (marjal or tanjir), and oven (al-tannur), as well as a cylindrical pot used for heating the matrass (mustawqid), different kinds of vessels (qarura), funnels, sieves, filters, and so on. Al-Razi’s clear descriptions of operations have made it possible to identify some of the alchemical procedures referred to in Arabic texts: tadbir is the word used in general for defining the treatment of bodies; sahq indicates grinding, decomposing, and the production of amalgams; hall or tahlil is solution; iqama is the procedure for solidifying; sabk is the fusion of metals; and taqtir means distillation and filtering.

As with the works attributed to Geber, many of the books attributed to Al-Razi— or the Latin author Rhazes—are pseudepigraphical. Given Al-Razi’s wide fame and the general medieval trend to fake the attribution of alchemical books, this should not come as a surprise.

The Legacy of Arabic Alchemy

Today no one doubts that Latin alchemy is mainly based on Arabic heritage. Before the first infiltrations of Arabic alchemical texts, the Latin West knew only a few translations of Greek books of recipes, largely out of context. The history of the influence of Arabic alchemy in the West faces some major problems directly connected with its sources: not all the Latin translations from Arabic are cataloged or identified, their handwritten tradition is scarcely known, and translators’ names are rarely specified.

Translations of complete Arabic alchemical treatises started to appear with regularity in the first half of the 12th century. Robert of Chester, Hugo of Santalla, Arnold of Villanove, Albert the Great, Gerard of Cremona, and Raymond of Marseille dedicated their efforts to the translation of Arabic alchemical treatises by Jabir, Al-Razi, and other known or anonymous Arabic alchemists. By the first decades of the 13th century, Arabic-language alchemical knowledge seems to have been completely absorbed by Latin authors who started to produce original works on alchemy strongly influenced by what they could read in previous translations. Alchemical passages in the works of Albert the Great, Roger Bacon, Michael Scot, and Hermann of Caryntia testify to the degree of assimilation of Arabic-language alchemical doctrines in the West. It was only in the Renaissance that Latin authors, in search of closer contacts with the ancients, started to recreate a line of tradition that reached back directly to the Greeks, skipping over the Islamic world altogether.