Introduction

History of science is a historical discipline as its name signifies, and as such it is sometimes coloured by historians’ biases. There are critical issues that defied attempts to reach the truth about them such as the first discovery of nitric acid, the discovery of alcohol, the first formulation of explosive gunpowder, the first cannon, how the compass reached Europe, and the question of whether the author of the Latin works of Geber was Jabir ibn Hayyan or a Latin author.

The problem of Geber is more than a century old,[1] and the hypothesis of a Latin author is still strongly entrenched. Even a specific Latin author is recently suggested to replace the pseudo-Geber.[2]

But despite the passage of time and the established status quo of the problem it is the duty of all historians of science not to acquiescence and to continue their search for the truth. History of science cannot be written on the basis of assumptions and conjectures. This paper will summarize the results of a recent research into this problem. It is hoped that it will alert historians of science into critically re-examining this whole question again.

Transmission of Arabic alchemy into the Latin West – a brief review

The works attributed to Jabir ibn Hayyan are very large in number. A considerable part of them was, no doubt, written by him; but it seems that some Arabic treatises may have been written at later dates and attributed to him. These works in their totality are called the Jabirian Corpus and they constitute a major collection of treatises in Arabic-Islamic science. Jabir's works cover nearly every field of learning especially alchemy. Not all of them came down to us. There are however several treatises on practical alchemy attributed to Jabir that became known only in the past few decades.[3]

Some of Jabir's works exist only in Latin and their Arabic originals in the libraries of the West were lost and cannot be located. This is what had happened to the majority of other early Arabic works that were translated into Latin.

The doubts raised by Paul Kraus [4] on the authorship of the Jabirian corpus and the debate that ensued [5] has no bearing on our research into the origin of Geber’s Latin works. It does not matter whether Jabir himself was the author or a later pseudo-Jabir. Similarly the religious beliefs of Jabir (whether he was a Shi’i,or an Isma’ili) do not affect his scholarship and has no impact on our investigation.

Before the translation of the Arabic works into Latin, alchemy was unknown in the Latin West. Robert of Chester finished in the year 1144 the first translation from Arabic of a book on alchemy, Liber de compositione alchimiae  “The Book of the Composition of Alchemy ”, which is the text of a dialogue between the monk Maryanus (Morienus) and Prince Khalid ibn Yazid (d. 704 AD). In the preface Robert says: "Since what Alchymia is, and what its composition is, your Latin World does not yet know, I will explain in the present book." [6]

Until recently, doubts were cast on the involvement of Khalid ibn Yazid in alchemy and on his encounter with Maryanus. Even the translation by Robert of Chester was questioned, and a historian of science like Julius Ruska attributed the Liber de compositione alchimiae to an Italian of much later date.[7]  But these doubts were expelled recently with the discovery of the Arabic original of the Latin work.[8]

Between the first translation of Robert of Chester in 1144, and 1300 some other important Arabic alchemical works were translated into Latin including, the Turba Philosophorum, The Secret of Creation of Apollonius of Tyana, the Tabula Smaragdina, De Perfecto Magisterio attributed to Aristotle, De anima in arte alchemiae attributed to Ibn Sina, De Aluminibus et Salibus and the Book of Secrets, both of al-Razi, and parts of Kitab al-Sab'in (the Book of Seventy) of Jabir. and several other works.[9]

The translation movement which started in the middle of the twelfth century continued into the thirteenth and beyond. One alchemical work, the Liber claritatis, ascribed to Geber, appeared in Latin in the last third of the thirteenth century.[10].During this period the Liber misericordiae (considered to be a translation of Kitab al-rahma) had appeared [11]  Also around the year 1300, the Summa Perfectionis magesterii (Sum of Perfection) bearing the name of Geber, emerged. This book is usually accompanied by three or four other treatises which carry the name of Geber: De investigatione perfectionis (The Investigation of Perfection), De inventione veritatis (The Invention of Verity), Liber fornacum (The Book of Furnaces), and sometimes the Testamentum (The Testament). These discourses were frequently printed together, in whole or in part, between the fifteenth and the seventeenth centuries

The name of Geber became known as a result of these treatises. The Summa, in particular, was so successful that it became the main chemical text book of medieval Europe. This book and the treatises associated with it were particularly notable for their clarity and freedom from mysticism and allegory. Naturally they appealed to practical alchemists and they exerted a great influence on Western chemists until the rise of modern chemistry. The name of Jabir in its Latin form "Geber" became suddenly a most celebrated one.

The emergence  of the Geber problem

Towards the end of the nineteenth century Berthelot came out with a hypothesis that the Latin works of Geber were written by a European alchemist who used the name of Jabir to give importance to his work.. These views were proclaimed in Berthelot’s massive work of three volumes (La Chimie au moyen âge, Paris, 1893). Berthelot was the most influential historian of chemistry in France, and he enjoyed great prestige and authority. As soon as he published his assumptions, they were adopted uncritically by later writers. Burg, in his introduction to the French translation of the Summa says: “nobody, after him, proved the opposite; at least, no one dared to question his frightening scholarship” [12]

Of course this statement is not true. Holmyard, as we shall presently see, had published several articles challenging the assumptions of Berthelot.  Others also like Stapleton and Partington raised doubts as well.. But despite the uncritical adoption of Berthelot’s views by many, his claims could not be established until now on a solid foundation.

The main assumptions of Berthelot

The assumptions of Berthelot are dispersed in the various chapters of his three volumes, especially in volumes I and III, but Chapter X of volume I, (pages 336-50), embody his main hypotheses. Several writers had summarized these assumptions and with the passage of time they were modified by his supporters without deviating from his line of thinking.

For the purpose of this study we have divided the main assumptions into the following categories. This itemization will help us in discussing each assumption independently.

First Assumption

There are no Arabic Originals

Second Assumption

The Latin Works were attributed to Jabir because of his high reputation among the Latins.

Under this assumption we shall discuss another statement of Berthelot:  that no direct quotation of Geber is made by Albertus Magnus or Vincent de Beauvais, (the pre-sumption being that the Latin works of Geber were therefore not known to these two alchemists.

This topic leads us also to discuss the translator of Liber fornacum

Third Assumption

a -In the Arabic treatises of Jabir the language is vague and allegoric.

b -One does not encounter in the Arab works of Jabir precise formulae for the preparation of metals, or salts, or some other substances.

Fourth Assumption

The style of the Summa recalls that of the schoolmen.

Fifth Assumption

The Summa does not contain any Muslim expressions while they are extravagant in Jabir’s Arabic texts.

Sixth Assumption

The Summa contains an account of the arguments of those who denied the possibility of transmutation. Of this one finds no trace of it in the Arabic treatises of jabir.

Seventh Assumption

a- (All the Summa) "is of a firmness of thought and expression, unknown to the former authors, in particular the Arab Jabir.”

b- It is impossible that the scientific character of Summa is that of the ninth century AD.;The state of knowledge and the chemical theories, on the other hand, are those which appeared towards the end of the thirteenth century.

c- No precise doctrines or facts are stated, no character is quoted.

d- While the Arabic texts mention the manifest and the occult doctrine yet another part of these doctrines misses completely and consequently appears to belong to a more modern period. There is no allusion in the Arabic texts to the sulphur and mercury theory of the generation of metals.

e- In the opinion of the author (of the Latin works), there exist three natural principles of metals: sulphur, arsenic and mercury.  This is, actually, a new theory subsequent to the theories of Avicenne. .

Eighth Assumption:

The shorter opuscula of Investigatione perfectionis, Inventione veritatis and Liber fornacum are merely extracts and summaries of the Summa. They reproduce the same preparations and operations with the addition of more modern materials and facts  such as the names of  (1) saltpetre, (2) salt of tartar, (3) rock alum, (4) feather alum; and (5)  the mineral acids (dissolving waters) especially nitric acid. These preparations do not appear in in any manuscript before the thirteenth century or the beginning of the fourteenth.

Also, the Arabic treatises which bear the name of Jabir ibn Hayyan do not mention these materials nor the preparation of nitric acid. These discoveries exist only in the Latin works of Geber, and they appear to be foreign to the Arabic works and subsequent. to them.

Refutations of Berthelot’s assumptions

Since Berthelot came out with his assumptions in 1893, there appeared several criticisms to that assumption by some eminent historians of chemistry and alchemy who were more knowledgeable in Arabic chemistry and alchemy than Berthelot himself.

One of the earliest criticisms was written in 1905 by Henry E. Stapleton who was a celebrated historian of Arabic chemistry and who contributed significantly, through his numerous papers, to the advancement of our knowledge of the subject. [13] We shall have the opportunity to discuss Stapleton’s paper in our present study.

But the greatest amount of criticism and refutation was launched by Eric John Holmyard who was also a noted historian of chemistry with a sound knowledge of Arabic. In a series of papers that appeared between 1922 and 1928 Holmyard dealt thoroughly with each of Berthelot’s assumptions proving their invalidity. [14]  Our work here can be considered as an extension of Holmyard’s work albeit the passage of about 80 years in between. We are reporting here the results of new findings based mainly on the discovery of new Arabic works of Jabir, while referring to Holmyard and others whenever necessary.

Most of the refutations of Holmyard are still valid today and they were not demolished, as was recently assumed, by the publication of Paul Kraus’s work on Jabir ibn Hayyan.[15]

Another celebrated historian of chemistry who sided with Holmyard was James. R. Partington. His paper on the “Identity of Geber” touched upon some important issues that will be discussed in this study also.[16]

Ernst Darmstaedter in an article that followed the publication of the German translation of Geber Latin texts in 1922, wondered  why the Liber fornacum cannot be a translation from Arabic ? [17]. In other articles he alluded to some works attributed to Geber as re-workings by the translators [18]  In one article he admitted  that some parts of the Summa appear, to be from Arabic origin, probably al-Razi.[19].  At that time (1923), Jabir’s works on practical alchemy were not yet known.

After the debate of the twenties had abated, some historians of chemistry  such as Robert Multhauf were not yey much convinced by the hypothesis of the Latin author and they declared in their monographs that the Geber works represent Arabic alchemy and chemistry,[20].  Some others took a neutral position by reporting the Latin works when discussing Jabir and reporting at the same time the assumptions of Berthelot.