We still have to determine what was on al-Nadīm's mind when he connected Khālid's translations of alchemical books and the administrative translations of the dīwān to the spread of philosophical and scientific books in the Islamic civilization. What was the connection between the translation of the dīwān and the translation of books on philosophy and science? If we are to gain some insights regarding these questions we must pursue the subtle hints that were already supplied by al-Nadīm himself.
The reason why those hints do not readily seem to connect the dots for us between the translation of the dīwān and the philosophical and scientific texts, and thus have deprived us so far from appreciating the real input of al-Nadīm on this matter, is to be sought in al-Nadīm's particular use of the word "dīwān" in this account. The term itself was also used in several earlier and later sources without ever specifying what was meant by it. The word is still used in modern Arabic, but has now come to designate a completely different entity, such as a government office (e.g., dīwān al-muḥāsaba) or a personal royal office (e.g., al-dīwān al-malakī). In some sense the word is at times still used in the classical sense when it referred to administrative offices that handled the affairs of the army as in dīwān al-jaysh, taxation bureau as in dīwān al-Kharāj, chancellery as in dīwān al-rasā'il, etc. If we restrict ourselves to those common meanings of the word, we then find it difficult to connect such government offices to the translation of philosophical and scientific books.
But when we return to the story of the arabization of the dīwān itself, we find that both al-Nadīm, in his Fihrist, and the earlier tenth-century author al-Jahshiyārī (d. 942) in his kitāb al-wuzarā' wa-l-kuttāb,[123] both tried to guide us to the correct meaning of the dīwān by giving us examples of the kind of activities they knew were taking place in it. The only example that they give, which has been slightly distorted in al-Nadīm's version that has come down to us, denotes that both authors intended the dīwān operations to mean the dīwān accounting procedures that Zādān Farrūkh was bragging about when he claimed that he was the only one who could carry them out. On the basis of that specialized knowledge he could assert that al-Ḥajjāj needed him more than he needed al-Ḥajjāj. The example of the kind of accounting both authors give obviously required handling arithmetical operations carried over fractions and the like, the kind of arithmetic that is still slightly complicated by our modern-day standards. Therefore the dīwān that needed translation was the dīwān in which such complicated operations were performed, and not as most people thought the government office in which records of personnel and their salaries were kept.
The second kind of dīwān, where salaries were meted out, did not need any translation for it was in Arabic in the first place. We are explicitly told by al-Jahshiyārī: "There were always two dīwāns in Kufa and Basra: one in Arabic, in which records of people and their grants were kept, that is the dīwān that was instituted by Umar [b. Al-Khaṭṭāb], and the other was for the purposes of revenues (li-wujūh al-amwāl) which was in Persian. The situation was similar in Syria, where there was a dīwān in Greek and another one in Arabic. Matters persisted in this fashion till the days of 'Abd al-Malik."[124]Therefore, the dīwān that al-Nadīm was talking about was the dīwān of revenues, and revenues were the backbone of any government then, as now.
Since operations dealing with revenues required arithmetical operations which in their turn necessitated at least other elementary operations such as the surveying of real estates, and the re-surveying when estates were passed on as inheritance, a dīwān officer, as a revenue collector should have the qualifications to carry out those procedures. Furthermore, the computation of time in solar years, when taxes should be paid, and as we know solar and lunar years are not always easy to coordinate without at least some elementary astronomical knowledge, that too must have forced the dīwān officer to learn some astronomy. Similarly, re-apportioning payments, especially after the distribution of inheritance, digging canals, trading, etc., all necessitated that the said officer acquire such operational skills for which Muḥammad b. Mūsā al-Khwārizmī had to compose a complete book on Algebra just for that same purpose.[125] Incidentally, that requirement seems to have led to the creation of the discipline of Algebra qua discipline,[126] which was not known to the Greeks in the fashion that was articulated by al-Khwārizmī.
All the operations a dīwān officer was supposed to perform were not easy, and there must have been some elementary texts or manuals that were used to train those who worked in the dīwān. It is rather unfortunate that no such documents seem to have survived from this early period, probably because they were thought of as simple enough to be learned and discarded, or because their contents were orally transmitted from father to son, and thus there was no need to publish them to the public. But we do have some slightly indirect information about their contents, and the kind of operations that were required in these dīwāns. For we do find in the work of Ibn Qutayba (d. 879), who preceded Jahshiyārī by a half a century and al-Nadīm by almost a full century, and who himself was a contemporary of the last period of translation that followed the translation of the dīwān, a short synopsis of the qualification of those who sought employment in the dīwān, or those who were then called kuttāb. Those kuttāb were undoubtedly the heirs of the dīwān employees whose functions we are now seeking.
In his book Adab al-kātib, he regrets in the introduction the neglect that had become the share of the Arabic sciences of his time. Ibn Qutayba went on to stress that the kātib must seek the following sciences, if he were to be worthy of the name kātib, and not be among those who are after the office of kātib in name only:
He must — in addition to our books — investigate matters relating to land surveying, so that he would know the right angled triangle, the acute, and the obtuse angled triangle; the vertical plumb lines (masāqiṭ al-aḥjār), the various squares (sic), the arcs and the curves, and the vertical lines. His knowledge should be tested on the land and not in books, for the one who reports is not like the eye-witness. And the non-Arabs ('ajam) used to say: 'whoever was not an expert in matters relating to water distribution (ijrā' al-miyāh), the digging of trenches for drinking water, the covering of ditches, and the succession of days in terms of length increase and decrease, the revolution of the sun, the rising of the stars, the conditions of the moon when it becomes a crescent as well as its other conditions, and the control of weights, and the surface measurement of the triangle, the square, and the polygons, the erection of arches and bridges as well as water lifting devices and the norias by water side, and the conditions of the artisans and the details of calculations, he would be defective in his craft.[127]
125
Khwārizmī says that he composed his book on Algebra in order to answer to the need of men who "constantly require in cases of inheritance, legacies, partition, lawsuits, and trade, and in all their dealings with one another, or where the measuring of lands, the digging of canals, geometrical computations, and other objects of various sorts and kinds are concerned..."