Lines and letters. Diagrams in Arabic scientific manuscripts

In mathematical works, diagrams are as instrumental as the text in conveying the content of propositions and demonstrations. Nevertheless, scholars usually give priority to the text, leaving the diagrams unstudied. But the tide is turning. Thanks to the work of pioneers, diagrams are finally beginning to receive the attention they deserve, and a new approach based on systematic and rigorous investigation is gradually gaining ground. Awareness of the centrality of diagrams has developed more rapidly and vigorously in the field of ancient mathematics, whose experts have already provided important contributions. As far as the Arabic tradition is concerned, the process has only just begun; the field still needs to be properly defined in terms of content and methodology. Within this general framework, the present paper, aims to provide new insights into the diagram phenomenon in the tradition of Arabic scientific texts. To this end, I have chosen the second chapter of the Kitāb nūr al-ʿuyūn wa-ǧāmiʿ al-funūn as a case study, analysing its diagrams on Euclidean geometrical optics by means of a codicological and palaeographical methodology. In an attempt to identify significant and problematical features, my main goal has been to study the diagrams in their individuality, and to find out what they can tell us about the process of copying an Arabic scientific manuscript. In entering this moving field, I will try to show that codicology and palaeography can contribute intriguing evidence worthy of further investigation.