Preface

Sometime between 1028 and 1038, Ibn al-Haytham completed his monumental optical synthesis, Kitab al-Manazir (»Book of Optics«). By no later than 1200, and perhaps somewhat earlier, this treatise appeared in Latin under the title De aspectibus. In that form it was attributed to a certain »Alhacen.«⁑  These differences in title and authorial designation are emblematic of the profound differences between the two versions of the treatise. In many ways, in fact, they can be regarded not simply as different versions of the same work, but as different works in their own right. The underlying point of this observation is so obvious that it can all-too-easily pass unremarked and thus unheeded: translation is not a straightforward conversion-process. To translate, as its Latin form interpretare suggests, is to interpret. Accordingly, the Arab author, Ibn al-Haytham, and his Latin incarnation, Alhacen, represent two distinct, sometimes even conflicting, interpretive voices. The same holds for their respective texts.

To complicate matters, »Alhacen« does not even represent a single interpretive voice. As we shall see in due course, there were at least two translators at work on the Latin text, one of them (Gerard of Cremona?) hewing as faithfully as possible to the Arabic original, the other content with distilling, even paraphrasing, the Arabic original.⁑ Consequently, the Latin text presents not one, but at least two faces to the reader. The Latin text is also markedly different from its Arabic source in organization. To start with, in lacking the first three chapters of book 1 of the Kitab al-Manazir, the Latin text is missing almost half of that book in its original form. This turns out to be a significant omission in terms not only of amount, but also of content, for it is in those three chapters that Ibn al-Haytham sets forth key methodological principles for later discussion. Furthermore, the internal structure of the Latin text—according to chapters, subsections, and even paragraphs—is often at variance with that of the Arabic original.⁑  The analytic flow is therefore not precisely the same in the two texts, a fact that has a significant, albeit subtle, impact upon how the treatise is assimilated by the reader.

The two texts also differ according to lectorial perspective. The conceptual prism through which a medieval Arab scholar would have read the Kitab al-Manazir is fundamentally different from that through which his scholastic Latin counterpart would have read the De aspectibus. To be sure, Arab and Latin were separated by language, but they were separated at an even more profound level by conceptual and cultural differences, some reflected by, some reflecting upon, language. Thus, for instance, the Arabic term sura (»form«) might have been taken in a relatively concrete sense as »similitude« or »image« by a medieval Arab reader, whereas its Latin rendering, forma, might well have been taken in a more abstract sense as »intentional species« by his scholastic Latin counterpart.⁑  Suffice it to say, examples of this sort abound.

All of this is to affirm that in many critical respects the Latin text can be, and indeed should be, analyzed on its own, wholly independent of the Arabic original. I emphasize this point to underline the fact that much of the analysis that follows in this book is slanted specifically toward Alhacen’s De aspectibus, not Ibn al-Haytham’s Kitab al-Manazir. Much of what is said about the former no doubt extends to the latter, but such extension is often incidental rather than essential. Why restrict my analytic scope in this way? For one thing, I am not an Arabist. Any attempt on my part to speak for the Arabic text would thus be presumptuous. For another thing, even were I competent in Arabic, I doubt that I would have much, if anything, worth adding to what A. I. Sabra has already established in typically judicious fashion in his critical edition and English translation of the first three books of the Kitab al-Manazir.⁑

At this point one might question the very need for a critical edition and translation of the Latin text when we already have Sabra’s Arabic edition and translation. We have already addressed this issue—at least partially—by pointing to the fundamentally different interpretive faces the two texts present. The Latin text, in short, is not just a replica, in different linguistic guise, of the Arabic. The same holds for my English translation. As an interpretatio of an interpretatio, it stands entirely apart from Sabra’s. My English translation is thus intended as a complement, not an alternative, to his.

There is yet another possible objection to this critical edition. A perfectly serviceable Latin text is already available in the form of Friedrich Risner’s editio princeps of 1572, a landmark of lateer Renaissance scholarship that has since been reprinted.⁑  What point is there, then, in publishing an essentially redundant modern edition? This objection is blunted in at least three ways. First, Risner’s edition is not at all »critical,« at least not in the proper sense of the term. His primary goal in publishing the De aspectibus was to create an up-to-date version that would appeal to contemporary readers interested in optics. To that end he not only modified terminology and phraseology according to humanist standards, but also restructured the text by subdividing the original narrative into theorematic chunks. Second, although Risner did provide annotation, primarily by interpolating sources and citations into the Latin text, his purpose in doing so was not to place the work into proper historical context. On the contrary, it was to modernize it out of proper historical context. Third, as we will later see, the two manuscripts from which Risner drew his edition fall within the least authentic of three basic family-traditions.⁑

Furthermore, with a critical Arabic text of books 1-3 of the Kitab al-Manazir at last available, it is now possible to compare that text at every level against its Latin counterpart. But any comparison of Arabic and Latin texts on the basis of Risner’s edition would be worse than useless; it would be downright misleading, because Risner transformed both terminology and phraseology in ways that do more to mask than to reveal proper textual links. In light of this consideration, the need for a new, critical edition of the De aspectibus is as obvious as is the insufficiency of Risner’s text to fulfill it.

The need for such an edition gains added urgency from the fact that most of the key derivative works produced by Alhacen’s Perspectivist disciples have come out in complete or partial form over the past three decades. The first to see print, John Pecham’s Perspectiva communis, was published by David Lindberg in 1970.⁑ Since then Lindberg has produced editions of Roger Bacon’s De multiplicatione specierum and Perspectiva.⁑ Meantime, critical editions of books 1-3 and 5 of Witelo’s Perspectiva have appeared between 1977 and 1991, and an edition of book 4 is currently underway.⁑ Finally, at the divide between Perspectivist and modern optics, Johan Kepler’s Ad Vitellionem paralipomena (»Emendations to Witelo«) of 1604 has finally appeared in a critical English translation.⁑ The time is thus ripe—indeed, overripe—for a critical edition of the work upon which all of these were ultimately based.

A final question that might be raised is why this edition is limited to the first three books of the De aspectibus rather than extending to all seven. There are two reasons. First, as I discuss later in the introduction (p. xviii), books 1-3 form a distinct and complete thematic unit. They therefore stand perfectly well on their own without any need of the remaining four books to provide context. The second reason has to do with the length of the De aspectibus, which runs to nearly 200,000 words. A proper edition of the remaining four books of the treatise, which can be subdivided into three thematic units, will be at least a decade in the making. I hope to have the next installment, a critical edition of books 4 and5, ready for publication within the next three or four years.

This edition has been some fourteen years in the making. Over those years I have accrued a sizeable debt to individuals and institutions for their help and support. I therefore wish to take this opportunity to acknowledge those debts and express my gratitude to those who have allowed me to run them up. Let me start at the institutional level. First, and foremost, I wish to thank my home institution, the University of Missouri, for its generous support of this project, support that has come at both the Columbia campus level, through the MU Research Council (1987, 1990-91, 1993, and 1996), and the system level, through the University of Missouri Research Board (1995). I owe thanks, in addition, to the American Philosophical Society, the American Council of Learned Societies, and the National Endowment for the Humanities for their support during the summers of 1989 (APS, ACLS) and 1990 (NEH). Finally, I wish to express my deep gratitude to the National Science Foundation for supporting me during calendar year 1999, while I put the finishing touches on this edition and rendered it into publishable form.

Still at the institutional level, I am pleased to acknowledge the administration and staff of various libraries and manuscript-collections to whose holdings I needed continual access for this project. In particular, I owe a profound debt of gratitude to the staff of the following libraries for making me welcome at various reprises during the past twelve years. Foremost among these are the Trinity College Library, Cambridge; the Crawford Library of the Royal Observatory, Edinburgh (special thanks to Angus Macdonald for his kindness); the Wissenschaftliche Allgemeinbibliothek, Erfurt; the Royal College of Physicians, London; the Bibliothèque Nationale, Paris; and the Bibliothèque Municipale, St-Omer.

At the personal level, finally, I wish to thank the following: Matthew Shaw, not only for his invaluable editorial help, but also his yeoman labor in generating the Latin-English glossary and index; Melinda Lockwood, for her many services in formatting the text and rendering the diagrams importable into that text; Kristi Keuhn for her editorial contributions; Dallas Denery, Bruce Eastwood, David Lindberg, Robert Hatch, Norman Land, and Ann Stanton for their critical reading of various sections of this introduction. I would also like to thank the two referees, Noel Swerdlow and an anonymous reader, for providing me with several useful suggestions about both content and sources. Thanks, too, to Carole LeFaivre-Rochester, senior editor at the American Philosophical Society, for her encouragement and advice over the past year or so as this edition took final shape.