All authors communicate in the intangible medium of language. Old authors, whose works have survived as manuscripts, present special problems of communication. Although the exact works of these authors may not be recoverable, documentary texts of their works, that is versions that have survived in manuscripts, may be studied and manipulated in order to reconstruct a reasonable approximation of them. This enterprise, although by nature inexact, is by no means so uncertain as to allow total confusion. Furthermore, although one cannot be always certain that one has re-established the exact words of a Virgil, a Shakespeare, or a Sa‘di from the surviving textual evidence of their works, this absence of certainty should neither lead to despair nor to doubt as far as the essential validity of the enterprise is concerned. Textual criticism of classical texts, as I have pointed out elsewhere, is much like paleontology. One may not be able to reconstruct the precise form of a dinosaur or an early primate from scant fossil evidence, but one can make a pretty good guess about the form based on that evidence. Therefore, the uncertain nature of the fossil evidence invalidates neither the inferences drawn from it, nor for that matter the discipline of paleontology. By the same token, although we may not always be able to determine the exact form of a classical Persian poem from its manuscript evidence, it is not true that we can therefore say nothing worthwhile about it.

As textual scholars, we are more fortunate than the paleontologist. Those who work with fossils of flora and fauna know that one tyrannosaurus rex is anatomically pretty much like another. But those of us who work with the creative activity of literary geniuses, by contrast, have the luxury of knowing that the object of our study is unique.