Grafton himself comes unnervingly close to this ideal. At the very least, he possesses an acute sense of how well-read people have responded to books throughout the past five centuries. Grafton's specialty is understanding how scholars and intellectuals dealt with the information overload spawned by Johannes Gutenberg. In recent years, historians have done a great deal of research into how books are published and circulated. Robert Darnton's Forbidden Best-Sellers of Pre-Revolutionary France -- with its account of the Enlightenment-era pornography industry -- is one of the more sensational examples. While the focus of Grafton's work is not as obviously titillating, in such works as Forgers and Critics or the improbably lively history of the footnote, he chronicles how certain books maintain -- or lose -- their centrality as authoritative sources of information about the world.

Five hundred years ago -- as Grafton writes in New Worlds, Ancient Texts -- there was a vogue for "comprehensive books that described the world as a whole, summing up between covers all intellectual disciplines and their results." Some of these encyclopedic references were written by contemporary authors, but, in general, the older a book, the more reliable it was considered to be. Even contemporary volumes were often attributed to classical scholars in order to bolster their credibility. It wasn't until Columbus and Galileo made discoveries not anticipated by the ancients that age itself stopped being a marker of credibility, and people looked to the new -- as purveyed by scientists and explorers -- for reliable information about the universe.

The tales that Grafton tells are about the great philologists who straddled this divide and precipitated a new way of understanding the texts that had been inherited from previous generations. Scholars such as Joseph Scalinger and F.A. Wolf were among the first to see classical works as products of another era, rather than as road maps to the cosmos. This sense of history might even be a spur to creativity. Grafton's new volume on Alberti shows the Renaissance athlete, architect, and writer carefully studying the geniuses of the past -- in order to surpass them. To be sure, most of these men make improbable candidates for heroic stature. The awesomely learned Isaac Casaubon proved that the "ancient" writings of Hermes Trismegistus were frauds; but today he is remembered only for his namesake, the bloodless old bookworm that Dorothea marries in Middlemarch. Yet Grafton is charting a crucial shift in attitude. Once no single text could be considered definitive, reading became an active process: a form of passionate attention that makes writing answerable to the critical mind. It's not an irreversible kind of progress, either; as we seek out ever newer information, we get farther away from that moment of possibility when old texts were open to question but still part of a shared intellectual culture. How much difference is there, really, between revering old books simply because they're old and ignoring them for the same reason?