MAY 2006 :: EDUCATION

In Search of History
Online Resources Are a Treasure for Scholars, but They Pose Risks, Too

By John Letzing
Dow Jones Newswires

Roy Rosenzweig thought he had unearthed a treasure. While clicking through the Internet five years ago, the George Mason University history professor found a letter posted on a Web site that was purportedly written in 1829 by Martin Van Buren, who would become president eight years later.

In the letter, Mr. Van Buren-whom historians view as a champion of free markets-appeared to urge then-President Andrew Jackson to shift policies and start regulating "a new form of transportation known as 'railroads.'"

For a historian, such a letter was like gold. It seemed to reveal a whole new side of a political figure whom academics thought they had pegged long ago. An excited Mr. Rosenzweig spent days with his colleagues trying to further decipher the importance of his find.

After more study, however, Mr. Rosenzweig realized that the letter could not be genuine. Subtle but incriminating errors, such as the word "unemployment," a term not used until much later in the 19th century, gave the ruse away. "People who know history" could tell the letter was a fake, he says.

Mr. Rosenzweig's discovery illustrates just one of the thorny issues facing scholars as they explore the vast new resources available on the Internet. Their options multiply daily, as Internet titans, libraries and universities team up to convert new troves of printed information into digitized, searchable form. But so do the pitfalls of relying too heavily on Internet-based research.

Gold Mine

Many of the emerging online resources are a gold mine for researchers. For instance, there's Google Book Search, a project that is scanning and posting online tens of millions of volumes from the libraries of Oxford, Stanford and Harvard universities, the University of Michigan and the New York Public Library. Similarly, Amazon.com has its "Search Inside" program, which allows anyone to search and read millions of pages from books on its site. Yahoo and Microsoft, meanwhile, are part of the Open Content Alliance, a project scanning material from sources including the university libraries of California and Toronto and the British national archives.

Scanned books represent only part of the online materials that researchers are finding so valuable. While scholars once had to troll through stacks of academic journals, now they can search through huge numbers of publications instantly. Generalized tools do this, such as Google Scholar, an index of various published academic work, and more specialized sites, such as arXiv.org, a repository of physics papers maintained by Cornell University.

More exciting is the growing amount of primary sources such as journals, letters, photographs and other original documents. Mr. Rosenzweig says his ability to sift through 19th-century newspapers digitized and indexed at ProQuest.com has shaved months from the time it used to take him to complete research projects.

Historical material heading onto the Web isn't all documents and images, either. Last November, 5,000 digitized wax-cylinder recordings dating back to 1895 were posted online by the Cylinder Preservation and Digitization Project at the University of California at Santa Barbara.

Geoffrey Bowker, who studies communication history at Santa Clara University, says a Web site he has found "enormously useful" is VictorianWeb.org, a database about Britain's Victorian era, where he has found material ranging from poetry to contemporary ads.

Mr. Bowker, however, does see dangers on the digital frontier. One that he mentions: the risk of shrinking horizons.

"If you have a digital catalog with 75% of the books in it in the field, and a card catalog with 100%, people will still choose the digital," Mr. Bowker says. The untapped 25% gets pushed further into the abyss. "There's material that used to be looked at that's not being looked at," Mr. Bowker says. "Many of us are hoeing the same kind of territory now."

Other scholars say services like Google Book Search will ultimately serve to broaden knowledge. "Stuff that's been lost from the foreground of understanding will be [brought] back," says John King, dean of the School of Information at the University of Michigan.

At Northwestern University, Pablo Boczkowski, an associate professor of communication studies, says that an Internet temptation his peers fall prey to is a simple failure to get up out of their chairs. "Faculty don't go to libraries very often anymore," Mr. Boczkowski laments. What they therefore miss out on, he says, is the serendipity of accidental discoveries: riffling through sources in search of one thing, and finding unexpected gems.

It was purely by chance, he says, that he happened on the topic for his first book, titled "Digitizing the News," a study of ways newspapers use technology to develop new products.

Peer Review

Fred Turner, an assistant professor of communication at Stanford, complains that what search engines do find is listed in order of the most heavily used sites first, not by what is most relevant. This "doesn't work as well in a scholarly setting," he says, adding that hits listed on Google Scholar do not quite match his own understanding of what is important in his field.

Another danger Mr. Bowker sees is the Web's potential harm to the peer-review process. This time-honored ritual, whereby scholars judge one another's work without knowing who the author is, is essential for advancement in academic circles. Shielding authors' identities prevents rivals from gratuitously trashing their work. But using a search engine, Mr. Bowker says, it's easy to enter a few key phrases from the work under review and discover who the author is. That taints the anonymity of peer review.

There is also the potential to be led astray. The bogus Van Buren letter, for example, seems to have first appeared in the early 1980s in a newspaper ad for coal-industry interests in the Washington Post. The fraud was quickly discovered and was reported by the Post itself.

But according to Mr. Rosenzweig, when he found the letter posted on a Web site, news about the hoax hadn't yet made it online, at least not in any widely accessible form.

No harm done, in this particular case. Indeed, the Web makes it possible for "group scrutiny" to provide a kind of self-policing function, he says. Mr. Rosenzweig continues to believe strongly that the burgeoning amounts of research material available online should be welcomed, despite the dangers.

"Even if not everything is good, it helps the aggregate get better," he says. Today, he adds, if one performs a Google search with the terms Van Buren, Jackson and railroads, the first two results that come up are exposes on the fraud.