From John Hennessy:
As any of you who have the historical hound dog’s desire to hunt know, the world of online research is expanding before our eyes. It’s an exciting time in some ways, as the source material available to us grows every day. I confess I have reveled in the chance to plow through the dozens of now-online newspapers from the Civil War period–papers that I have never seen before. I have learned a few things along the way–most notably that it’s not long before you strike a point of diminishing returns: the source material pours forth, but what it tells us that’s new narrows (the entire field of military history as it relates to the Civil War suffers so). I have found thousands of wartime letters in the last few months, and while some are highly quotable, it’s a rare day that I find something that really goes beyond the interesting to tell us something new or important. Still, I realize that sometimes the significant emerges from the assemblage of tiny pieces.
There have been some spirited debates about the limits of online research. We all know researchers and writers for whom the research world begins and ends at their keyboards. If something doesn’t exist online, then they’re not going to see it. Indeed, it’s easy to imagine that we have a whole generation of historical thinkers who will be conditioned to find their material online, and largely only online. What does that mean to our historical work? I have been pondering a way to gauge just how important online resources have become, and I offer this little tidbit. Definitive? No. But maybe a useful reminder.
I went through the footnotes in seven of the 25 chapters of my book Return to Bull Run, which I finished writing in 1992 on the eve of the digital age. I looked at every citation in every footnote to calculate what percentage of them could be had online today. Going in, I guessed about 25% of the citations I included in the book would be available online today (by that I mean available on a permanent website; I did not include ebooks in the calculation, unless they were available for free use at Google Books or other archival site). I was wrong.
The seven chapters I reviewed included 663 citations. Of those, the material cited in 419 of them can today be found online. That’s 63%.
Given the nature of the book–a battle book–it includes a goodly number of references to the Official Records, which of course are readily available online. Of the 663 citations, 191 were to the ORs. If we back those out, we are left with 228 out of 472 citations available online–still 48%, a much higher number than I expected.
Swapping over to the bibliography, the percentages drop quickly. I didn’t go through it item by item, but it’s apparent that just a tiny share of the manuscript materials I used are available online today–probably less than 5%, if that. It’s worth noting that the manuscript collections I used included some of the most critical material available–Fitz John Porter’s Papers at the LOC, T.C.H. Smith Papers at the Ohio Historical Society, John Warwick Daniel’s papers at UVA, or the unpublished US Army Generals Reports of Civil War Service at the National Archives are four examples among many. Without collections like those, little new would have emerged, and I daresay the book would hardly have been worth writing.
To me, the big news is this: The internet has done little to affect access to manuscript collections (beyond making catalogues and finding aids more readily available), and that’s unlikely to change anytime soon.
Of the newspapers I used in 1992, about half are today available online. Of books, again, it’s about half. All told, I estimate that about 20%-25% of the bibliography is available digitally.
What about the other side of the ledger: What’s out there today that couldn’t be had 25 years ago–that was too obscure, too restricted, or too far away for me to find in the early 1990s? The majority of material going online goes there because genealogists will buy or use it. Some of the material from the National Archives–widows pensions, Southern Claims–that’s finding its way online is a great boon. But those sources don’t begin to relieve anyone of the need to visit NARA. Mountains remain untouched by researchers and will remain so, largely because there’s no commercial benefit to digitizing most of it.
By far the greatest bonanza of new online material is in the realm of newspapers–again, much of it being done on various genealogy sites. In the last six months alone, I have found an additional 30 relevant newspapers–probably 70 additional accounts of the campaign–that I did not have access to way back when (I cite about 90 newspapers in the book).
Where does all this leave us? I think it’s a bit like one of those government reports that tells us eating too much sugar will make us fat. It’s obvious, and it’s clear too that digitization makes research easier. (And certainly we now have a legion of people doing research who would never have attempted it if not for co-axial cables.)
But the big question is, does the internet make research BETTER–does it ultimately improve the products historians are putting on the street? Based on my own work–then and now–my sense is that the answer is “no”–at least for the sort of narrative history embodied in Return to Bull Run. Writing it in 2012 would surely have been easier, but I wonder if I would not have been snared by the ease of online research and missed much that was important elsewhere. And looking at the fabric of the research and the construction of the book, I can’t say there’s anything out there in today’s digital universe that would have changed or improved the book markedly, while there is a great deal NOT online that would have spelled historiographical disaster had I missed it.
While it’s astonishing how fast the mass of materials online is growing, we are still far from the day when new, credible, comprehensive, and definitive history can be written from the digital domain alone. Covet your Ipads, but also hang on to those rolls of dimes for the copy machines, continue to make friends in your favorite repositories, and keep those laptops ready for transcription (barbaric though it may seem), because doing really good history requires all those things.
(A disclaimer: I’m not arguing here that there is nothing in Return to Bull Run I wouldn’t change. I would–including some revised thinking on big issues and players, like McClellan–but these have little to do with the fineries of new source material and more to do with my own evolving understanding of the war. But that’s a topic for another post on another blog.)