Analyzing Indices

I’ve been revising an article this week, and I needed to revisit an old and familiar source: James Thome and Horace Kimball, Emancipation in the West Indies (1838). The text praised the favorable results of emancipation in Antigua, Barbados, and Jamaica, critiqued the apprenticeship system and gradual emancipation, and it became a popular resource for abolitionists making the case for immediate emancipation in the United States.

I was interested to see how Thome (the primary author – Kimball died not long after the two returned to the United States) wrote about “amalgamation” and interracial marriage amid West Indian emancipation since this was a touchy subject even among the most radical white abolitionists. I did a keyword search on Google Books’ version of the text to see where “amalgamation” appeared, and among the few pages were several from the index.

I found that the book’s index had an entry for “Amalgamation,” as well as:

Promiscuous seating in church, 21 (See “Amalgamation” etc.)

Intermixture, 76, 79, 287 (See “Amalgamation.”)

The pages corresponding to the index entry for “Amalgamation” turned out to be passages related to interracial social scenes, not interracial marriage. Thome and Kimball described how white families sat next to colored and black families in church; as well as recounting how they had dined alongside white, black, and colored ministers and their wives. While the book does contain discussions of “licentiousness” during slavery (see p. 98-101 for example), the index entry for “Amalgamation” did not point to these pages.

This struck me as quite interesting — and it suggests the usefulness of looking at an index for insight into ideas about categories and language. In what I think might be a useful research move, it also led me to question how abolitionists used the term “amalgamation” in their own writings.

Before this, my sense had been that the term “amalgamation” was almost always used by the opponents of abolitionism. (For examples, see the American Antiquarian Society‘s online exhibit of the “Practical Amalgamation” series). Opponents of emancipation could take advantage of the slippery meaning of the word — it could refer to benign social mixing, a mixed labor force that threatened white workingmen’s jobs and social status, or the more illicit subjects of sex and interracial marriage.

Abolitionists – like James Thome and his editor, Theodore Weld – rarely used this term to describe themselves or their plans for emancipation in the United States – or so I had thought. In thinking about what I’ve read, it seems that abolitionists in the 1830s almost only referred to “amalgamation” when they were discrediting accusations that they were “amalgamationists.” Instead, they would point to licentious white southern slaveholders as the real culprits of “amalgamation.”

So why did Thome and Weld include “Amalgamation” in their index? Were they trying, in some small way, to reclaim the word and to distinguish it from its sexual connotations? It looks like they might have been trying to do just that, and if I hadn’t looked at the index, I would have missed this point. They didn’t use the word very much in the text of the book, but the pages in the index under “Amalgamation” included descriptions of families clearly distinguished by their race (negro/colored/white) who were not intermarrying but worshipping together in church, dining together in “civilized” dining rooms, and discussing the issues of the day together in middle-class parlors.

Teaching Research: Slave Narratives

For the second meeting of my research seminar, Defining Marriage, I assigned Patrick O’Neil’s article, “Bosses and Broomsticks: Ritual and Authority in Antebellum Slave Weddings,” Journal of Southern History (March 2009).

I assigned this alongside Joan Scott’s “Gender: A Useful Category of Historical Analysis,” and two other articles about marriage in early-antebellum America (Richard Godbeer, “‘Love Raptures,’ and Jan Lewis, “The Republican Wife”). The goal for class was to discuss the different sources and methods for each article, as well as how the three articles illustrated Scott’s definition of gender.

In addition to being a way to talk about gender, paternalism, and ritual, “Bosses and Broomsticks” offered a bonus: a perfect teaching moment about digital sources, research methods, and the need to question evidence.

O’Neil uses digital archives, including the North American Slave Narratives collected at Doc South and the Library of Congress WPA Slave Narratives. While this is not in and of itself unusual, O’Neil also cited the URLs of each source in his footnotes. This changed the way my students read the article: they could go find and read these sources themselves, and they were excited to hear that the LOC also had audio files that they could listen to because they were curious about the use of dialect in the transcribed interviews.

This seems like a pretty good reason for scholars to be more explicit about the digital nature of our research.

Additionally, O’Neil’s method (he used keyword searches and folded together a textual analysis with his statistical analysis) also led to a conversation about how digitized sources can open doors to new and different approaches to research. Hopefully these early idea seeds will grow into research papers in a few months.

The WPA slave narratives also presented an opportunity to talk about the limitations of primary sources. I hadn’t planned to go in this direction since this wasn’t really the point of this week’s seminar, but I’m glad we did because we’ve now established a set of questions for later discussions about the quirks and complications of sources.

What I used to do, Part III

Or, how I’ve adapted the old to fit with the new.

"Amanuensis," Scribes in the Abbey of Echternach (Luxembourg), Manuscript from the 11th century.

When I’m at an archive, I transcribe and I take copious notes. I recreate documents in huge Word documents that I can “find and search” later. I also keep a separate document for commentary — my notes, my research questions, and leads for future research.

On a recent trip to the New-York Historical Society, my confidence faded when one of the archivists tentatively told me that the speech I was reading and transcribing was also online. “You know . . . you can actually read this on archive.org . . .” she trailed off. Here I was, typing as fast as I could when everything I was entering into my laptop could be accessed at any time from my couch. A crisis of archival faith.

(An aside: this wasn’t the first time I experienced this. In Boston last spring, a very kind and helpful archivist boasted about how many abolitionists’ letters they had digitized. She told me that I could just look at those letters online. Pause. Yes, I said, but I’m here. In Boston. I could understand that she wanted to protect the documents from being handled too much, but what was I supposed to do now? Did she want me to sit in the reading room and read the letters on my computer? Did that even count as research? Historians are still supposed to go to archives, right? To paraphrase Mindy Kaling: was everyone else going online without me? (Seriously: what is everyone else doing? I know many archives aren’t online, but as so many become digitized, is this the way of the future? Is it naïve to still want to see and handle documents in person?))

After learning that the speeches I was reading at the New-York Historical Society were online, I debated whether I should just finish up early and go home. But. During the past few hours, I’d had more productive thoughts than I’d had in the past two weeks. I wasn’t just reading these speeches from 1864 — I was thinking about them, making connections, and my article’s outline had started to come together. The quiet room, the time crunch (. . . have to finish this speech before lunch), and the coffee I’d had earlier in the morning had created the ideal conditions for inspiration.

Later in the day, it hit me: this is what I need in my digital research! I need to find a way to create these conditions when I’m reading the 45 PDFs from the New York Tribune or the Google Book I’ve downloaded to my iPad.

So, to conclude this series of posts: how do I institute slowness in a research environment where speed and quantity rule? I’ve started to transcribe again. I organize the many newspaper articles I’ve been collecting in folders in DropBox, each titled with the date, the newspaper, and something of the headline or topic. I can then access them via my iPad with iAnnotate or Good Reader, programs that allows me to zoom in and out so that I can see the entire article or page. I also store Google Books (I’ve been reading nineteenth-century memoirs of dead missionaries most recently) on my iPad. The key is to treat these documents as if they aren’t digital. When I’m reading them, I open a Word document on my computer and I read through the documents as if I’m in a reading room at an archive. I transcribe long passages, I take notes, and I write out my thoughts.

This felt like unnecessary labor at first because it’s counterintuitive to the promise of accessibility. I don’t need to transcribe these documents. I have them with me basically all of the time. But this was beside the point. For me, the transcribed passages themselves are less important than the time I spend transcribing and taking notes. The process puts slowness back into my process and sets aside time to think.

Part I and Part II

What I used to do, Part II

Words I never thought I would say: I miss microfilm.

Not the faint chemical odor or the kind of sea sickness that inevitably results, of course, but the time it required of me. I came to this realization during a recent visit to the New-York Historical Society. I had to look at a few things on microfilm, and it felt like returning to an old friend. (More on this visit in a later post).

With a few exceptions (including my day at the NYHS), I haven’t used microfilm since my dissertation research because most of what I need has been digitized. Digitization has allowed more people to access historical documents, and the ability to do keyword searches of an entire run of a newspaper has led to new kinds of research questions, as Caleb McDaniel has discussed on his excellent Offprints blog.

At first, I didn’t put a lot of thought into how digital research differed from what I did before. When I began my first post-dissertation research project, I turned to databases like America’s Historical Newspapers to collect dozens of articles on a topic in a 30-minute break between classes. I saved the articles that seemed relevant, and, if I had time, I created an entry in Zotero. ”Search/scan/save” became my new research process by default. I rarely had time to read and take notes on the articles as I found them; nor did I read through complete issues of newspapers anymore. But I did have several brand new folders on my laptop’s desktop filled with PDFs. I would read through them later, I told myself — when I had time.

While this looked like research, it didn’t feel that way.

In the old days – the early 2000s – I lived in the microfilm room. I spent hours each week sitting in the basement of the library. The microfilm reader and the ten-cent charge for printing a page forced me to progress slowly. I read every handwritten letter then and there. Using the old notecard system, I wrote out an index card for each one – author, date, a few notes about the content. The already glacially slow process would ground to a halt when I decided to print. Later in the day, I would sort and staple the day’s stack of documents, at which time I had to re-read each letter again in order to make more detailed notes on the letter’s card.

I do not mean to romanticize microfilm. The microfilm room could feel like an over-air-conditioned prison. There were no windows, and time seemed to stand still. To make matters worse, while I had sentenced myself to this machine for three or four hours, I watched the editorial assistants for the Journal of Southern History come and go, fact-checking their one item, merrily rewinding their reel, and returning to the light.

Yet now I understand how vital this experience was to my work. The process of reading and printing from microfilm forced me to move slowly. It carved out a quiet portion of my day when I was alone with my thoughts and unable to multitask. I had to focus on each letter at that moment. I can’t remember this time in the library’s basement leading to any particular breakthrough – for me that happens when I write and revise – but I think this time away from books, my own writing, and, of course, the Internet was essential. It was a researcher’s version of what Pico Iyer describes in a recent New York Times piece about the importance of quiet and solitude to creative thinking.

One of running topics here will be to consider how the means of research shapes our process and the relationship we have to our sources. If you have any similar experiences or thoughts on this matter in your own research, please share!

Part III will examine how I’ve tried to reinsert slowness into my research process and to blend the good aspects of my old practices with the advantages offered by new technology.

What I Used to Do, Part I

Or, how exactly am I supposed to keep track of this
massive pile of papers?

In graduate school, my first major research projects involved hundreds of pages printed from microfilm. One paper on antebellum southern children’s literature (my first published article!) essentially involved me printing out the entire run of the magazine (about 8 years) from microfilm. For another paper, I printed out twenty years worth of speeches and articles from the WCTU’s annual meeting minutes.

Since my research as an undergraduate had not involved anything on this scale, my plan was pretty basic: keep the papers in stacks and mark important pages with Post-its. I’m fairly sure that there may have even been colored Post-its involved, so, you know, it was very sophisticated.

At some point soon after this semester, one of the most memorable and valuable moments of my academic career occurred. A professor took me into her office to show me her filing system. She had a couple of boxes filled with notecards, and each notecard referenced a printed-out document (a letter, a newspaper article) filed neatly away in color-coded file folders. The cards were arranged chronologically, and the documents were arranged by person (if they were letters) or by the newspaper (if they were articles), etc. Her system was and is fairly standard: she could move by year (what happened in April 1878?) but she could also read through a person’s papers or the editorials of a particular newspaper.

It all seems obvious now, but at the time, this was a revelation. Stacks that had colonized my living room, be gone!

For my dissertation, my research involved a large collection of letters covering around 40 years and written by 30 or so people. While I had other documents as well, this was the bulk of my research. To organize it, I assigned each letter-writer a folder  in which I filed his or her letters in chronological order. I also wrote a notecard for each letter – noting the date, the correspondents, and the topics of discussion – and I filed these cards chronologically. The act of making this filing system enriched my understanding of the evidence, and when it came time to revise — individual chapters and the entire manuscript — the system was invaluable.

But for various reasons, this amazing system doesn’t always work, and I find that this is the case for my current research.

• What if you don’t print out your documents because they’re saved as PDFs?
• How do you file or write one notecard for a 90-page pamphlet or 400-page book?

Computer programs like Zotero, Endnote, (see a Prof Hacker comparison of the two) and others try to replicate this system for a digital age, but while I want these research databases to work for me, and I’ve tried using them, I find that they don’t. I’ll explain why I think this is the case in the next post:

Part II: How I acquired all of those papers in the first place.