When I was first invited to participate in the “Notes to Narrative” forum, I was initially reluctant to publicize my messy notational practice. But friends told me that exposing the sheer inefficiency of the process was a socially useful act that might prove reassuring and enabling to historians in the making.

I travel a long distance—to London, England—to visit the archives I often use; so my research trips tend to be highly compressed and focused on the acquisition of documentary materials rather than on sorting and processing them as I go along. As a social and cultural historian, I am always engaged in a double process of interpreting documents. I assess a document as both a source of historical information—evidence about what “happened” in the past—and also as a rhetorically constructed text and cultural representation.

Before going to the archives, I try to read in advance the secondary literature on my research topics, including popular histories, for two reasons. I need to know the received narratives about my historical subject and master a lot of historical detail. I also mine these texts for their archival citations. It is a rare occasion when I discover a source that has not been previously consulted by someone else (it is a delight but a rare occasion).

Unless I am going to the U.K. National Archives in Kew, where it is unnecessary to book a place in advance, I write ahead to archivists, make an appointment to visit the archives, and request documents in advance. I note the names of the archivists with whom I have corresponded and spoken. When I arrive, I attempt to establish cordial relations with them.

I arrive, equipped with a laptop computer, a battery and plugs, an external hard drive with many gigabytes of storage space, and a digital camera with at least one extra battery and a cable connection to the computer. I open up a Word document file for each source. I do not use a bibliographic system such as “Endnotes,” but I keep a running bibliography. I try to take stock of the file before I dig into the specifics of summary note-taking. I try to determine the conditions that generated this file: in the case of governmental files, was it a result of routinized oversight or was it the result of the disruption of routines by an outside force, such as a complaint or a newspaper report about a scandal? I also try to describe the materiality of the file, its size, variety of paper, layers of communication, and chronological ordering.

I then take careful summary notes of what seem to be the most significant documents, with careful attention to their place in the sequence of the file, how much space they take up, and their rhetorical ordering and sequencing. I back up my files religiously on the external hard drive. I also open a second Word file, which I call “Idea File.” This is a space of free association, a form of uncensored writing that might well not see the light of day in any final text. This second file is intended (1) to engage in an active cognitive process or “engaged reading,” (2) to move to generalization while still staying very close to the textual particulars, and (3) to start the writing process as early as possible.

In this second file, I sometimes include a running commentary on what I think is going on in this document and how it might connect to other sources. I also write memos to myself about secondary or theoretical texts I need to consult in relation to this material. Most important, I record the details or features that strike me as surprising or telling: whether there are significant silences or elements that seem to be missing.

I also do a certain amount of photocopying and digital photography. I think it is a mistake to photograph or photocopy documents without simultaneously processing them into some form of notes and explanation. I treat both forms of reproduction as supplementary to the task of summary notations and engaged reading. I am selective about what I reproduce—I tend to reproduce the most important documents that would otherwise require a complete transcription, and then I also reproduce some supplementary documents. I keep a log of the documents I photograph or photocopy. Before I photograph, I place a rectangular piece of paper on the document that records its archival citation and maybe a keyword or two relating to its contents.

To keep my notes in order when I am in “writing mode,” I print out notes from the computer and file them in file cabinets, arranging them alphabetically according to chapters. I keep a running index of this file system on my computer. I read over my hard-copy notes, annotating them and adding comments to idea files.

I index these notes in multiple formats. If I have a really long source file (say I have been working on a huge scrapbook collection of newspaper cuttings), I type up a compressed index of that file, with highlighted elements, using a split screen. In addition, I organize my notes into topics or themes by cutting and pasting text fragments and quotations, using a keyword search on my computer.

I then assemble a two-inch loose-leaf notebook. This contains a paginated table of contents and three sections. The first section is a summary of relevant secondary materials on the subject and occasional, suggestive excerpts from theoretical texts.

The second section consists of my various indexes, idea files, and topical arrangement of notes. The third section often contains the full, long source notes that are at the heart of the chapter, accompanied by some of the photographed or photocopied texts.

This tends to come to about 300 pages but it is mostly what I will need to write a chapter.

The indexing and compression process serves two functions: it is a primitive, homemade information retrieval system that helps me with paper management; in other words, it is an effort to stave off disorientation and distraction from too many documents. It is also an aid to memory. The process of producing these indexes helps me regain an active command over the texts I have read over a long time. No retrieval system can substitute for the historian’s memory—but it can aid memory.

After all these maneuvers, my texts do not write themselves. I still have to engage in trial and error and multiple redrafts. I keep redrafting outlines. My organizational difficulties relate to negotiating foreground and dense, compressed background. I always have a lot of things going on in a text; or to put it more positively, I try to produce a multilayered text that does not appear to be too busy. It is a challenge for me to balance diachronic and synchronic elements in a text and sustain a narrative arc and build momentum, but I must say that my note-taking methods, unusual and idiosyncratic as they might seem, help me immensely in my writing.

—Judith Walkowitz is professor of modern European social and cultural history at Johns Hopkins University where she specializes in the history of Britain and of comparative women’s history. At Hopkins since 1989, she has served as both the director of graduate studies and the director of its women’s studies program. Prior to coming to Johns Hopkins, she taught at Rutgers University where she helped to develop its highly regarded women’s studies program. Walkowitz is the author of several books including City of Dreadful Delight: Narratives of Sexual Danger in Late-Victorian London and Prostitution and Victorian Society: Women, Class and the State, and was a founding history editor of the journal, Feminist Studies. She is currently working on a book project entitled Cosmopolitan London, 1880–1945, to be published by Yale University Press.