Triage in the Archives

I’m working on my last batch of documents in the provincial archives in Shandong. There are two challenges to doing my historical research here which I often think about. The first is the problem of access to both the archive and much of its contents. I have been very fortunate but I regret that it is more of a result of good fortune than anything else. This posting will focus on the other problem, the need for a kind of triage in the archives and the constant awareness of my own personal limits as a reader. It is a humbling experience, and I suspect many, if not most, historians, come to face it if they have spent much time doing archival research, especially dealing with documents not in a language they speak and read natively.

Language and Detailed Local Knowledge

I enter the archives here with a topic in mind, a relatively good understanding of the regional and chronological context for my topic of study, and a working knowledge of the terminology often used in the kinds of documents I will be looking at, in part thanks to the existence of a published collection of documents from the same archive (山东革命历史档案资料选编). However, I have two major disadvantages that I feel very acutely every day I come to the archives. One relates to my language ability; the other to the limits of my local knowledge.

Though I can read Chinese, especially when it comes to the materials in my particular field of study, I have two huge linguistic disadvantages compared to any native speaker of Chinese (and, to a lesser degree, native speakers of Japanese): 1) I read Chinese much slower, and more importantly, skim Chinese slower, than native speakers. I still have to occasionally look up words that cannot either be understood by context or safely ignored due to probable irrelevancy. 2) I do not have a lifetime of practice reading handwritten documents using cursive or radically simplified Chinese characters, which compose over half of the materials I’m looking at. This means that some of the many handwritten documents I look at here, where I do not have permission to photocopy or take photographs of the materials I am looking at, are partially or in a few cases completely impossible for me to read.

The second major kind of disadvantage I have relates to the fact that, as one archivist here put it to me sympathetically, “This must be overwhelming, since you have only had time to study Chinese history for a year or two before you came.” This makes it seem like every Chinese historian has studied Chinese history for decades and is thus many years ahead in terms of knowledge of the specifics of Communist party anti-treason campaigns in Shandong province, which is simply not the case. However, all other things being equal, I must come to terms with an obvious fact that lies at the heart of what the archivist was trying to point out to me: It is physically impossible for me to have found time to read more than a subset of the Chinese language secondary works or document collections that are related to my field in the short time I have worked on my dissertation, let alone read, as some graduate students and scholars here undoubtedly have, read the many other peripheral works that help one understand the context surrounding my topic. This is even more true since I am doing a transnational and comparative project that also incorporates Korea.

The only way people in my position can walk into the archive each day with some degree of self-respect is to convince ourselves that we have something unique to offer the study of our historical topic that gives us some kind of advantage relative to other scholars and students who might be working on a similar field here. Whatever this might be, our critical question, our comparative approach, our sensitivity to patterns etc. that might not be apparent to those working in other scholarly contexts, and so on, it gives us the confidence to go in and struggle through the historical materials and accept our weaknesses. In my case, I try to tell myself the contribution I can make is largely to be found in the way I “slice” the range of my inquiry and attempt to use that slice to answer particular questions. I remain open to the idea, however, that the “uniqueness of approach” claim may ultimately be an illusion, and as the quality of academic research here in China improves rapidly (I was really impressed with the breadth of reading and fresh approaches taken by some graduate students I have met here), some of the other advantages that foreign scholars coming to study might once have dared to claim are disappearing.

Even if one does avoid falling into complete despair, it remains an incredibly humbling experience to walk into the archive each day and be faced time and time again with one’s own all-so-apparent inadequacies. Below, let me share some aspects of that experience with some examples and the unfortunate but necessary steps I have to take in order to maximize the number of historical gemstones I can mine in the ocean of archival material available to me, despite my weaknesses.

Archive Triage

A man walked into the provincial archive here a few weeks ago and asked to see proof of his father’s selection as a “model worker” in 1952. He unrolls a crumbling certificate glued onto some old newspapers that he says is the original certificate. An archivist looks through some kind of a list they have for that year and find no mention of the man’s father. The man left dejected, “Everyone told me it was fake, but I still can’t believe it.”

Though this is a sad story, this shows a kind of ideal situation for a historian: to be able to walk into an archive with a detailed question, to find an authoritative source that can answer the question, and walk out with a relatively firm answer.

A policewoman walked into the archive last week and said she wanted to know more about her father’s case. Apparently, sometime during the 1940s (I can’t remember the exact year) he was accused of being a “traitor” and a “reactionary” for being a Nationalist party member in some Communist base area and had various trouble in the many anti-reactionary campaigns that followed in the decades thereafter. The archivists helped her look for information but found nothing that could be of help to her. She was offered several other places that she could go and look into things but still left disappointed.

Here we have a case where someone has a somewhat broader question, anything about her father’s case would have been helpful to her, but the archive was completely silent. This is similar to what many historians face, and I think they often change their topic, their sources, the archive in question, or the way they frame their questions in response.

However, there is another common problem, which I face here along with, surely, many of my fellow PhD cohort now camped out in various dusty reading rooms the world over: The challenge of what to do when the archive offers many hundreds of documents that each have a small possibility of offering a nugget or two that may be of use.

One simple and immediate strategy that a historian can then take is to immediately limit the scope of inquiry. That isn’t always the best first approach, however, and should probably only be attempted after getting a good sample of the whole range. Just because you have a huge potential source base, doesn’t guarantee you that selecting any subsection of it, based on region (limiting my study to treason elimination squads in the Jiaodong district), period (for example, the early formative period 1939-1941), or narrower topic (focusing just on how the squads attempted to get the ‘masses’ involved) will yield enough to be interesting.

It seems like the good results from archival research come in fits and starts. I can go for days without finding anything really useful, but then come across several fantastic finds in the course of a few hours. However, even in these cases, these fantastic finds may still only translate into a single paragraph of text or a footnote within the mammoth that is one’s dissertation. Depending on the kinds of source materials, you often have no idea if the next thing you pick up will be a total waste of time or will yield something wonderful.

Learning not to read. One of the skills that has been quite painful for me to learn is to overcome the urge to read everything. A Weihai police report from late 1945 that I looked at yesterday, for example, was over 80 pages long. Of those 80 pages, perhaps half a dozen distinct paragraphs, often separated by a dozen pages, are remotely useful to me. If I really read the full 80 pages of handwritten text, that document would take a whole day. I would probably have a much better understanding of Weihai in 1945 and could probably have possibly found more or even as much as twice the useful information, but very quickly one has to make a call about whether the potential gains are worth the time. Fortunately, the year long preparation for one’s oral exams in a PhD program, which involves the ‘reading’ of hundreds of books helps teach the lesson of not reading but effective combination of selective skimming and close reading of some sections. Unlike preparing for orals, however, the key here is not to extract the ‘main arguments’ of a report by a treason elimination squad in the Binhai district in 1944 or a Shandong police journal from 1947. The key information is very often in precisely the minute factual details and anecdotes that orals preparation teaches you to give only enough attention that you can evaluate whether they contribute or contradict the argument being made by the author of a work.

So what to do? Well, when the source base is quite large, the most useful strategy I have found is to quickly identify patterns in the structure of texts and calibrate your reading speed to locations most likely to yield results. Reading everything would, of course, yield more, but time is a very scarce resource. Early on, I found that many (but by no means all) treason elimination squad reports are divided roughly into sections, not always clearly identified, and that the kinds of meaty anecdotes I have found useful in the past are usually located in two of these sections as instructive examples. Village petitions to have certain people punished as traitors usually have long introductions and conclusions which are highly formulaic and can be skipped. North Korean trial records have extremely rigid structures that, while also not clearly marked, can be located easily by finding certain key phrases in the first sentence of paragraphs, and so on. One strategy I adopted was to familiarize myself quickly with document structures when looking them over as a whole before skimming them. To facilitate this process, try requesting similar kinds of documents in groups, even when they are separated by region and time, because similarity in the structure of these texts can significantly reduce the time it takes to process them.

This has risks too, however. If I request lots of different kinds of documents from the Tai’an district in 1942, for example, I will quickly come to understand the importance and power of something known as the “Tai’an incident” which ripples across other regions in that year and others that follow. Taking the document group approach, however, the importance and power of that event only becomes apparent when reviewing my notes from several weeks of reading. This teaches another lesson though: despite the extra time it takes, frequently review the notes one takes in order to identify new patterns, new keywords or documents to search for, and deepen one’s understanding of the chronology and institutional or regional context of the material.

The last and most painful thing I have had to do which is the best proof I have of the sad reality of my personal limits as a researcher is the kind of triage which is based purely on a linguistic evaluation: Last week I had at one point a dozen or so documents. One of these were detailed meeting minutes from a public security bureau meeting held in a Communist controlled but nominally Japanese occupied area. Given the fact it was “close to the ground” in terms of being a very “local” text, and clearly not edited before being bound and submitted, there probably would have been some good unfiltered information about what was going on in the area. Thus, the chance of finding “gemstones” of information in the source was relatively high. However, the handwriting was about 90% illegible to me at first glance, and even if I slowly worked through it, I doubt I would be able to determine more than 50% of the content with careful reading. If I was a Chinese native speaker with more experience working in these documents, I could probably do much better. However, since I’m not, and my time is scarce, I decided to use the several hours I would have spent on that document on two or three other documents with a lower chance of yielding good material but which I could read much more easily.

This is the kind of decision that has to be made all the time, and it is sad and frustrating. It is especially frustrating when one is looking directly at the gemstone in question. To take one example of many, I found an anecdote filled with rich detail in one report on an exchange between an accused traitor, some women who attended the mass trial that were yelling from the audience, and a man who got on the stage to confront the accused. I could make out bits and pieces of it, and have a theory about what transpired (I believe the accused was thrown into a well), but several key phrases were illegible to me—not because the text was smudged or the paper burnt, but because the handwriting was too difficult for me to read in a few sentences. Thus, I did not record the anecdote at all in my notes. Of course, native speakers also have a great deal of trouble with some of these texts but, all other things equal, have a huge advantage when trying to decipher things. I will still be able to write my chapters and have found great material to support my arguments, but I often lament the fact that I had to leave so many bright gemstones embedded in the rock because I couldn’t take the risk of having misunderstood a text based on a mere partial reading.

I have tried to shared some the humbling realities of doing research here and some of the triage I have had to perform while in the archive. As a closing comment: I often wish that historical research encouraged something akin to the practice of “pair programming” wherein two researchers work together on the same materials, side by side, checking for accuracy, misinterpretation, poor selection of material, etc. I know there are many good pair translators out there, but I think it is less common for historians to collaborate – especially at the research stage as oppose to the writing stage and it reminds me of the debates we had in seminars over whether history can ever be a discipline that truly encourages collaborative work.¹

1. During our discussions in seminar, the historians of the Annales School were seen as the major exception to this observation [↩]