keeping alvarez alive

One of the things that struck me the most about the readings this week was the simultaneous sense of overload and loss. It's an exciting and also uncertain time, when everyone seems to be scrambling and regrouping. Rosenzweig's article left me feeling particularly bewildered. I had never really thought about the life span of a digital document. I might need the help of an archeologist in ten years to access material that I created today! I'm glad that I read that first and Turkel's last, because he takes the conversation back to why digital projects open up new and compelling horizons for history in the first place. I was also wondering why people are taking so much trouble to scan books. The books are there. One of the greatest virtues of scanning all the books in the world is that people will be able to find more books. But couldn't a really good global card catalog do that anyway? I know it seems like a silly observation, but why do the books have to be entirely searchable? Shouldn't more emphasis be put on searching effectively rather than digitizing everything? Or does digitizing everything change the way we search for the better?

For my project this semester I've been thinking of creating what I hope to be a somewhat manageable website regarding the subject of my masters thesis. I will likely use what we've learned about databases so far in my larger research for my dissertation, but I just don't have access to all that material right now and I still have to think through the questions I want to ask properly before I start thinking about how to ask them. The good thing is that the material from my previous project is finite. It is a collection of letters, diaries, manuscripts, novels, and a radioplay, written by María Alvarez, a Salvadoran feminist whose activism prompting the passing of suffrage (constitutionally, if not effectively) in El Salvador in 1939.

I have some ideas that I'm quite excited about.

1) It happens that my fiance records music in his spare time, so we have some pretty nice recording equipment in the basement. I thought it would be great to ask a few friends, native Spanish speakers, to act out the widely successful radio play that Alvarez wrote in 1959, a turning point in Latin American history. I could divide the chapters into several podcasts and make them available on the website after a brief synopsis. The radio play is fascinating because the Cuban Revolution happened right in the middle of it and you can see Alvarez trying to make sense of it, torn between anti-communism and a kind of nationalist admiration. It also shows a time in Salvadoran history, now largely forgotten because we see everything through the lens of the civil war, when Christian Democracy was quite compelling to a lot of people, and moreover, that it was closely linked to the feminist movement. It will surprise people that there was a feminist movement in 1959 at all. This is the most exciting part of my project, I think, because it will enable me to try to approximate the original form. I hate reading plays as if they were prose. The original recording was lost in a bombing of the radio station in the 1970s (it was the Catholic YSAX through which Archbishop Romero broadcast his homilies).


2) I always thought it was interesting as I was reading all her novels that she referenced so many Latin American political works. I was curious about this network and looked up some of the key figures, but I never bothered to be thorough about it. I could actually go through her writings and write down all the names and books she mentioned and then do a Google search. In the end I could reconstruct the network and have links to fellow feminists, political figures, writers, etc.

3) I just don't know what to do with her novels and other documents. As I have never undertaken a project like this before, I don't know if it would be feasible to scan them and make them searchable. Given my lack of experience, I'm not sure if I will know how to keep this website alive.

But before addressing these question, I have to address the bigger question that is looming: copyright. She was my great grandmother and my entire family entrusted these documents to me when I was writing my thesis. She published her feminist magazines herself. All the books are out of print and didn't sell much in the first place. One of them was published in Buenos Aires, one in Madrid, and one in San Salvador. The radio station still exists but lost its archives in the fire. I know I can't use the letters because a friend of mine photocopied them for me at the Women's Archives at Radcliffe College in the papers of Doris Stevens -- but what about everything else?

My question about IUScholarWorks is also in this vein: are papers copyrighted? What if a paper becomes a part of a book later on?

Finding Short Cuts in Databases

The exercise this week was quite useful for me because I've been working on a research paper and I knew exactly what I needed to find. I have already done the bulk of the work and have my primary sources from my research in El Salvador over the summer. In a nutshell, my paper focuses on how the use of sacred space has changed in El Salvador from 1950 to 1980, using the Metropolitan Cathedral as a case study. It burnt down in 1951 and remained in a perpetual state of incompleteness throughout the decades in which there was political upheaval from within the ranks of the clergy. It was the place where hundreds of thousands of people gathered every year to celebrate the patron saint, the Divine Savior, which when translated is the name of the country--El Salvador. The patron saint was always invoked as a symbol of "salvadoranness" and I examine how the meanings inscribed in the image (and its "home" in the cathedral) changed throughout the years.

I've already done that part. But there was one key thing that I needed to find in order to make my argument. I knew for a fact (through talking to people and passing mention of this in my other sources) that the Cathedral had been "taken" by popular organizations, sometimes armed, sometimes peaceful, almost fifty times from 1970 to 1985. These takeovers usually went on for days; sometimes government forces bombed the Cathedral to get the people out; and the organizations' explicit aim was usually to make their voices heard when there was no other place to voice dissent. I knew this, but I needed to find sources to back me up, as well as the specific dates of the takeovers in order to compare them to the dairies and collections of homilies that I have of the archbishops.

Now, if I were in El Salvador I would probably just go to the national archive and skim through ten years' worth of newspapers. It would take a long time. The good thing about this time period is that it created a political climate in which the United States was clearly interested in the goings-on of the popular organizations. So I figured some sort of report would make it into the major newspapers. I went to ProQuest and did a search for "cathedral el salvador." Although I only got four hits (they only seemed to report on the takeovers when they involved some sort of bloodshed), it was a good place to start, since I then knew which ones had been the most significant. The filter of the U.S. press therefore made it easier to know which takeovers had loomed larger than the others. ProQuest newspapers, because it searches all the text, is extremely useful in this regard. It is incredibly user-friendly, and if you have a detailed enough search you probably won't be overwhelmed with hits. However, that was not enough. There had been 48 takeovers during the period I was interested in and I only had record of four.


Where to look next? Google was useless, and so were all the databases of scholarly articles, even the ones centered on Latin America. Finally I had an idea. The U.S. State Department archives! There are several and for my purposes FRUS just didn't work in terms of searchability. Then I found the Digital National Security Archive. It has four possible searches: documents, bibliography, chronology, and glossary. I did a search for documents with the key words "cathedral el salvador" and got a long string of results. The documents were mostly weekly reports sent from the Embassy to the State Department regarding the meaningful events of the week. The citation includes a list of the people mentioned, the subject-matter (such as violence, disappeared persons, political prisoners, takeovers, and YES, National Cathedral). You click on the document and you have a scanned image of the original report. Because the U.S. was carefully monitoring the situation, the reports just mentioned in passing what was going on in the Cathedral, who had said what, and the date. This was perfect, since I could then both refer back to the diary entries and homilies for that date, know what the U.S. Embassy was up to, and look up the newspapers for those specific dates.

In sum, I think that for my purposes, it is very difficult to construct any research based solely on online databases, because the "meat" of what I'm looking for is just not in a database. However, they really provide ways in which to direct and fine tune my research that I probably would not be able to accomplish with print sources alone. I can't imagine consulting print copies of the New York Times, etc, just to find one or two articles regarding the Cathedral, but because it's so easy, I can do it in ten minutes, and it enhances my searches elsewhere. Likewise, I would never think to look for State Department records for information on El Salvador (the 'filtering' effect would drive me crazy), but if I know exactly what I'm looking for, it provides invaluable information that it would be difficult to find so easily elsewhere.

data maps are like time machines...

...Click on the date, and there you are! When I first entered Social Explorer, I watched the slideshow of race maps of New York City over and over. Then I compulsively tried different combinations of dates with a certain set of categories. After a few hours of this, my fiancé walked in the room and asked me what I was doing. I said "Watch this!" and showed him where Austrians in the U.S. were mostly living in the year 2000 (he's half Austrian). I was disappointed to see that he did not share my enthusiasm. Then I showed him the race map from New York.

"What does that tell you?" he asked. "What did race mean to people in 1910, and what did it mean to people in 2000? Did they ask people the same questions?"

We had a very long discussion about this. I argued that, yes, there are previously constructed boxes in censuses into which reality is made to fit. But there are many things we can learn not only through the prism of censuses, but also from the shape of the prism itself. In "marital status" before the year 2000, for example, there was no category for "divorced," just "married" or "never married." Before 1960, as far as I can see, there was no category for "race" at a national level. In 1960, moreover, the race category had no subcategory for "Hispanic." Then I stopped and wondered when "Hispanic" became a "race," given that Latin Americans have African, Native American, European, Asian, and Middle Eastern ancestries. It became an official race in the United States in 1970. Given the slippery nature of the category, by1980 all races had the disclaimer "non-Hispanic" next to them. It remained the same through 1990, but by the year 2000, "Hispanic" had three modifiers next to it: Black, White, Other.

I find this fascinating--these efforts to more precisely define a group of people that share a vague something in common, something that is even shared with some Europeans (the Spanish were listed as Hispanic in the "foreign born" section in the 2000 census). I find it fascinating because these are matters of self-identification, but the choices largely dictate the reality that is reflected in the numbers. A census in El Salvador at mid-century left out "Indian" as a category, for example, and suddenly it became a "reality" that all Indians had assimilated into the larger culture somehow overnight. And yet, as flawed as the collection of data may be, it tells us something important. As I watch races shift around Manhattan, Queens, the Bronx and Brooklyn, or grow fainter or darker around a particular point as the date changes in the upper left hand corner, I know that I am watching imperfect data imperfectly reflecting a seismic shift. But the shift is there.

My fiance's questions point to a fundamental difficulty in comparing such vast amounts of information: the sources might differ in content and in kind. George Alter's articles demonstrate how one might try to control for such gaps when the information provided is pretty much bare bones. Data can never speak for itself; it is up to the researcher to take pre-conditioned data and interpret shifts while taking into account variables that were not recorded. In "Theories of Fertility Decline," I found his evaluation of the "rising cost of children" quite instructive: "Subjects in surveys and interviews overwhelmingly report that children cost more now than they did in the past... On close examination, however, they are not describing a change in the costs of children as much as a change in their own definition of appropriate childrearing.... Historically, children have become more expensive not because the prices of commodities consumed by children have risen, but rather because parents have decided to invest more in goods and services for their children" (16). How does one control for culture? Culture is one of those things that have proven almost impossible to define--yet there it is. He also points to the reasons why doing this kind of research with large amounts of such data is both necessary and a little unsettling: the negative result. When theoretical frameworks are informed by a particular vision of the past, historical analysis must be both nuanced and as thorough as possible. These theories are, in many cases, the drive behind policies.

All these questions have made me really think about my own research--and the possibility of a negative result. My starting point is now as shaky (and as vague) as ever: the Catholic Church became increasingly involved in matters of social justice throughout the twentieth century (I can come up with a more compelling way to put it, but when it comes down to it, that's what it is). It was controversial. Many people stopped being Catholic and many others embraced the conservative Opus Dei. The Church was not a monolith and there was a lot of disagreement (more than we usually think) from within. Okay. So far in my academic career, I have concentrated on interpreting sources written by the institutional clergy. Articles, books, homilies. But I have this nagging feeling that these sources will tell me just as much as what I've learned about the Jansenist controversies in 18th century France: the arguments in journals. How many people read those journals? How many people could read at all? Who cared?

How should I phrase my questions? What sources are available? One thing I might ask is, how many people came into contact with the Church at an institutional level throughout their lives and how did this change in the twentieth century? I can only track what the institutional church was saying or disagreeing about, but it is far more difficult to track people's reactions to what was being said. A priest can yell all he wants from a pulpit, but if there's only one person in the pews, what does that say? What if he yells and the Church is overflowing? Early on in my research I thought it might be a good idea to try to track people's use of sacred space as time went on, and see how it shifted with the changing ideology. The task, however, seemed too monumental. When I brought it up at my review, one of Peter Guardino's excellent questions was, "what if people were just going to different churches because of urban expansion?" How would I control for that?

Here is what I came up with.




So, by enlisting the help of several people (possibly many) for the data entry, I could enter the names of people as they go through the sacraments. Do they stay within their parish church throughout their lives? I can pretty much track where people are going to church (roughly), and whether there were massive shifts in affiliations at certain points in time that do not correspond to urban growth. This is just a rough idea of how the information might be organized. Of course, this would not substitute but complement my initial approach to the research. But that fundamental disconnect between the people doing the talking and the people hearing it might be bridged in this way.

I might also be able to do something with schools, but it is obviously still very unclear exactly what:

interpreting "two pigs, six silver spoons, and eight hoes"

Working with old newspapers is difficult because it is almost impossible to suppress the instinct to browse. After all, that's how we read the newspapers with today's date on them: we wander through them. Browsing is not a bad thing, of course, and can sometimes take your research to unexpectedly fruitful places. It also helps to put you in a similar frame of mind as the publication's original readers: they probably wanted to be informed but also entertained. When you're faced with thousands of issues, however, lack of focus can make you lose sight of your original questions. When you're faced with millions, lack of focus reaches entirely new levels. How can digitized sources help to keep our research on track?

I looked up databases with digitized newspapers and found that they only exacerbated the problems that I had faced before. They seem to have really opened up the breadth of possible research. But they do not provide ways to organize that research in any meaningful way. I have about 1,500 digital photographs taken from a weekly newspaper called Orientación, published by the Archdiocese of San Salvador. My sample spans from 1951 to 1984. I don't want any more for the time being. It is very easy to get lost in what I have. How do organize the various stories that seem to develop independently of each other through the decades? Should I be looking for other things aside from the explicit stories or arguments in the articles? How do I keep track?

Some of the questions I am asking would be easy to keep track of through time with a simple Excel work sheet, or even by tagging photographs on iPhoto with several key words in order to be able to do a quick search. With a little organization, I could trace how anti-communist rhetoric evolved into social justice rhetoric, for example, or a timeline of how the metropolitan cathedral was rebuilt after it burnt down in 1951 (the reconstruction took over 50 years). These things are related to the explicit content of the articles and hardly involve a very close reading. But in looking at the way people have used their sources to reinterpret the past, I was particularly impressed by Thatcher Ulrich's work on labor in the eighteenth century, and in my own field, Muriel Nazzari's Disappearance of the Dowry: Women, Families, and Social Change in Sao Paulo, Brazil, 1600-1900. Nazzari used "inventarios" spanning several centuries as her sole source; they are the same thing, pretty much, as probate inventories.


Nazzari uses the dowry as a focal point in evaluating how property-owning families (a corporate unit including both father and mother) changed their behavior toward their sons and daughters over the course of three centuries, 1600 to 1900. Her argument is straightforward: the disappearance of the dowry meant that women went from being the buyers of men to being bought. The development of individualism and market forces caused the splintering of the patriarchal clan and a subsequent shift of the family’s allocation of resources. The privatization of women’s roles was due not simply to ideological shifts, but the emergence of a vastly different economy.

All of this, Nazzari concludes after interpreting data like "two pigs, six silver spoons, eight hoes" and seeing how it changed slowly over time. She took advantage of the publication in several volumes of the inventarios in the archives of Sao Paulo until the late nineteenth century. These were not digitized, however, and she had to deal with a representative sample for each period to keep the study manageable. If she could have isolated the data and then been able to have links between different kinds of data, she could have probably had a much larger sample, but I don't think it would have significantly changed the research itself.

There are several things in my sources that I think could work in a similar way--details that only start to take shape as a larger picture when they are compared with each other through time. What I took from Nazzari for my own work is an approach to sources. Granted, a newspaper is a different kind of source. Articles are different from wills. But there are parts of the newspaper that matter-of-factly record certain yearly events, or the amount of money gathered for the rebuilding of the cathedral, or the amount of people who showed for the festivities of the patron saint. A database of some kind would really help me to isolate these and compare them to the broader stuff being discussed in the articles.

At this point, I think it's time to pick up a pencil and some paper, since there's only so much a person can do with a keyboard.