Spatial history & the interdisciplinary job market: some experiences

2011 February 10

Like most early-career scholars who blog under their real names, I don’t write a lot publicly about my encounters with the job market. Even so, an experience I recently had has convinced me to break that silence.

Background

There’s a lot going on these days with the use of mapping technologies in historical scholarship. As someone who started my dissertation early enough to not catch that wave, I know that I want to land at an institution that can support some mapping-related research while I work on my book manuscript. I’ve written before about the idea of mapping rural African-American midwives from my primary sources, which is one kind of exploratory project I’d like to pursue. I also have some correspondence I’d like to map, and I’d like to do some computer-assisted exploration of the historical social networks I study.

Usually when I talk to historians about my plans for digital projects, they’re fairly receptive. As a result, when I’m applying to postdoctoral fellowships and faculty positions, I pitch my interests to search committees as intriguing future projects and/or book-manuscript-revision work.

A recent interview with an interdisciplinary committee taught me a few things about how these ideas come across to people from quantitative social sciences, and I think they’re worth sharing and discussing as a larger community of digital-methods people in historically-oriented fields. And so, some Monday-morning quarterbacking:

Interviewer: “What’s your hypothesis? What do you expect to find by using maps?”

This question, asked about my interest in mapping some of my sources, is one I didn’t have a good answer for on-the-spot. At the time, I didn’t understand well enough that social scientists who use GIS or other mapping technologies do so to confirm a hypothesis. The idea of using mapping as an exploratory tool didn’t seem to be something intelligible to this faculty member, at least the way I was describing it. (I’ve since been told that people who do quantitative work usually diss such approaches as “fishing” for a theory in the data.) I think my interlocutor was assuming that “we” map data when “we” want to test a theory against the real world. From that perspective, I probably didn’t sound like I knew what I was talking about, and I regret that.

What I wish I had said:

I’m interested in applying tools that have been historically used by social scientists in hard-positivist ways, but I’m not a social scientist. For one thing, humanities-style data has lots of gaps, and most of my methods training in (women’s history/sexuality history/African-American history) is concerned in some way with how to interpret silences and gaps in the written record.

Humanities training is useful in capturing the texture and details of individual experiences, and I want to use mapping tools in an exploratory way to visualize things that I might see as trends. The kinds of analysis I’m interested in are more like how qualitative social scientists use interview-coding software to analyze their interviews with research subjects. It’s possible that GIS isn’t the tool I need and that Google Maps is a better option, but I want to be in an interdisciplinary community where I can meet people who do spatially-oriented work and learn from them.

(Edit: on the specific conjectures and questions I’m interested in exploring via mapping, see my response to a comment below.)

Interviewer: “But isn’t that more about networks?”

This question appeared slightly later, when I talked about using the Children’s Bureau correspondence collection as a data source for visualizing who was writing to whom, when, and how those patterns changed geographically over time. (I was thinking about some of the work that’s been done by Mapping the Republic of Letters, covered by the NYT here and here, which I think has a lot of potential for mapping feminist social-reform networks in the early 20th century.)

What I wish I had said:

Yes, it is about networks, and I think that being in interdisciplinary contact with social scientists who understand social-network-research tools will help me find what I need, tools-wise. I may venture into structural analysis of social networks, but that’s not my field of expertise, and I’m not using these tools as a way to prove a hypothesis or develop a new theory. I’d like to piggyback off existing tools as a way to visualize social networks over time, with analytical approaches that are qualitative and humanities-style.

If your institution has any researchers who work on algorithmic approaches to visualizing gappy data–probably in a computer science or library-and-information-studies department– I’m also interested in offering my data to them as an interesting set of real-world humanities materials. I’d love to help someone else figure out how to represent visually what I can’t know about, even qualitatively, based on the primary sources available to me. I don’t want to claim that my sources say more than they do, and I understand that some CS/LIS people are interested in these questions from their own disciplinary perspectives. (I learned a lot about this at MIT Hyperstudio’s h-digital conference last year.)

What I learned about my work:

I’ve apparently internalized the ideas Johanna Drucker gave in her talk at MIT last year. She argues that humanities people who think about our sources/texts through data visualization don’t need to try to be social scientists– and that we shouldn’t try. We have our own interesting, rich analytical traditions, and creative use of digital tools can give us new ways to apply that training. Here’s her analogy in a nutshell:

Anyone who’s ever studied public health, post-1800 urban history, or the history of medicine knows about the English epidemiologist John Snow, who pioneered the practice of mapping cases of epidemic disease as a way to figure out disease vectors. His maps looked like this:

Snow cholera map SM

The black bars extending back from the street at each address were cholera cases. Using this style of mapping, Snow figured out that the Broad Street pump (near the center of this map) was giving out cholera-infested water.

That is an early form of social science, and it was a distant ancestor of GIS-enabled epidemiology. Drucker, on the other hand, has a sketch that she uses, wherein each point on one of those city-block maps is transformed into a three-dimensional representation of a person. I wish I had a good image of it. This is a conceptual idea for her, not an implemented system, but it’s been fundamental to how I’ve started thinking about my future research.1

Distinguishing spatial history from other ways to use maps

For a spatially-oriented social historian, historical people aren’t solely dots on a map. They’re individuals living in families and neighborhoods and communities; doing certain kinds of labor; interacting with certain kinds of political systems, economies, and gender/race/class ideologies. They send letters. They write diaries. They scribble marginalia and family records in books, or maybe they don’t own any books. They read newspapers and see plays. Enslaved people in 18th-century New York talk to one another on their way to the neighborhood water pump, so much that their masters suspect them of plotting a slave rebellion. They’re born, grow up, get jobs, become disabled in industrial accidents, give birth, marry or cohabitate (or not), buy and sell sex, grow old, die. We read the documents and artifacts that they leave behind and try to reconstruct their world, and visualizations of that material are really good tools for thinking with.2

From a social-sciences viewpoint, the ways that some humanities scholars want to use mapping and social-networks ideas probably look about as methodologically rigorous as Florence Kelley and her colleagues’ maps of Chicago’s immigrant neighborhoods in the late 19th century. By modern standards of social-science method, their work seems relatively primitive, but historians still use the Hull House Maps and Papers whenever we want to understand late 19th century Chicago. That doesn’t make us social scientists manqué(e). It makes us scholars who want good visual tools, from whatever disciplinary context, that can help us ask and answer our own interesting questions.

Where do we go from here?

Early-career scholars who do digital-methods work in history and related fields: Let’s swap notes so that we’re all more likely to get those jobs and fellowships. Your comments are welcome on any of the following questions, either below or as links to your own blog posts.

  • When you describe your digital work/interests in CVs and cover letters, what phrasings are committees responding positively to?
  • What misunderstandings have you seen from hiring/fellowship committees about digital-methods work?
  • In our future interviews, how do you think we can address those questions more effectively when they come up?

Of course, if you’ve served on the other side of the interview table recently, we’d really love to hear from you too. Given the topic, pseudonymity and/or obscuring of identifying details is to be expected.

  1. Although I do understand that the NEH-funded Neatline project at the University of Virginia is building tools which look an awful lot like an implementation of Drucker’s ideas. That’s not a surprise, as Drucker was at UVA before she moved to UCLA, and UVA’s Scholars’ Lab has well-established institutional expertise in spatially-oriented digital projects.
  2. Yes, arguably this is the New Social History 2.0, or maybe it’s Cliometrics 2.0. Unlike the 1960s-70s, it doesn’t require roomfuls of punchcards and armies of mostly-male graduate students to come up with interesting conclusions. You do, however, have to know someone who’s doing it to understand that it’s possible.
20 Responses leave one →
  1. Bruce permalink
    February 10, 2011

    Being a geographer somewhere in the gray area between social science and humanities (although probably closer to the latter) …

    There is a pretty hot area of study in geography in qualitative GIS, usually/often coupled with a critical theoretical perspective (Critical GIS). Though I don’t do this work myself, I had a student who was working on mapping stories and video interviews onto GIS.

    • February 10, 2011

      Thanks for your comment. However, see my comment below about how rare geography departments are in universities with top history programs. (The published literature on geography as a discipline bears out my speculation.)

      So, extrapolating from that: if someone wants to go to the top history grad program that they can get into, they’re relatively unlikely to attend a university that has a geography department, which makes it hard to know that such a thing as “critical GIS” even exists.

      (I leave this as an exercise for someone else: cross-reference the top +/-20 graduate history departments, possibly by subfield, with universities that have even undergrad-level departments in geography. Let us know what you find. What GIS-using disciplinary departments do exist at those universities?)

      • February 10, 2011

        I can give the example of my university: we have a small geography dept (technically a cmte) that has a major but no grad students (geography.uchicago.edu). Nearly all courses, including the yearlong GIS sequence (available to undergrads and grads alike) is crosslisted and, hence, approached as inherently interdisciplinary; the head of the sequence takes pride in how he has taught students from nearly every corner of the university.

        As for learning about critical/qualitative GIS, as I was expanding my geography knowledge by reading critical geography journals like “Antipode,” I noticed that none of the articles looked like what we were producing in GIS class. So I simply started searching for geography articles about GIS that were also critical geography. That led me to work by Kwan and Pavlovskaya (as well as Pickles’s critiques from even earlier). A couple further steps led me to last year’s edited volume “Qualitative GIS” put out by Sage. So I offer myself as an example of someone from a humanities dept (English) who still learned about this subfield.

        As for your question about top 20 history depts., I have no idea where my school’s history dept. ranks. But I can say that every history grad student has the opportunity to take the yearlong GIS sequence, buy an ArcGIS license at a discount (which is unavailable to me), consult with the geography faculty on their projects, and participate in current debates in geography vicariously via journals.

  2. February 10, 2011

    I think you should take a look at the work done by historical archaeologists in spatial and landscape analysis. GIS has become an integral part of our analysis, but we also deal with a data set that is full of gaps and is inherently incomplete, while also dealing with historical data as well. A lot of the things you are discussing are things archaeologists are discussing. Historical Archaeology is on JSTOR now, so a couple searches there might get you pointed in the right direction. I’d also recommend James Delle’s work on slavery in Jamaica.

    • February 10, 2011

      Now that I think about it, my colleague who works on 15th century Europeans on the North American coast– and who’s the only dissertation-writer in my department doing heavy GIS work– is also reading heavily in archaeology. Thanks for your suggestions.

  3. February 10, 2011

    Hi, Shane — this is a terrific post, which could open helpful conversation not just among early-career scholars but with many of us who lacked training in spatial tools and methods as part of our graduate curricula in the humanities and are turning to them now. (On this, at our NEH Institute for Enabling Geospatial Scholarship, the Scholars’ Lab recommended a recent article in LLC by Martyn Jessop — on the roots of what he calls the “inhibition of GIS” in the digital humanities.)

    I want to comment on your footnote about Neatline, now proceeding with support from the Library of Congress and in collaboration with the Omeka team at CHNM. The Scholars’ Lab never benefited from conversation with Johanna on our spatial projects, but if you detect a common thread, I am certain it comes from my own partnership with her on our Temporal Modelling Project in the late ’90s & early Aughties. I do consider Neatline a direct descendant of that project, in which we prototyped timeline software that brought visualization to the early phases of DH content modelling (seeing viz as part of the thought process for knowledge representation, rather than as a secondary, algorithmic or analytical output.)

    • February 10, 2011

      Bethany, thanks for your illumination of the intellectual genealogies of Neatline. I’m interested to see how this conversation goes forward, as more people with spatial humanities interests hit the faculty job market and report on what they’re being asked.

      Link-parking for later (when I have time to read it) and for other readers:

      Martyn Jessop, “The Inhibition of Geographical Information in Digital Humanities Scholarship,” Literary and Linguistic Computing 23, no. 1 (2008): 39-50. doi: 10.1093/llc/fqm041 .

  4. February 10, 2011

    I’ve had my blogging and HNN work – not the same thing as the DH work you’re doing, but a lot of people think so – on my CV since I started, and I’ve almost never gotten a question about it. Despite putting it out there, sometimes even in my letters, people seem shy about talking about something that’s only marginally considered professional behavior.

    On the flip side, about the only DH stuff I ever remember seeing on a CV was teaching related – wiki projects, things like that. I have a colleague who was an early adopter of computers in history projects, but has let most of that fall by the wayside in the internet age. It’s hard to sustain. Still, we have a good core of people here who understand the value of technological experimentation: I’d expect DH work to get a good hearing from a committee in this department.

  5. February 10, 2011

    This is a great post, Shane. An essay that you might want to check out is Steve Ramsey’s “In Praise of Pattern,” from 2005. It speaks to a similar set issues, although Ramsey is most concerned with visual forms of statistical analysis and graph theory, not maps per se. Much like the John Snow example, Ramsey talks about how Leonhard Euler, the mathematician, sketched out the famous Konigsberg Bridge problem “in order to demonstrate the terms of his proof,” not– as we tend to assume– to serve as the proof itself. Ramsey goes on to say how most visualizations function so as to allow the viewer to see an otherwise familiar issue/problem/concern in a new light. They rarely offer a solution. Rather, they suggest new possibilities for– and point to possible implications of– future research. Ramsey’s conclusion is that the sciences and the humanities are more aligned, in their methodologies, than one might think. But the essay also might help to articulate some of the shared concerns– and the differences– between the disciplines that your project engages.

  6. Mike permalink
    February 10, 2011

    As a geographer that uses quantitative methods, I have to say that I would have the exact same response re: hypothesis and fishing. As it becomes easier to create maps more academic researchers are just mapping data for the sake of it. I’ve seen a ton of these projects and most don’t work out. I don’t think you need a hypothesis, but you do need a research question. Why are you mapping the data? Are you looking for certain spatial patterns or associations? Too, you don’t necessarily need to use quantitative methods to answer spatial questions. Geovisualization is a totally valid research method. I would recommend picking up an undergraduate level text book on map reading and analysis.

    • February 10, 2011

      Mike,

      You’re right that the post above doesn’t make clear what the specific nature of my questions is. I’ll try to clarify.

      My dissertation research explores the geographic spread of compulsory birth registration laws across US state-level jurisdictions– a topic that political scientists would refer to as policy diffusion, and which I mostly study through examining statutes and official records. I explore how the different political cultures of different states led them to hurry up or drag their feet in building effective bureaucracies for enforcing these laws. That’s one facet of my research, and that part doesn’t particularly call for mapping tools.

      The other big facet of my project is cultural and social history. I examine how ordinary people learned that they and their children needed birth certificates to interact with government and civil society. The correspondence corpus I use is letters from these ordinary people–all over the US, but particularly in rural areas– to government officials. I read them to determine how likely it was for people born in particular years and places to know about, and to be able to use, a birth certificate as an official identification document. This is the part that I’d like to use some mapping tools to get a handle on.

      I need a good way to be able to put hundreds of these letters onto a map, selectively display them by topic, year written, or demographic characteristics of the sender (year born, race, urban/rural birthplace or residence, etc.) I know from the other facet of my research that people born in cities were more likely to have access to birth certificates, but I don’t have a good sense of how that varied from state to state and over time. I’m hoping that a mapping project would give me a way both to explore that and to easily access groups of letters in a structured, visual way.

      Obviously, there would need to be a relational database underneath the map, and I could answer many of these questions by querying a good SQL database; but the ability to cross-reference a sender’s residence against historical census population-density maps would be very useful, and hard to do without GIS.

      One of the weaknesses of my education–and in my experience, that of many people educated in highly-ranked graduate history programs– is that relatively few of our universities have geography departments of any kind. The strong bias of elite private universities against perceived-to-be-”vocational” professional training means that a geology (or maybe architecture) department is where we’re most likely to be able to find courses that assign that “undergraduate level text book on map reading and analysis.” The GIS-related section in my university library, last I checked, was somewhere between 3 and 10 linear feet.

      Now that I think of it, it’s possible that the skeptical questions I referred to above have as much to do with that institutional background as anything else. That’s interesting to notice. Thanks for your comments and suggestions.

  7. February 11, 2011

    Thank you for this post Shane. As a second year grad student interested in GIS, I now have a great post and discussion to reference when bringing my interests to my adviser. I study economic change in Argentine soccer in the mid twentieth century and would like to incorporate spatial history into my project as a way to support my arguments on soccer’s role in the process of urbanization in Buenos Aires.

    I think Mike’s comment is spot on, and the questions he has suggested will be very helpful in formulating research questions for my own project.

  8. February 13, 2011

    The different approaches of social scientists and humanities scholars of spatial history was a topic highlighted in the Virtual Cities / Digital Histories conference that Bobby Allen and I organized last year, particularly in Ian Gregory’s closing remarks, which you can listen to here: http://virtualcitiesdigitalhistories.web.unc.edu/recordings/.
    It’s worth noting that the most interesting humanities projects don’t use ARC GIS software, the standard social science tool, but are built on new web friendly location-based tools such as Google Earth and Google Maps, which don’t offer the possibilities for statistical analysis that ARC GIS does, but do offer more web-friendly, accessible and user-friendly means of mapping and visualizing data of interest to humanities scholars. That’s the approach we’ve taken in the project in which I’m involved, Digital Harlem: http://www.acl.arts.usyd.edu.au/harlem/. So I’m with Johanna Drucker: in interdisciplinary settings to distinguish what we do from what social scientists do, rather than trying to articulate it in their terms. In what way we can also make clear the possibilities for collaboration that exist.

  9. February 16, 2011

    Shane,

    Thanks for the notice. The image of the Snow map with point of view will be published in DHQ as part of my From Data to Capta article, accepted and forthcoming. If I had the image file on my desktop, I’d send it to you. But it’s somewhere else and my file system is unruly.

    Best,

    Johanna

Trackbacks/Pingbacks

  1. Tweets that mention Spatial history & the interdisciplinary job market: some experiences | cliotropic -- Topsy.com
  2. Spatial history and interdisciplinarity | Brian Sarnacki | <!-- History Grad Student -->
  3. Weekend Reading: Etiquette Question Edition - ProfHacker - The Chronicle of Higher Education
  4. Links.Historische 02/12/2011 at Links. Historische.

Leave a Reply

Note: You can use basic XHTML in your comments. Your email address will never be published.

Subscribe to this comment feed via RSS