“Communities and Networks” is the theme for this year’s AHA meeting, so let’s talk about them from a digital humanities historian’s perspective.

In my own research I keep wishing there was a way that I could easily and dynamically visualize the various layers of relationship that exist within or across the groups of people I’m studying. Working on dissident Philadelphia Quakers during the American Revolution I’ve made some terrific discoveries by dipping into genealogical resources, for example: three of the dozen men involved in a Revolutionary-era business venture were closely related through marriage before any of it started because A had married the widowed mother of B while C had been married to B’s wife’s sister before she died quite young, perhaps in childbirth. And my interest in the business venture comes from noticing a correlation between individuals’ involvement in it (1775) and in a seemingly unrelated dispute on the religious front (1781). But that’s just one example within the community I’m studying, and the set of relationships quickly starts to get quite complicated. How do I see them all?

I’ve looked at (and experimented with) some genealogical software and it’s got room for lots of details as annotations, but is mostly geared toward showing trees: all someone’s ancestors or all someone’s descendants–which for my purposes is limited and flat. What I want to see is more like a rhizome or a social molecule or a Facebook for historians with multiple facets, different types of relationships, metadata, footnotes and visualization tools (and sure, GIS) built in.  And it’s not just about family connections, but all manner of social and cultural connections between people that may or may not end up being significant—but how will I know until I can see them?

Though the critical impulses may be similar, this poses a different methodological problem than working with specific texts via data mining or network analysis, because the bits of evidence that add up to layers of relationship are gathered from many different idiosyncratic and specific historical sources in a process that isn’t close to being mechanizable yet, if ever. (My research boundaries, in other words, are not delimited by a  set of texts or sources; given time and survival, everything that exists is fair game.)  And my interest at this point is less in displaying some final product on the open Web (though that’s surely a worthy goal as well) than in visualizing the networks so that I can make better sense of them, for the purposes of interpreting them to address historical-cultural questions.

I have some examples of experimental diagrams I’d like to share with interested colleagues to see if they’re interpretable or useful. I’d also love for us to talk about what historians might need in the way of purpose-built network-visualization software—and what we need to learn (in terms of computing, information theory, etc.) to create that most effectively. Or… are there projects out there that already address these needs? Jean Bauer, who I’m hoping will join the session, is working on Project Quincy and DAVILA. Is there existing multi-use software–I’ve wondered about DEVONthink for example–that can somehow bend to these purposes?

So, if any aspect of this ethnographically-inspired digital-historical prosopographical fantasy appeals to you–how to specify it, build it, use it, improve it–I’d love for us to chat.

(An earlier version of this session took place at THATCamp Virginia in December 2010.)