Some ruminations about using digital history tools to answer historiographical
question. The central query of my book, The Politics of Women’s Culture,
addresses one of the pivotal historiographical questions of women’s history, the notion of “women’s culture.” I’ve
obviously been working my way along using the traditional tools of the
historian, and so far I’ve argued that the tangled origins of the concept help
to explain its fraught usage in the field.
Although I feel I’m on strong ground, it seemed to me that
this issue represented a perfect topic for the application of different digital
history tools.
I began by attempting
to use google ngram to trace the trajectory of the term's usage. That didn’t work so well due to problems with
the digitized texts. I then turned to
Bookworm with fairly similar results.
My next foray involved using JSTOR to trace the rise and
fall in several different periodicals of related concepts.
That proved satisfying in some ways but not in others.
Finally, I began messing around with various text visualization methods.
Finally, I began messing around with various text visualization methods.
I feel like I’m on solid ground in terms of the origins of
the term, but now I want to dive deeper, to use topic modeling to parse out
greater associations say between “power” and “women’s culture,” or “politics”
and “women’s culture” which is proving to be one of the central issues
underlying the much-disputed concept.
Finally the tools started helping me get the answers I
wanted. Because David Newman
collaborated on the closest project I could find to what I wanted to do (Block and Newman (2011). What, Where, When and Sometimes Why: Data
Mining Twenty Years of Women's History Abstracts. Journal of
Women's History), I used
his very nice java based topic modeling tool.
Here are the results for the famed 1980 Feminist Studies Symposium, Politics and Culture in Women’s History, with
articles by Ellen Dubois, Mari Jo Buhle, Temma Kaplan, Gerda Lerner and a
response from Carroll Smith-Rosenberg.
I’m still learning how to interpret topic modeling, and I’m
still not sure how valid it is for a document of this size, but the
associations suggested by the clusters are intriguing to me, especially #3 which
indeed as I expected associates culture with power suffrage and oppression. My textual analysis reveals
that to be the central debate underlying all the articles. What is power and how do feminism and culture
relate to shifting power relations?

No comments:
Post a Comment