• If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • You already know Dokkio is an AI-powered assistant to organize & manage your digital files & messages. Very soon, Dokkio will support Outlook as well as One Drive. Check it out today!


Workshop (2015-07-31)

Page history last edited by Alan Liu 8 years, 10 months ago

Progress to Date (and Future Scheduling)



  • Next week (Project Development Calendar):
    • Monday (Aug 3), 3 pm -- meeting of the scraping crew. In preparation for this meeting, please do the following:

      1. Begin thinking about what our priorities should be in collecting other publications beyond the ones we already have.  For example, here are some possibilities:
      * Other U.S. cities (e.g., Washington Post)?
      * Other nations: Canada, Australia New Zealand, India?
      * Online media news/popular media
      * Social media
      * Middlebrow (e.g., USA Today) vs. highbrow (e.g., New Republic, LA Review of Books)?
      * Economic press (e.g., Forbes, Business Insider, The Economist)?
      * Higher-education press (e.g., Chronicle of Higher Education, Higher Ed)?
      * Campus papers (e.g., Harvard Crimson, Yale Daily News, UCLA Bruin)?
      * Commencement speeches.
      * Articles on "sciences"?

      2. Do some quick reading with human eyes in the NYT and other materials we have scraped (sampling, of course). The purpose is twofold:
      * To begin adding to our scrubbing list ("List of Fixes Needed for Raw Texts") for preprocessing the works;
      * To get an overall gestalt for what we have, which will help guide us in discussion of topic modeling strategy and ultimate goals.

      Wednesday (Aug. 5), 1 pm -- Planning meeting (joined by Scott Kleinman). The goals of this meeting will include:
      * Discussion of scrubbing and other preprocessing steps
      * Discussion of topic modeling strategy
      * Discussion of ideas for eventual public-facing interface


The U.K. and Commonwealth Nations Problem


  • Email from Melissa Terras:

    Humanities is very much an Americanism, yes. We'd call it the Arts in the UK until recently, when Humanities started to creep in (I just checked my alma maters - Glasgow: http://www.gla.ac.uk/colleges/arts/, Oxford has now changed to the "Humanities Division" but that is a relatively new restructure - http://www.humanities.ox.ac.uk/ - UCL is now Arts and Humanities, again, a fairly recent reshuffle away from just arts).

    The distinction is that english, history, etc are "arts degrees" but fine art is an "art degree" if that makes sense - small linguistic variance. So if something is described as "arts" then it = humanities, but if it is described as "the arts" then it is arts + art, and if something is "art" is it just art practice. Funny displinary shift going on with now calling "arts" humanities, coming over from you guys.

    So I wouldnt be searching for "the arts" I would be searching for just "arts" or "faculty of arts" or "school of arts" (or even "of arts").

    The arts in UK english isnt just specifically music, painting, etc - that is how it is defined in american english, but its different over here!

    Minority status also points to the small number of practising humanities scholars there are over here, compared to practising artists, or scientists.

    Hope that helps, clear as mud!

  • Email from Andrew Prescott:

    Indeed, the concept of the humanities was an alien one in Britain until the 1990s. I’ve always regarded it as an American imposition, I’m afraid, and view the term with deep suspicion. Even though I’ve used the term ‘humanist’ I’ve never understood what it meant after the sixreenth century. I studied the classic combination of History, English Literature and Geography at A Level in a South London grammar school from 1970-1972, and I was firmly in the ‘arts’ stream - the alternatives were science or economics, no mention of the humanities. When I was considering a university degree, the discussion was always whether to go for an arts degree or a science degree. I haven’t read CP Snow on the Two Cultures for some time, but I suspect the polarity there is arts v sciences. The humanities was a concept which as a scholar I hadn’t really heard of until the mid 1990s, when discussions about setting up a research council for the area were led by the British Academy. This immediately raised the problem of distinguishing these discussions from the work done by the Arts Council, and it was this definitional need whuich really first brought the term humanities into common parlance in the UK. This was confirmed by the creation of a Humanities Research Board by the British Academy in 1998. The growth of activities like Humanities Computing and the creation ofv services such as the AHDS added to the currency of the term. Some responsibilities were transferred from the Arts Council to the AHRC on its inception, and this made the bureaucratic need of the research councils to distingisgh betweeen the arts and humanities more pressing, and it is reallty from that point that the term started to be commonplace in the UK. I’m fascinated that Alan’s text mining seems to confirm my impressionistic memories, and reinforces my view that the term humanities is not a helpful one for us, and that the older usage of arts is better. The objections to humanities would be not only that it is a piece of thoughtless Americanisation, but also that it suggests stronger inks to the social sciences than are in fact there and that it isn’t a term understood by the general public in the UK.


  • Email form Stefan Sinclair:
    For what it's worth, I think in Canada we're stuck somewhere between our commonwealth heritage and our American overlords, like usual :)


  • Email from Alan
    Thanks, Melissa, Ted, Stefan (and, by extension, Andrew too):

    I'm going to bring all this into my meeting with my summer RA crew this afternoon to think about. We're going to try to figure out how to "operationalize" this issue at least in the limited way of collecting articles mentioning "arts" in UK, Canada, Australia, New Zealand, and probably India too (but not in the U.S., where "arts" has a different, more specific usage that would redefine the problem we are researching).

    It's becoming clear to me, though, that we're onto a fascinating related research problem (the rise of such general level terms as "humanities and arts" in public and institutional discourse in tandem) that my current project of collecting a corpus focused on "humanities" is not well designed to study. (This is somewhat analogous to the rise of "media" as a general concept around the time of McLuhan instead of a species-level concept [different kinds of mediums].)

    The general problem is related to the argument of Geoffrey Harpham's book, _The Humanities and the Dream of America_, which studies a sequence of influential foundation, government, and other documents from the 1960's on (leading, for example, to the establishment of the NEH and NEA) that raised national awareness of the humanities as part of the Cold War/Sputnik moment.. (I.e., Pride in the humanities was the logic that the U.S. needed to distinguish its arms race from the Soviet arms race as a defender of civilization.) So Harpham's thesis is that "humanities" in the U.S. rises in prominence after the 1960s as a policy and institutional discourse. I don't recall, though, that he has anything to say about how that is related to public, media discussion of the humanities.  To research the hypothesis that "humanities" as a genus concept rose in official and public discourse together from the 1960s on, we'd need a much more comprehensive corpus that would allow us to identify topics or clusters around such specific themes as "literature," "history," etc. and also the genus-level themes. Google Ngram viewer wouldn't work exactly because that's books. Corpus linguistics would be another approach, but of course the nature of a "corpus" in that field is different. It'd be interesting to use both Google books and the various American and British English linguistics corpora to get a quick snapshot of what we are dealing with, however.


  • Emails between Alan and Ted Underwood:
    • Alan: Perhaps the best solution is to collect for the commonwealth nations all articles mentioning "humanities," "arts" (and also "liberal arts"), and then hope that topic modeling will naturally separate out the two senses of "arts"? (The "the" is another problematic piece of the puzzle.)


    • Ted: I think the key open questions here have to do with how you're thinking of using this corpus downstream, especially what kind of cross-national comparisons you're thinking of making. And that's hard for me to anticipate.

      But in general your plan sounds great. Topic modeling is supposed to catch exactly the sort of nuance/polysemy Melissa mentions, without any need for hard-coded distinctions between e.g. "arts" and "the arts." If there are articles mixed in the corpus that are off-topic for the problem that interests you, there are pretty good odds that they'll wind up primarily associated with a topic that makes their off-centerness legible, and you can use that process to tighten the focus if you like.

      Starting with a relatively inclusive set of search terms also seems wise since there are many other ways to tighten later -- e.g. you can always re-run your original "dragnet" on your own corpus with a tighter filter if you want to explicitly finesse/filter out "the arts." Tends to be harder to expand a corpus if you've started narrow.



Today's Workshop (options):

  • Scraping Work
  • Inspection Work
  • "Reading" of corpus
  • Researching other publications to collect


Comments (0)

You don't have permission to comment on this page.