4 Responses to “The Old Bailey Proceedings: Big Data and the Expansion of Research Methods”

  1. Nick Sacco February 23, 2013 at 7:16 pm #

    Reblogged this on Exploring the Past and commented:

    Check out the full post at the IUPUI Digital History Blog!

  2. angelabpotter February 24, 2013 at 8:21 pm #

    Nick-
    I thought the op-ed that you linked to was brilliant and I hope we can discuss in class. I know that there are some tools that can hope to reach into some of these areas (such as social meaning) but to date they are not quite satisfying.
    Really thought provoking!
    Angie

  3. MDKenny February 25, 2013 at 8:39 pm #

    While I appreciate your point about the limited applicability of big data sets to micro-historical inquiry, I don’t share your enthusiasm for Brooks’ contribution. His point about Dante having met Beatrice only twice, for example, ignores the possibility that digital evaluation of Dante’s writing about Beatrice may be able to tell us more about what the author thought and felt about her, and about his sense of the significance of those meetings.
    As digital humanist and Professor of Italian at the University of Virginia Deborah Parker points out in her Guidelines for the Evaluation of Digital Humanities Projects, thanks to projects such as The World of Dante, (http://www.worldofdante.org/) “students can now quickly peruse the myriad ways in which recurring characters such as Vergil, Beatrice, or God are named in the poem. Rapid retrieval and organization of such data facilitate considerably our understanding of how Dante employs a wide range of expression to construct a character or characterize a place.” [1]
    This may seem like a minor point, but given Brooks’ ability to survive as a professional pundit in spite of his commitment to drawing conclusions that run counter to the available facts, I have a hard time taking seriously his perspective on what data is and is not good for. Take, for example, his conclusion that big data is not well suited to solving big problems:
    “If you are trying to figure out which e-mail produces the most campaign contributions, you can do a randomized control experiment. But let’s say you are trying to stimulate an economy in a recession. You don’t have an alternate society to use as a control group. For example, we’ve had huge debates over the best economic stimulus, with mountains of data, and as far as I know not a single major player in this debate has been persuaded by data to switch sides.”
    That he chooses to stick to his gut feeling that, in spite of ample evidence to the contrary (see Reagan, Bush II presidencies), major tax cuts for the wealthiest class help to strengthen the economy and result in a trickle-down effect, is not an argument for why big data is not helpful, but it may be a powerful argument for why David Brooks’ ruminations are largely ideological.
    His analogy with the haystack is also misleading: The haystack of historical data is already quite large. The question historians have to grapple with is how to construct plausible and illuminating arguments about the existence of patterns within the data. Historians expect data to lead to a variety of conclusions, given the perspective and the goals of each individual researcher, but it is the arrangement and presentation of the available facts around an argument that we have to evaluate. Digital methods allow researchers to work more easily at a variety of scales, but do not fundamentally change the size of the proverbial haystack of data.
    It seems to me that Brooks arguments, as usual, are self-serving. His anonymous banking magnate, who made a gut decision to stay in Italy despite the data, was determined to be a good guy, even if it meant the bank would suffer some short-term losses. Likewise, ten years ago this month David Brooks was just “try[ing] on the potential mentalités of our ‘friends’ and ‘allies’,” when he argued that French and German critics of the false U.S. intelligence on Saddam Hussein’s chemical and biological weapons program were not responding to the evidence or lack thereof, but were perhaps “simply not brave enough to confront Saddam,” or “would rather see American men and women – rather than French or German men and women – dying to preserve their safety. [2]
    But more than just self-serving, Brooks’ presentation is also internally inconsistent: In the column you cite, he claims that “data can help compensate for our overconfidence in our own intuitions and can help reduce the extent to which our desires distort our perceptions,” but then goes on to argue that “data obscures values,” “struggles with context,” produces false correlations – and that those correlations grow exponentially false as more data is collected – and is not generally useful in swaying opinions of experts on large problems. I suspect Brooks is not subject to rigorous editorial review at the Times, or else he might have had to answer questions about how this all squares up, or what purpose his concluding line served other than to lend some credibility to his ideas by associating them with someone who actually deals critically with data and its potential for analyzing and solving problems. [3]

    [1] Deborah Parker, “Guidelines for Evaluating Digital Humanities Projects,” ADFL Bulletin 41.1 (2009): 71. http://bit.ly/Zvuoka

    [2] David Brooks, “French Kiss-Off,” The Weekly Standard, February 6, 2003, accessed February 25, 2013,
    http://www.weeklystandard.com/Content/Public/Articles/000/000/002/215jfyfl.asp.

    [3] Edward R. Tufte, Data Analysis for Politics and Policy (Englewood Cliffs, N.J.: Prentice-Hall: 1974). http://www.edwardtufte.com/tufte/dapp/

    • Nick Sacco February 25, 2013 at 10:18 pm #

      Matthew,

      I appreciate the time and effort you’ve taken to share some of your thoughts and clarify some points regarding the David Brooks article. I think it merits repeating that I’m not coming from a point of expertise when it comes to discussing big data and textual analysis. I am not a regular reader of David Brooks, do not subscribe to all of his views, or consider him as the “authority” or an “authoritative” voice in the big data discussion. I cannot determine whether or not Brooks’s writing is “self-serving” because I just don’t have the sample size and understanding of his views to make such a claim. Furthermore, I had no idea the “World of Dante” existed, and had I known that it did exist, perhaps the blog post would have been different.

      I was asked to analyze some of the shortcomings of big data and textual analysis, and I attempted to use the Brooks article as a means to help me express some of those views. In sum, if I wasn’t asked that question, Brooks wouldn’t have played any part in my essay. There are probably some defects in Brooks’ argument, but I still firmly believe that big data sometimes struggles to help us understand the quality of relationships between people. The Brooks article helps provide a starting point for discussing these topics, as evinced by your passionate response. By all means, if you know of articles and essays that exist that do a better job of helping us understand some of the shortcomings of big data and textual analysis, share them with us.

      David Brooks aside, I hope MY content was worthy enough to be taken as serious scholarship and that I was able to address the questions posted in a decently effective manner.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: