It started on Twitter – doesn’t it always? Ted Underwood tweeted:
Highly recommend this, especially the ¶ around "DH should be wary of promises of ease." https://t.co/qSdoTemTMh— Ted Underwood (@Ted_Underwood) January 3, 2017
Which pointed to Andrew Goldstone:
“Teaching Quantitative Methods: What Makes it Hard”: my Debates in DH essay, and some liner notes. https://t.co/RzUVEF8UnT— Andrew Goldstone (@goldstoneandrew) January 3, 2017
I took a look at Goldstone’s piece:
I have argued that teaching quantitative methods is hard, but I am not suggesting that it needs to be made easy. On the contrary, DH should be wary of promises of ease: in prepackaged tools, in well-meaning introductory tutorials and workshops that necessarily stop short of what a researcher would need to draw conclusions, in rationalizations of inconclusive arguments as exploration, play, or productive failure. Having organized workshops, produced tools, and talked elsewhere about data-exploration myself, I’d like to think the desire for ease is understandable. These approaches can be useful in making the unfamiliar more accessible. Nonetheless, I learned from my students to be more skeptical that the easeful way can naturally lead to the production of new knowledge, because a gentle introduction simply cannot get very far in developing ways of interpreting quantitative results. Good programming skills actually exacerbate the problem when they make highly complex transformations of data—matrices of frequency correlations, probabilistic topic models, and so on—simpler than figuring out whether those transformations are appropriate to the data at hand. 
And Goldstone goes on to enumerate more problems. I gave it a quick read and offered up my 2 cents:
@bbenzon @goldstoneandrew In a way it seems obvious. But then the awkward question is, how/where *can* we teach that to literature majors?— Ted Underwood (@Ted_Underwood) January 3, 2017
And we’re off.
When and where
Competence in digital humanities methodology cannot be conveyed quickly whether we’re talking about educating undergraduate majors, graduate students, or full-time faculty. I got my doctoral training at SUNY Buffalo in the 1970s where, in lieu of the traditional competence in two (or even three) foreign languages, you could present competence in a minor field. That option is still on the books:
Candidates may also create a Minor Field, broadly defined as an area of knowledge or intellectual discipline other than English or American Literature. The range of possibilities is vast, and the only requirement is that the Minor Field have some defensible relation to the student’s dissertation.
I presented psycholinguistics, which I worked up in the Linguistics Department (on top of various undergraduate courses). I would assume two or three DH-relevant courses could now be slotted in there.
Just what those courses might be, who knows? One or two might by taught by the English Department. SUNY Buffalo also has a program in Library and Information Studies within the Graduate School of Education, and departments of computer science and sociology. Appropriate courses could easily come from there.
That, of course, is just one institution, but you get the idea. Educating undergraduate majors would be similar, draw on courses from various departments. Educating post-graduate scholars who already have full-time jobs, or a slate of adjunct positions, is a different matter. Summer programs is one answer, and that’s going on. Just fit things in wherever possible.
Think like a sociologist
Thinking like a sociologist – a theme Andrew Goldstone has been sounding for awhile  – may not come easily to someone whose core training is to literary critics. Sociologists are interested in causal explanations. Literary scholars, not so much.
Consider the passage from a recent article by Sharon Marcus:
Explanation designates the operation by which literary critics assign causality, though explanation can also signify description and interpretation, as when we “explain” a poem. Literary critics tend to downplay causality — “why?” is not our favorite question — and usually refer the sources of a text’s meaning or form to disciplines other than literary criticism, such as history, biography, economics, philosophy, or neuro- science. Thus scholars often relate specific features of literary works to general phenomena such as modernity, capitalism, imperialism, patriarchy, or the structure of our brains. But because explanation is an undervalued operation in literary criticism, one seen to depend on the kind of literalism that leads many critics to reject description as impossible, the exact nature of the link between general phenomena and specific works often remains nebulous. Literary critics are more likely to posit the relationship between the realist novel and capitalism as one of homology, analogy, or shared commitments (to, say, individualism) than they are to trace a clear line from one as cause to the other as effect. (pp. 304-305 )
Perhaps literary scholars need explicit training in causal thinking, in sociological theory and method.
As an undergraduate at Johns Hopkins I took a course in sociological theory from Arthur L. Stinchcombe, who maintained that, given some phenomenon to be explained, a good social theorist should, within an hour or two, be able to come up with: 1) two or three possible explanations, and 2) a half-dozen or so observational tests that would allow you to distinguish among the explanations. Our term-paper assignment was to pick some phenomenon, real or imaginary, come up with three alternative explanations, and propose observations that would discriminate between the alternatives. Why’d he allow us to pick imaginary phenomena, you know, make stuff up? Because he was more interested in how we thought than in actually executing the research.
A year or so later he published a book, Constructing Social Theories , that makes fascinating reading, though simply reading it will not make anyone a social theorist. Here’s a paragraph that floored me.
A second graduate-school conversation, with Philip Selznick, shaped my attitude toward what theory is for. He remarked that one felt satisfied that he understood something when he could summarize in a sentence the guts of a phenomenon. He gave the illustration that he felt satisfied when he realized that the achievement of the Bolshevik parties was “to turn a voluntary association into an administrative apparatus.” To use, as a criterion of judgment, the guts of a phenomenon—what is going on—is better than to use any logical or formal criterion.
There you have it, a single sentences, but the guts of a phenomenon. And not just sociology, but anything.
How often do you sit down and, in a single line, capture the guts of something you’re working on? Maybe you write it out in 200-300 words first and then, thinking about that, write that single line.
We’re now far away from the nuts and bolts of fitting another course, workshop, or book into a busy schedule. We’re now thinking about intellectual craft in a fundamental way. If anything is to come of digital humanities, it has to work at that level of craft. It cannot be just a high-tech ad-on to pad out a resume or attract grant funds.
 Andrew Goldstone, Teaching Quantitative Methods: What Makes It Hard (in Literary Studies). Forthcoming in Matt Gold and Lauren Klein, Debates in the Digital Humanities 2018 (University of Minnesota Press). Online PDF: https://andrewgoldstone.com/teaching-litdata.pdf
 Social Science and Profanity at DH 2014, blog post, July 26, 2014, URL: https://andrewgoldstone.com/blog/2014/07/26/dh-soc/
Distant Reading: More Work to be Done, blog post, August 8, 2015, URL: https://andrewgoldstone.com/blog/2015/08/08/distant/
 Sharon Marcus, Erich Auerbach’s Mimesis and the Value of Scale, Modern Language Quarterly 77.3 (September 2016) 297-319. pp. 304-305
 Arthur L. Stinchcombe. Constructing Social Theories. University of Chicago, 1968. URL: