Thread by @benmschmidt

Here's a great work of digital history in the new issue of the AHR: "Networks and Opportunities: A Digital History of Ireland’s Great Famine Refugees in NYC," by @TylerAnbinder, Cormac Ó Gráda, and Simone Wegge. Instantly adding to spring syllabi. academic.oup.com/ahr/article/12…

@TylerAnbinder This dovetails perfectly with a conversation about quantification and digital history that I've been having with @Ted_Underwood, @jtheibault, @Zoe_LeBlanc, et al about digital and quantitative history, so let me draft a blog post as a thread here. (+)

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc I stumbled on this AHA piece when @jtheibault linked @HistDem's SSHA address about the "Revival of Quantification" in history. (video: ), This slide is the kicker--although Ruggles does later put the "true reviv[al]" of quant, social history in the future.

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem Theibault rightly asks if this disproves the argument of a 2019 AHA talk I just posted saying that "Time on the Cross" permanently exiled computational history in the US to the social sciences, not history depts. It's a good question. benschmidt.org/post/2019-12-0…

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem I regret saying "computational" history. Too vague. I really mean the type of stats-dependent work that plays an outsize role in the way literary studies journals think of "digital humanities." Call it statistical history, cultural analytics, or historical data science.

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem This article stands in fascinating relation to that conversation. It has 5 tables (though no 'graphs'), blowing way past Ruggles' threshold. And it's in conversation w/ the economic history literature, citing an 2018 NBER paper which cites their work. nber.org/papers/w25287.

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem And the authorship team here is itself interdisciplinary enough that you could reasonably claim this piece is cliometrics returned to the highest place in the historical profession. Ó Gráda was even a fellow of the Cliometrics Society in 2016!

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem But (you knew I had one, right?) it's also worth noting some of things that *aren't* in this piece outside titles in footnotes. For example, the following word stems:

1. model
2. statistic
3. regress
4. counterfactual (and not just the word.)
5. predict
6. coefficient

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem "Network" is in the title and conceptually at the heart of the piece. But apparently they had some issues with the Java versions on their systems before installing Gephi, because is emphatically your grandfather's "network," not the thing with nodes and edges.

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem Nor is there R, Python, or Javascript in the methods or anything about reproduction; the primary software used here is Microsoft Excel.

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem The big exception to this is related to the core analytical intervention of the abstract; after a paragraph defining correlation, they report 2 correlation coefficients indicating that emigrants who lived near others from their Irish home counties tended to make more $ in NYC.

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem (emend: 'half a paragraph defining...') The correlations are pretty good (R=~.65), but I don't know how firmly to follow the causal link. This comes right after noting that ex-urban immigrants are poorer: why not control for that? Or for trade?) Etc.

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem I'll leave this as an exercise for my data analysis class in the spring. It really doesn't matter. While today's historical rhetoric requires dressing up a database with a thesis statement like this, I don't think any historians will mistake this for the work's true contribution.

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem The real point is that this is another step in the effort to make use a major data resource the authors have created: lots of bank records on Irish immigrants. Those are online now thanks to NYPL, but I think they worked from the microfilm. digitalcollections.nypl.org/items/33091bd0…

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem This article seems to be the one that goes into the AHR rather than their 'methods' piece (tandfonline.com/doi/full/10.10…) because it makes a careful effort to interlink these data with carefully selected stories from the genealogical work they do, like this one.

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem And even in the data explication parts, the things most likely to interest readers are maps like this one, which run essentially parallel to the argument.

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem The supporting digital apparatus for exploring the individuals in the dataset with reference to their Irish *and* New York locations seems well done and quite digitally sustainable, with fascinating linkages to underlying documents.
…dragstoriches.digital.library.gwu.edu/?annotation=6&…

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem I'm really interested to see how people use it. But also note that it's all in a gwu url; things like this still can't quite fit into our publication models.

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem Summing up: this article is, good especially in the way that it finds different ways (linkages to stories, historical data visualizations) to explore data, not as a piece of statistical or quantitative work. Strip all that away, and there's a decent chance I'd never have seen it.

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem And on that topic--why didn't I see it? This was a digital article in the flagship journal of the historical association, publicized by the journal, but I only stumbled upon it because I happened to go to their webpage on the right day.

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem In the first three days up, it had 14 pageviews; I think at least two or three of those were me. Of course it's *good* if historians can break free of the culture of simultaneity on the web today, and later promotion and the print edition will find this its readers.

@TylerAnbinder @Ted_Underwood @jtheibault @Zoe_LeBlanc @HistDem But I also can't shake the notion that we're doing something a little wrong when there's so little fanfare around work that explicitly describes the way they want the resources to be used in the classroom, which requires moving beyond specialists alone.

End thread.