”Digital turn” vs. ”computational turn”

While having a hunch usually is enough to make a tweet about something (often hoping someone will either prove you wrong or right, or ask interesting follow-up questions), it is sometimes necessary to look into things to discover if your hunch has any validity.

In this case, my hunch is that digital turn and computational turn are (or rather have become) empty phrases used by humanities researchers to sound hip and aware about the fact that the internet happened and that digital literacy has an impact of our way to research and understand things. Maybe my hunch is more of a prejudice, based on hearing too many complacent remarks from newly turned digital humanists, and maybe I should just leave this as a rant with a TL;DR I am a bitter and prejudiced librarian at the end. But… let’s look at some data instead.

DHQ – a landmark digital humanities journal

Digital Humanities Quarterly is one of the most well-known digital humanities journals and possibly a good material for looking into whether the phrases digital turn and computational turn are used/over-used/under-used in the field in general. I downloaded a batch of DHQ content from their website (yay open access and general accesibility!) in 2019 and have all articles from vol 1 issue 1 to vol 13 issue 3 on my computer.

I uploaded these articles in an xml format to the excellent online text mining tool Voyant and was pleased to find Voyant was great at handling xml files (I’m making a note for future reference as I use this tool when I teach digital tools and methods at Uppsala University Library). I got some immediate results and couldn’t immediately grasp how the Phrases tool work so here I’m posting the results shown in the Contexts tool. (Quick note: This is just because I’m lazy/not overly invested in this topic, but if you really want to learn one of the Voyant functions in-depth they have GREAT DOCUMENTATION.)

The phrases aren’t super common in DHQ articles, it seems. Here they are, presented in the context (embedding code provided graciously by Voyant’s export functions).

Digital turn


Computational turn

So, not very convincing results to prove my hunch is right, but maybe an insight into the original of digital turn? I could definitely follow-up on that David M. Berry thread and try to look into some citation network data. I won’t, at least not now, but I could.

Searching for the phrase in Summon (ub.uu.se)

The question that popped into my head after seeing the above-mentioned results was: Maybe recognised digital humanities scholars are less prone to using these terms? Maybe they are used by researchers who are ”new” in the field, or those who are using the term in another, non DH context to sound impressive or pedagogical about the digital transformation of science and society?

The next thing I did may not answer these questions, but it does give some insight on the overarching academic disciplines where the phrase digital turn is most common. So, what did I do? Simple, I did a phrase search on ”digital turn” in the online catalogue of Uppsala University Library, Summon.

I received 1441 results on this search and decided to filter them according to discipline (a facet in the left hand pane of the search interface). The disciplines are quite detailed, so I did a manual sorting of the disciplines into the high level categories used in DiVA (you can read more about these at the DiVA wiki on Advanced search, look under Controlled searches > Category ID).

A few of the disciplines were hard to sort (recreation & sports; parapsychology & occult sciences; applied sciences; geography) so for these I had a quick look on the search results and decided on category based on the articles and/or journals that were represented in the results.

(Side note: parapsychology & occults sciences had 1 result, and led me to the fascinating journal Aries, which on the publisher’s web page has the tantalising description ”the first professional academic journal specifically devoted to a long-neglected but now rapidly developing domain of research in the humanities, usually referred to as ”Western Esotericism”.” The question I know we’re all asking ourselves now is: is the ghost of Sir Arthur Conan Doyle sometimes guest editor, or even *drum roll* ghost writer for this journal?)

While I have a whole spreadsheet with this information, I will just post some highlights here. You can also get the link to the search results. You can obviously have the spreadsheet if you like (just send an e-mail), I’m just too lazy to post it here now since it’s quite basic and therefore not worth the effort.

Okay, so highlights. The ”digital turn” usage per high level category is:

Category Number of search results
Agricultural and Veterinary Sciences 
Engineering and Technology 146
Humanities and the Arts 773
Medical and Health Sciences 74
Natural Sciences 69
Social Sciences 823
All 1887

As you can see, All amounts to a larger amount of search results (1887) than I previously stated (1441). My guess is that since I imported the numbers per discipline, some of the results will have been tagged with two or more disciplines, meaning they count twice as examples of a discipline in the table.

You can also gather from the table that the phrase ”digital turn” is way more popular in humanities and social sciences – together they account for 1596 of 1887 results, i.e. 84,6 %!

Why is this? Is the phrase ”digital turn” more fascinating in social sciences and humanities? Did the other sciences simply not encounter a digital turn, but rather a digital transition spread out over such a long period of time that digitalisation wasn’t received as something revolutionary and sensational?

Where is it? Where is the ”digital turn”?

The temporal aspect is also interesting – the first 40 results were published before 2000 (with the oldest actually being a record from 1898, from Medical Times and Register – apparently the phrase was found in the full-text (which wasn’t available to me), but I couldn’t find it, despite reading through all 8 pages (and learning a lot about surgical practices)). A time graph of the result list would be interesting to see, but I’m not yet sure of the best way to export the result list as an easy-to-work-with data set, so that will have to wait.

Maybe I’ll look into computational turn tomorrow.

Lämna ett svar

Din e-postadress kommer inte publiceras. Obligatoriska fält är märkta *