Humanities, meet the sciences!

Karen Hellekson's picture

A recent spate of research, which has included some data input into Zotero, has only reaffirmed my belief that the sciences can teach the humanities much. I’m not just talking about quick peer review turnaround times and wait times to publication that don’t stretch into years. I’m talking about something simple, something basic: abstracts and titles.

Admittedly I am coming at this from the point of view of an unaffiliated scholar. Getting access to texts is a huge chore. I can’t just magically obtain something and flip through it to see if it’s what I need. I have to research it first, then decide if I want it, and then decide if it rates being one of the five books I can request at one time. I can’t possibly be the only person who wishes that I could figure out what something was about without actually having to read it.

Heed my call, journals and scholars in the humanities! Abstracts and titles. Please, I beg you, make them count. Let’s follow the example of the sciences here.


Abstracts in the humanities are relatively rare but are becoming more common now that many journals are transitioning online. The library databases I see squish all the data into a common presentation format, which includes a space for an abstract. Good start! Of course, if no abstract originally appeared—as is all too common in the humanities—we instead get a useless one- or two-sentence summary that some poor data-entry sod had to write up when submitting the data to the aggregator.

Contrast this with the sciences. They provide abstracts with everything, not to mention maybe also keywords, a two-sentence precis, and a “what’s new” box that indicates what this text is adding to the literature. You want to know the point of the article? It is right there. You want to know what this adds to the field? They say it in actual words. There are two kinds of abstracts: unstructured and structured. Unstructured abstracts are just narrative text, maybe 250 words long max. Structured abstracts are divided into sections such as background, methods, results, and conclusion, thus neatly summarizing the entire study.

Were the humanities to more widely implement abstracts, I imagine that most would use unstructured abstracts. But allow me to humbly suggest heads for a structured abstract: background, thesis, methodology, and significance. Background and thesis ought to be self-explanatory. By methodology, I mean the approach (queer theory; posthuman analysis) and an indication of the sort of study (close analysis of a single text; compare–contrast; manuscript analysis; online survey; application of existing theory to a new text). By significance, I mean what this is adding to the field.

At the very least, constructing an abstract along these lines will really focus the writer, forcing her to actually have a point. This can only be good. In addition, the information provided about approach/methodology will help scholars contextualize each other’s work, especially if people are working across disciplines.

The abstracts are then ideally put online, for free, to be endlessly reproduced by anybody who wants to, so that access to a locked-down library database is not necessary for one to learn of the mere existence of this bit of scholarship.


In the sciences, article titles are actually meaningful. Some titles, usually the ones that use verbs, actually summarize the results of the study: “Defective IL-10 signaling in hyper-IgE syndrome results in impaired generation of tolerogenic dendritic cells and induced regulatory T cells” (Saito et al., J Exp Med, 2011;208:235–249, chosen randomly). Were I a cell researcher, I would know instantly whether this paper was relevant to my current project.

Humanities journal article titles, on the other hand, tend to be informative only in their subtitles. The foretitles are often little discursive dances, often a pithy quotation.* As an extreme example, check out the table of contents for the March 2010 issue of PMLA (125, no. 2).† Out of the five titles provided, only one has the actual topic of the paper (Beowulf) listed in the foretitle. Here are the foretitles of the other four: “Swollen Women, Shifting Canon,” “‘As a leaf on a branch…,’” “Clustering and Curling Locks,” and “Beyond Sacrifice.” Go on, give a guess as to what these papers are about. No, try. Give up? From what I can infer from the subtitles, the topics are, respectively, midwifery and genre of the romance lyric; Dante; Paradise Lost; and Milton.

I call for boring, representative paper titles because the titles will reveal the topic, and what with the endemic lack of abstracts, we need all the help we can get. Many databases truncate titles after the foretitle, so the subtitle is not reproduced and is not searchable. The foretitle thus ought to include the actual topic of the paper. Think of it as a dense keyword dump. Something like this: “Midwifery, the Romance Lyric, and Tenth-Century Occitan Poetry.”


I’m not calling for keywords in the humanities, despite their extensive use in the sciences, because they aren’t that helpful without an infrastructure in place that would permit them to make meaning. The Web can do that now. If authors had titles that actually reflected the topic, as well as useful abstracts, keywords would be superfluous because hits via search engines would actually work, assuming the titles and abstracts were available for free online (and that’s a big if). Keywords in the sciences and in medicine are often related to subject terms, like MeSH terms, and assigning MeSH terms is so difficult that professionals have to do it.


I don’t think I’m asking for anything radical here. I’m not sure why the humanities has created an infrastructure so unfriendly to actually finding out what is going on in individual bits of scholarship. Of course, my mantra is, “What’s the point?” And if I can’t figure out what the point is, I move on. I thus urge writers in the humanities to not make me work so hard. Explaining yourself will also help justify the importance of your research—and if you can’t, well, maybe that bit of scholarship doesn’t need to be disseminated.



* In foretitles that are quotations, the quotation ought to be immediately recognizable: a famous line, a cliche, whatever. And if it’s not, the quotation ought to be used and sourced somewhere in the paper itself. Otherwise, it’s just puzzling. What is this quotation I have never heard of, and why are you using it?

† I am only picking on PMLA because I think it can take it. Apologies to the authors of the papers for implying disapproval; I’m confident that they had lots of fun coming up with their titles. Kudos to PMLA for having its tables of contents and abstracts available for free online.

This text is copyrighted under a Creative Commons Attribution 3.0 Unported License. This post was originally written on March 10, 2011. It may be freely copied anywhere. If you read this document at a site other than its original, I may not see any comments you might append, and I’d love to hear from you. Please comment at the original blog post if you wish me to see your remarks.

Filed under: essay

Karen Hellekson

Publication date (from feed): 

Thu, 10 Mar 2011 19:17:09 +0000