Guest blog post: LaTeX for the humanities

Introduction

Having used LaTeX in humanities research for nearly 8 years, I was recently asked by Overleaf to author a blog post about my experiences with using it in a field that is usually not associated with LaTeX. This post should be of interest to non-users of LaTeX who want to know more, but also for community stakeholders (such as Overleaf) who want to expand their user base.

LaTeX is for “tech-heads”, right?

First off, let’s deal with one of the most common misconceptions regarding LaTeX. When I first started hearing about LaTeX (some 9–10 years ago) it was always with the connotation that LaTeX is for mathematicians, engineers and computer scientists: people who need to display advanced mathematical formulas and algorithms. This preconception is still present among the researchers I work with today, which includes linguists, engineers, roboticists, philosophers, and social scientists. It’s easy to see why this view is so commonplace: my own university, for example, organizes LaTeX courses each semester—but only for engineers. Many of the templates you can find on Overleaf are for technical journals and conferences published by, for example, the IEEE and ACM, and the ability to show advanced mathematical formulas is often at the top of the list of why you should use LaTeX over traditional word processors.

While it’s true that LaTeX has its origins in computer science, and that a large part of the userbase are from technical fields, many of the advantages offered by LaTeX are just as relevant for linguists, sociologists, and interaction analysts. I hold degrees in language and communication and have never had the need to display mathematical formulas in any of my papers, yet I find LaTeX to be indispensable to my work.

No, LaTeX is for authors

Long-time users of LaTeX will easily recognize some of the advantages of LaTeX. For example, that it separates style from content; it handles large documents well; it handles references, cross-references and design decisions consistently, and it is compatible across platforms and versions. Many of these features are just as relevant for researchers working with social interaction or second language learning as they are for researchers working with biochemistry or mechatronics.

• Consistency: LaTeX handles most textual and graphical elements consistently throughout the document. Word processors also have the capability of producing consistent documents, but at least the people I work with don’t make use of this. For example, a few weeks ago I edited a paper I co-wrote with 3 other authors: the Word document featured 3 different font faces, 3 different font sizes (for similar elements that is), and 2 different styles for figure captions. When we had to resubmit the paper to a different journal we hired a student assistant to reformat all the references. Speaking of references, I also recently helped a colleague convert an article written in Word to LaTeX. During this work we found 13 in-text references that were missing in the bibliography. All of those problems could easily have been avoided and solved by LaTeX, and saved us precious time.
• Stability: While there are features in word processors that make it possible to modularize documents, most people (such as my students and colleagues) don’t use, or even know of, these features. When these documents grow into large theses and dissertations they often become unstable. A colleague of mine had this problem when handing in her doctoral dissertation, which she blogged about here.
• Flexibility: The separation of style and content is a known procedure in communication design. Specifically, we teach our design students to separate the content from the information design and from the graphical design of any publication (flyers, websites, posters, books, etc.). This approach is in place so that we are able to assign the right people for the right job, and to make it easy to reuse designs and content without them being dependent on each other. The same principle applies to academic writing as well.

Thus, the philosophy upon which LaTeX is built is not completely foreign to the humanities—in fact we often teach it to our students. What we’re not so good at, though, is applying the same philosophy to our own written outputs. So, why don’t we?

I want to use LaTeX, but I can’t!

Making the transition from Microsoft Word to LaTeX isn’t an easy one. The obstacles I most often encounter are:

1. LaTeX looks a whole lot like a programming language;
2. Collaborators and co-authors use Microsoft Word;
3. Many journals in the humanities do not have LaTeX templates, and even if they do, editors and production teams ask for Word files.

LaTeX looks like programming and I don’t do programming!

For a non-programmer LaTeX does look a lot like a programming language (hey, maybe it is—what do I know, I’m not a computer scientist), which tends to scare off some users. Here, I think that we often focus too much on the syntax and too little on the workflow. The difference between learning a new syntax and learning where to click, and when, is arbitrary. One may seem more “intuitive” or “natural” only because we’re more familiar with it. The real difference between WYSIWYG word processors and LaTeX is, I think, to be found in the different workflows. For example, for good or for worse, LaTeX forces you to make decisions about structure, design and layout, but offers consistency in return. Word processors make no such demands, but in return it’s up to you to make sure that your design decisions are consistent. In my experience most users fail to realize this “responsibility” and end up with messy documents.

Collaborators and co-authors use Word

Learning how to use LaTeX is one thing, but dealing with co-authors, supervisors, and other collaborators who don’t, is perhaps an even greater challenge. However, there are several resources available to make collaboration between LaTeX and non-LaTeX users a bit easier:

1. Share PDFs rather than doc files. Many PDF viewers have fairly advanced features for commenting and editing text in PDFs.
2. Convert PDFs to Word. This is possible in Adobe Acrobat Pro, and possibly also other viewers. The result is far from perfect, and certainly not anything that is ready for publication, but good enough if collaborators insist on a Word file. However, re-integrating “tracked changes” from a Word file back into a LaTeX source file is arduous work.
3. Overleaf’s Rich Text editor. The editor does a decent job at rendering documents in a WYSIWYG fashion, and is less daunting to look at than a .tex source file.
4. Use the LaTeX comment function. I’ve often used % to write comments or corrections to a chunk of text. Then you can have comments, discussions with co-authors (or yourself), and comment out chunks of text that you don’t yet know where to put—you want to keep that text in your source file whilst simultaneously maintaining an uncluttered output file.
5. Share the source file. I’ve had some success with sending my .tex source file to non-LaTeX-using co-authors. The file size is small so it’s easy to send via email, and can be opened by virtually any text editor (including Word). Often what we want is feedback on our text production, not on the layout design, which is often defined by the publisher anyway.

In spite of the many ways to share your work between LaTeX and other platforms, some collaborators may not know of these capabilities—or simply might not care. For example, a colleague of mine (who wrote a paper via Overleaf) could not get her co-author to comment on and review the paper unless she could send him a Word file.

Publishing issues

You will also hear the comment that most journals in the humanities simply don’t accept submissions in LaTeX. This might be true for some but, for the most part, I’ve found that editors are usually simply unaware of the possibility and therefore tell authors to submit Word documents. Users are usually also met with messages such as “Upload one Word file with all the information you want in the published document, and another Word file in which all identifying elements have been removed for blind review”. During the submission process files are often automatically converted to PDF, which makes the procedure even more absurd. However, many of the large publishers that publish journals and conference proceedings in the humanities—such as Elsevier, Springer, and Wiley—do in fact offer ready-made LaTeX templates.

Getting started

I won’t include a tutorial on how to use LaTeX because there are already so many good guides and tutorials available on the web. There is, for example, Andrew Robert’s Getting to grips with LaTeX, or Charlie Tanksley’s LaTeX for Philosophers (and other Humanitarians). However, an easy way to get started with LaTeX is to use Overleaf because you don’t have to worry about package installation or compilers—that is all handled by Overleaf. What my students like to do is to switch back and forth between source view and Rich Text mode to see how changes in one mode affect the other.

Using Overleaf I’ve given LaTeX introductions to linguists and social scientists and have found that as little as 2 hours instruction is enough to get people started and comfortable with using LaTeX. After one such session a colleague of mine decided—a few months before handing in—to write her doctoral dissertation in LaTeX, and did indeed do so via Overleaf. Once her decision was taken, she made the transition more or less from one day to the next. After her dissertation she was especially impressed by how easy it was to reuse content from her dissertation in journal templates. This is of course a bit extreme, so what I usually recommend for new users is to take a smaller paper or report that needs to be written or revised anyway, and do that in LaTeX, rather than go all-in at once. Taking a look at the templates available on Overleaf is also a good way to get a feel for how LaTeX works before you go all-in and write up your dissertation in LaTeX.

In conclusion

This article is written from a dual perspective: firstly, with the intention to show researchers in the humanities that there is indeed an alternative way to produce scholarly texts, other than the one offered by traditional word processors. Secondly, it (hopefully) also makes clear some of the issues that are important to humanistic researchers and what obstacles need to be cleared if LaTeX (and Overleaf as a professional enterprise) is to reach a wider acceptance outside the domains of the natural and technical sciences.