CDH-cookbook

What is the Role of Programming in the Humanities?

^ Main contents

Contents

• Could Your Project Benefit from Digital Tools?
• How to Find Methods and Code to Help Achieve Research Goals

If you are taking these courses, chances are you are a student or academic working in the Humanities who believes that you need coding skills to complete a project or analyse your data. Because most lessons one finds online are geared towards people from STEM backgrounds, we believe it is important to provide here some guidance and opportunities for reflection about how to conceptualise of your project in terms of digital tools, and how to see where code could potentially help your work grow.

Could Your Project Benefit from Digital Tools?

💡 It is probably helpful for you to read the following section even if you already have a clear idea of what you want to achieve, because this section is designed to challenge you to think about your research in different ways.

Digital Humanities is a broad umbrella term for multiple areas of enquiry and research, all of which are connected by the presence of ‘digital’ things: digital concepts, digital methods, and digital material. While today all scholars, whether they be in the Sciences or the Humanities, use digital tools, you may not think of yourself as a ‘Digital Humanist’. Fortunately, you don’t have to be in the field of Digital Humanities to make use of its advancements. This course takes the very practical perspective that code can be accessible and useful to any scholar who needs to use it, regardless of their affiliations.

Whether or not you are a Digital Humanist, it is possible that your research could benefit from code. One of the challenges that scholars in the Humanities face is that they often don’t know what is possible with code, and so they don’t know if they should learn it or try to use it in their research. The easiest way to navigate this impasse is to think about your project in very practical terms:

What aims do I want to achieve?

What are the tasks which I need to do to achieve these aims?

Could there be digital tools that exist which can help me reach these aims or complete these tasks?

For example, you may have the following (hypothetical) scenario: you have research aims which do not involve digital analytical tools, but in order to meet them, you have to process a lot of files of text in the same way, over and over. This is a prime candidate for a digital tool, because you can automate repetitive tasks using some code.

The answers to these questions vary infinitely by one’s project, but there are general guiding principles! While not covering all cases, the following principles can help you identify whether your project could benefit from automation:

[DO WE WANT A GRAPHIC FOR THIS LIST?]

• 💭 Do you have a lot of material, more than is manageable using analog methods within a day’s work?
• 💭 Do you need to repeat the same kind of analysis on your material?
• 💭 Will your research generate data which will need organisation and management?
• 💭 Are you tempted to use a spreadsheet to make sense of your data?

These are indicators that you either have data which can be handled with digital processes, or you have tasks which can be automated. If this is the case, your project can likely benefit from code, and there probably already exist software packages (organisations of code) which can be used to manage your data.

There is also a second potential scenario: the research questions or aims which you have could be either achieved, or enhanced, through the use of digital tools. We can use my own PhD thesis as an example. My fundamental research question was: within a large corpus of medieval manuscript material in Hebrew, are there verifiable patterns of codicology (physical format of the manuscripts) and linguistic features? People have attempted answers to similar questions in my field without the use of automation, but they were only able to do so either on a very small number of manuscripts at a time, or in projects which span decades and involve large groups of researchers. I decided to look for methods which would enable me to find patterns quickly and reliably in a larger number of manuscripts. Thus, my PhD became a ‘Digital Humanities’ project.

Again, while not covering every case, the following principles can be of assistance:

[AGAIN, WE MIGHT CONSIDER MAKING GRAPHICS OF LISTS]

• Would your research benefit from making use of more data to clarify the picture?
• Does your research question involve finding patterns, bridging fields, or finding/creating categories?
• Does the research problem deal with either too much data, or suffer from unprocessed data?
• Do you feel the need to think about or explain your data in statistical terms?

These are some (and certainly not all!) indicators that your research project could be enhanced, achieved, or even transformed with the use of digital tools.

A major theme appears here, which you may have already caught: a large part of Digital Humanities is making sense of and managing data. Many Humanities researchers think that they don’t have data, or that they don’t have a lot of data. In the vast majority of cases, this is simply untrue. Enlightening statistical tests can be run on small datasets of around 30 pieces of data, and anything that one describes, and even how they describe it, count as pieces of data which can be analysed by a computer in some way. Moreover, small datasets which are complex (involving multiple known or undiscovered relationships between entities) are also prime candidates for the use of digital tools.

How to Find Methods and Code to Help Achieve Research Goals

This section of today’s lesson is, unavoidably, generalised, simply because there are infinite iterations of research questions which one may have and to which one can apply digital methods. However, there are a few guidelines for finding methods:

1. Get your research question/task/objective into terms which are as specific as possible. What *exactly* do you want to do? What *specifically* do you want to know? Describe to yourself and write down the process you imagine going through to meet these objectives. The more specific you can be, the easier it will be to find exactly what you need.
2. Break down your project into discrete steps. It is highly likely that you will use different methods at different steps. For example, in my PhD I made use of relational databases, descriptive statistical methods, multiple machine learning methods, and data visualisation methods to achieve the outcome of my project.
3. Reach out to scholars in Computer Science, Digital Humanities, or Data Science, and ask them what kinds of methods might help. The majority of the time, people in these fields are more than happy to help one find the right method for one's research. It is important to know from the start that Digital Humanities is a highly collaborative field, and often the only way to progress is to interact with others.
4. The internet is your friend. Chances are, the research question you are asking has an analog in another project or question, which someone has probably written about on the internet. Do a search, or post in forums, and you will find that there are often plenty of resources that will appear to hand. Note: Using ChatGPT and other AI tools is risky for this particular kind of query: they tend to deliver haphazard or non-academic results, unless you get lucky. The best approach is to search in a variety of ways, rather than just rely on one method.
5. Use Digital Humanities resources, such as forums, websites, blogs (the Programming Historian mentioned above is an excellent resource for this, by the way), to find examples and pointers.

Reflection

How might you integrate coding into your own research project? Would it play a central role? Would it support peripheral things, like basic data management or analysis? Could it enable central things, like the analysis itself?

Written by Estara Arrant, 2025-04-16
Licence: CC BY-SA 4.0