Why do we want humanists to acquire digital skills?

I am travelling back home from the IV Congreso Internacional de la Asociación de Humanidades Digitales Hispánicas [Fourth International Conference of the Hispanic Digital Humanities Association] (HDH) in Toledo, Spain. I attended a very relevant panel on master’s level education in digital humanities. One of the issues that came up was why we want to teach digital skills to humanists, and whether this makes sense or not. In fact, this deep and foundational issue was sort of going around for the whole conference and at different panels and sessions, but unfortunately it didn’t get to be properly discussed. I will try to address it here, at least partially.

So, why do we want humanists to acquire digital skills? Let’s start by defining our terms. What are “digital skills”? Actually, what do we mean by “digital” in “digital humanities”? Does using a blogging platform like this one to reach out to others constitute a digital skill? Do digital skills include programming, scripting, modelling, managing databases, applying algorithms…? To answer this question, I will use the following distinction. Imagine that someone tells you that someone else knows a lot about cars. What idea forms in your mind?

1. The referred to person is a good driver, knows how to use their vehicle properly, can check the oil level, top up windshield washer fluid, or replace a flat tyre.
2. The referred to person has encyclopaedic knowledge of car makes and models, engine specs, and the history of car making.
3. The referred to person is an engineer who designs and builds cars for an automobile manufacturer.

Note that these possibilities are not incremental; in fact, they are quite independent of each other: a car engineer may even lack a driver’s license, and an excellent driver and car owner may be totally ignorant of the engineering under the hood (if you pardon the pun).

The problem

All of these possible answers are valid, as the statement that someone knows a lot about cars is quite vague. Similarly, stating that someone has great digital skills may mean different things:

1. They can use digital tools such as word processors, spreadsheets or blogging platforms proficiently. They know all their secrets and hidden tweaks, and usually use them very efficiently to produce great results.
2. They have an amazing knowledge of the history of digital tools, know every detail about how they evolved and changed, who created what and when, and what was the social and cultural context that surrounded each development.
3. They are software engineers who design and create digital tools at a software development firm.

Again, these possibilities are quite independent from each other: many professional software developers (case 3) are not particularly good at using many digital tools, and may ignore a lot of the history and social aspects of technology; many proficient users of digital tools are oblivious to what goes on behind the user interface.

In my experience, the “digital” qualifier in “digital humanities” is often used vaguely, sometimes referring to 1, 2 or 3 above without much clarification. When I attend conferences such as HDH or CAA, I see people who can specify, design, implement, test and deploy complex software systems (case 3), others who can’t do this but instead are experts in the history of the digital (case 2), and others who can’t do either but are amazing users of existing technologies (case 1). All of them are welcome and all of them are necessary, so having them all under a single umbrella term of “digital humanities” poses no problem.

The problem arises when we try to design a master’s degree in digital humanities. What do we want to teach humanists? Is it 1, 2, 3, or a combination? Do we want to obtain great users of technology, great digital historians, or great engineers? Obviously, the contents and skills to be taught would vary a lot depending on how we answer this question. And any prospective student would (or should) want to know in advance what kind of training a master’s degree will give him/her. I have looked at flyers and other materials for most digital humanities master’s degrees in Spain and none of them answers this in a clear and explicit manner.

The options

Let’s examine each option in turn. Let’s imagine for a moment that “digital humanities”, and master’s degrees about it, should focus on 1 above, i.e. educating students to become excellent users of existing digital technologies. Degrees like this would perhaps teach students how to use digital tools to gather, store, process and display data; and how to apply existing algorithms and techniques to transform data or obtain new information from it. Students would learn specific tools and technologies (such as MySQL, TEI, WordPress or R) and perhaps a bit about their foundations (how relational databases work, how to write a good blog post, etc.), but they would not necessarily learn much about how these tools work under the hood, or why these tools are the way they are, or what other options exist. In fact, one of the weakest points of this option is the fact that tools and technologies, especially in the software realm, are extremely volatile, and change very fast. If you spend a few thousand euros or dollars, and a year or two of your life, to learn R or TEI, what are you going to do in a few years when R or TEI go out of fashion or become obsolete and some new technologies come up? In a sense, you would be stuck with your tools of choice, those that you learnt and love, and you will probably stick to them (understandably) while feeling more and more out of place in a rapidly changing technological ecosystem.

Let’s instead imagine that “digital humanities”, and master’s degrees about them, should focus on 2 in the previous list, that is, they should emphasise the history, sociology, politics and/or anthropology of the digital. Degrees like this would teach students history of technology, perhaps some philosophy of science, and definitely some notions of the most relevant technological waves and innovations. Students would obtain a good understanding of how digital technologies and tools have evolved, what social and cultural issues have mediated technological change, and how digital tools have shaped our societies. This is not too far from many media studies programmes. Students in degrees like this would be able to describe and understand what has happened so far in relation to digital technologies, but they would not necessarily become proficient in either using these technologies or designing them. Also, they would become good at describing what has happened, but not necessarily good at predicting what should happen, which is necessary if you aspire to design and create things such as software systems.

Lastly, let’s imagine instead that “digital humanities”, and master’s degrees about them, should focus on 3 in the list above, i.e. they should emphasise the engineering, design and development of software systems and tools. Degrees like this would teach students how to assess problems, envision solutions, specify and design systems to address these solutions, and test and deploy them. Students would become de facto engineers, and would (at least in theory) be able to question existing technologies and develop new ones, thus changing the world (or a small fraction of it). Using the car analogy, they would not become good drivers necessarily, but they would definitely become good car designers and builders. The weakest point of this approach is that we already have engineers to do this, and that a 1- or 2-year master’s degree for humanities graduates would certainly not be enough to teach the necessary contents and skills that make a good engineer.

Are we stuck, then? At this point you may be thinking “yeah, well, but these options are leaving out intermediate or hybrid solutions”. For example, what about an option that combines 1 and 3? This would take an approach to train advanced users that are extremely proficient at using some tools and technologies, but who also can tweak them, customise them to perform around the edge of their intended purpose, and even question them. Digital humanists of this kind would be even capable of envisioning new technologies and tools that go beyond existing ones, but would perhaps be unable to develop them without the close collaboration with engineers, unless they are very small and simple ones.

So, why?

Any approach from the list above can work, and so may any combination. I am not trying here to favour one above the others, but just examine pros and cons of each. To pick one, I think that the question at the top of this article may help: why do we want humanists to acquire digital skills? Is it because we think that they should be able to squeeze as much power out of digital tools as possible? Is it because we believe that, by being proficient in digital technologies, they will be better at liaising and communicating with engineers? Is it because we think that (digital) humanists should replace engineers and develop their own tools?

If we (or you, as a master’s degree designer) want humanists to become good at using digital technologies and tools, then it makes sense to adopt option 1 above. But we should be aware of the risk, and very especially of the fact that learning specific tools and technologies without a deeper understanding of the underpinning principles and theories is likely to produce knowledge that will soon become stale and obsolete, thus requiring ongoing training and study. Most humanists are used to a slow pace of change in ideas and academic fashion, but the digital world is something else, as it moves much much faster. If you are a student or are thinking about enrolling in a degree that uses approach 1, don’t even think that your degree will give you the skills to become and stay employable for a long time. No way. It may give you an initial impulse, for sure, but you will need to keep studying on your own intensely and forever.

If we aim at humanists to become better at collaborating with engineers and communicating with them, then I don’t think you need to study a digital humanities degree at all. I have worked as a software engineer for many years, and I have had many clients who were excellent at communicating with me, and many who were extremely bad. In my experience, there is no correlation between technical knowledge and communication skills. I’ve worked with excellent communicators and collaborators who had no idea of the engineering under the hood, and I’ve worked with people who had quite a good understanding of what was going on (in fact, some were software developers like me) but were hopeless at expressing what they wanted or understanding what was possible or viable. If you want humanists to become better at liaising with engineers, ask them to study written and oral expression, improve their language skills, and practice their sketching abilities. It is the labor of the engineers to bridge the gap between disciplines, and good engineers are well trained to do that. If you are a humanist and the engineers you work with are disappointing at understanding you, perhaps you need to switch engineers (or improve your communication skills).

Finally, if we want humanists to become capable of developing their own digital tools and technologies, and replace engineers in doing so, then I don’t think you will succeed. A master’s degree is not even close to being enough to provide the necessary training to produce good engineers, especially if students come from an exclusively humanistic background with no previous university-level training in maths or logic. “But you can learn R in a few months”, you can say. Well, yeah. You can learn the syntax of R and some usual programming idioms to put together small scripts, but that’s far from being a solid, reusable and reliable skill. I’ll use an analogy. Imagine you want to learn how to play chess. It’s very easy to learn how to move each piece on the board, and what the basic moves are. You can probably master that in a couple of days. But that doesn’t make you a good chess player. Becoming a good player requires a lot of experience, playing with many different adversaries so you learn about different styles and techniques, and developing a tacit understanding of the underlying mechanics of the game that you can exploit almost instinctively. Likewise, learning how to write small scripts in R is a good starting point, but that’s very far away from becoming a good programmer. And programming is just a small part of software development! (I will write about this in a future post).

In other words, we should not confuse learning what can be done with learning what should be done. When we learn how to move each piece on the board, or how to write different commands or expressions in R, we are learning what we can do. But we have no idea of what we should do in each situation. Being able to decide how to respond to a Sicilian defence, or choosing between a property of an enumerated type and subclassing to implement subsumption in programming, require strategic thinking and a wide range of experiences in different settings. After all, you can always look up the syntax of a particular method or command in the documentation, but you cannot easily look up help to make a decision about what to do.

Don’t think that a master’s degree in digital humanities will teach you strategic thinking in engineering. It may teach you what to do, but not what you should do (no degree in engineering will probably teach you that, either). If you want to become skilled in this regard, do an engineering bachelor’s degree rather than a digital humanities master’s. In other words, if you want to become an engineer, study engineering.

Conclusion

I guess the take-home messages from this post are the following:

• If you are in charge of putting together a master’s degree in digital humanities, think why you want humanists to acquire digital skills, and act accordingly. Whatever option you choose, be explicit about it in your dissemination materials.
• If you are thinking about enrolling in a digital humanities master’s degree, think what your goals are. If you want to learn to use existing tools and technologies, go for it, but don’t forget that you will need to keep studying after you finish your degree if you want to stay current. If your major goal is to become better at communicating with engineers, don’t enrol; instead, develop your communication skills. If your goal is to become capable of developing your own tools beyond very simple ones, don’t enrol either, and find instead a good software engineering undergraduate degree.

One final word of caution. This post makes some gross generalisations about skills and training. Of course, every individual has a unique life history and skill set, and I’ve known a few people who clearly violate the assumptions that I make here. Still, I think that what I have described works as general advice. But, as usual, take it with a grain of salt.