Working Together: The Digital World and English Studies By Tiago Sousa Garcia

Posted by boo11 at Feb 25, 2019 08:30 AM |

Jenny invited me to speak because of my unusual background. I’m not entirely sure how unusual it is, particularly in current company. Although I currently have a degree, MA and doctorate in English Literature, before embarking on the Humanities train, I did a professional course — I think roughly equivalent to a higher apprenticeship, higher than an A level but below a degree — in Computer science, and started — but never finished — a degree in Computer Engineering. At the same time that I was doing my degree in Engineering, I was simultaneously working as the lead (and sole) coder for a small company in Portugal. When I changed my degree from engineering to literature and quit my job, I thought I had left that life behind me and would never use it, need it, or even refer to it again: like Shakespeare, I had my own lost years.


I was wrong. Fast forward ten years, past the degree, MA and PhD in Literature, and I find myself constantly referring to and drawing back on my previous digital life. I don’t, actually, code that much anymore — my skills are rusty, my memory of the various programming languages I used to know and love, faded — but the understanding of how computers work, and how computer people think, has been invaluable. In fact, even in my study of literature, I think I retained my problem-solving engineering approach. On the first year of my literature degree, one of my tutors told me I think like an engineer — a backhanded compliment if I ever heard one. Today, I am a Research Associate for the Animating Texts at Newcastle University (ATNU) — a project led by Jenny Richards that, in a nutshell, is bringing together Humanities and Computer Science researchers to think about, and find new ways of interacting with this crazy thing we call text. It is literally my job description to serve as a bridge between computer science and humanities colleagues: making sure everyone understands each other, translating not only the language each side uses, but also aims, objectives, and, ultimately, approaches and methodologies. I might be tooting my own horn a little too much here, but I like to think of myself as the linchpin that holds ATNU together (kidding, kidding).


From my point of view of in-betweenness, I have witnessed first-hand the opportunities arising from, and the barriers to, collaboration between such different camps. If I have gained any grain of wisdom from my time with ATNU is that the two are often one and the same thing, and that the challenge is knowing how to disentangle them.


The barriers to collaboration are everywhere, and often lie in the simplest things. Take language, for example. Sitting in meetings, exchanging emails, just talking the corridors, we use the same words, but we say different things, or we say different things, but mean the same thing. This is not new for anyone who studied post-structuralism, of course, which often delighted in the impossibility of interpersonal communication, while anyone who ever tried to learn how to code knows that there are as many ways of displaying a simple ‘Hello World’ message on screen as there are computer languages. I don’t mean to be a late-stage deconstructivist here: this is not an impossible barrier, and people from both sides can and do understand each other’s language, but the language barrier is often forgotten in everyday communication. There are many concrete examples of this, but my favourite is the concept of ‘data’, when today we cannot take two steps without running into it somewhere.


For computer scientists ’data’ is the underlying concept that underpins everything computers do, and it is so significant and so intrinsic to the process that there is a basic assumption that everyone, everywhere, must immediately know what ‘data’ is. I don’t mean, of course, the most high-minded discussions of data within computing science, but rather the simple referent of data as information, and information as something that can be stored, analysed, transformed, and used by computers. For humanists, particularly literary scholars, — and I don’t mean to be patronising here, and I hope that’s not how I sound right now — data is scary. Data is numbers, static and limited information that could not — cannot — accurately convey all that we know about literature (or history, or music, or…). Well meaning, but perhaps misguided, humanists sometimes throw the term around in ways that make little to no sense to computer scientists; well meaning, but perhaps misguided, computer scientists ask humanists to explain what type of data they work with, and how can they help treat it.


The corresponding term from the humanities that baffles computer science is, I think, text. Text, depending on your specific discipline or methodological approach, is either the conveyor of simple information, or the unknowable signifier that carries an unaccountable myriad of meanings: the symbolic representation of spoken language, the face-value meaning of words, the metaphoric meaning of words, as much the written as the unwritten; it is both the intellectual content and, if you are so inclined, the physical expression of that content. Think about codicology, for example, the study of manuscripts: text is both the words written, the paper or parchment that supports it, the script in which the symbols are formed, the ink, the smudges, the holes, the provenance, and so much more. For humanists of a literary bent, text is the single most complex concept in the world. For computer scientists, text is string of characters, each of those signified by (depending on the encoding) 8 or more consecutive bits. A capital A, for example, is 0100 0001 in UTF-8 encoding. It is, in other words, data.


For computer scientists, everything can be data; for literary scholars, everything can be text. Text and data are not the same thing — let me emphasise this — but they serve similar functions, and are treated in the same fashion, in both camps. The real barrier at the centre of this problem, it should be clear by now, is not just one of language, but a deeper, scarier, more challenging obstacle: it is an epistemological one. On the one side, the desire for clarity, exactness, uniquely defined categories, and one-faceted information; on the other, the love of fuzziness and ambiguity, the knowledge that the world is always more complicated than we think, and ultimately irreducible to human understanding because, for the most part, humans complicate things. This fundamental difference infiltrates even the smallest details, from the large questions to the tiniest work arrangements. For collaboration to be successful, there needs to be flexibility on both sides, of course, and allowance on both sides to withstand each other’s methodologies and approaches.


So, what are the opportunities? Well, as I see it now, it doesn’t mean overcoming this near unbridgeable gap, but embracing it (and yes, I do realise that this sounds a lot like corporate-speak, but let me try to talk it through). Interdisciplinarity as I understand it, has for years relied on a reciprocal deal: my discipline has a problem that requires your discipline’s help to solve; my discipline’s problem helps your discipline in tackling a different problem. Interdisciplinarity is a simple venn diagram in which we look for the area of intersection, tackle that area, and bring what we can back to the large single-discipline area. This is still, I have to emphasise, a good approach to a whole series of problems, and possibly the easiest most profitable way of generating benefit from interdisciplinarity. It still applies to the intersection between Humanities and Computer Science: how to extract meaningful data from billions of badly formatted historical records? Are computers able to sift through and figure out what is and what is not literature, and if so, could we help illuminate the black box of machine learning? All of these are questions that a traditional interdisciplinary approach can tackle: discipline A and discipline B both contribute to solve problem x, the different results are fed back to discipline A and B.


Increasingly, however, I have been thinking that this approach is not ideal: the separation between objectives and results is still too great, the collaborators still primarily turned towards their home disciplines. Instead, we need more questions, and I think more people, who can confidently stroll both sides to the point that there will be hardly a side at all. The humanist in me looks longingly back to the Renaissance, where polymaths, rather than specialists, were the norm —think of what would have been Da Vinci’s paintings without his interest and research in science. And where would Mathematics be if Descartes was not broadly interested in philosophy? A considerable part of this comes from undergraduate training, but — I sense my colleagues’ anxiety — this should not necessarily be a focus on STEM disciplines and approaches at the expense of the humanities, nor a blind focus on humanistic knowledge without at least a solid understanding of a STEM-like approach to real world problems.


For those of us who are already past the age of learning, it shouldn’t mean that we are hopeless dinosaurs, waiting for the asteroid that will eventually decimate us. In academia, there are broad tendencies that we should try and be able to address that can help make it a more friendly environment for those who would like to move between different fields. For example, despite all the talk of interdisciplinarity, appointments are still primarily made at a discipline level, to people who can make a case for a well-defined specialism. In my own humble experience, after a PhD that proclaimed the softest of interdisciplinarities (between history, Portuguese literature, and English literature), I struggled to find a comfortable home in schools of any discipline; and, in the end, it ended up being my lost years in computer science and engineering that became decisive for Newcastle. If we do things right, there are endless possibilities. We need to know what we don’t know to be able to ask the right questions; and we only know what we don’t know, if we learn it. Academia is already being left behind on this: people with humanities degrees are routinely employed by tech companies, big and small.


This also doesn’t mean we should try to force interdisciplinarity down the throats of everyone. Interdisciplinarity, particularly the cross-roads between computer science and the Humanities, is not a panacea capable of solving all of the world’s ills. There should remain ample space for specialism, for discipline-specific questions. Not all funding should be dedicated to commonalities: linguists will not cure cancer; historians will not be able to colonise Mars; and not all humanities questions can be helped by computers. In our drive to look to the future, and in our evangelical passion for working together, we should not forget that even the best of us can’t know everything about everything.


Thank you.

Share this page:

Contact us

The English Association
University of Leicester
Leicester LE1 7RH

Tel: 0116 229 7622

Follow us on twitter Find us on facebook