And that's more or less where I left the paper at the Digital Past conference. [Cue audience applause. Questions]
What follows is a version of a collaborative paper on Romanticism, biometrics and geotourism I gave at the Royal Commission on the Ancient and Historical Monuments of Wales's inspiring Digital Past conference in beautiful Llandudno on 11 February 2016. The paper reported on research conducted with computer scientist Prof Reyer Zwiggelaar and geographer Dr Hywel Griffiths. Now you might assume English literature, computer vision and geography have little to say to each other. Actually, our experience is that the most interesting things often happen at the intersection of different bodies of knowledge. So, what happened?
In eighteenth-century aesthetic theories, geology, geomorphology and physical geography are often key components of the supposed visual and emotional effect – the affective charge – of sublime or picturesque landscapes. Many Welsh sites were seen at the time as particularly productive of the sublime – that sense of being overwhelmed, physically and emotionally. Several of these Romantic sites continue to represent important tourist destinations, such as the glacially-sculpted landscapes of Snowdonia. See left for Turner's watercolour painted during a visit in 1799, at the beginnings of modern tourism. These days, the mountain receives some 400,000 visitors a year. Geotourism, originally a Romantic phenomenon, then, is now crucial to the economy and sustainability of these regions. While tourism in Wales is clearly doing a lot of things right, Welsh Government is committed to exceed 10% tourism growth by 2020. Perhaps there’s room – with the Digital Past conference's keyword, recalibration, in mind – to do more, and do it differently, to shift, to renovate, the visitor’s sense of relation, and connection, to spectacular Welsh landscapes and their layered histories. Which is where biometrics – devices that can measure our emotional states – comes in.
Anyone working at museums or galleries will be aware of the tensions that exist between the wish to enrich visitor experience through interpretation and informative signage – about the geomorphology of Welsh mountains, for example – and the desire to avoid over-mediating, or interrupting in any way, that experience. Many people visit Snowdonia and other National Park landscapes precisely to immerse themselves in – to “become one” with – nature, without distractions. Ecocriticism alerts us to problems with that organicist idea of “oneness”: Christa Grewe-Volpp cites such factors as pollution, toxicity and cultural differences that “act against notions of an undifferentiated merging”. At the same time, ecocritics would wish to preserve the value of that “active … relational process”. The concept of “oneness” also emerges, of course, from within a Romantic paradigm, famously articulated in Wordsworth’s and Coleridge’s notion of “the One life, within us and abroad”, in which a mysterious force, or unifying principle, runs through all humans and nature, connecting and uniting. We might regard this sense of contiguity with the natural world as yet another example of Romanticism’s seductive self-representations, which need to be resisted. On the other hand, it heralds our own understanding of ecological connectedness and ethical responsibility. Further, it remains a fact that many of us – still under the spell, or hex, of a Romantic aesthetic – continue to visit stunning sites such as the summit of Snowdon precisely for something akin to what the Romantics categorized as the Welsh sublime, for what Wordsworth in Tintern Abbey called “a sense sublime/ Of something far more deeply interfused”.
So what questions would a Romanticist, a computer scientist, and a physical geographer pose in this context? What can we contribute to the visitor economy? We’re interested in the quantifiable aspects of that mysterious interconnection between landscape and emotion. In other words, are its effects measurable using biometric sensor platforms? And can modern biometric technology be used to enhance that sought-after experience of immersion, of “oneness”, by making the interconnection between landscape and emotion available to (self-)analysis and interpretation? Our project isn’t simply an attempt to sidestep the issue of whether we should “educate” visitors about Wales’s geomorphology using signage. We think biometric analysis technology could emerge as a valuable tool in landscape heritage tourism in Wales, helping visitors become differently aware of, inward with, and able to interpret, their physical and emotional relationship with Welsh geoheritage. Put another way, we think biometrics can help us to become better tourists – better (post-Romantic) lovers of nature.
But let’s take a step back into the sublime. In 1757 Edmund Burke published his Philosophical Enquiry into the Origin of our Ideas of the Sublime and Beautiful. Burke argued that vast, spectacular, rugged scenes of nature – whether experienced first-hand, or mediated through art and literature – provoked painfully pleasurable symptoms of terror in the spectator. The idea was that terror produced by the sight of mountain summits, chasms, raging seas, thundering cataracts – preferably with a ruined castle or abbey – was visible, and legible, through its physical proxies, or symptoms, such as a racing pulse, breathlessness, a swooning sensation, fever. As well as a linguistic discourse, then, the sublime had its own biophysical vocabulary. Moreover, what begins as an eighteenth-century aesthetic theory quickly becomes for the Romantics a category, or cast, of mind, a mode of intellect.
For Wordsworth, mountainous terrain becomes the topographical centre of the sublime. Above is a well-known passage from The Prelude, where he relates climbing Snowdon by moonlight. You can hear in it, as Wordsworth broods on “the dark abyss”, that active connection Burke made between the “body and mind”. The last line – “and cannot choose but feel” – is key. Once the sublime has overwhelmed the intellect, and language itself, it leaves us with no choice but to feel the connection between landscape, the emotions and the body.
Now the Romantic sublime was always teetering at the rough precipice of parody. This satire by Gillray depicts four women reading Matthew Lewis’s insalubrious gothic novel, The Monk (1796), set in a ruined abbey in the middle of a dark forest – of course. The women are reading by candlelight, again, of course, and exhibiting physical symptoms of terror, which Gillray clearly thinks are rehearsed. But were Romantic audiences simply performing what was expected of them when they read sublime literature, or were they actually experiencing emotion-specific biophysical reactions? Until recently, your guess was as good as mine. But the advent of modern biometric technology allows us to measure – to quantify, and assign numbers and values to – exactly what’s going on.
Over the couple of years, interest in wearable biometric sensor platforms – think Apple Watch, Samsung Gear and Fitbits – has grown. The tech we’re developing as part of our Romanticism, biometrics and geotourism project shares abilities with those commercial devices that tend to concentrate on heart rate, movement and location. But our wristbands can also measure temperature and galvanic skin response – basically, how much we sweat, which is itself a proxy of anxiety – movement, blood pressure, location (if paired with a GPS-enabled device), and so on, and we hope to couple that with information about respiration. In addition, our methodology includes the use of a more finely calibrated baseline normalization approach to compensate for the variation in biometric response between individuals. So far, the application of biometrics in tourism contexts has overwhelmingly been in a regulatory capacity – for example, Disney’s use of fingerprint scanners to combat illegally traded park passes and tickets. We want to put biometrics to different work. The Romantic period saw an explosion of tourism opportunities in Wales. Perhaps this boom can be repeated in the biometric paradigm – as part of the biometric sublime.
So what have our pilot projects looked like? Reyer and I first met in the thermal domain, so to speak. Our initial collaboration back in 2010 used data from postgraduate English literature students reading poems by Romantic poet, John Keats, while being filmed with a thermal imaging camera. It was the run-up to Valentine’s Day. Reyer had just been given a new piece of kit, and obviously, I wanted to get my hands on it. The results surprised us both. Reyer had been using thermal imaging for lie detection, and it turns out that when you lie, the periorbital region of the face – the skin around the inner corners of the eyes – heats up, just as the cheeks cool down. See above for a view from Reyer’s and Rajoub's work on that all-important periorbital region.
Now – and here’s the interesting bit – we measured the exact opposite when our student volunteers read Romantic poetry: the corners of their eyes cooled down, and their cheeks warmed up. Pretty much exactly what Keats tells us in his letters when he talks about the “holiness of the heart’s affections” and the “beauty of truth”. So it was part publicity stunt in the run-up to Valentine’s Day, but there was some serious science sitting under it.
A more recent pilot project was the Quantified Romantics event, which ran as part of the UK-wide Being Human festival of the Humanities last October. It was staged in the wonderfully eclectic Ceredigion Museum in Aberystwyth, and basically involved placing people in The Vortex, our name for the darkened enclosure – event-horizon dark – inside which we projected images of sublime and gothic art from the Romantic period. We fitted members of the public with prototype wristbands, and showed them paintings such as “Mad” John Martin’s “The Great Day of His Wrath” and Fuseli’s famous “The Nightmare” while we captured their biometric data streams, which we analysed together afterwards, interpreting changes, and asking people to reflect on their own awareness, or otherwise, of their biophysical responses. We got some nice media coverage, wider interest, including an edition of Radio Wales’s Science Café, broadcast 8 February 2016, which is now on iPlayer, if you’d like to listen to it.
Looking forward, we’re now exploring applications for geoheritage and geotourism in Wales. One of our outputs there will be a biometric map that we’re tentatively calling the Quantified Life Guide to Wales. The idea is to use biometric equipment to measure visitor responses to spatial aspects of culture at eight Romantic sites, and by collating the data statistically predict the categories of “body and mind” experiences tourists can expect – moreover, at precise locations within these sites. In that respect, The Quantified Life Guide to Wales will be a biometrically enhanced version of Romantic-period tour guides – those foot-stepping descriptive maps written with the express purpose of leading others to the exact spots where they might experience sights that, according to J. Evans’s Letters Written During a Tour of South Wales During the Year 1803, would “please while they astonish the beholder”.
Here’s a tourist guide from 1796, A Descriptive Account of the Devil's Bridge, penned by the aptly named Mr Walker, which instructs tourists, step by step, how to work their way from the triple-stacked bridge itself, around some “frightfully projecting rocks”, Walker’s words, to exactly the right spot that would give them a “commanding view of the various falls, which burst upon the astonished sight, in all their sublimity and grandeur”. So the aim of such Romantic guides is to instruct us exactly where to stand to experience various physical symptoms of “terror and amazement”, without actually toppling into the chasm below. Amusingly, Walker points out that his instructions needed to be so precise, because of unscrupulous guides – or “lazy clowns”, in his words – whose own familiarity with Devil’s Bridge meant they couldn’t always be bothered to take visitors to the best viewing spots, when, quote, “half the trouble will secure the expected fee”. If the field data confirms our lab results, we’ll be able to create biometrically calibrated versions of tourist guides like Walker’s – biometric maps using non-traditional hierarchies, ideally more participatory, and crowd-sourced, rather than top down. It’s a new way of thinking about touristic field space.
What about wider applications? There are potential uses for museums and galleries. (Click here for an interesting Swiss project pioneering the use of biometrics in museums in 2009.) We’re used to thinking about gallery space narrativized according to genre, or chronology. We tend to group the old masters, the abstract art, and so on. But what if we arranged things according to the statistical probability of visitors experiencing a certain kind of embodied, bio-emotional response? We could reorganize museums and galleries according to a different conceptual rubric, where all the pulse-raising objects go "over there", and all the ones likely to induce calm "in that corner", and so on. We could have mood rooms, bringing together artists and sculptors from different periods and styles who would otherwise never rub shoulders. Sounds bonkers, but in recent years we’ve seen major rethinks around contextualization and participation – so why not biometrics?
Of course, there are ethical dimensions to consider. We’re particularly aware, after the Edward Snowden revelations – a different kind of Snowdon, a different kind of sublime – of the potential for abuse of personal biometric data. The use of remote bio-sensing technologies to take our data from us without our knowledge or consent is a growing problem. (Click here for the Chaos Computer Club's 2014 expose of how even your fingerprints can be captured remotely.) Thermal imaging cameras at airports monitor differential temperature in our faces, and can make good guesses about our emotional state, as well as our health. Linked up with face recognition software and tracking algorithms – well, you can see there’s a real problem around mass, warrantless surveillance. And on a smaller scale, there’s the thorny issue of our biometric devices knowing more about our biophysical condition than we do. On Monday 8 February, the Guardian carried a piece entitled “Your fitness tracker knows you’re pregnant before you do”. But don't get me started on surveillance.
And that's more or less where I left the paper at the Digital Past conference. [Cue audience applause. Questions]