Haptic media and the cultural techniques of touch: The sphygmograph,photoplethysmography and the Apple Watch

Touch and the intimacy of the medical encounter

Touch has a significant but fraught place within the cosmology of medical practices. New media technologies have produced an anxiety over the potential sensory deskilling of medical practitioners, especially as digital technologies are seen to take on much of the diagnostic work which was formerly constituted in the ‘hands on’ relationship between patient and doctor. The touch of the doctor has often stood as metonymic of the intimacy which is characteristic of the ideal relationship between doctor and patient, and this intimacy has often been construed as threatened by the technologisation of medical practice. Pulse palpation, in particular, has held an important place in the definition of the physician’s craft in the Western tradition since at least Roman times; Greco-Roman physician Galen produced 18 books on the techniques and analytic framework of pulsology, and his contribution to the Ars Sphygmica would be influential for millennia after his death. And yet, the standard history of cardiology has held that since the mid-nineteenth century, techniques of pulse taking have been progressively automated, with the tactus eruditis or ‘educated touch’ of the physician being obsolesced by technologies such as the sphygmograph, the sphygmomanometer and the electrocardiogram. As Anna Harris (2016) has noted in a recent article, ‘[t]he growing abundance of medical technologies has led to laments over doctors’ sensory de-skilling, technologies viewed as replacing diagnosis based on sensory acumen’ (p. 31). Medical historian Stanley Reiser (1993) has argued that ‘the decline in physical examination and the use of senses in diagnosis, as technological substitutes are invented, seems … inevitable’ (p. 272). This critique of the loss of intimacy in the modern medical encounter is in line with Nicholas Jewson’s seminal and oft-cited article on ‘The disappearance of the sick-man from medical cosmology, 1770–1870’, in which Jewson describes the pre-modern encounter between patient and physician as one of a ‘probing’ of the unique individuality of the patient by a caring and engaged doctor at the bedside. Jewson (2009) argues that the rise of ‘hospital medicine’ has led to a loss of the intimacy which characterised bedside medicine, as ‘[t]he special qualities of the individual case were swallowed up in vast statistical surveys … [and] the sick-man was no longer regarded as a singular synthesis of meaningful sensations’ (p. 628). This apparent triumph of the abstract and the scopic, and the degradation and decline of the tactile or haptic, is consistent with what David Parisi (2011) has recognised as a common discourse surrounding the historical transformation of the senses in the wake of the birth of the clinic and the rise of Industrial Capitalism; that it was in this period that the visual attained dominance as a technology of science and of governance, with the tactile being buried beneath the glass shards of collodion plates.

It is necessary to avoid a reductively teleological principle of technical development in examining the impact of wearable sensor technologies on the reconfiguration of the senses and of clinical authority more broadly. Far from abolishing the value of touch through an ocularcentric representational mode, clinical practitioners have drawn upon the representational power of heart rate–measuring technologies to extend and deepen the authority of doctorly touch, strengthening rather than undermining the authority of the doctor as a literal embodiment of medical knowledge. This point may seem somewhat obvious or otherwise banal; of course medical practitioners continue to touch their patients, and Roy Porter has even argued that doctors have touched their patients more rather than less since the technologisation of modern medical practice (1993). What is at stake then is not so much an empirical as much as an epistemological or historiographic questioning – how to give an account of this technical relation which would not unduly privilege the one who touches/is touched, over the apparatus which mediates/produces this relation, and vice versa? It is proposed that a method predicated upon the cultural techniques school of media theorisation may provide an appropriate conceptual grounding.

The cultural techniques of (proper and improper) touch

The intimacy of the bedside doctor is a double-edged sword; there is a fine line between the authorised probing of the physician and the erotic connotations which have been attached to touch, especially in the Christian tradition, fixated on the fall which transpired after Eve touched the apple (Gilman, 1993; Jütte, 2008: 6). Porter (1987) has described the ‘touch of danger’ which characterised especially the eighteenth-century medical encounter, ‘in a culture whose religion inculcated belief in the corruptions of the flesh, medicine suffered from guilt by association’ (p. 206). Hence, beyond a normative ethical argument justifying the touch of the doctor as metonymic of that bond taken as constitutive of the ideal doctor–patient relationship, there is a need to ‘probe’ the ways in which the touch of the doctor has been distinguished from less salubrious forms of touch, how it has been codified and ‘authorised’ as a legitimate means by which to ascertain diagnostic data, and how this authorisation has been transformed over time, and in changing technical and cultural contexts.

The sense of touch is defined by an extreme ambivalence; it is, on the one hand, that most ‘primordial’ or primitive sense (Hansen, 2006) which makes possible sensory perception tout court, at its core consisting of an undefined and undefinable capacity for feeling. On the other hand, it is touch which for Marshall McLuhan (1994) compromises the very structure of sense itself; it is ‘the interplay of senses, rather than the isolated contact of skin and object’ (p. 314). Touch comprises an irreducible physiological difference which eludes categorisation and representation; it has no obvious ‘seat’ or organ; it is both internal and external to the body. And yet, touch has also been understood as that sense which above all others allows us to ‘get a grip on things’; Fiona Candlin (2006) notes that for the Ancient Egyptians, touch provided the ‘assurance that objects were impenetrable and separate from one another’ (p. 39). Mark Paterson (2007) has described how the development of haptic media interfaces, from ‘virtual handshakes’ to surgical training tools and the use of force-feedback in video game controllers, have served as an experimental milieu in which designers have grappled with the project of producing a more convincing or satisfying ‘sense of presence’ than would be possible via visual media alone. The appeal of haptic media then would lie in their ‘ability to enhance the sense of presence of an object in a gameworld or on the desktop … [and in their] ability for this sense of presence to be communicated at a distance, to literally feel the presence of another’ (Paterson, 2007: 131). It is this tension between touch as point of physiological heterogeneity and touch as anchoring a sense of being in the world which has characterised the theoretical difficulties in defining touch as a sense; as Mika Elo (2012) has written,

As a sense touch has been considered too complex and obscure to be able to offer a clear model for cognitive classifications, which have traditionally been grounded on visual logic. On the other hand, we rely on Aristotle’s classifications even in the division into five senses where the sense of touch has its place like the thumb in the hand. This ‘full hand’ of the five senses has been apt to tame the structural heterogeneity of touch that threatens cognitive discourse. It has made it possible to present touch as a sense that can serve cognitive interests by guaranteeing an immediate, hands-on touch with reality.

As Elo notes (2012), there is nothing guaranteed about touch’s status as guarantor; the transformation of ‘touch as heterogeneous field’ into ‘touch as guarantor of certainty’ must be considered as a question of the historical transformation of touch through the technological apparatus. Clare Colebrook (2013: 4) refers to the ‘stabilising’ form of touch as ‘proper touch’; touch as a guarantor of meaning, as that which can overcome difference and separation and affirm proximity and intimacy between the toucher and the touched; and as distinct from an ‘improper’ touch, which would fail to bridge the gap between toucher and the object of the touch. While Colebrook questions the possibility of ever producing a purely ‘proper’ touch, the point here is that, especially in the clinical field, touch in its ‘natural’ state, as unbounded sensory heterogeneity, represents error and failure; it is undirected, it must be tamed, trained and broken in, in order to get it to reliably reproduce the qualities of intimacy, proximity and stability, for which it is valued.

In this context, we can consider the media-historical training of the sense of touch as an example of a cultural technique, concerned as it is with defining proper from improper touch and with preserving this distinction in the face of the threat to cognitive clarity posed by touch’s physiological heterogeneity. Geoffrey Winthrop-Young (2013) has described cultural techniques as ‘those skills and aptitudes necessary to master [a] new media ecology’ (p. 5), though their study always emphasises that this mastery is never unidirectional; the gestures and habits which form around the use of technologies also tend to master the user, to inscribe her in routines through which the media artefact comes to define the distinctions between self and other, which, in turn, define the subjectivity of the user (Winthrop-Young, 2013; see also Geoghegan, 2013; Parikka, 2013). Cultural techniques theory is concerned with investigating how ways of using media technologies serve to draw cultural boundaries and distinctions and with how this drawing of boundaries becomes alternately routinised and destabilised. As Bernhard Siegert observes, following Michel Serres’ (1982) theory of the parasite, to establish a communicative channel is always to seek to exclude that noise and error which threatens to corrupt the communicative channel. Cultural techniques theory draws attention to how we live with error, within the failure of communication; a cultural techniques approach to the history of media studies would then be a ‘history and theory of interruption, disturbance, deviation’ (Siegert, 2008: 35).

While media studies’ discussion of the haptic tends to privilege the interface as object of analysis, this article examines haptic media as media technologies which train or discipline touch and which serve to produce touch itself as a coherent communicative technology. To consider touch as a technology in this way is to begin from the assumption that the haptic ‘begins’ as a radical physiological heterogeneity which must be ultimately tamed and managed through the technological apparatus. A haptic media studies in this vein then would make the argument that touch is not abstracted away through technologisation, but redefined and revalorised within changing technological and institutional contexts. This study of the definition and exclusion of haptic error is ultimately a political one, examining how the right touch and the right to touch are produced and with what consequences.

The sphygmograph and the birth of the graphical method

The sphygmograph was the first means by which the body’s pulse could be represented graphically. A plate would rest upon the pulse point of the radial artery, which would absorb the fluctuations of the pulse, abetted by a flexible steel spring. These movements would then be transmitted to a recording arm, which would register the traces of the pulse upon a mechanically rotated strip of paper (see Lawrence, 1978: 197). While Karl Vierordt’s 1855 sphygmograph was a large and unwieldy affair, Étienne-Jules Marey’s 1859 sphygmograph incorporated a sophisticated mechanical motor designed by the master horologists Maison Breguet and was light enough to be worn on the wrist. Although there had been sporadic attempts to more precisely measure the pulse in preceding centuries (Reiser, 1978: 97), Marey’s ‘graphical method’ constitutes a key exemplar of what Lorraine Daston and Peter Gallison (1992) describe as the ideal of ‘objectivity’ as a new discourse of scientific representation; they cite his pronouncement in La Méthode Graphique dans les Sciences Expérimentales (1885) that

there [was] no doubt that graphical expression will soon replace all others whenever one has at hand a movement or change of state-in a word, any phenomenon. Born before science, language is often inappropriate to express exact measures or definite relations

as a tribute to the spirit of a new age and a new ‘wordless science’, which would be ‘wary of human intervention between nature and representation [and so] turned to mechanically produced images to eliminate suspect mediation’ (p. 81). There seems little doubt that Marey personally was devoted to something like the pursuit of the mechanical objectivity which Daston and Gallison describe; however, as Robert Brain has noted, the lines of the sphygmograph still invoked a certain affective, embodied relation between the device and its user; the graphic method ‘demanded a formal, cerebral, and intellectual calculation … [while a]t the same time, the observer perceived a sensible intuition of the tracing’ (Brain, 2015: 23–24; see also Brain, 2015: 120).

While Marey’s own writings and practice may support the contention that the development of the graphic method through the sphygmograph was in aid of a medical practice which aimed at eliminating the need for human interpretation, other uses, modifications and clinical applications of the sphygmograph reveal an altogether more ambiguous place for the instrument in terms of mediating the vital rhythms of the human body.

The proper touch of the educated finger

A number of prominent English physicians in the mid- to late nineteenth century took up the question of the place of doctorly touch in calibrating or cross referencing the traces of the sphygmograph, including W.H. Broadbent (The Pulse, 1890), John Burdon-Sanderson (Handbook of the Sphygmograph, 1867), Robert Dudgeon (The Sphygmograph: Its History and Use as an Aid to Diagnosis in Ordinary Practice, 1882) and William Ewart (The Pulse-Sensations: A Study in Tactile Sphygmology, 1894).

Rather than abjuring the physician’s capacity for subjective judgement, Broadbent argued that the cultivated touch or tactus eruditis of the physician remained necessary in order to give meaning to the traces of the sphygmograph and to give a more accurate reading of the ‘vehemence’ of the pulse than could be easily read from the sphygmograph’s tracings (Broadbent, 1890: 42). As Broadbent (1890) wrote,

While, then, I think that every student ought to be familiar with the sphygmograph, and will gain from a study of its indications a comprehension of the pulse in its different forms obtainable in no other way, I am of opinion that we learn by means of the educated finger all that the sphygmograph can teach, and more. This instrument is invaluable as a means of educating the sense of touch and of cultivating the faculty of observation. (p. 34)

This would be a form of touch which, rather than being removed from the interpretative process, would be remediated as a sensibilisation or corporealisation of new forms of medical authority, which would be able to rehabilitate the scope for doctorly intuition within the frame of the new prestige and authority accrued to the graphic method. While, following Daston and Gallison, the sphygmograph has generally been understood as a ‘replacement for observations formerly made by the tips of an experimenters’ fingers’ (Snyder, 1998: 385), an abjuration of doctorly intuition and all accompanying pitfalls and temptations of subjective thought, this alternate reading demonstrates a more complex and contested remediation of the physician’s touch within the new constellation of techniques of scientific representation.

The ‘educated finger’ in this context would constitute an aspect of what Parisi (2011) has called ‘tactile modernity’, by which the haptic is understood not as counter to the ocularcentrism often considered constitutive of modernity but as in itself a ‘distinctly modern formation, drawing legitimacy from the successful application of rational experimentation to touch within the confines of the experimental … laboratory’ (p. 191). The production of the educated finger may be considered comparable insofar as touch is not subordinated under this new metrological regime but instead valorised and given new predictive power. This can be interpreted as an attempt to solidify the authority of the hand of the clinician, whose power may have been threatened by the anonymising aura of the sphygmograph’s abstract representation.

The French sphygmograph did not enjoy an easy entrance into the clinic of English doctors, and even among those enthusiasts of the device, there was widespread agreement that Marey’s sphygmograph was difficult to learn to use and potentially prone to inaccurate tracings if the sphygmograph were inexpertly applied to the patient’s wrist. The early sphygmographist Dr John Burdon-Sanderson modified Marey’s invention through the addition of an adjustable spring, which could be used to more easily vary the pressure of the plate resting on the radial artery. In his 1867 Handbook of the Sphygmograph, Burdon-Sanderson gives instructions on the amount of pressure to apply when affixing the machine; he felt that 300 g was the standard for a normal pulse, but the utility of the spring was that it allowed for the physician to apply a differentiated amount of pressure, in cases where the ‘compressibility’ of the pulse differed according to the ‘hardness’ of the artery (p. 32). Burdon-Sanderson advised that clinicians work within a range of 100–300 g of pressure, in order to divine, through practised trial and error, the most legible and consistent sphygmic reading (p. 9). The correct amount of pressure to apply to the wrist in the application of the sphygmograph was also a concern of the later sphygmographer Robert Dudgeon (1882), who complained in his own sphygmograph handbook, that ‘All wrists are not equally well adapted for obtaining good characteristic sphygmograms’ (p. 62; Figure 1). The physician, according to Dudgeon (1882), must ensure ‘perfect steadiness’ of the arm (p. 68), though the appropriate tightness of the device will vary between patients, and particular care must be taken with

[t]he wrists of strong athletic young men [which] are generally so encumbered with stiff prominent tendons that it requires some practice to be able to get the instrument properly applied to the artery. In such cases it will often be found advantageous to flex the wrist a little, whereby the cord-like tendons are relaxed and access to the artery obtained. (p. 62)

OPEN IN VIEWER

Figure 1.

Depicting the ‘mode of applying Dr Dudgeon’s Sphygmograph’.

Other physicians felt that the difficulties in application and standardisation of sphygmographic traces were so intractable as to fundamentally compromise the applicability of the technology in the clinic. As late as 1894, William Ewart (1894) was advocating that the sphygmograph be disregarded by clinicians in favour of the finger as the sole vector of analysis, precisely due to the finger’s superior adaptability to differing physiological conditions:

The sphygmograph works against varying pressures with a constant energy – the finger, with varying energies. In this the finger enjoys theoretically a great advantage. Any given pressure of the sphygmograph which is most perfectly adapted for some one of the events of the pulse wave, will be much less adapted for all others; whereas the touch is not limited by any rigid scale of pressures. (pp. 13–14)

For Ewart then, it was precisely the finger’s ability to account for physiological heterogeneity which constituted its foremost strength; while critics of pulse palpation would have claimed this ‘adaptability’ as constitutive of touch’s suspect unruliness, this quality is transformed into a virtue insofar as it is conceptualised as isomorphic with the unruliness and variety of the body itself. Broadbent meanwhile advocated a synthesis of the sphygmograph and the Galenic tactus eruditis, the use of the sphygmographic tracings to produce an ‘educated finger’, which would combine the authoritative data provided by the sphygmograph on the pulse’s duration with the exquisitely differentiated qualitative aspects of the pulse, such as the firmness of the artery and the ‘force’ of its beat, which were sensible only to the finger of the learned practitioner.

Broadbent provided diagrams which cross-referenced sphygmographic traces with his own idiosyncratic representations of how different pulses might feel according to ‘hands on’ sphygmic techniques. Somewhat in contrast to Ewart’s argument, Broadbent (1890) was critical of Dudgeon’s innovations in the addition of the weighted spring, arguing that this variability produced only ‘a gratuitous provision for exaggerations and for extraneous jerks and vibrations’ (p. 33), thus positioning the learned finger as a bulwark against representational artefacting.

What is of interest here in terms of considering the sphygmograph as a haptic medium, concerned with disciplining touch and making it productive, is how touch is conceptualised as ‘filling in’ for the errors and blindnesses of the sphygmographic apparatus. The unruliness of touch as physiological heterogeneity becomes productive insofar as it is ‘tamed’ through the error of the measuring apparatus. The value of touch emerges as the correlate of the sphygmograph’s own capacity for error; that which the device is unable to account for, those ‘qualitative’ aspects of the pulse’s beat, are now valorised as the province of the educated finger. And yet, Broadbent’s (1890) own inscriptions still resemble a sort of quasi-graphical method which serves to produce medical touch as of value to the extent that it might be tied to the project of ‘cultivating the faculty of observation’ (p. 34). Broadbent (1890) criticised the old Galenic pulse palpation handbooks for being over-intricate, introducing distinctions between pulse types and ‘[confusing] the essential features of the important variations of the pulse by overwhelming them in minute distinctions of no practical significance’ (p. 6). Broadbent’s educated finger could feel elements of the pulse which the sphygmograph missed, its ‘firmness’ and ‘force’, but it had to learn to feel them as the sphygmograph did, translating them into lines which did not look out place next to the sphygmograph’s traces (Figure 2).

OPEN IN VIEWER

Figure 2.

Depicting the ‘felt’ experience of the pulse, alongside sphygmographic tracings.

The sphygmograph disciplined the physician’s touch at the same time as it served to expand its diagnostic powers, producing forms of touch which worked in symbiotic relationship to the technology itself; the deft facility with gauging the correct pressure to which the sphygmograph should be tightened on a patient’s wrist, and the manipulation of the arm necessary to prepare it for the attachment of the recording device in a ‘neutral state’. While others have noted the issues of doctorly intervention surrounding the sphygmograph, these have tended to be regarded merely as engineering snares, inhibiting the development of a more untroubled and truly ‘objective’ graphic method (see Borell, 1993; Moss, 2006: 586; Reiser, 1978: 103), rather than as defining the hand as itself a mediating and mediated technology of modern clinical investigation. To examine the rearticulation of touch and the production of the ‘educated finger’ to which the sphygmograph was put is not to attempt to unearth a buried sensuality somehow resistant to the emergent ocularcentrism of the times, but instead to examine how these new techniques of representation served to rearticulate touch as a diffusion and decentralisation of medical authority. If there is an intuition being drawn upon here, it is one which is not simply free-floating and inspired, but instead that which emerges as the very correlate of technologised measurement. Even if acolytes of the graphical method dreamed of a means by which to transcend the vagaries of human intervention, the physician’s touch remained necessary in order to grasp a coherent message from the low murmurs of the body (cf. Foucault, 2001). In the recursion to a remediated and rearticulated but still inescapable tactus eruditis, we witness the frustration of Marey’s desire for a universal and self-evident graphic language, as the stubborn intractability of physiological difference stymies the project of a totalising and fluid abstraction of vital capacity. These lines did not speak for themselves, instead they needed to be coaxed into speech through the careful prodding of the educated finger. The touch described here was not subordinated under the new metrological regime but instead valorised and given new predictive power. This new form of tactility, made ‘proper’ through its absorption of the analytical power of the sphygmograph, produced a kind of technologically enhanced physiological hermeneutics, empowering the one who touches, who decides to touch, who decides.

Photoplethysmography and error

The sphygmograph is now commonly regarded as a failed technology, its finicky application undermining its practical application in the clinic (Moss, 2006: 586). Reiser (1978) has argued that ultimately the sphygmograph’s clinical functions were superseded by Scipione Riva-Rocci’s 1896 sphygmomanometer (cf. Borell, 1993), the inflatable blood pressure cuff which remains in common use today, while in the experimental physiological milieu, the electrocardiograph or EKG attained a growing prestige and acceptance from the early twentieth century onwards (cf. Fleming, 1997: 22). There is a need, however, to avoid overly deterministic or reductively Whiggish accounts of technical or scientific development; it is rarely the case that once technologies and technological competencies become established, they are then irrevocably superseded or otherwise erased from history.

With the emergence of the practice of photoplethysmography, it is possible to detect a comparable remediation and transformation of the haptic as has been described emerging around the sphygmograph. Photoplethysmography is a technique for measuring the pulse which relies on the algorithmic processing of ‘tissue light propagation changes during cardiac cycle’ (Lemay et al., 2014: 107). The origins of the technique date to the 1930s, with pioneering experiments conducted by Alrick Hertzman and Clare Spealman (1937) ascertaining that ‘[v]ariations in the light transmission of a finger due to changes in the blood content may be detected by a photo-electric cell and recorded optically by means of an amplifier and string galvanometer’ (p. 334); the photoplethysmographic device would measure the variations in the ‘absorption of light by a transilluminated tissue … [in order] to detect vascular changes with the photoelectric cell’ (Hertzman, 1938: 328). Photoplethysmography has revealed itself as useful in measuring blood flow and pulse information in conditions where sphygmomanometry or electrocardiography would prove inadequate or otherwise suboptimal, such as in measuring genital blood flow in physiological sex studies (Drucker, 2014).

More prosaically, photoplethysmography has recently attained wide prominence due to its incorporation in mass-market medical devices, ‘driven by the demand for low cost, simple and portable technology for the primary care and community based clinical settings, the wide availability of low cost and small semiconductor components, and the advancement of computer-based pulse wave analysis technique’ (Allen, 2007: 1). The wide availability and affordability of wearable sensors has formed part of the technological milieu which has encouraged the growth of the Quantified Self movement, which preaches ‘self-knowledge through numbers’ and advocates for the use of wearable sensors as an aid to greater knowledge of and autonomy over one’s health and medical decision-making (Wolf, 2010). Self-quantification through wearable sensors has become increasingly banal (Pharabod et al., 2013) and mainstream as wearable medical sensors have been incorporated into a growing range of consumer products. Notably, a photoplethysmographic mechanism is included as part of the Apple Watch (in both its first and second iterations). Green and infrared light-emitting diodes (LEDs) on the underside of the watch rest on the wrist and record information on the wearer’s pulse every 10 minutes (Apple, 2015). Photoplethysmography is considered an especially useful mechanism for wearable heart rate monitors, due to its unobtrusiveness and portability (Allen, 2007: 1).

However, despite both the clinical and mass-market popularity of photoplethysmographical devices, the technology has been and remains compromised by issues of reliability and artefacting. In a 1938 article, Hertzman detailed several sources of error within photoplethysmographic measurements. Most important among these was movement, with Hertzman noting that vigorous movement ‘shifted the position of the skin with respect to the photoplethysmography’, making the measurements unreliable (p. 331). Another important variable was the ‘character of the contact of the plethysmograph with the skin’ (Hertzman, 1938: 331); there was a necessity to avoid ‘blanching’ of the skin through the restriction of blood flow, which would corrupt the legibility of the photoplethysmographic signal (Allen, 2007). Later studies revealed further concerns in the production of repeatable results through photoplethysmographic testing; a 2014 literature review revealed that motion artefacting and pressure disturbances continue to pose threats regarding the corruption of the photoplethysmographic signal, even in modern devices (Tamura et al., 2014: 285).

Indeed, these problems have not escaped the Apple Watch’s use of photoplethysmographic measurement, as many users and technology journalists have noted. Users have complained on online forums that the Apple Watch heart rate data seemed error ridden; the watch displaying, for example, an unusually sedate heart rate of 60 beats per minute after vigorous exercise, or producing data wildly out of alignment with that produced by (more expensive and cumbersome) chest-worn electrocardiograms (Alger, 2015; Singh, 2015). In response, Apple has posted an online ‘explainer’, detailing the workings of the watch’s photoplethysmograph and explaining those factors which might inhibit its successful functioning. These included motion artefacting, as well as skin perfusion (a measure of the deliverance of blood to the skin’s capillaries), with the advisory that ‘skin perfusion varies significantly from person to person and can also be impacted by the environment’ and might, for example, not work when exercising in the cold (Apple, 2015). Apple includes a diagram on their explainer page depicting how to tighten the band in order to produce a ‘snug but comfortable’ fit, an uncanny echo of illustrations of the correct application of the sphygmograph in nineteenth-century handbooks. Apple (2015) also delivers the caveat that ‘[e]ven under ideal conditions, Apple Watch may not be able to get a reliable heart rate reading every time for everybody’. The Fitbit Surge, another popular wearable sensor device which also uses photoplethysmography to measure heart rate activity, has recently had a class-action lawsuit brought against it by users, angered that it ‘[does] not and cannot consistently record accurate heart rates during … intense physical activity’ (Lieff, Cabrasser, Heimann & Bernstein, 2016).

Self-monitoring and improper touch

The significance of this capacity for error in some of the most popular self-tracking technologies on the market today lies in the tension between autonomy and surveillance which underlies the ideology of self-tracking. As noted, the popularity of the Apple Watch and other such wearable apparatuses can be seen to reflect the mainstreaming of the Quantified Self movement, dedicated to the pursuit of ‘self-knowledge through numbers’ (Wolf, 2010). While often caricatured as mindless automatons (Morozov, 2013), self-quantifiers have invoked a critique of medical diagnosis oddly congruent with the critique of the ‘disappearance of the patient’ described earlier in the article. One of the movement’s foremost advocates, the Wired columnist Gary Wolf, has advocated for the use of wearable sensors to produce ultra-detailed self-portraits of the user’s vital capacity, which could be used to confound diagnosis based on population-level medical abstractions; Wolf (2010) argues that ‘[p]eople are not assembly lines. We cannot be tuned to a known standard, because a universal standard for human experience does not exist’. Self-quantification in this light seeks to produce a ‘reappearance’ of the patient, a production of the self as a unique individual rather than an ‘average case’.

Accordingly, the Quantified Self movement has been much celebrated as a ‘resistant’ practice, capable, in Dawn Nafus and Jamie Sherman’s (2014) argument, of defining a new metrology which ‘does not cohere’ with established methods of monitoring and control:

The difference cannot be thought of as a standard deviation from a norm; rather, they are not on the same curve. Their differences are embodied in diverging practices that cannot be meaningfully assembled together using the categories in current circulation. (p. 1791)

That is, for Nafus and Sherman, the Quantified Self represents a challenge to the imposition of top-down ‘normative standards’, especially regarding the definition of health. The Quantified Selfer here is celebrated for her innovative approach to redefining metrology, raising the prospect of a new concept of measurement which is able to take into account the personal, the subjective and the idiosyncratic, all aspects which Nafus and Sherman would argue have been obscured by previous paradigms of measurement.

However, as Gina Neff (2013) argues, the growth in the use of self-tracking technologies has not led and likely will not lead to a ‘power shift’ from health-care providers to patients. Instead, this change has potentially tied patients ever more tightly to the medical establishment, insofar as the masses of data produced by methods of self-quantification must still be interpreted, or at the very least ‘verified’, by medical professionals. As such, rather than constituting anything like a ‘liberation’ from apparatuses of governmentality, one may see this emphasis on self-quantification as constituting part of the trend towards what Robert Aronowitz (2009) defines as the ‘convergence of risk and disease’, by which ‘early diagnosis and aggressive treatment have led to symptom-less and sign-less disease … in which treatments have largely been aimed at altering the disease’s future course’ (p. 417). The development of such a model of medical practice emphasises individual responsibility for detecting whether or not one’s biological state deviates from particular norms, norms which may be constantly revised by the appropriate medical bodies and hence require constant monitoring (Rose, 2007: 40). Hence, rather than ‘liberating’ patients from the grasp of the medical establishment, the emphasis on self-monitoring and ultra-early detection of disease imbricates patients ever more deeply within it, insofar as new methods for managing medical risk mean that one is able to begin undergoing treatment many years earlier than would have otherwise been the case.

Considering the Apple Watch then as a form of haptic media, as a technology for producing proper touch and distinguishing it from an improper touch, leads us into certain epistemological tensions. The Quantified Self movement can be read as a banalisation and extension of Marey’s dreams of a universal graphical method for translating the vital capacity of the body into a fluid and self-evidently ‘objective’ representational system. The educated finger theorised by Broadbent served to corporealise the project of the graphical method, disciplining touch as a component of a new diagnostic apparatus. The Apple Watch’s haptic training is far more ambiguous and fraught. Even as Broadbent’s educated finger was calibrated according to the error of the sphygmographic recording mechanism, he could still feel confident that this touch could be made to provide reliable sensory data which could be read alongside the traces of the sphygmograph, which would ‘fill in’ for the gaps in the sphygmograph and hold a comparable evidentiary authority. The error of photoplethysmography, however, rather than being ultimately corrected by the touch of the user, is compounded, dispersed. The knowledge of this error circulates in online discussion boards and collective action lawsuits; Apple’s guide to use of the Apple Watch seems only to confirm the unreliability of the recording mechanism, rather than to offer a conclusive solution. The user’s touch here then becomes a means of corporealising this failure; the user is instructed on how to correct the watch, but this touch can never be authoritative in the same way as the doctor’s educated finger, ‘even under ideal conditions’. The truly ‘proper’ touch lies elsewhere, in the hand of the doctor who remains authorised to give a correct reading of the patient’s vital capacity. In effect, the Apple Watch, while ostensibly offering users a form of empowerment through embodied self-knowledge, in fact serves to generalise a form of ‘improper’ touch, touch which fails to overcome error, which fails to produce the body as a coherent communicative medium. In this way, we can see the distinction between self-monitoring and self-knowledge; even if the instruments of self-quantification demand an intensive form of self-surveillance, such monitoring struggles to attain the status of a definitive knowledge, with which one might be able to truly get to grips with, to act upon one’s body. Deborah Lupton (2013) has written of the ambivalence and anxiety which accompanies the routinisation of self-monitoring practices; here, we might locate this ambivalence and anxiety within the cultural techniques of improper touch which are produced through the photoplethysmographic apparatus.

Conclusion

This article has defined an approach to the study of haptic media which considers the way a set of cultural techniques defining the distinction between ‘proper’ and ‘improper’ forms of touch coalesce around media technologies. While the birth of the graphical method and the sphygmograph has often been considered a threat to the embodied authority of the clinician, the examination of the literature surrounding the production of the ‘educated finger’ has demonstrated how the technology remediated the tactus eruditis of the physician, producing new haptic techniques which served to strengthen the authority of the hand of the clinician as the bearer of proper touch. While the contemporary proliferation of wearable sensor technologies has often been understood as generalising or banalising this capacity for knowing the body, the analysis of the role of error within the application of the photoplethysmographic sensor has demonstrated how technologies like the Apple Watch in fact serve to produce a distributed form of improper touch. Rather than producing the body as a coherent communicative medium, the Apple Watch serves to enmesh the body within a network of error, which the ordinary user is unable to ultimately correct, thus denying the promised capacity for bodily enquiry. Photoplethysmographic sensors’ capacity for error serves to produce an intimate co-dependency between self-quantifier and medical practitioner; even as the proliferation of wearable sensor technologies encourages users to practice an intense form of surveillance over their own vital capacity, the ‘truth’ of the data produced by wearable devices remains compromised and necessitates validation and confirmation by trained medical professionals. Even as self-quantification devices seem to offer the promise of empowering the user to define oneself in all of one’s individuated biological difference, the place of the doctor is reaffirmed as necessary in order to bridge the gaps in the recording apparatus. The significance of this comparison then serves to contest established discourses surrounding the place of embodied knowledge in the late nineteenth century and contemporary periods. Despite common understandings, the tactile remained a significant locus of medical authority post the introduction of graphical recording technologies, while in our own period, the attempt to generalise the project of bodily self-knowledge remains trammelled by a landscape of technical and discursive snares, which serve to generalise and enshrine a form of technologised improper touch.