- Personal Homepage
Jeffrey Bardzell is an Associate Professor of HCI/Design and new media at the School of Informatics in Indiana University - Bloomington. With a Ph.D. in Comparative Literature and Minor in Philosophy, Bardzell brings a humanist perspective to HCI and is known for developing a theory of interaction criticism. His other HCI specialties include aesthetic interaction, user experience design, amateur multimedia design theory and practice, and digital creativity.
- Publication period start
- Publication period end
- Number of co-authors
Number of publications with favourite co-authors
Most productive colleagues in number of publications
Bardzell, Shaowen, Bardzell, Jeffrey, Forlizzi, Jodi, Zimmerman, John, Antanitis, John (2012): Critical design and critical theory: the challenge of designing for provocation. In: Proceedings of DIS12 Designing Interactive Systems , 2012, . pp. 288-297. http://dx.doi.org/10.1145/2317956.2318001
Bardzell, Shaowen, Rosner, Daniela K., Bardzell, Jeffrey (2012): Crafting quality in design: integrity, creativity, and public sensibility. In: Proceedings of DIS12 Designing Interactive Systems , 2012, . pp. 11-20. http://dx.doi.org/10.1145/2317956.2317959
Bardzell, Jeffrey, Nichols, Jeffrey, Pace, Tyler, Bardzell, Shaowen (2012): Come meet me at Ulduar: progression raiding in World of Warcraft. In: Proceedings of ACM CSCW12 Conference on Computer-Supported Cooperative Work , 2012, . pp. 603-612. http://dx.doi.org/10.1145/2145204.2145296
Bardzell, Jeffrey (2011): Interaction criticism: An introduction to the practice. In Interacting with Computers, 23 (6) pp. 604-621.
Forlizzi, Jodi, DiSalvo, Carl, Bardzell, Jeffrey, Koskinen, Ilpo, Wensveen, Stephan (2011): Quality control: a panel on the critique and criticism of design research. In: Proceedings of ACM CHI 2011 Conference on Human Factors in Computing Systems , 2011, . pp. 823-826. http://dx.doi.org/10.1145/1979742.1979497
Bardzell, Jeffrey, Bardzell, Shaowen, Nardi, Bonnie A. (2011): World of Warcraft as a global artifact. In: Proceedings of ACM CHI 2011 Conference on Human Factors in Computing Systems , 2011, . pp. 169-172. http://dx.doi.org/10.1145/1979742.1979485
Bardzell, Shaowen, Churchill, Elizabeth, Bardzell, Jeffrey, Forlizzi, Jodi, Grinter, Rebecca, Tatar, Deborah (2011): Feminism and interaction design. In: Proceedings of ACM CHI 2011 Conference on Human Factors in Computing Systems , 2011, . pp. 1-4. http://dx.doi.org/10.1145/1979742.1979587
Kannabiran, Gopinaath, Bardzell, Jeffrey, Bardzell, Shaowen (2011): How HCI talks about sexuality: discursive strategies, blind spots, and opportunities for f. In: Proceedings of ACM CHI 2011 Conference on Human Factors in Computing Systems , 2011, . pp. 695-704. http://dx.doi.org/10.1145/1978942.1979043
Bardzell, Shaowen, Bardzell, Jeffrey (2011): Towards a feminist HCI methodology: social science, feminism, and HCI. In: Proceedings of ACM CHI 2011 Conference on Human Factors in Computing Systems , 2011, . pp. 675-684. http://dx.doi.org/10.1145/1978942.1979041
Bardzell, Jeffrey, Bardzell, Shaowen (2011): Pleasure is your birthright: digitally enabled designer sex toys as a case of third-wave H. In: Proceedings of ACM CHI 2011 Conference on Human Factors in Computing Systems , 2011, . pp. 257-266. http://dx.doi.org/10.1145/1978942.1978979
Jung, Heekyoung, Altieri, Youngsuk L., Bardzell, Jeffrey (2010): Computational objects and expressive forms: a design exploration. In: Proceedings of ACM CHI 2010 Conference on Human Factors in Computing Systems , 2010, . pp. 3433-3438. http://doi.acm.org/10.1145/1753846.1753997
Jung, Heekyoung, Youngsuk, Altieri, Bardzell, Jeffrey, Scheible, Jürgen, Pierce, James, Paulos, Eric, Yim, Ji-Dong, Shaw, Christopher (2010): Demo Hour. In Interactions, 17 (6) pp. 6-7. http://dx.doi.org/10.1145/1897239.1897241
Bardzell, Jeffrey, Bolter, Jay, Lowgren, Jonas (2010): Interaction criticism: three readings of an interaction design, and what they get us. In Interactions, 17 (2) pp. 32-37. http://dx.doi.org/10.1145/1699775.1699783
Blythe, Mark, McCarthy, John, Light, Ann, Bardzell, Shaowen, Wright, Peter, Bardzell, Jeffrey, Blackwell, Alan (2010): Critical dialogue: interaction, experience and cultural theory. In: Proceedings of ACM CHI 2010 Conference on Human Factors in Computing Systems , 2010, . pp. 4521-4524. http://doi.acm.org/10.1145/1753846.1754189
Cockton, Gilbert, Bardzell, Shaowen, Blythe, Mark, Bardzell, Jeffrey (2010): Can we all stand under our umbrella: the arts and design research in HCI. In: Proceedings of ACM CHI 2010 Conference on Human Factors in Computing Systems , 2010, . pp. 3163-3166. http://doi.acm.org/10.1145/1753846.1753944
Pace, Tyler, Bardzell, Shaowen, Bardzell, Jeffrey (2010): The rogue in the lovely black dress: intimacy in World of Warcraft. In: Proceedings of ACM CHI 2010 Conference on Human Factors in Computing Systems , 2010, . pp. 233-242. http://doi.acm.org/10.1145/1753326.1753361
Bardzell, Jeffrey, Pace, Tyler, Terrell, Jennifer (2010): Virtual fashion and avatar design: a survey of consumers and designers. In: Proceedings of the Sixth Nordic Conference on Human-Computer Interaction , 2010, . pp. 599-602. http://doi.acm.org/10.1145/1868914.1868983
Bardzell, Jeffrey (2009): Interaction criticism and aesthetics. In: Proceedings of ACM CHI 2009 Conference on Human Factors in Computing Systems , 2009, . pp. 2357-2366. http://doi.acm.org/10.1145/1518701.1519063
Jung, Heekyoung, Altieri, Youngsuk L., Bardzell, Jeffrey (2009): SKIN: designing aesthetic interactive surfaces. In: Proceedings of the 4th International Conference on Tangible and Embedded Interaction , 2009, . pp. 85-92. http://doi.acm.org/10.1145/1709886.1709903
Bardzell, Jeffrey, Pace, Tyler, Brunetti, Laura, Huang, Qian, Perry, Nina, Gim, Hyewon (2009): Emerging Standards in Virtual Fashion: An Analysis of Critical Strategies Used in Second L. In: HICSS 2009 - 42st Hawaii International International Conference on Systems Science 5-8 January, 2009, Waikoloa, Big Island, HI, USA. pp. 1-10. http://dx.doi.org/10.1109/HICSS.2009.184
Bardzell, Jeffrey, Bardzell, Shaowen, Pace, Tyler, Karnell, Jeremi (2008): Making user engagement visible: a multimodal strategy for interactive media experience res. In: Proceedings of ACM CHI 2008 Conference on Human Factors in Computing Systems April 5-10, 2008, . pp. 3663-3668. http://doi.acm.org/10.1145/1358628.1358909
Bardzell, Shaowen, Bardzell, Jeffrey, Pace, Tyler, Reed, Kayce (2008): Blissfully productive: grouping and cooperation in world of warcraft instance runs. In: Proceedings of ACM CSCW08 Conference on Computer-Supported Cooperative Work , 2008, . pp. 357-360. http://doi.acm.org/10.1145/1460563.1460621
Blythe, Mark, Bardzell, Jeffrey, Bardzell, Shaowen, Blackwell, Alan (2008): Critical Issues in Interaction Design. In: Proceedings of the HCI08 Conference on People and Computers XXII , 2008, . pp. 183-184. http://www.bcs.org/server.php?show=ConWebDoc.21491
Bardzell, Jeffrey, Bardzell, Shaowen (2008): Interaction criticism: a proposal and framework for a new discipline of HCI. In: Proceedings of ACM CHI 2008 Conference on Human Factors in Computing Systems April 5-10, 2008, . pp. 2463-2472. http://doi.acm.org/10.1145/1358628.1358703
Bardzell, Shaowen, Wu, Vicky, Bardzell, Jeffrey, Quagliara, Nick (2007): Transmedial interactions and digital games. In: Inakage, Masa, Lee, Newton, Tscheligi, Manfred, Bernhaupt, Regina, Natkin, Stéphane (eds.) Proceedings of the International Conference on Advances in Computer Entertainment Technology - ACE 2007 June 13-15, 2007, Salzburg, Austria. pp. 307-308. http://doi.acm.org/10.1145/1255047.1255141
Bardzell, Jeffrey, Bardzell, Shaowen, Birchler, Craig, Ryan, William (2007): Double dribble: illusionism, mixed reality, and the sports fan experience. In: Inakage, Masa, Lee, Newton, Tscheligi, Manfred, Bernhaupt, Regina, Natkin, Stéphane (eds.) Proceedings of the International Conference on Advances in Computer Entertainment Technology - ACE 2007 June 13-15, 2007, Salzburg, Austria. pp. 216-219. http://doi.acm.org/10.1145/1255047.1255093
Blevis, Shunying, Bardzell, Jeffrey, Wroblewski, Nancy (2007): Feed the Dragon Wisely: Designing for Childhood Awareness as a Means of Lifelong Obesity P. In: Stephanidis, Constantine (eds.) Universal Access in Human-Computer Interaction. Applications and Services, 4th International Conference on Universal Access in Human-Computer Interaction, UAHCI 2007 Held as Part of HCI International 2007 Beijing, China, July 22-27, 2007 Proceedings, Part July 22-27, 2007, Beijing, China. pp. 860-868. http://dx.doi.org/10.1007/978-3-540-73283-9_93
Bardzell, Jeffrey, Bardzell, Shaowen, Briggs, Christian, Makice, Kevin, Ryan, William, Weldon, Matt (2006): Machinima prototyping: an approach to evaluation. In: Proceedings of the Fourth Nordic Conference on Human-Computer Interaction , 2006, . pp. 433-436. http://doi.acm.org/10.1145/1182475.1182530
19.5 Commentary by Jeffrey Bardzell
King Azaz: “You couldn’t have tea for two without the tea, could you?”
The thesis of my commentary is that Noam Tractinsky’s chapter on Visual Aesthetics reflects and champions only a portion of work on aesthetics that has influenced HCI, that he has (hopefully unintentionally) marginalized alternative approaches, and that a more balanced picture needs to be offered. To defend this thesis, my commentary will do the following:
The most substantial criticism of Tractinsky’s article that I will make is that his account of aesthetics in general and visual aesthetics in HCI in particular is extremely limited, rather than comprehensive as he promises, and it excludes both major aesthetic ideas and also major aesthetic contributions to interaction design. Such a marginalization therefore offers, in my view, a distorted account of his putative topic—visual aesthetics in HCI—and might encourage readers to miss opportunities to work towards a goal that all of us have in common: a desire to make interaction more aesthetic.
While my commentary takes a critical position with regard to Tractinsky’s essay, I want to stress up front both that Tractinsky’s research, as well the work of those within his tradition (including nearly all my fellow commenters on this chapter), has had enormous positive influence over the years in HCI, that I myself both teach and use such work, and that I broadly agree with Tractinsky’s prescriptions for the future of such research. My purpose is not to attack what clearly is a rigorous, useful, and influential research approach; rather, it is to critique its positioning and its limits and to explore alternative formulations that complement and strengthen, but do not replace, its place in HCI.
At the end of the day, we are successful if we have helped designers make interactions that are more aesthetic, not if we win academic turf wars.
19.5.1 An anti-theory theory and its consequences
Tractinsky opens his chapter by defining and scoping his operational understanding of “visual aesthetics” by separately defining “visual” and “aesthetics”:
By using two standard dictionary definitions and by bracketing aside the “various other phenomena studied under the ‘aesthetics’ heading,” Tractinsky categorically disengages with millennia of aesthetic thinking in philosophy, art history, literature, architecture, and film. Yet by referencing brain sciences in his definition of the visual, he indicates his willingness to engage with scientific scholarship. Thus, Tractinsky has signaled his intention—one that he will carry out throughout the article—to be scholarly about the empirical science of aesthetics and strategically unscholarly about philosophy of aesthetics. The latter is not an accusation from me but rather how he describes himself:
Here we have the strong claim that interdisciplinary attempts at defining aesthetics confuse the scientific community, and that the way forward is deceptively simple: to base aesthetic research in HCI on what he calls “ordinary people’s intuitive interpretation of the terms,” which are reflected in the dictionary definitions and to disregard all that multidisciplinary handwringing. With that established, research is “primarily empirical and ... characteristically descriptive (i.e., ‘what is considered beautiful’) rather than normative (i.e., what should be considered ‘beautiful’)” Section 19.1. Here, Tractinsky is again positioning himself and other researchers outside of aesthetic debates: his job is simply to discover what is already out there in the world and not to take positions on (which would place him inside) aesthetic debates.
Thus, Tractinsky’s success with his “ordinary language” definition of aesthetics hinges whether the view of aesthetics that he presents reasonably reflects people’s ordinary views of aesthetics. If it does, then his categorical rejection of millennia of humanist scholarship on aesthetics seems reasonable, since it does not appear to provide anything useful for his research project and may even harm it through multiple and confusing technical definitions. Moreover, if he does hold to an ordinary definition of aesthetics, then he can justifiably argue that he is outside of aesthetic debates and simply discovering what’s already there (which is what he means when he talks about being “descriptive” and not “normative”).
But if Tractinsky’s work turns out not to be based on ordinary views of aesthetics, then he has set up two problems for himself: first, he is vulnerable to the criticism that his approach lacks scholarly rigor because he seems simply to have chickened out of engaging with the conceptual difficulty that most of the rest of us find intrinsic to aesthetic reasoning; and second, he loses this right to claim that he is merely descriptive and not normative, and therefore that his proposed science of aesthetics effectively becomes another candidate philosophical view of aesthetics, and hence subject to the very sort of philosophical critique that he seeks to circumvent by avoiding theory in the first place.
I argue that it is easy to see that the notions of aesthetics that Tractinsky cites with approval over the course of the article neither match the brief and nearly vacuous dictionary definitions he quotes, nor does the view of aesthetics that he promotes reflect the common, non-scholarly, intuitive views of aesthetics held by ordinary people. Therefore, I argue that what “guides research on visual beauty in HCI” is not a simple idea intuitively shared by most people, but rather is a sophisticated, technical, and robustly academic theory, that this theory inevitably has normative dimensions to it (and, incidentally, it’s hard for me to understand how anything that has “implications for design” is not intrinsically normative), and therefore deserves critical scrutiny in order to be used rationally. In short, I argue that Tractinsky introduces an aesthetic dogma that is cloaked in supposedly descriptive empirical science; as a field, we need to disentangle the two so that we don’t throw out the baby with the bathwater.
19.5.2 Tractinsky's extra-ordinary language definition of aesthetics
In the course of the article, Tractinsky cites with approval a diverse (and unabashedly scholarly) array of aesthetics concepts that anyone would be hard-pressed to claim are part of what he calls “ordinary people’s intuitive interpretation of the terms” and certainly won’t be found in collegiate dictionaries under “aesthetics.” These include the following:
There is no space here to evaluate these different concepts (each of which brings with it insights and difficulties), and Tractinsky is certainly justified in outlining them as influential and important aesthetic ideas that have been explored in psychology, design, and more recently HCI. But in embracing all of these ideas, Tractinsky has outlined a philosophical infrastructure for an academic theory of aesthetics and departed from a commonsense or dictionary notion of the aesthetic. I believe I can reasonably assert that none of the ideas in the list above are part of “ordinary people’s intuitive interpretation of the terms” and I can also reasonably assert that all of these ideas are foundational to Tractinsky’s research project. Therefore, his anti-theory stance is invalid: he has constructed for himself a theoretical apparatus constituted by a set of technical and interlocking ideas, and he is not relying on a simple dictionary definition as he claims.
And it is building from this apparatus that Tractinsky offers a particular academic theory of aesthetics in Section 19.2, where he outlines a flow model for what he calls “the aesthetic process.” According to this model, design variables (including “low-level” attributes such as use of color and symmetry and “high level” attributes such as novelty, typicality, and fluency) lead to (cause?) aesthetic valuations (such as visceral, behavioral, and reflective ones, following Norman) and in turn aesthetic valuations lead to (cause?) outcomes (such as affects and emotions, brand trust, perceived usability, perceived product character). Each of these relationships is modified by the type of system used, cultural inputs, domain, type of task, and aesthetic tastes. In this model, Tractinsky intermingles traditional aesthetic categories (e.g., symmetry, form, composition, balance of typicality and novelty, diversity, craftsmanship, expressivity) and the language of experimental psychology (e.g., the information processing metaphor, variables, moderators, input/output, performance, motivation).
This is a very sophisticated theory, and at this point I would like revisit Tractinsky’s foundational claims:
- The two dictionary definitions (which both assert that beauty is that which is pleasing) accurately reflect ordinary people’s intuitive understandings of aesthetics and these are that which “guides research on visual beauty in HCI”
- And therefore:
- No academic theory of beauty beyond a dictionary definition is needed to pursue this research and
- Empirical science on visual aesthetics for HCI is descriptive and not normative because it stays out of multidisciplinary aesthetic debates
But merely summarizing Tractinsky shows that what he has offered here is far more than the dictionary definition and adds up to an information processing theory of aesthetics in which design inputs yield evaluation outputs, and evaluation outputs become inputs for outcomes outputs. According to this theory, the process itself has “moderators” including use context, system attributes, culture, and individual differences as inputs. This is substantially more specific and more guiding than his dictionary definition (“a pleasing appearance or effect: Beauty”), a phrase that is vague to the point of meaninglessness and is obviously insufficient to guide the empirical research of aesthetics in HCI!
I have, I believe, established that in spite of what he claims, Tractinsky operates with a much more sophisticated aesthetic philosophy than a dictionary definition, but I have not commented on how his theory relates to ordinary people’s intuitive understandings. But I think some very simple reflection can disabuse us of that pretention as well. For example, I teach a course called Interaction Culture to HCI students within a School of Informatics and Computing. In the opening minutes of the first day of that class, before I even introduce myself or give students the syllabus, I show the first few minutes of an art film that has achieved some popular success. This year, I showed the first 5-6 minutes of Run Lola Run, a 1998 German action film with a philosophical subtext. The opening of the film includes 3D computer graphics, 2D cartoon animation, 2.5 post-production compositing (e.g., titles), live motion acting, heavy image manipulation, and an intense techno beat; in addition to the bewildering visual assault there is also a short but mysteriously philosophical verbal script whose relationship to the images is not obvious. After showing this introduction once, I ask students to simply talk out loud about their reactions to it. Some describe how it made them feel—excited, anxious, curious. Others talk about symbols that constitute the work as an artifact—how the heavy use of clock imagery and the metronomic beat of the techno soundtrack reinforced each other. Others offer suggestions about what the director may have intended or was trying to say or do, how this fits in the German cinematic tradition, etc. Still others talk about what was happening when the film was made (the 1990s in Germany or popular culture in general). Are these not ordinary, common, and intuitive aesthetic reactions? Such interpretative strategies—and not aesthetic processing theory—are taught to us as children in schools and at home and come almost naturally to us as adults. My sense is that if one really wants to understand what ordinary people intuitively do, all one needs to do is watch ordinary people intuitively encounter beautiful things. Tractinsky goes far beyond that in his research, and rightly so, but it is disingenuous to claim that he neither needs nor uses any disciplined academic theory of aesthetics.
So Tractinsky’s aesthetic processing theory is not intuitive or common. Nor is it an empirically discovered fact in the world. That theory is thus a philosophical theory of aesthetics, or, in the language of logical positivism (which I personally reject but am not certain that Tractinsky does), a dogma. Aesthetic processing is a theory of aesthetic response constructed out of the methodological and conceptual apparatus of information processing psychology and adapted using aesthetic vocabulary from both the sciences and the arts. For example, Tractinsky’s own seminal distinction between “classical” and “expressive” aesthetics are both derived from the history and philosophy of art, the former attributable to philosophers such as Hutcheson, Bell, and Beardsley and the latter attributable to philosophers such as Langer and Collingwood. (Both concepts have also been developed and critiqued for over a century, and the conceptual difficulties of each are well known among analytic philosophers of art, if not the HCI community.)
19.5.3 The aesthetic processing theory and its discontents
My argument thus far has been largely philosophical, seeking to show that the conceptual edifice on which Tractinsky builds his aesthetic processing theory is flawed inasmuch as it claims to be a-theoretical when it clearly is not.
However, the real point of all this is much more practical: I want to show that the theoretical blindness built into Tractinsky’s philosophy of visual aesthetics has important consequences for HCI that need to be dealt with. Two of them are as follows:
Of the two, the second is a more serious shortcoming, especially given that Tractinsky claims that “The objective of this paper is to survey the field of visual aesthetics in HCI,” when in fact it surveys only a favored subset of that field. I will address both of these practical consequences in what follows.
19.5.4 What do we want from visual aesthetics in HCI?
Scientific research is expensive, and one way or another the public pays for it, and so any scientific agenda should deliver some sort of public good. What is the social value of aesthetic research in HCI? Let us follow Tractinsky’s example by beginning with the dictionaries:
These definitions are not terribly helpful. The only public good I can imagine deriving from them is that this research will make interacting with digital systems more “pleasing.” But surely this is a weak argument: just as my insurance company won’t pay for voluntary cosmetic surgery to make me look more pleasing, I can’t imagine policymakers in this era of austerity investing in scientific research to make user interfaces more “pleasing.”
Tractinsky himself offers numerous and much better arguments in the course of his essay. He notes that aesthetics has long been integrated within design disciplines whose professional and socioeconomic success is beyond dispute and whose theories and methods can be leveraged in HCI and interaction design. He notes that Gestalt psychology has shown that aesthetic criteria are linked with other design values, including usefulness and suitability, his most powerful argument to the HCI community, which historically has had a orientation towards the useful. He adds that aesthetics satisfies human needs (not merely superficial desires), contributes to wellness, and seems to be linked to the formation and experience of the self, making the argument that aesthetics is good for people’s lives. He also notes that aesthetics helps otherwise similar products differentiate themselves, thus contributing economic value (which, in the case of Apple, has been substantial). I accept each of these (normative) arguments as stated and stress that he has here offered a number of social benefits that can emerge from this research, most of which are functional in nature: aesthetics supports usability, aesthetics satisfies needs, aesthetics contributes to the self, aesthetics contributes to economic prosperity.
If we turn to other philosophers of aesthetics besides Tractinsky (and I count him as one, whether or not he does), we can see many other arguments commonly made that would support the idea that this research contributes to the public good. Common claims in the aesthetic literature include the following statements about aesthetic response and/or aesthetic experience (synthesized from Bardzell, 2011):
Going back to Plato, aesthetics has been implicated not only in pleasures but also its role in contributing to (or detracting from) an educated and responsible public, and this predisposition is amply reflected in the list above. As interactive technologies continue to replace older media forms in mediating how people interact with themselves, each other, and the world, making interaction aesthetic in these senses seems to be imperative, rather than optional. The cultivation through aesthetic engagement of ourselves as perceptive, imaginative, and insightful citizens (an epistemological position) would seem to depend increasingly on human-computer interaction.
I have briefly sketched 3 simple arguments justifying aesthetic interaction: a hedonic argument, a functionalist argument, and an epistemological argument. While I personally support all three, it seems the second (Tractinsky’s functionalist argument) and the third (aesthetic philosophers’ epistemological argument) could be the most compelling for policymakers as well as researchers and practitioners within the field of HCI and interaction design.
19.5.5 Critically assessing the aesthetic processing theory in HCI
Now that I have sketched out three primary arguments in favor of pursuing visual aesthetics in HCI, and I have earlier established that Tractinsky offers a theory of aesthetic processing as a means to do so, so we are finally in a position to evaluate the strengths and weaknesses of this aesthetic processing theory for HCI.
22.214.171.124 Strengths of the aesthetic processing theory for HCI
Tractinsky does much of my work for me here, since his article systematically summarizes the achievements of this tradition, and by and large I accept his account of that at face value; as I said in my introduction, I use aesthetic processing research in both my own research and teaching. I’ll add a few points here for emphasis.
I begin with the point that although aesthetic philosophy, literary theory, and art history (etc.) have anticipated and expounded on many of the concepts used in aesthetic processing theory, nothing that the humanities tradition has done anticipated the exciting and specific findings about how quickly judgments and evaluations are made, how such judgments are causally linked to closely related phenomena (e.g., affect, human needs), and how all of the above influence behavior. If one accepts Norman’s three-part distinction between visceral, behavioral, and reflective aesthetic perception and evaluation (and I do not, but that’s beside the point here), it is clear that the aesthetic processing approach has offered unparalleled insight into the visceral processing—a fact that is not lost on marketers as well as designers. Some of the questions that this research is well positioned to answer include user perceptions of a system’s usability and beauty, behavioral consequences of aesthetic valuations, attitudes (such as trust) towards organizations represented in systems, brand personality, and perceived system performance.
Aesthetic processing research provides new discoveries about the very mechanisms of aesthetic perception and experience, and their implications go beyond HCI and should influence anyone interested in visual aesthetics in the humanities and sciences. In their critique of McCarthy & Wright for failing to clarify much specifically about the content of experience, Hassenzahl, Diefenbach, & Göritz (who in my view operate in a similar paradigm with Tractinsky) argue that this sort of approach can reveal much about how experience is constructed from perceptions, motivations, affect, and judgments and thereby offer useful implications for designers (Hassenzahl, Diefenbach, & Göritz, 2010); they are right about the strength of their approach compared to McCarthy & Wright’s, though I believe McCarthy & Wright also offer complementary strengths that Hassenzahl, Diefenbach, & Göritz do not acknowledge. Anyway, aesthetic processing approaches to aesthetics have revealed much about the nature of fast aesthetic judgments of interactive systems, have done so with useful implications for design, and have offered compelling evidence to support their findings.
Another benefit of this approach is that Tractinsky’s work in the 1990s decisively undermined prevailing attitudes (especially the high-profile urgings of Norman and Nielsen at the time) that viewed the aesthetic as inherently in conflict with the usable. Tractinsky helped change the field by offering evidence that usability and aesthetics were not, in fact, in conflict. In so doing, he helped create space for others of us interested in aesthetic interaction, not by making a nice argument about aesthetics, e.g., using Dewey or Heidegger to argue for a more robust aesthetic sensibility, but by proving the prevailing wisdom wrong on its own terms: experimental science. I am hardly a fan of scientism for its own sake, but the ability of this paradigm of research to leverage science to contribute to the scholarship of aesthetics (and aesthetic philosophers and literary theorists are beginning to read cognitive science on aesthetics), and specifically to advocate successfully for more work on aesthetics in HCI, is an enduring achievement.
Of the three arguments I offered supporting research on visual aesthetics in HCI (hedonic, functionalist, and epistemological), a strength of this tradition is that it speaks to the first two in powerful ways.
126.96.36.199 Weaknesses of the aesthetic processing theory for HCI
As with any synoptic theory of aesthetics, the aesthetic processing model as presented has a number of practical weaknesses.
The first weakness stems from the fact that the aesthetic processing model differs in important ways from our common experience of the aesthetic: as I have argued, it is too steeped in information processing theory to fit with an ordinary person’s experience. One aspect of that is the reductive approach that aesthetic processing theory relies on. Its level of analysis is often faster than what humans are consciously aware of: for example, the finding that broad aesthetic judgments are made within x number of milliseconds may have application for professional designers, but it certainly sits outside of an ordinary person’s experience. Likewise, the reduction of holistically experienced phenomena (e.g., emotion) into constituent, measurable parts is also alien to the common aesthetic experience. Finally, many of the example interfaces shown in these studies are frankly ugly (see Figure 19.3 in the main chapter, and Figures 19.1A-B in Lindgaard’s comment for typical examples), and the reason for this, I believe, is that the researchers’ intellectual goal tends to be something like “given that we have to design, e.g., a Windows email dialog box, what choices will make it be perceived as more rather than less beautiful?” rather than “how do we design a beautiful interaction?” It is a strength of aesthetic processing theory that it is able to answer the first question, but it is a weakness of the approach that it can’t put Humpty Dumpty back together again, that is, that once complex wholes are reduced into parts, the process can’t be reversed. Pragmatically, this research is not typically used to promote beautiful interactions, but rather more beautiful (than they otherwise would have been) interactions.
Second, the aesthetic processing model is also comparatively weak at ethical and socio-cultural considerations. I agree with the ancient Greeks in seeing ethics and aesthetics as so deeply intertwined as to be inseparable, but in doing so, one must move the aesthetic from the realm of the perceptual and into the hermeneutic. Tractinsky writes, “this chapter is not about ... our reactions to object qualities that do not immediately and primarily stem from its visual attributes” (Section 19.1), so he seems to be excluding attributes that influence how we perceive the visual in the first place. In the Strengths section, I mentioned some of the specific contributions of empirical approaches based on the aesthetic processing theory: user perceptions of a system’s usability and beauty, behavioral consequences of aesthetic valuations, attitudes (such as trust) towards organizations represented in systems, brand personality, and perceived system performance. Not only do none of these have a strong ethical dimension to them, but worse, armed with the findings of this research, marketers and designers are in a better position to manipulate users, because so much of this research provides practical guidance on how viscerally to influence perceptions, behaviors, and affects through design choices.
Aesthetic philosophers have long taken to task theories that, like this, focus tightly on the perception of objective visual qualities precisely because of concerns about manipulation. For example, analytic philosopher Mary Devereaux (1998) investigates this issue in Riefenstahl’s Triumph of the Will, a Nazi propaganda film that was made with such stunning cinematic vision and craftsmanship that it won awards across Europe in the late 1930s and continues to influence the language of cinema today. If we limit ourselves to “objects or qualities” that “immediately and primarily stem from its visual attributes” we have little mechanism for dealing with the seductive evil of Riefenstahl’s film; on the aesthetic processing theory, such a film is beautiful. Now, Triumph of the Will is an extreme case, but it calls into question the hermeneutic effects of moral sensibility within of aesthetic sensemaking and reflection. Consider a 2007 video game produced and distributed by Hezbollah called Special Force 2, a first-person shooter in which the player’s goal is to shoot and kill as many Israeli soldiers as possible: ethics and aesthetics will converge in human-computer interaction in diverse and socially important ways.
An example pertaining to the social context of aesthetic response is Duchamp’s Fountain, which is an otherwise ordinary urinal that the artist signed (under a nom de plume) and put on display in a museum, making it “art.” There is nothing visually interesting about this urinal: if it is art, it is only so on account of its having been placed in an art museum, on a pedestal as a work of art, by someone recognized by his community as an artist. All of the aesthetic processing involved in this case is at best loosely connected with visual stimuli in the sense that Tractinsky uses the term. In his Commentary to this chapter, Marc Hassenzahl touches on this issue, noting that “beauty” is socially constructed rather than individually processed, though I note that Hassenzahl’s idea has already been developed and debated for decades in philosophical aesthetics, yielding Arthur Danto’s influential notion of “artworld” and three versions of Georgie Dickie’s “institutional theory of art.” One risk of the a-theoretical position outlined by Tractinsky is that it runs the risk of reinventing wheels.
A third weakness is the lack of medium specificity in Tractinsky’s account. An important topic in many aesthetic fields is an effort to discover and/or articulate what a given medium is uniquely good at presenting for aesthetic attention: what can film do that painting cannot, or vice-versa? The aesthetic processing model applies equally to all visual forms, painting, (visual) HCI, film, sculpture, etc. But given that we are in HCI and interaction design, it would seem that interactive, rather than visual, aesthetics would be the target. This concern is raised by Jinwoo Kim is his Commentary to this chapter. This weakness can presumably be explained by the fact that the aesthetic processing model, by virtue of being a processing model, is fundamentally about the cognitive (i.e., processes internal to an individual body and its cognitive processing) and thus has less to say about the visual artifact itself, except inasmuch as it is perceived and rendered available to consciousness.
The final weakness I will mention is that the aesthetic processing theory struggles to deal with the notion of skilled or expert interpretation, e.g., the idea that a critic or a designer has a more robust or better understanding than a layperson—a claim that both critics and designers make, justifiably in my view, because such an understanding is the foundation of their professions and what they are trained for. Tractinsky himself acknowledges this problem in Section 19.3.2 when he calls for more work regarding “the disconnect between designers and users,” but he does not acknowledge the design research (e.g., Schön, 1983; Buchanan, 1995; Cross, 2007) or HCI research (e.g., Lowgren & Stolterman, 2004; Greenberg & Buxton, 2008; Bardzell, 2009; Bardzell, 2011) that has already explored this issue. Lacking a solid account of how legitimate subjective expert judgments are formed, exacerbated by a rejection of the normative aesthetics that they imply, the aesthetic processing account leaves little epistemological space for designers of aesthetic interactions to be anything other than ordinary people armed with empirical data—a characterization that is far from the self-perceptions of most designers and is all but senseless when applied to artists.
In short, the weaknesses of the aesthetic processing account is that it sheds little insight on the third argument in favor of visual aesthetics for HCI I described above: the epistemological one, which focuses on how aesthetic encounters have the long term effect of cultivating our capacities for imaginative perception, insight, critical thought, and empathy.
188.8.131.52 Critical evaluation summary
As I have stressed throughout this essay, the aesthetic processing theory has made contributions to HCI research and, I would argue, aesthetics research more generally. Its analysis of visceral perception and evaluation coupled with its ability to demonstrate subtle causal relations among diverse factors at the very heart of aesthetic perception are peerless as far as I know in any discipline. As Tractinsky notes, there is still much to do here, and as someone committed to aesthetic interaction and aesthetic life in general, I enthusiastically support the continuation of that agenda.
At the same time, as my list of weaknesses hopefully showed, there are major gaps the aesthetic processing theory’s coverage of the whole domain of aesthetics, and these matter. Its reductionism, struggles with ethics and social context, lack of medium specificity, and problems with subjective experts and normative criteria are not merely incidental gaps that just haven’t been filled in yet, but rather reflect intrinsic confounds in the theory itself. The existence of gaps and confounds is hardly a reason to reject a theory, since I can’t think of a theory that doesn’t have both.
But it is a good reason to be more epistemologically open-minded than Tractinsky portrays himself to be and to acknowledge as fellow travelers researchers who work on alternative formulations of aesthetics. Specifically, this encyclopedia chapter aiming to survey the whole field of visual aesthetic for HCI should acknowledge the history and philosophy of aesthetics as pursued throughout the humanities for millennia rather than replace them with vacuous dictionary entries, and also acknowledge work in HCI that builds on these traditions. The inclusion of such contributions would also enrich the prescriptions for the future of aesthetic research in HCI.
19.5.6 Aesthetics, according to the rest of the world
It has been a professional mystery to me, since moving to HCI from my doctoral work in comparative literature and philosophy, why so much of the work on aesthetics in HCI and design is so emphatically cut off from the rest of the aesthetic world. Indeed, reading this research, one might not even know that there is a massive domain of inquiry into aesthetics beyond aesthetic processing and other experimental traditions. Though I repeatedly stress that I find value in aesthetic processing, it is also worth pointing out to readers that aesthetic processing occupies the marginal position academically. If one consults the Wikipedia entry on aesthetics, or searches on the term “aesthetics” at Amazon or Google Scholar, what I am saying will become abundantly clear. Less than 10% of the Wikipedia article could even remotely be considered along the lines of what Tractinsky describes. Oxford’s 4-volume Encyclopedia of Aesthetics was edited by a philosopher (Michael Kelly) and is overwhelmingly not about aesthetic processing theory. Dozens of similar high profile examples in between these two extremes can be found.
I speculate that this self-imposed exile from millennia of interdisciplinary aesthetic thinking reflects a scientific habit that emerged in the Renaissance and came into its own in the Enlightenment and again in the Logical Postivism of the 1920s through 50s, which seeks to reject tradition as dogmatic and confused and seek instead to start anew, using rigorous science and empirically discovered facts to re-investigate phenomena that traditional culture muddled with its dogmas and pet theories. Yet the presence of traditional aesthetic concepts, categories, and systems of relations in aesthetic processing theory reveals the problem. If one uses vocabulary like “classical,” “expressive,” “aesthetic,” “experience,” and “judgment,” one is always already operating from inside the very tradition that is being rejected. And seeking to remove this vocabulary from aesthetic processing won’t work—it can’t be aesthetic processing unless some of this vocabulary is incorporated (and it always is).
So my argument instead is to accept the legitimacy of aesthetic processing but to end its self-imposed exile from the rest of aesthetics. (Humanists would benefit from a little empirical rigor as well—my argument cuts both ways.) What’s needed, then, is to reintroduce the rest of the aesthetic world and juxtapose it to Tractinsky’s essay, so that a fuller picture begins to emerge.
184.108.40.206 The aesthetic disciplines
As Tractinsky correctly argues, aesthetics is fundamental to human life and wellness. Not surprisingly, something so important to being human is going to get a lot of attention in human thought. One way to introduce it, then, is to focus primarily on contemporary aesthetic thought and the disciplines in which it unfolds. The following brief introduction is meant to sketch out what some of these disciplines are and the sorts of issues that people within them try to deal with. I also include a handful of introductory readings as starting points for those interested.
220.127.116.11.1 Analytic philosophical aesthetics
Analytic aesthetics takes as its problem the careful evaluation of aesthetic systems of thought or dogmas. By way of self-disclosure, the analytic tradition has been most influential on my own understanding of aesthetics, and my approach to critiquing Tractinsky’s essay is inspired by it. An analytic approach is typically strong at evaluating arguments about aesthetics, frequently (and frustratingly) demonstrating the irrationality of both everyday and sophisticated aesthetic positions. Nearly all of the core concepts of aesthetics (e.g., expression, pleasure, beauty, artworld, realism, experience, style, emotion, form, metaphor, representation, creativity, fiction) and aesthetic mediums (dance, poetry, film, theatre, painting, sculpture, etc.) are analyzed with logical rigor and definitional clarity in this discipline; it remains a mystery to me why analytic aesthetics isn’t more influential in HCI than it currently is. Introductory works include Dickie’s Introduction to Aesthetics: An Analytic Approach (1997), Eldridge’s An Introduction to the Philosophy of Art (2003), and Levinson’s Oxford Handbook of Aesthetics (2003), and for HCI readers I also recommend the first third of Carroll’s Beyond Aesthetics (2001), because it speaks directly to major aesthetic issues in HCI today (including and especially formalist theories like Tractinsky’s and experience theories like McCarthy & Wright’s).
18.104.22.168.2 Art history and theory
Aesthetics has historically been linked to reasoning about art. The art history and theory tradition is exceptionally strong at close analyses of art works, their experienced effects, the conditions of their creation, and the historical, national, cultural, and social contexts of their production and use. Beyond the innumerable large, full-color textbook histories of movements and traditions, I also recommend Julian Bell’s What is Painting? (1999), Gayford & Wright’s The Grove Book of Art Writing (1998), and any of the volumes in MIT Press’ Documents of Contemporary Art series edited by Blazwick to get a feel for how art historians and artists think about art and aesthetics.
22.214.171.124.3 Film aesthetics
Studies of film are of interest to interaction designers for several reasons. Film was the “new media” of the twentieth century, and one can see in the development of film and the social and intellectual reactions to it over time parallels to digital and interactive media and reactions today. As a dynamic, visual, and screen-based medium, film is also arguably closer to digital interaction than other cultural forms (such as novels or paintings). Finally, because film was so spectacularly implicated in the horrors of the twentieth century (Nazism in particular), film theorists and critics have intermingled aesthetic and ethical considerations in insightful ways that have similarly deep implications for HCI. Good starting points for interaction designers include Murch’s In the Blink of an Eye (2001, also cited by Hassenzahl in his Commentary to this chapter), Braudy and Cohen’s comprehensive Film Theory and Criticism (2004), and Monaco’s visually exemplified introduction to film theory, How to Read a Film (2000).
126.96.36.199.4 Literary theory
Literary theory has almost become synonymous with postmodernism and so-called “Grand Theory” (e.g., Deconstructionism, Psychoanalysis, Feminism, Marxism) and their excesses. One of the strengths (and problems!) with this tradition was its development of what has been called a “hermeneutics of suspicion,” that is, the development of an interpretive habit that rejects traditional notions of authorial intention and aesthetic pleasure, and replaces them with analyses that claim to expose the secret machinations of the selfish subconscious, the false consciousness-creating ideologies of capitalism, and/or the repressive effects of patriarchy. Less acknowledged is that a backlash against Grand Theory within literary studies began in the late 1980s, and careful/close readings of literary texts has been making something of a comeback. I recommend as starting points Barry’s Beginning Theory (2002), Tyson’s Critical Theory Today (2006), and The Cambridge History of Literary Criticism, volumes VIII (Seldon, 2005) and IX (Knellwolf and Norris, 2007) in particular.
188.8.131.52.5 Visual cultural studies
Visual cultural studies offers a cultural studies take on “both “high” and “low” forms of visual culture, including painting, product design, fashion, comic books, and advertising. These approaches are often interdisciplinary in their mix of critical interpretation, historical analysis, and sociological analysis and tend to remain well grounded on the visual artifacts and sociohistorical data, avoiding some of the dizzying flight of postmodernist “speculation to the death” (Baudrillard’s phrase) characteristic of work in some of the other fields. The best introduction to this work I have seen is Barnard’s Approaches to Understanding Visual Culture (2001); also interesting are design histories, such as Marcus’ Masters of Modern Design: A Critical Assessment (2005).
184.108.40.206 Cultural aesthetics and the human quality of life agenda
As I have spent much of my adult life reading about aesthetics from the fields listed above, I have my own (perhaps idiosyncratic) sense of the achievements of aesthetic thinking. I point out from the outset that these frequently comes in the form of theory—the kind of thing that Tranctinsky wants to rule out. But these theories are not, at least in the hands of the stronger writers, muddled and speculative dogmas; rather they are new concepts or new systems of concepts that empower us to perceive the experiential and socio-cultural significances of cultural works in much more diverse, nuanced, and personally fulfilling ways. These theoretical innovations are legitimated in at least two ways I can think of. First, they have to empower us to see and feel the sociocultural significance and experiential meanings of a work more robustly than we can without them. Second, they have to withstand, at least partly, the often brutal scrutiny of analytic philosophy, as described above.
I will briefly sketch some of the issues and related concept systems that have been developed to help us think more deeply about them—and to cultivate our appreciation for the aesthetic.
Whereas aesthetic processing theory formulates art as a "stimulus” that causes a “response” in our eyes and affective apparatus that in turn causes behavioral dispositions, other aesthetic theories position aesthetic encounters as the primary means by which an intelligent person works through a “learning-like process” to understand in an authentic and personal way “what it is like to live in the distinctive way of someone else,” helping overcome the barriers to “coping with others” and also to overcome the “impoverishment of the sensitivity on which moral competence often depends” (Miller, 1998, p50). Here, Miller links aesthetic perceptiveness (which, incidentally, is a concept at the very core of Baumgarten’s definition of the word “aesthetics” when he invented the term in the 18th century) to an empathic and holistic comprehension of the distinctive style of another’s life to moral competence. I can’t in this space to justice to Miller’s full reasoning, but his account is incredibly insightful and yet also participates elegantly in my own intuitive understanding of aesthetic experience. When I read Murakami’s Wind-Up Bird Chronicle or watch Claire Denis’ 35 Shots of Rum, I don’t merely enjoy the carefully crafted prose or sumptuous cinematography: I also feel like I am learning, expanding my horizons, growing in some way as a person. In short, Miller’s account helps clarify what I already know: that aesthetic experience can be good for me.
But it is not always good for me. I spoke earlier of the “hermeneutics of suspicion,” the interpretative strategy that with a dollop of paranoia investigates how aesthetic responses can be false pleasures. By false pleasure, I refer to pleasures that are harmful to us in ways that we fail to perceive or understand while we are enjoying them. Visually seducing desperate and fearful citizens to seek the paternal embrace of Nazism is a false pleasure, offering an ideological myth that stimulates the very fear it promises to assuage in place of actually providing a socially just system of government that would accomplish such security. Feminist film theorist Laura Mulvey 1975 showed in a seminal analysis how, in the hands of nearly exclusively male directors, cinema’s camera imposes on viewers a heterosexual male gaze, that the camera visually inspects and finds scopophilic pleasures in female actors’ bodies in ways that it does not for male actors, having diverse consequences for male and female film viewers.
Both of the preceding paragraphs have linked aesthetic experience with personal growth and the emergence of an intelligent and moral identity or the perversion of them. The following quote is from Richard Shusterman, an aesthetic philosopher who has influenced a considerable amount of HCI research to the point that Shusterman has been invited (and accepted) to be a featured speaker at CHI 2012. In the quote, Shusterman introduces his aesthetic vision:
Similarly, in the words of artist Nicolas Bourriaud:
My purpose in citing these quotes is not to assert that this is better or more important that the research ambitions of aesthetic processing theory, but simply to stress that this sort of agenda (a) is legitimately aesthetic HCI (including visually aesthetic HCI), because its aesthetic credentials are unassailable and it has influenced diverse HCI researchers (e.g., Schiphorst, 2009, 2011; Ferreira & Höök, 2009; Bardzell, 2011 among others), and (b) also can contribute to the same goals that aesthetic processing theorists themselves advocate: aesthetic interactions.
Aesthetic philosophers have also developed an extensive vocabulary to investigate artistic expression, including the development and articulation of sophisticated and intensely personal emotional insights (e.g., Collingwood, 1938) and the nature of creativity. Expression is important for HCI and interaction design, because unlike novels or paintings, interactions are made out of user expressions as much as designer choices. The aesthetic processing theory’s focus on visceral perception/response rather than expression, reductiveness rather than holism, and lack of medium-specificity make it difficult (though not necessarily impossible) for this theoretical approach to offer a strong account of self-expression. Jinwoo Kim, in his Commentary to this chapter, suggests that it would be good to explore YouTube creativity and the “social formation process of visual aesthetics,” and he is right. And, indeed, there is already HCI research on this that, again, is not acknowledged anywhere else in this chapter: Bardzell, 2007; Blythe & Cairns, 2009; Blythe & Cairns, 2010; Luther & Bruckman, 2008. I submit that this research agenda is strengthened, not confused, by a more holistic use of theory.
220.127.116.11 Cultural aesthetics in HCI
As I have already suggested, Tractinsky’s reference list notwithstanding, aesthetic theories from the humanities have influenced HCI and interaction design in significant and worthwhile ways. I briefly introduce a number of them below in the hopes of offering Encyclopedia of Human-Computer Interaction readers a more comprehensive and balanced set of references than they would have gotten had my Commentary not existed. I cannot here comprehensively cover all of the relevant work, but I do at least want to introduce four major themes of non-aesthetic processing approaches to aesthetics in HCI, themes that in many cases include seminal work.
The first research theme includes medium-specific theories of interactive aesthetics. As noted earlier, aesthetic processing does not explicitly distinguish between visual interaction and any other visual medium; it also scopes out non-visual digital interactions. Jonas Löwgren in (Löwgren, 2006, 2009) constructs an analytic vocabulary specifically for interaction, focusing on the notion of a interaction quality, which is a holistically understood description that intentionally blurs the boundaries between artifact descriptions and phenomenologically felt reactions. Examples include pliability, rhythm, dramaturgical structure, and fluency. In Bolter & Gromala (2006), the authors explore the ways that interactive technologies foreground and transform our understandings of transparency and reflectivity. In their analysis, these aesthetic qualities are also linked to dominant epistemologies in computer science. Lim et al. (2007) develop a concept of an interaction gestalt, leveraging both aesthetic processing (including Tractinsky) and pragmatist aesthetics (including Dewey and Shusterman) to articulate a set of attributes of an interaction gestalt, which include connectivity, continuity, movement, orderliness, pace, and time-depth among others. Common to all three medium-specific theories of interactive aesthetics are holistic understandings, explorations of design qualities, and efforts to link together interaction attributes with experience and understanding as they are consciously present to us.
The second research theme involves design and research methodologies surrounding aesthetic interaction. As interaction designers have gone from improving the performance of text editors to designing everyday technologies that are increasingly expected to be technologically robust, usable, sustainable, aesthetic, and socially just, design becomes an infinitely more complex problem. Standing in for empirical data about everything is an expert ability to read culture and to situate designs in appropriate and appealing ways within it (Kuutti, 2009). Supporting this ability to critically “read” design is a rising interest in the professional practice of criticism, an interest first introduced in HCI by Bertselsen & Pold (2004), and developed in my own work on interaction criticism (Bardzell & Bardzell, 2008; Bardzell 2009; and Bardzell, 2011). Also drawing on the arts and critical theory, but leveraging it in a different direction, is critical design, as developed in Dunne (2006), Dunne & Raby (2001), and Gaver et al. (2004). Critical designers develop provocative designs to challenge users by staging dilemmas that “force a decision onto the user, revealing how limited choices are usually hard-wired into products for us” (Dunne & Raby, 2001, pp45-46). Yet another approach, inspired by early twentieth-century literary theory, is defamiliarization, which seeks to find ways to enable designers to see beyond their own everyday assumptions by defamiliarizing themselves with and from them (Bell, Blythe, & Sengers, 2005).
A third research theme involves specific aesthetic design domains that touch upon HCI. In an anthology, a number of prominent HCI researchers from different intellectual traditions contributed a notion of funology as a new normative goal for HCI besides usability (Blythe et al., 2003). Another rising domain of aesthetic HCI interest is research on craft and DIY, which gets at a number of aesthetic issues, including creativity, the pleasures of things well made and making things well, and the relations among our pastimes, our heritage, and ourselves (e.g., Goodman & Rosner, 2011; Buechley et al., 2009; Bardzell, Rosner, & Bardzell, 2012). Responding to the explosion of non-WIMP-based user interfaces has been a rising interest in embodied interaction, and much of this work has also had an aesthetic dimension. Bardzell & Bardzell (2011) studied the designers and design processes of digitally enabled sex toys to understand designing for the confluence of sensual pleasure, intimate experience, social activism, and consumer electronics. HCI researchers have connected the performing arts, dance in particular, with interaction design research practices (Schiphorst, 2011) and user experience design and aesthetic response research (Latulipe, Carroll, & Lottridge, 2011).
The final, and arguably most important, of all the themes I briefly sketch here is aesthetic experience. Aesthetic experience is a major topic in nearly all domains that identify themselves as “aesthetic,” from recent analytic philosophy to the ancient Greeks, and from cognitive science to postmodern literary theory. Because it has been so tightly linked with user experience (UX) design—which remains the most common job title of my program’s graduates—this one is of signal importance. The seminal work in this area is McCarthy & Wright’s Technology as Experience, a book that constructs a theory of technology-mediated experience by combining American pragmatist philosopher John Dewey’s theories of aesthetic experience in his seminal book Art as Experience with Russian literary theorist Mikhael Bakhtin’s concept of dialogism. Their resulting theory has had extraordinary impact in our field not because (as Hassenzahl, Diefenbach, & Göritz complain) it seeks to tell us what the content of an experience is, but because the book has raised the imaginative perceptiveness, the insight, and the interpretative sensibility of thousands of UX designers (including all of my students)—a true achievement within the aesthetic HCI agenda. Others have used similar theory. In her recent My Life as Night Elf Priest, cultural anthropologist and HCI luminary Bonnie Nardi (2010) simultaneously constructs a theory of aesthetic experience and interpretively analyzes World of Warcraft play as an aesthetic experience. Her work helps interaction designers understand more analytically why World of Warcraft has been the smash hit that it has been, even as she contributes to the theory of aesthetic interaction experience. Finally, I want to point to Boehner, Sengers, & Warner’s (2008) “Interfaces with the ineffable: Meeting aesthetic experience on its own terms,” which among other contributions systematically explores the relationships between cognitive science and critical aesthetic conceptual systems. Though they position themselves as partisans on the critical side of that divide, and construct the two theoretical orientations as divided, nonetheless they can also be read subversively to explore opportunities to bridge that divide—which is increasingly what I think we should be doing.
19.5.7 A constructive conclusion
Given the unmistakeably increasing role of interaction in our everyday lives, mediating virtually every aspect of life, from work to the bedroom, interaction design simply must be aesthetic, just as our buildings must not only keep out rain but also be beautiful places to inhabit and our clothing must not only keep us warm but also help us express who we are or want to be. HCI is not mainly about high-performing text editors and aircraft controller interfaces any more. Whoever is seeking ways to make interaction more aesthetic I consider a fellow traveler.
We can’t have tea for two without the two or the tea.
I’d like to acknowledge several scholars for helping me formulate the thoughts I have developed in this Commentary. I’d like to begin by thanking Noam Tractinsky, whose article prompted the strong feelings that all but compelled me to work out the ideas in this essay. While this article took on, at times, a critical tone, scholarly dialogue is the means by which knowledge advances are made, and my own thinking is clearer and (in my view) more beautiful than it would have been without his work on aesthetics in HCI. Methodologically, beyond acknowledging analytic philosophy in general, I’d like to reference Gilbert Cockton’s careful critique of Gould & Lewis (2008) as an intellectual and rhetorical model for my analysis here. I’d also like to acknowledge Boehner et al.’s (2005) critique of the information processing model of affect in this, as that was the paper that introduced me to HCI's own "hermeneutics of suspicion" regarding the hidden operations of information processing theory in our research. Finally, I’d like to thank Mads Soegaard for convincing me to turn an emailed rant into a constructive commentary—and for his patience while I put it together.
21.9 Commentary by Jeffrey Bardzell
I begin my commentary by expressing my excitement to see an important philosopher directly engaging the HCI community. Up till now, philosophy’s participation in HCI has been mediated by HCI researchers’ interpretations and often (frankly) dumbed-down introductions to philosophical concepts. So this is a special moment for us, but it also introduces a practical challenge, and that is that because Shusterman is outside of our community, he can’t fully participate in our “language games,” which, in the case of HCI refers to the ways we frame and legitimate research, the sorts of questions we ask, the stock of examples we assume everyone knows, the history of the field we all think we hold in common, etc.
For a relationship between somaesthetics and HCI to prosper, linkages will need to be developed. Doing so is a book or at least a lengthy journal article, and I have here only a commentary to work with. So I will pursue the more modest goal of sketching out a position for somaesthetics in HCI, with the hope that others (from HCI, from philosophy) will join those of us who are already exploring those linkages.
My thesis in this commentary is that somaesthetics occupies a unique theoretical position in our field, able to connect pragmatist approaches to HCI (design theory, experience design) and embodied approaches to HCI (affective computing; mobile, pervasive, and ubiquitous computing). This has implications both for users, and in particular, norms for serving users in the deepest and most important ways possible; and also for interaction designers, and in particular, the cultivation of the professional self as an expert subject.
My argument will first sketch out recent calls for a more designerly HCI, exploring what those calls really mean, and specifying the challenges this community faces in responding to them; and second it will explore somaesthetics (and related pragmatist traditions) as a collection of useful concepts that has practical, pedagogical, and normative implications for a somaesthetic HCI. I will conclude by reflecting on some limitations of somaesthetics and HCI, which could guide future work.
21.9.1 The rise of design as an input for HCI
As is well documented elsewhere, since the 1990s there have been increasing calls for a more designerly approach to HCI, calls which have picked up steam since 2000. As computing has moved from the workplace and into everyday life, e.g., with the rise of mobile phones, pervasive computing, and the wild spread of digitally enabled consumer electronics, from DVRs to programmable sex toys, the distinctions between product design, service design, communications design, interaction design, and the emerging area of experience design have never been blurrier. Today design is a thriving subcommunity of HCI, with its own official subcommunity within CHI, a highly successful biannual conference in DIS, and ongoing calls for even more design contributions to the field. For example, McCarthy & Wright (2004) observe that
This turning towards design is often couched in critiques of the limits of social scientific approaches that have dominated HCI for decades. Greenberg & Buxton (2008) write,
There is a lot going on in this quote, including a critique of an HCI dogma that privileges objective measures over other legitimate forms of design knowledge-making, a prioritization of the social sciences over design and engineering innovation, and above all the practical consequences that we are too narrow to innovate. Restating this argument more simply, Greenberg and Buxton imply that innovation depends on subjective expertise and is hard to get at using the objective methods valorized by the HCI community.
Kuutti (2009) agrees: tracing the evolution of HCI as an academic-industry collaboration to deal with the explosion in demand for usable software during the rise of the PC in the 1980s, Kuutti describes how usability became the primary practical achievement to respond to this problem, with the unfortunate side-effect that “this also meant a certain intellectual impoverishment: calls to discuss about HCI theoretical foundations lost the audience, when the somewhat a-theoretical (and originally sometimes even anti-theoretical) usability movement took over” (Kuutti, 2009, p. 7); indeed this anti-theoretic stance is alive and well today, as can be seen in Tractinsky’s account of visual aesthetics in HCI and the portion of my commentary devoted to confronting his anti-theory position.
So the lack of design thinking inhibits innovation, and it is characterized by impoverished theory and a dogmatic fetish for objectivity. What then is a designerly approach to HCI? What sorts of theories, methodological strategies, and goals add up to a designerly HCI? A few choice quotes can at least offer some insights:
Design pedagogy is about “educating such personalities that can filter and crystallize cultural influences into effective and meaning-laden forms.” This characterization is a far cry from the traditional rhetoric of user-centered design. Whereas user-centered design positions the designer in an almost passive position of discovering existing needs using scientific methods and then designing around and for what is discovered in that activity, traditional design activates the designer as a perceptive, insightful, and imaginative meaning-maker, an ability that is individualistic to a certain degree and dependent on judgment rather than data, and offers a radically different view of the foundations of a design problematic.
Cockton offers a more specific account of the sorts of things that designers do as a part of their profession:
This is a different sort of list than one might expect from a traditional textbook on HCI: reflection, theory, a foregrounding of purpose, transdisciplinarity, thoughtfulness, exploration of alternative design means all characterize a designerly approach. Again, the role of a subjective expertise is unmistakeable in this list. It is quite easy to imagine how one designer could be superior at reflection, use of theory, creativity with alternative means, and so on than another. In contrast, user-centered design methods seem to presume that a designer’s strength is the quality of the data informing decisions, not anything internal to the designer herself.
My third quote comes from Greenberg & Buxton again, who offer a different sort of list and yet one that is fundamentally compatible with the vision of design that is emerging here:
Guiding all of this seems to be a holistic interpretation of what the design will be, accompanied by a rationale, and vetted by an intensely iterative and ongoing critical process involving stakeholders and other designers. This situates design in a dialogic and argumentative tradition. What is being argued for and against is the designer’s particular framing of the problem and speculative vision of its solution. Again, the subjective expertise of the designer—as an active meaning-maker and speculative reasoner—is the foundation of the whole activity. Others are brought in on similar terms: their own ability to interpret, frame, re-frame, and speculate determines the quality of critical feedback that they can provide.
In short, design professionals require a cultivated ability to read socio-cultural signs and trends; a creative and reasoned ability to explore alternative futures; a verbal ability to articulate these activities; a receptiveness to alternative framings and a willingness to explore highly variable alternative directions; and above all a personal identity or coherence that holds all of these moving parts together through a given process. Much more is personally demanded of designers than is personally demanded of traditional usability engineers.
So how to we get there?
There are several answers to this question. One is that a small number of individual designers cannot achieve critical mass: we need a design culture in HCI (Nelson & Stolterman, 2003). Creating such a culture has implications both for how interaction designers are trained and also for how the community legitimates certain knowledge practices. But however that happens, one thing is for certain: theory is going to be in the middle of it. Theory has historically been central to design and the humanities, inasmuch as each is concerned with the insightful and imaginative understanding of culture. Inasmuch as HCI now wants such accomplishments to be part of its discipline, HCI will have to get over its aversion to theory and fetish for methods. And this is where a philosophical program such as somaesthetics enters the picture.
21.9.2 Somaesthetics in/for HCI
Shusterman has offered here an accessible introduction to somaesthetics, effectively condensing and synthesizing earlier works, including “Somaesthetics: A Disciplinary Proposal,” which appeared as Chapter 10 in the second edition of Pragmatist Aesthetics: Living Beauty, Rethinking Art (2000) and his more recent book Body Consciousness (2008). I reference these books partly because they much more context for somaesthetics and Shusterman’s whole project. The first chapter of Pragmatist Aesthetics carefully situates somaesthetics within the context of Dewey’s pragmatism and in opposition to analytic aesthetics. This context translates to some core positions.
One core position is a formulation of somaesthetics as comprising a holistic and even organic perspective on life, work, and the self. Shusterman’s holism is evident in many places, not least in his efforts to undermine rigid distinctions between aesthetics and ethics, as when he writes, “aesthetic considerations are or should be crucial and ultimately perhaps paramount in determining how we choose to lead or shape our lives and how we assess what a good life is” (Shusterman, 2000, p. 237). HCI has long debated about the relative values of function versus aesthetics, with traditional HCI siding with function; but Shusterman follows Dewey in rejecting the distinction and seeing aesthetics as a holistic and inclusive term that encompasses the categories traditionally subsumed under function and form.
Dewey’s notion of an aesthetic life is not mere hedonism, but rather a sophisticated understanding of how the human as an organism purposefully and successfully copes with its environment (Berstein, 1971). Shusterman follows Dewey in rejecting divisions, distinctions, formalisms, and hierarchies. This is not merely an abstract chin-scratching philosophical position, but rather a practical position that has serious methodological implications: it basically rejects atomism or scientific reductionism of experience (Bernstein, 1971). Applied to HCI, such a view would reject the ways that affect researchers decompose affect into mood and emotion, emotion into positive and negative valence, positive valence into n number of positive emotions, and each of those into d degrees of intensity. Rather, a Deweyan view would construe emotion as a part of human’s purposive future-oriented disposition to the world, helping the organism orient itself in the best possible way.
Described holistically, somaesthetics is “a life-improving cognitive discipline that extends far beyond questions of beauty and fine arts and that involves both theory and practical exercise,” which seeks “to end the neglect of the body that [was] disastrously introduced into aesthetics,” with the ultimate goal to “contribute significantly to ... an art of living” (Shusterman, 2000, pp. 266-67). That’s a nice sounding agenda, but what does all that mean?
Somaesthetics, like any other philosophical position, can be characterized as comprising a system of relating concepts. I began with the context and overall view, because I don’t want an analysis of somaesthetic concepts applied to HCI to lose sight of what somaesthetics is supposed to do, which is to help us lead or shape our lives and to recognize what a good life is. I will consider this in two different directions, both of which are central to HCI:
- The training of interaction designers
- Normative criteria for user experience
18.104.22.168 Aesthetic perception, somatic training, and the interaction designer
In this section I want to argue that two of Shusterman’s key concepts, aesthetic perception and somatic training, contribute to both (a) a substantive epistemological account of the designer as expert subject and also (b) a useful set of norms to orient professionals cultivating designerly ways of doing (with implications for interaction design pedagogy as well).
22.214.171.124.1 Aesthetic perception
Among the most common views of aesthetic perception in HCI is the stimulus-response model. On this view, an object in the world, such as an interface, acts as a stimulus to the human cognitive system, which responds to it, e.g., by perceiving it, storing it in memory, understanding it, deciding to act based on it, etc. In HCI, for example, physiological data, such as breath rate and skin conductance, is collected as ways of measuring a person’s response to an input. Important advances to the field have been made with this model, and I certainly don’t want to suggest that it is somehow “wrong.”
But the stimulus-response model has epistemological limitations, and these have implications for interaction design professionals. The key limitation is that such a model assumes the existence of certain interpretative skill in the first place. It’s not always obvious to us, but understanding is a deeply cultural and learned ability. For instance, when we go to a museum or historical site, a docent not only relates historical backgrounds and contexts, but more importantly tells us what to look at. But if it is right before our eyes all along (i.e., a visual stimulus), then why don’t we respond the right way (i.e., with appreciation)? Why is it that a professional and amateur photographer standing side-by-side looking at the same thing will take very different pictures of it? How can a professional designer look at a given design material and come up with surprising and expressive new forms, where others simply rehash existing forms? Docents, accomplished photographers, and designers see and understand in richer ways than others do, and this is fundamental to their professional abilities.
Shusterman offers an explanation for these abilities by paraphrasing the work of 18th century philosopher Baumgarten, who first coined the term “aesthetics.” For Baumgarten, in Shusterman’s paraphrase,
In the stimulus-response model, and in most empirical science itself, such qualities have no meaningful place. Visual stimuli, and by extension empirical data, are seen to speak for themselves: the experiencing subject has only to perceive them to understand them. So stimulus-response is basically passive, and the model doesn’t differentiate among responders. Much UX research in this tradition assumes that research subjects are fundamentally interchangeable and simply seeks to average their physiological or reported emotional responses (e.g., using Likert scales).
But what Shusterman wants to emphasize is—and here he is leveraging phenomenological hermeneutics and reader-response theory as well as pragmatism—that understanding and hermeneutic skill must also exist before perception. Though we often speak commonsensically as if object, lightwaves, visual perception, mental image, understanding, judgment, and decisionmaking all happened in a causal linear sequence, in fact it cannot characterize what actually happens. Meaning-making is an active process; meaning is not a form stamped in our cognition like a seal ring to wax.
Some people can perceive more keenly than others; some have more penetrating insights, some have a greater imaginative capacity. Importantly, these are not static “faculties” that we are born with, but rather “sensory cognition” comprises cultivatable abilities or habits that we practice and can improve over a lifetime. As Dewey writes of intelligence, it is not “the faculty of intellect honored in textbooks” [but rather is] “the sum total impulses, habits, emotions, records, and discoveries which forecast what is desirable and undersirable in future possibilities, and which contrive ingeniously on behalf of imagined good” (Dewey, cited in Bernstein, 1971, p. 211).
Somaesthetics is substantially responsive to the calls for a design sensibility in HCI because it offers an epistemological account of what such a sensibility actually is: a sensitive, imaginative, penetrating, tasteful, poetic, and expressive habit or disposition to design problems, materials, processes, opportunities, and situations. But how does one achieve such habit or disposition?
126.96.36.199.2 Somatic training
One of the signature pieces of Shusterman’s somaesthetics is his call for somatic training: if it’s the case that the body is the “tool of tools,” then philosophers need to get out of their armchairs and cultivate their own somatic competencies. He defines this as “actually engaging in programs of disciplined, reflective, corporeal practice aimed at somatic self-improvement” (21.3.2).
To understand what he means by this, we might consider the practices that he uses to exemplify this: “diverse diets, body piercing and scarification, forms of dance and martial arts, yoga, massage, aerobics, bodybuilding, various erotic arts (including consensual sadomasochism), and such modern psychosomatic therapies as the Alexander Technique, the Feldenkrais Method, Bioenergetics, Rolfing, etc.” (Shusterman, 2000, p. 272). Common to each of these is a long-term commitment to body refinement. This is not a question of mastering the body by feeding and exercising it according to recommendations from the health sciences; rather, this is a practice of self-stylization for which the body is the locus of one’s individual stylized identity. The distinction I’m drawing here is akin to shaving one’s head versus becoming a “skinhead.” One is a mere physical description of a change to the body, whereas the other entails the same physical change but in such a way that it is inscribed in and defining a symbolic identity for the person doing it.
Thus, for Shusterman, the cultivation of somatic sensibility is an outcome of corporeal training. And surely he is right about this. I trained myself to skate well enough to join an amateur hockey league, and I also enjoy watching ice hockey. It is certainly the case that my appreciation for the watching the sport is partly based on a somatic understanding of the sheer skill of good players: people who have themselves tried to skate backwards while turning and accelerating and also while controlling a puck and keeping it from an opposing player can almost physically appreciate the somatic aesthetics of such movements as they are displayed gracefully and effortlessly in a professional game. Surely people who go to the ballet, many of whom have had some dance training themselves, also appreciate somatically, almost “feeling” in their own bodies the somatic near-perfection of a professional dancer’s alignments, body angles, and turnout. Indeed, how many somatic spectacles (e.g., professional sports, dance, rock concerts, etc.) do we enjoy that we haven’t tried out in one form or another, whether it’s backyard football, an attitude en pointe in front of the mirror, air guitars, or temporary tattoos?
Somatic training is also a part of HCI. We don’t just talk about designs: we sketch them, prototype them, try them out, put them in people’s hands and homes and watch what they do. These are all somatic exercises, and all of them require considerable training before anyone becomes good at them. Even among HCI researchers, the rising interest in critical design (e.g., Dunne, 2006; Dunne & Raby, 2001) and research through design (Zimmerman et al., 2007) extends this trend: such research uses design methodologies not in the hopes of creating new commercial products, but in order to generate knowledge and theory. Critical design is not easy to do, and seems to be an activity that requires iteration, practice, and training (Bardzell et al., 2012).
The other obvious area of HCI that involves somatic training is pedagogy. In our HCI/Design Master’s program at Indiana University, students work in groups and individually to generate sketches, prototypes, workbooks, and portfolios within a studio culture in which they are frequently presenting their work for critique by faculty and peers. I don’t think we’re unique in that: such pedagogy is a part of our field. Additionally, a majority of our students’ Master’s projects involve a domain with which—external to their participation in our program—they already have somatic training: we’ve seen projects building on prior experience in political activism, hardcore World of Warcraft play, fashion illustration, senior health care provision, and film production, among others.
Somatic training is substantially responsive to the call for a design sensibility because it relates design processes and practices to the underlying epistemology of an expert subject. Design processes are a form of somatic training: they entail disciplined embodied practices, and these practices eventually heighten perceptual and expressive sensitivities towards human needs, visual forms, problem reframings, socio-cultural meanings. It is by such mechanisms that designers train to become the kind of people who can, re-quoting Kuutti, “filter and crystallize cultural influences into effective and meaning-laden forms.”
188.8.131.52 Somaesthetic Experiences
Much of my commentary thus far has focused on the suitability of somaesthetics as a theory that offers a rich and useful account for training the specialized sensibilities expected of design professionals. However, neither Shusterman nor the other pragmatists were seeking to support specifically the design profession: the pragmatist project is fundamentally geared to improving all human quality of life by reminding us that humans are organisms purposefully engaged in their lived environments, and not information processors or “disembodied ratiocinators” (in the memorable phrasing of Bannon & Bødker, 1991). And that means that somaesthetics also provides normative criteria for the design of aesthetic experiences for users.
That is, if we want to reframe UX away from usability and towards something more robustly aesthetic, then we need to replace existing UX goals with new ones. Traditional ones include low task completion time, low error rates, high learnability, etc. Kutti (2001) proposes three alternative criteria that seem in the spirit of Deweyan pragmatism:
These sound like good goals, but their abstractness creates a practical challenge for designers. Here somaesthetic and pragmatist theory can begin to unpack some of these concepts. From a pragmatist perspective, aesthetic interaction should contribute to some combination of the following user experience-abilities. I combine “experience-abilities” into a single concept, because pragmatists stress how skills emerge in and from experiences; it is only by being challenged—not dumbed down with ease of use, transparency, and simplicity—that skills and aesthetic pleasure emerge.
The following are normative goods valorized by pragmatism in general and somaesthetics in particular. That is, an aesthetic interaction is one that adheres or contributes to some critical mass of the following:
- The experience and cultivation of Baumgarten’s “perfection of sensory cognition,” that is:
- keenness of sensation
- imaginative capacity
- penetrating insight
- good memory
- poetic disposition
- good taste
- expressive talent (from Shusterman, 2000, pp. 264-65)
- The expansion of people’s range of sympathetic identification with others (Guignon & Hiley, 2003, p. 36, paraphrasing American pragmatist Richard Rorty)
- A reformation of objects not as external to and in opposition to individual subjectivity, but rather seeing objects as contiguous with human consciousness, that is, seeing objects as human “activity in an objectified or congealed form” (Bernstein, 1971, p.46)
- An orientation towards the future (hope, intention, disposition), not the past (secure knowledge): “anticipation is ... more primary than recollection; projection than summoning of the past; the projective than the retrospective” (Dewey cited in Bernstein, 1971, 207).
- An appreciation of (and contributions toward) consciousness as dynamic and emergent, rather than static but wanting to collect more information. Consciousness is not a fixed form of mental seeing (i.e., disinterested understanding), but more along the lines of “the craftsman involved in doing and making.... The craftsman perfects his art not by comparing his product to some ‘ideal’ model, but by the cumulative results of experience—experience which benefits from tried and tested procedures, but always involves risk and novelty” (Bernstein, 1971, paraphrasing Dewey, p. 219)
- Ongoing somaesthetic training, self-improvement, and self-stylization. If experiences of art and beauty are distinctive for the powerfully gratifying ways they absorb our attention, unify our consciousness, and engage our emotions, then increasing our powers of awareness, focus, and feeling through better mastery of their somatic source could render more of our experience similarly rewarding in such ways. (Shusterman 21.1)
An aesthetic experience is one in which (a) the aesthetic goods listed above are experienced or felt, and also (b) the experience of them contributes to the long-term somaesthetic skills of insightful perception, imagination, meaningful self-stylization, and a disposition to do good.
All of this is not to suggest that every single interaction design must meet all of the above norms; rather, pragmatist holism would seem to suggest that interactions need only to contribute to and participate in lived ecologies where these qualities are experienced and subsequently cultivated through practice into skills. It is the lived world that ideally needs to have these qualities, not every single thing a person touches within the lived world. But inasmuch as our lived world is artificially designed—buildings, clothing, parks, appliances, furniture, and now interactive technologies—the burden is on us as designers to make that artificial world humane.
Somaesthetics offers normative criteria and a conceptual vocabulary to facilitate the design and evaluation of humane interactive products.
21.9.3 Criticisms and Limitations
It is probably obvious that I am sympathetic to somaesthetics and believe that it can contribute to HCI and interaction design. But one bad habit that we as a field have is that while we are eager to advocate for the introduction of a given theory, we often don’t acknowledge that this theory has confounds or limitations. I have tried to argue some of the specific ways that somaesthetics can contribute to HCI, and I think it’s also important to explore some of the ways it is not particularly well positioned to contribute to HCI. By exploring both strengths and limitations, as a field we can use theories more effectively and have some sense for when alternative theories are needed.
184.108.40.206 Somaesthetics is only loosely coupled with methods
HCI is a field that likes its methods, and it’s not clear how somaesthetics translates into methods. More fundamentally, it’s not even clear whether somaesthetics should translate into methods, at least not in the sense that the term is used in the sciences. At stake is an epistemological disagreement about how best to produce knowledge. A traditional experimental methodology, such as that described by Tractinsky in his interaction-design.org Encyclopedia entry on Visual Aesthetics, isolates variables in controlled experiments—classic methods from experimental psychology. For Kutti at least, such an approach is the antithesis of how designers operate:
Kuutti seems to establishing an exclusive opposition between expert judgment-based approaches and methods, a position also suggested by Greenberg & Buxton (cited earlier). Indeed, the very existence of methods seems to dumb scientists down into “barbarians.” Kuutti’s provocative language aside, it is easy to understand why scientists might view personal judgment as lacking any methodological rigor and thus barely any better than “mere opinion,” and why designers might view scientific methods as replacing human judgment with mechanistic algorithm-like recipes, which would seem to be a form of intellectual “barbarism.” Though it’s easy enough to understand these caricatures, it’s less clear whether they have any validity or in fact if they mainly just exaggerate differences. Dewey harmonized scientific and artistic approaches, but he did so by treating the sciences as an aesthetic practice, a position that might not appeal to practicing HCI scientists (though there is not as much daylight between Dewey’s position and that of post-positivist science, e.g., Quine, as one might expect).
Regardless of how apparent or real the opposition is between expert judgment and scientific method, it’s not clear to me that somaesthetics is going to resonate with interaction designers for whom data and methods are paramount.
220.127.116.11 Somaesthetics says little about the content of actual experiences
In a recent paper, Hassenzahl, Diefenbach, & Göritz (2010), who operate within a cognitive approach to UX, criticize McCarthy & Wright’s pragmatist take on UX as follows:
What Hassenzahl, Diefenbach, and Göritz are getting at is that there is an empirical dimension to experience, that is, users of a given design do have an experience e, it should be possible for research to discover the content of that e, and that McCarthy & Wright’s approach categorically fails to address that question, because it offers few strategies to capture “what these ‘values, needs, desires’ are”—which is an empirical, rather than critical, question.
I think there is a valid point here. I do see value in McCarthy & Wright that Hassenzahl, Diefenbach, and Göritz apparently do not, but I can understand why they read McCarthy & Wright in that way.
And now I will also add that I think a somaesthetic approach to HCI basically has the same shortcoming for basically the same reasons. Again, while somaesthetics is strong at offering an account of how an individual trains or cultivates the self as a perceiver and expresser (not just a thinker), it offers fewer tools to try to understand the content of particular experience x, and yet designers do have reason to want to know that.
18.104.22.168 Shusterman's stock of examples isn't particularly helpful
For an encyclopedia entry on interaction design, Shusterman not only used examples that take a lot of work for HCI readers to understand in the way that he wants them to, but he also missed opportunities to explore somaesthetic HCI with appropriate examples from the field. It is, of course, easy to explain this problem as a result of Shusterman’s outside status. Nonetheless, it is a huge missed opportunity, not just rhetorically (in terms of his ability to persuade HCI readers to engage with somaesthetics) but also substantively (some examples from HCI surely would help everyone think more deeply about somaesthetic HCI).
An obvious starting point is the field of robotics, in particular robotic work for domestic settings and everyday contexts, where the robots are designed to be appropriately meaningful as computational bodies in everyday life. Robot designer Tatsuya Matsui, for example, “believes that robots are like flowers. They can be delicate and beautiful. They are endearing and need nurturing” (Hornyak, 2007), a somaesthetic concept that is undeniably visible in his work:
Beyond robotics, several areas of HCI have explored embodied computational artifacts in relation to human embodiment. Another work is “Soft-Spikey Mouse,” created by artist Youngsuk Altieri working with Heekyoung Jung and myself:
So when Shusterman writes, “At this stage in somaesthetics research, we have only been concerned with somatic feelings of human bodies and thus with only one side of the HCI interaction. But, in principle, it may be possible to consider the somaesthetics of nonhuman somas, including computer bodies” it is clear that he simply has not yet engaged with the considerable amount of work in our field that has already been doing precisely that for decades.
As I imagine how such examples might influence Shusterman’s thinking, and how much I want to hear what he has to say about such work, it becomes clear to me that somaesthetics just might benefit as much from HCI as the other way around.
And I wouldn’t have it any other way.