Immodest Proposals:

Research Through Design and Knowledge

Jeffrey Bardzell
Informatics and Computing
Indiana University
Bloomington, IN 47408
jbardzel@indiana.edu

Shaowen Bardzell
Informatics and Computing
Indiana University
Bloomington, IN 47408
selu@indiana.edu

ABSTRACT

This paper offers theoretical support for research through

design (RtD) by arguing that to legitimize and make use of
research through design as research, HCI researchers need
to explore and clarify how RtD objects contribute to
knowledge. One way to pursue this goal is to leverage
knowledge-producing tactics of the arts and humanities
traditions of aesthetics, key among which is a communitywide and ongoing critical analysis of aesthetic objects.
Along these lines, we argue that while the intentions of the
objects designer are important and annotations are a good
mechanism to articulate them, the critical reception of objects can be equally generative of RtDs knowledge impacts. Such a scholarly critical reception is needed because
of the potential inexhaustibility of design objects meanings, their inability to be paraphrased adequately. Offering a
multilevel analysis of the (critical) design fiction Menstruation Machine by Sputniko!, the paper explores how design
objects co-produce knowledge, by working through complex design problem spaces in non-reductive ways, proposing new connections and distinctions, and embodying design ideas and processes across time and minds.
Author Keywords

Research through design; HCI; criticism; epistemology

ACM Classification Keywords

H.5.m. Information interfaces and presentation (e.g., HCI):

Miscellaneous.
INTRODUCTION

Research through design (RtD), the practice of using design

thinking, processes, and products as an inquiry methodology, has garnered considerable attention in both HCI and

Lone Koefoed Hansen

Aesthetics and Communication
& PIT Research Centre
Aarhus University, Denmark
koefoed@cavi.au.dk

design discourses [e.g., 3,4,6,9,11,12,15,20,22,28,29,32,37,

40,41,42]. As an emerging practice, there remains some
contradiction and confusion about what it is and what it
should be, which might [41] or might not be [22] a problem
in itself. Whether or not a unifying theory is the answer, our
position is that to support the theoretical development of
RtD as a research practice, to support the communitys uptake of its knowledge contributions, and to support more
HCI researchers and practitioners trying out this practice, at
a minimum more research needs to be done to explore and
to evaluate the epistemic potentials for RtD to contribute to
knowledge in HCI. Design is both a noun and a verb as [18]
writes; here, we focus on the noundesign as objects.
We begin by observing that while RtD as a practice is in
itself knowledge producing [1], so too can design objects
produce knowledge. However, what this means is unclear in
the research literature. Is the knowledge outcome of an RtD
object a special form of communication, one that is superior
in some sense to verbal discourse? Is it these objects job to
reveal true propositions about the world? To reveal the potentials of design materials? To reify design arguments? To
express emotional or subjectively felt experiences of the
artificial world and its apparent trajectory? To critique assumptions imbued in everyday designs? To reveal alternative ways of being to motivate us to pursue them?
We investigate RtD in its relation to the production of
knowledge; specifically, how design objects are knowledge
producers both for those that encounter them and those that
design them. In the arts and humanities, the ability of objects to produce knowledge is well understood, and the aim
of this paper is to characterize how we might understand the
kind of knowledge that RtD objects produce, how they produce it, and how we engage with it. We begin by characterizing the discourses on RtD in HCI vis--vis its knowledge
objectives. We observe that there is demand for good analytical skills to carve out core and recognizable important
new concepts [37, p14], but less on what the analytical
skills actually look like or do. To respond we turn to hermeneutical practice in aesthetics, which is based on the
view that a central value of the arts is that art objects and
phenomena co-produce meaning (as opposed to understanding aesthetics as primarily related to beauty, emotional expression, or aesthetic pleasure). In principle, this goes back

to the eighteenth-century German philosopher Baumgarten

and has been explored throughout the twentieth century in
both continental and analytic aesthetics (see [26,30,36] for
accounts of these histories). Following these traditions, we
argue that all objects can be critiqued from a critical and
aesthetic perspective, but that some objects are far more
complex than others, thus offering an (in principle) inexhaustible) source of knowledge production; this view offers
strong theoretical support for RtD in HCI. Also, RtDs influential formulation in Frayling [20] appears to be committed to this argument, which is currently (and sadly) underdeveloped in HCI. We also argue that the success of RtD
hinges not only on the development of a substantial collection of outstanding exemplars (as has already been proposed in HCI and RtD research [22,40] and in art-andtechnology research centers such as V2 and SymbioticA),
but also on the development of a hermeneutic community
capable of actively critiquing and interpreting the significance of these exemplars for HCI and design.
To explore how a detailed critique might work when we
understand objects as knowledge producers offering to the
viewer the possibility to engage in meaning-making practices unfolding a range of complex and multifaceted views,
we offer a multilevel analysis of a (critical) design fiction,
the Menstruation Machine by Sputniko! (aka Hiromi Ozaki)
[35]. Drawing on the analytic practices in the arts and aesthetics, we connect this analysis back to the issue of research through design and the contribution to knowledge
that certain design exemplars are able to make.
RESEARCH THROUGH DESIGN AND KNOWLEDGE

Although the terminology has since been redefined [32] and

added to [11], Fraylings highly influential working paper
Research in Art and Design [20] is typically cited as the
source of the term research through design, though others
from the RCA investigated the links and overlaps between
practitioners and researchers [1], and though the practice
had active precursors prior to the publication of Fraylings
paper [40]. In it, Frayling critiques a false dichotomy between research and art/designthat the former is cognitive
and the latter is expressiveconcluding that since many of
the motivations and practices of the two are alike, there is a
more productive distinction of the relations between research and art and design [20, p.5], including the following
(corresponding to a nearly identical triplet in [1]):
Research into art and design: historical research, aesthetic
and perceptual research, and research into theoretical perspectives on art and design (i.e., traditional art historical
and critical-humanist approaches to art and design).
Research through art and design: materials research (i.e.,
customising a piece of technology to do something no one
had considered before), and the explicit and detailed use of
an art/design research diary.
Research for art and design: research where the end product is an artefactwhere the thinking is [] embodied in

the artefact, where the goal is not primarily communicable

knowledge in the sense of verbal communication.
This is a dense passage in a short but dense article, and
some confusions have emerged. What HCI researchers call
research through design most closely maps onto what
Frayling in fact called research for design (cf. [32]). Additionally, Fraylings use of research into/through/for art
and design is commonly edited in HCI research to skip the
art and part, as if Frayling had no interest in art and was
only theorizing design research. However, the similarities
between art and design were foundational to his essay.
One consequence of eliding art and from Fraylings distinctions is to bury his commitment to art as a mode of
knowledge and understanding, in which the artist, like the
designer and the researcher, is determined to explore and
investigate (i.e. research) the world through interrogating,
designing, organizing, and makingresearch through
practitioner action is [1]s term for this. The designer as
well as the artist is, Frayling writes, a creator of meaning
and, like the researcher, this happens in an orderly way
and thus in a cognitive idiom (p3). The point of Fraylings paper is to argue that research for art and design
(i.e., what HCI calls HCI research through design) is not
merely an expressive mode where objects are personal expressions of the creators mind, devoid of cultural and societal significance. Instead, Frayling seeks to establish it as a
tradition which stands outside the artefact at the same time
as standing within it (p5), a view that echoes [Archer]s
shed[ding] light on [something] by attempting to construct
something (p11) as well as resembles contemporary art
theories arguing that art participates in culture/society by
commenting on it or looking at it from a particular position.
We will follow HCI in referring to research through design, even though in Frayling it is referred to as research
for art and design, but we will not follow HCI in suppressing the role of art and the way that artists operate quite
consciouslyin a cognitive idiom (p3) because this understanding of how artists workand thus of how art objects produce meaning as part of an aesthetic and cultural
exchange with the viewerprovides theoretical resources
that can contribute to research through design in HCI today.
Knowledge is unfolded in objects

As a field, HCI must answer what sorts of knowledge outcomes can come from objects in (art and) design projects; if
we cant, we cannot legitimize RtD as a way of doing HCI
research. Naming it design-oriented research, Fallman [15]
argues that it should include problem setting as an important part, the possibility of exploring possibilities outside
of the current paradigm (p231). Stolterman and Wiberg [37]
describe how a concept design carries and manifests all the
combined knowledge that has influenced the design
(p104). Such knowledge is available to designers as well as
to the user, we might add; in these objects, paradigms are
being challenged, complex ideas are being explored.

Similarly, [38] argues: art has a crucial democratic significance. Artists can experiment with the politics of things:
they draw attention to a matter, turn it into an experience,
rewrite it [] enable people to see through the [hidden]
political role of things and play with it (p26-7). Again,
some objects can shift our perspective on and understanding
of a situation, not by prescribing a particular way of looking, but by drawing attention to existing practices by designing something that questions them. This is a more open
ended version of a similar view by [40] for whom RtD is a
type of research practice focused on improving the world
by making new things that disrupt, complicate or transform
the current state of the world. This research approach
speculates on what the future could and should be by producing design exemplars. The knowledge contribution of
RtD is to offer concrete exemplars of preferred alterity,
which are generative of further design thinking.
Research through design, then, is a thing-making practice
whose objects can offer a critique of the present and reveal
alternative futures, while remaining grounded in empirical
science, behavioral theory, contemporary technological
possibility, and socio-cultural practices. Similarly, [7,12,13]
all argue that some objects lend themselves to an encounter
aimed at reflecting on contemporary technology discourses
as much as it is aimed at providing a sensual experience. In
other words, as [7] write, works of digital art are experiments in interaction design (p24) that invite us to reimagine and redefine our contexts (p28).
Gaver [22] offers another view of RtD: instead of being
extensible and verifiable, theory produced by research
through design tends to be provisional, contingent and aspirational. [..] an endless string of design examples is precisely at the core of how design research should operate, and
[..] the role of theory should be to annotate those examples
rather than replace them (p938). For Gaver, design exemplars are prioritized over theory. This suggests that rather
than viewing RtD projects as instrumentally serving better
HCI or design theory, the goal should be to build the
strongest possible collection of design exemplars and then
to annotate the exemplars to help bring forward their
knowledge contributions. It also suggests that the corpus of
research becomes not a collection of papers (as in the ACM
Digital Library) but rather a corpus of designs, with texts
existing to explain thema portfolio of annotated designs,
much like artists exhibit their work in galleries, attaching
manifestos and artists statements to the exhibition. [21] go
as far as to argue that digital artifacts themselves should
stand as peer-reviewable forms of research, worthy of professional credit and contestable as forms of argument
(p407). These calls echo structures found in other aesthetic
practices where critics, peers, and buyers assess the objects
in reviews, critiques, exhibitions, and concert bookings.
While we acknowledge important differences among the
authors cited above, we want to stress several commonalties
to the formulations offered. We note that none of these

formulations frame RtD as contributing to HCI theory in

the scientific sense of the word theory.. Instead, common
to all of the formulations is that RtD offers non-verbal design proposals in the form of artifacts, which support design
thinking: the interplay of object and tradition in a way that
opens up practice [20]; the ability of experimental objects
to aid designers wrestling with the boundaries of design
problem setting and framing [7,15]; the ability of experimental objects to provide material images of a speculative
future [13,40]; and the role of objects to encode knowledge
and to become themselves the primary research corpus,
moving the verbal into a supportive role [22,37].
Knowledge is unfolded in the interpretation of objects

Having emphasized the designers understanding of the

object, in the following we argue that the critical reception
of objects can also articulate and develop its knowledge
impacts. Although we all already perform critiques as part
of daily practice, there is little culture of sustained critical
analysis of objects in HCI. If there are too few exemplars of
RtD in HCI, there are even fewer exemplars of scholarship
on these exemplars [17,27]. Yet in the arts and humanities,
both types of exemplars are common, to the point of dominating those discourses. The role of critique in those discourses is to remove barriers to appreciation [10], and collectively they help artists and critics alike think more generatively and creatively about the arts, including genres,
materials, forms, meanings, and society more broadly, including social and political change. Recalling [1], we argue
that RtD in HCI would benefit from this.
How aesthetic appreciation has such epistemic power has
been theorized in the arts and humanities. In AngloAmerican aesthetics, [23] writes that literature (by extension, art) can be seen as in some significant respect informative of extra-textual reality [], declar[ing] to the
reader something about the way her world is (p82). In addition, art can also illuminate experience by making us
more sensitively aware of what it [experience] contains
[25, p71]. In continental aesthetics, Seel [36] develops a
theory of aesthetic appreciation (or interpretations):
first, aesthetic appreciation is an experience of appearing
in that the object cannot be determined completely. The
object is perceived in its constitutive indeterminacy, i.e.,
we deliberately do not search for one meaning but search
with greater attentiveness to what is indeterminable in
them (p54). The attraction lies in never being done with
the reading. Second, aesthetic appreciation results from a
complex interplay between the work, the spectator, and the
(cultural) context. While there is potentially no end to the
ways that an object can be interpreted, an interpretation is
also conditioned by knowledge of the type of object materially and formally, its histories of production and reception,
and commonly understood categories and meanings (cf.
Gadamers horizons of understanding, Isers modulating
operations, and in design literature Coynes notion of design criticism as constrained by a community of practice).

In other words, an important value of art is its contribution

to understanding, which comes through a meaningful encounter. Art is thus a meaning-making practice with meaning-making objects, and, like science, it helps us understand
the world. It differs from science, though, in the ways by
which it informs us about the world: oversimplifying, this
view holds that whereas art chiefly improves our thinking
(e.g., by enhancing our perceptiveness, imagination, intellectual sensitivity, ability to make fine discriminations,
etc.), science chiefly extends our knowledge through the
discovery of new facts and explanations [23,24,38]. Following this, we offer the pragmatic claim that taking seriously
the reading and interpretation of complex design objects
provides resources to support RtD. Designers are tasked
with understanding the relationships between present and
near-future technological possibility and future ways of
being, such that design solutions can be introduced.
What we are arguing here is that such a job is epistemically
akin to one of the roles played by the artist, as described by
aesthetic philosopher Eldridge [14]: The special pertinence
of [..] artistic achievement is that human life continually
presents materials about which we do not know exactly
how to feel and judge. [] We can wish to have our puzzlement worked through and clarified, and successful art
will do this in a way that does not resolve the puzzlement
into a dogmatic moral, but which clarifies its complex particularities (p226). It is not difficult to paraphrase Eldridge
from the standpoint of design. Technology and technologymediated behaviors are the materials about which we [as
designers] do not know how to feel and judge. And a successful RtD object might be understood as an object that
helps us work through and clarify a given puzzling problem
space, not to resolve it into a dogmatic theory, but rather to
clarify its complex particularities (e.g., the specific contours
of the design problem, to paraphrase Fallman).
The role of the work of art or RtD is not to present us with
new facts about the world, but rather to enrich our capabilities of perceiving and interpreting the world, that is, viewing the world through art, rather than checking art against
the world as [25] writes, before asking Does this make us
alive to new aspects of this sort of occasion? (p68-70).
When [40] write of RtD that The knowledge produced
functions as a proposal, not a prediction [40], they are getting at the same idea as [25]: the knowledge role of the object is not to offer an accurate representation of the world
(past, present, or future) but rather to make us alive to new
aspects of a human domain of interest. The key difference
is that whereas works of art serve these purposes for a public, RtD serves these purposes for designers. RtD makes
designers more alive to the design situation, who, like
artists, also must work with a view of human experience in
its widest sense: visual, aural, and tactile, as well as practical, emotional, and intellectual. And just as works of art
are better at this than verbal discourses, so good RtD exemplars are rightly prioritized over verbal discourse.

Knowledge is unfolded in interpretative communities

The notion that design exemplars are the primary body of

knowledge, and that verbal discourse does not stand over
that body but rather serves to remove barriers to the understanding of how these exemplars serve design thinking, is
present in all of our primary RtD sources. And it is specifically this role of the verbal to which we now turn. Consider
the following argument by American philosopher Goodman
[24] on theme and variation in architecture and other arts:
Establishment and modification of motifs, abstraction and
elaboration of patterns, differentiation and interrelation of
modes of transformation, all are processes of constructive
search; and the measures applicable are not those of passive
enjoyment but those of cognitive efficacy: delicacy of discrimination, power of integration, and justice of proportion
between recognition and discovery. Indeed, one typical way
of advancing knowledge is by progressive variation upon a
theme (p261). A passage like this provides strong theoretical support for Gavers call for portfolios. It is through collections of multiple exemplars that themes and variations
become visible, and progressive variation upon a theme is a
typical way of advancing knowledge, e.g., Goldsmiths
interrogation of themes of ludic interaction and ambiguity
over the years. Annotation of course can help render
themes, variations, and their interpreted significances visible to broader audiences, helping to articulate and disseminate knowledgebut knowledge that is primarily in the
designs themselves. This is why professor in Architecture
Flyvbjerg [18] calls for more case-study research, arguing
that the Kuhnian insight that a scientific discipline without
a large number of thoroughly executed case studies is a
discipline without systematic production of exemplars, and
a discipline without exemplars is an ineffective one.
But here we argue that annotated portfolios, important as
they are, are not enough. For not only does a given design
studio perform progressive variations on a theme to generate knowledge: the whole design community does. To
scope this inquiry to individual studios is to obfuscate the
collective achievements of the community at-large. Likewise, the notion of annotations also has an important limitation: in [8,22], the studio itself annotates its own portfolio.
But imagine if our knowledge of modernist painting were
limited to what Matisse said about Matisse, what Picasso
said about Picasso, and so forth? If the HCI and design research and practitioner communities are to leverage the
growing numbers of RtD objects into more generalizable
knowledge, these communities will need to collectively
practice the role that Gaver ascribes to annotation, which is
to remove the barriers of appreciation of individual and
collections of exemplars for the broader community
through writing about designs.
There have already been calls for more criticism in HCI
(e.g., [1,5,16] as well as examples of it in action [17,27]
though often without using the term [7,12,31]), but while it
has often used RtD as its objects of inquiry, the connection
has been tacit. Advocating for an expansion of design criti-

cism alongside the development of RtD exemplars and annotated portfolios maintains the hierarchy of design exemplars over verbal explications, a hierarchy that has been in
place in the humanities for centuries. But we stress that we
need more than first-person annotations of exemplars by
their own designers, although they remain very welcome.
THE MENSTRUATION MACHINE

To explore the benefits of critical analysis of RtDs, we offer

a critique of one such exemplar: Menstruation Machine
(MM hereafter) by Sputniko! [35]. In it, we summarize
available interpretations of this design and argue that they
have significant defects, offering our own analysis of how
MM significantly contributes to design research. In doing
so, we move the burden of identifying RtD knowledge contributions from solely the designer toward a more community-distributed modela hermeneutic community, as has
been tried and proven in the arts and humanities. MM (Figure 1) is a 2010 (critical) design fiction by Sputniko!, also
known as Hiromi Ozaki, a Japanese artist then at the RCA
and now at MIT. It is a metal chastity belt-like device,
which simulates menstruation for its wearer. The device is
accompanied by a 3-minute music video and the following
short artist statement (quoted in full):
Its 2010, so why are humans still menstruating? The pill free
interval was devised in the early days of the contraceptive pill
only because it was felt that women would find having a period more acceptablehowever, out of all mammals only humans, apes and bats need to bleed monthly for their reproductive cycle. What does Menstruation mean to humans? Who
might choose to have it, and how might they have it?
Fitted with a blood dispensing mechanism and lower-abdomenstimulating electrodes, the Menstruation Machine simulates the
pain and bleeding of an average 5 day menstruation process of
a human. Menstruation Machine (Takashi's Take) is a music
video about a boy Takashi, who builds the machine in an attempt to dress up as a female, biologically as well as aesthetically, to fulfill his desire to understand what it might feel like
to be a truely kawaii (cute) girl. He determinedly wears the
machine to hang out with his kawaii friend in Tokyo, but
(MIS-)READING MENSTRUATION MACHINE

We have shown this design in several design and health

informatics courses in the U.S. and Europe, and we ask
students what they think this design means. The response is
among the strongest we have ever seen to such prompts,
even when we take a step back and explain the particular
nature of the speculative and critical design tradition at
RCA [13]. What initially emerges is what we call a functionalist reading of MM, which states that the device helps
men empathize with women concerning menstruation. The
main view is that by facilitating this, men would be more
motivated to solve the problems of menstruation. One
graduate student appreciated the designs tactic, which he
said is to take something that he as a man has no access to
(menstruation) and convert it into a design challenge. Further, a debate about this design was held in 2014 at MoMA

Figure 1: Menstruation Machine

and documented online [39]. The debate centered around

the authenticity of MMs simulation, casting epistemological doubt on any empathy that arises from it. The debate
asked questions such as: Is empathy in design a good thing?
How do you know as a designer when you have real empathy? Does MM falsely promise to give men an experience
that they cant really have? These are important questions
with implications for design practice and education. Yet
they also carry forward the assumption thatdespite the
speculative backdropMMs primary purpose is to create
empathy among men by providing a simulated female experience. All three readings are predicated on a radical difference between men and women, and the function of the
MM is to simulate for one side the experience of the other.
The strengths of these functionalist readings are that they
focus on what seems to be the most salient feature of the
designits ability to simulate menstruationand that the
design makes us imagine new scenarios in which men
might want to use this device and why. As a functionalist
and even a critical design, this reading asserts, it prompts us
to imagine new scenarios of use and clearly supports design
thinking in the domain of gender, the body, and technology.
But this type of reading also raises a number of doubts. It
assumes that the device is for men. Neither the designer in
her statement nor the design itself says that the design is
for men: there is no reason it cannot be used by women,
trans individualsanyone. It seems revealing that everyoneeven feminist scholars at MoMAin one way or
another take for granted that the design is to be read in
terms of gender binaries and gender essentialism. Another
doubtful feature of this reading is its assumption that design
is always also problem-solving; that the device might be
ironic, satiric, or otherwise polysemous seems to be secondary or even excluded as a possibility. Finally, it is important to note that none of these readings take seriously the
devices very sophisticated aesthetics. Yet the design, the
artist statement, and the music video all collectively present
a highly sophisticated and stylized way of being (which
includes a futuristic, urbanized, Japanese, youthful, genderbending, electronic, high fashion, special night out)an
ironic oversight indeed for a debate staged at MoMA.
Collectively, the functionalist reading suggests some common assumptions concerning electronic devices: that they
are above all functional; that they solve problems; that they

reinforce but dont problematize social norms such as gender; that they do not attempt to offer cultural commentary.
This is a very limiting view of an objects power. If collections of RtD exemplars are to have impact on the research
community, we will collectively need to up our game.
RE-READING MENSTRUATION MACHINE

We now offer our own reading of Menstruation Machine,

informed by critical practices in art, philosophy, and design.

page. While another critic or reader will likely point to other details, everyone can agree that these details are there.
But by pulling these details into view, we have also taken
the first analytic step, because this selection of details has
implications for how we read the deviceif nothing else,
the details we have provided undermine the legitimacy of
the functionalist reading by demonstrating that that reading
contradicts the artists explicit intentions and fails to account for a very high number of concrete details that are
undeniably part of the device and its presentation to us.

A Descriptive Account of Menstruation Machine

We begin with a description of MMs aesthetics, that is,

how are we supposed to understand its role as a specific
style or way of being? With its polished steel and finished
form, its visual language connotes technology, high-end
product design, and futuristic fashion. It hides its machinery
but also makes no secret that it is a machine, like an Apple
product. In the music video, it also fits seamlessly with the
Japanese kawaii (cute girl) aesthetic (e.g., the protagonists
rainbow-colored hair, glamour make-up, clothes made with
metallic textiles, pop baseball cap, and designer purse). In
the statement that accompanies the design (reproduced
above), she asks, What does menstruation mean to humans? Thus, she avoids making the commitment to gender
essentialism by asserting that this is a design for men. She
does not ask about what its like to experience simulated
menstruation, but rather, what does menstruation mean?
She also foregrounds the presentations aesthetics, devoting
a third of her statement to the music video.
The video presents MM in an aestheticized way, using the
medium of the music video: MTV style editing, techno
soundtrack, song lyrics, a simple narrative, a rocknroll
protagonist. It introduces MM as part of the protagonists
impressive fashion; MM manages to look entirely of a piece
with the rest of the outfit. The protagonists putting on the
device is shown alongside makeup, doing her hair, and putting on her outfit: everything here suggests that this is a
particularly special night out, an experience to be cherished.
The protagonists narrative is cross-cut with footage of the
technology of the device, revealing its technical guts and
functions, almost like a stylized technology demonstration.
Finally, the lyrics of the song are important, because they
do not merely communicate the experience of cross-dresser
wearing the machinebut rather taunt the protagonist, and
by extension the viewer, to try it in the first place: You can
learn / If you really dare and I thought you really really
really really want to know how it feels, tellingly adding,
Right? The music video dares us to enter its fictive world,
and the protagonist should be read not as standing in for a
man who wants to know what its like to menstruate, but as
anyone in our world who wishes to enter a world where this
is a common consumer deviceincluding the pains (It
hurts, doesnt it?) and new perspectives (Tell me what
you can see / Can you ever feel the same?) it brings.
In professional criticism, the role of a descriptive account
such as this is to get the critic and her readers on the same

Menstruation Machines Propositional Content

MM expresses cultural as well as physiological aspects of

menstruation and makes them manifest in a physical form.
As previously stated, audience responses vary but frequently include laughter, puzzlement, disgust, and even dismissiveness, due to the devices clash with some of the cultural connotations to menstruation (for instance that it is a
curse). One way of understanding the effects of the device
is that it addresses cultural discourses on menstruation by
providing an interpretation of its effects, proposing that
menstruation is not only (or even primarily) a physiological
event, but has cultural, social and personal effects as well.
Following ethnographer Mols study of medical practices
[33], we might say that MM proposes that menstruation is
ontologically multistable; it is far from merely a simple
physiological event but must be understood from several,
equally important, perspectives. These include links between menstruation and age, in the cases of menarche and
menopause; menstruation as a taboo, even impure or disgusting; menstruation as controllable; views of menstruation as a super power that connects Earth and fertile women; and the menopausal woman as vulnerable according to
womens magazines. All of these accounts propose that
menstruation is a powerful and exclusively female phenomenon, something that MM is possibly circumventing.
So when MM brings menstruation with its blood and pain
to the forefront of the design, menstruation becomes a
choice but also a desirable, visible, festive, optional, ondemand, and fetishized event, that is no longer exclusively
available to women. In addition, MM makes menstruation a
spectacle for onlookers: it is a highly desirable object made
of polished chrome. All of this seems to go counter what we
expect of a technology that mediates menstruation, and it
seems to get at issues of gender, the body, and technology.
But what is the contribution of MM to design knowledge?
In the following we undertake an analysis to clarify ways
that MM contributes to design knowledge, pointing as well
to ways that such an analytical strategy might serve as a
way to strengthen RtDs research impacts. Specifically, we
explicate and analyze the propositional content of the device. If RtD objects function as design proposals, we can
ask, what exactly does Menstruation Machine propose? To
answer this, we identify a handful of propositions made by
the design, and we analyze them as knowledge contributions. By proposition we mean a claim that is offered as

true in the sense of correctly picking out some feature of

the world. Propositions in science, philosophy, and other
disciplines are commonly organized into hierarchies in support of one another, reaching for more general or higher
level propositions. So to analyze a design in this way is first
to generate a number of propositions that the design seems
to make, and second to assess what that group of propositions aggregates into. This is, of course, a critical practice,
and we make no claim that others would derive the same set
of propositions as if these and only these were objectively
present in the design, but we do believe that others would
accept that our outcome here is at least reasonable. We
number each proposition for easy reference in our analysis,
but this list is not intended to be read as ordered.
1. Menstruation is a design material. [34] define design
materials as what a designer puts together in compositional
relationships, creating real things. Materials are used to
bring a design into existence in the world, to make it appear
in a real sense (p227). Design materials are not limited to
physical materials but also include abstract material (a
process, or a symbol, or system) and to people as social,
cultural, and spiritual material. MM proposes that menstruation can be a design material, both in the physical
sense of milliliters of blood and electric force required to
induce cramps, and in the semantic sense of its various cultural significations. More subversively, it proposes that a
cultural taboo can be a design material.
2. Menstruation is desirable. This is the designs most
obvious and most critical proposition, the one that everyone
instantly grasps and feels compelled to respond to. But as
we have shown in our analysis so far, theres more than one
way to unpack the desirability of menstruation. We might
say: It is desirable for men to experience menstruation, so
that they can improve their empathy towards women. Or: It
is desirable for transgender women to have this device to
feel more feminine. It is visually desirable as an accessory
for kawaii (cute girl) fashion. It is desirable to have devices that can simulate sex-specific physiological experiences. It is intellectually and pedagogically desirable to
illustrate Judith Butlers difficult concept of performativity, giving this concept more intellectual traction. This list
could go on, not least because of its ability to scale from
very specific statements to very general onesall of which
invite us to imagine a world where a phenomenon colloquially known as the curse is re-construed as desirable.
3. Menstruation can be simulated with a device. People
can experience what it is like for another person to menstruate. The device proposes that anyone can experience a
menstrual cycle (measured physiologically). This is not
limited to men, but could include, for instance, premenstrual girls, postmenopausal women, or, for that matter, a menstruating woman who wants to double the symptoms. But
this proposition implies some unexpected propositions as
well, including the idea that menstruation can be turned on
or off at will, since consumer devices have this ability.

4. Menstruation is a physiological event. The design operationalizes its menstrual simulation in strictly physiological and even medicalized terms. Doing so is part of its
pointsee proposition 2 above. Still, it doesnt accommodate crucial physiological aspects like menstruations connection to pregnancy (e.g., the skipped periods implications for an unwanted pregnancy) or the feeling of bonding
when periods of women living together, synchronize. The
connotation that nature works in mysterious ways is left out
of the design, even if it focuses on physiology.
5. Technological innovation is regressively gendered.
The designer explicitly makes this point, I wanted to illustrate how technology doesn't necessarily evolve fairly and
contrasts the approval processes of drugs for birth control
versus erectile dysfunction. Part of the designs impact is
that it normalizes a sexually marginalized position, and the
act of doing so could be interpreted as shedding light on the
fact that the opposite is closer to the norm. In this way, it
encourages designers more generally to imagine design
problem spaces from the perspective of those who are sexually (or otherwise) marginalized, that is, to normalize their
world by inscribing their norms matter of factly in design.
6. A feminine performance is not real unless it has the
possibility of a full menstrual experience. This proposition comes through most explicitly in the video, which is
centrally about the performance of the feminine. Specifically, the video proposes that to be a kawaii involves more
than a visual performance: one also has to experience menstruation, and this experience must in some sense be transformative: I thought you really really really really want to
know how it feels and Can you ever feel the same? This
proposition rules out, for example, wearing a pair of pants
with simulated blood dyed into the crotch. This proposition
also exists in some tension with Proposition 4, that menstruation is physiological, because the video suggests that
menstruation for the protagonist is much more than that.
Analysis of Menstruation Machines Propositions

For reasons of space, we stop enumerating propositions at

six, though we could easily extend the list to include many
more, such as A menstrual simulator should achieve medically precise accuracy, which reflects the strong medicalization of the device and its design process as presented in
various supporting documents, suggesting that this is not
merely an art project even if it is based in the speculative
design paradigm. We turn instead to an analysis of these
propositions. Again, in philosophy and science, propositions are generally structured to support higher-level claims
or propositions. However, though MM may be a design
proposal, it does not paraphrase straightforwardly into a
coherent argument along the lines of science or philosophy.
Weve only stated six inferred propositions, and already we
see that they exist in tension with one another. Further
analysis will reveal more tension.

Let us situate MMs propositions in relation to the debate in

gender studies about gender essentialism. Gender essentialism is the view that men and women have irreducibly different experiences of their bodies that cannot be reconciled
under a single universal human, with diverse implications
for the social sciences, policy, institutions, etc. The view is
controversial in gender studies, but it is beyond the scope of
this paper to rehearse the various positions. We raise it here
because one of the central propositions of MM is that menstruation can be experienced by anyone who wears MM, a
position that on the surface appears to reject gender essentialismor to design a way out of it. Moreover, as we saw
in the debate on this design at MoMA, a concern was raised
that the device makes a promiseto help men feel empathy
for womenthat it cant deliver, an essentialist objection.
Let us analyze the propositions to see whether the device is
essentialist. Proposition 1 (menstruation as design material)
is gender non-essentialist, because the materialbleeding
and muscle contractionsare experienced by all humans.
Proposition 2 (m. as desirable) makes no commitment one
way or another as phrased, but its various unpackings do.
For example, the unpacking that says menstruation is desirable because it helps men empathize with a female experience is essentialist, while the unpacking that says its desirable because it exemplifies Butlers concept of performativity is not. Proposition 3 (m. as simulatable) also does
not commit to an essentialist position, because it makes
menstruation portable and therefore separable from womens bodies. Proposition 4 (m. as physiological) medicalizes
menstruation while denying its cultural connotations; if we
follow feminists in analytically distinguishing between (socially constructed) gender and (biological) sex, Proposition
4 gravitates towards biological sex and thus towards an
essentialist position. Proposition 5 (technological innovation is regressively gendered) frames technological innovation in a way that categorically distinguishes between men
and women; it is essentialist. Proposition 6 (feminine performance is only real if m.) is also essentialist, because it
suggests that to be subjectively real, a feminine experience has to include the experience of menstruation. In sum,
the propositions are all over the map on the essentialism
issue, and some of them are even internally ambivalent.
With our proposition-driven account of MM, we above all
wanted to experimentally try out the possibility of verbally
articulating ways that designs function as proposals open
to interpretative inquiry. Specifically, we wanted to bring
into view the propositions that constitute the design as a
proposal, as a way of gaining insight into how they might
contribute to design thinking. Identifying the propositions
also continued to undercut the functionalist reading by revealing a number of aspects of the design qua proposal unrelated to or even contradicting the functionalist one.
FROM DESIGNS TO KNOWLEDGE

The above reading of MM suggests that design can serve as

an open-ended cultural and social commentary as well as a

commentary on the role of design thinking in making technologies for particular purposes. In doing so, it reveals that
there is much more to be understood about a design than the
functionalist reading. This expansion of interpretive understanding is crucial, because a perspective limited to functionality is not very useful for RtD objects whose purpose,
among other things, is to work on having fewer blind spots
in design research, or, as [15] puts it exploring possibilities
outside of current paradigms. Let us return, then, to the
question of how RtD might contribute to HCI research.
Designs as Reified Arguments?

One possibility is that the knowledge offered by an RtD

object is one particular argument. Proposing a way of linking design objects to research and thus knowledge production, [21] argue that RtD objects reify arguments, which
in addition to contributing to knowledge can even anticipate
and preempt criticisms. While such an idea is provocative,
they do not drill down to the level of propositions and
chains of reasoning to show what that would look like, or
even establish convincingly that such a thing is possible.
Our experiment in analyzing MM through individual propositions gave us an opportunity to take a run at it.
But our proposition-driven account did not in the end provide a single, straightforward answer to what MM is or
means, and we remain dubious that a design like MM can,
by itself, reify an argument, at least not in the way that
we understand an argument. While we presented sufficient
evidence to reject the functionalist view, we were not able
to replace it with an alternative view at that level of simplicity. Rather, we demonstrated that the propositions do
not by themselves resolve neatly into a single coherent research argument, even though the design materials do resolve neatly into a single coherent design form.
The point that an RtDs formal coherence does not manifest
or signify an underlying coherence of argument is crucial to
a theoretical understanding of RtD as a research methodology. It explains why design exemplars, rather than annotations or critiques of them such as this one, are primary: like
art, RtDs are polysemous and even inexhaustible. It helps
men empathize with women captures the meaning of MM
as well as action is the chief end of existence captures the
meaning of Hamlet, which is to say: not well at all. We are
skeptical that we can expect this type of object to provide
an interesting, cogent, verbally paraphrasable research argument. Our analysis showed that MM, rather than making
an argument, instead opens a discussion of complex cultural
and social norms such as gender and taboos and on how
these are affected by the design of particular technologies.
Conceptual Polysemy and Design Understanding

If the knowledge provided by an RtD object cannot necessarily be viewed as a scientific explanation, a reified argument, or the sum of its functional meanings, then how does
it contribute to design research, knowledge, or understanding? Let us return to MM: once we started working with

possible meanings or propositions, it became clear that MM

is multistable and polysemous, offering perspectives from
many angles. Instead of offering a coherent position on the
issue of gender essentialism, we found it self-contradictory.
But this logical inconsistency does not imply a cognitive
failure of the work: As our analysis shows and as Seels
concept of an objects constitutive indeterminacy [36]
suggests, we look to the RtD in a way analogous to the way
that aesthetics says we look to art: to work through and
clarify an area of human life about which we do not know
exactly how to feel and judge [14]; to make us alive to
new aspects of this sort of occasion [25]; to develop
through theme and variation finer delicacy of discrimination, power of integration, and justice of proportion between recognition and discovery [24].
In saying this we agree with but also extend Stolterman and
Wiberg [37], who write of concept-driven designs,
The key here is that the concept carries and manifests all the
combined knowledge that has influenced the design. It is in this
way that knowledge is embodied in the artifact; it is not necessarily expressed in words or other descriptions, even though
such descriptions do complement and strengthen the experience
and understanding of the concept design. (p104)

The artifact embodies a considerable amount of combined

knowledge. However, it embodies more than the combined
knowledge that has influenced the design, which presumably could be elucidated via annotation; it also comes to embody the combined knowledge of those who engage it seriously and analytically once it is released into the world. In
this way, over time the object that merits such attention will
continue to accrete new meanings, entering into complex
relationships with prior understood meanings, the mastery
of which will eventually come to define RtD collections of
exemplars in relation to a hermeneutic community of practice. This hermeneutic community of practice, comprising
designers and researchers alike, will share an expert analytical vocabulary and sensibility to co-produce meanings with
RtD objects, whose symbolic functioning is too semantically, syntactically, and pragmatically dense to paraphrase, to
decode, or to interpretatively pin down or exhaust.
But Did She Intend It?

It is highly improbable that Ozaki intended viewers of her

design to use it to engage debates in aesthetics and philosophy of science about what constitutes knowledge. What
evidence we have of her intentions suggests that MM was a
speculative design, specifically intended to reimagine the
meanings of menstruation using design, rather than text, as
the medium of inquiry. And often RtD has more pragmatic
goalsto explore the expressive or functional capabilities
of a new material, say, or to help a designer creatively explore a problem domain. As MM was surely successful
within the scope of its intended speculations about menstruation, so can RtD projects be successful in their pragmatic
goals without immediately aspiring to scientific or philosophical contributions [22].

But as we hope has become clear, our objective here has not
been to decode Sputniko!s intentions or even to unpack
all the combined knowledge that has influenced the design. It has been, rather, to seek to identify and communicate ways that MM can contribute to HCI and design
knowledge beyond the immediate contexts and goals within
and for which it was undertaken. Specifically, we have argued that the design can be understood to interrogate relations among technology, design, gender, taboo, and the
body, without reducing these issues to a universal theory,
bullet list of design implications, or dogmaand that this
interrogation is a form of knowledge production.
As we, in a design critical role, begin to develop these understandings, introduce new connections, propose new links
between design materials and cultural meanings, and so on,
the device comes to mean more than Ozaki intended. This
is no more a criticism of Ozaki than it is a criticism of
Shakespeare to say that Hamlet has remained relevant
across centuries and world cultures by transforming in significance in ways he could never have begun to imagine, let
alone sanction. But this also means that design annotations,
while an important contribution to RtD and design
knowledge more generally, is the beginning, not the end, of
the articulation of a designs contributions to knowledge.
MM neither serves to make nor to illustrate a single or particular argument: it does much more work than that. By
entering into a polysemous dialogue with existing design
languages, cultural norms, user experiences, and research
literaturewhich are publicly available to design researchers and not locked in the private mind of the designerMM
can co-produce new design understandings, if we have the
ability to read it. This latter skill is why competence in design criticism is so important for HCI: without it, it will be
difficult to bridge between the compositional, formal, semantic, and experiential composition of the design and the
conceptual nature of design research. In Seels words, in
order for the object to appear, it needs to enter a qualified
and engaged dialogue with an interpretative community.
ACKNOWLEDGEMENTS

This research has been funded by Aarhus Universitys interdisciplinary research centre for Participatory Information
Technology, The Intel Science and Technology Center for
Social Computing, and the National Science Foundation
(#1002772). Thanks also to Jodi Forlizzi, Erik Stolterman,
and our reviewers for their constructive criticisms..
REFERENCES

1. Archer, B. (1995). The Nature of Research. Co-design,

interdisciplinary journal of design, Jan 1995, 6-13.
2. Bardzell, J. (2011). Interaction criticism: An introduction to the practice. Interacting with computers, 23(6),
604-621.
3. Bardzell, J., and Bardzell, S. (2013). What is critical
about critical design?. In Proc. of CHI13. ACM.

4. Bardzell, J., Bardzell, S., and Stolterman, E. (2014).

Reading critical designs: Supporting reasoned debates
about critical designs. In Proc. of CHI2014. ACM.

24. Goodman, N. (1976). Languages of Art. Hackett.

5. Bertelsen, O., and Pold, S. (2004). Criticism as an approach to interface aesthetics. In Proc. of NordiCHI04.
ACM.

26. Guyer, P. History of Modern Aesthetics. In J. Levinson,

ed., 2003, The Oxford Handbook of Aesthetics, Oxford
University Press, 25-62.

6. Blythe, M. (2014). Research through design fiction:

Narrative in real and imaginary abstracts. In Proc. of
CHI2014. ACM.

27. Hansen, L.K. (2011) The Interface at the Skin. In Andersen, C.U. and Pold, S. (eds.) Interface Criticism. Aesthetics Beyond Buttons. Aarhus University Press.

7. Bolter, J. D. and Gromala, D. (2003). Windows and Mirrors: Interaction Design, Digital Art, and the Myth of
Transparency. The MIT Press.

28. Hansen, L. K. and Kozel, S. (2007) Embodied imagination: a hybrid method of designing for intimacy. Digital
Creativity, 18 (4), 207-220.

8. Bowers, J. (2012). The logic of annotated portfolios:

Communicating the value of Research through Design. In Proc. of DIS2012. ACM.

29. Koskinen, I., Zimmerman, J., Binder, T., Redstrom, J.,

& Wensveen, S. (2011). Design research through practice: From the lab, field, and showroom. Elsevier.

9. Carroll, J. M. & Kellogg, W. A. (1989) Artifact as theory-nexus: hermeneutics meets theory-based design. In
Proc. of CHI '89, 7-14. ACM.

30. Kyndrup, M. (2008). Den stetiske relation. Gyldendal.

25. Graham, G. (2005). Philosophy of the Arts. 3rd ed.

Routledge.

10. Carroll, N. (2009). On Criticism. Routledge.

31. Lwgren, J. (2009). Toward an articulation of interaction esthetics. New Review of Hypermedia and Multimedia 15(2), 129-146

11. Dalsgaard, P. (2010). Research in and through design:

An interaction design research approach. In Proc. of
OzCHI '10. ACM.

32. Ludvigsen, M. (2006). Designing for social interaction:

physical, co-located social computing. PhD Dissertation,
Center for Interactive Spaces, Aarhus.

12. DiSalvo, C. (2012). Spectacles and Tropes: Speculative

Design and Contemporary Food Cultures. The Fiberculture Journal 2012 (20), 109-122.

33. Mol, A. (2002). The Body Multiple: Ontology in Medical Practice. Duke University Press: Durham.

13. Dunne, A. and Raby, F. (2013). Speculative Everything.

The MIT Press.

34. Nelson, H. and Stolterman, E. (2013). The Design Way.

2nd ed. The MIT Press.

14. Eldridge, R. (2003). An Introduction to the Philosophy

of Art. Cambridge: Cambridge University Press.

35. Ozaki, H. (2010). Menstruation Machine.

http://www.di10.rca.ac.uk/hiromiozaki/menstruationmachine.html (accessed 20 Sept 2014)

15. Fallman, D. (2003). Design-oriented human-computer

interaction. In Proc. of CHI03. ACM.

36. Seel, M. (2005). Aesthetics of Appearing. Stanford University Press.

16. Feinberg, M., Carter, D., & Bullard, J. Always somewhere, never there: Using critical design to understand
database interactions. Proc. CHI'14, ACM, 1941-1950.

37. Stolterman, E., and Wiberg, M. (2010). Concept-driven

interaction design research. HumanComputer Interaction, 25(2), 95-118.

17. Fernaeus, Y. and Vallgrda, A. (2014) Ajna: negotiating

forms in the making of a musical cabinet. In Proc. of
DIS'14. ACM.

38. Verbeek, P.-P. (2012). Politics at Issue. On Art and the

Democratization of Things. In Boomgaard, J. et al.
(eds.), Open 24: Politics of Things. nai010, Rotterdam.

18. Flusser, V. (1995) On the Word Design: An Etymological Essay. Design Issues, 11(3), 50-53.

39. Yang, H. (2014). Review of MoMA Design and Violence Debate. http://yanghanzi.com/2014/04/13/to-whatextent-can-design-transcend-human-experienceboundaries/ (accessed 20 Sept 2014)

19. Flyvbjerg, B. (2006). Five Misunderstandings about

case-study research. Qualitative Inquiry 12(2), 219-245.
20. Frayling, C. (1993/4). Research in art and design. Royal
College of Art Research Papers 1(1), 1-5.
21. Galey, A. & Reuker, S. (2010). How a prototype argues.
Literary and Linguistic Computing 25(4), 405-24.
22. Gaver, W. (2012). What should we expect from research
through design? In Proc. of CHI12. ACM.
23. Gibson, J. (2007). Fiction and the Weave of Life. Oxford: Oxford University Press.

40. Zimmerman, J. & Forlizzi, J. (2014) Research through

design in HCI. In Olson, J. & Kellogg, W., eds., Ways of
Knowing in HCI, Springer.
41. Zimmerman, J., Forlizzi, E., Evenson, S. (2007). Research through design as a method for interaction design
research in HCI. In Proc. of CHI'07. ACM.
42. Zimmerman, J., Stolterman, E., & Forlizzi, J. (2010). An
analysis and critique of Research through Design. In
Proc. of DIS10. ACM.