Elements of a shared theory of science for design

ART 5 (1) pp. 1.1–1.32 Intellect Limited 2018 Artifact: Journal of Design Practice Volume 5 Number 1 © 2018 The Author(s). Published by Intellect Limited. English language. doi: 10.1386/art.5.1.1.1_1 PER GALLE The Royal Danish Academy of Fine Arts, School of Design Elements of a shared theory of science for design1 ABSTRACT KEYWORDS Higher education for the design professions – designers, architects and engineers – has been combining traditional disciplines of research and design in new ways, into what is now known as design research. As a discipline in its own right, design research is still quite young, and the design professions need a theory of science that takes their specific character into account, thus contributing to their self-image, and locating them in the greater landscape of research. Using philosophically oriented conceptual analysis I develop a system of basic concepts as elements of such a theory of science for the design professions. The concepts of design, research and design research are analysed, clarified and coordinated, in relation to concepts such as knowledge, theory and practice. At the same time I propose a precise, yet intuitive, terminology that may be shared and foster cross-disciplinary understanding and communication among the design professions, and which each of them can expand for purposes of its own. design research design research theory of science philosophy of design interdisciplinarity 1. INTRODUCTION Design has long been practised in both artistic and technical professions, but has no firmly established tradition as a research discipline in its own right. Nowadays, however, design teaching is becoming increasingly research based. Therefore there is a need to understand how design and research can be united in a fruitful manner. In other words, the design professions need to 1. This article was originally published in Danish in FORMakademisk (Galle 2010). It was translated by the author in late 2017 and subsequently republished in Artifact: Journal of Design Practice with the kind permission of the Editor of FORMakademisk. To bring the article up to date, some supplementary notes, comments and updates of citations were made (and marked as such). Furthermore, some passages have been 1.1 Per Galle edited for improved clarity. Apart from that, the translation preserves the meaning and style of the original version as closely as possible. 1.2 develop a theory of science that takes their specific character into account. This article is intended as a step in that direction. Traditionally, the academic training of designers and architects aims at enabling them to perform creative design on an artistic basis, and to some extent craftsmanship as well. Given the demand for research-based education, and the emergence of the knowledge society in general, they are facing the challenge of making their own contributions to research, and integrate it in their professional tradition (Galle 2003; Davis 2008). Admittedly, educational programmes for students of engineering have been research based for quite some time, but with an emphasis on the ‘hard’ disciplines of natural and formal sciences, and their application in optimization of product performance (Clausen et al. 2009: 208–09). Physics, chemistry, mathematics and informatics in various combinations are all part of the engineering programmes; and ‘engineering science’ is a commonly used word. But there is a growing understanding that engineering cannot adequately be understood as ‘applied [natural and formal] science’ (Schön 1983; Jørgensen 2009); that sound judgement, experience and even a sense of aesthetics may be of crucial importance to engineering (Wengenroth 2004); and that technical artefacts have a ‘dual nature’ as both physical objects and objects in a social reality, shaped by the intention for them to serve a particular function (Kroes 2002; Jørgensen et al. 2009; Houkes and Vermaas 2010). Finally, certain recently established educational engineering programmes have an explicit focus on design, whether dealing with design of industrial products, or architecture. So the engineering professions, too, may have reason to contemplate the issue of how research and design may evolve in tandem. However, I will not offer any profession-specific reflections on the design/ research relationship in education, whether for design, architecture or engineering. Instead, I will consider education for these design professions collectively, and analyse the relationship between design and research in them as a whole. My aim is to achieve a clarification and coordination of a number of precise basic concepts that are or may become shared by the design professions – a conceptual frame of reference – and to propose correspondingly a common basic terminology that may serve both profession-internal reflections, and, I hope, facilitate mutual understanding and cooperation among the professions. I do not thereby attempt to enforce a linguistic standardization upon others; but on the other hand I wish to point out the importance of a shared and well thought-out professional language – and to exemplify what I mean by that. Should my specific proposals gain acceptance and thus facilitate profession-internal thinking and cross-professional traffic, it would be a considerable additional benefit. The central basic concepts to be considered below will be design, research and design research, in that order. Along the way we shall deal with related concepts such as knowledge, theory and practice. The results are proposed as elements of a shared theory of science for the design professions – a theory that each profession may elaborate subsequently and adapt to purposes of its own. So this is all about conceptual labour, and even though in principle it might be approached by empirical methods, a form of philosophical analysis is more suitable. Empirically, for instance, one might try to clarify concepts by looking for properties shared by specific instances of what a number of experts would tend to point out as design, research, design research, etc.; or Artifact: Journal of Design Practice Elements of a shared theory of science for design what is spoken of as such in the literature. But such empirical data processing would depend on contingent variations in the data, it would be extremely space consuming, and because of the amount of details involved it would be difficult to create the broad cross-disciplinary picture we are seeking. Another empirical approach would be to explore how a number of design experts define central concepts. Poggenpohl et al. (2004), who have also pointed out the need for a linguistic ‘infrastructure’ for design, have experimented with such a collective formation of definitions; but they found that it gave rise to a large number of problems. Salustri and Rogers (2008) propose a dictionary for design, developed by methods of lexicography, i.e. the theory of dictionaries. Such a dictionary might chart the professional language in actual use at a given time, but without the possibility of moulding the professional language through coordination of terms and concepts. The deliberate coordination of concepts and moulding of language is important for a young research area such as design. That is why in this article I will apply an analytical–philosophical method, suggesting motivated and coordinated definitions, while arguing for the properties and conditions by which the definitions determine and delimit the concepts at issue. Despite the abstract nature of such conceptual labour, the analysis is elementary in the sense that it is limited to the most basic concepts. In the following exposition I have attempted to achieve the clarity of a tutorial, but from time to time nuances and issues are suggested that call for further research. Thus I am addressing not only the general reader, who merely wishes to form a well-founded critical attitude towards the idea of design as a research area, but also researchers and other specialists who wish to explore the subject in greater depth. 2. DESIGN We shall concentrate on design as the act of designing, i.e. the design process. The word ‘design’ is sometimes used about products made on the basis of a design process – as in the sentence ‘the sales of our latest design exceed expectations’. Instead, we will call them design products, to make clear the difference between process and product. Design researchers are far from consensus about a particular definition of design (Dickson 2002, Chapter 7). On the contrary, there would seem to be a resigned realization that questions about a ‘correct’ definition are fruitless (Buchanan 2004). On the other hand, as design practitioners or design researchers, we cannot afford just to resign and shrug off the definition issue, for our professional self-image depends on it. We need a concept of design that not only captures something essential about the design professions, but also clarifies what design research (regardless of the various approaches to it) is or can be. 2.1 The so-called ‘expanded concept of design’ In Denmark design has traditionally been seen as the shaping and styling of material industrial products, because the design professions emerged in the late eighteenth century ‘as a consequence of a need for drawings and models that could serve as a basis for the industrial production’ (AhnfeldtMollerup 2005: 9, my translation). Among the professions concerned with shaping and styling, Ahnfeldt-Mollerup also counts architecture, town planning and handicraft; but she sees them as distinct from design. However, www.intellectbooks.com 1.3 Per Galle partly because of the influence from English, where ‘design’ is used in a broader range of contexts (including architecture), and partly because ‘[nowadays] a much larger proportion of everyday objects are produced industrially, the practice of design has expanded […]’, she notes. From this AhnfeldtMollerup concludes that ‘[t]he concept of design is undergoing considerable change’, and she therefore poses the question, ‘[h]ow do we define the professional nature of design under these circumstances?’ (2005: 8, my translations). Hjelm, too, observes that design is spreading to new types of products, both material and immaterial ones: ‘Today we design everything from drugs to welfare systems’. And much like Ahnfeldt-Mollerup, she raises the question of what nowadays constitutes ‘the particular competence of the practising professional’ (Hjelm 2005: 1, original emphasis). Under the heading of ‘The expanded concept of design’, Friedman discusses the same tendency to view design as having developed from ‘an appendix to the arts and crafts professions’ to a far-reaching ‘industrial discipline’ in its own right. According to Friedman, understanding design as such a discipline necessitates ‘the development of an overarching design theory’, which ‘ought to support applicable theories […]’ (2005: 7, my translations). I concur with Friedman’s desire for an overarching design theory that might create an understanding; as I said, my aim is precisely to outline the basics of such a theory. But retaining historically motivated distinctions between, say, design and architecture, does not contribute to this end. To create a coherent understanding of design research, we need a concept of design broad enough to bridge historical divisions. On the other hand, we should not define the concept of design so broadly as to include all sorts of irrelevant phenomena under it. For instance, let us consider Herbert Simon’s elegantly defined and wellknown design concept – which by the way Friedman (2005: 7) sees as the unifying central idea for ‘the design profession of today’: ‘Everyone designs who devises courses of action aimed at changing existing situations into preferred ones’ (Simon [1969] 1996: 111). This concept of design is ‘expanded’ indeed. The definition opens the floodgates for absurdities: If you contemplate kicking off your shoes under the conference table, because your feet are getting too hot, you would be designing in this sense! Even if the neighbour’s cat lies in wait to catch a mouse, it would be carrying out an act of design (Galle 2011: 92 [citation updated 2017]). At best, such a comprehensive and inhomogeneous class of phenomena may be of interest as a philosophical curiosity; but it is ill suited for the role as subject matter of the design professions, or of design research. 2.2 A timeless concept of design What Ahnfeldt-Mollerup, Hjelm and Friedman seem to be looking for, each in their own manner, is a way of strengthening and demarcating the basic professionalism of designers at a time where the production of more and more types of products are governed by a preceding design process: not only material products such as bridges, chairs and shoes, but also immaterial ones such as services, interfaces, etc. (Buchanan 1998, 2001, 2004). But even though design finds new areas of application, and therefore designers need to acquire new knowledge and competences, it does not follow that the very act of designing becomes essentially different. In a world in flux we do not advance our professional self-understanding by changing our most 1.4 Artifact: Journal of Design Practice Elements of a shared theory of science for design fundamental concepts, too. Rather we do it by sticking to a timeless essence of our profession – in this case a timeless concept of design – on which we can rest our professional identity, and relatively to which we can understand the changes that keep happening.2 – Which is not an argument against developing new concepts where they are needed to grasp new phenomena. Other professions seem able to evolve without thereby having to change the conception of their essence. In the educational system, for example, different and more numerous subjects are being taught than a hundred years ago, and pedagogy is an active area of research. Even so one does not, as far as I am aware, talk of an ‘expanded concept of teaching’. For the very essence of teaching, I suppose, is the same as always: helping other people develop their knowledge or skills through learning. And within the medical profession, new diagnoses, new knowledge about the causes of disease and new methods of treatment have changed the daily practice, without giving rise to any talk of an ‘expanded concept of medicine’. In Denmark no less than 38 authorized medical specialities exist; e.g. occupational medicine, geriatrics, clinical genetics and psychiatry (Indenrigs- og Sundhedsministeriet 2007). Even so, presumably being a medical doctor today is still, as always, essentially about preventing and curing disease, and alleviating its symptoms – or more generally, about fighting whatever afflictions threaten us in our capacity as biological beings. So, rather than changing and ‘expanding’ the concept of design itself, let us try to pin down an essence of design, i.e. attempt to define a timeless design concept to give us a firm and stable vantage point from which to observe and deal with the torrent of impermanence. Sargent beautifully describes the impermanence in a context of engineering. He explains how new design projects constantly create a need for un-anticipated combinations of knowledge; e.g. in the case of mechanical engineering design of bearings, which required knowledge about both materials and kinematics. After some time such syntheses of knowledge lend themselves to generalization and are expressed theoretically in what he calls ‘design idioms’ (Sargent 1994: 390–91). Thus technical knowledge and specialization never reach a state of completion. As he concludes, ‘[i]t could be that the only thing common to all design is the intention to produce something useful. That does not mean that design theory and methodology research ends, it means that it is unending’ (Sargent 1994: 400). Even though in this passage Sargent focuses on impermanence and the development of the profession, the first sentence hints at an essence of the kind we are seeking: ‘the intention to produce something useful’ as common to all design. The expression does not in itself constitute a definition, but it does capture something important that is by no means confined to engineering. Even jewellery and fashion design arguably originate in a wish to ‘to produce something useful’, when by that we mean something sumptuous, stunning, or seductive. However, merely having an intention is obviously not the same as designing, which involves some sort of action. Buchanan proposed the following proper definition, which is considerably closer to capturing an essence of design, regardless of professional specialities and product types: ‘Design is the human power of conceiving, planning, and making products that serve human beings in the accomplishment of their individual and collective purposes’ (2001: 9). One may object, however, that ‘power’ suggests potential rather than actual action, while design is something people do (a process they initiate and sustain), 2. Note added 2017: More recently, I developed a related argument in (Galle 2016: 336). www.intellectbooks.com 1.5 Per Galle 3. Bamford (1990: 234) developed a precise definition of design, which – unlike the one I propose here – does not delegate purpose-directedness to the concept of artefact. That makes his definition considerably longer and more complex, but it delimits largely the same concept as mine, and is based on related considerations. [Comment added 2017: Parsons develops a new definition of design by reformulating and supplementing some of Bamford’s ideas (2016: 11). For a critical discussion of Parsons’ definition, see (Galle 2016: 324–25, 336–39).] 4. The word ‘possible’ is philosophically controversial, as it might suggest that, besides actual objects (artefacts), there are also possible objects, which enjoy a particular kind of existence. [Comment added 2017: e.g., possible but non-actual objects might be conceived of as a kind of abstract entities that exist without location in time and space.] Competing theories about what exists, and how, are the business of ontology, a major branch of philosophy. Discussions of ontology in relation to design can be found in (Galle 2008, 2009). 5. Note added 2018: For a recent – and rather heated – debate on the relationship between science and design, see (Farrell and Hooker 2012, 2015; Galle and Kroes 2014, 2015). 1.6 and not merely something they are capable of. But apart from that, the concept should be simplified by eliminating the ‘making’ of the conceived products as a necessary condition for designing. There is no need for a designer always to be able to make the conceived product him- or herself; and quite often no such product is ever made. For example, that is usually the case when students of architecture try their hand on studio projects, but nevertheless such work should also count as design. In a more recent book on the theory of science of engineering, Clausen et al. formulated the following definition, which might very well apply to other design professions: ‘Design is to be understood as an organized preparation and completion of a deliberate but creative course of development, in order to create new products, processes and systems’ (2009: 216, my translation). Like Buchanan, these authors emphasize the preparation of an intended product (processes and systems may be understood as immaterial products), but without requiring the actual making of such a product. That presumably the product in question has a purpose is not explicitly mentioned. If we keep open the possibility that no actual product results from the effort, the deliberate but creative course of development must be construed as a development of knowledge, rather than development of the product itself. To make this clear, and to emphasize the fact that designer and maker need not be identical, I suggest the following definition as an attempt to capture a ‘timeless concept of design’:3 ฀ Design: to develop and express knowledge about a possible4 (new) artefact with the intent of enabling yourself or someone else to make it. So, design involves the production of knowledge as a crucial feature, and this knowledge must be expressed – say, by means of drawings and text – with the intent to make production of the artefact possible, which the designer had conceived, i.e. had developed knowledge about. However, it is not a decisive requirement that an artefact is in fact made, or can be made at all, on this basis; the process may fail, e.g. if the designer expresses him- or herself obscurely, or forgets to take practical circumstances into consideration. In that respect the definition resembles Simon’s broad definition, and the one by Clausen et al. (whereas in Buchanan’s definition, success seems taken for granted every time a designer makes use of the potential mentioned there). Just like design, research also involves the production of knowledge; but it is knowledge subjected to other standards and purposes, as we shall see in Section 3. Thus, even though the two activities are related, they are not identical, nor is one a special case of the other.5 Clausen et al.’s proposed definition above, as well as mine, capture Sargent’s ‘intention to produce something useful’, only using other words. And, importantly, the concept of design as per these definitions covers both the way in which nowadays we prepare for the making of countless products (material and immaterial products), and the way we did in the days of early industrialism, before anyone called himself or herself a ‘designer’. – For that matter, the concept covers what probably we have been doing since the Stone Age: down to and including our most primitive use of sticks and stones to outline the location of a hut. The design process is purpose oriented in two related ways: It prepares the making of an artefact, and it also thereby indirectly furthers the purpose Artifact: Journal of Design Practice Elements of a shared theory of science for design for which the artefact is to be made, and which guides the thinking of the designer. For, if we follow Hilpinen ([1999] 2011 [citation updated 2017]), an artefact is a product deliberately made for a particular purpose – ‘man-made answers to man-felt needs’ as Mollerup puts it (2007). (Note that Hilpinen also keeps open the possibility of fiasco, for according to his definition, an artefact may fail to fulfil its purpose.) As noted earlier, an artefact made according to prior design we shall call a design product. But not all artefacts are design products. If I take a branch in the forest and cut it to make myself a walking stick, or if I make a cheese sandwich for my lunch, then I have made an artefact, but directly as it were, i.e. without prior design. Due to its purpose-orientation design inescapably (essentially) involves an element of prediction (Galle 2008: Section 4.1); namely the prediction that an artefact made according to the knowledge expressed by the designer will fulfil its purpose. The purposes of artefacts are numerous like the ‘man-felt needs’ the artefacts must satisfy, according to Mollerup: for example, an artefact may fulfil its purpose by being usable in a certain way, by being able to evoke certain emotions,6 by signalling the social status of its owner, conveying an artistic expression and so forth.7 In my definition above the intended possible artefact is referred to as new. This is where creativity comes into design. It means that such an artefact, as far as the designer knows, has not previously been made. The degree of novelty ranges from the spectacular (Utzon’s Sydney Opera; Dyson’s bagless vacuum cleaner) to the marginal (another anonymous wellington boot; a document binder, padlock, paintbrush, etc.), just as design may very well unfold within a tradition (Jon Knudsen, personal communication, 2 September 2008). It is therefore disputable whether novelty is strictly necessary for a process to be regarded as design, which is the reason why ‘new’ is enclosed in parentheses. (As we saw, Buchanan does not mention novelty in his definition.) On the other hand, if novelty were entirely absent, and had not even been a concern of the designer, it would violate common parlance to speak of design; such a case would rather amount to plagiarism.8 2.3 Design in context However, no matter how we define our concept of design, a definition cannot capture the context in which design occurs, and in which it must be understood lest we miss out on important nuances. As we have chosen to define design as a process, it is natural to think of it as embedded in a context of other processes, with which it cooperates. Figure 1 is a grossly simplified model of such a system of processes (rounded boxes) that cooperate by exchanging products (rectangular boxes). The arrows indicate how the processes generate or use the products. (Evidently, the word ‘product’ is used here in a rather broad sense, as a collective name for various kinds of entities.) The design process depends on (‘uses’) a design task, i.e. a more or less explicit statement of the aim(s) of the design process.9 For an architectural project, for instance, the design task may take the shape of a design brief prescribing certain quantities of floor space for different purposes, stating accessibility conditions and so forth; in other situations the design task may be an order from a company wanting a new graphic identity; or the design task may emerge from an informal conversation about how the designer may be of assistance to a client. 6. Note added 2017: See (Fokkinga and Desmet 2013) for an interesting elaboration of this point. 7. Note added 2017: According to Parsons (2016), the Modernist movement, flourishing in design during the first half of the twentieth century, took function (i.e. intended use) to be the primary purpose of design products, and assumed that other qualities (including aesthetic and expressive ones) would follow more or less automatically once function was taken properly into account. He introduces this idea in his Chapter 2, and elaborates on it in subsequent chapters. For a guide to, and a critical discussion of, his analysis of Modernism, see (Galle 2016). 8. Note added 2017: Or else inadvertent replication. 9. Note added 2017: Borrowing a familiar term from research (and research-based higher education), the design task might be called a problem statement, to stress the similarity between design and research. However, not all designers may subscribe to the idea that their work is about ‘solving problems’; rather they may be motivated by a desire to express themselves and explore new possibilities (Sidse Ansberg Bordal, personal communication, 1 December 2017). www.intellectbooks.com 1.7 Per Galle Legend Task description A process Design task A product Design B Design representation B generates A A Making A Design product B uses A B Distribution & use Occasional relation Figure 1: Design in context, schematically illustrated as a system of processes that cooperate by exchanging products, as explained in the legend to the right. © Per Galle. Figure 1 1.8 The design task does not contain enough information to initiate the Figure 1 making of a design product. Only the design representation does that. It may take the form of shop drawings with written specifications, and it expresses the knowledge that the designer develops (Galle 1999; Cross 2006: 16; Krippendorff 2007). In disciplines such as architectural and technical design, the design representation and the design product are clearly separated, as are the processes of design and making. But in other design disciplines there is no such clear separation (see also Galle 1999: Section 2.2). For example, to a ceramicist manually kneading and forming the material, without making use of a working drawing or sketch, the wet lump of clay taking shape may function as a design representation; one that later, through firing and glazing, ends up as the design product, of which the wet lump expresses the designer’s knowledge. Design representation and design product are overlapping here; and so are the processes of design and making. Something similar we also find in graphic design, where for example a series of proposals for a book jacket or a home page imperceptibly end up constituting the finished design product (Anette Højlund, personal communication, 13 June 2008). Yet according to our definition, these cases are instances of design, since knowledge is being developed and expressed with the intent Artifact: Journal of Design Practice Elements of a shared theory of science for design to enable the making of the final product – even though the results of such knowledge expression are part of, or precursors to, the final product itself. In my definition of design two sub-processes are mentioned: to develop knowledge, and to express it. Just as we saw that design and making may overlap in some cases, these two sub-processes can occur alternatingly, especially during sketching (Herbert 1993). In other words, design is not necessarily a simple sequence of clearly separated processes (as the strong simplification of both the definition and the model might trick one into believing), but rather a complicated interplay of multiple processes (Jon Knudsen, personal communication, 2 September 2008). Several published studies suggest a ‘co-evolution’ of the design task (the ‘problem’) and the design representation (the ‘solution’) (see for instance Archer 1979: 17; Schön 1983: 100; Cross 2006: 80; Harfield 2007 [citation added 2017: Halstrøm and Galle 2014]). That’s why in Figure 1 it is suggested that – even though in principle design is carried out in reaction to a design task formulated by a client, employer or the designer him- or herself – the design task evolves under the influence of the design process, as shown by the dashed arrow from ‘Design’ to ‘Design task’. For example, along the way a designer may discover new attainable qualities in the design product, which were not anticipated, but are incorporated in the design task (opportunistically, and in retrospect [added 2017]), thus supporting the designer’s decisions (Schön 1983; Galle 1996). The model in Figure 1 does not specify who carries out the various processes. A given person may be involved in one or more of them. For example, the designer – apart from designing – may have described his or her own design task, alone or in concert with a manufacturer or the user of the design product; the user may have participated in the design process, and so forth. These social aspects of the system are not our focus here. 3. RESEARCH From design we now move on to research, thus preparing the ground for a discussion of design research. Research is closely intertwined with concepts such as knowledge and theory; and theory can hardly be fully understood without clarification of its relation to practice. Therefore, we shall consider these concepts too, but not in isolation from the context of design, so we shall keep relating them to the ‘timeless concept of design’ that we introduced above. 3.1 Academic method and research Research is a complex and multifarious activity, and it is hardly an exaggeration to say that there are many and divergent perceptions of how it should be carried out, and how it develops over time. The words ‘research’ and ‘science’ as names of an activity are used interchangeably, at least in academic discourse (Kragh 2003: 146), so we can re-phrase the question of what research is, as a question about what science as an activity is – without thereby making it any easier to answer. But at least the re-phrasing of the question sets us on the track of the philosophical discipline to which the question belongs, namely the philosophy of science. This is one part of the theory of science, along with its other parts: the history, and the sociology of science (Klausen 2005: 13–15). The literature of the field is dominated by clashes between competing ‘schools of thought’, such as positivism, logical positivism, critical rationalism, www.intellectbooks.com 1.9 Per Galle phenomenology, hermeneutics, social constructivism, postmodernism (Collin and Køppe 2003; Fuglsang and Olsen 2004) and more recently, feminism, as well (Stormhøj 2004; Anderson 2009). However, these schools of thought have not always succeeded each other like the kings of a country, one seizing power as the other dies. Even when a school becomes obsolete, some of its ideas may live on in others, and opposed views coexist at any time. This state of division in the philosophy of science may be seen as a result of disagreement about questions such as these: what role do our concepts play in our experiential understanding of the world? Should a researcher become involved in, or rather remain distanced from, his or her subject, and how do researchers in fact relate to their subjects? How are different research disciplines interrelated? Given that they all describe one and the same reality, will their combined contributions lead to an ever more coherent world-view? Can all phenomena ultimately be explained physically (or socially); and does physics (or sociology) therefore enjoy a privileged status? And, more recently: do men and women perceive reality in essentially different ways? If so, does the male approach to research repress the female approach? A clear and thorough exposition of the complex development of the philosophy of science, especially over the last 100 years, can be found in (Godfrey-Smith 2003). I will not attempt a recapitulation of it here, nor defend any one particular school of thought. However, by merely expounding and delimiting the concepts of design, research and design research in a coordinated manner, I hope to create the conditions for a fruitful debate on such questions in a context of design. To this end I have attempted – in a manner similar to the above analysis of the concept of design – to formulate a short general definition of research: one that intuitively is in keeping with common parlance, yet is precise enough to support a nuanced discussion, and does not – as far as I am aware – conflict with established theory of science. I have made the definition as neutral as possible, in the sense that it does not favour any particular philosophical school of thought, or any professional speciality. As we all know, the normal career path into research goes via a Ph.D. programme, following and supplementing a graduate programme. Likewise, the concept of research may be seen as ‘following and supplementing’ the concept of academic method. Both are about attaining and dealing with knowledge in a skilled manner: ฀ Academic method: to develop, express and disseminate non-trivial knowledge in a way that fosters justified confidence in that knowledge. ฀ Research: to develop, express and disseminate new non-trivial knowledge in a way that fosters justified confidence in that knowledge. Research so conceived is not different in nature from the kind of academic method taught at a bachelor’s or master’s degree programme (see e.g. Hammershøj 2008); rather there is merely a difference in the degree of originality expected. So – contrary to what my use of two separate definitions might suggest – it is not a matter of two sharply distinct steps. A student’s work may contain elements of proper research long before the master’s exam, and the work of established researchers is not always full of surprising new insights. But on the whole, the level of ambition gradually rises from the first written student assignment, over the bachelor’s, master’s and Ph.D. theses, to the most exquisite contributions at the Nobel Prize level. 1.10 Artifact: Journal of Design Practice Elements of a shared theory of science for design My attempt at defining a general research concept is quite consistent with that by Kragh: ‘Science is an intellectual and social process striving for, and actually resulting in, a form of knowledge that is […] (i) public, (ii) fallible, (iii) corrigible and (iv) testable’ (2003: 150–51, original emphasis, my translation). But his four characteristics are more specific than my requirement about fostering ‘justified confidence’. Perhaps that is why his definition, according to his own judgement, ‘does not quite cover the pure formal sciences such as mathematics and logic’. The knowledge being developed by research is expressed and disseminated through theory – a highly linguistic activity. (Both knowledge and theory are important concepts in relation to research, and we will take a closer look at them in Sections 3.2 and 3.3.) All of this happens in different ways, depending on profession, project, researcher and choice of method. In empirical research, knowledge is developed through accumulation of experience, for example through observation, interviews or experiments; in other kinds of research exclusively through argumentation from given assumptions, or thorough interpretation of aspects of reality, relatively to a given frame of understanding. Theory, too, may assume different forms, depending on the genre and style conventions of the professions, their systems of notation and their channels of publication. The import of the novelty requirement is, as Biggs and Büchler put it (2007: 66), that ‘[t]hat which is new in this context is that which is new for the community – not just for the researcher’, and as they furthermore explain, this necessitates a thorough and careful search for literature and other information. But the novelty requirement is not as clear-cut as one might think; for it raises a number of questions: Would it count as research if independently you develop, express and disseminate non-trivial knowledge, which others happen to have developed already, but which […] ฀ they have not yet expressed? Or which […] ฀ they have expressed in an internal and secret report – say, for a military or a commercial organization? Or which […] ฀ they have expressed, and to a limited extent disseminated in a ‘grey publication’ which almost nobody is aware of? Or which […] ฀ they have expressed and disseminated via a proper research journal among the many you do not know? Puzzling borderline cases such as these endow the concept of research with some vagueness. But vagueness in our most important concepts seems to be a fact of life. We can live with this vagueness, if only we are aware of it and clearly state what we mean when we use the concepts in borderline cases, where their applicability may be doubted. The requirement in the above two definitions about non-triviality of the knowledge developed was included in order to prevent the concept of research (and the concept of academic method) from becoming so broad that – like Simon’s design concept – it ends up comprising absurd banalities. For example, we cannot call it research that I measure the length of my worn-out shoelaces in order to acquire new ones of the same length – notwithstanding the fact that I thereby acquire new knowledge (at the time of writing, nobody actually knows how long my shoelaces are), and even though I am careful to ‘express and disseminate’ the result, say, by jotting it down on a shopping list. But precisely what knowledge should count as non-trivial will depend on www.intellectbooks.com 1.11 Per Galle the circumstances under which research is conducted; for example its purpose and academic context. The requirement that research should foster justified confidence in the resulting knowledge means that one should make a considerable effort to maximize the reliability and intelligibility of one’s results. To do so, the researcher works systematically, honestly and self-critically, takes relevant existing knowledge into account (as already expressed and disseminated through theory), states any relevant underlying assumptions, and argues explicitly for the conclusions reached, so that others can test them (Kragh 2003: 154–64; Jensen 2004: 137–38; Langergaard et al. 2006: 52–65). The justified-confidence requirement is important to set research apart from other kinds of knowledge development, seen in astrology, for example, or – less controversially – in the day-to-day accumulation of experience in practicing professionals. However, the research belonging to particular professions or their sub-disciplines may be subject to profession- or discipline-specific norms and traditions for what more precisely counts as confidence-fostering knowledge-development and production of theory. This is particularly pronounced in the humanities and social sciences, less so in the natural sciences (Kragh 2003: 170). Researchers spend many working hours on peer reviewing, i.e. on scrutinizing manuscripts of others and offering expert critique (often under mutual anonymity) – as well as revising their own manuscripts upon receiving such critique, before their research findings are allowed to go to press. Peer reviewing is an institutionalized way of ensuring that readers can have justified confidence in the results of research. But it is not infallible. In 1994, physicist Alan Sokal managed to publish a deliberately meaningless article, full of fashionable post-modern jargon, in a peer-reviewed literary-political journal. Subsequently he caused considerable outrage by publically declaring his article a hoax (Sokal and Bricmont [1997] 2003; Godfrey-Smith 2003: 146), thus exposing the journal as having betrayed its responsibility to ensure the readers’ justified confidence – which only goes to show the importance of justified confidence as a cornerstone of research. 3.2 Knowledge As noted above, both knowledge and theory are crucial to understanding research. Occasionally one hears the words ‘knowledge’ and ‘theory’ used interchangeably or with overlapping meanings. But in professional and academic language they are more useful if their meanings are separated, so that is what we will do here. The question of what knowledge is has been subject to philosophical speculation since antiquity. Disparate views have been proposed and defended, but no consensus has been reached. (A short but informative exposition can be found in Jensen 2003.) On that background it may seem foolhardy to launch yet another view on what knowledge is. Nevertheless, I venture to do so, because along with that I want to define theory in such a manner that the three concepts of research, knowledge and theory are attuned to each other, and to the design-related aims of this article. That the art (and design) professions need to develop a concept of knowledge that fits their own purposes is a point clearly stated by Reilly (2002: 3): ‘If the arts back away from this challenge they risk having to work with definitions of knowledge that do not relate well to their working methods’. In 1.12 Artifact: Journal of Design Practice Elements of a shared theory of science for design the following definition I have tried to balance this with the desideratum that inter-profession communication should be possible: ฀ Knowledge: a familiarity with something actual or merely10 possible, and its change or persistence over time. 10. ‘Merely’ was added 2017 for clarification, since everything that is actual (exists, perhaps in the concrete part of the world) is of course also possible. (See also note 4 above.) Whatever one is familiar with, in virtue of having knowledge, we shall call the object of (one’s) knowledge. Some everyday locutions for the same are: whatever one has (develops, acquires, etc.) knowledge about; or whatever one is knowledgeable about. (‘Knowledge about’ in this sense was used already in the definition of design in Section 2.2.) By ‘something actual’ I mean something that exists now or has existed; by ‘something merely possible’ I mean something that does not exist and has not existed – but might. The phrase ‘its change or persistence’ adds to such objects also the (actual or merely possible) processes in which they (might) participate. The definition implicitly presupposes a person (a subject) who has knowledge in virtue of a mental state, namely familiarity. We can think of knowledge both as this mental state in itself, and as the relation that the familiarity constitutes between the person and the object of the person’s knowledge. (On knowledge as a relation, see Zagzebski [1999: 92].) I deliberately did not specify what might play the part as the object of knowledge – apart from the two broad categories of the actual and the merely possible. This is to avoid unnecessarily tying up the definition with a particular ontology (i.e. a system of assumptions about what exists, and what existence is). The definition allows for both knowledge about something through experiential contact with it (such as my being familiar with my own living-room, my children, my current mood, the way to the nearest bus stop), and knowledge about something through familiarity with true propositions about it (such as when I know that aluminium is a good conductor, or that the sum of the angles in a triangle is 180 degrees). The two kinds of knowledge are called knowledge by acquaintance and propositional knowledge, respectively (Zagzebski 1999; Ishikawa and Steup 2017 [citation added 2017]). Another distinction that can be made within the proposed concept of knowledge is that between knowledge that has been expressed, and knowledge that has not – and which perhaps cannot be expressed at all: ‘tacit knowledge’. Polanyi is known for having pointed out the importance of tacit knowledge, in research and elsewhere, given ‘the fact that we can know more than we can tell’ (1967: 4, emphasis added). Furthermore, within the confines of our definition of knowledge it is possible to distinguish between knowledge that something is, was or can be in a particular way (i.e. propositional knowledge again); and knowledge how something can change or persist through our acting or passivity. This distinction corresponds to Ryle’s famous distinction between ‘knowing that’ and ‘knowing how’ ([1949] 2000, Chapter 2). Depending on the ontology one endorses or presupposes, it also makes sense to speak of knowledge about abstract entities. These include, for instance, mathematical entities, laws of nature – and, notably, concepts (e.g., dignity, beauty, tool, frock, user friendliness), what falls under them, and what might fall under them. Such familiarity with concepts is an important species of knowledge, which enables us to pronounce judgments about what something is or can become; ‘knowing what’ we might call it, in analogy to Ryle’s terms. But what a concept is, in and of itself, and whether it is ‘actual or merely www.intellectbooks.com 1.13 Per Galle 11. Note added 2017: A somewhat abstruse remark at this place in the original Danish version was left out. possible’ (e.g., if it is socially constructed, is of a mental nature, or exists mindindependently) are ontological questions that we will set aside. The same goes for epistemological questions about how we become familiar with various objects of knowledge. Notwithstanding these evasions, let it be quite clear that my concept of knowledge lends itself to criticism: ฀ One objection might be directed at the fact that the definition includes the merely possible as something we can have knowledge about (qua being familiar with it). It seems fair to say that there is no such mental state as being familiar, simpliciter. One has to be familiar with something (familiarity, as used in the definition, being a relation). So only what exists can be objects of knowledge. But since, according to the definition, the objects of knowledge include the merely possible, we have thereby said that the merely possible exists – which is controversial, if not downright incoherent. Thus we soon find ourselves bogged down in awkward ontological implications of our formulations (Quine [1948] 1961); or controversies over whether or not terms (nouns and other nominal phrases) can refer to something nonexistent; and if so, what that might mean (Vision 1986). Given that the subject matter of design would seem to be precisely the merely possible (designers concerning themselves with how the world can or should be, rather than how it is or has been), there is a need for a philosophy of design to clarify questions related to the merely possible. Some attempts are discussed in (Galle 2008). ฀ Another line of attack on the proposed concept of knowledge might be to argue that it is just as difficult to give necessary and sufficient conditions for familiarity with something to constitute knowledge, as it is to give such conditions for belief (in the traditional analysis of knowledge as ‘justified true belief’). 11 But disagreement and controversy is the rule rather than the exception when it comes to philosophical attempts at clarifying what knowledge is. Even if we limit the subject to propositional knowledge, whose traditional (but unsatisfactory) analysis as ‘justified true belief’ dates back to Plato, its proper analysis is under vigorous debate by contemporary philosophers (Pollock and Cruz 1999; Ishikawa and Steup 2017 [citation added 2017]; Steup 2017 [citation updated]). When despite these complications I venture to suggest a definition of knowledge it is, as noted earlier, because I wish to adapt the concept thus delimited to our purposes. That is why my definition contains the phrase ‘something […] merely possible’, suggesting that we can have knowledge about how, under certain circumstances, something can be or can happen. Without a concept of knowledge thus making room for the merely possible, we could hardly maintain the proposed definition of design, or claim that design products are being made based on knowledge. Knowledge of the merely possible seems to be the very foundation of the design professions: not necessarily complete or certain knowledge, but knowledge sufficient to justify decisions. For example, as architects we can and must know that if a building is organized in a certain way, it can be used 1.14 Artifact: Journal of Design Practice Elements of a shared theory of science for design for certain purposes; as mechanical engineers we can and must know that a mechanism with a certain geometry made from materials with certain properties, will be able to move in a certain manner and so forth. Besides, academic method and research outside the design professions also aim at production of knowledge about the merely possible (of the future); for instance national economy, climate research and management. Having defined knowledge as a species of familiarity, we keep open the possibility that one can know something with various degrees of certainty; for it is possible to be more or less familiar with something, and this leaves room for mistakes and hence doubt. This open-endedness is deliberate, for the kind of knowledge developed when designing is obviously not infallible; but acknowledged error is a valuable source of new knowledge (Petroski 2006). But what about research? Can we apply the same concept of knowledge in research as in design, or should we impose stricter requirements upon knowledge produced by research? For example, we might require that research produce hard knowledge, defined as knowledge complying with the following principle: if a person knows that something is the case, then it is the case (Rescher 2002: 480). (Classical propositional knowledge, conceived of as justified true belief, is ‘hard’ in this sense.) But this would be unsatisfactory, for two reasons. First, it would be confusing to have two concepts of knowledge at work; one for design and another for research – including design research. Second, there is nothing in the nature of research – such as we have tried to characterize it – that speaks for another and harder concept of knowledge for research. Quite the contrary. For even in research, where the quest for ‘justified confidence’ (Section 3.1) involves serious efforts to reduce mistakes and doubt, a margin of uncertainty is still necessary. Were we to insist on ‘hard knowledge’ without reservations, we would commit ourselves to a belief in infallibility and continuity of research, which historical experience renders untenable. At least since Kuhn’s influential book, The Structure of Scientific Revolutions ([1962] 1996), it has been virtually impossible to maintain such a degree of optimism.12 Or else, if we rejected the notion of infallibility but still insisted on ‘hard knowledge’, we would never be able to decide whether or not a given activity were research or not. For even if the activity might lead to certain apparently true claims as a basis for knowledge, we would not be able to eliminate the possibility that those claims would turn out eventually to be false. That would mean that the activity had failed to produce hard knowledge, and so it would have turned out not to have been research after all. So, there is no need for the idea of a special kind of knowledge developed by research. But there is a need for a reasonable ‘division of labour’ between the concepts of research and knowledge, as regards the reliability of research results. Suppose for example that we subscribed to the classical conception of knowledge as ‘justified true belief’. Then knowledge would be reliable by definition, regardless of its provenance, i.e. it would foster ‘justified confidence’ automatically. So the only difference between acquiring knowledge through research and acquiring knowledge by other means (say, by way of practical experience) would be that knowledge gained by research would have to be new and non-trivial (compare the definition of research in Section 3.1). Such a meagre concept of research would be rather useless.13 This is why, in the definitions of research and knowledge (in Sections 3.1 and 3.2), I have played down the reliability component of the concept of knowledge, but strengthened 12. Even though his idea about recurrent collapses and ‘revolutions’ of normal scientific thinking have later been modified, by Lakatos and Laudan, among others (GodfreySmith 2003, Chapter 7). See also GodfreySmith’s discussion (in his Section 12.3) of the need for a conception of science that accommodates mistakes and correction of theories. 13. A similar argument can be made on the basis of a more recent conception of knowledge such as ‘reliabilism’ (Steup 2017). www.intellectbooks.com 1.15 Per Galle 14. Note added 2017: Non-human beings, such as pigeons, dogs and horses, may ‘know’ the way home, where to find food, how to defend their offspring, etc. But we shall ignore this fact, which is irrelevant in the present context. 15. Other authors may use the words ‘knowledge’ and ‘theory’ differently. For instance, a point sometimes made in the theory of science is that observations are ‘theory-laden’, which is to say that our observations not only depend on the part of the world we observe, but also on the concepts we have about it (see, e.g., Godfrey-Smith 2003, Section 10.3; Koch 2004: 96–97; Langergaard et al. 2006: 84–85, 90–91.) In this connection, ‘theory’ means more or less the same as ‘concept’. It would be more consistent with our terminology to speak of observations as ‘concept-laden’, ‘concept-dependent’ or perhaps ‘conceptually coloured’. 1.16 it in the concept of research (requiring ‘justified confidence’). Thus the two concepts are mutually balanced and supplement each other. 3.3 Theory and practice But what is the relationship between knowledge and theory? In ordinary parlance, the word ‘theory’ carries several rather broad meanings, e.g. ‘system of assumptions or doctrines that explain given facts or phenomena’; and since Aristotle it has had a connotation of something hypothetical or abstract ‘without regard to action (practice)’ (Teori [‘theory’] n.d., my translation). For our purposes we delimit and specify a single meaning within this broad semantic field, drawing on the concept of knowledge we have already introduced: ฀ Theory: a detailed description of the object of knowledge. In other words, a theory is a detailed description of something with which one is familiar by virtue of one’s knowledge (cf. the meaning of ‘object of knowledge’ as stated in connection with the definition of knowledge). A theory, which is a description of the object of knowledge about a specific subject (lighting, for instance), we shall call a theory about this subject. A theory may be verbal, graphical, or a combination of these. The requirement that it be ‘detailed’ is included in order to avoid – in accordance with ordinary parlance – that an insubstantial description is called a ‘theory’. For example, the sentence ‘It is hot’ pronounced about the Sun is not a theory, even though it may expresses, correctly if imprecisely, an aspect of comprehensive knowledge about the Sun. Theory can be produced through research, or other forms of knowledge development. Even a design representation (Section 2.3) can be considered a kind of theory; namely theory about a particular (merely) possible artefact, or kind of artefact. This will be a somewhat unusual use of the concept of theory, but it is acceptable as long as it does not lead to conflation of design and research. According to our definition, knowledge is a mental state of an individual (human being).14 Therefore knowledge cannot exist without somebody having it. Theory, by contrast, is evidence of somebody’s knowledge, and such evidence, once produced, can exist independently of human beings. Under the right circumstances, other people than the author of a theory can understand it, and thereby acquire knowledge. Imagine, for example, that a researcher has acquired substantial new knowledge through his work, but dies before sharing this knowledge with others. But if he has expressed it in the form of theory that others are capable of understanding, they can acquire and elaborate their version of his knowledge after his death. So, metaphorically we might say that theory is seeds of knowledge in bloom – seeds that under good growth conditions may sprout and bloom anew.15 Ryle argued that ‘knowing how’, the ability of competent action, cannot be reduced to ‘knowing that’; and that, consequently, competent action does not require awareness of rules or propositions ([1949] 2000, Chapter 2). According to this view, the understanding of theory cannot in itself enable competent action. On the other hand, the understanding of theory might induce a person to practise certain actions – to master a particular practice (note the difference in spelling) – thus acquiring the ability of competent action. Artifact: Journal of Design Practice Elements of a shared theory of science for design But what are we to make of this practice, and how does it relate to theory? Is there an opposition between theory and practice? We will seek the answers on the basis of the following definition: ฀ Practice: to employ one’s knowledge in action (particularly everyday routine action, often professional). Such action obviously leaves experience in its wake, which gives rise to knowledge by acquaintance (Section 3.2). So knowledge breeds new knowledge through practice, gradually and through mutual interchange. Thus also in design, which is an important species of practice. In design we apply knowledge we already have (typically, nowadays, as a matter of professional routine), but we also develop new knowledge – notably in unexpected ways by paying heed to what Schön calls the ‘back-talk’ of the situation: ‘Through the unintended effects of action, the situation talks back’ (1983: 135). An example of the interchange between knowledge and practice is seen in the design case by ‘Clara’, on whom Schön and Wiggins comment, ‘she discovers to design, and designs to discover’ (1992: 154). Or, as Bunge more generally puts it, ‘knowledge considerably improves the chances of correct doing, and doing may lead to knowing more […], not because action is knowledge, but because, in inquisitive minds, action may trigger questioning’ (1966: 337, original emphasis). It is worth noting that practice is not merely a source of new knowledge, but also potentially a source of new theory; namely if the knowledge by acquaintance that is developed through practical experience is expressed as a ‘detailed description’, for other people to study. The knowledge they may gain by so doing is propositional to begin with.16 (Later, if they act upon it in their own practice, it may develop into knowledge by acquaintance as well.) So just like research, practice – including that of the design professions – is a source of knowledge and potentially of theory. Practice is merely a different source. Practice so conceived is in opposition neither to theory, nor to research. 16. Since theory is not necessarily verbal or purely verbal (but may involve graphics, as I mentioned), understanding it does not always, strictly speaking, lead to (purely) propositional knowledge. It would therefore be more accurate to speak of knowledge gained by understanding theory as theoretical knowledge – in contrast to experiential knowledge: the knowledge by acquaintance we can acquire through experience. (See Section 3.2 about Zagzebski’s terminology.) 4. DESIGN RESEARCH Having now dealt with the concepts of design and research, we are in a position to combine them into the concept of design research. 4.1 Designerly academic method and research To gain recognition and legitimacy in society at large, academic method and research in the design professions should live up to the same kinds of quality standards that apply to other academic professions (Galle 2003). So it is convenient to phrase the definitions of academic method and research in design as specialized variants of the general definitions of academic method and research from Section 3.1: ฀ Designerly academic method: to develop, express and disseminate non-trivial designerly knowledge in a way that fosters justified confidence in that knowledge. ฀ Designerly research (aka design research): to develop, express and disseminate new non-trivial designerly knowledge in a way that fosters justified confidence in that knowledge. www.intellectbooks.com 1.17 Per Galle 17. Using the terminology of the Frascati Manual (OECD 2002), we can characterize design research as applied research when it develops directly practice-relevant knowledge. When the results are of a more indirect relevance to practice (aiming at stimulating or enabling further research) we are dealing with basic research, at least of the kind that OECD calls [practice- or application-] oriented basic research. – Note, however, that since practice-relevance is difficult or sometimes even impossible to predict, it makes no sense to insist on maintaining a sharp distinction between basic and applied research, according to contemporary general theory of science (Langergaard et al. 2006: 195–98). As you can see, the text after the colon is unchanged in both definitions, except that ‘designerly’ is added to narrow down the scope of ‘non-trivial’: what we are after is not just non-trivial knowledge, but more specifically non-trivial knowledge that is also designerly. Such designerly non-trivial knowledge will have direct or indirect relevance to the practice of the design professions (such as designers, architects and engineers). Although ultimately design research as a whole should aim at developing knowledge of direct relevance to design practice, not every single research effort needs to do so. A design research effort may contribute knowledge of indirect relevance; namely, when the resulting knowledge can stimulate or enable further design research.17 The other requirements – concerning dissemination, novelty and justified confidence – are unchanged. Even though they are not specific to design research, design researchers have formulated similar requirements, when reflecting on the question of what good design research is. For example, the confidence requirement (as construed at the end of Section 3.1) is well aligned with the quality criteria originally proposed by Bruce Archer and later recommended by Nigel Cross (2006: 101–02). Cross also requires that the problems explored by design research be ‘worthy and capable of investigation’, which corresponds to our non-triviality clause (or practice-relevance). In the same sprit, Biggs and Büchler (2008: 91–92) propose the following hallmarks of good research: that the results are disseminated, so that others can be influenced by them (compare the similar condition in our definitions), that the results are original (compare our ‘new’) and that they are placed in a context of similar, already existing results. This last item implicitly matches our ‘justified confidence’, which (as noted in Section 3.1) implies that existing knowledge be taken into account. Despite these fairly precise and extensive requirements, design research is not tied by its definition to any particular tradition. It is free to move within and across the three major research traditions: the humanities, the social sciences, and technology and natural sciences. On these, see for instance (Kragh 2003: 166–67). Consequently, it is natural for design researchers to take an eclectic approach to the research methods and basic assumptions of other research disciplines and their traditions. Groat and Wang (2002) see this in a positive light, as a rich variety of possibilities. They offer a thorough survey of how the thinking of other disciplines can be utilized in, and adapted to, design research – with a main emphasis on architecture. One might worry, however, that eclecticism, if taken too far by a discipline such as design research, will lead to a lack of disciplinary identity. We will return to this theme in Section 4.3. 4.2 Designerly knowledge and theory Just as we delimited designerly specializations of academic method and research, so we can delimit a specialization of theory: ฀ Designerly theory (aka design theory): a detailed description of the object of designerly knowledge. Design theory in this sense is the product of designerly academic method or designerly research; or it can express knowledge gained through design practice. Design theory differs from the general concept of theory (Section 3.3) only by specialization: the knowledge whose object is being described is 1.18 Artifact: Journal of Design Practice Elements of a shared theory of science for design designerly knowledge. That is, it is of direct or indirect relevance to professional design work. But even with this restriction, the concept of design theory is very broad. It can be subdivided in many different ways, as evidenced by the classification systems employed by research libraries associated with architectural, engineering and design education. For our purposes it can suffice to distinguish two sub-concepts: ฀ Application theory for design: designerly theory on design processes, design products, and contextual conditions for design. ฀ Theory of science and meta-theory for design: theory on design research and design theory. The application theory for design expresses, as suggested by the name, the kind of knowledge most directly relevant to the designer’s professional practice. The enumeration of ‘design processes, design products and contextual conditions for design’ in the definition suggests a further trichotomy (among many possible subdivisions). I will adopt this trichotomy, because it is simple yet informative, and without bias towards any particular design profession. The theory of science and meta-theory for design can be thought of as the theory produced by design research, when it is reflecting on its own production of knowledge and theory. The knowledge expressed by the theory of science and meta-theory is relevant to design practice too, albeit more indirectly. The theme of relevance is elaborated in Section 4.4. The above definition suggests a conception of the theory of science and meta-theory for design as a specialization of a theory of science and metatheory in general – where the latter is understood as a theory of research and theory. This conception of general theory of science tallies with expositions of the concept in textbooks. For example, according to Langergaard et al, theory of science is ‘about what science is, and about how it is related to the concept of reality’ (2006: 35, my translation). Klausen sees ‘theory of science’ as ‘a collective name for various reflections on the nature of science and the individual branches of science’ (2005: 12–13, original emphasis, my translation): history of science, sociology of science and philosophy of science (2005: 13–15). Within this broad spectrum, we have merely touched upon the philosophy of science about a particular ‘branch of science’, namely design research. As for the name ‘theory of science and meta-theory for design’, the first link suggests that the theory is about design research. So, strictly speaking, ‘theory of research and meta theory for design’ would be more adequate as a name. But as a new term, ‘theory of research’ would have difficulties competing with ‘theory of science’, which means the same but is well entrenched in other disciplines. If in the design professions we are to speak a language that is understood outside our own ranks, we must respect common parlance. (But speaking about ‘science’ in connection with design is problematic in itself; see the Appendix for a discussion of this issue.) The second link of the name, ‘meta-theory for design’, merely expresses the fact that we are dealing with theory about theory; e.g., in the present article, where we are discussing what theory is, what it may be used for, how it relates to practice and so forth. www.intellectbooks.com 1.19 Per Galle Theory of science & meta theory for design On design research and design theory Process-oriented design theory On design processes, their participants, tools, and methods Product-oriented design theory On design products, their workings and interaction with people and environment Context-oriented design theory On historical, cultural, technical, and other conditions for design Figure 2: A graphical–metaphorical overview of the entire design theory. Hovering at the top level, we see the theory of science and meta-theory for design. Lying at the bottom level is the application theory for design, subdivided into three parts. The objects of knowledge expressed by the four kinds of theory are indicated to the right. © Per Galle. To sum up and create an overview, I have tried in Figure 2 to visualize metaphorically the concept of design theory and its subdivision as just described. Indirectly, this also yields an overview of what might be called the subject matter of design research, i.e. an overall survey of the objects of knowledge that design researchers explore. Other surveys can be found in the literature; e.g. in Horváth (2004) and Hubka and Eder ([1992] 1996, Section 5.2 and Chapter 7). The transparent disc hovering in mid-air in Figure 2 symbolizes the theory of science and meta-theory for design. It ‘covers’ (i.e. is about) the application theory for design (as well as being about itself). The application theory assumes a more ‘down-to-earth’ position in the lower part of the picture. Here, it is symbolized by the solid disc, which is divided into three named parts that correspond to the enumeration of objects of knowledge in the definition: ‘design processes, design products, and contextual conditions for design’. These are explained in the boxes to the right. The two central parts, process-oriented and product-oriented design theory, complement each other like Yin and Yang. Together, they form the ‘core’ of application theory that most directly affects design practice. (In a context of philosophy of technology, already Bunge [1966: 331] introduced a similar dichotomy: ‘Substantive technological theories’, ‘regarding the objects of action, for example, machines’; and ‘operative technological theories’, ‘concerned with action itself, for example, with the decisions that precede and steer the manufacture or use of machines’.) But just as important is the outer ring, skirting the perimeter of the core. It symbolizes the context-oriented design theory, which is about the conditions for design, i.e. the context in which design practice unfolds. The ring is conical, slanting down towards ‘the surrounding terrain’, to which it forms a gradual transition. This transition illustrates an important point (which we will consider more closely in Section 4.3), namely that it neither easy, nor desirable, 1.20 Artifact: Journal of Design Practice Elements of a shared theory of science for design to make a sharp demarcation between design theory and theory produced in adjacent disciplines. There may be gradual transitions internally in the system, too. One and the same text may contain several kinds of theory – much as one and the same foodstuff contains fat, carbohydrates, protein and dietary fibre mixed in particular proportions. 4.3 Demarcation between design theory and the theory of auxiliary disciplines Nowadays, numerous artefacts are made in accordance with the outcome of prior designing. Thus, the context of design may comprise virtually any aspect of nature or culture. Consequently, it is difficult to determine how far ‘outwards’ design researchers should extend design theory, and how much they should leave to researchers from other disciplines – which, for our purposes may be called the auxiliary disciplines. As Buchanan remarks, ‘[t]hose involved in design research are easily drawn into research in other fields’ (2001: 17). He even considers this issue of demarcation towards the theory of the auxiliary disciplines ‘the central dilemma of the new design research’, and he asks, ‘[w]hat is the nature of a discipline that brings together knowledge from so many other disciplines and integrates it for the creation of successful products […]?’. I have already hinted at one possible answer to this last question, by pointing out process- and product-oriented design theory as the ‘core’ of application theory for design, with context-oriented design theory forming a gradual transition to the subjects of the auxiliary disciplines. The ‘core’ we can point out as the ‘private’ area of design research, while contextoriented theory may contain design-relevant contributions from both design researchers and their colleagues in other disciplines. At worst, the worry that Buchanan seems to express – that the discipline of design research threatens to disintegrate – will apply only to this transition zone. The question is, however, if this is a genuine and serious problem? I would suggest, rather, that the presence of the transition zone simply means that application theory for design, like other important concepts, has a certain degree of vagueness. In other words, that for some theories one may be in doubt whether or not they fall under the concept. Let us take a look at an example. A student at The Danish Design School (now [2017] The School of Design, under the Royal Danish Academy of Fine Arts) was working on a design project, aiming at the design of a speculum. (An instrument used for gynaecological examinations.) A major objective of the project was to minimize the discomfort felt by the woman being examined; and one way to achieve this was to shape the instrument so as to avoid any reference to sexuality that might be embarrassing to her. Thus it became important for the student to consult anthropological theory about common notions of sexual taboos. (Karen Lisa Salamon, personal communication, 18 December 2006.) Does this mean that such anthropological theory should be subsumed under application theory for design? We might include the taboo theory to some degree, depending on how often it is made use of by designers. Or, if we insist on a yes-or-no answer, we might consider whether the taboo theory is so much in demand that it should be part of the permanent syllabus for design education. If so, we subsume it under (context-oriented) application theory; otherwise we do not. www.intellectbooks.com 1.21 Per Galle As these pragmatic decision procedures show, the example does not suggest any genuine and serious problem of disintegration in design theory. It just seems to confirm what I have noted earlier, that we can live with some vagueness in our most important concepts, as long as we are prepared to encounter borderline cases. The vagueness of application theory for design is even an advantage, if we are aware of it. For, as Fink puts it: Demarcations between the disciplines tend to blind their members in a peculiar way to the connection that always exists between the subject matter of their own discipline and that of the others. Reality does not present itself to us cut up in segments matching our department and faculty borders. (2003: 201, my translation) 4.4 The context and practice-relevance of design research Just as Figure 1 showed design in its context, and thus enabled us to nuance the concept of design, so Figure 3 shows design research in its context, to provide more nuances than conveyed by the definitions alone. The figure is an elaboration of the system from Figure 1, including three new boxes in a row extending to the right, and arrows showing relations of cooperation in the same way as in Figure 1. Design research is illustrated as a process that consumes information from all parts of the system, generating two kinds of ‘products’. One is the Legend Task description A process Design task A product Design (practice) Application theory for des. Design research Theory of science & meta theory for des. B Design repræsentation B generates A A Making A Design product B uses A B Distribution & use Occasional relation Figure 3: Design research in context, between its two ‘products’: application theory for design, and theory of science and meta-theory for design (horizontal row). The Figure 2 relation of research to design and the design context (vertical row) is suggested by arrows, as explained in the legend. © Per Galle. 1.22 Artifact: Journal of Design Practice Elements of a shared theory of science for design application theory for design that directly connects design research with the professional and educational practice of design. The other is the theory of science and meta-theory that results from the self-reflection of design research. This diagram, too, is a simplification. To avoid clutter, not all possible relations have been shown. For instance, design research might have been related to other processes, which are being studied by design researchers directly, and not only through their ‘products’. Nor does the figure show the many agents who carry out the processes, i.e. designers, researchers, manufacturers, users, etc. (see for instance Krippendorff 2006: Sections 2.4 and 5.2). The application theory for design does not necessarily take the form of handbooks featuring justified directions for action, even though this kind of theory exists (see, e.g., Alexander et al. 1977; Lidwell et al. 2003). The influence of the theory on practice can be subtler than that, however. Theory is a means of raising the designer’s general awareness of relevant phenomena – such as the influence of cultural differences on how people perceive design products (Boztepe 2007), perceptual and psychological factors that facilitate or hamper the use of a product (Norman 2004, 2013 [citation updated 2017]). Or theory can foster radically new ways of thinking (McDonough and Braungart 2002). As we saw in Section 3.3, design in its capacity of practice has itself a potential for producing theory; hence the dashed arrow from ‘Design (practice)’ to ‘Application theory for design’ in Figure 3. A clear example of this is the portion of Monö’s textbook on design semiotics (1997) that is based on his own experience as an industrial designer. Another example is Hove’s working paper (2010), in which she focuses on the aesthetic aspect of the designer’s work and the dissemination of ‘tacit knowledge’ about form. If sufficiently general, such practice-based theory can stimulate new practice, in a relationship of mutual support. As Schön and Wiggins say (1992: 155), ‘[t]he hard work of making explicit the discoveries gained through designing may help to make them more readily accessible and more subject to conscious control and choice’. In the same spirit, Sevaldson argues more systematically and thoroughly for the importance of design research based on design practice, i.e. research that can produce ‘new communicable knowledge that is only found within design practice’ (2010: 8). So design researchers have good reason to take an interest in experiential knowledge developed by practitioners of design: how it can be expressed and disseminated as application theory for design, and thereby be consolidated as an outcome of designerly academic method, by fostering ‘justified confidence’ (Section 4.1). Such consolidation might involve cooperation between designers and researchers – or formal research training of practitioners. Furthermore, it might involve generalization of experiential knowledge based on particular projects, critical examination of the reasons (if any) behind specific decisions, traditions or ‘best practices’ and exploration of relations to existing relevant theory. In an educational setting, a similar consolidation of the students’ own experiential knowledge seems suitable to train them in designerly academic method, and make them critically aware of the potentials and limitations of design research. For knowledge consolidation to fall under the concept of design research, the criterion of novelty (Section 4.1) must somehow be fulfilled. The designerly knowledge that is being developed, expressed and disseminated in the www.intellectbooks.com 1.23 Per Galle 18. Note added 2017: Several passages in the original Danish version of the appendix were concerned with the difference in meaning between ‘science’ and the Danish word ‘videnskab’, whose meaning overlaps that of ‘science’ but is broader. Since this is hardly of any interest to English-speaking readers, I have skipped those passages, and revised the remaining parts of the appendix to restore coherence. form of application theory, must be sufficiently new, even if its object remains the same. But the novelty criterion can be regarded as fulfilled, if the consolidation leads to an increased degree of familiarity with the object of knowledge; for instance through well-supported generalization, or new inquiries that confirm or disconfirm habitual assumptions. However, one difficulty that should be expected in connection with knowledge consolidation is that the knowledge by acquaintance generated by design practice will be ‘tacit knowledge’ to some extent, and will be associated with specific cases, ‘exemplars’, from which it may be difficult to generalize (Rolf 1995, particularly Chapters 7 and 8; Wengenroth 2004; Niedderer 2007; Hove 2010). As for the theory of science and meta-theory for design, it expresses knowledge whose relevance to practical designing is more indirect. Its raison d’être is, first, that it offers the design professions an overarching frame of understanding and a clarified terminology, for the benefit of teaching and research. Second, the theory of science and meta-theory may enable designers and design researchers to criticize, select and develop (application) theory for design in a competent and well-informed manner, for the benefit of design teaching and practice. The present article, for instance, is intended as a contribution to the theory of science and meta-theory for design, with precisely those two purposes in mind. 5. CONCLUSION It is up to the reader to judge whether this ambition has been achieved. To the extent that it has, we have taken a step towards the goal set forth in the introduction: a theory that enables us to understand how design and research can be united in a fruitful manner – within and across the design professions. In my efforts to clean up and clarify the requisite concepts and terminology, I have been committed to respecting intuition and common parlance. If the results seem familiar, that may be why. What transcends the accommodation of common parlance, however, is the attempt at developing, deliberately and systematically, a coordinated system of concepts, with associated terminology. This is a system, I submit, that takes the particular character of the design professions into account, but without thereby isolating them from the world of other professions with which the design professions coexist and must cooperate. APPENDIX: THE WORD ‘SCIENCE’ – A SOURCE OF CONFUSION18 The term ‘theory of science for design’ appears in the title of this article, and I have been using it extensively – though not without qualms. For using such a combination of the words ‘science’ and ‘design’ is likely to cause confusion, as we shall discuss below. Yet, despite these misgivings, I retained ‘theory of science’, because it is a term so ingrained in other disciplines that it is virtually indispensable if you want to theorize about research – whether in design or elsewhere. On the face of it, the term ‘theory of science for design’ seems to suggest that ‘design science’ is a meaningful term, presumably designating a new kind of science. That is controversial, however. 1.24 Artifact: Journal of Design Practice Elements of a shared theory of science for design Indeed, we might boldly (and somewhat at odds with common parlance) stipulate that science is whatever researchers do in their capacity as researchers, or that science is the body of theory thereby produced. And so, given that active design researchers exist as a matter of obvious fact, this leads us to the conclusion that something very naturally labelled ‘design science’ must also exist. On the other hand, design research is a much more recent and less widespread phenomenon than, say, research in the natural sciences, such as physics, chemistry, biology, etc. So it is still early days, and it would be far too pretentious to claim that design research has created an entire science. For whatever ‘science’ may mean, it has connotations of something rather dauntingly comprehensive. Furthermore, there is some evidence that ‘science’ as used in contemporary English is biased towards research or theory concerning the study of natural phenomena. For example, according to Oxford Dictionaries, the primary sense of ‘science’ is ‘[t]he intellectual and practical activity encompassing the systematic study of the structure and behaviour of the physical and natural world through observation and experiment’ (Science 2017). Only as a secondary sense of ‘science’ we find ‘[a] systematically organized body of knowledge on a particular subject’ (but I take this as licence to retain my term ‘theory of science for design’ after all). According to the WordNet lexical database for English (Princeton University 2015), the bias towards natural phenomena is not so strong, if at all present. One meaning (or ‘sense’, as it is called) of ‘science’ is described in WordNet (see Science n.d.) as ‘ability to produce solutions in some problem domain’. (This matches quite well what designers do, thus perhaps providing another licence to speak of a ‘theory of science for design’.) Another sense of ‘science’ is ‘a particular branch of scientific knowledge’ – where ‘scientific’, in turn, is defined in terms of ‘science’! This is not very informative, but if one turns on the display of hyponyms of ‘science’ – i.e. words with a sense that is a specialization of the current sense of ‘science’ (Science > S > direct hyponym n.d.) – then it turns out that no less than nineteen hyponyms are listed. These include ‘mathematics’, ‘architectonics’ (alias ‘tectonics’, i.e. ‘the science of architecture’ – but no science of design in general is mentioned), ‘psychology’, ‘information science’, ‘cognitive science’, ‘social science’, ‘strategics’, ‘systematics’ (on classification), ‘thanatology’ (on death, especially psychological and social aspects), ‘cryptology’ and ‘linguistics’. The remaining eight are all names of various natural sciences. That is hardly evidence of a bias towards the study of natural phenomena, but names of familiar disciplines of the humanities (such as ‘cultural studies’, ‘history’, ‘art history’, etc.) are conspicuously absent from the list of hyponyms. So at least the meaning of ‘science’ as recorded in WordNet has a bias towards non-humanist research (with the possible exception of psychology, depending on demarcation issues). Such a non-humanist conception of science would be quite appropriate if we subscribed to Simon’s idea of ‘a science of design, a body of intellectually tough, analytic, partly formalizable, partly empirical, teachable doctrine about the design process’ ([1969] 1996: 113, emphasis added). He points out artificial intelligence and formal methods of optimization as important techniques for design, and even ventures to claim that ‘[t]he need to make design theory explicit and precise in order to introduce computers into the process has been the key to establishing its academic respectability’ ([1969] 1996: 114). www.intellectbooks.com 1.25 Per Galle But it would be misleading to take Simon and his technocratic approach as representative for design research in general, since there are design researchers who work within the scholarly purview of the humanities. One prominent example is Donald Schön, whose classic, The Reflective Practitioner (1983), inaugurated quite a different kind of design research, which – in stark contrast to Simon’s ‘science of design’ – was centred around empirical studies of the actual ‘reflective practice’ of designers (for a review and comparison of Simon and Schön, see Galle 2011 [citation updated 2017]). The upshot of our discussion so far is that using the term ‘design science’ about what design researchers do or produce, is at best controversial, or rather misleading. Its intuitive meaning (and even more clearly Simon’s notion of it) is much too narrow to cover every kind of design research that is actually being conducted. (2006: Ch. 7) Nigel Cross has made a vigorous attempt to bring some order into the terminological confusion about the relationship between design and science. For example, he makes a distinction between ‘design science’ and ‘science of design’. ‘Design science’, he suggests, would refer to the highly controversial idea of an ‘explicitly organised, rational and wholly systematic approach to design’. That is, design as an inherently scientific activity, presumably much like Simon’s ‘science of design’. By ‘science of design’, on the other hand, Cross means the study of design as a phenomenon, using scientific – i.e. systematic and reliable – methods. This distinction may resolve the controversy; but it seems to me that neither ‘design science’ nor ‘science of design’ captures the important kinds of design research that produce what I called application theory for design. Thus the fact remains that speaking of science in connection with design tends to create confusion. What I suggest, therefore, is simply that we do not. In other words, that when speaking of design, we place the word ‘science’ under a taboo. As noted in Section 3.1, ‘science’ and ‘research’ can be taken to mean more or less the same. So we might as well speak of research, rather than science. Only when using the phrase ‘theory of science for design’ we should break the taboo, in order to ensure that people in other disciplines understand what we are talking about. ACKNOWLEDGEMENTS I would like to thank Anette Højlund, Cecilie Bendixen, Flemming Tvede Hansen, Jon Knudsen and Mogens Myrup Andreasen, all of whom have read and commented on earlier versions of the article. Helle Egsgaard offered valuable critical remarks on a tentative formulation of the definition of design; this led to a greater emphasis on the initial thought processes involved. [Added 2017:] I am grateful to the editors of Artifact: Journal of Design Practice for inviting me to publish the present updated version of the article in English, and to the editor of FormAkademisk, where the original appeared in Danish in 2010, for her permission to do this. REFERENCES Ahnfeldt-Mollerup, M. (2005), ‘Designfagets faglighed’ (‘The professionalism of design’), Magasinet Humaniora, 20:3, pp. 8–11. Alexander, C., Ishikawa, S., Silverstein, M., Jacobsen, M., Fiksdahl-King, I. and Angel, S. (1977), A Pattern Language, New York: Oxford University Press. 1.26 Artifact: Journal of Design Practice Elements of a shared theory of science for design Anderson, E. (2009), Feminist Epistemology and Philosophy of Science, Stanford Encyclopedia of Philosophy (Winter 2009 ed.), Stanford: Stanford University, http://plato.stanford.edu/archives/win2009/entries/feminismepistemology/. Accessed 1 July 2010. Archer, B. (1979), ‘Whatever became of design methodology?’, Design Studies, 1:1, pp. 17–18. Bamford, G. (1990), ‘Design, science, and conceptual analysis’, in J. Plume (ed.), Architectural Science and Design in Harmony. Joint ANZAScA/ADTRA Conference, Sydney, 10–12 July, Kensington, N. S. W.: School of Architecture, University of New South Wales, pp. 229–38. Biggs, M. A. R. and Büchler, D. (2007), ‘Rigor and practice-based research’, Design Issues, 23:3, pp. 62–69. —— (2008), ‘Architectural practice and academic research’, Nordisk Arkitekturforskning, 20:1, pp. 83–94. Boztepe, S. (2007), ‘Toward a framework of product development for global markets: A user-value-based approach’, Design Studies, 28:5, pp. 513–33. Buchanan, R. (1998), ‘Branzi’s dilemma: Design in contemporary culture’, Design Issues, 14:1, pp. 3–20. —— (2001), ‘Design research and the new learning’, Design Issues, 17:4, pp. 3–23. —— (2004), ‘Design as inquiry: The common, future and current ground of design’, in J. Redmond, D. Durling and A. Bono de (eds), DRS Futureground International Conference, Melbourne: Faculty of Art & Design, Monash University, pp. 1–21. Bunge, M. (1966), ‘Technology as applied science’, Technology and Culture, 7:3, pp. 329–47. Clausen, C., Jørgensen, U. and Yoshinaka, Y. (2009), ‘Design mellem teknologi, brug og form’ (‘Design bridging technology, use and form’), in U. Jørgensen (ed.), I teknologiens laboratorium – ingeniørfagets videnskabsteori (‘In the laboratory of technology – the theory of science for the engineering profession’), Lyngby: Polyteknisk Forlag, pp. 203–23. Collin, F. and Køppe, S. (eds) (2003), Humanistisk videnskabsteori (‘Humanist theory of science’), København: DR Multimedie. Cross, N. (2006), Designerly Ways of Knowing, London: Springer. Davis, M. (2008), ‘Why do we need doctoral study in design?’, International Journal of Design, 2:2, pp. 71–79. Dickson, T. (2002), Designforskning. En international oversigt (‘Design research. An international survey’), Aarhus: Arkitektskolens Forlag. Farrell, R. and Hooker, C. (2012), ‘The Simon–Kroes model of technical artifacts and the distinction between science and design’, Design Studies, 33:5, pp. 480–95. —— (2015), ‘Designing and sciencing: Response to Galle and Kroes’, Design Studies, 37, pp. 1–11. Fink, H. (2003), ‘Universitetsfagenes etik’ (‘The ethics of university disciplines’), in H. Fink, P. C. Kjærgaard, H. Kragh and J. E. Kristensen (eds), Universitet og videnskab. Universitetets idéhistorie, videnskabsteori og etik (‘University and science. The history of ideas, theory of science, and ethics of the university’), København: Hans Reitzels Forlag, pp. 193–221. Fokkinga, S. F. and Desmet, P. M. A. (2013), ‘Ten ways to design for disgust, sadness, and other enjoyments: A design approach to enrich product experiences with negative emotions’, International Journal of Design, 7:1, pp. 19–36. www.intellectbooks.com 1.27 Per Galle Friedman, K. (2005), ‘Det udvidede designbegreb’ (‘The expanded concept of design’), Magasinet Humaniora, 20:3, pp. 4–7. Fuglsang, L. and Olsen, P. B. (eds) (2004), Videnskabsteori i samfundsvidenskaberne. På tværs af fagkulturer og paradigmer (‘Theory of science in the social sciences. Across disciplinary cultures and paradigms’), 2 ed., Frederiksberg C: Roskilde Universitetsforlag. Galle, P. (1996), ‘Design rationalization and the logic of design: a case study’, Design Studies, 17:3, pp. 253–75. —— (1999), ‘Design as intentional action: a conceptual analysis’, Design Studies, 20:1, pp. 57–81. —— (2003), ‘Forskning i de kunstneriske designfag’ (‘Research in the artistic design professions’), Nordisk Arkitekturforskning, 16:3, pp. 81–88. —— (2008), ‘Candidate worldviews for design theory’, Design Studies, 29:3, pp. 267–303. —— (2009), ‘The ontology of Gero’s FBS model of designing’, Design Studies, 30:4, pp. 321–39. —— (2010), ‘Elementer af en fælles designfaglig videnskabsteori’, FORMakademisk, 3:2, pp. 51–76. —— (2011), ‘Foundational and instrumental design theory’, Design Issues, 27:4, pp. 81–94. —— (2016), ‘Self-Knowledge by proxy: Parsons on philosophy of design and the modernist vision’, She Ji: The Journal of Design, Economics, and Innovation, 2:4, pp. 322–42. Galle, P. and Kroes, P. (2014), ‘Science and design. Identical twins?’, Design Studies, 35:3, pp. 201–31. —— (2015), ‘Science and design revisited’, Design Studies, 37, pp. 67–72. Godfrey-Smith, P. (2003), Theory and Reality: An Introduction to the Philosophy of Science, Chicago: University of Chicago Press. Groat, L. and Wang, D. (2002), Architectural Research Methods, New York: John Wiley & Sons. Halstrøm, P. L. and Galle, P. (2014), ‘Design as co-evolution of problem, solution, and audience’, Artifact, 3:4, pp. 3.1–3.13. Hammershøj, L. G. (2008), At forholde sig akademisk. Om opgaveskrivning på lange videregående uddannelser (‘The academic attitude. On written assignments in long-cycle higher education’), Arbejdspapir (‘working paper’), Emdrup, DK: Institut for Pædagogisk Filosofi, DPU, http://www. dpu.dk/om/lgha. Accessed 25 November 2008. Harfield, S. (2007), ‘On design “problematization”: Theorising differences in designed outcomes’, Design Studies, 28:2, pp. 159–73. Herbert, D. M. (1993), Architectural Study Drawings, New York: Van Nostrand Reinhold. Hilpinen, R. ([1999] 2011), Artifacts, The Stanford Encyclopedia of Philosophy (Winter 2011 ed.), Stanford: Stanford University, http://plato.stanford.edu/ archives/win2011/entries/artifact/. Accessed 30 November 2017. Hjelm, S. I. (2005), ‘If everything is design, what then is a designer?’, In the Making, [1st] Nordic Design Research Conference, Copenhagen: 29–31 May, http://www.tii.se/reform/inthemaking/proceedings.htm. Accessed 1 June 2005. Horváth, I. (2004), ‘A treatise on order in engineering design research’, Research in Engineering Design, 15:3, pp. 155–81. Houkes, W. and Vermaas, P. E. (2010), Tecnnical Functions. On the Use and Design of Artefacts, Dordrecht: Springer. 1.28 Artifact: Journal of Design Practice Elements of a shared theory of science for design Hove, H. (2010), FORMSPROG. Formidling af designerens formgivningserfaringer (‘Language of form. Communicating the designer’s experience of form’), Copenhagen Working Papers on Design // 2010 // No. 4, Copenhagen: The Danish Design School, http://www.dkds.dk/Forskning/Publikationer/ Copenhagen_workingpapers_on_design. Accessed 28 July 2010. Hubka, V. and Eder, W. E. ([1992] 1996), Design Science. Introduction to the Needs, Scope and Organization of Engineering Design Knowledge, 2nd translated, revised and augmented ed., London: Springer-Verlag London, http:// deseng.ryerson.ca/DesignScience/. Indenrigs- og Sundhedsministeriet (2007), ‘Bekendtgørelse om speciallæger’ (‘Ministerial order on medical specialists’), BEK nr. 1248, 24 October. Ishikawa, J. J. and Steup, M. (2017), The Analysis of Knowledge, Stanford Encyclopedia of Philosophy (Fall 2017 ed.), Stanford: Stanford University, https://plato.stanford.edu/archives/fall2017/entries/knowledge-analysis/. Accessed 5 December 2017. Jensen, H. S. (2003), ‘Videnbegrebet’ (‘The concept of knowledge’), in P. N. Bukh, K. S. Christensen and J. Mouritsen (eds.) Videnledelse: et praksisfelt under etablering, København: Jurist- og Økonomforbundets Forlag, pp. 25–42. Jensen, L. B. (2004), Fra patos til logos. Videnskabsretorik for begyndere (From patos to logos. Science rhetoric for beginners), Frederiksberg C: Roskilde Universitetsforlag. Jørgensen, U. (2009), ‘Teknologi og ingeniørvidenskab’ (‘Technology and engineering science’), in U. Jørgensen (ed.), I teknologiens laboratorium – ingeniørfagets videnskabsteori (‘In the laboratory of technology – the theory of science for the engineering profession’), Lyngby: Polyteknisk Forlag, pp. 255–72. Jørgensen, U., Jakobsen, A. and Clausen, C. (2009), ‘Problemløsning i ingeniørarbejde’ (‘Problem solving in engineering’), in U. Jørgensen (ed.), I teknologiens laboratorium – ingeniørfagets videnskabsteori (‘In the laboratory of technology – the theory of science for the engineering profession’), Lyngby: Polyteknisk Forlag, pp. 131–54. Klausen, S. H. (2005), Hvad er videnskabsteori? (‘What is theory of science?’), København: Akademisk Forlag. Koch, C. A. (2004), ‘Kritisk rationalisme’ (‘Critical rationalism’), in L. Fuglsang and P. B. Olsen (eds), Videnskabsteori i samfundsvidenskaberne. På tværs af fagkulturer og paradigmer (‘Theory of science in the social sciences. Across disciplinary cultures and paradigms’), 2nd ed., Frederiksberg C: Roskilde Universietsforlag, pp. 79–111. Kragh, H. (2003), ‘Hvad er videnskab?’ (‘What is science?’), in H. Fink, P. C. Kjærgaard, H. Kragh and J. E. Kristensen (eds), Universitet og videnskab. Universitetets idéhistorie, videnskabsteori og etik (‘University and science. The history of ideas, theory of science, and ethics of the university’), København: Hans Reitzels Forlag, pp. 145–92. Krippendorff, K. (2006), The Semantic Turn. A New Foundation for Design, Boca Raton: Taylor & Francis. —— (2007), ‘An exploration of artificiality’, Artifact, 1:1, pp. 17–22. Kroes, P. (2002), ‘Design methodology and the nature of technical artefacts’, Design Studies, 23:3, pp. 287–302. Kuhn, T. S. ([1962] 1996), The Structure of Scientific Revolutions, 3rd ed., Chicago: University of Chicago Press. Langergaard, L. L., Rasmussen, S. B. and Sørensen, A. (2006), Viden videnskab og virkelighed (‘Knowledge, science and reality’), Frederiksberg: Samfundslitteratur. www.intellectbooks.com 1.29 Per Galle Lidwell, W., Holden, K. and Butler, J. (2003), Universal Principles of Design, Beverly, MA: Rockport Publishers. McDonough, W. and Braungart, M. (2002), Cradle to Cradle: Remaking the Way We Make Things, New York: North Point Press. Mollerup, P. (2007), ‘Man-made answers to man-felt needs’, Artifact, 1:1, pp. 37–39. Monö, R. (1997), Design for Product Understanding. The Aesthetics of Design from a Semiotic Approach, Stockholm: Liber. Niedderer, K. (2007), ‘Mapping the meaning of knowledge in design research’, Design Research Quarterly, 2:2, pp. 1, 5–13. Norman, D. A. (2004), Emotional Design, New York: Basic Books. —— (2013), The Design of Everyday Things, 3rd. rev. and expanded ed., New York: Basic Books. OECD (2002), Frascati Manual. Proposed Standard Practice for Surveys on Research and Experimental Development, Paris: OECD, Organization for Economic Co-operation and Development. Parsons, G. (2016), The Philosophy of Design, Cambridge: Polity Press. Petroski, H. (2006), Success Through Failure. The Paradox of Design, Princeton and Oxford: Princeton University Press. Poggenpohl, S., Chayutsahakij, P. and Jeamsinkul, C. (2004), ‘Language definition and its role in developing a design discourse’, Design Studies, 25:6, pp. 579–605. Polanyi, M. (1967), The Tacit Dimension, New York: Anchor Books. Pollock, J. L. and Cruz, J. (1999), Contemporary Theories of Knowledge, 2nd ed., Lanham: Rowman & Littlefield. Princeton University (2015), ‘What is WordNet?’, http://wordnet.princeton. edu/wordnet/. Accessed 22 December 2017. Quine, W. V. (ed.) ([1948] 1961), ‘On what there is’, in From a Logical Point of View. Logico-Philosophical Essays, Cambridge, MA: Harvard University Press, pp. 1–19. Reilly, L. (2002), ‘An alternative model of “knowledge” for the arts’, Working Papers in Art and Desgin 2, Hatfield: University of Hertfordshire, https:// www.herts.ac.uk/__data/assets/pdf_file/0015/12309/WPIAAD_vol2_reilly. pdf. Accessed 5 March 2018. Rescher, N. (2002), ‘Epistemic logic’, in D. Jaquette (ed.), A Companion to Philosophical Logic, Malden, MA: Blackwell, pp. 478–90. Rolf, B. (1995), Profession, Tradition och Tyst Kunskap. En studie i Michael Polanyis teori om den professionella kunskapens tysta dimension (‘Profession, tradition and tacit knowledge. A study in Michael Polanyi’s thory on the tacit dimension of professional knowledge’), Otta, Norge: Nya Doxa. Ryle, G. ([1949] 2000), The Concept of Mind, London: Penguin Books. Salustri, F. A. and Rogers, D. (2008), ‘Some thoughts on terminology and discipline in design’, in D. Durling, C. Rust, L.-L. Chen, P. Ashton and K. Friedman (eds), Undisciplined! Proceedings of the Design Research Society Conference, paper no. 299, http://shura.shu.ac.uk/530/. Accessed 5 March 2018. Sargent, P. (1994), ‘Design science or nonscience’, Design Studies, 15:4, pp. 389–402. Schön, D. (1983), The Reflective Practitioner. How Professionals Think in Action, New York: Basic Books. Schön, D. and Wiggins, G. (1992), ‘Kinds of seeing and their functions in designing’, Design Studies, 13:2, pp. 135–56. 1.30 Artifact: Journal of Design Practice Elements of a shared theory of science for design Science (2017), Oxford Dictionaries, Oxford: Oxford University Press, https:// en.oxforddictionaries.com/definition/science. Accessed 22 December 2017. Science (n.d.), WordNet 3.1, http://wordnetweb.princeton.edu/perl/webwn?s= science&sub=Search+WordNet&o2=&o0=1&o8=1&o1=1&o7=&o5=&o9 =&o6=&o3=&o4=&h=. Accessed 22 December 2017. Science >S> direct hyponym (n.d.), ‘WordNet 3.1’, http://wordnetweb.princeton.edu/perl/webwn?o2=&o0=1&o8=1&o1=1&o7=&o5=&o9=&o6=&o3 =&o4=&s=science&i=1&h=100000011013020 – c. Accessed 22 December 2017. Sevaldson, B. (2010), ‘Discussions & movements in design research. A systems approach to pracice research in design’, FORMakademisk, 3:1, pp. 8–35, http://www.formakademisk.org/index.php/formakademisk/article/view/62. Accessed 18 July 2010. Simon, H. A. ([1969] 1996), The Sciences of the Artificial, 3rd ed., Cambridge, MA: MIT Press. Sokal, A. and Bricmont, J. ([1997] 2003), Intellectual Impostures, New ed., London: Profile Books. Steup, M. (2017), Epistemology, Stanford Encyclopedia of Philosophy (Fall 2017 ed.), Stanford: Stanford University, https://plato.stanford.edu/archives/ fall2017/entries/epistemology/. Accessed 5 December 2017. Stormhøj, C. (2004), ‘Feministisk teori: Videnskabs-teoretiske positioner’ (‘Feminist theory: Positions in theory of science’), in L. Fuglsang and P. B. Olsen (eds), Videnskabsteori i samfundsvidenskaberne. På tværs af fagkulturer og paradigmer (‘Theory of science in the social sciences. Across disciplinary cultures and paradigms’), 2nd ed., Frederiksberg C: Roskilde Universietsforlag, pp. 473–516. Teori (‘theory’) (n.d.), Gyldendals Online Leksikon (‘Gyldendal’s Online Dictionary’), http://www.gyldendalsleksikon.dk/leksikon/leksikon.htm. Accessed 15 May 2008. Vision, G. (1986), ‘Reference and the ghost of Parmenides’, in R. Haller (ed), Non-Existence and Predication (Grazer philosophische Studien, Band 25 / 26), Amsterdam: Rodopi, pp. 297–326. Wengenroth, U. (2004), ‘Managing engineering complexity – an historical perspective’, Engineering Systems Symposium, MIT Engineering Systems Division, Cambridge, MA, 29–31 March, http://esd.mit.edu/resources/ symposium2004.html. Accessed 25 June 2009. Zagzebski, L. (1999), ‘What is knowledge?’, in J. Greco and E. Sosa (eds), The Blackwell Guide to Epistemology, Malden, MA: Blackwell Publishing, pp. 92–116. SUGGESTED CITATION Galle, P. (2018), ‘Elements of a shared theory of science for design’, Artifact: Journal of Design Practice, 5:1, pp. 1.1–1.32, doi: 10.1386/art.5.1.1.1_1 CONTRIBUTOR DETAILS Per Galle, born 1950 in Copenhagen, was trained as an architect at the Royal Danish Academy of Fine Arts, School of Architecture. He holds a Ph.D. in computer science from University of Copenhagen, on combinatorial methods for architectural space planning. Galle works at the Royal Danish Academy of Fine Arts, School of Design, where he teaches academic writing and other subjects, supervises a number of Ph.D. students, and pursues his research www.intellectbooks.com 1.31 Per Galle interests. These currently include the philosophy of design, and the prospects of a theory of science for design. He is a Fellow of the Design Research Society. Contact: The Royal Danish Academy of Fine Arts, School of Design, Philip de Langes Allé 10, DK1435 Copenhagen K, Denmark. E-mail: [email protected] Per Galle has asserted his right under the Copyright, Designs and Patents Act, 1988, to be identified as the author of this work in the format that was submitted to Intellect Ltd. 1.32 Artifact: Journal of Design Practice