Learning through drawing in Art and Science James McArdle and Russell Tytler
Abstract Science studies demonstrate the role of the visual in creating new knowledge, indicating potential parallels between creative processes in science and in art. Case studies show how art has intersected with science ideas, and the commonalities and differences between creative visual processes in art compared to science. We use the construct of affordances to illuminate these differences. We explore a school science pedagogy in which students construct visual representations, which more authentically demonstrates the imaginative processes through which knowledge is created in science. We discuss how, conversely, scientific impulses and processes may enrich art education. Art is centrally concerned with image production. Science is increasingly reliant on images for knowledge generation, analysis and communication (Elkins, 2011). We argue that a sharper conception of the commonalities and differences in the creation and harnessing of visual representations in art and science can inform productive directions for mutually supporting learning in science, and in art.
The image in science In science, Latour (1986, p. 3) argued that the emergence of scientific thought has depended on developing effective representational tools, and that changes over time to procedures for writing and imaging have altered the ways scientists argue and validate their case. From multidimensional modeling of galaxies to computer generated abstracted graphics, visuo-spatial imaging plays a critical generative role in scientific reasoning and knowledge building. We can find detailed early evidence of the central role of images in the work of Michael Faraday. Faraday, in his experimentations at the London Institute, was a careful documenter of his explorations and ideas, and a dedicated communicator. David Gooding (2004, 2006) analysed the role of Faraday’s diarized visualisations in generating new perspectives on electromagnetic phenomena, and establishing these as theory. Table 1, based on Gooding, shows a series of drawn entries in Faraday’s notebook entries over one day where he moved from observations of patterns of needle orientation around a wire, to 3D enhancement to imagine field lines in 3D, to the imaging of a process in time, and finally to an inference for construction of the first electric motor.
Table 1 (Gooding, 2004, p. 16). Visual reasoning by dimensional enhancement and reduction - - a day's work recorded in Faraday's manuscript for 3 September 1821
Gooding (2006, p. 60) argues for the central role of visual images, such as those generated by Faraday, in scientific reasoning: … while successful science does require a stable linguistic formulation, creative research cannot be conducted solely with well-formed linguistic representations. There are nonvisual ways of forging an isomorphism of words, images or symbols to what they denote, but images are particularly conducive to the essential, dialectical movement between the creative stages of discovery and the deliberative, rational stages in which rules and evaluative criteria are introduced to fix meanings and turn images from interpretations into evidence.
Thus, reasoning through visual images underpins the formal, rational, written language through which scientific knowledge is generally assumed to be secured. New developments in scientific thinking are very often fundamentally tied to newly generated visual inscriptions, such as with Feynmann diagrams of particle interactions, or the double helix structure of chromosomes. The general inferential process by which Faraday and other scientists generate new visual forms through reduction and dimensional enhancement exists in the realm of imaginative creation. The powerful suggestive visions of constructed images are active in progressing and expressing novel thought, forms and processes. The drawings act to channel perceptions to drive imaginative acts. They would seem to have much in common with exploratory creation and investigations of new visual languages in art. In this paper we are concerned to explore the links implied between knowledge production in art and science, and further to learning in art and science. We will begin exploring these relationships with an account of a school science pedagogy that draws on these insights.
A representation production pedagogy in science There is a substantial literature arguing that learning and knowing in a disciplinary area involves a process of enculturation into its discursive practices. These are shaped around a set of discipline specific and generic literacies to build and validate knowledge (Clement, 2008; Lemke, 2004; Moje, 2007). Explanatory and problem solving competency in science involves the generation and coordination of multimodal representations, including visual images and 3D models. Further studies have verified the defining, rather than supporting role, played by representations in knowledge generation and problem solving in school science (Lehrer & Schauble,
2006). There is thus increasing recognition of the central role of visual representations, including drawings and models, in science learning as in the epistemic practices of science. We have been developing a school science pedagogy that challenges students to produce representations, often visual, to explore their capacity to clarify and explain phenomena (Tytler et al., 2013). Students draw, and model, their interpretations of phenomena, supported by the teacher to evaluate and refine these to achieve consensus around the scientific representational canon. We advance distinct reasons for actively drawing in learning science (Ainsworth, Prain & Tytler, 2011). In further analysis we have linked these arguments for active construction of representations in school science to the role of representations in the knowledge-building practices of science itself, through the construct of ‘affordances’ (Gibson, 1979). Drawing enables the explicit representation of spatial relations while its constraints flow from a need to select and simplify the raw phenomenon. Affordances as ‘productive constraints’ account for how and why representation supports reasoning and learning in science (Prain & Tytler, 2012). The key point of these deliberations and research, for this paper, is that they promote understanding of and interest in a) the links between drawing as a scientific practice and as a support for learning science, and b) the role of processes of imagination in representation construction in science, and in school science inquiry, linked to imaginative processes of art. We analyse below the process by which two age 11 students, Jesse and Paul, observed, discussed, constructed design drawings, constructed a model, and communicated a developing understanding of centipede movement. The model construction task was to focus on explicating movement. As with Faraday, 2D drawing channeled attention on key elements of spatial patterns, limiting complexity, enabling sharper visualisations of essential structures and processes. Figure 1a shows the boys’ drawings of aspects of the centipede structure, including a design leading to a segmented model with elastic joints. The model, utilizing a construction kit with segments threaded with elastic, is shown in Figure 1b. As the boys describe their model, Jesse moves his hands in a sideways undulating movement: ‘… so instead of moving in straight lines, it moves [he gestures to signify the undulation] so we used elastic so it could move properly’.
Figure 1a: Drawings and design drawing of Figure 1b: Centipede model with segments centipede structure and movement joined by elastic.
Reporting to the class, they draw on gestures to indicate the undulating movement of the centipede’s legs and body, and move close behind each other to represent segments, using their hands and entire body, gesturing and moving in synch. In considering the processes by which the boys gained insight into the centipede movement, we would draw attention to: •
The perceptual organizing role of their talk, sketches, and decisions on model design, to ascertain how the legs and body move;
•
The way understanding deepens through successive transformations across representations, from labeled drawing, talk, design drawing focusing on the nature of joints, model construction, and embodied characterization, each actively channeling selection and focusing attention.
•
The specific affordances of the different representation tasks. Drawing specified the relation between segments and leg attachments. Design drawing characterizes the joints. The modeling forced attention on the material properties allowing movement (e.g. the choice of elastic, with hard sections). Embodied representation enacted the undulating movement of successive sections.
•
The selection by the boys of media they were familiar with, indicating enjoyment of working with the particular technology, and the exploitation of the affordances offered by the materials. We can presume their applied experimentation with the medium extends their past history of playful use of these media.
The boys’ successive creation of representations involved the imaginative creation of new visual symbolic tools and artifacts, following the process, described by Gooding (2004), from pattern to structure to process. Their crossing of modes in the development of new visual languages is suggestive of Faraday’s imaginative use of drawings and constructed exploratory models. The process is also similar to that described in a scientific paper on centipede movement which involved the use of
models for problem solving the energy advantages of the undulating movement (Zimmer, 1994).
Case studies exploring art and science links Art and science have been continuously linked historically through the work of illustrators, for instance of biological drawings or mechanical diagrams. However, we argue that even with illustration art has an important interpretive role beyond mere objective representation. In this paper we explore the wider role of creative visual processes in knowledge generation and cognition in science, and the converse case of scientific insights driving contemporary art. Art in the service of science, and science used for art-making, entail processes in which technology is linked with representation construction. Case studies from each area demonstrate particular interactions between visual representation and idea exploration and theory generation. In The Expression of the Emotions in Man and Animals (1872) Charles Darwin defended the argument that emotion expressions are evolved and adapted. The book was pioneering in its use of photographs (seven heliotype plates), and for extending the use of the medium into collection and analysis of scientific data. Julia Voss (2010) argues that Darwin “thought with his eyes” ... “to formulate his theories in the first place”. Darwin uniquely used visualizations, sketched (poorly, he estimated), collected and commissioned engravings and photographs. Evidently, the compactness of data contained in the portable and physical form of images, readily compared, was crucially important for the development of his theories, where an account of the stages between evolved features can be envisaged. Origin of the Species (1859) contains just one illustration; Darwin later develops a more sophisticated mode of visualisation in employing artistic manipulation to develop ideas, then communicate them to a bourgeois audience. Consider first the case of a crying baby: the image Oscar Gustav Rejlander contrived from a hasty, necessarily unposed, snapshot, enlarging it, redrawing it in chalk, then rephotographing it, using his established artistic method of montage and imagemanipulation. At that time, photography’s technical shortcomings exaggerated the gap between what a photograph would show, and what observers thought they saw, frustrating early photographers, and certainly Darwin, who was at pains to record fleeting expression. Differences between painting and photography, and whether photography was art, were disputed during this period (Sutton 1863). Rejlander's famous The Two Ways of Life (1857), based on Raphael’s The School of Athens (1510), montaged from thirty-two negatives, is intended for religious or philosophical contemplation. The baby has illustrative and scientific purpose, being commissioned by Charles Darwin. Prodger (2009) argues that Rejlander was an appropriate choice for Darwin's purpose, against popular modern expectations that the use of the medium in science is oriented toward 'objectivity' for scientific illustration, even though aesthetically mediated visual data now necessarily predominates where direct imaging is not possible (Tourney 2009). Given that Darwin required images that would be both ‘realistic’ and paradigmatic or ‘metaphorical’, a photographer whose credo “In all picture compositions the thought should take the first place . . . all else [is] the language which is to give it expression” (quoted in Wall 1896), provided material better than Darwin himself could produce. How much did the extension of Darwin’s theories into the less tangible realm of the socio-psychological evidence for selection and evolution owe to advances in artistic imaging of the time? How did art advance the application of his theory of evolution to expression and the evolution of mind? While Rejlander expresses what would appear to be a didactic program, his experimental 'pure' art-making attracts Darwin. Rejlander’s novel means of amplifying data sparks imagination and reaches a wider
audience. While such examples acknowledge contributions of art to science, the converse situation is also well represented in the literature. . Cameron Robbins’ methodology might be ‘scientific’ but his ‘sampling’ is distinctly quixotic. Studied longitudinally Robbins’ production develops from instruments that produce complex drawn forms from chaotic inputs, to installations elegant in their simplicity. Prodigious from the early 1990s, he creates and deploys a number of drawing machines to record phenomena of wind, tide, even movements of a ridden bicycle. His ‘Heath-Robinson’ instruments mimic chart recorders of the kind once used for weather and tide observations. Darwinian adaptivity appears in this artist’s development as the predominant creative characteristic of his serial production, in the construction of levers, paper drivers and styluses to record the inputs for a specific project (some involve converting whole buildings to contain the device). Adaptation and evolution of forms are processes of both scientific and artistic creation (Simonton, 1999). However not just any variation or innovation may be considered creative unless the product is evolved to, or useful to, a goal. ‘Wild ideas’ without such coherence may be judged insane rather than creative. Simonton points out that the creative effort does not exist in a vacuum but is appreciated according to practical or aesthetic standards. In Robbins’ case it is not the end product, the drawing, that demonstrates this coherence, but rather the instrument, its process and product together that are aesthetically pleasing to the artist’s audience. His works evoke haunting responses to social or political issues, such as The Sea Wailing (2002, with John Turpie), which commemorates an 1840s incident where Aboriginal people were pushed to their death from the Elliston cliffs in South Australia. Robbins’ tuned organ pipes installed in the blowholes of the cliffs resonated mournfully with each swell of the waves. Robbins collaborations, in such works as Sea Wailing, his involvement in artist-run spaces, and community art works are a demonstration of Gruber’s (1988) ‘evolving systems approach’ concept of creative work that reflects the complex interactions of people, processes, and knowledge. Robbins’ Merricks Beach House Installation, 2007, reverses his usual employment of random meteorological effects; he controls air circulation in a room to ‘draw’ with smoke a startling vortex that appears and disappears spontaneously. Creativity in any field produces novelty, and in art this may present as a surprise, revelation, manifestation or apparition. Such words share associations with magical, superstitious or religious experience as being on the ‘spooky’ outer edge of awareness. Artists develop novel ways to convey spatial information in 2D images, by tracing motion. Human vision, through binocular and stereo perception, as well as through other embodied means, enables comprehension of spatial/motion relations to the level of engagement encompassed in Gibson’s term ‘affordance’ (Gibson 1979) and as something which ‘points two ways’ between observer and environment. He sees pictorial, planar imagery as a construction into which we can build a representation or notation of these relationships with mathematical formulae, or by sectioning space as with a window, but he allows that there ‘may be some truth’ in the proposition that a still picture might yield an experience of motion. Ernst (1893) would have us understand that with our two eyes we stand at two places at the one time. One could see new potential to be derived from that idea; in movement, with our two eyes, we might exist in two moments simultaneously. In further experimentation with photography McArdle (in Vortex) seeks to evoke sensations of being on the ground, in the landscape. The imagery is the result of working in the environment to render the unique qualities of human perception with the camera. ‘Motion perspective’ is described by Herschel (1833);
“Let any one traveling rapidly along a high road fix his eye steadily on any object, but at the same time not entirely withdraw his attention from the general landscape, he will see…the whole landscape thrown into rotation, and moving round that object as a centre” These phenomena result from ‘optic flow’, a largely subconscious visceral perception arising from a connection between body and environment. Gibson (1979) proposes they are also a sign of mind and attention. Heightened attention is necessary in observing motion perspective as concentrated still points embedded within the surrounding confusion by the observer (photographer/viewer). These are not arbitrary abstractions of the landscape, but give rise to a sense of orientation through a condensation of space by time. The preparation for the construction of McArdle’s mural scale vertical montages involves collecting images of the landscape seen on the move on train and car journeys through the Bendigo region. A glance on such a journey may pick something up and this place is then visited on foot. Moving and swiveling of body and camera in dance-like and manual Figure: James McArdle (2007) gestures re-arranges the chaotic array of near and distant Gush Chromogenic Print forms; they are registered as the streak and blur and the 1500x770 mm. painterly swirl, the vortex, that form in the foreground and background around still points that remain sharp and distinct. The digital permits montage of momentary gestural representations into a seamless triptych of foreground close-up, medium shot and wide angled horizon, combining them to render space through the camera’s placement and panning movement over time. In such work, scientific insights into visual perception of motion are central to its aesthetic exploration. Reviewer Dirk de Bruyn proposes “that these are not random operations but document a bodily relationship to these spaces […]. Is this an indication of a spent and unsettled landscape, a space in crisis or are these the traces of emotion imparted from the body of the photographer himself?” (de Bruyn, 2007). Discussion From these case studies of professional practice in art and science, and learning, we argue that: 1. The core of the creative process in both art and science is the construction and refinement of representations to solve a problem. Opening up affordances through these representations is key in each case. 2. Beyond that foundational core, however, the process differs in two major respects: a. The purpose in science is to produce a generalizable solution that lies beyond individual sensibilities whereas the core purpose in art is to elicit individual insight and enhanced subjectivities. b. The focus in art is on the exploration of how affordances can be generated through the medium, compared to science where the refinement has a core aim external to the medium as such, which is subservient to abstracted ideas. 3. A better understanding of the points of convergence and divergence of art and science can open up enhanced collaborative possibilities for cooperation across the art/science boundary, and productive strategies for enhanced learning in both areas.
With regard to the first point concerning the common representational nature of the creative process in art and science we propose that we can see inspiration and intuition in the way a creative mind from either field makes leaps from one way of seeing to another. In our examples from science and art we are dealing with the problem of representing phenomena. In many of these cases the representation concerns time; how to represent time in a still photographic image, or in a drawing or model. In the science lab, and in the art studio, we distinguish a common pathway from phenomenon> curiosity>discernment of a problem>exploration and analysis> selection and orchestration of elements into a coherent synthesis. The generation, refinement and orchestration of visual and other representations is central to these processes, and the pathway can be discerned in the professional practice of Faraday and McArdle as it can in Jesse and Paul’s exploration of centipede movement or the artistic representation of a forensic investigation. Purposes in art and science However, the different fundamental purposes of representation generation, refinement and orchestration in the two disciplines affect the nature of each element in the process. In art, the burden of the task, and the focus of the analysis, is to create an artefact that for the participant observer offers insight or illumination, renewed or enriched sensibilities and new perceptions. The focus is on the individual response, but the process is such as to achieve new discursive possibilities that advance the field. It is not a random achievement. As with science, it is subject to peer review processes. In science, the exploration and analysis serves to build explanation supported by evidence in an explicit way. We would argue that this is achieved through coordination of the visual representation within explanatory narratives, which allow logical processes to be engaged with, in a formal process of argumentation. The burden of the scientific analysis thus rests on forms of evidence that test and affirm the visual innovation. Thus, in Faraday's case, practical and theoretical notions of fields, generated initially in visual form, are ultimately expressed in multiple forms, including written language and mathematical formulae, which can be reapplied as generalisation. This is contrasted with art in which the validation of the analysis rests on different grounds involving critique within a different evidential tradition. Analysis processes are central to both art and science. These have things in common but points at which the purpose diverges. In terms of commonality, we see for instance in McArdle's work the generative nature of analysis of the process of stereoscopic vision and how this affects the images across the field. However, while such insights may, in a science context, have fed into a convergent explanatory depiction of sight, and the way the camera exposes aspects of vision, the real aim of the exploration is rather to a more divergent end – in art, the analysis feeds back into a generative product and concerns the refinement of this to evoke and stimulate insight into perceptual processes. The central concern is to explore the affordances of the medium to achieve this. Affect is a central part of the process, whereas it is explicitly ruled out of the scientific end product. Affordance as a construct to distinguish practice in art and science In both cases the construct of affordances is fruitful in providing insight into the nature and purpose of the representation construction. In science, the different representations and different modes offer particular affordances through the way they constrain and focus attention on key features of the phenomenon, to build conceptual insight. Each representation offers an element of an explanatory narrative, and these are coordinated to construct a coherent and defensible account. Thus, the boys, in their exploration of centipede movement, make use of talk, diagrams, a model, and role-play, to capture different features of the centipede structures and the nature of the
movement. Each contributes to an understanding that is multiple and distributed across these discursive elements. Each, however, is convergent in its intent. While in science the affordances of visual and other representational elements serve instrumental purposes subservient to objective and abstracted ends, in art the ultimate purpose of the analysis is to explore and refine the possibilities of the medium. While these may be employed for scientific ends (e.g. Darwin and Rejlander, or illustrators like Andrew Howell (Russell, 2012) who created artworks that more accurately depict elephant health than photographs or measurements), ultimately for the artist the creation of affordances is the core purpose. Productively linking art and science Our cases demonstrate problem solving through representation construction that is common to both fields, involving dual exploration of the physical phenomenon and ways of representing it. This raises the question of whether, and under what circumstances, these parallel explorations of affordances might be productively harnessed to support practice in the disciplines, or to support learning. We have described how the creation of affordances central to the artistic purpose can be harnessed to serve scientific ends. This type of representational exploration has underpinned the work of biological illustrators. Enhanced possibilities for collaboration can be seen in a spate of engagements of artists-in-residence at scientific laboratories, of which Howells’ work (Russell, 2012) is an instance. How does the creative work of artists relate to those of scientists? How and at what point does the process of representation construction and refinement in art diverge from that in science? In the case of McArdle's work, as an instance, we would argue that the exploration of the phenomenon of visual perception leading to representation construction is a process that would be generative for both an artistic and a scientific analysis, up to the point that the focus coalesces around representational exploration. McArdle’s purpose is to explore the potential of particular photographic processes and forms to generate perceptual insight involving the nature of vision and embodiment of self in the landscape rather than to converge attention on a description of how vision works in these contexts. The analysis, nevertheless, draws on scientific knowledge and is in a number of respects itself scientific. This raises interesting questions about the possibility of joint scientific / artistic explorations in the classroom. We have been exploring the design of art curriculum at tertiary level in which students approach a range of media in ways that draw on science ideas, and interpret science methodologies, for artistic purposes. These might incorporate science fiction, science processes, or science in the media. Examples include the imaginative recreation of forensic investigations using old techniques from forensic science from the 1920s-1940s which interpret the symbolic potential of crime scene photographs such as the “god’s eye view” from a camera with wide-angle lens, mounted on a high tripod. In the centipede model construction, the boys moved through a series of representations to their model, and then to their role-play to create a multi modal explanatory account of how the centipede moved. If, however, we were to conceive of this as an artistic endeavour, the purpose of which was to create imaginative representations that might include moving models, video, or dance, then the process of analysis of movement would be similar. In each case, the task of the teacher would be to focus attention on the logic of the movement and the problem of representing this. In the science class the teacher continually encouraged and challenged students to represent the movement rather than the more usual, concrete task of creating a static model of the animal form. An art teacher with a commitment to analysis of movement as central to an artistic rendition would be encouraging similar exploration,
but perhaps widened to selection and exploration of the medium itself, and the evocation of further elements of the animal's behavior. We argue that combining these approaches not only would add to a richer and more satisfying understanding of animal movement, but that each of the scientific and artistic response would feed productively into the other to enrich each disciplinary perspective. Thus we speculate that there may be much to gain, in school classrooms focusing on scientific phenomena, from encouraging both 'scientific' and 'artistic' representation work, such that a fuller and more meaningful experience could be created for students, as well as an appreciation of the different but related discursive traditions of the two areas.
References Ainsworth, S., Prain, V., & Tytler, R. (2011). Drawing to learn in science. Science, 333 (26 August), 1096-1097. Clement, J. (2000). Model based learning as key research area for science education. International Journal of Science Education 22 (9), 1041-1053. De Bruyn, D. (2008). Visceral Intuition, Critical Thought : On The Power of the Technical Image in ‘Azimuth’, realtime 83. Doran P. M. (ed.) (1978) Conversations avec Cézanne. Artistes d'hier et d'aujourd'hui, Vol.1. Macula. Elkins, J. (2011). Visual practices across the University: A Report. In O. Grau (Ed.) Imagery in the 21st Century. Cambridge, MA: MIT Press. Gibson, J. (1979) The ecological approach to visual perception. Boston. Houghton Mifflin. Gooding, D. (2004). Visualization, inference and explanation in the sciences. In G. Malcolm (Ed.) Studies in Multidisciplinarity, Vol 2. Elsevier, 1- 25. Gooding, D. (2006). From phenomenology to field theory: Faraday’s visual reasoning. Perspectives on Science, 14(1), 40-65. Gruber, H. E. (1974). Darwin on man: A psychological study of scientific creativity. New York: EP Dutton & Co. Gruber, H. E. (1988) The evolving systems approach to creative work, Creativity Research Journal,1(1), 27-51 Herschel, J. F. W. ‘A Treatise on Astronomy’, Longman, Rees, Orme, Brown, Green & Longman, London (1833) Latour, B. (1999). Pandora’s hope: essays on the reality of science studies. Cambridge, MA, Harvard University Press. Lemke, J. (2004). The literacies of science. In E. W. Saul (Ed.), Crossing borders in literacy and science instruction: Perspectives on framework and practice. Newark DE: International Reading Association and National Science Teachers Association. Lehrer, R., & Schauble, L. (2006). Cultivating model-based reasoning in science education. In K. Sawyer (Ed.), Cambridge handbook of the learning sciences (pp. 371-388). Cambridge, MA: Cambridge University Press. Mach, E. (1893) Popular Scientific Lectures, Vol. 1 Open Court, Illinois, 66-88. Moje, E. (2007). Developing socially just subject-Matter instruction: A Review of the literature on disciplinary literacy learning. Review of Research in Education, 31, 1-44.
Prodger, Phillip (2009). Darwin's camera : art and photography in the theory of evolution. Oxford University Press, Oxford ; New York Prain, V., & Tytler, R. (2012). Learning through constructing representations in science: A framework of representational construction affordances. International Journal of Science Education, available online DOI:10.1080/09500693.2011.626462 Russell, A. (2012). An Australian illustrator has fused science and art to help save an endangered species - the Asian elephant. Australian Geographic, 109, JulyAugust, n.p.. http://www.australiangeographic.com.au/journal/elephantillustrations-the-art-of-science.htm. Simonton, D. K. (1999). Origins of genius: Darwinian perspectives on creativity. New York: Oxford University Press. Sutton, T. (1863). “On Some Uses and Abuses of Photography,” in Journal of the Photographic Society. 8:129 (January 15, 1863): 202-6. Toumey, C. (2009). Truth and Beauty at the Nanoscale. Leonardo 42(2), 151-155 Tytler, R., Prain, V., Hubber, P., & Waldrip, B. (Eds.). (2013). Constructing representations to learn science. Rotterdam, The Netherlands: Sense Publishers. Voss, Julia & Lantz, Lori. (Translator) (2010). Darwin's pictures : views of evolutionary theory, 1837-1874. Yale University Press, New Haven [Conn.] Wittgenstein, L. (1972). (G. E. M. Anscombe, Trans.). In G. E. M. Anscombe, & R. Rhees (Eds.), Philosophical investigations (2nd ed.). Oxford: Basil Blackwell.