Published in Cartographic Journal 32:6. pp.
A shorter version of this paper is available...
A shorter version of this paper is available...
ABSTRACT: Regardless of changing official definitions, many cartographers continue to think of cartography in terms of art and science. This paper critiques the use of the art/science dualism as a means of understanding cartography, particularly by those interested in reexamining the role of aesthetics, design, and visual expression in cartography. Two basic approaches to "art" and "science" in the context of cartography and information graphics are described along with their limitations. I argue that the manner in which the art/science dualism has been used in cartography does not stand up under close scrutiny and that attempts to strictly differentiate art and science have ended in confusion while simultaneously demeaning both art and science. I suggest that various and seemingly divergent trends including postmodern deconstruction, hypermedia, cognitive psychology, semiotics, geographical information systems, and visualization all point to a process oriented means of understanding cartography. Within this process, "art" and "science" serve a functionally similar role, informing the different ways in which we come to know and re-know our human and physical worlds.
Cartography is considered as the science of preparing all types of maps and charts and includes every operation from original survey to final printing of maps (United Nations 1949, cited in Freitag 1993).
Cartography is the art, science and technology of making maps, together with their study as scientific documents and works of art (I.C.A in Meynen 1973).
Cartography is the discipline dealing with the conception, production, dissemination and study of maps (I.C.A. in Anonymous 1992).
The four and a half decades between the first international definition of cartography adopted by the United Nations and the recent definition devised by the International Cartographic Association (I.C.A.) have witnessed a varying role for the often vaguely defined yet important ideas of "art" and "science" in cartography. Cartography, by definition, was a "science" in 1949, an "art, science, and technology" in 1973, and is now neither an art nor a science. Yet changing definitions do not necessarily change the way cartographers think about cartography. Recent discussions about the nature of cartography have yet to forsake the notion of cartography as an art or a science or some combination of the two. Keates (1993 p. 201) suggests a path for future inquiry by arguing that "the important relationship between cartographic design and art has never received the same degree of attention [as that between cartographic design and science]." This interest in aesthetics, design, and visual expression in cartography - commonly categorized as its "artistic side" - is more broadly reflected in the popularity of Tufte's books on information design (1983, 1992). In both of these cases cartography (and information graphics in general) are understood to have an important "artistic" component which has been undervalued. There is, then, an important future role for the complex idea of "art" - however vaguely defined - in cartography, regardless of its omission from the most current I.C.A. definition.
In this paper I suggest that despite the vacillating role of "art" and "science" in official definitions of cartography, many cartographers not only still think in terms of this dualism, but also explicitly desire a rekindled focus on the array of aesthetic and design issues which are often categorized as "the art of cartography." I seek to examine the history of the art/science dualism in cartography and information graphics in general. Such an examination reveals distinct and important differences in how the art/science dualism is used, differences which still inform and shape thinking about cartography. My goal is to provide a basis from which to critically evaluate the way cartographers have used the idea of art and science to think about maps and cartography, and to elucidate some of the limitations and problems of the art/science dualism. Before we can heed the call of Keates and others to reconsider the role of what has been defined as 'art' in cartography, we need to articulate how the idea of 'art' and 'science' have come to shape - and in some cases limit - our understanding of cartography.
Art and Science in Current Discussions about Cartography
Cartographers have long sought to understand the most general nature of cartography. Such ways of thinking often include both "art" and "science." The persistence of literature about the "art of cartography" implies that the relationship between art and science in cartography is seen as significant (Morris 1982, Woodward 1987, Marcus et al. 1992, Keates 1993). Such ways of thinking are important because they reflect fundamental assumptions about what cartography is and what cartographers do. We design maps, design courses, teach students, engage in research, evaluate publications, and write about cartography based on such distinctions.
That the art/science dualism still shapes our thinking about cartography is revealed by a recent discussion on the Geographic Information Systems electronic mail discussion group (GIS-L). Thierry Huet asked "does anybody know if cartography is an art?" (Huet 1994). Dozens of replies followed this query. The discussion highlighted three approaches to the question of art and science in cartography: cartography as an art or as a science, cartography as some mixture of art and science, and art and science as inadequate bases for understanding cartography. In nearly all the responses, the art/science dualism was uncritically accepted as a means of thinking about and understanding cartography. My goal in relating this discussion is to reveal the different ways that notions of art and science shape the current understanding of cartography.
Certain respondents to Huet's question see the technology of modern cartography as destroying the art of cartography: "Its a shame that it is often practiced by the artless" noted Ted Samsel (Samsel 1994). Sonny Parafina criticizes the "science" of cartography by arguing that cartography is an art "because it takes skill to present information graphically" ... and ... "this skill or talent cannot be reduced to a set of algorithms" (Parafina 1994). Edward Tufte's books on information graphics are then cited as evidence that cartography is surely an art (Tufte 1983, 1992).
On the other hand, some discussants excised the art from cartography. Steve King argues that "cartography is the technology of making maps," which consists of surveying and drafting (King 1994). Marinus Groenveld argues that
...generally a map is produced with the purpose of showing people where things are on the surface of the earth or within a geographical location...that is not art. If you produce a pretty map that fulfills that purpose...then you have a pretty map...the work that has gone into making the map "pretty" and the resulting artwork on the map I guess could be considered art...but cartography itself is by definition simply map-drawing. If it doesn't fulfill it's primary purpose it may be art but then it is not longer cartography (Groenveld 1994).
However, most of the responses to the query about "cartography as an art" regard it as both art and science. Bruce Davis sees the art of cartography in the "need for subjective judgement, particularly in presentation" and the science of cartography in its "standards and specific methods of doing some things" (Davis 1994). Steve Tomlinson sees the art of cartography as the "design and presentation of ... information" and the science of cartography as the "accurate relative representation of the mapped features using surveying/remote sensing techniques" (Tomlinson 1994). Steven Garner argues that there is a geographical dimension to the art/science dualism in cartography:
Traditionally cartography is both an art and a science. In North America we tend to emphasize the science side while European cartographers emphasize the artistic side of the discipline (i.e. Swiss maps). The argument rages as to which is better (Garner 1994).
Jim Petch responded to all the above statements, arguing that
There is a problem with this issue...the issue is ... false. The question is a waste of time. The science-art debate is ... sterile. The proper debate is about ideas (Petch 1994).
The above discussion reveals three basic approaches to the art/science issue in cartography. First are the respondents who understand the art and science of cartography as distinct and polarized opposites. Second are the respondents who understand the art and science of cartography as coexisting but serving different functions. And third are the respondents who question the viability of the art/science dualism as a means of understanding cartography.
The GIS-L episode reveals that regardless of changing definitions, the issue of art and science and their relation to cartography still resonates in the minds of cartographers. The issue regularly resurfaces during theoretical discussions about cartography and its history, or when the issue of map design is raised, or when the role of creativity and aesthetics in cartography is debated. However, while the discussion begins with cartography as an art and a science, art and science often end up at odds with each other. The perspective of cartography as an art is used to criticize attempts to automate map design, or to emphasize the creative, intuitive, and subjective nature of maps which makes them so "magical." The perspective of cartography as a science is used to criticise the subjective and "mapping as a craft" view of cartography, or to emphasize and justify the "rational" methodologies which have allowed cartographers to construct increasingly accurate and "objective" representations of the "real world," or to reveal map design rules which can be used to optimize the utility of maps. Thus it is often the tension between the polarized dualism of art and science which ends up shaping our thinking about cartography and mapping: cartography is an art because it is not a science and cartography is a science because it is not an art. Such a negative and circular way of thinking about cartography is problematic.
In order to understand such limitations and problems of the art/science dualism, it is important to examine the different ways the dualism has been and is used. This examination requires a discussion of the general issue of dualisms and some of their problems, as well as a discussion of the particular manner in which the art/science dualism has structured thinking about cartography and information graphics in general. My intent is to provoke thought about the limitations of uncritically using the dualism of art and science as a means of understanding cartography, aimed particularly at those who hope to reexamine the role and function of design and aesthetics in cartography.
Dualisms in Geography and Cartography
The use of dualisms in geography and cartography is common, and our thinking is crowded with them: theory/empirics, objective/subjective, unique/general, global/local, rural/urban, and, of course, art/science. Such dualisms are certainly useful as a means of thinking and conceptualizing but become problematical when they are used uncritically. Andrew Sayer has discussed the problem of an uncritical dependence on particular dualisms in geography (Sayer 1991). There has been much commentary on the tendency for Western thought to be structured by binary oppositions. Yet
...although binary oppositions such as new/old or north/south are the simplest, most minimal, way of registering differentiation, it would be surprising if everything in the world also conveniently happened to be two-sided and hence susceptible to analysis purely or largely in terms of dualistic conceptual systems (Sayer 1991 p. 285).
The problem is that "what impresses us about such thinking may have more to do with its simplicity and symmetry than its ability to interpret the world" (Sayer 1991 p. 284). Sayer's point is not that dualistic thinking should be avoided, but that we should use dualisms - such as art and science - critically, paying close attention to the manner in which they structure how we think, interpret, and understand the world.
Brian Harley has discussed the dualisms which have structured our understanding of the history of cartography (Harley 1989a). This "discourse of opposites" is based on a series of assumed dualisms such as art/science, inaccurate/accurate, subjective/objective, and manual/machine. Harley argues that the uncritical acceptance of such dualisms is detrimental to a viable history of cartography and that the historical origins of such dualisms should be carefully examined. Uncritical reliance on the simple dualism of art and science, thus, may be limiting our ability to make sense out of what cartography has been (its history) and what it is becoming. With a sense of the problem of uncritical dualistic thinking in mind, the next several sections examine the manner in which the art/science dualism has structured our understanding of maps and other information graphics.1
Art and Science in Cartography: Three Approaches
The relationship between art and science has been discussed in numerous contexts including monographs by Alpers (1983) and Hartal (1988), edited collections by Wechsler (1978) and Graubard (1988), and journals such as Leonardo, which is devoted to investigating the relations between the arts and the sciences. The focal point for this study will be the literature which addresses the relationship between art and science in cartography and other information graphics.
A review of pertinent literature reveals three basic approaches to the art/science question in cartography and information graphics which approximate the approaches taken by the GIS-L discussants. First, literature which implicitly assumes that such graphics are necessarily the outcome of a "scientific" process and that any "artistic" value they may have is separate from the science and of secondary importance. I call this the "Polarizing Theme." Second, literature which attempts to evaluate some difference between "art" and "science" that may help to establish what in a map or graphic can be considered "artistic" and what can be considered "scientific." As this inevitably ends up conceptualizing science as progressive and art as immutable (a-progressive), I call this the "Progressive Theme." The third approach, which eschews the art/science dualism, will be discussed in the last section of this paper. I argue that both the polarizing and progressive approaches to the question of art and science in cartography and information graphics are problematical upon close examination.
1. The "Polarizing Theme"
There are a number of textbooks on the topic of scientific illustration including Zbigniew Jastrzebski's Scientific Illustration: A Guide for the Beginning Artist (1985). Jastrzebski defines scientific illustration as an "art in the service of science," as visual "supportive material" for scientists, and as a "visual explanation of scientific studies and findings" (p. 5). It is clear that Jastrzebski considers information graphics to be products of an objective, scientific process. Such graphics are "primarily produced for the scientist and his research, secondarily for the whole scientific community. [They are] definitely not prepared for the artist himself or the accidental viewer" (Jastrzebski 1985 p. 5).
A similar view is expressed by Geoffrey Lapage in his Art and the Scientist (1961). Lapage defines the "artistic qualities" found in graphics used by scientists as "qualities added to a scientific illustration which are also found in works of art and make the scientific illustration valuable for its own sake. They are not essential to the scientific purpose for which the illustration was made" (Lapage 1961 p. 57).
Similar sentiments to those of Jastrzebski and Lapage can be found in the cartographic literature. A paper by Arpad Papp-Vary sums up a dominant viewpoint of the polarized nature of art and science in cartography:
The essential purpose of map-making has always been the creation of the most exact reflection of reality or the graphically true representation of space.... To achieve these aims, cartographers in the past made their maps to high artistic standards. To increase the aesthetic effect of their products, the titles and legends were surrounded with artistic figures and the map frame was also artistically drawn. The purpose of such artistic work was to help the recognition of reality on the basis of the use of the maps; at the same time, the attractive figures made the map readers interested in the map and its use. The basic purpose of the maps, however, was still to reflect reality as perfectly as was possible given the knowledge of the time. The scientific problems of the exact representation of the real world have always been the primary and determining factor in the process of map-making, while the artistic work has only been of secondary importance (Papp-Vary 1989 p. 106).
Jastrzebski, Lapage, and Papp-Vary understand information graphics as necessarily the end product of a "scientific" process with "artistic" and "aesthetic" value being strictly of secondary (if any) importance. Further, the "art" and the "science" in such graphics are distinct and can be easily separated. Science is objective and analytical, a reflection of reality. Art is subjective and intuitive, a reflection of subjective indulgence. Art and science are two different and separable things. Following a similar line of reasoning, the British Cartographic Society proposed a dual definition of cartography in 1989. The first definition of cartography is for the general public and "prospective" cartographers "not yet engaged in cartography" and defines cartography as the "art, science, and technology of making maps." The second definition, intended for practising cartographers, excises the art from cartography, defining it as "the science and technology of analyzing and interpreting geographic relationships, and communicating the results by means of maps" (Anonymous 1989 p. 4). While this distinction was later abandoned, the fact that it was considered illustrates how the "polarizing theme" can shape a general understanding of cartography.
All these views reiterate the polarizing view of art and science found in C.P. Snow's pervasive idea of the "Two Cultures:"
...the intellectual life of the whole of western society is increasingly split into two polar groups...Literary intellectuals at one pole - at the other scientists. ...In our society we have lost even the pretence of a common culture. Persons educated with the greatest intensity we know can no longer communicate with each other on the plane of their major intellectual concern (Snow 1964 p. 3-4, 60).
The assumptions about the polarity of art and science by Jastrzebski, Lapage, Papp-Vary, the British Cartographic Society, and Snow are simplifications which in essence do not consider the relations between art and science. Jastrzebski sees the art in scientific illustration but denies it a role; indeed in his text he hints at the possible corrupting influence of artistic expression on scientific graphics. Lapage appreciates the art more than Jastrzebski yet still sees the artistic aspect as independent and unnecessary to the scientific purpose of graphics made by or for scientists for their work. Papp-Vary argues, as Jastrzebski, that the art in the science of cartography is strictly of secondary importance. The British Cartographic Society formalizes the distinction by suggesting that practising cartographers need not engage in the "art" of cartography. All assume a polarizing model which simply skirts the more fundamental nature of their subject. "Art" is easily separated from "science" in an overly simplistic assumption about the dualistic nature of art and science.
2. The "Progressive Theme"
A second and more complex theme in the information graphics and cartography literature assumes that art and science necessarily exist together (albeit distinctly) in maps and information graphics, and seeks to evaluate some difference between "art" and "science" that may help to establish what can be considered "artistic" and what can be considered "scientific." The progressive theme consists of two arguments. The first concerns the different way that art and science treat their past, and how this can be used to differentiate the artistic from the scientific aspects of maps and other information graphics. The second argument suggests that art and science serve different functions in maps and information graphics.
2.1. Progressive Science and A-Progressive Art
Historian of science Thomas Kuhn articulates the different manner in which art and science deal with their past - what Kuhn calls the most obvious difference between art and science: "The past products of artistic activity are still vital parts of the artistic scene" while "in science new breakthrough do initiate the removal of suddenly outdated books and journals from their active position in a science library ... unlike art, science destroys its past" (Kuhn 1977 p. 345). David Knight links this idea of the progressive nature of science and the a-progressive nature of art to information graphics:
While the various (scientific) illustrations from the past continue to give pleasure, their usefulness tends to diminish with the passage of time because scientific language and the concerns associated with it changes, whether it be visual or ordinary language. If it is the work of a great artist, it may pass time's test and live on, passing into 'art' if it is no longer 'science,' rather than being a casualty of progress (Knight 1985 p. 124).
Such an argument is evident in the cartographic literature and is characterized by a belief in the progression in the "scientific" quality of maps from the "simple," "poor," and even "pathetic" maps of "primitive" peoples or of our predecessors up to the "accurate" and "objective" maps of today. This is particularly evident in older literature on the maps of "primitive" cultures such as the American Indian (Burland 1947/48), and of commentary on Medieval Mappaemundi (Beazley 1897). Old maps are interesting primarily because of their quaint "artistic" nature. Their "science," if there is any, is outdated and not important. Arthur Robinson notes that "[t]he older is a flat map or a globe map, the more likely it is to be called an art object" (Robinson 1989 p. 94). It is, then, the progressive nature of science which separates science from art. In an oft-repeated quote from his History of Cartography, Leo Bagrow wrote "This book ends where maps ceased to be works of art, the products of individual minds, and where craftsmanship was finally superseded by science and the machine; this came in the second half of the eighteenth century" (Bagrow 1964 p. 22). Bagrow's sentiments are common. Ronald Rees, in his discussion of the historical links between cartography and art, writes "until science claimed cartography, mapmaking and landscape painting were kindred activities, often performed by the same hand" (Rees 1980 p. 60). Bagrow and Rees, while obviously concerned with the "art" in cartography, certainly do art no service by implying that it has a static nature, so easily superseded and snuffed out by progressive science.
Robert Root-Bernstein has evaluated the differentiation of art from science by the way each treats their past and found it wanting:
...traditions do not change any more rapidly in science than elsewhere. Remaining adherents to passé theories are not necessarily disbarred from the profession. Science does not automatically reject its past for innovation. On the contrary, science is a selective process that weeds out bad ideas, irreproducible results, and incorrect problem solutions to leave fruitful ideas, reproducible results, and useful problem solutions. Science...is selective in what it retains of its past, building upon that which has been most useful (Root-Bernstein 1984 p. 111).
Indeed, there instead seems to be more similarity than difference in the way that art and science treat their past:
I suggest that artists reject earlier traditions of art for the same reason that scientists reject earlier traditions of science: the old problems are solved; new ones await. Certainly an artist could choose to paint like Rembrandt just as a scientist could choose to perform experiments on falling bodies similar to those conducted by Galileo. But painting like Rembrandt tells us no more about perception and solves no new problems of the use of light than Rembrandt already did, just as more data on falling bodies reveal nothing new about the nature of motion. To be successful, the artist, like the scientist, must introduce into his discipline new methods, new perceptions, or new phenomena that raise new problems for colleagues to address (Root-Bernstein 1984 p. 111).
A similar argument has been used to critique the manner in which cartography has relied on the assumption that the "art" in old maps is immutable while the science is outdated, superceded by more accurate and objective knowledge. Much of this critique developed in tandem with the reconceptualization of the history of cartography (Blakemore and Harley 1980, Harley and Woodward 1987). Traditional histories of cartography are criticized for their tendency to conceptualize the development of cartography as progressive and teleological while simultaneously ignoring a complex array of cultural, social, economic, and political issues. An alternative view of cartography can be formulated, where cartography is "not a neutral activity divorced from the power relations of any human society, past or present [and] there is no single nor necessarily best way in which to represent either the social or physical worlds" (Edney 1993 p. 54). Edney argues for understanding cartography as "composed of a number of modes" defined as historically contingent "sets of cultural social, and technological relations which define cartographic practices and which determine the character of cartographic information" (1993 p. 54). These contingent modes are related to the continual emergence of new problems, methods, and phenomena which drive developments in both "art" and "science" as discussed by Root-Bernstein. From this it follows that "There is not hard and fast distinction between the 'art' and the 'science' of cartography; nor is it that 'cartography is both an art and a science..." (Edney 1993 p. 54). Instead, each "cartographic mode" is the result of particular historical (cultural, social, political) circumstances, and the distinction between what is defined as "art" and "science" varies from mode to mode. To impose a particular notion of art and science (defined by our particular late 20th century Western cartographic mode) on all maps ignores both historical and contemporary differences in the manner in which art and science are defined. It is, then, problematic to sustain the argument that science is progressive and art is a-progressive and, thus, to use such alleged differences to distinguish art from science in information graphics and maps.
2.2. The Differing Function of Art and Science in Information Graphics and Cartography
The second argument of the progressive approach suggests that art and science serve a different function in maps and information graphics. Kuhn has also argued that there is a major difference between painting and information graphics:
...paintings are end-products of artistic activity. They are the sorts of object which the painter aims to produce, and his reputation is a function of their appeal. The scientific illustrations, on the other hand, are at best by-products of scientific activity (Kuhn 1977 p. 342).
Thus, Kuhn sees a difference between ends - the visual product or language of an artist, and means - the visual product or language of a scientist. Kuhn sees something similar in the idea of the aesthetic and how it differs in art and in science:
...in the arts, the aesthetic is itself the goal of the work. In the sciences it is, at best, again a tool: a criterion of choice between theories which are in other respects comparable, or a guide to the imagination seeking a key to the solution of an intractable technical puzzle. Only if it unlocks the puzzle, only if the scientist's aesthetic turns out to coincide with nature's, does it play a role in the development of science. In the sciences the aesthetic is seldom an end in itself and never the primary one (Kuhn 1977 p. 342).
Kuhn does discern similarities in art and science that are important. For instance, they both must deal with technical problems which must be solved for the end product to be realized. Kuhn continues:
...the scientist, like the artist, is guided by aesthetic considerations and governed by established modes of perception. But an exclusive emphasis upon those parallels obscures a vital difference. Whatever the term "aesthetic" may mean, the artist's goal is the production of aesthetic objects; technical puzzles are what he must resolve in order to produce such objects. For the scientist, on the other hand, the solved technical puzzle is the goal, and the aesthetic is a tool for its attainment. Whether in the realm of products or of activities, what are ends for the artist are means for the scientist, and vice versa (Kuhn 1977 p. 343).
Similar sentiments are expressed by cartographers. Keates sees an immutable aesthetic element in mapping, an "art" which cannot be accounted for by progressive "science." The aesthetic and art of cartography is then used to critique what Keates sees as the over-dependence on the "science" of cartography. "I like to draw attention to the fact that there are realms of human experience which are not - and never will be - products of science; and although few of us can aspire to create them, we can - and should, enjoy them" (Keates 1984 p. 43). Yet this aesthetic element in cartography is distinct from science, and it is possible "for a map [to] be 'well designed' in a functional sense without creating anything of the aesthetic property we can sense in other things" (Keates 1984 p. 41). Art, in the end, serves a different function than science and need not be part of the successfully designed map.
Robinson argues along similar lines, noting that there are two types of "man-made things responded to aesthetically": the fine arts, which create "non-purposeful art objects" such as paintings and sculptures, which are appreciated for their own sake as ends in and of themselves; and the useful arts, which create purposeful objects, such as maps, which are "enjoyed as a means to something else" (Robinson 1989 p. 92). In this case, the map is essentially a useful and purposeful object, employing certain "artistic means" to achieve an end which must be "something else," a something else which is, logically, non-artistic; ie., scientific. The cartographer, then, uses artistic means to achieve a broader scientific end; the art in a map, thus, serves a different function than the science in a map.
Robert Root-Bernstein has also evaluated the differentiation of art from science by asserting their different functions. Kuhn argues that the "aesthetic" is the artist's goal and the scientist's tool. Thus, the function of art is to produce aesthetic ends which may involve solving "technical puzzles" while the function of science is to produce solutions to "technical puzzles" which may involve aesthetic means. Keates and Robinson place a similar argument in the realm of cartography. Such an assertion of functional difference in turn must logically support the view that there is a structural similarity in the products of each endeavor ‹ a scientist's monograph (explaining a solution to some scientific puzzle) is analogous to an artist's painting, for example. Both are ends. To this analogy, it seems, Kuhn would agree.
Yet Root-Bernstein finds fault with this formulation of the analogous relationship between art and science. Kuhn's belief in functional differences found in art and science, as represented by his conception of the differing role of aesthetics, is at odds with the practices of certain scientists. Henri Poincare has argued that
The scientist does not study nature because it is useful to do so. He studies it because he takes pleasure in it; and he takes pleasure in it because it is beautiful. If nature were not beautiful, it would not be worth knowing and life would not be worth living....I mean the intimate beauty which comes from the harmonious order of its parts and which a pure intelligence can grasp (quoted in Root-Bernstein 1984 p. 112).
And Alexander von Humboldt:
In the uniform plain bounded only by a distant horizon, where the lowly heather, the cistus, or waving grasses, deck the soil; on the ocean shore, where the waves, softly rippling over the beach, leave a track, green with the weeds of the sea; everywhere, the mind is penetrated by the same sense of the grandeur and vast expanse of nature, revealing to the soul, by a mysterious inspiration, the existence of laws that regulate the forces of the universe (Humboldt 1844 p. 25).
In both cases, art and science seem to serve similar functions. Further, similarities between scientific and artistic activity have been commented upon quite frequently. The journal Leonardo features articles by "artistic scientists" and "scientific artists." Indeed, the articles in Leonardo often conclude that attempts to polarize art and science are bound to end in confusion. Instead, the focus is on goals and ideals and methods shared by artists and scientists: both art and science serve similar functions. Art historian E.H. Gombrich has noted that
...the task of setting down a pictorial likeness on a flat surface bears a startling resemblance to the method used by scientists in arriving at a theoretical picture of the natural world. In representing the appearance of things, the artist does not simply trace an outline of their visual contours, but prepares instead a hypothetical construction to be matched and then modified in light of further experience. Through an alternating sequence of "makings and matchings" the artist gradually eliminates the discrepancies between what is seen and what is drawn. . . . (Such) "makings and matchings" of the artist correspond to the "conjectures and refutations" of the natural scientist (quoted in Miller 1983 p. 222).
There is a case to be made for functional similarity, and thus structural difference in art and science. Root-Bernstein concludes that "both scientists and artists are engaged in the common pursuit of new ways of perceiving and of controlling nature" and that this common pursuit "is mirrored in common methods" (Root-Bernstein 1984 p. 109). Thus, paintings are analogous, structurally, to experiments, art galleries to scientific meetings. Paintings are not, then, an end but instead are a means, like an experiment. Established theory, in science as well as art, "is simply an approximation to perceived reality that permits predictions to be made about the unknown." Physicists deal with point masses and curved space, chemists with energy and atoms ‹ "science, like art, has its perceptual conventions ‹ its approximately-but-not-quite-true models of the world" (Root-Bernstein 1984 p. 113). Art and science - however defined - serve similar functions.
Judith Wechsler has come to a conclusion similar to Root-Bernstein's. Wechsler argues that art and science both strive for "fit" ‹ the "most appropriate, evocative and correspondent expression of reality heretofore unarticulated and unperceived, but strongly sensed and actively probed" (Wechsler 1978 p. 1). Wechsler further argues that aesthetics guide the search for "fit" in both art and science:
Viewed as a way of knowing, aesthetics in science is concerned with the metaphorical and analogical relationship between reality and concepts, theories and models. The search in science for models that illuminate nature seems to parallel certain crucial processes in art, as Cyril Smith points out: they share a fundamental evocative quality (Wechsler 1978 p. 6).
Relating this back to the subject at hand, it is plausible that the function of "art" and of "science" in information graphics and cartography are more similar than different: both provide a structurally varying means to perceive, interpret, conjecture, make statements about and control nature. They provide different means to the same end. What "art" does and what "science" does is not really that different - it is more a matter of how one goes about the process, and how one goes about any process differs as much within the various "arts" and the various "sciences" as it does between them (Stafford 1994). Art and science are by no means the same thing; however, their similarities and complex interactions are such to preclude any attempt at strict segregation via either the polarizing or progressive approaches. Adhering to the presumed dualism of art and science, then, is problematical as a means of understanding cartography.
The limitations of the art/science dualism as a way of thinking about cartography and map design seem particularly evident in literature which ponders and critiques the state of cartographic design research. Petchenik (1983), Keates (1984, 1993), and Wood (1993) have all written soul searching critiques of the "science" of cartographic design, and all found it lacking. Commenting on the era of "scientific" map design research (roughly late 1960s and 1970s), Keates states that
Looking back at this period, which lasted for little more than a decade, it is clear that it yielded virtually nothing substantive about cartographic design in the sense of influencing the maps which went on being produced. But the attention given to cartographic design was stimulating, and at least the questions raised were interesting. (Keates 1993 p. 200-201).
Such ambivalent critiques are seldom commented upon by practitioners of "scientific" cartographic design, but Dobson (1985) responded to Petchenik's (1983) critique of the science of cartographic design. Dobson in essence argues that such critiques are "misguided" and "naive" and not very scientific in their critique of the science of cartographic design (pp. 27-28). My reading of this debate suggests that "art" (aesthetics, intuition, creativity) is used to clobber science, and science (rationality, reason, analytical, objective) is used to clobber art. Art and science, whatever they may be, are in a sense both demeaned by such comparisons - they are, as Leo Steinberg argues, "yoked" through metaphoric and analogical negations of each other (Steinberg 1988).
In attempting to single out the artistic or scientific aspects of maps and information graphics based on the simple (polarizing or progressive) art/science dualism, then, cartographers and other individuals interested in maps and information graphics may be asking the wrong questions. I argue that assuming art and science as polarized oppositions to each other is problematical, that assuming art to be a-progressive and science progressive is problematical, and that assuming art and science serve differing functions in the context of information graphics and cartography is problematical. If cartographers are to reconsider the role of aesthetics, design, and creativity in cartography, as called for by Keates and others, then we are going to have to discard the problematic assumptions of the art/science dualism, turning instead to inquiries into the different ways that what we vaguely define as art and science together shape maps and cartography within different contexts and for different purposes.
3. Cartography as Not Art and Science
Debates over art and science in cartography seem to shed little light on the current dynamic state of cartography in the context of "postmodern" deconstruction, hypermedia, cognitive psychology, semiotics, geographical information systems, and visualization. While seemingly disparate and divergent trends, these developments all point towards an understanding of cartography as a process.
Brian Harley and Denis Wood have "deconstructed" and critiqued many of what they see as the fundamental assumptions underpinning cartography and have reconceptualized cartography as an actor in the process of social and cultural formation and transformation (Harley 1987, Harley 1989b, Wood 1992). Within this context, the continued use of the art/science dualism by cartographers is understood as a reflection of scient-ism - the misguided construction of a progressive, objective, and de-contextualized history of cartography (Harley 1989a, Edney 1993). As an alternative to such "scientism" Rundstrom has argued for the adoption of a more humane and process-oriented means of understanding cartography:
Process cartography consists of two concentric ideas. It situates the map artifact within the mapmaking process, and it places the entire mapmaking process within the context of intracultural and intercultural dialogues occurring over a much longer span of time (Rundstrom 1991 p. 6).
This process-oriented approach to cartography is revealed in studies such as those by Wood and Beck (1989), McCleary (1990), and Pandya (1990). Such a perspective allows cartographers to examine how maps function in different historical, cultural, social, and political contexts. We can reexamine materials such as Harrison's Look at the World (1944) and Raisz's Atlas of Global Geography (1944) not as quaint old maps but as spectacular new ways of seeing and understanding the world given the particular circumstances of the first global war (Hendrikson 1975, Cosgrove 1994). These new ways of seeing depend on a functional synthesis of "art" and "science" - a synthesis which itself questions the need and possibility of separating "art" from "science."
In a similar vein and from a critical and technological perspective, hypermedia is touted as a means of reconstructing the deconstructed; an everchangable, non-linear, unbounded, decenterable, and hyper-relational process for organizing and reorganizing knowledge (Krygier 1994). Remarkably, the most fertile work in hypermedia is developing from the convergence of work by literary theorists and computer scientists (Landow 1992). That the ideas and goals of literary "artists" and computer "scientists" should converge to create one of the more exciting theoretical and technological developments of the late 20th century lends support to the inadequacy of the art/science dualism.
Developments in cognitive psychology and semiotics are being applied to maps in order to explicate the processes of imbuing visual spatial representation with meaning and of understanding spatial information represented in visual form (MacEachren 1995). Alternative conceptualizations of cartography have attempted to expand it from the illustration and communication-orientation it adopted in the 1950s, 60s and 70s. Many of these approaches have been inspired by the process oriented approaches to cartography, information graphics, and statistics of Jacques Bertin (1981) and John Tukey (1977). Both GIS and geographic visualization at their core are concerned with the process of geographical analysis and understanding. Maps have an obvious role to play in this process, particularly as exploratory methods (Monmonier and MacEachren 1992, MacEachren and Taylor 1994).
Cognitive, semiologic, and geographical analysis components in GIS have all converged in the context of scientific visualization and geographic visualization in geography (Hearnshaw and Unwin 1994). It should not be surprising that, as with hypermedia, the most exciting developments in visualization are driven by the convergence of the functionally similar work of "artists" and "scientists." The "yoking" of art and science has been surpassed by the idea of "renaissance teams" in the practice of scientific visualization. Donna Cox has described the process of artists and scientists working together on complex scientific visualizations (Cox 1990, 1991). The product of such collaborations (including the ubiquitous NCSA Storm Cloud Visualization) are not easily describable as either art or science. They would not exist without the input of both artists and scientists and are something more than just a simplistic mixture of art and science. However defined, art and science serve similar functions in attempting to envision and understand complex ideas, theories, and data. There is, then, something "emergent" which arises out of such collaborations. Accordingly, the dualism of art and science - via the polarizing or progressive themes - is not particularly useful as a means of understanding the products of such collaborations. Similar approaches can be found in cartography, where (carto)graphic designers work in tandem with scientists in shaping visualization methods for research and education (DiBiase et. al 1994, Krygier et. al 1995). Attention is thus shifted to the process of understanding and knowledge construction, the manner in which ideas are shaped and clarified, and the ways in which we come to know and re-know our world. How this occurs, rather than "is it science?" or "is it art?" becomes the focus.
I began this paper by noting that official definitions of cartography have at different times included and excluded the idea of cartography as an art and a science. I show that regardless of current definitions, the idea of art and science - however vaguely defined - still shape current discussions and thinking about cartography. The reasons for this are two-fold. The idea of cartography as an art and a science has an extensive history and is well established to the degree that it is difficult to not refer to it in our general discussions of cartography. Changing the official definition of cartography does not necessarily change the way we think about cartography. Yet it also seems that the persistence of the art/science dualism reveals a desire to reconsider the role of aesthetics, design, and visual expression in cartography. Practicing and academic cartographers with an interest in such issues tend to channel their discussions into the art/science dualism, in effect posturing themselves against the "science" of cartography which is seen to have dominated the last several decades of cartographic inquiry. While I am sympathetic to the desire to reconsider aesthetics, design, and visual expression in cartography, I am dispute of the utility of using the art/science dualism as a basis for this reconsideration.
I examine two ways that cartographers and others have defined the relation between art and science in the context of information graphics and cartography: the "polarizing" and "progressive" themes. I argue that both of these conceptualizations do not stand up under close scrutiny and that attempts to strictly differentiate art and science based on such conceptualizations are bound to end in confusion while simultaneously demeaning both art and science. I suggest that we consider the consequences of thinking without the problematic art/science dualism.
Trends such as postmodern deconstruction, hypermedia, cognitive psychology, semiotics, geographical information systems, and visualization all point to a process oriented means of understanding cartography. Visual methods such as cartography aid in this process of understanding and knowledge construction, in shaping and clarifying ideas, and in the different ways in which we come to know and re-know our world. Such a process is culturally, historically, socially, and politically contingent and ever evolving, producing new questions, ideas, and issues which continually confront us. I suggest that within this process we consider the function of art and science - however defined - to be similar, discarding the problematical reliance on the art/science dualism. As an alternative, I suggest that we examine particular instances where "art" and "science" have converged - episodes such as the global cartographies of the 1940s, the theoretically infused technology of hypermedia, and the renaissance teams of scientific visualization - to guide a reconsideration of the role of aesthetics, design, and visual expression in cartography.
1. I am using the category "information graphics" to include maps, topographic illustrations, graphs, diagrams, and other scientific illustrations and graphics. I use this generic term primarily to avoid awkward references to the various different forms of graphics discussed in the paper.
Acknowledgements: For constructive comments and criticism thanks to Cindy Brewer, David DiBiase, Matthew Edney, David Green, Nik Huffman, Alan MacEachren, Annie Newman, David Woodward, and two anonymous reviewers.
Alpers, S. 1983. The Art of Describing: Dutch Art in the Seventeenth Century. Chicago: University of Chicago Press.
Anonymous. 1989. "Cartographic Perspectives on the News." Cartographic Perspectives 1.
Anonymous. 1992. "Report of the ICA Executive Committee." ICA Newsletter 20 (reprinted in Cartography and Geographical Information Systems 20:3. pp. 187-195).
Beazley, C. 1897. The Dawn of Modern Geography. London: J. Murray.
Bagrow, L. 1964. History of Cartography. London: C.A. Watts. (revised edition).
Bertin, J. 1981. Graphics and Graphic Information Processing. Berlin: de Gruyter.
Blakemore, M., and J.B. Harley. 1980. "Concepts in the History of Cartography: A Review and Perspective." Cartographica 17:4 (Cartographica Monograph 26).
Burland, C. 1947/48. "American Indian Map Makers." Geographical Magazine 20. pp. 285-292.
Cosgrove, D. 1994. "Contested Global Visions." Annals of the Association of American Geographers 84:2. pp. 270-294.
Cox, D. 1990. "Scientific Visualization: Collaborating to Predict the Future." Educom Review 25:4.
Cox, D. 1991. "Collaborations in Art/Science: Renaissance Teams." Journal of Biocommunication 18:2.
Davis, B. 1994. Posting to GIS-L Thursday October 6.
DiBiase, D., et. al. 1994. "Multivariate Display of Geographic Data: Applications in Earth System Science." In: MacEachren, A. and D. Taylor, eds. 1994. Visualization in Modern Cartography. Oxford: Pergamon. pp. 287-312.
Dobson, M. 1985. "The Future of Perceptual Cartography." Cartographica 22:2. pp. 27-43.
Edney, M. 1993. "Cartography Without Progress: Reinterpreting the Nature and Historical Development of Mapmaking." Cartographica 30:2/3. pp. 54-68.
Freitag, U. 1993. "Five Selected Main Theoretical Issues Facing Cartography: I. Map Functions." Cartographica 30:4. pp. 1-6.
Garner, S. 1994. Posting to GIS-L Wednesday October 5.
Graubard, S., ed. 1988. Art and Science. Lanham MD: University Press of America.
Groenveld, M. 1994. Posting to GIS-L Friday October 7.
Harley, J.B. 1987. "Maps, Knowledge, and Power." In: D. Cosgrove and S. Daniels, eds. The Iconography of the Landscape: Essays on the Symbolic Representation, design and use of past environments. Cambridge: Cambridge University Press. pp. 277-312.
Harley, J.B. 1989a. "'The Myth of the Divide: Art, Science, and Text in the History of Cartography." Paper presented at the 13th International Conference on the History of Cartography, Amsterdam.
Harley, J.B. 1989b. "Deconstructing the Map." Cartographica, 26:2. pp. 1-20.
Harley, J.B. and D. Woodward, eds. 1987. The History of Cartography, Volume One: Cartography in Prehistoric, Ancient, and Medieval Europe and the Mediterranean. Chicago: University of Chicago Press.
Harrison, R. 1944. Look at the World: The Fortune Atlas for World Strategy. New York: Knopf.
Hartal, P. 1988. The Brush and the Compass: The Interface Dynamics of Art and Science. Lanham MD: University Press of America.
Hearnshaw, H. and Unwin, D., eds. 1994. Visualization in Geographical Information Systems. New York: Wiley.
Hendrikson, A. 1975. "The Map as an 'Idea': The Role of Cartographic Imagery During the Second World War." The American Cartographer 2:1. pp. 19-53.
Huet, T. 1994. Posting to GIS-L Wednesday October 5
Humboldt, A. 1844. Cosmos. New York: Harper and Brothers.
Jastrzebski, Z. 1985. Scientific Illustration: A Guide for the Beginning Artist. Englewood Cliffs: Prentice-Hall.
Keates, J. 1984. "The Cartographic Art." Cartographica 21:1. pp. 37-43.
Keates, J. 1993. "Some Reflections on Cartographic Design." The Cartographic Journal 30:2. pp. 199-202.
King, S. 1994. Posting to GIS-L Thursday October 6.
Knight, D. 1985. "Scientific Theory and Visual Language." In The Natural Sciences and the Arts. Stockholm: Almqvist and Wiksell International.
Krygier, J. 1994. "Geography and Hypermedia." manuscript.
Krygier, J., C. Reeves, D. DiBiase, J. Cupp. 1995. "Multimedia in Geographic Education: Design, Implementation, and Evaluation of Multimedia Resources for Geography and Earth Science Education." Paper to be presented at the 17th International Cartographic Association Conference, Barcelona. Copy available at http://www.ems.psu.edu/Earth2/E2Top.html
Kuhn, T. 1977. The Essential Tension. Chicago: University of Chicago Press.
Landow, G. 1992. Hypertext. Baltimore: Johns Hopkins University Press.
Lapage, G. 1961. Art and the Scientist. Bristol: John Wright and Sons Ltd.
MacEachren, A. 1995. How Maps Work. New York: Guilford.
MacEachren, A. and D. Taylor, eds. 1994. Visualization in Modern Cartography. Oxford: Pergamon.
Marcus, M., J. Olson, and R. Abler. 1992. "Humanism and Science in Geography." In: R. Abler, M. Marcus, and J. Olson. Geography's Inner Worlds. New Brunswick: Rutgers University Press. pp. 327-341.
McCleary, G. 1990. "Pursuing the Cheyenne: Mapping Tribes, Trails, the 1857 Expedition and the Battle of Solomon's Fork." Meridian 4. pp. 3-28.
Meynen, E. 1973. Multilingual Dictionary of Technical Terms in Cartography. Stuttgart: International Cartographic Association.
Miller, J. 1983. States of Mind. New York: Pantheon.
Monmonier, M. and A. MacEachren. 1992. Special Content: Geographic Visualization." Cartography and Geographical Information Systems 19:4.
Morris, J. 1982. "The Magic of Maps: The Art of Cartography." Unpublished MA Thesis, University of Hawaii.
Pandya, V. 1990. "Movement and Space: Andamanese Cartography." American Ethnologist 17. pp. 775-797.
Petch, J. 1994. Posting to GIS-L Friday October 14.
Papp-Vary, A. 1989. "The Science of Cartography." In: D. Rhind and D. Taylor. Cartography Past, Present and Future: A Festschrift for F. J. Ormeling. London: Elsevier Applied Science Publishers.
Parafina, S. 1994. Posting to GIS-L Wednesday October 5.
Petchenik, B. 1983. "A Mapmaker's Perspective on Map Design Research, 1950-80." In: D. Taylor, ed. Graphic Communication and Design in Contemporary Cartography. New York: Wiley. pp. 37-70.
Raisz, E. 1944. Atlas of Global Geography. New York: Harper and Brothers.
Rees, R. 1980. "Historical Links Between Cartography and Art." The Geographical Review. 70:1.
Robinson, A. 1989. "Cartography as an Art." In: D. Rhind and D. Taylor. Cartography Past, Present and Future: A Festschrift for F. J. Ormeling. London: Elsevier Applied Science Publishers.
Root-Bernstein, R. 1984. "On Paradigms and Revolutions in Science and Art: The Challenge of Interpretation." Art Journal Summer Issue.
Rundstrom, R. 1991. "Mapping, Postmodernism, Indigenous People and the Changing Direction of North American Cartography." Cartographica 28:2. pp. 1-12.
Samsel, T. 1994. Posting to GIS-L Wednesday October 5.
Sayer, A. 1991. "Behind the Locality Debate: Deconstructing Geography's Dualisms." Environment and Planning A 23. pp. 283-308.
Snow, C. 1964. The Two Cultures. Cambridge: Cambridge University Press.
Stafford, B. 1994. Artful Science: Enlightenment Entertainment and the Eclipse of Visual Education. Cambridge: MIT Press.
Steinberg, L. 1988. "Art and Science: Do They Need to be Yoked?" In: S. Graubard, ed. Art and Science. Lanham MD: University Press of America. pp. 1-16.
Tomlinson, S. 1994. Posting to GIS-L Thursday October 6.
Tufte, E. 1983. The Visual Display of Quantitative Information. Cheshire: Graphics Press
Tufte, E. 1992. Envisioning Information. Cheshire: Graphics Press.
Tukey, J. 1977. Exploratory Data Analysis. Reading: Addison-Wesley.
Wechsler, J., ed. 1978. On Aesthetics in Science. Cambridge: MIT Press.
Wood, D. 1992. The Power of Maps. New York: Guilford.
Wood, D. and R. Beck. 1989. "Janine Eber Maps London: Individual Dimensions of Cognitive Imagery." Journal of Environmental Psychology 9. pp. 1-26.
Wood, M. 1993. "The User's Response to Map Design." The Cartographic Journal 30:2. pp. 149-153.
Woodward, D. 1987. "Introduction." In Art and Cartography: Six Historical Essays. Chicago: University of Chicago Press.