Colour Vision. A Study of Cognitive Science and the Philosophy of Perception.
Recently there has been growing interest in a bouquet of problems commonly subsumed under the term 'colour'. This literature spans the natural, life and human sciences as well as philosophy and the humanities.(1) There appears to be consensus that 'interdisciplinarity' is a good thing - as indeed it is for those committed to the thesis of the unity of science. This thesis, which constitutes the metaphysics of modern science and much of Western philosophy, posits a deterministic, law-governed and potentially intelligible structure that pervades the material universe (Dupre, 1993). The virtue of interdisciplinarity is that it holds out the promise of commensurability and convergence.
Within this tradition, Thompson offers an extremely interesting contribution to the philosophy and science of colour. He overcomes the myopia that sometimes accompanies adherence to any particular discipline, and takes both science and philosophy seriously. Indeed, he claims on his final page that he is retreating from the sterility of discussions of colour impaled on the centuries old opposition between matter and mind. His goal is to introduce a middle term 'life'. If Colour Vision is read from back to front, then just as Pride and Prejudice and Sense and Sensibility cannot define but must exemplify their themes, so too Thompson cannot define but must exemplify 'life'. This question is, what difference does Thompson's 'life' make to scientific and philosophical problems of colour?
Not wishing to confuse the reader, I will state at the outset my own commitments. As a field anthropologist I am directly concerned with what might be called 'life'. Furthermore, I place myself in a philosophical tradition that regards empirical inquiry as providing the most convincing accounts of how things are. It is against these commitments that my own research has led to the conclusion that normal(2) human beings are not compelled by anything to 'see' colour. This does not mean that there are (normal) people who cannot use, or use only randomly, spectral differences to live in the world (as a relativist might assert); rather it means that people do not explicitly 'see' or 'notice' an attribute or aspect possessed by all objects and lights from which to abstract and remember what scientists call 'colour'. These considerations have not led me to conclude that Modern Man is more evolved in his categorization of colour than Primitive Man (cf Berlin & Kay 1969).(3) Rather it has led me to question the nostrum that 'colour' is an indubitable sense-datum.
Thompson's theory of colour vision
'Seeing colour' has been the central example of foundational epistemology from Descartes to the logical positivists. In this tradition 'the world of colour' supposedly orders and measures reality, and serves to 'anchor' exemplary protocol sentences or epistemological primitives. The world of colour is focused by Thompson by the metaphorical tool of 'colour spaces' (on which more below). In contrast to an objectivist, dispositional, subjectivist, relativist or anthropocentric realist account, his aim is to show how the world of colour is the outcome of a biological process that renders just one world visible: for all humans and all other creatures, for each creature at different times, and for any one creature at a particular time. Recognizing the species-specific variation of colour perception, his task is to show how all actual manifolds fit into one world. His metaphysical goal is to characterize the continuity of the world lying behind the phenomenon of colour; his scientific goal is to characterize the role of colour for the animal in its environment. He argues for the first in 'the comparative argument', and for the second in 'the ecological view'. Before he can propose these new positions, however, he must clear the ground of sterile arguments.
Philosophical discussions of colour are at an impasse due to the etiolation of the naturalistic doctrines of subjectivism and objectivism. Subjectivism is the doctrine that the brain extrudes colour onto the world, objectivism that colours are perceiver-independent, physical properties (e.g. surface spectral reflectancies). Thompson proposes an 'ecological' synthesis to develop a 'relational ontology' to provide 'a more genuine middle way' (p. 140). His proposal draws from neuroscience, computational vision and ecological psychology and can be summarized by the slogan 'ecological cognitive science'. Specialist approaches to (colour) perception are not to be superseded but rather placed in 'a framework for connecting ... [them] to a more general understanding of the animal as the proper subject of perception and action' (p. 303). Claiming primary affiliation to philosophy he echoes Kant: cognitive science without philosophy is blind; philosophy without cognitive science is empty (p. xiii).
Cognitive science is defined as an interdisciplinary venture concerning the scientific study of mind, wherein colour vision plays a pre-eminent role. Recent discussion of colour in neuroscience, cellular biology, molecular biology, psychophysics, cognitive psychology, linguistics, artificial intelligence and philosophy are reviewed. Although sensitive throughout to the needs of prior justificatory arguments (hence his philosophical turn), with regard to neurophysiology and psychophysics, in contrast, he suspends this potentially critical stance. With this bias in place, he directs his full ire at computationalism (objectivism)(4) and sensationalism (subjectivism), which conflict with his ecological approach. His new theory of 'ecological cognitive science' can link the internal structure of the animal with the environment of which that animal is the subject. With regard to human beings this argument may dissolve 'the philosophical thicket known as the mind-body problem' (p. 251).
The comparative argument
Thompson takes the following four properties as fundamental to the capacity of discriminating colour: (i) wavelength discrimination; (ii) metamerism(5) in additive colour mixture experiments; (iii) sensitivity to simultaneous and successive colour contrast; (iv) some sort of colour constancy response (i.e. simultaneous sensitivity to 'localized' wavelength distribution and to ambient light).
Colour sensitivity defined thus has been experimentally established for many species (insects, birds, fish, primates). Differences in colour vision are modelled in terms of differences in 'colour space '. Most of Thompson's discussion is limited to the 'receptoral colour space' as distinct from phenomenal, postreceptoral and several other colour spaces.(6) For bees, most humans and many other primates the colour space has three dimensions at the receptoral level and so is 'trichromatic', other kinds of animals are dichromats, and yet others (e.g. pigeons, carp) are tetrachromats or even pentachromats. (Lack of correlation between complexity of animal and colour space is explained in terms of 'evolutionary degeneracy' of mammals in general). The three dimensions of the colour space of a trichromat correspond with the possibility of finding an additive mixture of three well chosen coloured lights, for example a red, a blue and a yellow (three 'primaries'), to match (for the particular animal) the colour of any light.
What could it mean to be a tetrachromat? 'Seeing' in four dimensions is not a better way of' seeing' in three dimensions: 'it is simply different' (p. 152). The metameric equivalent classes of a trichromat require three degrees of freedom - of a tetrachromat, four degrees. Where the normal human trichromat has one non-spectral colour range (purple), tetrachromats are likely to have three non-spectral colour ranges; for example, mixtures of red and blue, green and UV (ultraviolet), and red and UV.(7) This makes cross-species comparisons difficult: as Thompson says, '...two colour spaces of dimensionalities n and n + 1 will be incommensurable in the sense that there is no way to map the kinds of distinctions available in the higher-dimensional space into the kinds of distinctions available in the lower-dimension space ... the nonspectral colour classes available to the tetrachromat clearly cannot be mapped into the single nonspectral colour class available to a trichromat...' (p. 156).
In contrast, between two kinds of trichromats, say human and honey bee, the number of primary colours will be the same, although their respective ranges of wavelength sensitivities usually differ (p. 156). According to Thompson, human and honey bee receptoral colour spaces are commensurable because the kinds of distinctions in the bee space can be mapped into the human space. The comparative argument across all species is buttressed with the thought-experiment of 'novel colours' for humans (tetrachromatic human females are claimed to have been found). Starting from the assumption that 'the geometry of the colour space is in some sense constitutive of colour', were there to be 'some resemblance route from our colour space to the novel colour space' then there would be merely 'variations' in geometry and not 'incommensurability' (p. 277). It remains rather vague, however, whether there could be such a route. The importance of this argument is that it would also provide continuity and thus commensurability between humans and, for example, pigeons (tetra- or pentachromats). Then the thesis that colour vision is the evolutionary solution for integrating a physically heterogeneous collection of distal stimuli into sets of visually salient equivalence classes, deployed in a variety of perceptual conditions, would be vindicated. It is the set of equivalence classes that governs each specific colour space and the hope that eventually we will understand how all colour spaces are commensurable, that holds the comparative argument together.
Despite concentrating the discussion at the receptoral level, Thompson insists that 'seeing colour' involves much more than simply receptoral-level activity as probed by spectral stimulus mixtures. Unfortunately for the comparative argument '[t]o close the gap between the receptoral and perceptual levels would require more psychophysical and physiological knowledge of the postreceptoral processes involved in tetrachromatic colour vision than is currently available' (p. 158).
The ecological view
'Ecological cognitive science' presents the reader with an oxymoron. Cognitive science and ecological psychology have tended to be opposing orientations in the psychological mainstream. Cognitive science - the scientific study of awareness, thought and mental organization, in terms of the computational or information-processing model of mind - tends to be concerned with the revelation of pre-existing form (structures and functions) in the head of the Cartesian and/or Humean subject. Ecological psychology, usually drawing inspiration from the work of J. J. Gibson, is concerned with the processes wherein and whereby form might be generated and held in place. For example, perception in ecological psychology involves the active pick-up of information over time. This involves continuing transformations of perceiver and perceived and of opportunities that are 'afforded' for action, as well as meaningfully mediating these processes for further action. The stasis of much cognitive science stands in sharp contrast to the dynamic reticulations (or structured structurings) of ecological psychology. The point of departure of cognitive science is the privileging of the traditional individual subject (Popper's methodological individualism; Fodor's methodological solipsism), that of ecological psychology, the agent-in-its-environment.(8)
How does Thompson go about reconciling these opposing orientations? The inspiration for Thompson's ecological view is the characterization of colour as 'Janus-faced', that is, it faces both world and perceiver - an 'interactional' idea that is not as new as he presents it as being.(9) Once invoked he can reconcile opposing camps of philosophers concerned with colour science. He proposes that rather than deal either with world or animal, an animal - environment ecosystem become the proper unit of investigation. Environment (not 'world') is understood ethologically as implicating the sensorimotor capacities and evolutionary history of animal life. The neo-Darwinian notion of 'adaptation' is carefully skirted and, following Levins & Lewontin (1985), he claims evolutionary history should be imagined as 'activity' and 'co-determination' (though he eschews their word 'construction', p. 219). Organisms are not the passive object of endogenous/exogenous forces but their own co-producers from one generation to the next.
Thompson's starting point is Gibson's rejection of representationalism (of which Newton and Locke are the avatars, the computationalists its current exponents). Following many commentators, Thompson finds Gibson suggestive but lacking. But in contrast to, for example, Noble (1993), Thompson augments the theory with an account of the internal organization of an animal's perceptuomotor system - hence his unquestioning stance towards the neurophysiology and psychophysics of colour. In looking for internal constraints, however, he fails to grasp the limitations and metaphysical implications of this move. As his argument progresses, ecological considerations drop out and all that is needed to close the gap between receptoral and perceptual levels is more knowledge of postreceptoral processes.
I can only conclude that Thompson is redefining the philosophical problems of colour on so narrow a base as to make them ever-more congenial to mainstream methodological assumptions and aims of cognitive science. What he is not doing is enlarging cognitive science with ecological considerations.
Thompson's introduction of the middle term 'life' is perhaps one vital step in the right direction. But how far does it take us? In the concluding sections of the comparative argument he points to the literature on cross-cultural colour categorization of Berlin & Kay (1969). He claims this work supports his own conclusion that neurophysiological internalism is unsatisfactory on both philosophical and scientific grounds because of the subjectivism implied about chromatic perceptual content (p. 213). Had he probed further, however, he would have found that Berlin & Kay are committed to the deeply contested thesis of linking propositions between colour vision, categorization and biology (conceived in straightforward neo-Darwinian terms).(10) If Thompson accepts this programme then I can only regard him as a sociobiologist all the way to the wall.(11)
Nonetheless, I find his comparative and ecological arguments congenial enough to take seriously. In what follows I apply them to the colour science system itself. I develop what I call a 'historiography' (as distinct from 'history') of colour science in order to distance myself from the whiggish teleology and universal, foundationalist legitimations that underpin the orthodox colour science story. My purpose is to chart the emergence of systems and practices and their specific functionings in power/knowledge regimes. These are systems of ordered procedures for the production, regulation, distribution, circulation and operation of statements about 'colour' which are linked in a circular fashion with the productive, sustaining and extending systems of the institutions and practices of science and technology.
A historiography of colour science
Just as Thompson argues that generations of organisms shape and are shaped by the opportunities for action in their environment, so too, it could be argued, colour science shapes and is shaped by the opportunities for action in the laboratory environment. I will argue in this section that colour can be regarded as a co-production of colour science and laboratory environment, while colour-in-the-world is a co-production of colour science, laboratory and the techno-scientific bases of raw materials and commodities. In this way the 'truth' of colour (philosophical blueprints, instrumental realizations, experimental procedures, standardized products) brings the objectification of colour into the world by mobilizing the deep resources of this techno-scientific grid (see also Galison, 1997). Once 'in the world' it is but a small step to misrecognize 'colour' as the face of nature herself, as the pre-established ground-plan of genetic and cosmic harmony.(12)
It is impossible to separate out either a pure philosophical lineage or a pure tradition of instrumental and experimental practice in the emergence of 'colour' over the past few centuries. Empiricism and rationalism interacted, both being variously translated into instrumentation, experimentation and mathematicization, which themselves interacted and fed back into philosophy. Broadly speaking, the Baconian dictum (1620) that axioms deriving from the senses rise by a gradual and unbroken ascent to arrive at the most general axioms of all, merged with the Cartesian observer (always already male) who was likened to the camera obscura (Ihde, 1995). Like other instruments (microscope, telescope), the camera obscura selected, focused and enhanced a two-dimensional image-field of coloured patches outside any living context. ' Mondrians',(13) aperture mode experiments and the Munsell system are latter-day embodiments of this picture theory of perception. Hence the logical positivists' dictum that the world of colour orders and measures reality (Lenneberg & Roberts 1956:30, following Mach, 1900).
On Descartes' account, the picture or image of the coloured world was not mimetic but a reconstruction in-thought, situated in the space of chromatic figuration (the original of 'the colour space'). Nature being inherently mechanical, the certitude of compositional design (God's design) was guaranteed by mathematical constraint. Truth in colour vision was to see chromatic form in mathematical terms (Judovitz, 1993). Current interest in colour algebras (Hubey, 1997; Sokolov, 1997), and Thompson's enchantment with colour space geometry, can be directly traced back to Descartes' idea that truth in vision is to see chromatic form in mathematical terms.
After Kant(14) the Cartesian account was translated by Helmholtz and countless others as energy originating in the environment transduced ('encoded') into varying activities in the nervous system, which corresponded to nothing resembling phenomenal or mimetic experience. A cerebral transformation enriched and re-presented the input to provide a form intelligible to the organism.
For 'the Senses' (in this case, colour) to be amenable to quantitative measurement required defining 'sensation' in terms of a functional dependence on stimulation (Danziger, 1990) and with this the concept of a 'stimulus' emerged. When measurement required a numerically graded stimulus series, Newton's spectrum (itself an instrumentally mediated picture) was ready to hand. Consequently a pre-theoretic verbal label ('this light/surface is red-orange') was substituted by numerical magnitude ('this light/surface has a dominant wavelength of 605 nanometers'). Thus, the folk wisdom that 'colour-is-wavelength' was born, and with it the conflation (in colour science) of coloured lights and surfaces.(15)
In this way instrumentally mediated measurement came to define the 'sensation' of colour. Once the graded stimulus series was in place more dimensions could be added: the chromatic aspect, abstracted as 'hue', was first supplemented with 'lightness' or 'brightness' (physical stimulus: light intensity), and then with 'saturation' (physical stimulus: colorimetric 'purity'). The emergence of this new three-dimensional stimulus invited the claim to the 'psychological reality' or 'naturalness' of three dimensions of colour: hue, brightness and saturation (see Jameson's, 1997, reassertion of this dogma; cf. Thompson's 'phenomenal colour space'). With this the basic ingredients - the recursive interactions of philosophy, instrumentation and experiment - were in place for the emergence of the modern colour science system.
The co-production of colour in and out of the laboratory
To make 'colour' the phenomenon is to define it in a specific way, under a particular description, by removing from the experimental setting everything but the realization of a purified 'stimulus' comprised of various wavelength distributions. Relegating 'context' to the periphery, the phenomenon so defined by contextless procedure is placed at the core. Standing behind these procedures are colour space models. The notion of a colour space is the powerful theoretico-mathematical device that ratifies the procedures, objectivizes the data and reinforces the production of colour as an isolated phenomenon. Networks of different levels of theory constitute the truth-producing apparatus which also produce the phenomenon. By mutual adjustment of phenomenon and frame, a purified laboratory phenomenon - colour - emerges and is declared 'true to' the world. The frame itself is true to prior framing devices: the construction and use of instruments, the dialectic of theories and practice.
The colour scientist's network of theories, models, data and the workings of instruments and apparatus mesh to give a snug fit with the decontextualized laboratory reality. This fit is the result of whatever is admitted from the-world-out-there being tailored to fit the laboratory world. As in all laboratory science, but most especially in vision science, observation and measurement are conducted with apparatus that has co-evolved in conjunction with the phenomenon as seen through these observations and measurements and concomitant modes of data analysis (Danziger, 1990; Galison, 1987; Hacking, 1992). As a result colour science explores the (changing) definition of the colour science system itself. What characterizes this science is not merely that it is a laboratory science, but that its creation, the manufactured phenomenon 'colour' is created on the basis of the mathematically defined chromatic model-in-thought.(16)
If this is grasped then it is no longer possible for Thompson to argue the unity of science thesis with respect to colour. The terms of that debate - from genes and ganglion cells, through 'categories', to regularities of the terrestrial world - are either absolutely irrelevant or must defend their claim to relevance in terms of philosophy (which Thompson fails to do). Colour science might then admit its own contingency and set its gendered hubris and cosmic/genetic schemes aside. In turn this might open up quite general questions about how objects-for-knowing generated by experiment relate to objects-for-knowing generated by theory, and about what might count as 'reality' (Heelan, 1988, p. 515).
What could explain the pragmatic success of colour science? One answer is that the environment is remade by paints and dyes, reproductive technologies, coloured lights, television, cinema, computer screens and so on, in such a way that it reproduces the theory of colour that provided the techno-material base of colour technology for the production of the coloured world. The world then becomes a quasi-laboratory in which it is neither 'truth' nor 'world' that explains what makes colour science work, but the remade world that mirrors what is made in the laboratory. Scientific colour is raised in the laboratory both mechanically, physically, chemically, digitally and as a result of the negotiation and interpretation of the relevant phenomena (Johnston, 1996). 'Co-production' of colour science and 'colour' therefore stresses that what is currently called 'colour' is inseparable from its processes of manufacture in the laboratory and pragmatic realization in 'the world'. It is this process of cumulative technological inheritance and transmission that Thompson's and kindred accounts confuse with the grand design of Nature.(17)
Conclusion: what kind of entity is a trichromat?
The core of Thompson's comparative argument rests on the commensurability between human and bee colour vision and the promissory note that more will be 'discovered' about the female(18) human tetrachromat and postreceptoral processes of other tetrachromats. Together these will explain how all colour spaces are connected. What kind of being is it that embodies and enacts the 'colour vision' shared by bee and human? I can only conclude it is an automaton, that creature of the cognitive science landscape that has prompted Dennett (Thompson's mentor) to say that the mechanist world view is here to stay (Dennett, 1981, p. 233). Suitably characterized, it is automaton 'colour vision' that will eventually unit dichromats, trichromats, tetrachromats and pentachromats.
Automaton' colour vision' may be an appropriate characterization of bee' colour vision'. But in what could be the commensurability of bee and human colour vision lie? Perhaps human infants in the global techno-scientific world are destined to become automaton 'colour see-ers' even before they possess selective vision. Born into a world of automaton 'colour', that kind of 'colour' is what they are guided to 'see'. In due course to ask such an observer what the common sense norm of colour is, is precisely to ask for the conciliance with automaton and bee 'colour vision'.
Contra Thompson, I am sympathetic to a sufficiently 'thick' description that would show that for all its sophisticated quantitative methods and boundless hope for the future, colour science is the co-production of theory and practice (philosophy, instrumentation, experimentation, negotiation, interpretation) in the age of automaton 'colour'.
Where does this leave Thompson's tertium quid - ' life'? His account of automaton vision leaves only two options: either 'life' is the spark of Frankenstein's machine or it is the Divine spark itself. I leave it to the reader to decide what difference this makes to Thompson's scientific and philosophical account of colour.
BARBARA SAUNDERS (Departments of Philosophy and Anthropology, University of Leuven, Belgium)
1 See for example, Byrne & Hilbert (1997), Davidoff (1991), Gage (1993), Hardin (1988), Hardin & Maffi (1997), Hilbert (1987), Lamb & Bourriau (1995), MacLaury (1997), Maund (1995) and Westphal (1987).
2 'Normal' in this limited context is a measure of people according to 'trichromacy' in respect to 'visual pigments'.
3 For discussion of Berlin & Kay see Saunders (1992, 1995; Saunders & van Brakel, 1997).
4 Computationalism is 'objectivism' because it considers colour to be the result of a computation starting from surface reflectances.
5 Metamers are considered the spectrally different radiations that produce' the same colour' under' the same viewing conditions' for a particular creature.
6 What the relation is between these subordinate colour spaces is not directly addressed.
7 This description is a paraphrase of Thompson's quote of Burckhardt (1989). If Thompson's worries about incommensurability make any sense, then in his own terms, it is incoherent to use phrases like 'mixtures of red and blue' with respect to tetrachromats.
8 This is not to deny that recent developments in cognitive science and artificial intelligence referring to 'situated agents' or 'artificial life' may have started to blur the distinction. Thompson understands cognitive science in the more traditional sense, failing to do anything much with his ecological stance.
9 This kind of dialectic is found predominantly in the German tradition; for example, in Schelling, Hegel, Marx, as well as in early religious educationists (Froebel) and art historians (Panofsky).
10 Saunders & van Brakel (1997) summarize the criticisms.
11 'Sociobiology' here refers to the position staked out by Lumsden & Wilson's 'epigenetic rules' (1981) and Barkow, Cosmides & Tooby 'adapted mind' (1991).
12 See Shepard (1991) and Nathans (1989).
13 Mondrians are used in experiments to test and further develop Land's retinex theory (cf. Land, 1986). Though different in aims and purposes, nonetheless they are like the Munsell system in being a culturally constituted measuring device.
14 Hume (1888, p. 632) had already claimed that '...the understanding corrects the appearances of the senses....'
15 Many confusions were thrown up by the conflation of lights and surfaces in arriving at a definition of colour that would satisfy the different phenomena and practical goals of physicists and psychologists. Eventually committee consensus carefully delimited an aspect of light, that in turn was interpreted as a physiological response to radiant energy. The definition reads: 'Color consists of the characteristics of light other than spatial and temporal inhomogeneities; light being that aspect of radiant energy of which a human observer is aware through the visual sensations which arise from the stimulation of the retina of the eye' (Colorimetry Committee, 1953, quoted by Johnston, 1996, p. 412).
16 Colour science should not be thought an isolated example, however. Danziger (1990) amongst others regards the whole of psychology to be exploring the (changing) definition of the psychology system itself.
17 Cornford's (1935) suggestion that space and time cannot be classified as realities of nature but as frameworks of our own making, could be extended to colour. As space and time have 'turned out to be less like steel than india-rubber', so colour too could be regarded as an 'arbitrary figment of the human brain' as 'only one of many alternatives agreeable to nature' (p. 216).
18 Those informed by feminist arguments will be alert to the reinscription of 'sex' (fixity of gender identification and amplification of gender coherence) as the biological norm.
Bacon, F. (1620). Aphorism xix. Novum Organum.
Barkow, J., Cosmides L. & Tooby, J. (1991). The Adapted Mind. Evolutionary Psychology and the Generation of Culture. New York: Oxford University Press.
Berlin, B. & Kay, P. (1969). Basic Color Terms. Their Universality and Evolution. Berkeley, CA/Oxford: University of California Press (1991).
Burckhardt, D. (1989). UV vision: A bird's eye view of feathers. Journal of Comparative Physiology, 164: 787-796.
Byrne, A. & Hilbert, D. R. (Eds) (1997). Readings on Color, vols. I and II. Cambridge, MA: MIT Press.
Cornford, F. M. (1935). The invention of space. In Essays in Honour of Gilbert Murray. London: Allen & Unwin.
Danziger, K. (1990). Constructing the Subject. Historical Origins of Psychological Research. Cambridge: Cambridge University Press.
Davidoff, J. (1991). Cognition through Color. Cambridge, MA: MIT Press.
Dennett, D. (1981). Brainstorms. Cambridge, MA: MIT Press.
Dupre, J. (1993). The Disorder of Things. Metaphysical Foundations of the Disunity of Science. Cambridge, MA: Harvard University Press.
Gage, J. (1993). Colour and Culture. Practice and Meaning from Antiquity to Abstraction. London: Thames & Hudson.
Galison, P. (1987). How Experiments End. Chicago, IL/London: University of Chicago Press.
Galison, P. (1997). Image and Logic. A Material Culture of Microphysics. Chicago, IL: University of Chicago Press.
Hacking, I. (1992). The self-vindication of the laboratory sciences. In A. Pickering (Ed.), Science as Practice and Culture. Chicago, IL/London: University of Chicago Press.
Hardin, C. L. (1988). Color for Philosophers. Unweaving the Rainbow. Indianapolis, IN/Cambridge: Hackett.
Hardin, C. L. & Maffi L. (Eds) (1997). Color Categories in Thought and Language. Cambridge: Cambridge University Press.
Heelan, P. (1988). Experiment and theory: Constitution and reality. Journal of Philosophy, 85, 515-524.
Hilbert, D. R. (1987). Color and Color Perception. A Study in Anthropocentric Realism. Stanford, CA: Center for the Study of Language and Information.
Hubey, H. M. (1997). Logic, physics, physiology, and the topology of color. Behavioral and Brain Sciences, 20, 191-194.
Hume, D. (1888). In A Treatise on Human Nature. L. A. Selby-Bigge (Ed.), Oxford: Oxford University Press.
Ihde, D. (1995). Image technologies and traditional culture. In A. Feenberg & A. Hannay (Eds), Technology and the Politics of Knowledge. Bloomington, IN: University of Indiana Press.
Jameson, K. (1997). What Saunders and van Brakel chose to ignore in color and cognition research. Behavioral and Brain Sciences, 20, 195-196.
Johnston, S. F. (1996). The construction of colorimetry by committee. Science in Context, 9, 387-420.
Judovitz, D. (1993). Vision, representation, and technology in Descartes. In D. M. Levin (Ed.), Modernity and the Hegemony of Vision, pp. 63-86. Berkeley, CA: University of California Press.
Lamb, T. & Bourriau, J. (1995). Colour, Art and Science, Cambridge: Cambridge University Press.
Land, E. H. (1986). Recent advances in retinex theory. Vision Research, 26, 7-21.
Lenneberg, E. H. & Roberts, J. (1956). The Language of Experience: A Study in Methodology. Memoir 13, Indiana Publications in Anthropology and Linguistics, Baltimore, MD: Waverly Press.
Levins, R. & Lewontin, R. (1985). The organism as the subject and object of evolution. In R. Levins & R. Lewontin (Eds), The Dialectical Biologist. Cambridge, MA: Harvard University Press.
Lorenz, K. (1975). Kant's doctrine of the a priori in the light of contemporary biology. In R. I. Evans (Ed.), Konrad Lorenz. The Man and His Ideas. New York/London: Harcourt Brace Jovanovich. Reprinted from General Systems. Yearbook of the Society for General Systems Research (Eds L. von Bertalanffy & A. Rapoport), vol. VII, New York: Society for General Systems Research. First published (1941) Blatter fur Deutsche Philosophie, 15, 94-125.
Lumsden, C. & Wilson, E. (1981). Genes, Mind, and Culture. The Coevolutionary Process. Cambridge, MA: Harvard University Press.
Mach, E. (1900). Die Analyse der Empfindungen und das Verhaltnis des Physischen zum Psychischen. Jena: Gustav Fischer, 1919.
MacLaury, R. (1997). Color and Cognition in Mesoamerica. Constructing Categories as Vantages. Austin, TX: University of Texas Press.
Maund, B. (1995). Colours. Their Nature and Representation. Cambridge: Cambridge University Press.
Nathans, J. (1989). The genes for color vision. Scientific American, February, 42-49.
Noble, W. (1993). Meaning and the 'discursive ecology': Further to the debate on ecological perceptual theory. Journal for the Theory of Social Behaviour, 23, 375-398.
Saunders, B. A. C. (1992). The Invention of Basic Colour Terms. Utrecht: ISOR.
Saunders, B. A. C. (1995). Disinterring basic color terms: A study in the mystique of cognitivism. History of the Human Sciences, 8 (4), 19-38.
Saunders, B. A. C. & van Brakel, J. (1997). Are there non-trivial constraints on colour categorisation? Behavioral and Brain Sciences, 20, 167-232.
Shepard, R. (1991). The perceptual organization of colors: An adaptation to regularities of the terrestrial world? In J. Barkow, L. Cosmides & J. Tooby (Eds), The Adapted Mind. Evolutionary Psychology and the Generation of Culture, pp. 495-532. New York: Oxford University Press.
Sokolov, E. N. (1997). Four-dimensional colour space. Behavioral and Brain Sciences, 20, 207-208.
Westphal, J. (1987). Colour: Some Philosophical Problems from Wittgenstein. Oxford: Blackwell.
|Printer friendly Cite/link Email Feedback|
|Publication:||British Journal of Psychology|
|Article Type:||Book Review|
|Date:||Nov 1, 1998|
|Previous Article:||Perceptions of risk in motorcyclists: unrealistic optimism, relative realism and predictions of behaviour.|
|Next Article:||What is colour?|