Contrary to argument that unique hues are undemonstrated, the World Color Survey shows that speakers of more than 100 minor and tribal languages focus color categories predominantly on 4 of the 40 hue columns of the ethnographic Munsell array. The pattern is not conditioned by saturation levels or other arbitrary structures among the color chips, nor is Western influence likely to be the cause. Moreover, all evidence suggests that color cognition is autonomous despite the connotations and polysemies of color terms.

Saunders & van Brakel's criticism of Berlin & Kay's methodology misunderstands the fact that scientific hypotheses are tested by generating new, replicable data with novel explanatory power. Thus, although Berlin and Kay studied differences in colour words between languages, the same patterns are also present in colour word usage within languages, in a range of literary and other textual databases.

Saunders & van Brakel's analysis of the phenomenal categorization and subsequent experimental research in unique hues fails to include contemporary methodological improvements. Alternative strategies are offered from the author's research that rely less on language and world knowledge and provide strong evidence for the general theoretical constructs of elemental hue, nonbasic, and basic color terms.

Two points are made on the basis of (mainly) the cross- cultural psychological record. The first is that cross-cultural data indicate at least weak, nontrivial constraints on colour classification. The second is that exceptions to cross-cultural regularities as described by Saunders & van Brakel are compatible with the view that constraints on colour categories are probabilistic rather than deterministic.

The longevity of the Berlin and Kay theses results from the way in which they were formulated, contrasted with extreme relativism. Saunders & van Brakel need not reject colour opponency to reject universal colour categories. Colour opponency does not manifest itself in language, even when dealing directly with spectral colours.

I take issue with Saunders & van Brakel's claim that neural processes play no interesting role in determining color categorizations. I distinguish an aspect of color categorization, namely, color discrimination, from other aspects. The law of trichromacy describes conditions under which physical properties cannot be discriminated in terms of color. Trichromacy is explained by properties of neural processes.

Multidimensional scaling of subjective color differences has shown that color stimuli are located on a hypersphere in four-dimensional space. The semantic space of color names is isomorphic with perceptual color space. A spherical four-dimensional space revealed in monkeys and fish suggests the primacy of common neuronal basis.

Saunders & van Brakel prematurely reject the idea of nontrivial constraints in color nomenclature. Their claim that the universality of color naming is caused by Western contact and cultural dominance is inadequate because of the great variety of terminology systems still found in the world. The complex interactions of hue, brightness, and saturation can be studied rigorously. If we discard the standard models of color nomenclature because of some discrepancies and anomalies, we will not be able to explain the vast array of remaining data that is consistent with current theories.

Confusion concerning the issue of universality of colour categorization would greatly diminish if context regains its fundamental status in psychological research and we give up on the reductionist notion that biological universality implies behavioral universality.

Saunders and van Brakel propose that we rethink or reject much of current colour theory. Many of the problems they cite appear to call for further research rather than a radical rethinking of colour theory. The controversy described in this target article appears to be itself a case of “ekphrasis,” or something that does not exist.

The terms red, green, yellow, and blue are both necessary and sufficient to describe our chromatic experience. Their uniqueness and opponent nature is supported by evidence obtained under supra- threshold conditions, especially hue cancellation. These constraints are nontrivial. How some electrophysiologically identified mechanisms contribute to colour appearance is not known, but their complexities do not refute our experience of elemental hues.

Evidence from molecular genetics bolsters the claim that color is not a perceptuolinguistic and behavioral universal. Neurobiology continues to fill in many details about the flow of color information from photon reception to central processing in the brain. Humans have the most acute color vision in the biosphere because of natural selection and adaptation, not coincidence.

Sections R1 to R3 attempt to take the sting out of hostile commentaries. Sections R4 to R5 engage Berlin and Kay and the World Color Survey to correct the record. Section R6 begins the formulation of a new theory of colour as an engineering project with a technological developmental trajectory. It is recommended that the colour space be abandoned.

Starting from macroscopic and microscopic facts of cerebellar histology, we propose a new functional interpretation that may elucidate the role of the cerebellum in movement control. The idea is that the cerebellum is a large collection of individual lines (Eccles's “beams”: Eccles et al. 1967a) that respond specifically to certain sequences of events in the input and in turn produce sequences of signals in the output. We believe that the sequence-in/sequence-out mode of operation is as typical for the cerebellar cortex as the transformation of sets into sets of active neurons is typical for the cerebral cortex, and that both the histological differences between the two and their reciprocal functional interactions become understandable in the light of this dichotomy. The response of Purkinje cells to sequences of stimuli in the mossy fiber system was shown experimentally by Heck on surviving slices of rat and guinea pig cerebellum. Sequential activation of a row of eleven stimulating electrodes in the granular layer, imitating a “movement” of the stimuli along the folium, produces a powerful volley in the parallel fibers that strongly excites Purkinje cells, as evidenced by intracellular recording. The volley, or “tidal wave,” has maximal amplitude when the stimulus moves toward the recording site at the speed of conduction in parallel fibers, and much smaller amplitudes for lower or higher “velocities.” The succession of stimuli has no effect when they “move” in the opposite direction. Synchronous activation of the stimulus electrodes also had hardly any effect. We believe that the sequences of mossy fiber activation that normally produce this effect in the intact cerebellum are a combination of motor planning relayed to the cerebellum by the cerebral cortex, and information about ongoing movement, reaching the cerebellum from the spinal cord. The output elicited by the specific sequence to which a “beam” is tuned may well be a succession of well timed inhibitory volleys “sculpting” the motor sequences so as to adapt them to the complicated requirements of the physics of a multijointed system.

We offer a critique of the role of the parallel fiber beam as the unit of cerebellar computation, with the “tidal wave” as its mode of operation. Instead we see the microcomplex linking cerebellar cortex and nuclei as the unit, with parallel fibers providing the means to coordinate the effects of microcomplexes in modulating various motor pattern generators (MPGs).

We discuss anatomical data that may represent the substrate for very diverse input to a single folium, indirectly supporting the notion of a laterolateral beam spreading along the long axis of a folium. We also raise the question of whether the more linear body representation in somatosensory cortico-ponto-cerebellar pathways represent an adaptation to sequential processing of information from contiguous body parts.

Braitenberg et al.'s proposal, like most previous theories of cerebellar function (see Bower 1997, for review), is fundamentally based on the striking geometric relationship between parallel fibers and Purkinje cells. As in previous models, the current theory assumes that the activation of granule cells results in a “beam” of activated Purkinje cells, although it adds the new requirement that the granule cell layer itself have a particular spatial/temporal pattern of activation. I believe there is clear evidence that parallel fibers do not have the type of excitatory effects on Purkinje cells required by this theory.

Braitenberg et al.'s target article presents the best current integration of anatomical and physiological data, and provides a qualitative description of cerebellar function in terms of the dynamics of processes based on the geometry of the cerebellar cortex. We compare the proposed model to our own quantitative model based on the concept of Purkinje unit.

Braitenberg et al.'s view that the cerebellum contributes “mechanical efficiency” to multijoint sequences of movement is too narrow to account adequately for results from new anatomical, neurobehavioral, and neuroimaging studies. A broader view is that the cerebellum modulates attention, sensory, motor, and other neural systems in order to accomplish its prime function, which is to learn to predict and prepare for imminent information acquisition, analysis, or action.

The main assumptions on which the tidal wave hypothesis rests will be questioned. First, since focal synchronous mossy fibre input is sufficient to ensure spread of activity along the parallel fibres, the tidal wave is redundant. Second, spatial and temporal characteristics of mossy fibre input make spatio-temporal sequences appropriate for setting up a tidal wave unlikely in the behaving animal.