Art Neuro Zeki
Art Neuro Zeki
Art Neuro Zeki
In his paper, Neural Concept Formation and Art: Dante, Michelangelo, Wagner
Semir Zeki writes we can trace the origins of art to a fundamental characteristic
of the brain, namely its capacity to form concepts (Zeki, 2002, p. 53). He pro-
poses that this capacity is itself the by-product of an essential characteristic of
the brain. That characteristic is abstraction, and is imposed upon the brain by one
of its chief functions, namely the acquisition of knowledge. (ibid., p. 53). Then,
centring his argument around the ideal of love, he claims that Dante, Michelan-
gelo and Wagner had created in their brains, he further asserts that none of the
three found that ideal in real life, and each was impelled in a different way to cre-
ate works of art in response to that gap (ibid., pp. 534).
After reading through the piece several times I find the speculative hypothesis
weak on data and scientifically unconvincing. This is not just a result of Zekis
ill-conceived definition of art. His theoretical view also puts aside the degree to
which developing and appreciating artworks involves a number of complex
operations. Indeed, perhaps the most noteworthy aspect of Zekis theory is the
degree to which it undermines his own premise that a real theory of art and aes-
thetics must be neurologically based. Instead, after asserting that his approach is
dictated by a truth that he believes to be axiomatic that all human activity is
dictated by the organization and laws of the brain: that therefore, there can be no
real theory of art and aesthetics unless neurologically based (Zeki, 2002, p. 54),
he presents a tautological and psychologically-based argument best summed up
by the phrase and then a miracle occurs.
abstraction in the visual brain. His discussion of abstraction and the visual brain
is then abstractly extended to discuss opera and literary work. His explanation for
the sweep of the argument that the kind of detail we would need for the argu-
ment to hold together can be found in his earlier work ignores two key points.
First, his earlier work is similarly focused on vision. In addition, as I have explained
elsewhere (Ione, 2000; 2001), and as others have reiterated (Levy, 2000; McMahon,
2000; Nash, 2000; Shortess, 2000), his earlier work, too, is speculative and removed
from the nuts and bolts of art practice and art appreciation. In summary, the theory
is both ill-formed in and of itself and does not apply to art practice and appreciation.
The conclusions he draws from his conjectures highlight two other major limi-
tations that are ingrained in his theory. One is, as he writes: How the brain
abstracts is only partially known, for simpler constructs only. (Zeki, 2002, p. 59)
It seems, with so little known, one should draw sweeping conclusions with more
care. The second limitation results from his decision to speak about art in terms
of ideals and universals. Coupling his view of ideals and universals with his view
of abstraction he gives us the impression we should assume the subsystems of the
brain are homologous (in the sense of employing similar principles) when we
discuss neural concept formation and art.1 I would argue that painting with a
visually-biased brush blurs neural distinctions that researchers know occur when
processing language, visual art and music. It would have been helpful to clarify
why he favours a homologous view, if he does, and to include some discussion
that related his visually-biased argument regarding abstraction to known pro-
cessing distinctions among visual art, music and literary works.
While it may well be, as he infers, that abstraction and knowledge-acquisition
are homologous insofar as visual art, music and language are concerned, we need
to know with more specificity how his view integrates empirical studies that have
demonstrated modularity including his own scientific research. Again, had he
discussed some of the work now available on the literary and musical brain, the
reader would be better equipped to understand how he comes to draw his conclu-
sions. We know that words and music are primarily processed in different parts of
the brain although the entire brain is used in this processing. It is also well
known that subtle language distinctions are evident in brain processing when we
compare spoken words to those written or heard. These kinds of distinctions have
been repeatedly recorded in basic studies. He does not say how particular cortical
areas and binding should be explored in regard to the neural concept formation.
Without some indication of precisely how the binding mechanism under discus-
sion operates, his theory has neither a clear nor a firm neurological foundation. In
short, as the theory now stands, it is unclear what Zeki is saying and why he
ignored the many published studies that would have allowed him to offer a more
robust exposition. A top-down, homologous theory that does not address known
[1] Overall, even general studies of brain research cover basic differences among language, auditory and
visual processing (see Carter, 1999; Greenfield, 1997; Posner and Raichle, 1997). These sources also
include recommendations for further reading. One of the most interesting comparative studies com-
pares the portrait painter Humphrey Ocean with non-artists performing the same task (see Miall and
Tchalenko, 2001; Solso, 1999; 2001).
60 A. IONE
distinctions and how different areas of the brain contribute to a mental abstrac-
tion, when our data show discernable differences, is at best vast mystification.
Zekis failure to critically engage with data is disappointing. His decision to
look only at the visual brain is inexplicable because the kinds of studies that
would help clarify his position are far-reaching. For example, while work on the
musical brain is not as abundant as work devoted to vision, a growing number of
researchers are exploring the musical brain and studies of knowledge-acquisition
in this area are growing.2 Studies by Robert Zatorre, a neurologist and a musi-
cian, would have been an excellent touchstone since Zatorre pioneered investiga-
tions into the neurobiology of music about two decades ago. Isabelle Peretz, also
a neurologist and a musician, studies amusia (tone deafness) in her lab at the Uni-
versity of Montreal. Her research has provided evidence of the distinctive nature
of the pathways involved in perceiving music. Andrea Halpern, a singer and cog-
nitive psychologist, uses PET imaging to observe the brains of subjects as
they imagine specific melodies (Halpern and Zatorre, 1999). Another excellent
resource is Christo Pantevs work. While at the University of Mnster in Ger-
many he and his colleagues showed that a much larger area of the auditory cortex
was activated in musicians on hearing a piano tone compared with a non-
musician (see Pantev et al., 1998; 2001). He is now using magnetoencephalo-
graphy to track neural development in young musicians taught by the Suzuki
method. This work is directly related to knowledge acquisition.
Equally unsettling is that Zeki excludes key neurological studies and, instead,
peppers his argument with philosophical notations that reference historical
views of art and aesthetics. It is striking to see how he places his argument in
terms defined by Plato, Kant, Schopenhauer and other philosophers rather than
neurological research that directly relates art and the brain. It is also ironic to see
this approach in light of the number of contemporary philosophers who view
recent scientific research as a tool that will aid us in distinguishing philosophical
arguments from actual brain operations. Given Zekis dependence on philosophi-
cal sources, rather than empirical data, it is not surprising to find his conclusions
are largely unrelated to current research in this area. What his inclusion of histor-
ical philosophers does, however, is lend support to a monolithic view of art,
favoured historically, that aligns art with universals. Adopting this historical
foundation allows him to easily frame his views in terms of universals and ideals,
rather than explaining particular operations. Unfortunately, the sweeping asser-
tions that result do not rest on scientific analysis, although they do allow him to
treat the work of Dante, Michelangelo and Wagner as if the distinct differences
among them are inconsequential.
Finally, and of great concern, is the way Zeki evaluates the data he does present
when he turns to the three artists named in the articles title. The neural concept for-
mation argument is essentially segregated from the psychological interpretation he
[2] Music, maestro, please! offers a good overview of this research (Abbott, 2002). In addition, each of
the scientists discussed in the text contributed to The Biological Foundations of Music, published by
the New York Academy of Science. Their articles include bibliographies that allow topics to be pur-
sued with more specificity (see Zatorre and Peretz, 2001).
SEMIR ZEKIS NEURAL CONCEPT FORMATION 61
offers to explain how an unrequited love was given form in the art of Dante,
Michelangelo and Wagner. The specific cases are studied in the next section.
[3] Reproductions of the Botticelli interpretation were recently re-published (see Botticelli, 2000).
62 A. IONE
Again I reiterate that Zeki, too, has offered an artistic interpretation of Dantes
work. He, too, engages with the philosophical notions, rather than presenting an
analysis based on empirical scientific data. Although there is no indication that
Zekis theoretical response was inspired by a romantic relationship similar to
Dantes with Beatrice, it is clear when reviewing Zekis speculative argument
that it is grounded in philosophical views, not scientific parameters.
Similarly there are too many unanswered questions and leaps of faith when he
turns to Michelangelo. According to Zeki, what made Michelangelo leave so
much of his work unfinished can be traced to the same source as that which made
Dante create his art the impossibility of realizing the ideals formed by the
brain in the experience of particulars in real life (Zeki, 2002, p. 65). Then, when
introducing Michelangelo, Zeki explains that Michelangelos unfinished work
attests to arts great power, and that we can surmise that Michelangelos unfin-
ished pieces, being ambiguous, speak to us in neurological terms. He further
explains, the ambiguous work of great art speaks to us in a neurobiological
sense, because when a work of art is unfinished it offers several solutions, all
of equal validity. Reading this it was unclear to me what is neurological about his
definition and how we should apply this neurobiological interpretation to unfin-
ished artwork by artists whose work and life clearly refute Zekis assumptions.
An artist whose work raises questions about the credibility of Zekis thesis
regarding the conjunction of art, love and unfinished works of art is Rembrandt
Harmenszoon van Rijn (160669).4 More specifically, we know that love was
often expressed in Rembrandts art and that he didnt find that unrequited love
was necessary whether producing finished or unfinished work. Rembrandts ren-
ditions of the women with whom he shared his life show this exceptionally well.
Moreover, reviewing the catalogue for the recent exhibition Rembrandts Women
reveals a number of easily accessible examples of how love was portrayed both
in his finished and unfinished work (Williams, 2001). A Woman in Bed, a particu-
larly poignant painting, has the peculiar honour of being said to represent all
three women in Rembrandts life. While we dont know who the model was, we
can easily see that Rembrandt has invited the viewer to share an intimate
moment. Similarly, the face of another sensual work, A Woman Bathing in a
Stream, depicts Hendrieckie, one of the women with whom Rembrandt shared
his life. Looking closely we find the handling of the paint is unusually spontane-
ous and the picture appears unfinished in some parts. This is most evident in the
shadow at the hem of the raised chemise, the right arm and the left shoulder. This
paintings earthy sensitivity, its ability to capture a personal moment, and its
unfinished areas all, in effect, demonstrate that Zekis correlation of unrequited
love and unfinished works of art is too glibly applied.
Zekis discussion of Wagner, his third artist, is more developed than his inter-
pretations of Dante and Michelangelo. I was impressed to see he gave a bit more
thought to how Wagners process of composing the music translated to the lis-
teners brain. I believe the Wagner section would have been stronger had he
incorporated some of the exciting work being done on synesthesia into the dis-
cussion. Briefly, one of the outstanding features of Wagners work is its reach,
and another is that he approached his work with the intention of stimulating
cross-modal experience. He believed that opera was the highest form of art
because it combined visual, dramatic, vocal, musical, choreographic and textual
elements into a single form. Juxtaposing contemporary scientific experiments
that have deepened our understanding of cross-sensory connections with artistic
experimentation would have allowed Zeki to examine Wagners work from a
number of rich perspectives. Had he looked at them he might have also recog-
nized that a number of recent scientific studies have raised questions about the
very philosophical and historical assumptions he relies on in his exposition of
art. As Christopher W. Tyler and I have discussed (Ione and Tyler, in press),
what is most impressive about contemporary research in this area is the way the
idea of five distinct senses, as codified by Aristotle, has been re-examined and
found to be too black and white. Although the details are too complex to re-state
here, suffice to say that scientists are opening neurological doors that will aid
further study of sensory integration. This includes our ability to conceptualize
how an artist like Wagner developed work capable of stimulating cross-modal
sensory experiences within his audience, among other things.
Historical Perspective
Zekis paper, in my opinion, is perhaps most fascinating in its unintentional dem-
onstration of a number of implicit assumptions often assumed to be axiomatic
truths when art and cognitive science are brought together. While historical ideas
cannot be well-developed in this response essay, I do want to bring some of the
historical biases adopted by Zeki to the fore.
The School of Athens by Raphael Sanzio (14831520) offers a means to
quickly introduce some historical foundation.5 One of the four main fresco walls
in the Stanza della Segnatura, this painting is often described as a complex alle-
gory of secular philosophy and is said to illustrate the historical continuity of Pla-
tonic thought. Compositionally, the work depicts Plato and Aristotle surrounded
by philosophers, past and present, in a splendid architectural setting. The space
in which the figures congregate is said to be based on Bramantes design for the
new St Peters in Rome.
Zekis argument brings this painting to mind because it is carved out of the tra-
dition enshrined in the famous painting. He, too, glorifies the people and ideas
represented and aligns those of the Italian Renaissance with the ideas of the
ancient Greeks. Looking closely at the painting, we find Raphael used the stars
of the Italian art world to stand in for Greek figures. Michelangelo, one of the
three Titanic figures in Western culture that Zeki examines, is placed in the
[5] Images of the fresco can be found on the Vatican page, see http://www.vatican.va/museums/
patrons/documents/vm_pat_doc_12101999_raphael_en.html. Information about the people repre-
sented is available at http://hypo.ge-dip.etat-ge.ch/www/athena/raphael/raf_ath4.html. Excellent
images, identification of the figures and a brief history can also be found in James Becks Raphael:
The Stanza della Segnatura, Rome (1993).
64 A. IONE
[6] Views reported by Portuguese artist Francesco de Holanda, and often cited in the literature to show
how the Italian view has been given preference in art history and theory (see Klein & Zerner, 1989).
SEMIR ZEKIS NEURAL CONCEPT FORMATION 65
aesthetic theories in the West. Ive introduced these sixteenth- century debates to
note that it was the Italian point of view that became foundational to later studies
and the philosophical texts. Particular philosophical and religious predisposi-
tions were inscribed in this view, and these predispositions influenced how our
contemporary art and science dialogue took form.
Art was increasingly defined in terms similar to the Italian view, and its terms
are very much embedded in the philosophical theories that equated art with uni-
versal Truth. The very theories Zeki incorporates into his discussion. Art was
also seen as a modality in the service of philosophy and spiritual pursuits. Cogni-
tive Science and I believe consciousness studies as well have largely
adapted this framework to their studies of art, despite the fresh empirical data
that has been brought to the study of the mind. This adaptation has taken form
implicitly, as a part of how cognitive science continues to define its methodol-
ogy. Generally the question what is cognitive science is answered in a way that
characterizes it as an extension of the theories of mind we date back to antiquity,
and Platos Dialogues are usually referenced as the key starting point. Plato, of
course, brought contradictory views on art to the table.
Perhaps of greater importance when reviewing Zekis paper is that he tells us
he is not a Platonist, and nonetheless presents an argument reminiscent of Pla-
tonic ideas rather than a neurological exposition. In light of the way Plato is both
cited and rejected by Zeki, it is unfortunate Zeki never considers where his views
fit in relation to the particular questions that have always plagued Platos legacy,
and the way many of Platos views were implicitly absorbed into philosophical
theories of art. Nor does Zeki show any concern about the degree to which Plato
has influenced Western philosophy and theories of mind. Instead Zeki simply
recognizes Platos authority. While it is perhaps not relevant to note that,
despite his own artistry, Plato bans the artists from his Republic, Platos views of
the brain, as expressed in the Timeaus, are important in light of Zekis critique.
As is well known, the Timeaus was particularly successful in spreading Platos
teleology and his rejection of sensation and observation in favour of reason. Of
greater importance when applying cognitive science research to art today is that
while Platos ideas about the brain have been re-examined over time, the
approach he outlined is still employed to ask and answer questions about the
mind, sensory experience, perception and the relationship between empirical
knowledge and what is often designated true knowledge. One might say that
the way in which methodological derivations have been re-configured under-
scores Platos influence on how theorists discuss art in relation to theories of the
mind today. Yet, as is frequently noted, art practice and appreciation cannot be
encapsulated with a theory.
Conclusion
This response to Semir Zekis Neural Concept Formation and Art: Dante,
Michelangelo, Wagner barely begins to touch upon the complex issues inter-
woven with the theory Professor Zeki presents. As I have sketched above, and
discussed elsewhere (Ione, 2000; 2001), Zeki defines art in a way that deletes
66 A. IONE
much of what artists do and the complexity of how humans relate to art. His com-
mitment to particular axiomatic assumptions about art also limits his analysis.
When this limited purview is the framework adopted in studies of art its con-
stricted vision in effect emphasizes the art historian E.H. Gombrichs view that
the wellspring of art lies in the multifarious contributions of individual artists.
While cognitive studies can enhance our understanding of art, sweeping specula-
tions cast as science have the potential to detract from efforts to understand and
appreciate art. Theoretical conjectures, moreover, have the potential to compli-
cate the equation. With this in mind, I would urge Semir Zeki to develop his the-
ory further and, as he does so, to fully examine groundbreaking neurological
studies that produce data related to the exciting ways artists actively develop and
cast the various art forms he examines.7
References
Abbott, A. (2002). Music, maestro, please!, Nature, 416 (7 March 2002), pp. 1214.
Beck, J. (1993), Raphael: The Stanza della Segnatura, Rome (New York: George Braziller).
Bindman, D. (2000), Divine Comedy (Bibliotheque de lImage).
Botticelli, S. (2000), The Drawings for Dantes Divine Comedy (London: Royal Academy of Art).
Carter, R. (1999), Mapping the Mind (Berkeley, Los Angeles, London: University of California Press).
Graves, G. (2002), Divine Comedy: The Dante Project, March 2002 performance at the Zellerbach Playhouse.
Greenfield, S.A. (1997), The Human Brain (New York: Basic Books).
Halpern, A.R. and Zatorre, R.J. (1999), When that tune runs through your head: A PET investigation of
auditory imagery for familiar melodies, Cerebral Cortex, 9, pp. 697704.
Ione, A. (2000), An inquiry into Paul Czanne: defining the role of the artist in studies of perception and
consciousness, Journal of Consciousness Studies, 7 (89), pp. 5774.
Ione, A. (2001), Innovation in Art and Science: Reply to Semi Zeki, Trends in Cognitive Science, 5 (4),
p. 140.
Ione, A. and Tyler, C.W. (in press), Synesthesia: is F-Sharp Colored Violet?, Journal of the History of
the Neurosciences.
Klein, R. & Zerner, H. (ed. 1989), Italian Art, 15001600 (Evanston, IL: Northwestern University Press).
Levy, E. (2000), Review: Inner Vision: Art and the Brain, Endeavor, 24 (3), pp. 137, 138.
McMahon, J. (2000), Review: Inner Vision, An Exploration of Art and the Brain, Leonardo On-line
(September 2000), http://mitpress2.mit.edu/e-journals/Leonardo/reviews/sep2000/
bk_INNVIS_mcmahon.html.
Miall, R.C. and Tchalenko, J. (2001), A painters eye movements: A study of eye and hand movement
during portrait drawing, Leonardo, 34.
Nash, J. (2000), Unfinished portrait of the artist, Nature, 404 (09 March 2000), pp. 1234.
Pantev, C., Oostenveld, R., Engelien, A., Ross, B., Roberts, L.E. and Hoke, M. (1998), Increased audi-
tory cortical representation in musicians, Nature, 392, pp. 81114.
Pantev, C., Engelien, A., et al. (2001), Representational cortex in musicians: Plastic alterations in
response to musical practice, in Zatorre and Peretz (2001).
Phillips, T. (2001), Dantes vision, The Royal Academy Magazine, 70 (Spring), pp. 503.
Posner, M. I. and Raichle, M.E. (1997), Images of Mind (New York: Scientific American Library).
Shortess, G.K. (2000), Review: Inner vision, an exploration of art and the brain, Leonardo On-line
(Sept.), http://mitpress2.mit.edu/e-journals/Leonardo/reviews/sep2000/ bk_INVISION_shortess.html.
Solso, R.L. (2000), Brain activities in an expert versus a novice artist: An fMRI study, Journal of Con-
sciousness Studies, 7 (89), Special issue on Art and the Brain, Part II, ed. J.A. Goguen and E. Myin
(Exeter: Imprint Academic), pp. 7585.
Solso, R.L. (2001), Brain activities in a skilled versus a novice artist: An fMRI study, Leonardo, 34.
Williams, J.L. (2001), Rembrandts Women (Munich, London, New York: Prestel).
Zatorre, R.J. and Peretz, I. (ed. 2001),The Biological Foundations of Music (New York: NYAS).
Zeki, S. (2002), Neural concept formation and art: Dante, Michelangelo, Wagner, Journal of Con-
sciousness Studies, 9 (3), 2002, pp. 5376
[7] Special thanks to Christopher Tyler and Kevin Flanagan for feedback as this piece developed.