Eric R. Kandel, Reductionism in Art and Brain Science: Bridging the Two Cultures. New York: Columbia University Press, ©2016 Eric R. Kandel.
Dr. Eric Kandel, whose most intriguing book I am going to discuss, is a tremendously distinguished neuroscientist. He is known in particular for his "discoveries concerning signal transduction in the nervous system," to quote his citation upon winning a Nobel Prize for physiology or medicine in 2000.
When he won this award, he was already a member of a half-dozen learned societies, had won 20 other prizes and awards, and been the recipient of nine honorary degrees from universities in the U.S. and abroad.
"Reductionism in Art and Brain Science" is his seventh published book, in addition to which there must be who-knows-how-many published scientific articles.
BACKGROUND ON THE AUTHOR
Born in 1929 in Vienna, Dr. Kandel immigrated with his family to Brooklyn in 1939 and attended Erasmus Hall High School, one of the shining lights in New York City's then mostly-excellent public school system. From there, he went to Harvard, majored in 19th and 20th century European history and literature, and thought of going on to do graduate work in European intellectual history.
However, at Harvard, he met Anna Kris, a fellow student and the daughter of two émigré psychoanalysts who came from Alt Wien and hence shared something of the Kandel family history.
Moreover, Anna's father, Ernst Kris, was not only one of Sigmund Freud's intimates but also had a previous career as an art historian and curator. The family as a whole inspired in Eric a wish to become a psychoanalyst, too. Who knows? Ernst Kris may also have encouraged in him an interest in the visual arts.
In the early 1950s, psychoanalysis was at the peak of its influence in the United States. Though it was still hotly controversial, it was much admired by many thinking people. To practice it, though, one had to become an M.D., so Eric Kandel decided to enter medical school.
Another reason for his interest in psychology, if not psychoanalysis, may conceivably have been the influence of B. F. Skinner (1904-1990). Skinner had been teaching psychology at Harvard since 1948, and was a proponent of behavior modification, as originally examined and defined by Ivan Pavlov (1849-1936), the Russian physiologist known primarily for his work in "classical conditioning."
Pavlov had discovered that while even untrained dogs would salivate at the approach of food, a dog could be trained – or conditioned -- to salivate at the sound of a bell, buzzer, or metronome – if these were succeeded by the arrival of food in their dishes.
With Skinner, it was pigeons that he most famously trained to play ping-pong -- once again, by offering food as a reward. Skinner had developed this into a whole school of psychology, predicated upon the assumption that human beings, too, could be conditioned by a system of punishments and rewards into altering their behavior.
Of course, neither Pavlov nor Skinner had a clue as to how the brains of these animals learned their behavior, let alone how humans did. That would be left for Dr. Kandel and his associates to discover – but I am getting ahead of my narrative. At this point, all I need say is that in the 50s, the psychoanalytic approach of Freud and behaviorism of Skinner were at opposite ends of the discipline of psychology, with no love lost between them.
I say this upon the basis of my own experience, having had two years of psychoanalysis by 1952, when I entered Barnard only four years after Dr. Kandel had entered Harvard. I was interested enough in psychology to inquire about psychology courses I could take, but was told that the Barnard psychology department was dominated by behaviorists, not a Freudian in the bunch.
(This would be more than ever the case forty years later, when I finally got around to auditing an introductory psychology course at Barnard, though the '90s variety of anti-Freudian pedagogy at Barnard had a Marxist tinge and my teacher defined herself as a "social psychologist").
DR. KANDEL'S BREAKTHROUGH: THE BIOLOGY OF LEARNING & MEMORY
In the speech Dr. Kandel gave when accepting his Nobel Prize, he said that "I entered N. Y. U. Medical School dedicated to studying psychiatry and becoming a psychoanalyst. Although I stayed with this career plan through my internship and psychiatric residency, by my senior year in medical school I had become so interested in the biological basis of medical practice…that I decided I had to learn something about the biology of the mind."
There follows in this speech a description of the topnotch academic and governmental laboratories he worked in, the eminent neurobiologists he worked for and with, and the series of laboratory experiments that other scientists had performed in pursuit of ambitions similar to his own.
However, to summarize as best I can, he began to focus on the study of perception, as it might be witnessed and described through the biology of the brain. To rehearse what little I remember of that introductory psychology course at Barnard, the brain is constructed of neurons, or cells, and synapses, or the chemical and electrical signals that travel over the extremities of these neurons to connect them with each other.
Dr. Kandel wanted to illuminate the biology of learning and memory. He felt that all or most of the studies by other neuroscientists didn't provide enough information because the brains chosen to investigate were larger and more complex, being mostly those of humans or at least vertebrates.
Though not all neuroscientists agreed with him, Dr. Kandel believed that the brains of simple invertebrates, while containing far fewer neurons and synapses, were composed of the same kinds of neurons and synapses as the higher vertebrates, up to and including humans.
Therefore he adapted the principle of what is known as "reductionism," and looked around until he found a creature whose brain was smaller and therefore simpler than that of any creature so far studied, so simple that its learning processes would be clear and unequivocal.
He eventually settled upon a sea snail known as Aplysia, whose brain has only 20,000 (somewhat larger) neurons, as opposed to the human brain, which may contain as many as 100 billion (somewhat smaller) ones.
Aplysia couldn't do much with its tiny brain, but it could slightly retract the covering of its gill at the touch of a hand. When it was then given an electric shock in its tail, it retracted the covering of its gill much more dramatically.
The breakthrough came in 1970. Dr. Kandel and his team were able to train Aplysia to withdraw its gill much more dramatically with only the touch of a hand—using the same sort of classical conditioning Pavlov had used with dogs and Skinner, with pigeons.
Furthermore, Dr. Kandel and his associates were able to observe and describe how this "learning" process took place -- not in the neurons, but in the synapses, whose compositions ever so slightly altered in the learning process.
In other words, when you learn something by committing it to memory, you – ever so slightly -- alter your brain.
To summarize; the human brain is a subject that Dr. Kandel has been studying in one way or another for at least 70 years, or pretty much ever since he entered med school in 1952. He displays his knowledge of it in a most illuminating manner in this book. As I have virtually no prior experience with this subject, I can only hope I have gotten my synopsis of his discoveries more or less right.
DR. KANDEL ON ART
On the other hand, he mentions no formal training or prior experience in writing on the visual arts, and this is a subject I have been studying and writing about since 1967. That was the year when I was assigned to handle the Art section of Time magazine (in those days that Art section appeared weekly). I received my Ph.D. in art history in 1982, from Columbia University (where Dr. Kandel is a professor in the departments of neuroscience and psychiatry).
My dissertation dealt with painting in the 1940s in New York, the decade which saw the birth of abstract expressionism. Among the hundreds of painters working in New York during that period that I wrote about in the dissertation were three first-generation abstract expressionists who also figure prominently in Dr. Kandel's book – Jackson Pollock, Willem de Kooning & Mark Rothko.
In addition, he deals with Morris Louis, the abstract color-field painter who, although born in 1912, the same year as Pollock, didn't develop a signature style until the mid-1950s. Louis was only beginning to become really well-known when he died at the age of only 49 of lung cancer in 1962.
The color-field school of painters -- of the 1960s and since -- has also been a specialty of mine since I was at Time. So maybe I do know a bit more about this subject than Dr. Kandel does, though again I believe that art is more about interpretation than science is, and I would be the first to agree with Clement Greenberg when he said that there can be no objective proof of esthetic judgments.
Accordingly, I won't comment at length upon Dr. Kandel's esthetic judgments, beyond saying that his discussion of abstract expressionism appears to have been influenced more by his taste and/or desire to advance his theories about abstract painting than by any desire to give a chronologically accurate history of the period.
Most notably, he deals first with de Kooning, and only secondarily with Pollock. This might suggest to the average reader that de Kooning was the real trail-blazer, and more important than Pollock. I would disagree with both of these suggestions.
Pollock was not only ahead of de Kooning in developing a purely abstract style, but also the first to become widely known for it. De Kooning had only one solo exhibition in New York during the 1940s, in 1948. This was the first solo show he had of pure abstractions. Pollock had seven solo exhibitions in New York in the 1940s – one each in 1943, 1945, 1946, 1947, 1948 and two in 1949. He was already showing abstractions by January 1947.
Dr. Kandel has polite superlatives for Pollock, too, but his decision to place de Kooning ahead of him becomes more understandable when you know that (according to his speech at the Nobel Prize awards) he collects prints by Austrian and German Expressionists.
De Kooning's famous but figurative "Woman" paintings are dealt with in "Reductionism in Art and Brain Science" enthusiastically --- and certainly they are lot closer to the demon goddesses of Klimt, Kokoschka & Schiele than anything Pollock ever painted.
Introducing de Kooning before Pollock also enables Dr. Kandel to introduce Harold Rosenberg before Greenberg, and to treat Rosenberg's 1952 article on "action painting" in Art News as the first writing ever on abstract expressionism.
This ignores the fact that Greenberg had been writing about Pollock and other abstract expressionists since 1943, and at such length that the first two volumes of his 4-volume collected writings are filled with reviews and articles published in the 1940s (Dr. Kandel's bibliography doesn't list Greenberg's collected writings).
As nearly as I could determine, when working on my dissertation, Rosenberg published nothing at all on the visual arts prior to his 1952 article, except for a couple of catalogue essays for galleries.
The treatment of Morris Louis and Louis's relationship to Helen Frankenthaler is confusing. The passage begins on Page 130 by defining Rothko as a "color-field painter" and then observing, "Other color-field painters include Helen Frankenthaler, Kenneth Noland, Morris Louis and Agnes Martin." (Agnes Martin? First time I've heard of her categorized as a color-field painter. I would call her a minimalist, particularly since every painting by her I've seen was close to monochromatic).
To continue to Page 134 of Dr. Kandel's book, we read that when Louis was living in New York between 1936 and 1943, he "came under the influence of Pollock and the New York School, but he soon turned from action painting to Color Field painting, inspired by Rothko and, more directly, by Frankenthaler."
Between 1936 and 1943? Frankenthaler was born in 1928. In 1943, she was still in high school. Even Rothko didn't achieve his signature style until the end of the 1940s.
The page continues with a description of Louis's pouring technique, and says that with Noland he formed the nucleus of the Washington Color School. Then it says that "In April 1953 Noland brought Clement Greenberg to Washington to meet Louis." Next it describes how Greenberg "looked over Louis's work and was immensely impressed." This is not what happened.
Instead Louis and Noland came up to New York in April 1953, and went with Greenberg (and others) to Frankenthaler's studio to see her revolutionary painting, "Mountains and Sea" (1952). In this painting, she had inaugurated the technique of thinning her paints and staining her unprimed canvases with them, instead of allowing thicker paint to sit on the surface of a primed canvas.
Both Noland and Louis adopted this new technique and it became the basis for their best-known paintings. I could quote any number of books that say this but maybe it is enough to point out that the gist of it even appears on the official Morris Louis website.
This staining technique of Frankenthaler's was derived not from Rothko but from Pollock—specifically his "black paintings" of 1951. To quote Louis: Frankenthaler was "a bridge between Pollock and what was possible."
As far as I know, Greenberg never applied the term "color-field painter" to Rothko. Indeed, the only time he dealt with him at any length was in "American-Type Painting" (1955). He didn't review Rothko's solo shows, not because he didn't admire his work but – as he himself once told me – because Rothko asked him not to.
Again, as far as I know, the first writer to call Rothko a "color-field" painter was Irving Sandler, in "The Triumph of American Painting: A History of Abstract Expressionism" (1971). Sandler divided the major figures of abstract expressionism into two branches, "gesture painters" (including Hans Hofmann as well as Pollock and de Kooning), and "color-field painters" (including Clyfford Still & Barnett Newman as well as Rothko).
But enough of these no-doubt nit-picking complaints! Where I am particularly eager to disagree with Dr. Kandel involves an issue half-way between art and science. It is an issue that goes to the heart of his argument that abstract art is a form of "reductionism" and as such constitutes an esthetic equivalent to his own Nobel Prize-winning neuroscientific discoveries.
I may as well confess, however, that my own argument on this subject cannot be substantiated by any appeal to standard authorities in the art world. This notion that to move from representation to abstraction constitutes a loss of meaning, content and subject matter – in short, a reduction to "the cylinder, sphere and cone" – is already standard not to say whiskery dogma in the art world.
Indeed, the notion that abstract art depicts nothing and hence means nothing has been the accepted view ever since Picasso & Braque first mystified Paris with Analytic Cubism around 1910, and Marcel Duchamp accordingly began to complain that abstraction was purely "retinal" art shortly thereafter.
As far as I'm concerned, this notion is wrong. Far from being divorced from subject matter, I find the best abstract painting richly allusive, with resemblances to not one but many different objects both in the natural and humanly-made world. The simplest way to describe this phenomenon is to say that abstract painting is ambiguous.
I call the phenomenon "multireferential imagery," as opposed to the "uni-referential" imagery of representational painting. For the benefit of those readers new to my writing, I will now trace my development of this theory as briefly as I can (regular readers may want to skip to where I return to neuroscience and its relationship to the visual arts.).
TO RECAP: THE ORIGINS OF "MULTIREFERENTIAL IMAGERY"
The original basis for my theory occurred to me initially in late 1982, shortly after I'd deposited my dissertation. I first published this theory in an article in Arts Magazine in September 1983. But its genesis goes back many years, maybe even to when I still in high school and undergoing my first psychoanalysis.
Anyway, in analysis (as it was practiced in those far-off days), one learns that everybody has a conscious and an unconscious mind, and that a patient, with the aid of the analyst, can discover what is going on in her unconscious.
She does this by describing her dreams and then, by the process of "free association," linking them to many other subjects, objects and experiences that occur to her and that seem to be memories hitherto hidden in her unconscious. Hence I learned how one object in a dream could be related to many other objects and subjects in my mind.
After four years in college, when I was supposed to be developing on my own into a mature woman but didn't, I went back into analysis (with a different, younger and less expensive analyst). At this point, I was working at Time as a researcher, and in my spare time, trying to write plays.
This analyst (God bless him) didn't try to analyze those plays. Instead, he urged me to channel my "feelings" into them. In analyzing my dreams, he underlined the notion that the feelings I had in dreams were prompted by actual things that I had seen or that had happened to me – in short, that we do not "feel" in a vacuum but in response to actual experience.
In those days, I still took a very dispassionate view of abstract art. While I respected it, I couldn't see that it was "about" anything. This changed after I became a writer on Time and was assigned to the Art section.
What changed my mind in 1968 was a curator who argued (passionately) that abstract art and great art in general were not about subject matter but about "feeling." This corresponded not only to what the analyst had said about "feeling" in my own writing but also from my study of T. S. Eliot in college. There I had learned that although Eliot's great poem "The Waste Land" didn't appear to have any plot, it nonetheless had a powerful ability to convey feeling to its readers
From that moment on, I began to relate to abstraction on a gut level (not least because I had a crush on the curator).
In the case of Pollock, I began to get a warm, exciting feeling when I looked at his (best and most abstract) paintings. Mindful of the analyst's advice that "we do not feel in a vacuum" I tried to figure out what I was reminded of by those paintings.
I concluded that his liquid swirls and swoops of paint reminded me of – semen, and the happy feelings I associated with that body product. I wouldn't have dreamed of publishing this association in Time, though. Our (predominantly less sophisticated) readers would have been scandalized or – worse still – laughed.
By the early 1980s, when I was working on my dissertation, my interpretation had changed. I was now reminded by Pollock's paintings of a nuclear holocaust, but I couldn't see how the same painting could be about two such different things at the same time, so I pushed it all to the back of my mind.
Instead I focused on what had been written about Pollock's life and times, and on what had been written about his paintings-- both by the Manhattan art critics of the 1940s, and by all the biographers and art historians who had come along since.
What I found was that Clement Greenberg stuck to describing the paintings in terms of their composition and colors, that is to say, purely on the basis of their style -- the way that Pollock wanted him to.
Although Greenberg wrote that Pollock's paintings had "meaning" and "content," he denied that abstract art in general could have "subject matter." Most critics, though, took a less sophisticated approach, and compared Pollock's paintings to one thing or another in the external world.
Henry McBride, a senior modernist critic with the New York Sun, shared my notion of a nuclear holocaust. He wrote, rather poetically, that one Pollock reminded him of "a flat, war-shattered city, possibly Hiroshima, as seen from a great height in moonlight." Emily Genauer, a coyly sarcastic & more conservative critic writing in the New York World Telegram, was reminded of "a mop of tangled hair."
An article in Life on Pollock likened his paintings to "macaroni," and Sam Hunter, then a younger critic on the New York Times, may even have prefigured my associations with semen: he wrote of Pollock's "pure calligraphic metaphor for a ravaging aggressive virility."
Nor had the biographers & art historians, writing decades later, necessarily abandoned examining these paintings for subjects: one likened them to jazz, while more than one was reminded of landscape imagery.
I was able to link up all these allusions with things that Pollock had seen, heard – or at least (in the case of nuclear holocaust) with things of which he could have seen photographs (in Life, most notably—I was able to confirm from Greenberg that the artist did read Life, even before it wrote about him).
My references to Pollock's biography moreover showed that these various images had had emotional resonance for him – though I also agreed that he hadn't been aware of or responsible for any of these allusions – consciously that is.
My explanation for this phenomenon was based upon Freudian dream psychology. I argued that all of these allusions had coalesced or synthesized in the artist's unconscious mind, and, because they had, viewers picked up on them in observing the finished painting.
Different observers were reminded of different things, I maintained, because their own differing visual and emotional experience corresponded to different sights that the artist had seen and recorded in his or her memory.
The first person on whom I tried out this theory was the mind doctor I was going to in the 1980s. He was a psychiatrist, not a psychoanalyst, and very much into group therapy and cognitive psychology – an 80s model of psychotherapist, not one from the '40s, '50s or '60s. His (incredulous) response to my theory was something like this: "But that's just like the Rorschach tests – you can see anything you want in them, too!"
I answered something like this: "Are you sure? Don't some of these tests look more like one thing, and some look more like another?" He admitted that yes, this was true, and I could check it out in any psychology book on the tests. I did, and my 1983 article has a footnote mentioning one such book, and its list of "more usual whole responses."
TO RETURN TO NEUROSCIENCE
Dr. Kandel appears to share a tiny bit of my thinking on this, though very far from the whole of it (as nearly as I can tell). More than once, he suggests in his book that a viewer responds to an abstract painting upon the basis of "personal experience."
However, he never defines what kind of personal experience he means—beyond suggesting that this might be an emotional response and/or reflect having seen other (presumably also abstract) paintings.
He never suggests that such experience could include memories of objects that the viewer has actually seen in the natural world—or other humanly-made objects either. My own associations with the imagery of Mondrian, for example, include humanly-made objects, as straight lines and perfect squares or rectangles do not occur in nature.
(Among other things, I am reminded by a Mondrian of photographs I have seen of the clear square window panes in old Dutch Reformed churches in The Netherlands, as opposed to the brightly-colored, and often figurative window glass in most other Christian churches.
(I also know that a) the artist was raised in a rigorously religious family and even thought of becoming a clergyman before he decided to become an artist and b) the Dutch Reformed church is Calvinist in origin, and hence takes very seriously the Second Commandment.
(That's the one about "Thou shalt not make unto thee any graven image, or any likeness of any thing that is in heaven above, or that is in the earth beneath, or that is in the water under the earth." But I digress.)
Of course, attempting to plumb an abstract artist's unconscious is a perilous activity, because the undertaking is so speculative and even more because abstract artists prefer to think of themselves as depicting nothing and conveying nothing beyond at most an indefinable emotion or "content."
These are two big reasons that I have long sworn off attempting to interpret contemporary paintings and the work of living artists in this way. A third and even more important reason is that my current profession is primarily as an art critic, not an art historian, and such speculations are irrelevant when it comes to assessing the quality of a painting -- which is an art critic's main job.
Art historians may have their reasons for wanting to reconstruct an artist's physical, psychological and sociological origins and context, but a critic's job is to single out artists and paintings most likely to remain admirable to future generations, and that must be primarily upon the basis of their style, not their subject matter.
How many hundreds of Madonnas were painted in Venice in the 16th century – and why do we only care about those painted by a very small handful of artists: Giorgione, Titian, Tintoretto, Veronese and maybe one or two others? Style, not subject matter, is the ultimate determinant of quality.
As for Dr. Kandel, the biggest single reason that he doesn't undertake to find any associations between what abstract artists may or may not have seen and what they paint must be that he is only concerned with perception, and never creation – always with what the viewer sees, never with what the artist has made.
He apparently assumes that communicating the visual appearance of observable objects or specific subject matter can't occur if the painting under study is abstract – or at least, that it shouldn't, if the viewer has been properly conditioned to ignore them.
And maybe I missed it, but I found no evidence in this book that neuroscience has yet solved the mystery of how the human mind – or brain, if you prefer – abstracts and generalizes.
How for example does it get from a) a dog named Lola to b) a cocker spaniel, including not only Lola but all the other cocker spaniels in the world, to c) a dog, meaning not only all the cocker spaniels but also all the poodles, schnauzers, Labrador retrievers, shih tzus and so on, to d) an animal, meaning not only all the dogs, but also all the cats, cows, tigers, elephants and so forth?
How did the mind or brain – of observers as well as artists -- ever get from the sun, moon, and all the planets (including Earth) to the generalized concept of a perfect circle or globe?
Perhaps the explanation for this can be found somewhere in the mysterious way that information processing progresses from "bottom-up" to "top-down." Dr. Kandel introduced me to this seeming dichotomy in his book.
I had never heard of it before, but if I understood him correctly, this is a widely-used means of scientifically classifying – and contrasting -- the difference between the more inherent and natural ways of identifying what the optic nerve conveys to the brain in the act of seeing (bottom-up) and the more sophisticated ways of identifying what one sees, based on one's more educated ability to place it in one or more relevant contexts (top-down)
He argues that representational art relies mostly on bottom-up information processing, since it depicts objects in the natural world. Top-down is more appropriate for abstract paintings, since it relies more on the experience of more educated observers. Such observers can appreciate an abstract painting because they will be able to compare it with other paintings and their own personal or emotional experiences.
To me, however, pondering this apparent dichotomy as I read Dr. Kandel's book, the problem is not quite that simple. As indicated by my discussion of Pollock's paintings above, one can be reminded of many things in the natural world by an abstract painting.
Similarly, I would argue that a representational painting is perfectly capable of reminding observers of both of emotional experiences and of other paintings (a late Cézanne landscape can make me quiver in delight, nor is my pleasure lessened by knowing that both Matisse and Picasso loved such work too).
Finally, what mental faculties are used to respond to a semi-abstract painting, whether we are talking about a seascape by Turner or a comic fantasy by Miró? Can it be that we have a meeting of bottom-up and top-down when we contemplate a picture that is partially abstract and partially representational?
Putting the matter into more general terms, I see a steady continuum between bottom-up and top-down, not a dichotomy. To be fair, Dr. Kandel occasionally hints at this possibility, but for the most part, he denies it.
It is conceivable that in setting up this duo as primarily oppositional, he has been influenced by his fellow Viennese, the distinguished art historian Ernst Gombrich (1909-2001).
I heard Dr. Kandel give a talk in October 2018 at the New York Psychoanalytic Society & Institute as part of a conference on "The Mind of the Artist." His topic was "The Beholder's Share: Looking at the Brain Looking at Art," and in this talk he mentioned Gombrich – though I blush to admit that I wasn't taking notes, so I can't state exactly what he said.
However, on my own I see a similarity between "top-down" and "bottom-up" and Gombrich's ideas about "making" and "matching." Gombrich discussed this concept in his influential book, "Art and Illusion: A Study in the Psychology of Pictorial Representation."
This book – to which I was introduced in my first semester of grad school – was first published in 1960, but it was based on a series of seven lectures that Gombrich delivered at the National Gallery in Washington in the spring of 1956.
1956 was at or near the high water mark in the prevalence of abstract expressionism. From what I have heard, this was during the period when from one end of Fifty-Seventh Street to the other, nearly every gallery in town was showing abstract expressionist paintings, both from the giants of the first generation and from second- and third-generation followers, mostly of lesser strength.
Within the next few years, the Neo-Dada antics of Robert Rauschenberg, Larry Rivers & Jasper Johns would begin to attract the attention of all the good folk who had never been able to relate to abstraction and welcomed this return to figuration if not representation. But Gombrich was more of a purist.
He doesn't seem to have been much interested in Neo-Dada's objets d'art (ranging as they did from a red-stained bed to a series of vulgarly-embellished targets). Instead, he was passionate about the representational painting of earlier eras, and "Art and Illusion" can be read as an extended defense of such work in the face of abstract expressionism.
However, what Gombrich really liked about the representational art of ages past wasn't only its powers to create an illusion of reality (in one or another way). It was also the capacity of these old painters to create works of art completely worth looking at for their own sakes -- as art in other words, as well as illusion.
He therefore set up a dichotomy of Art, on the one hand, and Illusion, on the other (hence the title of his book). He defined this pair in terms of another dichotomy: "Making" (art) and "Matching" (illusion)."
As far as he was concerned, a landscape by Constable was worth looking both for the way it depicted something outside of itself ("matching") and as a beautiful painting in its own right ("making").
But he also insisted on the ambiguity of his concept. He argued that you could look at a painting as a painting (as art) OR you could look at it as a simulation of external reality (as illusion) but you couldn't see BOTH at the same time. He compared this paradox to the celebrated duck-rabbit image from an old Viennese humor magazine. And it's true that if you look at this image, you can see the duck OR the rabbit, but never both at the same time.
Gombrich used these concepts to justify his admiration for representational painting, which as he saw it could both be esthetically rewarding in its own right, and also offer a satisfying replica of the external world. He also seems to have loathed abstract expressionism, using a particularly ugly painting by Pollock to illustrate his brief discussion of him as "the action painter."
He also chose an unusually nasty (though still valid) association with the external world to describe the Pollock, suggesting that it resembled "the intricate and ugly shapes with which industrial civilization surrounds us…twisted wire and complex machinery…." What he resented most was the fact that Pollock had seemingly telescoped Art and Illusion into a single image: "Making," Gombrich wrote, "results in matching."
The dichotomy between Gombrich's concepts of art and illusion, especially as reinforced by his ideas about making and matching, suggests to me a third pair of concepts that I only became familiar with in grad school—but that I have subsequently found to be well-known elsewhere.
This is two very much opposed ways of looking at and discussing art, the formal and the iconographic (or, to bring that last term more up to date, the socio-iconographic).
Formal analysis is used to look at and analyze a painting purely as art, how well it "works" in terms of its drawing and color, how the paint is applied and so on. In other words, it responds primarily to the "making" of a work of art, and only secondarily (if at all) to its subject matter.
Socio-iconography, on the other hand, deals with the "matching" properties of a work of art -- imagery of a painting, its subject matter, its socio-economic context, and so forth. A postmodernist scholar applying socio-iconography to a work may be completely blind to the work's "making" or esthetic quality – to its formal attributes, in other words.
In my experience, this third duo is ambiguous in the same way that the duck-rabbit image is—you can either look for the imagery in a painting OR try to evaluate it on a formal basis. Except to the extent that a painter has employed formal devices to clarify iconography, it is near impossible to see both at the same time.
I could be wrong about this, but it seems to me that Dr. Kandel's duo of "bottom-up" and "top-down" borrows from all three of these earlier dualities.
He associates "bottom-up" with illusionistic (i.e. representational) painting that "matches" the external world, and can be analyzed by iconographic methods. Similarly, he associates "top-down" with "abstract" (i.e. non-representational) art that has been "made" without any intention of "matching" the external world and can be analyzed only by formal methods.
The biggest difference between Gombrich and Dr. Kandel is that Dr. Kandel really likes abstract painting --- his book can be read as an apologia for it, as well as an explanation of (or even a recommendation for) how it is to be viewed.
But he may have overlooked Gombrich's point, which is that traditional representational painting is superior to abstraction because it can be appreciated from the standpoint of both "making" and "matching," whereas with Pollock, "Making results in matching."
I agree that a representational painting can be appreciated both from the standpoint of "making" and "matching." But I would add that, on the basis of my theory, abstract painting can also be appreciated from both those standpoints, too.
How does this figure into the concepts of "bottom-up" and "top-down?" Well, it suggests that some of the more sophisticated capacities of the "top-down" brain are necessary to fully appreciate representational painting, and that some of the most rudimentary features of the "bottom-up" brain are necessary to fully appreciate abstract painting.
In other words, these dualities may be not as oppositional as they appear.
ANOTHER POINT OF VIEW
I may be dreaming here, and I'm definitely not a neuroscientist or even a psychologist, but in trawling the web, I came across a somewhat speculative scientific paper that appears to share some of my own difficulties with a flat-out opposition between "bottom-up" with "top-down."
This paper suggests that "our current understanding of bottom-up and top-down" has "problems and limitations." It was written by Karsten Rauss, of the University of Tùbingen in Germany, and Gilles Pourtois, of Ghent University in Belgium. It was published in the 17 May 2013 issue of Frontiers in Psychology, and the authors "propose a reformulation of this distinction in terms of predictive coding."
The paper begins with a history of the original development of the concepts of "bottom-up" and "top-down." It then considers their "problems and limitations" and finally presents "predictive coding" as the solution to these problems.
However, writing my review of Dr. Kandel's book has already taken me far too long, and it would take me another month or so to really understand what "predictive coding" is all about, let alone how it can be used to "reformulate" the distinction between "bottom-up" and "top-down."
So I will just try and cut to the chase, as I really want to get this show on the road.
To judge from the diagrams included in this paper, "predictive coding" means that impulses in "top-down" perception travel down a kind of ladder to help with "bottom-up" perception, and the reverse may also be true. The diagram suggests that these impulses skip some steps on the ladder, for reasons that are not clear to me.
The authors imply that more work is needed, and that one experiment essential to proving their hypothesis might be difficult to carry out "with current methods." However, they suggest that other "more realistic" approaches might offer "converging evidence" to support their view.
"Based on a predictive-coding model," they conclude, "we have outlined a conceptually unequivocal distinction between bottom-up and top-down processes that addresses some of the limitations of our current understanding of these terms.
"Our proposals highlight the mutual interdependence and constant interaction between bottom-up and top-down processes.
"Thus, rather than searching for cases of pure bottom-up or top-down processing, future efforts should address their relative contributions as well as the mechanisms of their interaction in the context of a given task."
I couldn't have said it better myself.
This is not to say that I didn't enjoy reading "Reductionism in Art and Brain Science," because I certainly did. It has many sympathetic thoughts and observations. After all, how often in these postmodernist days do we get a fervent endorsement of truly greatmodernist abstraction?
Or, for that matter, how often do we get what amounts to a formalist defense of abstract painting and a presentation of Greenberg that (by comparison with what else has been written about him) is surprisingly sympathetic?
I am also grateful to Dr. Kandel for forcing me once again to think through and present my theory of multi-referential imagery. And to show how my position corresponds to or disagrees with not only his insights but also those of Gombrich.
Finally, I am most grateful to Dr. Kandel for enabling me to argue my defense of multireferential imagery in a neuroscientific context, as opposed to a purely art-historical or psychological one.
We all know that Sigmund Freud is considered terribly unscientific these days – though I would still maintain that one day neuroscience will get around to substantiating many if not all of his insights.