Franklin Toker On the Science Of Art History

Issue Date: 
January 19, 2010

Periodically, Chancellor Mark A. Nordenberg asks a Pitt faculty member to speak during a University Board of Trustees meeting about his or her work. The following is an abridged version of Prof. Franklin Toker’s talk, “The Science of Art History,” at the Board meeting of Oct. 30, 2009. Toker is a professor of the history of art and architecture.

Franklin TokerFranklin Toker

The years of Chancellor Nordenberg’s stewardship have produced a happy conjunction of three phenomena: Undergraduate instruction is now regarded as important at the University of Pittsburgh; art history is now accorded an important place in our undergraduate and graduate programs; and the University now values its interaction with the city of Pittsburgh. These are all positive developments for someone who teaches and lives here and who uses the city for part of his research.

There is a related phenomenon, too. On this campus we used to have what you can still find on many campuses in the United States—a sense of us versus them, with faculty on one side and the Board of Trustees on the other. But what I sense at Pitt today is a remarkable unity of purpose that binds together students, faculty, staff, and administration. That is one climate change we can all believe in, and one for which the faculty is grateful.

Art historians generally work with images, but I am speaking without any today because it’s just not possible to show you any significant part of my work and that of my colleagues. Our teaching covers the 5,000 years from Stonehenge to  skyscrapers, and my particular research involves a scope that is similarly extensive, from the founding of Roman Florence 21 centuries ago to the completion of Pittsburgh’s new Children’s Hospital a few months back. But the most powerful projector of images ever invented is the human brain, so you’ll have no problem keeping up with the images I will talk about.

My talk title proposes that art history is a science, even though when I studied the discipline at McGill and Harvard universities and began teaching it here, the three departments in which I worked all called themselves “Fine Arts.” It’s significant that all three subsequently dropped that name in favor of “History of Art.”

“Fine Arts” comes from the French Beaux-Arts, which alludes to the princely tradition of art collecting. But today even the French have dropped “Fine Arts” in favor of “Art History,” which is a literal translation of the German term Kunstgeschichte. There was an even earlier name for what my colleagues and I do: This was Kunstwissenschaft, an 18th-century German term that translates as the science of art. That earlier term is still amply used in Germany today. It’s the name of the art history department at the University of Bremen, for example, and it forms part of the journal title for the distinguished Marburger Jahrbuch für Kunstwissenschaft (The Marburg Yearbook for the Science of Art).

The idea of the “science of art” has not entirely disappeared in English, either. But, we think of it in the limited sense of detecting forgeries—a sort of CSI: Art History, as when a Leonardo da Vinci drawing was recently confirmed as authentic because his fingerprint on the back, in watercolor, matched a fingerprint in paint on the back of a documented Leonardo painting.

I regard this kind of investigation as low-grade science, if it is science at all, and it’s certainly not new. For the last 200 years, the first thing art historians do when they investigate paintings on wood panels is to analyze the wood itself: If the image is on cedar, it was painted in one part of Europe; if on poplar, it comes from a second region; if on oak, from a third.

These CSI: Art History-type investigations almost always involve the art market, which involves only a very few art historians. What I mean by the science of art history is that we art historians follow a rigorous intellectual matrix that parallels the way scientists work. Actually, what are art and science but parallel modes of thought? The Latin expression scientia means knowledge, and the Latin term ars means technique, so the two demand to be understood as complementary halves.

I regard everything I write as having a scientific basis, even though some of my books carry the dreaded adjective of best seller. Would it do any good to assure you that my newest book will never be a best seller? It is the first of four volumes on the archaeological excavations I ran under the Cathedral of Florence. The cost of the book is high—150 Euros [$219]—and it carries 122 pages of Latin, mostly untranslated, plus about 400 dense footnotes. But to say that these attributes make the book scientific is to fall into exactly the trap we need to avoid: the false idea that if a study is difficult to follow, then it’s scientific. It is neither difficulty nor obscurity that makes something scientific: It is the rigor of the method that it follows. Nor, incidentally, does entertainment disqualify a work as science. James Watson’s memoir The Double Helix, about his discovery of the structure of DNA, is highly entertaining, superbly well written, and educational.

So I would advance the heretical idea that even my recent Pittsburgh: A New Portrait is a kind of science. The long review in the Sept. 22, 2009, Wall Street Journal called Pittsburgh an architectural guide of a quality that few other American cities possess. That was kind of the Journal to say, but I do not see the book as an architectural guide at all. If it were, why would I have devoted so many pages to distressed neighborhoods like Larimer, Lincoln-Lemington, and Homewood?

No: I see the book instead as an analysis of Pittsburgh’s neighborhoods as urban organisms, as though they were each put under a microscope. I specify in the introduction what I regard as the marks of a flourishing neighborhood, and I state why various Pittsburgh neighborhoods are not flourishing. The book avoids nonscientific judgments like “Isn’t Shadyside wonderful?” or “I find the Hill so depressing.” (My personal sentiments—not in the book—run to the reverse.)

My method is visual or visual-spatial analysis, which I regard as scientifically valid even when it may not seem so. For example, when a scholar looks at a work of art and says, “I have a gut feeling this painting is by Rembrandt,” he or she may appear to be utterly nonscientific. But not so fast. The science of artificial intelligence tells us that the brain gets its answers by sifting through a computer-like comparison of images. Connoisseurship (the system that allows a scholar to declare an undocumented painting to be a Rembrandt) is consequently an important methodology when no reliable records are available. My stress here is on reliable, since even the valid signature “Rembrandt” does not make a painting a Rembrandt, because that painter had many assistants, and he occasionally signed his name to their paintings—a sort of Andy Warhol before the times.

Connoisseurship is not my specialty, but it paid off in my first book, for which I had to immerse myself in the work of James O’Donnell, an Irish immigrant architect working in New York around 1820. Researching in New York, I examined a house at number 2 Oliver Street. “That looks so much like O’Donnell’s work,” I said to myself, even though I knew those long-destroyed buildings only from prints and photographs. With due diligence I studied deeds and tenant records for that modest house in the stiflingly hot Municipal Records Building in Brooklyn, and I discovered that 2 Oliver Street was indeed a house O’Donnell had designed, some 190 years ago. The chances of such an identification were minuscule: 2 Oliver Street is one of just two Manhattan buildings of the era that can be definitively linked to a known architect.

So was it a “gut” feeling, or was it a simulation of a pixel lineup that allowed me to nail O’Donnell’s house? I’d say the latter: I was simulating what the computer does, but better, because I ignored the top floor of the house, which I could see had been changed, while a computer would have gone crazy trying to reconcile a roof of the 1860s with a house from the 1820s.

If it is a science, why did art history not become one of the “ologies,” like psychology or sociology? One reason is that “artology” is so clunky sounding, but another is that we existed so much earlier than those other disciplines. The first prominent art historian was Pliny the Elder, who composed his vast Natural History in the first century CE (CE being the neutral-based term for what we used to call AD). Pliny’s discussion of Greek sculpture and painting acknowledges as one of his sources a study written four centuries earlier by Xenokrates of Sicyon, so Xenokrates was the first real art historian, at least in the West.

I do not normally use the term “science of art history” because it is too late for it to be adopted now, and I try to keep my claims for the discipline modest. Medicine, after all, saves lives, while Rembrandt drawings do not. I do not even argue that art history is central to the humanities, as other disciplines claim for themselves. Nevertheless, in the four volumes I am currently publishing on Florence Cathedral, who but an art historian could work in all the disciplines I bring to bear, from liturgy, economics (13th-century economics, but economics all the same), and church politics, to archaeology and social and cultural history?

The geopolitics were very different but my method was the same in my first book, The Church of Notre-Dame in Montreal, set in the context of French Canada in the 1820s, and in my Fallingwater Rising, set in the context of the Great Depression of the 1930s.

Still, you are free to challenge me: Is art history truly important? I’d answer, “Yes,” because we now see the world of visual images as ever more complex. Howard Gardner’s 1983 book Frames of Mind: The Theory of Multiple Intelligences listed not just traditional intelligence (which Gardner simply calls “logical-mathematical ability”), but seven additional variations that also qualify as “intelligence.” One of these is the visual-spatial analysis that my colleagues and I practice. Years before Howard Gardner recognized visual-spatial ability as a valid measurement of intelligence, Rudolf Arnheim had already made the case for it in his 1969 book, Visual Thinking.

Howard Gardner may be right in individuating eight different kinds of intelligence, or he may be in error, but there is no question how powerful images are. I refer occasionally in my classes to the iconoclastic controversy of the eighth and ninth centuries, when suddenly, in Constantinople, certain Byzantine emperors attempted to ban and destroy all the venerated images of Christ and the saints, and thousands of monks were put to death defending the holy icons. For years we could say, “How barbaric those people were back in the eighth and ninth centuries, to cause bloodshed merely over some images.” But our self-congratulations of living in an enlightened century evaporated in September 2005, when the Muslim world erupted in fury after a Danish newspaper published 12 cartoons depicting Mohammed. A minimum of 100 people died in various Muslim countries in protests, and the Danish embassies in Syria, Lebanon, and Iran were set ablaze. This controversy also involved the United States, since Yale University Press deleted those cartoons from a recent book it published, in fear of Muslim backlash.

Visual-spatial analysis is no less important in our daily lives. A famous photograph that was published around the world in February 1968 showed South Vietnam’s police chief shooting a Vietcong insurgent at point-blank range on a Saigon street. That one photograph by Eddie Adams did more to make Americans question their involvement in Vietnam than did any single battle, won or lost. Or conversely, think of what more than anything else healed the rage lasting from that war: Maya Lin’s Vietnam Memorial on the National Mall in Washington, D.C. We now have millions of walkabout photographers in the United States, since cell-phone photographs are now routinely taken in situations like the 1991 videotaped beating of Rodney King that inflamed Los Angeles. The whole question of the propriety of taking and disseminating images is now a flourishing subset of the legal profession.

Art history is a science not only because it is the discipline that analyzes everything visual, but also because visual production is a uniquely valuable way of advancing information. It is commonplace to call Leonardo da Vinci a scientist as well as an artist, but we now realize that his anatomical drawings not only recorded the body, but also actually assisted in the advancement of medicine by capturing in line and wash what textual descriptions could not. Andreas Vesalius was indebted to Leonardo when he published On the Fabric of the Human Body, his pathbreaking 1543 volume on anatomy. Tellingly, the illustrations were entrusted to a student of the Venetian painter Titian.

And a related point: When researching the past, the visual arts have a huge documentary value when no other kinds of documents have survived. In the fourth of my volumes on the Florence Cathedral, I am revising or even rewriting centuries of Florentine history through visual evidence, since the textual record is so poor. Or take a parallel case in the history of medieval Bohemia, today the Czech Republic. In 1382, Princess Anne of Bohemia married Richard II, King of England. The childless marriage lasted just a dozen years, so it did not change English history. But it profoundly affected Czech history, since the religious reform ideas being propagated in England by John Wycliffe percolated back to Bohemia and were picked up by the Czech priest Jan Hus, who was put to death for heresy in 1415.

Being a secret transfer of heretical ideas, it is a matter of some conjecture exactly how Wycliffe in England influenced Hus in Bohemia. But while we learn little from the paper trail, we still have paintings that visually demonstrate that there were indeed tight links between the two countries in those dozen years, and later. What England notably exported to Bohemia was heresy, but what Bohemia notably exported to England was what we call the International Gothic style of painting. In fact, I regard the masterpiece of International Gothic to be the Wilton Diptych, most likely a portrait of Anne of Bohemia’s husband, Richard II, and quite possibly by a Bohemian artist that Anne brought with her to England. So images can bear witness when other witnesses remain silent.

Art historians also make themselves useful in judging images that have been manipulated for propagandistic objectives. David King’s 1997 book The Commisar Vanishes documents how Joseph Stalin had his early collaborators cropped out of official photographs after he purged them. (Stalin’s underlings probably labored weeks to excise those nonpersons from the negatives; today we can do it in five minutes with Photoshop.) Elsewhere in the 1930s, Robert Capa fed American and European outrage at the suppression of the Loyalists by the Fascists in the Spanish Civil War with a photograph of the death of a Republican soldier. But this photograph, too, has been shown to have been faked.

Every political power has manipulated visual imagery, from Hitler back to Napoleon and Emperor Augustus. I personally have an investment in two manipulated images that form the core of two books I am working on as future projects. One is the bronze statue of St. Peter in St. Peter’s Basilica in Rome: There Peter sits enthroned and stiff, with one extended foot worn smooth by pilgrims who have kissed it for half a millennium. Scholars always regarded the St. Peter as a rare survivor  of Early Christian art from the fourth century, until the restorer Bruno Bearzi analyzed the bronze and found it contained alloys only used in the 13th century. Now I and other specialists are convinced that St. Peter was cast by a certain Arnolfo di Cambio, about whom I recently published a long paper in Italy. Now my task is to figure out how Arnolfo so cunningly fashioned his statue so that a work of around the year 1290 came out in the style of around 390—and why. The answer presumably lies in what propaganda value the Pope of the time was trying to achieve with this exceptional commission.

Another image I am working on is Las Meninas, the masterpiece of the 17th-century Spanish master Diego Velázquez and one of the half-dozen great paintings on earth. In this huge painting, the princess Margarita, about 6 years old, stands at the center, with courtiers left and right. Velázquez himself paints at an enormous easel in the corner, recording the scene. This is not a manipulated image, exactly, but it is certainly a manufactured image, since no such gathering ever took place.  My task here is to prove my theory of why the queen of Spain—not the king, contrary to usual custom—got Velázquez to paint it. (My theory involves incest and a bastard half-brother who sought to marry the princess and grab the Spanish crown; it’s not for the faint-hearted.)

All the images I have discussed above—Pittsburgh, the house in New York, Florence Cathedral, the Wilton Diptych, the St. Peter statue, and Las Meninas—all these subjects of past or future research involve the biggest change to come over art history in recent years, which is technology. In interviews, I am always asked what is the biggest difference I see in Pittsburgh between the book I published in 1986 and today, but the real difference between 1986 and 2009 is not in what I see but in how I work, with computer applications. Around 1995, I urged the Digital Research Library at Hillman Library to digitize all the street atlases of Pittsburgh, from the 1870s until the 1940s. Digitally speaking, Pittsburgh is now the best-recorded city in the United States, so that (as I say in my book) my students and I can get answers in minutes to questions that took me a whole day to resolve in 1986. My research has been profoundly changed through computer applications: textual, visual, and graphics elements; computer-aided-drawings for my illustrations; dictating directly to my computer (since 1995); and digitizing thousands of photographs. In terms not just of speed or work capacity, but also in the quality and depth of the answers I can now get, the cybernetics revolution has far exceeded any science fiction I might once have dreamed of.

This means I have great responsibilities, too. I am obliged to keep exploiting technology, and, in everything I do, I am obliged to move to a deeper, more essential level of work, going from mere facts to insight, which I think is the true distinction between my 1986 and 2009 books on Pittsburgh. Not only can you not go home again (as Tom Wolfe tells us in his novel), you cannot stand still, either.

The New York Times recently carried an absorbing interview with Sir Tim Berners-Lee, the MIT professor who assembled the core software that drives the World Wide Web. Berners-Lee believes that making information and online tools freely available has to free innovation. “If you liberate the data,” Berners-Lee says, “who knows what applications people will create” That is exactly how I, working in Allegheny County, can make research progress on Las Meninas ahead of my colleagues in Spain.

Research and teaching have never been more exciting or challenging, and the science of art history plays an important part in it—not least at the University of Pittsburgh.