On Monday, 8 May, in Berlin, Schwartz heard a top connoisseur account for differences in finish between two paintings by Hugo van der Goes as acceptable variations within a single artistic personality, and on 12 May, in Den Bosch, heard another top connoisseur denying the very possibility of such a thing concerning two paintings by Jheronimus Bosch. What a week!
What circularity! Not so much concerning the question of whether a painting is accepted as the autograph work of a particular master or not. What’s circular is that the framing of the problem determines the nature of the conclusion, bringing you back to where you started. Have a look:
Stephan Kemperdick on Hugo van der Goes (ca. 1460-1482/83): “Compare the sheaves of wheat on the threshold of the Portinari Altarpiece [above] to the same detail in the Nativity. The one in the Portinari took ten hours to paint, the one in the Nativity was tossed off in ten minutes.” We were standing in front of the Nativity, in the Berlin Gemäldegalerie, where Kemperdick is curator of German, Netherlandish and French painting before 1600. The occasion was a CODARTfocus visit to the first exhibition ever devoted to the work of Hugo van der Goes, curated by Kemperdick and his junior colleague Erik Eising.
Hugo van der Goes, Adoration of the Magi (The Monforte Altarpiece), ca. 1470-75
Oak panel in original frame, ca. 170 x 263 cm
Berlin, Gemäldegalerie (1718). Click to enlarge
Hugo van der Goes, The Nativity, ca. 1480
Oak panel, 99.2 x 249 cm
Berlin, Gemäldegalerie (1622A). Click to enlarge
Owning two of the master’s largest and least transportable paintings, the Nativity and the Monforte Altarpiece, an Adoration of the Magi, the Berlin Gemäldegalerie is uniquely qualified to mount this exhibition.
“Compare any two equivalent details.” I do so, in Kemperdick’s spirit. Baby Jesus’s legs in the Monforte (left) are indeed firm and fully packed, rounded and modelled subtly in light and shade, his footsies waiting to be tickled. In the Nativity Jesus has flat feet, outlined rather mechanically, with toes that look like they were stuck into his feet. Having always gazed at the painting in awe, I was brought up short, confronted with my inattentiveness and my disinclination to look at old masters as if I were an art critic.
Jheronimus Bosch, The garden of delights, central panel, ca. 1500-05
Oil on panel, 190 x 176.7 cm
Madrid, Museo Nacional del Prado (2823)
Jheronimus Bosch – or not, The temptation of St. Anthony, central panel, ca. 1500 (if his), or after 1516 (if not)
Oil on panel, 131.5 x 119 cm
Lisbon, Museo Nacional de Arte Antiga (1498 pint)
Details of the above
Fritz Koreny on Jheronimus Bosch (ca. 1450-1516): “The goldfinch in the Garden of earthly delights could not be painted better. The observation of the bird was taken from life, with dedicated attention to the particularities of form and coloring. By comparison, the same bird in the Temptation of St. Anthony is far cruder, with roughly brushed in passages where in the Garden the greatest finesse is exercised. This comparison alone makes you look more critically at the rest of the painting.” This occasion was a three-day conference on Jheronimus Bosch, the fifth to be held in the admirable Jheronimus Bosch Art Center in the city that sometimes seems to be named for the artist.
Thus far, most of the gathered art historians in Berlin and Den Bosch could go along with the observations of Kemperdick and Koreny. But discussion did not stop there. We wanted to know how the two connoisseurs accounted for these differences.
Kemperdick, in accordance with general opinion, dismisses the possibility that the Nativity was painted by a different artist than the Monforte Altarpiece. He explains (away) the differences thus: “We know that in middle age Hugo came under great stress. He moved into a monastery as a lay brother, where one of his fellows wrote an account of the mental problems that afflicted him. One thing that was bothering him was the work load he brought with him. People said that he would need more than nine years to complete the commissions he had taken on. This led him to develop a more streamlined technique, taking shortcuts to finish paintings more quickly. That’s what we see in the Nativity.”
Koreny takes the differences he observes far more seriously. Braving prevailing opinion, he follows his observations to what he sees as their inevitable consequence: “The temptation of St. Anthony could not have been painted by the same artist who made the Garden of earthly delights. The Boschian work to which it comes closest is the Haywain. I assign it to an artist I called the Master of the Haywain, working in the Bosch workshop. He was not a bad artist; we could even call him more forward-looking than Jheronimus, not taking visual impressions so literally. I compare him to Edouard Manet, for his freedom of spirit. But he is not Jheronimus Bosch.”
Scholarly opinion in these two cases could not be more different. Concerning the oeuvre of Hugo van der Goes there has been a broad consensus for a century. That is all the more remarkable since not a single one of the paintings in that oeuvre is signed or documented as having been made by an artist of that name. The background of the artist himself was confused from the start. In 1550, in the first mention in print of any of the works ascribed to him – nearly seventy years after his death in 1482 or 1483 – Giorgio Vasari wrote that the Portinari Altarpiece was painted by a “Ugo d’Anversa,” Hugo of Antwerp. In 1604, Karel van Mander said he was born in Bruges. Other sources put his birthplace in Ghent, in the Zeeland town of Goes and as far afield as Leiden. As evidence, these statements are utterly valueless. Nonetheless, the conviction has taken hold that the Portinari Altarpiece could only have been painted by the prominent artist who is richly documented under various related spellings (none of them Hugo van der Goes) in Ghent in the years 1467-77 and in the Roode Klooster monastery from 1477 on.
While this gives us shaky testimony concerning the authorship of the Portinari Altarpiece, where do we go from there? The rationale allowing for the construction of an oeuvre by the maker of the Portinari is put into words thus by Catherine Reynolds, in her entry on the artist in the Grove Dictionary of Art: “The idiosyncrasies of the Portinari Altarpiece have left little disagreement about the other major works to be attributed to Hugo.” Disagreement has prevailed in the dating of his paintings and drawings, which resist being lined up stylistically and technically into a succession of works that could reasonably have been made by one artist in the assigned time slots. (The Berlin curators do not let this bother them.) The discrepancies thus revealed are nonetheless never taken as a reason to doubt that those “major works” – thirteen paintings and two drawings – were necessarily made by one artist.
Well, almost never. There is one art historian who respects the documentation more than his own take on artistic personalities. In the first volume of R.H. Wilenski’s book Flemish painters (1960), with its “historical survey,” we read all about the documented doings of Hugo van der Goes, while in volume 2, “Reproductions of pictures,” not one painting is captioned with his name. Only the Portinari Altarpiece is illustrated, ascribed, with deference to Vasari, to “Hugo of Antwerp.” Had he included the Berlin Nativity, he would have captioned it as the work of “The Master of the Berlin Nativity.” I often wish I had his nerve.
With Jheronimus Bosch something like the opposite prevails. Concerning the thirty-eight paintings attached more or less tenuously to his name – again, none documented in his time, but many signed – the leading experts agree on his authorship of fewer than half.
To sum up: when it comes to Hugo, the experts are so impressed by his idiosyncrasies that they are willing to explain away discrepancies between one painting and the next. Bosch’s thematic idiosyncrasies, on the other hand, can be found in so many paintings that the experts look not for resemblances but for distinctions, in which one goes further than another.
Without agreeing or disagreeing with Kemperdick or Koreny, I want to pose a question that challenges the certainty of both of them. It has to do with numbers. It has been estimated that more than ninety percent of the paintings made in the Low Countries in the fifteenth and sixteenth centuries, the time of Hugo and Jheronimus, have been lost. The figure may be over ninety-five percent. The surviving documentation tells a different story. Far more names of artists are recorded in authentic sources than oeuvres that can be attached to them. That ratio may also be above ninety percent. In the full panoply of Early Netherlandish paintings, we can be fairly sure, there will have been oeuvres to which all those divergent variations will have better belonged than to the small number of known masters to whom we give them.
What then are the odds that the connections we make between the surviving works and documents correspond to historical reality? Is there a statistician out there who would confirm my suspicion that our linkages of the paltry number of surviving works to a small fraction of the documented names have little claim to credibility, that the artistic personalities we live with are likely stories of our own creation? If there is, please speak up, so we can start retooling art history.
© Gary Schwartz 2023. Published on the Schwartzlist on 18 May 2023. With kind thanks to the Gemäldegalerie for the images of the Monforte Altarpiece and the Nativity.
My visit to Berlin with Loekie and my participation in the Jheronimus Bosch conference did not go all that smoothly. Getting around in Berlin with public transportation cost more effort and annoyance than it should have, especially all those stairways where there should have been escalators. On our return, I was able only to attend a little less than half of the Bosch conference. But the proceedings will be published, as have those of the previous four meetings of the kind. This is due to the exemplary patronage of the Bosch entrepreneur Jo Timmermans, who funds the Jheronimus Bosch Art Center where all of this takes place. In the meanwhile, Loekie and I are back at cask strength.
In Berlin I underwent another confrontation with myself. At the Humboldt Forum I went through the exhibition Berlin Global.
In a large variety of forms, including interactive features, and in a magnitude that I found overwhelming, I went from theme to theme, trying to make sense of things. The resolution of the displays varied from sound bites sketching a decade in ten lines to exhaustive coverage of single incidents. I decided that although a city like Berlin may be endlessly dynamic, I am not. City dwellers cannot inhabit more than a small part of the place’s conceptual as well as physical terrain. It’s a matter not only of human limitations but also of sheer self-preservation. Who can stand being confronted with it all?
The exhibition gives visitors a device that tracks their movements and lets them vote on charged issues. Some of the choices are imposed on you. You cannot move from space to space without choosing between alternative gates. This I found overbearing and didn’t like. I did like the videos with personal recollections and experiences, but to listen to them I had to hold up two earphones that after eight minutes turned out to be unexpectedly heavy. Two videos was my limit. A younger audience may be able to make better use of the offerings, which are undeniably rich and worthwhile, but I hope that the museum does not take this as encouragement to pile it on even higher.
On the top floor I expended the rest of my energy in the galleries of African art, which interested me in particular, since two weeks before Loekie and I had been to the Africa Museum in Tervuren, Belgium. (This was by way of preparation for attending a lecture in Brussels on 25 April by Debora Silverman, who has brilliantly sharpened discussion of artistic appropriation in the West of colonial seizures.) Both museums bend over backwards to apologize for even being in possession of the objects on display. Tervuren, to which more time and resources were dedicated, is a more accomplished and engrossing presentation. Both are provisional waystations in global and transhistorical developments that wish to mitigate past harms. I quote from an article on the Humboldt Forum by Ian Johnson in the current (25 May 2023) issue of the New York Review of Books:
The Humboldt Forum and most other universal museums around the world, however, still largely follow what [the Oxford archaeologist Dan] Hicks calls a “retain and explain” method: the objects are kept on display but given a sometimes hand-wringing origin story. As we walked around the exhibits, he said, “It’s a performance of white fragility. What should we do? Oh my.”
In fact, these explanations don’t heal the traumas but add to the sense of injury.
I’m not sure I agree with that judgment of Johnson’s. At least, you read in accounts of courtroom confrontations that victims even of violent attacks on their person are helped by the personal admission of guilt by the perpetrator. Shouldn’t this effect also take hold for most if not all inhabitants of depredated countries in Africa? I would say “Time will tell,” but Time has a way of telling completely different things than answers to the questions we think we are asking it.
Responses in the Reply box below (these will be viewed by all visitors to the site) or personally to Gary.Schwartz@xs4all.nl are always appreciated and will be answered.
So will donations.
Your donations help defray the costs of the Schwartzlist and encourage Gary Schwartz to write more columns.
5 thoughts on “417 The transparent connoisseur 7: Explaining away Early Netherlandish discrepancies”
I believe Charles Darwin once said, when talking about how to tell two species apart, when often species of plants and animals are visually similar, that the best distinction is whatever a good naturalist decides. Replace species with attributions of art and naturalist with connoisseurs and I believe one ends up at the same point of logic, that distinctions are not absolute but relate to perceptions. The more informed and intelligent the connoisseur, the more likely it is that their idea of what an artist did corresponds to what an artist actually did. However, the correspondence is always a likelihood rather than absolute; even a historical record from the time of the object begs the question of why the recorder believed their record to be accurate and why we assume the record is a certainty. Our conception of the artist affects the weight we give to any historical record, which in turn may change our conception of the artist, and we must always be careful not to hedge an artist’s intellect irrevocably round with our own. I conclude that all attributions are actually the result of connoisseurship, which may seem an unsatisfactory and ‘unscientific’ method but is actually the only method open to us. Therefore recent attempts to use image analysis, core groups, etc (and even historical records) to be more ‘scientific’ are only scientific when used simply as extra information for consideration by the connoisseur.
About Darwin, allow me to quote from an unpublished manuscript of mine:
In the history of attributionism one name above all others comes to the fore. Giovanni Morelli (1816-1891) was the most forthright proponent, starting in the 1870s, of an analytic system of attribution, which since then few have tried and none has succeeded in replacing. Listen to this orotund judgment by Claude Phillips of 1890: “It would be as absurd to return to a pre-Morellian period of criticism, as it would be to study natural science without profiting by the discoveries of Darwin.” A year earlier, Bernard Berenson (1865–1959) piled the snow even higher, writing that Morelli’s “services to the science of pictures are greater than Winckelmann’s to antique sculpture or Darwin’s to biology.”
The well-phrased comments by John Brown and Martin Royalton-Kisch I see as idealized versions of acce[ted wisdom. But what if accepted wisdom is demonstrably erroneous? What if it can be shown that the artistic personalities to which we attach Early Netherlandish paintings are constructs based on insufficient information? Are we not then obliged to rephrase out question so that the answers can be closer to the statistical (I wouldn’t call it “scientific”) likelihood? This would bring us closer to R.H. Wilenski.
I am reminded of the current debate about the book by Rosemary Sullivan, “The Betrayal of Anne Frank: A Cold Case Investigation.” The question posed there was: In light of available evidence, who is most likely to have betrayed the hiding place of the Frank family? Naturally, an answer came out. The most likely person, the investigators concluded, was a certain Jewish notary. This chain of reasoning remains upright even if the Franks were not betrayed, but uncovered through a chain of circumstances. That is what other researchers think happened. Upright – but if the critics are right, utterly false.
That is in my opinion what connoisseurs are doing with debatable cases – asking who is the most likely perpetrator. There will always be an answer. But if it is likely to be unbased, shouldn’t we look critically at ourselves and ask whether there is a more crediable tack to be taken?
Thank you indeed for a very interesting article!
Many thanks to Gary for the nice article and to you for your interesting contribution, which holds many of the keys to what we are talking about, and I like the Darwinian analogy.
The connoissurship topic is a vast one, ranging form history to psytchology and much else besides, but it’s correct to take the latest and most informed/respected view(s) into consideration before attempting to refute or alter them. There’s a bit about the problem in the introduction to my website at http://www.rembrandtcatalogue.net, and history suggests that we do gradually move closer to the truth with each generation of experts, even if there will never be unanimity about a particular attribution in many cases. But the website also carries a disclaimer, based on the RRP’s, that states “The opinions expressed by the author are just that – opinions. Formulating opinions about the authenticity of Rembrandt’s work (and work by other artists), and perhaps especially of his drawings, is often extremely difficult, and older opinions are frequently revised in the light of new facts or insights that emerge later. No warranty of any kind is attached to the opinions expressed here, except that they constitute the author’s honest assessment at the time of writing. I do not accept any liability for damages and anybody is free to disagree with the opinions expressed here.” If anyone thinks something else needs to be added, please let me know.
In the article I would completely agree, from what I can see in these images, with the opinions of both Kemperdick and Koreny…unless or until new facts or insights should emerge. But connoisseurs need to be aware, too, that differences in touch, even of works of the same period, do NOT always mean that they are by different artists. With Rembrandt we have a number of ‘documentary’ drawings, connected with other works or even indented for one of his etchings, all made in chalk and that we think we know all date from within a short period, but differ sustantially (Benesch nos.7, 12, 15, 20 and 21, all from around 1628-31). There are similar cases with his paintings and etchings. In the studios of the best, like the finest chefs in their kitchens, experiments are regularly undertaken that push the boundaries of the known or possible, in ways that can end up in failure – the sunken soufflé or the rather bad drawing (some of which might survive the dustbin).
At all times, as history endlessly reveals, connoisseurs beware!
Martin, the RRP did not start out thinking that they were offering mere opinions. In the introduction to vol. 1, they wrote: “Our attempt to define and purify Rembrandt’s oeuvre amounts to an effort to find rational, communicable arguments to support our opinions.” Only in vol. IV did a lengthy disclaimer appear, beginning “This is a publication of the Stichting Foundation Rembrandt Research Project. The opinions expressed in this volume (IV), and the previously published volumes I-III in the Series A Corpus of Rembrandt Paintings, should be understood as ‘opinions’ that are meant for academic use only.” I see this as a defensive retreat on the advice of lawyers. Prudent, as you are being, but at odds with the mentality of the Founders. And I do not agree that we “gradually move closer to the truth with each generation of experts.” We may be moving closer to consensus, but as I argue in my response to John Brown, this has no necessary relation to the truth.
With kindest thanks as ever,
Also, I do not see the constant shifting of opinion concerning the authorship of Rembrandt paintings as a