Deaccessioning John Constable:
presented in the session
Annual Meeting of the College
Although this paper includes a good deal for Constable experts, the main purpose is not primarily to contribute to Constable studies but rather to use the experience of over thirty years of questioning the authenticity of every work attributed to Constable to make a case for a general concept of authenticity, which I hope will be of service to this session as a whole.
Photo - Photo
Like any respectable art historian, I was initially amused by the audacious claim in this advertisement: "Every Treasure . . . a Certified Genuine Historical Artifact." No doubt the catalogue did include an occasional miner's pick from gold rush days, but what of all the imitation merchandise? As I reached for the phone to contact the Federal Trade Commission, it struck me that the claim was true. Every object in the catalogue was indeed made by human beings at some time and place. Every object was quite literally "a genuine historical artifact",
Nothing has brought home to me more strikingly the all encompassing nature of the concept we are attempting to address in this session. Everything is what it is. Therefore, everything is an authentic something.
Although this statement may initially seem to avoid the issue, I prefer to think of it as a clarifying insight with practical application for our day-by-day work as art historians. Because everything is an authentic something, our question should be not "Is it authentic?", but "What is it?"
To take Constable as a case study, which I shall be doing throughout this paper, no scholar has ever doubted the authenticity of this oil sketch. It was almost certainly in the artist's studio at his death, and was inherited by Miss Isabel Constable, his last surviving daughter. In 1888 she donated it, along with nearly four hundred other items by her father, to the South Kensington Museum (now the Victoria & Albert), as a permanent memorial to her father's achievement. Isabel cared deeply about and was intimately familiar with her father's art and can be assumed to have given only works which she considered by her father and, with few exceptions, representative of his best work. Add to this the facts that the sketch agrees almost exactly in size with nine other similar oil sketches and that this sketch is inscribed in ink on the back of the paper on which it is painted, in Constable's distinctive hand with a characteristic inscription. Above all, of course, the sketch itself is visually so similar in conception and technique to other sketches by Constable, and is so different from any oil sketches reasonably attributed to any other artist, that all scholars would accept it as an authentic Constable without question, even if it were to be uncovered in someone's attic today, on different size paper, without inscription or impressive provenance.
Since the time it was painted in 1821, it has been subjected to minimal intrusion. The fact that the sketch is on paper that has not been fixed to another surface has spared it's vibrant brushwork the compression that, until recently, regularly resulted from relining.
Indeed, it is difficult to imagine a work by Constable embodying more richly the complex attributes of authenticity: conceived by him, painted entirely by his own hand, and with his own hand documented on the object itself, passing directly from the artist through his daughter to the Victoria and Albert Museum collection, a type of sketch recognized as one of his distinctive contributions to the history of landscape painting; relatively unfaded and relatively untouched by other human beings.
But what of this Salisbury Cathedral in the permanent collection of the Metropolitan Museum of Art? Is this an authentic Constable?
Because the painting is unfinished, it is possible for us to see, even under normal gallery conditions, the meticulous build-up characteristic of Constable's finished paintings of the 1820s.
We see this even more clearly when viewer in the conservation studio as here. Constable generally began work on the canvas with a ground and light toned imprimatura, followed by pencil underdrawing and thin underpainting. He the put in areas of relatively flat, opaque paint; then developed the form with touches of impasto; and finally detailed the architecture and exquisitely finished, vibrant figures. In this painting, all of these are visible on one canvas - a crucial document for the study of Constable's technique. But this painting is not entirely in Constable's hand. During the mid and late 1820's, Constable was joined in his studio by Johnny Dunthorne, the son of Constable's early sketching companion in his home village of East Bergholt. Johnny became the only other human being who, as far as we know, Constable ever allowed to work on his paintings. In Constable's correspondence, we read of Johnny “squaring,” “tracing,” and “putting in outlines” for Constable's paintings. Restrained pencil outlines, unlike Constable's own, but copied from Constable's earlier versions of the same scene, are visible in this painting under infra-red reflectography, and are clearly I think by Dunthorne.
The initial buildup of forms in the unfinished portions of the Met. Salisbury is much too even and workmanlike for Constable, as in this area of the cows, indicating that the underpainting also is by Dunthorne.
In contrast, the final vibrant painting in finished areas, (such as this detail of Dolly Fisher with an umbrella and surrounding foliage) is by Constable himself.
Is this painting an authentic Constable? Graham Reynolds and I think so and have independently published it as such. Robert Hoozee, in his 1979 complete catalogue of Constable's paintings included it in his group of "doubtful works." But is this really a disagreement or merely a difference in catalogue categories? The painting is a collaborative effort of two artists, and the specific personal relationship between the two and their individual contributions to this painting can be specified in considerable detail. Label it what we will, this is what it is.
And what of this mezzotint at the Fitzwilliam? Is this an authentic Constable? Unlike the previous example, the hand here is not Constable's at all but rather that of his collaborator, David Lucas. But is the hand always our primary concern in authentication? What matters is not whether or not we label the object "authentic" but that we understand, as intimately as possible, what the object is. Here we know this in considerable detail. During the last decade of his life, Constable determined to produce a series of prints with accompanying text to explain the central concerns of his art to a broader public. These prints constituted a kind of pictorial review of his subject matter and emphasized what he called "the chiar'oscuro of nature," the changing play of light and dark over the landscape, which he felt gave natural landscape its essential character. With this in mind, Constable chose an able printmaker, not quite thirty years old, to collaborate with him in producing a series of mezzotints, the print medium most condusive to dramatic effects of chiaroscuro. Constable provided Lucas with a variety of drawings, sketches, and finished paintings, as the basis for the mezzotints.
In this case, he provided a powerfully expressive oil sketch (in a private American collection) probably painted especially for the mezzotint, but the sketch supplied few specific details as a starting point for the realization of the print.
Photo - Photo
Based on this, Lucas produced progress proofs, such as we see here, on which Constable then touched extensively, with white chalk and pen and ink, indicating changes he wished Lucas to make, which we see more clearly in this detail.
Even after it's publication in the first edition of English Landscape Scenery in 1830, Constable and Lucas continued to make changes.
In the 2nd edition, published three years later, we see additions such as the fisherman at the lower left and the introduction of Dedham Church.
The development of the print followed a process roughly comparable to that through which Constable typically elaborated initial oil sketches into finished exhibition paintings, and the letters and notes exchanged between Constable and Lucas demonstrate the degree to which Constable controlled the evolution of the print and, indeed, educated Lucas in his original view of landscape.
Andrew Shirley's classic catalogue of the prints is titled The Published Mezzotints of David Lucas after John Constable, and this traditional title has been followed until the most recent publications, when, happily, the prints have been described as "the Constable-Lucas mezzotints." Perhaps because of practical constraints, Graham Reynolds' authoritative catalogue raisonné, The Later Paintings and Drawings of John Constable, includes only the occasional drawing on the margins of some of the progress proofs, as on this proof of a different print.
But surely, not only the marginal drawings, but all of the proofs touched in Constable's hand, and, in my view, also the finished, untouched prints themselves should be considered integral to Constable's art and eventually incorporated in the catalogue raisonné.
Other hands were not always under Constable's supervision.
This early academy study (currently in the collection of Richard Constable, great-great grandson of the artist) initially seems to superimpose, rather tantalizingly, two male nudes with what appears to be a female nude; but this is clearly not Constable's doing.
The reclining nude, which is in Constable's hand, is part of an earlier, horizontal study, covered by Constable in order to reuse the canvas. In cleaning the vertical study, an early restorer clearly uncovered the third figure, which still retains small areas of the olive-like overpainting, then stopped when he got in trouble. The sketch may be entirely in Constable's hand, but the most striking thing about the image as we see it today, the superimposition of male and female figures, was never intended by the artist. Constable regularly reused canvases, but, as far as I can tell, never allowed his earlier image to show.
The slides on the screen were taken in 1981 in the owner's home in Somerset. In it's recent 1988 cleaning here in New York, the two studies were slightly disengaged spatially, by toning the reclining female and fully cleaning the two males, and (I feel regrettably) by enhancing the features of the standing figure.
We have now a third image, in which at least two restorers have been active participants. Is what we see authentic?
The deaccessing of a painting, because it is not quote "authentic,” often results in the object's banishment to an inaccessible storage area, where it cannot participate in the ongoing process of reevaluation of either the painting or the artist to whom it was previously attributed. Such was the case with this copy (at the Museum of Fine Arts, Boston) of Constable's 1819 exhibition painting, The White Horse, now in the Frick Collection. Identified on a label on the back as a study by Constable, for his six-foot exhibition picture, and purchased as such for the Museum of Fine Arts in 1895, it was gradually recognized as a copy and relegated to the stacks in basement storage, from which I finally managed to quite literally dig it out, with the help of a curator, in 1980. Although somewhat faded (we see the blue strip that has survived beneath the frame along the top), it is clearly by the hand of Johnny Dunthorne and, in my view, the most accurate and sympathetic copy of any Constable painting. In contrast to Constable's six footer, Dunthorne's copy is 4 1/2 feet long.
Done as an apprentice work, under Constable's close supervision, it mimics (except in size) Constable's own attempts at near facsimile copies of Claude during these same years.
Photo - Photo
Dunthorne's attempt at a near stroke-by-stroke copy is clearer when we compare details. We see Constable's original, with signature and date, on the right; and Dunthorne's copy on the left. As the only Dunthorne copy of a Constable painting so far identified, it will serve as our best guide in identifying Dunthorne's hand in Constable's paintings, produced for the French market at the time of his Paris success in the mid-1820's.
This Yarmouth Jetty, at the Tate Gallery, is I think probably entirely in Dunthorne's hand - even though his name is not mentioned in Leslie Parris' three page entry in his otherwise excellent 1981 catalogue of The Tate Gallery Constable Collection .
At the far end of the progression we have been following are works attributed to Constable in which he had no direct involvement at all.
Photo - Photo
These include, of course, deliberate fakes, a difficult category to establish because, unlike a copy, a fake presumes the conscious intention to deceive. Yet one this group of crude oil sketches and large pastiches of Constable subjects, all by the same hand and many deriving from the studio of James Orrock, seem to me to qualify. Judging by Orrock's own paintings, these are not in his hand but rather commissioned works from an as yet anonymous forger.
As a sampling, I show you these two sketches at the Castle Museum, Nottingham.
Photo - Photo
In addition, here are two slightly less obnoxious attempts to imitate
Stour Valley canal scenes, the one on the left at the Indianapolis Museum of Art, the one on the right in an American private collection.
Everything is an authentic something; and so, although it may be difficult for us to say the sentence, "These are authentic fakes."
This brief progression of examples, from which Constable has become increasingly absent, is meant to demonstrate a few of the varying degrees and types of involvement of a major artist in works that have been attributed to him.
The categories "authentic' and "not authentic," (even if we add the category "uncertain or questionable") are too generalized for getting at the widely varying relationships of artists to the objects associated with them. And the varying usage of the term "authentic," forced on us by the range of meanings it must encompass, makes comparisons among judgments of scholars often misleading, not to mention confusing to the public. The examples we have just reviewed make clear that we need to study and describe the frequently complex authorship of each object as fully as possible, without being overly concerned with the categories “authentic” and “not authentic.”
This requires us to distinguish between the often complex authorship of a work and, on the other hand, what we choose to value. The key to understanding the concept of authenticity is not to confuse the two. What human beings choose to value about authorship varies not only from culture to culture and from time to time, but also within a given culture. We should recognize and respect the varying types of authorship that have been valued at other times and places. In our own time, we as scholars, should not allow the market-driven and media-driven need for a simplistic and reassuring definition of "authenticity" to be so controlling. In our research, we should pursue, and in our publications and exhibitions we should project a richer and more complex view of authenticity, corresponding more closely to the nature of the works of art themselves.
In the first portion of this paper, I have attempted to review a few of the complex ways in which individual artists have been involved or not in the works of art attributed to them.
It would be equally instructive to review the complexity of evidence on which our reading of that involvement is based. The history of Constable studies provides revealing examples to demonstrate: first, that there is no single piece of evidence that can establish the authenticity of any work of art beyond question; second, that all evidence is incomplete; third, that each piece of evidence must be judged in the context of the date of and type of object on which it appears; fourth, that any new evidence about any work by the artist affects the balance of weight for evidence of all objects by the artist; fifth, that new evidence and new contexts for the evidence is unpredictably emerging; so that sixth, we are constantly reevaluating the degree of reliability of any evidence and of the form in which it has been perceived, recorded, and interpreted in the past; and finally that, for every object we are considering we must constantly hold in suspension a large number of separate pieces of evidence, of different types, with different degrees of persuasiveness, balancing, for example, what we know of the artist's personality against evidence from paint samples, and the object's provenance, and much else.
To witness the complexity and interdependence of this evidence in Constable's work would require examination of a much more diverse selection of examples than we can manage in so short a paper, but let us at least trace this process of continuing reevaluation in one major painting, a picture with no documentary grounding and with conflicting visual evidence.
We see here two paintings by Constable, at the right Wivenhoe Park, made famous as the centerpiece of Ernst Gombrich's Art and Illusion , and, at the left, the painting now accepted by all scholars as Constable's Full-Size Sketch for The White Horse ; both on display in the National Gallery of Art, almost ten years ago (April 1984). Two months later, at my request, the White Horse sketch was taken to the conservation lab for study and, quite reasonably I think, has not been on public display at any time since.
Although Constable's authorship of the White Horse sketch is now accepted by all scholars, as recently as the late 1970's it was doubted by three leading Constable experts. In their catalogue of the 1976 Constable bicentenary exhibition at the Tate Gallery, the organizers, Leslie Parris and Ian Fleming-Williams, wrote: "What is usually supposed to be a full-scale study for 'The White Horse,' a painting in the National Gallery of Art, Washington, does, however, have some curious features which suggest that the question should be looked at again," and indeed they did not request the painting for the exhibition, even though the finished painting, at the Frick, cannot be loaned, so that Constable's first six-foot exhibition painting went unrepresented.
The same year, in his book on Constable's Landscape Watercolours and Drawings , Fleming-William omitted the Washington sketch from his list of Constable's full-size oil sketches. In his 1979 catalogue of all Constable's paintings, Robert Hoozee placed the Washington painting in his category of "Dubious Works" and wrote: "Considered the full-scale study for [The White Horse], it is most probably an imitation."
That year, 1979, I was able to examine the painting for two days in the excellent light of the galleries used then as the textile conservation lab (where this slide was taken), the first time I was able to examine the painting out of its frame.
There were indeed reasons to be skeptical. There is no provenance for the painting previous to its first mention in 1872, over 50 years after its supposed date of creation, and a time when we know many imitation and fake Constable's were coming onto the market. It is especially surprising that there is no mention of so large a sketch for The White Horse in any known document during Constable's life, because we have extensive correspondence between Constable and his close friend, John Fisher, who purchased the exhibited painting (now at the Frick) immediately following its first public showing at the 1819 Royal Academy exhibition. The finished exhibition painting is discussed at length in their correspondence but there is no mention of a large sketch.
There are also problems topographically. We may accept the fact that the fence at the right, dovecote beside Willy Lott's House, and horns on the cows do not exist in the exhibition painting. But it is less easy to accept the position of the boatshed in the sketch, because Constable was meticulously accurate in his topography at this time, especially when describing his father's mill property at Flatford. In all other drawings and sketches of this scene, the boat shed appears in the position seen in the finished painting. Most problematic (and I suspect the deciding evidence for those who rejected the picture) was the form of Willy Lott's Cottage. This cottage appears in many drawings, watercolors, oil sketches and finished paintings by Constable, seen from several different angles. With the exception of a few very late paintings, in which he aggrandized the size of the cottage, Constable always represented its form accurately, as we can still confirm on the site today, where the cottage has been preserved with minimal restoration. The view here, from the south-west, was one of Constable's favorites, and he always represented the basic form of Willy Lott's cottage form this angle as we see it in the Frick painting - except it seemed in the Washington sketch.
A final objection, not voiced by the critics at the time, but one that had concerned me greatly before I had the change to examine the picture carefully, is that much of the handling seems crude for Constable, especially in comparison with his other confident six-foot sketches.
In the entry in his 1984 catalogue raisonné of the later paintings and drawings, Graham Reynolds reaffirmed his belief in the authenticity of the Washington sketch, calling attention to the heavy discolored varnish, but attributing to Constable all of the features that appear different in the final painting, including the incorrect and eccentric form of Willy Lott's cottage.
Up until this time, no one had mentioned, certainly not in print, the possibility that parts of the painting might be by Constable and other parts by other hands. Both critics and supporters wrote as if the painting must stand or fall as a whole.
Then early 1984, just in time for a "stop press" notice in the catalogue rainsonné, we found evidence that at least the early layers of the sketch were convincingly by Constable. At the end of a blessed year at CASVA, I managed to get the picture into the conservation lab in the National Gallery for extended study.
The x-radiographs (which we see here taped together on a lightbox) revealed immediately, underneath, the presence of an earlier, unfinished painting of Dedham Vale from the Coombs, previously unsuspected by anyone and not detected in the previous 1944 x-radiographs.
Photo - Photo
The exact correspondence of portions of the underlying image in the x-radiographs to the comparable portions of a Constable sketch in the V&A (notice the river hooking back exactly at the end of the roof, the chimney reaching exactly to the line of the river, the brilliant white at the end of the gable), this convinced everyone that the image underlying the White Horse Sketch was by Constable, and therefore that the White Horse Sketch also was almost certainly by Constable.
Photo - Photo
One other conspicuous detail in the x-radiograph was the detail of Willy Lott's cottage, which showed that under the eccentric form visible in the Washington sketch was the correct form, as we see here in the x-ray detail, just as it appears in the Frick painting.
In my article on the painting, for the National Gallery Studies in the History of Art (not published until 1990, but completed in 1984 and shared at that time with other Constable scholars), I distinguished three layers of paint: first Dedham Vale, then the White Horse, and on top later reworking and retouching.
The judgment that there was later reworking and retouching was based not just on a visual reading of the brushwork but also on microscopic examination of the surface of the painting, where we could see pronounced cracks filled with paint only from the surface layer, which therefore must have been added well after Constable had completed his sketch. (The examination of paint layers with binocular microscope and of paint cross-sections is one of the most consistently informative means of technical analysis of paintings, and protects us from discussing much later additions as if they were part of the original painting). In my article, I recommended a slow, studious cleaning, comparable to that carried out by David Bull on The Feast of the Gods. In 1992, Michael Swicklik, paintings conservator at the National Gallery of Art, began the long awaited cleaning, which he is carrying out with the highest professional standards.
In the remainder of this paper, I am much in debt to Michael and to Barbara Berry, conservation scientists at the NGA, who took and analyzed the recent paint samples.
After thorough research and study of related Constable paintings in other collections, Swicklick gradually removed most of the heavy layer of discolored varnish, to make possible a more reliable visual reading before any cleaning of the painting itself.
In this slide, taken in March 1993, not quite a year ago, we see the outermost coat of varnish removed, except for an area at the upper right, making clear the extent of previous discoloration.
The removal of varnish allowed Swicklick to study the surface of the painting more carefully and with more confidence. Over time it became possible for Swicklik to recognize the distinctive crackle pattern and differences in handling of the top layer of paint and to realize how extensive it was. Fortunately, new paint samples also revealed a thick layer of discolored varnish just under the top, finish layer of paint; that is, covering Constable's White Horse Sketch. Moreover, small cleaning trials showed that this outer layer of paint was quite different in composition from Constable's White Horse Sketch layer and separated rather cleanly from it.
Not all conservators would consent to having slides of a cleaning in process shown publicly; but Michael has lent his support as a mark of the increased openness of conservation and the recognition that the cure to the continuing misunderstandings of the profession is an educated art world and gradually an educated public.
To date, Swicklick has very carefully taken back the later overpainting in two areas: the tree area at the upper right, and the sky at the upper left.
Photo - Photo
In the left slide, we see the tree before even the discolored varnish was removed and, in the right slide, the upper part of the same tree as it appeared last week, with both varnish and overpaint removed. I have no doubt that what we see on the left is overpainted by a later hand, and that on the right we see what remains of Constable's hand, still with a bit of residue from the overpainting.
Let us be clear that the later finishing on the left is also authentic, and that if someone were to undertake a study of nineteenth century overpainting, important evidence will have been destroyed and available only in the carefully documentation accompanying the cleaning.
But I am convinced that the National Gallery of Art is wise to so prefer Constable's authentic White Horse Sketch, that it is removing the authentic nineteenth century overpaint; to allow us to see, for the first time since at least 1883, what remains of this landmark sketch.
Photo - Photo
As we see in this recently cleaned sky detail, the National Gallery of Art's six-foot sketch is emerging with cloud forms and colors that correspond more closely than anyone had imagined to the Frick exhibition picture.
I hope that the emphasis in this paper on the complexity of authenticity will make it possible for the concept to function more intelligently in the way we think about, write about, and treat works of art.