Menu Close

12. Summary of Evidence and Conclusions

The “Nantucket painting” is an exact duplicate of Cézanne’s well-known view of L’Estaque in the Musée d’Orsay (RF2761=FWN 119), painted in 1878 or ’79.  The two paintings differ slightly in size and platform.  The present painting is done in oils on a paper sheet glued to stretched canvas, and FWN 119 is done in oils on stretched canvas.  The first appears to be a preliminary study for FWN 119, for FWN 119 is the more highly finished.  Both paintings appear to be the same age, and the stretchers of each are made from identical store-bought components (See Part 5 and figs. 38a, b).  The subject of the painting, the seaside village of L’Estaque opposite Marseille, was one of Cézanne’s favorite retreats.  He resided there for many months at a time during the 1870’s and ‘80’s and painted or sketched some sixty views of the locale (See Part 9).

Fig. 38a: (Left) the stretcher of FWN 119 (photo courtesy of the Centre recherche et de restauration des musees de France; © C2RMF-36795). (Right) the stretcher of the Nantucket painting, shown at about actual relative scale. (Fig. 38a: left,

 Fig. 38 b. Comparison details of the upper left corners of the two paintings (Left, © C2RMF-36795).

What makes this pair of paintings remarkable is that one appears to have been carefully traced from the other by means of a camera lucida (See Part 4). One painting (FWN 119) is a well-documented work by Cézanne.  The other has been unknown until now.  The latter, thus, poses an important question:  Is it a long lost preliminary study by Cézanne for FWN 119, or is it a copy of FWN 119 by an unknown imitator of the artist seeking to understand Cézanne’s painterly technique (as the CR Committee has claimed)?  Whichever case is true, it is certainly clear that at one time both pictures were in the possession of the same artist.

Fig. 39 a.  FWN 119 and the Nantucket painting, both prior to cleaning.

Fig. 39 b. FWN 119 and the Nantucket painting after cleaning.

After reviewing the data sent them in January 2022, the Cézanne CR Committee of Walter Feilchenfeldt, Jane Warman, and David Nash responded with an opinion stating that our painting was not by Cézanne but was only a copy of FWN 119 by an unknown copyist “trying to understand Cezanne’s methodology” (See Part 2).  This position, however, is difficult to defend for the following reasons:

  1. There are no known copyists of Cézanne between the period 1878-79  (when FWN 119 was painted) and 1895 (i.e., when the first public exhibition of Cézanne’s works took place in Paris). In fact, there are  no documented copyists of Cézanne at all until after his death in 1906 – or, in fact, after his posthumous retrospective exhibition in 1907 in Paris at the Salon d’Automne (See Part 4).
  2. Had there been a copyist of Cézanne before 1895, it is hard to imagine what this individual’s interest would have been in making an exact, traced copy of a painting by Cézanne, an artist who was then very little known and poorly regarded (See Part 9).
  3. It is even more difficult to imagine a copyist wanting to duplicate FWN 119 at all, for between 1879 and 1894 this painting was owned by Gustave Caillebotte, and it hung in his house outside Paris. The copyist would have had to paint the picture there, setting up a camera lucida (See Parts 2 and 4).
  4. If the copyist’s aim was simply to study “Cézanne’s methodology,” as the Committee has suggested, why does our painting exhibit so many differences from FWN 119? Although the two paintings share the same composition, exact in every detail, our painting is not a passive copy of the other.  In fact, it does not appear to be a copy at all!  FWN 119 is highly finished; ours is not.  FWN 119 was painted on canvas; ours was painted on paper.  Whoever our artist was, he added corrections to his picture, even after the painting was varnished.   The same details do not appear in FWN 119!
  5. Why would someone copying another painting varnish his work and then, after the varnish was dry, go back to it and overpaint it rather clumsily? The heavy, uneven lines of blue and grey oil paint, carelessly applied to the sea surface along the near and far shorelines, actually mar the appearance of the picture.  If the artist had been copying the highly finished FWN 119, he would never have added such “corrections.”  There would have been nothing to correct!  Such details are without parallel on FWN 119 and indicate that our painting, thinking logically, must be the original version of the composition and FWN 119 the copy.  This would seem to prove that Cezanne painted both pictures in his studio in L’Estaque and that both were painted during the period 1878-’79.)
  6. Another novel feature of our painting is the heavy, rippled impasto primer layer of white paint applied as the base layer of the sloping foreground, which was applied with a palette knife. This layer (See Part 3) also does not exist in FWN 119.  If the painter was only a copyist, desiring to copy FWN 119 in Caillebotte’s house, why would he have added this heavy primer which was not present on the other?  Obviously, it served as a sealant for the paper, but it would have taken days or weeks for this layer to dry before the artist could have begun painting the pictorial details from the other painting.  This hardly sounds like the behavior of a copyist!  (See Parts 2 and 3).
  7. Then there is the brushwork. When the surface of our painting was dry, the artist added to it a coat of natural resin varnish (See Parts 2, 6 and 8).   When applied, this varnish would have been clear in color, but, in its fully oxidized state, we can see that he applied it in a pattern of short, diagonal brushstrokes. If this painting had been executed by an artist other than Cézanne, is it likely that, to apply a clear varnish, he would have emulated Cézanne’s distinctive personal brush technique?  It seems doubtful.  For all these reasons we have to suspect that the artist was Cézanne himself.

Evidence supporting the identification of the artist as Cézanne:

  1. The subject of the “Nantucket painting” (a view of the sea at L’Estaque) ties the picture closely to Cézanne. Logically, only Cézanne would have been so attached to this place (some 760 km south of Paris) and so obsessed with his art that he would have wished to create more than one exact versions of the same view (See Parts 6 and 9).
  2. Cézanne, in fact, had a well-known habit of making multiple copies – even exact tracings – of his own works (See Part 4), so the existence of two or more exact versions of the same composition, made with a camera lucida, only more strongly points to Cézanne as the artist. (Caitlin Haskell of the Art Institute of Chicago, in her joint YouTube lecture with Gloria Groom, “Reflections on Cézanne,” comments on the artist’s peculiar “desire of painting pictures again, painting pictures after pictures, reworking a picture that he has already done before, trying that again and learning from it…”). (See Part 4).
  3. To discover which of the two paintings is the original and which is the traced copy, one need only look carefully at both paintings to find evidence of traced outlines. Our painting shows no such outlines, but FWN 119 shows many.  For example, we can see black lines outlining the profile of the hills and the horizontal bars along the shoreline (i.e. the stylized Marseille waterfront), and also the diagonal quay (fig. 40), or the smokestack of the village brick factory, the railroad tracks, and the vineyard at right  (figs. 43 d-e).  These outlines seem to offer proof that FWN 119 was the traced version and that our painting was the primary work.  (For comparable traced “twins,” see the duplicate portraits of Mme Cezanne, in which one shows traced lines: Part 4, fig. 8b)

Fig. 40.  The hills above Marseille in the Nantucket painting (top) and FWN 119 (below), in which both paintings are shown in their pre-cleaned state.  Here, in FWN 119, the traced lines are clearly seen.  Note that a dark line follows the contours of the hills; dark lines also outline the shoreline features and the diagonal quay.  No similar lines appear in our painting above, indicating that FWN 119 was traced from the other.  (Compare the videos in Part 2, fig. 3, and Part 6, fig. 34c). This, of course, would seem to be proof that Cezanne painted both pictures.    

  1. It is remarkable how precisely the compositions of these two paintings match. All the details visible in our picture were replicated exactly in FWN 119 – with one exception! This detail appears in FWN 119 (fig. 41, right) but not in the other.  This is the large dark spot (a sailing ship?) which appears at the juncture of the quay and the shoreline.  If our painting had been a copy of FWN 119, the artist would never have left it out!  Since this spot does not appear in our painting, we can only conclude that it was added to FWN 119 only after it had been completely traced and copied from the first.  This again proves that our painting is the primary version of the picture and that FWN 119 is the secondary – meaning, again, of course, that Cezanne painted both pictures.

Fig. 41.  Detail of the two paintings after cleaning.  Note that the dark form at the junction of the quay and shoreline in FWN 119 (at right) does not appear in the Nantucket painting (at left).  If the latter had been a copy of FWN 119, the artist would not have left it out!  Similarly, if the latter had been a copy of FWN 119, the artist would never have added the clumsy “corrections” to the sea surface, which do not appear on FWN 119. 

  1. The distinctive brushwork of small slanted parallel strokes visible in the varnish on the distant hills, and in the foreground of our painting is an unmistakable characteristic of Cézanne’s painterly technique and points to him as the artist (figs. 42, 43a-e). These strokes appear on both our painting and in FWN 119 in the same spots and at the same angle (See Part 2, fig. 3; Part 6, fig. 34c)

Fig. 42.  Detail of the Nantucket painting before cleaning, showing Cezanne’s distinctive rows of parallel, diagonal brushstrokes in the yellowed varnish and on the hillsides.

Figs. 43 a-c.  Examples of Cezanne’s characteristic brush technique on other paintings:  FWN 913; 104; 145.

Fig. 43 d-e. Details of the Nantucket painting (left) compared with FWN 119 (right), after cleaning. Both paintings feature rows of short diagonal green brushstrokes to the right of the smokestack.  FWN 119 also shows dark outlines on the sides of the smokestack and ground details, and the slope line of the vineyard at right, suggesting that this painting was traced from the other.

  1. The overpainting of the varnish layer with new oil “corrections” (figs. 41 [left] and 42) is evidence that our painting was experimental and was completed before FWN 119. These flourishes are also evidence of Cézanne’s famous habit of going back repeatedly to a “finished” painting and reworking it, or of never being able to decide when a painting was finished. (As stated by U. Becks-Marlorney: “He constantly over painted and corrected his paintings ….  He was never satisfied and never really felt he had completed the painting.  He would keep bringing out the canvas again and again and try to record what he had seen in even greater detail.” Or: “Cézanne constantly reworked his paintings, spending months or – with breaks – even years on some of them.  But if he found that he could not achieve exactly the balance of color which he sought, he would simply abandon paintings where they stood, or even destroy the canvas in fury at himself and his lack of success.” In Cézanne 1839-1906. Pioneer of Modernism. [Köln 2020], pp. 24 and 60).
  2. The overpainted details on the surface of the water (fig. 41 [left]) hardly improved the appearance of the painting and, in fact, rather injured it. This would seem to be evidence of a painter who had no thought of selling this picture or of showing it to anyone, but who saw it only as a useful step in the production of a more finished work (FWN 119). (As said by a pair of his contemporaries: “Canvas was nothing more for this Provençal master than a blackboard on which a mathematician works out the solutions to a problem. Perhaps it’s as much due to this idea of work as to the lack of concern for making his work known that such a great number of Cézanne’s paintings were left unfinished. For Cézanne, the one condition indispensable to true work was to be able to carry it out without material worries.”  In R. P. Riviere and J. F. S. Schnerb, “L’atelier de Cézanne,” in La Grande Revue Dec. 25, 1907, 811-817. Quoted in Michael Doran, Conversations with Cézanne. Berkeley: U. of California Press, 2001, p. 89).
  3. The x-ray of the painting (fig. 44) shows that the artist applied a heavy, rippled primer of white paint with a palette knife to the lower half of the painting before he painted the details of the landscape. This suggests that he may have copied the outline and mass of the sloping terrain from an even earlier version of this picture before attempting to paint this sketch in his studio. The heavy primer would also seem to be good evidence of the artist’s creative genius working toward an idea for a painting (FWN 119) he had not yet begun – even though, in the end, this rippled impasto layer was an element that he chose to abandon.

Fig. 44.  The x-ray image of the Nantucket painting, showing that before painting the foreground landscape, he applied to the paper a heavy rippled primer, made with a palette knife.  The shape of the this layer suggests it may have been copied from an even earlier version of this landscape, perhaps one made “en plain air.”

  1. The same impasto primer, laid on with a palette knife, would also seem to betray the hand of Cézanne, for during the 1860’s and 1870’s, he often used the palette knife as a substitute for the paintbrush (See Parts 3 and 5).
  2. Unlike FWN 119, which was evenly painted in oils, the upper half of our painting was painted with thinned oils used like watercolors. In this regard it closely resembles another painting by Cézanne, also depicting the area near L’Estaque, called “Saint-Henri et le golfe de Marseille” (ca. 1883) in the Philadelphia Museum of Art (FWN 190).[1] Here the same distant hills and quay are rendered in a similar loose brush style (figs. 44 a, b).   If the artist of our painting was not Cézanne, how would he have known to paint the hills and quay in this manner if he was only copying the highly finished FWN 119?  The Philadelphia painting is presumed to have been painted four or five years after our painting.

Fig. 45 a, b.  The Nantucket painting above, after cleaning, contrasted with Cezanne’s “Saint-Henri et le golfe de Marseille” (ca. 1883) (FWN 190) in the Philadelphia Museum of Art.  The similar manner in which the same hills and quay are rendered in both paintings strongly suggests that the artist of the two paintings was the same.

  1. The same loose brush style visible in figs. 45 a, b can also be observed in Cézanne’s self-portrait, now in the Ny Carlsberg Glyptotek in Copenhagen (FWN 478) (fig. 46), where it appears in his bowler hat and coat surface. In his beard we can also observe the same rows of short, parallel, diagonal brushstrokes that appear on the hills of both our painting and FWN 119.

Fig. 46.  Self-portrait by Cezanne in the Ny Carlsburg Glyptotek, Copenhagen (FWN 478). The artist has rendered his hat and coat using the same loose brush technique that appears in the hills in figs. 45 a, b. The rows of parallel brush strokes highlighting his beard are also identical to those appearing on the hillsides of both our painting and FWN 119 (See Part 2, fig. 3; Part 6, fig. 34 c).

  1. In typical Cézanne fashion the painting is unsigned. Cézanne was a “habitual non-signer,” who signed his pictures only when the client demanded it.[2]
  2. When the heavy darkened varnish was removed (after probably 144 years), the colors of the painting closely matched those of FWN 119 (See figs. 39 a, b; Part 6).

Circumstantial Evidence for Provenance

  1. The painting was found in a Connecticut antique shop, which suggests that at some point in the past it was acquired by an American in France and brought to the States. Given the painting’s aged condition (See Part 5), the likelihood is that this American brought it over in the late 19th century.  The only place where such a painting could have been seen and made available for purchase would have been at one of the first public exhibitions of Cézanne’s work – in the Paris gallery of Ambroise Vollard in 1895 or 1898 (See Parts 9 and 10).  (Note: It would have been far easier for an American in Paris in the 1890’s to have bought an original Cézanne at one of these exhibitions than a copy of a Cézanne by an unknown artist!)
  2. The painting’s unusual platform – a sheet of heavy paper glued to a stretched canvas with taped edges – is paralleled by other comparable works by Cézanne. [3] This suggests that most of these works were prepared for exhibition in this way by Vollard for inclusion in one or the other of his exhibitions. (See Parts 7, 8).
  3. The painting appears to have come down to us coated with its original varnish and mounted in its original frame (See Parts 7 and 8). There appear to have been no changes or alterations to either since its manufacture, which suggests that, when found in the antique shop, it was an intact relic from the period ca.1878 (the painting) and 1895 (the likely framing).  The obvious age of the painting, the rough condition of the frame, and its fully oxidized varnish suggests it was early taken to an isolated place (a summer home on Nantucket?), where it would have escaped notice and where no cleaning of the painting could easily have taken place (See Parts 10 and 11).
  4. Vollard’s 1895 show included 150 paintings, but since he published no catalogue, and since no record of his 1895 inventory exists, less than half of the paintings in this exhibition have been identified, leaving the remainder unknown – and leaving open the possibility that our painting was one of those included.[4] (See Part 9).
  5. Vollard held a second Cézanne show in 1898 for which he did publish a catalogue, but the titles given to the featured paintings in the checklist are so brief or vague that in most cases they cannot be matched with known paintings.[5] It is intriguing, though, that cat. nos. 51 and 53 were both labeled “Marine,” and since Cézanne’s only “marine” paintings were views of L’Estaque, it is equally possible that one of these numbers was our painting.  (There would naturally have been some interest in it, for Vollard would have known that its mate, FWN 119, bequeathed to the French state by Caillebotte in 1894, had just been accessioned by the Musee de Luxembourg.)
  6. The curious mounting of the picture – a paper sheet glued to a stretched canvas to make it look more important than it originally was, and a modest frame – shows that it was prepared for exhibition somewhere. Because the frame was so plain, it is also clear that the painting and its sale predated the period of Cézanne’s fame (before about 1898), when his paintings were still modestly priced.  If the sale of the painting had taken place in the early twentieth century,  we would expect to find the painting in a gilded frame. The most likely possibility is that the painting was sold to an American who visited Vollard’s gallery between 1895 and 1898, who then took it to the States and shut it away in a private summer home on Nantucket, where, over the years, the importance of the unsigned, deeply darkened painting was forgotten by the original owner’s heirs and eventually sold as a piece of bric-a-brac.

The Nantucket painting, while no masterpiece, offers a fascinating window into Cézanne’s most famous artistic eccentricities:  his drive to copy and re-copy his own works; his difficulty in deciding when a painting was finished and his need to return to it repeatedly to modify it; his use of thinned oils; his use of the palette knife; his application of paint in small parallel diagonal strokes; and his representation of objects as “sensations” rather than as definable things.  As two collaborating observers of Cézanne wrote just before he died: “The spectacle is moving in a Cézanne painting, most often unfinished, scraped with a knife, overlaid with turpentine-thinned pentamenti, repainted several times, entrusted to a state of relief. One finds in this labor the struggle for style and the passion for nature, the acquiescence to certain classical laws and the revolt of an unedited sensitivity; reason and inexperience, the need for harmony and the fever of original expression.” (R. P. Riviere and J. F. S. Schnerb, “L’atelier de Cézanne,” in La Grande Revue Dec. 25, 1907, 811-817, quoted by Michael Doran, Conversations with Cézanne.  Berkeley: U. of California Press, 2001, p. 173).  These words could be describing this very painting!


[2] John Rewald, Cézanne and America: Dealers, Collectors, Artists, and Critics: 1891-1921. Princeton: 1989, p. 211; Gloria Groom in Cézanne (Chicago Art Institute 2022), p. 24.
[3] According to FWN (, there are 10 known paintings by Cezanne done in oil on paper glued to canvas:  FWN 24 (ca. 1865): 20 x 21.5 cm; FWN 45-TA (ca. 1867): 14.5 x 21 cm; FWN 124 (c. 1879): 53.5 x 72.4 cm; FWN 429 (ca. 1870): 55 x 38.8 cm; FWN 593 (ca. 1867): 22.5 x 33 cm; FWN 629 (1873-74): 21.9 x 19.4 cm; FWN 692 (ca. 1895): 70 x 57 cm; FWN 827 (1890-94): 72.5 x 42 cm; FWN 907 (ca. 1876-77): 19.5 x 12 cm; FWN 1003 (ca. 1865): 19 x 15.2 cm.