Cézanne wrote to the artist Emile Bernard on 26 May 1904 that ‘The litterateur expresses himself in abstractions while the painter gives concrete expression to his sensory experiences, his perceptions, by means of drawing and colour. One cannot be too scrupulous, too sincere or too submissive to nature; but one is more or less master of one’s model, and especially of one’s means of expression’ (note 1).
There are scattered references in Cézanne’s letters to the purchase of brushes, tubes of paint and canvases, but very few references to ‘his means of expression’ in the sense of his methods of painting. Those interested have had to rely primarily on information gleaned from their own viewing of Cézanne’s paintings and on second-hand contemporary accounts of his palette, aims and working practices (note 2).
This article will present the findings of investigations into the materials of eight paintings by Cézanne in the National Gallery: The Stove in the Studio (NG 6509), c.1865, Self Portrait (NG 4135), c.1881, Landscape with Poplars (NG 6457), c.1885 –7, Avenue at Chantilly (NG 6525), c.1888, Hillside in Provence (NG 4136), c.1890–2, An Old Woman with a Rosary (NG 6195), c.1895–6 (probably), The Grounds of the Château Noir (NG 6342), c.1900–4, Bathers (Les Grandes Baigneuses) (NG 6359), c.1894–1905 (see plates 1–8) (note 3). It will then examine the way in which Cézanne chose to use his materials between 1880 and his death in 1906. This will inform our understanding of his conception of the nature of painting and our assessment of his originality.
Cézanne’s letters provide evidence that he used Julian Tanguy as a colour merchant (note 4). Tanguy traded paints for works of art for many years and his shop in the rue Clauzel in Paris was the only place where Cézanne’s paintings could be seen. ‘Père Tanguy’, as he was known, ground pigments in the back of the shop (note 5). There is no evidence to date that Tanguy stamped the reverse of canvases that he supplied. According to Pissarro they were of inferior quality (note 6). However, in a recent study it was found that although Tanguy charged ten per cent less than major suppliers, there was nothing to differentiate his materials (note 7). Correspondence in 1878 and 1885 reveals that Cézanne owed money to Tanguy for painting supplies and was in debt to him for many years (note 8). Tanguy died in 1894 but there is material evidence to suggest that Cézanne used other colour merchants, or at least other suppliers of canvas supports, before that date.
Of the National Gallery Cézannes only one, An Old Woman with a Rosary, is unlined. It is therefore possible to see the canvas stamp of the colour merchant Chabod on its reverse. It reads, ‘Rue Jacob, M Chabod, Md de Couleurs, Extra Fines, Toiles et Tableaux, Rue Jacob’, indicating that he supplied colours and canvases. Cézanne’s Route Tournante, c.1905, has an identical stamp and Seurat’s Woman powdering her Nose of 1886 (note 9) has a stamp with the same address but different wording and format (note 10). A Chabod stamp, this time with a different address, was found on the reverse of Cézanne’s Dr Gachet’s House at Auvers (Paris, Musée d’Orsay) painted in 1872–5. It read ‘fournisseur Chabod, successeur de Bovard, marchand de couleurs et de toiles et tableaux, rue de Bucy 15’ (note 11). Chabod appears in the commercial almanacs for the years 1870–2 and then also for the years 1888–90 at 20 rue Jacob. Cézanne also mentions him in his sketchbooks (note 12). Among the sums of money he owed for sundry items is 4 – no unit of currency given – to Chabod. In a letter dated after 1900 (the exact year in the text is unclear) Cézanne writes to a ‘monsieur’ asking for burnt lake pigment from ‘maison Chabod’ (note 13).
There are a number of other references to colour merchants in Cézanne’s letters. In October 1866 he wrote to Pissarro from Aix-en-Provence saying that the paints were hard to come by and very expensive. While at Vernon in 1885 Cézanne asked his friend Zola to accept delivery of some canvases. In 1894 he wrote to ‘a dealer in art supplies in Melun’ about some canvases and in two letters of 1905 and 1906 he wrote from Aix and Fontainebleau to ‘art supply dealers’ regarding the delivery of canvases, paints and a palette. Finally on 28 September 1906 he wrote to his son that he had sent five tubes of paint back to ‘Vignol’ (note 14). (The reference to Vignol is unclear but he may have been an art supply dealer.) There is also evidence in the form of invoices and a letter that Cézanne used pigments and canvases supplied by Sennelier, Lefranc et Cie and also Bourgeois Aîné (note 15). In fact, it seems that Cézanne used paints from a variety of manufacturers, sometimes purchased through a single dealer and sometimes directly from the manufacturer.
There is both documentary and physical evidence that it was Cézanne’s practice to use standard-sized canvases (note 16). In his letters he mentions canvases sizes 20, 25 and 40. Cézanne most regularly used the ‘figure’ sizes 30 (92 x 73 cm), 25 (81 x 65 cm), 20 (73 x 60 cm) and 8 (46 x 38 cm), and his large-format paintings were executed on canvases ranging from no. 40 to no. 120 (note 17). Among the National Gallery Cézannes there is a figure 5 (35 x 27 cm), a figure 20, three figure 25s and a figure 30. These sizes support the idea that Cézanne tended to work on larger canvases later in his career (note 18). Bathers and the other two versions of the subject painted in the last years of his life are Cézanne’s largest paintings on canvas. Only The Stove in the Studio and Bathers have non-standard dimensions (note 19). It is probable that Bathers was the version inventoried after the artist’s death when found off its stretcher in the studio at the rue Boulegon (note 20). Photographs such as the one taken in his studio in Paris in 1894 show Cézanne working on stretched canvases (fig. 1) (note 21). Where Cézanne depicts his own paintings they are also seen on stretchers (note 22). However, Bernard does recall seeing canvases (which in his view had been abandoned) drying in the studio off their stretchers and tacked to the wall (note 23).
Cézanne mainly uses the commercial étude canvas before his financial situation improved in the 1880s, but he does not stop using them completely later in his career (note 24). The étude canvas is medium weight, with a loose, open weave giving a fairly bland texture. It is comprised of fine, thin and irregular threads of linen. Of the National Gallery paintings only An Old Woman with a Rosary and Bathers have more finely woven canvas. Cézanne does not appear to have favoured strongly textured supports (note 25).
There is evidence that the canvases that Cézanne bought from Chabod had been commercially primed on a large bolt prior to stretching, either by Chabod himself or by another manufacturer (note 26). The Chabod canvas stamp on An Old Woman with a Rosary is placed in the bottom left-hand corner, and is partly obscured by the stretcher bar, indicating that it was put on by the colour merchant before he attached it to the stretcher. It is also possible that Cézanne bought the stretcher and a roll of pre-primed canvas separately and assembled them himself. The ground extends into the tacking margins on all sides of the canvas except the top, showing that it was not primed after stretching. It is likely that the unprimed upper edge was also the edge of the large bolt from which this canvas was cut.
Scallop-shaped deformation of the canvas, related to the way it is secured at the time of priming, is another indicator of the point at which a canvas is primed. Cusping on all four edges of a canvas is visible only on Hillside in Provence and Landscape with Poplars, evidence that they were primed after the canvas had been attached to the stretcher. Cusping visible on only one side can indicate that this edge was close to the frame used to stretch large bolts of canvas that were being commercially primed. Avenue at Chantilly is an example of this. The absence of cusping can be due to the canvas being cut from the centre of a large bolt, before attachment to a stretcher or strainer, or to the fact that it was removed from its initial stretcher or strainer and reduced in size. A third possibility is a practice that Cézanne’s dealer Vollard attributes to Cézanne (note 27). He recounts how Cézanne occasionally painted a number of small studies on a large canvas which he then gave to Tanguy to cut up and sell to buyers who could not afford to pay very much (note 28). The non-standard-sized Stove in the Studio has been cut down. There is no cusping and the fact that the horizontal weave of the canvas is not lined up with the stretcher edge is clearly visible on the X-radiograph (fig. 2). It is not possible to know if this was because of the practice Vollard describes, or the result of an alteration in size after completion. However, it is interesting to note that the signature aligns both vertically and horizontally with the current stretcher bars, indicating that the painting was signed in its current dimensions.