JAIC 2004, Volume 43, Number 3, Article 3 (pp. 237 to 254)
JAIC online
Journal of the American Institute for Conservation
JAIC 2004, Volume 43, Number 3, Article 3 (pp. 237 to 254)




Some of the most important written evidence about the preference for unvarnished, matte surfaces comes from an American who was not at all sympathetic toward Impressionism. In the book Art-Talks with Ranger, the Tonalist painter Henry Ward Ranger (1858–1916) was quoted by Ralcy Husted Bell as referring disparagingly to the fashion for what Ranger called “flat” surfaces in imitation of Monet and other French painters:“Following the theory that a ‘flat’ surface gave the effect of more air than a shiny one, there came a desire on the part of some painters to make their pictures look like pastels.” Ranger said that “nearly thirty years” had passed since the fad for matte paintings began, which would put its beginning, according to Ranger, in the 1880s (quoted in Bell 1914, 74).

Another American writer who (like Ranger) disliked Impressionism showed awareness of the practice of nonvarnishing in 1893 in remarks prompted by the exhibition in New York of paintings by the French painters Claude Monet and Paul Albert Besnard (1849–1934) and the Americans J. Alden Weir (1852–1919) and John H. Twachtman (1853–1902). The author claimed that Impressionism was not “great art,” and in support of this view wrote: “I have contended that it is only half an art, and its advocates, unconsciously, corroborate my opinion. ‘These pictures should never be varnished,’ says one” ([Trumble] 1893, 1).

Other written sources confirm the understanding on the part of some Americans that a matte, unvarnished effect might be an option that an artist would choose. An 1898 book by Daniel Burleigh Parkhurst may be the first American instruction book to tell artists that they might elect to not varnish an oil painting. Parkhurst was well acquainted with prominent American Impressionists like Theodore Robinson (1852–1896)—Robinson's diary describes several meetings between Parkhurst and himself (Robinson 1892–96)—and Parkhurst's book is illustrated with reproductions of paintings by Monet and Robinson as well as reproductions of works by earlier and more conservative painters. In the section of the book that deals with adding media to paint, Parkhurst tells painters:

Prefer turpentine to oil, and expect your color to dry rather “dead,” or without gloss, by its use. If you intend to varnish, this is all right. If you do not intend to varnish the picture, keep the color as near the pure tones as you can. The grayer the color, the more the “dead” or “flat” drying will make it look colorless. (Parkhurst 1898, 62)

The New York writer Albert Abendschein, who was not as sympathetic to Impressionism as Parkhurst but whose main interest was in rediscovering the lost secrets of the Old Masters, documented the efforts that some painters made to achieve matte surfaces when he wrote in 1906: “I do not remember ever to have seen a picture of the Masters that led me to believe it had that dead, dull, lackluster, nontransparent look to the surface so much prized by some modern painters, who take special pains to bring it about” (Abendschein [1906]1909, 63).


It has been noted that some French Impressionist and Post-Impressionist painters used canvases primed with absorbent grounds, and it has also been pointed out that an absorbent ground could contribute to the matteness of an unvarnished painting (Callen 1994; Jirat-Wasiutynski and Newton 1998; Callen 2000). Some American sources document this connection between absorbent grounds and matteness. For example, Ranger, in his discussion of the fad for “flat” surfaces, said of a matte painting by Alexander Wyant (1836–1892) (who was not an Impressionist but a more conservative Tonal landscape painter): “This particular picture was an experiment of Wyant's, which he made one season when he painted on absorbent canvas” (quoted in Bell 1914, 74). Even more to the point, a ca. 1900 catalog published by A. H. Abbott & Co. of Chicago describes “Absorbent Canvas” as follows:“The ‘preparation’ on this canvas is such that pictures in oil on it are dull or ‘flat’ when finished, like so many of the modern French school” (quoted in Katlan 1992, 311).

It should be pointed out, however, that there is not necessarily a connection between absorbent grounds and the intention to achieve a matte surface. Absorbent grounds were available long before the advent of Impressionism. For instance, John Trumbull (1756–1843) specified absorbent grounds when he ordered canvases from Philadelphia and London in 1817 (Trumbull 1817a, 1817b, and 1817c), and Thomas Sully (1783–1872) recorded in his notebook that he used absorbent grounds on various occasions between 1810 and the 1840s (Sully 1809–71). A British instruction book from 1839 and an American instruction book first published in 1879 and reprinted in 1887 say that the principal advantage of an absorbent ground is that it will absorb a portion of the oil and thereby leave the colors more “pure” (Carlyle 2001, 168). The testimony of the American painter William Morris Hunt (1824–1879) and Abendschein's book from 1906 echo this argument (Hunt 1898, 30; Abendschein [1906]1909, 57–60), showing that during the period when Impressionism was being introduced into America, at least some American painters may have seen the primary purpose of an absorbent ground as preventing yellowing by absorbing excess oil rather than producing a matte effect. Other motives may have been operating as well: a contemporary landscape painter told us that he liked to use an absorbent ground mainly because the absorption of part of the oil made his paint dry faster (White 1992).

Evidence about Theodore Robinson's use of absorbent grounds is somewhat equivocal. Robinson mentioned in his diary that he used a white, absorbent canvas for one of his paintings in 1895; this is the only reference to the absorbency of a ground in any of his diary entries, implying that it may have been unusual (Robinson 1892–96, entry for December 17, 1895). Robinson used the French word absorbante for the canvas, although he had been in New York for some time when he wrote this diary entry and most likely bought the absorbent canvas in New York. (Robinson had ordered paints from France six months earlier, but he complained in his diary about the exorbitant duty he had to pay [Robinson 1892–96, entry for June 14, 1895].) Eliot Clark's description of Robinson's technique, written ca. 1918–1920, shows that Clark agreed with earlier writers who believed that one advantage of an absorbent ground was that it preserved the freshness of colors. Clark wrote that Robinson typically painted on a canvas “sufficiently absorbent to draw the superfluous oil … and in consequence the oil, which is accountable for the sallowing of the color, is reduced to an absolute minimum. This was probably not calculated or preconceived on the part of our painter, but it is accountable for the splendid preservation of his pictures.” However, Clark immediately followed this statement with the sentence: “No picture by Robinson should be varnished,” implying that—whatever Robinson's motivation—the leanness of paint caused by the absorbent ground was something that Clark thought should be preserved in the finished painting (Clark 1979, 31–32).


French painters were said to have, on occasion, squeezed paint out onto a blotter to absorb some of the oil, a step that would make it more matte (Callen 2000, 101). Ranger was aware of this practice (Bell 1914), but he did not identify any American painter who actually did it, and we have not found any other cases of an American painter taking this step. In fact, there is evidence that a number of American painters added medium to their paint. This is true even of a painter like John Twachtman, who can be documented as having wanted a matte final effect but who was also said to have added mastic varnish to his oil paint (Mayer and Myers 1993). Of course, if added in sufficient quantities, this additional medium could give some gloss to a painting even if the painting remained unvarnished, and experience shows that many unvarnished American Impressionist paintings (including paintings by Twachtman) are not uniformly matte. The consequences of this lack of uniformity are discussed in detail below, in section 4.

Roger Dennis (1902–1997), a painter and conservator who paid a great deal of attention to the media that various artists added to their paints, said in an interview that Charles Ebert (1873–1959) was unusual because Ebert did not add any medium to his paint, but used it straight from the tube. The authors have treated many paintings by Ebert, and they are almost always unvarnished and often quite matte. One painting by Ebert (Snow Scene, 1917, private collection) is particularly interesting because it was painted mainly in oil, but Ebert used a very matte gouache paint for some strokes in the branches of trees and even signed the painting in gouache. An 1887 painting by Childe Hassam (1859–1935) (At the Café, collection of Rhoda and David Chase) appears to have also been done using gouache (and possibly pastel) in addition to oil. Some French painters also combined oil with gouache and other water-soluble media (Callen 2000, 210). In each of these cases, the combination of oil with gouache makes sense only if the paintings were intended to be unvarnished and matte.

The desire for matteness may have been the motivation behind some American painters' use of other unusual media in addition to gouache. A reviewer of the 1886 exhibition of the Water Color Society called William Merritt Chase's (1849–1916) large painting “distemper” but clearly described egg tempera—one of the earliest instances of the use of egg tempera in America (Boyle 2002, 23). In 1914, Frederick Frieseke (1874–1930) said,“I usually make my first notes and impressions with dashes of tempera, then I paint over this with small strokes [of oil] as I have to keep it as pure as possible or the effects of brilliancy will be lost” (quoted in Sellin 2001, 88). Theodore Robinson's diary shows that he was aware that American painters were experimenting with both tempera and wax in the 1890s. Robinson wrote as follows about Robert Blum's (1857–1903) use of tempera:“Very light and brilliant. … He likes much the medium, and is doing it in a very agreeable pastellish way” (Robinson 1892–96, entry for December 16, 1894). Robinson mentions a painting “done in wax” by Charles A. Platt (1861–1933) in 1894; Thomas Wilmer Dewing (1851–1938) was also using wax in some of his paintings in the 1890s (Robinson 1892–96, entry for December 19, 1894; Hobbs 1996a, 138, 147n). At least part of the motivation for the use of wax may be inferred from American writer Abendschein's comment that wax added to resin or oil has “the additional advantage of eliminating the glassy surface” (Abendschein [1906]1909, 21).1

Copyright © 2004 American Institution for Conservation of Historic & Artistic Works