The Languages of Landscape - Review

University Park, Pa.: Pennsylvania State University Press, 1997. 303 pp.; 13 color ills., 80 b/w. $65.00.

Almost at the very end of his book on The Languages of Landscape, Mark Roskill rehearses the passage from John Ruskin's Modern Painters vol. 4 that discusses J.M.W. Turner's 1843 watercolor The Pass of Faido (Pierpont Morgan Library, New York). Ruskin's dilemma was one that faces all commentators on landscape painting: the relation of the image to its origins in the physical place, or the primacy of subject matter (as perhaps in portraiture) over "matters of invention dependent upon the artist" (Ernest Govett, cited p. 194).

In Ruskin's case, this dilemma was-exacerbated by his long-standing determination to prove that Turner's greatness as a landscape painter depended on his accuracy in representing the natural world. Here, in the Faido, Ruskin suddenly realized that the master had significantly altered the facts of the scene; when Ruskin compared a graphic transcription of what lay before him (a "topographical outline") with his own line drawing after Turner's original study, the latter's work seemed considerably removed from the scene itself. Ruskin brought himself to conclude that Turner's watercolor was "an arrangement of remembrances" not just of that one spot but of the whole journey that the artist had undertaken to reach it.

Roskill uses Ruskin's passage to announce (albeit somewhat belatedly in The Languages of Landscape) the role of memory in landscape art. But, crucial as is that topic, it is not the only issue addressed by Ruskin's chapter on "Turnerian Topography." It alerts us also to the role of the viewer as opposed to that of the artist, privileged in Ruskin's case by his personal relationship and discussions with Turner; to the responsibility of an artist vis-a-vis his subject matter on a variety of counts - but, especially, as with portraiture, in regard to accuracy or recognizability; to the appeal and significance of land of whatever sort - in this case, wild alpine terrain - for both the artist and his possible clients and viewers (that is: Why paint land at all?); above all, what makes land into landscape?(1) This last issue is fundamental, since, as Roskill explains in the conclusion, The Languages of Landscape has argued "from section to section and across time, that landscape art is to be understood as a means of transmitting a certain view or awareness of the natural world" (p. 232).

We are faced, then, with central questions about landscape painting. As Roskill sets out the scheme of his inquiry - both in the introduction (p. 8) and again in rather different terms in the conclusion (p. 228) - this involves a first chapter in which representations of outdoor settings are said to "provide, or support, an equivalence to the unfolding of stories"; then a chapter discusses the role of the viewer and the access he or she has to the imagery; the third chapter explores connections between landscape art and "issues of social and political moment in which nature and culture interact" (p. 8); chapter 4 treats ways in which artistic practice, "the force of social commentary and the cult of spectacle expand upon the meaning of being a landscape artist" (p. 8); a fifth chapter addresses "the modern loss of faith in the scope of an integrative vision" (p. 8); while the conclusion returns to the "hypostatized viewer" (p. 8).

It is a rich and varied agenda, none of it to be gainsaid, and all of it is of exceptional interest. At almost every point Roskill has useful things to tell us, some unusual examples of landscape art to select for analysis, or some methods of proceeding that are challenging. However, it must be noted that his argumentation is often very murky, leaving the impression that perhaps he has rushed on without working out his ideas sufficiently for himself or that the book certainly has not been well edited, as the following representative example shows:

This is not to say, then, that such a landscape of Fragonard [Blindman's Bluff in the National Gallery] courted female viewers, through the way in which it represents what is taken here as being a strongly feminine side to his art; but that nature, in the intimacy and playfulness of the scene depicted, defines itself here in opposition to the receptive and bountiful nature that is mastered by masculine force, to the point of being implicitly under the control or dominance of the male gaze. In particular, the sense of a specific time and of a governing viewpoint is displaced by the role given to memory; if there is an invitation here to enter the landscape, it is not a landscape altered to make it comply with a dream of possession, but one that is appropriate in its scenic resonances to an idealized intimacy and domesticity. In these ways, in contrast to more constrained dependencies on an interior, the "feminine" qualifies of Fragonard's park setting take on a seemingly "natural" signification. (p. 35)

It is not that there is nothing useful there; but readers are disadvantaged who wish to follow Roskill's arguments, relate them to paintings (this one is illustrated in both color and black-and-white), and see them as having a role to play in the larger intellectual structure of the book.

But what tends to blur at a local level of sentence and paragraph is also prone to uncertainty at the structural level. Roskill is haunted, of course (as was Ruskin), by a traditional discrimination against landscape as a prime mode of painting and by its relatively recent advance as subject matter from background to foreground and as what Ruskin considered the modern genre par excellence. So Roskill wants to make large claims for this most popular and various of art forms. And as a contemporary art historian, he is also obliged to situate himself within theoretical discourse, a task he accomplishes with what seems a mild distaste for its more rebarbative procedures (he explains that his own "seasoned inclination" is for using theory "only lightly and as a supporting framework" [p. 9]).

This mild theoretical inclination competes with a more conventional historical bent for control of his argumentation. The book progresses (logically enough) from early Hellenistic and then medieval examples of landscape to Willem de Kooning or Richard Diebenkorn via major considerations of British art in the 18th century and then a more international range during the 19th. But Roskill's instinct (rightly, I think) seems to be for the synchronic approach, and there are many occasions on which what would be the next episode in the historical narrative is frustrated by the invocation of something that, though it has theoretical aptness, does not occur chronogically at that point. Conversely, the thematic analyses can be sidetracked for too long by apparently irrelevant considerations that appear to have been lifted from a survey-course synopsis of lives and works of artists who painted landscapes. It is also perhaps worth remarking at this junction that Roskill's intended audience is not apparent: as a survey and introduction to the study of landscape painting, it offers challenges and fresh perspectives; yet it takes much for granted and more often than not seems to advance more nuanced positions than the survey can readily sustain.

Roskill himself presents different "takes" on his chapters. The fourth, as we have seen, is initially announced as treating the role and practice of the landscape artist as he emerged into prominence at the end of the 18th century; this is clearly a gloss on the historically driven inquiry. In the conclusion, however, the chapter is now described as treating "viewers' capacity to take what is offered in a figurative light, in accordance with four standard figures of rhetoric" (p. 228), which are identified as metonymy, metaphor (notably musicality), irony, and synecdoche.

It is pointless to linger on the uncertainties of approach and exposition; maybe some readers will even find them rewardingly tentative and exploratory. Instead, it is useful to insist on two theoretical ideas that seem to provide the mainstay of Roskill's approach to landscape painting. One is the notion of intertextuality. The other involves the role of inference by the viewer. (Roskill himself invokes semiotics as the large field within which he is working; I think that in fact this should be semiosis - less the structure of preordained meanings embedded in an artifact than the more turbulent process of making and interpreting texts and images.)

Intertextuality involves, inevitably, establishing the landscape painting as a "text" in the first place. This claim, sadly, Roskill never really argues, taking for granted that it functions in discursive ways that are equivalent to other texts, which he then adduces. The mode of proceeding is then analogic: the invocation of other "texts," literally so in the case of novels and poetry; implied texts in the case of the sublime, positivism, or irony; metaphorical ones in the case of parts of speech. Roskill certainly offers suggestive parallels between paintings and a range of modern literary works; like all such analogies, they compel in proportion to the assent a reader can give to their aptness in the first place.

A prime text not often invoked by Roskill for its intertextual relationship with painting is the natural world itself, an "awareness" of which (as we have seen) is what he claims is transmitted by visual art. But other transmissions of that awareness exist, and these might have been used to fix the special contribution of landscape painters at a given period: scientists, environmentalists, antiquarians, local historians, cultural geographers, students of "place" and genius loci, landscape architects, all these have their different takes on land and landscape. What is common to them all, though, is a recognition that even land, and certainly landscape, is culturally and socially determined. In that case, we need to have the contrary argument - that "landscape [namely, landscape painting?] represents traditionally the domain of nature as opposed to culture" (p. 1) - presented more forcibly.

It follows from Roskill's concern with the intertextual condition of landscape paintings that the inferences to be drawn from viewing them become if not their subject, at least the means of access to it. Here he is at his best, notwithstanding his own fears of undue subjectivity. Some readers/viewers will object to the dominance given to the "verbal construal" of these meanings, to the privilege generally accorded "critical contextualism" (p. 9) in the analysis. But they yield some instructive analyses - and here we can only lament the publisher's failure to provide the author with sufficient figures to sustain his (necessarily) verbal analysis with adequate visual support.

Yet in the final resort, Roskill himself finds wanting the "critical contextualism," subscribed to at the start; he ends his book hesitating before the "too open-ended conclusion" that response to landscape is "self-adjusting and aesthetically changeable" (pp. 235-36). What is missing, we are told, is "what verbal descriptions conveying the visual experience of landscape and memories preserving it in imaged form share with treatments that offer access to it in its framed pictorial guise" (p. 236). While that pronouncement remains somewhat gnomic, the final pages of this book deliver the reader to two famous designed landscapes - Monet's Giverny, now restored, and the Sissinghurst of Vita Sackville-West and Harold Nicolson. This somewhat surprising maneuvre allows Roskill to celebrate the associations, the "figurative implications" (p. 236), and the vicarious opportunities of "wandering through" landscape paintings.

This last appeal, of course, is not new: it was famously elaborated by Denis Diderot in his Salon of 1767,(2) and it reminds us of the decisive demarche by which landscape painting, during the 18th century, came to challenge the hierarchy of the genres and gave to its viewers, by the same process, a role that they had not previously or so widely enjoyed. Mark Roskill, in reviewing the languages by which painted landscapes speak to us, reaffirms both the importance of landscape and our need to attend to its infinite variety.

JOHN DIXON HUNT Graduate School of Fine Arts University of Pennsylvania Philadelphia, Pa. 19104

Notes

1. Roskill does not cite Alain Roger's Nus et paysages: Essai sur la fonction de l'art (Paris: Aubier, 1978), but it is a convincing account of this metamorphosis; Roger rehearses his thesis on the invention of landscape again and briefly in Court traite du paysage (Paris: Gallimard, 1997). A current concern with landscape theory in France can also be tracked conveniently in Roger's anthology La theorie du paysage en France (1974-1994) (Seyssel: Champs Vallon, 1995).

2. Diderot is not, as far as I can find, mentioned by Roskill (but the index has much missing from it and there is no bibliography); interested readers might turn to Ian J. Lochhead, The Spectator and the Landscape in the Art Criticism of Diderot and His Contemporaries (Ann Arbor, Mich: UMI Research Press, 1982).

Statement of Ownership, Management, and Circulation (Required by 39 U.S.C. 3685). Tide of publication: The Art Bulletin. Publication no.: 032-620. Date of filing: August 1, 1998. Frequency of issue: quarterly. No. of issues published annually: 4. Annual subscription price: $25.00-$145.00. Complete mailing address of known office of publication: 275 Seventh Ave., New York, NY 10001-6797. Complete mailing address of the headquarters of general business offices of the publisher: 275 Seventh Ave., New York, NY 10001-6797. Publisher: College Art Association, Inc., 275 Seventh Ave., New York, NY 10001-6797. Editor: John T. Paoletti, CAA, 275 Seventh Ave., New York, NY 10001-6797. Managing editor: Elaine Koss, CAA, 275 Seventh Ave., New York, NY 10001-6797. Owner: College Art Association, Inc., 275 Seventh Ave., New York, NY 10001-6797. Known bondholders, mortgagees, and other security holders owning or holding 1 percent or more of total amount of bonds, mortgages, or other securities: none. The purpose, function, and nonprofit status of this organization and the exempt status for federal income-tax purposes have not changed during the preceding 12 months.

Extent and nature of circulation: A. Total no. copies. Average no. copies each issue during preceding 12 months, 9,690. Actual no. copies of single issue published nearest to fling date, 9,531. B. Paid and/or requested circulation. 1. Sales through dealers and carriers, street vendors, and counter sales. Average no. copies each issue during preceding 12 months, 0. Actual no. copies of single issue published nearest to filing date, 0. 2. Mail subscriptions. Average no. copies each issue during preceding 12 months, 8,842. Actual no. copies of single issue published nearest to filing date, 8,829. C. Total paid and/or requested circulation. Average no. copies each issue during preceding 12 months, 8,842. Actual no. copies of single issue published nearest to filing date, 8,829. D. Free distribution by mail (samples, complimentary, and other free). Average no. copies each issue during preceding 12 months, 20. Actual no. copies of single issue published nearest to filing date, 15. E. Free distribution by mail. Average no. copies each issue during preceding 12 months, 0. Actual no. copies of single issue published nearest to filing date, O. F. Total free distribution. Average no. copies each issue during preceding 12 months, 20. Actual no. copies of single issue published nearest to filing date, 15. G. Total distribution. Average no. copies each issue during preceding 12 months, 8,862. Actual no. copies of single issue published nearest to filing date, 8,844. H. Copies not distributed. 1. Office use, left over, spoiled. Average no. copies each issue during preceding 12 months, 828. Actual no. copies of single issue published nearest to filing date, 687.2. Return from news agents. Average no. copies each issue during preceding 12 months, 0. Actual no. copies of single issue published nearest to filing date, 0. I. Total. Average no. copies each issue during preceding 12 months, 9,690. Actual no. copies of single issue published nearest to filing date, 9,531. Percent paid and/or requested circulation. Average no. copies each issue during preceding 12 months, 99.8. Actual no. copies of single issue published nearest to filing date, 99.8. I certify that the statements made by me above are correct and complete. Elaine Koss, Deputy Director.

COPYRIGHT 1998 College Art Association
COPYRIGHT 2004 Gale Group