“Alexander Cozen’s ‘New Method’ [...]” by Charles A. Cramer

CRAMER, Charles A. (1997), “Alexander Cozen’s ‘New Method’: the blot and general nature – painter”, The Art Bulletin, March 1997.


Alexander Cozen’s ‘New Method’: the blot and general nature – painter

Charles A. Cramer

Born in 1717 to English parents in Peter the Great’s Russia, Alexander Cozens was educated from the age of ten in London, where he remained until a brief return to Russia in the late 1730s or early 1740s.(1) In 1746 he was among the first English artists to go to Italy, where he studied under Claude-Joseph Vernet. Between 1749 and 1754 Cozens was drawing master to Christ’s Hospital in London, and by the mid-1760s he was at Eton, where he taught both Sir George Beaumont and the famous autobiographer Henry Angelo. Although only nine oil paintings can be attributed to him with any certainty, Cozens exhibited regularly at the Society of Arts, the Free Society of Artists, and the Royal Academy. It is not for his finished oils, however, that Cozens is remembered today, but for his drawings and theories, and for his direct or indirect influence on many of the most celebrated landscape artists of the English tradition.

Cozens was the author of four major treatises on what could be called “practical aesthetics.” In his treatise on The Shape, Skeleton and Foliage of Thirty-two Species of Trees for the Use of Painting and Drawing (1771) he attempted to fix the basic forms characterizing trees for the use of landscape painters. His Principles of Beauty Relative to the Human Head (1777-78) was an exemplar of expression, consisting mostly of nineteen plates, one showing the “Simple Beauty” of a woman in profile [ILLUSTRATION FOR FIGURE 1 OMITTED] and the other eighteen showing various formal modifications to that profile that would serve to illustrate some specific “character” (for example, “The Artful,” [ILLUSTRATION FOR FIGURE 2 OMITTED]). Another of his major treatises, The Various Species of Composition of Landscape in Nature, has survived only in fragments.

Today, the best known of Cozens’s theories is his description of how to form landscape compositions starting from a blot. This method was first published by Cozens in 1759 as An Essay to Facilitate the Inventing of Landskips, Intended for Students in the Arts, and again with a more thorough theoretical apparatus in 1785 under the title A New Method of Assisting the Invention in Drawing Original Compositions of Landscape.(2) It is with the last-mentioned that I will be principally concerned here. The blot [ILLUSTRATION FOR FIGURES 3, 7, 8, 9 OMITTED] was a sort of first indication of a landscape, hastily produced in thick, black ink on white paper, from which the artist could later elaborate a finished composition. In Cozens’s own words, the blot is “swift,” “suggestive,” “instantaneous,” “accidental,” “casual,” and “rude.” Our eyes confirm these adjectives; his blots are among the most surprising artistic products of the century. Both the method of blotting and the product seem utterly anomalous to the contemporaneous context of the highly rational classicism of the newly founded Royal Academy under the presidency of Joshua Reynolds. Indeed, the method and the product are only slightly less surprising when considered in the context of the highly inventive and fruitful school of British landscape painting around the turn of the century.

Symptomatically, art historians have had the utmost difficulty in dealing with Cozens and his blots, tending to displace them to nearly any period and culture other than his own.(3) In the 1950s in France, Henri Lemaitre asserted that “Alexander Cozens’s experiments of the 1740s and 50s presaged the art that we would later call ‘abstract,’” and also cited the influence of “the Far East” on his technique.(4) Louis Hawes in 1969 compared Cozens’s depictions of cloudy skies to “a Clyfford Still canvas turned on its side.”(5) E. H. Gombrich and Henri Zerner both linked the blots to the famous psychological test introduced by Hermann Rorschach in 1921, and Zerner complemented this anachronism by citing the roots of Cozens’s method in the seventeenth-century practice of Baroque artists who would “throw the first thoughts of their compositions in masses of light and shade.”(6) The problem underlying these anachronistic attempts to account for Cozens’s work is the emphasis placed on the final formal appearance of the blots, to the exclusion of the theory of artistic process that produced their forms.(7)

Recent theoretical treatments have not been much more successful. In a double fallacy, James B. Twitchell’s Romantic Horizons suggests that Cozens’s method resulted from “youthful impetuosity” (Cozens was already forty-two by the time he first codified his system of blotting in 1759) and repeats the commonplace assertion that “there is something singularly romantic about Cozens.”(8) Even Jean-Claude Lebensztejn, in his exhaustive “Introduction to the New Method of Alexander Cozens,” finds it necessary to account for Cozens from outside his “historical anchorage.” Although meticulous in its primary research, Lebensztejn’s first chapter concedes, “if the object that we are attempting to situate is first and foremost an object of surprise, of eccentricity, we must not forget that ’surprise’ brings with it the depths [fond] in which that object is anchored.”(9) Cozens’s work, it seems, is so theoretically and formally suggestive that at any moment, with or perhaps without the intention of the historian, it evades the context of its own culture and turns up in other centuries and other places.

I will argue here that it is possible to understand Cozens’s technique of blotting within its “historical anchorage,” that of the burgeoning picturesque school of English landscape and even that of the classically inclined Royal Academy. In fact, understanding the relation of the blot to contemporaneous British academic classicism – counterintuitive as it seems – will help to restore to that most long-lived of Western art styles some of the adaptive flexibility that was certainly a condition of its longevity. The displacement of Cozens from his classicizing culture is symptomatic of the tendency of modernism and modernist historiography to arrogate all formal innovation to itself and its privileged history, leaving academic classicism an improbably stale and spent art form for much of its life. My reconsideration of Cozens in this article will center on the importance of what I will call the “techniques of generalization” proper to classicism in the context of eighteenth-century empirical epistemology (the theory of knowledge). Cozens’s New Method is a rather disjunctive mixture of very concrete technical issues (such as how to make drawing ink) and very abstruse philosophical issues (such as the nature of artistic genius). The concept of “generalization” links these issues, describing Cozens’s interventions in both issues of technique (and hence form) and epistemological issues common also to contemporaneous academic classical art.

The “Blot,” the “Sketch” and the Landscape Drawing

We will begin here with the technical aspects of Cozens’s method in order to shift our emphasis from the blot as final formal product to the blot as a stage in an overall artistic process. We will then better be able to compare that process with the academic classical process and the picturesque process, even if its product seems at first sight incompatible with both The technical aspects of blotting are outlined by Cozens in five “Rules” following the epistemological apparatus of the New Method (20-31). Rule I describes how “to make Drawing Ink.” Rule II tells how “to make Transparent Paper,” which would be laid over the blot to make the sketch based on its forms. Rule III gives instruction on how to make a blot, Rule IV how to extrapolate the blot into a landscape sketch, and Rule V how to “finish” that sketch.

Cozens lists three steps to the formation of a blot in Rule III. The first, “Possess your mind strongly with a [general] subject,” is deferred by Cozens to “The Descriptions of the Kinds of Landscape Composition” (23 n), and we will also consider it below. The second step is described as follows (23):

2. Take a camel’s hair brush, as large as can be conveniently used, dip it in a mixture of drawing ink and water . . . and with the swiftest hand make all possible variety of shapes and strokes upon your paper, confining the disposition of the whole to the general subject in your mind.

Fig. 3 is an aquatint of one such blot that accompanied his methodological text. The printmaker made every attempt to maintain the gestural origin of the marks and the thick liquid quality of their medium.

The third step of Rule III suggests composing a large number of blots from which to select “whenever you are disposed to make a composition of landscape from any one of them” (24). The unfinished sketch after the blot is defined as “a landscape drawing without sky or keeping” (25 n). In eighteenth- and nineteenth-century British criticism, keeping indicates the “subordination of lights” in a work (Cozens’s definition, 29), or the transitional values that define mass and unify the lighting of a work. Cozens’s example of a sketch ([ILLUSTRATION FOR FIGURE 4 OMITTED], which is derived from the blot in [ILLUSTRATION FOR FIGURE 3 OMITTED]) includes two ground planes defined by light to dark gradations that are not established in the blot. This aspect of keeping is permitted when forming a sketch from a blot, as detailed in the third step of Rule IV. Once the original blot is dry, a new sheet of semitransparent paper is placed on it. Starting with a nearly black drawing ink, the practitioner should “consider which way the general light should come on the scene most properly” (that is, from left to right, front to back, or vice versa) and then should “make out and improve the light and dark masses that appear in the first or fore ground of the blot,” which is visible through the semitransparent paper. This step translates the original “accidental” forms of the blot into the landscape components that they suggest: you should “[study] every individual form with attention till you produce some proper meaning, such as the blot suggests.” The foreground color, once dry, should be reinforced and deepened in parts – ” (especially the trees and shrubs, &c.)” – with a second pass of the brush, and then the ink should be thinned a little with water and the “second ground” made out, and so on for “the rest of the divisions or grounds in the drawing” (26-27).

Up to this point the sketch is broadly defined in black and white. Although the reinforcement of certain areas and the articulation of ground planes allow some definition of middle tones (keeping), the unfinished sketch appears mostly as a series of stage flats. The drawing, or “finished sketch,” is to be made by adding the sky and the more subtle aspects of the keeping, which serve to “destroy flatness” (30) and to unify the atmosphere and lighting of all the ground planes. The process of “finishing” starts with the thinnest mixture of ink and water, with which the practitioner is to “wash the whole landscape [all the ground planes], except those parts which are intended to be in the first degree of light.” Once dry, this process is to be repeated as often as the ink will still mark the paper, each time leaving more and more parts untouched; and then a little more ink is to be added to the mixture and the process repeated again. The overriding principle to this Rule is that “whatever colour or degree of shade is in use, retain it as long as you can; that is, shade as much of the drawing with it as possible before you make the tint that is in the cup darker” (28-30) – that is, work the whole of the composition, not the individual parts. The “finished sketch” [ILLUSTRATION FOR FIGURE 5 OMITTED] defines its forms purely through chiaroscuro, the description of the masses and shapes of objects through value alone (although the aquatint differentiates the texture of the rocks with thick lines and breaks the masses of foliage with thinner lines).

Eighteenth-Century Classical Art Theory and Empirical

Epistemology Even contextualizing the blot as a moment in an overall process rather than isolating it as a final formal product, this brief sketch of the technical aspects of Cozens’s New Method has raised several points that seem directly contrary to received notions of academic classical process. Perhaps the most evident include the “suggestiveness” of form that mediates between the blot and the sketch (and “suggestion” is, of course, a major trope of the Romantic aesthetic) and the description of form through “masses” rather than the traditional classical terrain of “line.” Nonetheless, Cozens’s epistemological apparatus from the beginning is explicitly directed toward a contemporaneous classical aesthetic (2):

Composing landscapes by invention, is not the art of imitating individual nature; it is more; it is forming artificial representations of landscape on the general principles of nature, founded in unity of character, which is true simplicity; concentring in each individual composition the beauties, which judicious imitation would select from those which are dispersed in nature.(10)

Cozens’s idea of “invention” is neither about the artist’s fantasy (the terrain of Romanticism) nor about “individual nature” (the terrain of Realism), but about the “general principles” of nature (the terrain of classicism). Unity, simplicity, and the selection and concentrating of dispersed beauties through “judicious” imitation is the aim of Cozens’s new method. Compare Reynolds’s ideas of the “Grand Manner,” the highest style of art:

If it has been proved that the Painter, by attending to the invariable and general ideas of Nature, produce[s] beauty, he must, by regarding minute particularities, and accidental discriminations, deviate from the universal rule, and pollute his canvass [sic] with deformity.(11)

[The artist's] eye being enabled to distinguish the accidental deficiencies, excrescences, and deformities of things, from their general figures, he makes out an abstract idea of their forms more perfect than any one original. . . .

Thus it is from a reiterated experience, and a close comparison of the objects in nature, that an artist becomes possessed of the idea of that central form, if I may so express it, from which every deviation is deformity.(12)

[P]erfect form is produced by leaving out particularities, and retaining only general ideas. (4, 57)

Is Cozens’s insistence on the “general” as opposed to the “particular” in nature a mere concession to the rhetoric of the time, designed to lend ethos to his method, or did Cozens truly direct that method toward what we could understand as a classical aesthetic? The answer to this question is best explored through a consideration of the pedagogical and epistemological underpinnings of British academic classicism, and particularly through the widespread calls for “generalization” in art and epistemology.

Cozens’s pedagogical aims closely parallel those of Reynolds’s three “periods” of artistic study. Cozens directs his New Method against three specific causes to which “bad, or indifferently good” artistic productions may be attributed (2-3):

1. To the deficiency of a stock of ideas originally laid up in the mind, from which might be selected such as suit any particular occasion;

2. To an incapacity of distinguishing and connecting ideas so treasured up;

3. To a want of facility, or quickness, in execution; so that the composition, how perfect soever in conception, grows faint and dies away before the hand of the artist can fix it upon the paper, or canvas.

The three periods of artistic pedagogy outlined in Reynolds’s second Discourse (delivered in December 1769) are nearly identical: 1) learning “the rudiments”: “the power of drawing, modelling, and using colours”; 2) amassing a “stock of ideas” from “perfections which lie scattered among various masters,” which are later to be “combined and varied as occasion may require”; and, finally, 3) (conditional) “emancipation,” in which the student learns “to discriminate perfections that are incompatible with each other” (2, 25-27). Curiously, for both theorists, only one of the three aspects of pedagogy seems to pertain specifically to art. The other two are the domain of epistemology, or what the eighteenth century called “human understanding”: for both theorists, the art product must be created by a mind possessed of a “stock of ideas,” which it is capable of properly “distinguishing” and “combin[ing]” (or “distinguishing and connecting” and “discriminat[ing]“) and from which might be “selected” (or “combined and varied”) such ideas as “occasion may require” (or “as suit any particular occasion”).

Indeed, for eighteenth-century British empirical epistemology in general, the ability of “distinguishing and connecting” ideas from a “stock of ideas originally laid up in the mind” defines the minimum parameters of rationality. John Locke, in his highly influential Essay Concerning Human Understanding (1690), had outlined three basic functions that the mind could apply to perceived reality (particular nature). The role of these functions was to “sort” perceptions into an ordered “stock of general ideas,” which constitutes knowledge. The first thing the mind must do is “discern” a given idea or sense impression from the welter of contextual ideas and/or impressions.(13) “Discerning” gives the mind a “simple idea,” the fundamental unit of empirical epistemology. Once this has been achieved, the mind can do only one of two things: it can put ideas together through the additive faculty of “wit” or separate them out through the subtractive faculty of “judgment” (2, 11, 2, 156):

Wit l[ies] most in the assemblage of Ideas, and putting those together with quickness and variety, wherein can be found any resemblance or congruity, thereby to make up pleasant Pictures, and agreeable Visions in the Fancy: Judgment, on the contrary, Iies quite on the other side, in separating carefully, one from another, Ideas, wherein can be found the least difference, thereby to avoid being misled by Similitude, and by affinity to take one thing for another.

There is a qualitative as well as functional difference between the two faculties: although it makes “pleasant Pictures” through its additive operation, wit “misleads”; it is judgment that carries the burden of “carefully” establishing truth. This qualitative difference between “wit” (associated with the “fancy” or “imagination”) and “judgment” is the foundation of much eighteenth-century criticism and aesthetic theory. Concomitant with the mistrustful attitude toward “wit” is the denigration of metaphor (which explains one object through reference to another with which it has some resemblance) as a highly suspicious if also highly pleasing rhetorical device.(14) The poetic devices of alliteration and punning (which operate on the formal similarities among words) are similarly regarded with suspicion: Thomas Hobbes remarks that in formal contexts, and outside familiar company, “there is no jingling of words that will not be accounted folly.”(15) In general, any discursive or aesthetic technique that operates additively, on the “likeness” of objects or forms, is denigrated (at least in epistemological, if not in aesthetic considerations), while techniques that operate subtractively, carefully establishing “difference,” are lauded. Indeed, for Locke (among others), the faculty of the additive “wit” is the cause of both errors and madness, the opposites of rational understanding: “Mad Men . . . having joined together Ideas very wrongly . . . mistake them for truths.”(16) This creates a “disorderly jumbling” (2, 11, 13, 161) in the stock of ideas in the mind, which, Locke elsewhere notes, “gives Sence [sic] to Jargon, Demonstration to Absurdities, and Consistency to Nonsense, and is the foundation of the greatest, I had almost said, of all the Errors in the World” (2, 33, 18, 401).

As the philosophers’ metaphors suggest (the “additive” and “subtractive” functions of the mind, the “jumbling” of ideas), the contemporaneous theory of the operation of the mind was largely materialist. Ideas were said to arise through the physical stimulus of perception and were “stored” in the mind in fluid “sacs” of “animal spirits” or at the nodes of a network of nervous fibers. In turn, these ideas were thought to be connected through the stimulation of certain of these sacs or nodes, causing the “animal spirits” to run from one node to another, or causing a distribution of tensions in the network of nervous fibers, making up “trains of thought.”(17) The eighteenth century formulated a distinct branch of epistemology that explored the way in which “trains” of ideas could be connected through the additive faculty of “wit,” piloted by the subtractive “judgment,” in a way more or less conducive to rationality and to aesthetic process: the doctrine of associationism. First formulated by Hobbes in his epistemological-political treatise Leviathan (1651), associationism took on increasing importance to the critical and aesthetic thought of the subsequent century, as Walter Jackson Bate noted in his seminal text From Classic to Romantic:

The psychological criticism of this period was either determined or, more often, strongly colored by the doctrine of “the association of ideas”; it may be questioned, indeed, whether any philosophical or psychological doctrine has since permeated critical thought in so great a degree as did that of the association of ideas at this time.(18)

Martin Kallich has most thoroughly traced the extensive influence of associationism from its roots in empirical epistemology to its popular application by such critics as Isaac Watts, George Turnbull, Henry Home (Lord Kames), James Beattie, Alexander Gerard, Archibald Alison, and Francis Jeffrey.(19)

It is to this context of “associationism” that Cozens’s and Reynolds’s insistence on the proper “connections” of ideas must be related Since our main concern here is with aesthetic practice, we will consolidate and explore the relevance of associationism to both the production of art and to contemporaneous assessments of the mind’s basically antirational tendencies through that most famous of associationist texts, Lawrence Sterne’s Life and Opinions of Tristram Shandy (published in installments between 1759 and 1767). The very first chapter contains a covert reference to Locke’s doctrine of associationism and cites rampant associationism as the creative device behind the text:

. . . you have all, I dare say, heard of the animal spirits, and how they are transfused from father to son &c. &c. – and a great deal to that purpose: – Well, you may take my word, that nine parts in ten of a man’s sense or his nonsense, his successes and miscarriages in this world depend upon their motions and activity, and the different tracts and trains you put them into, so that when they are once set a-going, whether right or wrong, ’tis not a halfpenny matter, – away they go cluttering like hey-go-mad; and by treading the same steps over and over again, they presently make a road of it, as plain and smooth as a garden-walk, which, when they are once used to, the Devil himself sometimes shall not be able to drive them off it.(20)

Anyone who has read Tristram Shandy will confirm that the narrative trajectory is as “cluttering” and “hey-go-mad” as “animal spirits . . . set a-going.” Tristram is not even born until the fourth book of the serial publication, as he is constantly distracted by the sidetracks and groundwork necessary to explain his odd stance toward life and narrative. The text is a precise index of the “disorderly jumbling” (as Locke put it) of Sterne/Shandy’s random mental associations.

Significantly, Tristram’s odd stance toward narrative is specifically located to an exaggerated Lockean crisis of associationism. The last phrase of the above citation is directly derived from Locke’s Essay Concerning Human Understanding (2, 33, 6, 396):

Custom settles habits of Thinking in the Understanding . . . which seems to be but Trains of Motion in the Animal Spirits, which once set a going continue on in the same steps they have been used to, which by often treading are worn into a smooth path, and the Motion in it becomes easy and as it were Natural.

There is a dangerous similarity between the “as it were Natural” connections of ideas that are the result of mere “custom” and Locke’s just previous (but largely unexplored) assertion that some ideas are “ally’d by Nature” (2, 33, 5, 395-96). Determining the difference between the two formed a large part of the eighteenth-century epistemological project.

Tristram’s “animal spirits” – and hence his text – have been set a-going through a particularly disjunctive association of ideas. His father had formed the habit of winding the clock on the same night of the month that he made love to his wife. As a result of her husband’s odd association of these ideas in the dubiously rational category of “family concernments,” Tristram’s “poor mother could never hear the said clock wound up, but the thoughts of some other things unavoidably popped into her head – & vice versa.”(21) At the climax of Mr. Shandy’s lovemaking, then, Mrs. Shandy pops out with the apparently unrelated question, “Pray, my dear . . . have you not forgot to wind up the clock?” Mr. Shandy is startled, and as a result, the “animal spirits” that were to conduct Tristram’s preembryonic “HOMUNCULUS” to its place of reception were “scattered and dispersed.” The homunculus arrived at its destination “ruffled beyond description, – and . . . in this sad and disordered state of nerves, he had laid down a prey to sudden starts, or a series of melancholy dreams and fancies for nine long, long months together.”(22) Again, the joke is related to Locke’s pedagogy, where dire results were said to attend on the improper association of ideas, and it was accordingly the responsibility of the parents “diligently to watch, and carefully to prevent the undue Connexion of Ideas in the Minds of young People” (2, 33, 18, 401). The startlingly disjunctive association, inculcated at such an unfortunately formative moment, resulted in Tristram’s own permanent state of distraction and hence in his digressive style.

Tristram Shandy, of course, is a fictional text, but the complex stance toward and knowledge of associationism required for appreciating it establishes at once the logical extreme of and the general cultural knowledge of the “associationist problem.” The issue is one of “distraction” and the fundamental tendency of the mind to make multiple irrational leaps, especially when the (additive) “fancy” is uncontrolled by the (subtractive) “judgment.” It is in direct relation to the associationist problem that we should understand the need to generalize our ideas, which, as we have seen, both Reynolds and Cozens emphasize.

This necessity is predicated first of all on the overabundance and fundamental particularity of sense data and “simple ideas.” As Locke noted (2, 32, 6, 385-86),

The natural tendency of the Mind being toward Knowledge; and finding that, if it should proceed by, and dwell upon only particular Things, its Progress would be very slow, and its Work endless: Therefore to shorten its way to Knowledge, and to make each Perception the more comprehensive; the first Thing it does, as the Foundation of the easier enlarging its Knowledge, either by Contemplation of the things themselves, that it would know; or conference with others about them, is to bind them into Bundles, and rank them so into sorts, that what Knowledge it gets of any of them, it may thereby with assurance extend to all of that sort; and so advance by larger steps in that, which is its great Business, Knowledge. This, as I have elsewhere shewed, is the Reason, why we collect Things under comprehensive Ideas, with Names annexed to them into Genera and Species; i.e. into kinds, and sorts.

The “particular” ideas of sensation are variously grouped and ranked into more comprehensive ideas; it is only thus that an inductive knowledge of horses-in-general can be achieved in lieu of the particular habits and form of every specific horse. By means of this “bundling,” various levels of knowledge can be reached. Fig. 6 shows a graphic representation of how the mind can impose a structure of general categories on its ideas, a structure corresponding to the general order of nature that underlies the particular perceived instances. The mind at its most basic level is a disorderly jumble of current sense impressions and the random associative deflections arising out of those sense impressions. The duty of the mind is to re-form that chaotic network of ideas into a rational pyramid of levels of knowledge, by first discerning distinct ideas from the complex wholes of sense perceptions, then regrouping similar ideas into bundles and ranks presided over (as we shall see) by “general ideas” that represent the distilled essence of the ideas falling under a given category. The categories “species, genus, family,” and so on, utilized in the binomial taxonomy of the eighteenth-century Swedish scientist Carl von Linnaeus, were used by Locke to rank other sorts of ideas as well. Thus, the red of a particular piece of cloth is a simple idea; it is part of the “species” red, which is part of the “genus” color, which is part of the “family” quality, and so on (see 3, 4, 16, 428).

In the pursuit of knowledge, the mind will refer to its “general ideas” rather than to any given particular idea or to an exhaustive survey of all similar particular ideas. Now the question becomes: How does the mind form the “general ideas” that are the ambassadors of a bundle of similar “particular ideas”? The formation of general ideas of whatever referential level is predicated on the function of “abstraction.” The following passage from Locke shows the relation between “abstraction” and the “general ideas” that are necessary to broaden empirical knowledge and gives us an idea of how the function of abstraction operates on the data of the senses (2, 11, 9, 159):

The use of Words then being to stand as outward Marks of our internal Ideas, and those Ideas being taken from particular things, if every particular Idea that we take in, would have a distinct Name, Names must be endless. To prevent this, the Mind makes the particular Ideas, received from particular Objects, to become general; which is done by considering them as they are in the Mind such Appearances, separate from all other Existences, and the circumstances of real Existence, as Time, Place, or any other concomitant Ideas. This is called ABSTRACTION, whereby Ideas taken from particular Beings, become general Representatives of all of the same kind, and their Names general Names, applicable to whatever exists conformable to such abstract Ideas.

Even though its primary aim is (additively) to “bundle” similar simple ideas under the aegis of a representative type, the “abstractive” formation of the general idea is founded on the process of subtraction – the subtraction of contextual ideas and of accidental and contingent qualities from the perceived object. We are not concerned here with the names used to mark such general/abstract ideas – the name is arbitrary (as Locke recognized), and thus there is no motivation obtaining between signifier (word) and signified (general idea) in verbal language, and therefore there is no motivation between the signifier (word) and the referent (general nature). The case is otherwise for the visual arts, where there can be conformity between the signifier (visual re-presentation) and the signified (idea), inasmuch as the signified is itself visual, and therefore between the signifier (visual representation) and the referent (general nature, or the ideal). Along these lines it is significant that Locke notes that abstract ideas “are in the Mind such Appearances.” The name marks the appearance of the image-based abstract/general idea; nor is this appearance vague and equivocal (2, 11, 9, 159):

Such precise, naked Appearances in the Mind, without considering, how, whence, or with what others they came there, the Understanding lays up (with Names commonly annexed to them) as the Standards to rank real [i.e., particular] Existences into sorts [i.e., general ideas], as they agree with these Patterns [i.e., the image-based ideas]. . . .

The “general idea,” according to Locke, is formed by the process of subtraction and is stored as an image in the mind. (Indeed, the image-based model of the idea in eighteenth-century epistemology is one of the most persistent and characteristic tenets of the philosophy of the period.) It is on this limited collection of “general ideas” and the relatively fewer, and more controlled, associative deflections possible at the higher levels of the referential pyramid that rational epistemology and rational art are predicated.

Reynolds’s Theory of “Generalization” and the Problem of Associationism

Reynolds essentially took over this epistemology and applied it to a cogent art theory founded on “generalization” – again, specifically responsive to the dangers of “associationism” – in which the only remaining problem is the “fixing” of the general idea on the canvas, a process that for Reynolds was largely mechanical. We have already noted that his three-stage pedagogy is concerned more with issues of “human understanding” than with issues of technique. Reynolds explicitly applies the Lockean concept of “abstraction” to his theory of the generalization of form (3, 50):

[The Grand Manner painter] will permit the lower painter, like the florist or collector of shells, to exhibit the minute discriminations, which distinguish one object of the same species from another; while he, like the philosopher, will consider nature in the abstract, and represent in every one of his figures the character of its species [emphasis added].

And for Reynolds, this function of abstraction/generalization is explicitly mobilized against the problems of associationism. The goal of art is to help the audience to control the mad, disjunctive metonymic deflections of the stimulated stock of ideas. His advice to the artist to learn from privileged past artworks (6, 99-100) is aimed not simply at adding to the “stock of ideas” but also at reducing that stock to a more limited and controlled distillation:

A mind enriched by an assemblage of all the treasures of ancient and modern art, will be more elevated and fruitful in resources, in proportion to the number of ideas which have been carefully collected and thoroughly digested. . . . The addition of other men’s judgment is so far from weakening our own, as is the opinion of many, that it will fashion and consolidate those ideas of excellence which lay in embryo, feeble, ill-shaped, and confused, but which are finished and put in order by the authority and practice of those whose works may be said to have been consecrated by having stood the test of ages [emphasis added].

Art’s object is thus to teach the audience (the general public as well as the tyro artist) to “fashion and consolidate” the particular, empirical contents of their minds. The artist is to produce images of “general nature,” which teach the audience the scope and application of “general ideas,” as opposed to images of “particular nature,” which merely contribute to and stimulate random associative deflections. This, then, is why Reynolds constantly inveighs against the “particular” in art:

[The painter will not] waste a moment upon those smaller objects which only serve to catch the sense, to divide the attention, and to counteract his great design of speaking to the heart. (3, 50)

The attention should never be drawn aside by trifles. (4, 69)

The mind is apt to be distracted by a multiplicity of objects; and that scale of perfection which I wish always to be preserved, is in the greatest danger of being totally disordered, and even inverted. (5, 77)

The detail of particulars, which does not assist the expression of the main characteristick, is worse than useless, it is mischievous, as it dissipates the attention, and draws it from the principal point. (11,192)

Indeed, the concept of “General Ideas” is “the presiding principle which regulates every part of Art” (4, 5). It is for this reason that ideas of “general nature” are of crucial importance to Reynolds’s theory, and to the eighteenth-century desires for a rational epistemology and a rational art. Following Reynolds and the eighteenth-century empirical philosophers, I am contrasting “general nature” with the “particular nature” that is formed by sense perception and is problematically inflected by the (usually) irrational metonymic deflections of the associationist mind. General nature is the truth; and the art that exhibits it is valuable inasmuch as it helps to consolidate and order the “disorderly jumble” of particular nature in the public’s mind.

Cozens’s Epistemology: The Blot and “The Art of Seeing Properly”

Above, I have suggested a way of understanding the classicism of the late-eighteenth-century British Royal Academy that is not transhistorical but highly specific to contemporaneous epistemological concerns. The empirical concept of “generalization” preserves many of the traditional aims of classical art – its supposedly cross-cultural and transhistorical ontological status (as we subtract out the contingent data of time and place), its postulate of improving the audience of art, and even its insistence on formal simplicity – but at the same time it is a highly temporally and culturally specific description of how to meet those alms. With this redescription of the idea of classicism in late-eighteenth-century Britain, we can better understand the historical intent of Cozens’s method of blotting.

As in Reynolds’s Discourses, issues of human understanding – the ability of the mind to distinguish and select such ideas as suit any particular occasion – are given a central place in Cozens’s pedagogy. After explaining what a blot is and how the idea of blotting came to him (through a casual elaboration of stains on a piece of paper, a practice later validated by his reading of Leonardo’s observations on wall stains), Cozens returns to his three “problems” (12-13):

. . . to each of these defects, the art of blotting, here explained, affords, in some degree, a remedy. For it increases the original stock of picturesque ideas;

It soon enables the practitioner to distinguish those which are capable of being connected, from those which seem not naturally related; and

It necessarily gives a quickness and freedom of hand in expressing the parts of a composition, beyond any other method whatever.

Let us consider these three achievements separately. The first relates the art of blotting to the “picturesque” – a connection we shall consider below. The third relates to the formal appearance of the blot, its gestural “swiftness” and “rudeness,” which we are ultimately attempting to account for within the context of the contemporary classical aesthetic of “generalization.” We shall here consider how that odd form is conducive to the solution of Cozens’s second problem, the empirical/associationist issue of the due “connections” and “relations” of ideas.

Cozens’s first step to the formation of the blot, we will recall, was, “Possess your mind strongly with a [general] subject.” The beginning artist, who may not be possessed of a “general subject,” is referred by Cozens to his “Descriptions of the Kinds of Landscape Composition,” a series of sixteen aquatints appended to the end of the New Method. These types represent the minimal degree of “design” that is to be introduced in the making of the blot. Although Cozens does not explicitly cite the problem, a more spontaneous blot would probably center itself on the page, avoiding the edges – not an easy beginning for a landscape composition. So the sixteen “Descriptions of the Kinds of Landscape Composition . . . will be of use in furnishing the mind with an idea of a subject” (24 n). Perhaps the most remarkable aspect of these “kinds” is the generality that they represent, which is highly referentially inclusive. Type nine, for example [ILLUSTRATION FOR FIGURE 7 OMITTED], is described as “Two hills, mountains, or rocks, near each other” (32). The words chosen to define the forms on the paper determine the significance of the details: if they are mountains, the masses of light and dark represent huge areas of outcropping or vegetation; if they are rocks, then those masses represent small geological strata or faults. As Cozens asserts, the general forms of these blots, operating together with the intention to derive geomorphic objects from those forms, have “a direct tendency to recal [sic] landscape ideas” (16).

Although they are obviously “general,” then, it is this suggestiveness of the blot that has offered the greatest impediment to locating Cozens within his intellectual milieu. Since “one artificial blot will suggest different ideas to different persons” and, indeed, since “one and the same designer likewise may make a different drawing from the same blot” (11), then the blot can indeed be related to a Romantic aesthetic of “suggestion.” Or, if we posit a subconscious motivation to the individual’s interpretation of the blot, it is related to the Rorschachian inkblot test. Although such associations hold out a seductive theoretical interest for current thinkers, their historical validity is dubious. Cozens’s idea of the utility of the blot is thoroughly empirical and explicitly designed to “order” the parts into their proper “classes,” as in Lockean empirical epistemology (14-15):

[I]t may be remarked, that there is to be discerned, in all whole compositions in nature, a gradation of parts, which may be divided into several classes; for instance, the class of the smaller parts, the class of those larger dimensions, and so on to the largest. The curious spectator of landscape insensibly acquires a habit of taking notice, or observing all the parts of nature, which is strengthened by exercise. It may be perceived, that the application of this notice in youth is directed to the smaller parts . . . from which it is gradually transferred to the larger as we approach to age, at which time we generally take notice of whole compositions. By the means of this propensity we lay up a store of ideas in the memory, from whence the imagination selects those which are best adapted to the nature of her operations. All the particular parts of each object may not be preserved in the memory, yet general ideas of the whole may be thrown as it were into the repository, and there retained.

What Cozens is describing is the formation of the rational mind, which starts with particular ideas of “parts” (toward which the mind of youth is naturally directed) and progressively forms larger ideas of “wholes,” for example, gradually relating and subsuming the minutiae of leaves to the general pattern of their forms and growth on a branch, and then the pattern of branches to the general form of the tree, and so on, until the idea of the (compositional) “whole” of a species of landscape is reached that is adequate to but does not enumerate its individual parts. Reynolds implies the same gradual attainment of rationality through the formation of the subtractive “judgment.” Whereas in the second period of study, the amassing of a “stock of ideas,” the student “must still be afraid of trusting his own judgment,” the third period marks the conditional emancipation from authority: the student may “confid[e] now in his own judgment” (2, 26). This emancipation is conditional because the rules of art are not to be dispensed with, they have simply been internalized; the judgment has been trained; the rational ordering of ideas has taken place; and the artist may be trusted to make valid connections between general ideas.

However, where Reynolds’s pedagogy is founded on the “long, laborious” training of the judgment, after which the proper art product follows almost as a matter of course, Cozens’s method offers more immediate gratification. Clearly, anyone can make a blot, and, as Cozens notes, the elaboration of the blot into the finished landscape does not require any special capability. This is because such elaboration is an additive exercise easily performed by the fundamentally associationist mind; it does not require any of those difficult “abstract” or “general ideas,” but only particular ideas (16):

On the foregoing principles, very few can have reason to suspect in themselves a want of capacity sufficient to apply the use of blotting to the practice of drawing, nor can they be totally ignorant of the parts of composition in nature, for as they are previously prepared with ideas of parts, as before proved, so [blotting] affords an opportunity of calling them forth. . . . Previous ideas, however acquired (of which every person is possessed more or less) will assist the imagination in the use of blotting [i.e., in elaborating a composition from the blot]. . . .

Blotting succeeds in artificially fixing the whole, the general, to which the relatively unproblematic “stock of ideas of particular nature” can be referred. Indeed, Cozens describes the elaboration of the blot into the landscape drawing as the successive “embellishment and consolidation” of the work (13). “Embellishment” is doubtless performed when the practitioner extrapolates landscape elements from the rude forms of the blot (Rule IV); and “consolidation” when the practitioner unifies the “keeping” of the work through overall washes (Rule V). That the blot represents the crucial “general whole” is confirmed in Cozens’s description of the formation of the blot (6-7), where “the attention of the performer must be employed on the whole, or the general form of the composition, and upon this only; whilst the subordinate parts are left to the casual motion of the hand.” There is always the danger, of course, that “the person is inclined to direct his thoughts to the object, or particular parts, which constitute the scene or subject, as well as to the general disposition of the whole.” In itself, this “superabundance of design” is not bad, but it means that the drawing later to be made from the blot must be done “with judgment” – that is, the performer has in this case destroyed the very purpose of blotting, which is to serve the function of judgment (or, rather, to elide its necessity).

Here we can confirm the basically classical tenor of Cozens’s epistemology of blotting. The proper blot will exhibit only one idea, one “whole” at a time, and thus there is no need to reduce its deflections through the subtractive judgment (17).

A true blot is an assemblage of dark shapes or masses made with ink . . . and likewise of light ones produced by the paper being left blank. All the shapes are rude and unmeaning. . . . But at the same time there appears a general disposition of these masses, producing one comprehensive form [emphasis added]. . . .

Cozens always insists on this singularity of the idea or the subject suggested by the blot, as indeed he must, for if the same constituent mark in a blot were in the same mind at once a group of trees and a fissure in a rock, then his blots would be ineffective, associationist, and would produce nonsense landscapes. This limited “suggestive” interpretation of the blots is thoroughly compatible with the eighteenth-century counterassociationist classical aesthetic as I am construing it, for the blot does not suggest different ideas to the same mind at once (like punning, Hobbes’s “jingling” of words) and thus confuse that mind, or constituent parts of different wholes at once (like metaphor), and thus distract the mind, stimulating an ever expanding web of less and less compatible associations. If it is justifiable to consider the classical project of the time as a response to the dangers of associationism to the discovery of the truth (a response best seen in the “classing” of objects into their “general” referential types), then Cozens’s blots are consistent with that project.

Indeed, although the method requires no special capacities or training on the part of the practitioner, Cozens argues that it is thoroughly consonant with the principles of artistic genius (17-20). He even avers that the practice of blotting helps to form the rational mind: the “art of blotting” is “extremely conducive to the acquisition of a theory, which will always conduct the artist in copying nature with taste and propriety.” He goes on to claim, “This theory is, in fact, the art of seeing properly” (13). The very practice of blotting trains the eye to subsume the details of observed nature to the “whole” of the composition. Reynolds’s art process is thus reversed: instead of gradually learning to “distinguish the accidental deficiencies, excrescences, and deformities of things” through “long laborious” comparative observation of particular parts (3, 44), the artist starts with the “general” blot and learns to view particular nature as a blot and thus reform its minute digressiveness. The practice of making physical blots enables the eye to blot “when taking views from nature” (13-14):

In short, whoever has been used to compose landscapes by blotting, can also draw from nature with practice. But he cannot arrive at a power of composing by invention, by the means of drawing views from nature, without a much greater degree of time and practice.

Blotting is asserted by Cozens as an immediate, if slightly surreptitious, way to teach the mind to “generalize” empirical objects (either the immediately perceived or the “stock of ideas” in the memory).(23)

Classical Practice: The Blot as Technique of Generalization

The argument that I have presented above, which exposes the practice of blotting as a solution to the same associationist problem as that which Reynolds was addressing in his theory of the Grand Manner, may account for Cozens’s method in its contemporaneous context – but the solution of blotting may still seem surprising, eccentric, idiosyncratic to Cozens. This is not so. Indeed, the contemporaneous meaning of to blot was to edit, with strong emphasis on the subtraction of excess. The first definition for the word in Samuel Johnson’s Dictionary (1755) is metaphoric: “To obliterate; to make writing invisible, by covering it with ink.” Johnson illustrates the meaning with a significant example from Alexander Pope, “Even copious Dryden wanted, or forgot, / The last and greatest art, the art to blot.”(24) The Oxford English Dictionary lists a specifically eighteenth-century usage of blot as “an obliteration by way of correction.” One of its examples (expanded here to give context) is from Jonathan Swift’s “Author’s Apology” to his 1704 Tale of the Tub:

He [Swift] was then a young gentleman much in the world, and wrote to the taste of those who were like himself; therefore, in order to allure them, he gave liberty to his pen, which might not suit with maturer years, or graver characters, and which he could have easily corrected with a very few blots, had he been master of his papers, for a year or two before their publication.(25)

We recognize several motifs in these citations: that the younger are pleased by more associative, particular, unblotted texts, whereas the more mature appreciate the value of a good blotting; that blotting is a subtractive process applied to copious, overly free expression; and, curiously, that although Swift notes that the Tale required only “a very few blots,” it would require “a year or two” to execute those blots – that is, blotting is a long, laborious task.

In his Life of Johnson, John Boswell suggests the exact place of blotting in the production of the art product. Discussing literary imposture, Boswell refers to a Mr. Eccles’s appropriation of Henry Mackenzie’s novel A Man of Feeling for his own. Eccles “had been at the pains” of providing fraudulent proof of his authorship in the form of a falsified manuscript transcribed from the finished text. As evidence of the mental processes by which he supposedly authored the novel, Eccles’s manuscript included “blottings, interlineations, and corrections, that it may be shown to several people as an original.”(26) Reynolds makes precisely the same triadic assessment of the functions of the mind as they are applied to particular empirical data: the mind of the mature artist is “enabled to distinguish the accidental deficiencies, excrescences, and deformities of things, from their general figures… [to] make out an abstract idea of their forms more perfect than any one original” (3, 44). “Interlineation” is a tool applicable to “deficient” nature; “correction” to “deformed” nature; and “blotting” to “excrescent” nature.

The idea that the blot was applicable to the problem of overabundant “particular” nature and the excessive associative deflections of the mind was not, then, idiosyncratic to Cozens but a current figure of speech at the time. In fact, we can also account for the painterly form of the blot with reference to contemporaneous art theory. Again, curiously, it is Reynolds who can illuminate Cozens’s technique of generalization through the formal attributes of the blot. I noted that one of the aspects of the blot that we would have to account for was its painterliness – the definition of form through visible brushwork, in this case, a brushwork of chiaroscuro. The painterly chiaroscuro of the blot contradicts our received ideas of the “linearity” and “transparency” of classical art, which should offer access to the realm of the ideal unproblematized by issues of the artist’s handiwork.(27) It was Heinrich Wolfflin who most thoroughly inculcated the view of classical art as inevitably characterized by a “linear” and “multiplicitous” (as opposed to the Baroque “painterly” and “unified”) formal appearance.(28) In the eleventh Discourse, however, Reynolds follows a chain of reasoning connecting the smooth (Wolfflin’s “linear”) surface to a “minute” reality of parts (Wolfflin’s “multiplicity”), which is a dangerous agent of distraction, of associative deflection. Indeed, for Reynolds, the “high finish” of the linear paradoxically implies more manual labor – is less transparent to higher reality – than a painterly generalization (11,202):

As [painting] is an art, [the ignorant] think they ought to be pleased in proportion as they see that art ostentatiously displayed; they will, from this supposition, prefer neatness, high-finishing, and gaudy colouring, to the truth, simplicity, and unity of nature.

Here is a place, then, where the epistemological necessity of “simplification” – a constant trope of the classical aesthetic – is connected to a formal product rather contrary to our received ideas of the classical. It is possible, in other words, to speak of a painterly classical form; for Reynolds, the expression of the “true, simple and unified whole” is best performed through a “few, well-chosen strokes” (11, 193-94):

… the pleasure we receive from imitation is not increased merely in proportion as it approaches to minute and detailed reality; we are pleased, on the contrary, by seeing ends accomplished by seemingly inadequate means….

Carry this principle a step further. Suppose the effect of imitation to be fully compassed by means still more inadequate; let the power of a few, well-chosen strokes, which supercede labour by judgment and direction [i.e., the direction of means to an end], produce a complete impression of all that the mind demands in an object; we are charmed with such an unexpected happiness of execution….

It is of the utmost significance to remark that here the visibility of the “few, well-chosen strokes” does not indicate the manual labor of the painter, as is problematic in most Renaissance and academic classical “antimechanic” theories of art and of the “liberal” status of the artist in society. Instead, these marks “supercede labour” (the labor evident in “high-finishing”) in favor of mental “judgment” – and we will recall the subtractive function of the “judgment” in eighteenth-century epistemology. The application of paint can in this case be seen as a “directed” index of the mind’s judgment, as performing an elision of detail, as gliding over or through the minute particularities or “accidents” overlying the “substantial” form of nature.

So a “few lines or touches” are not mere manual labor but are the act whereby the artist performs – and the visible guarantor of the artist’s formation of – “one whole [out of] what nature had made multifarious” (11, 201):

It is by this, and this alone, that the mechanical power is ennobled, and raised much above its natural rank….

The great advantage of this idea of a whole is, that a greater quantity of truth may be said to be contained and expressed in a few lines or touches, than in the most laborious finishing of the parts where this is not regarded.

The important thing for Reynolds is not what to represent but what to leave out: “When [the artist] knows his subject, he will know not only what to describe, but what to omit: and this skill in leaving out, is, in all things, a great part of knowledge and wisdom” (11, 199). “Judgment” and the “skill in leaving out” – aspects of fundamental necessity to eighteenth-century empirical epistemology in general, and to Reynolds’s theory in particular – are best performed by a few elisive/comprehensive strokes.

Curiously – and this fact is underrepresented in the scholarship – contemporaneous criticism makes much of Reynolds’s “painterliness.” Thomas Warton’s “Verses on Joshua Reynolds’s Painted Window at New-College Oxford” (1782) very significantly singles out the handwork, the brushwork, as a major agent of that work’s ability to “reform” the viewer from the “capricious maze” (compare with the “associationist problem”) of traditional Gothic stained glass to the “chaste Design” of Reynolds’s classicism. The first lines of the poem already contain a reference to Reynolds’s hand (although the author is at first reluctant to quit his seductive Gothic taste):

Ah, stay thy treacherous hand, forbear to trace Those faultless forms of elegance and grace! Ah, cease to spread the bright transparent mass, With Titian’s pencil, o’er the speaking glass! Nor steal, by strokes of art with truth combin’d, The fond illusions of my wayward mind!

[emphasis added]

Toward the end of the poem Warton, his mind reformed, again returns to the brushwork:

Behold, [Beauty] prints upon the crystal plain, With her own energy, th’ expressive stain! The mighty Master spreads his mimic toil More wide… While in the warm enamel Nature lives [emphasis added].(29)

By the early nineteenth century, indeed, William Hazlitt was able to trace a direct cause-and-effect relationship between the ideas expressed in Reynolds’s Discourses and the fact that “the English school of painting is universally reproached by foreigners with the slovenly and unfinished state in which they send their productions into the world…. “He locates the reason for this “slovenliness” in the suggestion that was widely recognized as fundamental to Reynolds’s theory: “[I]f the rule here objected to, – that the artful imitation of the parts injures the effect of the whole, – be at once admitted, slovenliness would become another name for genius, and the most unfinished performance would necessarily be the best.”(30) In addition to the meaning of to blot as to edit, we have, I think, lost the rather painterly approach to the performance of the ideal that Reynolds was seen as advocating, and which is better able to account for Cozens’s proposal of the act of blotting as a technique of generalization.

Cozens and the Picturesque

Academic classicism is not the only contemporaneous context into which Cozens’s New Method fits both theoretically and formally. In his second reference to the three achievements of blotting, Cozens refers to the “picturesque.” The tradition of the picturesque is usually described in the scholarship as the gradual emancipation of native landscape painting from the confining rules of Neoclassical (for example, Claudean) compositional techniques, especially when that emancipation culminates in the celebration of the native English landscape.(31) We can locate Cozens in this tradition without disrupting the general academic classical logic that, as we have seen above, was integral to Cozens’s theory of the blot. The Claudean landscape idiom represents the conventional, or the “active,” knowledge of the artist (which can more or less distort even the empirical contemplation of the landscape, as the tourist device of the “Claude Glass” dramatically demonstrates).(32) The first achievement of blotting, we will recall, is that “it increases the original stock of picturesque ideas.” The increase in the crucial empirical stock of ideas is the result of the practitioner’s efforts to extrapolate landscape elements from the rude, unmeaning blot; confronted with “unmeaning” form, the mind ransacks its store of ideas to find a plausible meaning. The artist thus draws on his or her passive knowledge as well as his or her active (conventional) knowledge. By confronting the rude, unmeaning forms of the blot, the artist is forced to account for compositional possibilities and landscape “furniture” not part of the Claudean idiom, and thus Cozens’s new method expands the conventional ideas of the “picturesque,” or “that which is like a picture.”

In this way and in one other, Cozens’s New Method can be understood as expanding the conventional landscape idiom. Typical of his systematizing tendencies, Cozens published a treatise on The Various Species of Composition of Landscape in Nature: 16 Subjects in four Plates together with some Observations and Instructions (probably dating to the mid-1770s, as Kim Sloan suggests).(33) The plates of that essay have been recovered, and Constable himself made copies of the sixteen “species,” but the accompanying text remains lost. The idea, however, is almost certainly the same as that included in the New Method as “Descriptions of the various Kinds of Composition of Landscape.” In the New Method there are again sixteen “kinds,” nearly identical to the surviving plate list of the Various Species’s “styles” of landscape.(34) The kinds are described in such terms as “2. The tops of hills or mountains, the horizon below the bottom of the view” [ILLUSTRATION FOR FIGURE 8 OMITTED]; and “11. Objects, or groups of objects, placed alternately on both hands, and gradually retiring from the eye. The horizon above the bottom of the view” [ILLUSTRATION FOR FIGURE 9 OMITTED]. The numbers correlate to sixteen plates showing exemplary blots of these types. The typical Claudean composition can be described as a balanced, gradual recession into space, formed by landscape elements projecting from the sides of the view, which culminates in a (hazy, golden) distance [ILLUSTRATION FOR FIGURE 10 OMITTED]. This most recognizable “picturesque” (”like a picture”) composition – allowing for Claude’s expansion of the left-hand side to incorporate the narrative – is probably that shown in Cozens’s eleventh “type.” If we accept the scholarly view that British landscape painting gradually liberated itself from this dogmatic type (as it was practiced by Richard Wilson and his students around midcentury), then Cozens’s fifteen other types are literally an expansion of the Claudean convention. Type two, for example, with its high point of view and its emphasis on rugged form, is perhaps recognizable as a “sublime” landscape, although Cozens does not cite it as such.

The formal products of Cozens’s method, and especially the idea that the landscape drawing is an elaboration on a previous, highly generalized form, can also be accounted for through reference to picturesque theory. In plates accompanying the first of his Three Essays (”On Picturesque Beauty”),(35) William Gilpin shows two landscapes, one simple, smooth, and monotonous [ILLUSTRATION FOR FIGURE 11 OMITTED], the other varied and roughened with landscape “furniture” [ILLUSTRATION FOR FIGURE 12 OMITTED]. The figures illustrate the following passage:

… in landscape painting smooth objects would produce no composition at all. In a mountain-scene what composition could arise from the corner of a smooth knoll coming forward on one side, intersected by a smooth knoll on the other; with a smooth plain perhaps in the middle, and a smooth mountain in the distance. The very idea is disgusting. Picturesque composition consists in uniting in one whole a variety of parts; and these parts can only be obtained from rough objects. If the smooth mountains, and plains were broken by different objects, the composition may be good, on a supposition the great lines of it were so before.(36)

Gilpin thus draws the formal conclusion that the “picturesque” is characterized by “roughness,” differentiating it from the “smooth” (which is associated with the “beautiful”).(37) This formal distinction presages Uvedale Price’s more unequivocal distinction between the “picturesque” and the classical “beautiful” as the difference between “variety” and “unity” (or, rather, between variety and the “insipidity” of failed beauty).(38) But the process that Gilpin cites as mediating between the “smooth” and the “picturesque,” like the figures illustrating that difference, bears a remarkable similarity to Cozens’s method: the “picturesque” is an elaboration or variegation of a basically monotonous form, just as Cozens’s landscape sketch is the “embellishment and consolidation” of the blot. Although Gilpin cites the former figure as “without composition,” the aesthetic effect of the latter figure is nonetheless predicated, he notes, on the general composition shown in the “great lines” of the smooth rendition, just as the plausibility of the blot is dependent on its conformity to some valid” type” or “species” of landscape.

So for Cozens as for Gilpin, the foundation of the landscape drawing, the underlying condition of its success, is a “general idea” of great simplicity, illustrating a highly referentially inclusive “species” of landscape composition. For Cozens as for Reynolds, this “simplicity” is strongly associated with a broad, painterly manner, and we could draw Gilpin into the group by describing this painterly manner as more or less “monotonous.” Picturesque variegation is best performed only after and is predicated on the simple whole, to which all parts must be subsumed – just as the large gestures of the blot describe a whole, leaving what will become the “particular parts” to the incidental markings of the brush.(39)

Conclusions: Cozens in Context

An odd conclusion begins to emerge here – a conclusion that suggests that Gozens’s blots are perhaps among the best examples of the classical ideal of the eighteenth century. Founded on the crucial epistemological concept of “generalization,” the blot is coterminous with an idea of a “species” of landscape, one of the more general of general ideas. At the same time, it is not inappropriate to describe the blot as “abstract,” as we have seen, as long as we apply a Lockean, subtractive-mimetic meaning to that word, as opposed to a modernist, additive-formal meaning. The picturesque theory of the 1790s, like Cozens’s New Method, incorporates this highly generalized “abstract” form into its art process, as though the dynamic, procedural revariegation of a basic “monotonous” form guarantees to the picturesque “particular” its share in classical ideality.

If we return to the Royal Academy, we can in fact consolidate the odd connection between the monotonous and the classical ideal that appears to have been the starting point for Gilpin’s and Price’s interventions in picturesque theory. John Opie (professor of painting at the academy, 1805-7) makes reference to the associationist problem and the role of reductive generalization in reforming that problem in a lecture delivered in February 1807:

[A]ll possible license may be granted [to the artist], and a work elevated to any degree of the extraordinary without incurring the censure of being extravagant, provided – but here the mighty labour lies, which may well deter any attempt much above the ordinary course of nature – provided that the trains of ideas are perfectly connected, and the whole perfectly consistent with itself; that there is no break or opening between them, nothing of discordant nature suffered to interpose, to check the progress of the imagination, expose the illusion, and recall a different set of principles to the mind: this is all that is meant by probability in the imitative arts….

This reductive response to associationism, however, courts an odd problem for Opie:

Instances have occurred of some [artists], who have even been so absurd as to think colouring, chiaroscuro, and all that contributes to illusion in painting, as beneath their attention; who, because they have heard that nature might be improved upon in some particulars, have fondly imagined that their compositions approached the heroic and poetical in proportion as they receded from nature and became muddy, tame, and monotonous in the effect…. (40)

William Hazlitt, we will recall, made the same objection to the “painterly” theory of classical generalization, tracing a direct connection between Reynolds’s theory and the “slovenly and unfinished” appearance of early-nineteenth-century English painting.

In his Lectures on Painting, James Barry (professor of painting at the academy, 1782-99) is less troubled by the counterassociationist epistemological logic that associates “beauty” with the “muddy, tame, and monotonous”:

[T]he desideratum (at least in all matters of elevated compositions) is, that the artist should possess a great and noble mind, of ability to penetrate the depth, entire compass, and capability of his subject; to discern in one view all its possible circumstances, to select and unite whatever is most essential, most interesting, and of the greatest consequence to its energetic and happy elucidation, and to be able at the same time judiciously and severely to reject and suppress whatever useless exuberances may have arisen from the heat and fertility of his imagination.(41)

In another, though similar context, and with no sense of drama, Barry concludes that “Mere Beauty then, (though always interesting) is notwithstanding vague and indeterminate…. “(42) Not only the anti-academic Hazlitt, but even the academic theorists explicitly describe the classical ideal as muddy, tame, monotonous, indeterminate, and vague.

Reynolds’s eleventh Discourse on the “technique of generalization” through a “few, well-chosen strokes” was delivered in December 1782, the same year that Warton produced the poetic ode to Reynolds’s (mediated) handwork in the New College painted windows. The academic lectures of James Barry that I have cited were delivered in the mid-1780s, and Gilpin’s “Essay on Picturesque Beauty” was published in 1792. By the early nineteenth century, both Opie and Hazlitt feared that the idea of painterly generalization would be carried too far and result in muddy, unfinished productions. This sequence roughly sketches a history of painterly classical “techniques of generalization,” techniques which originate as a response to the midcentury associationist crisis best illustrated in Sterne’s playful Tristram Shandy (1759-67). These techniques, which climaxed in the 1780s, required the dialectical corrective of picturesque “variegation” in the 1790s; by the early nineteenth century they were seen as conventionalized and overapplied. It is doubtless significant to this history that Cozens felt that a republication of his method would be well received in the mid-1780s; indeed, there are significant formal differences between the “black Sketches and Outlines” of the 1759 Essay and the broader, less determinate “blots” of the 1785 New Method.(43) I have attempted here to locate the epistemology underlying the New Method in the context of the late-eighteenth-century “associationist problem” through specific comparison with Joshua Reynolds’s epistemology and pedagogy, and to locate both textual justifications for and figurative parallels with the odd formal appearance of the blot from inside the academy and from within the picturesque theory of the 1780s and 1790s. If this exploration has (justifiably, I think) somewhat confused our expectations of the formal appearance of the classical, I hope it compensates by locating Cozens well within the parameters of contemporary debate and artistic process.

The ideas in this essay are further elaborated in my Ph.D. dissertation, “Formal Reduction and the Empirical Ideal, 1750-1914,” intended for completion by the summer of 1997. Unless otherwise indicated, translations are mine.
1. The following brief biography is based on Kim Sloan’s extensive primary research in her two-part “New Chronology for Alexander Cozens,” Burlington Magazine, CXXVN, nos. 983 and 987, 1985, 70-75 and 354-63; and in her book Alexander and John Robert Cozens: The Poetry of Landscape, New Haven/London, 1986.
2. A transcription of Cozens’s New Method is available in Adolph Paul Oppe, Alexander and John Robert Cozens, Cambridge, Mass., 1954, 165-87; in Joshua C. Taylor, ed., Nineteenth-Century Theories of Art, Berkeley/Los Angeles/London, 1987, 63-71; and in Jean-Claude Lebensztejn’s L’art de la tache: Introduction a la Nouvelle methode d’Alexander Cozens, Paris, 1990, 467-84 and plates. Citations from the New Method in this essay are taken from Lebensztejn’s text but will carry Cozens’s page numbers (included in the margins of Lebensztejn’s reprint). Lebensztejn’s text is the preferable source, since it also includes reproductions of all the aquatint plates accompanying the original publication and a very extensive primary and secondary bibliography, which is invaluable to the general scholarship of the eighteenth century as well as to the “Cozens problem.”
3. Barbara Maria Stafford’s work on Cozens is a significant exception to this rule, although the context she explores is a very broad contemporaneous European debate on the semiotics of the accidental, and is perhaps less aesthetic than scientific; Stafford, “Characters in Stones, Marks on Paper: Enlightenment Discourse on Natural and Artificial Taches,” Art Journal, XLIV, no. 3, 1984, 233-40.
4. Henri Lemaitre, Le paysage anglais a l’aquarelle, 1760-1851, Paris, 1955, 11, 93-94.
5. Louis Hawes, “Constable’s Sky Sketches,” Journal of the Warburg and Courtauld Institutes, XXXII, 1969, 349 n. 20.
6. E. H. Gombrich, Art and Illusion: A Study in the Psychology of Pictorial Representation, Princeton, NJ., 1989, 183; Henri Zerner, “Alexander Cozens et sa methode pour l’invention des paysages,” L’Oeil, no. 137, 1966, 29, 32-33.
7. The methodological underpinnings of this article, which can broadly be seen as an attempt to describe “style” through a typology of artistic “process” rather than a taxonomy of the final formal appearance of artistic “products,” is suggested in Kendall L. Walton, “Style and the Products and Processes of Art,” in The Concept of Style, ed. Berel Lang, rev. ed., Ithaca, N.Y./London, 1987, 72-103.
8. James B. Twitchell, Romantic Horizons: Aspects of the Sublime in English Poetry and Painting; 1770-1850, Columbia, Mo., 1983, 138, 157. The New Method, which Twitchell cites, was published when Cozens was sixty-eight.
9. Lebensztejn (as in n. 2). 29.
10. As Lebensztejn notes (ibid., 46), Cozens has in fact borrowed this definition from William Burgh’s commentary to the 1783 edition of William Mason’s poem “The English Garden.” Although not Cozens’s own words, then, the genealogy of this passage suggests how widespread and precisely formulated the contrast between “individual” and “general” nature was by the 1780s.
11. Joshua Reynolds, Letter to the Idler, no. 82, Nov. 10, 1759, reprinted in Reynolds, Discourses, London, 1992, 358 (app. B).
12. Reynolds, Discourse 3, 44-45. Further references to the Discourses will be cited in the body of the text with a figure indicating the particular discourse followed by Wark’s page numbers.
13. Locke, bk. 2, ch. 11, sec. 1, 155-63. Further references to Locke’s Essay will occur in the text, in the form (2, 11, 1, 155-63). Locke points out that sense impressions are also epistemologically to be considered ideas, since despite their presumed extramental origin we really know them only as they appear to our minds.
14. For example, Oliver Goldsmith, in his essay “On the Use of Metaphors,” notes that while metaphor can be used with “almost every verb, noun substantive, or term of art . . . with admirable effect,” its extreme is hazardous: “the danger is in sowing metaphors too thick, so as to distract the imagination of the reader, and incur the imputation of deserting nature, in order to hunt after conceits”; Works, I, New York/Chicago, 1837, 301-2.
15. Thomas Hobbes, Leviathan, Indianapolis, 1994, bk. 1, chap. 8, 40.
16. Cf. ibid., bk. 1, chap. 8, 39: while “fancy” (Hobbes’s technical term for “wit”) can please by adorning discourse, “without steadiness, and direction to some end, a great fancy is one kind of madness. . . .” Cf.-also Etienne Bonnot, abbe de Condillac, Essai sur l’origine des connoissances humaines (1746), bk. 1, chap. 2, sec. 34, in Oeuvres philosophiques de Condillac, I, Paris, 1947, 18: the man whose wit is too strong “will have too much memory and too much imagination. . . . He will barely have the ability to reflect; he will be a fool [fou].”
17. Instances of this type of explanation include: (”animal spirits”) Rene Descartes’s Les passions de l’ame, Amsterdam/Paris, 1649; and (”pressing”) Hobbes’s Leviathan, London, 1651. The most thorough subsumption of aesthetic considerations to this materialist philosophy (”nervous fibers”) is that of Edmund Burke, A Philosophical Enquiry into the Origin of Our Ideas of the Sublime and Beautiful (London, 1757), rev. ed., London, 1759, although Joseph Addison had already considered the operation of Cartesian “animal spirits” in aesthetic phenomena in his essays “The Pleasures of the Imagination,” Spectator,, nos. 411-21, June 21-July 3, 1712.
18. Walter Jackson Bate, From Classic to Romantic: Premises of Taste in Eighteenth-Century England, New York/London, 1961, 96.
19. See Martin Kallich, The Association of Ideas and Critical Theory in Eighteenth-Century England: A History of a Psychological Method in English Criticism, The Hague, 1970. See also Killich’s articles “The Associationist Crisis of Frances Hutcheson and David Hume,” Studies in Philology, XLIII, no. 4, 1946, 644-67; and “The Argument against the Association of Ideas in Eighteenth-Century Aesthetics,” Modern Language Quarterly, XV, 1954, 125-36. My one major objection to Kallich’s work is that he fails to allow for the incorporation of “associationism” into “classical” aesthetics: he only grudgingly and in passing notes that “it must be admitted, the psychology only reinforced neoclassic rules of criticism and standards of taste” (Association of Ideas, 34).
20. Lawrence Sterne, The Life and Opinions of Tristram Shandy, New York, 1985bk. 1, chap. 1, 35.
21. Ibid., bk. 1, chap. 4, 39.
22. Ibid., bk. 1, chap. 2, 36-37.
23. Nor is the art of blotting confined in utility to the genre of landscape. Although Cozens does not elaborate on the thought, he notes (10), “I conceive, that this method of blotting may be found to be a considerable improvement to the arts of design in general; for the idea or conception of any subject, in any branch of the art, may be first formed into a blot. Even the historical, which is the noblest branch of painting, may be assisted by it; because it is the speediest and surest means of fixing a rude whole of the most transient and complicated image of any subject in the painter’s mind.”
24. The reference is to Pope’s “First Epistle of the Second Book of Horace.”
25. Jonathan Swift, The Works of Jonathan Swift, x, London, 1883, 15; Swift referred to himself in the third person here.
26. John Boswell, The Life of Johnson, Oxford, 1989, 300 (discussed in reference to the year 1761).
27. I borrow the term “transparency,” signifying the exclusion of all evidence of handwork from the surface of the artwork, from Philippe Junod, Transparence et opacite: Essai sur les fondements theoriques de l’art moderne, Lausanne, 1976.
28. Heinrich Wolfflin, Kunstgeschichtliche Grundbegriffe, 1915; translated by M.D. Hottinger in 1932 as Principles of Art History: The Problem of the Development of Style in Later Art, New York, 1950.
29. Thomas Warton, Verses on Joshua Reynolds’s Painted Window at New-College Oxford (1782), Oxford, 1930, 1, 7-8. However, the brushwork that Warton makes so much of was not, in fact, executed by Reynolds but by Thomas Jervais. Although Jervais’s name is also not mentioned in the poem, in a letter thanking Warton for his tribute Reynolds insists that it is to his hand that this work be attributed: “I am very sorry therefore my name was not hitchd in the body of the Poem, if the title page should be lost it will appear to be addressd to Mr Jervais”; ibid., iii.
30. William Hazlitt, “Essays on Reynolds’s Discourses” (1814-15), reprinted in Reynolds, 320, 327, app. 2.
31. Malcolm Andrews’s The Search for the Picturesque: Landscape Aesthetics and Tourism in Britain, 1760-1800, Stanford, Calif., 1989, is a recent example of this assessment of the picturesque; he follows the assessment of Christopher Hussey, The Picturesque: Studies in a Point of View, London, 1927.
32. The “Claude Glass” was usually a convex mirror through which the tourist could frame a landscape and bring it within a smaller purview. It was frequently tinted, both to reduce the details of the composition and to simulate the mellow patina of old painting. See Andrews (as in n. 31), 68-69.
33. Sloan, “A New Chronology: Part II” (as in n. 1), 356.
34. Sloan reproduces the plate list of The Various Species in facsimile in ibid., 358, fig. 18; cf. the plate list to the New Method, 31-33.
35. William Gilpin, Three Essays: On Picturesque Beauty, on Picturesque Travel; and on Sketching Landscape…, London, 1792.
36. Ibid., 19.
37. Ibid., 4, 5, 36.
38. Uvedale Price, Essays on the Picturesque, as compared with the Sublime and the Beautiful; and, on the use of Studying Pictures, for the Purpose of Improving Real Landscape, 3 vols., London, 1810. On the interrelation of “beauty,” the “insipid” and the “picturesque,” see 1, 9-10, 72-74, 92, 103-5. For Price as for Cozens and Gilpin, the picturesque can be described through a dialectical process of, or as a medium between, generalization and variegation: “As the excess of those qualities which chiefly constitute beauty produces insipidity, so likewise the excess of those which constitute picturesqueness produces deformity” (1, 175).
39. Cozens’s Principles of Beauty Relative to the Human Head, although predicated on a linear rather than painterly generalization, uses the same dynamic of first defining “Simple Beauty” [ILLUSTRATION FOR FIGURE 1 OMITTED] and then proposing the slight emendations to or variations on that “simplicity” that illustrate some more specific “character” [ILLUSTRATION FOR FIGURE 2 OMITTED].
40. John Opie, Lectures on Painting…, London, 1809, 78, 17.
41 James Barry, “Lecture on Composition,” in The Works of James Barry…, I, London, 1794, 456.
42. Barry, “Lecture on Design,” in ibid., 398.
43. Cozens, An Essay… (1759), in facsimile in Sloan, “A New Chronology: Part Il” (as in n. 1), 354.
Frequently Cited Sources
Cozens, Alexander, A New Method of Assisting the Invention in Drawing Original Compositions of Landscape (1785), in L’art de la tache: Introduction a la Nouvelle methode d’Alexander Cozens, by Jean-Claude Lebensztejn, Paris, 1990, 467-84.
Locke, John, An Essay Concerning Human Understanding (1690), ed. Peter H. Nidditch, Oxford, 1990.
Reynolds, Joshua, Discourses on Art, ed. Robert R. Wark, New Haven/London, 1981.
Charles A. Cramer is a Ph.D. candidate at the University of Texas at Austin, specializing in aesthetic theory and issues of classicism and modernism in art. He is currently completing his dissertation, “Formal Reduction and the Empirical Ideal, 1750-1914,” under the advice of Prof. Richard Shiff [Department of Art and Art History, University of Texas at Austin, Austin, Tex. 78712].
COPYRIGHT 1997 College Art Association
COPYRIGHT 2004 Gale Group

Comments are closed.