William Blake’s Designs for Edward Young’s Night Thoughts *

Dennis Welch and Joseph Viscomi **

From 1795 to 1797, Blake, undertaking the largest project in his artistic career, produced a series of 537 watercolor designs for Young’s Night Thoughts, and engraved 43 of them for the folio edition of the first four Nights. As Richard Edwards, the publisher, suggested in his Advertisement, the project was prompted by the greater interest of the late eighteenth century in illustrate editions of popular classics, the most note worthy being John Boydell’s Shakespeare Gallery. But even in responding to popular taste, Blake used his own format. As Edwards mentioned in the Prospectus, the illustrations “are in a perfectly new style of decoration, surrounding the text which they are designed to elucidate.” Individual plates from the first and second editions of Night Thoughts (1742-45) were glued into a rectangular window cut off-center in large sheets of Whatman paper (about 16 1/2 x 13 inches). Blake drew the illustrations in the large margins surrounding the inlaid pages, and he worked on both sides of his sheet. Originally planned for a four-volume deluxe edition, 150 to 200 select designs were to be engraved, mostly by Blake. But only the first volume, with its 43 engravings, was issued (1797). Only a fraction of Blake’s work, therefore, has ever reached the Public. With the publication of this two-volume Oxford edition, monochrome reproductions of the 537 watercolors, as well as 62 proofs and a “facsimile” of the 1797 edition (from the collection of Robert N. Essick), are now available. Seventy eight of the designs and two sets of the engraved title pages from the colored copies of Nights I-IV are reproduced in color.

Since this two-volume set consists mainly of reproductions, we will discuss them first, leaving specific aspects of its introduction for discussion later. In general, the quality of the reproductions Is not as high as we should like, especially after having been spoiled by the Trianon Press facsimiles. The colors are duller and darker and with edges sharper than in the originals. Yellows look dirty, greens and light blues are darkened rose and flesh tones washed out. NT 81, for example, is typical: the greens should be bright, the rose wash of the hills should be glowing, the yellow streaks should be bright lime, the body white, and the sky lighter.

Much of the inaccuracy in color and tone is due to the reproductions being reduced about 30% from the originals (11 3/4 x 9 in vs. 16 1/2 x 13 in). Watercolor, a medium difficult to reproduce in the best of photomechanical processes, simply cannot be reproduced accurately when its entire spectrum of color and tone is compressed into a smaller area. Because there is less room for gradations, there are losses in tonal and color values, and the colors appear darker and muddy. Another reason the colors appear less radiant than the are is that the paper on which they were washed is reproduced as darker than it actual is. Because watercolor’s characteristic brightness is due to the whiteness of the paper showing through the transparent washes, the darker appearance of the paper means that the overlaying colors, which may be bright, clear, and fresh, will reproduce as muddy, dark, and even as new colors: yellows turn into yellow ochres, warm greys into greyish violets, and flesh tones into pale washes. And because the intensity and hue of the original colors are reproduced in a lower register, the reader can barely recognize, let alone appreciate, Blake’s innovative use of painting directly on the white support instead of over conventional monochrome washes.

Although the color reproductions are disappointing, the editors’ selection of color plates is intelligent. NT 361, 415, 426, and 430, for example, are reproduced in color and are so different from their monochrome versions that they are evidence of the importance of reproducing Blake’s art in color—and of how misleading the black and white reproductions are. Perhaps it is unfair to criticize the monochrome reproductions for not giving us good pictures of the watercolor designs since Blake’s own engravings also fail to do so. But his engravings are better than their reproductions here would have us believe. Because the engravings are reduced, the white interstices of the line system are also reduced. As a result, the engravings appear darker, sharper, and more tightly composed than they actually are. These qualities are most obvious in 9E: the reader’s face and the garments of the small figures are not as densely drawn or dark as they appear.

Although the editors call the reproduction oft the 1797 engraved edition a “facsimile,” it is not. The engravings are not reproduced to size and, like the paintings, they are given an inch margin all around, making them look like individually matted plates rather than pages in a book.

Perhaps we cannot expect from an artbook to be given an accurate sense of the works of art as artifacts; even the best facsimile is only second best, and so what we have here is a good, very handsome, intelligently edited reference book that all students of Blake and Romantic art can use to great advantage. As such, the book is an indispensable to the reproduction of Blake’s artistic oeuvre as David Bindman’s Complete Graphic Works of William Blake (1978) and Martin Butlin’s The Paintings and Drawings of William Blake (1981).

The introduction to the Oxford Night Thoughts is especially useful. Ninety-seven pages long with notes and a helpful index, the introduction clarifies, resolves, and raises a number of points that deserve mention. For example, in a general discussion of the designs and engravings, the editors offer the following important cautions: (1) The “evolution” of the designs does not imply a rigid “progression” from preliminary drawings to watercolors to copper plates, but possibly a dialectical process between plates and designs. (2) The unfinished proofs often present odd facial expressions that develop more normal characteristics as the lines are refined; hence, refinements in a graphic image are not necessarily evidence of changes in the artist’s ideas or mood. (3) The reversal of images between watercolors and engravings was pragmatic, “largely a . . . response to the necessities of production,” and therefore we should not theorize about symbolic differences between left and right in the Night Thoughts designs (pp. 16-17).

The editors tabulate and compare the known surviving versions of the forty-three designs engraved for the 1797 edition, including sixty-two proofs, twenty-four of which are only now made available for study. Variations between the proofs and the published states appear to be minimal although there are a few exceptions as in 34E which has four states. (Christ’s head was changed from a profile to a 3/4 view.) Variations between watercolor designs and their engravings could be significant, however. For example, the bottom figure that was only suggested in NT 52 was brought out and finished in 17E. The face of the horizontal figure in the bottom left corner of 5E was added, and the woman’s genitals covered in 25E. Perhaps another significant, albeit subtle, variation in 25E—one that the editors do not mention—is the change from five to at least six pearls or beads on each wrist of Sense. According to Thomas Helmstadter, the change is from five to seven (“Blake’s Night Thoughts: Interpretations of Edward Young,” Texas Studies in Literature and Language, 12 [1970], 32, n. 15). Admittedly, this change may seem trivial, but the editors do note equally minute variations. Moreover, if as Robert Gleckner suggests in “Blake and the Senses (Studies in Romanticism, 5 [1965], 6-7, 12-15), the “improvement of sensual enjoyment” prophesied in The Marriage of Heaven and Hell (1793) means the expansion of sensory capacities, then the change in the number of pearls or beads (which are symbolic of the senses, according to Helmstadter) may be quite significant.

With regard to drawings and sketches related to specific Night Thoughts designs, the editors’ discussion is informative. However, there are a couple of unpersuasive associations. For example, the relation between Night in NT 103 and the original unerased sketch of a pensive woman on the back of a sheet well known for the drawing “An Angel Striding among the Stars” is not very convincing. (Perhaps it would be if the original sketch were more legible in figure 4 on page 42.) Furthermore, the association between Christ in NT 127 and the rough sketch superimposed on the pensive woman seems a bit stretched. As the editors admit, “the crudely drawn . . . figure has no distinctive marks of Christ and can only be identified by reference to the final watercolour” (p. 43).

The editors’ association of NT 452 with “Warring Angels,” a pencil drawing from the late 1780s of a subject from Paradise Lost, is more convincing. This drawing went through three versions, two of which were traced and combined as a “paste-up” into a new composition. (The drawing was not “mutilated.”) NT 452 is the reverse of the paste-up. That Blake returned to an old design is characteristic of him, but such detailed preparatory work for Night Thoughts is rare. Compositional and thematic prototypes, however, abound. The figure at the bottom of NT 190 is from the frontispiece to The Book of Los, itself from The Gates of Paradise, pl. 16, and from an earlier watercolor drawing, “A Crouching Woman.” The figure NT 312 is a combination of Los’s face and Urizen’s bodily position in Urizen (pls. 7 and 22; numeration from Blake Books). NT 378 is an elaborate version of “Little Black Boy,” p1. 2, and NT 479 is another version of “Laughing Song.” The suicide in NT 411 and the walker with cane and hat in NT 419 are from thumbnail sketches in Blake’s Notebook. But these prototypes are not studies. In fact, there are few direct or to-size preliminary drawings for the Night Thoughts designs. Some of the designs themselves, however, may have been preliminary to work that Followed. The drawing “Journey of Life” (c. 1805), which was used in Jerusalem, pl. 97, is already seen in NT 11 and, reversed, possibly, NT 168. The figure in NT 316 has the torso and face of Cain in the 1809 watercolor “Cain and Abel.” NT 341, 437, and 474 include the wall of angels used in Job, pl. 14 (c. 1805).

As the editors correctly acknowledge, most of the Night Thoughts designs seem to have been freely and quickly invented—the results, no doubt of Blake’s returning to images and themes and borrowing freely from his own repository of characters, subjects, and designs (borrowing, in effect, from a visual repository of ideas absorbed by him yet continually developing). Indeed, only by having such a repertoire could he execute 537 large scale designs in less than two years. He could work quickly because he had so much in mind—and so much to say. That there are a few designs with lines erased, as the editors point out, does not mean Bake varied much in practice from the advice he gave Palmer and Tatham about not departing from “first lines” of a drawing (p. 10-11). Everything points to Blake working quickly and intensely, having, quite literally, ideas and images at his fingertips.

Regarding the many colored engraved copies of Night Thoughts, the editors make an important contribution by studying carefully twenty-two of the known twenty-three copies and distinguishing three color types (on the basis of the different coloration of Death in 1E: White Death, I; Green Death, II; and Grey Death, III), and by detecting at least “thirteen variations in engraving, printing, binding, or colouring . . . among copies of the same type” (p 54). Of the thirteen variations, three are especially notable: (1) the two distinct states, heretofore unnoticed, of 11E, the title page of Night the Second; (2) a small monogram—probably JC—in at least five copies of Type I, suggesting perhaps a colorist claiming credit for the tinting of the plate; and (3) the grotesque coloring of some details in 6E, usually in 20E, and almost as often in 18E. The editors provide a much needed new census of the colored engraved copies, describing at least three previously unknown ones. Copy I-1 is the coloring archetype most likely by b Blake because of the general excellence of its coloring, the inscription declaring that Blake colored it, and the absence of any grotesque coloring.

Minor mishaps in the text of this edition are scarce—almost too scarce to mention. Nevertheless, a comma should be in place of the semicolon in the last line of page 38. To avoid any confusion the reference on page 47 to a preceding discussion relating NT 264 to pl. 10 of All Religions are One should be to page 38, note 58. The misnumbering of lines in Night VII begins on page 28 of the poem (NT 374), and not page 29. And in the reproductions of engraved proofs, the different states of 11E as well as 13E should be sequenced properly.

This edition was originally intended to be published with two volumes of commentary, but as the reproductions were completed it seemed inadvisable to the editors to delay publication of them. Nevertheless, the introduction does include illuminating comments on the frontispieces (NT 1 and 264), on the title page of Night the Third (NT 78), and on the illustration concluding the series (NT 537). With a taste of the commentary, we look forward enthusiastically to the full-length work. We hope that the commentators will elucidate such matters, among other things, as the following: (1) Blake’s use of iconography; (2) his Christology; (3) his transformation of Young’s similes, picture images, and personifications into symbols and myth; (4) the relationships between the Night Thoughts Illustrations and the Vala manuscript; (5) the use of facing pages to create diptychs as in NT 45-46, 71-72, 93-94, and other pairs; and (6) the use of spatial effects within designs as in NT 45, where they seem to reflect Young’s image of the telescope of time being turned aright so that we can see that what once appeared distant (the running out of time, suggested by the distant hourglass at the top of the text-box) is now creeping upon us with a sickle.

The 537 designs for Night Thoughts constitute one-quarter of Blake’s pictorial output. They have been in the British Museum since 1929 and consequently available to only a lucky few. And although there will always be the need for some scholars, especially those interested in technique and descriptive bibliography, to see the original designs and engravings, the Oxford edition now gives interested students and scholars access to pictures of all the designs, prints, and proofs. Critical questions and discussions, heretofore the property of a few, can only be given new life, and Blake’s art and development as an artist can only be better understood and appreciated with this edition.

[*] 2 vols. Edited by John E. Grant, Edward J. Rose, and Michael J. Tolley; coordinating editor: David V. Erdman. Oxford: Clarendon Press, 1980.

[**] Philological Quarterly (Fall, 1982): 539-4.