How often have I said to you that when you have eliminated the impossible, whatever remains, however improbable, must be the truth?
(Sherlock Holmes to Dr. Watson in The Sign of Four)

With his letter of 18 December 1804, Blake sent two engravings to his Sussex patron, William Hayley. These were proof impressions of “the Shipwreck,” after George Romney’s now lost The Shipwreck at the Cape of Good Hope, and “the Portrait,” a portrait engraving of Romney.The Complete Poetry and Prose of William Blake, ed. David V. Erdman, newly rev. ed. (New York: Anchor–Random House, 1988) [hereafter cited as E] 758. Both were commissioned by Hayley to illustrate his Life of George Romney (1809). Only Blake’s “Sketch of a Shipwreck after Romney,” however, was included in the published biography.Hayley’s Life of Romney includes twelve engravings, seven by Caroline Watson and the other five by Blake, Abraham Raimbach, Robert Mitchell Meadows, William Haines, and Robert Cooper. For Blake’s involvement with the biography, see Morton D. Paley, “William Blake, George Romney, and The Life of George Romney, Esq.,” Blake 45.2 (fall 2011): 50-65. Untraced since the 1939 Blake exhibition in Philadelphia, the rejected portrait engraving derives from Romney’s large, unfinished self-portrait executed in his forty-ninth year, which is now in the National Portrait Gallery, London.The evidence that Blake’s engraving is based on this self-portrait by Romney is set forth in Mark Crosby and Robert N. Essick, “‘the fiends of Commerce’: Blake’s Letter to William Hayley, 7 August 1804,” Blake 44.2 (fall 2010): 61-63. The brief catalogue entry for the Philadelphia exhibition describes Blake’s “portrait of Romney” as a “line engraving” that was “lent by Lessing J. Rosenwald.”William Blake 1757–1827: A Descriptive Catalogue of an Exhibition of the Works of William Blake Selected from Collections in the United States (Philadelphia: Philadelphia Museum of Art, 1939) 59, no. 104. Unfortunately, no such print has been located in the Rosenwald Collection, which is now divided between the Library of Congress and the National Gallery of Art, Washington.Ruthven Todd also refers to this print’s being in the Rosenwald Collection in his edition of Alexander Gilchrist’s biography of Blake. See Gilchrist, Life of William Blake, ed. Todd (London: J. M. Dent & Sons; New York: E. P. Dutton, 1942) 380. He repeats this information in his revised edition of 1945 (381). In a typescript catalogue of Blake’s paintings and drawings, Todd also claims that “a copy of this rejected plate” is “in the collection of Mr. Lessing J. Rosenwald … shewn at Philadelphia in 1939” (catalogue now in the Rosenwald Collection, Library of Congress). For the suggestion that Todd’s statements are based on the entry in the Philadelphia exhibition catalogue, see Crosby and Essick 63 and n30.

An anonymous engraving of Romney after his unfinished self-portrait has come to light.The print was purchased in 2011 from Grosvenor Prints, London, by the book dealer John Windle acting for Robert N. Essick. In their online catalogue, Grosvenor attributed the print to Watson based on her frontispiece to Hayley’s Life of Romney, which includes a stipple engraving of the portrait. Watson’s version is very different from the anonymous print discussed here. This recently discovered print is before all letters, suggesting that it is a proof impression.

The print (illus. 1) is on wove paper that has been trimmed inside the platemark.There is no watermark. For full bibliographical information, see Robert N. Essick, “Blake in the Marketplace, 2011,” Blake 45.4 (spring 2012): 141. The image has been heavily worked using a variety of etching/engraving techniques, including line and stipple. Below the image is the pencil inscription “George Romney, the Portrait Painter Self-Portrait” in an unknown but probably modern hand. The graphic syntax (the linear patterns, dots, and flicks used to represent objects, textures, and tonality) recalls engravings of the second half of the eighteenth century, which suggests that it was executed no later than c. 1820. Indeed, the physical, technical, and aesthetic properties of the print are consistent with other eighteenth- and early nineteenth-century reproductive engravings, including those by Blake.

Significantly, the model for the print—Romney’s unfinished self-portrait—was in Hayley’s possession until it passed to Romney’s son, John, around 1820.John Romney claims that Hayley “did not allow him [George Romney] to finish it, but hurried it off to Eartham without delay” (John Romney, Memoirs of the Life and Works of George Romney [London: Baldwin and Cradock, 1830] 192). For John Romney’s acquisition of the painting, see note 9, below. The only visual sources for this painting that were in public circulation prior to 1820 were Blake’s portrait miniature of Romney that Hayley had given to William Long (now untraced), an engraving by William Ridley after Blake’s miniature, published as the frontispiece to the April 1803 issue of the European Magazine, and London Review, and Watson’s frontispiece to Hayley’s Life of Romney, published in 1809.For Blake’s miniature of Romney given to Long and Ridley’s engraving, see Blake to Hayley, 16 March 1804 (E 744) and Crosby and Essick 62-63. Alex Kidson suggests that the self-portrait was presented to John Romney by Hayley “about 1820” (Kidson, George Romney 1734–1802 [London: National Portrait Gallery Publications, 2002] 180). An undated stipple engraving by Thomas Wright after Romney’s unfinished self-portrait is the frontispiece for John Romney’s Memoirs. For a list of Romney’s self-portraits and engravings deriving from them, see Humphry Ward and W. Roberts, Romney: A Biographical and Critical Essay with a Catalogue Raisonné of His Works, vol. 2 (London: Agnew & Sons; New York: Charles Scribner’s Sons, 1904) 134. A number of crucial differences between the Ridley and Watson engravings and the recently discovered print, particularly in the delineation of the background, clothing, and character of Romney’s face, discount them as sources.For example, the anonymous print depicts a smile line extending vertically from the left side of the mouth, consistent with the original painting. The Ridley and Watson prints do not show this line. The lapel of Romney’s coat also differs from Ridley’s and Watson’s prints. If the print was produced in the late eighteenth or early nineteenth century, as suggested by the graphic syntax, then whoever was responsible for its execution had access to the original painting. Apart from Watson, before 1820 only one engraver had direct access to, and more importantly was commissioned to engrave, this painting. That engraver was William Blake.

In Blake’s correspondence there are more references to a portrait engraving of Romney than to any other reproductive engraving he executed. The epistolary evidence indicates that he began the engraving during the final months of his stay in Felpham and continued to work on the plate until at least late 1804. In a letter to John Flaxman of 7 August 1803, Hayley reveals that Blake had “made two excellent drawings of Romney” after the unfinished self-portrait and Thomas Alphonso Hayley’s medallion, and that he intends for both to be “engraved for a single quarto volume of [Romney’s] Life.”G. E. Bentley, Jr., Blake Records, 2nd ed. (New Haven: Yale UP, 2004) [hereafter cited as BR(2)] 157. An initial fee of thirty guineas was agreed for the engraving after the self-portrait before Blake increased his fee to forty guineas.Hayley relates his negotiations with Blake to Flaxman: “Blake surprised me a little in saying (after we had settled the price of 30 Guineas for the first [portrait engraving of Romney] the price which He had for the Cowper) that Romneys head would require much Labor & he must have 40 for it—startled as I was I replied I will not stint you in behalf of Romney” (BR[2] 157). Because of this rise and the amount Blake had asked for engraving the medallion, Hayley did not finalize the medallion commission. Shortly after he returned to London, Blake wrote to Hayley, “I lose no moment to complete Romney to satisfaction” (7 October 1803, E 736). Over the next few weeks he appears to have made significant progress: “I go on finishing Romney with spirit” (26 October, E 738). In a letter of 13 December, he implies that the plate was sufficiently finished for proofs to be taken: “M^r Romney[s] Portrait goes on with spirit. I do not send a proof because I cannot get one the Printers [being] <having been this afternoon> unable or unwilling & my Press not yet being put up” (E 739). On 16 March 1804, Blake sent Ridley’s engraving of Romney to Hayley, describing it as “a copy from that Miniature; you kindly sufferd me to make, from the Picture of Romney which I am now Engraving” (E 744). By the beginning of May, Blake had either assembled his own press or convinced the previously unable/unwilling copperplate printer to pull a proof impression of the plate, which was shown to one of Romney’s oldest friends, Adam Walker:For Romney and Walker, see Paley, “William Blake, George Romney, and The Life of George Romney, Esq.” 54-55. I have seen the elder Mr. Walker. He knew and admired without any preface my print of Romney, and when his daughter came in he gave the print into her hand without a word, and she immediately said, “Ah! Romney! younger than I have known him, but very like indeed.” (E 748) Despite these testimonials, Blake did not send a proof to Hayley for another two months. On 22 June he offered an update on his progress, citing his former business partner, James Parker, to emphasize the quality of the engraving: “Head of Romney is in very great forwardness. Parker commends it highly. Flaxman has not yet seen it. but shall soon, & then you shall have a Proof of it for your remarks also” (E 753). Flaxman may have seen the print shortly afterward, for on 16 July Blake sent a proof, “still [in] an unfinishd state” (E 753), to Hayley.

Part of Blake’s dilatoriness may be explained by other projects he was then working on for Hayley, such as the two plates for the third volume of The Life of Cowper (1803–04), as well as other commercial work, including the frontispiece to Prince Hoare’s Academic Correspondence (1804).Blake appears to have prioritized the two Cowper plates over the Romney portrait engraving, probably beginning work on them at the beginning of 1804; on 23 February, he informed Hayley that the Cowper plates “are both in great forwardness” (E 742). They were published on 25 March. In a letter to Hoare of 25 December 1803 (BR[2] 174-75), Flaxman recommended Blake for the Academic Correspondence engraving, which was evidently finished by the beginning of February 1804, for a review of the pamphlet in the Literary Journal opines that “the Royal Academy of England might have offered an engraving worthy of the subject, and of the country” (BR[2] 190). For the Cowper and Hoare plates, see Robert N. Essick, William Blake’s Commercial Book Illustrations (Oxford: Clarendon P, 1991) 89-91. In addition, Blake was acting as Hayley’s representative in London, locating, inspecting, and, with Flaxman’s help, compiling a list of paintings to be engraved as illustrations for the biography of Romney. During this period, he also began relief-etching his two epic poems, Milton and Jerusalem, according to the 1804 dates on their title pages.

Almost a month before Blake sent a proof, Hayley had solicited Flaxman’s advice on hiring other engravers to work on the prospective illustrations for the biography.In a letter to Flaxman of 18 June 1804, Hayley seeks advice on employing “your Freand Cromak” (BR[2] 196). Flaxman had earlier advised against employing Watson for book illustrations (see BR[2] 194). By August, Blake was aware that Hayley was considering Watson, whose delicate stippling style appealed to fashionable tastes, to engrave Thomas Alphonso’s medallion of Romney.On 23 February 1804, Blake relates Daniel Braithwaite’s “wish that you [Hayley] would give the Public an Engraving of that Medallion by your Sons matchless hand” (E 742). He may have been attempting to revive the medallion commission, which had been left unresolved in August 1803 owing to his increase in fees (see note 12, above). On 7 August, Blake approves of Watson, stating that “Miss. W. would Engrave your Sons Medallion of Romney most delicately.”For a full transcription of this letter, see Crosby and Essick 53-54. While Hayley’s decision to commission Watson may have been informed by the criticisms that Blake’s engravings for The Life of Cowper and his six plates for The Triumphs of Temper had received from Lady Hesketh and her circle (Hesketh favored the softer aesthetic associated with Watson’s style of stippling), there were also practical reasons for his employing other engravers on the Romney biography.For a discussion of contemporary criticisms of Blake’s plates for The Triumphs of Temper, see Crosby, “‘a Ladys Book’: Blake’s Engravings for Hayley’s The Triumphs of Temper,” Blake in Our Time: Essays in Honour of G. E. Bentley Jr., ed. Karen Mulhallen (Toronto: U of Toronto P, 2010) 105-30 (especially 112). See also note 40, below. During 1804 Hayley, Flaxman, and Blake discussed the logistics of selecting paintings for reproduction and also the style that would best replicate Romney’s aesthetic. For example, on 2 January Flaxman suggests to Hayley that Romney’s chalk drawings “are all well worth etching in a bold manner which I think Blake is likely to do with great success” (BR[2] 177). In a letter to Hayley of 4 May, Blake proposes altering his linear style for the Romney commissions: “Mr. Flaxman agrees with me that somewhat more than outline is necessary to the execution of Romney’s designs, because his merit is eminent in the art of massing his lights and shades” (E 749). He goes on to say that he can etch the designs in a stipple manner similar to his 1797 portrait engraving of Leonard Euler.Blake’s engraving is the frontispiece to vol. 1 of Euler’s Elements of Algebra (1797). See Essick, Commercial Book Illustrations 75. Because of Hayley’s initial deadline of November 1804, however, more engravers were needed to complete the illustrations. In his letter to Hayley of 22 June, Blake cites Parker to suggest that even “Eight different Engravers” (E 752) would not be able to execute by November the eight designs listed. He revisits this point in his letter of 7 August: “It is certainly necessary that the best Artists that can be engaged should be Employd in the Work of Romneys Life.” At the beginning of the next paragraph, he describes working on the plates of the Shipwreck and “Romneys Head” with “Spirit & neatness.”See Crosby and Essick 53.

Blake was still working on “the Head of Romney” at the end of September, promising to send a second proof to Hayley: “You shall soon see a Proof of Him in a very advancd state  I have not yet proved it but shall soon when I will send you one” (28 September 1804, E 755). The following month, he claims that he will send proofs of “the Head of R and the Shipwreck, which you shall soon see in a much more perfect state” (23 October, E 756). Later in the same letter, Blake describes a transformative visit to the Truchsessian Gallery before connecting this experience to “the plates I am now engraving after Romney”: Dear Sir, excuse my enthusiasm or rather madness, for I am really drunk with intellectual vision whenever I take a pencil or graver into my hand, even as I used to be in my youth, and as I have not been for twenty dark, but very profitable years. I thank God that I courageously pursued my course through darkness. In a short time I shall make my assertion good that I am become suddenly as I was at first, by producing the Head of Romney and the Shipwreck quite another thing from what you or I ever expected them to be. (E 757)For Blake’s experiences at the gallery, see Paley, “The Truchsessian Gallery Revisited,” Studies in Romanticism 16.2 (spring 1977): 165-77. In a letter to Hayley of 4 December, he once more claims that the two Romney plates exemplify the recuperated artistic vision of his youth: “Hope in a few days to send Proofs of Plates which I must say are far beyond Any thing I have ever done. … I have lost my Confusion of Thought while at work & am as much myself when I take the Pencil or Graver into my hand as I used to be in my Youth” (E 758).

Blake finally sent a second proof of his portrait engraving on 18 December. It appears either that he harbored some concerns about his patron’s reaction or that Hayley replied in an untraced letter shortly afterward, for on 28 December Blake offers the following assurances: “I am very far from shewing the Portrait of Romney as a finishd Proof. be assured that with our Good Flaxmans good help & with your remarks on it in addition I hope to make it a Supernaculum” (E 760). He uses the mock Latinate superlative “supernaculum” to anticipate the high quality of his finished print.“Supernaculum” derives from the German phrase “auf den Nagel” (on to the nail) to describe liquor drunk to the last drop. The OED cites Blake’s letter to Hayley for the use of supernaculum to mean an excellent example of its kind. In a postscript, he reiterates his claim that the engraving required more work: “I made a very high finishd Drawing <of Romney> as a Companion to my <drawing of the> head of Cowper (you remember) with which Flaxman is very much satisfied & says that when my Print is like that I need wish it no better. & I am determind to make it so at least” (E 760). Hayley apparently responded positively to the print, or at least to Blake’s commitment to producing a “supernaculum,” because on 19 January 1805 Blake states: “I feel extremely happy that you think My Prints will do me Credit” (E 762). This is the final reference in the extant correspondence to his engraving Romney’s unfinished self-portrait.In the penultimate sentence of the letter, Blake refers to the projected engravings of Romney’s paintings: “I feel it is necessary to be very circumspect how we advance with Romney his best Works only, ought to be engraved for your Work” (E 762). On 11 December 1805, he mentions that the plate of the shipwreck requires work: “A very few touches will finish the Shipwreck Those few I have added upon a Proof before I parted with the Picture. It is a Print that I feel proud of on a New inspection” (E 767).

The epistolary evidence indicates that Blake worked on this portrait engraving between 7 August 1803 and at least 28 December 1804. He sent two proof impressions to Hayley: the first on 16 July 1804 and the second on 18 December. The second one was sent after he knew that Hayley had Watson in mind to work on at least one illustration for the Romney biography, and also after his transformative experience at the Truchsessian Gallery.

The evidence also suggests that the engraving was intended to be the frontispiece for the biography. In his letter to Flaxman of 7 August 1803, Hayley compares Blake’s portrait engraving of Romney to the frontispiece for the first volume of The Life of Cowper when discussing prices.See note 12, above. In addition, Blake’s comparison of his “very high finishd Drawing <of Romney>”  (E 760) to the drawing of Cowper after Romney (the intermediary model for the Cowper frontispiece) in his letter of 28 December 1804 suggests that the engraving of Romney was intended to be the frontispiece. Furthermore, the intended placement is supported by several inconsistencies in Hayley’s biography of Romney. In 1806 Hayley invited Watson to Felpham to make preparatory drawings of the Romney paintings he owned, including three self-portraits.Hayley, Memoirs of the Life and Writings of William Hayley, Esq., ed. John Johnson, vol. 2 (London: Henry Colburn, 1823) 59. She engraved them on a single plate for the frontispiece to The Life of Romney.For a discussion of this frontispiece, see Crosby and Essick 60, 63-64, and Paley, “William Blake, George Romney, and The Life of George Romney, Esq.” 60. According to Hayley’s preface, “the Introduction and a considerable part of [the biography], were written in the year 1803.”Hayley, The Life of George Romney, Esq. (Chichester: T. Payne, 1809) n. pag. In his Memoirs, Hayley claims that he began the biography “about the middle of December, 1803” (2: 45). He did not, however, fully correct his text by the time the work was published to take account of Watson’s tripartite frontispiece.According to his Memoirs, Hayley was “retouching the Life of Romney” and revising proofs during the first half of 1808 (2: 66). The directions to the binder, for example, describe the frontispiece as “Portrait of Romney at the age of forty-nine.”Hayley, The Life of George Romney, Esq. [421]. He also refers in the main text to a single image: “The portrait, that forms a frontispiece to this volume, represents him as he appeared in the most active season of his existence.”Hayley, The Life of George Romney, Esq. 254. In both instances, Hayley appears to be referring to Blake’s engraving, which was commissioned during the period that the majority of the biography was composed, rather than Watson’s frontispiece. We know that he initially negotiated a price of thirty guineas for Blake’s engraving—the same amount Blake received for the Cowper frontispiece—suggesting a similar size and amount of work for each plate. The image size of the Cowper frontispiece is 185 x 144 mm. The image size of the recently discovered print is 184 x 145 mm. The almost identical dimensions indicate a correspondence in format. Was the anonymous engraving designed to be the frontispiece for a quarto-sized volume on George Romney? To help answer this question, we must turn our attention to its aesthetic qualities.

There are significant challenges in attributing unsigned reproductive engravings to specific engravers. For booksellers and printsellers, accuracy of replication and uniformity of product were paramount. Professional copy engravers, such as Blake, possessed a standard armory of graphic syntax that facilitated the accurate reproduction of original images on copper, which were then printed as uniform objects: book illustrations and/or separate prints.For a detailed discussion, see Essick, William Blake Printmaker (Princeton: Princeton UP, 1980) 25-28. The emphasis on pictorial fidelity via a standardized graphic vocabulary drastically restricts individual artistic expression, making it extremely difficult to identify the authors of unsigned reproductive engravings based on graphic syntax alone. Indeed, it would be very difficult to identify Blake as the engraver of a number of his commercial reproductive engravings without a sculpsit inscription. For example, his portrait engravings of David Hartley (1791), Earl Spencer (1813), and Robert Hawker (1820) are highly conventional, executed with the standard linear patterns common to all copy engravers of the period.Another example is Blake’s “Beggar’s Opera” plate (1790), which Essick describes as “untouched by Blake’s own personal imagery or graphic techniques” (Printmaker 59). Without Blake’s signature on these prints, attribution would rely on circumstantial evidence, as is the case for his apprenticeship engravings.According to Benjamin Heath Malkin, Blake engraved “from those drawings” he made in Westminster Abbey during his apprenticeship (BR[2] 563). These engravings, all signed by Blake’s master, James Basire, as was the conventional practice, include portraits of numerous English kings and queens. See also Essick, Commercial Book Illustrations 115-20. With this caveat in mind, there are visual clues in the recently discovered print that can help in identifying its author.

Unlike the Ridley and Watson engravings, which are pure stipple, the anonymous engraving evinces an unusual combination of the standard graphic syntax associated with reproductive engraving. The background is rendered with dense lines interspersed with heavy stippling, probably made with a mattoir, or mace-headed tool.For Blake’s use of a mattoir in his commercial engravings, see Essick, Printmaker 56. Similarly, Romney’s coat is delineated with a bold mix of roulette (a fine-toothed wheel used to make dotted patterns) and mattoir work and combinations of heavy lines to achieve a rough texture. Certain areas of the coat, such as above the right arm and both lapels, are exceedingly dense. Under magnification, the interstices between the lines merge, dissolving the graphic syntax into rugged blocks. A combination of fine stipple and line is used to depict the hair, while sparse stippling provides form and texture to the necktie. The face is rendered with exceptionally delicate stippling, and demonstrates a sophisticated treatment of light and shadow on the forehead, nose, and around the mouth. This deployment of graphic techniques is striking, providing a particularly strong contrast between the soft face and the rougher clothing and background that is very different from the other engravings after Romney’s self-portrait.Blake uses a similar combination of delicate stippling to render facial features and dense linear strokes for clothing in his extant miniature portraits. See Crosby, “A Minute Skirmish: Blake, Hayley and the Art of Miniature Painting,” Blake and Conflict, ed. Sarah Haggarty and Jon Mee (Basingstoke: Palgrave Macmillan, 2009) 164-84.

The techniques discernible in the print were all within Blake’s range of graphic styles. For example, he uses comparable combinations of line and fine stipple to delineate hair in his portrait engraving of Edmund Pitts (c. 1793-96) (illus. 2).For the Pitts engraving, see Essick, The Separate Plates of William Blake (Princeton: Princeton UP, 1983) 178-80.

In addition, as with the depiction of Romney’s coat in the anonymous print, Blake uses extensive roulette and mattoir work to give a rough texture to Pitts’s coat that is juxtaposed to the subtle stippling on the face. The width of the roulette lines in the recently discovered print, approximately 0.9 mm., corresponds to the Pitts portrait. When we examine Blake’s early commercial engravings for roulette work, we find the same width of lines in, for example, the roundel borders of “Robin Hood & Clorinda” (1783), “Calisto” (1784), and “Zephyrus and Flora” (1784). These early commercial works also exhibit a notable contrast between bold patterns of lines and dots and delicate stippling.For a discussion of these early commercial engravings, see Essick, Printmaker 55-57. In “Morning Amusement” and “Evening Amusement” (both 1782) after Watteau, Blake deploys extremely dense linear patterns, with roulette and mattoir work in the foreground, while the facial features of the main group of figures, though showing less proficiency in the treatment of light and shadow than the anonymous portrait engraving, are rendered with fine stippling.For another example of extremely fine stippling, see “Venus Dissuades Adonis from Hunting” (1787). In this print, Blake uses coarse stippling on background foliage and delicate stippling on the central figures. See also Essick, Separate Plates 145-49. Another such contrast is evident in “The Idle Laundress” (1788), where dense linear patterns with heavy roulette and mattoir work on the boy’s coat represent a rough texture that is juxtaposed to the stipple shading on the face.

The technical and stylistic correspondences between the graphic syntax in the anonymous engraving and Blake’s early commercial prints prompt us to recall his claim in October 1804 that his experiences at the Truchsessian Gallery had inspired him to return to his aesthetic of twenty years earlier. Blake’s rhetoric in his letter of 23 October may seem hyperbolic, but for much of 1804 he was attempting to secure engraving commissions from Hayley.See Crosby and Essick, especially 68-72. No doubt recalling Hesketh’s criticism of his Cowper frontispiece (she believed that it hinted at her cousin’s insanity) and The Triumphs of Temper plates, Blake repeatedly informs Hayley that he is capable of altering his linear style to better represent Romney’s designs.Hesketh requested that Hayley not include an engraving of Romney’s portrait of Cowper in The Life of Cowper (BR[2] 105). In his letter to Flaxman of 7 August 1803, Hayley refers to Blake’s plates for The Triumphs of Temper, “I am sorry to say that the Ladies (& it is a Ladys Book) find Fault with the Engravings,” before revealing that “the Engravings of Cowper have also been heavily censur’d” (BR[2] 157). See note 18, above, and Paley, “Cowper as Blake’s Spectre,” Eighteenth-Century Studies 1 (1968): 236-52. In the letter of 4 May 1804, for example, he begins by claiming that he is developing a more fashionable aesthetic: “I thank you sincerely for Falconer, an admirable poet, and the admirable prints to it by Fittler. Whether you intended it or not, they have given me some excellent hints in engraving; his manner of working is what I shall endeavour to adopt in many points” (E 748).Blake may have been referring to James Fittler’s engraving of a shipwreck in Falconer’s poem. See Essick, Printmaker 173. By 7 August, he knew that Hayley was considering Watson to engrave Thomas Alphonso’s medallion. As we have seen, Hayley initially offered this commission to Blake the previous year. It is possible that, despite approving of Watson in his letter of 7 August, Blake may have felt threatened, particularly as her delicate stippling style was popular with Hayley’s circle.For a discussion of Watson in the context of Blake’s configuration of “Female Will,” see Essick, “William Blake’s ‘Female Will’ and Its Biographical Context,” Studies in English Literature 1500–1900 31 (1991): 615-30. The similarly delicate and skillful stippling evident in the recently discovered print certainly evokes Watson, although the clothing and background are very different from her work.

There are a number of idiosyncrasies in the characterization of Romney that are suggestive of Blake’s hand. Unlike the Ridley and Watson engravings, the print depicts him with a symmetrically bowed mouth.In both Watson’s and Ridley’s engravings, Romney has a slightly asymmetrical bowed mouth. The same visual trope can be seen in many of Blake’s paintings, including Adam Naming the Beasts, Eve Naming the Birds, and Christ Blessing.Martin Butlin, The Paintings and Drawings of William Blake, 2 vols. (New Haven: Yale UP, 1981) #667-68, 670, all datable to c. 1810. In the print, there are lines extending vertically from the left corner of the mouth, which, combined with the bowed mouth, hint at a playful smile that is absent from Ridley and Watson. Another distinctive feature is the enlarged and slightly protruding eyes. These differ significantly from the Ridley and Watson plates, but are similar to the prominent eyes in Blake’s own self-portrait of c. 1802.See Essick, “A (Self?) Portrait of William Blake,” Blake 39.3 (winter 2005–06): 126-39. Rather than the introspective, melancholic subject of the original painting and the Ridley and Watson engravings, the recently discovered engraving presents an artist who, if not “drunk with intellectual vision” (E 757), certainly possesses an outward-looking character.

If this print is by Blake, the slightly bulging eyes and bowed mouth may explain why Hayley rejected it. The expressive character departs not only from Romney’s original, but also from the tone of Hayley’s biography, which portrays Romney as melancholic.In his Life of Romney, Hayley follows the same autobiographical metanarrative that he establishes in the biography of Cowper, casting himself as curative for Romney’s increasingly frequent bouts of depression. See, for example, Life of Romney 255. John Romney addresses this characterization of his father in Memoirs of the Life and Works of George Romney: “Mr. Romney was naturally of a placid and easy disposition, and it was only in the decline of life, when his health was impaired by application, and his feelings ruffled by peculiar circumstances, that he manifested that morbidness of feeling, which Mr. Hayley has been so particular in noticing” (246). Hayley may have found the engraving excessive in its depiction of the artist as genius, a personality seized with excited imagination rather than inward reflection.For Hayley’s decision to choose Watson’s frontispiece over Blake’s, see also Paley, “William Blake, George Romney, and The Life of George Romney, Esq.” 58-59.

Without a sculpsit inscription, a definitive attribution of the recently discovered print is not possible, but it is certainly not improbable that the engraving is by Blake. To summarize the evidence: the graphic style used to delineate the print, and particularly the rather old-fashioned treatments of the clothing and background, provide a terminus ante quem of no later than c. 1820. Prior to 1820, the model for the print was available only via the following sources: Romney’s original painting, which was in Hayley’s possession, Blake’s miniature, which was in Long’s possession, and Ridley’s and Watson’s engravings. There are significant compositional and stylistic differences between the print and the Ridley and Watson engravings. These differences enable us to eliminate them as engravers and their respective engravings as models. Therefore, whoever executed the portrait engraving must have had access to the original self-portrait.

The correspondence shows that Blake worked on a quarto-sized portrait engraving of Romney after his own drawing of the painter’s self-portrait for at least sixteen months and sent Hayley two proof impressions. The recently discovered print is before all letters, indicating that it is a proof impression. The image size is within 1 mm. of Blake’s frontispiece for the first volume of Hayley’s quarto edition of The Life of Cowper.Watson engraved the frontispiece for the octavo edition of Hayley’s Life of Cowper (1806). Hayley consulted Blake, because in a letter of March 1805 Blake states that “the Idea of Seeing an Engraving of Cowper by the hand of Caroline Watson is I assure you a pleasing one to me” (E 764). At the time, Blake was negotiating with the publisher Richard Phillips about an octavo edition of Hayley’s animal ballads, illustrated with new engravings. The prospect of producing engravings after his own designs rather than a reduced version of a copy engraving that had already attracted significant criticism may have influenced Blake’s generous reaction to Watson’s gaining the Cowper commission. Until 1806, Blake’s engraving of Romney was intended to be the frontispiece for The Life of Romney, a quarto book.

The graphic syntax of the print is within the range of Blake’s techniques, including both the bold linear, roulette, and mattoir work on the clothing and the delicate stippling on the facial features. We know that during 1804 Blake repeatedly offered to adapt his linear style to include stippling and provided an example—his portrait of Euler—for how he could execute the Romney plates. By 7 August, he was also aware that Hayley was considering Watson—one of the most fashionable stipple engravers of the period—for the Romney biography, suggesting that if the print is by Blake, it is almost certainly not the proof state sent in July 1804, but either the later state sent to Hayley in December or an even more finished state executed in 1805. Finally, after his visit to the Truchsessian Gallery in October 1804, Blake informed Hayley that he had reclaimed the “intellectual vision” of his youth, which he had not experienced “for twenty dark, but very profitable years,” and stated: “I shall make my assertion good that I am become suddenly as I was at first, by producing the Head of Romney … quite another thing from what you or I ever expected [it] to be” (E 757). The recently discovered print exhibits the same combination of rough linear patterns, with roulette and mattoir work, contrasted with fine stippling conspicuous in the commercial engravings that Blake produced approximately twenty years before his visit to the gallery.

Considered together, both circumstantial and graphic evidence make it probable that the recently discovered print is a proof impression before all letters of Blake’s portrait engraving of Romney—a plate that was, in Blake’s own words, “quite another thing from what” both patron and craftsman expected. Unfortunately for Blake, Hayley appears to have had different expectations, as well as a different conception of “supernaculum.”