This handsome small variant of Gilbert Stuart's famous life-size, full-length portrait of George Washington (c. 1797-1801) is a very important addition to the body of works spawned by Stuart's iconic creation.(1) Some small painted copies are cited in older literature, but this is the only extant small version of the "Lansdowne" portrait known to us, and moreover it is significantly a variant, not a literal copy. It was painted by the Massachusetts-born Edward Savage who had already painted the President from life in New York City in 1789-90 at the request and by the arrangement of the President of Harvard University. (The portrait still resides at Harvard, to which Savage presented it.) Josiah Quincy, an American statesman and later president of Harvard, "always declared that the portrait by Savage in the College . . . was the best likeness he had ever seen of Washington."(2) At the time he was painting the first president, Savage had already begun to plan The Washington Family (1789-1796), his remarkably ambitious (8 feet wide) life-size group portrait (National Gallery of Art, Andrew W. Mellon Collection) (fig. 3). From 1791 to 1794 Savage was in London, and he had taken this painting-in-progress with him. Upon his return he made changes, in part to update the ages of the grandchildren of Martha Washington. His subsequent print after the painting, issued in 1798, became the most widely distributed of its time and made his portrait famous-the authoritative image of the first President and his family. But already, as Savage was finishing the painting and then exhibiting it (beginning 22 February 1796, Washington's birthday) in his own Columbian Gallery in Philadelphia, Gilbert Stuart was only about to begin work in the same city on his most famous painting, the so-called "Lansdowne" portrait of George Washington. Several replicas followed in quick succession.
It is important here to briefly summarize the history of Stuart's "Lansdowne" portrait of Washington and its replicas. The famous 1796 "Athenaeum" portrait of Washington-showing the proper left side of his head-was painted from life in preparation for a portrait commissioned by a Philadelphia merchant, William Bingham, recently elected to the U. S. Senate. This full-length portrait (1796; National Portrait Gallery) was given by Bingham and his wife to their friend William Petty, first Marquis of Lansdowne, whose title came to denominate all the Washington portraits of this type, this composition (fig. 2). Senator Bingham then ordered a replica for himself (1796-97; Pennsylvania Academy of the Fine Arts). This is the only one of the Lansdowne portraits that is signed and dated, probably a bow by Stuart to the Binghams for their essential patronage.(3)
By this time Stuart had transferred his studio to nearby Germantown (to avoid the yellow fever epidemic) where he continued working on the replicas that were being requested. The second replica (1796-97; Brooklyn Museum) was commissioned by William Kerin Constable, a New York merchant, who saw the Lansdowne original still in progress and requested that his replica be painted simultaneously so that it would be equally an "original." A third replica was commissioned by Charles Cotesworth Pickney, who had arrived in Philadelphia in September 1796 to receive his credentials as minister to France and was having his own likeness taken by Stuart. Since none of the Lansdowne portraits had yet been delivered Pickney would have seen all three in various stages of completion. This replica, purchased by the U. S. Government for the American residence in Paris, was completed by July 1797. The painting was never delivered when Pinckney was not credentialed by the French, and the present writer believes that this is the replica that has been in the White House since 1800.(4)
Edward Savage could thus have seen any of these paintings had he visited Stuart's studio in 1796-97, or seen the Bingham replica in the senator's collection. Savage was resident in Philadelphia from 1795 until 1801 so he had the time, opportunity, and need to see this painting. The sittings given by Washington to Stuart were public knowledge and newsworthy, and it seems more than likely that Savage saw one or more of the Lansdowne paintings earlier rather than later. Moreover, he would have had strong motivation to paint a reduced variant as soon as possible for exhibition in his Columbian Gallery-to trade on the publicity attending the painting that had quickly become the standard image of Washington as President. Since the Stuart original and replicas rapidly passed into private hands, to be seen by relatively few, Savage's small version would have given the public an opportunity to experience Stuart's portrait at second-hand and, not incidentally, to see Savage's recent and larger painting of The Washington Family in the same visit. In addition, as will be seen in a moment, Savage intended to make an engraving of the portrait and his small painting would serve as the modello for it. He had previously painted a small oil as a modello for his stipple engraving of The Washington Family (Winterthur; see note 4). But it is essential to recognize that, as we shall see later, neither Savage's painting of the Lansdowne type nor his mezzotint are close copies of Stuart, nor were they intended to be. They are carefully calculated variants.
When he moved to New York in 1801 he transferred his Columbian Gallery there. Documents for the Savage variant of the Lansdowne are limited. Most important is a list of contents of the Columbian Gallery dated April 6, 1802 (original, Ryerson Library, Art Institute of Chicago) which includes "A Whole Length Portrait of George Washington G. Stewart." Since this cannot possibly be an original Stuart, and since the entry does not say "life size," we may conclude that it was Savage's small "whole length" variant.
The clearest demonstration of Savage's authorship of this small "whole length" version is in fact the visual document already referred to: the mezzotint engraving of his Lansdowne variant published by Savage in 1801 (fig. 1). Impressions of this engraving are quite rare, perhaps because the copper plate was too soft to withstand numerous printings. It was reproduced in 1932 by Eisen who noted differences between it and the known Lansdowne painted portraits. He then illogically concluded that Stuart must have provided Savage with a new model to work from.(5) Instead, comparison of the engraving with our painting makes it clear that our painting is his modello for the print, and therefore the changes in design from Stuart's Lansdowne were Savage's intention. The print, an undoubted Savage, demonstrates the authenticity of the small painting, and these two works are mutually reinforcing. Stuart, on the other hand, would never have produced a new model that was so distinct from his original painting.
In his painting Savage incisively departs from Stuart's physical type of Washington. The head of the president seems small in proportion to his height, he is less compact than Stuart's Washington. It is well known that Stuart had used a stand-in for the president while painting the body, and that contemporaries had noted the discrepancy between the generalized proportions of the body and the fidelity of the head. Although many writers do not closely address the matter, Washington's idiosyncratic physique-large feet, wide hips, narrow shoulders, proportionately small head-are a matter of record. I have pointed out elsewhere that the most reliable physical portraits of Washington are those of Charles Willson Peale, whose life painting and many variants of George Washington at Princeton are the work of a skillful artist who knew and painted Washington over a longer period than any other. In his thoughtful variant Edward Savage has restored those proportions to the Lansdowne pose, because he had also given probing attention to the president's stature and seemed determined to preserve it even in a small painting.(6)
It is in the mezzotint (fig. 1) that Savage's purposeful alteration of Stuart's design is most immediately apparent to the viewer: The picture space is wider in the print, showing more of the table and of the books at the left and the entire chair at the right. In its increased width it differs from every other painting or print, whether by Stuart or his copyists, all of whom, for instance, crop the chair in just the same way. Savage may have felt that his change was an improvement that gave greater amplitude to the image. But in his small painting he had also widened the space, not as dramatically but quite noticeably. Apparently pleased with the change, he carries it further in the print. Another small, but telling contrast between Stuart's several Lansdowne portraits and Savage's painted interpretation is the treatment of the drapery tassel at the base of the column above Washington's hand. Stuart's tassel has a certain painterly vigor and variety while Savage's expires limply on the wall. (Savage rethought this in the mezzotint, revivifying it somewhat, in Stuart's manner.)(7)
Savage's painting and the print also agree in the slightly puzzling elimination of the rainbow in Stuart's painting, signifying the new political covenant. Savage may not have fully grasped its symbolic meaning. But Stuart introduced another complex and resonant contemporary political reference into his painting, embodied in the oratorical pose and gesture of the President, which has only recently been persuasively revealed to our modern eyes and minds.
The original Stuart full-length of Washington was commissioned by Senator Bingham at the moment when the treaty negotiated by John Jay that resolved long-simmering trade problems left unresolved by the Treaty of Paris was being debated by both the British and American legislatures. President Washington, who had extraordinarily dispatched the sitting Chief Justice on this mission, is now understood as addressing Congress in his last annual message in December 1795, on which occasion he urged the passage of the Jay Treaty. A London newspaper later described the original Stuart Lansdowne portrait thusly: "The figure is standing and addressing the Hall of Assembly. The point of time is that when he recommended inviolable union between America and Great Britain."(8)
Edward Savage's small full-length variant portrait, excellently preserved, retains the highly charged significance of Stuart's imaginative invention, while insisting on the accurate physical record of George Washington's stature that Savage had carefully represented in his earlier masterpiece, The Washington Family.
(1) The painting has received careful study and its excellent condition has been noted while it was on loan to the National Portrait Gallery and the Metropolitan Museum of Art, where it was on extended exhibition.
(2) Ellen G. Miles, George and Martha Washington: Portraits from the Presidential Years (Washington, D. C.: Smithsonian Institution, National Portrait Gallery, 1999), p. 50.
(3) Because it is signed and dated it had usually been assumed that this was the first painting (later willed by Bingham to the Pennsylvania Academy), but the most recent scholarship firmly demonstrates that the painting in the National Portrait Gallery is the original. See Carrie Rebora Barratt and Ellen G. Miles, Gilbert Stuart (New York: The Metropolitan Museum of Art, 2004), pp. 175-76.
(4) A discussion of the history of this painting and the Lansdowne sequence is in William Kloss, Art in the White House: A Nation's Pride, 2nd ed. (Washington, D. C., 2007), pp. 68-69 and pp. 269-70, notes. See also Ellen G. Miles in Barratt and Miles, pp. 180-183, for significant additional documention.
(5) Gustavus A. Eisen, Portraits of Washington (New York, 1932), p. 107 & p. 243, pl. XLVI.
(6) William Kloss and Diane K. Skvarla, United States Senate Catalogue of Fine Art (Washington, D. C., 2002), p. 412. It is also observed there that John Trumbull's full-length painting and Jean-Antoine Houdon's sculpture second the evidence of Peale's likenesses. To them can now also be added Savage's rendering.
(7) Savage had previously painted a small modello for use in preparing his stipple engraving of his The Washington Family (now at Winterthur, 18 1/8 x 24 1/8"). It has been asserted since the early nineteenth century that while Savage "drew the outlines on copper" for his Washington Family, it was largely completed by the Englishman David Edwin and perhaps by his apprentice John Wesley Jarvis. No such assertion has been made for the mezzotint after his small Lansdowne copy which shows a high degree of technical proficiency.
(8) Barratt and Miles, p. 172.
William Kloss, November 2007