Charles I in Three Positions

A face seen from the front can leave much concealed. Anthony van Dyck’s Charles I in Three Positions lets us move around the king without moving our feet: a profile, a frontal likeness and a three-quarter view occupy the same narrow space. Each offers another piece of the head’s shape. The sequence invites us to compare the changing appearance of the same features.

Three views of one face

At the left, Charles appears in profile, looking towards the centre. The nose, lips and pointed beard form a broken outline against the background. This view gives the face a sharp external boundary: the projecting nose, the small recess above the upper lip and the forward tip of the beard can all be followed as a continuous edge. The eye is partly concealed by the angle, while the ear and the fall of hair behind the head become more conspicuous.

The central head faces us. Here the silhouette tells us less about the nose, but the relationship between the eyes, cheeks and mouth becomes clear. Light spreads across the forehead and the bridge of the nose; shadows gather beside it and underneath the lower lip. The moustache stretches sideways, and the beard narrows to a point beneath a broad, pale face. We can compare the two sides in a way the profile prevents.

On the right, the head turns back towards the other two. Both eyes are visible, although one lies close to the edge of the face. The angle lets the nose project while preserving much of the breadth of the cheeks. It is the view that most readily reconciles the other two: the arrangement of features in the centre acquires depth, and the distinctive outline on the left becomes part of a rounded head.

The three faces do not form a conversational group. Their bodies occupy overlapping space, and the repeated features keep drawing us back to their common identity. Yet it is difficult to look at the picture without briefly imagining three people. The central gaze meets ours, while the outer heads seem to turn inward. The inward turns guide our attention back towards the central face. Our attention passes from one face to another, testing likeness as we go.

Try following a single feature across all three views. The moustache, for example, changes from an outward curl on the profile to a broad horizontal accent at the centre and a foreshortened shape on the right. The same exercise works with the eyebrow or the pointed beard. Recognition depends on relationships between features, even as their apparent shapes change.

Clothes that help the eye travel

The collars provide three bright, irregular platforms beneath the heads. Their lace borders are sufficiently varied to reward close attention: scallops, small openings and pointed edges interrupt the darker clothing. Against the pale skin, the lace can seem almost continuous with the head; against the coats, it becomes a distinct, intricate surface.

Below those collars, changes of colour keep the repeated bodies legible. The left coat is dark and cool, with narrow vertical highlights. The central red garment offers a warmer interval, crossed by a broad blue ribbon. At the right, purple fabric opens into large folds, with a hand resting across the chest. The cloth’s crests catch pale light, while deep shadows mark the hollows between them. A second hand appears lower down beside the left figure. These gestures give the lower half of the composition a different pace from the measured succession of heads.

Hair also changes the outline. In the left view, it spreads behind the neck in a dark mass; at the centre, curls frame the face and descend beside the collar. On the right, the turn exposes the separation between cheek, ear and the heavier fall of hair behind. The small glint at the ear offers a precise bright point among these dark shapes.

The background remains open and indistinct. Its blue-grey patches vary, but they never establish a room or a landscape that demands explanation. This leaves the spaces beside noses and cheeks especially useful. The background allows us to read an edge; the light and shadow within each face then help us imagine what lies behind it.

A painting sent to a sculptor

Charles wrote to Bernini on 17 March 1636 requesting a marble likeness. The bust was carved in summer 1636 and presented at Oatlands on 17 July 1637. Urban VIII intended it as a gift to Henrietta Maria, amid papal hopes of England’s Catholic return. Bernini received an £800 diamond ring in 1638. The bust was destroyed in the Whitehall fire of 1698. ^[1]

For a sculptor, each angle supplies information missing from the others. A frontal portrait can describe the arrangement of the eyes and mouth beautifully while leaving the depth of the head uncertain. A profile supplies information about that depth but conceals much of the face’s breadth. By placing the views together, Van Dyck offers points of comparison across the forehead, nose, jaw, neck and shoulders. The viewer can mentally connect them into a volume.

There are still limits to what a flat picture can provide. We cannot walk behind any of these heads, and the changing fall of light does not give a complete measurement of every surface. Turning such information into marble would require judgement. The painter has organised appearances; a sculptor must make a form that continues to work when the spectator changes position.

Colour adds another distinction. The blue ribbon, warm flesh and different coats help us separate areas on the canvas. In an uncoloured marble likeness, those divisions would need to be conveyed through shaping, depth and the behaviour of light. The fine pattern of lace presents a particularly clear challenge: in paint, pale marks and dark intervals can suggest delicacy within a shallow surface. In sculpture, a corresponding effect depends on carved relief and shadow.

The Royal Collection proposes Lotto’s triple portrait, once owned by Charles, as a possible precedent for Van Dyck. ^[1] The suggestion places this unusual arrangement within a practice of bringing several views together. It remains a proposed connection; the visual usefulness of Van Dyck’s design is apparent without requiring us to reconstruct a particular moment of borrowing.

A possible preparatory drawing

A black-chalk Portrait Study of Charles I in the Rijksmuseum offers a possible connection with the central head. The Frick Collection’s exhibition entry suggests that it may have been used for this part of the painting. It also observes that detailed drawings of heads are unusual among Van Dyck’s surviving studies, and reads the searching lines as evidence of work from life. ^[2]

The proposed relationship brings another stage of looking into view. The finished canvas presents three carefully organised likenesses; the drawing may preserve the process of finding one of them. The possible connection concerns one head; the preparation of the other two remains an open question.

Henrietta Maria’s response

The bust’s reception is given a personal voice in a letter from Henrietta Maria to Bernini, dated at Whitehall on 26 June 1639. William Hookham Carpenter printed an Italian transcription and English translation in his Pictorial Notices of 1844. In that published version, the queen expresses her and the king’s admiration for Charles’s bust and requests a similar likeness of herself, to be made from portraits delivered through Signor Lomes. ^[3]

The letter connects approval of one likeness with the desire for another. Henrietta Maria describes a work that the royal couple had seen and appreciated, then imagines extending the exchange. It also shows how portraits could serve a further act of portrait making: painted likenesses would travel in place of the sitter. Her request establishes an intention, rather than proof that a finished bust of the queen followed. ^[3]

The lost marble and the surviving painting now invite different kinds of attention. Van Dyck’s canvas preserves the visual information supplied to Bernini, but it also preserves colour, the contrasting textures of hair and lace, and the movement between separate likenesses. Those qualities can be enjoyed directly, even where the resulting sculpture can no longer be examined.

The Richelieu comparison

Another celebrated triple portrait makes the practical purpose of such images particularly clear. Philippe de Champaigne and his studio’s Triple Portrait of Cardinal de Richelieu , probably 1642, now in the National Gallery, repeats the cardinal’s head frontally and in opposing profiles. It was sent to Francesco Mochi in Rome for a full-length statue. It was not the source for Bernini’s earlier bust of Richelieu: that bust depended on other painted profiles, and its resemblance drew criticism. The Gallery also records that Van Dyck was approached about supplying a portrait for the proposed statue, but died before the commission was confirmed. ^[4]

The distinction matters when comparing the works. A similar format does not establish the same sculptor, commission or outcome. Richelieu’s triple portrait belongs to a related problem of obtaining a likeness across distance, with its own sequence of decisions. The National Gallery’s account draws attention to the uncertainty built into that process: the success of a sculpture’s likeness could be judged through the reliability of the paintings from which it had been made. ^[4]

From Rome to the Royal Collection

After ownership by Bernini’s family in Rome and subsequent collectors, George IV bought the painting in 1822 for 1,000 guineas. ^[1]

The Royal Academy’s 2018 exhibition guide records that Bernini kept the canvas and that it was exhibited at the Pantheon. It apparently attracted the admiration of Roman painters in the later seventeenth century. ^[5]

Its independent survival allows us to encounter a work made for another artist’s use as a portrait with its own compelling structure. The unequal turns prevent a mechanical symmetry, while the central face gives the eye a steady point of return. Every move towards the profile or three-quarter head tests that central likeness again.

Readers interested in Charles’s wider use of portraiture can explore Charles I as a patron .