Isabella Brant

The woman on the sheet is Isabella Brant. She was born in 1591, married Peter Paul Rubens, modelled for his portraits, and died in 1626. The man holding the chalk was her husband. In 1621, when he drew her, Rubens was the busiest painter in Antwerp, and also worked as a diplomat, carrying messages between courts. She was 30. The sheet he used is small and nearly square: 16.9 centimetres (7 in) tall and 15.8 centimetres (6 in) wide, roughly the span of a hand. That size tells you the setting. This was not made for a wall. It was made across a table, with the sitter present, in one sustained stretch of looking.

Before he drew a line, the paper was prepared. It was brushed with a brown wash — a thin, watered-down layer of brownish colour that turns the whole page a middle tone. The wash is the engine of the method, and the reason is simple. On white paper an artist can only darken: every mark takes light away, and the brightest thing on the page is the paper itself. On brown paper he starts in the middle and can move both ways. Black chalk pushes down into shadow. White chalk sits on top and stands for the places light strikes. The untouched brown does everything in between, for free. The technique has a French name, trois crayons, which means exactly what it says: 3 chalks — black, red and white — used together on one sheet. The red is the ingredient that makes the face read as skin rather than stone. It warms the lips, the cheeks and the rims of the eyelids — the places where blood sits close to the surface in a living face. Black is kept for the hair, the eyes and the dark of the dress. The division of labour is visible if you look for where the red stops.

To understand why the drawing exists at all, forget the frame it hangs in now. In Rubens's workshop a drawing like this was not a finished product for sale. It was stored information: a particular face, recorded accurately once from life, then kept and consulted whenever that face was needed for a painting. The sitter did not have to come back; the sheet stood in for her. This sheet proves the point 3 times over, because 3 oil paintings were built from it. The first came in the same year, 1621, and not from Rubens's hand. Anthony van Dyck, then Rubens's pupil, painted Isabella Brant in oils and gave the picture to his teacher as a gift; it now hangs in the National Gallery of Art in Washington. Consider what that transaction means. A pupil painted his master's wife without needing her to sit for him, because the record of her face already existed on his master's sheet, and he was allowed to use it. Rubens then painted his own versions: one between 1620 and 1625, now in the Cleveland Museum of Art, and another in 1625, now in the Uffizi in Florence. Isabella Brant sat once, for a drawing. 4 images of her came of it.

The dating is uneven, and the unevenness is honest. The drawing and the Uffizi painting carry firm years, 1621 and 1625. The Cleveland painting can only be placed within a 5-year span, 1620 to 1625. That is normal for workshop pictures: a drawing made from life in one sitting is easier to pin down than a canvas that may have been begun, set aside and finished as commissions allowed.

After Rubens died in 1640, the sheet began a second life as a collector's object. It passed through 5 collectors whose names are recorded, and the British Museum bought it in 1893. The purchase date survives inside the object's own catalogue label: the museum number, 1893,0731.21, begins with the year of acquisition. What the record does not say is just as telling — not what anyone paid, nor what each owner thought they had. Between 1640 and 1893 the drawing's history is a list of names.

The strangest chapter came in 1964, during re-mounting. A mount is the stiff backing a drawing is glued to so it can be handled and displayed; re-mounting is the routine conservation job of lifting the sheet off a worn-out backing and fitting a fresh one. A mount holds one face outwards and seals the other out of sight — so for as long as the old mount had been in place, nobody could see the back of this sheet. When conservators lifted it free and turned it over, they found a rough sketch there, also in red and black chalk. It is presumed to show Rubens himself with Hélène Fourment (1614–1673), the woman he married after Isabella Brant's death, together with a child of his. Presumed is the right word, used deliberately. The sketch is rough, nobody labelled it, and the identification rests on what the drawing appears to show, not on any statement by the people involved.

Hold the two sides together, because the pairing is the point. The front is a careful, warm portrait of a first wife, drawn from life in 1621 and praised ever since for its immediacy. The back carries a hurried sketch understood to show the second wife and a child, made later by a man who turned over a used sheet because paper was worth reusing. The two images travelled back to back through 5 collectors, a museum purchase and 7 decades of institutional care — and they were only ever seen as a pair because a backing board wore out.