Infanta Maria Teresa c.1651 [Metropolitan Museum of Art, New York]

Inconsequential mysteries are sometimes intriguing. Diego Velazquez (1599-1660) was nothing if not adept at describing both features and spirit in his portraits–as here, the Infanta Maria Teresa. How then to explain the odd face of his Rokeby Venus?

“Vague” is not the word for Venus’ face;  Cupid’s face is only suggested, but quite in keeping with his body; Venus’ face is lumpy and unresolved. It’s not the effect of the mirror. The back of her head suggests a graceful do, but from the front her hair looks as if she just got done toweling it. Her forehead is higher than it seems from behind. It’s not the face one would expect to find associated with one of the suavest derrieres in the history of painting.

Jonathan Brown* notices the oddness of the face, and attributes it to Velazquez’s effort to avoid resemblance to any actual woman. But surely Velazquez, of all people, could invent a face both anonymous and convincing.

c. 1646 122x177cm (Wikipedia)

Venus and Cupid (the Rokeby Venus) c. 1646   48 x 70″   [Wikipedia]

One possibility, I suppose, is that what we’re looking at is underpainting. The painting has a history of inept cleaning; the face might have been eroded in the process. I throw that in because if one writes about this sort of thing one ought to have a theory to contribute, however far-fetched.

But if overcleaning seems unconvincing, and delicacy about the identity of the lady unlikely, how do we explain that face? Another possibility is that Velazquez simply didn’t get it quite right. Maybe he was distracted–he was about to leave for a long trip to Italy. Maybe he was packing, wondering how many shirts he’d need, and didn’t have time to go back and contemplate the finished piece. Dealing with an artist of his grandeur, the normal assumption would be that he had good reasons for doing what he did, and if you don’t get it you should keep trying. But even great artists are human, and sometimes that assumption is contradicted by the evidence.


* Velazquez, Yale University Press 1986


Here we have two pieces–one by Cy Twombly (1928 – 2011), the other by Rembrandt van Rijn (1606 – 1669)–that begin with opposite intentions, but end with similar results.

Cy Twombly "untitled" 2005

Twombly “untitled” 2005 []

The Twombly is not, at first, surprising; swirling and splashing is what his painting is about. But while it’s certainly abstract, it’s not quite non-representational. His swirlings have an intricate and repetitive tension about them that implies a narrative. The swoops  draw the eye around and around and toward the right, while the background veil of thin vertical drips provide stability for the swirls to play against.

So: abstract, but inclining to narrative.



Rembrandt - "Landskab med hytter under store træer" c.1650 (W'ped)

Rembrandt – “Landskab med hytter under store træer” c.1650 (W’ped)


While the Rembrandt, also predominantly composed of swirls,  proceeds from the opposite direction, from narrative to abstraction. We spot the subject, a view of cottage and trees, right away, but it isn’t a formal illustration of picturesque cottage and tree. What Rembrandt was responding to was the lashing, constantly changing shapes as the branches whip this way and that in a strong wind blowing from left to right. It’s the wildness, the disorderliness of it, that he’s after. The action is in the swirls of foliage. The straight, mostly vertical lines of the cottage, and the oblique foreground ditch, provide the same still grounding that Twombly gets from his thin vertical drips. And for a contrasting negative, each artist provides a blank space beginning at the upper left–not as spacious in the Twombly as in the Rembrandt, but present in both pieces, and for the same reason.


For a discussion of similar issues with respect to Turner, Constable, and Kline, go >blog>archive>May 12, 2012: abstract and not.


Hans Bol - "Landscape with the Fall of Phaethon" 1569

Hans Bol – “Landscape with the Fall of Phaethon” 1569  (Courtauld)

We grow so accustomed to seeing really good art that it’s easy to miss what makes it good–hence the comparison here between treatments by Hans Bol (1534-1593) and Rodolphe Bresdin (1822-1885) of figures in a big landscape.

The vital difference here is coherence. Nothing in Bol’s piece quite relates to anything else. It’s a jumble of elements that might be found in a stylishly grand landscape, though usually in pursuit of some overall narrative which is lacking here. The foreground group, four figures and a fish, is drawn as if each were a separate little study that has been cut out with scissors and pasted down wherever there is room. The whole group on its little island has the same lack of relation to the rest of the composition as  the pieces do to each other. It kisses the farther shore in a few places, enough to muddle, for example, the woman’s outline, without really connecting foreground and distance. Follow the shoreline as it drifts aimlessly, rising and falling. Middle tones baffle Bol. The head of the fish, for example, is fully rendered from dark to light, but the body is left blankly white. Similarly, the ground that the figures inhabit is left white unless there is some immediate shadow, when some intermediate tone would begin to unify them. The water behind the woman’s head, and the shoreline beyond it–same thing. They are seen one by one, rather than as contributory pieces of a whole with a whole meaning. As for poor Phaethon and his horses, lost in a muddle of shapeless clouds–they are too insignificant to mention.

Rodolphe Bresdin - "Bathers in a Mountain Pool" 1865 (Art Institute of Chicago)

Rodolphe Bresdin – “Bathers in a Mountain Pool” 1865 (Art Institute of Chicago)

Bresdin’s landscape, on the other hand, while quite similar in design, is seen as a whole first, in which we can distinguish parts. The large mountain shapes to left and right form a V, joining with the foreground copse of trees. The trees gather into strong darks above the water–so dark that the eye jumps to the whites of the bathers. Contrasted with the middle tone of the pond they are drawn very simply (compare with Bol’s muddled lady).

From the V the hills fold back, first in a long recession on the right, then a more distant formation on the left, and finally to the far horizon. The richly worked sky–a pattern of horizontals–links the mountains to right and left, and by its weight keeps it all from floating away.

Within this grand design there are many unifying themes. Take the diagonal of the leftmost bather, then let your eye range up and back: you will find that same diagonal several times in the nearest trees, in the main line of the V at the top of the copse, and again and again in the central hills (and distant formation). Pick another diagonal, and look for its rhymes. Then another. Look for bright white shapes. Look for textures. Again and again they are varied and repeated. The horizontal shading of the water reflects the sky. The light, meanwhile, comes consistently from the left.

In a word, Bresdin’s is a wonderfully intricate but in no way mechanical composition. Compare with Bol. You will, in fact, find resemblances, but the results are very different.