Before I go on, let me reassure you that the exhibition is in many respects brilliant. It is well hung and the lighting works (as you know I have reservations about LEDs in other contexts, but here they are fine). The size is ideal for a monographic exhibition. It would have been pleasanter if fewer people were present during my two visits (the National Gallery dropped me from their private views some time ago, but I imagine that was pretty crowded too). The catalogue is excellent, taking the form of an extended essay into which the exhibits are interspersed with other relevant portraits (almost all by Goya); biographies of the sitters are given in a separate section, while the summary details of the exhibits are banished to the back. (Personally I’d have left some of the noble titles in Spanish – it’s less confusing – but I’ll follow them here; and Carlos III as “amigo de sus vasallos” is not quite as Revolutionary as “friend of his people”.) It’s not the standard format for an exhibition catalogue, and it won’t please everyone, but for non-specialists (of whom I am certainly one when it comes to Goya) it has the merit of readability, and the assurance that comes from a single voice – that of the curator, Xavier Bray. He is always intelligent, informative and often persuasive. Unsurprisingly, as a Goya specialist, Bray is a believer.
But when you enter the first room (or at least when I did), the first reaction was one of shock, at just how bad these masterpieces were. Several things came to mind immediately. The first was the tale of the Emperor’s clothes. So often this is misquoted today as an illustration of cognitive error, when it is nothing of the sort: the courtiers are all well aware that the Emperor is naked, it’s just that they are intimidated into not saying so by the fear of being thought unsophisticated. You can’t say that Goya wasn’t great, that every inch of canvas or scrap of paper he produced wasn’t infused with genius if you want to be taken seriously as an art critic. It’s a bit like attacking Bach (although a better parallel might be Berlioz). Another much misquoted parable is the curate’s egg, which of course can’t be how the bishop’s subordinate felt obliged to describe it. But I was also reminded of a period in my life nearly forty years ago, when I taught mathematics at Oxford. Marking undergraduate scripts in what were called “collections” was a relatively routine business, with a familiar grading system based on the Greek alphabet, modified with pluses or minuses for borderline cases. But occasionally (and as far as I am aware this only ever arose in mathematics) you would get a script in which the student had supplied an absolutely brilliant solution to one problem, while the rest of his answers were so poor as to question how he had arrived at it. (Those were the days before internet plagiarism.) Those very rare cases were rewarded with an αδ grade. (Postscript: I note that Derek Parfit, in On what matters, discusses an Oxford alpha-gamma, a mark he awards to Kant; perhaps the possible discrepancy was even more extreme in mathematics.)
My notes continue, and not just in the early rooms, with comments like “poor”, “appalling”, “out of drawing”, “stiff”, “wooden”, “no anatomy”, “risible”. Each time, the insistent question: if this were not by Goya, what would we think? We are told (p. 59) that the portrait of Cabarrús is “strikingly modern” – I doubt if this was intended in the Churchillian sense – although the debt to Velázquez is acknowledged (and seems to me the entire font of the modernity); but while the latter’s Pablo de Valladolid strides convincingly, the gap between this pose and the leaden feet and empty gesture of Goya’s sitter is enormous. In the group portrait of The Duke and Duchess of Osuna and their children, the duque (who was incidentally an amateur pastellist, although this is not mentioned) leans “at a slight angle, giving the overall composition a natural and informal air”: perhaps, if you are a believer; for others it gives the composition the rigidity of folk art, notwithstanding the charm of the children’s toys.
It is helpful to be told that the portrait of Juan de Villanueva [cat. 39] is “one of Goya’s greatest portraits”, as that is candidly not immediately evident to me. The portrait of Ferdinand VII (shown above) is so bad, and so insulting, that the wonder is not that the new king preferred Vicente López (who was actually rather good, as a recent acquisition by the Fitzwilliam demonstrates), but that Goya wasn’t immediately locked up and the picture burnt. Or the other way round, as an auto-da-fé.
Even when we come to the some of the most famous portraits – the Duchess of Alba, from the Hispanic Society of New York (the loan much trailed in the publicity) – there are problems: supposedly a great beauty with “not a single hair on her head that does not awaken desire”, she comes across, with her plastered eyebrows and a vacant expression unassisted by rather flat modelling, more as a fishwife in her best clothes than a duchess. Does she have a severe anatomical deformity of the lower spine? And the gesture, supposedly pointing to Goya’s signature in the sand, might as well ask “whose feet are those?” – for they cannot be hers. Solo Goya indeed.
The claims that Goya brought a special psychological insight to each portrait run into some difficulties with the discussion on p. 166 about the artist’s willingness to paint in a new head on an old body – just as so many court artists have had to do over the centuries. When the text gets to The assembly of the Royal Company of the Philippines (mercifully too big to travel from Castres), the author becomes curiously ambiguous, as if this picture is a step too far even for the true believer.
And yet, when we come to the really great portraits – the National Gallery’s own Wellington and Andrés del Peral, the Prado’s Marquis of Villafranca, Dublin’s Antonia Zárate etc. – it is hard to say precisely why apparently similar ingredients suddenly work. It is certainly not simply the question of finish, although that is a large part of the problem to which I return below.
Bray tells us that Goya was “the greatest portrait painter of the day” (p. 23; on the back cover, that is diluted to “one of the greatest portraitists of his time”). That thesis is taken for granted in this show rather than challenged or investigated. The exhibition is literally “solo Goya”, in that no works by other artists are included. Thus it differs significantly from the aims of previous exhibitions such as the Prado’s own 2004 show, The Spanish portrait from El Greco to Picasso (which I think also included the Hispanic Society’s duchess), or the even broader Paris show the following year, which came to London as Citizens and kings in 2006. In both of these we were treated to some of the best Goya portraits, but up against the competition. “Of his time” runs from the end of the careers of Chardin and Perronneau to Ingres and Delacroix: in between there is some fairly heavy competition, notably from Jacques-Louis David. (The organizers almost certainly regard Vigée Le Brun as several levels below Goya; but you might remember her Calonne from the Royal Academy show in 2006 even if you haven’t been to her current retrospective in the Grand Palais. And it goes without saying that there is no mention of Nick Penny’s observation that the Rosalba pastel the National Gallery brought out of storage recently held its own against Wellington on the opposite wall in Room 39.)
Of David the catalogue merely reproduces (fig. 54) his equestrian portrait of Napoleon crossing the Alps, which, despite what may strike us today as over the top, is still an infinitely more powerful icon than the Goya equestrian portrait with which it is compared (fig. 52): a vacant Charles IV sitting astride a cart horse (more used to carrying grocery than kings) in a dull autumn landscape. David’s real genius is not conveyed – but you can see it for yourself in the National Gallery’s own collections. For me Jacobus Blauw and the comtesse Vilain XIIII are at least as great as anything Goya ever did.
Another point of comparison which the catalogue only hints at is that of Anton Raphael Mengs, the dominant figure in Spanish court portraiture from 1761 to 1777. Paintings such as his Marquesa de Llano (shown here) are surely vital to understanding and assessing what Goya was trying to do in his full length portraits, and I cannot see how the catalogue can have omitted it.
One of Mengs’s self-portraits is included, as a parallel to the early self-portrait, catalogue 1 in the exhibition. It’s tucked round behind the entrance door. Don’t miss it: it’s key to the discussion below. Dated to c.1780 by Bray (although the excellent essay by Manuela Mena Marqués gives a wider c.1775–80), it shows that by the time of his late start as a portraitist, he was quite capable of painting a straightforward, well-finished portrait in line with expected standards of the day. It’s curiously prescient of the late self-portrait, cat. 60. But one suspects that Goya would not have set the world alight had he continued in this vein.
What we also learn (catalogue p. 24) is that Goya made two attempts to join the San Fernando academy, aged 17 and 20, and failed miserably on both occasions. (No doubt the Beckmessers of that institution left score sheets marked with some of the observations above.) He was in effect an autodidact – a sort of Liotard perhaps, although Liotard’s approach to academic requirements was to over-finish his work, the opposite of Goya’s response. Goya’s dislike of Mengs’s high finish may even have had a political dimension: the style is what was taught in Paris, and practised in Spain by the stream of French artists from Ranc to Van Loo, but (particularly after the Seven Years’ War) was rejected in Britain, as I have discussed elsewhere. Curiously in this catalogue the word “realism” seems to refer to the robustness of this lack of finish.
So for me the essential question – and the fascination of this show – is whether the explanation for the shortcomings of Goya’s portraits is “can’t” or “won’t”, and “won’t” subdivides into “didn’t want to”, “didn’t need to” or “didn’t trouble to” (as in “clients too stupid to care”: see de La Houlyère’s remarks to Danloux in my Prolegomena). I don’t think “can’t” will wash: “can’t” couldn’t produce the dozen or so spectacular examples that do appear in this show (to which one can surely add the Prado’s Countess of Chinchón, fig. 58) or the many more that are good in parts or the couple of portrait drawings that are excellent. There is surely more to this than a bad egg, or a peculiar combination of Greek letters. And the show makes a good case for Goya’s forward-looking with a number of late portraits that strikingly resemble the work of Manet.
But the crucial point is the influence of Velázquez. T. S. Eliot said that after Shakespeare, it was impossible for anyone else to write verse drama for several centuries, such was his dominance; and when Goya so conspicuously adopted Velázquez as his model, there was perhaps a danger that he did so too soon (perhaps we had to wait for Francis Bacon). What he took most conspicuously from his idol was a broad-brush technique, but he did so without the security of a formal training that could have allowed to him to use this freedom safely. So we find the hit and miss record exposed so acutely in this show (but evident enough in the 2004/06 shows if we had looked with sufficient impartiality).
The question this raises is this: does a painter have to be good to be great? Academic art history loves itself to take a broad brush to its story. An artist like Goya who provides the single stepping-stone between Velázquez and Manet is an ideal component in the narrative. It matters little whether Goya himself was looking backwards if he can be slotted into this neat sequence, and if we can retrospectively project the vision onto him. He certainly had enough talent to bear that burden credibly, but for me this is an opportunistic, reductionist narrative that is unfair to a good many competitors who don’t fit so neatly. A painting isn’t only good because it leads somewhere.
The acid test is in Room 1, a separate display (why?) in which the attribution of the National Gallery’s own Doña Isabel de Porcel is called into question. Science cannot decide the matter; connoisseurship alone can do so. We are invited to ask if the quality is good enough, as though only Goya could paint that well. I’m not going to offer a guess about the author, but I don’t think quality alone is the point: the question is not whether it is weak, but whether the weaknesses are precisely those that Goya shows elsewhere. If this were not by Goya,…