The last couple of days have been spent working through the fabulous new catalogue raisonné of George Romney by Alex Kidson (Yale). Anyone will immediately recognise the extraordinary achievement this represents: a lifetime’s work, not least because of the amount of travel involved – the author has bravely indicated which pictures he has personally seen, and it is a great many indeed. I’m certainly not a Romney specialist, and wouldn’t presume to review the book, but it provoked a few thoughts about what a catalogue raisonné should be.
Kidson’s approach is to strip the concept right down to exactly what it says – a list of the paintings that Romney made. With superb colour illustrations all the way through, on the same page as the entry (bravo Yale, and for going to three volumes so that we can handle the results).
What he doesn’t do is take up room with academic theories that really belong elsewhere (indeed he has created excellent space for others to do so). This is rather refreshing, as is the candid introduction in which he confesses to some of the difficulties of the task – ones with which any cataloguer of historic portraits will readily sympathise. And if there are shortcomings, they don’t really impinge on the real function of the book or in any way undermine its authority as arbiter of the œuvre.
I should also make my own confession: I’ve never really much liked Romney, regarding the well known pictures of Emma Hamilton as pretty awful (Kidson has had the sense to relegate them to a separate section so that you can easily avoid them). He very rarely used pastel, and then in such a personal way that he hardly needs my professional attention. But Kidson provides a strong argument for putting Romney back on the level he merits. For me that is still one rung below Gainsborough (for outright beauty and natural talent) and Reynolds (for interest and invention). And one can debate where he ranks against Zoffany and Wright of Derby.
Much of the success of a catalogue raisonné lies in pedestrian matters such as the layout. Here he broadly follows Mannings’s 2000 Reynolds in using an alphabetical sequence by sitter’s name. One can debate the minutiae (Kidson goes for the sitter’s name as at the time of the picture, which results in divided entries for ladies who marry or nobles who succeed: I wouldn’t do that), but the bigger discussion is whether that is the best way to understand an artist’s development. I took the same approach for the Dictionary of pastellists, largely because for so many portraitists the absence of progression makes it difficult to establish a reliable chronology (for the major artists I have however added supplementary files in chronological order, a luxury not possible with printed books); but Kidson tells us that the evolution of Romney’s style makes dating to within a few years quite feasible. It is a shame he didn’t allow us to see this.
One mechanical oddity which perplexed me is the treatment of “versions”. I think users will assume that those listed (with no particular warning) are intended to be autograph, but the note on p. 11 suggests not. Perhaps I am misreading the note but it seems rather curious (ambiguity is acceptable, indeed necessary, where a picture is known only from a bad photograph, but a great many of these versions have been seen).
There is a very brief biography of the artist (another book to follow?), and some details are picked up, but without discussion, in the entries. Kidson follows John Johnson in identifying Tom Hayley’s mother as Mary Cockerel, but this is disputed by some modern sources (e.g. the Oxford DNB or James King in his biography of William Cowper), and it isn’t clear on what basis Kidson repeats the traditional assertion. He may well be right to do so; I have no idea.
Kidson is also apologetic about the incomplete biographical details of the sitters. This is the area where academics look down their noses at portrait historians for bothering at all when they should be thinking about abstractions. In fact the biographical details are uncommonly accurate (anyone who works with standard genealogies knows that the appearance of certainty is a delusion). A few days in the archives would no doubt throw up some missing details: Miss Forbes, for example, was no doubt the illegitimate daughter named in her father’s will of 1773 as Jean (she married Weston in 1792); Lady Albinia Cumberland’s will was proved on 6 November 1850 when she must have been dead; while Thomas Keate married Emma Browne in 1784 in St George’s Hanover Square. But so what? It matters not at all that Lady Caroline Montagu was married to the 3rd not the 4th Duke of Montrose, but it may be of passing interest that she was a keen amateur pastellist: a knowledgeable sitter can influence a portrait. The Misses Harris’ dates are easily found – for example on the iconographical genealogies on my site – but the dimension those files offer that I believe is of real interest is in the comparative iconographies of families. Time and again Romney’s sitters are revealed also to have been painted by Reynolds, Russell or Gardner, and this Kidson does not tell us. There are numerous Ph.D. theses lurking in that database.
There is also little or no discussion of Romney’s visual influences, either of those he drew upon or those who drew upon him. Here’s an amusing example in which John Russell may have used Romney’s portrait of his friend William Hayley (Kidson 635), probably taken from the print by Jacobé, 1779, in the Chalk-writer which he exhibited in 1789:
Another case where Russell comes to our aid is with the San Francisco portrait of Colonel Thornton (Kidson 1290).
The author reports that this work from 1780 has traditionally been regarded as a portrait of Colonel Thomas Thornton (the rather colourful sportsman whose will in favour of his illegitimate daughter Thornvillia Diana Rockingham Thornton was the subject of cases before the Cour de cassation and the Prerogative Court of Canterbury), a suggestion Kidson dismisses because the sitter obviously cannot have been born in 1757, Colonel Thomas Thornton’s date of birth as given in the Dictionary of national biography. In fact that date is wrong: Thomas was born on 22 April 1750 (and baptised some days later at St James’s Westminster). But he is still too young, and in any case he looks nothing like the portrait of him by John Russell (of which there are a number of versions in pastel and miniature):
When the Romney portrait appeared in the Sir William Clavering sale at Christie’s in 1920 (omitted from Kidson, but reported in The Connoisseur, May 1920, LX, p. 46, where they lamented that it only fetched 420 guineas), it was described as of Thomas’s father, Colonel William Thornton MP, commander of the loyal volunteer force which his son inherited. A sensible suggestion until you discover that this Colonel Thornton died in 1769. In fact there is a much simpler solution with the Clavering provenance as the strong hint: the portrait is surely of another Colonel William Thornton (1728–1782), the younger son of an unrelated Thomas Thornton of Brockhall. A colonel in the first regiment of Foot Guards, he was aide-de-camp to General Clavering in India 1774 before fighting in North America in 1776. He rose to the rank of major general in the army, and is shown here two years before his death.