| National Gallery Technical Bulletin Volume 6, 1982
Hogarth's 'Marriage à la Mode'
History, condition and treatment
In ‘Shortly after the Marriage’, the second of Hogarth's six paintings satirizing high life, called ‘Marriage à la Mode’, the steward carries a single receipt with the date 1743 on it. The series is generally assumed to have been more or less completed by the April of that year when advertisements appeared in the ‘Daily Post and General Advertiser’ announcing the opening of a subscription for 'six prints from Copper-Plates, engrav'd by the best Masters in Paris' after Hogarth's paintings (note 1).
In May 1743 Hogarth went to Paris to engage engravers for the project. It was his intention that each plate should be done by a different Master and in June he entered into agreements with six French engravers. However, soon after, war broke out with France and Hogarth, not wishing to risk sending the pictures to Paris, was able to persuade only three of the engravers to come to London, where they each undertook two plates. These reasons for the delay were explained in an advertisement in November 1744, when the prints were said to be 'in great Forwardness'.
It has been suggested (note 2) as a further explanation for the delay that Hogarth may have become dissatisfied with the paintings as a result of his trip to Paris and that he may have painted the whole series over again on his return. What has become clear from the present study is that minor changes were made in all of the pictures, and that one of them (‘The Killing of the Earl’) was considerably reworked: but it is not possible to say whether the alterations were done before or after Paris, or to speculate on the existence of another set of paintings by Hogarth.
The next mention of the paintings and prints occurs in Hogarth's 'Proposals' (dated 25 January 1745) for sale by auction of the paintings for his earlier series ‘The Harlot's Progress’ and ‘The Rake's Progress’ from which prints had been made in the 1730s. In the 'Proposals' elaborate conditions of sale are set out and a footnote says 'the Six Pictures call'd ‘Marriage a-la- Mode’, will be sold in the same manner, but the Book for that Purpose cannot be closed till about a week after the Plates now Engraving from them are finish'd, of which public Notice will be given.' On the reverse side of the 'Proposals' are the titles of the individual pictures: those given for 'Marriage à la Mode’ are substantially different from the titles now in common use (note 3).
It is not clear whether Hogarth ever intended that the paintings of ‘Marriage à la Mode’ should be sold along with those of ‘The Harlot's Progress’ and ‘The Rake's Progress’ (note 4). In the event, the auction took place on 28 February 1745 and ‘Marriage à la Mode’ was not offered for sale.
Shortly afterwards, just two years after the original advertisement had appeared, the prints were finished (note 5). They are reversed with respect to the paintings (Fig.1). This follows Hogarth's practice in earlier series which he had engraved himself: rather than engraving using a mirror, he would make his plate directly from a ‘modello’ which had been deliberately painted the wrong way round. As Paulson suggests (note 6), this may mean that Hogarth had originally intended to do the engraving himself (note 7): normally a professional engraver would produce an unreversed image of the original and would have to have been specially instructed in this case. It may be assumed that the 'final' compositions intended by Hogarth are those of the prints: certainly if we are to 'read* from left to right the construction is more logical than in the paintings.
Little was heard of the paintings for the next six years. They were for a time shown in the auction room of a friend, but no buyer was found. Eventually, in 1751, at the end of an announcement of a subscription for some quite unconnected engravings, Hogarth mentioned almost casually, 'the Author's six Pictures of ‘Marriage A-la-Mode’, which are to be disposed of within the said Time, by a new Method, to the best Bidder.'
The sale was to be on 6 June 1751, at Hogarth's own house (note 8). A complex system of bidding by written notes was proposed. Hogarth expected five- or six-hundred pounds, but only one note was received by the appointed hour, for £120 from a Mr Chas. Perry who was not present. Hogarth was outraged and dismayed, 'curs'd and damned the publick', and promptly sold the paintings to Mr John Lane from Hillingdon (the only member of the 'publick' who had turned up) who offered to make the pounds of Perry's bid into guineas (note 9). Hogarth made Lane promise that he would not dispose of the pictures without informing him of his intention, and that 'he would never let any person, by way of cleaning, meddle with them, as he always desired to take care of them himself.'
The twenty guineas apiece that Lane paid included four guineas for each of the 'elegant Carlo Maratt frames'. It must be assumed that they are the same ones in which the paintings are exhibited today: certainly these are eighteenth century English 'Maratta' frames (note 10), and it is unlikely that they would have been substituted in the seventy years or so before they entered the National Gallery.
John Lane kept ‘Marriage à la Mode’ until his death in 1791, when the pictures passed to his nephew. Six years later they were bought at auction by Angerstein whose collection was acquired by the nation in 1824 to form the basis of the National Gallery.
Hogarth had regarded the series as his masterpiece and it is evident that the public who now visited the new National Gallery were fascinated by it. Indeed, by 1860 concern was being expressed at the crowds which surrounded it. A memorandum, dated 22 November 1860, signed by W.E. Alldridge, the principal attendant in the Sheepshanks Gallery, reads:
‘The Paintings of the 'Marriage-a-la-Mode' by Hogarth attract on the free evenings a great number of persons to them so much so that I have found it necessary to place an attendant to look after them. The rails are so low and near that they are often in danger from hats of persons leaning over and the fingers of others pointing out the several parts. Might I suggest their being glazed, as an accident might happen to them in an instant however carefully looked after.’ (note 11)
Alldridge's memorandum was occasioned by an incident two days earlier involving one of the Hogarths (it is not clear which one). His report says:
‘Last night a Female leaning over the rails looking at one of Hogarth's paintings of the 'Marriage-a-la-Mode' was taken with sickness and vomited on to the board covering the hot water pipes and partly on the Picture. I immediately attended to it and cleaned it off with a damp sponge wool and silk handkerchief and no damage has accrued to the Picture.’
The incident was the subject of correspondence (note 12) from the Science and Art Department of the Committee of Council on Education at South Kensington to Eastlake, first director of the Gallery. The pictures subsequently were glazed, although today they are again shown without glass.
In the present century, all or some of the paintings have travelled to international exhibitions in Amsterdam, Paris, Chicago, New York and Toronto. For a time they formed part of the British Collection at the Tate Gallery. In 1950 they returned permanently to Trafalgar Square.
Condition and treatment
There is no record of the condition of the paintings until 1859 when all six were examined at Eastlake's instigation. Nothing is known of what happened to them while they were in the possession of Lane or Angerstein, or what state they were in when they entered the National Gallery in 1824.
Perhaps Lane obeyed Hogarth's strictures about not letting anybody 'meddle with them' but it is unlikely that he would have felt bound by his promise after Hogarth's death, twenty-seven years before his own. In any case, the pictures certainly were treated at some stage, either in his ownership, his nephew's or Angerstein's because three of them have repaired tears of which there are no records in the Gallery archives. The damages are all in an approximately similar position in the upper left quarter of each picture: it may be coincidence, or it may be that the three pictures were stacked together when the damages occurred.
All six canvases were lined by 1859; the three torn ones would undoubtedly have been lined during repair and it may be that the others were done at the same time. Five of the canvases were relined in the years 1964–65; the sixth (‘The Visit to the Quack Doctor’) was relined during the recent treatment of 1980–81, when all six paintings were cleaned, restored and examined by a variety of methods including X-radiography and paint sampling.
The most striking feature of the condition of the six paintings is the variety (both in type and degree) of cracking in the paint and ground layers. The 1859 examination report describes the paint in all of them as 'wire-cracked': this refers to a narrow but deep type of cracking which exposes the light-coloured ground and is especially disfiguring in dark areas.
Cracking of paint films is of two main types: drying cracks, initiated during the drying process, and age (or mechanical) cracks which occur after paint and ground have ceased to be flexible (note 13). Both types occur in the ‘Marriage à la Mode’ pictures, but in such complex networks in places that it is difficult to distinguish between them.
On a microscopic scale, drying cracks tend to have rounded edges, since they are formed while the paint is still plastic, but their overall pattern is unpredictable. In these pictures, many of them are long and straight, but others are short and exhibit the familiar branched 'alligator-skin' pattern (Fig.2); most of them are deep and penetrate to the ground, but some are shallower and the ground is not exposed. Many of the deep linear cracks run approximately horizontally or vertically (Fig.3), and seem to have been caused or aggravated by stretching or restretching the canvases while the paint was still relatively new.
Whether or not drying cracks develop depends on the properties of both the paint and the ground or underlayer to which it is applied. Considerable shrinkage of a paint film can result from its own drying mechanism, from use of excessive volatile diluent and from migration of the medium into a ground which is only partly cross-linked. If the ground or underlayer is still plastic, or its surface is glossy and rich in medium, then adhesive forces between layers are insufficient to counteract the contraction of the paint film. Formation of drying cracks may be caused by a number of factors such as these, and may be further aggravated by external influences such as exposure to light.
Many eighteenth and nineteenth century pictures show this phenomenon, usually resulting from the use of unsuitable commercially prepared grounds. It is by no means as pronounced in the ‘Marriage à la Mode’ paintings as it is in others of the period, but it is nevertheless clear that Hogarth painted on grounds that were poorly prepared or insufficiently aged and that he painted over paint layers that were not yet dry.
Superimposed on the network of drying cracks is an overall system of age cracks. These are formed when the paint and ground have hardened and become brittle, and therefore penetrate both layers. Some tension remains in the paint and ground layers, even when dried, and this results in the formation of concavities (cupping) between the age cracks. This tension is strong enough to set up distortions throughout the entire structure of a picture and leads ultimately to deformation of the support or to cleavage from it. In the six ‘Marriage à la Mode’ paintings, the canvas ground and paint layers appear to have remained firmly attached to each other, but the whole structure of each picture has cupped.
That much is clear, and, to a degree, predictable. What is not clear is why two of the pictures should have been affected differently and should have reacted differently to treatment. ‘The Visit to the Quack Doctor’ and ‘The Killing of the Earl’ are both markedly cracked (both drying and age cracks) and have clearly not responded to traditional lining methods which aim to reduce cupping. The other four appear less cracked, but have in any case responded well to lining.
These differences do not correspond with detectable variations in the canvases or the materials of the grounds, which are thick and multi-layered (see p.60). Only ‘The Marriage Contract’ differs appreciably from the others in having a finer canvas and thinner ground, but in appearance it is no different to the other better preserved pictures of the series. Presumably the explanation lies in factors of environment or treatment. Position of hanging, storage conditions, lining methods and cleaning reagents might all have varied from picture to picture within the series and could all have contributed significantly to the overall state.
There was little to be done in the way of treatment to the cracked areas of the ‘Marriage à la Mode’ pictures. The prominence of age cracks and cupping can be reduced by suitable lining treatments. ‘The Visit to the Quack Doctor’ was the only canvas lined in the recent treatment, and the improvement was only slight: however, it had had beeswax ironed into the back of the canvas as a conservation treatment at some earlier stage and this diminished the effectiveness of the lining method.
All the pictures were cleaned during 1980–81. There is no record of any previous cleaning, but it is unlikely that this was the first, since the varnish was not excessively discoloured. It was grey rather than yellow, and had acquired an opacity and mattness of surface which made satisfactory viewing of the pictures impossible. There were retouchings on the three damaged pictures to conceal the tears, and on all the pictures to disguise areas of cracking. These were removed during cleaning.
Retouching after cleaning was confined to the areas of damage and to the more prominent cracks. Where drying cracks are especially disfiguring, the only way of reducing their effect is to inpaint them using a very fine brush.
During the cleaning and restoration of the ‘Marriage à la Mode’ pictures, the opportunity was taken of carrying out a detailed examination of each, principally by methods of paint analysis (the results of which are reported in the section which follows on p.59), by X-radiography and by infra-red photography. It was possible by a combination of these techniques to assemble a reasonably complete account of Hogarth's materials and method of working in each picture.
X-rays were particularly useful in elucidating changes in composition made during the painting of the pictures. In most cases the changes were relatively minor – adjustment of details or alteration of outlines. Hogarth must have calculated his compositions very precisely in order to have organized within them the wealth of detail and allusion which surrounds the main theme. However, the fifth picture (‘The Killing of the Earl’) has been radically altered and X-rays suggest that Hogarth began it with a rather different scene in mind, a scene even more melodramatic than the one seen now.
The changes to each picture deduced from the X-ray mosaics are summarized below. Infra-red photography and reflectography are widely used to detect under-drawings but none was found in any of the ‘Marriage à la Mode’ paintings; they were, however, instrumental in revealing an amusing alteration in ‘Shortly after the Marriage’.
‘The Marriage Contract’ (Fig.4): This X-ray (Fig.5) is clearer and easier to interpret than the others of the series, presumably because the ground is appreciably thinner and the masking effect of the lead white content is less (see p.60).
The tear is seen as a black diagonal stroke above Silvertongue's head; changes to the composition are as follows:
The black-robed figure holding the plan and looking out of the window originally faced towards the Earl and appears to have been leaning over, both hands held up in front of him, to whisper to the man standing at the table.
The entire window was painted before the ‘baldacchino’ was put in.
Both dogs were originally lying down.
In place of the oval head of Medusa in its elaborate frame was a tall rectangular picture (unidentifiable) in a plain narrow frame. The top edge of the long horizontal picture on the right-hand wall was originally lower.
‘Shortly after the Marriage’ (Plate 8, p.44 and Fig.6): X-rays (Fig.7) show that the Viscountess's right arm was originally lower and that she was probably not holding a mirror. Her eyes appear to be looking the other way, but this is unlikely and interpretation is ambiguous here.
The table at which she is seated was originally without a cloth, and presumably without crockery – since the pieces are painted over the cloth. This is significant in view of Martin Davies's suggestion (note 14) that it would have been strange for the Viscountess to have been downstairs for breakfast. Hogarth probably included the breakfast things as an afterthought.
There are ‘pentimenti’ in the fingers of the Steward's upraised hand and a painted-out chair beyond the columns beside the Viscountess.
Infra-red photographic techniques were used to investigate the curtained picture with the foot on the bed (Fig.8). It was clear from a simple visual inspection that the picture-frame continued beneath the curtain and so presumably the picture itself did too. Was the rest of the figure present beneath the curtain? Was it indeed supposed to represent Hogarth's own ‘Danae’? (note 15). X-rays showed nothing; conventional infra-red photography penetrated the curtain to some degree (Fig.9), but is normally not successful in areas of green paint.
The picture beneath was finally revealed by infra-red reflectography, which uses an infra-red vidicon connected to a television monitor; it is sensitive to longer wavelength infra-red radiation than infra-red film and allows greater penetration of green areas. The image, photographed directly from the television screen (Fig. 10) showed that the foot and ankle continued no further than the edge of the curtain, and that originally the picture had been of a 'Madonna and Child'; just visible between the columns was the back of a kneeling man – suggesting an 'Adoration of the Shepherds (or Kings)'.
A 'Madonna and Child' would have been the logical companion picture for the four Saints that hang nearby. But obviously it was too logical for Hogarth, too solemn by half, and he painted it out and substituted the curtain and the suggestive foot.
‘The Visit to the Quack Doctor’ (Fig. 11): The paint layers in this picture are generally thin in relation to the thickness of the ground and for the most part contain pigments relatively transparent to X-rays. Only areas of densest paint – the woman, the Viscount, and the girl – show significantly in the X-ray mosaic (Fig. 12).
A small hole shows up black in the woman's upper skirt. The only major alteration which shows is that the woman originally stood nearer to the Viscount – a shadowy earlier head is just visible alongside the present one.
A few ‘pentimenti’ are visible on the picture itself. A retort under the table at the left has been painted out with the table-cloth, but now shows through. The woman's black skirt has been made longer: originally, most of her shoe could be seen. The upright screw on the machine at the right edge has been lowered so that the last lines of the inscription on the book can be read.
‘The Countess's Morning Levée’ (Fig. 13): X-rays (Fig. 14) show that the hairdresser attending to the curling of the Countess's hair originally had a fantastic hairstyle himself. His hair seems to have been piled up (so high that it reached the picture frame above him) and topped with curling papers rather like those on the man sipping chocolate, but even more elaborate than his.
The position of Silvertongue's head was changed twice. Originally, he was much nearer the screen than he is now, and may possibly have been sitting rather than lying on the sofa. In that position, more of the dressing-table would have been visible: it was indeed painted, but is now covered by his arm. The head in that first position can be seen in the X-ray mid-way between the present head and the upraised hand. In its second position, the head was slightly higher than it is now.
The plate in the basket with the 'Leda and the Swan' design appears not to have beeen present originally: in its place were two jars.
The catalogue in front of the basket was first painted with two white pages showing, but one is now painted out to form the inside of the dark coloured cover.
‘The Killing of the Earl’ (Plate 9, p.44 and Fig.15): Major alterations were made during the painting of this picture, all of which are visible in the X-ray mosaic (Fig. 16).
At the left-hand edge originally was the figure of a screaming woman, turning away or fleeing (and actually painted in two slightly different positions). As well as showing in X-rays, the figure can also be seen in an infra-red photograph (Fig. 17). The woman appears to be holding a sword diagonally across her body, the hilt alongside her face.
The presence of this figure (but not of the sword) has been known since infra-red photographs were taken in the 1950s. Davies (note 16) thought then that it was of an 'oldish' woman and that it was part of another, unconnected picture which Hogarth had painted over. However, careful examination of the X-ray suggests that the figure resembles the kneeling Countess whom we now see, and wears a similar cap. Could this actually be the Countess rushing away in panic with Silvertongue's sword to conceal his guilt? If so, Hogarth must subsequently have thought better of it, painted the Countess out and then placed her kneeling at her husband's feet.
There are only two other possibilities. One is the suggestion mentioned above that the figure is unconnected with the present picture, but the significance of the sword and other alterations described below now make this appear unlikely. The second possibility is that it is of a woman other than the Countess, perhaps a servant: but it is most unlikely that a second woman would have been present with the lovers in the bagnio and even less likely that she would have felt compelled to rush off with Silvertongue's sword. That this is indeed the Countess seems the most probable explanation.
The other major alteration to the composition occurs between the Earl and the doorway where there is a figure of a further woman, also subsequently painted over. (In this case, infra-red photographs show nothing.) This woman is wrapped in a shawl and a voluminous dress and holds up a lantern in her right hand. She appears to be the landlady, just come in through the doorway with the men of the watch. The door has been burst off its hinges and lies at an angle against the wall; in the final version the door merely stands open, although the lock has been broken off and lies on the floor.
We cannot tell now whether both painted-out figures were present at the same stage in the evolution of the painting. For example, the fleeing Countess may have been replaced by the kneeling Countess before the landlady was even started.
However, if both elements did coexist at some stage, the picture then would have been very different to that seen now. A speculative reconstruction of it is shown in Fig. 18. At the left, the Countess runs or turns, open-mouthed, with the sword clasped in both hands; Silvertongue climbs through the window; in the centre the Earl dies alone; at the right the landlady, lantern in hand, leads the men of the watch in past the wrecked door.
If it seems an awkward and unbalanced arrangement (and perhaps that is why Hogarth abandoned it), it should be remembered that it was being painted with an engraving in mind. In the reversed print the eye would be led naturally from left to right, from the incoming watch and the landlady to the dying Earl and the escaping lawyer and finally to the Countess fleeing with the sword.
Nevertheless, Hogarth did change the composition into an altogether more symmetrical one. Perhaps it was the asymmetry that displeased him; perhaps he was dissatisfied with the figures themselves (from the X-ray, the landlady does not appear a successful creation); or perhaps he had introduced a discordant detail, a complication of plot that he did not wish to pursue.
The painting out of the two figures and the consequent complexity of the layer structure undoubtedly contributed to the cracked state of this picture. But even without the ‘pentimenti’, the layer structure proved to be more complex than expected (see p.65 below) and this must be seen as a cause of many of the more superficial drying cracks.
The X-ray mosaic also shows the repaired tear (white, because it is filled with a dense putty) and a number of lesser ‘pentimenti’. There are changes, for example, in the positions of the Earl's head and legs: his body may originally have been twisted at a slightly different angle. Also there is a faint indication of a sword lying across his feet rather than falling.
Finally, there is a minor but intriguing alteration to the window. In the picture, Silvertongue climbs out into darkness, but in the X-ray there is light beyond the window. Can Hogarth once have intended dawn to be breaking outside?
‘The Suicide of the Countess’ (Fig. 19): X-rays (Fig.20) show the tear in the upper part of the picture, but reveal little in the way of ‘pentimenti’.
It appears that the table may originally have been without its cloth, as the table was in ‘Shortly after the Marriage’, but this interpretation is less certain here.
The still-life picture with its black and gold frame was painted entirely before the picture hanging in front of it was added.
The overturned chair once had a high rounded back like the one by the window. Minor changes such as this are easily visible on the picture itself. Alterations to outlines such as the Countess's cap and her father's shoulder also show. One of the alderman's buttons has been painted out, presumably to indicate his parsimony at not having a replacement sewn on.
Notes and references
For details of 'Marriage à la Mode' in particular and Hogarth's life and works in general, see Davies, M., ‘The British School’, National Gallery Catalogues, 2nd ed. (London 1959), and Paulson, R., ‘Hogarth: His Life, Art and Times’ (2 vols.), Yale University Press (1971). The definitive work on 'Marriage à la Mode' is likely to be Cowley, R., ‘Marriage A-la-Mode: A Review of Hogarth's Narrative Art’, Manchester University Press (to be published 1982 or 1983).
Beckett, R.B., ‘Hogarth’, Routledge and Kegan Paul (London 1949), p.16.
For instance, the second picture is 'The Tête à Tête,' which conveys a rather greater irony than the present title of 'Shortly after the Marriage'. Also Hogarth invariably wrote of 'Marriage A-la-Mode' rather than 'Marriage à la Mode' which is used today and which to Hogarth's contemporaries would not have had the same satirical meaning. For drawing his attention to this, the author is grateful to Robert Cowley (see Note 1) who discusses the sale 'Proposals' in his forthcoming note, 'Hogarth's Titles in his Progresses and Other Picture Series' (to be published in ‘Notes and Queries’, 1982 or 1983).
Although Hogarth had included a small reproduction of 'Shortly after the Marriage' in the engraving called 'The Battle of the Pictures' which was the ticket for bidding at this sale.
All dated 1 April 1745. The date on the Steward's receipt in the second print is June 4, 1744, which presumably indicates when this particular one was completed.
Paulson, R., ‘op. cit’., Vol.1, p.478.
In fact, in the second announcement of the publication of the prints (dated 4 April 1743). Hogarth had inserted '(the Heads for the better preservation of the Characters and Expressions to be done by the Author)' but it is not generally thought that he took any part in the engraving.
Described on p.156, Vol. Ill of the notebooks of George Vertue, published in six volumes by The Walpole Society (Oxford 1930–55).
Lane's own account, quoted in Paulson, R., ‘op. cit’., Vol.11, pp.124–6.
See Grimm, C., ‘The Book of Picture Frames’, Abaris (New York 1981), p.241.
National Gallery, conservation archives.
National Gallery archives.
For a fuller discussion, see Keck, S., 'Mechanical Alteration of the Paint Film', ‘Studies in Conservation’.
1 (1969), pp.9–30.
Davies, M., ‘op. cit’., p.50.
Davies, M., ‘op. cit’., p.62, note 45.
Davies, M., ‘op. cit’., p.54.