Perfecting a rainbow: John Constable and David Lucas, ‘Salisbury Cathedral from the Meadows’, 1834-7. Part 4 – Catalogue of states (01e)

This is the fourth part of an examination of the various states of Constable’s mezzotint of Salisbury Cathedral from the Meadows. This was engraved by David Lucas under the artist’s supervision and instruction between late 1834 and Constable’s death on 31 March 1837. Here we continue the first section of a catalogue of states:

John Constable
Salisbury Cathedral from the Meadows, 1829- exh RA 1831
Oil on canvas, 1537 × 1920 mm
Tate Britain, London (T13896)
Image courtesy of Tate

Impressions taken before the addition of title and first publication line (continued)

R B Beckett [Constable’s Correspondence, IV, p.457, XI (b)] published a list of alterations that he dated to late 1836. These changes can here be situated here between the previously discussed impression 01d (ii) sold at Christie’s in 2009 and the next, 01e (i) at the Victoria and Albert Museum, London.

The lightening & a cluster of foliage near the spire – to take off monotony.

[Constable seems to have required Lucas to take down the tone of the foliage over the north transept. In the V&A proof he appears to reinstate some of the highlights].

Right hand turret of west end and more light edges in centre bit.

[Lucas darkened the sky beyond to achieve more contrast. If he altered the tower or gable itself that is lost by Constable’s vigorous scratching out in the V&A proof].

Waggon spokes of wheels a little.

[Lucas seem to have added a few highlights, but Constable called for more on the proof].

Elder bush – & additional blossom.

[Again Lucas appears to have been tentative with these changes. Constable called for more in the V&A proof].

Dog – darker neath chest

[Lucas deepened the tone of the bank by the dog’s chest]

Post of hand rail – light top.

[This is the diagnostic change that positions these instructions here in the sequence. This detail is not present in the Christie’s 2009 impression 01d (ii), and first occurs in the V&A proof 01e (i)].

Window in St Edmond’s tower.

[The church of St Thomas and St Edmond to the left. Lucas seems to have added several highlights, later subdued].

Window in gable end of house.

[This presumably relates to the cottage below left of the church tower, but Constable’s suggestion does not appear to have been implemented].

Cow a little – nose light spot.

[The cow’s nose is better defined by toning down the background and adding a highlight].

On 10 November 1836 Lucas visited Constable’s studio and watched the artist demonstratively work over a proof, and put the result in a glazed frame. The following day Constable wrote: ‘I send you a few notes, of what we did yesterday. I have not taken it out of the frame today as I felt satisfied & more, delighted, at the degree of perfection it now presents.’

He was still far from reconciled to the publishing trade: ‘All that I wish is that this print may be made as perfect as possible – before it is shown to the Sharks again. The better the bait, the better for all – though I think with you, that the Sharks have a greater appetite for the gross than for delicate food, they must play about the ship & amuse themselves as well as they can till we have the hook ready for them again. I should, however, prefer the “City” Sharks as the most ravenous – how abominable is their conduct! How like the booksellers & the poor authors. ‘

He concluded ‘Joseph [Constable’s man] shall leave the frame & glass if you like – it is of no use to me, & I should like to have this proof preserved as it is for me.’ The V&A proof may be the very same as that discussed here by Constable for it descended in the Constable family and was bequeathed by his daughter Miss Isabel Constable to the museum in 1888:

01e (i)

David Lucas with John Constable
Salisbury Cathedral from the Meadows, 1836
Mezzotint, (image) 438 x 555 mm
Victoria and Albert Museum, London, 1250-1888

This was catalogued by Shirley as his 39e:

Same size. ‘J Constable RA  D Lucas’ in scratched letters below the framing line. Much etching added on tree, and l. side generally. With the birds added. The rainbow averages 7/8 inch in section instead of 1 1/2 inches. Before the rainbow is returned in the top l. corner. Freely scratched and touched with white chalk and wash to show the enlargement of the rainbow lights in foreground and the water more disturbed by the cart. – V&A.

The addition of the names of the artist and engraver would suggest that Constable thought the print worth distribution by this point. The prospect of putting it before a wider audience rather over-provoked Constable’s critical faculties. Constable so comprehensively covered this proof with suggested alterations in scratching-out, Chinese white and grey bodycolour that it is difficult to see through to the original detail in order to make comparison with the Christie’s 2009 impression but some significant changes can be made out.

Click on any image to open in gallery view and flick between images. Close gallery to return to this page.

The most obvious change is that Constable has completely revised the rainbow fading it to evanescence towards the top.

There is another one clear and diagnostic innovation in detail. A figure has been introduced into the meadow in the middle-distance right, and the poplar has been brought forward and given a shadow. The animals that first appeared in the Fogg proof suggested a string of cattle, at least to my eyes, but here their scale is diminished to sheep, and Constable’s overworking in that area seems intended to clarify that decision.

Constable drew so many improvements on the V&A proof that he felt compelled send a written list to Lucas on 11 November 1836. We may work through this list list and its implementation in relation to a progress proof at the Art Institute of Chicago which represents the first instance of these changes being put into effect. It does not seem to have been known to Shirley.

01e (ii)

David Lucas with John Constable
Salisbury Cathedral from the Meadows, 1836
Mezzotint, (image) 438 x 555 mm
Art Institute of Chicago, USA (1929.276)

It must have been a cause of wonderment to Lucas that Constable could find so many changes to make in an already astonishing engraving. Despite their number, the changes are worth consideration for they are all real improvements. Such powers of discrimination must seem nearly incomprehensible to contemporary image consumers. In the following passage, Constable’s instructions in the list are given in italics.

SKY. Light clouds put in at top on the left, flying and detached

Click on any image to open in gallery view and flick between images. Close gallery to return to this page.

The V&A impression has dabs of white at the top of the tree left and right, duly implemented in the next state 01e (ii), and absent in the previous state, 01d (ii). In many ways this change typifies Constable’s aesthetic sensibility. It is a slight change but it gives the main tree mass much greater definition against the sky. It gives greater depth of form and this in turn signifies a truer conception. This might be taken as characteristic of what he understood to be ‘the moral feeling in art’ (see Part 3).

Do. On top of the great cloud (see through the flying dark bits) so as to heighten the great cloud.

Lucas brightened the top edge of the billowing cloud left of the cathedral spire. By the Chicago impression this section has gained considerably in contrast, the darks being deepened and the lights given greater definition.

Lightening to the left, a little more broad light diffused over it so as to show the dark cloud’s edges.

Constable worked over the lightning bolt with some quite vigorous scratching. Lucas seem to have taken Constable’s intention to broaden the brightness of the track a little, and to increase its contrast against the dark cloud behind.

Noble shower dropping from the main cloud – connecting the west end with the main light on the horizon to the extreme right.

A little cloud up to the west end & over the roof, of the cathedral

The definition in the lower part of the sky is enhanced to carry the eye behind the bow at this level across towards the brightness at the right. Lucas also darkens the sky around the south-west tower and over the roof, which gives the cathedral a slightly less busy silhouette.

West end, & cathedral; &c greatly increased in splendour & the petite work must be destroyed to enrich it & increase the size

Light on tower subdued – and also the bright light on north transept.

In the V&A proof [01e (i)] Constable goes over the west end, north side of the nave and lower part of the spire with vigorous scratching out. He seems to mean that Lucas can afford to sacrifice some of the fine architectural detail that he had already made out in favour of the increased brilliance of effect. In the event Lucas manages to preserve the sense of the detail even though he much heightens the illumination. Lucas also subdues the light on the angle of the tower, whilst dramatically highlighting the lower angle between the roofs and the spire above. He also tones down the whole end of the north transept, especially the west finial, the north-west angle, and the gable.

All the lights on meadow and west end, trees &c &c increased to make the mass. (Dark patch softened, under west end.)

Lucas introduces a shower of new highlights on the meadow and water seen through the bushes below the west end, as well as a highlight streak below the principal building to the right of the cathedral which is additionally lightened. The top of the dark mass of shrubbery below the north-west tower is softened at the top, which makes the transition gentler.

Horses bright off, by making the willows light and mass also. Wheels & waggon looked to, also whole of white horse cleaned & lighted & broader.

The highlights on the horses backs are toned down – most obviously in the middle horse – and the bushes immediately above the line of their backs are lightened and sprinkled with new highlights to stems and leaves. The spokes have their highlights gently emphasised and there is some careful attention to the sparkles in the water between the spokes of the large wheel. Constable’s idea with regard to brightening the white horse does not in the event seem to have been adopted. Its brilliance is instead heightened by darkening the tone of the water around it.

Water much alivened – by darkening the main lights & bringing the light near the post so as to make a point – & to graduate from stump well lighted at edges. Man & boat improved. Willow that grows on the rails is much improved by bringing it off & light on darkened water.

Lucas’s reworking of the surface of the water is one of the triumphs of the plate. The general darkening allows him to give much greater depth of effect and variety of tone than before, and to introduce the very beautiful detail of the concentric ripples radiating from the legs of the lead horse. The stump below the head of the second horse is lit down both sides, which gives a sense of parallax. It is hard perhaps to see that the man and boat have been improved significantly. There is a new stroke that seems to represent the starboard gunwale, but this is dropped in subsequent impressions. The willow growing over the rails to the lower right is given more substance and detail, and now contrasts more with the darkened water beyond it.

Light on posts, & rails, & making a good angle line – to assist by its diagonal, the level, or horizontal line of meadow, water &c &c. Water up to foot of dark posts attended to carefully.

Head of said post, cleared off, the light water sharply.

Plants and sedges attended to.

The Chicago state [01e (ii)] has a lot of new highlights around the posts and rails at the lower right, and the reworking of light and shade across the meadow levels it more. The water around the foot of the clutch of dark posts aligned to the end of the rainbow is given a greater variety of sparkling lights.

The ‘said post’ is presumably that below the rainbow. In any case the head of that immediately below the swallows gets a very subtle sprinkling of lights, and one one below it is brightened on the top and right side.

This perhaps requires more vigilance than most viewers would be prepared to give, but the highlights of plants below the white horse are distinctly darker in the next proof, to create in give more depth. A distinct new clump is added to the right of the dog, as well as a few flicks across it.

Dog enriched & a light on the water brought near to his head or beyond his nose.

The highlight on the dog’s back and flank is subdued a little, and the line of its left hind leg clarified. The most striking change in the next proof is to its tail. It is not clear that this was occasioned by anything on Constable’s marked proof, and it is clearly a mistake. It was immediately rectified. The highlight in the water below the waggon driver is lengthened towards the dog.

Famous alteration of plants on the left, particularly attended to – they consist of brambles running over high grass & mixed in with plantains.

It is not at all clear why the alteration of plants was ‘famous’. Possibly it had been the subject of some discussion between artist and engraver. Looking back through the various states, we can see that there was new work at the lower left between states 01d and 01d (i), gaining somewhat in beauty and frothiness of effect, but none between states 01d (i) and (ii). Looking between 01d (ii) and the V&A worked proof 01e (i), we can see that Lucas had already reworked this area to give a sense of long grass growing beneath the brambles and indeed extending beyond along the bank. Constable was here looking for further clarification of that, and we can see in the next state that Lucas did his best to comply. The alterations might have been ‘famous’ in that Lucas had been reluctant to meddle with an area that was already rather beautiful. Certainly we may wonder why Constable was so concerned to interfere. The precise botanical definition was obviously significant to him. Long grass and plantains would signify late summer, reinforced by the overgrowth of brambles. Constable obviously wanted to be specific about, and botanically appropriate to a particular time of year, which in this case would be the late summer. The most recent commentary on the painting by Amy Concannon and John Thornes has argued that Constable set the date at 25 August 1832, to mark the death of Archdeacon John Fisher.

Path with dog improved

This is a little difficult to construe, but the ground above the dog’s back is darkened and given a little more texture.

Much improved about old woman – the bank & post

There is careful new work on the figure of the old woman. A ribbon is added to her bonnet and her jacket lightened and given more definition. The silhouette of her skirt is sharpened by adding light to the water beyond. The most obvious changes are to the two posts on the bank; that to the right is much brightened and now becomes a pivotal dynamic in the composition.

Also on the water over her to the left – & about the old cow, her horns and nose.

The water around the woman is lightened with the effect, especially, of sharpening the silhouette of her bonnet. Any changes to the cow are more subtle. The light on its shoulder is taken down, and the lights beneath its belly and in the grasses around its head are eliminated. The changes to the cow are exceptionally fine; the effect rendered much more subtle. As slight as the alteration is, and as lost, probably, even to the engraver, it entirely encapsulates Constable’s quality as an artist.

Light on cottage gable half put out at least

There is no indication of this on the V&A proof, nor does any change appear to have been implemented

Church tower much bettered by the low flat tone

This is marked on the V&A proof, and Lucas duly noticeably lowered the tone.

Trees about “arse end” of waggon much put down & at foot of ash, also about the gardener’s “shitten house” – so as to clear the edge of the waggon well off.

Speaking of lowering the tone, Constable now introduces a distinctly Rabelaisian note into the conversation. The profile of the back of the waggon was previously rather confused by the trunk of the tree beyond it, so the tree is darkened to better define it. All the background along the top edge of the waggon is darkened to the same effect.

The identification of the small building as the ‘gardener’s shitten house’ is especially noteworthy. This could be worth an essay in its own right, but here it is probably best just to sketch in a few remarks. Amy Concannon takes the terracotta-roofed buildings in this part of the painting to be proof of Constable’s disregard for topographical accuracy. No such small building appears in the scale survey reproduced by Concannon showing the area in 1833

and neither roof appears in any of the related pencil sketches. But a small red-roofed building is a key detail in the initial oil sketch for the painting, and in the full scale oil study at the Guildhall, London is clarified in its final form.

It was obviously intended as a foil to the cathedral. Constable, however is still more specific as to its function. It is the gardener’s shitten house, and presumably the house beyond is that of the gardener. The house might well be identified with one of the buildings in the 1833 map, and presumably its occupancy and the holder of the office might be investigated.


A shitten house would no doubt have been a boon best sited away from the house, and probably not worth marking on any map. Amy Concannon complains that ‘a confused sense of perspective arises from the disparity of scale between the red-roofed buildings and the figure in the cart’, but this disappears when one accepts that the near building is just as small as it needs to be. The identification, however, does introduce a major eschatological theme into the painting. Constable’s deliberate coarseness in referring to the ‘arse end” of the waggon, and putting it in quotes to suggest that this was a part of a discourse already established as an aspect of the painting. It very much looks as if all this would stand very much more consideration. Suffice to leave it here by observing that the spire and the necessary point in opposite directions, and between them they encompass a broad spectrum of the human condition. It is a nice detail that Lucas gives the building a little window by the time of the next proof.

And also the head rail &man’s arm divided, to ensure that the main point water inriched & more lights & darks.

The first part of this relates to a slight sharpening of the shadow defining the arm in relation to the front rail of the cart. Constable seem to imply that this is in some way related to the enriching of the water surface, but that is perhaps just a slip in haste at the end of a long list.

But he wasn’t quite finished yet. He added a postscript:

 I took the proof out of the frame again – to add to the above remarks. Still more brilliancy on the willows over the horses and about the boat & man – both of which are more distinct – as well as the team horses, & light on collars. The waggon tail more sharp off. The wheels better drawn and clearer – more light on water – perhaps too much on right hand of post. A few lights on the upper branches of the large ash tree.

The alterations to the tree went rather further than ’a few lights’. The whole canopy behind the knot of boughs to the left is darkened, right the way up to the top edge. Similarly to the right the large canopy is considerably darkened and much more depth and contrast put into the background foliage all the way down to the ground. The ash reaches perhaps the peak of its affective dramatization at this stage.

Lucas painstakingly worked over the plate once more to implement Constable’s requirements. He might certainly be forgiven if by now he had begun to wonder whether his taskmaster would ever be satisfied.

Next: Conclusion of states before title – state 01f

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s