CHAPTER 9

PICTURES AND PHOTOGRAPHS

How strange that among all this painting of delicate detail there is not a true one of English spring!

JOHN RUSKIN ON THE PRE-RAPHAELITES, c.1860

Here I am concerned with the humble practicalities of what paintings, drawings and photographs (rarely sculpture) have to say about trees, woods and the landscape themselves. Constable’s paintings show that trees and other vegetation have dramatically increased in almost every one of his scenes on the Suffolk–Essex border in the 200 years since his time. The scattered pines in Cézanne’s pictures of southern France have, over a century, turned into vast, fire-promoting extents of pinewood.

Art has another significance. Much of what is written on landscape history is really the history of what literate people wrote about landscape and what people are supposed to have thought about landscape. Paintings reveal another aspect of the relation between people’s attitudes to landscape and the actual landscape that they were attitudinising about. Writers on trees and woods have waxed eloquent on how pictures reflect the artist’s supposed ideology or prejudice: Gainsborough has been rebuked for not being a socialist before his time. However, this kind of writing is seldom based on anything that the artist actually said, and often reveals more about the writer than the artist or the landscape.

WORKS OF ART

Landscape pictures may be topographical, of a known place; or generic (A Wooded Landscape); or forming the background to a portrait, battle, martyrdom etc. There are landscapes designed as works of art; maps as works of art; portraits of trees; and botanical illustration as a branch of art.

The earliest European landscape paintings are frescoes, some 3,500 years old, excavated on the Ægean island of Santoríni. They appear to depict the cavernous red, yellow and black cliffs unique to that sea-volcano. One shows Phœnix theophrasti, the Cretan palm, now one of the world’s rarest trees. However, many of the plants are stylised to the point at which, for example, commentators argue whether sea-daffodil (Pancratium) or papyrus is meant.¹

In 336 BC the warrior-king Philip of Macedon was buried in the great barrow of Vérgina. On the façade of his tomb is a wall painting, still just recognisable, which shows the king and his courtiers in the field, pig-sticking and spearing other fierce beasts.² A luckless lion fights a losing battle against the combined assaults of Philip and Alexander the Great. Probably the king wished to be remembered by an actual deed of valour: there were still wild lions in north Greece. As in many pictures, the interesting part is the background: a grassy savanna with rock outcrops and scattered trees, apparently beeches, either in clumps or as massive individuals. There are still savannas with coppice-like clumps of massive beeches in remote parts of Macedonia today. Most of the trees are shown as dead. Savanna-like landscapes and dead trees were to be an artistic tradition for over 2,000 years.

Artists add or omit trees for reasons of taste (or to save time?), especially if they are working up studio pictures from field sketches. If, in the studio versions of The Hay Wain, Constable left out one bay of the ancient house now called Willy Lott’s Cottage, he could easily have omitted a tree or two. But this hardly accounts for his views, like The White Horse, that now cannot be painted at all because trees have grown up and hidden them. Nor is it likely with pictures painted specifically to depict recognisable places, as in Humphry Repton’s Red Books.

Topographical views

Pictures of cities or country mansions begin around 1500, for example Hoefnagel’s views of Norwich and of Nonsuch Palace (Surrey) in c.1580. They are useful for hedgerow and park trees, less so for woods. Probably they are more accurate than ‘artistic’ pictures, being commissioned for people who could check the details.

Oblique aerial views were invented probably in Venice in the late fifteenth century, and continued until just before it became possible to draw real aerial pictures from balloons. They can be works of surprising accuracy, intermediate between pictures and maps. Examples are the ‘map’ of north Dorset c.1570 and Kip’s view of Grimsthorpe Park c.1710, which shows the layout of that most important of designed parks already complete by then.

Can artists draw trees?

One of the greatest European landscape artists, Simon Bening or Benninck (c.1483–c.1565), worked mainly in what is now Belgium. John Hunter drew my attention to ‘November’ in a series of the Labours of the Months (Fig. 59). Bening shows two spearmen going out and slaying a wild swine and blowing a horn and letting the dogs eat it – as one did in November on the Continent. But the landscape is the point of the picture: a hilly land of dispersed settlement, hedges and small fields and scattered farmsteads. It could be Essex or Herefordshire (but not Cambridgeshire). In the foreground a coppice-wood shows three stages of felling and regrowth. There are recognisable oaks, aspens and elms; the coppice stools are small-leaved lime, with its floppy habit and heart-shaped leaves. Lime at one year’s growth still retains its leaves, but grown-up lime has already lost them (as indeed it does). Bening puts in coppicing plants, such as broom and brambles. All this – on a leaf of parchment smaller than this page – is not just a ‘A Wooded Landscape’, but an individual wood with a personality still recognisable now. One day someone may identify the spot.

Bening is exceptional. Go into a gallery, take a landscape painting at random, and ask ‘What is that tree?’ Surprisingly seldom can you give a definite answer. Representing trees is perhaps the most difficult task in art, and few artists succeed. No picture (or photograph) of a big tree can be naturalistic: life is too short to depict the complex reality. Any tree picture is a caricature. The art of caricature is to identify the distinctive features (the ‘jizz’ of a tree, as bird-watchers would say) and discard the non-distinctive ones. Most artists keep the non-distinctive features and get no further than the traditional Army classification into Fir-trees, Poplars and Bushy-topped trees.

Pictures of trees in leaf are more problematic than of leafless trees, and even Bening seldom made them recognisable (Fig. 60). Out of 41 book illustrations by him that show trees, I can recognise the trees in 16 out of 25 that depict leafless or dead trees, but in only one or two out of 16 that depict fully leafy trees.³

A Belgian contemporary (or maybe Bening himself?) sketched a summer woodland interior. Although his purpose was to record a woodbank, he shows coppice stools and timber trees, and nearly (but not quite) makes them identifiable to species.⁴ Another rare exception is the Louvre tapestry of the Hunt of Maximilian, c.1500 (Fig. 61). A luckless wild pig, as big as a horse, fights a losing battle against the mounted hero and his courtiers; his rear is assaulted by a rabble of poodles, rottweilers and a greyhound in armour. The December landscape is unmistakeably an opening in a coppiced alder-wood with tufts of holly, clonal patches of aspen, and a beech-tree on a knoll. It looks like a park that has been made by partly grubbing out a coppice-wood. I have never met this plant community, but would recognise it.

Maybe artists could portray trees if they wished, but thought it unimportant. One hardly expects El Greco or Turner or Picasso to get the trees right: that was not their job. For many others trees are mere fillers of unoccupied spaces. But an artist may take immense pains with the details, yet still fail to draw a convincing tree, especially in a studio painting. Even pictures specially commissioned for identifying trees do not always capture their jizz: for many years a tree-recognition poster hung in conservationists’ offices without anyone noticing that the trees on it were not recognisable.

I once tried to locate Gainsborough’s picture, today called Cornard Wood, in the actual woods around eighteenth-century Great Cornard (Suffolk) (Fig. 62). Taking the recognisable church in the picture as a point of reference, I could not reconcile Gainsborough’s woodland with any of the woods known in contemporary maps. The trees, though beautifully detailed, are barely identifiable. The scene lacks infrastructure: coppice stools, banks and ditches bordering the lane, etc. Is it a work of imagination bearing the sort of relation to real landscapes that Piranesi’s Carceri d’Invenzione did to real Italian jails? No: Susan Foister has shown that it is derived in detail from a landscape by Ruisdael.⁵ A picture that has appeared on the cover of a book about English woodland is thus of a foreign wood. Ruisdael himself (who often depicted recognisable trees, even in leaf) was part of a long tradition among Dutch and Flemish Old Masters: whether it was true to the landscapes of their countries I cannot say. But Gainsborough was no mere plagiarist: he made the scene bland and generic by leaving out the dead tops and eroded roots that gave personality to Ruisdael’s oaks. These were not to be long out of fashion: in Leicester Museum is a view of the Blean near Canterbury, by Thomas Sidney Cooper in 1832, which has two great oaks with the shattered tops and dead branch-ends so characteristic of oaks on the Blean.

John Constable sketched trees convincingly in the field. He loved elms, and on a good day could master the features that distinguish the many kinds of elm. In his studio pictures, however, trees lose much of their distinctiveness – except for Lombardy poplar, then just entering the English landscape with its very distinctive shape. A remarkable exception is his painting of Salisbury Cathedral for the Bishop (Fig. 63). The spire is framed between two magnificent leafy elms – but they are East Anglian elms, not English elm, the common species around Salisbury. Were they a real outlier of East Anglian elm? Or did Constable import them from an East Anglian sketchbook? His original sketch of 1811 shows some of the other trees in the picture, but not these.

Painters of the Barbizon School, in mid-nineteenth-century France, lived outside the Forêt de Fontainebleau, then (before the foresters got at it) a majestic and distinctive landscape, with its extraordinary sandstone rocks and multitude of ancient trees. They rarely appreciated it. Most of their paintings are unresponsive to the majesty and distinction, giving the impression that Barbizon could have been anywhere. Narcisse Diaz, indeed, went to the trouble of reworking Cornard Wood or a Flemish predecessor.

In Britain, the Glasgow Scots and the Pre-Raphaelites were contemporary schools to Barbizon. Horatio McCulloch appreciated the ancient and majestic trees of Cadzow Park, and spent many an hour portraying them.⁶ The mid-nineteenth-century Pre-Raphaelites lavished their skill on plants and trees, influenced by the popular craft of botanical illustration. John Everett Millais and Ford Madox Brown got every leaf and petal right, and every lenticel on oak marble-galls (then a new curiosity). John Samuel Raven could depict a sainfoin field with 19 species of weed. The Tate Gallery exhibition in 2004 brought together about 140 of their works.⁷ Trees are recognisable if drawn from close enough to show individual leaves. More distant trees are seldom identifiable, except for Lombardy poplars and Scots pines, a few beeches, and some elms (but the artists turned a blind eye to the Elm Disease then raging!). Great spreading hedgerow oaks never appear. Only one picture, John Brett’s The Hedger, is of a convincing wood, with stools of oak and ash (but the cataloguer called them ‘birch’!) and in the foreground a woodbank with a man plashing a hedge on it. Pre-Raphaelites painted in the field, usually in summer, and encountered the extra difficulty of depicting leafy trees.

Portraits of trees

This art form, widespread in Europe from the sixteenth century, was especially developed in England. Strutt published a book in 1822, Sylva Britannica, portraying 48 famous trees and groups.⁸ Those still extant include the Tortworth Chestnut, which had already been a famous tree for more than a century and has not greatly changed since Strutt’s time. In Savernake Forest (Wiltshire) I identified what he portrayed as the King Oak shortly before it collapsed in 1990. There was apparently another King Oak nearby that collapsed in the nineteenth century and of which no trace now remains; thus are the identities of even famous trees sometimes confused.⁹ Strutt’s, like most veteran trees, are freestanding or in wood-pasture, not in woodland. Many are pollards. He measured many of them: the giant trees of 200 years ago were not very different in size from giant trees today. Chance once led me to lecture in the rooms of the Royal Agricultural Society in London, beneath a portrait of King George V as Patron. The king was shown under some of his ancient oaks, symbols of majesty, in Windsor Great Park. I happened to illustrate my lecture with those same oaks, this time as relicts of Windsor Forest and home of its special insects.

Artistic preferences

Artists choose some features of landscape and shun others. They love dead trees (because dead trees are easier to draw?). Dead elms in paintings are part of the evidence for early Dutch Elm Disease. They love badlands, landscapes of erosion gullies that are common in Italy but adopted by French artists such as Fragonard and even by Flemings, who would have seen badlands only on their travels. Many love pollards, which occur from the medieval Très Riches Heures du Duc de Berry through Dürer, Rembrandt, even Gainsborough (though rarely), and on to Arthur Rackham and later.

Artists mysteriously avoid certain features known to have been common, such as open-field strip-cultivation. In Italy they shun cultivation terraces, preferring to depict sloping hedged fields. (Was there a textbook which said ‘Never paint terraces’?) Except for a few Belgians and Pre-Raphaelites, they avert their eyes from coppice stools and trees with more than one stem. Before Dürer they seldom show recognisable mountains: Europe was slow to appreciate the spiritual qualities of mountains. They rarely get timber-framed buildings right, even though timber-framing was itself an art form.

Landscape imitating art

People remake the real world in the image of art. Park landscapes of the eighteenth century (p.118f) are supposed to imitate the works of Claude Lorrain and Poussin. Landscape designers cited these Old Masters, whose pictures they might have seen on their patrons’ walls, but the reality is more complex. Parks were rarely set up on a tabula rasa without pre-existing features. The designers ‘Capability’ Brown, Humphry Repton and their colleagues worked round and incorporated existing woods, hedges and ancient trees. Repton’s famous Red Books included ‘before and after’ pictures for his clients’ benefit.

Baroque artists such as Poussin (1594–1665) established the belief in the ‘ruined landscape’ of Southern Europe. They depicted the nymphs and heroes of ancient Greece disporting in noble forests and crystal fountains; the scenery was typical of the lush badlands of middle Italy, which they knew. Travellers who, a century later, reached the real, dry, ‘barren’ Greece concluded that the landscape had gone to the bad since Classical times by human mismanagement. They could not know that ancient Italy was rather like Poussin’s Italy, but very unlike ancient Greece and ancient Spain. This deep-seated misunderstanding contributed to the modern belief in desertification. (The Spanish Forestry Commission has striven, so far with little success, to remake Spain in the image of Italy.)¹⁰

Artistic cultures independent of Europe have different priorities and values. The Chinese and Japanese have for centuries understood the personalities of trees. Not for them the bland, straight-grown, bushy-topped: their space-fillers are golden clouds, not trees. They love dead boughs and eroded roots. A few flicks of a tiny brush define the jizz of a pine and its difference from other species of pine. European artists in Australia set about learning the individualities of a thousand species of eucalyptus, and used them to define the different landscapes of the new continent. (Some of them were clanking around Australia for forgery, unfettered by European artistic tradition.) Euro-Australian tree art is still very much alive, for example, in the works of John Duncan Firth.

PHOTOGRAPHS

Photography began as an art form, concerned with people, buildings and gardens. Most of the many volumes compiled on the lines of Barsetshire in Early Photographs are disappointing for woodland. However, the tradition of portraits of great trees continued: Menzies’ book on Windsor Great Park in 1864 has photographic prints of ancient trees pasted into the text that can be compared with the same trees now.¹¹

Many professional photographers, like artists, do not appreciate what makes a tree distinctive, and have an unerring nose for the commonplace. A standard composition is of a big, young tree against the sky, telling nothing. I once contributed to an article on Grimsthorpe Park, that amazing landscape of picturesque trees which, surely, even the least appreciative cameraman could hardly miss. The professional photographer that we sent – alas, without detailed instructions – ignored the wonderful ancient trees and sought out some horsechestnuts that could have been anywhere. (An exception is Thomas Pakenham, whose Meetings with Remarkable Trees and Remarkable Trees of the World are a superb continuation of the tradition of tree portraits.)

Woodland photography can be difficult. As with paintings, it is impossible to record all the complexity of a big tree. Branches against the sky tend to disappear. Matters are worse in summer, when the light is weak and green, and especially on a sunny day when sunflecks create awkward contrasts. (Will electronic photography make it possible to correct for these hazards?)

Woodland crafts were illustrated in Edlin’s Woodland Crafts of Britain, with its wealth of pictures taken in the 1930s or earlier. Woods themselves began to be illustrated in the 1920s: the photographs published in the Journal of Ecology and in Tansley’s books reveal what English woods looked like before deer attacked them, or the boundary between chalk grassland and beech- or yew-wood before the woods began to spread in the 1950s.

Better than any early woodland photography in Britain is the magnificent record of Californian redwood forests by A.W. Ericson and others.¹² The task of reducing the world’s grandest trees to roof-shingles and railway sleepers was a world-celebrated triumph of technology. In recording it, these photographs explain the initiation of what are now the world’s grandest coppice-woods.

As well as recording woods and trees that have disappeared, photographs document stability and change within existing woods and trees. They show that dead branches in oaks can persist for well over 50 years and are not necessarily the result of recent pollution. For this purpose it is important to relocate the original viewpoint. Fixed-point photography is dealt with in Chapter 21.

Hemispherical photographs

The fish-eye lens, invented by Robin Hill the biochemist,¹³ takes a picture of approximately an entire hemisphere projected on to a flat surface. Originally intended for studying clouds, it was applied by Clifford Evans, David Coombe and Margaret Anderson to the measurement of shade.¹⁴

Shade is measured by taking a photograph of the canopy and overlaying it with one of three grids, dealing with the three sources of light: direct sunlight, light diffused from blue sky, and light reflected from clouds. The first grid gives the sun’s position at different times of day and of year, from which one can record how much of its track is obscured by leaves or branches. The second and third grids divide the hemisphere into sections each giving the same amount of light from a standard blue or overcast sky, on which one counts how many sections are clear or obscured. Two photographs, one with the trees in leaf, the other leafless, measure shade with a precision that could otherwise be got only by round-the-clock light measurements lasting a whole year. This method avoids the great uncertainties that arise from using photoelectric cells, which respond very unevenly to light of different spectral compositions, or coming from different directions in different weather and at different times of day.

Hemispherical photographs are a simple way of recording changes in woodland canopies, for example round a coppice cycle or following disuse of coppicing (Figs 195, 196). The camera needs to be returned to a fixed point year after year, preferably to a permanent cradle fixed in the ground.¹⁵

Fish-eye lenses have been made for various models of camera. It is convenient to mount the camera, facing upwards, on a board with arrows at the ends that can be levelled and oriented north and south. These arrows will appear on the photograph and will indicate where the horizon lies as well as the orientation.¹⁶

Aerial photographs¹⁷

Photographers could have taken to the air almost as soon as photography was invented, but they seldom went up in balloons, and air photography became a source for woodland only in the 1920s.

In late summer 1940, during the Battle of Britain, the German Luftwaffe flew what amounted to a great survey of England, mainly the east and southeast, but sometimes extending even into Wales. The excellent original prints (and those of many other countries) are now in the Cartographic Division, United States Archives in Washington.¹⁸ They complement the Royal Air Force’s wartime photography of Germany, Austria, Italy and other countries in Europe and beyond, now in the Department of Geography, University of Keele.

Wartime aerial photography is the best record of woods just before the Locust Years (Fig. 197). The German photographs (a triumph of both valour and technology) record the last phases of coppicing in many woods. The replanting movement, however, had barely started. The usurpation of woods by military airfields mostly came after 1940 and was not recorded by the Luftwaffe, but sometimes was by the raf.

Air photographs are an excellent record of the structure of woods. An early example, taken in 1930 in the beechwoods of the Oxfordshire Chilterns, shows the effects of selective logging, drought and fungal attack, as well as distinguishing woods from plantations and picking up abandoned pasture invaded by trees.¹⁹

Black-and-white photographs seldom identify tree species, apart from the circular patches produced by elms and other clonal trees. Colour photographs, which begin about 1960, are better (especially in autumn); infrared false colour are better still. Earthworks are normally visible only in felled areas. Crop-marks and soil-marks, natural or artificial, such as the ‘stripes and polygons’ produced by freezing and thawing in glacial times, or those due to Iron Age field-systems and Roman villas, can normally be identified only in grubbed-out woodland, but can sometimes be observed extending into a wood from adjacent fields.

Distinguishing woods from plantations should be easy because most plantations are in rows. In practice it is less simple. As a plantation gets older or is neglected, its canopy flattens and the tree rows become indistinct. Worse, photographs of natural woods, with their uneven canopies, give the illusion of rows that are not there. Especially on slightly oblique photographs taken in sunshine, the parallel shadows and the apparent tilt of the taller trees seem to line up into rows. Having looked in vain on the ground for such rows as seen in aerial photographs of ancient woods, and having seen apparent rows in an air photograph of New Guinea rainforest (and when overlooking a wood from a mountain), I distrust such evidence without confirmation.

Thermal infrared images, which indicate the surface temperature of objects, measure the microclimate in woods, an important influence on both trees and woodland plants and animals. They reveal, for instance, the warmer temperatures of recently coppiced areas, or the difference in temperature between the two sides of a woodbank.²⁰ (A hand-held apparatus is used on the ground to detect deer.)

Air photographs and satellite images are used to measure ecological change and deforestation. This is more difficult than it looks. ‘Deforestation’ itself is hard to define in many tropical countries, where forests do not have sharp edges, farmland can be full of trees, and changes are more complex than acres of forest turning into acres of non-forest. What is to be done statistically with forests that have been depleted or burnt but still remain forest?