This is Chapter Seven of E.H. Gombrich's "Preference for the Primitive"
Quoted text is in YELLOW. Text quoted from other authors is in GREEN *************************************************************************** ****************************************************************************
In 1930 Herbert Read wrote that
‘from the study of negro art and the Bushman we are led to an understanding of art in its most elementary form and the elementary is always the most vital’
He clearly shared the opinion discussed in the preceding chapter, that these artists and their kinsmen all over the world were unspoilt children of nature and that their creations could teach us what art meant to mankind before the academies had put it in straitjackets.
I can't get from "most elementary form" to "unspoilt children" as easily as Gombrich did - but it does make sense if we look at Read's previous sentence:
Indeed it has been said with truth by Count Gobineau the Negro possesses in the highest degree that sensual faculty which our civilised instincts tends to destroy, but without which no art is possible
Possibly Gombrich was wary of quoting such a notorious racist.
I believe that, among students of art, the great anthropologist Franz Boas was the first to repudiate this picture of primitive man, in his classic volume on Primitive Art, first published in 1927. In the preface to that great book, he defines his position from the outset. He concedes that
‘there must have been a time when man’s mental equipment was different from what it is no when it was evolving from a condition similar to that found among the higher apes.’ But ‘that period’, he emphasizes, ‘lies far behind us and no trace of a lower mental organization is found in any of the extant races of man ... Some theorists assume a mental equipment of primitive man distinct from that of civilized man. I have never seen a person in primitive life to whom this theory would apply ... The behaviour of everybody, no matter to what culture he may belong ... can be understood only as an historical growth .‘
Thus Boas dismisses any attempt to arrange all manifold cultural lines in an ‘ascending scale’ in which to each can be assigned its proper place.
By ‘ascending scale’, Franz Boas was, of course, thinking of the idea of progress that had also dominated the history of art since classical antiquity the idea that artists got better and better in the imitation of nature, and that one could therefore assign a date to an image simply by measuring how close it had come to its model in reality.
On the contrary -- the Boas quote appears to refer to the "mental equipment" that distinguishes humans from apes, rather than one culture from another (or one race from another, as Gobineau had once asserted)
It is worth noting that Franz Boas was reluctant to accept this interpretation. He did not believe that the images of tribal art he had studied with such care could in any way be described as the product of a limited skill. Having analysed the immensely complex system of representation practiced by the Indian tribes of the North West Pacific, he remarks that ‘when the artist desires realistic truth he is quite capable of attaining it’, and he instances a carved head made by the Kwakiutl Indians of Vancouver Island ‘which is used in ceremony and intended to deceive the spectators who are made to believe that it is the head of a decapitated dancer’.
I've had a similar thought in response to naturalistic drawings of animals on cave walls that date back 20,000 years.
It is indeed an important example which should make those of us pause who have spoken or even written about the evolution of representational skill, for it reminds us of the dependence of such skills on the function the image is intended to meet. Only when illusion is demanded are artists hkely to acquire the skill to create
But was it wholly legitimate for Boas to generalize on the example he used, and to conclude that the skills of mimesis were always at hand for the asking? It is a question that naturally concerns me, because in Art and Illusion I argued the opposite, claiming that even given the will, mimetic skill was never developed overnight, but tended to progress in the course of several generations.
But the skill I was speaking of was that of drawing or painting on a flat surface; I should have remembered that the medium of sculpture is different.
Of course, the fact that a naturalistic drawing was made 20,000 years ago does not mean that it did not take several generations, if not several centuries, to achieve such a skill.
The same could be said regarding non-naturalistic images -- and actually, I would guess that highly stylized shapes - like the characters in Chinese calligraphy - are more likely to require many generations to achieve perceived virtuosity.
This chapter is dedicated to the proposition that "primitive art" does not need scare quotes any more than "primitive technology" needs them.
There is no question that an abacus is more primitive than a computer, so why can't we accept that one-point perspective is a technical advance in the representation of three-dimensional space onto two dimensions, and attempts made without it are more primitive ?
I find it incredible that Gombrich, or anyone else familiar with art history, would not suspect that a variety of intentions might explain the variety of pictorial representation.
But I find his following rumination even more incredible:
What we have learnt from the psychology of perception is that, if asked, we would all underrate the degree to which they appear to diminish. Trace the distant one and transfer it to the front and you will be surprised by the difference in scale.We are surprised because the objects in our proximity appear to be relatively constant in size regardless of their distance. These so—called ‘perceptual constancies’ which influence the way we see our environment are in conflict with the optical laws of perspective which govern the images created by mimetic art, no less than those recorded by the camera. At this point we must proceed with some caution to avoid the trap of misjudging the function and achievement of optical perspective as untrue to our visual perception. We are free to say that we do not see a tiled floor as it must be drawn in perspective, but it remains also true that if it is not so drawn it fails to look like a level floor paved with tiles of identical size.
To come to the point at last: what the mimetic artist must learn is not to concentrate on how he sees the world, but to ignore his visual experience in favour of a geometrical construction based on the laws of optics which tell us that light rays are propagated in straight lines and will result in an orthogonal projection on our retinas.
Only if an artist applies these laws will his image convey the correct information to any viewer.
As Norris Kelly Smith pointed out in his discussion of Brunelleschi's lost tavolette , one does not need Albertian geometry to draw things in accurate perspective -- one only needs to observe spatial relationships as they actually appear. And I do this all the time when drawing models at the art club. There is certainly a discipline to visually measuring relative sizes and locations -- but it does not require the applied knowledge of optical geometry - any more than a sparrow needs to know aerodynamics in order to fly.
Only if an artist applies these laws will his image convey the correct information to any viewer. It is easy to test this assertion, because applying the same laws we can retranslate the flat, perspective image into the 3-D motifs the artist had in front of him. In the case of Vermeer's painting this experiment has actually been made
Similarly, this experiment claims to prove that Vermeer used a lens when making this painting.
But both of these experiments only suggest that Vermeer was painting something like what he saw -- and the photographic reconstructions are jumbled messes compared with the delightful stillness and harmony of even a reproduction of Vermeer's painting.
To complicate matters further it turns out that for the painter, perspective and proportion are inextricably linked. We all know of photographs in which a person’s hands or feet look monstrously distorted because the camera has recorded their perspectival appearance. A moment’s reflection will show that this influence of perspective will diminish, but not disappear, at a greater distance. It was precisely Vermeer who took account of this effect in his painting of The Astronomer at the Louvre, where the left hand, closer to the observer, is represented considerably larger than the right hand. For figures as far removed as they are in The Music Lesson, the effect may not be noticeable, but theoretically it remains true that, when projected onto a plane, proportions must fluctuate with the perspectival setting.
On the contrary - it appears to me that the astronomer's left hand is not large enough, considering that it is so much closer to the viewer -- and judging by the sharp angle of convergence in the presumably parallel lines in the back of the chair behind him, the viewer is less than 10 feet from that left hand.
Same thing with these two hands in Caravaggio's "Supper at Emmaus" -- the right hand is roughly the same size as the left, but since it is much further away from the viewer, it appears to be too large. It has the effect of disrupting a single P.O.V. and pulling the viewer into the center of the action -- which is not necessarily a bad thing.
It is well known that these unexpected deviations from our visual experience can only be registered by the artist through the device of measuring — that is, by comparing the apparent shape of a distant object with one close by, either by looking through a grid or screen close to the observation point, or at least by using the handle of a brush held in the outstretched hand. True that mimetic artists can train themselves to do without such artificial devices, but the fact is that mimetic rendering is based on an art of construction — we may even call it of reconstruction — that by-passes the perceptual experience.
Yes -- no measuring devices and no geometric theory is required -- accurate relationships can be established by a self-trained eye.
Once we understand the true complexity of this achievement we shall cease to be surprised that it was the fruit of gradual developments. We can no more expect every human being to be born with this capacity than we can expect everybody to be able to construct and use a photographic camera.
But we might expect some people to develop this ability on their own without special training, without theory, without any special equipment. . It's an eye/hand/mind discipline, like accurately throwing a rock -- so it can arise independently in many times and places.
Though it is true that Western Europe seems to have the only culture whose artists would consistently show size proportionally diminishing with distance throughout the visual field.
Albrecht Durer, 'St John Devouring the Book', 1498
Russian adaptation, 19th C.
Knowing that an 'ascending' order in the method of representation does not play well among either anthropologists or art historians, Gombrich goes to some length to argue his case -- comparing the prose of a great novelist (Henry James) with the pidgin English of an immigrant, and comparing Baroque paintings with copies made by children.
More interesting to me is the above comparison of a Russian woodcut with the Durer on which it was loosely based --- because the Russian artist, though unknown, could not be considered either a child or an inexperienced visual artist.
Fraenger has shown in a masterly analysis to what extent the copyist has eliminated the indications of the third dimension, how he has brought the distant ship forward, and how this transformation results in a flat but satisfying design in which the whole page is filled with shapes disregarding natural proportions. Mark that the Russian woodcut is not really two-dimensional, much less a child’s drawing. It only follows the gravitational pull in that field of force I am trying to describe.
This is the kind of fantastic/visionary subject matter that seems, at least to me, least suitable for a realistically optical pictorial space.
I can see why a pious viewer might want Durer's treatment for a scene from the life of Christ -- where the a single point-of-view places the viewer's feet on the ground near Jesus.
But scenes from Revelation are highly charged, prophetic visions . Wouldn't realistic details within a realistic space feel out of place?
Henri Rousseau, Child with Puppet, 1903
It is in this reduction of depth, this filling of the surface regardless of proportion which we also know from portraits of the folk tradition. It is not uninstructive here to compare these woodcuts with the work of a late nineteenth-century primitive, Henri Rousseau — ‘Le Douanier’ — whom, as you know, Picasso helped to discover. What they share is precisely the yielding to the pull away from the third dimension, the absence of convincing modelling in favour of constant tones, and the absence of complex aspects, though it may be argued that Rousseau’s superior talent shows in the poetic qualities of his work which transcend the more primitive mode.
Could those specific "poetic qualities" be achieved within a deeper pictorial space?
Tombstone of Rev. Grindall Rawson, 1744
(as pictured in Gombrich's book)
Tombstone of Rev. Grindall Rawson
(as found on the internet)
Once we have established this uniformity or law in the transformation of mimetic images, nothing prevents us from speaking of a structural kinship between various images which we can call less sophisticated or more primitive. Belknap has collected many examples of American portraits of the early nineteenth century which are in fact adaptations of English mezzotints, where the story repeats itself. It also does this on a New England tombstone, where the problem of defining the aspect of the eye in half profile landed our sculptor in trouble.
The New England sculptor apparently had less trouble when his work is lit from a different angle. It looks like Gombrich has playfully turned the tables on those enthusiasts who use photography to make artifacts look good.
Donatello, Cristo Morto, 1453
Donatello, Compianto, 1455
Other than in profile, it's hard to design a human head into low relief -- so the Egyptians and Assyrians wisely avoided doing that. But here's a few more good examples.
But knowledge of the most interesting transformation in American folk art I owe to Professor Nelson of the University of Minnesota. He investigated an altar carved by a Norwegian immigrant at the beginning of the twentieth century which is a masterpiece in its own right.Yet it turns out that Christenson, the craftsman, did not invent the biblical scenes represented. He took them from a Bible illustrated by the once fashionable French master Gustave Doré, whose dramatic and theatrical plates he again reduced in the predictable direction, into regularly patterned, rather flat relief.
I never would have guessed that these two pieces were related in any way other than subject matter. Even if Dore's print was the only image of the crucifixion that Christenson had ever seen - I still would not say that he drew anything from it. He expresses rustic Christian life in Norway and the northern Great Plains - Dore expresses the Romantic era in Paris. (BTW -- shouldn't the figure of Christ be smaller relative to the Magdalen so far in front of him? Dore has given Him the proportions of a giant )
I find this Dore version of the Crucifixion far more satisfying - though it's probably not as useful as an object of simple-minded veneration that might hang on the walls of a parochial grade school.
Arch of Constantine
I am convinced that this common fate of three-dimensional works in the hands of primitive craftsmen has a bearing on one of the most hotly debated issues in the history of Western art: the transition from antiquity to the Middle Ages. The reduction of the individual figures to types — not to say stereotypes — arranged in serried ranks, the filling of the whole area and the lack of articulation in Christenson’s relief bear a more than superficial resemblance to that much-discussed monument of late antiquity the Arch of Constantine, from the early fourth century of our era — or more precisely, to the reliefs which date from the time of Constantine.
For this hastily erected triumphal arch also shows reliefs of earlier periods put to novel use in this victory monument. The contrast between these earlier, naturalistic relief and the squat, immobile figures of the later ones could not, and did not, escape early observers, and thus the Arch of Constantine became the prime witness from the days of the Renaissance on to what was considered the decline of ancient art and the coming of medieval barbarism.
It was only this century which challenged this interpretation of the reliefs as symptoms of decline. Were they not rather symptoms of a deliberate abandonment of classical standards and the expression of a preference for the primitive, comparable to the works of twentieth-century masters? I have come to think that the question has been wrongly put, and that the parallel between the art of the twentieth century and that of the fourth century AD — as far as it exists — does not justify us in celebrating the Arch of Constantine as a precursor of modern primitivism. If the craftsmen of the period followed the natural pull towards the two—dimensional base line they were neither to blame nor to praise. They had no use for flying, and so their art stays closer to the ground. The social and historical reasons which led to this shift of function and of style, also exemplified in Roman provincial tombstones cannot concern us here. All I want to emphasize is the uniformity of structure between their art and other two-dimensional products in folk art of the kind we have briefly reviewed.
Though I last saw it 40 years ago, I can still remember my dismay at the contrast between the delightful, spacious monumentality of the older sculpture in the rondels - as opposed to the stiff and clunky frieze that was set beneath it in Constantine's day. What a depressing transformation to mark the birth of Christianity as a European state religion.
The contrast here is not quite so painful
Here's an area that seems engaging - but I wonder how it appears in situ.
Here's a frieze from Ghandara from roughly the same era - it['s a bit more lyrical.
Arch of Constantine, detail
Here is one of the fragments taken from the 2nd century Trajanic era (what Gombrich identifies as a "Roman provincial tombstone")
Gombrich categorizes the sculpture on the arch as either Classical or Primitive -- but I think it would be more helpful to observe the greater variety of intentions and abilities, that run from desperation to virtuosity.
Gombrich doesn't come out and say that the Constantine-era sculptors were incompetent bunglers -- but it's rather disparaging to say that they could not resist "the natural pull toward a two-dimensional base line".
Isn't resistance to "natural pulls" the base line for human social life, whether in small tribes or great cities ?
Do the conflicting styles on the Arch of Constantine mark the transition from Classical to Medieval art ?
Do pieces like the above demonstrate a deliberate abandonment of classical styles, just as artists did in the early 20th Century ?
Or does the arch just demonstrate what happens when chisels are given to untrained incompetents? That would seem to be where Gombrich would stand --butd he backs off and allows that these artists "had no need of flying, so they stayed on the ground"
Given the variety of work on this monument - I would hazard to select "all of the above" - while admitting that I do not share an enthusiasm for a sharply defined, historical sequence of styles that is the legacy of the German academic tradition of art history.
Gingerbread moulds, "The Three Holy Virgins", Austrian Museum of Folk Art, Vienna
Hercules with deities, Dura-Europos, Palmyra
All I want to emphasize is the uniformity of structure between their art and other two-dimensional products in folk art of the kind we have briefly reviewed. It was that perceptive student of late antique art, Ernst von Garger, who once observed in this context that ‘gingerbread figures should not be regarded as great works of art, even when they happen to be of stone.’A comparison between a sculpture from Dura Europos and real moulds for gingerbread shows that he was not exaggerating
Hah! Hats off to Gombrich for coming up with a gingerbread mould that so closely resembles a sculptural program.-- though how did the notion of "great art" enter this discussion ? I'm not sure, but this may be the first time it has appeared in this book.
And I fail to find much "uniformity of structure" between them.
Lindisfarne Gospels: Matthew, 8th C.
Codex Amiatinus, Ezra, 6th C.
I am convinced that one can be an admirer of late antique and medieval art without losing sight of the fact that the skill of high flying was no longer needed or practiced in that period. Not that this tendency towards reduction of naturalistic images implies a lack of skill in other directions. There is no more astounding a monument to the skill of hand and mind than the marvelous pages of the Lindisfarne Gospels of the late eighth century of our era. The sureness of hand and eye to which these pages bear witness needs no emphasis. And yet, when the artist of that gospel book copied a naturalistic model ;he followed the same pull we have now observed so often. The famous image of the Prophet Ezra in the sixth-century Codex Amiatinus , or another very much like it, served as a model for his Evangelist. Once more the psychological law of reduction can be seen to have been in force: the rounded, plausible figure of the scribe has been flattened out as if it were composed of strips of paper, the head has been reduced to a formula. All indications of the setting have been dropped.
Their sense of space/design certainly feels different, but I'm puzzled how Ezra in the Codex Amiatinus might serve as a "naturalistic model" any more than Matthew in the Lindisfarne Gospels.
And if either exemplifies "high flying" -- Matthew seems to be floating in heaven, while Ezra's chair is firmly resting on the ground.
I'd also like to note that "skill of hand and mind" also distinguishes one good baseball player from another - while the Lindisfarne Matthew might be distinguished by qualities of design, spirit, and lyricism - especially from the Amiatinus Ezra. But those are qualities that Gombrich never mentions regarding anything.
Reiner Van Der Huy, brass font, St. Bartholomew's, 1107-18
Font, Castle Frome parish church, 1170
Admittedly the style or styles of what we call ‘medieval art’ extended over more than five hundred years and over the whole of Europe. Generalizations about such a range of monuments are therefore bound to be superficial, if not misleading. And yet we can say that the action of gravitational pull, the problem of reduction from three—dimensional to two—dimensional, can never be left out of account when examining and assessing these monuments. For even within medieval art there is a gamut of styles from a classical, individualizing, eye—witness conception to a two-dimensional manner — witness two fonts decorated with reliefs.
The one from Liege, by Reiner van der Huy, of the early twelfth century, shows the Baptism of Christ in a highly articulate classical manner. The other font, from a small English church, shows all the marks of reduction I have discussed. Surely we have a right here to call the one more primitive than the other.
Gombrich has combined "highly articulate" with "classical manner", and I'm not sure whether one or both are required by him to declare that the St. Bartholomew's font is less primitive than the one from Castle Frome.
Judging from the photos, both are interesting, but neither one would get me to travel very far to see it - while I'd spend a day in London to pour over the Lindisfarne Gospels.
Schwäbisch-Gmünd, St. Johannis-Kirche, Madonna and Child, 12th Century
Tombstone of Francesco Gugale, C. 1800
This kinship of structure between various medieval styles and other monuments from other periods confirms the need to pay attention to these modes of reduction. The Romanesque relief from Gmflnd in Germany does not look all that distant from a sculptured Pieta , but the latter was carved around 1800 by an Italian peasant — it is a piece of folk art.
Unfortunately, I couldn't get a good image of that tombstone. From what I've got, it seems more concerned with the rhythm of angular edges than balanced masses. Does Gombrich suggest that the peasant who carved it might also have been able to produce something like that Romanesque Madonna? I find that hard to believe.
Or take this Maori statuette, a work of missionary art showing the Virgin Mary entangled in ornamentation, Is it all that different from the sacred figures of the Book of Kells? . Both are products of traditions which mastered the art of complex patterns but were new to the task of rendering the human figure.
A fascinating pair of examples from opposite ends of the planet -- and it does seem that the human figures have some difficulty emerging from the strong, sinuous linear patterns on their surfaces.
But is that because the artists were "new to the task of rendering the human figure" --- or --- because that's how people in those times and places felt that the human figure should be rendered ?
Given the limited autonomy that people may have experienced living in small groups in hostile environments over the first million years of humanity -- it would seem appropriate to visualize people as tightly locked into a network of patterns. Which would locate the primitivism in society rather than just its art.
Ivory plaque, Byzantine, 6th C.
Leonardo Da Vinci, 1481
I hope this series of examples, however cursory, has tended to confirm my claim of what might be called the ‘law of gravitation’ in image-making. Given the fact that there are very few psychological laws of any validity I think we must regard it with some interest. It justifies, does it not, speaking of certain structural features in images which we are entitled to describe as ‘primitive’? The reader will not find it hard to recognize these features in the late antique ivory plaque representing the Holy Virgin enthroned between the three adoring Magi and an angel, all arranged in a symmetrical pattern . Obviously the version of the same theme by Leonardo daVinci is further away from that base.
With this example, Gombrich adds yet another feature to his notion of 'primitive' which now includes:
1 Reduction from three dimensional space to two
2. Shapes taken from rounded to flattened
3. backgrounds taken from complex to simple
4. presenting types instead of individuals
5. figures tangled in ornamentation
None of which can found in Leonardo's "Adoration" which presumably epitomizes the advanced technology of image making -- as if its specific pictorial qualities needed no explanation beyond the same drive for technological improvement that replaced the vellum scroll with the printed book.
. That marvellous panel is unfinished, and therefore not easily legible, but when you stand in front of that work in the Uffizi in Florence you soon come to recognize that, in my terminology, the composition is not only three-dimensional in space, but extends the psychological and emotional depth through the way Leonardo explores and renders the relationship and the reactions of each of the participants in the miraculous event. The master’s many studies for this composition convince me of the choices he pondered and of the efforts he made to evoke the sacred story in all its mystery, showing the Kings humbling themselves before the Christ Child, and the Virgin smiling her gracious smile.
Here is the Filippino Lippi panel that was commissioned to replace the one that Leonardo never finished.
Why did Gombrich prefer to use Leonardo's version as an example ? It's a drawing that was painted, years later, by other hands -- so it's rather useless if you're concerned with the artist's original intentions.
As it appears now, it's a mishmash - with the architectural and narrative space in the background apparently done just to show that he could do them.
Maybe that's the point -- pictorialism for it's own sake.
The Expressive Gamut
I know there are plenty of people who deny that there are standards by which to judge quality. I am not one of them. I believe that great art is rare in painting as it is in music or poetry, but that where we find it we confront a wealth and mastery of resources which transcends ordinary human comprehension. Even so, the preference for the primitive is an understandable reaction, for the increase of artistic resources also increases the risk of failure. Base line art is safer and all the more lovable for that.Yet it is human to want to transcend such limitations and to improve the language of art, the instruments of expression, towards ever more subtle articulation. This is, I think, what the twentieth century attempted to achieve by absorbing into its resources the modes and methods of primitive image-making. It extended the range of the expressive gamut to include both regression and refinement. Let me recall the two great artists of the twentieth century mentioned in the preceding pages: Picasso, who played on the resources of style as on a grand organ, using classical and primitive conceptions as the spirit moved him, and Paul KJee, the more gentle explorer of the means of artistic expression, who indeed made good his intention of learning from the art of children, without ever becoming childish. The more you prefer the primitive, the less you can become primitive.
After detailing how the "primitive" is the base line of human nature from which sophisticated expression has thankfully arisen, Gombrich is doing some serious backtracking in his final paragraph that ends by telling us "The more you prefer the primitive, the less you can become primitive."
Ultimately, he doesn't want take a stand against 20th C. Modernism, despite his repeated disparagement of all its facets: expressionism, surrealism, formalism.