Quoted text is in RED. Text quoted from other authors is in GREEN)
"Cimabue is the artist with whom Vasari begins his story of the resurrection of Italian art from the depths of barbarism"
Which is, apparently, why White begins at the same place, even if the modern historian is "shy of aesthetic value judgments.
Though, not too shy to note that "a sense of decorative unity is one of the first and most exciting impressions gained on walking round the upper church at Assisi"
White then walks us around the nave and transepts, detailing the stylistic unity achieved by Cimabue and his assistants.
(the complete cycle can be seen here )
The system created by this repetition of individually symmetrical compositions is historically most imponant, as it is seemingly unique. It is not a question of parallelism of subject matter, which is extremely common. Nor is it a matter of those subtle compositional harmonies which form a visual undercurrent in almost any great cycle designed by sensitive anists. Whilst there are occasional cases of the symmetrical possibilities of a pair of scenes being seized upon and emphasized, there are no earlier examples of a complete pictorial series undergoing such a rigid discipline. In particular, the arrangement seems to be entirely foreign to native Italian decorative schemes.
I wish he would have offered an example of what he considers typical of "native Italian decorative schemes" done prior to this date.
The understanding of the urufying function of the perspective of the painted brackets supporting the passageway broaches the question of the constructions used by Cimabue in his narrative compositions. This necessitates a glance at the general problem of the representation of solid objects, and at some of the historical antecedents of Cimabue's methods, which, afrer all, appear relatively late in the history of the representation of space and solidity.
So then, White offers the following graphic to categorize the representation of cubic objects:
In all primitive arts the first stage in the representation of any cubic object is invariably to show only a single side of it
But again, I wish he had offered an example. How many primitive societies even possess cubic objects? Is White referring only to ancient Egypt?
Observing that it is "the foreshortened frontal setting that is consistently used by Cimabue at Assisi", White gives us an analysis of "St. Peter Healing the lame":
The large polygonal building in the centre of the fresco is seen frontally. All the remaining buildings on either side are rectangular in plan, and all retain a single undistorted plane lying parallel to the surface. The effect of the receding faces is, however, softened by the bird's eye view employed for the whole scene. As a result all the planes running into depth and establishing the solidity of the architecture also run sharply upwards. The in running eye is at the same time allowed to continue the upward movement implied by the repeated verticals of the composition without any interruption. In this way the conflict between surface and solidity is reduced to a minimum.
Another important feature is the way in which the buildings recede, not towards the centre, but away from it. This has three main consequences. The first is that the eye, instead of being urged towards the middle of the composition by the orthogonals, or receding lines, is held there by the verticals lying parallel to the surface. The tendency is to read the design in the opposite way to that of a typical renaissance picture which draws attention inwards from all sides. Instead, the gaze moves from the centre outwards, much as the real world is scanned by turning the head from side to side away from the norm, or central axis, represented by a direct glance straight ahead. The tendency to organize his compositions consistently in this way marks one of Cimabue's major contributions to the development of space representation. It is principally this factor of organization which distinguishes his designs from those of many of his contemporaries whose buildings, while individually constructed on similar principles, recede haphazardly to left or right, regardless of their position in the picture. Cimabue's composition is, however, only in harmony with the physiological facts insofar as the eye is led to scan the scene from the centre. There is no question of a scientific interpretation of the facts of vision, or of conformity with them, whether as a whole, or as represented by a single focused glance.
To begin with, no, this is not an example of what White calls "foreshortened frontal" because the horizontals remain parallel so that the verticals do not become shorter as they recede into the distance.
Nor does White mention the role played by an arch from the architecture of the basilica itself.
But most importantly, he does not address the pyschological/spiritual effect of the design -- i.e. how the angles of those flanking buildings crank up the drama of the central event, and how the baptistery in the center regulates the figure of St. Peter whose magic touch occurs within the compass of its doorway.
Norris Kelly Smith would have begun his discussion by asking what this painting was supposed to do for the 13th C. Francisicans who commissioned it -- while White, in the spirit of Modernism, begins with how cubes are represented and affect the surface of the frontal plane. 
The diagram also demonstrates the third point inherent in Cimabue's design. Just as the high viewpoint minimizes the conflict between solidity and the surface, so the divergence towards the wings means that almost the whole width of the fresco is occupied by frontal planes harmonizing with the flat wall. Only at the very edge of the composition, where the framing bands emphatically re..assert the surface, does the eye begin to slide into depth, to be brought to a stop almost immediately by the confines of the pictorial world.
There is no evidence to suggest that the facades off all three buildings are in the same plane. The buildings on the two sides might be in front of the central baptistery and facing inward towards the square in front of it.
A viewer who feels that arrangement would be standing on the piazza along with the miracle working Peter -- and wouldn't that be exciting!
White's multiple viewpoints are, of course, possible, as are many other speculative arrangements. But why choose them?
****************
The Four Doctors of the Church,
and
the Scenes from the Old and New Testaments
White introduces these "Doctors of the Church"
in the vaulting at Assisi
as exemplary of another
"stage in the development of architectural space"
Isaac Master: St. Jerome
He points out that the orthogonals on the bottom
recede upwards,
while the orthogonals on the top
recede downwards,
which "increases the depth-creating qualities of all receding lines in the composition" and "the realism of the architecture is as a result greatly increased"
Though, he also points out that this is not the case with St.Gregory.
Unfortunately, a large reproduction of that one could not be found on the internet, so we'll just have to guess how much difference those downward recessions really make in those vault paintings. The Cimabue scene shown way above (Peter healing) has receding orthogonals that only point upward - and its sense of depth and realism seems no less to me.
To counter the argument that such "divergent recession" was connected to the curvature of the vaults, White points out that same formula can be found in the Isaac Master's depiction of "Brothers Interceding with Joseph", but again that one could not be found on the internet. (it is curious why no one has yet posted images of all the frescoes at Assisi - except for Giotto's)
White does not discuss what seems to be the main feature of this architectural apace: the sense of envelopment.
He asserts that "no definite caesura separates the two halves of the scene, since the various architectural elements stand upon a single unbroken base"
But the bottom line of the triangular frame is bent upward in the middle, leading me to feel that the two sides of architecture are facing each other, with the viewer standing in between them.
Feast at Cana
White also notes that "The lowering of the viewpoint, besides increasing the impression of solidity and depth, has the second important effect of immediately making a convincing interior possible--- the process begins with the early master who painted the two bays -- (which) include the story of the Feast at Cana"
Except that there are no divergent recessions here.
There are three horizontal bands in this scene, and each one is seen from a different point of view - which might tell the story, but makes no sense as a picture window.
Here's what White has to say:
The conflict between the various viewpoints is much more noticeable in this scene than in any of those already discussed. In Cimabue's type of composition the different points of view are consistently distributed across the
picture surface, whilst here a table seen from the left is placed inside a building seen from the right. The footstool on the left conforms with the table, and the now invisible throne on which the figure of Christ is sitting probably also receded to the left. It is as if the two halves of a Cimabue composition had been pressed together and telescoped in the process.
The placing of the servitors and the wine jars in front of the foremost columns of the architectural canopy greatly limits the sense of enclosure. The relatively early stage in the transition from a true exterior to a true interior, represented by this building, is further underlined by the foreshortened frontal setting. The latter allows its whole right end as well as its forward surfaces to be seen from the outside. On the other hand the receding beams and the coffering of the ceiling are clearly shown, and despite the rather sharp tilting of the table to display the plentiful feast, there is a definite sense of looking into a spatial enclosure.
I hadn't even noticed the framing architecture of the dining room, it seemed as extraneous to the scene as the architectural features of the basilica itself. But it is cute how it's seen from the right, while the table and footstool are seen from the left. While Mary, the large woman on the far left, is both behind and in front of the table.
Regardless of its weird perspective, this does seem to be a flaccid, clunky, un-inspired, poorly designed painting that's important only because of its location. 
Especially when compared to Veronese's version
from about 300 years later.
Though, as Norris Kelly Smith might have noted,
the version at Assisi was painted for the home church
of the Franciscans
for whom food was to be shared with the poor
rather than consumed by the rich.
The tipping table at Assisi
seems to be offering it to pilgrims
who are presumably the intended viewers
while the servants are busy
handling the miraculous wine
that keeps on flowing.
Jesus seems to be lost in the crowd
who have partying on their mind.
The feeling for the interior increases greatly in the scenes of 'Isaac and Jacob' and 'Isaac and Esau' in the second bay from the entrance. The building is still seen in a foreshortened frontal setting that allows the exterior of the right wall and the outer boundaries of the open forward wall to be inspected. The scale in relation to the picture space as a whole is completely changed, however, and the edges of the building squeeze against the frame on all sides. There is no longer room for figures to move between the roomspace and the spectator, whilst the avoidance of conflicting viewpoints and the more naturalistic foreshortening of the upper surface of the bed combine to increase the reality of the enclosure.
Yes, this does feel like a window
onto a separate, imagined world.
There's even a curtain
that has been partially drawn
so that we can look inside.
In the last example, the 'Pentecost' on the end wall of the entrance bay, the beginnings of a centrally viewed interior are seen for the first time at Assisi. Here the rectangle of figures, those seated nearest to the onlooker facing inwards with their backs completely turned, gives an impressive sense of reality to the space which they enclose. The rather shrunken quality of the half Roman, half Gothic triple vaulted building, which scarcely seems to overhang the figure space at all, is at least partly explained by the difficult L-shaped field. In spite of this the solidity of the architecture and the strongly emphasized, if disparate figure room, together with the centralized setting which places the perspective emphasis within the space enclosed, mark an important step towards the many fine interiors of 'The Legend of St. Francis', which completes the decoration of the nave.
Briefly scanning the internet, I can't find any depictions of this event that are quite like this one, with the room enclosed on three sides and completely open to the viewer - as if you might join the disciples sitting on the near bench, which is made to protrude forward as it echoes the protrusion of the three vaults above them.
This special moment is not just for them. Just as with the feast at Cana, you, the pilgrim, are invited to join. A big empty space on the bench has been left for you, right in front of the holy Virgin.
The top edge of the side walls of the partition behind the disciples recede to a much higher horizon line than the receding lines in the vaulting behind them, so White might suggest that this painting is only an intermediary "step" towards true linear perspective.
But that also has the effect of tipping the room of disciples towards the viewer, making that viewer feel more included.
In contrast with the above contemporary version by Giotto.
A wall of columns separated old man Scrovegni from the holy spirit that had already descended upon the blessed apostles.
And there is no room for him on the bench.
*******************
The Legend of St. Francis 
The clearest evidence that the master responsible for the main body of 'The Legend of St. Francis' starts from that vision of reality shared by the painters of the middle register and vaults of the two entrance bays is to be seen in 'The Vision of the Thrones'. The design is entirely composed of foreshortened frontal objects, their forward surfaces emphasizing the smooth plane of the wall, and the heavenly thrones are all seen from the left, whilst the altar and its canopy recede towards the right. There is even a good deal of structural uncertainty in the latter, since the verticals supporting the roof brackets have their bases set at widely different depths, although their tops lie in a single plane.
The heavenly thrones are seen from the left because that's how the monk on the left would see them. The chapel is seen from the right to establish it as a separate scene with its own protagonist, Francis.
The "structural uncertainty" of the chapel results from an attempt to create an interior volume for the chapel (which is also established by the candelabra that looks a lot like the one in Giotto's mural in Padua)
And, I'd like to suggest that it's only in the last two centuries that artists "started with a vision of reality". 
It is, perhaps, interesting that the only other fresco in which the orthogonals diverge towards the edges of the composition is the next-door 'Vision of the Fiery Chariot'. In both cases the radical change in viewpoint is accompanied by a change from one reality to another, from a material to a visionary world. This may easily be no coincidence, as these are also the only two frescoes in which objects placed in the upper part of the picture are seen from above, in contradiction of the normal viewpoint of the buildings below. The distinction between the way in which earthly and heavenly realities are seen is therefore made in the vertical, as well as in the horizontal sense.
Agreed!
But both of these paintings bother me.
The overall plan seems inspiring,
but the drawing feels so clunky.
A fine example of a building seen from a normal viewpoint with its sides foreshortened and the frontal surface held firmly in the plane, occurs on the right of 'The Dream of Innocent III', the final fresco in the second bay . The sloping lines of the Lateran supported by St. Francis, on the left of this same scene do not reflect a different type of spacial setting, but the tilting of the entire body of the collapsing church
This "early stage of empirical perspective" might also be called an advanced stage of visionary perspective.
The sense of drama/mystery within the empty space is delicious, and such a contrast with the overstuffed "Feast at Cana".
The change is not fortuitous, based merely on a desire to introduce a new decorative pattern and vary the monotony of repeated frontal and foreshortened frontal settings. It is, instead, one of the many symptoms of the growing excitement aroused by the direct observation and imitation of the natural world.
It is a simple optical truth that as soon as it is possible to see more than one side of any cubic object, all the visible surfaces will recede from the observer and will be foreshortened. ..... The convention, considered up to now, which allows one face to remain unforeshortened, parallel to the representational plane, was, for a time, progressively discarded by those artists who were inquiring most acutely into what it was that they could actually see, who were looking most intensely at the individual objects in the world about them, and who were trying to represent these objects more faithfully than their predecessors.
A rather succinct statement of White's ideology that privileges scientific observation over mythic vision:
"progressively"
"inquiring most acutely"
"represent more faithfully"
In the extreme and most characteristic case of the oblique setting, all the visible surfaces of a cube recede with equal sharpness at fortyfive degrees to the observer’s line of sight. In pictorial terms, this means that all the visible faces of the object will also be receding as sharply as possible from the picture plane. This situation is diagrammatically represented in Fig. 2, b
If, in this drawing, the rectangular group of buildings shown in plan and continued in the converging dotted lines is considered as a solid block, it is essentially a representation of the ground plan of a building such as the ruined church in the fresco under discussion.
If, on the other hand, the extension is ignored and the buildings in plan are considered as separate units, then the rectangular inlet with its jutting wings shows in essence the position of the houses on either side of the fresco of ‘St. Francis Giving Away his Cloak’ 
or of ’St. Francis Repudiating his Father’
In either interpretation the extreme nature of the conflict between the individual solid object and the picture plane against which it thrusts its wedge-like shape is dramatically emphasized. The diagram, besides stressing the intensification of the element of conflict between the real and the pictorial worlds, also reveals a certain connection between the new representational method and the earlier artistic solution found by Cimabue. 
The two flanking buildings shown in Figure 2, a, have been retained in Figure 2, b. All that has happened is that the short, forward surfaces in the first drawing have been pulled back into steep recession in the second, making them correspond with what would actually be seen by looking at such buildings. The inference is not that the second solution is a correction of the first, from which it derives. This is historically unlikely, as the Cimabue stage does not always precede the development of the second visual pattern. The comparison does, however, underline the conclusion, reached when discussing Cimabue’s painting, that his compositions represent a first, tentative move towards the recognition of those same visual facts which also underlie the later development.
Rheims, 13th Century
Bonano Pisano, 1180's
BTW, here's a few sculptures from preceding decades.
They also represent some cubic forms (among other things, and show a similar contrast between forms that lie flat against a background plane (Bonano) and forms that are emerging from it (Rheims)
********
It's so wrong-headed
to look at that painting
of "Francis Repudiating his Father"
and consider the architectural background
as a conflict between "an individual solid object and the picture plane"
instead of an enhancement
of the conflict between the
two figures in the foreground,
Francis and his father,
as expressed by the gestures
of both the people and the buildings.
Even if we do notice
the stronger assertion of pictorial space
in Giotto than in Cimabue.
It is noticeable that, to begin with, it is very largely the extreme version of the oblique pattern which is used by progressive painters. It seems as if, in the first excitement of discovery, it was impossible for the adventurous late thirteenth-century artist to resist making every object just as solid as it could be whenever he used the new construction. This is particularly interesting, since it is generally agreed that both in primitive art and in the reviving naturalism of the period leading up to the Renaissance an interest in the object itself precedes any interest in space as such. The interval, or nothingness, which separates one solid from the next, is relatively unimportant.
White definitely does not feel the drama
created by the space between Francis and his father,
no do his notes indicate who else
has joined in that "general agreement"
that intervening spaces are relatively unimportant
or which examples have been offered.
Such abstract statements take on concrete form when the essential qualities of the oblique construction are investigated.
The dominating feature of any building shown obliquely is the vertical of the forward jutting corner. Just as it is upon this main axis that the observer focuses his attention in reality, since it enables him to judge at a glance the apparent slope of the evenly receding pairs of horizontals, so it is this central axis that holds the spectator’s attention in the pictorial world.
In any focused system of perspective, such as that invented in the Renaissance, the vanishing point provides the all important centre of attention. In such a system the most important architectural elements are the orthogonals, or receding lines, which lead the eye towards the focal point lying in space beyond the confines of the buildings themselves.
In the empirical system based on the oblique setting the situation is radically different. It is clear that the main accent in the building of the ‘St. Francis Before the Crucifix’ is indeed the vertical of the forward corner, softened though it is by the crumbling of the masonry. It is equally significant that not only at this point, but throughout the composition, attention is controlled by the verticals rather than by the receding horizontals. These merely move the eye outwards to either side of the central axis. It is stopped, and concentrated on the important elements in the story by the verticals which frame both saint and crucifix. It is, moreover, the magnetic force of the building’s forward jutting axis which holds back the eye, returning it to the centre, and preventing it from following the receding lines out into nothingness beyond the confines of the composition.
Which is just to say that if receding orthogonal lines
point to a vanishing point that is outside the picture,
that point cannot be a center of attention within the picture.
That's rather obvious, isn't it?
But I'm not sure how attention is controlled
within a painting.
What if we cropped off the front of the building
so that we no longer had an "oblique setting"
with the "magnetic force of the forward jutting axis"
Is your eye any more free to wander down
those receding horizontal lines?
What if we also cropped off the bottom
so that we no longer have any converging horizontals.
What is now missing?
I think it's the sense of anticipation and mystery
created by the rhythm of volumes within the empty church,
as well as the emphasis on the praying hands
which echoed a rear broken wall on the left
and pointed to the empty volume above the crucifix.
The fundamental quality of the extreme oblique construction, the control of attention by the feature which epitomizes its aggressive solidity, can be tested in any of the innumerable examples which occur throughout Italian fourteenth century painting, or in the art of any other period in which it appears. It is inherent in the construction, and quite independent of the individual composition. Reduced to diagrammatic essentials, the effect is shown by the two cubes in Fig. 1, d 
If the rest of the diagram is covered up, it is very noticeable that attention tends to centre on the two cubes themselves rather than on the intervening space. 
If the foreshortened frontal cubes in Fig. 1, c, on the other hand, are isolated, it is on the space between, and not upon the solid objects that the eye is chiefly concentrated. This is particularly interesting as the diagram does not illustrate a focused system which would greatly increase the effect.
I wonder how this thesis could be tested?
(and if White is going to privelege scientific thinking,
how could be make such an assertion
with some kind of test?)
It seems to me that the bilateral symmetry of both those diagrams
equally focuses attention to the center.
The tendency of the extreme oblique construction to concentrate attention on the isolated solid object is accentuated by the fact, illustrated in Fig. 2, b, that the use of an extreme oblique setting for a number of buildings in a design entails the retention of the multiple viewpoints of the older system (Fig. 2, a).
In the frescoes of 'St. Francis Repudiating his Father’ , ‘St. Francis Giving Away his Cloak’ , or ‘St. Francis and the Demons at Arezzo’ , the tendency of each architectural block to isolate itself from the other similar elements of the composition is clearly shown. But if the onlooker capitulates, and allows himself to concentrate for a moment on either of the two towns shown in the last two frescoes, an additional fact immediately becomes apparent. In either case, the spaces separating the houses have completely disappeared. Once again, this is not peculiar to these particular compositions. It is inherent in the construction itself. There is no possibility, no matter how logically a town may be organized, of looking down a street, of letting the eye rove through the houses directly into depth, as in a typical renaissance picture. 
(Masolino's "Feast of Herod")
All the streets veer away diagonally, hidden by the houses. These congeal into a single, wedge shaped mass, as solid as the individual buildings which compose it.
White chose that Masolino painting as a "typical Renaissance picture", and we might note that the architectural constructions on either side feel much more independent of each other, with no emphasis on the center, than do any of the Giotto paintings.
This same phenomenon accentuates the tendency to isolation noticed already, which is partly a direct result of the constructional dominance of verticals that, by their nature, cannot lead the eye across from one part of a composition to another. In any of the three examples chosen from ‘The Legend of St. Francis’ it is very difficult indeed to associate the two halves of the design as far as the architecture is concerned.
But how far is the architecture concerned? The two halves of a design can also be associated with each other using other visual or narrative elements - it's just that White has chosen to ignore them all in this book.
This quality of emphatic, se1f iso1ating solidity, which restricts the creation of space to the dimensions of the block-like objects trying to force their way through the flat surface of the wall, underlies many of the characteristic achievements and limitations of these frescoes of the story of St. Francis.
The interest in direct observation shown by the increasing use of the oblique setting; by the portrait quality of so many of the buildings; by the descriptive naturalism and spatial grouping of the figures; and finally by the multiplicity of carefully reported detail, is equally visible in the representation of interiors. Centrally seen interiors are now more numerous than foreshortened frontal compromises. The increased authority with which the problem is tackled can be seen in ‘The Approval of the Rule’.
The audience chamber extends almost to the full height of the design, and only at the top are the forward surfaces of the bracketed vaulting of the side wall left in view. Below this point the side walls are cut by the frame. The figure on the extreme right, standing within the room space, is only partly visible, whilst the lower border cuts across a floor which does not stop within the confines of the scene. The realism of this limited space is further stressed by the double line of figures ranged in depth upon the right, although the spatial content of the left-hand group is less convincingly described. In the upper part of the fresco a wealth of firmly constructed vaults and distinctively coloured orthogonals make both the plastic and the spatial content of the architecture absolutely plain. Nowhere is there confusion between lines leading firmly into depth, and transverse, horizontal cornices. There is no doubt about the growing conquest of the enclosed space of the interior, as there is none about that of the solid object limited space which, for these artists, lay outside it.
White does not mention
the absence of a proportional relationship
between size and distance.
The placing of the figures within this new, interior space is of particular interest throughout ‘The Legend of St. Francis’. Their relation to the architecture in ‘The Apparition at Aries’ is a direct extension of the idea already present in ‘The Pentecost’ on the upper wall.
Apparition at ArlesPentecost
Was the painter trying to make
architectural details as they
actually might appear,
or have they been made to echo/enhance
the expressions of the figures?
It is this same close union of figures and architecture in the definition of space which is continued in the Arena Chapel at Padua. 
The 'Last Supper' provides the finest example of this arrangement.
There, no place is found, amongst the narrative scenes, for the Gothic realism of ‘The Preaching of St. Francis before Honorius III’, but the space defining semi-circle of the figures is re—stated in still more magisterial terms.
I like how the five sets of tall, narrow windows
near the ceiling
seem to express the inquisitiveness
of the papal court
as it listens to Francis preaching.
And I wonder about the terms
that White uses to distinguish
the "magisterial" frescoes at Padua from the "Gothic Realism" at Assisi.
This is where he steps away, momentarily,
from his examination of realism in architectural renderings.
I'd agree about the
"close union of figures and architecture".
Each church has it own variant
of figure and architecture.
Assisi's is appropriate for monks and pilgrims,
while Padua is appropriate for a retired banker.
Two of the remaining interiors at Assisi, however, present a different problem.
In the fresco of ‘The Crib at Greccio’, the interior of the church consists of part of a choir screen and pulpit, sharply cut at either end by the frame, and seen from the choir itself so that only the back of the large crucifix is visible together with a free-standing, canopied altar, and a lectern on a table base. There is nothing else. Above, undifferentiated blue spreads out a seeming sky. Nowhere is any trace of the enclosing architectural shell apparent.
In ‘The Funeral of St. Francis’ the view is taken looking from the have towards the choir. A rood beam, surmounted by a crucifix and thirteenth century panels, runs across the scene. Its end supports, just visible on the extreme left and right, implicitly reveal the architectural limits. This is confirmed by a low wall, of the same height as the rood beam, receding gently on the left, its top line dipping only very slightly below the horizontal. This leads to a small apse surmounted by a semi-dome that shows its blue cowering above the level of the beam. Overhead, and to either side, undifferentiated blue once more takes over. But although no attempt is made to show the roof or upper walls of the church, the artist has tried to suggest their presence by hopefully hanging panels, crucifix, and even the sanctuary lamps from the amorphous blue of the sky.
These two scenes, and especially ‘The Crib at Greccio,’ have been put forward as convincing evidence that the frescoes at Assisi must at least be later in date than those in the Arena Chapel. This conclusion, based on the argument that the ‘slice of life’ impression, the sense of spatial continuity beyond the confines of the frame, is far in advance of anything to be seen in the late thirteenth century, is by no means generally accepted. (Friedrich Rintelen)The fallacy in the argument itself is not so clear, however. The sense of space is indeed remarkable, particularly in the more coherent Christmas scene, which gives less indication of the architectural entity of the church itself. The modern imagination is left free, and the mind’s eye sees visions far beyond the descriptive scope of the men who painted at Assisi.
The description of the structure of the two scenes shows that the starting point is not the church as a whole, but its internal furniture. In each case the artist stops short at the point beyond which it seemed impossible for him to carry conviction. In ‘The Crib at Greccio’ no effort is made to indicate the enclosing body, and in the funeral scene the attempt is made, but perforce abandoned halfway with results that are less, not more convincing, since the imagination is brought down to earth with something of a bump. This is a completely different approach to the problem of the interior from that in which the building, however ill equipped to contain the figures, is the essential starting point. It is, moreover, historically the more primitive of the two.
Why is the absence
of an architectural frame
such a "problem"?
Why is the word "primitive"
applied?
And why is no consideration given
to either the overall decor of the church
or the specific narratives involved
in those two paintings?
I can't recall how this design works
in the actual church,
but in the reproduction
it's use of a single beam
cutting horizontally across the middle
is fascinating
and, so far as I can tell, unique.
Veneration of the True Cross
Two tenth-century illustrations from the Menologion of Basil II are particularly revealing in the present context, as they provide a remarkably close parallel to the late thirteenth century compositions.....they reveal quite clearly that the artists concerned had little or no feeling for the closed interior
Unfortunately, only one of the two pages
described by White is available on the internet.
(the Vatican owns it, why don't they publish it?)
With that wall curving back
and that distorted head on the right,
this almost appears to be a reflection
on a fun-house mirror.
While White's response
reveals quite clearly that he had
little or no feeling for either spatial design
or sacred narrative.
With the Menologion in mind, the significance of the two frescoes at Assisi is less difficult to grasp. They represent the furthest point of development of a very old idea, not the beginning of a new conception. They belong,in type, to the group of works including all the early and mid-thirteenth century scenes of churches indicated only by the presence of an altar or some other part of the internal furnishings. This compositional type, the influence of which survives throughout the fourteenth century, precedes the already partially described process of constructing an interior by means of a clearly defined building that gradually moves forward, expanding until its borders run beyond the confines of the frame and all the figures are contained.
But how important are those "old ideas" or "new conceptions"
to the total experience of those paintings
and whatever their original purposes might have been?
And if we're going to be assigning a "compositional type",
wouldn't the bilateral symmetry of the above scene
be the most fundamental way it differs from the
two Assisi paintings to which it is being compared?
The realization that ‘The Crib at Greccio’ and ‘The Funeral of St. Francis’ reflect an application of the new sense of spatial composition to an old scheme based on the individual interior object is, perhaps, confirmed by the fact that they lead nowhere. No further development of the idea occurs in the succeeding years, and it is clear that it did not have the same effect upon the contemporary artists and their public as upon the modern observer. Already at Assisi, the conception of the interior as an enclosed, and enclosing box is well enough advanced to carry full conviction, and it is this idea which is not merely echoed, but continuously developed throughout the fourteenth century. The placing of these two outstanding compositions at the end, instead of at the start of a long development, does nothing to detract from this remarkable achievement. It only shows that they cannot be used as evidence in an evolutionary, date-determining argument which ignores their historical antecedents and their subsequent influence or lack of it.
If we're going to examine "the old scheme"
what's really needed here
is a website that covers Byzantine painting
in a comprehensive way.
Was Assisi really the first example
of a wall divided into deep picture boxes
each of which tells a story?
Was it also the first example
of a space dedicated to telling the story
of a modern saint?
And, as White then asserts,
was there also "an unprecedented
effort to tie the whole ensemble together" ?
In something of a digression,
White details out that effort
over the next eight pages (40-47).
It's a fascinating topic,
but how can one understand his argument
without actually standing in the church itself?
*******************
CAVALLINI AND ROME
The striking coherence in the development in the representation of space at Assisi raises an immediate question as to whether this is an isolated phenomenon, deceptive in its orderliness, or a true reflection of the main stages in the growth of spatial realism in 13th C. Italian art. The answer lies in Rome, and more especially in the art of Pietro Cavallini 
Presentation at the Temple,
Santa Maria in Trastevere
Most of White's attention
goes to Cavallini's cycle of frescos
at S. Paolo Fuore le Mura,
but since that only survives
as sketches made by Cardinal Barberini,
I'll skip that discussion
and focus instead on a later mosaic
shown above.
Quite a difference
when this wall is lit by the window
instead of light fixtures in the church.
The window light shining
on the reflective curved surface
creates a sense of depth
that is completely lost at night
and in the black and white reproduction
in White's book.
(and one might note
that this depth has not been created
just by diagonal lines receding into space)
Here's the entire apse.
Regarding this piece, White says:
The architecture is so constructed that the base is in the extreme oblique setting, whilst the horizontal main line of the top appears to reflect a compromise with the foreshortened frontal type.
I'm really doubting
that the artist
was aware of any such compromise.
But it is a rather unusual arrangement.
The roof of the structure in the foreground
is seen from below,
while the apparently taller structure in the rear
is seen from above.
So the only place the viewer
could possibly be standing
is outside the imaginary scene
and inside the real church's apse itself.
The imaginary architecture
only functions to complement
the actions of the figures
set before them,
just as it does at Assisi
Cavallini - Last Judgment
The facial expressions
are similar to Assisi also.
Cavallini - Last Judgment
Assisi - Judas
While both make an interesting comparison
with the following
that dates from about 1261
in Constantinople:
Christ Pantokrator, Hagia Sophia
Christ with the Pharisees, Duccio
and with this one from around 1310.
Wouldn't that kind of comparison
be more interesting
than an analysis of the representation
of cubes?
