The square plan is the main one for the cross vault. The pure form of this vault, consisting of four equal strips with two mutually perpendicular axes, dictates the square shape of the plan. On the contrary, a barrel vault on a square plan produces an unpleasant impression due to the contradiction between its one longitudinal axis and the two axes of the plan. Just as bad is the cross vault on a rectangular plan; with horizontal strips, the formworks in this case have different shapes, with an elliptical curve along the long side of the plan. The cross vault on a rectangular plan loses its spatial integrity.
The formwork of the cross vault is directed from the center of the plan outward, to its perimeter. Not relying on the steppes, but only touching them, the formworks do not close the space, but dismember it in four directions. Under these conditions, deaf closing walls would contradict the spatial image of the vault; therefore, it is more expedient to fill the tympanums of the formworks of the cross vault not with blank walls, but with glazed surfaces that do not close the space. We find such a technique in the middle nave of the basilica of Maxentius and Constantine (Fig. 180).
If the barrel vault rests on the walls everywhere, forming a single whole with them, then the cross vault, resting on the corner columns, denies the walls and can exist without them. It is not very suitable for enclosed spaces and is rarely found in indoor
halls of buildings both of the Renaissance and later. Its form was, as it were, created for open spaces, and it was readily used by the best architects for loggias (Loggia dei Lanzi in Florence), outdoor arcades (Brunellesco's Educational House). Predominating in the lush coverings of the basilicas and thermae of Rome, the cross vault gave way to the sail vault in the architecture of Byzantium and again gained dominant importance in the architecture of Western feudalism in the 11th-14th centuries. The Renaissance found a true place for it in the architecture of outdoor open structures.
In the inner halls of Roman basilicas and thermae, the diagonal ribs of the cross vault are conjugated with the capitals of the columns (although there is an insertion of a piece of entablature), but the columns do not carry the vault and are attached after its erection (Fig. 181). The pressure curve of the vault runs in the thickness of the massive walls much higher than the fictitious support - the capital of the colony. The builders of the basilica knew this, creating special buttresses on top of the roof to absorb the pressure and installing columns of expensive marbles after the construction of the building, during its decoration (Fig. 181).
In visual perception, the columns seem to carry the vault, and thus, the requirements of tectonics are met, but in reality the columns do not perform the functions of supporting the vault.
Among the Romans, engineering art was in the service of architecture, but did not switch to the highest level of approach to the synthesis of technology and art. The Renaissance, using the cross vault in loggias and outdoor open arcades, correctly used its basic, fundamental architectural properties. In the absence of walls, the diagonal ribs of the vault clearly rest on the corner columns, which are their only supports. Without making false attached columns and without masking the buttresses, which perceive the thrust, the architects of the Renaissance frankly used open metal connections.
Thus, we have two compositional systems: the ancient system of cross vaults in the enfilade of the thermae hall, which gives the aesthetic impression of a light floating cover, independent of the walls, but structurally false, and a truthful architecturally complete system of the Renaissance.
The first of these two systems, the ancient one, Sedlmayr called the canopy system. One can agree with this name, but it must be pointed out that the covering of an open structure in the form of a rotunda, canopy (canopy), arcade, etc., where there are no walls and decorative columns at all, but only really loaded columns, such a covering should be called true canopies . Vaulted coverings, in which, by means of attached columns, only the impression of free, unloaded walls is created, should be called a false canopy.
In all architectural styles, one can trace this technique of spatial composition - this canopy system, but only in different versions and interpretations. The geometric shape and constructive solution of the canopy system can be very diverse. For its spatial construction, the forms of a cross sailing vault, sail-closed, fan, and also a dome can be used.
1 As is the case in St. Sophia in Constantinople, created by Anthimius of Traless.
On fig. 182 shows possible solutions for the canopy system.
Fig. 1 gives the basic system of the canopy, with the groin vaults of Roman thermae.
Fig. 2 depicts a possible variant in the form of a mirror vault on the columns. The mirror vault of the Renaissance was used mainly in enclosed spaces, along the walls; examples of mirror vaulting on columns are very rare. On fig. 164 shows a mirror vault on the columns in the Munich art gallery Shack, in fig. 176-mirror ceiling of the old town hall in Paris.
Fig. 3 gives the third form of the canopy, from sail-closed vaults. This form is extremely rare. As an example, let's name the Gothic vault in the French chapel of the Holy Spirit (St. Esprit) in the city of Rue (Fig. 273).
Fig. 4 depicts a canopy system of sailing spherical vaults. Along with the system of cross vaults, it is the most valuable in the architectural sense. Ancient Rome did not use this system. Byzantium in the Justinian era gave already completed canopy systems: the side galleries of St. Sophia in Constantinople form suites of freely floating spherical canopies on columns. In the Renaissance, there are separate cases of the use of a sail vault - in the Pazzi Chapel, in the portal of the Uffizi Palace in Florence (architect Vasari), in the Doria Palace in Genoa (1564), in the Lateran Palace in Rome (1588) etc., - however, this era did not leave us architecturally complete canopy systems. In later epochs, systems of canopies made of sail vaults in two buildings of French art of the 18th-19th centuries are especially interesting: in the Madeleine church in Paris, built by Vignon (1762-1828), the canopy system consists of three sail vaults, based on attached columns (Fig. 371); in the Parisian Panthéon built by Soufflot (1709-1780), four separate cruciform sail canopies rest on free-standing columns (Fig. 370).
Fig. 5 depicts a canopy of fan vaults. Amazing in their lightness and boldness, examples of such canopies are known in England in the 14th-15th centuries. The best of them are in the chapel of Henry VII.
Fig. 6 gives Viollet le Duc's scheme with conical funnels.
These examples show the variety of compositional and constructive solutions of the canopy. We will return to this problem in later chapters.
Now, after we have noted the general architectural and compositional features of the cross vault, its main forms, its place in the overall composition of the building and its interpretation in different eras, let's move on to a detailed study of possible geometric forms.
II. FORMS OF THE CROSS Vault
The basic, strictly geometric form of Roman cross vaults underwent significant changes in the process of technical and decorative (stylistic) development. Some changes have affected the main guiding curves of the vault of both strippings and diagonal ribs, others relate to the shape of the surface of the strippings themselves.
Already Byzantine masters increase the diagonal curve and simplify the shape of its curve. Instead of a low elliptical curve, a diagonal edge is drawn with one radius from the lowered center C (Fig. 183). The height of the vault is greater than with simple cylindrical formwork (equal to half the side of the base plus a certain value h, less than the half-diagoal). The formwork strips are raised (raised), their surface from cylindrical becomes spheroidal. Cross-section of the vault of spheroidal formworks by horizontal planes gives a four-petalled outline (Fig. 183).
From this example, it is clear how a change in the forming curves of the vault entails a change in the shape of the formwork due to the close relationship between these elements. In addition to curves and stripping, the angle of the diagonal rib is also subject to change: at the support it is 90°, as it rises along the rib to the slat, it increases and disappears at the top (180°); with a raised curve, the flattening of the angle to the shelyga goes faster. This softening of the angle was used by Renaissance architects for decorative purposes - to obtain a large smooth surface (plafond) in the shell.
Undoubtedly, Gothic, for which this type of arch is the main one, gave the greatest variety of forms of the cross vault. The change in the shape of the Gothic cross vault went hand in hand with changes in the curves and arrangement of the ribs, creating intricate decorative patterns. Medieval architects extracted artistic effects from the structural frame of the vault, while improving its technical side, masonry methods and material selection. In view of the very complex interweaving of all these constructive and compositional factors and the dependence of the forms of cross Gothic vaults on the decorative design of the frame, the study of these forms must be postponed to the end of this chapter. Now we will analyze the main geo-
On fig. 184 shows four types of surfaces: FIG. 1 - cylindrical, fig. 2 - cooper, fig. 3 - conical, fig. 4 - ellipsoidal.
The first type has ADO strippings in the form of cylindrical surfaces inclined at an angle a, and the stripping sheaves - straight DO - are also inclined at an angle a. To construct a line of diagonal ribs in the vertical projection of the arch, we divide the main stripping curve AB into nine parts. Let us draw on the horizontal and vertical projections the generating cylinders from division points 1, 2, 3 and 4. Points 1", 2", 3" and 4" of the diagonal edge are obtained on the vertical projection as the intersection of the generating cylinders. Curve AO of the rib of the vault represents, as a section of the cylinder by a plane, an elliptic curve, as well as the curve BO, starting from the support B. Both diagonal curves AO and BO intersect in the crest of the vault at a certain angle. In view of this, plowing along straight inclined shelygs does not give a flat surface at the top, convenient for a picturesque ceiling. This served as an obstacle to the use of such a vault in the Renaissance, but it is found in Gothic.
In FIG. 2 shows a cross vault, ripped along the curve DO, with radius R and an arbitrary center C. When moving the semicircular curve ADB along the ripping curve DOC, the surface of the formwork will be spheroidal, double curvature, which is usually called a cooper. The cross section of such a surface by a horizontal plane gives a four-lobed shape in plan, as indicated earlier in Fig. 183. In this case, the horizontal projections of the seams 1-1", 2-2" and 3-3" will be the same as with cylindrical stripping in Fig. 1. The vertical projections 1 - 1", 2-2", 3- 3" and 4-4" will be outlined from the same center C. The curve of the diagonal rib will be of an indefinite (close to elliptical) shape, but without a break in the crest of the arch O.
In FIG. 3 shows the case with conical stripping. Having chosen the top of the cone at point M (on the left side of the drawing), we draw through points 1, 2, 3 and 4 of the side arch forming M-1, M-2, M-3, etc. both in horizontal and vertical projections. In addition, from another vertex M "of the frontal stripping, we draw vertical projections of the generators, in the form of radii M"-1, M"-2", M"-3, etc. At the intersection of the generators of two adjacent strippings, we get on the diagonal of the plan and on elevation point 1", 2" and 3" and 4" elevation edges.
The section of the cone of the left stripping with a diagonal plane will give an ellipse with a major axis AB and a minor axis CC "(shown in the drawing by a dotted line in combination with the horizontal plane). The vertical projection of this diagonal ellipse DAB will also be an ellipse A" CB (drawn by a dotted line on a vertical projection). The vertex C of the ellipse is higher than the point O, i.e. the top of the arch, however, the generatrix of the MC of the cone intersects the diagonal edge at the point K, which lies below the top of the arch O. The section of the diagonal curve AO (from the heel to the shell) represents a segment of the diagonal ellipse, less than its quarters, rising from the base of the vault to the height H. Similarly, the other section 0D of the diagonal edge will be the same segment of the ellipse as AO. At the top of the arch O, both of these sections of the ellipse will be mated at an angle, without forming a smooth curve. Consequently, plowing along a straight line in the case of a cylindrical (Fig. 1) or conical (Fig. 3) stripping results in a fracture of the diagonal rib at the top of the arch.
The fourth, most interesting way of constructing an ellipsoidal formwork is shown in Fig. 4.
Having described an arbitrary ellipse with the C-C axis around the square plan ABB, we will rotate it around its own C-C axis. The surfaces of the left and right stripping will then be the surfaces of an ellipsoid of revolution (see vertical projection). The vertex O of the ellipsoid will be the top of the vault lying at the height H. In the same way, the upper and lower strippings will be formed by the surfaces of another ellipsoid with the axis OE. To obtain a vertical projection of the line of intersection of two mutually perpendicular ellipsoids, we use a horizontal projection in the form of two diagonals AB. Next, we cut the ellipsoid with planes passing through points 1, 2, 3 and 4, lying on the wall arches, and through its axis SS. To draw intersection curves, we draw transverse planes /, // (coinciding with the side of the square) and /// along the center of the arch. These sections are shown on the vertical projection as circles /, // and ///. The planes of the section of the ellipsoid OE will be depicted on a vertical projection with radii O-1, O-2, O-3 and O-4. The planes of the section of the ellipsoid С-О-С are depicted on the vertical projection by the curves С "-1-С", С "-2-С", С "-3-С", С "-4-С" and C "DOC" " ". The points of the vertical projection of the diagonal ribs are determined by the intersection of the radial lines O-1, O-2, O-3, etc. with the curves of the section of the ellipsoid. On the horizontal projection of the diagonal edge, points 1", 2", 3", etc. will be obtained by the intersection of the diagonals of the plan with the horizontal projections of the sections of the ellipsoid. In all four types of stripping (Fig. 1 - 4 Fig. 184) lines in plan and vertical projection give an image of working beds of masonry.
Of all four types of strippings, cylindrical (Fig. 1) and conical (Fig. 3) give a broken diagonal curve and a rigid geometric form of stripping. In the other two solutions, we have strippings of double curvature - a barrel surface (Fig. 2) and an ellipsoid surface (Fig. 4). Of course, the surface of an ellipsoid, approaching a spherical one, is more pleasing to the eye, but its implementation is difficult, requiring a variety of circles built by points. The barrel surface is easier to do, since here all the circles are drawn with two radii r and R. Both solutions are good in that they give smooth curves for the diagonal ribs, without a break in the shell (see vertical projections).
From the foregoing, it can be concluded that for stripping, based on diagonal ribs, you can choose any convex surfaces, as well as spherical ones, with vertices at any points of the vault plan. As we will see later, Gothic architects used various swollen spherical formwork. In FIG. 4 fig. 184, one can trace the transition of the cross vault into the sail vault. If the major axis of the ellipse C-C is shortened, then the minor one will lengthen. In the limit, both intersecting ellipsoids will turn into one ball circumscribed around the plan with a radius equal to the half-diagonal. In plan, the ball is shown as a circle drawn with a solid line. The sharp diagonal edge of the intersection of ellipsoidal formworks will completely disappear, since all four formworks will lie on the same spherical surface. The arch will turn from a cross into a spherical sail.
The considered geometric forms of formwork are the main factors influencing the spatial image of the cross vault. Form stripping also determines the shape of the curve of the diagonal rib, which is the line of their intersection. Given the surfaces of strippings, we get diagonal edges as their derivatives. The Gothic masters, on the contrary, set themselves on the crooked edges of the frame, which was the main shaping and decorative factor of the entire vault, and the stripping between them, greatly crushed, served only as a secondary, local filling. The construction of the decorative and constructive frame of the Gothic vaults will be analyzed below, but here it remains to consider the change in the profile of the diagonal rib protruding at an angle and the formation of various forms of plafonds in the shell.
Fig. 2 fig. 1851 depicts a groin vault with a diagonal rib rounded off. Renaissance masters often resorted to such a means, especially when painting on the vault. This is how the ribs of the vault in the stanza della Senyatura, painted by Raphael, are rounded (Fig. 209).
1 Fig. 1 fig. 185 depicts the basic form of the cross vault.
It is possible to cut the rib with a straight chamfer, as shown in Fig. 3 fig. 185. A greatly enlarged chamfer will be read as an independent part of the vaulted surface, namely, as the surface of a sail-somkiut vault. Instead of one diagonal rib, in this case, two ribs diverge from the support, representing the lines of intersection of the inserted surface of the sail-closed vault (chamfer) with the remaining reduced formwork of the main cross vault. It is possible to increase the surface of the sail-closed vault in such a way that it becomes the main element, and the formwork of the cross vault becomes secondary (this will be discussed in the chapter on the closed vault).
With a large rounding of the diagonal edge of the cross vault with a radius equal to half the side of the plan, the cross vault turns into a fan vault (Fig. 4, Fig. 185). Thus, the introduction of the surfaces of other vaults into the limits of the diagonal rib greatly changes the basic form of the cross vault and even destroys it.
In these examples, we see a number of intermediate and mixed forms of vaults and observe the transition from one form to another.
In the history of architecture, one can find many examples of interesting and beautiful combinations, in which elements of different vaults are combined into one new vaulted covering. Baroque gave rise to especially complex combined forms of vaults, where even combinations of a cross vault with a dome in the form of a ceiling are found.
In all deviations from the main form of the cross vault, the architect's desire to get a flat figure in the shelyga, in the form of a ceiling, suitable for painting and sculptural images, is noticeable. Already in the Roman vault of the tomb of the Pankratiev brothers (Fig. 200), the shelyga is occupied by a square ceiling, which is part of the general pictorial geometric decoration of the vault. In the vaults of the Renaissance, a round medallion is more often found in the shelyge according to scheme 1 in Fig. 186, for example, in the vestibule of the Palazzo Vecchio (Fig. 187) or in the Elliodoro station according to the project
Peruzzi (Fig. 206). In Villa Madama, the ceiling has the shape of a square with concave sides, according to scheme 3 in Fig. 186 (see also fig. 212). Such medallions use only a more or less flat area of the vault and have little organic connection with its shape. On fig. 186 shows a variety of compositions in which the shape of the ceiling is organically linked with the structure and shape of the formwork and the ribs of the vault. One follows from the other, and all taken together gives a holistic concept.
Fig. 5 repeats the variant already known to us with a blunted edge of the vault (Fig. 3 Fig. 185). The plafond in the form of a square, rotated by 45° to the axis of the vault, is clearly connected with the faces of the blunt rib. The vault plan (see side) can be read in two ways. If we single out four strippings, the rest of the body of the vault can be considered a sail-closed vault; if we take the three triangular faces at the support as one whole, for a faceted funnel, we can call the vault a faceted fan (compare with Fig. 4, Fig. 185). The size of the square ceiling can be increased arbitrarily. In monuments, this form is rare.
Having broken the diagonal edge of the vault shown in Fig. 5 fig. 186, we get a vault with an octagonal ceiling (see Fig. 4 Fig. 186). In the direction from the corner of the octahedron to the support, a third diagonal edge will appear. This middle edge will, however, flow in, as in a closed vault (see below), and the two extreme ones will protrude (inside the vault). In the cross vault of the mosque in Ephesus (Fig. 188), these protruding ribs and the middle diagonal, flowing in, are clearly visible. This small vault, with a span of 2-3 m, is very skillfully made of solid blocks of white marble. The octagonal lock of the vault is made in the form of a compressed ring-drum, topped with an ornamented domed slab.
The inexhaustible fantasy of the East enriched the shape of the vaults in Ephesus with an additional detail, which gave the vault the character of a faceted crystal (Fig. 7, Fig. 186 - Mohammed el-Gauli Mosque in Cairo). The architect introduced small rhombic medallions into the pile of formworks, thanks to which additional ribs were obtained. The kink-fold of the diagonal rib was extended to the side faces of the strippings. The result was a new form of folded cross vault with seven ribs and three folds. On fig. 189 gives a plan and a section of such a vault in Okella Kajt-Bai. The seams of masonry shown in the plan give a clear idea of the folded surface of the vault. The flat recessed plafond is adorned with a stalactite pattern. The same form of vault can be interpreted as a fan vault with a folded funnel (see plans), especially if we take into account the absence of the main through diagonal rib and a round depression in the shelyga. Such vaulted forms of the cross vault, as we will see below, are found in the Gothic.
A variant with an octagonal ceiling located along the axes of the arch is also possible (Fig. 8, Fig. 186). The formwork strips, in accordance with the face of the octagonal ceiling, received small square ceilings. This form is most consistent with the calm and clear interpretation of the surface of the vault in the Renaissance. Most of the Renaissance decorations of a simple cross vault analyzed below have five medallions - one in the center and four in the row of strippings.
Special shapes of shades can be deduced from the constructive solutions of formwork forms and from their execution from stone. If the filling of the formwork with masonry is carried out normally to the bisector of the corner (Fig. 6, Fig. 186), according to the English Gothic method, then by bringing the masonry to the stripping sheath, we will get a hole in the middle of the arch in the form of a four-pointed star. Making this hole in the form of a ceiling, we will get a new form of it, closely related to the lines of laying the formwork.
Fig. 9 fig. 186 represents a cross vault with swollen spheroidal formwork. This Byzantine technique of a vaulted arch along a curve has already been analyzed by us above (Fig. 183). As is known, the section of such a vault by a horizontal plane gives in plan a figure in the form of a quatrefoil; this form of ceiling is very interesting for painting. In the decoration of formworks, the surface of rotation can be emphasized with horizontal lines of parallels. One of the possible decor options, with the division of the middle four-bladed ceiling into four sectors, is shown in Fig. 4 fig.214. The same motif was used in the decoration of the cross vault of the Frugg Chapel (XVI century, fig. 228); the ribbed gothic pattern and details of the arches were created during the transitional period and already bring in the style elements of the renaissance.
III. CASSONS ON THE CROSS Vault
Any caisson of strict rhythmic construction from geometric figures fits freely on the cylindrical surface of the vault, as well as on a flat ceiling. It would seem that the surface of the cylindrical formwork also allows the free use of the caisson. However, the development of the stripping surface gives diagonal ribs in the form of curved lines OA (see Fig. 195), the adjacency to which the geometric patterns of the caisson can never be correct, but is always random. By adjusting the pattern and additional inserts, it is partly possible to mask the defect of the junction, but even with this assumption, the inflection of the geometric figures of the caisson through the diagonal rib gives an unacceptable solution, with deep depressions and ruptures of the diagonal rib, as can be seen on the vault of the Basilica of Maxentius and Constantine (Fig. 180) .
Roman architects did not consider it necessary to decorate the diagonal rib, which was designed constructively in the form of a brick frame in the concrete body of the vault (Fig. 190, Durma reconstruction). The complex octagonal caisson gives an ugly joint at the edge with the introduction of random hexagonal and round caisson figures. This is especially clearly seen in the reconstruction of the middle heel of the vault of the Basilica of Maxentius and Konstantin (Fig. 191, left; reconstruction of Ronchevsky), where D is the preserved part of the heel, and the reconstructed part is shown in dotted lines.
 |
| Rice. 190. Details of Roman concrete cross vaults with a brick frame |
 |
| Rice. 191. Caisson decoration of the heels of the cross vault |
 |
| Rice. 192. Interior of the baths of Diocletian (reconstruction of Auer) |
 |
| Rice. 193. Interior of the baths of Caracalla (reconstruction of Tirsh) |
 |
| Rice. 194. Interior of the lobby of Pennsylvania Station in New York |
 |
| Rice. 195. Schemes of coffered decor on the cross vault |
 |
| Rice. 196. Decor of the heel of the vault of the term of Hadrian's villa (according to Ronchevsky) |
 |
| Rice. 197. Development of the coffered decoration of the cross vault of the thermae of Hadrian's Villa |
 |
| Rice. 198. Circle-mesh wooden cross vault |
 |
| Rice. 199. Cross vault of mixed masonry in the Louvre |
The same picture of the joint of the caisson on the rib is given in the reconstruction of the heel of the vault, the terms of Diocletian (Fig. 191, right; reconstruction of Paulinus); here the second row of octagonal caissons ugly cuts into the rib of the vault.
Other reconstructions of the coffered décor of the cross vaults of Roman thermae do not provide a sweep of the vault, but perspective drawings of the interiors, and the caisson joint on the edge is clearly shown incorrectly with an artificial adjustment of patterns. So, on the reconstruction of the tepidarium, or the so-called "Cela media", performed by Blue (Fig. 179), the lenticular-shaped caissons on the rib of the vault are, of course, false in shape and cannot be the same along the entire length of the rib.
Auer during the reconstruction of the tepidarium of the baths of Diocletian1 (Fig. 192)
The surviving parts of the baths of Diocletian were rebuilt by Michelangelo into the church of Santa Maria degli Angeli.
applied the most complex false-constructive caisson found in late Renaissance plafonds1. The use by the Romans of such a caisson, unsuitable for a cross vault, seems unlikely, especially since another reenactor, Paulinus, gives a completely different version of the decoration of the same baths. The joint of the coffered pattern on the edge was also drawn by Auer arbitrarily, not in accordance with the correct construction.
Finally, we also note the reconstruction of the Baths of Caracalla, made by Thiersch (Fig. 193). Here the design of the caisson has been restored, the remains of which, in the form of a preserved piece of stucco decoration in the heel of the vault, are found in the ruins of the baths of Hadrian's villa in Tivoli (Fig. 196). As will be shown below, Thiersch also makes a mistake in adjoining the coffered pattern to the edge of the vault.
In connection with the unsuccessful solutions of the coffered decoration of the cross vault in the proposed structures, it is interesting to note the mistakes of modern architects.
The lobby of the Pennsylvania Station in New York (Fig. 194) is an almost exact copy of the Basilica of Maxentius and Constantine, only the radius of the curved strippings is somewhat less than the radius of the main arch. As a result, the rows of caissons of the formwork and the main vault do not coincide at all with each other on the edge of the vaults, and the latter takes the form of a thin, ugly partition between them. A greater architectural mess is hard to imagine. The Americans distorted the Roman basilica and illiterately decided on the coffered decoration of the vault.
So, a number of unsuccessful attempts to reconstruct the coffered decor of the cross vaults and the mistakes made at the same time confirm the difficulty of adjoining the geometric caisson to the edge of the vault, which we indicated. By accurately constructing the caisson on the development of the formwork surface, it is necessary to find out all the defects in the adjoining of the caisson to the rib and give possible architecturally correct solutions.
For analysis, let's take that decorative rectangular caisson, traces of which remained on the vault of Hadrian's baths in Tivoli.
In FIG. 1 fig. 195, on the left side, a breakdown of the caisson is made by segments a and b along the stripping arc and its horizontal projection is drawn on the cross vault. On the right side, the formwork has been unfolded. The wall arc of the stripping will straighten into the segment CA. The diagonal edge of the arch will give in the sweep the OKA curve, the points of which 11, 21, 31, 41, 51 are obtained by straightening the corresponding arcs of the caisson. On the unfolding of the formwork in the form of a triangular cutout of the OCA, we apply the correct drawing of the caisson, setting aside the dimensions of its squares a and b.
At points 11 and 21, the diagonal rib OA in the development deviates so slightly from its horizontal projection OB that the corners of the caisson almost lie on the development of the diagonal rib. Points 31 and 41 of the rib in the development move to the right and do not coincide with the corners of the caisson 3 and 4. The angle 5 of the caisson has moved away from the development of the diagonal edge of the curve OA already by a significant value 5-51. The vertical face of the caisson 5-5 meets the unfolding of the diagonal rib at point K. Minor discrepancies between points 3 and 4 of the corners of the caissons and points 31 and 41 of the ribs of the arch are hardly noticeable, and they can always be adjusted. But the faces of 5-5 caissons lying in different strippings cannot coincide and form an incoming angle K on the diagonal edge and a caisson in the form of a hook around it (see the upper left corner of Fig. 1, Fig. 195).
Ronchevsky's sketches from the surviving remains of the caisson (Fig. 196) give exactly the same picture of the reentrant angle K, which we also obtained with the correct delineation of the caisson on the development of the stripping. On the development of this decor, also executed by Ronchevsky (Fig. 197), we see the same reentrant angle K
In FIG. 2 fig. 195 reversed construction. On the horizontal projection of the vault (in the left part of the drawing), a regular grid of caissons is applied, the corners of which lie on the projection of the diagonal rib. On the development of the stripping (on the right side of the drawing), the angles of the caissons also remained, of course, on the development of the rib, but the width of the caissons a1, a2: etc., as well as b and b1, would increase in the direction towards point B. Thus , the coincidence of the corners of the caisson with the diagonal rib, which was obtained in the reconstruction of Thiersch (Fig. 193), is possible only with such caissons, the dimensions of which increase towards the support, which is unacceptable.
It is clear that the solution of the caisson on the vault of Hadrian's Villa (Fig. 196-197) must be recognized as the only possible and correct one. Random forms of caissons are inevitably obtained in the supporting parts of the vault, as a result of the intersection of cylindrical decors, and to a certain extent violate the integrity of the rib. To avoid accidental joining of caissons on the edge, only one way is possible - the use of such a caisson, the pattern of which includes a diagonal edge of the vault. Such is the constructive caisson of an oblique grid, used by the Romans for the domed surface of the apses of the temple of Venus and Roma (Fig. 14-15). Fig. 3 and 4 fig. 195 show the construction of such a caisson on the surface of the cross vault.
In FIG. 3 on the right is a sweep of the formwork, and six parts of the arc (1, 2, 3, 4, 5, 6) are deposited on the SA line. At points 1 and 2, the corners of the caissons almost lie on the development OA of the rib, but point 3 already departs significantly from the corner a of the caisson 2. From point 3 to the heel A, no caisson can fit.
On the left side of Fig. 3 shows a dotted line projection of the grid of the caisson, applied to the extended cylindrical surface of the stripping. On the surface of the cross vault, the grid will have to stop at caisson 2, near point 3 of the diagonal rib, and caisson 2 will have the shape of a somewhat distorted square, since point a (see the development) will have to be pulled up to point 3 of the diagonal rib. In FIG. 4 shows the reverse construction. On the projection (on the left) the correct grid is applied, on the scan (on the right) the squares of the caissons turned out to be elongated to the support of the form.
In FIG. 9 shows an axonometric image of a slanting caisson on a cross vault. The grid of the caisson ends (as in the developments, Fig. 3) at point 3 of the diagonal rib. The flatter the vault, the better the mesh of the caisson fits on the surface of the vault stripping. With an oblique coffered mesh, the diagonal edge of the vault receives a completely correct architectural and constructive meaning of the main working element of the vault, which carries the mesh frames of the formwork. The decorative pattern in this case is an organized constructive frame that performs the synthesis of decor and construction, which is inherent in the best examples of Gothic vaults.
The latest systems of circular-mesh wooden cross vaults, shown in fig. 198, also serve as examples of decorative and constructive frames that meet the challenges of modern architecture. All other coffered decorations of the cross vault, which do not take into account the diagonal rib and arbitrarily bend their pattern through it, must be recognized as false decorative.
A special place is occupied by the decor of the cross vault, which reproduces the normal cutting of stones and the laying of them. In FIG. 8 fig. 195 shows the usual masonry of hooked stones K and L, with a lock-plafond in the form of a cross. An example of such masonry made of mixed materials - hewn stone and brick - is the arches of one of the halls (sale de Manege) in the Louvre in Paris, built by L. Visconti in 1852-1857. (Fig. 199). In FIG. 7 Fig. 195 shows another masonry - from hexagonal stones K and L with an octagonal ceiling M convenient for painting. Both solutions can serve as a good motive for artistic processing.
As a result of our analysis, we have to state that the rhythmic coffered pattern in the broad sense of the word (as we understood it in the chapter on the barrel vault) cannot be successfully deployed on the surface of the cross vault.
IV. CROSS Vaults of Ancient Rome and the Renaissance
The principle of the rhythmic pattern in the form of the so-called "endless field" fully corresponds to the extended monotonous surface of the cylinder, but is in conflict with the surface of the cross vault, dissected by diagonal ribs and consisting of four segments of the cylinder. The correct construction of the decor of the cross vault should be subordinated to its diagonal ribs. The composition should group all decorative elements around a central spot (square or round in the vault), building the entire scheme crosswise along the axes and diagonals of the plan. All framing and secondary plots should concentrically cover the central motif. Thus, a single closed "diagonal" composition will be created, in harmony with the forms of the vault and revealing them.
If the Romans did not find special decorative forms for the giant cross vaults of their terms and were content with random intersections of cylindrical decors, then among the cross vaults of Roman tombs and small vaults of the terms we find a number of the best decors built on the principle of a single diagonal composition. Among them is the decor of a small cross vault - the tomb of the Pankratiev brothers on the Latin road, near Rome (Fig. 200, 201). The cross vault of a square plan (4.28x4.28 m) occupies the middle of the covering; side arches with a width of 0.6 m with a wicker ornament coincide with the surface of the formwork (a common technique of Roman architects with a rectangular plan). The vault is decorated with stucco frames of small relief, made with a special alprimo technique, i.e. stamping on the top wet layer of plaster. Frames and plafonds are filled with stucco ornaments and figurines, made by hand, and picturesque motifs.
 |
| Rice. 200. Decoration of the cross vault of the tomb of the Pankratiev brothers near Rome |
 |
| Rice. 201. Decor of the heel of the vault of the tomb of the Pankratiev brothers near Rome |
 |
| Rice. 202. Decor of the vaults of the term of Hadrian's villa (according to Cameron) |
 |
| Rice. 203. Reconstruction of the terms of Diocletian (according to Paulinus) |
 |
| Rice. 204. Heel of the cross vault of one of the halls of the Winter Palace in St. Petersburg |
 |
| Rice. 205. Decor of the cross vault of the stanza del Inchendio |
 |
| Rice. 206. Decor of the cross vault of the stanza del Elliodoro |
 |
| Rice. 207. Decor of the cross vault of the "Hall of Heroes" in the Munich Glyptothek |
 |
| Rice. 208. Decoration of the cross vault of the stanza della Senyatura |
 |
| Rice. 209. Interior of the stanza della Senyatura |
 |
| Rice. 210. Decoration of the cross vault of the chapel del Pallio in the Palazzo Cancellaria |
 |
| Rice. 211. Decor of the cross vault of the portal of Peter's Cathedral |
 |
| Rice. 212. Decoration of the cross vault of Villa Madama |
 |
| Rice. 213. Decoration of the cross vault of Villa Belcaro |
 |
| Rice. 214. Examples of the composition of decors of the cross vault |
 |
| Rice. 215. Cross vault of Amiens Cathedral |
The whole composition has a strictly geometric scheme along two axes. Successfully used all the flat places of the arch. In the middle of the shelyga there is a large square plafond, the strippings are marked with rectangular plafonds. All plafonds are framed with stucco frames that form a common pattern throughout the vault. The beginning of the diagonal rib, the support of the vault, is highlighted by a picturesque dark rhombus and a small sculptural figure (Fig. 201). With all the artistic merits of the decor composition, it should be noted, as a minus, that the significance of the diagonal rib is poorly expressed by decorative means.
In Cameron's work "Roman Baths" there are two decorations of the cross vaults of the baths of Hadrian's Villa (Fig. 202), built according to the same principle of symmetrical diagonal composition and made with the same stucco technique as the vault of the Pankratiev tomb (Fig. 200). One solution (at the bottom of the drawing) is very reminiscent of the decor of the Pankratiev tomb. It should only be noted that the middle plafond is very large for a cross vault and goes onto the curved surfaces of the formwork; T-shape stripping lights have lost their meaning. An attempt was made to emphasize the diagonal rib decoratively, albeit in a small area. The second solution (at the top of Fig. 202), with round medallions, is more interesting. (Note that round medallions in formwork will often be found in Renaissance decors.) The location of four round medallions on a sharp diagonal edge of the vault is unfortunate due to the fracture of the medallion, and also because of the rupture of the diagonal edge, from which only small pieces remain. A composition with eight round medallions is possible, as we will see below, with a sail vault.
Let us dwell on one more interpretation of the decoration of the cross vault, which does not have specific historical examples, but is given by Paulin in a sketch reconstruction of the baths of Diocletian (Fig. 203). This new form of décor, with five square plafonds arranged crosswise in the flat upper parts of the vault, is shown in scheme 6 of fig. 195. The corners of the vault, clearly distinguished by the decor, are interpreted as square cantilevered heels, laid out with an overlap of horizontal rows of masonry. The decor of the corner feet corresponds to the direction of the masonry joints.
The heel of the cross vault of one of the small halls of the Winter Palace in St. Petersburg (Fig. 204) gives a clear idea of this decor scheme. Despite the ugly stucco forms of the lower part, the entire heel gives the impression of a well-formed and strongly designed load-bearing element of the vault. Plafonds of strippings, in the form of an octagon with inscribed circles, are unsuccessful in form and pattern.
Having considered the meager remains of the Roman decorations of the cross vaults, which do not allow us to accurately establish the main line of development, we still have to note the slight accent of the diagonal rib in the decoration of the vault. The artists of ancient Rome decided the decor of the cross vault in most cases as an intersection of the decors of barrel vaults.
We consider the orientation of the decoration of the cross vault on its diagonal to be the most truthful and organic solution.
The above two sets of terms of Hadrian already give something positive in this direction, and we will see the echoes of these techniques in the decorations of the Renaissance. The Renaissance did not blindly follow antiquity, it searched for its own ways in the decoration of the vaults, showing its own taste. Using the techniques of stucco decorations with tint and picturesque decorations - with molding, common in Roman practice, the masters of the Renaissance were looking, however, for surfaces for the free deployment of frescoes. Rejecting the caisson in the decoration of the cross vault1, they gave a number of beautiful compositions in a free decorative interpretation, adhering to the geometric diagonal scheme.
The theoretician of the Renaissance, the architect Leon-Battista Alberti (XV century), does not at all dwell on the problem of vault decoration and does not give a theory of its composition. In chapter 2 of book VII of Alberti's treatise, there are only the following lines: “The vault also has its decorations. Among the ancients, the same decorations that jewelers made on sacrificial bowls were also used by architects to decorate spherical vaults. And those decorations that were made on fabrics were imitated in cylindrical and cross vaults. Therefore, one can see quadrangular, octagonal and similar figures, located in the arch at equal angles and along even lines, in different rays and circles, so that nothing is more charming. This also includes those decorations of the vaults, which, without a doubt, are the most worthy, namely the caissons, which we see everywhere, both in other places and in the Pantheon. Next comes a story about the construction of boxes for caissons made of bricks on clay.
Let us turn to the analysis of the best solutions for the decor of the Renaissance.
Evaluation of the artistic painting of the vaults is not included in our task. The analysis will concern exclusively the architectonic side of the decor composition, which is the main one in the work of the architect.
1 Quite deliberately, the masters of the Renaissance did not follow the example of Rome in this case and did not repeat the unsuccessful coffered decors of the Basilica of Constantine and the baths of Diocletian and Caracalla.
2 Color scenic decoration in architecture is the most difficult moment. The main techniques and rules of past centuries are almost lost.
As the main solutions of the cross vault, we take the ceilings of three stations (hall) of the Vatican: del Incendio, della Senyatura, del El Liodoro.
In the stanza del Inchendio, the cross vault was painted by Pietro Perugino (1446-1556); Raphael, performing his frescoes on the walls, completely preserved the work of his teacher. The decor of the Perugino arch is simple and clear to the point of naivety. The diagonal ribs are clearly decorated with rods, the triangles of the formwork are filled with the favorite form of a round medallion (Raphael, creating his famous wall frescoes, apparently partially preserved the decor of Sodom), which was often used also on triangular sails.
The decor of the same scheme was repeated by Perugino on the vault in Cambio in Perugia, but there seven planets are depicted in the medallions.
We find the same decor scheme in French Gothic. The star vault of the chapel of the castle of Huaron has five large round locks carved from stone in the form of medallions (Fig. 232).
The main feature of this scheme is the absence of the middle ceiling. This is typical for a cross vault, as it does not destroy the through diagonal ribs. This decor decision must be recognized as the only correct one, worthy of imitation.
In the stanza del Elliodoro, the great Sienese master Baldassare Peruzzi (1481-1537), a contemporary of Raphael, created a new original decor (Fig. 206). Diagonal ribs are clearly marked by a picturesque ornamental ribbon. The triangular strippings are interrupted by an annular belt of the same pattern as the diagonal ribs. The sector-shaped frames thus formed provide a lot of space on which to unfold large plot pictures. According to Burkhardt, Raphael painted the main background of the four paintings in a blue tone, which gave the paintings considerable lightness. Despite the participation of two great painters in the creation of this decor, the framing ring belt remains the weak point of the composition: it is completely inconsistent with the shape of the cross vault, artificially transferred here from the sail vault, where the circle separates the sails from the skufia. On the edge of the vault, the ring gives a break, which looks especially unpleasant if the vault is viewed from an angle in perspective. As already mentioned earlier, a four-blade belt is more appropriate here, located along the parallels of swollen spherical strippings (see Fig. 4, Fig. 214).
For comparison, we present the decoration of the cross vault of the "Hall of Heroes" in the Munich Glyptothek, executed by Cornelius (Fig. 207). With a decor scheme completely similar to that of Peruzzi, Cornelius divided the painting of the formwork into two plots, and the entire decor into sections, depriving it of the unity of the composition. We find the most complex composition in the decor of the cross vault of the stanza della Senyatura, executed by Giovanni Sodoma (1477 - 1550) in 1511 (Fig. 208)1. In this composition, four round medallions are conveniently located in the strips of formwork, similar to the decor of Perugino in the stanza del Incendio. However, despite the presence of ribs, large square frames with pictorial scenes were introduced, passing along the ribs of the vault, for which the latter had to be hewn and rounded (Fig. 209). Such violence against the form of the vault cannot be justified in any way. In addition to this main drawback, the whole grid of stucco frames is a random, unorganized accumulation of various geometric shapes adjacent to each other at the corners and squeezed one by the other (especially the corner squares). No skill of a brilliant brush could save the false composition of Sodom.
The same false decor scheme was carried out on the cross vault of the chapel del Pallio in the Palazzo Cancellaria in Rome (Fig. 210). The authors - Perino del Vaga, a student of Raphael, and Federico Zuccheri (1542-1609) - tried to link the frame figures more organically than in the Senyatura stanza. Picturesque paintings in frames are placed only in demouldings. Narrow, long frames on diagonal ribs are filled with stucco figured plots and, together with a round middle shield, form a four-pointed star. This artificial, dry organization of the scheme introduces some order, but cannot create a truly artistic composition. Due to the fact that the rounded ribs are knocked down, it is difficult to read even the basic shape of the cross vault in the picture.
Somewhat later (1619), the baroque master G. B. Ricci from Novarra decorates the cross vault of the portico of Peter's Cathedral in Rome (Fig. 211). Its scheme clearly reveals the shape of the cross vault, the medallions are correctly located in the formwork. There is a stucco coat of arms in the shelyge. Only the dryness of the forms and the baroque broken frames can be considered among the shortcomings of this decor.
Exceptional skill in the decoration of the vaults was shown by the students of Raphael in the Roman villa Madama, built under the direction of Giulio Romano (1492-1546) according to the project of Raphael. Another student of Raphael, Giovanni da Udine, took part in the picturesque decoration of the premises. The decoration of the villa continued for another five years after the death of Raphael, from 1520 to 1525. The decoration of domes and niches will be discussed below, but here we will dwell on the decor of the cross vault of the villa's salon (Fig. 212). With exceptional clarity, the artist emphasized with painting not only the form, but also the meaning of the elements of the vault, without depriving the composition of its general decorative value. The vault supports are marked with four ornamented heel stones, similar to the heels of the decor of the vaulted vault of the Pankratiev tomb (Fig. 201). The castle in the vault is given in the form of a cruciform stone. Heavily decorated and painted diagonal ribs, like tense struts, connect the castle to the heels, forming a cruciform frame. The body of the formwork between the ribs is filled with light arabesques, in imitation of a stretched awning (velum); the lines of the arabesques follow the main lines of the vault and emphasize the character of the surface. The oval medallions are large, well placed and well connected with the arabesques. Interior decor of Villa Ma-
the lady is a subtle artistic study of Perugino's simple and valuable scheme in the Stanza del Inchendio (Fig. 205).
Finishing the review of the most original decorations of the Renaissance, we note the clean, lyrical decoration of the cross vault in the Villa Belcaro near Siena, built by Peruzzi (Fig. 213). The whole composition is made in the style of "grotesque" and depicts a light trellis dome of a garden arbor with climbing plants and fluttering birds. The design of the trellis strictly adheres to the main lines of the arch, the diagonal ribs are highlighted; the demouldings are filled with medallions in the form of stretched scarves with mythological scenes. This charming, joyful decoration sins only in that, by creating the illusion of open space with its transparency, it seems to abolish the vault.
In addition to the best works of the Renaissance masters cited, we give in fig. 214 a number of compositions performed on the basis of the material analyzed by us.
In FIG. 1, a constructive and decorative coffered pattern, built according to the scheme of the coffered plafond of the palace in Pastrana, was successfully laid on the surface of four formworks (Fig. 72). The pattern is crosswise inscribed along the axes of the formwork. The diagonal rib is included in the main pattern of the pattern. The roof supports, due to their limited dimensions, must be machined as independent parts of the roof, like exhaust heels.
In FIG. 2, the decor of the cylindrical vault of the Saint Bernardo chapel in the Palazzo Vecchio is used (Fig. 110). The frame system of this décor is well placed in the formwork strips. The corners of the cross vault are processed with horizontal rods, corresponding to the rows of masonry. The diagonal rib was not included in the overall decor pattern, which is a defect in this scheme.
The motif of the composition shown in Fig. 3, taken from the coffered barrel vault of the Palazzo Reale in Venice (Fig. 111). The diagonal rib, as in Fig. 1, is an element of the general decor pattern and is successfully combined with octagonal shades.
Fig. 4 depicts a variant of the decoration made by Peruzzi in the stanza del Elliodoro (Fig. 206). The general round belt, with which Peruzzi unsuccessfully covered the entire vault, was replaced by a four-lobed belt corresponding to four swollen spherical formworks and located along their parallels. The decor of the vault feet is also coordinated with the rows of masonry of spherical formwork.
In Roman buildings, as in the Renaissance, the surface of the cross vault was treated with plaster and used for pictorial decoration. The artist, not constrained by the construction, by virtue of creative intuition, emphasized the working elements of the vault with pictorial means and modeling. We noted this as a healthy, exemplary example.
V. GOTHIC CROSS Vaults
It is necessary to learn the synthesis of construction and decor on those samples in which it is not an accidental artistic motif, but an integral organic element of the composition, namely, on the works of Gothic masters, on the Gothic cross vault.
Representing a perfect structure made of stone (Stone Gothic vaults of the 12th and 13th centuries were already approaching modern reinforced concrete vaults in thickness. With ribs of 40-50 cm, the stripping of the vaults had a thickness of only 10 cm), built on the principle of a visible, revealed frame, and at the same time an organic constructive stone decor, the Gothic vault meets the main modern requirements of the composition. The flexible, elastic frame system, within certain limits, allows the architect, in search of the best decor, to direct the ribs (ribs) of the vault at his discretion.
When analyzing Gothic vaults, we will leave aside the question of cutting stones and the method of laying, as having no significance for the present, and we will indicate the lines (seams) of the masonry only to visually clarify the shape of the surface and the direction of the acting forces.
The basis of the Gothic cross vault, as you know, is an active frame that carries the filling in the form of small vaults. The Romans hid the same brick frame in the mass of a cast concrete vault and did not give it a decorative relief design.
The constructive method of constructing an independent frame was completed already in the 12th century. in early French Gothic (Saint Denis Cathedral, 1140). The vault of the Amiens Cathedral (1218) is an example of the same design (Fig. 215).
The frame of the Gothic vault is built according to clear and practical techniques and methods of stone art. The diagonal elliptical rib of Roman cross vaults requires stones of various shapes and difficult to work with. In Gothic, it is replaced by a simple semi-circular rib, made of identical stones. In later buildings of the 13th century, for example, in Reims, the diagonal arch has a lancet elevated form. The formworks were laid out as independent small vaults, based on the ribs of the main vault. The latter, to give rigidity to the frame, was laid out from long, durable stones with a small number of seams. Stripping, on the contrary, was made of small, light limestone stones; The swollen spheroidal shape, which reduces the thickness, also contributed to the facilitation of stripping.
 |
| Rice. 216. Star vault of the cathedral in Beverley (England) |
 |
| Rice. 217. Schemes of Gothic vaults |
 |
| Rice. 218. Cross vaults of the middle nave of the cathedral in Exeter (England) |
 |
| Rice. 219. Various forms of cross vaults and their stripping |
 |
| Rice. 220. Schemes of mesh vaults |
 |
| Rice. 221. So-called honeycomb vaults (Wabbengewolbe) |
 |
| Rice. 222. Schemes of working ribs of Gothic vaults |
 |
| Rice. 223. Schemes of Spanish Gothic vaults |
 |
| Rice. 224. Star vault of Worcester Cathedral |
 |
| Rice. 225. Cross vault of Christchurch Church |
 |
| Rice. 226. Cross vault of the church in Warwick |
| Rice. 227. Cross vault of the church in Vulpit |
 |
| Rice. 228. Cross vault of the Frugg Chapel |
 |
| Rice. 229. Interior of the Vladislav Hall in the Prague Palace |
 |
| Rice. 230. Interior of Anna's Church in Annaberg |
 |
| Rice. 231. Decor scheme for the Vladislav Hall in the Prague Palace |
 |
| Rice. 232. Decor of the vault of the chapel in the castle of Huaron in France |
Due to the difficulty of laying large formworks, additional working ribs are introduced, the so-called tiercerons (in French tiercerons, in German Dienste - helpers), directed, like the main diagonal rib, to the support. In shelygi and in secondary areas, craftsmen began to introduce horizontal, long, even stones, called lierns, to simplify and beauty the joint of stones. With the development of the frame, the piers also began to serve as supports for the upper ends of the tierserons and received a curved vaulted outline.
Such a system received its full constructive development in the so-called star cross vault, first used at the intersection of the naves of the Amiens Cathedral (1220-1288). This vault, in its form and construction, represents a truly complete architectural concept.
On fig. 216 shows such a star vault of the cathedral in Beverley, in England; LB and LD, MB and MA, etc. - tierserons, El, EH, EF and EG - lierna. The British call such a system Complex quadripartite Yaults.
The main feature of the cross Gothic vault is a clearly defined profiled diagonal rib.
On fig. 217 shows the most characteristic Gothic vaults. The bottom row shows cross vaults: Fig. 4 - ordinary stellate arch, fig. 5 - arch of a more complex shape, in the form of an eight-pointed star, fig. 6 - vault, in the middle of which an octahedron is formed by the intersection of ribs (tiercerons). In all three forms of the patterned frame, the main diagonal ribs of the cross vault are clearly distinguished.
All three vaults in the top row of Fig. 217 are similar to the lower cross ones, but they do not have the diagonal ribs of the cross vault and are other forms called mesh. Fig. 1 represents the basic shape of the mesh vault; here, in place of the diagonal ribs, four cylindrical sails appear. The vault shown in Fig. 2, even closer in outline to the cross frame (Fig. 5), but the diagonal ribs are also absent. Finally, in FIG. 3, the vault also has a diagonal rib, but it does not pass through the shelyga and rests against the ring. If the three ribs coming out of the supports (diagonal and two tierserons) are outlined with the same radius and form a regular funnel, then in this case a new form of fan arch is obtained.
These examples show how free the architect is in creating spatial schemes. The desire to enrich these forms of frameworks was expressed primarily in an increase in the number of tiercerons.
So, in Exeter Cathedral in England in 1270 (Fig. 218), the middle nave has a typical vault with a large number of tierserons converging on a support into one bundle. 13 ribs converge on the support, of which there are two diagonal CB and CZ, one transverse CD, two window ribs - SA and SU and eight tierserons. This bundle of ribs forms a square basket on a support, typical of English Gothic.
With such a number and arrangement of tiercerons, the value of the diagonal rib decreases, since the tiercerons perform its own function and remove part of the load from it. At one glance at the square ribbed basket of the vault of Exeter Cathedral, where all the ribs (ribs) are almost equivalent, the work of entire sections of the cylindrical longitudinal vault, in the form of CEDY rhombuses, resting with their vertices C and D on the wall abutments, becomes clear. The meaning that the diagonal rib has in a simple groin vault is nullified here, and such a vault can hardly be called a full-fledged groin vault. Rather, it is a special kind of cylindrical vault, the efforts of which are directed by a bundle of ribs to individual reference points, instead of a solid heel. Thus, the mere presence of diagonal ribs does not yet create a normal groin vault if there are other uniform ribs.
Another case of changing the basic working scheme of the cross vault frame is given by the previously considered ploughing of it and the change in the forms of formwork. Even with a small sparging along a curve (see Byzantine vaults above), the vault transfers part of the effort to the side walls and removes the load from the diagonal ribs. With a strong sparging along a curve, when the surface of the vault becomes spheroidal or spherical (sail vault), the diagonal rib becomes flatter, almost disappears, and forces are transmitted along the meridians, in all directions.
Changing the cylindrical surface of formworks into a double curvature (spheroidal) surface also redistributes the forces in the ribs. The most commonly used forms of the surfaces of cross vaults and their strippings are compared in fig. 219 indicating the change in the work of the ribs.
In FIG. 1 shows the main scheme of the cross vault with cylindrical formwork. Acting forces in strippings, shown by arrows, are transferred to the diagonal ribs. The walls (side arches) are free from load.
In FIG. 2 rows of masonry strippings are laid out in the form of arches along the movable circles on the ribs, with the addition of each stone in place. This scheme gives an idea of the technical perfection of early French Gothic masonry, which is so brilliantly described by Viollet le Duc in his encyclopedia. With this system, the wall arches perceive the spacer and part of the weight of the formwork (see arrows in Fig. 219), and the diagonal ribs are unloaded.
In FIG. 3, swollen spheroidal formworks were used (in German - Bussige Carren). Formwork strips can be located
wives at an arbitrary point on the plan, closer or further from the center of the vault. The stripping sheaves can be higher than the vault shely (crossings of diagonal ribs). As in the dome, the pressure of the formwork is transmitted along the meridians (see arrows) in all directions, on the ribs, on the walls and to the vault support.
In FIG. 4, the vault, in addition to the diagonal ribs, has along the axes of the stripping, in their shell, working curved ribs AC and BD. The vault is divided
thus, into eight strippings, which can have an arbitrary shape - swollen, spheroidal, etc. The surface of each stripping can be divided into three small triangles (there will be only 24 strippings in the vault). At the same time, 16 ribs converge in the lock of the vault: 4 main diagonal, 4 lierna and 8 branches of tierseron. Thus, this system represents another example (the first is Exeter Cathedral) of a plurality of ribs, but already concentrated at the castle vault. In the supports, there are only 3 ribs each (diagonal and two tierserons), as in a normal stellate vault. The whole composition takes the form of an eight-pointed star.
In FIG. 5 the entire surface of the vault is so strongly vaulted that it turns into a sail vault. All ribs diverge from the shelyga, like meridians, and work like the ribs of the dome. The formworks either have an independent curvature and rest on the ribs, or coincide with the spherical surface of the vault. In the latter case, if the main shell of the vault is strong enough (in thickness), the ribs are purely decorative elements. The diagonal edge loses all meaning here. The pattern of the rib frame can take on more free forms.
Such a rib cage system without a diagonal working rib is known as a "net vault". The main forms of the German mesh vaults are shown in fig. 220; fig. 3 fig. 220 gives a plan of the mesh dome shown in FIG. 5 fig. 219 (figs. 1, 2 and 3 fig. 217 show mesh vaults in axonometric projection).
All the six forms mentioned in Fig. 220 (Fig. 1 - 6), have two working edges at the supports; the dotted lines complete the frame pattern. Fig. 6 fig. 219 represents a special decorative form of folded formwork, found in isolated cases in late Germanic Gothic and known as Zeilengewolbe (cellular vault).
This arch does not have an independent profiled rib frame - the sharp edges of the folds already form a fairly strong frame. Therefore, it would be more correct to call such a cross folded vault. This form of the vault is fully consistent with modern folded reinforced concrete structures and is easily performed in reinforced concrete. It is interesting to note the repetition of the same folded form of the cross vault in the architecture of Islam, for example, Okella Kait-Bai (fig. 7 fig. 186 and fig. 189).
A variety of folded vault with closed (blocked) folds in the form of rhombic faceted depressions (Fig. 221) - the vault of the bishop's castle Altenstein (late 15th - early 16th century) is called in German Wabbengewolbe (honeycomb vault). There is no pure cross vault form here. Twelve ribs, diverging in a star-like manner from the top of the vault, are laid on a spherical surface. It is rather a sailing ribbed vault with faceted pyramidal formwork.
Thus, any change in the shape of the cross vault or its stripping leads to a redistribution of effort. It is impossible to say anything about the work of the grid of ribs based on their location in the plan, without knowing the spatial shape of the vault.
In Gothic, the basic form of the cross vault is transformed in such a variety of ways that it is extremely difficult to give a clear definition of each of the derivative forms. It can be said that the pure geometric cylindrical form of the cross vault is almost never found in Gothic. All forms of Gothic vaults occupy an intermediate position between the cross and sail vaults. Therefore, the concept of the shape of the cross vault should be clarified and somewhat expanded. We will call a cross vault not one where there are diagonal ribs, but one in which the diagonal ribs make up the main working frame. We also agree that with swollen or spheroidal strippings (Bussige Curren), the vault does not lose the name of the cross, if the general ploughing does not remove the main load from the diagonal ribs.
In addition to working ribs (ribs), the surface of the cross vault is often saturated with a whole network of non-working ribs that make up decorative patterns and stars. Their directions do not coincide with the direction of the current efforts; in addition, they often have curvilinear shapes in plan. Being purely decorative elements of the frame, they can be easily distinguished from the general grid of vault ribs.
On fig. 222 an attempt was made to decipher the meaning of the edges. The thick line shows the working diagonal ribs, the thin solid line shows the auxiliary ribs (tiercerons), which are also involved in the work, and, finally, the dotted line shows non-working, purely decorative, shaping the pattern of the vault.
The vaults shown in Fig. 1 - 4, have, in addition to diagonal ribs, a pair of tierserons, working together with diagonal ones on a support. The dotted line shows the lierns connecting the points of mutual intersection of the tierserons. In FIG. 5-8, the vaults have only diagonal ribs and a grid of non-working lierns: only in the vault of FIG. 5 have one tierseron per support.
Finally, the bottom row (fig. 222, figs. 9-11) gives special figurative solutions in which the working diagonal ribs break off in the middle of the vault, occupied by a complex star pattern. Such a solution assumes a strong vaulting of the cross vault, turning its middle into a spheroidal surface. The diagonal rib, slightly protruding from the support, is lost in the middle spherical part of the vault (Compare the schemes of Fig. 222 with the schemes of mesh vaults (Fig. 220). Fig. 1 Fig. 220 differs from Fig. 1 Fig. 222 only in the absence of diagonal ribs. The same Fig. 6 Fig. 220 differs from Fig. 3 Fig. 222). Forms 9-11 are transitional to mesh vaults (Fig. 220). So, fig. 3 fig. 220 gives a frame diagram almost identical to FIG. 11 fig.222.
The architecture of Spain gives even richer forms of vaulted patterns. The Spanish masters were not inferior to the best architects of England, who created openwork lace fan vaults of the 14th and 15th centuries. This skill is so deeply rooted in the architecture of Spain that Gothic vaults are found in baroque temples of the 17th century. The complex pattern of crooked liernes was successfully combined with the bizarre forms of the Baroque style, and only the ribs of the vaults received a different profile.
The vault of the cathedral in Segovia (figs. 1 and 2, fig. 223) has diagonal ribs and tierserons, as in fig. 4 fig. 222; its pattern is decorative, curvilinear. In FIG. 3 fig. 223 shows the vault of the cathedral in Salamanca. A complex curvilinear pattern is woven into the normal grid of this star-shaped cross vault.
The solutions of the cross vault in English Gothic deserve special attention. The traditions of the French Angevin school were transferred to England and gave a number of special local solutions. In addition to the mentioned main stellate vault of the cathedral in Beverley (Fig. 216), we note the vault of the cathedral in Worcester, built in 1372 (Fig. 224).
Intersecting tierserons form an octagon in the roof of the vault (cf. fig. 4, fig. 222) with sculptural capstones at the intersection of the ribs.
The vault of the Christchurch church is interesting (Fig. 225). Each demoulding of it has a pair of incomplete tierserons AQ and AJ, BP and BO, etc.
Partial liernas HI and FG intersect in the shelyga. The star is outlined by additional ribs (depicted by a dotted line), or counter-liers LG, CM, N1, 10, etc. and liers KL, MN, OP, etc., connecting the ends of the tiercerons with diagonals. In addition to the peculiar pattern of the star, the vault is interesting for hanging supports, which were then fully developed in Oxford Cathedral and in the Henry VII Chapel in Westminster. In the vault under consideration, the outlines of the diagonal ribs and the tierserons are so close to each other that the result is not a square basket, typical of a cross vault, but rounded, as in a fan vault. The arch is a transitional form to the fan, brilliant examples of which are given by the already named arches of Oxford Cathedral and the chapel of Henry VII; however, strong diagonal ribs and weak tierserons make it possible to classify it as a cross vault.
The arch of the Church of Mary in Warwick, built in 1439 (Fig. 226) has the same grid of tierserons and lierns as the Christchurch vault. A significant difference is in the star-shaped keystones: in the arch of the vault, the asterisk-lock E has an eight-pointed shape, in strippings - a six-pointed one. Both of these vaults are examples of the strong development of the ribs in the 15th century.
In the church in Vulpit of the end of the 15th century (Fig. 227), the stellate vault has already reached the limit of enrichment and complexity of the framework. A bundle of four tier serons emerges from the corner supports near the diagonal rib.
Between them, additional spacers PVQ, PWS, TXU, etc. form a large eight-pointed star. At the end of the 15th century, the ribs lose their constructive significance and in the last years of the century, when fan vaults appear (see below), they turn into a surface decoration on the vault stones. In the last two examples of English cross vaults of the late 15th century, of particular interest is the development in the shelyge of the vault of a rich cruciform decorative spot that stands out from the general
background of working ribs. There is already a diagonal composition of the decorative pattern, but still made of straight rods.
A special place in the decoration of the vaults is occupied by the German Frugg Chapel of the 16th century (Fig. 228). Its vault is the same diagonal cruciform composition with a purely decorative plafond of curved arcs. In addition to a large four-petal pattern that captures the formwork, the same ceiling is inserted in the middle, within which the diagonal ribs are decorated with molding. The entire composition uses the four-lobed form of the ceiling, which we considered earlier (see fig. 4, fig. 214), and with its crooked ribs is close to the Spanish decorations shown in fig. 223. The Frugg Chapel, built in the transition to the Renaissance, contains new style elements not only in the decoration of the walls, but also in the decoration of the ceiling, where there is already a desire to create a Renaissance "diagonal" composition.
We will finish our review of Gothic vaults and their decor with two late Gothic monuments that are exceptional in terms of fantastic curvilinear forms of decoration. On fig. 229 shows the hall of Vladislav in the palace in Prague, built in 1486-1502. German master Riet; hall size 60x16 m. 230 shows the interior of the Church of Anna in Annaberg (Germany), built by Durbach and Büttingen in 1499-1520. In both cases, we encounter constructions and decor that are clearly contrary to logic. The drawing of the ribs of the Vladislav Hall was deliberately built from arcs of the same radius, with one opening of the compass (Fig. 231); all edges of double curvature and have no structural significance. The shape of the vault is mainly cylindrical with stripping, of an indefinite, crumpled character. The vault of the Church of Anna is the same, the ribs of which helically cover the pillars.
Finishing with this analysis of the construction of Gothic cross vaults, we note the main stages in the development of constructive techniques. The main period - High Classical Gothic - gives the most structurally perfect systems built on the full use of the frame. In the next period, the frame is enriched with figured false-constructive decorative ribs. The last period - late Germanic Gothic (Sondergotik) - completely ignores the main ribs of the frame, turning it into a purely decorative carpet pattern of ribs. The last remnants of the articulation of individual vaults are removed, and the roof of the hall churches (Hallenkirche) is transformed into a continuous vaulted ceiling on columns.
In the history of Gothic cross vaults, we see a series of successive concessions to decorativeness at the expense of constructive principles. The farther Gothic went from strict constructive logic, the more arbitrary the composition became to the detriment of artistic qualities. High and late Gothic created various and perfect variants of "canopy systems".
The supports of the arches, in the form of beams, extremely thin pillars (columns) directly pass into the ribs (ribs) of the arch. There are no capitals and even part of the cornice of ancient canopies; the support of the vault fuses with the rib. Through supports, passing through a series of horizontal articulations of the wall, create the verticalism of the Gothic. With thin supports and stone bindings of window openings, the wall becomes a transparent lattice. The canopy on pillars and arcades is the main element of the architecture, covering the space, while the walls are secondary.
In the late Gothic hall churches, with equal heights of the middle and side aisles, the canopies of the individual aisles seem to dissolve mutually, the boundaries of the individual vaults are erased, the latter grow together into a common canopy on free-standing pillars. The cross vault in this case gives the simplest and most logical merging of homogeneous vaults into a single stone ceiling along the columns.
VI. conclusions
It is necessary to draw practical conclusions for modern architectural thought from the analysis of the infinitely diverse surface forms, rib schemes, frame and spatial solutions of Gothic architecture.
Everyone admires the wonderful art of the Gothic, extensive studies are written about it, but the ways of using this wealth in new architecture have not been discovered. At the same time, all paths are open to the heritage of the Renaissance, and it is often used blindly, without proper criticism. In this regard, our final conclusions about the system of ceilings with cross vaults are based on a comparison of the forms and decorations of Gothic and Renaissance vaults; such a method contributes, in our opinion, to the identification of organic, truthful, truly beautiful elements in the work of the Gothic and Renaissance, which retain their value for the present.
1. Of the three modern varieties of vaults - ribbed frame vaults, solid smooth shells and folded - in Gothic we find mainly ribbed frame vaults or with stiffeners and rarely folded. In the Renaissance, almost exclusively smooth solid vaults are used.
2. Cross vaults of the Renaissance are poor in form; dominated by a regular cylindrical shape. Gothic cross vaults reveal an exceptional wealth of forms and are a valuable material for modern shaping.
3. The cross vaults of the Renaissance, erected in brick, have clothes made of smooth plaster, requiring a picturesque decor; here, therefore, three production processes are needed. Gothic arches are created by the hands of a master architect from hewn stones without picturesque decor; here everything is reduced to a single creative process. In modern construction conditions, vaults can be made in the following forms:
a) solid reinforced concrete vaults-shells, requiring, like the vaults of the classics and the Renaissance, stucco clothing, low-relief stucco decoration (antique al primo) and painting;
b) ribbed vaults with a reinforced concrete ribbed frame and filling the strippings with any other material (The vault of the hall of the Kazansky railway station restaurant in Moscow was made by the author of this book (Fig. 359)) - brick, majolica, finished decorative boards, etc .;
c) ribbed vaults, but based on the basic positive principles of Gothic masonry - from hewn stone or artificial stone with any surface texture.
In cases b) and c), the ribs, as working elements of the structural frame, must be strongly profiled.
4. In the Renaissance, the decoration of the vault by modeling and painting was often the main task of the artist-architect; the decorator and the painter did not always understand the forms of the vault and often did not consider it necessary to reveal them with pictorial means. In Gothic, decor and construction represent a single compositional whole in the work of the master. Modern architecture must continue the development of the brilliant decoration of the Renaissance, both artistically and technically, but with strict observance of the basic principle of the Gothic, i.e., the synthesis of decoration and construction.
5. Due to the difficult linkage of the geometric coffered pattern with the shape of the cross vault, the Renaissance did not use it for decoration. Gothic, using an oblique grid of ribs in the form of a caisson for a barrel vault, did not transfer it to the cross vault. With the current state of technology, this problem can be solved. Following the example of a wooden circular mesh cross vault (Fig. 198), it is possible to implement a coffered structure in reinforced concrete and metal. Work in this direction will give new, modern forms of the cross vault.
6. The Renaissance developed on the surface of the cross vault the form of a diagonal composition of decor with picturesque medallions along the axes of the strippings and a ceiling in the center of the vault, weakly revealing diagonal ribs (A rare example of a “diagonal composition” on a cross Gothic vault with openwork medallions carved from stone is given by the vault of the castle chapel Huaron in France (Fig. 232)).
In Gothic, diagonal ribs and auxiliary tierserons are always included in the decor composition, in which the middle ceiling is not feasible or is in its infancy (In German Gothic: the decors shown in Fig. 9-11 Fig. 222; the ceiling of the Frugg Chapel Fig. 228, in English Gothic - the arches of churches in Worcester Fig. 224, Vulpit Fig. 227 and Warwick Fig. 226). The multiplicity of ribs in English Gothic (See Exeter Cathedral, Fig. 218) destroys the significance of stripping and leads to an understanding of space that is alien to us.
The modern composition of the decor should reveal the main diagonal ribs, avoiding the multiplicity of false ribs (lierns). When decorating diagonal ribs, it is more correct to develop plafonds in strippings, avoiding the middle plafond in the shelyga. Basically, with a clear expression of the form of the cross vault, it is necessary to use the strength and variety of the picturesque decor of the Renaissance (an example is the decor of the vault of Villa Madama, Fig. 212).
7. In Gothic, the minor elements of the frame (mainly lier), which make up a curvilinear pattern on the surface of the vault, do not lie in a vertical plane and have a double curvature. The same arcs of double curvature, in the form of curves in the plan of arches, are known in baroque vaults.
When constructing a decor from arcs, rods and frames on the curved surfaces of the vault, the main edges of the same curvature, lying in a vertical plane, should be profiled more strongly. The decor of the rods of double curvature should be made with a light low relief.
8. The complexity and variety of forms and decors of the vaults make us recall that the composition of the decor should be created simultaneously with the solution of the spatial form of the vault. To build the latter, a thorough knowledge of the codes of the past and modern and an understanding of their work is necessary. Vaulted coverings are one of the most difficult and
interesting spatial problems of architecture.
Gothic architecture.
Gothic- This is a period in the development of medieval art, covering almost all areas of material culture and developing in Western, Central and partly Eastern Europe from the 12th to the 15th centuries. Gothic came to replace the Romanesque style, gradually replacing it. Although the term "Gothic style" is most often applied to architectural structures, Gothic also embraced sculpture, painting, book miniature, costume, ornament, etc.
Gothic evolution.
Gothic originated in the 12th century in northern France, in the 13th century it spread to the territory of modern Germany, Austria, the Czech Republic, Spain, and England. Gothic penetrated into Italy later, with great difficulty and a strong transformation, which led to the emergence of "Italian Gothic". At the end of the XIV century, the so-called "international Gothic" swept Europe. Gothic penetrated into the countries of Eastern Europe later and stayed there a little longer - until the 16th century. For buildings and works of art that contained characteristic Gothic elements, but created during the eclectic period (mixing different styles of different cultures), in the middle of the 19th century, and later, the term "neo-Gothic" is used. In the 1980s, the term "gothic" began to be used to refer to a subculture ("gothic subculture"), including a musical direction ("gothic music"). The word comes from the Italian gotico - unusual, barbaric. At first, this word was used as a swear word. It should be noted that many believe that the name of the style comes from Goten - barbarians. But do not be confused, this style has nothing to do with the historical Goths. For the first time, the concept in the modern sense was applied by Giorgio Vasari in order to separate the Renaissance from the Middle Ages. Gothic completed the development of European medieval art, having arisen on the basis of the achievements of Romanesque culture. Gothic art was cult in purpose and religious in subject matter. It appealed to the highest divine powers, eternity, the Christian worldview. Gothic in its development is divided into 3 periods:
1) Early Gothic;
2) The heyday;
3) Late Gothic.
Gothic style.
Basically, it manifested itself in the architecture of temples, cathedrals, churches, monasteries. It developed on the basis of Romanesque, more precisely, Burgundian architecture. In contrast to the Romanesque style, with its round arches, massive walls and small windows, the Gothic style is characterized by pointed arches, narrow and high towers and columns, a richly decorated facade with carved details (vimpergas, tympanums, archivolts) and multicolored stained-glass lancet windows. . All elements of this style emphasize the vertical. As in all Gothic, there are three stages of development in Gothic architecture:
1) Early;
2) Mature (high gothic);
3) Late (flaming gothic).
With the advent of the Renaissance north and west of the Alps at the beginning of the 16th century, the Gothic style lost its significance.
Almost all the architecture of Gothic cathedrals is due to one major invention of the time - a new frame structure, which makes these cathedrals easily recognizable.
System of flying buttresses and buttresses.
The frame system of Gothic architecture is a set of constructive building techniques that appeared in Gothic, which made it possible to change the loads in the building and significantly lighten its walls and ceilings. Thanks to this invention, the architects of the Middle Ages were able to significantly increase the area and height of the structures being built. The main structural elements are buttresses, flying buttresses and ribs. The main and most striking feature of Gothic cathedrals is their openwork structure, which is a sharp contrast to the massive structures of the previous Romanesque architecture.
The main and most striking feature of Gothic cathedrals is their openwork structure, which is a sharp contrast to the massive structures of the previous Romanesque architecture.
Gothic vaults.
The most important element, the invention of which gave impetus to other achievements of Gothic engineering, was the rib vault. It also became the main structural unit in the construction of cathedrals. The main feature of the Gothic vault is the clearly defined profiled diagonal ribs that make up the main working frame that takes the main loads.
Load distribution.
The technical breakthrough of the Gothic architects was their discovery of a new way of distributing the load. It must be said that any free-standing building experiences two types of loads: from its own weight (including ceilings) and weather (wind, rain, snow, etc.). Then it (the building) transmits them down the walls - to the foundation, then neutralizing them in the ground. That is why stone buildings are built more solidly than wooden ones, since stone, being heavier than wood, is at greater risk of collapse in the event of an error in calculations. In Romanesque architecture, partly the heir to ancient Roman architecture, the entire walls were the load-bearing parts of the building. If the architect wanted to increase the size of the vault, then its weight also increased, and the wall had to be thickened so that it could withstand the weight of such a vault. But in Gothic architecture this method was abandoned. Crucial to the development of Gothic was the idea that the weight and pressure of masonry could be concentrated at certain points, and if maintained in these places, other elements of the building no longer needed to be load-bearing. This is how the Gothic frame arose - although the prerequisites for it appeared a little earlier: "Historically, this constructive technique arose from the improvement of the Romanesque cross vault. Already Romanesque architects in some cases laid out seams between the formwork of the cross vaults, protruding outward stones. However, such seams then had a purely decorative value ; the vault still remained heavy and massive". The innovation of the technical solution was as follows: the vault was no longer supported on the solid walls of the building, the massive cylindrical vault was replaced with a lighter openwork one, the pressure of this vault is transmitted by ribs and arches to the pillars (columns). The resulting lateral thrust is perceived by flying buttresses and buttresses. "The rib vault was much lighter than the Roman one: both the vertical pressure and the lateral thrust were reduced. The rib vault rested with its heels on the pillars-abutments, and not on the walls; its thrust was clearly identified and strictly localized, and it was clear to the builder where and how the thrust must be "redeemed". In addition, the rib vault had a certain flexibility. The shrinkage of the soil, catastrophic for Romanesque vaults, was relatively safe for him. Finally, the rib vault had the advantage of allowing irregular spaces to be covered." Thus, the design is greatly facilitated due to the redistribution of loads. The previously load-bearing, thick wall turned into a simple "light" shell, whose thickness no longer affected the building's load-bearing capacity. From a thick-walled building, the cathedral turned into a thin-walled one, but "supported" along the entire perimeter by reliable and elegant "props". In addition, Gothic abandoned the semicircular, conventional arch, replacing it wherever possible with a lancet. The use of a vaulted arch in the vaults made it possible to reduce their lateral thrust, directing a significant part of the pressure directly onto the support - moreover, the higher and more pointed the arch, the less it creates a lateral thrust on the walls and supports. The massive arch was replaced by a ribbed arch, these ribs - ribs diagonally crossed and perceived the load. The space between them was filled with a simple demoulding - a light laying of brick or stone.
flying butt- this is an external stone persistent arch, which transfers the thrust of the vaults of the main nave to the supporting pillars, spaced from the main body of the building - buttresses. The flying buttress ends with an inclined plane in the direction of the roof slope. In the early period of the development of the Gothic, there are flying buttresses hidden under the roofs, but they prevented the illumination of the cathedrals, so they were soon pushed out and became open to the outside. Flying buttresses are two-span, two-tier, and combining both of these options.
Buttress- in Gothic, a vertical structure, a powerful pillar that contributes to the stability of the wall by counteracting the expansion of the vaults with its mass. In medieval architecture, they guessed not to lean it against the wall of the building, but to take it outside, at a distance of several meters, connecting it with the building with arches - flying buttresses.
This was enough to effectively transfer the load from the wall to the support columns. The outer surface of the buttress could be vertical, stepped or continuously inclined.
pinnacle- a pointed turret, which was used to load the top of the buttress at the junction of the flying buttress. This was done to prevent shear forces.
Post-abutment- could be a simple section or represent a "bundle of columns".
Rib- the edge of the vault arch, protruding from the masonry and profiled. The system of ribs forms a frame that supports the lightweight masonry of the vault. The nerves are divided into:
1)cheek arches- four arches along the perimeter of a square cell at the base of the vault.
2)Ozhiva- diagonal arch. Almost always semicircular.
3)Tierseron- an additional rib coming from the support and supporting the lier in the middle.
4)Lierny- an additional rib running from the point of intersection of the revival to the gap of the cheek arches.
5)counterlierny- transverse ribs connecting the main ribs (i.e. revivals, lierns and tierserons).
6)formwork- in the rib vault filling between the ribs.
7)Keystone(power socket)
Decor.
The technical solution of structural problems was not the only task of the Gothic architect. The enrichment of textures and the embellishment of the structure proceeded simultaneously with the evolution of constructive solutions and were almost inseparable from them. The buttresses were crowned with lanceolate turrets-pinnacles, in turn decorated with serrated protrusions. Spillways with the help of a sculptor turned into a fantastic combination of animal and plant forms. The tides of the portals going deep into the ledges are supported by thin columns alternately with elongated figures of angels and saints, and the arched outline of the tympanum above the doors was covered with reliefs on the themes of the Last Judgment or similar subjects and painted in bright colors. Thus, all forms of art played their part in enlightening the flock, warning the faithful of the dangers of a sinful life and visually depicting the bliss of a holy life.
In the solution of window openings, the same merging of constructive evolution and ornamentation took place. Initially, the case was limited to the grouping of two or three medium-sized windows in a single architectural frame. Then the partition between such windows was successively reduced, while the number of openings increased, until the effect of a completely dissected wall surface was achieved. Further reduction in the size of the stone piers between the smaller windows led to the emergence of a lacy window structure, the ornamental pattern of which was created by thin stone ribs. Initially assembled in the simplest geometric forms, the lacy structures of the windows became more and more complex over time. In England, such a "decorated" style of the late 14th-15th centuries. was replaced by "perpendicular", which in France corresponded to the style of "flaming gothic".
The multi-coloured stained-glass windows in these windows were assembled from small pieces of glass, clamped with an H-shaped lead profile to provide moisture insulation. However, the lead casings were not strong enough to withstand the pressure of the wind on the large surface of the glass, which subsequently required the use of frames made of iron rods or rebar.
Over time, curly stone ribs began to be used instead of iron fittings, which paved the way for freer lace compositions. V stained glass windows 12th c. the dominant colors were shades of blue, complemented by red, bringing warmth to the whole. Yellow, green, white and purple were used extremely sparingly. In the same century, the builders of Cistercian churches, abandoning the abundance of flowers, began to use grisaille for decorative purposes (painting in different shades of the same color, often gray) on a simple greenish-white glass surface. In the 13th century the size of the pieces of colored glass is increasing, and red is used much more widely. In the 15th century stained glass art begins to decline.
Gothic rose/rosette
Rib vault options.
Schemes of various variants of the rib vault.
In Gothic cathedrals, one can find many variations of rib interlacing, many of which are unnamed. Several main types:
1) Cross vault (quadripartite rib vault)- the simplest version of the rib vault, which has six arches and four fields of formwork.
Arched cross vault.
2) Hexagon vault (sexpartite rib vault)- a complicated version of the cross vault, due to the introduction of an additional rib, dividing the vault into 6 decks.
3) Star vault (lierne vauit, Stellar vault)- the next stage of complication, thanks to the introduction of lierns, the number of which can increase. The location of the ribs takes on the shape of a star.
Star vault. Photo below.
A star vault is a form of cross Gothic vault. Has auxiliary ribs - tierserons and lierny. The main diagonal ribs of the cross vault are clearly distinguished in the frame.
4) Fan vault (fan vault)- is created by ribs emanating from one corner, having the same curvature, forming equal angles and forming a funnel-shaped surface resembling a fan. Typical of England ("spread gothic").
5) Net vault (netvault)- ribs create a grid of ribs with cells that are approximately the same in size.
Castles, manors and houses.
In the civil architecture of the Gothic era, it is necessary to distinguish between the early castle, which served as both a dwelling and a citadel, from the later country residence, which was built in an era of relative reduction in the need for individual defense of each from everyone. In both the first and second types, one can find signs originally developed in church architecture.
A house typical of the 13th century. had three floors and was placed to the street either with a side wall or with an end face. The ground floor was usually occupied by a shop and warehouse; on the second there were living rooms, of which the main one was facing the street; sleeping quarters were located in the third or in the attic. The shop facing the front and the kitchen behind were usually separated by a courtyard. Already in the 13th century. decorative design of chimneys came into fashion, and carved decor was widely used.
The most popular materials in residential construction were wood and plaster, but in some regions stone or brick was preferred. The wooden frame was usually assembled from powerful beams, the joints of which were carefully fitted and hemmed. The frame was left open from the outside, it brought a clear decorative pattern to the facade. The pattern was formed by vertical and horizontal rods, in some places connected by diagonal ties (in some regions - by crossing diagonals). Filling between the elements of the frame was made of plaster on wooden shingles or brick, then covered with plaster. Window coverings generally followed church fashion, but, of course, in simplified forms.
In the 14th-15th centuries. there are no significant changes in the general layout or in the structural scheme of the residential building, however, the number of windows increases, and they themselves become larger. By 1500, the former "lace" bindings are usually replaced by rectangular windows with straight imposts and rods.
civil architecture.
The Gothic architecture of France is not limited to churches, castles and residential buildings, also covering city halls, city bell towers, hospitals, schools of various levels and all other public buildings necessary for the life of a medieval person.
The city bell tower usually served as a symbol of the independence of the city. Several bells were hung on it, among which was a signal bell, and in the 14th century. clocks were set on it. In Moulins, a tower of this kind has been preserved, on which the clock is called by mechanical figures.
Most of the medieval hospitals were built in the Gothic era. Both the church and the feudal lords were their founders, but the management of the hospital was usually transferred to the hands of the church. The hospitals of that time had broader functions than modern ones, since in them, along with the treatment of the sick, they provided shelter and food for pilgrims, the elderly, the homeless and the needy. Their planning, constructive system and decor were borrowed equally from church architecture and from the architecture of a residential building. The first "lazarettos", or leper colonies for patients with leprosy, were also the first hospitals in the narrow sense of the word. In such infirmaries, lepers lived in separate houses, and those who looked after them lived in a separate building. Around 1270 in France there were up to 800 infirmaries, but by the 15th century. the need for them was reduced so much that the funds allocated for their maintenance were directed to other purposes. The Hospital Maladredi du Tortoire gives an idea of the type of this institution. Three buildings are located on a rectangular plot: a two-story building for patients, a chapel and a two-story staff building, which housed the kitchen. On each of the two floors of the hospital building there was one long hall, lit by eight lace-laced windows. Fireplaces heated the hall and provided its ventilation, and mobile wooden screens between the beds made it possible to separate the patients from each other.
Monastic orders, specialized in helping the sick, created a different type of hospital. The best preserved medieval hospital in Beaune allows you to see the classic hospital layout of the 15th century. On the sides of the courtyard surrounded by an arcade are large halls (one for men, the other for women) and two side wings. Initially, at the end of each hall, an altar was arranged, illuminated by a large window. The halls were covered with wooden vaults. Glazed tiles on the outside, murals and tapestries inside brought intense color to the overall solution. Wooden galleries surrounding the yard gave patients the opportunity to walk under the fresh air.
Milan Cathedral. height from the ground (with a spire) - 108, 50 m; height of the central facade -56, 50 m.; length of the main facade: 67.90 m; width: 93 m; area: 11.700 sq. m; spiers: 135; 2245 statues on the facades.
Cathedral in Reims (Notre-Dame de Reims) in the French province of Champagne (Champagne). The Archbishop of Reims, Aubry de Humbert, founded the Cathedral of Our Lady in 1211. Architects Jean d'Orbais 1211, Jean-le-Loup 1231-1237, Gaucher de Reims 1247-1255, Bernard de Soissons 1255- 1285
Abbey of Saint Denis near Paris. France. 1137-1150
Gothic style. Cathedral in Chartres - Cathédrale Notre-Dame de Chartres - Catholic cathedral in the city of Chartres (1194-1260)
Gothic Ulm Cathedral. Ulm in Germany, 161.5 m high (1377-1890)
Roman Catholic Gothic Cologne Cathedral of the Blessed Virgin Mary and St. Peter (Kölner Dom). 1248-1437; 1842-1880 It was built on the model of the French cathedral in Amiens.
Romanesque vault
An important task of Romanesque building art was the transformation of a basilica with a flat wooden ceiling into a vaulted one. At first, small spans of the side aisles and apses were covered with a vault, later the main aisles were also covered with a vault. The thickness of the vault was sometimes quite significant, so the walls and pylons were designed thick with a large margin of safety. In connection with the need for large covered spaces and the development of technical construction ideas, the design of the initially heavy vaults and walls began to be gradually lightened.
The vault makes it possible to cover larger spaces than wooden beams. The simplest in form and design is a cylindrical vault, which, without pushing the walls apart, presses on them from above with a huge weight, and therefore requires especially massive walls. This vault is most suitable for covering rooms with a small span, but it was also often used in the main nave - in France in the Provence and Auvergne regions (Notre Dame du Port Cathedral in Clermont).
Later, the semicircular shape of the vault arch was replaced by a lancet one. Thus, the nave of the cathedral in Otun (beginning of the 12th century) is covered with an ogival vault with the so-called edge arches.
The basis for new types of vaults was the old Roman cross vault over a square in plan room, obtained by the intersection of two half-cylinders. The loads arising from this arch are distributed along the diagonal ribs, and from them are transferred to four supports at the corners of the overlapped space. Initially, the ribs that appeared at the intersection of the half-cylinders played the role of arches, which made it possible to lighten the entire structure (St.
Monastery church in Cluny. Romanesque cross vault:
1 - end ribs; 2 - diagonal ribs; 3 - lock; 4 - stripping.
If you increase the height of the vault so that the diagonal intersection curve from elliptical to semicircular, you can get the so-called elevated groin vault.
The vaults most often had solid masonry, which, as we said, required the construction of massive pylons. Therefore, a big step forward was Romanesque composite pylon: half-columns were added to the main pylon, on which edge arches rested, and as a result, the expansion of the vault was reduced. A significant constructive achievement was the distribution of the load from the vault to several specific points due to the rigid connection of transverse edge arches, ribs and pylons. The rib and edge arch become the frame of the vault, and the pylon becomes the frame of the wall.
At a later time, end (cheek) arches and ribs were laid out first. This design is called ribbed cross vault. During the heyday of the Romanesque style, this vault became elevated, and its diagonal arch acquired a pointed shape (Church of the Holy Trinity in Cana, 1062-1066).
To cover the side aisles, instead of the cross vault, sometimes they used semi-cylindrical vaults, very often used in civil engineering.
Romanesque structures are, first of all, an elevated rib vault, a pointed arch and the offset of oblique lateral braces from the vaults by a system of supports. They form the basis for the subsequent Gothic style in architecture.
Stone vaults were nowhere so common as among the Romans: the ruins are full of their remnants, everywhere vaults made of rubble and mortar, boldly thrown over space, cover the ancient halls; or at least the remains of a stone structure in the form of a ledge hanging over the surface of the walls have been preserved as witnesses of the original structure and reveal to us the structure of vaults destroyed by time. These vaults of small-sized material vary, so to speak, to infinity; they blocked either rectangular fences, then round, then polygonal in terms of area, then exedra. Made on formwork, they adapted equally well to the most diverse plans and to the most diverse requirements for the location of premises. In addition, many of them seemed to have been designed to exist for centuries, and the noble simplicity of their forms gave the buildings a strict, majestic appearance. Building techniques have never corresponded so well to the material and spiritual needs of the people; and it becomes clear to us why the Romans based their whole architecture on the use of such a structural system.
The problem of replacing wooden structures with others that are both stronger and more durable is as old as the art of building; but before the advent of arches, made in the form of a monolithic stone structure, no truly practical solution was known. The flooring of slabs and stone architrave ceilings of Egyptian and Greek temples required materials that were obtained at the cost of hard labor and used at great expense. We find in the construction of primitive architecture several vaults, made of horizontal rows of stones, gradually overhanging one another; we even find vaults composed of wedge-shaped stones, the lines of the seams of which converge at one point on the horizontal axis; but, out of ignorance or in obedience to a common system, the builders of these first years almost always laid the wedge-shaped stones of their vaults dry, placing no cement, no mortar, no substance to compensate for the irregularities of the pavement between the two blocks. Hence the need arose to give the stones used a very regular shape, hence the practical difficulties arose, which, undoubtedly, must have limited the possibilities inherent in hewn stone vaults. Among all the ancient peoples, hewn stone vaults were most common among the Etruscans; however, even with them their use was very limited; they covered with vaults sewers, underground spillways used to drain damp plains, aqueducts, city gates, but in Etruscan buildings designed to satisfy ordinary living needs, and even in religious buildings, the vaulted structure never received permanent use; wooden logs were used, similar to those described by Vitruvius in the Tuscan temple, or stone architraves, similar to those reproduced in the facades of several buildings that have come down to us, carved into the rock.
As for the Greeks, despite their constant connection with Etruria, they apparently never thought of reproducing the varieties of Etruscan vaults, the seam lines of which intersect at one point. We find in the original Greek buildings, at Mycenae, and especially on the island of Euboea, false vaults, made of loose masonry, but vaults of wedge-shaped stones, the seams of which converge at one point, were not used by the Greeks before the Roman conquest; in the forms of building with a flat ceiling, their architects gave the highest expression of the ideas of proportionality and regularity; and the Greeks cherished these forms as the most beautiful creation of their genius; they were, as it were, part of their national glory and held out all the time that Greek independence lasted. Therefore, the Greeks, being witnesses to the appearance of vaulted buildings, did not take part in them and left them to the Romans. architects are honored to spread this structural system, which they made simple and practical, thanks to the use of small-sized materials, artificially combined into a single whole.
Whether the Romans were the inventors of vaults made in the form of a monolithic stone structure, that is, made of small stones tightly cemented with mortar, or not, but be that as it may, before them, no people had thought of building from small stone materials vaults of large spans. The Romans themselves, apparently, for a long time neglected the possibilities that such a construction could give, or did not know them; and we see that it is constantly applied only towards the end of the last century BC; it seems to have developed during the period of material prosperity that followed the end of conquests in distant lands and the end of civil strife. Her successes were at that time swift; a real revolution was taking place in the art of building. The use of arches in the large halls of public buildings entailed a complete change in plans; the supports, which were now subjected to a new kind of effort, had to assume hitherto unknown forms; had to change the grouping of the hall to ensure a clear perceived strut of the vaults. Until now, builders have lived, as it were, at the expense of the funds of Greece and Etruria, only during this period building techniques are freed from the fetters of tradition; a whole constructive system, truly Roman, is born, or at least receives a correct and widespread development.
This transformation, which took place in the last years of the republic, was, of course, prepared for a long time; but whether the first examples of monolithic vaulted structures disappeared during the long period of time separating us from the Romans, or rather these primitive buildings were demolished and gave way to magnificent structures erected by the emperors, and the traces of this interesting series of experiments and improvements that preceded the era of Augustus were as if erased by time?
Be that as it may, the Pantheon stands before us at the same time as a masterpiece of Roman architecture and as one of the first monuments of its history; and examples of earlier times are too rare and doubtful to testify to the successive successes of the building art of the Roman Republic. We will not attempt to restore the picture of its origin by guesswork - we immediately began to study the vaults, made in the form of a monolithic stone structure and reached complete completion; we will describe the conditions under which they were built, and we will try to connect these collected facts with a small number of simple ideas that seem to have dominated the Romans in the whole system of vaulted buildings.
If we turn to some Roman building with stone vaults, if we examine, for example, one of the rows of aqueducts that furrowed the environs of Rome, then we will notice at the ends the main arches made of bricks or stones, the lines of the seams of the masonry of which converge in one common center, and behind these main arches, a rough masonry of pieces of tuff or tiles, similar to concrete. A compact mass of crushed stone and mortar, enclosed between two arched facings, the lines of the masonry seams of which converge at one point - such is the structure that is revealed during a superficial examination of the ruins. But upon closer examination of these outwardly homogeneous rough massifs, we will find ribs of a completely different structure embedded in them, real built-in ribs, sometimes whole brick lattices that form an internal skeleton in the body of the fillings, a light frame branching, subdividing and extending inside the rough stone structures that clothe.
One should not look at the skeleton of the vault as a system of rigid arches, erected simultaneously with the masonry of the structure, made of crushed stone and mortar, and intended to strengthen it, in a word, as something similar to stone pillars in the walls of modern buildings. Brick frames placed in the masonry of the Roman vaults were erected earlier, and the rough masonry was made later, as evidenced by the discrepancy between the rows of infill and frame masonry (Fig. 8).
This light frame, this frame, embedded in the vault, consists, like the main arches with which it ends, of bricks; the lines of the seams of its masonry converge at one point, and in this respect it is somewhat similar in construction to our stone vaults; but here is an analogy. ends, and if we leave aside the internal structure of the vaults and consider the filling itself, we will be convinced of the simplicity of the construction, which is completely uncharacteristic of modern buildings.
| Rice. eight. |
The very name of the vault evokes the idea of a structure of stones laid in such a way that the lines of the masonry seams converge in one common center; and this idea actually corresponds to the design of Roman vaults of hewn stones laid dry; likewise, this representation is correct, as we have just said, in relation to arches of brick, placed in the form of a strong frame within arrays; but to extend it to the arrays themselves would be completely erroneous; the rows that make up the laying of the filling of the Roman vault, made in the form of a monolithic stone structure, retain a strictly horizontal position from the base to the top; and seeing how the traces of these rows are marked in the ruined parts of the ruins, one involuntarily recalls even layers, sometimes outlined quite clearly in the dimensions of the layered soils. Such an arrangement of seams is a rather unusual phenomenon, and therefore it seems useful to us to explain it graphically. I give in two comparative sketches cross-sections of vaults erected according to either system.
In the modern vault, the seams are located as shown in Fig. 9.
In the same way, the seams are located in the Roman vault of hewn stone, laid dry.
On the contrary, in a Roman vault made of small-sized material, forming a monolithic stone structure, the seams invariably have the direction shown by hatching in the second of the above-mentioned schematic sections (Fig. 10). Thus, the Romans, depending on whether they built from hewn stone or from rubble cemented with mortar, constantly placed the lines of seams either converging in one common center, or strictly parallel. These two opposite methods, however, do not contain any discrepancy, any contradiction in ancient methods, because there is a profound difference between the conditions for balancing vaults of stones laid dry and vaults made in the form of a solid monolithic stone structure.
 |
 |
| Rice. 9 | Rice. 10 |
In one case, the stones are held together only by their shape, and it is necessary to give the lines of seams a direction leading to their intersection at one point; in the second case, the binder turns everything into a whole single block in which layers of mortar and rows of stones are mixed
into one continuous homogeneous mass; therefore, the direction of these rows does not really matter from the point of view of the strength of the structure; and the Romans took advantage of this circumstance to introduce into their work a significant simplification: they decisively freed themselves from all complications that could be caused by the arrangement of the seams so that their lines converged at one point. Thus, the masonry of their arches is nothing more than a continuation of the supports, which, as it were, hang over the span; abolish the skeleton embedded in the infill, and a stone structure remains, with respect to the direction of the rows, very similar to the structure of the walls that carry it.
We said about ancient walls that the Romans used two types of monolithic masonry, namely, without tamping and with tamping; and we noticed that only the first was used to build walls with a lining of thin bricks, because only it can be done without auxiliary devices and continuous formwork. Considerations of the same order apply to vaults, and they allow us to predict which of the two types of masonry should have been used in them. For the vaults, it was inevitable that there was an arrangement of internal formwork, which gave the filling a proper shape, but if this formwork was necessary, in other words, if circles were needed for the vaults, then at least it was necessary to try to make these circles as cheap as possible, and this condition should have affected the choice of the Romans between two masonry structures known to them. If they used masonry that required tamping, they would subject the circles to shocks that could loosen their mates, but first of all, this method would cause strong deformations in the supporting frame of the circles: the circles would be clamped in places located close to the supports of the vault (Fig. 11 ), and at the same time, the outer facings would begin to burst outwards.
To accept such efforts, it would be necessary, in addition to circling, to arrange formwork for them; circling and formwork, all this temporary wooden structure, would have to be extremely strong in order to withstand the bursting forces and the incessant action of blows: in the face of these difficulties, the best solution was to abandon the rammed masonry.
So reasoned the Roman architects; the masonry of their vaults, wherever I could establish its structure, was made with exact observance of those methods that were used in the construction of ordinary walls. Sometimes lighter materials are taken for the vaults than for the walls, but the method of execution is the same in both cases - the masonry of the vaults is never rammed.
 |
 |
| Rice. eleven | Rice. 12 |
Although when choosing the type of construction, they had in mind to save on circles, the effect exerted by the vaults on their supports was, nevertheless, in some cases very strong. As long as the masonry of the vault rose slightly above the heels, it held almost by itself; its rows with successive overlaps really joined the vertical, continuation of the supports, like some kind of process ABS in the form of a protrusion attached along the line AB(Fig. 12); - the shape of this protrusion ABS did not differ significantly from the theoretical profile of a beam of equal resistance, suitable for a solid body, embedded in the wall at one end and loaded only by its own weight, and therefore, these parts of the vaults did not require difficult and expensive supports for their construction. In an extreme case, the vault in this lower part could do without scaffolding - a template was enough to give its lower surface the curvature and shape that it should have had.
But this ease of execution diminishes as the vault gets higher; its overhanging parts, the farther, the more they press on the circles, and the load near the top of the arch increases with extreme speed.
Soon the vault is like a semi-liquid mass, resting with all its weight on the supporting devices; from the circles, which, just now, were almost superfluous, energetic resistance is now required, the higher, the more dense and massive the vaults should be; Roman vaults were never light: the rough structure of their infills forced them to sometimes be given enormous dimensions.
Moreover, it was necessary to support this heap of materials, which had not yet reached a sufficiently strong connection, with supports that were unable to bend.
This was a serious difficulty: the slightest settling, as a result of which the stone structure would have to work, would, just during the setting process, cause internal displacements in the masonry, which consisted of crushed stone and mortar, and maybe even cracks.
In an ordinary vault, the lines of the masonry seams of which are directed to one point, the draft circled, although annoying, but rarely causes a catastrophe: maybe cracks form in several seams, but the stability of the building does not depend solely on the integrity of these seams, the mortar in this type of vault serves first of all, for adjustment, for distribution of pressure, it is not an astringent, it is just a layer between wedge-shaped stones; even if this mortar cracks or disappears, it will not necessarily endanger the integrity of the vault, and its presence is so little necessary that the ancients never used mortar in their cut stone constructions.
But in the vaults, made in the form of a monolithic stone structure, as the ancients thought, the role of the mortar ceases to be auxiliary; here he, and he alone, provides a connection between the elements of the structure; as soon as this connection is broken, only something similar to a broken, collapsed, formerly monolithic massif will remain of the structure.
Thus, in order to derive a Roman vault from small materials, it was necessary to ensure that the circles were completely immutable: this, so to speak, was the first condition for success, and this condition could only be met with great difficulty when simple wooden circles were used. But even using more wood, multiplying the number of pairings, giving them impeccable accuracy, it is impossible to solve all the difficulties: the tree, with the best pairing, is bent, warped, deformed, and a monolithic vault, unable to follow all the deformations of the wooden structure that serves as its formwork, will constantly be under threat. lose support due to possible precipitation or shear circled.
It should be added that it would be too unusual for Roman builders to attach such importance to temporary devices: it would be surprising if they, who usually considered useful only those works that are designed for a long existence, and especially those accustomed to always look for simple solutions, suddenly in in a single case, such complex and expensive ancillary work would be used.
Finally, if we pay attention to the composition of workers employed on construction sites, then, although in different ways, we will come to the same conclusion. The Romans, who had an unlimited number of workers in all places of their empire, did not find workers everywhere with equal ease, who could be entrusted with responsible carpentry work. When the structures being erected require a simple expenditure of physical effort, it is easy to recruit labor among the conquered peoples, in the armies, among the slaves. But as soon as complex and difficult structures are involved, such as circles that are strong and not deformable, the possibilities of execution become more limited; architects will have to muster, at great expense, many skilled craftsmen, and besides, they will have to put up with inevitable delays. And when, having spent money and time, they are able to erect entire scaffolding to support the filling of their huge vaults without the risk of their settlement, then the next day after the end of the work, all the expenses for these temporary devices, so to speak, will be wasted, all this expensive equipment without a trace. will disappear. It was, of course, unprofitable to sacrifice expensive and hard labor in vain, the ancient builders tried to avoid this, and their efforts to partially get rid of their dependence on temporary forests inspired them with an idea as witty as simple - to introduce this a semblance of an internal brick frame that supported the mass of the filling masonry during the construction and thereby unloaded the circle,
 |
| Rice. thirteen. |
In the first tables appended to this work, a general view is given of the various frames of the vaults, and they are shown embedded in the infill they supported, and the drawings placed in the text explain some structural details, and we can immediately understand, at least in general features, the nature and significance of the functions performed by them.
I took a simple type from various framework systems and tried to reproduce it in fig. 13 view of the structure in the process of construction.
The figure shows temporary circles C, lightweight frame made of bricks D, placed directly on the circles, and finally, filling M from crushed stone and mortar, from which, at the end of the work, a vault is formed in the proper sense of the word.
In accordance with modern building techniques, temporary circles C would carry the entire vault, they would have to be made extremely strong, and therefore they would be very expensive. Here, on the contrary, the wooden circles carry, so to speak, only the skeleton of the arch, this is a significant difference, which makes it possible to reduce the bearing capacity of the circles, that is, to make them much less powerful, which will entail a significant reduction in costs.
Thanks to the addition of this strong frame structure, which covers and protects them, the temporary circles are secured from any danger of breaking, they form the desired shape for filling, without experiencing the gravity of its weight; once erected, the frame of bricks becomes a real system of circles, extremely durable vaults remaining in the body of the masonry, merging with it into one and contributing, along with rough monolithic masonry, to the strength and durability of the structure.
These second brick circles, thus included in the body of the masonry, are undoubtedly more expensive than the amount of fill material they occupy; but how insignificant these extra costs will seem, when compared with the savings achieved by the device of a temporary wooden structure. In addition, this extra expense in itself was very insignificant.
As a material for the frames, a simple brick was used, although large in size, but its manufacture in the outskirts of Rome was inexpensive.
On the other hand, this brick, despite its cheapness, was used truly remarkably economically.
Instead of making this frame solid, we see that the Romans made it through, thus eliminating about half of the bricks that would be necessary to make such a load-bearing continuous shell over the circles (Plate I).
Often they were limited to individual ribs, so to speak, spring arches, immersed in the thickness of the filling of crushed stone and mortar (Plates IΙ, III, VII, VIII, IX, X, XI). And these friendly arches are made of ordinary masonry; they were never made solid, but openwork in all directions; these are lattice structures made of bricks covering narrow strips of the vault at a certain distance.
Finally, in some cases, in order to reduce the costs that, given the thickness of the vault, the installation of bricks on the edge requires, the Romans used frames made of bricks laid flat and forming a kind of curved flooring on the surface (Plate IV, Fig. 1). Sometimes two floorings of this kind were laid one above the other, but then the second was usually no longer continuous (Plate IV, Fig. 3). It was impossible to go further in the economical use of materials.
As for labor costs, they were less than expected, judging by the ingenious and in some cases sophisticated combinations that we see in the figures: everything was done quickly, I would even say that it was perhaps very rough work. Looking around a Roman building, you feel that the ancient builders learned, by practice, to lay out the brick frames of vaults in haste and achieved in them all the economy of time and labor compatible with such work; the appearance of these auxiliary devices speaks of the most hasty execution, and the irregularity of forms in them is sometimes so striking that I was forced, in order to make clear the idea of the builders, to give these frameworks in my drawings a regularity that was often far from being detected by the most careful examination of the ruins.
However, in no case should the Romans be reproached for unreasonable carelessness; in this case, the speed of work at the expense of its accuracy was more an advantage than a disadvantage. Any waste of time in auxiliary building work, unless it is justified by the strict requirements of urgent necessity, must be considered useless; and the rough appearance given by the Roman builders to the skeletons of their vaults, testifies to the fact that they correctly understood their purpose. It was enough to install the brick frame so securely that it only lasted until the completion of the laying of the filling: as soon as the stone monolithic structure was ready, everything turned out to be embedded, immured in its mass; and during the decorative work, the last traces of the frame, which could still be visible from the inside, disappeared under a thick layer of plaster; What advantage, then, would have been given under these conditions by more thorough execution? The rather carelessly made frames of the Roman vaults were good enough; and trying to do them more carefully would be a waste of time.
 |
| Rice. 14 |
But besides economic considerations, the Romans had another more important reason for avoiding delay. In order to fully understand the reason for their haste in completing the frames of the vaults, one must clearly imagine the state of the building at the moment when the auxiliary brick structures are to be installed. The masonry of the supports has been completed and the circle has just been installed in place. The architect then faces a difficult choice. Continuing the laying of the infill, he runs the risk of crushing the circle; if, on the contrary, he suspends the work of laying the infill in order to take it up again when the laying of the vault frame is completed, this forces him to leave the entire artel of workers and slaves unoccupied.
His only means of coordinating everything is for him to hurriedly set up these frames and finish their laying, while the filling does not yet put pressure on the circles. If, for example, AB indicates the level where the pressure begins, it is necessary that at the moment when the filling stack reaches the level AB, the arches of the frame were brought out under the castle and the structure would have the form shown in fig. 14.
Thus, the laying of frames and the filling of the entire structure as a whole begins and is carried out simultaneously, but the frames must be brought out and finished with masonry so that they can already fulfill their purpose during that short period of time while the laying of the filling holds on its own. Hence this so conspicuous haste; the cause of it, as we see, was serious, otherwise there would have been a temporary inactivity of the numerous laborers whom the Romans used to carry out the simplest and most laborious part of the construction work in their large structures.
This initial period, when the inner frame of the vaults had to be erected in its entirety and with great haste, was, however, the only critical moment in the work: the masonry of the vaults ended on these rigid supports as easily as ordinary masonry; and when, finally, the moment of their uncircling came (an operation rather complicated with other constructive systems), it was carried out without any danger, or rather, the uncircling did not represent any serious operation. It was possible without any risk to remove the wooden structure that carried the formwork: he was a real circle. the frame itself; and hidden in the masonry filling of crushed stone and mortar, these brick circles blocked the span, bearing the weight of the vaults until the mortar was completely hardened.
Now we can cover in general both the course of Roman construction and the advantages associated with the constructive system of ancient vaults: it, as we see, is based on very simple and practical principles; some of the principles underlying it are so natural and come to mind so easily that they can be found in a different form in architecture, most outwardly different from Roman; I'm talking about French architecture of the Middle Ages. Rib vaults in our (French) cathedrals, of course, do not resemble the vaults of the Romans either in their external appearance or in the static conditions of their work; some are held by a deliberately created complex combination of forces and thrusts; in others, stability is created simply by the monolithic structure of their masonry; but with regard to methods of erection, the analogy is indicative, and all the more remarkable that it may be accidental. Indeed, who will not be struck by the fact that the ribs of medieval vaults are equivalent to antique frames. In one case, the ribs are made of bricks and placed in a mass of masonry infill of crushed stone and mortar, in the other they protrude in relief and support the infills of real masonry. But differences in forms and material are not important here: the main one. the installation is the same on both sides; hidden or protruding ribs play, at least during the execution of work, the same role; and the less similarity in their appearance, the more one feels how natural and understandable the idea of erecting vaults on the second row of circles built of stone material is. I do not undertake to predict the transformations that this ingenious idea will experience in the future; but the applications that it has consistently received in two radically different architectures, in my opinion, speak for the fact that it is fruitful; and the study of the possibilities that such a solution can give in our day is certainly worthy of the full attention of the builders.
At the end of this first study of the vaults on the monuments, it would be useful to compare all our hypotheses as a whole with the indications in the texts. Unfortunately, the positive information on this subject is very incomplete, and the hints are very obscure.
Vitruvius mentions the names of the vaults several times, but does not give any details about the methods of their construction; if we analyze his entire treatise, we will hardly find at least one place in it that seriously illuminates this, perhaps the most important issue in the entire history of ancient architecture. He talks about the way. reproduction of the design of the arch with the help of a wooden structure made of boards arranged along a curve, intertwined with reeds; and plastered; as for real vaults, it is in vain to look for their description from him. Is it necessary to see in this strange gap an omission on the part of the author, or the result of a complete distortion of his works? Or, finally, is it a sign indicating the state of building art in the time of Vitruvius? I would willingly lean towards this last suggestion; and the date of construction of the most ancient vaults of large dimensions that have survived to our time makes it, it must be admitted, very plausible.
 |
| Rice. 15. |
Despite these gaps and ambiguities, Vitruvius always remained an authority among the Romans; and later authors" were for the most part content to repeat in a less ponderous, less lengthy, but often less precise form, the indications of his text. methods of their construction; the expert in agriculture, Palladius, and the anonymous author who shortened Vitruvius, keep the same silence about building techniques regarding vaults in the proper sense of the word, but they spread, following the example of the original author, from whom they copied, about these very little interesting constructions, outwardly imitating the curvature of the vaults, without possessing either their strength or their durability.
 |
| Rice. sixteen. |
But, if we are deprived of the possibility of verification by texts, then at least we can find out what the traditions say. The Italians even now use temporary wooden structures very sparingly when it comes to circles for the construction of vaults; thus, it is not uncommon to see that they use such a construction as shown in fig. 15.
The permanent brick circles of the Romans are presented here in the form of a row of bricks laid flat, resting on a cross-beam made of defective wood, and several bricks placed on edge; sometimes the Italians remove the flat bricks during the circling, while the Romans usually left them in place. However, even in modern Italian buildings, I have repeatedly met already completed vaults, covered inside with such a curved brick flooring, which originally served as their formwork and circles.
Here (Fig. 16) is another system of brick circles, conceived in approximately the same spirit.
The circles, on which the vault is displayed, consist of two curved protruding ribs starting at the heels, passing at the top into a through brick wall located on a wooden beam.
Finally, I will give as a last example (Fig. 17) a circling structure consisting of two wooden beams abutting one another and carrying a through wall of bricks, a kind of tympanum made with irregular masonry, the purpose of which is to support the masonry of the vault during the production of work.
 |
| Rice. 17. |
It is probable that none of these three types of circling corresponds exactly to the ancient designs; but it seems to me that it is impossible not to recognize here and there a remarkable identity of principles: for example, the same desire to confine ourselves to the simplest wooden structures, brick, which in both cases plays an important role as a material for circles, and its use for the sake of economy and ease flat in masonry decking or in the laying of through walls. However, in further study, the observation of modern methods will repeatedly help us to understand the practical methods of the Romans, obscurely visible in the ruins, or at least add new evidence in favor of the explanatory hypotheses that we set out above.
Let us now return to the Roman frame designs. They are divided, as we see, into two groups, of which one covers all structures based on the use of arches or lattice frames of such masonry, the seam lines of which converge in one center, and the other - all those based on the use of brick flooring, laid flat. We will deal in turn with both solutions in various types of vaults and the first duty - in barrel vaults.
a) Arches on frames with radial seams.
Frames whose masonry seam lines intersect in one center are usually made of two types of bricks: square bricks with a side length of 2 Roman feet (slightly less than 0.60 m) and rectangular bricks with side dimensions of 2 feet and about 1/2 foot ( 0.15 m).
Arches were made of rectangular bricks, ribs, placing the latter at a distance of 2 feet between the axes, and with large square bricks, with a side length of 2 Roman feet, these ribs were connected to each other in the same way as shown in Fig. eighteen.
 |
| Rice. eighteen. |
In this way, a kind of lattice was obtained, which can be considered the most complete type of Roman frame with radial seams.
Sometimes (but this is an exception and seems to be the result of negligence rather than deliberate calculation) large square bricks used for communication, instead of being laid as shown in fig. 18, that is, one after the other along one line - along the generatrix of the cylindrical vault, they overlap one another so that each square brick covers the entire width of both arches connected by it (Fig. 19).
The layout is doubly defective - because a) a much smaller part of the vault can be covered with the same amount of materials, and b) it is more difficult to fit the filling into the reduced cages of the framework.
Perhaps a little more strength is provided by the greater number of these arches; but with another system, apparently, strength is obtained that is quite sufficient even for the widest vaults; and since the frameworks here were essentially auxiliary structures, the ancients acted wisely, sacrificing this slight increase in strength, for the sake of more important conditions of economy and lightness.
A remarkable example of a building made according to the first method (Fig. 18) we find in the hall of the Caesars' palace in Rome, which is part of a group of buildings surrounding the Circus Maximus. I present this summary in Table. I; in order to give a clearer idea of its general structure and to show how it relates to its supports, I have drawn a series of sections in which all the details of the building are revealed and at the same time summarized. the ideas that we have been able to form up to now both about the design of vaults and about the usual structure of Roman massive stone structures. These drawings will make it possible to establish an identity between the construction of the masonry of the vaults and supports, the horizontal arrangement of rows in the infills of the vault, and, finally, in particular, the presence of a common frame, which changes from the inside when moving from the cylindrical vault to the supports with a lining of triangular bricks.
This table gives perhaps the most complete type of antique structural framing system: the brick framing shown here combines the valuable qualities of a rigid support and continuous cladding.
 |
| Rice. nineteen. |
But this construction still required a quantity of bricks that may seem enormous, and the Romans, sacrificing this too costly advantage, gradually abandoned such a construction in order to move from a solid brick frame to free-standing, arched ribs hidden in the masonry. I will try to show the consequences of these simplifications and variations. But, connecting with the first type of construction all the subsequent examples that I want to give, I, of course, do not pretend to restore the historical chain of events and the way in which the changes in constructive methods went in reality: the relative dates of the construction of various vaults that we will have to compare, usually little known; and therefore it would be too bold to set out to find, in the current state of archaeological knowledge [the end of the 19th century. - Ed.] real continuity of Roman ideas; my intention is only to identify among the many different forms the leading basic idea underlying the design of the permanent circles - the frames of ancient vaults.
Going after this reservation to compare the code shown in Table. I, with various vaults shown on the same scale in Table. II and III, we will see that they are obviously connected by one common idea, which found the most complete expression in the code of the Palatine.
On fig. 1 tab. II frame arches are no longer connected directly. with each other by large square bonding bricks: instead of this common bonding, the arches are simply placed closer together.
 |
 |
| Rice. twenty. | Rice. 21. |
The frame of the vault is now, so to speak, reduced to a system of free-standing ribs; these ribs are not more than 0.15 m wide, in the direction of the generatrix of the vault, and the gaps between them exceed the dimensions of ordinary Roman square bricks. Thus, the space between the arches is not divided into cells; but on the other hand, to the right and to the left, on each side of the arch, the ends of large square bricks interspersed with bricks 0.15 m wide protrude; without dividing the space between the arches into separate cells, they nevertheless clearly outline these divisions in it and, so to speak, compensate for the discontinuity of the frame structure. Each arch, taken separately, would have the form shown in fig. 20: these protrusions of large bricks, as it were, captured the mass of the filling and did not allow it to put pressure on the circles; at any rate, it is certain that the tight connection of the infill with these small protrusions of the frame ribs helped to transfer most of its weight to the arches, instead of allowing them to bear the full weight of the temporary circling structure.
The vaults shown in fig. 1 tab. II, are a characteristic example of the builders' attempts to get rid of the dependency and expense associated with the construction of a solid lattice frame, while retaining almost all the advantages given by the integrity of the structure: this vault is taken from the arcades of the aqueduct, which is considered the aqueduct of Nero and the remains of which are built into the walls of the gardens. stretching on both sides of the street leading to the Church of S. Stefano Rotondo in Rome.
In order to distinguish the structure shown in our drawing on the spot, rather intense attention is needed: the filling of the vault consists of fragments of tiles of the same color as the frames, and the frames themselves are so crudely made that, without knowing in advance about their existence, it is very difficult to notice them. in a mass reminiscent of veined rock, rock of the same shade, which envelops them and further complicates the examination, already difficult due to the ruined state and barbaric execution. I have already warned at the beginning that I must, for the sake of clarity, give in my drawings a certain regularity to the supporting structures that the Romans arranged; in this case, more than anywhere else, I had to allow myself this liberties; and more than anywhere else, this curious aqueduct shows how important the Romans attached to the speed of erecting these frames. We already know enough about the reasons for this extreme haste, but nowhere is it more clearly reflected than in the irregular forms of this arcade.
Such free-standing arches, as shown in our sketch (Fig. 20), were easily feasible, but due to their small cross section (about 0.15 m), their stability was in doubt: these arches could be deformed from a longitudinal bend in their plane or out of the plane; the Romans devised a way to compensate for their lack of resilience; they began to pair these arches, replacing the design shown in fig. 20, the one we see in Fig. 21.
A rib made of two arches paired in this way is nothing more than a narrow strip cut from a lattice frame, similar to that found in the Palatine: the grouping of arches, which increased their cross-sectional area, reduced the possibility of longitudinal bending. The advantages of the new design compared to the previous one were significant, and we see that these paired arches are widely used in a number of structures, of which we will name at least the Colosseum (Plate II, Fig. 2).
Figure occupying the top half of the table. II, depicts part of the galleries that form the outer enclosure of the amphitheater. The drawing shows two parallel and adjacent galleries simultaneously, the spans of which are almost identical; only one of them was erected on frames, while the monolithic masonry of the other was made directly on the circles.
Therefore, one should not consider the constructive technique that interests us as systematically used by the builders of the Colosseum: The Colosseum in relation to its structures is, so to speak, a huge summary of all the achievements of ancient building art, where all ancient constructive techniques were used in turn. Were the vaults rebuilt at different times, was its construction entrusted: simultaneously to several contractors who were given a certain freedom in the application of certain methods, be that as it may, but in different vaults of this structure, and sometimes in different parts of one and the same same vault, you can see the most opposite construction techniques. In general, the barrel vaults appear to have been erected over arches hidden in the masonry, the shape and placement of which are quite clearly shown in our drawing. However, no absolute law prevails either in the placement of these ribs or in their design: sometimes they start at the level of the height of the heels, sometimes, on the contrary, much higher; either their axes correspond to the axes of large architectural articulations, or (Plate II, Fig. 2) the arches resting on stone pilasters are located eccentrically with respect to the axis of the supports on which their heels are placed. With some diligence, architects could use these arches as a decorative element for their vaults, but they prefer, at the cost of inaccuracies in work, to eliminate the risk associated with too slow execution of these elements, intended only to ensure strength, so that then, after construction, to hide their irregularity. structures under a thick layer of plaster. This sloppiness is common to most of the frameworks we'll look at next; but before going any further, the true purpose of the framework we have just described must be analyzed more carefully.
It may be said to me that the functions of the brick frames of the Palatine (Plate I), which can serve as circles during the construction of the vault, are self-explanatory: it is a one-piece lattice structure that works as a whole; there is nothing more logical. Even in the aqueduct of Nero (Plate II, Fig. 1), where the arches, although very close together, but the brick ledges emerging from one arch, still do not meet with the ledges of the neighboring arch, it is clear that the brick frame can withstand the great weight of the masonry filling during the construction of the vault; but will everything be just as clear when the frame of the vault is reduced to a series of arches hidden in the masonry filling, to ribs not only located separately, but separated by intervals of about 3 m? Doesn't it seem that the arches here will simply bear the load of only the part of the infill that is located above them? But won't the filling, being in a semi-liquid state, rest on the formwork laid along temporary circles in the gaps between two arches, just as it would lie on it if these free-standing arches were absent? That's the doubt; I believe that it can be solved in the following way.
 |
| Rice. 22. |
Imagine (Fig. 22) a vault of a similar design, bounded at the top by a horizontal plane; in other words, imagine a vault whose construction has been suspended; let's assume that D and Ε
- two of its arched ribs.
It is clear that both of these arches, despite the empty gap DE, left between them, will be enough to carry the masonry filling the vault, if each of the horizontal rows of this array does not end with a straight line NAR, but a curve like an arc DBE: the result will then be achieved, no matter how irregular the rough fragments of which the horizontal rows of the vault are composed, provided that a sufficient lifting arrow is given to the various curves, like AB in an arc DBE. Accordingly, one can mentally divide each row of monolithic masonry into two parts: the part of the row located behind some imaginary line DBE, will hold on to itself, forming, as it were, a kind of horizontal arch, the seam lines of which converge in one common center and which rests on the ribs D and E. Fill part S between the curve DBE and the inner surface of the arch, it will be, as it were, suspended from the first, in some way stuck to it, thanks to the adhesion that the solution has until it completely hardens.
This explanation puts an end to the objections that could be based on the lack of integrity of the frames, and proves how little importance the Romans attached to the thickness and regularity of the wooden formwork boards, even when there was a very large distance between the ribs of the frame: for the formwork, the shape of the elements of which can be imagined in the many places where they left an imprint, long thin boards, which had many defects, were usually taken, as if carelessly thrown onto small circle farms. Their purpose was indeed not so much to support the stone structure as to serve as a form for it: at most, what they had to bear until the mortar hardened was an insignificant load on that part of the array, which is indicated by the letter S on our last schematic sketch.
The same construction of a frame made of individual ribs, but on a larger scale, we find in the Basilica of Constantine (Plate III). Above, vaults were considered that covered the gallery with a span of about 5 m, while the largest span of the vaults of the Basilica of Constantine is 23 m; this is almost the width of the nave of St. Peter in Rome.
With such a span, the vaults required load-bearing ribs of exceptional power; therefore, the architect, obviously fearing the insufficiency of such simple arches as in the Colosseum, attached to them the same additional girth arches, so that the ribs of the frame in the Basilica of Constantine consist of two brick arches located one above the other (Plate III and Fig. 24 ). This idea of such an arrangement of the ribs of the frame, in order to correspondingly increase the bearing capacity of the vaults of a very large span, was quite natural; meanwhile, would it not be better, instead of placing the arches one above the other, to place them directly next to each other, carefully bandaging them. In this case, the lining of the inner surface of the vault could be more satisfactorily completed and both a large bearing area and greater stability of the ribs would be provided, while the amount of brick used would remain the same.
This is true: it is true that such an arrangement of arches directly next to it did not change anything in relation to the consumption of bricks, but the situation was different with the costs of temporary circles. When two arches are located one above the other, as in the Basilica of Constantine, then only for one, the lower one, circles are needed; when this inner arch is installed, it can already serve as a support for the one that is thrown over it. On the contrary, if these arches are paired, placed side by side instead of one above the other, then both of them will simultaneously load the circles; and since their weight is approximately the same, the strength of temporary devices should be doubled. Thus, for the sake of saving on temporary circles, it was advantageous to do as the Romans did, that is, to make each edge of two overlapping brick arches.
It remains to be seen whether this achievement of saving on rims is depreciated by the fact that, with a given arrangement of arches, the danger of buckling increases.
There is no doubt that the arch with a span of more than 23 m and having a cross section of 0.60 m wide at the distance of the circles should collapse and collapse from its own weight. But when determining the strength that the frame of a vault made of crushed stone and mortar should have, it should not be set as a condition that the frame must maintain stability and flatter an additional load immediately after its construction is completed.
Indeed, it is not so important that the frame has sufficient strength at the moment when it is already built and completed, as long as it has sufficient strength and stability at the moment when it will be loaded with masonry infill of crushed stone and mortar? Meanwhile, if we consider the issue from this point of view, undoubtedly the only correct one, then we will make sure that the arches with a cross section width of 0.60 m fully satisfied their purpose, and here is why:
 |
| Rice. 23 |
1. During the entire period, while the masonry of the filling held by itself and did not load yet circled, the frame, of course, was not exposed to any risk, being, moreover, sandwiched between the beams by a wooden formwork, the purpose of which was to serve as a form forming the octagonal caissons of the vault ( Table III and Fig. 25).
2. Later, when the pressure from the weight of the masonry began to be transmitted, it gradually increased, very slowly at first, and then more and more vigorously, as the structure was raised.
By the time the pressure is transferred from the weight of the filling masonry to the arch (Fig. 23), the actual span of the arch AB was already significantly less than the span of the entire barrel vault. In addition, as the masonry of the vault rose higher, the working part of the arches of the frame gradually decreased and remained only on that segment of them that was not yet hidden in the masonry of the filling, and we see that the bearing capacity of the frame was constantly growing along with the load, which she had to endure; and it is quite possible that at the moment when the upper, still raw masses of the infill masonry were in great need of support, the span of those parts of the arches that were not yet hidden by the infill masonry from. crushed stone and mortar, decreased so much that at this value, the strength of the frame fully corresponded to the magnitude of the load.
In short, the strength of these ribs and their resistance to buckling increased as the span decreased. AB, i.e., as the need for resistance increased. Thus it is explained that such thin arches could serve as the ribs of the frame in the erection of one of the colossal vaults built by the ancients: such a result is undoubtedly a most remarkable achievement.
 |
|
| Rice. 24. | Rice. 25. |
If the design of the vault is perfect, then it must be admitted that the caissons decorating this vault are not linked to the distribution of the frame elements, which played an important role in its construction. I give on a large scale (Fig. 24 and 25) a detail of one part of the ribs of the vault of the Basilica of Constantine.
To the left (Fig. 24) there is an exposed rib, to the right (Fig. 25) the same rib embedded in the filling masonry. As can be seen in the figure, the ribs ran along the protrusions on the surface of the vault, separating the large octagonal caissons of the vault from each other, and in this respect their location was well chosen. But the architects who were entrusted with the decorative finishing of the building came up with the idea to fill the gaps between the large caissons with small square recesses, and for the sake of this fantasy, the builder was forced to make recesses in these ribs to a depth corresponding to the depth of the small square caissons falling on the ribs (Fig. 25) . He got out of the difficulty by a device which at first glance seems strange, but I think that he should not be condemned too harshly for these liberties in relation to architecture. To allow a contradiction between the architectural forms and the main structure of the building, to hide the skeleton, which is essential for the stability of the masses, is to create a work that condemns the mind, it is to show a lack of taste, offending the mind with a spectacle of obvious deception. But do we have sufficient reason to say that by hiding the brick frames of their rows, the Romans are hiding from the viewer one of the main structural elements of these vaults? I don't think so. What is, in fact, the structural system of frames of the Roman vault? Just a witty trick used during the production of work: these internal frames served only during the construction, they made it possible to bring out the vault, to give it the masonry of the vault acquires solidity; finally, after the solution has hardened, their independent existence, as it were, ceases, and they appear in the vault only as an integral part of it. From this moment on, the Roman architect no longer sees in this whole either a frame or fillings, but a homogeneous monolithic mass, and it is truly permissible for him not to emphasize the difference in the external decorative finish, which, in his opinion, has disappeared in the structure of the vault.
 |
| Rice. 26 |
That is why the cases when the ancients reveal the frame of the vault in the exterior of the completed structure are extremely rare; as an example of vaults in which complete agreement between the skeleton and external forms is achieved, I can only cite the barrel vault in the temple of Venus and Roma. Unfortunately, the entire upper part of this remarkable vault is destroyed, and the fragments of the lower part are insufficient and too much affected by time to build any assumptions on the basis of which it would be possible to restore its original appearance. I therefore cite, not as certain, but at least as very probable, those constructive elements that could be revealed to some extent by examining this code and which appear to me as they are shown in fig. 26.
The caissons were square in shape, and the directions of the ribs of the caisson coincided with the direction of the sides of the caissons, which are located continuously, some in the direction of the axis of the vault, and others perpendicular to this axis: all of them together formed a continuous lattice of large cells, some of the longitudinal sides of which are horizontal, and others coincide with the direction of the sections normal to the axis of the dome.
The transverse ribs of this vault have a smaller width than those in the Basilica of Constantine, but they are solid and not through, as in most other Roman buildings.
As for the method of erecting these brick frames, these protruding ribs, outlined in relief on the inside of the vault, it is self-evident. As shown in our drawing, the brick ribs, together with the formwork, probably wooden, formed a solid whole before the filling was laid: horizontal ribs reinforced the transverse arches; both of them, retaining their position thanks to the formwork used for the manufacture of caissons, formed a light vault between the circles and the filling masonry, partly wooden, partly stone, which played the role of a frame, similar to the role of a through-frame structure made of brick, shown in Table. I. Here we find complete agreement between the structural system and architectural forms; the architect accidentally used a brick frame as a decoration, but nothing forced him to do this, he was free to choose the architectural design; and the consistency of external architectural forms with construction, observed in the temple of Venus and Roma, is not, in my opinion, a serious evidence of the superiority of this building in comparison with others.
We have now examined the main types of frames, the masonry seam lines of which converge in one common center. Having now taken a general look at their uses, it will be possible, without needing any additional data, to evaluate both their useful functions and the results that they provide in the construction of vaults. But along with the advantages that they give, is there still no reason to consider their use associated with some dangers? These frames, immersed in the thickness of the masonry of the vaults, apparently formed in the still damp mass of crushed stone and mortar, as it were, an incompressible core; included in a monolithic stone structure, which settles on its own, without external influence, they may have interfered with the course of shrinkage and caused the appearance of large and small cracks. If this were indeed the case, then the frame systems that facilitated the erection of vaults would hasten or cause their destruction, but fortunately the situation is quite different. In fact, the masonry of the infill of the vaults is not a mass laid in one step, and it is curious how the gradual progress of erection in even, very thin layers reduces the risk of shrinkage; each layer very quickly acquires its final volume, each row shrinks in turn; and since the general shrinkage is eliminated, there is no longer any fear of cracking. However, this remark does not apply specifically to the type of frameworks that we have described below: it is applicable to another kind of structures, which we are now considering, and therefore we will not repeat it further.
b) Arches on frames made of bricks laid flat.
Compared with solid brick frames shown in Table. I, frameworks of free-standing brick arches, like those of the vault of the Basilica of Constantine, had the advantage of requiring less material; moreover, they fulfilled their purpose quite satisfactorily. However, even with the same costs, a solid frame is easier to implement, and therefore it was natural to strive to create a structure that, having all the advantages of a structure from free-standing arches, at the same time would create a continuous bearing surface; this seems to be the origin of the new frame construction, the use of which is found in Roman vaults.
These large bricks, laid on high-quality gypsum or quick-setting mortar, formed, as it were, a thin continuous shell over the entire convex surface of the formwork; this shell, reproducing the shape of the inner surface of the vault, was a kind of curved brick flooring (Fig. 27).
In some cases, the entire frame of the vault consisted of one such flooring, but usually another one was laid over it, similar to it, but consisting of smaller bricks, forming a second shell, firmly connected to the first layer of gypsum or mortar.
Thanks to this layering, a kind of protective crust was created over the entire surface of the formwork, a kind of light vault. ABCDE(Fig. 28), which could not be uncircled immediately after the completion of its construction without the danger of its destruction from its own weight (Fig. 29); it gained strength as the main vault was erected, until it was strong enough to bear the load of the infill masonry lying on top of it.
In fact, the reason preventing the immediate circling of this auxiliary vault was not so much the small thickness of its strong walls as its semicircular shape. The stability of a brick vault laid flat is ensured by two conditions: firstly, the outline of the vault in the form of a gentle circular arch with a very small lifting arrow, and secondly, its pinching in two unshakable supports. In the case of a semicircular outline, the rigidity of the arch is insufficient; to give it sufficient rigidity, you need to fill in the side parts of the arch AB and DE(Fig. 28). This backfill counteracts the bending of the vault and prevents its thin walls from collapsing under its own weight. In Roman vaults, similar vaulted floorings of bricks laid flat were apparently used precisely in such cases.
 |
| Rice. 29. |
The masonry of the vault did not yet load the circles, while its first rows were already clamping the brick auxiliary flooring to a certain level. BD(Fig. 28); that part of the brick vaulted flooring that was supposed to actually carry the load, i.e. its working part, reduced to a simple circular arch BCD, was in the best working conditions. At the moment when the masonry of the main vault reached the level BD, it was already possible to remove the circles and, if necessary, transfer them to another place, i.e., in other words, build a vault in parts and use the same circles in the construction of subsequent parts of the vault.
The Romans really often used this technique. To be convinced of this, it is enough to pay attention to the fact that the bricks of the vaulted flooring, instead of being laid alternately and forming a masonry with tied seams, are laid with through seams, like checkerboard cells (Fig. 27). This circumstance is quite consistent with the idea of erecting a vault in separate links: if we assume that the bricks were laid in a bandage, then the edge of each link would be jagged; this would cause some difficulty in connecting the links together. By abolishing in this way any connection, the Roman builders thereby eliminated all the difficulties of fitting.
Savings on circles do not require proof - it is obvious.
In accordance with the remark made above on a similar case, it is sufficient that the circles withstand the load of the weight of the vault alone; the first row of bricks serves as a formwork for the second row, and both together form a solid frame that carries the load from the masonry of the entire vault.
The vault detail shown in the following figure (Fig. 30) illustrates the application of the described vault design. This example is taken from the baths of Caracalla, which are perhaps the most significant building of all built according to this constructive system.
In this example, the first of two vaults is made of square bricks, with sides measuring 2 Roman feet (0.60 m) and 4 to 5 cm thick; the second flooring is made of smaller bricks - with sides of ⅔ of an antique foot or approximately 20 cm. In addition, a number of bricks are placed on edge in the thickness of the second flooring; these bricks form, as it were, butts or anchor protrusions on the outer surface of the vaulted flooring.
 |
| Rice. thirty. |
The purpose of the various parts of this peculiar construction is given in the preceding description, and the order of the work is quite obvious.
Instead of a continuous formwork, separate boards were stuffed onto a circular truss at a distance of 2 feet from axis to axis (Fig. 30); on these milkings, a flooring of large square bricks was hastily laid. Thus, the cost of the wooden planking of the circles was low, due to the large size of the bricks, the first course of the decking could be laid extremely quickly.
At the end of the laying of the first row, the second row could be laid with less haste from smaller bricks. Indeed, the second flooring is always made of small bricks; I know of only one example of the use of bricks of the same large size for both rows in the vaults of the Pantheon (covering wall niches, pl. XIII). The second row of bricks was, as we shall see later, to overlap the seams of the first row; the dimensions of the brick of the second row - 20 × 20 cm - corresponded well to this purpose.
However, it was necessary not only to design a supporting frame for the masonry infill of the vault: it was also necessary to provide some connection between this frame and the infill, so that after unrolling the whole structure would be a single monolithic array; it was for this purpose that bricks were used, placed on the edge, which were included in the masonry of the lower vaulted flooring at a certain distance from each other (Fig. 31). These bricks placed on edge, which served for communication, tended to tip over under the influence of their own weight; in some constructions of Hadrian's villa, they tried to prevent them from tipping over by laying small bricks leaning against the butts (Fig. 31).
 |
| Rice. 31. |
Such was the design of the vaults at the time of their erection; we should not, however, expect that in their ruins we will find it untouched. The vaulting of flat bricks has mostly disappeared; remnants of it can be found at the heels of the vault, in the incoming corners formed at the junction of the vault with the walls, in a word, in those places where these fragile vaulted floorings were best protected from destruction. In the spans of the arch, the double vaulted flooring collapsed; the original placement of square bricks can only be judged by the more or less clear imprints left by them in the monolithic masonry of the vault filling; everywhere only bricks have survived, placed on edge, now sticking out of the surface of the surviving remains of the vault (Plate IV, Fig. 2); in some cases, these butts and brick linings, embedded in the masonry filling the vault, survived and remained in their places, while only fragments of the entire flooring-frame were preserved.
Turning to the conclusions, we can say that, using a frame of brick laid flat, the ancient builders pursued two goals: first, to provide the masonry filling the vault with a solid and solid supporting surface; secondly, to ensure a strong connection between the frame and the masonry. We have just considered how they fulfilled this double condition in the vaults of two of the most famous buildings - the Villa of Hadrian and the baths of Caracalla; in ordinary cases, the type of frame used in their gigantic vaults could be greatly simplified, since its advantages could be achieved at a lower cost.
Let us now turn to the study of the improvements introduced by the Romans into this design in order to achieve greater economies in labor or materials.
On fig. 32 shows a frame that is closest in type to the two previous examples. The first vaulted deck is still solid, and the bricks of the second row only cover the seams of the first deck; in such a simplified way, the vaults of some halls of the Caesars' palace were laid out. Judging by the prints, the vaults of Sette Sale (a reservoir near the baths of Titus) were of approximately the same type. Such placement of bricks in the second row of the vaulted flooring combined the advantages that, while requiring less brick consumption, it provided a good connection between the frame and the masonry filling the vault.
 |
| Rice. 32. |
 |
| Rice. 33. |
Roman builders went further - instead of blocking all the seams of the lower vaulted flooring, they limited themselves to laying bricks only along the seams perpendicular to the axis of the vault (Fig. 33). Thus, the frame as a whole is a solid brick flooring, reinforced with smaller brick ribs, which, according to the builders, served at the same time to cover the seams and stiffeners.
This design is found in the vaults of several tombs on the Appian Way; on the table. IV, fig. 3 shows a perfectly preserved detail of the vault of one of the tombs. The size of the bottom deck bricks is 45 cm (11/2 ft) aside; the size of the bricks of the ribs covering the seams is only 22 cm. The gypsum, which served as an astringent, leached out over time, so that traces of the bricks of the vaulted flooring can hardly be detected. Its remains are more easily found in the ruins of the so-called Quintilian villa, preserved to the left of the Appian Way, not far from the tombs just mentioned.
In several other monuments of the Appian Way, the idea of using the upper vaulting only to bridge the seams is expressed even more clearly and openly; in these structures, the bricks of the upper flooring are no longer laid in a continuous layer, but are located at distances from each other (Fig. 34) and precisely in those places where the action of shaking or too much load could be destructive, i.e., at a common junction point four adjacent corners of the bricks of the bottom row of decking.
 |
| Rice. 34. |
To achieve even greater savings, it was necessary to completely abolish the upper deck. The Romans took this last bold step towards simplifying the design and achieved that they began to build vaults with a single-row flooring; however, the use of such a frame, consisting of a single layer of flooring, is relatively rare: in Roman barrel vaults, I have been able to find only one clearly expressed example in the so-called circus of Maxentius outside the gates of St. Sebastian (Porta San Sebastiano) (Plate IV, Fig. 1), where all the vaults on which the amphitheater was erected are made with a single-row flooring of large bricks.
 |
| Rice. 35. |
The use of vaulted brick flooring was widespread in ancient vaults; such frames are found not only in simple cylindrical vaults, but also in vaults of the most complex outlines; they were equally used in the vaults covering vast halls, as, for example, in the baths of Caracalla, as well as in the most modest vaults of the narrow aqueducts of the galleries; in this last case, the flooring is often reduced to two brick slabs measuring 60 x 60 cm, set at an angle and supporting each other; in fig. 35 shows the design of one of the many galleries of the aqueduct overlooking the arena of the Colosseum.
In other cases, instead of two inclined square bricks, they were limited to one horizontally laid slab, which served as a ceiling (Plate XIII).
The vaulted flooring of flat bricks served as a supporting structure not only for the vaults, laid out in horizontal rows of rubble and mortar; in cases where the Romans built even free-standing arches with radial seams, they invariably provided them from below for reinforcement with a similar brick flooring. As an example of the use of such an arch with radial seams, brought out with an auxiliary flooring, one can point to the porticoes of the amphitheater near the Church of the Cross in Jerusalem.
 |
| Rice. 36 |
The ceiling above the water pipe in the Baths of Caracalla belongs to the same type (Fig. 36).
Finally, I must pay attention to the four large vaults covering the lowered sides of the huge central hall in the baths of Caracalla. In the whole building, these vaults alone are made of masonry, the seam lines of which intersect at one point; it can be said that these four barrel vaults are the only ones not only in this building, but also among all the vaults of Roman structures that I examined in Italy. Their masonry consists of alternating rows of large bricks and tiles laid in mortar. On the table V depicts one of these vaults: the radial brickwork of this vault, as well as the layered monolithic masonry of crushed stone and mortar of other vaults, is brought out on a double vaulted flooring, which is similar in everything to the floorings of brick laid flat described above.
By all the examples given, one can judge the general nature of the auxiliary vaulted flooring used in ancient architecture as a frame - the supporting structure of the vault. These frames, so common in antiquity, are still used in Italy today. I have often been present at the laying of such vaulted floorings in those localities where they were used two thousand years ago with success, which the remaining ruins sufficiently testify to.
Such vaulted floorings are still often used and still are even in Rome itself; the closed vaults, which adorn modern villas, are mostly laid out on a floor of bricks laid flat, as are the vaults in the baths of Caracalla; the inner surface of the vault is usually formed by one row of bricks laid flat on a gypsum mortar, the rest of the masonry of the vault is a monolithic masonry of fragments of rubble stone and mortar.
 |
| Rice. 37. |
Over time, the meaning of the frame and backfill in the design of the vaults has changed. The Romans considered the brick frame only as an auxiliary structural element supporting the main body of masonry filling the vault; the latter was the main part of the structure, ensuring its strength and durability. Now the vaulted flooring has become the main, load-bearing structural element; in some modern vaults, this clearly expressed purpose of the main masonry of the vault - to serve only as filling - was revealed especially clearly: these vaults are only made from the bottom at the heels with the correct masonry on the mortar, while the upper parts of the masonry filling the vaults are simply backfilled with rubble. Italian masons call this kind of vault construction volte alla volterrana and sometimes give it the expressive name volte a foglio (leaf vaults).
In France, this vault design is now rarely used, but in the last century it was used frequently. The detailed description of these vaults given by Blondel deserves mention (see "Cours d" architecture ", t VI, chap. II) Flat, lowered vaults, which are the subject of our study, suddenly began to be used in French architecture in the 18th century In fact, their use was only the revival of an old tradition that has been preserved from time immemorial in the building techniques of Roussillon masons, a description of these techniques, see below.
Along the walls of the room, covered by a vault, longitudinal bars were laid, which served as supports for mobile circles, 21/2 feet wide (Fig. 37); along these circles a double flooring of brick laid flat was laid; the bricks of each row and both rows were firmly bound together with gypsum mortar in exactly the same way as in Italy, and in the same way as the ancient Romans did. When the part of the masonry attributable to the circular link was completed, the link moved along the guide bars for an insignificant distance (Fig. 37); then, on the same circular link, the next part of the vaulted flooring was laid, etc. all this, apparently, corresponded, being much more modest in size, to the masonry of ancient vaults.
It is quite obvious that such a design is fully consistent with the ancient Roman principles of laying vaults. Since the area where these vaults were used borders on the Roman colonies in Provence, it is possible that this method of laying the vaults is only a memory of Roman techniques. This similarity is so obvious that the above description of a completely modern masonry system is of great interest, especially in that it exhaustively confirms our conclusions based on the study of the ruins of Roman monuments.
2. Cross vaults.
So far, we have considered examples of barrel vaults. Turning now to the study of cross vaults, I would like to note their significance in Roman architecture, to clarify the question under what circumstances they were used, and to show examples of the use of the methods of laying vaults described above in them.
We know that, as a rule, the Romans avoided crossing vaults. In the amphitheaters at Arles and Nîmes we do not find a single groin vault, although their annular corridors and radial passages intersect in all directions; in the Verona circus, only a few cases of small barrel vaults intersecting can be noted; in the ruins of the Colosseum one is surprised at the negligible number of intersecting vaults with such a large number of intersections of countless galleries.
To avoid the intersection of the vaults with each other, the Romans usually placed the heels of one of the vaults over the top of the other vault (Fig. 38).
 |
| Rice. 38. |
In cases where such a solution was feasible, it eliminated all difficulties; but often the insufficient height of the galleries did not make it possible to arrange intersecting vaults at different levels, and involuntarily one had to resort to cross vaults.
Another circumstance in itself entailed the use of cross vaults: the Romans often had to overlap buildings with vaults, which consisted of a central and two side aisles. With this solution, there are only two possibilities to give access to natural light in the middle nave: either the vault should be raised at a sufficient height to place light openings below the level of the heels, or they should be punched in the vault itself. The Romans usually settled on the second solution: this is the origin of the cross vaults over the large nave of the Basilica of Constantine (Plate III) and the vaults over the two halls of the baths of Caracalla - over the central and another, perfectly preserved hall, which in the 16th century. was converted into the church of Santa Maria degli Angeli. In some cases, the use of cross vaults was not caused by structural requirements, but by the desire to add variety to the architectural composition. However, such cases are extremely rare, almost always the use of cross vaults was justified both by aesthetic considerations and by constructive requirements.
 |
 |
| Rice. 39. | Rice. 40. |
But we will not touch on the question in which cases the Romans used cross vaults - our task is to indicate what methods they resorted to when choosing their outlines and during construction.
Let us first of all consider what the outline of the ancient test vault was.
Giving preference everywhere to simpler solutions, the Romans sought to solve the cross vault in the form of the intersection of two cylindrical vaults of equal spans. Thanks to this decision, they could take circular curves for the outlines of the vaults and thereby avoid the elliptical outlines of circular trusses.
The Romans in rare cases strove for strict equality of spans of intersecting vaults; if there was a slight difference in the sizes of their diameters, they neglected it and limited themselves to placing the shelygi on the same level, preserving semicircular outlines in both vaults.
The central nave of the Basilica of Constantine was covered in this way (Fig. 39). For the total height of the intersecting arches, the size of the wider of them is taken; the section of the other arch is a semicircle with a raised center, the total length of the lifting boom of which AB is equal to CD. The fact that the heel of the less wide vault was somewhat raised did not in the least damage the appearance of the vault and even gave it a more elegant appearance. However, the difference in the dimensions of the sides of the vaulted building was often too great to allow this technique to be applied. In these cases, the Romans tried to bring the solution of the design of the cross vault to the solution of the design of the vault on a square plan; in doing so, they resorted to a very simple technique, shown in fig. 40.
Actually, only the square was covered with a cross vault ABCD allocated in the middle part of the room; the size of the side of this square was equal to the size of the smaller side of the rectangle, overlapped by a vault; parts of the rectangle not covered by the cross vault were covered by a continuation of the longitudinal barrel vault ( AE).
 |
| Rice. 41. |
This solution was very common, but it should not be considered the only one: the Romans did not at all abandon either the solution of cross vaults on rectangular plans, or the vaults with an elliptical section, which are the result of this solution. In the baths of Diocletian, three sections of one well-preserved hall are covered with cross vaults, the ratio of spans of which was approximately 2:3; in fig. 41 shows the plan of these arches, and their general view is given in Table. IX.
This vault is the most remarkable example known to me of solutions of cross vaults over an elongated rectangular plan; this example, however, is not the only one. The vaults of an elliptical outline lasted until the Byzantine architects, heirs of the traditions and aspirations of Roman art, applied in the classical cross vaults a very expedient technique, depicted in fig. 42.
Thanks to the new ingenious design of the vaults, greater or lesser irregularities in the plan no longer led to the complication of the outline of the vaults. The cheek curves could be semicircles (regardless of whether the sides of the overlapped rectangle were equal or not equal to each other); the vaulted room could be a quadrangle with unequal angles; the curves at the intersections of the vaults became arbitrary, and nothing prevented them from giving them a semicircular shape; all circles could be made in the form of semicircular trusses.
Having noted the connection that exists between Roman principles and Byzantine innovation, let us return to the study of ancient cross vaults and consider the methods of their construction.
Whatever the outline of the cross vault, the Romans simplified their construction, using techniques very close, at least in their basic principles, to the techniques used by them in the construction of barrel vaults. The design of the cross vaults, as well as the cylindrical ones, consisted of two independent parts: from a monolithic filling masonry and from a through brick frame or from a light vaulted brick flooring, which supported the filling masonry during the construction of the vault and thereby replaced, at least partially, temporary circles. .
 |
| Rice. 42. |
In those cases where the Romans carried out the laying of the cross vault on the vaulted flooring, they performed the corner ribs of the vault from large brick slabs; no matter how small the dimensions of the flooring bricks, these slabs never had less than 45 cm to the side; usually the size of their sides was 60 cm and the thickness was 5 cm. These ribbed slabs in most cases have not been preserved, but their size and shape can be judged from their imprints; mentally, you can reproduce the general view of the frame. On fig. 43 shows such a design of the vaulted flooring before laying the main masonry of the vault filling.
 |
| Rice. 43. |
This example is taken from the ceiling structure of one of the baths of Caracalla. Very similar variants of such a solution are found in the Caesars' Palace, Hadrian's Villa, etc. The issue of the design of the intersection of the vaults is resolved even more simply in cases where through frames are used. Ribs M and N were located along the lines of intersection of the vaults (Table IX), and, if necessary, additional arches were introduced R in the transverse direction from one abutment to another. The latter did not differ in any way from the brick arches used in the laying of cylindrical vaults. In the future, we will consider only the design of the corner ribs of the cross vaults (Fig. 44).
Three parallel brick arches, connected in pairs by baked clay tiles, formed a load-bearing skeleton located along the corner rib. For the final completion of this construction, it was only necessary to slightly trim the bricks so that the rib in shape corresponded to the protruding corner of the cross vault. The bricks were not pre-hewn according to the pattern, but were simply hewn on the spot. This simple processing cost almost nothing and did not delay the work.
Difficulties arose only when laying the upper parts of the diagonal arches. Without much difficulty, it was possible to close one of the arches, for example, an arch M(Table IX); but at the moment when it was necessary to carry out the adjunction of the arch to it N, difficulties inevitably arose: both parts of this second arch press on the arch from two sides M threatening to crush her. Obviously, before laying the last bricks of the arch N, it was necessary to fill the upper cells of the through arch M. Arch M with filled cells could already withstand the pressure from the adjoining parts of the arch N. Thus, the construction of the vault was completed without further difficulty.
 |
| Rice. 44. |
In this way, vaults were made in the baths of Diocletian. Typically, this design was used for vaults with a span of at least 15 m. In vaults with smaller spans, the load-bearing part of the structure becomes correspondingly lighter and lighter; the brick frame is gradually simplified in accordance with the reduction in the weight of the main masonry filling the vault. Following a logical series of possible changes, the Romans first abolished the intermediate paired arches of the arch type R shown in Table. IX; next they eliminated one of the three arches that formed the compound diagonal arches; finally, of these three arches, the Roman builders destroyed two, so that the frame of the vault was reduced to arches of a single section, running along each rib. Thus, in Roman architecture one can find all possible variants of the construction of vaults, which are transitional from a frame system to a system of vaults made of monolithic masonry without any frame.
Let's try to give examples of various types of brick frame structures found in this successively changing series:
1. In one of the galleries of the Palatina, located in the southern part of the hill (see Table VIII), there is a frame construction that, in its appearance, is closest to the frame that we have adopted as the main type. Rib arches are located exactly the same as in the baths of Diocletian; they consist of the same number of arches, interconnected in the same way. But in this case, due to the smaller size of the hall, the intermediate arches were considered superfluous. In other words, the design is reduced to that shown in Table. IX, minus intermediate arches R.
2. As an example of the use of diagonal ribs, consisting of only two arches, I will give a cross vault over the central part of the Janus Quadrifrons arch in Rome. The general view of the vault is shown in Table. VII, fig. one ; in fig. 45 shows a detail of a rib freed from filling masonry. After the previous detailed study, the order of the work is quite obvious: first, one diagonal arch was erected, without finishing the laying of the other; then two or three upper cells were filled with concrete, after which the laying of the second arch was completed.
 |
 |
| Rice. 45. | Rice. 46. |
3. Let us finally consider structures in which diagonal arches were allowed, consisting of only one row of bricks. An example of such a decision is found in the vaults of one of the halls of the Caesars' palace, the ruins of which, located separately on the Palatine site, rise above the recess of the Circus Maximus. Each of these diagonal arches (Fig. 46) consists of one row of narrow bricks, and the masonry of the arches includes large square tiles hewn in place. These nlites protrude from the arch to the right and left, and, entering the thickness of the monolithic masonry of the vault, thus provide a strong connection between it and the brick frame.
Having gone through a series of transformations, the design of the frame of the ancient vault came to its simplest form. A study of its further development during the succeeding centuries up to the present would take this work out of the study of Roman building art; we would have to go to the Middle Ages and consider the vaults of Western Europe, erected between the 11th and 17th centuries. In these vaults we find the same diagonal ribs and protruding double girth arches; but in this case the purpose of these arches is different. In Roman vaults, the framework matters only during the time when the masonry is not yet fully strengthened and needs additional support; after the final hardening of the frame masonry, it merges with the surrounding infill masonry and works in the same way with the entire masonry due to the adhesion of all parts. The Gothic frame, which was no less important during the construction of the vault, retained its independent significance even after the rounding; it fully bears the load from the filling of large hewn stone between the ribs and transfers this load in the form of a thrust, which is perceived by massive buttresses or backwater of flying buttresses. The systems of balancing in ancient vaults and in Gothic vaults are essentially different. The similarity between these types of vaults can only be established by comparing them at the time of construction; but under these conditions, the similarity is undeniable. Gothic arches give only a new interpretation of the main elements of the cross vaults of the times of the Roman Empire. A detailed study of the common features and differences between ancient and Gothic vaults goes beyond the scope of the task set in our work. We have given the main variants of frame structures in Roman vaults and will indicate in the next section how the same constructive principles were extended to vaults with a circular plan, that is, to domes and semi-domes.
3. Vaults on round bases.
Of all types of vaults, spherical vaults load the circles the least. Each horizontal section of such a vault is a closed ring, which itself tends to maintain balance. It is obvious that a dome with a plan in the form of a regular circle requires less construction of a strong frame than with an arbitrary plan consisting of irregular curves.
A number of ancient domes were erected with the help of simple wooden circles alone; an example is the vault of a large building erected at the gates of Rome in honor of the mother of Emperor Constantine.
However, these properties, which are a consequence of surface curvature, decrease as the radius increases. In domes with a span approaching that of the Pantheon in Rome, the curvature is so small that all the advantages arising from it lose any meaning. Even at smaller spans, the Romans appear to have been wary of the possibility of circling breaking under the load from the weight of the masonry; in cases where the span reached 20 m, they resorted to the construction of a frame, considering it capable of facilitating the work of temporary circles.
To facilitate the work of the circling, the Romans in some cases used a brick frame, similar to that depicted on panel I.
The implementation of this frame was hampered by the convex shape of the vault. I had to lay rows of bricks along meridians with changing directions. The dimensions of the framework cells changed all the time, successively decreasing. Obviously, these difficulties should have limited the application of this system. Domes of this design are extremely rare; of these, the most interesting is the dome of the building known as the Torre de Schiavi, to the left of the road leading from Rome to Praenesta. To avoid the difficulty caused by the reduction of cells, the use of a frame laid over the entire surface of the vault was replaced by individual meridional ribs dividing the vault into a number of sections in the form of spherical wedges.
An example of a vault of such a design is the vault of ancient terms adjoining the Pantheon in Rome; on the table. X shows part of the frame of the lower part of the vault; the upper part is difficult to recover due to the lack of accurate data. It is difficult to determine whether these belts of bricks suddenly broke off, resting against rings, as in the Pantheon (Fig. 49), or whether they intersected like ribs in cross vaults. Now the vault is cut in half by the street, and its surviving ruins do not provide more data than those that formed the basis of the schematic reconstruction of the vault shown in Table. X. These ruins are also of great interest from another point of view: it can be assumed that they are the remains of the baths of Agrippa and, therefore, date back approximately to the time, speaking of which Vitruvius barely mentions building materials from baked clay. If this assumption is correct, then the described example of the use of a brick frame in vaults is one of the oldest in the history of building art. The general appearance of the ruins does not contradict this: the entire construction, down to the smallest details, is made extremely carefully - the caring attitude and painstaking attention of the builder are felt in everything; caution in execution indicates the use of a new construction technique. With the acquisition of sufficient skills, the Romans began to pay less attention to the thoroughness of work; in this case, the successful solution of the design of the vaults fully corresponds to the excellent execution; in vaults of a later period one can find frames of lighter construction, but we will not find such careful finishing and forms of such impeccable regularity.
The dome of the building, bearing the controversial name of the temple of Minerva the Physician, is an example of the same vault solution, but differs sharply from that described in its rough execution. Part of this code is shown in Table. XI, and the general plan - in fig. 47; from this figure one can fully judge the incorrectness of this plan.
 |
| Rice. 47. |
The general composition of the building is quite clear: in front of us is a vault supported by small sails on a decagonal drum. The vertices of the polygon serve as the bases of ten arches dividing the dome into ten equal parts. Some of these spherical triangles are in turn separated by secondary arches. The entire construction as a whole is a well-decided frame scheme, understandable at first glance and does not need additional explanations.
However, upon closer examination, we will notice some uncertainty in the implementation of such a simple design and find strange errors in its details. The frame at the heel is extremely massive, exactly when determining its dimensions, an error was made in the calculation; then, at a height of several meters above the heel, it becomes much lighter - obviously, during work, the builders noticed the excessive strength of the frame and abandoned their original intentions for reasons of economy. The main arches, the supports of which are located at the vertices of the polygonal plan, are composed of five yoke-branches at the heel, and only three at the apex. The decrease in the number of branches could be explained by the desire to increase the cross section of the main arches in accordance with the increase in the cross section of the arch at the heel. This explanation in itself would be quite reasonable, but, given: the totality of the facts, the first assumption should be recognized as the only true one. In other words, the construction of the frame was undoubtedly > mutilated, due to the fact that its original design during construction underwent fundamental changes. This deviation from the main idea is especially clearly expressed in the execution of secondary arches located in separate sections of the dome.
In some sections we see two arches, breaking off almost at the very beginning; they have no constructive value because they are not closed; in other sections, only one arch is drawn, rising to an insignificant height and suddenly breaking off, and therefore just as unnecessary as in the first case; finally, in a number of sections, the builders, convinced of the uselessness of these auxiliary arches, completely abandoned them. Thus, in the case under consideration, we find sections in the same vault, subdivided by two open arches, separated by one arch, and, finally, sections without any articulating arches. In these arches, begun with masonry at the heel, then changed or finally interrupted, an indecisiveness uncharacteristic of Roman architecture manifested itself. The temple of Minerva the Physician was obviously built in the last years of the existence of the Roman state; both in terms of the plan and in the external appearance of this building, there are many features characteristic of an era close to the heyday of Byzantium. In the vaults of the terms of Agrippa, we see the emergence of new building techniques, and in the vault of the temple of Minerva the Healer, a decline. These vaults seem to embody. represents the extreme limits in the development of the building tradition, which lasted with amazing constancy throughout the long period of the Roman Empire.
It is worth mentioning how the techniques considered in relation to spherical domes changed in semi-dome vaults and vaulted ceilings of niches, and how structures with vaulted brick flooring were carried out in them. Tab. XI, XII and XIII give fairly clear answers to these questions: in Table. XII and XIII depict two different structures for covering niches with vaulted brick floors; on the table. XI - construction of ceilings of large niches with a frame of individual arches.
Attention should be paid to how successfully the expansion of the meridional arch directed at the mouth of the semi-arch is perceived by means of its end resting against the powerful cheek arch.
In spherical vaults, the execution of the frame is always a difficult job, and therefore the Roman builders, less than any other, considered it necessary to start it from the very heel of the vault; the entire lower part of the masonry was brought out to a certain level without any brick frame, sometimes even without any circles; at the same time, the curvature of the dome was controlled using only one cord, fixed in the center of the dome, the length of which was equal to the radius of the dome.
 |
| Rice. 48. |
Among other examples, one should cite the vaulted ceilings of the niches of the baths of Caracalla - it is very likely that they were erected in the same way (Fig. 48).
In order not to deviate from the task set before me - to get acquainted with the design of ancient vaults through a personal study of individual monuments - I should not mention the Pantheon, since its dome, being covered with a thick layer of plaster, is a system of caissons without any visible indication of the presence of a framework. However, in view of the extraordinary significance of this structure, I will nevertheless turn to this example, using the testimony of another person.
During the work on the repair of the vault under Pope Boniface, Piranesi took the opportunity to study the details. It was necessary to beat off and restore the plaster, damaged and crumbling over time in various parts of the vault; for this, movable scaffolds were installed that moved along the ledge of the cornice and rotated around an axis fixed at the top of the dome. This ingenious device made it possible for Piranesi, who immortalized the monuments of ancient Rome in his drawings, to study the entire inner surface of the vault to the smallest detail. In Piranesi's writings, we often find too loose assumptions, but in this case, his testimony deserves more credibility. The position from which Piranesi had the opportunity to examine the vault, to a certain extent, ensures the veracity of his image. The accuracy of the reproduction of parts that are visible today only partly confirms the accuracy of the image and those details that we are not able to see.
 |
| Rice. 49. |
Rice. 49 faithfully reproduces Piranesi's drawing of the construction of the inner frame of one-eighth of the dome.
In the Pantheon, as well as in the temple of Minerva the Physician, the framework of the vault consists of meridional arches. CC(Fig. 49). On unloading arches BB the load from them is transferred, which makes it possible to leave voids that facilitate the laying of the drum, and, finally, intermediate arches divide the part of the dome surface enclosed between two meridional arches into smaller parts. Thus, the purpose of the frame elements in the lower part of the dome is clearly visible from their designs.
Let us now consider the construction of a brick frame in the upper part of the dome. Comparison of two drawings (50 and 51), depicting two successive views of the construction of the upper part of the dome, shows the order of construction of the structure, apparently carried out in two steps.
Above the meridional arches CC usually ended as shown in the left figure (Fig. 50). Their desire to get closer was extinguished by a brick ring framing a round hole at the top of the vault, and the pressure from them was transmitted to the ring through eight touching arches.
The upper ring compressed by these eight arches could withstand the pressure of the meridional arches only up to a certain time; as the filling was laid, the force grew and threatened to crush the ring Ε
. Ring strength Ε
was considered sufficient as long as the masonry filling the vault did not reach the level Ν
; from that moment it was considered necessary to strengthen the entire frame structure of the upper part of the vault; laid out the second concentric ring SSS, which, like the ring that bordered the upper opening, was supported by arches OO, - was also supported by a system of arches, indicated in the right figure by the letters TT.
 |
 |
| Rice. 50. | Rice. 51. |
This is the origin of the arches TT and rings S, which make up the difference in figures 50 and 51. This interpretation is quite reasonable: the ring S, concentric to the ring bordering the upper opening, could not be carried out without auxiliary arches T; the latter, in turn, could not be erected until the filling reached the level N, because otherwise there would be no way to install them and how to perceive their thrust. In other words, the necessary sequence of erection of the upper part of the dome is quite justified and justified. At the beginning, the meridional arches rested with their upper ends only on the ring E; as soon as the masonry filling the dome reached the level N, this ring was strengthened by a ring S placed at some distance from it. With the adoption of such a sequence in the construction of the frame, its purpose and its entire structure, as well as the very order of work, become quite clear.
I cite this explanation as an assumption to be further verified, and draw the attention of researchers to those circumstances that can serve as an explanation for the questions that arise when studying this huge dome: nineteen centuries of its existence serve as the best proof of the correctness of the methods used; reliable knowledge and study of these methods would contribute to the development of building art and would illuminate an important fact in the history of ancient architecture.
The dome of the Pantheon rests directly on a round drum; such was the solution of the first Roman domes, such as, for example, the domes over the round hall of the baths of Agrippa (Plate X) and the domes over all the round rooms in the first years of the empire. The design on sails, which we mentioned when describing the dome of the temple of Minerva the Physician, penetrated into Roman architecture very late. Examples of its application are mostly related to the period of decline that came after the reign of Diocletian and preceded the heyday of Byzantium. In the temple of Minerva the Physician, sails are used - to move from a spherical arch to a ten-sided base; at Torre de Schiavi the dome was erected with rather crude sails on an octagonal plan. The dome of the central part of the tomb of Placidia in Ravenna, a monument closer to ancient than to Byzantine art, was erected on a square plan.
Thus, ceilings in the form of domes on sails gradually appeared in Roman buildings, from which in the 6th century, under Justinian, architects created a completely new, independent structural system.
4. Special types of vault construction; ways to give arches greater strength: the use of buttresses, etc.
The auxiliary structures of the frame type that we have considered, which were used by the Romans in the construction of vaults, can be divided into two types: we can include arched-type brick frames with radial seams, brick lattice ones into one type. frames and free-standing brick arches; the second includes vaulted flooring made of brick laid flat, and other types of auxiliary structures of this type. This classification, due to its great imperfections, cannot fully cover all possible solutions.
Often the Romans used only one of the indicated types of vault frame structures; sometimes we find in their buildings a combination of both types; an example of such a solution is the vault covering one of the halls of the Palatine (Plate VI) and representing a system of spring arches, brought out along the vaulted flooring from slabs laid flat. These two constructive systems complement each other, and the architect combined a solid flooring with a rigid frame of brick arches with radial seams in the vault design.
It can be assumed that the Romans did not recognize in their constructive decisions universal and rigid rules; they did not consider it possible, under the endlessly changing conditions of construction and the requirements for buildings, to use the same unshakable methods. In this regard, it is impossible not to notice a clear preference in the choice of certain building materials or methods in the performance of construction work: in Rome, brick frames are used in the construction of vaults; in Pompeii, for example, the frame is made of completely different materials, and the appearance of the vaults changes dramatically. The architect does not limit himself to using brick frames or vaulting with flat bricks; he introduces an auxiliary structure between the formwork and the masonry filling the vault, in which one should not, however, look for the likeness of that skillfully lightweight frame that we have described above. This construction is a continuous layer of tuff fragments and mortar, covering the formwork in the form of a shell, the process of which is similar to paving with crushed stone. The purpose of the framework of the vault here is an auxiliary thin vault made of almost unfinished materials, which bears the weight of the infill masonry, as in the case of a flat brick vault. This type of vault construction, most often found in Pompeii, is most clearly expressed in the vaults of the corridors of the arena, the galleries of both theaters and in the halls of the lower floor of the so-called house of Diomede, etc.
In Verona we will no longer find the use of tuff or brick; they are replaced by pebbles quarried in the river Ech (Adiga), from which a similar thin-walled vault is laid out, used to support the masonry filling the vaults of the corridors of the amphitheater.
In cases where the vaults have small spans and are at an insignificant height from the ground, the Romans change the methods of their construction and refuse to use circles and frames; they erect vaults directly on an earthen embankment, serving as a kind of formwork; in this way the arch found in the ancient cemetery in Vienna was built, the construction of arches in the basement of one of the main temples on the Palatine was carried out using the same method. In this case, the earth embankment, which served as a formwork during the construction of the vault, remained unremoved and was preserved in the form in which it was made by the builders.
We see how the ways of achieving savings on auxiliary devices are changing, while the basic principles of the construction of vaults remain unchanged; I want to show by a number of examples what various forms this idea took among the Romans when it was resolved.
So far I have described vaults with a curved bottom surface; the curvilinearity of the outlines of the circling in itself presented difficulties in work, and the Romans began to look for more economical solutions in the rejection of curvilinear outlines. We meet an attempt at such a solution in the theater in Taormina. The overlapping of large niches was made in the form of a lintel with a broken outline, which replaced the cylindrical vault (Pl. XV, Fig. 5). The easiest way to understand this extraordinary design is to imagine a lancet arch, made up of straight elements resting against each other; it is clear that with such a outline of the overlap, two thick boards resting against each other could serve as circles. This trick cannot be called an exception in the Roman building art: in the plain surrounding Rome, near the rounded end of the circus of Maxentius, I discovered antique structures that are modest in appearance, in which the section of the vaults, oblong in plan, is similar to these ceilings of niches in Taormina. The circles of such a simplified vault exactly correspond to the rafters of gable roofs. It seems to me that it is difficult to find a better example of the freedom with which the Romans found solutions on the basis of the principle of economy that I have tried to highlight.
Freely choosing examples of the implementation of this idea, the Romans did not miss any opportunity from which they could benefit. Realizing that the pressure on the circles from the weight of the masonry is much greater at the top of the vault than at its supports, they tried to apply masonry of various designs in the corresponding parts of the vault.
An example of such a solution is the double arch shown in Fig. 2 tab. XV; its lower part is made of solid masonry of large bricks, and the upper part is a brick frame filled with crushed stone and mortar. On fig. 1 of the same table shows the large arches of the lower floor of the Pantheon, the lower parts of which are tied together; the upper parts are three separate arches, laid out independently, without dressing; the lower arch was used as a circle for laying the upper arches.
The Romans, in addition, used the adhesive force of the solution and erected small vaults without any circles; in some plumbing galleries in Greece we find such a solution, and the overlapping of plumbing galleries in the porticos of Eleusis (Fig. 52) can serve as an example.
 |
| Rice. 52. |
Sectorial-shaped bricks here were laid in thick layers of mortar; the two lower bricks were laid quite simply; after they have already been installed in place and the mortar fastening them with the previously laid out part of the masonry has hardened, the capstone was laid in the place prepared for it; in this way, the masonry of the vault could be carried out without any auxiliary devices.
In the case of a concentrated load or the need to create a support for a transverse wall, it was necessary to strengthen a certain section of the vault structure; in these cases, the Roman builders abandoned the usual frame, hidden in the filling masonry, and resorted to the device of the girth arches protruding from the masonry; sometimes the heels of these arches rested on pilasters, but more often the Romans limited themselves to the fact that the arches protruded from the surface of the vault only in the upper part of the vault, while the lower parts of the girth arches remained hidden in the filling masonry (Fig. 53).
Thanks to this technique, in an overloaded area, the arch receives the necessary reinforcement; at the same time, pilasters are completely abolished, and the room is freed from unnecessary ledges, while the walls around the entire perimeter are given a continuously even surface.
There is no need to enlarge here the number of examples of these special devices and their application in particular cases; they clearly manifest the principle of reasonable economy, which is visible in all cases with the same clarity, despite all the variety of methods.
Considering that the questions about the methods of erecting vaults are sufficiently clarified, let's move on to considering the issue of the arrangement of supporting elements that perceive thrust. At first glance, it seems that this issue does not apply to the constructive systems of the vault that we are considering. Indeed, in these constructions, it is not so important for special devices to perceive that thrust, which usually occurs in an arch of wedge-shaped stones; the whole vault is a monolithic massive body, and the main task is to create strong enough supports that can withstand the pressure from the weight of the vault.
 |
| Rice. 53. |
The ability of monolithic arches to maintain their shape without any additional supporting abutments was, it would seem, their main advantage; this property of them is too elementary for the Roman builders not to notice it; they, however, did not lose sight of the dangers that this construction of vaults concealed. The erected vault is loaded gradually, and its deformations sometimes proceed for quite a long time; the top of the vault gradually descends, and its lower lateral parts tend to disperse. If the possibility of these movements is not prevented, there is a danger of serious damage as a result of these deformations; after their completion, internal stresses accumulate in the masonry of the vault, and the vault can be compared with a loaded powerful spring resting on two supports. It is clear that it is not necessary to put the masonry of the vault in such working conditions; it is necessary to deal with the appearance of deformations, and the best way to do this is to firmly fix the bursting elements of the vault with powerful buttresses. Such, in my opinion, is the origin of the buttresses used in ancient vaults. The fig shown here. 54 gives a clear idea of their shape, size and location.
The buttresses of the Church of Santa Maria degli Angeli, the Temple of Peace, and almost all great Roman cross vaults, with a few exceptions, have a similar appearance. In buildings with cylindrical vaults, buttresses are spaced less often and have a shorter overhang; in buildings with a round plan, the use of buttresses is an exception. This sequence, however, is so natural that it does not require additional explanations.
In general, the Romans used external buttresses on very rare occasions; taking care to ensure the stability and strength of the vaults, as well as other parts of buildings, they avoided such devices; instead of erecting special buttresses, they looked for solutions that would ensure the stability of the vaults by appropriately arranging the individual parts of the building. In this regard, a number of fruitful lessons can be learned from the study of the layout of large Roman structures.
 |
| Rice. 54. |
We will not give here a number of examples of such devices, equally understandable as well as ingenious, which, however, are not amenable to exact calculations; the direction of thought that guided the Romans can be considered fairly established. The essence of their methods is easy to understand in a detailed study of the plans of such large structures as the baths of Caracalla, Diocletian and Titus, Palatine and the like; you are convinced with what perseverance and with what various methods the Romans avoided work intended solely to ensure the stability of vaults; in almost all cases, the structural elements intended for this purpose are simultaneously used in connection with the main purpose of the structure.
In the case, for example, when a rectangular room is covered with a cross vault, the Romans place the heels A of the vault not exactly in the corners of the room, which would cause the device of protruding buttresses, but at some distance from the outer walls BC, as shown in Fig. 55.
 |
| Rice. 55. |
With this decision, the areas AB transverse walls were replaced by buttresses; in the case when the width of the room was less than its depth, the advantages of this solution are further complemented by the advantages of the design of the cross vault with a square plan of the room (see Fig. 40); buttresses are inserted into the premises, being part of the internal walls and increasing the usable area of the premises without additional costs. We meet such a solution in almost all cases of intersecting barrel vaults; a large number of remarkable examples of such a solution can be found in the Baths of Caracalla.
The plan of the Basilica of Constantine is an example of a different kind of solution to the same problem: the groin vaults of the middle nave had too large a span to not be strengthened by the construction of strong buttresses. Such buttresses are the transverse walls indicated in Fig. 56 letters A, B, C and D.
 |
| Rice. 56. |
However, these walls are not given the appearance of ordinary buttresses attached to the supporting pylons of a large cross vault; cylindrical vaults are thrown from one wall to another, which formed a space AB used as a side nave.
In this way, they ensured that the buttresses ceased to clutter up the building from the outside; they were no longer elements that were specifically designed to give strength to the structure, but were included in the usual solution in which the individual parts of the building mutually support each other, without causing the need for additional and unnecessary devices.
In those cases where there was a possibility of a free choice of means, Roman architects still instinctively settled on the simplest solution, which consisted in increasing the size of the vault supports, arranging, however, extensive voids in the thickness of these supports to save masonry when building larger stone massifs; this method was used in the construction of Agrippa's Pantheon (Plate XIII).
The walls of the Pantheon around the entire perimeter are a solid stone drum, lightened by a series of internal voids located one above the other, the placement of which I try to make clear by showing them without wall cladding hiding them.
In the gaps between these voids, which facilitate the masonry of the walls, and the covered arches, there are recesses in the form of niches covered with vaults, facing the bulge in the direction opposite to the direction of the thrust.
The Romans lightened their stone structures subject to thrust in two ways; they either left voids inside them, covered with cylindrical vaults, or arranged niches in them with semi-dome ceilings; similar constructive techniques can be found in the supporting walls of ancient vaults, in retaining walls (Plate XIV, Fig. 1).
In all these cases, their purpose is the same: by allowing to increase the total thickness and base area of the wall, they increase its stability without a significant increase in its cost.
Simultaneously with the construction of powerful stone massive supports, the Romans tried to reduce the risk of thrusting by using very light materials for the construction of vaults; in the construction of ancient vaults, pumice was constantly used; a large number of examples that confirm the use of pumice in precisely those parts of the vault where weight reduction is especially important, does not give us the right to consider this an accident. Most of the vaults in the Colosseum, in the baths of Titus and Caracalla are built of very porous volcanic tuff, from which all stones of dense rock have been carefully removed.
The brief description given in the compilation work of Isidore of Seville, apparently borrowed from one of the Roman authors, quite accurately sets out the custom of leaving the lightest building materials for laying vaults.
Another circumstance is often associated with the idea of lightening the vaults, but it, in my opinion, was given too much importance. This is the presence in the masonry of monolithic fillings of vaults of clay pots.
The insignificant part of the total volume of the masonry of the vaults that the pots usually occupy, and mainly the way they are placed, rather make it seem that their use is completely inconsistent with theoretical considerations based on the use of the light weight of these hollow pots. Indeed, if the Romans expected to reduce the weight, and hence the thrust, by introducing these pots into the masonry, we should find them in the upper parts of the vault, where the heavy weight of materials should be most avoided.
In reality, we do not observe this; moreover, most often we see quite the opposite.
The use of these earthenware pots can be studied from a 4th-century site named in connection with this Torre Pignatarra (Pot Tower); baked clay pots embedded in masonry were also found in the vault of the temple of Minerva the Healer (Minerva Medica) (Plate XI); Finally, I have examined the use of these pots in a number of tombs located along the Via Labicana, and mainly in the vaults of the circus of Maxentius, located behind the gates of St. Sebastian: in all these cases they were found placed in the lateral parts of the vaults. On fig. 1 tab. IV shows the placement of pots in the masonry of the last of the mentioned monuments; sometimes they are found in the masonry of the openings, but more often they are located directly above the supporting walls, and their number is increasing! in those places where their main quality - light weight - cannot be used at all. I met them even in the thickness of the wall; I will give one of many examples of such an unexpected placement: when studying the main facade of the temple of Minerva the Healer (Minerva Medica), you can find such a pot on its right side, slightly above the arch of the doorway, hidden in the masonry of the wall, directly behind the lining. In a word, it can be concluded from the above facts that when placing these clay pots, the possibility of using their light weight was not taken into account.
Apparently, the origin of the use of pots found in the masonry of Roman monuments can be explained as follows.
Liquid foodstuffs for the population of Rome were delivered to the city in clay pots; the townspeople did not have anything to send to them in exchange for the products they received, and a large number of such already used and little valuable dishes greatly embarrassed them. Together with the rest of the rubbish, they took these pots to what is now called Monte Testaccio (Pot Hill); this hill with such a characteristic name consists entirely of fragments of pottery. Builders came up with the idea to use this earthenware as a building material; these pots were an artificial material of excellent quality, not exceeding the cost of the rubble stone that they replaced. Due to the significantly lower weight of pots compared to ordinary stone, they were used mainly in the laying of the upper parts of the building. However, the desire to achieve their use of reducing the weight and load of the vaults seems alien to the Romans; we find such a solution in the buildings of Ravenna and Milan; it is difficult to decide whether the vaults, lightened by embedding clay pots in the masonry, are the Lombard architects' own invention, but in any case it can be considered most likely that this ingenious solution was not borrowed by them from the Romans. More plausible can be considered the assumption that this solution, applied in the dome of the church of St. Vitaliy (San Vitale), came to Italy in the same way as the architectural solution of this temple. This assumption thus attributes all the merit of the first conscious use of clay pots in the laying of vaults to the architects of the Byzantine school.
In general, when studying purely Roman buildings, it should be recognized that the use of clay pots in their history is secondary, and. the study of their application does not give grounds for any important conclusions that would supplement or clarify the principles we have outlined in our study.
 |
 |
| Rice. 57. | Rice. 58 |
One of the above drawings (Fig. 54) reveals one essential feature of ancient vaults: these vaults also serve as the top cover for the buildings covered by them; the Romans never built wooden rafter roofs over vaults. Roman builders, apparently, considered the protection of stone vaults by roofing over wooden rafters, that is, the use of construction from expensive, unstable and short-lived material, as a vicious system of duplication of construction. The Roman architect either uses roofing on wooden rafters, refusing vaults, or resorts to vaulted structures; in this case, yun does not make a wooden roof; vaults perform all functions: metal sheets or tiles are laid on their outer surface to protect against rain; sometimes the leveled flat surface of the vault is covered with a thin layer of greasy dense cement mortar (Fig. 57).
A number of vaults in the Baths of Caracalla belong to this type: the laying of the vaults at the top ends with an almost horizontal platform; the last layer of masonry is covered with a mosaic of colored marble and serves as the floor of a magnificent terrace.
In cases where the outer surface of the vault is covered with tiles or metal sheets, it is given the form of a roof with slopes, which it replaces.
An interesting example of such a decision is the vault of the temple of Santa Maria degli Angeli (Fig. 54). Inside it is covered with a number of cross vaults; if we imagine a special roof above each of the cylindrical vaults, then their mutual intersections will create just the same shape that is given to the outer surfaces of the vaults; the location of the valleys exactly corresponds to the ribs of the cross vaults; this solution most naturally and best of all ensures the free flow of rainwater. A similar solution is found in the Paris baths, in the Basilica of Constantine and others; only in the case of spherical domes, the shape of the outer surface corresponds to the convex shape of the dome, and the section along such a dome has the form shown in Fig. 58.
Such an exception to the general decision is quite justified, if we take into account that in order to create a horizontal external surface, it would be necessary to bring the volume of masonry to a volume significantly exceeding half the usable volume of the dome. The Romans saw in such a decision an unacceptable excess; in this we see one of the most characteristic expressions for the Romans of how, having a certain system of views, the principles of which cannot be absolute, they were able to refrain from extreme decisions arising from their usual methods.
In our study of ancient vaults, only the following questions remained unanswered. What ensured the safety of a number of vaults? What reasons led to the destruction of other vaults? Finally, by what methods did the Romans restore partial damage to the vaults and prevent their final destruction?
Among the reasons for the destruction of vaults made in monolithic masonry of crushed stone and mortar, one should first of all mention the influence of underground tremors and uneven ground settlement. As the next reason in order, it is necessary to note the destructive effect of large plants growing on the vaults; at first glance, it seems insignificant, but the Romans attached it very serious importance. The Roman laws reflect the measures that attempted to prevent this danger by establishing gaps between green spaces and aqueducts, for which the occurrence of cracks is especially dangerous. The Senate adopted a resolution prohibiting, starting from 11 BC. e., to plant plants at a distance of less than 15 feet from aqueducts; we learn about this from the treatise "On Aqueducts" by Frontinus, and three centuries later this decision is confirmed and receives even more clarification in the constitutions of Emperor Constantine.
Indeed, the danger they were trying to avert was very serious; it is difficult to imagine the size of those parts of the masonry that are peeled off by the action of the roots of plants. Perhaps only the devastation wrought by human hands can compare with the destructive effect of these imperceptibly acting forces.
Regardless of the causes of damage, the restoration of Roman vaults was carried out by summing up the second brick vault with radial seams.
In the vicinity of Rome there are a number of examples of aqueduct vaults reinforced with such an additional vault, erected from the inside and making up for the insufficient strength of the frame bearing the damaged masonry of the vault; rice. 2 on the table. XIV depicts such an arch erected from below, reinforcing the arch of the aqueduct.
The example shown in the figure is taken from the arcade near the Lateran, the ruins of which adjoin the Scale Santa chapel.
The method of erecting these auxiliary arches is as simple as it is ingenious. A new arch to support the cracked arch was erected without an exact fit to the surface of the old arch; a gap was deliberately left between the upper surface of the new and the lower surface of the damaged arch; this gap was laid only on one front side in such a way that a void was preserved between both arches, which was then filled with dense concrete, which formed between them, as it were, a gasket.
Such was the technique, which was sometimes simplified by the fact that additional arches were brought close to the cracked one - without this gasket. In this way, in my opinion, a number of monuments in Pompeii, damaged during earthquakes that preceded the great eruption, were restored. Apparently, the terms and the amphitheater were also restored in the same way. As a last example, I will give an ancient vault, known only by description, which was, as the original says, "supported by supporting arches" of double thickness, set on independent supports (Orelli, n° 3328). Another explanation of the arches of Pompeii could, if desired, be given, but the document I have just referred to dispenses with the need for a discussion on this subject, the results of which may not have been sufficiently definite; one may doubt the choice of interpretation of the purpose of the Pompeian arches, but with even greater right it can be argued that exactly the same arches were used by ancient architects to protect damaged vaults from collapse.
Regarding the sense in which the generality of scale must be understood here, as well as regarding our use of the conditional method of representation, see at the end of this work - notes to the tables.
Minerva Medica.
In proof of the authenticity of his image, Piranesi cites the following: he says that he depicted. the interior view of the dome (Fig. 49) as the dome appeared to him when it was cleared of antique plaster.
This vault has now been destroyed, for more on it see Le Blant, Monuments of Christian Literature in Gaul, vol. II, p. 125. , by which one can judge the methods of construction.
The described arches serve as unloading arches, transferring the load from the overlying parts of the wall to the solid parts of the base. They are almost completely filled with masonry, and it is quite obvious that they were laid by masonry after the laying of the arches along the circles was completed. Using this masonry as formwork would be a mistake; outwardly, it would give the impression of achieving unloading, but in fact we would have a single monolithic masonry, in which all efforts are transmitted vertically, as in the absence of an unloading arch.
"Sfungia, lapis creatus ex aqua, levis ac fistulosus et cameris aptus" ("Spongy stone formed in water, light and porous, suitable for vaulting"). Origin., lib. XIX, cap. X.
When studying the use of these clay pots in ancient masonry, one should remember about clay vases, which, along with metal vessels, according to Vitruvius, served to improve the resonance of large meeting rooms.
Such a comparison would, in my opinion, be purely coincidental. Indeed, as understandable as attempts to improve acoustics in theaters, they are just as redundant in the construction of tombs like Torre Pignatarra, or monuments along the road to Praenesta. In addition, Vitruvius does not say that these vases were walled up in the thickness of the walls of theatrical buildings; they were simply installed under the stepped seats of the amphitheater (Vitruvius, book V, 5, 1). Thus, drawing an analogy between these two cases of using clay pots is devoid of any basis.
See the description of hollow tube vaults in de Dartein on Lombard architecture, who has made available to me the results of his research, which helped me to shed light on the origin of hollow clay pot vaults. De Dartein believes that the beginning of this system of construction dates back at least to the 4th century; he notes its use not only in the church of St. Vitalius in Ravenna, but also in the Baptistery of Ravenna, restored and decorated by Archbishop Neon (423-430) and in a very ancient chapel near the church of St. Ambrose in Milan in the chapel of St. Satire.
frontin. Deaquaed., n. 126 and 127; Cod. Theod., lib. XV, tit. II, i. one ; cf. Cassiοd. Variaruir. lib. II, ep. 39; lib. V, ep. 38; lib. VII, form. 6.
Compare these indications of ancient authors with those of Alberti in the sixth chapter of the tenth book of his treatise On Architecture.
CYLINDRICAL Vault
In the architectural schools of antiquity, the outer surface of the cylindrical vaults covering the building, aligned with the slope with the help of plastering, directly bears the roof tiles; only in the Byzantine architecture of Ravenna can one hardly find a few examples of light vaults protected by a wooden roof. The latter device, a Romanesque innovation, probably of the Cluniac school, became common; the consequence of this was the saving of material and a decrease in thrust; the Cluniac barrel vault is predominantly a light vault covered with a roof.
outlines
Until the end of the XI century. the outline of the vaults is semicircular; when there is a need to increase the lifting arrow, they are content with raising the level of the heels of the arch. The only examples of an elevated oval arch known to us are found in Tournus and are probably inspired by some Asian pattern.
Arched vault, which is mistakenly dated Church Saint-Front attributed to the 10th century, is not found in any building that could be dated with certainty before the 12th century. In Issoire ( rice. 98, V), where the roof lies directly on the coating of the vault, the lancet arch was a means of reducing the massiveness of the masonry, and only in Burgundy was it used to reduce the thrust (C).
Note: The question of the time of the appearance of the lancet vault in the buildings of the Romanesque style remains controversial. If the church of Saint-Front at Perigueux cannot be attributed to the 10th century, then examples of earlier monuments are known than in Issoire: the old cathedral in Digne, which can be attributed to the end of the 11th century, has an ogival vault. Some researchers of Romanesque art, such as Kishera, prove on a number of monuments that the appearance of lancet vaults in France should be attributed precisely to the 11th, and not to the 12th century. See Lasteurie, cit. cit., p. 240.
|
Rice. 98 |
Cluniacs accept it from the moment (beginning of the thirteenth century) they begin to build barrel vaults, boldly placed on very high foundations, as in the main naves of the churches of the Charito on the Loire or the Paray le Monial; the balance of these vaults was unstable, and any decrease in thrust proved to be extremely important. The Cluniacs appreciated the advantages of the lancet arch in a static relation, to them we owe its use, which marked an era in the history of architecture, since this arch is a progress in balancing systems.
Methods of implementation and masonry
Romanesque architects, using stone as a material for arches, deprive themselves of one advantage - the ability to build without circles; The main difference between the Romanesque vaults and the Byzantine ones lies precisely in the fact that they were erected on circles.
The vault is never laid out in vertical successive rows; such masonry is justified only for brick vaults made without circles. Also, rows of masonry are never horizontal, like rows of small stone in Roman vaults; masonry in horizontal rows is associated with the idea of an artificial monolith, and the Romanesque vault is always lined with wedge-shaped stones.
Cylindrical vault with ribs
In most cases, the inner surface of the Romanesque box vaults is divided at some intervals by girth arches ( rice. 99). Usually these arches are independent of the vault masonry (drawing C); sometimes at the heels they merge with the masonry of the vault and then gradually protrude until, finally, at the very top of the vault, their lower surface becomes parallel to the inner surface of the vault itself (B). In some cases, girth arches even cut through the vault (A).
Obviously, these arches served to strengthen the rigidity of thin vaults. But they rendered a special service when laying the vaults: now it was necessary to fear not the breakage of the circles, but their deformation, while the presence of girth arches guaranteed against this.
First, girth arches were removed; they gave the circle system extreme rigidity, and the vault was erected already on the circles strengthened in this way. Rice. one hundred clarifies this explanation by showing the main cases from the practice of setting circled.
August Choisy. History of architecture. August Choisy. Histoire De L "Architecture
