University of Virginia Library

Search this document 
The Plan of St. Gall

a study of the architecture & economy of & life in a paradigmatic Carolingian monastery
  
  
  
  
  
  
  
  
 I. 
  
  
collapse section 
  
collapse section 
  
  
  

collapse sectionI. 
collapse sectionI. 1. 
  
 I.1.1. 
 I.1.2. 
 I.1.3. 
 I.1.4. 
collapse sectionI.1.5. 
  
collapse sectionI.1.6. 
  
 I.1.7. 
collapse sectionI. 2. 
 I.2.1. 
collapse sectionI. 3. 
 I.3.1. 
 I.3.2. 
 I.3.3. 
collapse sectionI. 4. 
 I.4.1. 
 I.4.2. 
collapse sectionI. 5. 
 I.5.1. 
 I.5.2. 
 I.5.3. 
collapse sectionI. 6. 
collapse sectionI.6.1. 
  
  
  
  
  
  
collapse sectionI. 7. 
 I.7.1. 
 I.7.2. 
collapse sectionI.7.3. 
  
  
  
 I.7.4. 
 I. 8. 
collapse sectionI. 9. 
collapse sectionI.9.1. 
  
  
  
  
collapse sectionI. 10. 
 I.10.1. 
 I.10.2. 
collapse sectionI. 11. 
collapse sectionI.11.1. 
  
  
  
 I.11.2. 
collapse sectionI. 12. 
 I.12.1. 
 I.12.2. 
 I.12.3. 
 I.12.4. 
 I.12.5. 
 I.12.6. 
 I.12.7. 
collapse sectionI. 13. 
 I.13.1. 
 I.13.2. 
 I.13.3. 
 I.13.4. 
 I.13.5. 
 I.13.6. 
 I.13.7. 
 I.13.8. 
collapse sectionI. 14. 
 I.14.1. 
collapse sectionI.14.2. 
  
  
  
  
  
  
  
  
collapse sectionI.14.3. 
  
  
  
  
  
 I.14.4. 
 I.14.5. 
 I.14.6. 
collapse sectionI.14.7. 
  
  
  
  
 I.14.8. 
 I.14.9. 
I.14.9
collapse sectionI. 15. 
collapse sectionI.15.1. 
  
 I. 16. 
 I. 17. 
collapse sectionII. 
collapse sectionII. 1. 
  
 II.1.1. 
 II.1.2. 
collapse sectionII.1.3. 
  
  
  
  
collapse sectionII.1.4. 
  
 II.1.5. 
collapse sectionII.1.6. 
  
  
  
  
  
collapse sectionII.1.7. 
  
  
  
collapse sectionII.1.8. 
  
  
  
collapse sectionII.1.9. 
  
  
collapse sectionII.1.10. 
  
  
 II.1.11. 
collapse sectionII.1.12. 
  
  
  
  
  
  
  
  
  
  
 II.1.13. 
collapse sectionII. 2. 
collapse sectionII.2.1. 
  
  
  
  
  
  
  
  
collapse sectionII.2.2. 
  
  
  
  
  
collapse sectionII. 3. 
 II.3.1. 
 II.3.2. 
 II.3.3. 
 II.3.4. 
 II.3.5. 
 II.3.6. 
 II.3.7. 
 II.3.8. 
 II.3.9. 
collapse sectionII.3.10. 
  
  
  
  
  
  
  
  
collapse sectionIII. 
collapse sectionIII. 1. 
 III.1.1. 
 III.1.2. 
 III.1.3. 
collapse sectionIII.1.4. 
  
  
  
collapse sectionIII.1.5. 
collapse section 
  
  
  
  
collapse section 
  
  
  
collapse sectionIII.1.6. 
  
  
  
collapse sectionIII.1.7. 
  
  
collapse sectionIII.1.8. 
collapse section 
  
  
  
  
  
  
  
  
  
  
  
collapse section 
  
collapse sectionIII.1.9. 
  
  
  
  
  
collapse sectionIII.1.30. 
collapse section 
  
collapse section 
  
  
  
  
collapse section 
  
  
  
  
  
  
  
  
  
collapse sectionIII.1.11. 
  
  
collapse sectionIII. 2. 
 III.2.1. 
 III.2.2. 
collapse sectionIII.2.3. 
  
  
  
collapse sectionIII.2.4. 
  
  
  
 III.2.5. 
collapse sectionIII.2.6. 
  
collapse section 
  
  
  
  
collapse sectionIII.2.7. 
  
  
  
 III.2.8. 
collapse sectionIII. 3. 
 III.3.1. 
 III.3.2. 
 III.3.3. 
 III.3.4. 
 III.3.5. 
collapse sectionIV. 
  
collapse sectionIV. 1. 
collapse sectionIV.1.1. 
  
  
  
 IV.1.2. 
 IV.1.3. 
 IV.1.4. 
 IV.1.5. 
 IV.1.6. 
 IV.1.7. 
 IV.1.8. 
 IV.1.9. 
 IV.1.10. 
 IV.1.11. 
 IV.1.12. 
collapse sectionIV. 2. 
 IV.2.1. 
 IV.2.2. 
collapse sectionIV.2.3. 
  
  
  
collapse sectionIV. 3. 
collapse sectionIV.3.1. 
  
  
collapse sectionIV. 4. 
 IV.4.1. 
 IV.4.2. 
collapse sectionIV. 5. 
 IV.5.1. 
collapse sectionIV. 6. 
collapse sectionIV.6.1. 
  
  
  
collapse sectionIV. 7. 
collapse sectionIV.7.1. 
  
  
  
collapse sectionIV.7.2. 
  
  
 IV.7.3. 
 IV.7.4. 
 IV.7.5. 
 IV.7.6. 
 IV.7.7. 

I.14.9

CONFIRMING EVIDENCE:
THE PALACE GROUNDS AT AACHEN

We do not know whether or to what extent, if ever, the
modular construction methods used on the Plan of St. Gall
were implemented in any of the larger monasteries built or
rebuilt during the reign of Louis the Pious. But one other
site, a physical and historical reality, can be said to have
been organized along similar lines: Charlemagne's Palace
at Aachen. While Ernest Born and I were working on the
Plan of St. Gall, Leo Hugot of Aachen made a meticulous
dimensional survey of the palace grounds and its buildings.
None knew of the other's work, which was simultaneously
and for the first time displayed in 1965 at the Council of
Europe Exhibition Karl der Grosse in Aachen, each in the
form of a model, together with two brief explanatory statements
that formed part of the official Exhibition Catalogue.[387]

Charlemagne's Palace at Aachen was established on the
ground of an old Roman settlement that—like most other
Roman provincial towns or military camps—was inscribed
into a large rectangle, internally divided into four equal
quarters by two streets intersecting each other at right
angles, conditions that are even today mirrored in the
course of certain streets of the city of Aachen (fig. 71.X).
In the southeast corner of that rectangle were the famous
hot springs, the amenities of which were one of the primary
reasons for the selection of this site as the first "permanent"
residence of the great ruler. In laying out his residence,
Charlemagne did not follow the Roman dispositions
blindly. He changed the alignment of his buildings so that
the axis of the Palace Chapel would run from west to east,
as the liturgy demanded, and since all of the other buildings
of the Palace were either parallel or at right angles to the
church, this system of building came to lie athwart, at an
angle of 38 degrees, the Roman street system. As the
Romans had done with their settlement, so Charlemagne
also inscribed his residence and its buildings into an area
of rectangular shape (fig. 71.Y). In laying out his grounds
he availed himself, as the dimensional survey of the site by
Leo Hugot shows, of a modular base value consisting of a
rod 12 feet long. He placed the Palace Chapel (fig. 71.Y, 1)
against the southern edge, the Audience Hall (fig. 71.Y, 5)
against the northern edge of a large open square, each side
of which had a length of 30 rods = 360 feet. Internally
this square was composed of 16 smaller modules, each of
which formed an equilateral square of 7 rods = 84 feet.
The Palace grounds were intersected by two streets which
crossed each other at right angles dividing the site into an
outer and an inner court. These streets were each 2 rods
broad = 24 feet, bringing each side of the square to a total
of 30 rods = 360 feet. Hugot's analysis of the square grid
of the Palace grounds (Hugot, 1965, fig. 2 facing p. 524)


105

Page 105
did not include the site of the Emperor's Audience Hall
(fig. 71.Y, 5) or the site of the southern annex to the Palace
Chapel; the so-called Secretarium (fig. 71.Y, 4). Ernest
Born's analysis, superimposed in red on Hugot's plan
shows that with these two buildings included, the whole of
the Palace grounds could be conceived as being inscribed
into a rectangle, measuring 30 rods in width (30 × 12 =
360 feet) and 52 rods in length (52 × 12 = 624 feet). He
emphasizes in his caption to fig. 71.Y the importance of the
use of the sacred numbers 3, 4, 10, 12, 30 and 40 in the
construction of this grid.

The proportions of the Palace Chapel, Hugot's analysis
has shown, were as carefully and consistently regulated as
the layout of the entire Palace grounds (figs. 71.Za, b, c).
The base module, again, is a rod 12 feet long. The chapel
itself was fitted into a square, each side of which measured
84 feet = 7 rods (fig. 71.Za). To this cube was added in the
east a choir 24 feet deep (2 rods), and on the entrance side
a westwork of identical depth. In the vertical plane the
84-foot cube reaches from ground floor to base of the
pyramid. The total elevation is composed of: height of
outer wall, 48 feet (4 rods); height of gallery roof, 12 feet
(1 rod); height of drum, 24 feet (2 rods).

The height of the pyramidal roof of the octagon, in this
context (the original pyramid has disappeared) can only
have measured 2 rods = 24 feet. The total height from the
ground to the apex of the structure, accordingly would be
9 rods = 108 feet.

This is not the time to go into the structural aesthetics of
this important building.[388] But I cannot forego the pleasure
of remarking on the implications Hugot's findings had with
regard to our own work. First of all, it removed whatever
residual doubt Ernest Born and I may still have entertained
concerning the correctness of our interpretation of the Plan
of St. Gall. Second, it added new weight to the arguments
which we have advanced in a preceding chapter concerning
the provenance of the original scheme from sources close
to the Court School as well as to Bishop Hildebold, the
titular head of that school (791-819).

Bishop Hildebold's church at Cologne, as has been shown
in a preceding chapter, served as model for the Church of
the Plan of St. Gall.[389] The site organization methods, which
Leo Hugot has shown were used for the Palace of Aachen,
are greatly akin in spirit to those which we have shown to
have been used in the layout of the Plan of St. Gall. This
kinship suggests that the designer of the latter was not only
familiar with, but in all likelihood inspired by the former.
He might even have had access to the original drawings
used for the Palace grounds and its buildings.

The importance of Leo Hugot's findings about the
modular construction methods used in the layout of
Charlemagne's Palace at Aachen can hardly be overemphasized.
It is here for the first time in the history of
medieval (and possibly Western) architecture and site
organization that not only the grounds, but also the most
important building on these grounds, the Palace Chapel,
are controlled by a binding and all-pervasive rule of modular
prime relationships. We shall return to this problem
later on in a discussion of the possible historical roots of
this concept.[390]

[ILLUSTRATION]

AACHEN, PALACE CHAPEL. TRIBUNE

DETAIL, BRONZE RAILING, CA. 800

Eight such railings, each 4 feet high and nearly 14 feet long, were each cast in one piece:
an astonishing accomplishment of the Carolingian Renaissance. Roman grille work patterns,
Byzantine acanthus leaves, and Hiberno-Saxon scroll-and-grid motifs are subsumed in
a sophisticated medieval linearism.


106

Page 106
[ILLUSTRATION]

71.X AACHEN: THE CITY CENTER. A CADASTRAL PLAN AFTER 1800 WITH
CHARLEMAGNE'S PALACE GROUNDS

Superimposed in red are the grounds and buildings of Charlemagne's Palace as well as the reconstructed street
system of the Roman town of Aquae Granis
(redrawn in part by Ernest Born, after Hugot, 1965, fig. 1 facing
p. 534, and with the aid of Hugot's original drawing which he generously made available to us for that
purpose
).

Like most Roman towns or military camps this settlement was internally divided into four quarters by two
main arterials intersecting each other at right angles. The course of these, as Leo Hugot has shown, are
recognizable in the street alignment of the modern city of Aachen. They are: from north to south, the
alignment Kockerell Strasse—Klostergasse and Kleine Marschier Strasse
(A-B); from west to east, the
alignment Jakob Strasse and Grosse Koln-Strasse
(C-D). The latter was part of a Roman road that led from
Herleen to Kornelismünster into the Eifel Mountains; the former of a road that connected Liège with Julich
and Cologne.

The city of Aachen (French: Aix-la-Chapelle; Italian: Aquisgrana) owes its name to a Celtic settlement and
sanctuary that had sprung up in the vicinity of a cluster of sulphur hot springs which the Romans after their
conquest of this territory, in the first century A.D. converted into a watering place for legionaries, pensioners,
and other civilians visiting or settling there for recreational purposes or for reasons of health. The Romans
referred to this location as Aquae Granis =
"the Waters of Granus" (a Celtic deity worshipped in connection
with hot springs
). In ancient times, as well as in the Middle Ages and up to our own days, this kind of spring
was believed to have a curative effect on such afflictions as gout, arthritis and scrofula.

At the collapse of the Roman empire the city of Aquae Granis was destroyed (presumably around 375) but the
life of the native population appears to have continued. King Pepin bathed in the springs of Aachen in 765
and ordered the baths to be cleaned. Charlemagne signed deeds in
AQUIS PALATIO PUBLICO in 768 and 769,
and between 777 and 786 rebuilt and enlarged his father's palace. The site became his favourite winter
residence from 794 onwards, a date which marks the turning point from ambulant government to rule from a
central seat of government, at least during the winter months, the summers continuing to be taken up by
warfare. Charlemagne furnished the site with a royal Audience Hall and a monumental Palace Chapel
(fig.
71.Z
), a residence for himself (location and details of construction unknown) as well as a considerable number
of lesser buildings to house his court as well as his bodyguards.

The Palace, as was to be expected, gave rise to the growth of a vast cluster of subsidiary establishments,
mainly to the north and to the west of the royal court, and acquired the appearance of the town when all of
these
VICI, together with the Palace, were surrounded in 1172-1176 with a wall by order of Emperor Frederic
Barbarossa.


107

Page 107

108

Page 108
[ILLUSTRATION]

71.Y SITE PLAN OF THE EMPEROR'S AUDIENCE HALL AND THE PALACE CHAPEL AT AACHEN,
BUILT BETWEEN 796 and 804

KEY TO NUMBERS OF PLAN

1. THE PALACE CHAPEL

2. ATRIUM

3. NORTHERN ANNEX

Metatorium where the emperor changed his attire before entering the chapel

4. SOUTHERN ANNEX

Secretarium for assembly of the clergy and the holding of synods in contemporary sources referred to as In Laterano

5. THE EMPEROR'S AUDIENCE HALL

6. ENTRANCE HALL at ground level connects the OUTER COURT with the INNER COURT

7. BARRACKS for the emperor's guard

Superimposed on Hugot's drawing of a part of the Emperor's Palace at Aachen is shown the rectangular schematism on which its plan is
based. The basic module, as Hugot's dimensional survey of the site has demonstrated, measures 12 feet. Seven basic modules comprise one
major unit. The length of the plan is 7 major units, plus 2 basic modules for the east-west street, plus 1 basic unit for the projection of the
north apse of the Emperor's Audience Hall. Thus the length of the rectangle is 7 × 7 plus 3 units, or 52 units; its width is 4 × 7 units plus
2 units, or 30 units—a rectangle 624 feet long by 360 feet wide.

What captures our attention is the prevailing scheme of numerical and space relationships. The sacred numbers 3, 4, and 7, with 10, 12,
and 40, represent values of measurement that govern or control lines and critical grid relationships and are suffused into the fabric of the
plan to form interrelated kinships. For example, the length of the plan, 52 units, saturated with 75
[(7 × 7) + 3] is the sum of 40 plus 12.
The width of the plan, 30 units, is the product of 3 × 10 and, with 12 as a multiplier, creates 360 feet
[(3 + 3 + 3) × 40]. Such a collection
of
NUMERI SACRI offers striking evidence of the presence of sacred numbers as a dominating influence in the mind of the Carolingian
planner at the heart of the Empire, at this period in the development of western civilization. The aesthetic consequences of this pervasive
geometric and numerical schematism is another and different problem. Too, it is not without interest that the 3-4-5 triangular relationship
for the formation of a right angle is consonant with the 12-foot grid and would facilitate, in the field, layout for building foundations. In
monumental building schemes in particular, and all building in general, this would be advantageous to both the architect and his director
and construction foreman on the site.

Thus, rectangular schematism, identifiable with esoteric sacred numbers, was well-tailored in some respects to the practical needs of a
builder whose responsibilities, far removed from finely woven webs of theology, were characterized by mundane objectives, intolerant of
hocus-pocus.

E. B.


109

Page 109

110

Page 110
[ILLUSTRATION]

71.Za AACHEN, PALACE CHAPEL, BUILT BETWEEN 796 and 804

PLAN AND ELEVATIONS BY LEO HUGOT
GRID SUPERIMPOSED BY AUTHORS

The interesting and developmentally crucial significance of Leo Hugot's discovery of modular principles governing the organization of the Palace
Chapel at Aachen is its demonstration, by implication, that even a centrally planned Carolingian building cannot escape the power of
transformation which, at the age of Charlemagne, converts the Early Christian basilica into a modular structure.

The design of the Palace Chapel at Aachen is based on that of the church of San Vitale at Ravenna, begun around 532 and finished in 546.
In both buildings the shape and composition of the primary spaces as well as the formation of the basic morphological features are essentially the
same. In each case a tall octagonal center space with fenestrated and dome-surmounted drum is encircled by a peripheral envelope of outer
spaces, which, although internally divided into two stories
(ambulatory and gallery) are, in their combined height, lower than the center space.
In each case the drum with its dome rests on eight huge arches, rising from piers erected in the eight angles of the octagon
(cf. fig. 162.B).

These are the basic compositional and structural features. Yet in style and spatial concept the two buildings differ distinctly. In San Vitale the
shell separating the octagon from its spatial perimeter is made up of semicircular niches that billow out into the body of ambulatory and gallery.
Despite their intensive perforation
(three arched openings on each level) these curved niches, together with the piers to which they are attached,
are aesthetically perceived as a continuous sheet of masonry stretched around the center space. The movement is encircling, not divisive, and the
enveloped space, being uninvaded by any of the enveloping features, retains its full corporeal solidity and homogeneity. It is sculpture springing
from a concept of spatial mass akin in spirit to the monolithic self-containment of the component spaces of the Early Christian basilica
(Cf.
figs. 174 and 177.A-C
).

The stylistic and conceptual archetype and prototype of this manner of molding space is the Roman Pantheon, a body of incomparable globular
perfection and beauty contained in a masonry shell of simple cylindrical shape whose surfaces pass in unbroken planar continuity into those of
the semicircular dome by which it is surmounted, with no intrusion at any point. The camera—one-eyed and stationary—is incapable of
capturing this quality of style, but Giovanni Piranesi, with his uncanny sensitivity for such matters, has portrayed it with great perspicuity in a
series of masterful engravings.

The design of the Palace Chapel at Aachen by contrast is based upon the concept of spatial divisibility. This quality is strikingly reflected in the
manner in which the eight component surfaces of the octagon meet and connect with one another. Instead of billowing niches swinging inward and
outward, yet never losing their encircling hold, the Chapel's straight surfaces, separated by sharp lines, rise in undisrupted ascent from the
ground to the apex of the vault by which it is covered. The dome over the octagon of San Vitale is circular
(or nearly so) and therefore
detaches itself distinctly from the octagonal shape of the body of space lying beneath it, the transition from octagonal drum to circle of the dome
being achieved by means of squinches. It rests or hovers like a protective lid over the space it covers. The dome over the octagon of the Palace
Chapel, by contrast—a cloister vault, not a hemicycle!—is segmented into eight separate parts, like the eight sides of the octagonal shell that
supports it. The emphasis thus is shifted from connecting surfaces to separating lines. In Aachen, for this reason, the center space conveys the
feeling of being
"sliced" or "sliceable" rather than "whole" and "rounded." It could aesthetically be defined as an aggregate of triangular
prisms, meeting with their sharp inner edges in the center axis of the building. This is divisive Carolingian modularity: the conceptual equivalent
of the modular square division of the Carolingian basilica
(figs. 166-173); medieval divisionalism versus Classical corporeality.

The differences are discernible with even more striking sharpness in the structural articulation of the outer spaces. In San Vitale, ambulatory
and gallery were covered by timber roofs formed by continuous sequences of beams or trusses all lying on the same level, and therefore visually
perceived as flat and continuous annular planes
(the present vaults are medieval; see Krautheimer, 1965, 170). In the Palace Chapel at Aachen
the same spaces are covered, on ground floor level by groin vaults of alternating square and triangular shape; and on gallery level, by rampant
barrel vaults alternating with sharply defined triangular spaces. This is cellular medieval organization of space, springing from the same
conceptual sources that in the longitudinal layout of the basilica lead to the arch-framed bay division of the Romanesque and the Gothic

(fig. 177); and for more visual demonstration, Horn and Born, Viator, 1975, figs. 38, 39.A-B.

Centrally planned buildings do not lend themselves with the same ease to modular division and, for that reason, are not part of the mainstream
of medieval development. One hundred and fifty years after the construction of the Palace Chapel, a sophisticated Florentine architect created in the
Baptistery of that city, a synthesis between classical and medieval, disclosing that even south of the Alps materials inherited from antiquity are
reshaped in a similar manner
(cf. Horn, 1938, 99-155; reprinted 1973).


111

Page 111
[ILLUSTRATION]

71.Zc FRONT ELEVATION

[ILLUSTRATION]

71.Zb SIDE ELEVATION

 
[387]

For the Plan of St. Gall, see Horn, 1965, 402-10 ("Das Modell eines
Karolingischen Idealkloster") and idem, 1965, 391-400 ("La Maquette
d'après le plan de St. Gall"). For the Palace grounds at Aachen, see
Hugot, ibid., 395-400 ("Die Pfalz Karls der Grossen in Aachen") and
385-390 ("Le palais de Charlemagne à Aix-la-Chapelle") as well as the
more detailed and more comprehensive analysis in Hugot, 1966, 534-72.
We acknowledge with profound gratitude Dr. Hugot's generosity in
allowing us to make use of his original drawings in the preparation of
the red overlays reproduced in figures 41.Y, 71.Y and 71.Za, b, c.

[388]

For a recent discussion of the Palace Chapel, see Kreusch, 1966,
463-533 where all previous literature is cited.

[389]

See our chapter "Speculations about the Prototype Plan," above,
pp. 27ff.

[390]

See our chapter "Square Schematism," below, pp. 212ff.