The Architectural Theory of Francesco di Giorgio

Author(s): Henry Millon

Source: The Art Bulletin , Sep., 1958, Vol. 40, No. 3 (Sep., 1958), pp. 257-261
Published by: CAA

Stable URL: https://www.jstor.org/stable/3047782

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

CAA is collaborating with JSTOR to digitize, preserve and extend access to The Art Bulletin

This content downloaded from

128.103.147.149 on Fri, 31 Mar 2023 01:59:44 UTC
All use subject to https://about.jstor.org/terms
 NOTES 257

applied to in the
so distinguish
contract that payment could be made to an agent.
all, 1272, in
Moreover, any work onMaestro
the altar of San Jacopo was
simple work for
not something that even the
a man of his importance could al
posite sideregard
oflightly. Whatthe
he was asked to dochurch
was intended
On July to 0o,
change its appearance completely and to make it wh
1273,
tis and Ranuccio Bonaju
compatible with the place that the altar already held in
contract the devotion of the
was people of the city-then very im-wit
signed
the Imperial
portant as a bankingnotary,
center. The expense involved was in
the presence of
large enough so that the operai the
of 1272 did not findprio
the church for
means to provide which,
for it-their successors were evidently t
mo had made
more resourceful andamet the pulpit.
expenses promptly. Per- A
priest, Grandebene,
haps the explanation lies with the bishop, Guidaloste wh
actions in Vergiolesi,
1265 a man of vigor and in power. conne
on the great chalice
It is true that the cost of the work, one hundred lire, an
chaplain of San
is given in terms Jacopo
of common Pisan currency so that it
In October is not as much are recorde
as it sounds. Calculated by the rate for
was on hand
the year 1270, whichand we know in terms of theagain
Sienese
piece of florin, one hundred "small" Pisan
marble lire would still be
furnishe
However short we
over sixty gold florins--a considerable sum at that may
time.
ber 13 ofMasterthatBuono, whose work as asamemason we have men- y
Petri, de tioned,
Cappella sanc
in the latest expense account I could check, was
scultore di pietre
getting for himself and a workman foure marm
soldi a day-at
for, in that rate it would take the two of them, master
accordance withand
is then paid one
manual laborer, a year hund
and eight months to make what
fixing Niccolo wasor
over paid for four rather
months' work. Why, then, r
added a smaller amount
should we minimize its importance?
lead whichIt is hard
he to discover supplied
what became of Niccolo Pisa- b
action took
no's altar. place
Presumably, again, it was in the
only added to,
der and a i.e.,messenger
covered up, in 1314 when the silver altar frontal o
as two of four witnesses. was made, and still later, when the ends were added.
Finally in December we have records of the pay- There is, in fact, evidence for this in the later docu-
ments to two different iron workers, one for the ironments. Why, then, when these reliefs were removed
ties which secured the pieced-out stonework of the and the altar was uncovered, when everything was
altar (33 soldi, io denari) and the other, 3 soldi, for
taken to the chapel of San Rocco in 1787, did not the
simple tenons."x altar emerge from under them? It must be to this
It is customary to minimize the importance of Nic-period that we should give the blame that attaches to
colo Pisano's work on the altar of San Jacopo. The the loss of all but the description derived from the doc-
theory often advanced is that he did not finish ituments
(a of a significant work of a major master.
mistake inherited from a misinterpretation of the evi-
dence in the past) or that what he did was merelySMITH
a COLLEGE

question of repairs (a wrong emphasis, considering the

nature of the work he actually completed) or, as To-
esca put it, that "it is reasonable to suppose that theTHE ARCHITECTURAL THEORY OF
master often gave only the design for his works and FRANCESCO DI GIORGIO*
that they were carried out by others under his direction,
HENRY MILLON
as in the case of the altar which, in 1273 he undertook
to repair ('riattare') for the cathedral of Pistoia." It
does not seem to me that the facts bear this out. It is "... questa porzione sard modulo a tutto l'edi
not as though Niccolo were at that time a resident of Francesco di Giorgio (1439-1501) is unique
Pistoia, or doing some important work there; he had fifteenth century architectural theorists for hav
to come from a distance for this specific purpose al- corded the mathematical procedures he used in
though the work was finished in a relatively short time. ing a building. His writings contain detailed
He was present in person both to make the contract tions of methods for determining plans, elevati
and to receive payment, although it had been specified volumes.' Francesco's advocacy of a modular

in the dell'Arte,
9. Peleo Bacci, Documenti toscani per la storia spring of I,
x956. I am deeply indebted to Profess
Florence, x9xo, pp. 85-87, doc. 2. kower for the basic idea as here expressed and to P
xo. Ibid., p. 88, doc. 3. John Coolidge for his valuable criticisms and sust
terest.
* This article is the result of an investigation begun
I. Francesco's under
Trattato di architettura civile e militare
Professor Rudolf Wittkower at Harvard University
(printed in 11.1.141
from MS the in the Biblioteca Nazionale in
summer of x954 in a seminar on Renaissance Architectural
Florence) was published as part of the military history of
Theory. It was delivered as a paper at the Frick Symposium

This content downloaded from

128.103.147.149 on Fri, 31 Mar 2023 01:59:44 UTC
All use subject to https://about.jstor.org/terms
 ,i , i iD L E r

-A-A

2. Fol. 2v

j. Fol. 42V 3. Fol. 39

x-3. Francesco di Giorgio, Trattato. Florence, Bibl. Naz., Ms I1..1-14

P op

6. Fig. .superimposed faade elevation

4. S. M. delle Grazie, Calcinaio near Cortona, 5. Fig. z superimposed on Fig. 4 of S. M. delle Grazie, Calcinaio

architect, Francesco di Giorgio. Plan (from (author's drawing) (author's drawing)

Papini, Francesco di Giorgio 4rchitetto, III,
pl. LVI)

This content downloaded from

128.103.147.149 on Fri, 31 Mar 2023 01:59:44 UTC
All use subject to https://about.jstor.org/terms
 258 THE ART BULLETIN

which orders the entire building

which I will demonstrate is
briefly."" The two drawings sympto
age, but the century never
(Figs. I, 2) represent produced
two different ways of dividing ano
ment of its modular
the humanprinciples as
body. One system is the seven-head overt and
height,
ous. Neither Alberti nor
the other, Filarete
the nine-and-a-third-face gave any d
height. The seven
crete information module
about division (Fig.
goodi) produces aproportions
longitudinal o
modules although they
church clearly
plan with a dome imply th
diameter equal to the width
module and good of proportions are
the church. A double module gives the diameter of necessa
tecture.2 the apse. In a more detailed explanation of the nine-
All fifteenth century architectural theorists spoke of and-a-third-module division (Fig. 2), Francesco notes
proportions in terms of mathematics or in terms of the that a circle touching the nose and bottom of the torso
human body, which may mean there were understood ".. . will give the width of the temple," and the dis-
methods that architects employed when designing, but tance from the nose "... to the crown of the head ...
no attempt to find such schemes precisely applicable to will give the diameter of the hemicycles" which sur-
fifteenth century buildings has been successful. An ex- round the domed area.' In a like manner the location
amination of Francesco di Giorgio's theoretical writings of the elements of the plan are outlined according to
may help to explain the reason since it is possible to de- the anthropomorphic concept.
fine the relationship between his theory and his archi- The two methods represent different and, therefore,
tectural practice. flexible canons of proportion since a comparative ex-
amination shows relative parts of the body to be un-
equal. For example, if we assume the body to be 5' 10o"
Francesco describes several methods of obtaining tall, the head in the seven-part division would be ten
modules and proportions,8 but the most interesting is inches. This same body in the nine-and-a-third-part
the anthropomorphically derived modular grid in which breakdown would have a head longer by I 4 inches,
the proportions of the human body are used to deter- roughly the distance from the lips to the nose-a sig-
mine a temple plan (Figs. I, 2).' The head or the nificant change of proportion in any head! It is the
face is the fundamental unit and the divided body in- system and not the body that determines the propor-
dicates the location of the major features in the plan tions. Francesco, like Procrustes, stretches or amputates
of the building. Proportions of a faqade are also de- (note the feet of the nine-and-a-third division) the
rived from the human body (Fig. 3).6 Intersecting human figure to conform to his abstractly conceived
circles are superimposed on a grid generated by using bed of modules.s Is not this method merely a sophisti-
the height of the head as the module. cated combination of mediaeval geometry with Ren-
Francesco devotes an entire chapter to the derivation aissance arithmetic, adapting the prevailing geometrical
of the proportions of temples from those of a man which method so ably illustrated in Villard de Honnecourt's
he illustrates with two examples. The seriousness of his sketchbook?
intentions can be seen in the modest statement which The anthropomorphic approach is also used to deter-
prefaces this section. "... . finding many varied opinions mine the proportions and articulation of a fagade (Fig.
exist about this body I have determined something 3)V. Francesco's drawing of the fagade is accompanied

Italy by Carlo Promis in Turin in 1841 simultaneously with
an extract that contained the Vita di Francesco di Giorgio
Martini, the Trattato, and the Atlas, but not the military his-
tory. Page numbers used here are from the shorter version.
2. Alberti makes his recommendations for proportions and
ratios in a very matter-of-fact way. He suggests that the
height to width ratio of a door be z2:i or the square root of 2:i
(L. B. Alberti, Ten Books on Architecture, Bartoli translation,
Leoni edition, London, 1755, Bk. I, Ch. xii), or that in a
room "if the length be five times the breadth, make the height
the same as where it is four times, only with the addition of
one sixth part of that height" (Ix, iii). The implication of a
modular system is explicit in the demand for "regularity . . .
size and situation equality . . ." (IX, viii). Further evidence
of a similar nature can be found in vI, ii; ix, ix.
3. For Francesco, modules seem to be units of measure
large enough to be used for the basic ordering of a building.
Modules are obtained in three different ways. By the first, or
arithmetical method (Trattato, Bk. II, Ch. 2, p. 39), a story is
divided vertically into a number of equal divisions in which
cornices, string courses, windows, and doors are allotted a
predetermined number of parts. The method involves only an
arithmetical proportional division. The second, or geometrical
method, found in both the second and fourth books, uses a
construction to find a unit for a modular grid (Trattato, II,
9, PP. 59, 6o; IV, 3, pp. io6-io8). The constructions are of
two different types. One type, used to determine the proportions
of a room, utilizes two perfect figures (a circle inscribed in a
square), and determines a module that is a portion of the
radius which is roughly commensurable with the side of the
square. This type of construction is also used in the often re-
produced drawing of a facade-section for a domed building.
The other type of construction utilizes the diagonal of the
square (square root of 2) and the diagonal of the double
square (square root of 5) to establish an approximate relation
between the diagonals and the side of the square. The drawings
are attempts to find by a simple mediaeval geometrical con-
struction a non-empiric Renaissance module which may be
used arithmetically to approximate ratios and proportions
which are incommensurable. The result is a measuring unit of
considerable practical value, for it avoids the problem of the
relation of the length of a diagonal to the side of a square or
the length of the radius of an inscribed circle to the side of
the square.
4. Trattato, Iv, 4, pp. 109-I I.
5. Trattato, IV, , p. o101.
6. Trattato, Iv, 4, pp. 109-111.
7. Trattato, p. i i o.
8. The anthropomorphic view is employed not only in plan
but in the design of entablatures (Biblioteca Nazionale, MS
II.I.14, fol. 38v), column capitals (ibid., fol. 34v), and
columns (ibid., II.1.x4x, fols. 32v, 33r).
9. Trattato, IV, i, contains the textual matter relating to
this drawing but Promis does not reproduce the drawing,

This content downloaded from

128.103.147.149 on Fri, 31 Mar 2023 01:59:44 UTC
All use subject to https://about.jstor.org/terms
 NOTES 259

by specificcal sketch. Should the wall be placed on the outside or

recommendati
tions, by on the inside of the grid
what line? Is a combination of these
Francesco
or human two figure
indicated for different situations? How thick
applie
should the wall be? To
"proportional andwhat extent is thisgeome
determined
cesco by what the
argues architects thought was
that thestructural necessity,
rela
for a i.e., the supposed
temple that compressive strength
is of masonry,
eithe the
be forces at work on theby
determined masonry, the size of the building,
dividin
equal parts
and to whatwith the
extent by geometrical calculation? he
unit. The Fortunately, there areis
base in Francesco's manuscripts a
given
height of few examples ofdoor
the plans that contain gridis
lines and also
dete
two show wall thicknesses."1
diagonals, In several cases the
and wall is
wher
next to last
placed with itsgrid line
inside face on the grid line with pilasters "w
door." projecting from the wall on the opposite
Apparently the side of the ins
in determining grid line." Columns are usually
the placed on theheigh
grid line
lished by theor centered onintersection
the intersection of two lines, as they are
circle."1 Half-unit divisions indicate the location of on the nine-and-a-third-module height plan.1" Un-
fortunately, there is no such drawing extant to aid in
architrave, frieze, and cornice. One half-unit is arbi-
the determination of wall thickness or structural nature
trarily added to the summit "because the head is some-
of a fagade which would enable us to relate it to an
thing superior."l2 This deviation is useful for it gives
the rise of the pediment. Even though it is only dia-actual project.
grammatic, this drawing describes the major features Since the buildings designed in Francesco's Trat-
of a fagade rather completely. tato are of a different size from those he actually con-
structed,"1 some difficulty may be expected in the direct
II transferal of his theoretical proportions to his extant
work. Also, information concerning the attributions of
The chief problems resident in the application of a
buildings to Francesco di Giorgio and what part he
theoretical diagram to an actual building are concerned
played in them is still inexact." The one building that
with those matters that are usually left unstatedisin attributable to Francesco with some degree of cer-
diagram. For example, there is usually no indicationtainty is the church of Santa Maria delle Grazie in
Calcinaio, near Cortona.xs
of either wall thickness or wall placement in a theoreti-

which appears in Bibl. Naz., op.cit., fol. 39v. A reproductionand thus to determine within certain limits the absolute size of
may be found in Stegmann and Geymiiller, Architektur der any example.
Renaissance in Toscana, xI, 3, p. 19. A drawing of a similar17. Just what Francesco actually constructed remains some-
nature though less geometrical, and probably earlier, can what be mysterious. There is immense difference of opinion in
found in the Codex Saluzziano, Turin, Library of the Dukedistinguishing
of his own work from that done after his designs.
Genoa, MS 148, fol. 2 xv. A. Venturi (Storia dell'arte Italiana, vIII, i, pp. 737-883)
io. Trattato, p. ioi. would ascribe some thirty buildings or parts of buildings to
Si. Francesco states (loc.cit.) that the base height is deter-
Francesco, while A. Weller (Francesco di Giorgio, z439-
S5oz,
mined by the intersection of the diagonals and the circles. How- Chicago, 1943, pp. I99-21o) lists only six. For a
ever, the drawing in the Biblioteca Nazionale, op.cit., here fairly complete bibliography on Francesco's architecture see
reproduced, shows the base height determined by the inter- Sections A, B, and E of bibliography in Weller, op.cit., pp.
section of the circle and the interior vertical grid lines. This4o5ff.,
is or the bibliography in the handsome monograph by
probably an error of the copyist and speaks strongly against Roberto Papini, Francesco di Giorgio architetto, Florence,
Francesco as the author of the drawing. It is unusual that the 1946, II, pp. 267-275. Publications which appeared after
major intersections of the two circles or the intersections of Papini's
the book that should be noted include Giovanni Canestrini,
circles and the diagonals in the middle region do not establish
Arte militare meccanica medioevale, Milan, 1945 (?) Mario
any significant point. The upper circle serves no function Salmi,
at Piero della Francesca e il Palazzo Ducale di Urbino,
all. Florence, 1945; and Mario Salmi, Disegni di Francesco di
12. loc.cit. Giorgio nella Collezione Chigi Saracini, Siena, 1947.
13. For example, in the Codex Saluzziano, Biblioteca Reale, 18. The documents that pertain to the church have been
Turin, fol. iiv, all three plans; fol. x2r, all five plans; fol.
assembled in Appendix II of Weller, op.cit., pp. 352-354. The
2zv, the plan in the lower center of the folio. decision to build the votive church in Calcinaio came soon
14. For example, in the Codex Saluzziano, fol. i Ir, plan after Easter in 1484 when there were evidences of some mir-
second from the left; fol. i v, plan on the lower right; fol.aculous happenings. On July 1, 1484, Francesco, who was in
2zr, the two full plans; fol. i2v, the plan in the lower center.
Gubbio working for the Duke of Urbino, was paid a certain
I5. Also in the Codex Saluzziano, fol. xviv fol. x3r, the sum for a drawing and model of the church. In April of the
two plans in the upper right. following year Francesco was in Cortona and on the eighth
16. Most of the buildings illustrated in the manuscripts of
arethe month received payment as the designer of the church.
considerably larger, as evidenced by the great number of sup-Another document mentions Francesco as still in Cortona on
porting columns, the length of the naves in relation to the April 30, 1485. The church was dedicated in June 1485 and
transepts, the spacing of the columns, the size of the entrances,
the long account which is preserved in the Cortona archives
and the complex structural nature of the apse and transepts. mentions Francesco by name. What may be a drawing of a
While it would not be impossible to design structures that church of a similar nature can be found in the Codex Saluz-
make scale determinations difficult (as was so often done ziano,
in fol. 14r, and in the Codex Laurenziano-Ashburnham
the next century) it is probable that the intercolumniation dis-
361 in the Laurentian Library in Florence on fol. 12v. Fran-
tances are intended to be governed by structural considerations
cesco died in 15o1 and left the building unfinished. The work

This content downloaded from

128.103.147.149 on Fri, 31 Mar 2023 01:59:44 UTC
All use subject to https://about.jstor.org/terms
 260 THE ART BULLETIN

The plan of the building is which

I. The height of the base, a standard
is found at the L
with the nave divided
heightinto three
of the intersection equal
of the lower circle and bays
and transept containingthe two diagonalsone bay
as Francesco instructs each
in the (F
plan is perfectly symmetrical
text. about the
2. The height
axis with the exception of of the
the top ofsmall
the prominentsacristy
lower
of the apse. The initial impression is one of
string course.
consistency. By placing over
3. The height of the Fig.
top of the main 4 a trac
cornice. Also,
theoretical plan divided
the main anthropomorphicall
cornice and frieze together equal one-
and-a-third parts half
(Fig. 2)
module (that a tocorrespond
allocated the frieze in the
tained that is too precise to
manuscript drawing). be accidental
4. The rise of the pediment is exactly one unit.
I. The entire building rests within a rectan
by 6 squares. Perhaps the grid determines some of the following also:
2. The transept arms in
I. The door is aboth
standard 1:2 of the
ratio but plans
the height
same width.
and consequently the width may be taken from
3. The naves in both plans are of the same width. the diagonal lines.
4. The circle of the crossing in the Francesco draw- 2. The cornice of the pediment is one-quarter mod-
ing corresponds in size to the octagonal dome of ule high instead of one-half module, but since
the church.
the entablature of the building is about one-half
5. The distance from the center line of the nave to the size of the one in the manuscript drawing this
the exterior face of the nave wall equals the may be a justifiably noted correspondence.
radius of the dome carried on the ring of columns. 3. The placement of the circular window may have
6. The pilaster-wall-grid line relationships are iden- been intended to coincide with the module line
tical to the manuscript plans noted earlier."' at its lower extremity.
The differences are: If one labored over the point, many more cor-
respondences could be alleged. The admittedly variable
i. The absence of the side aisles. nature of Francesco's method of design might make
2. At the extremities of the arms of the transepts,
such a comparison deceptive and the attempt foolhardy.
the walls are inside module lines, rather than
For instance, the upper circle in the manuscript draw-
outside, but they are consistent in this respect.
ing is apparently there only to form a logical whole,
3. The dome does not rest on a ring of columns.
and on the faqade it determines nothing. Likewise, at
4. Asymmetrical sacristy placement.
the points where one would expect something to oc-
5. The hemicycle width at the apse indicated by the
cur--such as at the intersections of the two circles or
head is wider than the existing niche.
at the intersections of the circles and the diagonals in
the middle region-nothing happens. Although the
The correspondences are clear in every part except
for the one-third-module at the west front.20 Could
placing of the circular window, the secondary cornice,
this mean there were a series of steps leading down
and the height of the arch over the doors may not have
which were intended or have since been covered been or determined from such a drawing as this, the cor-
lost? Or is this another example of Francesco's pro-
respondence of the theoretical drawing and the actual
crustean adaptation? project indicate a very dose relationship. The degree of
The west elevation of the building presents a fagade
correspondence is about all that should be expected of
Francesco.
that is divided in three vertically. Uniform string
course and entablature heights are used throughout III
the building. The west front is surmounted by a pedi-
ment, while the two arms and the apse are covered by The definable correspondence between the architec-
a hipped roof. The whole is surmounted by a pointedtural theory and practice of Francesco di Giorgio gives
octagonal dome set on an octagonal drum. a clue as to the standard practices of the fifteenth cen-
When the anthropomorphic design for a fagade ex-tury. Francesco di Giorgio was certainly not an archi-
tect generously endowed with talent, and any indica-
amined earlier (Fig. 3) is placed over this fagade the
tion about his methods should shed light on everyday
height to width ratio is identical (Fig. 6). The manu-
script drawing is for a somewhat smaller church and practice. It cannot be denied that the family of build-
the door size is, correspondingly, much larger. Ac- ing forms and proportions of the last half of the fif-
teenth century indicate some geometrical or mathe-
cordingly, the architrave, frieze, and cornice in the
theoretical drawing occupy far too large a portion matical
of method was used in their creation. A great
many studies have been made which attempt to show by
the facade of Santa Maria delle Grazie. However, the
constructions with right and isosceles triangles, root
regulating lines determine the following:

was carried on after him by Pietro di Domenico who cesco to alter awkwardly the column spacing in the last bay
com-
pleted the dome in 15 5. of the 9 1/3-face height plan. He never knew what to do with
19. See note 14 for manuscript and folio citation. the extra 1/3 module, and perhaps it was ignored in this plan.
20. It is this same extra 1/3 module which caused Fran-

This content downloaded from

128.103.147.149 on Fri, 31 Mar 2023 01:59:44 UTC
All use subject to https://about.jstor.org/terms
 NOTES 261

triangles, goldenAll othersection

arch
tional lines spoke of theabout
build m
structions There usually is entail
a gen
the placement that of the
proportio reg
than not thereare is
the a cons
result
to make the matics, buildingmusic, fi
guishable ratios,
tectural propor
harm
used in the the human
buildings? b
Francesco di
mony Giorgio
of the pru
ings, what may
service be toone ansoi
a mathematically lying coheren
on visua
way. While The Francesco's
flexible p
ing the figure immutable to the me
rel
what should tury be architect
the "per
figure, he exhibits
may help anexplequ
plication of logicalhis abstract
system m
The geometrical cesco constr
did not
building and serves me
trinaire mann
proportionality. owes itsPropor
heigh
which is decreased
dependent som
at it
arithmetic artisan."28
but final adju
of the architect and do no
of the compass.
[HARVARD UNIVERSITY]

2 1. The following is a partial listing, in chronological order,

Saggi sulla architettura del Rinascimento, Milan, 1931, p.
of attempts at proportional derivations: A. Thiersch, Handbuchx2ff.; Theodor Fischer, Zwei Vortrdge iiber Proportionen,
der Architektur, Iv, I, Stuttgart, 1883; G. Dehio, Ein Pro- Munich, i934; Marcel Texier, Giometrie de l'Architec-
tortiongesetz der antiken Baukunst und sein Nachleben in Mit- ture. . . , Paris, 1934; Louis Hautecoeur, De l'Architecture,
telaltar und in der Renaissance, Strasbourg, 1895; L. Denina Paris (1939 ); Robert Gardner, A Primer of Proportion,
and A. Proto, "La real chiesa di San Lorenzo in Torino" New York, 1945; Matila Ghyka, The Geometry of Art and
Architettura italiana, Turin, 92zo; Jay Hambidge, Dynamic Life, New York, 1946; Bruno Zevi, Architettura e storio-
Symmetry, New Haven, x 92o; F. M. Lund, Ad Quadratum, graphia, Milan, 195o, p. 5i; George Jouven, Rythme et
London, 1921; Jay Hambidge, The Parthenon and other Architecture, Paris, 195i; Charles Funck-Hellet, De la pro-
Greek Temples; Their Dynamic Symmetry, New Haven, 1924 portion. . . , Paris, I951; Cesare Bairati, La simmetria dina-
Le Corbusier, Vers wne Architecture, Paris, 1924, pp. 58-62;mica, Milan, 1952; M. Borissavlievitch, Le nombre d'or,
M. Borissavlievitch, Essai critique sur les principaux doc-Paris, 1952; and idem, Traiti de l'Architecture, 2 vols., Paris,
trines relatives a l'esthitique de l'Architecture, Paris, 1925i
1954.
Robert W. Gardner, The Parthenon, Its Science of Forms, 22. This is the major thesis of the book by R. Wittkower,
New York, x9z25 Jay Hambidge, Elements of Dynamic Architectural Principles in the Age of Humanism, London,
Symmetry, New York (c. 1926); Arthur Butler, The Sub- 1952. See also, e.g., Alberti, op.cit., ix, v, vi.
stance of Architecture, New York, 1927; Julien Gaudet, Ell- 23. Promis, Trattato, p. o101.
ments et thiorie de l'Architecture, Paris, n.d.; G. Giovannoni,

This content downloaded from

128.103.147.149 on Fri, 31 Mar 2023 01:59:44 UTC
All use subject to https://about.jstor.org/terms