The Tradition of Public Patronage at the Florentine Cathedral
D- , uring the fourteenthcenturythe governingboard of the
Florentine Duomo evolved from a series of sporadic su-
pervisory committees to an institution whose authority had
been tested and structuresrefined in response to the gigantic prob-
lems of funding and design that arose during the construction
of the new cathedral. This remarkableinstitution, the Opera of
S. Maria del Fiore, has long been the object of scholarly interest.
In the last century, the illustration of its institutional history
constituted the ultimate, if unrealized, aim of Cesare Guasti, con-
ditioning the choice of sources assembled in his two fundamental
volumes on the cathedral.!Modern studies of the Duomo, how-
ever specific their aim, have benefitted from the breadth of
Guasti's view, and some scholars have followed his lead with
1 C. GUASTI, La Cupola di Santa Maria del Fiore illustrata con i documenti
dell'Archivio dell'Opera secolare. Saggio di una compiuta illustrazione dell'Ope-
ra secolare e del tempio di Santa Maria del Fiore, Florence, 1857. The broader
intentions of the author, clearly indicated in the subtitle of this volume, are
explained in its dedication, p. V. Guasti's second documentarycorpus, treat-
ing the history of the cathedral from 1293 until the period of the cupola,
Santa Maria del Fiore. La costruzione della chiesa e del campanile secondo i
documenti tratti dall'Archivio dell'Opera secolare e da quello di Stato, Florence,
1887, was prefaced by a fuller description of his ideal plan, embracing the
administrative history of the Opera, pp. XVI-XXI. Although the accompanying
"Discorsoanalitico su' documenti",pp. XXXIII-CXIV, was ostensibly limited to
the history of the monument,it in fact constantlyreveals Guasti's interest in the
Juridicaland administrativepolicies of the Opera and as such has constituted
an importantguide to subsequent studies. For convenience, in this essay the
frequent referencesto the two Guasti volumes will appearin abbreviatedform:
Cup. (for 1857) and SMF (for 1887), each followed by the appropriatedocu-
ment number (and by the page number only in cases of extremely lengthy do-
Casa Editrice Leo S. Olschki s.r.l. and Villa I Tatti, The Harvard Center for Italian Renaissance Studies
is collaborating with JSTOR to digitize, preserve, and extend access to
I Tatti Studies: Essays in the Renaissance ®
inquiries into the Opera's administrativestructures!2 Nevertheless
it seems possible to improve our grasp of the political mentality
behind the institutional framework if we are prepared to pause
over those aspects of the documentationwhich have perhaps least
interested art historians: rhetoric and attention to form. In
these eddies in the flow of events, a vocabulary emerges which
was formed together with the institution, perpetratingits own ideo-
logical and constitutional bases.3 Specifically, in the face of de-
2 The fullest study of the Opera for the period up to 1370 is A. GROTE,
Das Dombauamtin Florenz, 1285-1370. Studien zur Geschichte der Opera di
Santa Reparata,Munich, [1959/61]; for the trecento see also L. MUSTARI, The
Sculptor in the Fourteenth-Century Florentine Opera del Duomo, PhD disser-
tation, University of Iowa, 1975, esp. pp. 12ff, "The Opera of Santa Maria del
Fiore". A survey of the institution from the thirteenth century up through the
period of the cupola is in H. SAALMAN, Filippo Brunelleschi. The Cupola of
Santa Maria del Fiore, London, 1980, esp. pp. 173ff, "Guild Control after 1331:
The Opera of SantaMariadel Fiore". Interestingfor their comparativeapproach,
if extremely cursory, are W. BRAUNFELS, MittelalterlicheStadtbaukunstin der
Toskana, Berlin, 1953, esp. 151ff, "Die Domopera"; and C. MACCARI, "Le
Opere del Duomo in Toscana",Fede e arte, VIII, 1960, pp. 286-301. Fuller
comparisonswith the administrationof other cathedral churches, beyond the
scope of the present essay, would yield both analogies and contrasts to the Flo-
rentine situation. See, for example, the study of the civic and spiritual signi-
ficance of the Milanese Fabbricadel Duomo: G. SOLDI RONDININI, "La fabbrica
del Duomo come espressionedello spirito religioso e civile della societa milanese"
(Studi e testi di storia
in idem, Saggi di storia e storiografiavisconteo-sforzesche
medioevale, n. 7), Bologna, 1984, pp. 49-64.
The vast generalbibliographyon the Florentinecathedralcan be approached
through W. and E. PAATZ, Die Kirchen von Florenz, III, Frankfurtam Main,
1952, pp. 320ff; an updating and overview of this material is provided by
SAALMAN, op. cit., while the matter of Brunelleschiand the cupola is exhaus-
tively treated in C. BOZZONI-G. CARBONARA, Filippo Brunelleschi:saggio di bi-
bliografia,2 vols., Rome, 1977-1978. Those scholars who have worked closest
to the documentary sourceshave typicallydisplayedat least a subconsciousaware-
ness of the Opera's structuresand practices. Since nearly all of the sources dis-
cussed in the present paper have been considered in previous studies, my debt
to this body of scholarshipwill be evident even where my interpretationdiverges
from earlier formulations. No attempt will be made here, however, to register
systematicallyprevious interpretationsof the documentationdiscussed when the
matter seems to be well consideredby the recent bibliographycited.
3 For the applicationof the concept of "normativevocabulary" the study
of Florentinepolitical institutions see the importantstudy of J. NAJEMY, Corpo-
ratism and Consensus in Florentine Electoral Politics, 1280-1400, Chapel Hill,
1982, especially the introductorycomments on pp. 15-16: "The language that
legitimates and determinespolitical behavior is most likely to be found in the
records documenting the evolution of those institutions that gave shape and
cisions of vast responsibility, the Opera developed a tradition of
consultation and consensus-buildingwhich explicitly evoked the
perception of its relationship to the city's government and ulti-
mately to the Florentine populace.
The concept of representationalgovernment accountableto its
constituency is, of course, inherent in the spirit of republicanism,
so important and persistent in Florentine political ideology. The
oscillations of the communal regime from broadly democratic to
narrowly oligarchic were negotiated in terms of the proportional
representationallotted to constitutionally defined categories of ci-
tizens, and the rapid rotation of all offices guaranteed plurality
of participationwithin the politically enfranchisedclass. Further-
more, although the highest office of the Commune, the Signoria,
enjoyed vast powers, while the councils' role was limited to voting
on legislative proposals, public consultation on the formulation
of policy thrived in the form of the "Consulte e pratiche", ad
hoc advisory committees whose proceedings constitute a precious
record of the decision-makingprocess for a span of two centu-
ries.' These meetings were in reality no more representativethan
the choice of the citizens convoked, but their very existence tes-
tifies to the governmental officials' awareness of their responsi-
bility to sound out and consider a variety of opinions on matters
of public concern.
Since the Opera del Duomo was, as we shall see, a delegate
of the Florentine Commune and its members were typically expo-
nents of the city's ruling class, it is not surprising that its prac-
tices bore a marked resemblanceto those of the civic government.
This is not to affirmthat the Opera directly mimicked its parent
substance to political assumptions,ideas and goals"; so that "the meaning of
events was built into the languageand structureof the institutions themselves".
4 E. CONTI, Le "Consulte" "Pratiche"
e della Repubblicafiorentinanel Quat-
trocento, I, 1401, Cancellierato Coluccio Salutati, ed. idem et al., Pisa, 1981,
pp. Vff. The importanceof these consultative processes in the fourteenth and
early fifteenth centuriesis underscored G. BRUCKERin FlorentinePolitics and
Society, 1343-1378, Princeton, 1962, esp. pp. VI, 72ff, and idem, The Civic
World of Early RenaissanceFlorence, Princeton, 1977, esp. p. 13. Their signi-
ficance during the Medicean period is illustrated by N. RUBINSTEIN, The Go-
vernmentof Florence under the Medici (1434 to 1494), Oxford, 1966. For an
analysis of the same practicesfrom 1494 to 1512 see F. GILBERT, "Florentine
Political Assumptionsin the Period of Savonarolaand Soderini",Journal of the
Warburgand CourtauldInstitutes, XX, 1957, pp. 187-214.
institutions, but rather that its development was conditioned by
the same republican mentality and by a constitutionally derived
sense of public accountability. As an institution charged with
the management of an enterprise of universal significance to -the
citizens of Florence, it may even have preceded and exceeded the
Commune in the evolution of consensus-buildingprocesses in the
1360s.5 The sense of legitimacy and purpose consolidated and
articulated in this period was bequeathed to subsequent genera-
tions and constitutes a vital key to the interpretation of Brunel-
leschi's relationship to the Opera di S. Maria del Fiore in the
next century. The following analysis of how experience forged
tradition over the first century and a half of the cathedral's
building history will, of necessity, proceed in a highly condensed
form, galloping through intensely studied terrain, while avoiding,
whenever possible, controversial questions of design and chrono-
logy which do not seem crucial to the definition of the Opera's
role vis-a-vis its professional consultants and collaborators,on the
one hand, and the public and its institutions, on the other.
An early phase of the rebuilding of the city's ancient cathedral,
dedicated to S. Reparata, was plagued by unstable and discon-
tinuous funding from both public and ecclesiastical sources. Com-
munal subsidies began with frequent, if timid, allotments for the
repair of the old church.6 These soon grew into more substantial
serial allocations which accompanied the decision to rebuild the
cathedral, with support reaching an annual level of 4,000 lire
up through 1303.7 All these measures were voted by overwhelm-
5 BRUCKER, Florentine Politics (see note 4), pp. 76-77; cf. the observations
on consensus politics in the electoral scrutinies in the 1350s and 1360s and in
the period after 1382 in NAJEMY,op. cit. (see note 3), pp. 166ff, 263ff.
6 SMF 1, 3-4, 6-8, allotments recorded from 1293. Proposals for public
funding are known from 1285 and 1292: GROTE, op. cit. (see note 2), p. 30
with correcteddating of SMF 2.
7 SMF 9, 13, 17, 18, 22, 26. Additional measureslegislated in 1295 and
1296 (SMF 10, 16) were aimed at collecting monies directly from the populace.
It would be difficult to estimate their success, for there was apparentlyno effi-
cient enforcementagency: the first was an offering of atonement invited from
those who knew themselves to be guilty of fraud with public funds; the se-
cond, a tax on wills and a surtax prorated to the amount of communal lira
tax assessed to individuals in the city and territory, was to be exacted by
un-salariedrectors of the bishop. For estimates of Opera income in this and
the following period see D. FINIELLO ZERVAS, "Un nuovo documento per la
ing majorities the city councilswith the insistent and explicit
motivation the work regarded only the reverence
that not and
praisedue to God and the patronsaints,but also the honorof
the Commune the decorum the city.8 Meanwhile 1296
and of in
Boniface VIII had allowed to 3,000 florins the income
up of of
courtsfor the "sumptuous" churchnew which,in
his formulation, been undertaken the addressees the
had by of
bull, the bishopand the cathedral chapter.9 Thesefirstfunding
effortswere however quickly, to
exhausted, judgefromthe affir-
mationmotivating indulgences by
accorded Clement in 1310
for ten yearsto individuals contributing the construction
saidto be moreor less interrupted lackof money.'0 sit-
uationwas worsestill in 1318, when, aftera long silence, the
communal councils declaredthat the site was nearlyabandoned
and that workcouldnot proceed withoutpublicsubsidy."Ob-
serving that the bishopand chapter decreeda tax on the
clergyfor the benefitof the new church, councils
suit by approving generous
a new allotment five years. Six
months later,in support suddenly
of stepped-up construction,the
councilsfurther allocated it onethird theInquisitor's
to of revenues
dueto the Commune the nextten years.12
Not surprisingly, the earlyperiodboth church Com- -and
munemaintained control overtheirfinancial contributions ap-
pointing theirown representatives administer
to them. The civic
directly fourOperai elected the Priors
to supervise theirexpenditure,'3 the samemenindependently
storia del Duomo e del Campanile Firenze, 1333-1359",Rivista d'Arte, XXXIX,
1987, pp. 3-53: 7ff.
8 See the various formulations of this invocationin SMF 3, 6, 7, 9, 10, 14,
16, 17, 18, 22, 26.
9 SMF 12.
10 SMF, Discorso analitico,p. XLII.
11 SIMF 29.
12 SMF 30. The petition
presented by the officialsof the Opera explained
that to keep up the marble supply and increased work force ("ut in eodem
opere plus solito viriliter laboretur")it was necessary"quod comune Florentie,
more solito, manum porrigeretadiutricem". Appeal was made, as usual, to the
"honoreet decore comunis et populi Florentini".
3, 4, 6, 7, 9, 17, 18.
solicited further legislation to augment the Opera's income."4Even
before Boniface VIII's indulgence, the bishop had also nominated
his own Operai.'5 It is not clear whether both groups of repre-
sentatives comprised the Operai of S. Reparata who, together
with the Calimala consuls for the Baptistery, petitioned the gov-
ernment for the enlargementof the piazza in 1296,16but another
proposal of the same year was presented jointly by the bishop
and his two Operai, on the one hand, and the Commune's two
Operai, on the other.'7 Although the first instance of the Com-
mune's delegating its supervisory powers to a guild dates from
1303, when Por S. Maria was drawn by lot to fill this role for
one year,18 evidence of joint ecclesiastical and lay representation
on the board of works is still available in 1318, when the new
public funds were put into the hands of three officials elected
separately by the bishop, the chapter, and the Commune."9 The
situation recorded in the 1321 Statute of the Capitano del Po-
polo consists in the annual rotation of five major guilds in the
supervisory role, each to appoint three officials empowered to
spend money with the assent of a fourth, who was to be a priest
nominated by the chapter.'
Such an institution was inevitably hobbled by lack of con-
tinuity in personnel and funding. According to Giovanni Villani,
in 1331 the work on S. Reparatahad long been interruptedbecause
of the diversion of communal funds to military expenses.2" This
state of affairswas assessed by the communal councils in October
of that year. The traditional legislative rhetoric had now to be
14SMF 10, describedin note 7 above.
15 SMF 11.
16 SMF 14.
17SMF 16, describedin note 7 above. See the bishop's Operai still elected
in 1299 (SMF 19, 20).
18 SMF 27.
19SMF 29. Cf. also the petition presented to the Signoria in 1313 "pro
parte Capituli ecclesie Florentine ac etiam pro parte operis eiusdem ecclesie":
GROTE, op. cit. (see note 2), p. 39.
2 SMF 31.
21SMF 33: "si ricominci6a lavorarela chiesa maggiore di Santa.Reparata
di Firenze, ch'era stata lungo tempo vacua e sanza nulla operazioneper le va-
rie e diverse guerre e ispese avute la nostra citta".
inverted: long interruption the church, beautifully
the of so be-
gun,was declared no uncertain termsto be a publicdisgrace.'
Since,as it was againemphasized, work appertained the to
dignity the government the decorum the city, a better
administrative formula and
was required, it was found at this
time.Thecouncils approved firstpermanent
ing,automatic of all communal payments, in
by the attribution a similar of
percentage the saleof gabelles.'
Thesepublicfundswereassigned the consuls the powerful
NVool Guild,who werein turnempowered electOperai re-
ceiveandexpend themin the construction the cathedral.We
hear no more aboutsystematic funds or represen-
tationon the Opera A of
afterthis time.24 variety offering boxes
scatteredaround city, somein the wool merchants'
modestamounts the comingyears
while perhaps servingto keep alive the conceptof the direct
betweenthe populace the cathedral, parallel to
the linesof delegated responsibility whichnow descended through
the councils the guild and its Opera.' For publicsubsidies,
'2 SMF 35: "cacthedralis ecclesiaFlorentinacepta fuit tam formosaet pulcra,
quod ad honorem comunis Florentie et decorem civitatis ipsius cedit non mo-
dicum...; post cuius temporis lapsum nulla fuit in dictum subsidium assignata
pecunia per comune Florentie, ita quod remansit iam est longum tempus et est
absque hedificatione aliqua, quod redundat in grande dedecus obbrobrium et
abominationen Comunis iamdicti".
24 An attempt to reactivate the right to income from the episcopal court
may be documentedduring a temporarycollapse of communalincome in 1353:
SMF 70, p. 76.
25 VILLANI (SMF 33) indicates that the collection boxes in wool shops pro-
duced a revenue of 2,000 lire for the first year, but such contributionsseemed
to have taperedoff radically. ZERVAS, Ioc. cit. (see note 7), pp. 12-15, concludes
that the Opera income recordedfor 1333-1359 was largely from public sources.
When these funds were drasticallycut in 1353, the Opera's salaried manager
(provveditore)was able to scrape together only about 158 lire from boxes
in the Mercanzia,the Camera del Comune, and the cathedral and from the
"cassettadal pane" and that of the "isconti dei maestri"(SMF 70, p. 74). To
his repeated queries about reinstating the boxes in the wool shops, the Operai
responded that the guild was not in a position to do so (SMF 70 pp. 76,
80, 81). The marked preponderance public funding over income from pri-
vate offerings,pious donations, and legacies at S. Maria del Fiore is in contrast
with the results of analysesavailablefor other cathedrals;compare,for example,
L. RICCETTI, cantiere edile negli anni della Peste Nera", in II Duomo di
however, an amendment to the 1331 legislation, holding the
Opera accountableonly to the consuls, and not to the Commune
as previously decreed, served to reinforce the intermediary po-
sition of the guild.26
The basic composition of the Opera was written into the
Wool Guild's new constitution of 1333: four Operai, a treasurer,
and a notary. All but the last were unsalaried guild members,
elected on a representationalbasis. All would serve brief four-
month terms and be ineligible for re-election for the span of four
years.27 The number and duration of offices and of eligibility,
as well as the method of selection by lot from the pool of can-
didates approved in qualifying scrutinies, would be adjusted re-
peatedly throughout the history of the institution,' but the guid-
ing principle remained that the Opera should effectively represent
the entire guild. It therefore constituted, despite the rotation of
individuals, a standing, prestigious entity, able to solicit, consol-
idate, and defend public subsidies within the limits of the Com-
mune's ability to contribute.29 Thanks to a recent study of the
Opera's income for the years 1333-1359, we now know a good
deal more about how this worked.' Cutbacks were inevitable in
time of civic emergency or strife, but even in the worst moments
the guild could bail out the Opera with short-term loans while,
Orvieto, ed., idem., Rome-Bari,1988, pp. 142-148; SOLDIRONDININI, Ioc. cit.,
(see note 2), pp. 55-59.
26 SMF 36.
Although the revision of Opera accountsfor 1333-1359, entrusted
to eight officials elected by the Signoria and colleges, constitutes a major ex-
ception to this policy (ZERVAS,Ioc. cit. [see note 7] pp, 31ff), the norm, still
lively in the following century, was that the books of each Opera treasurer
were audited by syndics elected by the guild.
27 SMF 42.
28 See GROTE,op. cit. (see note 2), pp. 99ff; MUSTARI,op. cit. (see note 2),
29After 1333, petitions were normally presented by the Operai (SMF 51,
52, 61, 67, 77; cf. petitions by consuls in 1333: SMF 40, 41), who repeatedly
played upon the Commune'sinterest and duty in the enterprise with rhetoric
styled upon established formulations,e.g., SMF 52: "cum dicta cathedralis ec-
clesia incepta fuerit tam formosa et pulcra, que cedit in honorem Dei et Matris
sue et beate Reparate virginis, et in honorem ac decus comunis et civitatis
Florentie; ad hoc ut tam pulcrum et honorabile opus iam inceptum melius per-
fici et compleri possit, ed quod gratia iam facta per dictum Comune videatur
fuisse et esse liberalis et gratiosa, prout in talibus esse debet; placeat vobis.."
30 ZERVAS, loc. cit. (see note 7).
1. Detail of the Misericordia fresco, dated 1342, (Florence, Bigallo)
showing the cathedral complex.
2. Reconstruction of the Arnolfian plan for S. Maria del Fiore, superimposed
on the outline of the plan of the existing cathedral (after Toker).
A - rising masonry over foundations still in service today.
B - rising masonry over abandoned foundations excavated 1965-1980.
C - rising masonry reconstructed as probable extension of A and B.
D - rising masonry reconstructed as possible extension of A, B and C.
- 3' o Maria d a t .S
I /\, 4 A -ea-X--=v"
I ,~ sd p .1
3. Plan of S. Maria del Fiore (after Sgrilli).
4. Exteriorelevationof S. Mariadel Fiore (after Sgrilli).
-- ' A t......-0 _-_ S l
ii/'. rah -~ Z,4 rn/i .diwn A ./4 . ft_
5. Longitudinal section of S. Maria del Fiore (after Sgrilli).
*o ; j s ut
6.S Mra e Foe,cpoasenfrmsoteat
A.-A A. A A- dI - -_
;rA!>-^^'-^ + ... ..........
S. t r .. a -
8. Analytical bird's-eye view of the cupola of S. Maria del Fiore
A, -AF 4.. i
8.Pa,eevto the_ n scino lanter of S.Mrade ir
9. Tomb of Filippo Brunelleschi S. Mariadel Fiore, portraitof the
architectcarvedby Andreadi LazzaroCavalcanti,1447.
10. Tombof FilippoBrunelleschi S. Mariadel Fiore,detail of the portrait.
no doubt,lobbying support its petitions the government
in of to
in defense cathedral
of funding. In sucha situation, fromabout
1332, it was at last possible gearup the workshop a long-
Although lackof internal documention before1353 makes
perilous,3' seemslikelythatin the firstyears
of its existence new Opera's
the role was conditioned that
of its predecessors, defined,in the narrowest sense, simplyas
assuring honest,efficient of publicfundsfor the city's
cathedral. powers matters design artistic
Its in of and supervision
defined, it is noteworthy in 1334 the city
authorized Signoria, the headsof the guildand
the Opera, appoint
to Giottosupervisor all civic construction
projects including, on the list, the worksof S. Reparata.'
LikeArnolfodi Cambio, whomthe Commune rewarded
1300 with tax exemption his magnificent
new cathedral which,it was hoped,wouldbe the mostbeautiful
and honorable templein Tuscany," Giotto was cherished aas
professional presence whose abilityto bringdistinction the to
city was a matter publicconcern.Presumably decision
of the to
divertthe Opera's resources fromthe halting church effortto the
less urgent"campanile Giotto"was connected
di with his own
preferences. DianeZervas' of
analysis the Operaaccounts from
1333-1359indicates duringthoseyearsthe bell towerrose
steadily largely, not entirely, the expense the church
if at of
itself.3 No documentation to
survives relatehow Andrea Pisano,
afterGiotto'sdeathin 1337, introduced in
changes the master's
plan,or exactly why andhow he was replaced Francesco
lenti, who, even in the absence scholarly
chronology the successive stages,is generally recognized as
31 We rely until that time primarily upon communal and guild records,
supplementedby occasionalnotarial acts, chronicles,etc.
32 SMF 44. See the relative observationsof SAALMAN, op. cit. (see note 2),
33 SMF 24: "comune et populus Florentie ex magnifico et visibili princi-
pio dicti operis ecclesie iamdicte inchoacti per ipsum magistrum Arnolphum
habere sperat venustius et honorabilius templum aliquo alio quod sit in par-
'4 ZERVAS, loc. cit. (see note 7), pp. 24-29.
the designer behind the ever more open and ornate bifora and
trifora stories.35 From the 1350s, when in-house documentation
begins to be available in the notes of the Opera's provveditore,
Filippo Marsili, we can observe occasional meetings of masons to
advise on the final stages of the tower; these are normally pre-
sented as routine affairs.6 When opinions are recorded, they
either approve the work as proceeding or offer unanimous and
uncontroversialadvice on technical problems.
The same notebook of Marsili and that of his successor, which
extends to early 1359, tell quite another story when, from 1355,
the Opera's attention again focused on the project for the church.
A well-known detail of the fresco of the Misericordia in the Bi-
gallo, dated 1342, (Fig. 1) is generally believed to be a useful
record of the state of the building: the faade breaks off at about
the same height reached at that time by Pisano's part of the cam-
panile, and the old S. Reparata rises behind that screen and that
of the lower south wall. But recent archeologicaldiscoveries now
tell us that there was more to the new S. Reparata than can be
seen poking out of the medieval city in this view: the south
35 Contrastingviews of the building history of the campanile are polarized
around the theses sustained by M. TRACHTENBERG, The Campanileof Florence
Cathedral, "Giotto's Tower", New York, 1971, and by G. KREYTENBERG,"Der
Campanile von Giotto," Mitteilungen des Kunsthistorischen Institutes in Florenz,
XXII, 1978, pp. 147-184. Their divergent chronologies in turn depend on the
parallel histories of the revetments of the cathedral's aisles sustained by H.
SAALMAN, "Santa Maria del Fiore: 1294-1418", Art Bulletin, XLVI, 1964,
pp. 471-500 (reprinted, except pp. 493-500, in idem, op. cit. [see note 2], pp.
32-57), on the one hand, and on the other by G. KIESOW, "Zur Baugeschichte
des Florentiner Domes", Mitteilungen des Kunsthistorischen Institutes in Florenz,
X, 1961, pp. 1-22, and by G. KREYTENBERG, Der Dom zu Florenz, Berlin, 1974.
ZERVAS' study (loc. cit. [see note 7]) brings new evidence in favor of the
36 The Ricordanze of Marsili, 18 March 1353/4-23 March 1357/8 (SMF
70, pp. 72-117) consist in briefly noted points in agenda, normally followed by
the annotation of decisions handed down. It is not always stated whether the
numerous "consigli" mentioned concerned the campanile or the design problems
of the church, to be discussed shortly, but the following can be safely ascribed
to the tower: p. 82, 8 June 1355 - "consiglio usato del campanile"; p. 85, 5
Feb. 1355/6 - "consiglio de' maestri sul campanile" (4 masons approve work and
give technical advice); p. 86, 9 Mar. 1355/6 - "consiglio usato del campanile";
p. 88, 10 June 1356 - "consiglio della ruota"; p. 89, 28 Nov. 1356 - "consigli
usati del campanile"; pp. 90-91, 12-31 May 1357 - "consigli usati del campanile"
(use of hoists for tower); p. 92, 10 June 1357 - "consigli usati"; p. 113, 7 Dec.
1357 - "consigli usati del campanile".
wall, withoutany provision its foundations vaulting,
been extended a crossing areawherefoundations some
risingmasonry been executed a laterally
in Toker's of
reconstruction, a smaller-scale,Arnolfian prototype
of the dometriconch that was eventuallybuilt.' (Fig. 2) Al-
thoughthe areawithinthe foundations' perimeter still lit-
teredin the 1350s with the pre-existingurbanfabric,therewas
visibleprogress the of
towards realization a new cathedral which
corresponds neither the articulation its navenorin the scale
andextension its eastendto the finalmodelapproved 1367,
according which,as we know, the familiar
to building was ac-
tuallyexecuted. (Figs. 3, 4, 5) From1355, when the Operai
ordered capomaestro Francesco Talenti makea woodmod-
el demostrating the east chapels
how oughtto be and how to
a dozenyearsof heated
correct defectin the navefenestraticn,39
discussionswould bring the Operaback face to face with its
exposed it was to disapproval the waste
of publicmoniesif existingwork were to be undone,criticism
37 An early formulationof the data from the excavationsconductedin 1965-
1974 (F. TOKER,"Florence Cathedral: The Design Stage", Art Bulletin, LX,
1978, pp. 214-231) is supersededby a fuller study incorporatingthe results of
a 1980 excavation at the area of the joining of the Arnolfian south transept
and crossing: idem, "Arnolfo's S. Maria del Fiore: A Working Hypothesis",
Journalof the Society of Architectural Historians,XLII, 1983, pp. 101-120.
38 In proposing this reconstruction,whose main purpose is to demonstrate
the existence of a coherent and compellingArnolfianproject which, though -mo-
dified and expanded, constituted the basic concept of S. Maria del Fiore as it
was constructed,TOKER ("Hypothesis"cit. [siee note 37]) accuratelydistinguishes
amongst types of evidence: archeological, documentary, iconographical,and com-
parative. The archeologicalevidence alone proves that the early construction
presupposed a five-bayed, timber roofed nave, a vaulted octagon, and lateral
transeptsbeginningwith a vaulted sacristyor chapel oriented longitudinally. The
belief that there is documentaryevidence to support the elaboration of this
chapel fragment into the prototype for the Duomo's characteristicradial tri-
conch (ibid., p. 109) is, however, apparentlyunfounded, since the 1366 text,
cited (SMF 141) as referring to the Arnolfian remains as a /"croce", in fact
regardsone of the new projectsunder considerationat that time. This attractive
feature of the proposed reconstruction rests then primarilyon visual evidence -
above all the representation a similar structurein Taddeo Gaddi's frescoes in
the Baroncelli Chapel (S. Croce, ca. 1330) - and on the argument ex silentio
that the documentsof the 1360's do not describe the radical transformation of
some other scheme into that eventually approvedfor the east end.
39SMF 70, p. 81.
if the edifice were not as beautiful and perfect as hoped, disgrace
if it were not structurallysound.'
The variations in the design of the Florentine Duomo ap-
proved between 1355 and 1357 involved the transformationof the
nave from an exposed-timberstructure, articulatedin five double-
windowed bays, (Fig. 2) to a much more modern arrangement
of three giant vaulted bays leaping down the same space. Since
only the perimeter walls of the new cathedral had been begun,
this revolution could feasibly have been introduced within the
existing scale. The major drawback to the otherwise magnificent
scheme was, however, that in the first bays, at least, the aisle
walls had reached a height that already incorporated the fenes-
tration according to the former plan, in conflict both with the
axes and increased height of the vaulted scheme. Thus even this
design change, which was to take place within the limits of all
existing foundations, presented the first dilemma between imper-
fection and demolition the Opera is known to have faced.4"
It proceeded cautiously at first, obtaining for the vaulted pro-
posal illustrated in Talenti's model the unanimous approval of
enlargedconsultationsof master masons, both secular and religious,
with strong representationfrom the seasoned vaulting experts of
40 The timid would not have been encouraged by the example of the
audacious project for the Duomo Nuovo in Siena. In the years around 1355
two Florentine protagonists in the discussion on S. Maria del Fiore, the capo-
maestro Talenti and the master builder, Benci di Cione, answered the summons
for opinions on the statics of the defective new Sienese structure, and their
opinions must have sealed the fate of the ill-advised enterprise, already inter-
rupted for economic difficulties after the Black Death: V. LUSINI, II Duomo
di Siena, I, Siena, 1911, pp. 179ff; E. CARLI, II Duomo di Siena, Siena, 1979,
41 On the problem of the windows see SAALMAN, Ioc. cit. (see note 35), pp.
475-476, 480-481. The stylistic considerations of the controversy over whether
the marble incrustation over the raw masonry was in place by 1355 need not be
recapitulated in the present context (sources cited in note 35 and continuing
debate by M. TRACHTENBERG, review article of KREYTENBERG, op. cit. [ see note
35], in Art Bulletin, LXI, 1979, pp. 112-131; and by G. KREYTENBERG, Letter to
the editor: "The Cathedral of Florence", Art Bulletin, LXII, 1980, pp. 340-341).
My perception of the Opera's dilemma is however based upon full agreement
with SAALMAN'S affirmations (loc. cit. [see note 35], pp. 476, 496) that the
alteration or demolition of even the raw masonry to adapt it to the new vaulted
program would have been considered a drastic undertaking whose singularity
and expense, we might add, would have placed the Opera in the position of
having to justify the decision to its sources of income.
S. MariaNovella.42A hundred citizensand ecclesiastics were
alsocalledto confer, it is significant the Operai, con- in
sultationwith the guildconsuls,decided againstproceeding with
a meeting be held in the Palazzo Priori,presumably
to dei pro-
posedto obtainthe official governmental stampof approval for
the variantplan.43Whileworkon the trifora zoneof the campa-
nile proceeded consultations
peacefully, with expertscontinued,
culminating no less thanthirty-eight meetings before19 June
1357, whenthe new bays'dimensions were at last fixedand,to
the soundof bells, organs song,ground broken the
foundation the firstgreatpier.44By this time, however,
Opera doorswereopento experts. Bencidi Cione,for example,
was heardout on a matter the pierfoundations,45 for the
actualdesignof the piers,both the capomaestro Talentiand the
outsider executed drawings gessomodels
in competition.' Sincein this case evaluating councilsshowed
divisions opinion,the Operaiordered new composite
posalwhichwas to not
submitted the scrutiny only of invited
experts of anyone who wishedto comment the publicly
42 SMF 70, pp. 83-84 (15, 16, 17 June 1355), 84 (5, 7, 31 Aug. 1355).
SAALMAN, loc. cit. (see note 35), p. 477, note 25, proposes the likelihood of a
"prior and unrecorded consensus" around the desire for vaulted nave bays.
43 SMF 70, pp. 84 (7 Aug. 1355), 85 (Jan. 1355/6), 86 (9 Mar. 1355/6).
44 SMF 70, pp. 88 (7, 10 June 1356: subject not stated), 92 (10, 13
June 1357), 93 (15 June 1357), 94-95 (19 June 1357), 96 (21 June 1357:
"Metti a uscita per salario dato a' frati e a' maestri, per 38 volte che si sono
raunatial consiglio della chiesa, in piu volte lire 19").
45 SMF, 70, pp. 97 (3 July 1357), 98 (7 July 1357).
16 SMF, 70, pp. 100-101 (17, 19 July 1357). Only the judgments of
experts are recorded,but both the cathedralcanons and a few of many invited
citizens were present at the 17 July meeting.
47 SMF 70, pp. 102-103 (28 July, 3 Aug. 1357). The third model was
executed by Talenti together with Giovanni Ghini, who, as vaulting expert,
would soon join his colleague, the standing capomaestro,in the supervision of
the pier construction: SMF 70, p. 110 (26 Oct. 1357). Both men would in
turn contributedesigns for the pier capitals: SMF 70, pp. 111 (17 Nov. 1357),
116 (30 Jan. 1357/8). The Operai went so far as to order that the two
capomaestri's proposals for the piers and the capitals be frescoed in full scale
on the premises of the Servites of the SS. Annunziatafor final selection: SMF
70, pp. 116 (16 Feb. 1357/8) 117 (23 Mar. 1357/8). The final decision appa-
rently took place in the interval (23 Mar. 1357/8 - 28 Sept. 1358) between the
recordsof the provveditore, Marsili,and those of his successor,CambinoSignorini.
The old fenestration problem remained however an embarrass-
ment to the Opera. In 1358 its consultants continued to favor
the ideal solution, which involved demolition, in order to have
a single window, "bela e grande" in each bay; the Operai, on the
other hand, could not be insensitive to the urgency of building
up both side walls to the definitive height in order to vault at
least the first bay which, it was said, would be "a great con-
solation for all the citizens".48 This conflict between the con-
sultants' advice and the Opera's responsibilityto its public finances
and the populace was apparently resolved at this time in favor
of the latter. (Figs. 4, 5) It is therefore perhaps not a coincidence
that in the same period the Opera was continually involved in
attempts to improve its public funding.49A petition of early 1357
to triple its standing percentage of communal finances, revised
to postulate instead its doubling, eventually found the s-upport
of the Priors in office in the last two months of that year.' Their
legislative proposal was passed by the councils in December al-
though its application probably awaited the new, corrected for-
mulation of April 1358.51 In this text the familiar motivations,
the honor of God and the decorum of the city, are corroborated
by the observation that the renewed vigor of construction activity
demanded such funding, and more.
Work apparently proceeded uneventfully on the new vaulted
scheme in the following years.52From late 1364, however, a new
48 SMF 72, p. 120 (13 Nov. 1358), 123 (10 Jan. 1358/9); SAALMAN, Ioc.
cit. (see note 35), p. 481.
49 SMF 70, pp. 74 (3 May 1353), 75 (June 1353), 77 (12 July - 4 Oct.
1353), 79 (15 Dec. 1354), 81 (8 June 1355), 84 (31 Aug. 1355), 85 (Jan.
1355/6), 89 (28 Nov. 1356), 90 (11 Jan. 1356/7 - 12 May 1357), 91 (10 June
1357), 98 (7 July 1357), 110 (26 Oct. - 6 Nov. 1357), 112 (7 Dec. 1357). The
recovery of the repeatedly mentioned 950 lire, one sixth of the Opera's income
diverted to military expenses in 1352, was still pending in 1359: ZERVAS,loc.
cit. (see note 7), p. 10.
50 The original proposal (SMF 70, p. 90) predicted that with a cut of six
denari per lira the church could be completed in four years at a total cost of
40,000 florins, while at the existing rate of two denari, 'si penera XX anni, e
costera 70000". Cf. ZERVAS, loc. cit. (see note 7), p. 29.
51 SMF 77.
52 The Ricordanze of Signorini (SMF 72) terminate with 30 January 1358/9,
and Opera documentation resumes only in 1362 with the first extant book of
deliberations (SMF 85ff). A third book of ricordanze regarding the interval was
of meetings lay andreligious
and examined important matters the nave
design(clerestory fenestration) structure, closeattention
beingpaidto thatof the firsttwo largenavevaultsfinally be to
built in 1365 and 1366.53But whatwas to comebeyondthese
bayswas already matter discussion.The finaldesign,as we
know, extended nave througha fourthbay, displaced
octagon elevated cupola
and overa drum.
(Figs. 3, 4, 5) This was, unlikethe 1357 project,incompatible
with the existing east-end foundations, vastlymoreexpensive be-
causeof its sheersize andstructurally frighteningbecause the
increased weightand spanabovethe crossing. How was such a
relativelywild scheme reconciled the Opera's
economy, sobriety, publicresponsibility?
and I believethat we
can observea seriesof consummately diplomaticmaneuvers on
the partof the Operawhichgenerated consensus
a around what
mighthavefirstseemed little morethana designer's fantasy.
While wvithin Opera ranks the co-capomaestro,
Ghini,was constructing new modelof the church, Operai
sent out for furtheradvicealongtraditionalrepublican by
turning the guildswhose rolls included
to artists:Seta for the
goldsmiths,Medicie Speziali the painters,
for Pietrae Legname
for the masons.54Theirrespective complied nominating
consultants the Opera,whose proposals
for coalesced arounda
single,alternativeplan, knownin the documents the design
of the "maestri pittori" which,followingSaalman, shallcall
the committee project.Compared Ghini's
to modelit was,in the
lost during the last century (SMF, pp. LXXV-LXXVI). See the interpretation
of progress during this hiatus in SAALMAN, Ioc. cit. (see note 35), pp. 481-482.
53 SMF 119 (27 Sept. 1364), 120 (4 Oct. 1364) 121 (22 Nov. 1364), 125
(20 Dec. 1364: decision to hold monthly meetings), 126 (20 Dec. 1364), 128
(23 May 1365: expenses for consultation with lunch in the residence of Wool
Guild on previous 22 March), 131 (18 July 1365). In October 1365 the Com-
mune decreed that the affluent societies of Orsanmichele and the Misericordia loan
2000 florins to the Opera for the construction of the two great vaults (SMF 133);
the second was completed in March 1365/6 (SA'IF 135).
54 The invitation of 12 July 1366 (SMF 140) mentions only the first two,
but the presentation of representatives from all three (SMF 141, 142) suggests
that the masons' participation, if not solicited, was at least welcome. Ghini's
"chiesa pichola" was said to be completed by 29 July (SMF 146), but further
work on it was interrupted on 3 August (SMF 148).
early stages, a paper project, apparently little more than a plan,
which certainly contained the four-bay nave and presumably the
triconch crossing.55Within just a month from when, in July 1366
the artists had begun to work on their plan, a first semblance of
consensus chose it hands down. This approval derived from a
mass meeting of over eighty citizens convoked with the consuls,
the Operai, the capomaestri, and the professional consultants,
where it was advised that the matter, together with the vaulting
problems, be referred to a board of eight citizens, elected two
each from the quarters of the city.' The board members, imme-
diately named by the consuls and Operai,57having interrogated
55 In SAALMAN'S analysis (loc. cit. [see note 35], pp. 485ff) the principal
innovation of the committee project was the introductionof the drum between
the body of the crossingand the cupola. His argumentsare carefullyconsidered,
but are expressly based on the assumption that earliet and alternative plans
included a similar arrangementof the chapels and sacristies in the east end.
The much discussed ideal representationof the Duomo frescoed in the chapter
house of S. Maria Novella precisely in this period (commissioned30 Dec. 1365
to be completed within two years) by one of the leading committee members.
Andrea Bonaiuti, would, however, seem to favor the hypothesis that the com-
pelling feature of the new proposal was the externally articulatedtriconch itself.
Although the main dome has been demonstratedto be an Arnolfian heritage,
there seems as yet to be no absolute proof that a multi-domedcrossing characte-
rized the first plan (cf. note 38, above), the Talentian model of 1355-1357,
nor Ghini's "chiesa piccola". Bonaiuti's fresco already shows a four-bayednave
(with wishfully idealized fenestration) as well as a triconch which differs from
that eventually approvedby its smaller scale and the absence of the drum. The
result is remarkably harmonious,and it seems that the desirabilityof the drum
could only have been generated by the accompanyingproposal to inflate the
scale of the crossing area, whose dating, without Saalman'sassumption of the
fundamentalsimilarity of competing projects, is problematic.Although the de-
finitive maximumdiameter of the octagon was fixed only in August 1367 (SMF
178), the matter of east-end dimensions was under considerationa year before
(SMF150). The examinationof the implicationsof this problem goes beyond
the intent of the present study, in whose context it will be sufficientto bear in
mind that the committee proposal was perceived as strongly characterizedand
extremelyattractive,but riddled with structuraldifficulties.
`6 SMF 146 (29 July 1366). The most precise suggestion to this effect
was that of Andrea Rondinelli. All five opinions recorded suggested that the
citizens work together with a committee of masons.
57 SMF 147 (29 July 1366), listing the names. It is not possible to re-
print such lists in the confines of this essay, whose focus is institutions rather
than individuals. Nevertheless the question of the social and political identity
and of the architectural competenceof the cathedral'scitizen advisors is a pro-
mising area of study which deserves attention. Of this genre for the time
being we have only the pioneering analysis of D. FINIELLO ZERVAS, The Parte
the design committee, expressed preference for its project with
some reservations, particularlywith regard to its stability, a mat-
ter in which they took the word of the masons. That very day
the consuls and Operai jointly decreed, for the honor of God
and the city, that construction henceforth follow the committee
plan, and the consuls proceeded to the Palazzo dei Priori to inform
(but not consult) the city's highest office. They were commended
and offered support in their endeavors."
Although the committee plan was now a fait accompli, the
elevation to be built upon it was not. During the following year
not only its authors, but the capomaestri and others prepared
schemes for the undefined triconch area, and when the committee's
elevation came up for evaluation before a new, mixed consulta-
tive board of thirteen, it was approved only on the condition that
it have great strength.59 Tempers flared, and the Wool consuls
had to pass a special act to protect the honor of the capomaestro
Ghini, as an extension of their own, from threats and insults.'"
He in fact, together with Talenti, was author of an alternative
elevation based on the new plan. In July another board of eight
citizens and a further group of twelve savi, having been advised
by the capomaestri that the committee project was not structur-
ally sound, withheld their approval and expressed the desire to
see the elevation proposed by Ghini and Talenti executed as a
model.61 When four expert churchmen also declared the com-
mittee project impossible,62 the Operai scrambled back up their
delegative ladder for advice from the consuls. The answer, though
sibylline, was intelligible to them: " build what is most pleasing
Guelfa, Brunelleschi and Donatello, Locust Valley, N.Y., 1987, pp. 87ff and
327ff, of the administrative 'elite' of institutions concerned with artistic super-
vision including the Wool Guild and the Opera, from the 1390s into the first
decades of the quattrocento.
-8 SMF 150 (13 Aug. 1366).
59 SMF 170 (31 May 1367).
60 SMF 173 (9 July 1367). At the same time the consuls decreed that the
competent governing of the Opera was so much a matter of guild honor that
the designated Operai could not renounce the office for any other except the
61 SMF 174 (24 July 1367), 175 (25 July 1367).
b2 SMF 176 (31 July 1367).
to the citizens and yet fitting, beautiful, strong, and honorable for
The Operai in fact went back to the ecclesiastical consultants
requesting modificationof the committee scheme in order to for-
tify it without vilifying its attractions.'4 They also installed three
of these churchmen as permanent advisors and paid a custodian
to keep the building site and Opera open on Sundays so that the
people could see the alternative designs.65 By October 1367 two
scale models were available for viewing in separate houses near
the church: that of Ghini (apparently the new alternative) and
that of the committee, which, fortified as advised, had been exe-
cuted by Ghini and Talenti; work could proceed no further with-
out a decision. The leading citizens were summoned in groups by
invitation, and all but one found the committee model unequiv-
ocally more beautiful, magnificent and honorable for the city.
The doors were then thrown open to the whole populace, con-
voked by crier, and in the space of two days over four hundred
individuals of all classes and trades had their names recorded as
favoring the irresistible committee model. The notary, ser Bru-
nellesco Lippi, father of Filippo, was among them.'
For their confirming decree the Operai were now probably
awaiting one final step, which is particularly interesting in the
present context because it recalls how their relationship with the
people had originated. Three weeks later the Signoria summoned
the Opera to the city hall in the presence of yet another advisory
council, a group of seventeen appointed by the Priors themselves
and composed of eminent Florentine citizens judiciously larded
with a few artisans.67This board proceeded to the building site,
where it solemnly examined and approved the committee model
as having all the advantages: it was more beautiful, fitting, hono-
63 SMF 177 (3
Aug. 1367): "inveniant et invenire debeant omnem modum
quod fiat illud laborerium dicte ecclesie, de quo putent quod la citadinanza plus
contentetur, et quod sit utile pulcrum et forte et honorabile pro comuni Flo-
64 SMF 178 (9 Aug. 1367).
65 SMF 179 (12 Aug. 1367), 186 (30 Aug. 1367).
66 SMF 189 (25 Oct. 1367), 190 (26-27 Oct. 1367).
67 SMF 192 (19 Nov.
rable,and strong.'6 Somepractical advicewas also proffered:
buildit up in stone,leaving marble revetments later,and-
musicto the Opera's - the Priorsshouldsupport by in-
creasing the publicfunding. With this authoritative opinionthe
consultations closed, and the Operaiimmediately decreedthat
henceforth construction shouldfollow this modelalone. What
the lengthy had
process constructing consensus signified
of a to
the Operaand Wool Guild was expressed with only a little
exaggeration yearlaterwhen new guildlegislation
a forbade all
Opera officials deviate
to fromthe approved project. The pream-
ble to the act statedthat the decisionhad been madeby the
Priors,theiradvisory colleges, the consulsof all the guilds,
as well as a multitude lay and religious
it couldbe saidthatthe entireCommune passed judgment".69
Throughout restof the century the earlyyearsof the
quattrocento, construction in
consisted the gradual
lization the approved
of design;but whenever, of
or error,new difficulties arose,consultationonce againassured
consensus. For example, 1404 it was discovered the
70 in that
capomnaestro Giovanni d'Ambrogio pitched of the tribune
buttresses (Fig. 4, partB) too high, invading areadestined
for the parapet according the model.Its correction,
eitherbecause accumulated erroror ambiguity the model,
68 The adjectives,"piiubello e pituutile e onorevole e forte", echo the pre-
scriptionsdictated by the consuls in July (see note 63).
69 SMF 214 (15 Dec. 1368): "pe gli singnori Priori e collegi e consoli di
tutte l'Arti, che allotta (sic in Guasti) crano, e grande numero di cittadini re-
ligiosi e secolari,che si puo dire e reputareessere istato tutto il comune di Fi-
renze..." John Najemy has pointed out the similarity of this passage with
the advice offered to the Signoria in 1378 to discuss legislative proposals with
the consuls of all the guilds so that, if passed, they could be said to have the
consent of the whole city (J. NAJEMY, "Guild Republicanismin Trecento Flo-
rence: The Successesand Ultimate Failure of CorporatePolitics", The American
Historical Review, 84, 1979, pp. 53-71: 66). Although the circumstancesand
the procedures are different, in both cases there is an attempt to claim the
strength of universal approval for measures which had passed the scrutiny of
representational bodies. In the case of the cathedralthis process was far from
systematic; but even if the representationof quarters was arbitrary and of
guilds selective, and if the referendumrecords the opinions of a fraction of the
city's politically enfranchisedpopulation, the intent was to legitimize a monu-
mentally expensive decision as truly representative the will of the city.
70 SAALMAN, loc. cit. (see note 35), p. 492.
had implications for the clerestory windows and related revet-
ments. The Operai duly convened the Wool consuls together with
a group of nineteen citizen and artists.7" Amongst the goldsmiths
represented were Brunelleschi and Ghiberti, the proverbial rivals,
who knew something about corporate patronage themselves, hav-
ing both competed three years before in the contest for the Bap-
Like the Opera of the cathedral, that of San Giovanni, re-
sponsible for the Baptistery, derived from the Commune via the
delegation to a guild, in this case the ancient establishment of
cloth merchants known as Calimala. Since the original records
are lost, our knowledge of the competition derives from two po-
tentially very biased sources: 7 Lorenzo Ghiberti's own Commen-
tari 7 and the celebratory Life of Filippo Brunelleschi written by
Antonio Manetti; the latter, although composed only about 1480
as a gigantic footnote to the same author's version of the No-
vella del Grasso Legnaiuolo, was based on information obtained
years before from the protagonist and on the same sort of dili-
gent researchthat went into the most detailed extant reconstruction
of Filippo's burla.74 Ghiberti gives the fuller account of the
juridical data of the contest: seven competitors; thirty-four con-
sulting judges, all of whom ultimately gave written opinions in
his favor, seconded by the Opera, the consuls and the entire Ca-
limala guild. Manetti, on the other hand, concentrates, presu-
mably for narrative purposes, on the rivalry between Brunelleschi
and Ghiberti alone, the leading contestants, but he gives a behind-
the-scenes account which is extremely interesting in the light of
what we have observed so far on corporate patronage. Brunel-
71 SMF 425 (10 Nov. 1404).
72 For the available documents and sources as well as a full discussion of
the competition see R. KRAUTHEIMER, Lorenzo Ghiberti, 2 vols., 2nd ed., Prin-
ceton, 1970: I, pp. 34ff; II, pp. 360ff.
73 L. Gi-IBERTI, I Commentari,ed. 0. MORISANI, Naples, 1947, p. 42.
7 A. MANETTI, Vita di Filippo Brunelleschipreceduta da La Novella del
Grasso, ed. D. DE ROBERTIS - G. TANTURLI, Milan, 1976, pp. 60-64. For the
basis in truth, even when distorted by partisansentiments, of Manetti's account,
see U. PROCACCI,"Chi era Filippo di ser Brunellesco?"in Filippo Brunelleschi.
La sua opera e il suo tempo (Acts of the Convegno Internazionalein Florence,
16-22 October 1977), Florence, 1980, I, pp. 37-64: 38.
leschi,he says,quicklyexecuted trialpieceandquietly
the moment confrontation.
Ghiberti, the otherhand,cun-
ninglyraced aboutthe city asking of
advice all thosehe expected
mightbe namedjudges,repeatedly amending wax modelto
beforecasting panelin bronze.
The judgeseventually nominated were in fact the sameLoren-
zo had supposed, and manyof them had already openly ex-
pressed theirsupport his entry. Confronted,
for however, with
Filippo'sonlyat the moment judging, wereastounded
of they by
the mastery displayed.Unableto recant
it entirely theirformer
position,they conferred, the
voted,and advised Operaithat the
commission should conferred
jointly the two sculptors. The
Operaisummoned two contestants,
offering contract as
advised.Ghiberti not flinch, Brunelleschi inflexible:
all or nothing. As one unwittingly for
destined "bigger things",
he stoodfirmwhenthe Operai to the
threatened award commis-
sionfor the doorsto his rival;andwhenthis in fact took place,
presumably the finalwrittenacts cited in Ghibertis
mentari, departed Rome,leavingbehindhim a divided
Although Manettipatently and
Ghiberti, severalaspects his account
of ringtrue in the context
of publicadministration. work must not only be beautiful,
but mustalsomeetthe approval the consultative
of processes in-
to fairness the publicinterest. Whenmore
thanone competitor something
valuable offer,the ideal
wasto obtain best of both: by creating modified,
modelin the caseof architectural the
design, sharing contract
in the caseof autographworks. It seemsentirely likelythat the
Operaandguildhad attempted gaina consensus the joint
75 MANETTI, op. cit. (see note 74), p. 64: "e rimasenenello oppenione del
vulgo la citta tutta divisa". There seems little need to ascribethe sharp division
of opinion in the city describedby Manetti to later gossip (cf. KRAUTHEIMER,
op. cit. in note 72, I, pp. 41-2). The very scale, importance,and duration of
the competition (to early 1403), as well as the likelihood that the selection was
narroweddown to two contestants before the final decisions (ibid., pp. 42-43)
would justify a lively public interest in its outcome.
conferralof this costly and prestigiouscommissionbut were thwarted
in their aim by the intransigenceof a hot-headed young Filippo.6
The "maggiorecosa" for which Brunelleschi'stalents were des-
tined was of course the cupola of the Florentine Duomo, in 1412
officially rededicated to S. Maria del Fiore.' As the work on the
church at last proceeded up through the drum level, the time had
come to confront the great riddle contained in the 1367 model:
how to construct the enormous octagonal cloister vault which it
had mandated. (Fig. 6) If a few cracks in the nave vaults in the
1360s had sent the Operai off in a flurry of consultations, it is
easy to imagine the tensions and fear of disaster gathering over
the Opera and the guild at this time. The rich and intensely
studied documentation of both institutions is a precious source
for the study of Brunelleschi's relationship with these patrons
in the years of his remarkableachievement. Manetti himself was
the first to study and quote the Opera records, weaving them into
his account of the events, which also benefitted from the oral
tradition of Filippo and his circle as well as from a still lively
understandingof republican institutions and attitudes. With due
allowance for his narrativelicense, we can use Manetti as a knowl-
edgeable guide and subtle informer for the interpretation of the
cupola documents in the context of established Opera practices.
Manetti's tale of the cupola opens with a reference to the
honorarium of ten florins paid to Brunelleschi for a first, inten-
sive round of consultations with the Opera personnel.78 Medieval
76 This interpretation is not necessarily incompatible with the observations
of A. MIDDELDORF KOSEGARTEN ("The Origins of Artistic Competitions in Ita-
ly" in Lorenzo Ghiberti nel suo tempo, Acts of the Convegno Internazionale
in Florence, 18-21 October 1978, Florence, 1980, I, pp. 167-186: 186) that
the Baptistery competition involved the "grafting on to the competitive practice
of medieval communes the Antique concept of the contest as agon between
great masters". My emphasis is, however, on the strength and tenacity of re-
publican ideals, an ethic which was enthusiastically defended by the new 'civic
7' SMF 464, 465 (29 Mar. - 29 Apr. 1412). The provvisione of the Con-
siglio del Popolo recalls that the church had been so dedicated at the time of
its foundation, and thus also Giovanni Villani (SMF 5: 1294). The acts of
1412 revitalized this tradition and, among other things, ordered the use of the
new name where habit had kept alive, even in public documents, the dedica-
tion of the earlier church to S. Reparata.
78 MANETTI, op. cit. (see note 74), p. 78; he cites a lost uscita or cassa
vaultswerebuilton centering; the construction a wooden
armature the vastandloftycupola
for seemed the headmasons
not just difficult, outright
but impossible, according the bio-
grapher, the acutereasoning Filippopointedout to them
the full rangeof problems involved. The documents show the
Operaand guild behaving the time-honored
in manner the
face of the dilemma. In 1418 the Operai sent out publicbans
inviting kindof modelor designregarding cupola's
any the cen-
tering,offering expenses audience all participants a
and to and
prizeof two hundred florins the winner.79 two mostex-
pensive responses this callwereGhiberti's
to modelof smallun-
bakedbricksand Brunelleschi's, whichwas a surprisingly large
mock-up cupola, constructed ordinary
with bricks finished
detailsexecuted collaboration Nannidi Bancoand Do-
natello.' A firstgreatconsultation, whoseproceedings un- are
fortunately not recorded, examined the entries,8' in the
following the responsible
year institutions geared for the final
In November 1419, the Wool consulslegislated intensify
supervision the difficult by creating new magistracy
of task a com-
prisedof four expressly electedguild members who, with the
consent the consuls Operai,
of and wouldsolicitandconduct the
affairsrelating the cupola. The motivation this provision
leavesno doubtaboutthe importance attached the matter
the guild:"theadministration the cathedral beenentrusted
to us by the peopleand the Commune, bringing the greatest
honor and perpetual fame, above all if it is properly run".82
book (26 May 1417) which corresponds to the two extant stanziamenti published
by Guasti: Cup. 16 (19 May 1417).
79Cup. 11 (19 Aug. 1418).
80This analysis of the documents follows SAALMAN, Op. Cit. (see note 2),
pp. 61-62, who sets straight the results of Manetti's incomplete archival research.
MANETTI, op. cit. (see note 74), p. 91, scandalized by the waste of money, states
that Ghiberti's "useless" model cost 300 lire and Brunelleschi's only 50, but
the expenses in several entries for the latter in fact total 310 lire.
81 Cup. 15 (7 Dec. 1418).
82 Cup. 1 (15 Nov. 1419): "per populum et Commune Florentie gubernatio
et cura Opere Sancte Marie del Fiore, cathedralis ecclesie florentine, est univer-
sitati Artis lane conmissa; quod cedit in honorem maximum et famam perpetuam
In the spring of 1420 these new officials apparentlycommissioned
not a new compromise model, as was once thought, but further
elaborationof the Brunelleschimodel to test whether its continuous
fabric allowed adequate lighting into the crossing.' New public
proclamationsinvited all to come and discuss the work, and further
payments were made for consultations and designs.'
Manetti conflates these and the meetings of 1418 into a single
dramatic narrative portraying Brunelleschi's behavior in the face
of the doubts of his interlocutors."5Only Filippo, having proved
physically and economically unfeasible all others' attempts to de-
vise a supporting structure over which to build the vault, main-
tained that the cupola could and must be erected without this
traditional crutch. His proposal, derived from the intensely mo-
tivated contemplation of classical architecture, was so eccentric
that it initially provoked the ridicule of the Operai and their peers
as well as of the experts convened. To Filippo's great embarrass-
ment, the Operai, in the presence of the consuls, literally had
him carried out of two meetings, presumablywhen his passionate
(lefense of his idea exceeded all the bounds of protocol. But
the rash youth of 1401 was now a very determined man in his
Artis prefate, maxime si dicta fabrica suo recto ordine gubernabitur, convenit
honori Communiset dicte Artis".
83 This is the interpretation of SAALMAN, op. cit. (see note 2), pp. 63-69,
who demonstratesthat materialsacquiredfor the "novo et ultimo modello" from
8 March to 22 April 1420 (ibid., doc. 95; cf. extracts from the Italian version
in Cup. 50) would correspondto a model of the drum alone, executed in the
same 1: 8 scale as Brunelleschi'sbrick model. The case for this hypothesis
is strong, but it cannot be supported, as Saalman proposed, by his doc. 164
of 3 Oct. 1421 ("s.5 s.6 per fare dipigniere il tanburo"), which almost cer-
tainly regards the box for denunciations ordered to be placed in the church
(Archivio dell'Opera del Duomo, henceforth AOD, II-1-79, c. 25v, 15 Sept.
1421: "unum tamburumin quod scribatur quod tamburaripossit... quicum-
que de pecunia et rebus dicte opere habuisset indebite... et similiter qui-
cumque delinquens dicte opere..."). The same reservationsare applicable to
his docs. 169 and 213.3 (the latter doc. of 21 July 1425 again follows the
deliberationof 14 April 1425 to reinstate the "tamburo"for the Opera's em-
ployees and masons: AOD, II-1-86, c. 10v). On the theft of the denunciation
box in March 1421 see BRUCKER, Civic World, cit. (see note 4), p. 309.
84 Cup. 44 (27 Mar. 1420), 45 (3 Apr. 1420), 46 (1 Apr. 1420), 47 (1
Apr. 1420), 48, 49 (24 Apr. 1420).
85 MANETTI, op. cit. (see note 74), pp. 80-85.
forties, able to bend his characterto the politics of the situation'.
While patiently seeking allies amongst the other consultants by
extolling any possible praiseworthy detail in their proposals, he
also set to work to remedy his lack of concrete experience as an
architect and his still insufficient connections with the influential
wool merchants responsible for the final decision. One coup
exemplifies his adroit diplomacy: asked whether he could demon-
strate his methods on a smaller scale, he convinced the prominent
guild member, Schiatta Ridolfi, to allow him to vault his private
chapel (now lost) in S. Jacopo sopr'Arno without centering. Still
according to Manetti, further demonstrations and persistent per-
suasion eventually won the day for Filippo. His methods were
at length judged valid, and he agreed to put them in writing
in a summary document, which was eventually approved and re-
corded by a joint meeting of the consuls, Operai and four cupola
officials in July 1420.'
But already these three magistracies had elected Brunelleschi
and Ghiberti, together with the Opera capomaestroBattista d'An-
tonio, to the new office of provveditori della cupola.' Battista, a
trusted in-house presence, would receive a bonus over his ordi-
nary daily wage, while Filippo and Lorenzo were salaried at thirty-
six florins each per year. Ghiberti in his Commentari proudly
declares that he and Brunelleschiwere colleagues in this office for
eighteen years at the same salary.88 Manetti, on the other hand,
affirms that Ghiberti's presence was a useless waste of public
money and that it was eventually terminated.' We shall see that
86 Ibid., pp. 85-88, and Cup. 51, from a lost registration in the Opera
archives;anotherversion of the report is conservedin the archivesof the Wool
Guild and was published by A. DOREN, "Zum Bau der FlorentinerDomkuppel",
Repertorium fur Kunstwissenschaft, XXI, 1898, pp. 249-262: 258-261. SAALMAN
(op. cit. [see note 2], p. 79) believes that the report was already composed in
connection with the model of 1418, executed by Brunelleschiin collaboration
with Donatello and Nanni. The three would constitute the first person plural
subject of its incipit, "Qui appresso faremo memoria/menzione...". Cf. the
interpretationof TANTURLI (in MANETTI, op. cit. [ see note 74], p. 85 note 2),
who considers the plural a linguistic convention of the substantiallyanonymous
technicaldocument,whose subject, if any, could only be the four cupola officials.
87 Cup. 71 (16 Apr. 1420) and DOREN, Ioc. cit. (see note 86), pp. 261-262.
8 GHIBERTI, op. cit. (see note 73), p. 47.
8 MANETTI, op. cit. (see note 74), pp. 91, 95.
neither of these accounts is exact. The contrast between them
has prompted scholars to speculate on the personal rivalry and/or
friendship of the two masters and their relative roles in the cu-
pola. Manetti's polemics and scrambledchronologieshave exposed
his version to general scepticism, prompting a number of writers
to re-evaluateGhiberti's contribution to the design and execution
of the dome.9' Nevertheless, a critical reading of his text, seen
against the background of patronage practices so far observed at
the Opera del Duomo, supplies a remarkablyplausible key for the
interpretationof documented events.
According to Manetti, in 1420 the city was again divided as
at the time of the competition for the Baptistery doors.9' The
consuls and Operai had determined to offer the principal super-
vision of the cupola to Filippo at the unprecedentedlymeager rate
of thirty-six florins per year. Motivated more by desire for honor
than for profit, he surprised the Ghiberti faction by accepting.
But Lorenzo's supporters pressed the Opera with scarcely veiled
threats, pointing out that it would be prudent to give Brunelle-
schi a companion in such a risky and publicly exposed enterprise,
and so Ghiberti was also instated at the same salary. Filippo
and his advocates were disappointed by the final outcome, but
had to accept it.
Two observations are possible regarding this account. First,
the election and subsequent documents describe Brunelleschi and
Ghiberti as "provveditori della cupola" and not "capomaestri".
In the past there had been no guarantee that the author of a
winning design would also be employed in its execution, and the
offices of capomaestri had continued, for example, undisturbed
90 P. SANPAOLESI, concorso del 1418-20 per la Cupola di S. Maria del
Fiore", Rivista d'Arte, XVIII, 1936, pp. 321-344; R. BONELLI,"Antonio Ma-
netti, 'tendenzioso fino a travisarei fatti"' in Filippo Brunelleschi. La sua opera
e il suo tempo, cit. (see note 74), II, pp. 923-932; F. BORSI, 'E specialmente
nella edificazionedella tribuna fummo concorrentiFilippo e io anni diciotto a
un medesimo salario..." in Lorenzo Ghiberti nel suo tempo, cit. (see note 76),
II, pp. 541-552; A. BOVE-S.BRICCOLI BATI-S. Di PASQUALE-B. LEGGERI, "Que-
stioni marginali concernenti il rapporto Ghiberti-Brunelleschi nella costruzione
della cupola di Santa Maria del Fiore", ibid., pp. 553-560.
91 MANETTI, op. cit. (see note 74), pp. 88-90. On the significanceof this
division see PROCACCI, cit. (see note 74), p. 42.
afterthe approval the committee
of model. In the case of the
cupola,however, whoseself-supporting principle depended upon
innovative construction it
techniques, must have seemedadvis-
ableto institutionalize bondbetween
the invention execution
through new supervisory office. The actingcapomaestro was
alsoincluded the roll of threeprovveditori a suitable
in as finan-
cial arrangement his working
above wages,'but the role of Bru-
nelleschi Ghiberti musthavebeenconceived part-time
pervisory consultation, whoserecompense therefore dis-
proportionate. the of
Furthermore, presence Ghiberti thiscapa-in
city cannothavebeen astonishing.Plurality supervision
responsibility a long-established
tradition the Opera:we
can recallthe not always harmonious of
Ghiniin the 1360s, as well as the advisory boardswhichhad
neverceasedto exist. In the face of a dividedcity, the Ope-
ra, a civic delegate spending publicfundsfor the realization of
the city'smostprominent monument, neededto involveany or-
opposition a strongmajority consensus.Wide appro-
val for the fundamentally Brunelleschian modeland start-up pro-
gramwas beingsought,andone of the best waysto accomplish
this was to offer supervisory to
representation other strongly
contestants. consuls Operaiand desired plurality
at anycost,with or withoutGhiberti, the electiondocument
namestwo alternates, both of whomhad produced modelsand
designs:the painter, Pesello,andthe dabbling scholar, Giovanni
di Gherardo Prato.
Manettidescribes Brunelleschi first grudgingly
but soonchafing the
against unwelcome partnership Ghiberti.
Eventually Filippocouldnot resistforcingthe situation with a
verypointedpractical joke,his feignedillnessand absence from
the worksite its calculated
and devastating consequences Lo-
renzo,a tale told with the samewit and psychological acumen
that pervade Novelladel Grasso.93
the Brunelleschi successfully
92His name appears added as an afterthought at the end of the text
conserved in the guild records, while it directly follows that of the other two
in the Opera rendition (respectivelyDOREN, Ioc. cit. [see note 86], p. 262, and
Cup. 71). Subsequently,however, his office is usually describedsimply as "capo.
93 MANETTI, op. cit. (see note 74), pp. 92-95.
exposed his rival, who was trapped into taking on and inevitably
botching the design of one of the cupola's internal chain systems.
But if the challenger was gratified in the records with the addi-
tional epithet of "inventor" of the cupola, his partner's support-
ing faction was still strong enough to maintain him in office.'
As the dome rose, inclining its self-bonded masonry over the
great void of the crossing, (Fig. 7) the risks, the perils, the sheer
vertigo of the situation nagged all those concerned, from the
lowliest manovale to the loftiest merchant statesman. Under this
tension the semblance of equality between the two provveditori
finally crumbled. There is in fact an interruption of Ghiberti's
salary from July 1425 through February 1425/6.'5 Now, he had
more lucrative projects to attend to, such as the third set of
Baptistery doors recently commissioned from him at the annual
salary of two hundred florins.' Brunelleschi, on the other hand,
had of necessity behaved very much like a full-time employee of
the Opera, closely following every step of the work, checking con-
struction materials,devising the myriadsecuritymeasuresembedded
in the fabric of the cupola: the success of his invention depended
on these details. From the summer of 1425 to early 1426, a gen-
94 MANETTI (ibid., p. 95) dates this episode - which he certainly learned
by word of mouth, probably Brunelleschi's - to 1423 on the basis of his
research, which turned up an uscita document of 31 August of that year re-
warding Filippo, "inventoree governatoredella muragliadella maggiorecupola",
for invention, particularly new model for the great wood chain ("catenagrande
del legname della detta cupola"). This correspondsto the two versions of the
stanziamentoin Guasti (Cup. 177): in the former, however, "chatenammagniam
ligaminis" should read "chatenammagniamligniaminis".(AOD, II-1-83, c. 68).
This was the first and only wood chain installed. Since the story also says
that Ghiberti hoped to copy his chain from that around the Baptistery vault,
which is in wood, the dating is probably correct. Manetti does not treat the
amendments to the 1420 program approved after consultation in early 1422
(SAALMAN, op. cit. [see note 2], p. 74; Cup. 52).
95 Cup. 74, 28 June 1425. Payments of his salary are lacking not only
through 1425, but through the first two months of the following year (AOD,
II-4-9 and II-4-12).
96 KRAUTHEIMER, op. cit. (see note 72), II, p. 370, doc. 36 (2 Jan.
1424/5). The contract included an exclusive clause, "che finche non sara
finita, non possa pigliare a fare altro lavoro". Its intent was certainly that the
artist take on no new commissions,but it might also have been interpreted as
being incompatiblewith his position at the Opera del Duomo. Nevertheless, he
remained active in the context of the cupola, producing five bronze bosses for
the hoist, paid to him in October 1425 (Cup. 141).
eralreview plans thecompletion thevault's
of for of upper reaches
involved new roundof extensive
a consultation citizens
experts over a seriesof modelsanddesigns.'7 Clearly, guild
andthe Opera the needof outside
felt support the last,breath-
taking leap,andonce againthey drummed a consensus a
new,detailed program for the remaining of the dome,com-
from Ghiberti,Battista and
of the four cupolaofficials.98 to
The decision proceed according
to these specifications takenjointlyby the consuls,Operai,
and cupola officials,who technically assumed for
the text theirappointees submitted.If this is the juridical
content theirdeliberation, practical
of its significance never-
thelessclearto the Opera's notary,who succinctly indicatedin
the relative thatthe cupola
title wouldproceed "according the
designof Filippodi ser Brunellescho".9 sameact returned
to the matter the cupola's
of provveditori: 1 March
Brunelleschi, whose laborand diligence was praised, would be
raisedto one hundred florinsper year for full time; Ghiberti,
whosediligence aloneis cited,was reinstated a salary thirty-
six florinswiththeobligation putin onehoureachworking
Manetti attributes partial
this victoryto the good officesof
Filippo's friends,10 it is well to bearin mindthat the archi-
tect'sown little politicalcareer was at its apex in this period.
97 SAALMAN, op. cit. (see note 2), P. 119 and docs. 213.4-8; Cup. 58-60.
98 Cup. 75: deliberation of 4 Feb. 1425/6 containing and sanctioning the
report dated 24 Jan. 1425/6.
99 "Quod laborerium cupole magne sequatur secundum designum Filippi ser
Brunelleschi" (AOD, II-2-1, c. 170v). This important phrase, omitted by
Guasti who systematically excluded such marginalia from his transcriptions, was
published in my entry on the document in Filippo Brunelleschi: l'uomo e l'artista,
catalogue of the documentary exhibition at the Biblioteca Medicea Laurenziana.
Florence, 28 May-31 December 1977 (Ministero per i Beni culturali e ambientali,
pubblicazioni degli Archivi di Stato, XCIV), Florence, 1977, pp. 46-47. It must
have been common knowledge that the cupola's design and execution were es-
sentially Brunelleschi's. Cf. the registration in the Priorista of Paolo di Matteo
Fastelli Petriboni for July-August 1420, published by U. PROCACCI, "Sulla cro-
nologia delle opere di Masaccio e di Masolino tra il 1425 e il 1428", Rivista
d'Arte, XXVIII, 1953, pp. 3-55: 33, note 46, and the context provided for it
by G. TANTURLI, "Rapporti del Brunelleschi con gli ambienti letterari fiorentini"
in Filippo Brunelleschi. La sua opera e il suo tempo, cit. (see note 74), I, pp.
1'X MANETTI, op. cit. (see note 74), p. 95.
Having repeatedly sat on several city councils, he served his
one term as Prior in May-June 1425.101 If politics were a factor,
then temporaryreduction of Brunelleschi's wage in the spring of
1431 may be related to the decline of his appearancesin civic
elective offices, a circumstance which, it has been in turn sug-
gested, may have depended on his ill-fated attempt to flood Lucca
in support of the beseiging Florentine troops a year earlier."
With this exception, however, the pacts of 1426 remained in
force until the completion of the dome in 1436. Filippo was
de facto in charge, but never alone; he was legally bound to the
execution of a program which, though widely underwritten, was
essentially his own, pushed and prodded through, and perhaps
refined by, the traditional consultative processes of the Opera.
Even his own corrections to his calculations for the diameter of
101 D. FINIELLOZERVAS, "Brunelleschi's Political Career", Burlington Magazine,
CXXI, 1979, pp. 630-639, where it is also suggestedthat the architect'srepeated
appearances civic councils from 1418 may have won him status and credibility
in the cupola competition. Recent historical studies have emphasized the im-
portance of both personal and political connections in the partisan society of
quattrocentoFlorence. See, for example, D. KENT, The Rise of the Medici:
Faction in Florence, 1426-1434, Oxford, 1978. As a curiosity it may be noted
that during Brunelleschi's priorate,when he could not have been present on the
constructionsite, his salarywas neverthelesspaid in full (AOD, II-1-86, c. 51v);
the full-time clause, however, took effect only on 1 March 1425/6.
102 For an analysis of Brunelleschi's remarkable military strategy which
exonerateshis calculationsfrom the responsibilityof its failure see P. BENIGNI-
P. RUSCHI, "Il contributodi Filippo Brunelleschiall'assediodi Lucca"in Filippo
Brunelleschi. La sua opera e il suo tempo, cit. (see note 74), II, pp. 517-
533. Nevertheless the debacle may have influenced his subsequent political
fortunes, as first suggested by N. RUBINSTEIN, "Palazzi pubblici e palazzi pri-
vati al tempo di Brunelleschi",ibid., I, pp. 27-36: 36, note 67, and further
elaboratedby ZERVAS, lOc. cit. (see note 101), pp. 635-636. His Duomo wages
were withheld not only for the period of his presence in Lucca (20 March-12
June 1430), but through 15 September 1430, when the Operai had to send
for him in Figline (Cup. 101 and AOD, II-4-12, c. 130; II-4-13, cc. 2v, 5v).
On 16 Feb. 1430/1 the Operai "pro bona et iusta causa" reduced his wage
to 50 florins per year (Cup. 82). Although the deliberationstates that it is to
be effective from 1 April 1430, hence retroactively,the year is probablyan error
for 1431, since paymentsare recordedat the 100 florin rate from 15 September
1430 to 13 March 1430/1 (AOD, II-4-13, cc. 5v, 10v), while the 50 florin rate
applies from then to 30 June 1431 (ibid., c. 13v). Ghiberti had meanwhilebeen
entirelyoff the payrollfrom 1 July 1430 through30 June 1431. From 1 July 1431
both provveditoriwere reinstatedaccordingto the pacts of 1426, and they remain-
ed in their positions throughJune 1436 (Cup. 83, 84, and "Correzioni", 188).
oculushadto be approved the prescribed
the cupola's by
From 1420 to 1436, then, Brunelleschi was co-provveditore
of the cupola, with no official mandate or role for the rest of
the church. Nevertheless he was always there and, as work pro-
gressed, ever more respected. It was therefore perfectly natural
that he figure as a frequent and authoritative consultant of the
Opera on other matters. The space of this essay does not permit
examination of these many concerns, ranging from the interior
decorations of the church to the fortification of outlying towns,
entrusted by the Commune in times of emergencyto the seasoned
personnel of the cathedral. Limiting our inquiry to the archi-
tecture of S. Maria del Fiore, we can observe how already, during
the years of his mandate for the cupola, Brunelleschi was drawn
into the discussion of broader questions of priority for the com-
pletion of the church.
In 1429 the Operai ordered Brunelleschi and Ghiberti to exe-
cute a model of the whole church, including of course the cupola
and the new east end, but also the new fasade and chapels off
the nave aisles proposed as buttressing in response to new cracks
that had opened in the nave, presumablyunder the thrust of dis-
placement from the rising dome."' The Operai reported to the
consuls about this in 1430.105 But with all efforts directed towards
the cupola, there can be little surprise that the Operai, having
repeatedly conferred about these grave dangers with the provve-
ditori, Battista d'Antonio, and numerous other experts, opted for
the less elegant solution of a new system of tie rods which was
103 The consuls had delegated the power to modify the 1426 program to
the Operai and the cupola officials (Cup. 75). Since the latter office was fused
with that of the sacristy officials a year later (Cup. 10), the Operai alone were
competent to approve adjustments proposed jointly by Brunelleschi, Ghiberti
and Battista d'Antonio in 1432 (Cup. 247-8 and SAALMAN, op. cit. [see note 2],
pp. 132, 138). Since a new reduction of the diameter from 10 to 9 2/3 brac-
cia proposed by Filippo and Battista the following year (Cup. 251) implied the
reversal of the Operai's previous deliberation, it had to be enacted by them
jointly with the consuls.
104 Cup. 61 (22 Sept. 1429). Payments to the carpenters who executed
the model are recorded from 29 November 1429 to 2 May 1430: Cup. 62-66.
The interpretation of the documentation follows the analysis of SAALMAN, op. cit.
(see note 2), pp. 128-131.
105 Cup. 67 (2 May 1430).
in fact installed to Brunelleschi'sspecificationsfrom 1431.'1? Still,
the architect preferred the more radical chapel solution, and in
1434 the Operai directed him to complete the ideal model so that
the whole "veritas" could be known of the dome, the lantern,
the rest of the church, and the proposed chapels."0
This truth exceeded the Opera's immediate resources. Seve-
ral matters pressed as the closing of the cupola approached.The
lantern would of course be required for the perfection of the
dome. Under the cupola, the whole east end, which had served as
long as most people could remember as a busy, unpaved, clut-
tered construction yard set off from the officiated nave by rough
partitions, now presented itself as the biggest and barest imagin-
able ecclesiastical container. With Brunelleschi's completed mod-
els before them, the consuls and Operai sat down in August of
1434 to work out their priorities. Nothing was underratedin im-
portance; but the preference of the populace, said to be crying
out for the entire church to be opened, was clear."0 Having as
usual solicited counsel from many quarters, the officials could
only conclude that the chapel scheme was not a popular priority.
After the cupola, the next effort would be to finish the interior
of the east end so that divine offices could take place there as
soon as possible. After that, the lantern would be begun, and
only when it was finished could other structuralmatters be taken
in hand. The convenient, if pressing, deadline for the rushed
preparationof the east end was provided by Pope Eugenius IV,
who had agreed to preside at the consecration of the new cathe-
dral on 25 March 1436 before departing from his first Florentine
sojourn."? When the event took place, some work was still pend-
ing on the cupola, which received its finishing touches and the
benediction of the bishop of Fiesole only in August of that year.'10
106 The fullest body of transcriptions concerning these catene is now avail-
able in G. POGGI, Duomo di Firenze, II, posthumousedition, ed. M. HAINES,
Florence, 1988, pp. 173-181. Additional references in SAALMAN, cit. (see
note 2), docs. 253.9-10, 253.16, 253.20-22.
107 Cup. 70 (17 June 1434).
108 Cup. 259 (12 Aug. 1434): "totum populum clamare magno desiderio,
quod detur modus ad habitandumtotum templum".
'09 POGGI,op. cit. (see note 106), II, pp. 195-201.
110Cup. 261 (31 Aug. 1436).
Technically, was now possible proceed with the lantern.
(Fig. 8) Although already 1432 the Operaihad incitedBru-
nelleschi makea modelof it,"'in 1436 the Opera
to once again
paidnot only for another modelby him, but also for those of
othercontenders, notably of his inevitable
that rival,Ghiberti, and
no less thanthreeentries submitted a new challenger, car-
penterand aspiring architect,Antoniodi ManettoCiaccheri."2
In thesummer consuls already
the had cometo the Opera exam-
ine Filippo's modelover lunch,but the othercontestants could
be with an insidedecision. The Operai fact in
proceeded through remarkable seriesof consultations
a new be-
fore the last day of 1436, when they finallyapproved Brunelle-
schi'smodel,statingtheirwish to beginthe lantern desiredby
the wholeFlorentine populace invoking powersconceded
themby the councils, people,andCommune Florence by
the Wool Guild."3Firsttherehad been a greatcouncilof the-
ologians, learned men, architects, goldsmiths, otherartisans, and
Florentine citizens.Thenthreerounds votingby separate
missions, eachcomprising architects, painters, gold-
two two two
smiths, mathematician, two capable
one and citizens, handed
downopinions in writing. Finally, committee eightpromi-
nent citizens,including Cosimode' Medici,draftedthe report
thatwas sanctioned the finalact of the Operai. The lantern
shouldbe constructed according Filippo's
to model- described
as the strongest, lightest,best lit, andmostwater-resistentand-
underhis personal supervision. But the Operaishouldconvoke
andexhorthim to put aside"omnes rancores" to consent
incorporate his work any improvements,
111 Cup. 250 and 264 (30 Oct. 1432). This act emerged in the context of
the decisions over the span of the opening to be allowed at the summit of
112 Cup. 265-272 (19 Mar. 1435/6-19 Apr. 1437), including delayed pay-
ments for the models submitted by Antonio, who is of course not to be confused
with the biographer of Brunelleschi. I have attempted a fuller analysis of the
documents and their relationship to Manetti's Life in The "Sacrestia delle Messe"
of the Florentine Cathedral, Florence, 1983, pp. 64-66. In the same period the
Operai had called for models for a new hoist to lift the lantern materials
above the cupola (Cup. 129: 7 Sept. 1436).
113 Cup. 273 (31 Dec. 1436).
be derived from his competitors' entries. Once again, the archi-
tect's design had become a universal project.
During the deliberations over the models for the lantern, Bru-
nelleschi had not always been without a job. Already in October
the Operai had found a formula to reinstate his salary by com-
missioning from him the tile roofing of the vaults of the three
tribunes at a fixed rate plus the personal recompense of one
hundred florins per year."'4 The interim arrangementapparently
remained in force for over a year after the approval of his lantern
model. Finally, in February 1437/8 the Operai installed him
officially, at the same pay and with the comprehensive title of
provveditore of the cupola, its lantern, and the whole church; as
required by statute, the consuls approved this provision and ex-
tended it for another year in December."5 Although work on the
lantern had started slowly with the search for the vast marble
blocks required, Brunelleschi's stable position in the Opera would
not again be challenged."16 joint deliberation of the consuls and
Operai in 1443 went so far as to authorize the extension of the
traditional annual term to a permanent election for the rest of
his life.117 No one else, according to the justification of this ex-
ceptional act, could be found to take on the seemingly impossible
task of devising how to hoist and position the great marble blocks
for the lantern.'18But there was a hitch. Filippo was to provide
114 26 Oct. 1436, published in C. VON FABRICZY, "Brunelleschiana", Jahrbuch
der Kdniglich Preussischen Kunstsammlungen, XXVIII, Beiheft, 1907, pp. 1-84:
38; cf. Filippo Brunelleschi: l'uomo e l'artista, cit. (see note 99), p. 54.
115 Cup. 85 (13 Feb. 1437/8), 87 (16 Dec. 1438).
116 Temporary reductions in 1439 and 1440, regarding all salaried employees,
were dictated by the financial straits of the Opera: Cup. 88-91. For the marble
supply, see Cup. 277ff and SAALMAN, op. cit. (see note 2), pp. 141ff.
117 Cup. 93 (12 Apr. 1443). The only precedent for this act was the consuls'
and Operai's offer (26 Apr. 1439) of a life appointment to Battista d'Antonio
if his son would agree to rescind the arrangement whereby he received his father's
pay. This proposal was not enacted. The consuls finally appointed Battista
d'Antonio as capomaestro of the Opera for life on 17 June 1450. He died on
11 December 1451. SAALMAN, op. cit. (see note 2), pp. 188-189 and docs.
118 The new hoist was being devised in the spring of 1443 (ibid., p. 164).
The text of the election authorization closely recalls Manetti's account of Bru-
nelleschi's tactics when the vaulting of the dome was under discussion. The
demonstrated difficulty or impossibility of the task in the hands of others made
a full andclearexplanation the methods intended use.
of he to
Perhaps old man,with his notoriously
the cageypersonality, de-
murred this clause,for a similar
offerwas repeated 1445
withoutthe provisoand with the addedjustification his that
ingenuity inventions savedthe Operaand Comune an
There wouldhavebeenlittle timeto enjoythis unwonted job
security, FilippoBrunelleschi on 15 April 1446.1w The
following December guildconsuls
the convened remember
in an extraordinary way."2'Countless and
colleagues citizens had
reminded themof the architect's and
achievements inventions and
of all the expenses had thereby
that been saved. It was fitting,
not only for his personal honorand famebut also for that of
the Commune, the guild,at whosebehesthe had worked,
demonstrate gratitude.The consulstherefore
its decreedthat a
site in or near the church designated his final resting
place."2 Opera a
wouldsupply suitable amount marble
be carved the expense Filippo's (his adopted
at of heir son, the
sculptor, di and
Andrea Lazzero), an appropriate would
be composed. In early 1447, the Operai,togetherwith two
consuls, willingly to for
attended theirmandate the burial honors
and eternal fameof the most "eloquent ingenious"
leschi,who was saidto havebeen assisted onlyby God and the
Virginin the realization of
withoutarmature the greatdome."
A marble withthe inscription
slab "FILIPPUS ARCHITECTOR"
him the winner by exclusion: "... intellecto quantum non tantum difficile creditur,
sed difficilimum, conduci super dicta Cupola lapides et marmora... et quod
nullus offert, nec reperitur qui predicta incipere et ad perfectionem conducere
dicat..., et de periculis maximis dubitatur, prout per operarios Opere dicte ec-
clesie ipsis consulibus relatum fuit: et auditis ipsis operariis referentibus, quod
Filippus ser Brunelleschi infrascriptus obtulit eisdem perpluries dictum laborerium
incipere et ad perfectionem deducere vere credit, et hoc experientia demostrare,
si eidem de infrascripta provisione provideatur..." (Cup. 93).
119 Cup. 95 (7 Dec. 1445). His accomplishments "non solum cedunt ad
ornatum et decus dicte ecclesie, sed etiam magnifici Comunis Florentie".
Cup. 96 and p. 201.
121 CUp. 119 (30 Dec. 1446).
122 Private burials in the church had long been subject to the approval of
the consuls: cf. SMF 70, p. 94 (16 June 1357).
123 Cup. 120 (18 Feb. 1446/7).
would mark the site of his tomb under the brick pavement, while
nearby in the first righthand bay the wall monument would com-
prise his portrait and verses commissioned from Carlo Marsup-
pini.'24(Figs. 9, 10) The Florentine chancellor rose to the occa-
sion with the familiar text which still greets the public entering
S. Maria del Fiore: "Not only the marvellous dome of this cele-
brated temple but also the many machines he invented with di-
vine genius stand to prove how Filippo the architect excelled in
the Daedalian art. Wherefore, because of the distinguished gifts
and singular virtues of his mind, on the 15th of the Calends of
May, 1446, a grateful fatherland decreed that his worthy body
be buried in this grave". 15
A monument in the cathedral was the highest civic recog-
nition, reserved for a rarified circle of public heroes, including
Luigi Marsili, John Hawkwood, and Niccolo Tolentino, whose
number the Commune had hoped to extend posthumously to such
luminaries as Dante and Petrarch."2 An excellent example of
how Brunelleschi's tomb was regarded by Florentines in the de-
cades after his death is to be found at the very outset of his
biography by Manetti. The Life is in fact addressed to Girolamo
Benivieni, who, admiring the true story of the intricate practical
joke excogitated at the expense of I1 Grasso, wants to know more
124 The slab was lost when the church was repaved in marble, but the
tomb was found in 1972 during excavations under the cathedral: G. MOROZZI
in S. Reparata. L'antica cattedrale fiorentina, Florence, 1974, pp. 14-15, Figs.
40-41. Its inscription reads "CORPUS MAGNI INGENII VIRI/PHILIPPI
S. BRUNELLESCHI FLORENTINI". For the artist's death mask and further
documentation regarding the execution of the tomb see G. POGGI, "La 'maschera'
di Filippo Brunelleschi nel Museo dell'Opera del Duomo", Rivista d'Arte, XII,
1930, pp. 533-540.
125 "D. S. Quantum Philippus architectus arte daedalaea valuerit, cum huius
celeberrimi templi mira testudo, tum plures machinae divino ingenio ab eo ad-
inventae documento esse possunt. Quapropter ob eximias sui animi dotes singu-
laresque virtutes, XV Kl. Maias anno MCCCCXLVI, eius b. m. corpus in hac
humo supposita grata patria sepelliri iussit". The Operai had approved the
inscription with an act of 19 May 1447, where the text appears with minor
variants (Cup. 121 and POGGI, op. cit. [see note 106], II, p. 131).
126 See the summary of a program of patriotic tombs planned by the
Commune between 1393 and 1396 with the suggestion that it was inspired
by Coluccio Salutati in E. BORSOOK, Mural Painters of Tuscany from Cimabue
to Andrea del Sarto, 2nd ed., Oxford, 1980, p. 76; many of the relative docu-
ments in PoGGI, op. cit. (see note 106), II, pp. 123ff.
aboutits inventor why "he received
and suchan honoras to be
buriedin SantaMariadel Fiore,and his very likeness,as they
say, placedthere sculpted marblefor eternalmemory, with
suchan epitaph With thesewordsManetti
"." launches his
laudatoryaccount, whileourshas comefull circle. Although the
architect's might seem to have been
constrained the conventions publicpatronage,
by of theseproce-
dures alonecouldultimately the
provide highest formof individual
recognition, publicstatement.At the sametime the Opera del
emerges a remarkable institution. in
Bound on all sides
by its juridicalpositionand publicresponsibility, found in
theseveryties the strength promote mostdaring projects,
constructing maintaining civic consensus which alone,
with sustainedfinancialeffortoverdecades, could
'7 MANETTI, op. cit. (see note 74), p. 47.