Suger's Miracles, Branner's Bourges: Reflections on "Gothic Architecture" As Medieval Modernism Author(S): Marvin Trachtenberg Reviewed Work(S): Source: Gesta, Vol

Home , Le Mans Cathedral, Medieval architecture, Romanesque architecture, Suger

Suger's Miracles, Branner's Bourges: Reflections on "Gothic Architecture" as Medieval Modernism Author(s): Marvin Trachtenberg Reviewed work(s): Source: Gesta, Vol. 39, No. 2, Robert Branner and the Gothic (2000), pp. 183-205 Published by: International Center of Medieval Art Stable URL: http://www.jstor.org/stable/767145 . Accessed: 15/03/2012 06:50

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

International Center of Medieval Art is collaborating with JSTOR to digitize, preserve and extend access to Gesta.

http://www.jstor.org Suger's Miracles, Branner's Bourges: Reflections on "Gothic Architecture" as Medieval Modernism

MARVIN TRACHTENBERG Institute of Fine Arts, New York University

Abstract eval text and image: a highly focused reexaminationof Abbot Suger's controversial writings on the abbey of Saint-Denis, When it becomes clear that a given interpretative model and a new look at the that launched Robert set contains irreducible contradictions and is great building (or of criteria) Branner'scareer, the cathedral of incapable of resolving certain stubborn problems, it is time Bourges. for a new paradigm. Recognizing that scholarship on Gothic There is, however, an issue that first needs to be aired. architecture has long since reached this impasse, this study Some readers may wonder why I believe a redefinition nec- identifies new terms of understanding in the oppositional essary at all, for surely, one might imagine, the huge literature pair "modernism"and "historicism."These two transhistor- on the Gothic somehow terms for its under- ical are to numerous ad- provides adequate concepts found offer interpretative however, tells us other- vantages over such traditional categories as "skeletalform," standing. Historiographic analysis, "diaphaneity," "linearity," "diagonality," "square schema- wise. It is not that readings of the Gothic in the familiar terms tism," and the like, all of which are severely burdened by the of rib vaulting, skeletal structure, diaphaneity, diagonality, well known constrictive problematics of "style," among other linearity,baldachins, geometry, scholasticism and so forth are difficulties. Properly defined and understood, the modernism/ in themselves uninformativeor Rather, as historicism is shown to numerous necessarily wrong. paradigm offer advantages Louis Grodecki observed in 1977, none of the inter- in studying medieval architecture across a wide range of con- existing cerns, from descriptive, componential, and formal analysis to pretative models offers "a firm rigorous definition" of the social, intellectual, and contextual integration. The entire ap- Gothic.2 As a group they cannot be assembled into anything paratus of the great cathedrals is seen in a radically new, more than an unwieldy and finally self-contradictory inter- comprehensive way, and such intractable historical problems of and as that the "Gothic column" are re- pretative bricolage mainly nineteenth-century early of seemingly oxymoronic methods and discourse. This unsat- solved. Theproposed paradigm is further explored and tested twentieth-centuryanalytic regarding both word and image in medieval architecture, in a isfactory situation has not improved recently, in fact quite the new reading of the controversial texts of Abbot Suger on Saint- opposite. In the past two decades or so Gothic architectural Denis, and in a revised solution to the Bourges "problem." studies have turned sharply away from the broad questions that preoccupied earliergenerations (no matterhow limited or The great trans-Europeanarchitectural movement endur- irrelevant their methods may now seem) toward the study of ing some four centuries called "Gothic" has remained veiled the particular:the functions of individual spaces, questions of and perplexing despite many generations of solid and often patronage and liturgy, the specific social, political, material, brilliant scholarship. In an earlier essay I proposed a new con- and representationaldimensions of architecture.3In this recent ceptual frameworkthat I hoped might serve as a point of de- work (much of it highly laudable on its own terms) the old parture for understanding this enigmatic architecture.1This question, "What is Gothic?" which runs as a central theme adumbrationof my idea, however, occupied only a few intro- through Paul Frankl'sexhaustive survey of its long historiog- ductory pages of an article that was mainly dedicated to re- raphy, rarely is explicitly asked, because implicitly it is con- vising our conception of Italianarchitecture during the period. sidered somehow already answered, or conversely, because it This initiatory statementwas thus extremely compressed, and is deemed irrelevant and in any case impossible to answer.4I left much undefined,unresolved, and unexplored. The follow- refuse this attitude of semantic innocence or terminological ing pages are directed not to Italy but to the essentially French futility, and emphasize that our definition of "Gothic"remains genesis and primarydevelopment of the "Gothic."Here I re- a central point of reference in virtually all discussions of formulate,interrogate, clarify, and above all extend my argu- European monumental architecture of the twelfth through ment. I pursue certain necessary considerations regardingthe fifteenth centuries. The question hangs heavily in the air as problematics of its terminology as well as its many possible soon as the word "Gothic"emerges from our mouths or in our implications, some of which lead toward ratherradical histo- writings, not to mention our explications of the architectureto riographicand interpretativepositions. Following the study of nonspecialists and the public at large. We function as histo- a numberof key theoreticalissues, I will test the viability of the rians of this architecturalzone in a most curious way, con- new interpretativemodel through practical analysis of medi- stantly speaking of the "Gothic"as if we securely knew what

GESTAXXXIX/2 ? The International Center of Medieval Art 2000 183 it was and that it actually existed, knowing all the while that in any case.' As with Gothic we tend to use this term either the one thing certain about the subject is our very lack of any unthinkinglyor disparagingly,that is, with the idea that it rep- such articulated consensus. Whether we like it or not, the resents, if anything, a rathernaive understandingof the pre- "Gothic"question remains with us, ratherlike an elephant in Gothic. I propose instead, in a mannersomewhat analogousto a room that won't go away no matterhow much we pretendto my etymological revaluationof "Gothic,"that the originalcore ignore it. meaning of the word "Romanesque"also has a certain powerful validity (at least for immediate purposes at this point in A new paradigm: Gothic as medieval modernism the argument). In fact, I would argue that it provides a more accurateassessment of the period it denotes than all later aca- Rather than reviewing how problematic are most of the demic analysis in terms of square-schematism,bay systems, concepts that we currentlyuse to study the period (especially radiating chapels and the like, which, like the usual terms for how they almost inevitably are shipwreckedon the rocks and "Gothic,"do not hold up under hard scrutiny in regardto ac- shoals of stubborn exceptions and contradictorymanifesta- curacy, compatibility,or comprehensivenessof applicationto tions of form), I shall directly interrogatethe term "Gothic" this highly varied architecture.Instead, the early nineteenth- itself, by asking why we persist in calling Europeanarchitec- century term Romanesque was on the mark, or nearly so, or ture of the twelfth throughfifteenth centuries after a barbarian at the very least more historically accurate generally than la- tribe of late antiquity.Are we here merely being creaturesof ter nineteenth- and early twentieth-centuryrumination on the meaningless, ingrainedhabit, or might the word "Gothic"not architecturein question.Pre-Gothic medieval architecturewas, contain a germ of etymological truth?I suggest that the word, quite simply, Roman-esque.It was, in other words, deeply his- despite the dimension of blatant absurdity in its usage, in- toricizing. That it often embodied "modernist"tendencies as volves a key to understandingthe period. Here we must look well is also true. In fact, perhaps the fairest characterization to the linguistic practice of the Renaissance, which first made of what we call the Romanesque period would be in terms of the connection between the word "Gothic" and the architec- a conflict, instability, an unresolved tension between the two ture in question. Essentially the connection is simple. In the opposing currentsof historicism and modernism,in which the eyes of the Renaissance, the Goths were the destroyers of former tended to predominate (although not in any progres- Rome and its architecture.They were, in other words, the lit- sive way or with any clear pattern).In this reading, the Gothic eral embodimentof anticlassicism. This belief, of course, was turn would amount to a shift in orientation, a move towards bound closely with the Renaissance view of medieval archi- the resolution of the contest, away from historicism and in tecture: its essence (by necessity) was its anticlassicism. But favor of an ascendant, eventually dominant modernism.8 the Renaissance also used another term for the recent post/ The components of historicism and modernism varied non-antique architecturalphenomenon, "lavori moderni."5If widely in usage and strengththroughout Romanesque Europe, we put this term together with "Gothic"-that is, put together a stubborn fact that serves to invalidate the usual models of the two earliest descriptive terms for the movement-we descriptive analysis (which cannot accommodate this com- have an architecturethat is both "modern"and "anticlassical," plexity) but instead fuels my proposed approach.Some areas or going a step further,"modernist" and "antihistoricist,"which were nearly purist (through various chronological spans and are two ways of saying nearly the same thing.6 I submit that in varying density of occurrence) in stressing one tendency these early sources were closer to the heart of the matterthan over the other. For example, the medieval churches of Rome, most later, purportedlyscientific scholarshippreoccupied with such as the nave of San Lorenzofuori le mura,were so faithful rib vaulting, skeletal structure,scholasticism, diaphaneity,ge- to early Christianmodels, that is, so immaculatelyhistoricist, ometry, diagonality, and so forth. In fact, I propose that were that to "Romanesque"scholarship they represent intractable it possible to give later medieval architecturea name more exceptions to standardinterpretative models, while to the non- descriptively accurate and less loaded with misinformative specialist, according to the usually received explanation, they connotations than "Gothic"-while at the same time retain- are Romanesqueonly by their date (Fig. 1).9Normandy, on the ing the historically and conceptually legitimate, hidden mean- other hand, was in certain aspects sometimes so strongly mod- ing of that term-that name would be "medieval modernism." ernist that instances such as Saint-Etienne in Caen (Fig. 2) This term admittedly will strike many readers as strange could be reasonably included by Ernst Gall in his admirable and problematic, and much of this paper is dedicated to its book on early Gothic (which, however, implicitly posed the clarification and justification. By this notion I refer to some- conundrumof premature,anachronistic manifestation).10 More thing far more concrete and motivated, conscious and indeed characteristic of the period, or considered central to it, are self-conscious than any Focillonesque "spirit of modernism" works like Autun Cathedral(Fig. 3) or its model, Cluny III, in the Gothic, an essential differentiationthat I will elaborate which embody a complex, often tense or conflictedrelationship later. For the moment the idea becomes clearer if we go back between historicizing elements, such as classicizing columns, another chronological step to the period generally called pilasters, vaulting, ornament, and "normative"proportions, Romanesque, a term with which "Gothic" usually is opposi- on the one hand, and on the other modernist tendencies to- tionally paired, hence requiring some critical attention here wards the bay system, spatial fragmentation,schematization, 184 ...... X ...... M...... Jl??l K, ?XXXXXO?..:...... 117?t :XX X.: MR

...... 'M ...... :::.A ...... X, ...... K?K ...... 77S FIGURE 1. Rome, San Lorenzo fuori-le-mura, ZN view of nave towards east (photo: author).

......

...... ? ...... aa~ : ...... OZ ...... F . .. . .

......

FIGURE 2. Caen, Saint-Etienne, nave elevation (photo: author). FIGURE 3. Autun, Saint-Lazare,forechoir (photo: author). 185 S...... W/;m

...... •!ii!i~i••? i iiixiiwoof ...... ii:iiiiii~~ Itiiit~ii?k~iiiigii:~~~s~iiii ::::-4'1111!'Aiiiii~~ ~ : iiiiiiiiif-iiiii _ _ _ : iiiiiiiiiiiiiiiiiii 1~4~918111~E~:%ii;:-~l;:--~~1~PI::&& ~ni,%~IN5~

1 ... i71, r ii :ii iiNor

I ...... X...... ii i f i ii i i i i i i - i i i ii ii~i'iii'iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii iiR ...... Aml?i ......

...... ::ii ......

41 .

FIGURE 4. Speyer, cathedral, early version (drawing: Wenzel Hollar). FIGURE 5. Speyer, cathedral,second version, nineteenth-centurylithograph.

structural rationalism, and attenuated proportions, a tension "transitional"movement towards the Gothic, a misinterpreta- that, of course, was part of the pervasive and continuing am- tion that never has been quite put to rest.12The Romanesque biguity of the conflicted medieval attitude toward antiquity.11 was not driven by any unconscious process of stylistic evolu- The great exemplar of this interaction was the imperial tion but rather was guided by a strongly self-conscious view cathedralat Speyer. The first Speyer of the early eleventh cen- of history, of the present in relationship to the past, in which tury took up the nearby fourth-centuryConstantinian basilica the latter was not to be relinquishedin architecturebut instead at Trier, turned it outside in, and added an attenuated half- emphasized (the strengthof the historicizing impulse depend- columnar layer of bay-dividing elements (Figs. 4, 6). What is ing on specific circumstances). The Romanesque did not want particularlytelling is that the second Speyer as rebuilt a half- vainly to be "Gothic" but-not unlike our own recent post- century later is more rather than less historicizing (Fig. 5). modernist architecture,as well as much of the architectureof Not only were huge Roman-style groin vaults erected, but the the nineteenth century-was both modernist and historicist at main piers received a double order of massive columns that the same time. are far more antique in proportions than the original attenu- In the period of medieval modernism-or what we usu- ated forms. As a whole the interior of this prodigious cathe- ally call the Gothic-the historicizing elements disappear,or dral of the Holy Roman Empire thereby recalled still-famous more accurately,are deliberately, self-consciously, and point- large-scale vaulted ancient interiors such as the imperial ther- edly made either to disappear or to lose, by degrees, their mae or the Basilica of Maxentius (true no matter how such historicist presence and embodiment of historicist desire. The buildings were misidentified in the middle ages). Speyer tells meaning-laden historicist language taken from Rome, the us that what we call the Romanesque was never an inexorable classical apparatusof groin and barrelvaults, columns, pilas- 186 ij, ITI

•i iiiiiiiiiii:ii i i i ii ji+!!iiiii iii: i i ii :

.....i 7iiiii!i:

ii ii • '• iii!!•!ii i:; • iiiiii iii•!iiiiiiiiii*iii~iiiiiiiiiiiii:•:•:!•:• • ::iiti•71ii•

i ii

MIL, ...... iii.ii/ii:i:/ii I' :

FIGURE 6. Trier, Basilica (photo: Foto Marburg). ters, trabeation,and load-bearingwalls is now in diverse ways much of its logic) worked essentially as follows. Whereas the and to varying degrees, over time, negated throughexclusion, groin vault was conceived as two interpenetrating barrel suppression, or subversion. In its place a medieval mode of vaults, producing groin lines of difficult-to-manageparabolic architectural modernism takes hold: not the scattered, epi- curvaturesand folds, the rib vault instead began conceptually sodic, fragmentary,often superficial modernist gestures made and in constructionas an integralframework of discrete arches, previously (for example, at Speyer or Cluny III), but a com- each with an independently generated curvature defined not prehensive, vital, new deep-rooted system of structure and by a mere edge but by strong plastic form. The formerly sec- form, which is inherently both modernist and anticlassical in ondary groin line thus becomes the visually primaryrib, with logic and effect, and I argue, in significationand self-conscious the vaulting surface now appearing as the visually secondary motivation. infill, the web, which serves as the ground for the dominant The paradigmaticcenter of this transformativeprocess is figure of the ribs. All that is left of the historical model is the the ruthless, powerful and, I maintain, highly self-conscious idea of masonry vaulting with a four-partcross-diagonal di- and knowing critique of the logic, geometry, technique, and vision (and even that quadripartiteorganization is frequently appearance of the Roman-and Romanesque-groin vault, exploded); in its physical realization all else is displayed as which leads not simply to "progress"but, in the rib vault, to dramatically and conversely transformed. a radical mutation and reversal of virtually all of its traits Perhaps even more boldly revealing of the deeply criti- (results that could not possibly have emerged without a high cal, iconoclastic thrust of the rising movement is the closely degree of techno-historical knowledge and intellectual moti- associated modernist element known as the pointed arch. Al- vation). In formal terms, this critique (deeply Euclidian in though it has been variously interpretedin terms of bent tree 187 branches, heavenward thrust, Islamic precedent, and so forth, I propose that to the contemporary spectator it would have looked most of all not "pointed" (a post-medieval English term) but like a broken arch (i.e., as in the relatively recent French arc brisd). That is, regardless of its key role in resolving geometric and structural problems of the rib vault, the form would have been seen, especially in the early period of its use, as a literal breaking-and reconstitution-of the semi- circular, unbroken arch that through the entire middle ages had been one of the primary historicist elements, a central ...-...... point of reference to antiquity, whose architecturalauthority XN: was now being shattered and replaced by a new modernist system. (Technically, the broken arch was realized by a de- centeringoperation, in which twin centersof curvaturereplaced ...... a single central point.) One might say that the broken arch was ... an indexical of the revolution.13 ...... sign ...... The full architecturalturn to modernism does not happen instantaneously,of course, but in a multi-generationalprocess that sometimes might appear paradoxical in its development (a misleading illusion, we shall find, eventually dispelled by the reading proposed here). Perhaps this was truest of the treatmentof the column, that great bearer of antique allusion which (unlike the immediatelysuppressed trabeation) persisted deep into the period of the turn to modernism (centuries later FIGURE 7. Paris, Notre-Dame, nave (photo: author). sowing controversy and confusion among architecturalhistorians working with the usual methodologies).14 Thus at Suger's Saint-Denis the conceptually point-like supportsystem of the it were, incarceratingthe key historicist architecturalidol ac- linear, skeletal rib vaulting takes the form of a revival, or more cording to its own devices (Fig. 8).17A generation later, for accuratelya reemphasisof classicizing columns (Fig. 17). Such example at Troyes Cathedral or in the nave of Saint-Denis, columns become prominent in the main arcades of "early bundles of such colonnettes dominate the pier completely: Gothic" cathedrals-in most other respects highly modernist they are now virtually all that we see, or are meant to see of buildings-including Laon and especially Paris (Fig. 7).15 the building's supports (Fig. 9). All along, these colonnettes Highly indicative of the progressive takeover of the building are progressively and radically thinned down and stretched by the modernist vision is the mannerin which these conspic- out-as if on a rack-to the point that all connection with an- uous historicist forms subsequently are gradually subverted, tique supportive substance and iconic presence is dissolved, transmuted,and ultimately eliminated (not in linear progress all affirmationof their antique columnar origins negated: part but by a far more complex developmental process). I propose of a process in which the classical element itself is physically that what tended to be displayed and seen in this process was forced (cumulatively imprisoned, broken, stretched) to repre- not so much the column itself (as at Saint-Denis and the main sent the anticlassical (Fig. 10). Similarly, the capital, that cru- arcade at Paris) but its anticlassical, modernisttransformation, cial sign of the classical orders, is abstractedinto the crocket its degree of differencefrom a classically authenticcolumn and type and shrivels to a mere speck in the gigantic elevation (as column-usage(which the evidence of such structuresas Speyer, if burnt to a crisp on a spit), becoming at the higher levels Saint-Denis, Paris, and numerous others suggests was known nothing more than a faint sign of the terminal point of the to architectsand clients of the time). The column, or columnar columnar schema. In many cases the capital eventually disap- schema, I submit, was now retained, indeed made newly to pears altogether, as does, in the still more radical late Gothic, proliferate densely throughoutthe church interior, in order to the attic base and indeed, the entire independent colonnette, make possible the systematic representation of the critical which, having exhausted its possibilities of modernization, progress of modernism through its expository self-distancing finally gives way to a bundle of continuous vertical moldings, from authentic historicism, in readable, dramatic (and often at which point the suppression of historicism is finally com- hyperrationalized)16architectural language. pleted (Figs. 11, 12).18 A pivotal step in this movement occurs in the High Whereas the interior was modernized piecemeal-in a Gothic pilier cantonnd, which literally imprisons the central process that from Reims and the nave of Saint-Denis onward column-earlier displayed freestanding and, by comparison, came increasingly under the domination of tracery,that most in good classical form at Paris and Laon-in a modernist cage proteanof intensely modernistdevices (Fig. 13), which I need of attenuatedcolonnettes, turning classicism against itself, as only mention to conjure its rising omnipresence-the exterior 188 ...-15...... nalized reinforcement system-equally in place at Cluny III, Saint-Etienne in Nevers, the Pantheon, and the Basilica of Maxentius-was now exfoliated, as it were, replaced by the , explosive new mode of openwork reinforcement that I would VXX ?141 term modernist structuralexhibitionism, in opposition to his- XXX.:31? toricist "Roman-esque"structural decorum (Figs. 14, 15, 16). So radically anticlassicist was this new external formatting Ki?j.g that columns were not needed to representthe transformative M...... process (as on the interior), and thus tended to be used only as ...... minor, decorative accents, at most, on flying buttresses. This MIN dramatictransformation of the cathedralexterior, as I read it, N , Al, W'...... was made possible not merely by an internal structurallogic, X6.

1.5s? '01 compelling though that logic may have seemed, but also by the powerful, iconoclastic antihistoricisturge of the medieval modernist movement, the power of modernist consciousness

...... Hi ...... and desire to follow through in practice the radical direction ...... ::]x that rational analysis indicated. y.l ...... xl:. Some further definitions, clarifications, and implications ...... The above account in more de- ...... represents considerably veloped and autocriticalform the core argumentof my earlier ...... article (before it turnedto the spread of medieval modernism ::xxxxx=0...... beyond northernFrance20 and then to a considerationof Italy's ?,.MMm:- ...... place in the medieval scheme of things). The first point I now ...... make the issues raised thesis the ...... ?ri ..... K...... concerning by my regarding ...... "Gothic"-and I stress that the following as well as the pre- ?M. in ...... ME,frIMMI:01* .1...... ,...... vious observations are all work progress, open to revision, ...... X ...... and offered in the knowledge that my reading poses many ...... critical-historical addressed here-is ...... problems incompletely ...... that my approach to medieval architectureis derived from a ...... ,...... K? ...... familiar methodology applied to the architecture (and other ...... x cultural production) of certain non-medieval fields, in partic- ...... ular the Renaissance and modern periods. These periods are generally understood not in terms of style, as in virtually all ON, current readings of the middle ages, but rather in terms of what, for lack of a more concise term, I would call "modalities of cultural-historicalconsciousness." That is, the central factor underlying culturalproduction and reception in the Re- naissance and modern periods is seen as a mode of historical consciousness, the period's sense of the relationship of the FIGURE 8. Reims, cathedral,pilier cantonne(photo: Anne-Marie Sankovitch). self, the institution, the community, or a particulardiscursive formation to the past. The characterof such periods hinges on the dependence of culturalambition and desire on the past or, of the cathedral rapidly, in a swift radical turn, became the conversely, on the sense of independence from the past; in most pure and powerful antihistoricist, modernist aspect of other words, the oppositional pair that we know as historicism the building, in concept as well as detailing. If on the interior and modernism. the classical language of form was agonistically turnedagainst In the historicist mode, cultural production is categori- itself, step by exhilaratingstep, on the exteriorit was exploded cally grounded in and legitimated by reference to historical and scatteredto the winds. The signal event here was, of course, precedent,often though not always classical antiquity.In mod- the dramaticappearance of the flying buttress.The main issue ernism, this historicist groundingis suppressed-some would is not precisely when or where this occurred but the way in say repressed-and culturalproduction is considered,however which the brilliant modernist radicalism in a few decades' self-deceptively, as both self-generating and self-legitimizing, practice destroyed the classical probity of closed volumes in groundedin the present (/future) without recourse to history.21 which buttressing either was entirely contained within the Thus, at least since Burckhardt,the Renaissance has been de- building or closely adhered to it.19This largely hidden, inter- fined principally not as a style but as a movement driven by 189 Mf

lk' ...... '_X' ...... giiii:__V

FIGURE 9. Troyes, cathedral (photo: author). FIGURE 10. Rouen, Saint-Ouen (photo: author).

a turn in historical self-consciousness and desire, by a deep new historicism seen as pervasive throughout cultural production. But it is the nineteenth and twentieth centuries that provide the more relevant parallel to my reading of the medieval. In Labrouste'sBibliothbque Ste-Genevieve, for example, the nineteenthcentury is seen, much like the Romanesque, as a time committed to exploring both historicism and moder- :- nity, a project believed to be compromised in the eyes of the early twentieth century, which turnedtoward a purist modernism, as in Le Corbusier'sVilla Savoie. My second point of commentary will be to address the necessity, evident from observedpatterns of misusage, to refine ...... and reinforce certain distinctions (working if not absolute) in the here, the terminology employed especially regarding aw • -i--:- : • ii ii:ii•!•'•i:'• iiiiiii~ !!!!•• i!ii word "modern/ism."Although one differentiationshould be obvious, it is often not kept in mind, and that is the difference between modernism and Modernism: respectively between a FIGURE 11. Abbeville, Saint-Wulfram, pier base (photo: Anne-Marie transhistoricallypotential mode of consciousness and experi- Sankovitch).

190 INI Xw: ...... x~: ......

...... k "X0 , ...... r...... -...... i 1 , Xxx.:?? P.--

FIGURE 12. Gisors, Saint-Gervais, nave aisle pier (photo: Anne-Marie FIGURE 13. Rouen, Saint-Ouen, nave triforium (photo: author). Sankovitch).

ence and its specific historical realization, i.e., the Modernist Renaissance, when one use of the term modernwas not to des- movement of the twentieth century (or, indeed, the wider ignate the "Gothic"but ratherthe "current"classical revival.25 "Modernist" project for civilization formulated by the En- Although in practice both medieval and twentieth-century lightenment).22Medieval Modernism would share with recent versions of modernism vitally incorporatedsuch "newness," Modernism a common antihistoricistgrounding in the "pres- this importantmeaning is not at the core of my redefinitionof ent," an emphasis on the critical power of reason over pre- the medieval passage known as "Gothic."My usage focuses cedent and authority, and ultimately to some degree an instead on the evident, underlyingshift in historical conscious- empowering of individual subjectivity, but-and this is a cru- ness, grounding,and desire ratherthan on the associatedformal cial distinction-not necessarily any particular formal fea- novelty of the cultural products of the shift, that is, their tures of its cultural production.23Unlike recent Modernism, specific formal difference from immediately preceding works. which has been so transparentand omnipresent in culture, Furthermore,I would point out analogously that histori- medieval Modernism experienced a much more problematic, cism is to be distinguishedfrom what might be called tradition- often submerged or disguised existence as a cultural project, alism, with which it is sometimes conflated. By traditionalism especially in terms of the textual record.24 I mean the sheer continuationof practice.26Its opposite is not But modernism also entails a third meaning, for "mod- modernism,but simply innovation, or modern-as-newness(on ern" may and commonly does signify, of course, simply what whatever terms of consciousness). is (or seems) new, recent, current, contemporary, or up-to- With the above distinctions in mind, it is possible to dif- date. In this usage historicism can also be "modern,"as in the ferentiate my approachfrom others that it might misleadingly

191 seem to resemble, for example that of Jean Bony (the bril- terms, and to explain that complexity unburdenedby the de- liant Focillonesque "Gothicist" who would appear to come mands of style, yet within a coherent view of the general for- closest). When he writes, "The art we call Gothic was the mal characterof these great works. The paradigmpermits the assertion of a spirit of modernity which went on renewing unproblematic interpretationof periods of instability and of itself for centuries, almost ceaselessly," what he intends by tension between historicist and modernist desire, like the "modernity"is essentially "a critical dissatisfaction with the eleventh and early twelfth centuries, in much the same way immediate past" allied with a ceaseless "power of inven- that it allows for the complexity of the period of the turn to tion."27 He means, in other words, modernism as sheer inno- medieval Modernism-a.k.a. the Gothic-in which the rising vation and radical refusal of tradition. But when he moves tide of modernist desire is rarely unaccompanied by a strain from such introductoryremarks to historical analysis, it soon of historicist consciousness, strongly manifest in certain spe- becomes clear that the architectureis being seen specifically cific situations and sites. In other words, the paradigmenables through the eyes of twentieth-century Modernism, with an us to understandthose "problematic"classicizing columns in extreme emphasis on abstract, formalist criteria including Saint-Denis or Notre-Dame in Paris (whose presence has so space, surface, luminosity, and grids.28If neither of Bony's befuddled scholars from the nineteenth century to the present) M/modernisms thus resembles what I have outlined, they do in the same terms as the predominant modernism of these understandablyrecall his great nineteenth-centurypredeces- buildings, without compromising or subordinatingeither as- sor, Viollet-le-Duc, who similarly emphasizes a spirit of inno- pect in the manner unavoidable within the old paradigm of vation, although in primarily structuralrather than formalist Romanesque-vs-Gothic,according to which by definition such terms.29 Of course these two preeminent interpretersof the classical elements should never appear in the latter style.31It Gothic practiced at very different historical moments and allows us to understandthe nave of San Lorenzo in Rome, the with very different agendas: whereas Viollet-le-Duc was a Florence Baptistery,Autun and Speyer Cathedralson compat- scholarly protoModernist who posited an ideal historical ible, meaningfulterms of identity and difference(which other- Gothic structure as the inspirational model for a modernist wise has provenimpossible), indeed to understandthe Florence contemporary practice, Bony was an academic scholar prac- Baptistery and Bourges Cathedral on those same congruous ticing within the interpretive paradigm of Modernist formal- terms. It lets us see Saint-Etienne in Caen as unproblemati- ism, writing primarily about individual medieval buildings cally including an eleventh-century quasi-modernist interior almost as if they were Modern ones according to formal elevation, sparingus the agony of deciding whether the build- criteria. Although I would not assert that my post-Modernist ing is proto-Gothicor a Romanesquebuilding with some proto- perspective is necessarily postmodernist, I am aware that my Gothic features. Similarly, it permits us to read the modernist interpretation has emerged at the end of the twentieth cen- vaulting of Sant'Ambrogioin Milan and its Lombardcontem- tury in the wake of the historicizing of the Modernist move- poraries for what it is, freeing us from the old nationalistic ment, and that it involves a critique virtually unthinkable for debate between francophile and italianist scholars about its Viollet-le-Duc and Bony who wrote from within the perspec- priority or nonpriorityto French "Gothic." tive of that living movement. Ultimately, I feel, the paradigm may even allow us to Turningfrom such problematics of definition and mean- develop a readingof medieval architecturenot as stratifiedhor- ing, a third point that I would make here concerns certain advantages of the paradigmof cultural-historicalconsciousness over a style-centered methodology. The paradigm spares us the many well known difficulties associated with "style."30 Its modalities of modernism and historicism can occur in infinite pure and impure states of existence and coexistence. These modalities are about conscious desire, and consciousness, as we know, is inherently accommodating,expansive, and multilayered ratherthan intolerant,reductive, or monadic (as "style" II in insistence on coherent sets of a tends to be its globalizing ..i.. r:i• i!i • limited number of formal traits in individual works, artists, ;;, and periods). The paradigm allows for irrationality and ex- ...... ception, for the messy complexity and contradiction of life and of art. Thereby the cultural-historicalconsciousness paradigm permits and even fosters the explication of conflicted, multilayered works; it encourages us to see works in non- 7r, reductive, anti-hegemonicterms, to see the cathedralsformally as the complex and even self-contradictoryentities that much contemporaryresearch is proving them to have been in social FIGURE 14. Rome, Pantheon (photo: author).

192 Aft,

j A&t

FIGURE 15. Nevers, Saint-Etienne, choir (photo: author).

, ifi

"CPs8?~*P~~?k~::I -P C ~-- TIMM_

FIGURE 16. Le Mans, cathedral, choir buttresses (photo: author).

193 izontally into "Romanesque"and "Gothic"layers at all (which ing spirituality), that of conceptual incompatibility;whereas inevitably solicits stylistic categorization), but vertically, as a the Renaissance fits, or seemingly can be made to fit current continuum of culturalproduction organized arounddominant architecturalconcerns, the medieval, especially the Gothic, is twin strands of historicist and modernist discourse, distinct generally understood according an assemblage of terms- yet often entangled in ever-varying relationships. To pursue masonry vaulting techniques and geometry, profile diagrams, such a dialectical genetic logic to its conclusion, it implies a diaphaneityand linearity,scholasticism, etc.-that are difficult, certainfundamental reorganization not only of the "medieval" obscure, and understandablyirrelevant to the pressing con- (an archaic, now rather paradoxical term that would neces- cerns of most young architects and their mentors in the late sarily disappear)32but conceivably of architectural history twentieth century. Here I would simply point out that these as a whole (at least within the European-Westernsphere): a concerns often involve questions of modernism, historicism, 90' reorientationof history from horizontal stratificationto a and their conflicted and unresolvedrelationships, which raises consistently vertical dialectical structure,producing a trans- an intriguing possibility: that a reformulated understanding parency and compatibilitybetween various architecturaltime of the medieval in terms of the paradigmsof historicism and zones-those called medieval, Renaissance, Baroque, and modernism would possibly open up the great epoch to the ac- Modernm-which currently are regarded as incompatible in tive interest of today's architects, who might have as much to their terms of interpretation.33 gain from an understandingof Saint-Denis and Bourges (and Moreover, such transparencywould result not only dia- theirmasters) as from Palladioand the Villa Rotunda.And con- chronicallybut potentially synchronicallyas well. As has been versely, the medieval architecturalfield of scholarship would recently demonstrated,the social tends to become involved in surely stand to benefit from a renewed interest in it within the style-based narrativesmainly to explain away aberrationsand field of architecturalpractice. inconsistencies in stylistic patternsand development, and this These are some of the issues and speculations that I find precludesmeaningful contextual interpretation.34Just as style- emergingfrom my thesis, tentativeindications of certaindirec- based criticism is inherently asocial in its abstraction,so the tions of theoretical and practical thinking about medieval ar- paradigmof cultural-historicalconsciousness by definition is chitecture in which I am currentlyengaged, and which I hope engaged in the social field, for what are consciousness and to realize eventually in an expansive form that will more fully desire if not those specifically of individuals, groups, institu- engage those many critical and historical problems raised by tions, and communities?That is, althoughhistoricism and mod- my reading. These would include the core idea, not easily ac- ernism are transhistoricalconcepts, they nevertheless compel cepted, of attributingaspects of what is regarded as a mod- close attentionto the historical specificity of the period under ernist mode of consciousness to the "pre-Modern"period; the study. The cultural products of cultural-historicalconscious- problem of where to situate such a modernist consciousness ness are inevitably representationsof that consciousness; as in that period, whether in the larger community, the work- representationsthey bridge the gap between artistic form and shop, the individual architect, etc.; how to reconcile consum- desire. This indicates that the center of the paradigm I am mated modernist desire in architecture with the entrenched suggesting for medieval architecturemight be displaced from and resistant authoritarianismand historicism of the period, the matrix of consciousness toward the field of spatio-visual or with scholasticism, problematic in its modernist strands. representation, which would shift the methodology toward One thing already can be suggested here. The terms of the alignment with many specific forms of currentresearch in the approachI am outlining will seem "anachronistic"to certain period-and in art history in general. readers only because the traditionalmethodologies to which It is not only academic research that stands in potential they subscribehave so naturalizedtheir own terms of analysis benefitof this interpretativeparadigm, but also the architectural that they are falsely and misleadingly seen as resident in the profession, its pedagogy, and its sense of historical ground- historical period itself. We can of course never meaningfully ing. Currentlywhen premodernarchitectural history is at all understandthe past strictly on its discursive terms, but only a viable presence in schools of architecture,it tends to be fo- on our own: this is what is meant by the iron law that every cused on the Renaissance. This highly valorized period pro- generationmust rewritehistory. The approachoutlined in these vides codified theory and rules to study, esoteric illustrated pages offers, as I have indicated, the possibility of closely de- texts (like the Hypnerotomachia Poliphili) to pretend to under- scribing and accounting for the complexity of architectural stand, splendidornamental forms to imitate in the Orders,and, phenomena in the "medieval" period of prevailing moder- perhapsmost of all, vivid, heroic humanmodels of legendary nity (as well as in the previous epoch) in a comprehensive, status to reinforce the Modernist hypervaluationof the archi- historically grounded, descriptively positive way that does tect as autonomous genius. That the medieval has so little not require,as do all currentinterpretative models, the neglect, presence in the currentprogram of architecturalstudy may be suppression or rationalized distortion of importantaspects of due largely to its lack of these Renaissance attractions(it has, this extraordinaryarchitecture, which was surely producedby alas, essentially only its buildings for the most part).But there one of the most powerful and sustained currentsof desire and may be anotherimportant obstacle (in additionto a contaminat- intellectual drive of all time.

194 Suger's miracles sarily thoroughly understood all of its technical and formal refinements. But the evidence certainly indicates that Suger Because the above discussion has mostly been so theo- was acutely aware of and in a deep way responsible for, and retical and generalized, it is useful now to apply my approach more importantly,that he felt responsible for its specific ar- in greater detail to two buildings, testing theory, as it were, chitectural character.37 through the practical analysis of specific medieval texts and Most importantly,I would argue, he felt a need to defend images. The first building is Saint-Denis-the new westwork its radical modernity.Although architecturalhistorians often (1130/35-1140) and choir (1141-1144)-and more particu- tend to emphasize evolutionist continuity,the derivationof the larly, the still perplexing problem of how they are treated in project's modernist design from post-Romanesque develop- Suger's celebrated texts (which all postdate the construction). ments of the early twelfth century, these developments, at I already have referredto the complexity of architecturaldis- least in the Ile-de-France, were manifested mostly in modest course in the choir, that is, the incorporation of historicist churchesor partsof such churches,such as the choir of Morien- monolithic columns in a radically modernist design, a multi- val and side aisles of Saint-Etienne in Beauvais. Despite the layering of modalities that would also include the historicist weight of these fabrics on the pages of our art histories, for crypt. As already noted, however, their coexistence is no Suger and his contemporariesthey were but minor features of longer problematic,as there is no necessity to explain away or a vast architecturallandscape of hundredsof non-modernist, rationalize the crypt or the columns in order to make the or slightly modernistchurches. Nor, obviously, did these men building "Gothic." Rather, each element becomes a distinct know of modernism's great future toward which the new voice in a complex discourse, with modernism, to be sure, church was such a decisive step, and which would vindicate playing the dominant, ascendant role, at least in the view of Suger's patronage of the new. In the eyes of contemporaries our long historical perspective. But rather than directly pur- the modernity of the great Saint-Denis project can only have suing the representationaldimensions of these voices, I will represented not continuity with the past (or possible future) offer some commentary on Suger and his writings. but sheer discontinuity; it was a definitive break, a radical These complex and singular texts are problematic not swerve that in itself-quite apartfrom its replacementof hal- only in their content but by their very existence. So excep- lowed ancient fabric-needed legitimation at a moment when tional are they in a period with so little writing about architec- architecturalmodernism was just on the cusp of ascendancy ture that scholarship has long puzzled over Suger's agenda.35 and had not yet widely acquiredits own sustainingforce of self- Generally and quite fairly, the texts are regardedas an apolo- legitimation. (As Panofsky insightfully put it, "It was as if a gia, Suger's defense against real and potential criticisms of Presidentof the United Stateswere to have hadthe White House the spectacular project. On the one hand, it is argued, there rebuilt by FrankLloyd Wright"-before 1920, I would add.)38 was the impious destructionof the sacred ancient fabric of the If this is so, then by what means did Suger seek to legit- abbey to contend with, and on the other what might have been imize the modernismof his enterprise(which he actually calls regarded as the architecturaland artistic extravagance of the not merely opus novum but modernum)?39 The most radically new work as a monastic structure,especially in the context of modernist part of the building was, of course, not the west- the ascetic current of monastic thought led in the period by work but the choir, and it was on this fabric that Suger-know- Bernardof Clairvaux, with whom Suger was in close contact. ingly, it would seem-concentrated his literaryefforts regarding Although internal and external evidence suggests that Suger modernism. Scattered through his text are statements that, was probably sensitive to both issues, they do not adequately when read in context and, as they used to say, between the account for his textual enterprise. Many importantmedieval lines, constitute three lines of defense.40The first of these is projects involved the demolition of highly valued ancient the most familiar to us, namely the functionalism of the choir church fabric, and many monastic projects, especially in the in terms of the circulatoryand visual advantages of its new Benedictineorder, were and continuedto be lavish, but none, so luminous openness to the movement and vision of pilgrims far as we know, provoked anythingfaintly resembling Suger's visiting the great display of relics in the chapels (Fig. 17).41 multi-volume apologia for dismantling what probably was Suger's second defense is what we might call aesthetic, indeed a rather decrepit Carolingian church and replacing it and it is produced most acutely by the famous line regarding with a new structureadequate to its currentrole as a major "the circularstring of chapels" that cause the church"to shine pilgrimage site and royal foundationof great political impor- with the wonderful and uninterruptedlight of most sacred tance. Thus, in additionto the issues of demolitionand decorum windows, pervading the interior beauty" (Fig. 18).42 Even if anotherexplanatory factor is necessary,and I suggest thatit was Suger is not referring here specifically to Pseudo-Dionysian the dominant,radical, problematicmodernity of the project. light metaphysics, he surely is emphasizing the positive new It has been argued that Suger may not have invested his aesthetic effects that only the open modernist design could project with the Neoplatonic connotations that Panofsky so have produced,the wonderful luminosity it provides the inte- ardently believed were there, and certainly he was not in any rior, heightening its beauty, hence its worthiness to be the way the architect of the building.36Nor need he have neces- church of such a prestigious foundation.43

195 But it is the third line of defense that Suger may have regarded as the most potent. It involved authenticationof modernity not by functionalism or aesthetics, but by the approval of God (and the saints), the ultimate source of authority and legitimacy. I refer to one of the miracles that Suger relates in De consecratione concerning divine assistance to the fabric, namely the one in which God himself protects the incomplete choir from a violent storm. In reading this passage, please note that the story centers on the key aspect of the structural apparatus of architectural modernism, the independent ribs (Fig. 19)-which Suger of course accurately calls arches- IN standing free of centering awaiting the construction of the webs, and note how often they are mentioned (in passages that I have italicized).44He writes,

when the work on the new addition with its capitals and aw:. upper arches was being carried forward to the peak of its height, but the main arches-vaulted independently-were not yet held together by the bulk of the webs, there sud- ...... :_: ---: --- arose a terrible and almost unbearable storm with ...... :::::::::::::: ::: ...... denly A :::::::::::: --- ?

"PIF: -- - an obfuscation of clouds, an inundationof rain, and a most O'Nlii, violent rush of wind. So mighty did this [storm] become

...... ii that it blew down, not only well-built houses but even 'HW xv: ag stone towers and wooden bulwarks. At this time ... when ...... ii::ii-:::l i;l i the venerable Bishop of Chartres,Geoffroy, was solemnly ...... celebrating at the main altar a conventual Mass ... such a force of contrary gales hurled itself against the aforesaid arches, not supported by any scaffolding nor resting on any props, that they threatened baneful ruin at any moment, miserably trembling and, as it were, swaying hither FIGURE 17. Saint-Denis, author). and thither. The Bishop, alarmed by the strong vibration ambulatory (photo: of these [arches] and the roofing, frequently extended his blessing hand in the direction of that part [and made other gestures] ... so that he escaped disaster, manifestly not God's rescue of the modernist choir vaults is the most through his own strength of mind but by the grace of God vivid of Suger's miracle stories, which are so numerous (at and the merit of the Saints. Thus [the tempest], while it least seven)46 as to give the impression, together with other broughtcalamitous ruin in many places to buildings thought passing remarks,that the entire new fabric, both westwork and to be firm, was unable to damage these isolated and newly choir, advanced only under divine inspiration, guidance, gen- made arches, tottering in midair, because it was repulsed erosity, intervention, protection, or approval. It is almost as if by the power of God.45 God himself were the builder, and Suger merely his instru- ment (an inference surely intended:just as miracles made men Three times in this extraordinarypassage Suger calls our at- saints, so Suger's enterprise might be sanctified).47This im- tention to the "isolated,""unsupported" ribs, threatenedby the pression is justified, for divine assistance to the new fabric is storm. God himself protects these "newly made" ribs, surely not merely randomly and spontaneously recollected, as one the key componential novelty of Saint-Denis's modernity in might first think when reading the dense, colorful, agitated the eyes of a non-architect, from the destruction that older text of De consecratione, but is effectively depicted as com- buildings suffer. If we allow that the ribs function synec- prehensive. Suger's string of miracles, far from being casually dochically for the entire modernist structure,Suger is saying thrown together, is highly organized narratologicallyand con- that that structureis sanctioned as worthy, granted legitimacy ceptually. The miracles appearin the actual order of construc- by God himself, with the saints thrown in for good measure. tion, beginning with the discovery of a quarryof exceptional In this manner Suger would appear to reveal that at a certain stone and the arrivalof accomplished workmen,48 continuing level and point of reception (if not also production) the self- with miracles concerning supports, roof, and-here shifting grounding of modernity as yet provided inadequate legitima- tellingly from westwork to choir-vaulting,49 and ending with tion, and some supplementaryretrofitting of external authority the provision of sheep for the final consecration feast.50 More- was still regarded as necessary. over, the miracles comprise two categories: one (all-purpose 196 I ~iiiii~r~PC~ ^iiii'i :ct~?? r iiiiiii Z "'''-- ,?e X ?J ~ r I Sr - ;t:-? ~ ~x1?-~ : ?-?a~i~?:????~-n:-~i~ il~plg~p~gy~g~i:?:?:~~:-r tB % -~?:i;;---? ??JI/ ? .. It :ii: j .h. ;:I iii cc~1 ii iiiii' ?~-i ~C~11

:i::?iiii:a?l'i?ii????:?'?l~~~l?j~i?.~,~~'~"?~-~ ~i;;?i???a~~;Z::~?lIr-~:::::~~l::~~

i ii-iii-iii-?i?~ :I-:--_ ~g~p~l : __i_ii iii iiiii ~lllllllll1lli1iii;;?;Bi:-:-::-i:iiiiit iiiiiiiiiii -:-iii?: -::-:ii ;I

FIGURE 18. Saint-Denis, panoramic view of ambulatory from hemicycle center, 1970 (photo: author).

......

...... ?: . "'ii'-- ::: ?--:: ......

...... ixx ...... v e ...... i ~, l:-:--i_:, -~ iiii i_ ii ...iii-i- M I ...... :?X. -OR ..mii ...... ?ii-i

...... ioid.... i~", xM A: -iiiixri,

FIGURE 19. Saint-Denis, ambulatory vaulting (photo: author).

miracles, as it were) affects the fabric as a whole (stone, work- fabric. It is this latter group to which I would draw our atten- men, sheep), while the other concerns three specific major tion. What I propose is that the vault miracle, alreadydiscussed building components, i.e., columns, rib vaults, and roofing, as a defense of modernism,should be seen in conjunctionwith which I suggest serve synecdochically to representthe entire the other miracles of this "componential"group, as effectively 197 forming a coordinated, cross-referentialdefense of the entire acle itself-would have been the same. Apart from stressing project,reflexively strengtheningmy reading of the vault mir- again that nothing was too good for his project and the acle as a self-conscious tactic to defend modernism.51 difficulties of procuring such materials, the story serves to Suger dedicates two miracles to the historicist,monolithic emphasize the historicism of the classical simulacramanufac- columns (Fig. 17). Although these two miracles occurreddur- tured from local materialsprovided by God. It informs us that ing the construction of the westwork-when columns were Suger regardedthe columns as classicizing-certainly ex post needed for the extension of the old columnarnave connecting facto and probablyalso at their manufacture-and, moreover, it with the new facade structure52-they would have provided that while he imagined that others would probablybe in agree- similar, even more prominentcolumns also for the new choir ment, he wanted to make sure. Generic divine blessing of the a few years later. Given this fact and the way the text seems new columns, and the divine procurationof their materialev- generally to conjure out of fragmentarymiraculous episodes idently were not enough for Suger. Seeking redundantlayers a comprehensive umbrella of divine assistance for the entire of legitimacy, he seems to have wanted to certify that the new fabric, I suggest that it would be difficult not to interpretthe monolithic shafts of his fabric, in the westwork connection two column miracles synecdochically, as applying to all parts and including the supports of his modernist choir,58 were of the building where monoliths were used, i.e., including the explicitly, unavoidably seen-like those of the Carolingian new choir.53 nave, which he evidently regarded as virtually antique59-as The second of these miracles is unproblematic:a group classically historicist, historicism being in itself a powerful of only seventeen workers manages, with God's help, to haul legitimizing agent. one of the columns up from the bottom of the quarry,a feat To go a step further,what Suger appearsto be saying with normally requiring "a hundredforty or at least one hundred his vault and column miracles, I propose, is that his project men."54 comprises both modernism and an important,counterbalanc- The first column miracle, which concerns the discovery ing form of historicism, by means of elements that are sanc- of the quarryitself (evidently a differentquarry from the one tioned by God not only generically but intrinsically and providing ordinary stone), is the more intriguing and telling. specifically as historicism. Thus the set of historicist choir Suger's story is that, unable to find a local source for mono- monoliths (which he evidently had polychromed, and some lithic columns, he had been hoping, in desperation,to procure capped with stronglyclassicizing capitals)60would have served them as spoils from Rome, perhapsfrom the "Palace"(Baths) as a doubly legitimized and metonymically legitimizing agent of Diocletian, when through "Divine mercy" a local quarry of the radical modernist apparatusthat it literally and meta- sufficientto his needs (having "reserved"its best materialsfor phorically supports. Saint-Denis) was unexpectedlyidentified at nearbyPontoise.55 Although the final miracle of this group does not overtly The curious thing about the passage is not the (almost participate in the historicist-modernist dialectic of Suger's anticipated)miracle of the quarrydiscovery, but Suger's rev- churchinterior, it nevertheless may have been directedtoward elation that he seriously considered obtaining the monoliths the problematicsof legitimacy as I have construedthem. Here, from Rome, going so far as to describe the arduous sea-and- in an associative parallelto the miracleof the quarrydiscovery, river routethat they would take. It is hardto know exactly what God guides Suger througha supposedly depleted forest to the to make of this story.56Some readersmight find it difficult to giant trees needed for the great tie-beams of the church roof. imagine that the great administratorSuger, who took such Although the reference here appears to be the roofing of the pride in the rapidconstruction of his project,believed even for central narthexarea, by extension-as with the columns-the a moment and under duress (as he claims) that he ever could miracle would also apply indirectly to the high choir roof, or would procurecolumns all the way from Rome (as much as which requiredbeams of the same exceptionally large dimen- he might have liked to), quite apart from the inappropriate sions: in other words, the miracle would appear to have per- scale of Diocletian's shafts for his building or, for that matter, tained to the new roofing as a whole.61 their likely unavailability, points which might be considered In modern scholarshipnearly all interest in the materials giveaways of a certainlack of authorialsincerity. On the other and facture of medieval church architectureis focused on the hand, for all its structuringof the miracle narrationand other masons and their "secrets,"the roof remainingvirtually invis- details, Suger's text is full of the unexpected, as might be an- ible (except to dendrochronologistsand roof specialists); but ticipated from the pen of such a complex, dynamic individual, the medieval observer--especially the patron--knew better, who was writing somethingclose to a medieval romance,epic, that carpenters and timber were a major and indispensable or indeed hagiography (considering the many miracles). It is part of the enterprise,represented by the elaboratescaffolding thus entirely possible, if not in fact more likely that the story apparatus,lifting machinery, massive and precise centering, was not made up (Suger did incorporate"spoils" from the old huge doors and other furnishings, and, of great final impor- nave in the new choir, in the form of six slender marble col- tance, the roof, comparable in the effort and expense it re- umns on the outer piers).57What I would emphasize is that in quired to the masonry of columns and vaults.62Similarly, the either case, Suger's purpose in telling the Rome/Diocletian obvious point needs to be stressed that the functional impor- story--which is extraneous and quite unnecessary to the mir- tance of the roof was hardly nominal. Thus, in providing es- 198 the crypt below to the summit of the vaults above, elabo- rated with the variety of so many arches and columns, including even the consummation of the roof.63 ...... The Bourges problem ...... In conclusion, I turn briefly to Bourges, specifically to what might be called the "Bourges problem":the question as to why the spectacular building had a relatively meager following, especially compared with Chartres,its exact contem- ?Wf. porary (Fig. 20). In other words, in terms of reception and 'W4 imitation, why was Bourges, which virtually all critics agree is a stunningmasterwork of the highest rankin originality and formal power, relatively such a "failure"?64Why was, in Bran- ner's words, in contrast to Chartresthe "model,"Bourges the "outsider"?65The answer commonly given is that essentially it "Yn embodied too ambitious and singular a scheme to be easily imitated or adapted;it was too visionary, "a whole that had to ...... be imitated in its totality or not at all" according to Branner, with Bony concurringthat it was "a ratherexacting prototype to follow."66True enough, to comprehensively and faithfully reproduceits five-aisled, transeptlessplan, staggeredcross section, and self-repeating elevation, would certainly have been an "exacting" brief, for the combination, which produced a "church-within-a-church,"was viable only at a scale approx- imating that of the prototypewhile requiring,for the complete effect, that the patronforgo a transeptand accept, at best, a severely restricted,nominal programof chapels. These practical problems were probably accompanied by importantsymbolic issues: the lack of a transeptmeant the absence of cruciform spatial iconography; and together with the continuous five- aisled plan, the missing transept produced an absence of architectural distinction and between choir and nave FIGURE 20. Bourges, cathedral, choir area (photo: author). hierarchy that was achieved in traditional planning and brought to full realization in the Chartrianscheme (with its three-aisled nave, transept,and five-aisled choir). The few followers of Bourges sential materials for the roof, God was not merely giving tend to confirm this reading by "normalizing"its scheme to generic sanction to the rebuilt church by providing one more varying degrees; at Le Mans, for example, the new Bourges- random building element. Rather he miraculously provided type choir includes a full ring of chapels as well as an adjacent that very element which protected the entire new enterprise transept,with the preexisting three-aislednave remainingsub- beneathit, modernistand historicistcomponents alike (whether ordinate to the new five-aisled eastern structure. in the narthex or choir areas); and this, I tentatively propose, Yet Le Mans (like the other followers) also tells us some- in the context of the seeming cross-referentialityof the other thing else: Bourges was rather more flexible and open to ad- miracles in question and the synoptic, synecdochic structure aptation than is commonly held; its intractability, in other of the text, is what Suger may have meant in telling the words, has been exaggerated.67That the "problem"of Bourges miracle of the forest. That he firmly understoodthe intercon- must have involved more than simply an inflexible scheme nectivity of all three components (columns, vault, roof) in mi- may also be inferred from the reception of its highly original raculous terms that applied to the choir is indicated by the details-or rather, the absence of active reception, for virtu- miracle of rapid construction which he had earlier related in ally none of these details, including the singular piers and De administratione: flying buttresses, which might have been employed readily in non-Bourges schemes, were taken up outside the Bourges How much the Hand Divine which operates in such matters group. The Bourges "problem,"in other words, runs deeper has protected this glorious work is also surely proven by than typology. the fact that It allowed that the whole magnificentbuilding I propose that the real "problem"with Bourges, which [would be completed] in three years and three months, from informed virtuallyall of its traitsand was centralto its meager 199 ...... -

K., ...... xxxxxx??,;...... EX.:. X.,.1 On. a...... II X. x:

...V...... WIN, a xH ?N. m R Ixx.T, J41 77 93 X: gow U.-t. .-Off

SAD %MXvl?: -X ...... iiH? X- ...... X.-g:: ...... -

_4 e...... 01 Al.lev.a -Mslawell mor xg X.14 M...... 3: xi::::::M 'T - :- :...... :mw. i--A El. x"V ...... A. _ ...... ww.ii so :~ . ?l M.. X: W N. A)".W. /Z ......

WK.1 V...... l k:

FIGURE 21. Bourges, cathedral, view including main and aisle piers FIGURE 22. Bourges, cathedral, respond system, detail (photo: author). (photo: author). reception, resided in its exceptional degree of modernity. of its primaryelements of pier, vaulting, and buttressingwith Bourges was "visionary"not merely in some generalized sense those of Chartres,whose receptive success was of course as but in a profoundly modernist way. The building, in fact, is powerful as that of Bourges was wanting. The pilier cantonnd to be recognized as unquestionably by far the most uncom- of Chartresor its immediatefollower, Reims, as alreadynoted, promisingly radical and comprehensively advanced modernist placed within a cage of attenuated colonnettes a full central design of its time-indeed, in virtually every respect so far column, amply classical in its proportions, including its cap- ahead of its time that it might be fairly called futurist.68 ital size (Fig. 8). At Bourges, despite a certain schematic sim- The modernismof Bourges is so thoroughgoingthat a full ilarity, no such agonistic interaction of column/colonnette account of it would involve almost a complete description of occurs, for the central pier element is not the ratheraccurately the building. A few essential observations, however, should classicizing column of Chartres but such a column already make the point. Its combined five-aisled, transeptless plan, transformed,declassicized, modernized,converted into a huge staggered cross-section, and self-repeating elevation produces pier-like cylindrical mass with only a vestigial capital. More- a synthesis generating the spectacular illusion of a church- over, taking exaggeratedlyaberrant proportions, in an accordi- within-a-church, a sensational effect without historicist refer- onlike movement these supports alternatebetween extremely ence or grounding (despite partial models at Cluny, Sens, squat and highly elongated forms, respectively, in the aisles Paris, etc.). The radically modernist, antihistoricist(as well as and main arcade (Fig. 21). On these gargantuan masses, non-traditional)character of the physical forms in which this which slitherup past their capitals as an undulationin the upper visionary scheme is realized becomes clear in a comparison wall, are placed-with an almost vine-like effect-a widely

200 ~ix

:?iS

io.. ..?: ......

r mg ~~saasal ~P, sB

......

mada

lx:i p: i'i i i:~~'i :-::.xi ze: ?1- P: All-- ,,-dew ei

FIGURE 23. Bourges, cathedral, ambulatory vaulting (photo: author).

201 Denis, it was possible to build a work of virtually pure modernity. Yet in a very real sense the strain of resistance to absolute modernism had not vanished but merely moved to a ...... L .. ..? ii!!iiii!?i• higher level, as it were. Whereas at Saint-Denis (as still at

V: Chartres)patron and architect had crafted a multivoiced work a~ , OXX that accommodated competing currents of cultural-historical Ye:I-- consciousness and desire, their counterpartsat Bourges evidently were single-mindedmen possessed by an uncompromis- ing modernist vision of architecture,who chose to disregard any such voices of "resistance."Bourges thereby became the first building in which all was sacrificedon the altarof modernity regardless of consequences.71

NOTES FIGURE 24. Bourges, cathedral, choir buttressing (photo: author). 1. "Gothic/Italian'Gothic': Toward a Redefinition,"JSAH, L (1991), 22-37. 2. L. Grodecki,Gothic Architecture, trans. I. M. Paris(New York, 1977), 24. spaced set of hyperattenuatedresponds, which at arcade level 3. E.g., H. Kraus, Gold Was The Mortar: the Economics of Cathedral ratherthan outwardin the usual are filled- Building (Boston, 1979); B. Abou-el-Haj, "The Urban Setting for Late multiplying pattern Medieval Church Reims and its Cathedralbetween 1210 and in with shafts at Building: inwardly supplementary (Fig. 22). Moreover, 1240,"AH, XI (1988), 17-41; A. Erlande-Brandenburg,The Cathedral, Bourges recognizably classical columns are denied even to the the Social and ArchitecturalDynamics of Construction,trans. M. Thorn hemicycle (where they are regularly found, as at Chartres), (Cambridge,1994); many of the articles in Artistic Integrationin Gothic aroundwhich the complex anticlassicalpiers of nave and choir Buildings, ed. V. Chieffo Raguin, K. Brush, and P Draper (Toronto, march unaltered. 1995). See also the comments of W. Sauerlinder on this displacement, in "'Premiere architecture or 'Renaissance of the twelfth to if we look first to Chartreswe find gothique' Turning vaulting, century'?Changing perspectives in the evaluationof architecturalhistory," the pointed arch pervasively realized in firm, strong profiles Sewanee Medieval Colloquium Occasional Papers, II (1985), 25-29. unwaveringly straight in plan. At Bourges, not only are the The shift in interest is starkly evident in comparing P. Frankl'sPelican ribs now extremely thin (more so even than at Paris), approach- textbook, Gothic Architecture (Harmondsworth,1962) with D. Kimpel and R. Suckale, Die Architektur in Frankreich 1130-1270 ing true linearity, but in the ambulatorybays they sometimes gotische (Munich, 1985). take a double curvature:the arch, in other words, here is not 4. P Frankl, The Gothic Wilson's just broken but twisted three-dimensionallyinto sinuous lines (Princeton, 1960). Christopher insightful recent survey, The Gothic Cathedral,Architecture of the Great Church, (Fig. 23). 1130-1530 (London, 1990), despite its title dismisses or avoids the old It is in its buttressing system, particularly in the choir question, shifting the center of discussion to the issue of the "Great flyers, that Bourges attains perhaps the extreme point of its Church" typology. Nor is there in Kimpel's and Suckale's masterful radical modernism (Fig. 24 and front cover). In sharpcontrast survey of French Gothic (Die gotische Architektur) any sustained en- defined as a to the nave of Chartres,where of gagement with the issue (Gothic being implicitly new, buttressing pier-towers Egyp- or architectural without review of the toid sustain massive linked post-classical style language, any weight double, full-quadrantflyers problematics of definition). by a wheel-like arcade of semicirculararches, at Bourges tall, 5. E.g., Filarete (cf. Frankl, The Gothic, 256-257, 858-859). This should emaciated, deep piers launch lean, stripped-down, extremely not be confused with the later Renaissance use of the term "modern" steep flyers that often take only segmental curves (at the upper for the Renaissance itself (for example, by the Pseudo-Raphael and levels) and thereby give the impression of almost straight Vasari). On the phenomenon of Renaissance period-naming, see E. H. strutsrather than archedunits. Although the futuristicBourges Gombrich, "Norm and Form: The Stylistic Categories of Art History and their in Renaissance Ideals," in Norm and Form choir flyers were structurallythe most efficient ever built in Origins (London, 1966), 81-88. the period, like the building's other details they were never im- and were somewhat "normalized"in the nave.69 6. During the Gothic period itself, so far as we know, architecture was itated, indeed, as The was In other the futurism of came under the crit- qualified only geographically, "opus francigenum." phrase words, Bourges used by Burchardvon Hall around 1280 in reference to the Germanab- ical fire even of its own workshop:part of a patternthat should bey church of St. Peter at Wimpfen im Tal (Frankl, The Gothic, 55). As perhapsbe entitled "the resistance to Bourges" (which was so Frankl points out (ibid.), Gervase of Canterburywas similarly aware much strongerand more widespreadthan the critical reception that the Gothic choir of CanterburyCathedral, which he chronicled, derived from France. In late medieval there is no reason to believe of Chartres-clearly a building with many strong historicist Italy, that the Gothic would not have been seen as French, not unlike something although traits, Suger's choir-emphasized by Bony).70 it may have become associated also with Germanyalready in the trecento What this reading seems finally to tell us is that fifty (H. Klotz, "Deutsche und italienische Baukunstim Trecento,"Mitteilun- years after Suger's inspired, yet divided patronage of Saint- gen des KunsthistorischenInstituts in Florenz, XI [1966], 173-206). By

202 the cinquecento the geographical locus of the Gothic had completely ed. P L. Gerson (New York, 1986), 136-137. The notion that previous shifted to Germany in the eyes of the Italians (who often called it the to Saint-Denis "Romanesque"architecture had suppressed the column maniera tedesca); but for Philibert Delorme, presumably reflecting is advanced in some detail by Sauerlinder ("Premiere architecture current French notions, the Gothic remained French, as well as being gothique,"); but the fact is that the column typically is retained in the "modern"(Frankl, The Gothic, 297). apse ("the most sacred part,"as Sauerlinder admits) even at Cluny and similar monumentally vaulted churches, and is replaced by piers only 7. On the term Romanesque, see Frankl, The Gothic, passim; the percep- in the nave. Moreover, it should be noted that typically these piers em- tive analysis of L. Seidel, Songs of Glory, The RomanesqueFacades of ploy columnar responds (sometimes pilasters), which, although less Aquitaine (Chicago, 1981), 4-7; and T. Bizzarro, Romanesque Archi- authentically antique than free-standing columns, could nevertheless tectural Criticism: A Prehistory (Cambridge, 1992). have been considered classicizing elements, which often are even of 8. Such a volte-face would closely parallel developments in literary-intel- relatively classical proportions. Furthermore,Sauerlinder's statement lectual circles centered in France, where a shift occurred around 1150 that "by the beginning of the twelfth century the column had disap- from a strong proto-humanisticrevival of antique material to self-con- peared from the nave of the Romanesque basilica throughoutEurope" sciously new methods and interests emphasizing novelty, promoted by is qualified by the authorhimself to exempt Italy and the Holy Roman scholars who sometimes were to call themselves see prone "moderns"; Empire-not exactly minor areas (despite Sauerlinder's calling them S. C. Ferruolo,"The Twelfth-CenturyRenaissance," in Renaissances Be- "backwaters");and one might add numerous large-scale examples of fore the Renaissance, ed. W. Treadgold(Stanford, 1984) 139-140, 144. columnarnaves in England and Franceitself (not just tiny ones as Sau- 9. I seek to explain this Roman purism in more positive terms in "On erlinder implies), representedby Tournus, Saint-Savin-sur-Gartempe, and Durham In Brunelleschi'sChoice: Speculations on Medieval Rome and the Origins Gloucester, (the intermediate piers). this light, the use of Renaissance Architecture,"in Architectural Studies in Memory of of the column at Notre-Dame in Paris (which has befuddled so many Richard Krautheimer,ed. C. L. Striker (Mainz, 1996), 169-174. historians [see n. 14]) or the Saint-Denis ambulatory(where the hemicycle ring, found, e.g., at Cluny, is simply doubled) seems consider- 10. E. Gall, Die Baukunstin Frankreichund Deutschland gotische (Leipzig, ably less anomalous or innovative. More innovative, to my mind, is the 1925). proliferation of the column throughout the interior, as pointed out 11. On this huge subject, see above all, E. Panofsky,Renaissance and Rena- below. See also the section below discussing Suger's writings. scences in WesternArt would that one (New York, 1965). I emphasize 16. The "scholastic"reasoning of Gothic design emphasizedby E. Panofsky and might think of, for example, the attenuated columns pilasters of (Gothic Architectureand Scholasticism [Latrobe,PA, 1950]) obviously in Speyer, Saint-Sernin,Cluny III, and Autunnot the usual terms of con- would have been virtuallyimpossible withoutthe mediumof the column. fusion and misunderstandingof the classical but more as self-conscious, 17. Of course the pilier cantonnmwas also foreshadowed in both Paris (in knowing, modernizing distortions of the very antiquity to which these the clustered aisle piers) and Laon (the cage-like eastern nave piers). same buildings are so powerfully attached. On the theologically problematic medieval status of the column see 12. On the problematics of the nineteenth-centuryconcept of the "transi- J. Onians, Bearers of Meaning, The Classical Orders in Antiquity, the tional" in modern architectural discourse, see A. M. Sankovitch, Middle Ages, and the Renaissance (Princeton, 1990), ch. 6. In medi- "Structure/Ornamentand the Modern Figuration of Architecture,"AB, eval illustrations idols often stand on columns, and when they fall the LXXX (1998), 687-717, especially n. 34. columns sometimes break or fall with them; a notable instance occurs in the socle relief on the west facade at discussed M. Ca- 13. The pointed arch, which first occurs in eighth-century Islamic archi- Amiens, by mille Gothic Idol the "columns of tecture, appears toward 1100 in Burgundy (for example, Cluny III) and (The [Cambridge, 1989], 2, Fig. 3): the like before the of the elsewhere (e.g., Durham). Its adoption is generally attributedto struc- Egyptian temple snap twigs onslaught Holy ... as if as of an elaborate edifice tural advantages; see, e.g., J. Bony, French Gothic Architecture of the Family representing, part sculptural of the total destruction of an alien and edifice and 12th and 13th Centuries (Berkeley, 1983), 17-21. That it appears first images, competing its in a "Romanesque"context would seem to confirm my reading of that images." period as one of not infrequent modernist episodes. In any case, resis- 18. For a clarification of this final modernist transformation, see A.-M. tance to the broken arch can still be seen in the early phases of Ile-de- Sankovitch, "A Reconsideration of French Renaissance ChurchArchi- France modernism, for example, Saint-Etienne at Beauvais (ca. 1120/ tecture,"in L'eglise dans I'architecturede la Renaissance, ed. J. Guil- 30), which continues to employ stilting instead of pointing to solve the laume (Paris, 1995), 161-180. The transformations of the classical geometric problem of equivalent arch heights in rib vaults; only in the column that I have outlined might be compared with Panofsky's anal- Saint-Denis choir (apart from the minuscule Morienval "ambulatory") ysis of contemporary sculpture, in which "the classical element is so does the broken arch first proliferate to near-consistent use. completely absorbed as to become invisible," and his reading of analo- gous trends in philosophy, historiography,and poetry (Renaissance and 14. This disorientationis chronicled by Sauerlinder, "Premierearchitecture Renascences, 102-103). gothique," 36-40. The author himself, however, overvalues the columns at Notre-Dame in Paris (and other classicizing episodes), virtu- 19. For a critical review and advancement of the research on early flying ally ignoring (as observed by S. Murray,"Notre-Dame of Paris and the buttresses, S. Murray,"Notre-Dame of Paris,"229-253. Anticipation of Gothic,"AB, LXXX [1998], 249 n. 17) the highly mod- 20. The crucial point in understanding the diffusion of modernism is to ernist forms of the rest of the building (except to criticize-or "smile distinguish between the main (Chartrian) typology of the northern at"-those who have sought to explain them). He proposes, "There is a French cathedraland the modernistorientation with its specific method- strong possibility that the Early Gothic architectureof northernFrance ologies. In the spread of "Gothic,"it was mostly not the former but the after St-Denis is just another aspect of the movement which Charles liberation from historicism and the design possibilities of the modern- Homer Haskins called 'The Renaissanceof the TwelfthCentury'" (p. 37; ist system that seized the architecturalimagination. The French typol- my emphasis). The discussion below suggests that the considerable im- ogy could be resisted, but not French modernism, except in Italy. portance of the columnar motif in the cathedrals involved more than 21. On the concept and problematicsof modernity,see above all J. Habermas, mere "revival"; see also nn. 8, 15. The Philosophical Discourse of Modernity, trans. E Lawrence (Cam- 15. On the Saint-Denis columns, J. Bony, "What Possible Sources for the bridge,MA, 1986);on the medieval use of the term,W. Freund,Modernus Chevet of Saint-Denis,"in Abbot Suger and Saint-Denis, A Symposium, und andere Zeitbegriffedes Mittelalters (Cologne, 1957); cf. n. 8 above.

203 22. See Habermas, The Philosophical Discourse of Modernity. Compare 33. This is offered as pointing to a restructuring of diachronics that is Terry Eagleton's recent remarks on transhistorical categories consti- possibly more historically grounded and centered than the approachof tuted, in my terms, by modernism and historicism: "Why should mo- G. Kubler in The Shape of Time (New Haven, 1962). define itself in terms, ratherthan reference dernity purely temporal by 34. Sankovitch, "Structure/Ornament,"694, 700-701. to a cultural style, a mode of production,an intellectual climate ... All 35. E. Abbot on the Church St-Denis and its periods are modern, but not all of them live their experience in this Panofsky, Suger Abbey of Art 2nd ed. G. is the central mode. Indeed the classical is a way of living one's experience as Treasures, Panofsky Soergel (Princeton, 1979) with its extensive to for though it were simply a reprise of the past, so that only those bits of it interpretativetext, bibliography updated 1978; more recent see W. W. "'The Recollection of the Past is which bear the legitimating seal of tradition can be regarded as literature, Clark, the Promise of the Future.' and authentic.... Modernity, by contrast, sees itself not just as one more Continuity Contextuality:Saint-Denis, and in Artistic in Gothic phase of time, but as a phase of time which re-evaluates the very no- Merovingians, Capetians, Paris," Integration 107-113. See also various in Abbot and Saint- tion of temporality. . . What strikes it as most typical about itself is Buildings, essays Suger Denis; P and L the dazzling, dismaying experience of time, which no longer comes Kidson,"Panofsky, Suger, St-Denis,"JWCI, (1987), 1-17; and C. Artistic at St-Denis wrapped in history or habit or custom but is now becoming almost Rudolph, Change (Princeton, 1990). their opposite. The modernis that which reduces everything which hap- 36. Kidson, "Panofsky, Suger, and St-Denis." I am in full agreement with pened up to half an hour ago to an oppressive traditionalism;it is less Kidson on Suger's status as non-architect,but not necessarily with all a continuation of history than an abolition of it" ("Newsreel History," items of his ratherharsh rejection of Suger as Neoplatonist (in terms review of P. Conrad, Modern Times, Modern Places [London, 1998], that call Panofsky "silly" [p. 8]). in London Review Books, 12 November of 1998, 8). 37. Panofsky, Abbot Suger, 34-37. 23. This distinction evidently has been overlooked by at least one reader 38. Ibid., 27. of my original article; see n. 28. 39. Ibid., 36-37, 52. 24. Cf. n. 8, however. 40. My analysis overlaps Panofsky's identification of four lines of defense 25. See nn. 5, 6. of the entire project (ibid., 27-28): due deliberation with the brethren; 26. Such traditionalismis, of course, to be distinguished from "invented" divine approval; preservation of old fabric insofar as possible; func- tradition, which is an historicist (or retrohistoricist)construct conjured tional necessity. of real and imaginary elements of the past; see E. H. Hobsbawm, "In- 41. Suger does not make the functionalist argument explicitly but indi- venting Tradition,"in The Invention of Tradition,ed. E. H. Hobsbawm rectly throughrepeated discussion of the crowding of the "narrow"old and T. Ranger (Cambridge, 1983), 1-14. choir and his "enlargement"(ibid., 86-89, 134-135). 27. Bony, French Gothic Architecture, 1. 42. Ibid., 100-101. To set the recordstraight about the composite/panoramic view of the Saint-Denis I made the 28. This aspect of Bony's work is clarified by W. Sauerlander's review, ambulatory(Fig. 18): photograph(s) in in the 1970s I a to Sumner for his "Mod Gothic," New YorkReview of Books, 8 November 1984, 43-44. 1970; early gave copy Crosby on Saint-Denis; it there without credit, and S. Murray ("Notre-Dame of Paris," 249 n. 17) does not distinguish monograph appeared proper has been between my concept of "medieval modernism" and Bony's "Modern- subsequently republished crediting Crosby. ist" reading. For a deeply insightful analysis of Modernist criteria in 43. Even Kidson, Panofsky'ssharpest critic, accepts the importancefor Suger twentieth-centuryarchitectural history, see A. Payne, "Rudolf Wittkower of the great chain of windows in terms of "light"("Panofsky, Suger, and and Architectural Principles in the Age of Modernism,"JSAH, LIII St-Denis," 10); see also Rudolph(Artistic Change, 65-68), who shifts the (1994), 322-342. effect to one of "saturatingthe church with imagery" in the brilliant stained glass "and the visual effect of that 29. This spirit pervades Viollet's work; see, for example, the article "Con- imagery." struction" in the Dictionnaire raisonne de larchitecture frangaise du 44. Panofsky (Abbot Suger, 242-243) demonstratesthat Suger is referring XIe au XVIe siecle, 10 vols. (Paris, 1858-1868). Cf. Sauerlfnder, "Mod not just to the transverse arches, but to the open ribwork as a whole in Gothic." this passage. Because the high vaults to which Suger referswere replaced in the thirteenth-centurycampaign, I illustratethis passage with Suger's 30. On the problematic concept of style, see above all M. Schapiro, ambulatoryvaulting (Fig. 19). The term ogive does not appearuntil the "Style" (1955), reedited in idem, Theory and Philosophy of Art: Style, thirteenthcentury. Artist, and Society (New York, 1994), 51-101, especially 62-65 regarding the critical position taken here. 45. Ibid., 108-109. 31. As in the statement, "The only other explanation for the [Gothic] fond- 46. To the six specified in De consecratione (and discussed below) must be ness for columns would be a respect for it as an Antique form. But this added the miracle of the speed of completion of the choir mentioned in seems excluded by the very nature of the shift from Romanesque to De administratione(Panofsky, Abbot Suger, 50-53) and quoted below. Gothic, involving as it did the rejection of a Classical vocabulary" Also many passing remarks allude to divine assistance ("the help of (Onians, Bearers of Meaning, 90). Cf. nn. 14, 15 above. God"), to the point that virtually nothing seems possible without it (e.g., ibid., 88-89). 32. "Medieval,"coined from Latin medius and aevum to designate a "middle age" between ancient and modern periods, would seem to make the 47. The miracle stories take up almost a third of De consecratione. term "medieval modernism" an oxymoron. This problem may be 48. Panofsky, Abbot Suger, 90-91. avoided, however, to the distinctions made above, accord- by adhering 49. Ibid., 90-97, 108-109. Panofsky notes the erection of the roof prior to to which "medieval" would not inherent ing designate any qualities (or the vaulting (243-244), which was, of course, the normal procedure. cultural "style") but only an "intermediate"period of time, while the 50. generic "modern"signifies the particularcharacter of the epoch. Admit- Ibid., 110-111. tedly, this would demand agreement on specific usage, a difficulty per- 51. In "Suger'sLiterary Style and Vision" (in Abbot Suger and Saint-Denis) haps best resolved by avoiding the term "medieval" altogether, with its R. W. Hanning presents strong evidence for "complexities of intention seemingly ineradicable connotations of cultural subordination to the and characterunderlying the apparentlysimple strategies of Suger's lit- framing classical-historicist periods. erary style" at the level of "rhetoricalstrategies, syntax, juxtaposition

204 of sentences" (149) and other literary-conceptualterms in ways sug- Spires," in Architectural Technology up to the Scientific Revolution, gestive of my reading (e.g., the way in which a "complicated set of tem- ed. R. Mark (Cambridge,MA, 1993), 182-231; and J. Munby, "Cathe- poral relations [is] resolved into a set of focal artifacts,"147). Compare dral Carpentry,"in The Archaeology of Cathedrals, ed. T Tatton- also G. M. Spiegel, "Historyas Enlightenment:Suger and the Mos Ana- Brown and J. Munby (Oxford, 1996), 165-182. (in the same volume, 151-158); she finds in subtle and gogicus" Suger 63. Panofsky,Abbot Suger, 48-51. It may be observed that the vault miracle and modes of elaboration"and, in- sophisticated "expressive purposes puts the vault and roof together.Moreover, I cannot resist the temptation a certain emphasis on triadic structures. triguingly, to observe that according to William Durandusand other medieval writ- 52. Panofsky, Abbot Suger, 92-93. ers (including St. Bernard) the roof of a church represented Charity, which covers a multitude of sins; Gulielmus Durandus, The Symbolism 53. This is implicitly the reading of Bony ("What Possible Sources") and of Churches and Church Ornaments, trans. J. M. Neale and B. Webb Clark ("The Recollection of the Past"), who associate the column- (1843; rpt. New York, 1973), 25; H. de Lubac, mddievale: Les events discussed below with the choir supports. Exegkse quatre sens de l'Ecriture, II (Paris, 1993), 41-60. Alternatively the roof 54. Panofsky, Abbot Suger, 92-93. tiles were the soldiers protecting the church against paganism and other J. des 118. 55. Ibid., 90-93. Although Panofsky (230) argues that the needed mono- enemies; Sauer, Symbolik Kirchengebdudes(Mtinster, 1964), liths "must have been those between the new narthex and the Carolin- 64. Whereas the post-Chartresprogeny seems numberless, the "school of gian nave," the quarrycertainly would have supplied those needed for Bourges" is generally considered to be limited to five (by no means neg- the new ambulatory as well. Although my reading of Suger's text is ligible) buildings, all begun in the second quarterof the thirteenthcen- directed primarily at the latter columns, it might also apply to the set tury: Saint-Martinin Tours,Le Mans, Coutances,Burgos, and Toledo. mentioned by Panofsky. 65. R. Branner, The Cathedral of Bourges and Its Place in Gothic Archi- 56. Panofsky accepts the Rome/Diocletian story at face value, and relates tecture, ed. S. P. Branner (Cambridge, 1989), 161. it to two occasions of the transportationof columns which he suggests 66. Ibid., 161; Bony, "What Possible Sources," 248. Suger may have known, without mentioning the disparity of shipping distances involved (ibid., 230-231). 67. Branner in fact appears to contradict his own estimate of the Bourges problem in his detailed, insightful analysis of its followers and their 57. Clark,"The Recollection of the Past,"94-95; the spoils were replacedin variations of the model (The Cathedral of Bourges, 177-201). the nineteenthcentury. W. D. Wixom compares Suger'sRome story with his extraordinaryproduction of ancient gems, vases, etc., in new "Chris- 68. Branner sees Bourges (although not quite as futurist) as "a significant tian" settings (e.g., the great chalice in Washington, D.C.), insightfully precursor of the later evolution of Gothic art" (ibid., 203). That, as observing, "If Suger had carried out his idea of transportingthe great Branner and others also show, the design of Bourges derives in many columns from Diocletian's palace [sic] in Rome, he would have suc- details and devices from earlier examples of the movement does not ceeded in outdoing the scale of all these antique insertions. Perhaps in contradict the hypermodernityof its transformationsof such models. his own mind he would have fulfilled his own renovatio more com- 69. R. Mark,Experiments in Gothic Structure(Cambridge, MA, 1982), 41- pletely if he had done so" ("TraditionalForms in Suger's Contributions 49. to the Treasureof Saint-Denis," in Abbot Suger and Saint-Denis, 303). 70. J. Bony, "The Resistance to Chartresin Early 13th-CenturyArchitec- 58. See n. 54. ture,"JBAA, XX-XXI (1957/1958), 35-52. 59. Bony, "What Possible Sources," 136; Clark, "The Recollection of the 71. A present-day parallel to Bourges and its critical context of "resis- Past,"95. tance" to absolute modernismappears as a central theme in R. Banham, 60. Kimpel and Suckale, Die gotische Architektur,92. Theory and Design in the First Machine Age, 2nd ed. (Cambridge, MA, 1980): in the twentieth century, Buckminster Fuller was virtually 61. Abbot 94-97. at first mentions Panofsky, Suger, Although Suger repairs alone (and without much influence) in actually following through in as the use for the timbers, at the end of the are story they explicitly practice the modernist "machine age" doctrines that were compro- used for the roof of the "new structure";cf. ibid., 235-236. mised by historicist and idealist strains in the work of other architects 62. Among the specialized publications on carpentry and roofs see, for (Gropius, Le Corbusier, et al.). Lest the reader remain unpersuaded example, M. Aubert,"La charpentede Notre-Damede Paris,"CAF, Paris that medieval modernism is anything more than a trendy neologism for (Paris, 1920), 397-406; J.-E Blondel, Les Fils de Nod: Les charpen- "Gothic,"I recommend a rereadingof the Bourges analysis substituting tiers du tempspasse (Paris, 1993); L. T Courtenay,"Timber Roofs and "Gothic" everywhere for modern/ism/ity.

205