• Author / Uploaded
• shumispace319

• Categories
• Posmodernismo
• Modernismo
• Narrativa
• Museo
• Paradigma

Citation preview

Ten Canonical Buildings: 1950–2000

Editorial Layout 02 08.indd 1

4/9/2008 10:43:34 AM

Peter Eisenman

TEN CANONICAL BUILDINGS 1950–2000

Foreword by Stan Allen Edited by Ariane Lourie

Editorial Layout 02 08.indd 3

4/9/2008 10:43:46 AM

First published in the United States of America in 2008 by Rizzoli International Publications, Inc. 300 Park Avenue South New York, NY 10010 www.rizzoliusa.com ISBN-13: 978-0-8478-3048-0 LCCN: 2007921092 © 2008 Rizzoli International Publications © 2008 Peter Eisenman “Eisenman’s Canon: A Counter-Memory of the Modern” © Stan Allen All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without prior consent of the publisher. Distributed to the U.S. trade by Random House, New York This book was developed with the support and cooperation of the School of Architecture, Princeton University.

DESIGNER Andrew Heid Printed and bound in China 2008 2009 2010 2011 2012 / 10 9 8 7 6 5 4 3 2 1 Eisenman, Peter Ten Canonical Buildings ISBN-13: 978-0-8478-3048-0 (alk. paper) 1. Postmodern Architecture 2. Critical Architecture II. Title. NA2760. E45 2006 720.1--dc22 2007921092

Editorial Layout 02 08.indd 4

4/9/2008 10:43:47 AM

Contents Acknowledgments

6

Eisenman’s Canon: A Counter-Memory of the Modern Foreword by Stan Allen

9

Introduction

15

1. Profiles of Text Luigi Moretti, Casa “Il Girasole,” 1947–50

26

2. The Umbrella Diagram Ludwig Mies van der Rohe, Farnsworth House, 1946–51

50

3. Textual Heresies Le Corbusier, Palais des Congrès-Strasbourg, 1962–64

72

4. From Plaid Grid to Diachronic Space Louis I. Kahn, Adler House and DeVore House, 1954–55

102

5. The Nine-Square Diagram and its Contradictions Robert Venturi, Vanna Venturi House, 1959–64

128

6. Material Inversions James Stirling, Leicester Engineering Building, 1959–63

154

7. Texts of Analogy Aldo Rossi, Cemetery of San Cataldo, 1971–78

178

8. Strategies of the Void Rem Koolhaas, Jussieu Libraries, 1992–93

200

9. The Deconstruction of the Axis Daniel Libeskind, Jewish Museum, 1989–99

230

10. The Soft Umbrella Diagram Frank O. Gehry, Peter B. Lewis Building, 1997–2002

256

Bibliography

288

Editorial Layout 02 08.indd 5

4/9/2008 10:43:47 AM

Acknowledgments The ideas and arguments presented in Ten Canonical Buildings were developed in seminars I gave over four years while a visiting lecturer at Princeton University’s School of Architecture. The school’s support, in particular the support of the dean, Stan Allen, made this book possible. I especially want to recognize the Princeton students who both participated in the seminars and spent summers producing drawings to illustrate these building analyses: John Bassett, Andrew Heid, Ajay Manthripragada, Michael Wang, Carolyn Yerkes and, later, Matthew Roman. Andrew Heid also stayed on to design this book. Clearly this book is the result of a team effort. Ariane Lourie endured numerous drafts and rewrites to help me bring this manuscript to its final form—even editing and repairing drawings—and Cynthia Davidson reviewed it for clarity. Jeffrey Kipnis made insightful comments on drafts of the introduction. Elise Jaffe + Jeffrey Brown helped to obtain the best historical images from the archives of Le Corbusier, Luigi Moretti, Mies van der Rohe, John Hejduk, Louis Kahn, Aldo Rossi, and James Stirling necessary to illustrate each building. I want to thank the architects who lent images from their offices: Venturi, Scott Brown and Associates, OMA, Studio Daniel Libeskind, and Gehry Partners. Finally, I would also like to thank David Morton and the editorial staff at Rizzoli New York for their patience and for reproducing these drawings with such care. –P.E.

Editorial Layout 02 08.indd 7

4/9/2008 10:43:47 AM

Eisenman’s Canon: A Counter-Memory of the Modern Stan Allen ‘Effective’ history deprives the self of the reassuring stability of life and nature, and it will not permit itself to be transported by a voiceless obstinacy toward a millennial ending. It will uproot its traditional foundations and relentlessly disrupt its pretended continuity. This is because history is not made for understanding; it is made for cutting. —Michel Foucault

The title of Peter Eisenman’s new book, Ten Canonical Buildings, suggests the construction of a new orthodoxy. Indeed, there is something didactic about Eisenman’s canon, and it is important to remember that these meticulous formal readings were developed in the context of seminars taught at Princeton from 2003 to 2006. At one level what is proposed here is nothing less than a new pedagogy, which would have at its center the close reading of exemplary twentieth-century buildings. In the past Eisenman has often been criticized for his reliance on concepts from outside of architecture. With this analytical work he declares explicitly that it is buildings themselves that are the source of ideas in architecture, and not applied philosophical concepts from outside the discipline. But to leave it at that would be to miss the force of his argument. His title, I would suggest, is something of a ruse; a sly bit of misdirection to distract the reader while he palms another ace off the bottom of the deck. Eisenman is operating on the basis of a rather unorthodox notion of the canonical, which places him much closer to Foucault’s idea of an “effective” history than to the conservative idea of maintaining a timeless, undeviating canon.

Editorial Layout 02 08.indd 9

4/9/2008 10:43:47 AM

10

It is the “liberating divergence” of architecture’s marginal or apparently insignificant moments that Eisenman has identified as canonical in this collection. In other words, innovation occurs when the previously marginal is absorbed into the discipline, triggering internal adjustments to the logic of the discipline itself. Eisenman understands the modern condition as shot through with contradiction, which is in turn manifested in formal discontinuity and historical rupture. “The purpose of history,” Foucault writes, “is not to discover the roots of our identity, but to commit itself to its dissipation.” “A canonical work,” Eisenman writes here, “is a hinge, a rupture, a premonition, in other words, of something that necessarily signals a change.” For Eisenman (an attentive reader of Foucault) the task of history is to make contradiction and discontinuity visible. He is searching for those moments when the ground of the discipline changes and the paradigm shifts. In this sense, Eisenman’s canon is the opposite of an eternal canon: it is precisely bound to the historical moment of rupture, meaningless outside the horizon of possibilities that it opens up at that particular time. To identify discontinuity as the primary analytical trope of this collection is also to take note of a conspicuous counterpoint to Eisenman’s mentor Colin Rowe. To mention Rowe here (as Eisenman does in his own introduction) is both to acknowledge the intellectual debt that Eisenman owes to Rowe and to measure the distance between the two. Rowe had famously postulated an underlying geometrical continuity between the classical and the modern. For Eisenman, Rowe’s emphasis on continuity locked modern architecture into a humanist tradition. To disengage modern architecture from its humanist tradition it was necessary to construct an alternative genealogy, in which fragmentation and discontinuity would now take precedence. Eisenman takes Rowe’s method of close formal readings and sets it to a

Editorial Layout 02 08.indd 10

Eisenman’s Canon

distinct task: the identification of breaks, ruptures and divergent pathways. He remains, however, indebted to Rowe for his analytical methodology: “Colin Rowe first taught me how to see what was written into the building but was not thematic of seeing as opticality.” Eisenman takes Frank Stella’s famous literalist dictum—“What you see is what you see”—and turns it on its head. Like his mentor Rowe, he is not interested in “what is literally there, but what is implied by what is there.” Perhaps the best known example of this method, and the essay that declared in the strongest possible terms his ideological distance from Rowe, is the article “Aspects of Modernism: Maison Dom-ino and the Self-Referential Sign.” In this essay (which opens with an epigraph from Foucault), Eisenman identifies the idea of the self-referential sign as the aspect of Dom-ino that makes it “truly modernist.” Eisenman’s starting point is the iconic perspective drawing of the Dom-ino system. Ostensibly the demonstration of a construction system, it is often taken as a diagram of the basic principles of the free plan. Eisenman reads the drawing against the grain, teasing out a series of small but significant formal moves that produce a kind of degreezero of architectural form: the minimum formal differentiation necessary to define the artifact as architecture, as opposed to mere structural diagram. All of the elements of the Eisenman methodology are here: the ostentatious disregard of structure, site, and program in favor of a nuanced formal reading, and the extension of that analysis as a more generalized proposition, which Eisenman calls a “diagram.” Maison Domino is one of the key diagrams discussed in this book. As in the previous articles, it is a privileged point of departure, a conceptual lever to open up the field of modern and postmodern architecture. The reference to Dom-ino here is then both to a method of analysis and to an exemplary modern-

4/9/2008 10:43:47 AM

Eisenman’s Canon

ist work. It is emblematic of the democratization of space under modernity and of postmodern architecture’s turn toward self-referentiality. For Eisenman, it remains “a true and seminal break from the 400-year-old tradition of Western humanist architecture.” Many of these same arguments are present in an earlier essay that prefigures the analytical method here: Eisenman’s brilliant, counterintuitive, formal deconstruction of James Stirling’s Leicester Engineering Building in “Real and English: The Destruction of the Box,” published in the first issue of Oppositions (1974), although written a decade earlier. Stirling stands in as Eisenman’s avatar in the intellectual tug-of-war with Rowe’s interpretive models: “In his need to clear a kind of ‘turf’ for himself, Stirling had to take on not only Le Corbusier but also the received interpretation of Le Corbusier provided by Stirling’s own tutor, Colin Rowe.” In a key passage and a sequence of diagrams that anticipate the more fully developed argument of the Dom-ino essay, Eisenman teases out the formal consequences of the Dom-ino diagram. While setting the structural support back from the edge of the horizontal plane of the slab emphasized the horizontal flow of space (sponsoring the free plan), it also freed the vertical surface from its structural support and allowed a layering of space in the vertical dimension. Eisenman locates Stirling’s formal innovation in an alternative proposition for the vertical surface that “implies the potential for presenting the vertical plane as a dominant spatial datum, while using a vocabulary which runs counter to the by-now traditional dematerialized cubist aesthetic.” The accuracy of this formal reading is perhaps less significant than its methodological implication. To me, the real force of the essay is to foreground the formal characteristics of Stirling’s architecture against the then-dominant interpretations of his work—as well as Stirling’s

Editorial Layout 02 08.indd 11

11

own explanatory framework. At that time, as is still the case today, Leicester was interpreted almost exclusively in terms of the clarity of its functional arrangements, its direct (not to say “brutalist”) use of industrial materials, and a series of quotations of canonical modernist precedents (the echo of Melnikov, for example, in the thrusting angle of the auditorium). To claim early Stirling instead for the camp of self-referentiality and formal innovation is provocatively counterintuitive. It opens the work up to wider interpretation, and serves to confirm the idea that a complex work like Leicester will always exceed definitive explanation. The analysis of Leicester is reprised in the current volume, with added anecdotal background, which makes it a better read, and newly drawn diagrams, which make the argument clearer. More important than chronology and precedence is the method itself: Eisenman’s dogged determination to read certain of these buildings against the grain of the received interpretation, through the primary vehicle of the cut-away axonometric diagram. This has its awkward moments. The drawings of Frank Gehry’s Case Western Reserve project, which emphasize the roof geometries seen from above, seem inadequate to the sculptural effects of an architect who designs almost exclusively in model form, and with a close attention to the experience of the building from street level. It does, however, yield brilliant formal insights in the analysis of the Jussieu Libraries, and reminds us that Rem Koolhaas, for all his engagement with architecture as social/cultural prop, is an architect of subtle and sophisticated formal invention. After all, would we really be so interested in Koolhaas if he were simply using architecture as an instrument of social criticism? Similarly, in his patient explication of the changing plan strategies for Robert Venturi’s mother’s house, Eisenman reminds us that Venturi, although usually asso-

4/9/2008 10:43:48 AM

12

ciated with the semiotic capacity of the vertical plane, is a brilliant plan maker, whose buildings can stand up to extended formal analysis. A final point of reference, perhaps less immediately obvious; it seems to me that Eisenman has internalized Harold Bloom’s idea that, as an author takes on his predecessors, rather than confront a fully realized, mature masterwork, it is often the early work, or the slightly marginal and unresolved aspects of the mature work, that offer a kind of handhold, or a crack to open up the field, and clear space for working. While Mies van der Rohe’s Farnsworth House and Stirling’s Leicester Engineering Building are inarguably central to postwar architectural history, Luigi Moretti is a less obvious choice. To examine Louis Kahn’s Adler and DeVore Houses, rather than his better-known public buildings, is similarly counterintuitive. We understand Le Corbusier’s Palais des Congrès in Strasbourg as canonical today primarily because it has sponsored several generations of work on the warped surface. In this case, Koolhaas’s Jussieu Libraries confer a retrospective “canonical” status to this previously somewhat overlooked building. But there is more here than a pursuit of the obscure for its own sake. Eisenman finds and zeros in on those moments—in well-known and in less well-known buildings—that still offer room for working. Eisenman’s canon is definitively not a new orthodoxy. A canon usually implies looking back to validate history’s great, untouchable monuments. Eisenman’s canon is instead anticipatory—it lays the groundwork for future monuments. It is also—in contradistinction to the notion of an anonymous canon handed down from on high—somewhat idiosyncratic, and in the end, highly personal. For all that, it is surely not a teleology with “Eisenman” at its endpoint. It is neither a universal canon nor an individual genealogy. It is both the record of one architect’s intellectual trajectory and a method that sug-

Editorial Layout 02 08.indd 12

Eisenman’s Canon

gests other, future trajectories. These buildings are precisely where the possibility of the new becomes evident for the first time, even if in a tentative and incomplete manner. This may be Eisenman’s most telling insight. He presents here a collection of suggestive possibilities, of architectural problems opened up and provisionally addressed, but always leaving room for the next author to complete the work, and create a new break, which will in turn open up new territory for generations of architects to follow.

4/9/2008 10:43:48 AM

Editorial Layout 02 08.indd 13

4/9/2008 10:43:48 AM

Introduction

In reading Harold Bloom’s book The Western Canon, I discovered that the term canon has more mobility than might have been at first assumed, and that it could help to structure my thinking about the fundamental project of this book, which is to address the necessary evolution of close reading in architecture. While The Western Canon looks at what constitutes canon in Western literature, some of Bloom’s various and perhaps subtle uses of the term help to clarify my thinking concerning this period of time. Bloom says in different contexts that canon refers to the experience of limits, which are extended or broken (74), or which are vital, original, arbitrary and personal (75). For Bloom, canon refers to authors and their entire oeuvre; in the context of this book, however, canonical buildings are singular works without reference to their authorial provenance. For Bloom, canon has centers; in this book, the edges and cusps are of interest. For Bloom, canon also has a heretical intensity (72), which is useful in distinguishing canon from its use in religious, as opposed to artistic or scientific contexts. The idea of canon would refer to an operative dogma in a religious context: an orthodoxy, as in canon law. In science, a canonic pattern—such as canonical coordinates or canonical conjugates—contains an uncertainty. A canonical pattern in music is contrapuntal, repeating but also constantly changing. In the context of this book, the term canonical encompasses the potential heretical and transgressive nature of ways of close reading architecture. If, as Bloom suggests, political correctness can be considered a polemic against difficult art, then the canonical is a combination of difficult and popular (56); in this book, it is the distinction between the easy and the difficult in terms of readings that will be made. Finally, no less an author than Michel Foucault rails against the idea of canon

16

and replaces it instead with the idea of the archive, which reorganizes hierarchies. I did not set out to define or co-opt the term canonical for architecture. In fact, even though I have attempted a provisional definition here, this is not the purpose of what follows. Rather, the idea of the canonical informs my interest in reading architecture, and also explains the inclusion of each building in this book, which lays out their roles in defining today’s particular historical moment in architecture. If part of the meaning of the term canon is to contravene its own accepted definition, then its use here represents that possibility. More specifically, the term canonical begins to define the history of architecture as a continual and unremitting assault on what has been thought to be the persistencies of architecture: subject/object, figure/ground, solid/void, and part-to-whole relationships. These concepts become canonical over time; therefore, in their attack on the canon, these buildings become canonical in themselves. But as a group, the buildings herein do not represent a canon. Rather, the idea of the canonical begins to describe potential methods of analysis, which derive from an interest in reading architecture in a more flexible and less dogmatic way. While this is a personal selection of architectural works, the ten buildings in this book do not represent my personal canon. Rather, they were chosen, in retrospect, for two reasons: they represent both a necessary evolution in the terms of close reading and an evolution in the nature of that close reading, from the formal to the textual and perhaps even the more phenomenal. Perhaps most importantly, these buildings not only challenge the canons of architecture, they also challenge our received idea of the canon of close reading. All of the architects discussed here represent different ideological, theoretical, and stylistic points of view, as well as different attitudes toward site, material, and program. What defines them, however loosely, in my opinion, is

Introduction

at the core of a postmodern practice, as distinct from a modernist practice and from the current state of architectural practice. This book seeks to locate the core ideas that form the basis for their argumentation. Ultimately this will be seen to involve both a rethinking of the reading strategies which sustained modern architecture and, at the same time, reiterate a demand for other forms of close reading. *** Colin Rowe first taught me how to see what was not present in a building. Rowe did not want me to describe what I could actually see: for example, a three-story building with a rusticated base, increasingly less rustication in each of its upper stories, and with ABCBA proportional harmonics across the facade, etc. Rather, Rowe wanted me to see what ideas were implied by what was physically present. In other words, less a concern for what the eye sees—the optical—and more for what the mind sees—the visual. This latter idea of “seeing with the mind” is called here “close reading.” Each of the buildings discussed here requires one to see in a different way, particular to the building under consideration. While these ten buildings may reduce the effects or thematic of opticality, each in turn organizes a different demand on visuality. Visuality does not refer to a prima facie response to image, but rather to what is apparent and implied by aspects of the building’s formal organization. Each of these buildings requires close reading. Close reading can be said to define what has been known until now as the history of architecture. But for our purposes here, close reading also suggests that a building has been “written” in such a way as to demand such a reading. If the first question posed in this book is: “close reading of what?” then one of the answers proposed in the following chapters involves the

Introduction

close reading of critical architectural ideas. The readings proposed in this book would not have been possible before 1968, with the effect of Jacques Derrida’s Of Grammatology and the idea of the undecidability of any single reading invoked therein. The use of the term undecidable in the context of this book is no mere wordplay between ambiguity, indeterminate, multiple and undecidable. The differences between these terms are crucial. Modernism was perhaps best defined by William Empson’s seven types of ambiguity. The idea of ambiguity lodges itself in a dialectical notion of either/or and determinate/indeterminate, which, as decidable characteristics, possess a supposed clarity which belies any need to examine their repressions. Undecidability questions the very nature of the notion of ambiguity itself. It is in this context that Derrida’s work remains underexamined in today’s architectural culture, which has gravitated toward the more facile interpretations of a Deleuzian schema of the multiple. If since 1968, undecidability is an aspect of criticality, and since undecidability as opposed to ambiguity is perhaps more difficult to tease out in architecture as opposed to say in literature, then today, more than ever, a close reading comes to terms with undecidability. The idea of undecidability makes it possible to look back and see changes in work which in turn demand a new kind of close reading, which, it will be argued, responds to the evolution of canonical in architecture. It is first necessary to distinguish between a canonical period in history and the period from 1950 to 2000 covered here. One way to study the discipline of architecture is to use a particular period in history as a master exemplar, to use the historical conditions of a particular period to stand for history per se. For example, instead of using history as a narrative structure, it is possible to take the period in northern Italy from 1520 to 1570 to describe a canonical moment in the history of architecture. This specific canonical

17

moment could serve to shed light on other such canonical moments in architecture’s history. It may not be necessary therefore to study many such moments to understand what is meant by a canonical moment. Canon in that sense requires a specific historical context, but it is not necessarily an expression of such a moment, a Geist, or a comparable historicizing imperative. It could be argued that a canonical moment describes what could also be called a paradigm shift. But the idea of a paradigm shift does not necessarily implicate the critical content latent in the idea of the canonical. The purpose of distinguishing a canonical moment from a history is that while history provides a narrative flow to the discipline of architecture, it does not in itself provide a necessary basis for close reading and for opening the discipline to question its own history, and thus to alternative interpretations of that history. As used here, the term canonical initially provides a possible basis for an alternative reading of what today constitutes the critical in architecture. Rather than focusing on history qua history—this building was built at this time, used in this way by this architect, etc.—the idea of canon in architecture also makes possible the recording of the changes in close reading, in issues that range from the formal to the textual, or from the phenomenological to the performative. Thus canon is a way of opening up a particular discourse to reading its own history as something other than a narrative of facts. These readings are the wedge that allows postwar modernism to be seen, absent its former ideology and clichéd rhetoric, as imbued with other powerful concepts. If canon establishes a perimeter to the center of the discipline, then such readings suggest that a critique of canon ultimately displaces this perimeter with a new canonic idea. It will be argued that the canonical will inevitably be a critique of what at any moment is termed the canon.

18

Introduction

1. Luigi Moretti, Casa “Il Girasole.”

2. Mies van der Rohe, Farnsworth House.

These ten buildings do not so much describe a history as they define the evolution of canonical works that eventually became known as postmodernism. Close readings that are other than formal or conceptual remain within and are at once canonical to a postmodernism and at the same time heretical to mainstream modernism. The term postmodernism is not used here to denote a style but rather refers to the period of time after modernism. Postmodernism reflects an attitude concerning ideas about architecture which are articulated as a critique of modernism and particularly of abstraction, modernism’s dominant mode of close reading. Not all buildings in the years 1950 to 2000 describe this moment. The ten buildings here are read, each in their own way, through a different lens, producing arguments which, taken together, define a series of canonical moments that loosely identify some of the transgressive concepts of the postmodern period.

The idea of the canonical is often confused with the idea of a so-called great work. In the context of this book, canon is not necessarily a list of great work, nor is it necessary for a canonical building to be a great work. In one sense, canon and great work have little to do with each other. A great building may be just that, requiring no more than an initial look that defines a single, directed reading, while a canonical building presupposes in this context undecidable, often diffuse readings as a necessary condition of the critical. As will be seen here, a close reading of a great building is complete unto itself, like JØrn Utzon’s Sydney Opera House or Frank Gehry’s Guggenheim Museum Bilbao, which requires little or no outside references in order to be read. This is not the case with a canonical building, which requires a reading forward to what the building inspired, as well as backward to what the building denoted.

Introduction

19

3. Le Corbusier, Palais des Congrès-Strasbourg.

4. Louis Kahn, DeVore House.

In this sense, great buildings are timeless, while canonical buildings are identified with specific moments in time. For example, in the eighteenth century, Palladio’s Villa Rotunda was considered canonical because its close reading produced an interpretation of his Villa Malcontenta. Yet in the twentieth century, Villa Rotunda was seen as a great work, and Malcontenta came to be called canonical because its close reading spawned an interpretation in Le Corbusier’s Villa Stein at Garches. A canonical building requires study, not in and of itself as an isolated object, but in terms of its capacity to reflect on its particular moment in time and its relation to buildings which both precede it and come after it. In the study of the buildings collected here, each canonical work impinges on those works created in its wake, works that in turn redefine what is considered canonical. Thus canon is intimately linked to and dependent on both the concept of close reading

operative at a particular moment in time and on the specific works which at the time provoke such a close reading. The canonical both places in doubt previous work and demands new interpretations, not only of the individual work, but also of architecture in general. In short, while the canonical building requires close reading, it also problematizes the idea of a great building or masterwork as a historically sedimented concept, without the mobility and flexibility that canonical implies. For example, one of the buildings discussed here is Gehry’s Peter B. Lewis Building for the Weatherhead School of Management at Case Western Reserve, which is neither as well known nor as great as his Guggenheim Museum Bilbao. There can be little doubt that Bilbao changed the architectural face of the ensuing decade, and it is certainly what can be called a great building or a masterpiece of its time. The first question is, then, why Case Western rather than Bilbao?

20

Introduction

5. Robert Venturi, Vanna Venturi House.

6. James Stirling, Leicester Engineering Building.

While Bilbao is effectively the most well-known and popular of Gehry’s buildings, this building was not so much concerned with reading and producing a critical stance on modernism as it was the reflection of a personal sensibility, albeit about the siting of a building in the city. Bilbao may be a great postmodern building, and its quality establishes Gehry’s personal view of the object in the city, but it does not embody an argument about its relation to history in the critical terms that characterize the Lewis Building. The argument set forth in this book considers the Lewis Building to be a canonical, as opposed to a great building, in that it organizes a demand for a close reading of a different kind, one that differs from the formal and conceptual readings that dominate architecture’s recent past. The Lewis Building can be considered canonical in defining more clearly its theoretical rupture with classic modernist readings than does Bilbao, because it refers back to the history of the discipline, espe-

cially to the plan of Schinkel’s Altes Museum, as a progenitor of the modern. Also discussed here are the Adler and DeVore Houses by Louis Kahn, as opposed to his betterknown, even seminal projects such as the Indian School of Management at Ahmedabad, the Exeter Library, or the Yale University Art Gallery. The Adler and DeVore Houses are an obscure pair of houses that were never built, yet they demonstrate certain of Kahn’s ideas in what was a crucial turning point in his career. They represent a moment in Kahn’s career between the Trenton Bathhouse, which preceded these houses, and the Richards Medical Building; the houses represent a moment which articulated several possible directions for Kahn’s architecture. The Adler and DeVore Houses also contain the origins of his eventual career direction; in fact, his next major project, the Richards Medical Building at the University of Pennsylvania, evolves as a stylized Kahnian trope that is clearly derived from these two houses. The

Introduction

same can be said for each building in this book, some more obvious than others. While individually these buildings may each be canonical, there is no intention in their collection here to define any socalled postmodern canon. A canonical building also spawns subsequent interpretations by other architects as a commentary on that particular moment. For example, Le Corbusier’s Palais des Congrès-Strasbourg both manifests his own critique of his earlier “Five Points,” and serves as a model for Rem Koolhaas’s Jussieu Libraries project. Thus each project discussed here represents a moment in architecture in which there is an acknowledgment of the past, a break with the past, and simultaneously a juncture with a possible future. While a great building perhaps is self-sufficient, a canonical building is not. Its outward references, forward and back, make it contingent on external factors. The difference between a canonical work, as defined here, and a critical work is more nuanced. All canonical works are per se critical, but not all critical works are canonical. The critical can be considered a necessary but not sufficient component of the canonical. In this book, the term critical refers to the capacity to open up to questioning problems which are essentially architectural. In the sense that it is used here, critical is a concept that distances the object or subject from the terms of the analysis at the same time that the analysis is also part of the subject or object. The important distinctions between critical and canonical are twofold: first, a canonical work is a hinge as well as a rupture, while a critical work can function principally as a break with its precedents. Canonical in this context refers to a rupture that helps to define a moment in history; it is a constant reevaluation in the present as to what constitutes such a rupture. Of course, a rupture can only be seen in hindsight, looking back rather than looking at the present. Second, a canonical work is time-bound: it depends on a particular

21

moment in history in order for it to be seen as a hinge/rupture in either the architect’s career or the architectural discourse. A building’s function, structure, and type—its instrumentality—are not the criteria for understanding its importance in the discipline of architecture, nor would these be considered aspects of its criticality. All buildings stand up; all buildings function; all buildings enclose. These qualities comprise neither the central characteristics nor the thematic of the buildings analyzed in this book. Canonical buildings are not considered canonical because they have functioned well; their instrumentality has never been the cause of their canonical role in the discipline. For example, whether or not Borromini’s churches functioned well has not been a concern in history, because the functioning of the church was not necessarily its thematic. Rather, the representation of those functions in the artifact was important. Whether the mass could be heard or whether Easter service was crowded was not the issue for Borromini or for his patron; in fact, these matters have never been the issues for the history of architecture. Equally, very few people care whether Gehry’s Guggenheim Museum Bilbao functions well or not; and many great museums—the Louvre in Paris, the Frick Museum in New York, and others—were not designed as such. There is no such thing as a good plan for a museum, because there is no plan for a museum. If canon is commonly associated with the critical as a reference to prior work, canon is also commonly associated with the textual, that is, an internal critique or questioning of its own status as a narrative. For the textual, I am referring to the Derridean idea that texts manifest the legible dimensions of ideas and objects while linking them with preexisting ideas and objects. In the context of this book, the textual will also connect ideas—for example, in the form of a diagram, an explanatory or analytical device aims to uncover

22

Introduction

7. Aldo Rossi, San Cataldo Cemetery.

8. OMA/Rem Koolhaas, Jussieu Libraries.

latent organizations. The textual becomes a tissue of marks that are no longer only representational as the three types of sign identified by the American pragmatist philosopher Charles Sanders Peirce: the icon, the symbol, and the sign. According to Peirce, the icon has a visual similitude to its object; the symbol establishes a visual convention for the relationship of the symbol to some object, and the index, which does not rely on the thematic of the optical, functions as a record or a trace. Each of these ten buildings will be situated as the fulcrum of an argument that the building defines, an argument that can be grasped through a close reading of textual, formal, and conceptual strategies. These will not always be the most well-known buildings, but they will stand for a moment when the relationship between the sign and the signified, the relationship between the

subject and the object, the relationship between form and meaning, and between instrumentality and discourse come into sharp focus. The period from 1950 to 1968 could be characterized by a rethinking of modernist abstraction. Thus the first four buildings shown here, each in its own way, define and critique previous invocations of close reading affiliated with modernism. For example, while Luigi Moretti’s Casa “Il Girasole” demands a formalist close reading, it begins to introduce concerns such as historical references and materiality, which later become known as postmodernism. Mies van der Rohe’s Farnsworth House, while continuing Mies’s investigation of the column grid in relation to interior space, external surface, and the corner, is still the most abstract—if not the most overtly modernist—of the four buildings, but it also becomes a manifestation of Mies’s first diagram.

Introduction

23

9. Daniel Libeskind, Jewish Museum.

10. Frank Gehry, Peter B. Lewis Building.

Le Corbusier’s Palais des Congrès-Strasbourg requires a reading beyond formalist close reading because at root it is a reversal of his own “Five Points,” but more importantly because it introduces a centripetal energy as well as a centrifugal energy that moves attention away from the center to the periphery and thus away from any classical, centric, deep-space composition. The last of the early projects is Kahn’s Adler and DeVore Houses, which represent both a rejection of the modernist free plan but also deny traditional part-to-whole relationships. Instead they introduce a play of readings, which ultimately are undecidable. This pair of houses is thus a hinge in Kahn’s career, but also a hinge between the first phase of postwar building and the second, transitional postmodern phase in America. The three buildings that characterize the second generation, from 1968 to 1988, exhibit

similar characteristics in orienting their critique of modernism toward a new realism, expressed in structure, materiality, and iconography. Aldo Rossi’s San Cataldo Cemetery presents a critique involving surreal or superreal shifts; James Stirling’s Leicester Engineering Building reverses the conventional solid/void characteristics of material, and Robert Venturi’s Vanna Venturi House evokes the form of an American shingle style with European overtones. However, none of these buildings lapse into a simple phenomenology. In fact, the characteristics of these buildings have less in common with the pure materiality of Kahn’s Trenton Bathhouse than they do with each other in broaching the conceptual implications of organization, type, and material. The three projects in the last section of this book, from 1988 to 2000, not only require close reading but also mark sufficient changes in what

24

constitutes the idea of close reading. This period begins with the Deconstructivist Architecture exhibition at the Museum of Modern Art, which included, if not the specific projects of Rem Koolhaas, Daniel Libeskind, and Frank Gehry discussed in this book, at least their sensibilities. What links them is a concern for diagram, as opposed to type, but each architect employs the diagram in a different way. In Koolhaas’s Jussieu Libraries, the diagram is an iconic device where the building displays a visual similitude to the animating diagram. As such Koolhaas’s work begins to define another reading strategy, one that Jeffrey Kipnis defines as performative rather than conceptual. In the performative strategy, the human subject becomes involved in the architectural object in a way similar to the minimalist sculptors’ involvement with the subject, the object, and the site specificity of the work in the late 1960s and early 1970s. Libeskind’s Jewish Museum also invokes the diagram, but to indexical ends, where the building marks a series of traces of its process of becoming. This organizes the demand for a close reading of not only the traces within the building but also the traces of its own origins in a prior project. It will be argued that Gehry’s Lewis Building for the Weatherhead School also relies on a diagram that invokes a shift in reading from the formal or conceptual to the phenomenological. These three projects, more than the other buildings assembled herein, best describe the dilemma of close reading today. Equally, it is perhaps too early in the architects’ respective careers, and in time, to assess which buildings in their oeuvre could be considered canonical, although it is certainly possible to understand their effect on the idea of close reading. In each case, the buildings herein disturb the complacency of the act of reading. The idea of undecidability suggests that readings are no longer necessarily dialectical. Ultimately it is not buildings but their readings which are undecid-

Introduction

able. These buildings not only challenge the formal and conceptual conventions sedimented in the history of close reading, but also challenge what constitutes the persistencies of any architecture: part-to-whole, subject/object, Cartesian coordinates, and abstraction/modernism. In attacking the clichés of modernism, these buildings of the postmodern period remain engaged in a challenge to opticality and the metaphysics of presence. In suggesting that the challenge posed by one era becomes clichéd in the next, this book offers neither solutions nor instructions for contemporary architecture, but rather presents a slice in time that is part of an endless cycle of becoming, and as such an idea of infinite displacement.

1. Luigi Moretti, Casa “Il Girasole.” Rome, Italy, 1947−50.

1. Profiles of Text Luigi Moretti, Casa “Il Girasole,” 1947–50 One of the first critical articles to appear in English on Luigi Moretti’s Casa “Il Girasole” was written by Peter Reyner Banham in 1953. Banham’s article, published in the February issue of Architectural Review, labeled Casa “Il Girasole” the defining monument of “Roman eclecticism,” which was an eclecticism that Banham considered operated within the confines of the vestiges of modernism. If the label eclecticism has different connotations today, in 1953 it implied that Moretti’s work could be seen as a haphazard collection of classical tropes and architectural strategies lacking any single organizing principle other than having been assembled by Moretti in a single building. In this sense Banham’s argument was prophetic, though his use of the term eclecticism, it will be argued here, was flawed. It is interesting to note that as early as 1953, Banham proposed that modern architecture had already become a style, and thus he was able to cite Moretti as deviating from its formal and supposed social imperatives. Moretti’s Casa “Il Girasole” would subsequently earn an important citation in Robert Venturi’s 1966 book Complexity and Contradiction in Architecture, a citation that would become physically manifest in Venturi’s own Vanna Venturi House (see chapter 5). One important distinction between Banham’s conclusion and a possible present reading is that prior to 1968, and the rethinking of the idea of a text proposed by Jacques Derrida’s Of Grammatology, it was not possible to propose a textual reading of what appeared to Banham to be mere eclecticism. Post-structuralism offered methods of analysis and composition as a new lens through which to understand complex phenomena; in certain cases, these phenomena defy a clear reading altogether, and instead represent a condition of what can be now called undecidability.

28

Casa “Il Girasole”

2. Casa “Il Girasole,” south elevation.

3. Casa “Il Girasole,” north elevation.

In this context, Moretti becomes neither an eclectic nor a modernist; rather, his work defies any easy categorization, even as one of the first, if rarely acknowledged, postmodern architects. It is this condition of what can be termed undecidability that emerges in his Casa “Il Girasole” and will develop as one of the defining themes of this book. Completed in 1950, Moretti’s Casa “Il Girasole” incorporated the first appearances of historical allusion in the wake of modernist abstraction. This overture to history is not, however, why Casa “Il Girasole” is the first building in this book. Rather, it is because Casa “Il Girasole” represents one of the first postwar buildings to manifest a hybrid condition of both abstraction and literal figured representation. These simultaneous yet seemingly antithetical positions are never resolved as a single narrative, meaning, or image. Rather, it is the dialectical relationship between the two positions that is questioned in a postwar climate that challenged the innate value of such a dialectic. Furthermore, it could be argued that Casa “Il Girasole” represents one of the first buildings after World War II to embody the undecidable nature of truths in attempting the parallel use of both abstract and figured tropes. It is here that an idea of what might be considered a text in architecture might be introduced. While the abstract and the figured refer to what is usually

described as the formal, the distinctions between the formal and the textual in what follows will be seen to be important. The term formal describes conditions in architecture that can be read not necessarily in terms of meaning or aesthetics, but in terms of their own internal consistency. This internal coherence involves strategies that have nothing to do with the primary optical aspects of the aesthetic (proportion, shape, color, texture, materiality) but rather have to do with the internal structure governing their interrelation. Formal analysis looks at architecture outside of its necessarily historical, programmatic, and symbolic context. The term textual can be defined in relationship to one of post-structuralism’s key concepts in the Derridian idea of text. Derrida suggests that a text is not a single linear narrative, but a web or a tissue of traces. While a narrative is unitary, continuous, and directional, a text is multivalent, discontinuous, and nondirectional. In the context of this book, the idea of a text, like the idea of a diagram, helps to initiate a change from the idea of reading a work as a unitary entity to understanding a work as an undecidable result of varying forces. In my work on Giuseppe Terragni, for example, the idea of a text reoriented my analysis of Casa Giuliani-Frigerio from essentially formalist interpretations to a more textual reading.

Casa “Il Girasole”

29

4. Casa “Il Girasole,” west elevation.

5. Casa “Il Girasole,” section, north-south.

Texts, therefore, do not deploy the same internal consistency as in the formal. In addition to provoking formal reading, buildings can equally be read as textual, offering different modes of reading, which may challenge established architectural vocabularies. For example, Alberti’s superposition of the Arch of Titus over the vernacular Greek temple-front at Sant’Andrea becomes textual, because this montage of architectural forms from different historical periods destabilizes a singular meaning. The textual provokes a reading outside of the facts of an object’s physical presence, or the underlying structures which govern its being; in the case of Alberti’s Sant’Andrea, the superposition of historical tropes creates this disturbance in presence that takes the building out of the category of the conventionally formal. If the formal begins from a conception of presence that is both a linear narrative and what can be called fixed or decidable, then the textual suspends the narrative of presence, in which a hierarchy is implicit, and offers instead undecidable relations rather than a single static condition. It is this undecidability of relations with both historical and modernist tropes that Moretti invokes to produce an initial critique of modernism. The abstract languages of cubism and futurism were subjected to a critique, which first took

form in Italy through neorealist cinema and its unvarnished view of Italy and the detritus of five years of war. Neorealist films like Open City and The Bicycle Thief were a form of empirical existentialism, in that they represented attempts to move the language of abstraction toward a language more closely associated with what could be considered “the real.” Moretti’s postwar work, which also proposed a didactic view of architecture that now critiqued abstraction, evolved out of such a neorealist sensibility. However, it is to Moretti’s credit that little of his first postwar work can be considered neorealist, just as it cannot be dismissed as eclectic. The subtlety of Moretti’s critique of modernist abstraction was articulated in his now much sought-after magazine Spazio (Space) in the early 1950s. Spazio followed in the tradition of architects’ little magazines, which began with Le Corbusier’s magazine L’Esprit Nouveau in 1920 and Mies van der Rohe’s magazine G, with Theo van Doesberg and El Lissitzky, in 1923. While Le Corbusier’s magazine referred to a new spirit, and the G of Mies’s magazine stood for Gegenstand (object) and effectively addressed ideas about objecthood, Moretti’s Spazio made an important distinction between the objectthing and the object of containment as space or volume. An object can be seen and analyzed as

30

Casa “Il Girasole”

6. Casa “Il Girasole,” ground-floor plan.

7. Casa “Il Girasole,” second-floor plan.

a geometric abstraction, but space is difficult to analyze as a physical entity because it is usually defined by other things. While space is a conceptual entity, its container is formal. Such a redefinition of the modeling of space was among the issues Moretti broached in Spazio. It was Moretti’s article “Valori della Modanatura,” (The Value of Modeling) in Spazio 6 (1952) that challenged the modernist conception of space. The article suggested that surface had the capacity to be modeled in such a way as to create a dialogue between volume and flatness, and therefore that the modeled surface could engage the affective potential of light and shadow. The article challenged the boxlike abstractions of modern architecture by raising the issue of profile, which is articulated through both hard edge and figured form. Profile is the edge of a figure—in other words, how a surface in architecture meets space:

the edge of a volume seen against the sky is a literal profile. This means that all architecture, because it is three-dimensional, will have some sort of profile. While in architecture a profile is the edge of a plane or the edge of a surface, it is also either the edge of the containing surface or the edge of the exterior space in relationship to the containing surface of the interior. In either case, profile tends to be the result of figured form, which in turn produces shadows. Moretti was not referring to a literal profile per se but to a conceptual profile, which was made thematic in the design. Moretti made profile thematic in his work by suggesting that profile becomes more than just the edge of a three-dimensional volume and instead serves to question the clarity of boundaries between edge and volume. In Moretti’s terms, profile is not a narrative device, revealing shape or figure, but rather can be disassociated from any shape or figure; this disassociation is not

Casa “Il Girasole”

31

8. Casa “Il Girasole,” third-floor plan.

9. Casa “Il Girasole,” roof plan.

merely a line but can be, for example, the dark edge of cast shadows. By calling attention to profile in architecture, Moretti suggests its role as a marker of undecidable relationships and engages space as an object for close reading. As hierarchy and singularity of meaning are made problematic, the rhetoric becomes textual rather than formal. The idea of space as volume was illustrated in Spazio by Moretti’s series of cast models of historical buildings, churches, and villas. Moretti broke with the conventions of architectural models by representing a building’s interior space as a solid volume and dispensing entirely with its exterior enclosure, structure, facades, or any other indications of an exterior skin. These volumetric models seemed to deny a relationship to the exterior. Rather, they embodied space itself, conceptualizing space by turning void into solid. In the history of architecture, analysis usually begins from the geometric, and from elements that can be touched

and defined physically—linear elements such as structure and walls—and subsequently broaches the spatial, that which is contained within physical boundaries. The history of architecture has been largely defined by this progression from object or geometry to space. Moretti’s models inverted this convention by taking space, rather than its enclosing surface, as a starting point for analysis. On the one hand Moretti deals with the edge of the surface—its profile—and on the other he engages volume without surface in these model studies. Moretti’s notion of profile and space, as articulated in his volumetric models, raises formal and conceptual issues that refuse resolution as a single narrative or meaning. These models prefigure a radically new diagram of space that Moretti further developed in Casa “Il Girasole.” The first impression of Casa “Il Girasole” is a dynamic tension between volume and edge. The cut in the center of the front facade is the

32

Casa “Il Girasole”

10. Casa “Il Girasole,” northwest corner.

first postwar use of the aedicular motif, whereby a spatial division occurs between two solids, which nevertheless remain related across its void. Moretti’s use of the aedicule comes out of an historical tradition, from the Palladian window to Carlo Rainaldi’s Santa Maria in Campitelli. Moretti’s facade cannot be considered a pastiche of history, however, because he uses historical motifs in a new way. The aedicule divides the planar surface of the facade of Casa “Il Girasole” into two volumetric pieces which, though paired, are not identical, nor do their edges align across the void. The physicality of the facade is equally ambiguous, in that it appears to be a cleft volume when viewed frontally, but when viewed obliquely, the facade becomes attenuated at the edges, resembling a screen. The tension between the facade seen as a screen and as a volume is further developed at the corners of the facade. If the corner was a

dominant motif of the neoclassical, and if the frontal picture plane was a dominant motif of the modern, then Moretti’s Casa “Il Girasole” uses elements of each while breaking with both traditions. The corners of Casa “Il Girasole” are sites of fracture: both the front and rear facades overhang the main mass of the building as thin screens, separated from the main volume of the building. The corner is also shadowed by an undecidability as an assembly of concrete solids and voids. This develops from the idea of profile that Moretti put forward in Spazio, yet the layered character of the facade creates a different understanding of profile. Casa “Il Girasole” is no longer a building where profile can be said to define a continuity, as would be the case in classical architecture where profile and shape were one and the same thing. One of the important theoretical propositions set into play at Casa “Il Girasole” is that the profile does not equate to the shape of the building.

Casa “Il Girasole”

Another theoretical proposition resides in the problematic of the corner: Casa “Il Girasole” does not present a clearly subjective view of the object, seen perspectively as Greek space, nor does it offer a frontal view as modern Roman space. It is something other, and makes an argument of its otherness, similar to the manner in which Adolf Loos disarticulated the exterior envelope from inner volumes. For Moretti, the play of solid, void, and edge are simultaneous conditions. Thus Casa “Il Girasole” is one of the first didactic examples of the idea of the profile as breaking up the regular outline of the modernist box: the modernist envelope is confronted by its opposite in the idea of contained volume. In modern architecture’s free plan, columns were usually the same size and shape as functional grounding elements. At Casa “Il Girasole,” the columns become figured, changing shape and size as they move through the building, signaling difference. The paired volumes and paired sets of columns speak to a formal order that is different from an abstract or neutral column grid. The pairing of the columns creates a play between symmetries in two different axes while at the same time disrupting an abstract nine-square grid and a plaid grid of servant and served spaces. In this, Moretti’s plan critiques the uniformity of space in the free plan. The importance of these two forms of notation lies in the breaking down of historical continuity, which for Moretti was the Renaissance villa, the baroque palazzo, and the nineteenth-century hôtel-de-ville. This is an evolution of the idea of the whole as a consistent relationship of parts, as would be the case with any idea of type to a condition no longer described by a dominant whole. The materiality of Casa “Il Girasole” lodges another critique of modernist abstraction. Material here is used rhetorically, but not in the tradition of formal rhetoric, as material in and of itself, nor for its purely phenomenological value, as in Peter

33

11. Casa “Il Girasole,” front facade profile.

Zumthor’s use of stone or wood. Rather, material functions here as notation, articulating difference in a manner reminiscent of Loos’s turn-of-the-century Viennese interiors. Loos juxtaposed marbles, granites, woods, metals, and stuccos to articulate their iconic value as individual materials. Loos’s interiors are not about the richness of the materials but their juxtaposition. The lobby of Casa “Il Girasole” is a riot of materials—metal, stone, glass, wood—that obeys no structural or compositional logic. No dominant material system can be discerned, and there is no governing color palette. The use of material is both notational and didactic, to call attention to the possibility of material as text. Material elements refer back and forth to one another, yet they do not represent anything other than the mere fact of their existence. While this could be considered a form of neorealism in architecture, in their refusal to refer to any external systems

34

Casa “Il Girasole”

12. Casa “Il Girasole,” base of west facade.

13. Casa “Il Girasole,” entrance.

of material meaning, the materials function textually. The stonework of the base takes on a notational quality in its use of false rustication, varied patterns, and sculptural motifs. In Casa “Il Girasole,” the “rusticated” base turns out to be a play on rustication. Rustication in a Florentine palazzo follows a logic of mass: heaviest at the base and increasingly thinner at upper levels. Countering this convention, the rustication at Casa “Il Girasole” harkens back to Giulio Romano’s sixteenth-century Palazzo del Te in Mantua, whose paper-thin rustication does not look like stone and whose keystones seem to drop out of their holding positions, questioning how the stone arch is structurally supported. The state of suspension between support and collapse, between heavy and paper-thin rustication, calls the materiality of stone into question. Moretti inverts the conventions of rustication by putting heavy stones on thin stones, incorporating stony blocks within window openings, or cutting rusticated stone in chevron

patterns that deny their structural logic. The sculpted remnant of a human leg is incorporated into a window jamb as if a relic from an early classical sculpture had found its way into the fabric of Casa “Il Girasole.” This historicizing motif triggers a thought about the past, but it is not aimed at a nostalgic or adulatory remembrance. Rather these sculptural elements are archaic and anarchic, as if the arbitrariness of everyday life, as portrayed in neorealist film, informs what Banham might consider the arbitrary, whimsical, and unsystematic use of materials. The sculptural leg has no meaning and could be considered purely arbitrary, but this is an order of arbitrariness divorced from an expression of will, historicism, and expressionism. Moretti’s calibrated arbitrariness calls attention to its own condition as arbitrary in an internal referencing that is textual rather than purely meaningful. Moretti’s Casa “Il Girasole” uses historical motifs to make a critical commentary on the formal coherence of architecture. Historicizing

Casa “Il Girasole”

35

14. Casa “Il Girasole,” rusticated base of west facade.

references such as the aedicular motif of the facade and the rusticated textures of the base point toward postmodern practices, yet at Casa “Il Girasole” these belong to a wholly different order. Such conditions make Casa “Il Girasole” both formal and textual; certain formal coherences are emphasized and simultaneously displaced. In Casa “Il Girasole” Moretti does not thematize proportions, materials do not cohere into narrative, and the masses of the building remain a series of juxtaposed volumes and screens, if not random notations, which replace the formal conventions of the plan. Many of the possible readings are undercut by other readings, and therefore do not provide any synthesis. If the notion of a text posits the breakdown of a decidability leading to closure or synthesis, then the textual in architecture suggests a breakdown in the notion of the meaningful organization of a single narrative. Casa “Il Girasole” has many possible contingent readings as a textual work; it does not sustain a single, dominant view of architecture,

which may explain one reason why Moretti’s work has gone almost unnoticed in the intervening years. Moretti’s Casa “Il Girasole” rewrites the conditions that suggest architecture itself, and which this book argues, relate canonic buildings to close reading. While Moretti’s building transitions from the abstractions of modernism to a sensibility more closely related to neorealism, it proposes methods of close reading of a different kind, methods no longer tied to modernism’s formal lexicon but rather to an undecidability of the text. Casa “Il Girasole” is the first and perhaps the earliest exemplar of such a discourse.

36

Casa “Il Girasole”

a Vi Vi a

le

Br

li

un

o

el

ar

Bu

ap

oz

hi

zi

Sc

15. Casa “Il Girasole” in Rome sits on a nearly rectangular block bounded by two major streets, Viale Bruno Buozzi to the south and Via Schiaparelli to the west. While the front facade is orthogonal to Viale Bruno Buozzi, the rear facade of the building is parallel to its street, thus deviating at a slight angle from the front facade. Other disruptions of symmetry that occur in the building include the central north-south axis, which is not a continuous axis and bends at the stairs.

16. The mass of the building is cut in two through most of its center, essentially creating a U-shaped building condition. The central void creates the initial appearance of an axial symmetry running through the building, but the implied symmetry is belied by the actual configuration of the side blocks, which are not parallel to each other. Rather, the volumetric sidepieces are splayed from the central axis of the building. In addition to marking this destabilized symmetry, the void registers as a vertical cut in the facade.

Casa “Il Girasole”

17. The massing of Casa “Il Girasole” alludes to certain classical ideas: its tripartite organization comprises a seemingly rusticated base; a middle portion that is accentuated in the facade as a glazed zone; and an upper zone that resembles a pediment crowning the upper portions of the building. The pediment is divided by a central cut that recalls a classical aedicule. The broken pediment is asymmetrical in that the right piece rises slightly higher than its corresponding segment on the left.

37

18. The vertical division in the facade, as well as the facade’s extension beyond the body of the building, produces a profile. The vertical cut creates the idea that the facade is volumetric, revealing the corner and inboard edge at its center. Yet at the outer edges of the facade, this presumed mass becomes an attenuated screen. On the upper three residential floors, the building’s two long sides are fractured by three minor cuts. The building thus presents a series of conditions which literally and conceptually cut into the modernist box.

38

19. The analysis of the ground plan reveals that the front and rear facades extend beyond the building base. Both facades are screenlike, but the front facade resembles a screen cleaved in two, while the rear facade hangs off an intermediate boxlike volume. Immediately apparent in the ground-floor plan are the two curved walls, which disrupt the axis of symmetry and appear to displace the staircase.

Casa “Il Girasole”

20. In Casa “Il Girasole” profile no longer defines a continuity; this contrasts with classical architecture, where profile and shape were conceptualized as one and the same thing. Here profile and shape are disjuncted from one another; that is, the profile is not the shape of the building.

Casa “Il Girasole”

21. The facade of Casa “Il Girasole” breaks down the unity of the modernist frontal plane into a series of compressed layers. The complex articulation of these layers is apparent at the corners, which are no longer legible as singular entities. An oblique view demonstrates that the facade is not just a thin plane but rather is composed of three layers: a screen as the outermost layer, a void slot between the screens, and a glazing layer.

39

The void between the screen and the building mass articulates the edge of the facade as a distinct element, and creates what could be considered a gasket space especially apparent in the side views of the building. This layering, along with the deep cut in the front facade, further erodes the physical presence of these layers, since they fluctuate between two volumes and a series of layered planes.

40

22. For analytical purposes, it is necessary to examine the columnar organization. Columns are numbered 1 to 4, from left to right, and A through K from front to back. Column line 1 initially appears reciprocal to column line 4, and column line 2 reciprocal to column line 3. This sets up an initial symmetry. However, column lines 3 and 4 relate to each other because they are skewed at the same angle from the orthogonal, while column lines 1 and 2 are related because they remain on the orthogonal. In column lines 2 and 3, the A column is a slab column. Columns 2B and 3B are also slab columns that on three sides still read much as columns. Columns 2C and 3C are different: 2C is a square column; 3C is a freestanding slab.

Casa “Il Girasole”

23. Other pairings involve columns in line 1 and line 4: columns 1A/1B and 4A/4B are thin rectangles. Columns 1C/1D and 4C/4D are square columns, which are slightly smaller in column line 4. In both cases they are attached in a way that makes them seem to bleed into an external wall poché. Column lines 1E/1F (4E/4F) and 1G/1H (4G/4H) consist of paired rectangles, which alternately extend out into wall poché or bend into a splayed exterior plane. Columns 2D and 3D, 2E-F and 3E-F, and 2G-H and 3G-H are each small square paired columns, except for the additional column beside 2D. In 2J and 3J there remains the slight trace of a column, provided by a slight articulation in what is otherwise a seemingly solid wall.

Casa “Il Girasole”

24. An organization of paired columns occurs from the front to the back. This begins with the freestanding columns 1A and 4A. Columns 3A and 3B begin as a pair with 2A and 2B as orthogonal and freestanding. There is no longer an orthogonal alignment between 2A and 3A. Rather, 3A is slipped toward the right while remaining the same distance from both exterior faces. Further pairings occur among square columns. In modern architecture’s free plan, columns were usually the same size and shape; they were ground elements. Here the columns have become figural, changing shape and size as they move through the building, signaling their internal differences.

41

25. The paired columns can be read as reinforcing the rhythmic progressions from the wider column groupings in A and B at the front of the building to the more tightly paired groupings at the rear of the building. While this progression can be read in plan, it has little to do with the organization of the functional spaces. As evidenced in the ground-floor plan, column line 3 is where much of the wracking, splaying, and distorting is concentrated. This column line serves not so much as a reading datum as a receiving datum, not so much the static place where vectors originate as the dynamic place where vectors are recorded.

42

26. Ground-floor vector analyses. An analysis of the interior volumes following the column subdivisions allows one to track several vectors. An erasing arc or force (A) seems to push against the mass defined by column lines 3 and 4 until only column 3C remains, but in a flattened and distorted state. This erasing arc (A) is joined by the partial S-curve of a second curved surface (B), which is also dislocated from its former linear position. This conjunction of forces creates a figure that seems to have been compressed to the rear and expanded outward to the center. The bulging part of the figure seems to affect the alignment of the main staircase with the central axis.

Casa “Il Girasole”

These forces suggest two different ideas of form: one as the product of a vector coming from the inside and causing a convex form; the other as produced by a vector originating outside of the space, which carves away the solid to create a convex form. Space is simultaneously positive and negative. The two curves play against one another, as the result of these forces. This is purposeful, typical of Moretti’s articulation of the active nature of space as carved away or compressed by a solid. The play between the carved out and projecting space can be seen as two opposing ideas embodied in the same form.

Casa “Il Girasole”

27. The organization of columns, alternately paired and single, creates an ABABA rhythm that suggests a compression at the back of the building and a sense of extension at the front. The columnar relationships are both partial orders and symmetries.

43

The pairing of the columns also creates a play between two abstract nine-square grids and a plaid grid of servant and served spaces. Moretti’s idea was clearly a critique of the free plan, where space was uniform.

44

Casa “Il Girasole”

a.

c.

b.

d.

28 a-d. Certain conditions on the south or front facade on Viale Bruno Buozzi complicate a more traditional reading. The facade (a) can be read as a classical, vertically tripartite, rusticated base, fenestrated body, and solid cornice. However, once this general type is accepted, deviations can be seen, for example, in the facade (b) in which the middle zone actually sits on steel columns rather than on the base. There is an articulated slot between the base and the main body. Moretti exposes the actual structural elements between the rustication and the underside of the floor (c).

Traditional rustication in a Florentine palazzo obeys a structural logic: heavy at the base, with increasingly refined rustication in the higher floors. Moretti confounds these conventions by placing heavy stones on thin stones, and by adopting a vertical chevron pattern for the implied rustication (d). This chevron pattern indicates that the rustication is not structural, but iconic. The stone base is rhetorical: it is not a Greek plinth, which implies a datum, nor is it in the modern idiom of piloti.

Casa “Il Girasole”

45

a.

a.

b.

b.

29 a-b. The side elevation on Via Schiaparelli complicates the readings already established on the front elevation. First, the heavy rustication continues around the corner, again marking the line of the structural columns behind. The same paper-thin chevronlike stone pattern appears, echoing the patterning on the right front base element. Second, the columns are again revealed, this time in the horizontal slot that runs across the top of the facade. Moreover, the alignment of windows is partially determined by the implied line of columns running behind the screenlike plane of the facade.

30 a-b. The various types of rustication, both smooth and rough, at Casa “Il Girasole” deny a structural role for one that is notational. The diagonals of the chevron-shaped rustication reappear in the geometry of several textured blocks (a). The windows in the back facade register the cut of the front facade, and seem to compress the space toward the center (b).

46

31. Casa “Il Girasole,” second floor, axonometric view. .

Casa “Il Girasole”

Casa “Il Girasole”

32. Casa “Il Girasole,” fourth floor, axonometric view.

47

48

33. Casa “Il Girasole,” axonometric view.

Casa “Il Girasole”

1. Ludwig Mies van der Rohe, Farnsworth House. Plano, Illinois, 1951.

Mies Layout 02 08.indd 50

4/9/2008 10:52:47 AM

2. The Umbrella Diagram Ludwig Mies van der Rohe, Farnsworth House, 1946–51 According to Philip Johnson, Ludwig Mies van der Rohe lived by his aphorism “Less is more.” Some years later, Robert Venturi, in a reply to Mies, said, “Less is a bore.” While Venturi meant this as a pejorative comment, it resonates differently when read through Roland Barthes’s citation of “the boring” as a locus of resistance; the boring was a way to stand against the rampant consumption of art by a postwar consumerist culture. Mies’s “less is more” is a key statement for architecture; it makes its first appearance in the Farnsworth House, where less is more in the sense that this is not an architecture of modernist abstraction, but one which provokes another kind of close reading. Of all the works in this book, the Farnsworth House is the most abstract, while seemingly retaining a modernist vocabulary and conception of space. But a close reading of the Farnsworth House reveals important deviations from the modernist conventions of the open plan and the expression of structure. Together these point toward what could be considered Mies van der Rohe’s first diagram. All houses are traditionally thought of as a unity. The Farnsworth House is a tour de force that denies this idea. From its detached and oversized entry “portico” to its pervasive yet disrupted symmetries, the Farnsworth House marks one of the beginnings of the breakdown of the classical part-to-whole unity of the house. While for the early modernists the house was often a place for the study of radical innovation, from Le Corbusier’s two canonical diagrams—the Maison Dom-ino and the Maison Citrohan—to Gerrit Rietveld’s De Stijl Schroeder House, these were still single, definable entities. The early houses of Mies were no exceptions to this attitude.

Mies Layout 02 08.indd 51

4/9/2008 10:53:01 AM

52

Farnsworth House

2. Farnsworth House, north elevation, 1946.

From his early Brick and Concrete Country Houses to the Lange House, Mies worked out many of his later large-scale projects at a residential scale. But the Farnsworth House disrupts this cycle; it is no longer a single, definable entity, and the little-mentioned detached entry platform produces the most poignant clue to this idea. Mies’s rejection of the part-to-whole unity is more subtle than Walter Gropius’s and Marcel Breuer’s obvious bi-nuclear houses, which are conceptually two-thirds of a palazzo type. The Farnsworth House does not function as a fragment, but proposes another type of reading altogether. Mies’s idea of building—and in particular, of building a house—can be contrasted with Heidegger’s idea of dwelling as an object in a specific place. Heidegger’s notion of dwelling concerned the rootedness to a place: site specificity, the grounding of the subject, and ultimately the presentness of presence. For Mies, dwelling is an abstract series of conditions and, in the case of the Farnsworth House, the “dwelling” itself offers the opportunity to enact a critical reading of modernity. The Farnsworth House can be seen as a transition from Mies’s earlier work to his later work; it is a hinge between what modernism was in Mies and what will appear postmodern in his work. The shift with the Farnsworth House also sets up the difference between a scenographic, or

Mies Layout 02 08.indd 52

postmodernist, use of architectural elements to create a visual illusion, and the alternative use of the column and wall to provoke a critical reading of modernity. This confrontation, from what had been containers in Mies’s early abstract building denying the idea of dwelling, to containers that were no longer only abstractions, produces a diagram of a different sort—one which is metaphorically figured—initiated at the Farnsworth House. It is the interplay between column, wall, and horizontal plane that marks the evolution of Mies’s thinking, beginning with his early houses, which emphasized the formal and organizational role of the vertical wall plane. The Brick Country House, for example, used vertical walls extending and pinwheeling out from a central vortex (à la De Stijl), while the later houses of the 1930s—such as the Tugendhat House in Brno and the prototypical courtyard houses—were composed of vertical planes which no longer extended out from the main volume, but rather defined and enclosed space. The first two houses, the Brick Country House and the Concrete Country House, were both load-bearing concepts without columns. These houses were essentially walls that did not enclose volumes in a boxlike rectangle; the space is fractured by the way the walls extended out into the landscape. Following from these two

4/9/2008 10:53:01 AM

Farnsworth House

53

3. Barcelona Pavilion, 1929.

4. Tugendhat House, Brno, 1928-30.

houses, Mies’s Barcelona Pavilion introduced a new set of questions regarding the relationship of column, wall, and roof. The walls here are no longer load-bearing, rather the columns become the load-bearing elements; the enclosing elements are distinguished from the tectonic elements. The pavilion could be called an open plan, as opposed to a Raumplan or even a free plan, because the column in this space is conceived differently from Corbusian columns, which allowed for the free movement of enclosing walls. The Farnsworth House is a transitional point that moves Mies’s idea in several new directions. First, unlike Le Corbusier, Mies had no diagram until the Farnsworth House. This, it could be argued, is an important distinction between the two architects. The Farnsworth House, however, sets the groundwork for a diagram, and in this sense it functions as an incipient diagram. Second, at the Farnsworth House Mies is no longer dealing with the corner or the column in space; rather, at Farnsworth he introduces the use of outboard columns, which rethinks structure in proposing the idea of the sign of the column. Mies’s use of the column suggests a movement from the abstract to the real: the sign of the column is a real column, exposed on the outside of a real floor slab. Thus the Farnsworth House poses two questions: one, the question of the representation of structure as

opposed to structure itself; and two, the disassociation of the column from its use as a spatial integer. The Farnsworth House is the first of Mies’s many projects to follow that questions the truth of what is seen as structure. Such use of the column can be related to Alberti’s critique of Vitruvius, which Alberti articulated in his De Re Aedificatoria (Ten Books on Architecture), regarding Vitruvius’s three basic principles of architecture: commodity, firmness, and delight. Commodity was usefulness, firmness was structural utility, and delight was beauty. Alberti said that all architecture is firmitas because all architecture must stand up, and suggests that Vitruvius was stressing firmitas not in reference to standing up, but in reference to the appearance of standing up—in other words, as the sign of structure. Thus a column or a wall has two functions: it stands up, and it represents the idea of standing up. The three categories of signs proposed by C.S. Peirce are useful in characterizing Mies’s use of the column: the icon, which has a visual and formal similitude to its object; the symbol, which has a cultural and an agreed-upon conventional meaning in reference to its object; and the index, which describes a prior activity of the object. Peirce also is one of the first to use the term diagram, which for him is an icon having a visual

Mies Layout 02 08.indd 53

4/9/2008 10:53:01 AM

54

Farnsworth House

5. Le Corbusier, Maison Dom-ino, 1914.

6. Barcelona Pavillion, plan, 1929.

similitude with its object. As a sign of standing up, the column embodies a double condition: a column is an icon that looks like a column, and it also is the sign or index of being a column. In the Peircian triad of icon, symbol, and index, a column is both an icon and an index. This condition of simultaneity—the column used simultaneously as a critique and a representation of structure—disrupts a single reading and provokes both formal (as a representation of structure) and conceptual (as a critique of structure) readings. These simultaneous readings of the column informed what could be considered Mies’s incipient diagram. This diagram responds on several levels to two other preexisting diagrams in modern architecture: the Dom-ino and Citrohan diagrams proposed by Le Corbusier. The Maison Dom-ino illustrated Le Corbusier’s “Five Points” as well as instituted an idea of the possibility of a spatial continuum in the horizontal dimension. The Maison Dom-ino presents a diagram as a horizontal sandwich of space, in that the floor and the roof are conceptually equivalent integers. Mies’s architectural development is in one sense a sustained critique of the Dom-ino diagram’s notion of a horizontal continuum of space. The Farnsworth House proposes what could be considered Mies’s first diagram: the umbrella, a critical diagram dis-

tinguished from those of Le Corbusier in that it makes a conceptual distinction between the horizontal floor plane and the horizontal roof plane while at the same time denying any horizontal continuum. Mies’s evolution of the column section can also be distinguished from Le Corbusier’s use of the column. In Le Corbusier’s work, the column was a didactic mark that punctuated space in the free plan. Usually these punctuations were round, allowing space to flow freely around them. The Dom-ino diagram does not reveal much about structural intention, but expresses intentionality about the continuum of space, set up in part by the locations of columns, which are flush from the ends and set back from the sides equally, implying a cut on both ends. Le Corbusier, for the most part, used round and square columns relative to their placement. If he wanted to stress the edge, he would use a square column flush with the facade; if he set the column back from the glass plane, he would typically use a round column. Mies’s columns are set back from the wall plane in the Barcelona Pavilion, but are also cruciform in section. The cruciform column section illustrates Mies’s position between Adolf Loos’s Raumplan and Le Corbusier’s free plan: the cruciform stainless steel columns define a series of cubic volumes

Mies Layout 02 08.indd 54

4/9/2008 10:53:01 AM

Farnsworth House

55

7. Farnsworth House, sketch plan.

in articulating the corners of each spatial unit. The chrome plating on the columns serves as a mirror, inverting conventional square columns: that which is typically solid—the actual corner of a space defined by a real column—becomes a mirror or a reflection of the space and thus becomes a void. The real column in some sense becomes a virtual column, even while it continues to define a spatial unit. For Mies, columns define and circumscribe spatial units; for Le Corbusier, columns allow space to pivot and act as a fulcrum rather than as corners. For Mies, the column and the corner become one didactic model, from the Barcelona Pavilion to the buildings at the Illinois Institute of Technology (IIT). The position in space and the sectional properties of the column at the corner frames a conceptual discourse for Mies. Yet at the Farnsworth House, the corner would seem to be a nonthematic element: the columns are no longer at the corner, neither gridding space internally nor holding the outboard corners. Mies’s initial sketches for the Farnsworth House demonstrate his intention to use the columns in a different way, namely outboard of the floor slab. It is possible to assume that the outboard columns are more of a structural expression, that the columns are functioning as structural elements. But this is not the case. This is the first time that

Mies Layout 02 08.indd 55

Mies places the columns outboard. He seems to suspend the roof between the columns, suggesting that another strategy is intended—one which occurs in many of the buildings that follow. At the Farnsworth House, the horizontal floor slab and roof are framed between the columns, so the columns are no longer supporting the roof, but rather the roof and floor are slung like hammocks between the columns. Mies’s postwar work represents a transition from the column as either load-bearing or marking a spatial quadrant to a condition where the column is the support of a suspension structure, in which the horizontal members are hung from the outboard structural columns and the overhead roof beams. This will lead to a subsequent development, in which the beams are articulated above and the roof hung from these beams, giving rise to what will become the Miesian umbrella diagram. The metaphorical umbrella is a diagram in which the roof and its appended columns seem to be hovering above a podium base. The Farnsworth House is the first realization of this umbrella diagram. The Farnsworth House is also perhaps the most didactic critique of the column and the wall as merely structural elements. This building has often been misread as an articulation of the principles of Le Corbusier because of its seem-

4/9/2008 10:53:02 AM

56

Farnsworth House

8. Crown Hall, IIT Chicago, 1950–56.

9. Alumni Memorial Hall, IIT Chicago, 1947.

ing evolution of the Dom-ino diagram, or as the progenitor of Philip Johnson’s Glass House. In either case, these attributions are problematic, if not superficial. The Farnsworth House is not about trabeation, but rather engages the look of trabeation in bringing the columns outboard and suspending the floor and roof slabs between the columns. The idea of the sign of structure at the Farnsworth House is also a precursor of the column-over-column detail at the Seagram Building and also at IIT, where Mies adds I-sections and H-sections at the corner and on the facade to mask the actual structure. This thematic in Mies’s postwar work engages structure that is the sign of structure; what is seen is not the actual column, but a mask of the structure. The Farnsworth House initiates this argumentation: when the column is placed outboard of the slabs, it still acts as a column, but not as straightforwardly, as in the case of directly countering vertical load. Because the slabs are being held up through the suspension of plates coming off the column, this allows the box-frame of the house to straddle and be suspended between the columnar

structure. This is a radical idea for the late 1940s and early 1950s; it is a radical idea for Mies, and breaks with his use of the idea of the column as a clear indication of tectonics. Instead the column reads as both structure and the sign of its diagrammatic condition. The plinth and the horizontal roof plane are again conceptually different in Mies’s space as opposed to Le Corbusier’s space. Whereas Le Corbusier’s ground plane is separated from the ground conceptually and floats, like the roof plane, Mies’s ground plane is tied to the ground while the roof floats free. If there are precedents for the differentiation of space between ground and roof at the Farnsworth House, one would include the Resor House of 1937−8. The model of the Resor House is the first indication of a new attitude in Mies’s work. The house seems to float above the ground, though it actually spans a ravine and is anchored at both ends. The house itself is a virtual podium that reappears in the Farnsworth House, with its suspension a few feet off the ground. This lifting of the house has a different value than Le Corbusier’s Dom-ino diagram. For Le Corbusier it signals the infinite horizontal extension of

Mies Layout 02 08.indd 56

4/9/2008 10:53:03 AM

Farnsworth House

57

10. Farnsworth House, plan 1946-51.

11. 50 by 50 House, plan, 1950-1951.

space; for Mies, it sets out the ultimate distinction between the ground and the roof, leading eventually to the umbrella diagram. The Farnsworth House also had important implications for the 50 by 50 House that immediately followed in 1951. First, at the 50 by 50 House, there are only four outboard columns, which appear at the centers of the sides of the square, providing the building with a clear rotational quality while framing the corners on the diagonal. Second, the ground-floor plane is no longer articulated; the glass box sits on what seems to be a natural plinth, which is clearly distinguished from the pristine white line of the roof. Together, the roof line and single columns produce an image of an umbrella-like structure. What follows, less literally but no less conceptually, are the Mannheim Theater project, Crown and Alumni Memorial Halls at IIT. In these projects, the columns are brought outboard, not so much to show them holding the roof up but rather to show them as representing another kind of spatial attitude articulated in Mies’s umbrella diagram. The 50 by 50 House also manifests the transition to the exposed steel truss running

above the roof line at IIT’s Crown Hall and at Mannheim, suspending the roof like a giant parachute. Clearly the National Gallery in Berlin is the last, and perhaps most subtle, in the line of progeny from the Farnsworth House. Given the gallery’s stone base and projecting roof line flush with the exterior column line, the umbrella effect is finally presented as concept and not image. In both the National Gallery and IIT, the idea of dwelling, or use, is clearly not what is at stake, since Mies sinks the primary functions below ground. Their envelopes function as an icon of a building that will be used as an architectural school or as a museum. On entering Crown Hall at IIT, for example, one notices little on the iconic plane that involves its use as an architectural school: all of the offices and studios, whether they need light or not, are placed below the plinth. Similarly, there is little at the plinth level at the National Gallery that represents its use as a museum. At the Farnsworth House, with Mies’s careful manipulation and placement of the forms, it becomes clear that a scenographic condition between the viewer and the building is not what

Mies Layout 02 08.indd 57

4/9/2008 10:53:04 AM

58

Farnsworth House

12. Neue Nationalgalerie, Berlin, 1966.

is desired. Rather, a reading of the relationships between column, floor, plinth, and roof other than as a series of modernist abstractions gives this building its critical dimension. To further this idea, Mies establishes shifting axes of symmetry deployed among the three disparate entities of the Farnsworth House: the entry platform, the house platform, and the glass box. While the stairs of the entry platform are aligned with the stairs to the house, the intermediate platform itself is slipped off this potential axis. Similarly, the glass enclosure is asymmetrically placed in relation to the floor slab, yet symmetrically placed with respect to the centerline of the column grid. This sliding or oscillating movement between the glass, mullion, columns, floor slab, and plinth produces

Mies Layout 02 08.indd 58

a complex and dense relationship of elements that, while they appear scenographic, produce a critique of any single reading. The different axes formed by a series of symmetrical parts indicate that the parts do not create a whole. What seems to be a classical and symmetrical whole is rather broken down into asymmetrical dynamic parts. Mies’s play against classical symmetries continues with his treatment of the glass surfaces. At the Farnsworth House, the glass is dematerialized and there are no horizontals to articulate a wall plane. The outboard columns of the Farnsworth House do not go above the line of the finished roof but just up to it, articulating the difference between structure and finished surface. This can be contrasted with Johnson’s Glass House, which,

4/9/2008 10:53:04 AM

Farnsworth House

59

whether trying to break away from a Miesian or a Corbusian space, basically defines a classical vertical surface with the marking of a chair rail in the vertical plane. There is no chair rail in the vertical plane at the Farnsworth House. Johnson, both for the sake of his furniture but also to differentiate his ideas, is interested in the glass as a plane or membrane, as opposed to Mies’s interest in glass as a void. Johnson’s intent is to render the surface as a vertical plane, while at the Farnsworth House, Mies renders it as an absence. The shift at the Farnsworth House registers the difference between a scenographic representation linked to postmodernism and the use of the column as a critical reading of modernity’s idea of a spatial continuum. The Farnsworth House stages this confrontation between what in Mies’s early building was used to deny the image of dwelling, and the Farnsworth House, where the elements are no longer abstractions. The organization of column, walls, and slabs become real but no less critical counters in the design, which proposes an implied real structure against the sign of structure; columns read not for their tectonic truthfulness, or for their visual composition, but for their condition as a sign of a conceptual diagram.

Mies Layout 02 08.indd 59

4/9/2008 10:53:05 AM

60

13. The organization of the column grid at the Farnsworth House and its entry platform produces an AAAA four-bay sequence. Each bay is equal in size. The edge of the enclosing glass wall falls symmetrically between the right two bays, extending a half-module beyond each column line.

Mies Layout 02 08.indd 60

Farnsworth House

14. Each slab is also positioned symmetrically within the column grid, yet the center of the main slab is not aligned with the column grid. The columns supporting the entry platform are aligned with the column grid of the house, yet the platform itself is slid off the axis established by the floor plane.

4/9/2008 10:53:05 AM

Farnsworth House

15. The glass enclosure is symmetrically placed within the framework of the columns. The center of the glass enclosure is aligned with the column line. The glass enclosure, therefore, produces a second center located along the middle column line, while the center of the floor plane lies along the mullion line. This establishes a tension between the glass enclosure’s center and that of the floor plane.

Mies Layout 02 08.indd 61

61

16. The play of the two centers is an aspect of the dynamic of the Farnsworth House. The location of the glass enclosure, in coming to the edge of one end of the floor slab but not to the other, while seemingly assymetrical, defines a symmetry about the middle column. These two symmetries also define closure, and anchor what would seem to be the potential extendability of the building.

4/9/2008 10:53:06 AM

62

17. The Farnsworth House does not present a horizontal sandwich of space in the manner of Le Corbusier’s Dom-ino diagram. The lower slab is raised off the ground on stub columns, but it does not echo the precepts of the Dom-ino diagram because the floor slab

Mies Layout 02 08.indd 62

Farnsworth House

is connected to the ground by the entry platform. The movement from the ground level up the stairs to the entry platform and up to entrance on the main floor is perpendicular to the grain of the house.

4/9/2008 10:53:06 AM

Farnsworth House

63

a.

b.

18 (a-b). The progression of movement at the Farnsworth House is significant, emphasizing Mies’s transformation of vertical and horizontal surfaces in response to Le Corbusier’s diagrams.

At the Farnsworth House, the perpendicular movement resembles that of the subject entering the Maison Dom-ino (a), as opposed to that of the Maison Citrohan (b), which is parallel to the movement of entry.

Mies Layout 02 08.indd 63

4/9/2008 10:53:06 AM

64

19. The traditional relationship of post to beam is confounded by the outboard columns, with no obvious fastening system, but rather a carefully detailed set of discreet connections, holds the horizontal slabs in place.

Mies Layout 02 08.indd 64

Farnsworth House

20. The attachment of the mullions to the corner produces what could be considered a positive inboard corner, thereby inverting the conventional form of the corner.

4/9/2008 10:53:06 AM

Farnsworth House

21. Mies’s corner reads as two entities fused together, with the trace of their joining still legible: the mullions are compressed together to articulate the corner.

Mies Layout 02 08.indd 65

65

22. The two sets of mullions at the corner produce an outboard L-shaped condition, creating a void at the outboard corner.

4/9/2008 10:53:07 AM

66

Farnsworth House

a.

b.

c.

Mies Layout 02 08.indd 66

4/9/2008 10:53:07 AM

Farnsworth House

67

d.

e.

23 (a-e). The existing column arrangement at the Farnsworth House (a). Possible alternate arrangements include columns aligned with the inner edge

Mies Layout 02 08.indd 67

of the fascia plane (b and c); columns organized similarly to those of the Dom-ino diagram (d); and doubled columns at the corners of the slab (e).

4/9/2008 10:53:08 AM

68

Farnsworth House

24. Farnsworth House, exploded axonometric of slabs and columns.

Mies Layout 02 08.indd 68

4/9/2008 10:53:08 AM

Farnsworth House

69

25. Farnsworth House, exploded axonometric view.

Mies Layout 02 08.indd 69

4/9/2008 10:53:08 AM

70

Farnsworth House

26. Farnsworth House, axonometric view.

Mies Layout 02 08.indd 70

4/9/2008 10:53:08 AM

Mies Layout 02 08.indd 71

4/9/2008 10:53:09 AM

1. Louis I. Kahn, Adler House. Philadelphia, Pennsylvania, 1954–1955.

4. From Plaid Grid to Diachronic Space Louis I. Kahn, Adler & DeVore Houses, 1954–55 In an essay on Marcel Proust’s Remembrance of Things Past, Maurice Blanchot raises the question of narrative time and its disruption. While there is a chronological time in narrative, Proust interweaves another form of time which Blanchot describes as another possibility of time brought back not as a memory but as an actual event. Blanchot quotes from Proust: “The footsteps that stumble on the irregular cobblestones of the Guermantes Way are suddenly the same footsteps that stumbled over the uneven flagstones of the Piazza of San Marco.” These footsteps are not just a double, or an echo of a past traverse. They evoke another sensation, one which does not take the form of a synchronic linear memory, but becomes a diachronic, nonlinear, and simultaneous experience. According to Blanchot, the Venice and Guermantes moments should not be considered separately, as a past and a present, but as a single presence that harbors a sense of absence. The incompatibility of these two moments creates a sense of simultaneity, which Blanchot suggests is a sensation that suspends and neutralizes even narrative time itself. The simultaneity, according to Blanchot, comprises the “then” of the past and the “here” of the present. These times resemble two instances of the “now,” superposed in the conjunction of two simultaneous presents which alter time in a narrative sense. This diachronic time disrupts the traditional synchronic condition of both a linear time of reading and the linear time of the story. In literature, the time of the action of reading and the time of the narrative are not the same. Yet unlike literature, architecture is thought to presume a single time: the experience of the building and the conceptualizing of the building are understood as one and the same.

104

Adler & DeVore Houses

2. Adler and DeVore Houses, elevations, 1954.

A building unfolds in a linear manner as a person walks in and around the space to come to understand the building. The time of “reading” is different for a reader of literature than it is for a reader of architecture. Time is only imagined in the space of literature, but in architecture it is actually experienced in space, and because the time of experience in architecture is linear, architecture is associated with synchronic time. Blanchot suggests that the disruption elicited by a narrative’s simultaneous moments represents such a diachronic moment in actual space. The question for architecture involves eliciting that disruptive moment or diachronic time: in other words, considering Blanchot’s reading of Proust in architectural terms, can architecture, like literature, propose affective moments in which the viewer is suddenly freed of the ultimate movement of time toward death, where one can experience some other kind of time, a more pure state that exists somewhere between the viewing subject and the object itself?

The narrative of time that exists in a building will always be constructed in real space, which is experienced as a narrative space—that is, by walking in and around a building. One way that architecture can suspend the time of the narrative is by superposing on it another time. While this may seem an appropriate concept for an architecture of close reading, such an idea has rarely been considered in the work of Louis Kahn. In the Adler and DeVore Houses of 1954– 55, unlike in many of his other projects, Kahn achieves what could be considered an architectural text in diachronic space. This is brought about by the superposition of classical and modern space; that neither of these “times” dominates results in a dislocation of moments or, in other terms, a disjunction that is experienced in space. In the Adler and DeVore Houses, Kahn presents architecture both as a complex object and as the potential for the subject to experience the object as both a real space and an imaginary space. Both conditions are present and can be read, each in

Adler & DeVore Houses

105

3. Trenton Bathhouse, New Jersey, 1954–59.

4. Trenton Bathhouse, axonometric view.

turn displacing the other. It is this unresolved moment in the Adler and DeVore Houses, which are themselves suspended in real time between the Trenton Bathhouse and the Richards Medical Center, that makes these two houses different from much of Kahn’s other work. It is in the context of the denial of axial symmetries and partto-whole relationships evident in much of Kahn’s later work that these differences lie. Thus the Adler and DeVore Houses can be seen to articulate an alternate internal logic: first, as a conscious, didactic proposal against the free plan of modern architecture, and second, as a critique of modern architecture. It is significant that the drawings for the Adler and DeVore Houses were published in Perspecta 3 (1955) in a short article titled “Two Houses,” which stressed the underlying geometric order of both projects. Each was conceived as a cluster of squares, according to Kahn, and each is represented with an emphasis on its columns, as if the houses were essentially an abstract

pattern of square columns within square enclosures. Kahn suggests that the houses “grow out of the same order,” but that their designs are different. These statements by Kahn imply an origin as a unity and a sameness, which belies the disjunctive conditions lodged in these two projects. Kahn actually produces two diagrams in each of these houses, one a classical, tripartite, nine-square diagram and the other a modernist asymmetrical diagram. The dual diagram counters the idea of a singular origin, just as these superposed organizations deny a beginning in a particular historical moment. The denial of a single and identifiable point of origin also begins to critique the notion of the classical part-towhole relationship in its denial of a single unified whole. A series of potential points of origin suggests the undecidability of relationships between parts, which therefore no longer can be subsumed within a clearly definable whole. A variety of historical moments can be discerned in Kahn’s plans for these houses. The

106

Adler & DeVore Houses

5. Adler House, preliminary plan.

6. John Hejduk, Texas House 4, 1954–63.

European influence of the late 1940s and early 1950s is evident; Le Corbusier’s Maison Jaoul, for example, is one such possible model, as a brick, wall-bearing structure whose barrel vault has dispensed with the flat roof of modernism. The hipped roofs of Kahn’s Trenton Bathhouse, as well as its emphatic materiality, are clearly antecedents to the Adler and DeVore Houses, given that initial sketches of both houses similarly have hipped roofs. The Trenton Bathhouse is the first example in America of a massive brick and concrete structure denying the free plan and dynamic asymmetries of modernism with a classicizing nine-square plan. While only a small portion of the Trenton Bathhouse was built, its master plan was radical in deploying a plaid grid, a BeauxArts grid of servant and served spaces with an ABABAB rhythm in its overall organization, rather than the homogenous space of the modernist grid. Kahn uses the particular alignment of the column within its masonry enclosure to differentiate the variegated bays that constitute the plaid grid. The structure is articulated with a certain redundancy because the large masonry units,

despite their appearance, are not load-bearing; the actual steel structure is just visible between the masonry and the roof. This is a self-referencing notation of the disjunction between the section of the roof and that of the plan; in other words, an articulated system in section, which evolves out of an extrusion in plan. There is no sectional displacement. If in Le Corbusier the section is often the site of the displacement of the subject, Kahn’s Trenton Bathhouse produces no such disturbance in section, which, it could be argued, fits within the pragmatic tradition and utilitarian organization of space in American architecture. Materials are expressed in the Trenton project, which eschews the use of a surface veneer. The corner pier structures and wall are made of concrete blocks. The wall plane of the corner pier structures is made of the same material as the corner, and contributes to the sense that the pavilions appear the same from the corner as from a frontal view, denying the picture or frontal plane. The bathhouse pavilions are neither Greek (conceptualized from a perspectival view) nor Roman (conceptualized from a fron-

Adler & DeVore Houses

107

7. DeVore House, plan.

8. John Hejduk, Wall House, 1968–74.

tal plane); here the point of view of the subject becomes irrelevant. This disruption of specific points of view in the Trenton Bathhouse’s ninesquare plan foregrounds the destabilization that becomes manifest in the plans of the Adler and DeVore Houses. The Adler House stands as a critique of the bi-nuclear houses of Walter Gropius and Marcel Breuer. In these bi-nuclear schemes, one enters in the middle space between two pavilions; on one side is the public space and on the other is the private space. These modernist bi-nuclear houses were essentially a misreading of classical architecture, in that in a classical parti, the void between the two pavilions would never have been entered. Kahn, who understood this, maintains vestiges of a bi-nuclear notation in the Adler House, which is a combination of a series of pavilions along the lines of the Trenton Bathhouse with a fracturing that does not occur in the work of Gropius or Breuer. A first sketch with nine-square and axial symmetry clearly has echoes of the Trenton Bathhouse. Later sketches demonstrate the fracturing of this organization,

producing what seems to be the fragmentation of a former nine-square grid. Yet attempts to fit the pavilion units back into a unified organization such as a nine-square are frustrated and elude any stable originary part-to-whole relationship. The Adler House maintains both of these ideas: the whole as the sum of its parts, and the impossibility of the whole; the whole is made impossible by the different shearings and slippages resulting from the superposition of a modernist plan and a classical nine-square parti. That neither plan is made dominant recalls the disruption generated by the diachronic idea of time in Blanchot’s reading of Proust. The plan of the Adler House registers this superposition of the modernist and Beaux-Arts schemas, which implies a transformation not from a single original state but from several possible originary conditions. This transformation leaves traces that can be read in the resulting plan. The movement in the Adler House’s square units produces a shearing motion and introduces two concepts: the idea of a grain to the space and the idea of time in its process. Yet while the square

108

Adler & DeVore Houses

9. Adler House, plans.

units of the plan themselves have no directionality, their varied motions away from possible points of origin always occur along a horizontal axis. The square columns and their groupings to form rectangles imply a grain and directionality to the implied movement of the pavilion units. In the Adler House, the horizontal motion of the units remains discontinuous, as if several dislocations from a seeming origin have occurred over time, even though that origin in itself cannot be fixed. The overall arrangement of square units comprising the Adler House resembles an organization of pavilion units intermittently sliding off the nine-square grid, yet their asymmetrical placement confirms a modern spatial arrangement. The interior grid of the house reverts to a Beaux-Arts plaid grid, and thereby decenters the nine-square grid parti. This is one of many of the embedded oppositions at work in the Adler House. If the house at first seems to be haphaz-

ard or an arbitrary organization of pavilions, this is not the case. Through a purposeful manipulation, Kahn produces the dislocations that articulate a text of diachronic spaces in the architecture of the Adler House. The DeVore House similarly resembles the record of a process that has been frozen at a moment in time; it alludes to a possible origin but frustrates any direct reading of such origins. This can be best understood in comparison to both John Hejduk’s Wall Houses and his Texas Houses, with which the Adler and DeVore Houses are contemporaries, though which came first is of little relevance in this context. Their similarities reflect a shared set of ideas also present in the early drawings for the Vanna Venturi House (chapter 5). Like the Adler House, Hejduk’s Texas Houses use a classical nine-square parti as their basis. Alternatively, Hejduk’s Wall Houses and Kahn’s DeVore House focus on the relationship of the

Adler & DeVore Houses

109

10. DeVore House, partial plan.

pavilion to the wall, which in and of itself is thematized as a didactic element to which the pavilions respond. In the DeVore House, the placement of each of its pavilion units seems to respond to the wall, and the shape of its columns further implies a directionality in relation to the wall. In both Hejduk’s Wall Houses and Kahn’s DeVore House, the wall performs as a divider, separating public from private and bucolic from built; it also becomes a threshold marking the moment of crossing. The wall simultaneously differentiates and makes intelligible; it separates but also links. Despite its obvious materiality, it embodies several seemingly contradictory abstract principles, establishing a narrative sequence for the subject passing through the house: while the subject is aware of a continual breach of that threshold, this awareness focuses the spatio-temporal emphasis on the moment of crossing. Thus, dividing different times and spaces, the datum of the wall

assumes a metaphysical presence as the central element in the formulation of the perceptual and navigational intelligibility of the design. In drawing attention to this moment, Hejduk and Kahn set up a tension in the architecture that questions the common understanding of the dialectical difference between inside and outside. Kahn’s DeVore House also fuses two different geometries and two different moments in history, the classical and the modern. What can be more evocative of the modern than a wall knifing through the heart of the house, which is not a modern house, but one that has classical echoes in its nine-square grid? If a wall is usually read as a divider between an interior and an exterior, once a person passes through the wall at DeVore, that conditioning has been problematized: has one left the structure or entered it? How space is experienced in an interval of time is part of how the time of the object is usually revealed to the

110

11. Richards Medical Center, 1957–65.

subject. Because the subject is continually operating around the wall and making reference to it, the only “inside” is that point in the interior of the wall itself. Everywhere else is outside of the wall, outside of its inside, but constantly aware of that moment of inside. A “time of inside” is thereby established in relation to the “time of outside.” The actual “time of enclosure” then becomes infinitesimally small relative to the continuum of which it is a part. Thus, one of the main characteristics of both of these works is the collision of time narratives within the boundary of the wall itself. The Adler and DeVore Houses can be considered an inflection point in Kahn’s work, for they stand both outside of the traditional interpretations of his work and become the starting

Adler & DeVore Houses

point for future work. From these two houses it is a short jump in scale to the Richards Medical Center, 1957–65, which is a project made up of a sequence of pavilions, extruded and positioned as if unwinding out of a tight spiral. Richards is also a pavilion project of servant and served elements extruded into the third dimension. If the pavilions of the Adler and DeVore Houses functioned as units marking tactical shifts across the physical threshold of a wall or the implied/conceptual threshold of a nine-square grid, then the pavilions of the Richards Medical Center are made to serve a picturesque rather than didactic function. Kahn’s sketches of the Richards Medical Center transform and extrude the pavilions to produce a romantic skyline. Each of the three main towers is a volume articulated by thin columns, with the servant spaces pulled off into smaller, separate towers. The H-shaped columns reiterate the nine-square organization of each volume and form a plaid grid. Unlike the Adler and DeVore Houses, the structure frames the middle tier of each side, leaving a void at the corner. This voided condition recalls the Exeter Library and other Kahn projects. At Richards, the two conditions contradict each other: the alignment of the pavilions sets up a frontal organization in plan, yet the entry occurs at the corner. The arrangement is both orthogonal and diagonal, a combination of Greek and Roman space that becomes a Kahn trope. The countermanding pavilion alignments, operating systematically in the Adler and DeVore Houses, become at Richards more graphic, and ultimately expressionistic. The legibility of the Richards servant and served spaces marks another shift from the undecidability of the pavilions in the houses. Kahn participates in the pre-1968 attempt to rethink the originary premises of modernism that characterize works belonging to this first paradigm shift. His work here represents a split between the unconscious theoretical propositions apparent in the work of both Mies and Moretti, and the seem-

Adler & DeVore Houses

ingly conscious theoretical reversals articulated in that of Le Corbusier. The movement toward real materials as well as the reintegration of classical schemata, is an expression in each of these houses of a critique of abstraction. This critique involves a new interest in what resembles incompleteness and fragmentation of form. With today’s hindsight it is possible to suggest that the shifts, dislocations, and superpositions in the Adler and DeVore Houses ultimately could be considered a questioning of the classical part-to-whole relationship.

111

112

12. The organization of the Adler House suggests that the house has a conceptual origin in a nine-square grid, five pavilion units and five square outdoor spaces.

Adler & DeVore Houses

13. The organization of the pavilions seem to originate in the nine-square grid. While the lowest row (open space, pavilion unit, open space) can be conceptually returned to such an origin, the center row (open space, pavilion, pavilion, open space) and the upper row (pavilion, halved open space, pavilion, open space) cannot.

Adler & DeVore Houses

113

B B

A B

B A

B

B

A A

A

B

A

A

A

A

A

A

A

B

B

B

B

B

A

14. The organization of the pavilions thus enables multiple readings. The Adler House’s internal logic depends on the simultaneity of two dissimilar systems involving the open spaces (A) and the pavilion units (B). The ideal ABA nine-square is adapted to accommodate the Adler House plan by removing a pavilion and shifting them. Although the plaid ABA plan derives directly from the Beaux-Arts tradition, this reading emphasizes the house’s modernist asymmetry.

B

B

A

B

B

A

15. Equally, the modern BBA arrangement emphasizes an asymmetry of organization and can be adapted to the Adler House by removing and shifting pavilions. The overall arrangement of the pavilion units presents a superposition of Beaux-Arts and modern plans. Their combined product allows for a double reading of the house as an ABA arrangement and a BBA arrangement, in other words, as both Beaux-Arts and modern organizations.

114

16–17. There are two ways to locate the Adler House within a classical nine-square-grid; yet each is an imperfect fit, with all or a portion of a pavilion falling outside of this idealized schema. When one interpretation

Adler & DeVore Houses

is chosen, one part of the scheme is outside of the ninesquare diagram. When the outside part is the basis for the basic diagram, the other part moves outside, thus there is no stable single diagram for the Adler House.

Adler & DeVore Houses

18. In general, the piers provide points of alignment for one nine-square diagram, which responds to the single, double, and triple pier conditions.

115

19. However, an anomalous condition (highlighted in red) can be defined by the single piers of the uppermost row. While this space remains a void in the project, there is a conceptual overlapping of an implied unit bounded by piers and open space.

Adler & DeVore Houses

yax is

116

is

ax

x20. The column grid of the Adler House has a problematic pier arrangement within the nine-square grid. There are four different pier organizations which contribute to a striation of space in the Adler House: the

pier by itself, the pier doubled along the x-axis, the pier doubled on the y-axis, and the intersection of the two doubled axes, which creates an L-shaped corner pier condition.

Adler & DeVore Houses

21. The dimensions of the window mullions and piers of each pavilion unit constitute a notational system for subdividing each pavilion. This minor grid is composed of thirty-six squares.

117

22. This thirty-six square sub-grid accommodates both nine-square and four-square organizations and thereby allows a double-reading of interior spaces that is also possible for the overall organization of the pavilion units. This allows the doubled piers, for example, to remain within the definition of the overall sub-grid.

118

23. Adler House, ground floor, axonometric view.

Adler & DeVore Houses

Adler & DeVore Houses

24. Adler House, roof level, axonometric view.

119

120

a.

25 (a-b). The pavilion units of the DeVore House are organized in relation to a wall. Origins can be attributed to classical ABCBA (a) and modernist asymmetrical schemes (b).

Adler & DeVore Houses

b.

26. Four actual units (B–E) and one implied unit (A) stand on one side of the wall, while a single similar pavilion unit (F) is located on the other side of the wall. The leftmost two units (A and B) are aligned but separated by a gap that functions as an implied wall.

Adler & DeVore Houses

a.

27 (a-b). Unit (E) is separated by a space equivalent to half of the unit dimension (a) from the adjacent unit D, but is the only unit physically attached to the wall. Units B, C, and D are aligned in a parallel manner along a wall that is breached by the C and F blocks. The outermost units (A, E) are aligned with the wall. Units B and D shear from the mass created by unit C, establishing a grain perpendicular to the wall.

121

b.

28. Units A and B are separated by a narrow slot of space, while unit D is separated from the wall by a space (b) that is equivalent to half of space (a). The dimension of a pavilion unit is double that of space (a). A logic of spacing emerges in which spaces of dimension (a) interlock and frame the units in a plaid grid.

122

Adler & DeVore Houses

a.

b.

a.

b.

29 (a-b). Units A and B are separated by a space, yet it is a pochéd area (the fireplace) that maintains their physical separation (a). Rectangular columns define the relationship between units B, C, and D and establish a grain running in parallel to the wall (b).

30 (a-b). A single square column located centrally between units C and F indexes the presence of the existing wall. One of the middle columns is unaligned, but in another reading it aligns with an outboard column to produce a double column in unit C (a). Unit F is “rotated” so that its two open sides suggest a movement producing a third shearing condition.

Adler & DeVore Houses

31. A strategy of mirroring and rotation emerges in the play of units, spaces, and walls. Unit C is mirrored in unit F across the presence of the virtual wall that is established by the square column aligned with the existing wall. However, the grain produced by the orientation of rectangular columns also suggests a rotational relationship.

123

Units A and E seem to mirror each other across a line perpendicular to the existing wall and established by the square column between units C and F. The units in the DeVore House can be read as a series of shearing and rotational movements. The effect of the superposition of these multiple systems both reinforces and displaces the relationships of unit to wall to conceptual grid.

124

32. DeVore House, ground-floor plan and sublevel, axonometric view.

Adler & DeVore Houses

Adler & DeVore Houses

33. DeVore House, ground floor, axonometric view.

125

126

34. DeVore House, roof plan, axonometric view.

Adler & DeVore Houses

1. Venturi & Rauch, Vanna Venturi House VI. Chestnut Hill, Philadelphia, Pennsylvania, 1964.

Venturi Layout 02 08.indd 128

4/9/2008 11:35:04 AM

5. The Nine-Square Diagram and its Contradictions Robert Venturi, Vanna Venturi House, 1959–64 Robert Venturi’s 1966 book Complexity and Contradiction in Architecture is one of the few texts by an American architect that responded to the shifts in culture occurring in Europe around 1968. Architectural texts that marked this change included Aldo Rossi’s The Architecture of the City and Manfredo Tafuri’s Teorie e Storia dell’architettura. If seminal theoretical texts of this period, such as Jacques Derrida’s Of Grammatology, Gilles Deleuze’s Difference and Repetition, and Guy Debord’s The Society of the Spectacle, provide a further indication, then it is clear that the period around 1968 represents a generational if not a paradigm shift. The texts of this period, both within and outside of architecture, made a profound impact on culture in general, and specifically on architecture in America. These books begin to question the internal conditions of the discipline of architecture, particularly in America, which had, until 1968, been a relatively nontheoretical and professional one, focused primarily on the pragmatics of the architectural practice. Texts by Venturi, Rossi, and Tafuri questioned architecture’s capacity for social reform, a thematic of mainstream modern architecture. If Le Corbusier had stated that the plan is the generator—of the building, the city, and on some level, of modern society—then these texts initiated a profound critique of the part-to-whole relationship implicit in his vision of the plan. No longer adherents of the polemical style of Le Corbusier’s Vers Une Architecture or the equally polemical historical perspective of Sigfried Giedion’s Space Time and Architecture, on which the generation of the 1940s and 1950s was weaned, these books of 1966–68 were didactic in their reevaluation of modernist principles. No longer was the ethos of CIAM (which held that modern architecture was a vehicle toward a better society), nor its rebirth later in Team Ten, thought to have much currency.

Venturi Layout 02 08.indd 129

4/9/2008 11:35:11 AM

130

In contrast to Europe, the significant social changes occurring in America, such as the breaking down of ethnic and religious barriers after the war, did not engage architectural polemics. Many American corporations hired new firms of young architects to design what was considered to be modern architecture for such projects as public housing; the Federal Housing Authority, which allowed for returning GIs to buy apartments at low cost, financed buildings designed according to Le Corbusier’s proposals for the Ville Radieuse but stripped them of their ideological component. Rather than social reform according to Corbusian principles, American cities became blighted by versions of towers-in-the-park-schemes, which, along with the flight to the suburbs, ate away at the formerly dense fabric of the city. Such changes in the city fabric reverberated through the architectural profession. Between 1945 and 1950, new practitioners and new offices were created to accommodate America’s postwar building boom. Instead of adopting modernism’s ideology, which was directed toward creating what was called “the good society,” these new practices adopted the modernist style as a manifestation of “the good life,” an ideal almost antithetical to the social utopian goals of modernism. Yet modernism’s relatively “new” look appealed to corporations, given their production for an openly acquisitive and relatively affluent society. In this period of architecture’s capitalist expansion, engaging an ideology of the left was certainly not a pressing issue. The architectural profession grew rapidly, with corporate practices founded by numerous architects, including Edward Barnes, Gordon Bunshaft, Harry Cobb, Ulrich Franzen, John Johansen, Philip Johnson, and I.M. Pei. The next generation, Michael Graves, John Hejduk, Don Lyndon, Charles Moore, Jaquelin Robertson, Robert Venturi, and Tim Vreeland, among others, found few openings for this type of practice and turned instead to teaching

Venturi Layout 02 08.indd 130

Vanna Venturi House

2. John Hejduk, Texas House 4, 1954–63.

and writing—in some cases by choice, in others for purely practical reasons. Venturi was clearly the most articulate spokesman for this younger generation and brought what was perhaps the first theoretical and ideological approach to American architecture. Complexity and Contradiction in Architecture, which was published by the Museum of Modern Art, launched an attack on modernist abstraction by reintroducing the idea of history in contemporary architecture. Venturi, who was educated under Jean Labatut at Princeton, had continued his research at the American Academy in Rome. Perhaps because of his stay in Rome, Venturi examined Italian buildings in relationship to issues faced by contemporary American architects. For example, Luigi Moretti’s Casa “Il Girasole” in Rome used historical architectural tropes that would play a subtle role in the arguments in Complexity and Contradiction. Venturi saw himself not as a postmodernist, but rather as a new American realist, bringing historical traditions into the present by way of American architectural traditions described, for example, in Vincent Scully’s The Shingle Style and the Stick Style. Complexity

4/9/2008 11:35:11 AM

Vanna Venturi House

3. Vanna Venturi House I, plan, 1959.

and Contradiction, as the title indicates, addresses questions of meaning in architecture, but not in the sense that would parallel the renewal of analytical work on language and signification by European structuralists and post-structuralists. Venturi is perhaps one of the first architects to make the important distinction between what C. S. Peirce had earlier labeled an icon and a symbol. Venturi’s famous dictum categorizing buildings as either “a duck” or “a decorated shed” casts the difference between an icon and a symbol in architectural terms. A duck is a building that looks like its object: a hotdog stand in the form of a giant hotdog, or in Venturi’s terms, a place that sells ducks taking the very shape of a duck. Each of these examples has a direct visual relationship to its object. This visual similitude produces what Peirce calls an icon, which can be understood at first glance; it does not require, for example, close reading. Venturi’s other term, the decorated shed, refers to a public facade for what amounts to a generic boxlike building. For example, a bank building with a classical facade fronting a rather ordinary enclosure is a decorated shed. The decorated shed is more of a symbol

Venturi Layout 02 08.indd 131

131

in Peirce’s terms, in that it has a conventionalized meaning: the classical facade symbolizes a public building, whether it is a bank, a library, or a school. Icons and symbols become related when an overuse of an icon produces its degradation as it becomes a symbol, which in turn becomes clichéd and thus drained of the necessity for any form of close reading. If the issue of meaning is introduced in Complexity and Contradiction, it is first articulated in built form with the Vanna Venturi House, built by Robert Venturi for his mother between 1959 and 1964. It could be argued that the Vanna Venturi House is the first American building to propose an ideological break with modern abstraction at the same time that it is rooted in this tradition. Like the citations in Venturi’s book, the Vanna Venturi House requires close reading—and actually questions what constitutes such a close reading, both at the time of its construction as well as today. No building has more completely symbolized a new American vernacular, yet its placement in this book, it will be argued, is haunted by a little-acknowledged set of origins in both Italian Renaissance tropes and in the nine-square grid of modernist abstraction. To locate the origins of the Vanna Venturi House, two projects are worth comparing: John Hejduk’s Texas House 4 from the mid-1950s, and Venturi’s first study of the Vanna Venturi House from 1959. Hejduk’s Texas Houses are nine-square exercises, which bear a speculative relationship to Louis Kahn’s Adler and DeVore Houses (chapter 4) in that there are a series of servant and served spaces, but in the Texas Houses, the thick corner piers of Kahn’s Adler House are dematerialized and stretched into linear wall elements. Hejduk’s projects modify the Palladian schema of the ninesquare grid. Yet as Venturi’s scheme for the Vanna Venturi House evolves, it confounds the nine-square schema differently from the works of either Kahn or Hejduk.

4/9/2008 11:35:11 AM

132

Vanna Venturi House

4. Vanna Venturi House IIa, plan, 1959.

5. Vanna Venturi House IIb, plan, 1959.

Venturi’s first scheme (House I), with its compressed center bays and the external columns on the centerline, invokes a nine-square grid. Yet, because of the compression of the center bays, it can also be read as a four-square grid. An articulated corner and freestanding columns are also overlaid on the nine-square grid, yet neither scheme dominates. This overlay of two potentially different interpretations is reinforced by the absence of a column centerline running from the top to the bottom of the scheme, while the continuity of both the vertical and horizontal axes is interrupted by the central fireplace element. Another interpretation would read the lateral striation of the plan into three zones, creating an ABA demarcation of servant and served spaces. The lateral center zone can be read as both a void, given that it lacks the defining walls of the upper and lower zones, and as a solid whose rectangular form is bounded by six large piers. In yet a further analysis, the mullions, columns, and wall-bearing

elements can be read as a cruciform organization. This play of multiple interior grids confounds any single interpretation of origin. Close readings of the plan of Venturi’s first scheme do not produce a dominant diagram or a primary organization, but rather several diagrams which mark the beginning of a shift from a single reading to one which can be called undecidable. Each of the six schemes for the Vanna Venturi House provokes such a reading of undecidability, which must be understood as a Derridean, post-structuralist notion, an idea that was not available until after 1968. A second reading in Venturi’s House I involves the disjunction of the exterior walls from the internal volumes. First the spacing between the exterior walls and the internal volume is unequal, the right-hand void being larger than the left-hand void. The mullions and columns of the exterior walls, while they align and are symmetrical across the building, have little relationship with the articulations of the inner volumes. A compari-

Venturi Layout 02 08.indd 132

4/9/2008 11:35:12 AM

Vanna Venturi House

133

6. Vanna Venturi House IIIa, plan.

7. Vanna Venturi House IIIb, plan.

son of the walls enclosing the inner volume reveals minor shifts and offsets in what would initially seem to be a symmetrical alignment of column and walls about a central axis. It is also important to notice the articulation of the four corners of the house’s inner enclosure; whether wrapping around the corners or coming together as a seam, the joints at the corners are articulated as notches. This reinforces the independent and distinct nature of each facade, brought literally, but not conceptually, together at the corner. These notched corners of the interior volume can be read as indications of further subversions of the whole into a multiplicity of parts. The notched corners of the enclosure mark the boundaries of a set of three inner “parts,” demarcated by rectangular zones. For example, in the central area these notches are integrated into the piers, which then take on an irregular form as mullions joined to the piers. In this scheme, wall fragment, pier, and mullion become a new

kind of pochéd “part” that cannot be read back to an originary whole. While there is a relationship of this first scheme to Hejduk’s Texas Houses, the subtractive notching and additive poché of these early studies give the walls a figured quality distinct from the linear quality of Hejduk’s Texas Houses: if the wall/space relationship in Hejduk’s Texas Houses is produced by a series of squares pinned at the corners by steel columns and defined by infill walls; in the Vanna Venturi house, the walls begin to take on a volumetric presence that undermines their relationship to a classical nine-square plan. In the second versions of the scheme (House IIa and IIb), the outside frame walls are still present, as is the central horizontal core; however, the play between a tripartite and quadripartite scheme is less evident. For the first time, a figured central body is clearly manifest in the form of two chamfered corners which now extend beyond the ends of the two exterior freestanding

Venturi Layout 02 08.indd 133

4/9/2008 11:35:13 AM

134

Vanna Venturi House

8. Vanna Venturi House IIIa, model.

9. Vanna Venturi House IIIb, model.

walls. The cutout in the lower right quadrant creates a directional vector, moving into the center. The changes in the second project have little to do with the preceding scheme in terms of the critique of nine-square or four-square organizations. The second scheme breaks with the central symmetry of the prior scheme, and can be read as responding to the potential of a figured condition through chamfering and cutting; this incision produces a dynamic directionality in the internal volume. Throughout the plan, elements defining the inner volume—walls, columns, and mullions—are intermittently aligned and misaligned with the external perimeter condition, as if to portray a contradictory condition of simultaneous affirmation and denial. These misalignments acknowledge the dislocation of the internal volume from the free standing wall planes. Diagonals, whether actual or implied, seem to converge on the central fireplace. These countermand the orthogonal organization of rooms symmetrically across the central vertical band of service spaces. The diagonals animate the organization of the plan with a volumetric sense. The large cut that defines the kitchen is echoed across the vertical axis by a much

smaller cut, which nevertheless follows the dominant diagonal pointing toward the central hearth. There is a shear about the center, as the diagonals are not aligned to meet there; this causes the fireplace to become a volumetric element that at one moment is in rotation, and at another moment is the object of an implied shearing. Subtle differences between the projects for House IIa and House IIb begin with the four columns that frame the center, which become increasingly figured, thickened by poché-like attachments. In other areas, symmetry is reestablished: for example, the small diagonal wall segment defining the upper bathroom is straightened out, mirroring the other side of the bathroom across the vertical axis. Clearly these are notational rather than functional changes. In versions IIIa and IIIb, the project retains the central element of the fireplace and the two external chamfered corners, yet here the central zone becomes increasingly articulated. The central horizontal walls are also thickened and a secondary chamfering is introduced, as if the dominant exterior volume were reduced and rotated, responding to the spiraling motion of the fireplaces, which

Venturi Layout 02 08.indd 134

4/9/2008 11:35:14 AM

Vanna Venturi House

135

10. Vanna Venturi House IVb, plan, 1961.

11. Vanna Venturi House V, plan, 1962.

shear off of the central axis in a yin-yang form. Finally, what was formerly a straight-run stair, aligned on the vertical axis, suddenly bends: this produces a third level of chamfering. This is the first appearance of such an oddly angled stair in the project, and it will become one of the signature figures of the final house. In the early plan studies, the seeds of the arguments of Complexity and Contradiction are present as a rethinking of the nine-square and four-square diagrams. The constant play between elements that are in reciprocal and symmetrical relationships in one reading and are displaced by another reading marks the beginning of what can be read as Venturi’s implied critique of any classical part-to-whole relationship. Vanna Venturi House IVb is perhaps the most subdued of the schemes, returning to a self-contained symmetry. A four-square parti is evident in the compact pairs of rooms flanking the compressed central service spine. Yet there is a vestige of the nine-square organization, in that the paired fireplaces are incorporated into a thickened band of services, as a third zone running vertically. The central fireplace is offset

toward the projecting semi-circular termination of the interior volume. This semi-circular external enclosure softens the blunt form of the chamfered end. A disjunction remains in what will eventually become the union of fireplace and staircase. In IVb there is a compression of space that modulates the internal organization. This compression is a thematic that will animate the final scheme. The fireplace is no longer in the center of the four-square plan, displacing the ideal of the centralized hearth. In IVb, the “decorated shed” is made of isolated walls that wrap around an internal “duck” with what becomes, after modernism, a stereotype of the gabled, symmetrical, double-chimneyed house. In this fourth project, the internal “duck” has a different profile than its exterior shell. The slot between the exterior shell and the interior object makes the play between the two legible. It is only in House V that the strategies animating the final scheme appear, beginning with a radical reorientation of the house on the site by ninety degrees. The long axis of the house is now made perpendicular to the entry from the street, reinforcing the idea of a facade across the grain

Venturi Layout 02 08.indd 135

4/9/2008 11:35:14 AM

136

Vanna Venturi House

12. Vanna Venturi House IV, model, 1962.

13. Vanna Venturi House V, model, 1962.

of the house. The exterior envelope of House V appears the same as that of IVb, but now a faceted figure extends out to the left beyond what is clearly a four-square plan. This figure is simultaneously compressed by some internal vector pressing from the back to the front. This shape seems to extend into the exterior courtyard without any internal logic from the main volume, and will become a central feature of the final project. In the sixth and final version of the Vanna Venturi House (VI), what remains is only a part of a previously existing and implied whole as if House V were cut in half. The sixth version initially appears to have been divided in two: the previously semi-circular end becomes a quartercircle, and the tripartite condition in the middle has vanished, compressed in the faceted figure of the stair/fireplace. The compressive energy internal to the previous plans is registered in the fireplace/stair element, which has been relocated to a central position. The resonances between Hejduk and Venturi reappear in this version: if the nine-square scheme of Hejduk’s Texas House had appeared in the first scheme of the Vanna Venturi House, then it is Venturi’s final scheme, as a half-house, that seems to influence Hejduk’s subsequent Half-House and Quarter-House, both of which are intended to be read from a prior

condition of unity. These changes all occur behind a complex facade notation—the articulation of a horizontal beam, a fallen arch, and a broken pediment involving an aedicular motif. Venturi had observed this aedicular motif in Rome, in Moretti’s Casa “Il Girasole” as well as Carlo Rainaldi’s Santa Maria in Campitelli. In baroque architecture, the upward thrust of the aedicule is usually a voided condition. In the Vanna Venturi House, the pediment is broken, but the upward thrust of the central fireplace is both volumetric and a void. As a centerpiece, the fireplace both affirms historical precedence and simultaneously denies this precedence. While Venturi employs historical elements in plan and section, he does so in a way that denies their historicity, and therefore questions the value of any historical precedent. Moreover, the vertical cut on the facade bears little relation to the horizontal compression that appear in the interior. Venturi’s critical reading of Moretti’s Casa “Il Girasole” can be discerned here: if the latter can be seen as layers of space compressed toward the rear of the building and away from the facade, then the Vanna Venturi House articulates these compressive vectors in a thickened, condensed facade apparent in the final plans. This disjunction between the vertical plane

Venturi Layout 02 08.indd 136

4/9/2008 11:35:15 AM

Vanna Venturi House

137

14. Vanna Venturi House VI, plan, 1962.

and the horizontal space of the plan looks forward to the decorated shed of Venturi’s 1967 National Football Foundation Hall of Fame competition. Other aspects of the design involve a gabled roof, which extends and interweaves with the facade, causing it to function as a mask. What is significant in this respect is the purposeful extension of the facade as a surface denying any volumetric corner. The treatment of the facade resembles Venturi’s notion of the decorated shed, in calling attention to the facade as simply that: a screen masking a shed. This single element conflates two of Venturi’s signature ideas. He combines the energy of the bent staircase, the central diagonal rotation, and the projective vectors, but contains these within an exterior condition that highlights its own function as a screen. A reading of the Vanna Venturi House requires a reading of its previous incarnations before any appreciation of its undecidable condition is possible. This is one of the first American houses that can be read as process, and as such becomes an expression of the process of reading in the object. To attempt to argue that any final project, the built building, carries all of the energy of its earlier studies can rarely be sustained. Most final projects move to establish certain, but not all, of the elements that evolved in the previous

Venturi Layout 02 08.indd 137

versions. For example, much of the complexity of other versions of the Vanna Venturi House—the play between the tripartite and quadripartite, the occupied center and the voided center, the center and the edge—disappear with the clarity of the final version’s articulation of a conceptual front and back to the house. The earlier schemes involving a nine- or four-square parti deny the possibility of any resolution. The final house is both more classical and less modern than the earlier projects. One version may not necessarily be better than the other, they merely attribute differing importance to different ideas. But in neither case is one single dominant idea operating. Classical and modern tropes operate simultaneously; the heritage of this house is one where that evolution is constantly at odds, and ultimately remains undecidable. This studied undecidability is one of the qualities that emphasizes the need for this project to be closely read, and simultaneously its demand for a different kind of close reading than its formal predecessors. Such a close reading no longer produces a single formal whole, but rather allows the unreconciled differences of this project to remain legible. Many of Venturi’s houses that followed became rhetorical devices, references, and ges-

4/9/2008 11:35:16 AM

138

Vanna Venturi House

tures which lack the inherent undecidability of the Vanna Venturi House. Even in the late 1950s and early 1960s, the idea of the house carried the possibility of theoretical and critical weight, and the possibility of manifesting ideologies other than that of merely a single-family house, as it did for Mies, Le Corbusier, Loos, and many of the early modern architects. There is a truism in architecture that books are sometimes more important than buildings. This could be said for Palladio, and perhaps for Le Corbusier. Yet the Vanna Venturi House is a writing, in architectural terms, of Venturi’s Complexity and Contradiction; no American house or building before or after can make that claim.

Venturi Layout 02 08.indd 138

4/9/2008 11:35:16 AM

Vanna Venturi House

139

15. Vanna Venturi House, elevations, ground-floor and upper-floor plans.

Venturi Layout 02 08.indd 139

4/9/2008 11:35:17 AM

140

16. The facade acts as a screen for the volume of the house. This motif similarly has resonances with the screenlike north facade of Luigi Moretti’s Casa “Il Girasole.”

Venturi Layout 02 08.indd 140

Vanna Venturi House

The notational arch, an indexical mark framing the opening, is a classical resonance.

4/9/2008 11:35:17 AM

Vanna Venturi House

17. The aedicule appears in the slot in the front facade. The half-round aedicular window in the rear facade and the split in the front facade play together.

Venturi Layout 02 08.indd 141

141

Similar to the Casa “Il Girasole,” a dialogue between the back and front facades is evident.

4/9/2008 11:35:18 AM

142

18. The front and rear facades of the Vanna Venturi House become screens or planes bracketing the interior space. The final version of the house has returned to the classical idea of a front and a back.

Venturi Layout 02 08.indd 142

Vanna Venturi House

19. A center axis running laterally is defined by a partition on one side, and a wall on the other.

4/9/2008 11:35:19 AM

Vanna Venturi House

143

20. There is a figural compression against the front facade. The rear corners can be read as voided.

Venturi Layout 02 08.indd 143

4/9/2008 11:35:19 AM

144

21. The fusion of the fireplace and stair elements registers important changes in each version of the

Venturi Layout 02 08.indd 144

Vanna Venturi House

house. This element becomes a single, central form, one that is cranked and bent in the final project.

4/9/2008 11:35:20 AM

Vanna Venturi House

22. Together the stair and fireplace spiral upward and outward. The staircase with the fireplace causes space to be articulated as a series of centrifugal vectors,

Venturi Layout 02 08.indd 145

145

generated by the most central fireplace element in the house. The inclusion of a staircase that leads nowhere is significant. It is an index of the denial of function.

4/9/2008 11:35:20 AM

146

23–25. The lateral cross section shows the thickened poché defining the vertical fireplace sitting within a frame. Sections cut across the width of the house reflect a compression toward the front facade and a thickened frontal datum.

Venturi Layout 02 08.indd 146

Vanna Venturi House

26 (Opposite page). The facade is split into two parts, with the flue off center. Both the broken arch of the facade and the arch on the rear facade reiterate the tripartite division between base, middle, and pediment. The windows break the lower chair rail into sections. A five-part horizontal window plays between symmetry and asymmetry with its central bars; another fourpart square window is divided symmetrically.

4/9/2008 11:35:21 AM

Vanna Venturi House

Venturi Layout 02 08.indd 147

147

4/9/2008 11:35:22 AM

148

Vanna Venturi House

27. Vanna Venturi House, exploded axonometric view.

Venturi Layout 02 08.indd 148

4/9/2008 11:35:24 AM

Vanna Venturi House

149

28. Vanna Venturi House, exploded axonometric view.

Venturi Layout 02 08.indd 149

4/9/2008 11:35:24 AM

150

Vanna Venturi House

29. Vanna Venturi House, ground floor and facade, axonometric view.

Venturi Layout 02 08.indd 150

4/9/2008 11:35:24 AM

Vanna Venturi House

151

30. Vanna Venturi House, front axonometric view.

Venturi Layout 02 08.indd 151

4/9/2008 11:35:24 AM

152

Vanna Venturi House

31. Vanna Venturi House, rear axonometric view.

Venturi Layout 02 08.indd 152

4/9/2008 11:35:25 AM

Venturi Layout 02 08.indd 153

4/9/2008 11:35:26 AM

1. James Stirling and James Gowan, Leicester Engineering Building, Leicester, England, 1959–63.

6. Material Inversions James Stirling, Leicester Engineering Building, 1959–63 James Stirling’s Leicester Engineering Building is one of the early postwar buildings to stage a confrontation between modernist abstraction and an incipient postmodern “reality” embodied in a material presence. It can be said that this confrontation is manifest not only in the forms of the building, which owe their paternity to the Modern Movement, but also in its materiality, which initiates a critique of the modernist palette as orchestrated within an abstract framework. Postwar realism took different forms across continental Europe, and in this sense, the English context immediately after World War II is significant in understanding the critique of modernist abstraction embodied both materially and conceptually in the Leicester Engineering Building. England was relatively removed from prewar mainstream modern architecture, which was essentially a continental phenomenon with varying political aims. In Germany in the 1920s, modernism was a left-oriented and Marxist-inspired movement; while in Italy after 1933, under Mussolini, modern architecture—in many cases even in the late 1920s—represented the aesthetics of the fascist regime, classical rhetoric, and monumentality. In France, modern architecture maintained an unclear relationship between left and right. Le Corbusier, for example, alternated between an entreaty to Mussolini and French Syndicalism on the one hand, and a radical program for urban change on the other. In England, as in America, the context for architecture could essentially be called pragmatic. But unlike in the United States, which remained under the sway of a prewar BeauxArts influence, architectural culture in England was profoundly affected by the war. First, because of the refugees from Polish architectural schools; this influx would have particular influence on the young student

156

Leicester Engineering Building

2. Leicester Engineering Building, site plan, 1959–63.

3. Leicester Engineering Building, axonometric, 1959.

James Stirling. Second, the war clearly disrupted the education of an entire generation of students. Of that generation, James Stirling and Colin Rowe first met while serving in the British army in Scotland at the Queen’s Barracks, Perth, in 1942. They would find each other again after the war at the Liverpool School of Architecture. Rowe had begun his studies at Liverpool in 1939, which had been a conservative school of architecture before the war, but had radically changed with the influx in 1938–39 of Polish refugees, many of whom had been active in the Modern Movement and introduced the school to Corbusian modernism and a form of Russian constructivism. When the war became critical after Dunkirk and the fall of France, both Stirling and Rowe volunteered and joined the army parachute corps. In training, Rowe made a practice jump out of a plane during which his parachute did not open, and his back was broken; Stirling, on the other hand, continued through his training and eventually jumped at Remagen Bridge. By the time he returned to

Liverpool in 1949, Rowe was his teacher, and Stirling completed his final thesis project in 1949 under Rowe and the profound influence of Le Corbusier. The 1950s signaled a change in the climate for modernity in England as architects, artists, and sculptors focused on alternatives to modernist abstraction. Collaborative efforts such as that of the Independent Group proclaimed an interest in everyday materials and an “as found” aesthetic, which Stirling captured in a papier-mâché model of soap bubbles in the famous 1956 exhibition “This is Tomorrow.” The sculptor Eduardo Paolozzi showed work in raw iron materials and Cor-ten steel; the pop painter Richard Hamilton incorporated collaged imagery of contemporary consumer culture; Nigel Henderson introduced his documentary photographs of the workingclass street; and the architects Peter and Alison Smithson used corrugated plastic and rough plywood for their Patio and Pavilion installation. “This is Tomorrow” brought together work by

Leicester Engineering Building

157

4. Melnikov, Russakov Worker’s Club, 1927.

5. Le Corbusier, Maison Jaoul, 1951.

members of the self-styled Independent Group (which included architects Stirling, the Smithsons, and the critic Reyner Banham). Some of the participants—including the Smithsons— were equally involved in a post-CIAM (Congrès International d’Architecture Moderne) group, Team Ten, which was dedicated to reviving the principles of modern architecture after the war. As a member of the Independent Group, Stirling was critical of Team Ten’s late modernist ideology. If “This is Tomorrow” drew attention to the cozy comforts of the postwar British consumer culture, in turning attention to the material of the everyday—advertising, furniture, the street— the exhibition also led to several widely divergent offshoots. One was in urban planning as envisioned by Gordon Cullen’s “Townscape” drawings, which resembled picture-postcard views of cities. Another direction was pop art, which developed in the work of Peter Blake, Hamilton, and Paolozzi as a celebration of technology, and would ultimately lead to the work of Cedric Price

and Archigram in London during the 1960s. The impact of “This is Tomorrow” created an impetus not only toward pop, but also toward a tough form of neorealism, different in England from that of Italy; it was named “New Brutalism” by the critic for the Architectural Review, Reyner Banham. New Brutalism was a reaction to the image of a comfortable British lifestyle and the Townscape movement. It was oriented instead toward an idea figured in blunt materials and forms such as the Martello towers on the south coast of England. The obvious materiality of these forms can in some sense be related to the role of materials in Italian realism, of which Stirling was aware. An important later influence can be gathered from his article “‘The Functional Tradition’ and Expression” in Perspecta 6, 1960. Here Stirling discussed Luigi Moretti’s plaster casts, which created what Stirling called “solidified space.” This seemingly paradoxical inversion of the material qualities of solid and void became a theme that Stirling would develop more didac-

158

Leicester Engineering Building

6. Leicester Engineering Building, office tower.

7. Leicester Engineering Building, office tower.

tically in his early works, including a series of flats in Ham Common outside of London and a row house project in Preston, which prefigured his conceptualization of an inversion of materials at Leicester. In another context, Stirling was also much taken with Le Corbusier’s reintroduction of materials, especially in the Maison Jaoul’s low, vaulted brick arches, which he described as almost primitive in character in his September 1955 article, “Garches to Jaoul,” in the Architectural Review. Stirling emphasized the materiality of the Maison Jaoul, which he contrasted to the “neutralized”

surfaces of Garches. Stirling noted that “it is disturbing to find little reference to the rational principles which are the basis of the modern movement,” and he saw in the Maison Jaoul not just a romantic or picturesque notion of postwar modern architecture but also, in its use of varied materials and a barrel vault, a profound critique of modern architecture. Such attention to materials reflected not only Stirling’s interest in a tough realism, in some sense accommodating his “north country” origins, but also his own reappraisal of postwar Le Corbusier. Yet Stirling’s return to material was different from that of Le Corbusier

Leicester Engineering Building

159

8. Leicester Engineering Building, stair towers.

9. Leicester Engineering Building, workshops.

and his English contemporaries in that, as will be argued here, Stirling used material both critically and conceptually. Stirling’s Leicester Engineering Building, done in partnership with James Gowan, marked a significant change from their earlier work. Leicester is one of the first manifold critiques of modernism and the first in a series of Stirling’s major university buildings in England, which include the Cambridge History Faculty Library, the Florey Building at Queens College, Oxford, and the Saint Andrew’s Dormitory project in Scotland. Of these four university commissions,

the Leicester Engineering Building is the most articulate in its critique of modernist abstraction. This critique is manifested in three different ways: first, in the use of glass; second, in the use of modular ceramic units (brick and tile); and third, in the compositional organization of the building’s masses. In modern architecture, glass was conceived and used as a literal void as well as a phenomenal transparent material, as discussed by Colin Rowe and Robert Slutzky in their seminal article “Transparency: Literal and Phenomenal,” published in Perspecta 8, 1963. In the Leicester

160

Leicester Engineering Building

10. Leicester Engineering Building, workshops.

11. Leicester Engineering Building, workshop roof.

Engineering Building, glass is used to suggest that it functions conceptually both as a solid and volumetrically. Contrasting with its implicit function as a negative or void in modernist architecture, glass serves here as a positive integer in what has been called a textual sense. Leicester marks the movement of glass from void to solid, in other words, a reversal of the conception of glass’s materiality, from literal void to conceptual solid. Stirling’s drawings of the building illustrate the evolution of this treatment of glass from the first early sketches to the final building. The often published early axonometric drawing by Stirling is important in understanding the conceptual development of the material inversions. The drawing marks four significant changes which will appear in the executed project. First, and probably the most noticeable, are the changes involving the idea of solidity in glass. These are the addition of volumetric diamond-shaped elements, which form both levels of the laboratory roof, and the horizontal glass projections replacing the banded glass striations of the office block. Second, and less obvious but no less significant,

changes can be seen in the curtain wall element in the tower block. In the drawing, this element is flush with the brick fascia above it (and actually recessed back from a vertical concrete element that is no longer present in the final building), while in the executed tower, the entire glass curtain wall is set forward of the brick fascia. A third and minor change occurs in the glass under the parallel (to the dominant grain of the building) lecture theater, where the glass element underneath the theater now has chamfered corners. All of those taken together have the same effect: what was seen in modernist abstraction as transparent, planar, and void is now to be read as more opaque, volumetric, and solid. A fourth change involves the actual material solids. In the drawing, these are rendered accurately, denoting the difference between the ceramic and running bond brick. All the surfaces which run vertically (denoting their nonstructural condition) are rendered in the drawing with vertical hatching, while the bearing-wall brick surfaces are rendered with horizontal hatching. It is also necessary to look at some of the other aspects of the building to understand its

Leicester Engineering Building

161

12. Leicester Engineering Building, glass stairwell.

13. Leicester Engineering Building, spiral stair.

conceptual density. At the core of the building are two stair towers whose forms have chamfered corners. The stair towers are not the same height and thus produce something of a romantic skyline similar to that of Kahn’s towers in the Richards Medical Center. The vertical gridding of the brick tile units on the stair towers is also significant in Stirling’s inversion of the conventional qualities of material. These tiles are laid end-to-end vertically; though these units have a real materiality and physical presence, they are not treated as structural. There is a play between the glass elements, which are not structural but appear volumetric and structural, and the brick units, which are laid vertically and made to resemble a surface veneer. This undecidable quality is reiterated in several details of the office tower. This volume is also supported by a concrete column that seems to run through the auditorium element, yet in becoming a concrete haunch it supports another glass element which can also be read as volumetric, since it projects forward of the tiled elements. This didactic use of materials clearly demonstrates the difference between glass as a plane,

glass as part of a continuous surface, and glass as a volume that is clearly interrupted by and articulated around this concrete haunch. These numerous inversions may be considered textual rather than formal since they are less aesthetically or visually conditioned. For example, one of the stair cores is shaped to read as a prism; however, its lower portion is cut away to reveal a spiral stair clad in the same translucent glass and opaque metal panels as the stair tower. The staircase becomes a figured element, twisting like a corkscrew and driving its way up through the floor surface. Chamfering and twisting are certain of Stirling’s strategies that suggest the glass again has become conceptually more solid than the concrete structure of the building. This stair core can be contrasted with a second stair encased primarily in clear glass. Here material inversions undermine conventional associations of structure: the transparent glass stair tower becomes a void that seems to hold up the large cantilevered mass of the auditorium, while the metal-panelled stair tower—the solid—is cut away, revealing the corkscrew of a concrete staircase. Such a play of materiality—

162

Leicester Engineering Building

14. Leicester Engineering Building, section, 1959–63.

making the glass staircase appear to support a massive volume while the concrete staircase is dematerialized into a spiraling vector—confounds the properties conventionally associated with each material. The didactic character of Leicester is also manifest in its reversal of what could be considered a modernist idea of a centrifugal composition, that is, one that moves its energy away from the center to the periphery. At Leicester, the organization of the volumes is centripetal, in other words, collapsing or being sucked down into the center of the volumetric massing. The diagonals of the ramps and beveled lecture hall volumes slant toward this center as they pivot about the central stair tower. There is a second, countermanding energy, a dynamic rotation that reveals the influence of Melnikov’s Russakov Worker’s Club in these two discrete volumes. The significant differences between the two buildings must be registered: while Melnikov’s project proposes a collapse toward the center, there is a lack of rotation in the composition of the projecting

volume of the Russakov Club, while rotation is a primary characteristic of Leicester’s juxtaposed volumes. Melnikov’s volumes seem to float free, while Stirling’s volumes are pinned by the towers, which introduce a dynamic thrust downward. Stirling further emphasized the undecidable nature of these materials by producing over the shed building (workshops) a volumetric glass unit, diamond-shaped and translucent, which is extended and repeated over the entire roof. It seems to slide or hover in an unstable position over the structure. Such inversions of materiality also continue throughout the workshop building: brick-tiled units are structural at the base, yet are surmounted by a concrete, lintel-like element above a reveal, which suggests that the concrete element is floating over the brick. In this sense, the “reality” of the void is articulated as a slot or cut-away—in other words, as real space—while the representation of void in glass—in other words, the “unreal” voids—are treated volumetrically. This constant displacement in meaning and function of materials provokes the need to read

Leicester Engineering Building

15. Leicester Engineering Building, ground-floor and mezzanine plans, 1959–63.

materials as conceptual rather than phenomenal physical integers, producing a building that is neither picturesque nor expressionist but rather defines a textual use of materials. The implied flows and forces in the volumes deny a static relationship between the viewer and the building. In Leicester the sense of arrested rotation gives a sense of space and time that is no longer merely formal as a dominant mode of discourse. Thus, Leicester, in its numerous inversions, denies the traditional architectural interpretations of facade, stasis, and literal materiality.

163

164

16. The Leicester Engineering Building is articulated as an assemblage of component elements—atria, stair towers, auditoria, offices, lab towers and workshops— each remaining separate from the other and treated volumetrically. The central tower of lobbies, elevator

Leicester Engineering Building

and stairs acts as a vertical fulcrum around which the masses of the building rotate. These rotating elements produce a centripetal thrust downward, similar to the directionality implied by the chamfered glass units and beveled volumes of the laboratory building.

Leicester Engineering Building

17. Stirling’s use of chamfered corners in both glass and masonry units reveals these elements to be conceptually solid and volumetric.

165

18. The stair tower is cut vertically at the corners and undercut at the top, thereby rendering them volumetric in both axes.

166

19. Leicester Engineering Building, interior spatial diagram of auditorium, office, and lab towers.

Leicester Engineering Building

Leicester Engineering Building

20. Leicester Engineering Building, diagram of circulation and stair towers.

167

168

21. Leicester Engineering Building, diagram of glass elements.

Leicester Engineering Building

Leicester Engineering Building

22. Leicester Engineering Building, diagram showing the structural elements.

169

170

23. Leicester Engineering Building, worm’s-eye axonometric view.

Leicester Engineering Building

Leicester Engineering Building

24. Leicester Engineering Building, ground floor, axonometric view.

171

172

25. Leicester Engineering Building, lobby floor, axonometric view.

Leicester Engineering Building

Leicester Engineering Building

26. Leicester Engineering Building, fourth floor, axonometric view.

173

174

27. Leicester Engineering Building, ninth floor, axonometric view.

Leicester Engineering Building

Leicester Engineering Building

28. Leicester Engineering Building, roof from north-east, axonometric view.

175

176

29. Leicester Engineering Building, roof from south-east, axonometric view.

Leicester Engineering Building

1. Aldo Rossi, Composition with saint and the Modena cemetery, 1979.

7. Texts of Analogy Aldo Rossi, Cemetery of San Cataldo, 1971–78 In 1945, just after the end of World War II, Georges Bataille published a book titled Le Bleu du Ciel, which translates as “the blue of the sky,” yet curiously enough was published in English as The Blue of Noon. The book was actually written in May of 1935. Its plot, set during the general strike in Spain and rise of Nazism, is a metaphor for the hopelessness of the left’s ideology in the face of the oncoming world war, with necrophilia as one of its central metaphors. It is therefore not without some relevance that Aldo Rossi’s competition project for the Cemetery of San Cataldo in Modena was entered under the title “The Blue of the Sky.” Its ossuary is an empty walled cube reminiscent of the stark geometries of a De Chirico painting or those of Ernesto Lapadula’s Palazzo della Civiltà Italiana at the Esposizione Universale di Roma planned for 1942. Rossi’s project is also a metaphor for the futility of redemption in the sanctuary; instead, the only hope is the ever-present but mockingly distant and unachievable blue of the sky. In the project for the Cemetery of San Cataldo, the metaphor not only derives from the typological explorations of Rossi’s analogical drawings, but also emerges as a polemical statement drawn from postwar literature signaling the political exhaustion of modernism. Unlike Rossi’s earlier projects or those which became more illustrative at the end of his career, the Cemetery of San Cataldo is both a political and architectural critique of modernism in which the ideas broached in Rossi’s book The Architecture of the City ultimately take physical form in the partial realization of the cemetery. Rossi’s critique of modernism is located in the grim reality of postwar Italy and its multiple reactions to the fascist monumentality in aspects of Italian modernism. One such reaction took form in the escapist aesthetic

180

Cemetery of San Cataldo

2. Central Business District proposal, Turin, 1962 .

3. Study for the Segrate Monument, 1967.

of the neoliberty style, referring to Italy’s ties to England in the late nineteenth century and adopting the name “Liberty” after the English manufacturer of Art Nouveau fabrics. Neoliberty called for a return to patterned figuration, evocative materiality, and a type of softness to counter both the harshness of modernist abstraction and the overblown imperial scale of Italian fascism. A second reaction was manifest in Italian neorealist film of the same period. Neorealism is a term derived from literature and film and, when extended to architecture, reflected the climate of the Italian liberation and its turn away from modernist abstraction. A documentary attention to everyday life and an abundance of details were among the mimetic techniques used to produce the “realistic” effects of neorealism. Neorealism involved a double mimesis in architecture. For example, the rebuilding of the Tiburtino district in Rome during the early 1950s produced buildings that were new by necessity but also needed to resemble the product of historical sedimentation. This mimicking of historical forms resulted in an architecture whose neorealist effect was

marked by a nostalgic quality. This was countered by another form of realist thought which worked against the grain of neorealism’s descriptive effects. Rossi’s concept of realism departs from neorealism’s humanist values, as Pier Vittorio Aureli describes in discussing the terms of Rossi’s “realist education,” and turns a new critical attention to what Rossi considered to be the “facts” of the city. In his early work and writings, Rossi initiates a critique of the scenographic effects of neorealism by pointing toward a more structuralist notion of realism in architecture that is grounded in typological studies. Rossi’s critique of the modernist canon—of both the abstractions of late modernism and the monumentality of fascist Italy—could be considered most evident in his drawings and his important first book, The Architecture of the City, published in 1966. The Cemetery of San Cataldo at Modena, and perhaps to a similar degree his Gallaratese housing complex, are among the few of Rossi’s realized buildings in this period that integrate his critique of abstraction with his interest in typology, analogy, and scale.

Cemetery of San Cataldo

181

4. Gallaratese 2, Milan, 1969–73.

Another critique of modernism in architecture was represented at the time by the magazine Casabella, which, under Ernesto Rogers’ direction, included many of Rossi’s early articles on Adolf Loos and Louis Kahn, as well as on modernist buildings such as Mies’s Seagram Building and Le Corbusier’s La Tourette. In addition to critical writings, Rossi’s early work as an architect included his participation in several competitions. The most famous of these was the project for a regional government center in Turin, which consisted of a giant four-sided square, a megabuilding on giant columns spaced 100 meters apart with a vast square courtyard in the center. Rossi placed this massive form outside of the city of Turin as a new kind of over-sized civic marker. In the Turin project, as well as in competition projects for monuments in Cuneo and Segrate, the juxtaposition of scales becomes an important theme. These early works also deploy pure geometrical forms—circles, isosceles triangles, and squares—which are extruded to form cylindrical, cubic, and triangular structures. The cubic form of his Turin project and the extruded trian-

gular form on a circular column for his Segrate Monument to the Partisans exemplify his interest in forms reduced to their geometric archetypes. Such forms would reappear in the Cemetery of San Cataldo competition in different guises. Rossi published The Architecture of the City before any of his work had been built, much like Robert Venturi’s publication of Complexity and Contradiction in Architecture, published the same year, before Venturi completed any major built work. Venturi and Rossi also shared an interest, new at the time and expressed in theoretical postulates, in describing the irreducibility of the city to any of modernism’s totalizing visions. Yet, where Venturi’s populist embrace of the city and its hallmark strip includes its temporary signage in the city’s symbolic language, Rossi instead adopts an analytic method to isolate what he considered the city’s urban artefacts. Such urban artefacts include elements of the city whose continuities, be they functional, such as housing, or symbolic, such as monuments, account for their permanence within the history of the city. In Rossi’s analysis, these artefacts can

182

Cemetery of San Cataldo

5. Domestic Architecture with Monuments, 1974.

6. Courtyard and tower, Fagnano, 1973. Detail.

also be considered catalysts for new buildings. This dialectic of permanence and growth defines Rossi’s understanding of the city as occupying different moments in time and suggests that the urban artefact records diachronic moments and histories. As one of the books that was critical in rethinking the relationship of architecture and the city, Rossi’s The Architecture of the City shares some similarities with Colin Rowe’s book Collage City, first published in 1978, yet it is their differences that are important. A fundamental premise of Collage City is that what existed—the buildings embodying the history of architecture—had an intrinsic value and could be considered truthful as well as foundational. Rowe gave a value to origin, and therefore any urban project had to respond to these pre-existing or, in Rowe’s terms, “set pieces” of the city. In Collage City, Rowe selected such set pieces, like a rotunda or a

square or even a mega-building like the Hofburg in Vienna, and inserted them into other contexts in a strategy that resembles Piranesi’s Campo Marzio project. Rowe’s idea of set piece, taken out of its original context and reinserted into a new context, linked contextualism to the idea of collage. Yet Rowe’s and Piranesi’s strategies differ in respect to the value of origins. Whereas Rowe assigns an a priori value to what exists and adds structures to reinforce this concept, Piranesi assigns no value to the existing context and creates set pieces with no a priori context as a grounding idea. Rowe’s method of collage reuses preexisting meaningful fragments, while Piranesi maintains the juxtaposition of elements without being beholden to an idea of the whole. Rossi’s approach could be likened to that of Piranesi in terms of retaining a tension between urban elements, denying a singular narrative, meaning, or origin. Instead of set pieces, fragments, or

Cemetery of San Cataldo

183

7. For Peter Eisenman, 1978. Detail.

8. Studio, 1980.

collaged elements of the city, Rossi conceived of the city as an ensemble of typological elements, whose simple geometries could be read as the result of removing their layers of historical accretions. The process of reduction is identified in Rossi’s typological analysis as the study of types of urban elements distilled to their most simple geometric form. This produced geometric figures with a level of figuration unlike the abstract entities of modernism and unlike the contextual character of Rowe’s urban fragments. In this operation, Rossi rethought the entire notion of typology developed in the nineteenth century by J.N.L. Durand as a series of type conditions for certain buildings. Rossi was perhaps the first postwar architect to reintroduce the notion of typology in architecture. In attacking the tradition of typology related to function as well as to the formal, Rossi used type as an analytical instrument with which to generate form as well

as to generate a critique of modernist abstraction. He reintroduced instead a typology which dealt not only with the problem of scale but also with the problem of meaning. Rossi envisioned typology as standard elements that were scaleless and only meaningful when understood in a particular context. This idea of typology raised the issue of repetition, suggesting that the city is given form by a repetition of certain archetypal elements or urban artefacts. The issue of repetition was also important in minimalist sculpture as a critique of narrative—the repeated series lacks beginning, middle, and end—and as a critique of origin, as the individual or starting unit is subsumed by other identical units. The repetition of an urban artefact destabilizes the relationship between these elements and their perceived aesthetic and functional value as cultural icons. Rossi uses iconic forms but strips them of their iconicity through

184

repetition, a technique that undermines the aura and uniqueness of architectural elements. When these elements are taken out of their aesthetic and functional context, they can potentially be deployed as textual elements. Their visual importance is undermined through repetition of archetypal elements that have no fixed or determinate scale. Rossi achieves shifts in scale in several different ways, in both drawing and building—for example, architectural elements such as pilotis. The image of the building on giant pilotis is one that Rossi repeats both in drawings and in buildings such as the Gallaratese housing project and subsequently in the Modena cemetery. Rossi’s drawings are also a locus for his critique of contextualism. His drawings, for example, register both the dislocation of place through the repetition of typological elements, and the dissolution of scale through the introduction of domestic objects into the urban environment. In these drawings there are recurring formal themes: types based on Platonic forms are scaleless, placed in different contexts, and thus estranged from any classical concept of a partto-whole unity; types drawn from domestic environments are envisaged at architectural scales; these objects of domestic use, in their figured condition, capture Rossi’s questioning of scale related to typology. Rossi’s urban artefacts could exist at any scale, in an interior as well as in a cityscape, as is suggested in the drawing called Domestic Architecture. While the drawing seems to present a table top with a cup, goblet and coffeepot in the center, all of which are household items, when the top is removed from the pot, the domestic object becomes an architectural form that reappears as Rossi’s Teatro del Mondo, his floating theater for the 1980 Venice Biennale. His play with scale allows the table, along with the fork and the spoon, to disappear into the city: the table top becomes the ground and the coffeepot becomes a building. Other elements, such

Cemetery of San Cataldo

as lighthouses or the giant figure of a saint, from the hills outside of Turin, continue to defamiliarize the sense of a unitary urban scale. If a lighthouse is a coffeepot at one scale and a coffeepot is a lighthouse at another scale, Rossi suggests that familiar objects have their own autonomous condition inscribed in their type, yet his concept of typology remains resolutely open-ended. One of Rossi’s first buildings to deploy this shift of scale and repetition of elements is his Gallaratese housing project, whose colonnade is less a classical organization than it is a repetition of typical elements. Gallaratese’s heavy pilotis return in a number of drawings, and when juxtaposed with the coffeepot, the mortar-and-pestlelike elements enter Rossi’s vocabulary as a means of describing an estrangement through scale. Rossi’s drawings also combine aspects of the Modena project into new relationships with the city: Modena’s conical shrine resembles an industrial tower, and occupies the same landscape as an archetypal Tower of Babel. In Rossi’s painting of the courtyard of Fagnano, Modena’s square cruciform window forms the backdrop for an arcade reminiscent of the Gallaratese housing block. Other drawings equate the punched-out square windows of Modena’s cubic ossuary to those of a house. It is when these elements are taken out of a real or built context that they become both analogic and textual, in that they do not conform to a single idea nor to any manifestation of reality. There is a play between the real and the abstract, between the scale of objects, and between the familiarity of objects which breaks down conventions that are attached to meaning, abstraction, form, and scale. Drawings, for Rossi, are not intended as artwork, nor are they examples of metaphysical or surreal content like De Chirico’s urban landscapes. While the deep shadows, black windows, and white surfaces of structures within Rossi’s drawings have De Chirico-esque characteristics, Rossi’s drawings are analogic as well as textual; they are a cri-

Cemetery of San Cataldo

185

9. Composition with plans, elevations, and sections, Cemetery of San Cataldo, Modena, 1971.

tique of architecture that cannot be made in the medium of architecture itself. The Cemetery of San Cataldo project focuses the energy of Rossi’s drawings and the ideas in The Architecture of the City to render the cemetery as another type of city. In the drawings that Rossi submitted for his competition entry, the conception of the cemetery as a series of parts becomes clear: rows of columbaria and objectlike ossuaries are the locus for the symbolic burning of the bodies. The “town square” occupying the center of the cemetery houses the artefacts culled from the interchangeably urban and domestic realms: the conical shrine recalls the coffeepot as well as the industrial tower, and the columbaria and ossuaries blend the typologies of house and memorial. Rossi’s cemetery also draws on Enlightenment models such as Fischer von Erlach’s cemetery and Boullée’s

funerary monuments, yet transposes the themes of life and death through the symbol of the house. In the project’s columbaria blocks, Rossi maintains the formal condition of the house through the use of a pitched roof and windows, yet strips the windows of the elements—the frames, mullions, and glass—which signify occupation. As an emptied opening, the windows of the columbaria instead register absence. Rossi’s project for the Cemetery of San Cataldo engages aspects of its context without resorting to a contextualist strategy, yet plays off of its position as an addition to a cemetery complex comprising a small Jewish cemetery and the existing Costa cemetery. The existing cemeteries—the campo santo (holy ground)—are traditionally enclosed by an external wall. One of Rossi’s decisions involved using a wall to join the cemeteries, which sets into play the vertical and

186

Cemetery of San Cataldo

10. Cemetery of San Cataldo, columbarium exterior.

11. Cemetery of San Cataldo, columbarium interior.

horizontal axes of the traditional Roman town, which grids the cemetery complex. The plan of Rossi’s cemetery project can be read as a diptych with the existing Costa cemetery, and bears both symmetrical and asymmetrical relationships to this context, aligning and slipping out of alignment with different portions of the plan. While Rossi’s plan also responds in its geometric order to the Jewish cemetery, its multiple misalignments further disrupt the classic idea of a part-to-whole relationship. Alternatively, the cemetery project can be seen to take the theme of the enclosure of figured objects from Le Corbusier’s Mundaneum project. Yet Rossi’s plan rethinks Le Corbusier’s utopian gesture within the context of a more problematic relationship of drawing to building. The competition drawings of the Modena cemetery depict the symmetrical axis with an entrance arcade leading to the cubic ossuary, the U-shaped columbaria leading down a central axis to the conical shrine, and the enclosing structures with pitched roofs—many of which can be considered traditional Rossian elements. These reappear within the context of the cemetery project, but have nothing to do with religious symbol-

ism, and instead become urban secular symbols brought into a sacred burial ground. The confusion of symbols between the sacred and profane is part of the textual nature of the project. The context of the idea of Le Corbusier’s plan as generator is called into question as Rossi puts both sectional and perspectival elevations into the plan; these, too, become typological elements deployed without scale and without the context of a single place or time. Another relationship presented at Rossi’s Cemetery of San Cataldo relates to scale, both of the city and of the individual building. This critique is proposed through a single element: the window. Le Corbusier suggested that when a window is too large or too small for a room— that is, when it is not the right size—then one is in the presence of architecture. Therefore an excess in the relationship signifies architecture as an excess in relationship to the functionality of the object. Rossi’s strategy differs slightly from Le Corbusier’s and is more akin to that in Adolf Loos’s house projects, in which the exterior of the house was conceived as different and separate from the interior. The facade in Loos’s case

Cemetery of San Cataldo

187

12. Cemetery of San Cataldo, niches, ossuary.

13. Cemetery of San Cataldo, columbarium.

was a double-sided membrane that articulated the urban scale of the city on one side and the domestic scale of the house on the other. Rossi also developed a similar strategy at Gallaratese, where the standard square window is sized in relationship to the scale of the square outside rather than for the room within, for which it is too large. This distortion of scale indicates that the room can be read as having been tacked onto the facade of the square. Thus the actual facade plane of the building is not to be read as the exterior of a building, but rather as the exterior facade enclosure of the public space. This play of scale articulates an idea about a disjunction in the relationship between facade-part and public spacewhole in terms of the city. The punched square window openings that reappear in the Modena cemetery similarly question the relationship of part-to-whole. The window functions as both the outside and inside of urban scale: the exterior scale of the window differs from that of its interior, from which one can read the slightly smaller frame of the exterior window. The legible change in dimension between inner and outer windows is poignant, in

that the wall thickness houses the square slots for urns. Moreover, the dimension of these square spaces is related to that of the window, a relationship reiterated by windows with a cruciform subdivision. The multiple scales at which the square operates are legible as a honeycombed effect, with the square window as an insert, reproducing the windows at many different scales. The window becomes the register of several repetitions: that of the square form, and that of the cruciform subdivision. Rossi also introduces an uncanny effect produced by other typologies—for example, that of the Tuscan farmhouse, with stucco surfaces, columns, and pitched roof. The square opening punched into the wall is a traditional type of window, yet in this context the window becomes a register of a space of absence and emptiness. Modena is important as much for its drawings as it is for the building, even though it was never built or completed as drawn. Many of the drawings are partial plan views rendered as flattened, one-point perspective views, similar to a cubist still life. Only the heavy outline of Rossi’s drawings attest to another sensibility. The drawings are diagrams and the buildings, in many

188

cases, are built as illustrations of the drawings. For example, the plan, section, and elevation are all deployed in a flattened treatment in the drawing. The shadows become important as the sole indicator of relief. Whether the architecture of the Cemetery of San Cataldo resonates in the same way as the drawing is an issue in much of Rossi’s work. As in the work of John Hejduk or Daniel Libeskind, or perhaps even Palladio, who redrew all of his buildings late in life, it is possible to say that building is a representation of an idea first proposed in a drawing. It could be argued that the portion of the Cemetery of San Cataldo that was built is not as deeply evocative of the ideas that exist in the drawings of the cemetery. Certain of Rossi’s drawings for the cemetery depict the ground plane becoming a skylight, as if describing an interchangeability of ground and sky, of plan and window. In this sense, Rossi’s title “The Blue of the Sky” speaks of a condition in which the sacredness of the ground has disappeared—dissolved, in some sense—into the vast emptiness of the sky. Modena’s physical buildings are powerful in their austerity and reticence; their primitive structural systems and cruciform windows, set within the wall of square burial niches, can be seen as a summation of a Rossian trope: the frame within the frame within the frame. The interrelatedness of these frames is textual as opposed to visual in nature, playing on associations related to typology and analogical forms. The Modena project, especially the drawings, also points toward the development of what Rossi would call his Citta analoga or Analogous City of 1976. In his essay, “L’architettura analoga,” Rossi draws on Jung’s notion of analogical thought as archaic, unconscious, and practically inexpressible in words, a counterpoint to rationalist logic. If analogous thought for Rossi is an interior monologue, then it also offers the possibility of understanding architecture as the product of a

Cemetery of San Cataldo

process of reasoning from parallel cases. The analogical method attempts to understand the city from its urban artefacts, which are elements from different places and different times, and thereby removes architecture from a historical form of logic to another condition of logic, which could be considered textual. Rossi said that to understand his drawings it is necessary to read the text of The Architecture of the City. The drawings of the analogous cities contain the primary elements, the monuments, the special places (or loci) similar to the written ideas presented in his text. Drawings then become another means of architectural thought, not illustrations of or metaphors for architecture. The movement from Modena’s competition drawings to its built components and subsequently to the Citta analoga drawings suggests that the relationship from drawn idea to built form is recursive, and ultimately textual, thus undecidable. Its most important moment is in the Modena cemetery, a work that can be seen as sited between drawing and building.

Cemetery of San Cataldo

14. Plan, persectival sections, and elevations for the columbarium, Cemetery of San Cataldo, Modena, 1971.

189

190

15. The organization of Rossi’s Cemetery of San Cataldo responds to an existing cemetery complex in Modena. The original site plan is important in the context of questioning part-to-whole relationships. Rossi’s cemetery can be seen as a diptych with the preexisting Costa cemetery (on the right-hand side); these two cemeteries flank the smaller Jewish cemetery, which functions as a hinge. While the areas within the walled enclosures of Rossi’s San Cataldo and the Costa cemetery are roughly the same size, Rossi’s San Cataldo establishes a distinct difference from the existing cemeteries. While Rossi’s San Cataldo project becomes part of an entire cemetery complex, it can also be characterized in terms of its dislocation from its specific context.

Cemetery of San Cataldo

Rossi’s San Cataldo cemetery is gridded by vertical and horizontal axes, which recall the cardo and decumanus of the traditional Roman city. There is an impression of symmetry about the hinge of the Jewish cemetery centered between the San Cataldo and Costa cemeteries. This symmetry is reiterated in the symmetry of the Rossi scheme, which has a central axis with square, pyramid, and conical structures in its center. Yet the organization of Rossi’s project can also be read as bi-nuclear about two main volumes (the cube and the cone), producing a constant play between symmetry and asymmetry. The project’s symmetry and use of Platonic forms recalls the Italian ideal cities, as well as cemeteries by Ledoux and funerary monuments by Boullée and Fischer von Erlach. The project can thus be seen as a palimpsest of Roman town, neoclassical framework, and utopian modernist schemes like that proposed, for example, in Le Corbusier’s Mundaneum project.

Cemetery of San Cataldo

16. Rossi’s Cemetery of San Cataldo project is aligned along the top and bottom with the Jewish cemetery, yet not with the dimensions of the top and bottom of the Costa cemetery. The central axis of the Costa cemetery, which is the dominant cross-axial circulation of the plans, aligns through the Jewish cemetery with the major cross axis of the Rossi project, an axis not

191

located in the center of the interior space of the Rossi project, but which runs across the bottom third of the internal divisions. The midpoint of the Costa cemetery articulates, both vertically and horizontally, what can be viewed as eight squares gridding the cemetery space (the four central spaces being true squares, the four peripheral spaces being less than squares).

192

17. Rossi’s project for the Cemetery of San Cataldo appears to be a complete rectilinear enclosure. More specifically, it is articulated as a flattened three-sided U-shape with a top element distinguished from the U-shape with small gaps between the building blocks. This exterior U-shape is then repeated in the interior of the space as the single scaffold organizing each of the symbolic elements along the central axis. This repetition of the outer U-shape and an inner U-shape maintains a tension between elements.

Cemetery of San Cataldo

The exterior rectangular block recalls the Siedlung (housing blocks) by Ludwig Hilberseimer, which have no identifiable front or back, as each side is identical. Rossi confounds the idea of the Siedlung by adding a pitched roof, which effectively mixes different typologies in combining different characteristics from different moments in time. The questioning initiated by buildings at different scales challenges conventional typologies and prevents a single reading of the object.

Cemetery of San Cataldo

18. In the Cemetery of San Cataldo the voided cube, while reminiscent of the voided cube of Rossi’s Turin project, also bears a relationship to his Cuneo monument and serves a similar function as a monument to war dead. Rossi also describes the voided cube as an abandoned house, lacking functional windows and a roof.

193

The square and cube are repeated at various scales throughout the project, from the large scale of the building to the individual scales of the ossuaries and the smaller scale of the squares of the cruciform windows. Rossi is clearly interested in the restoration of figure, but one of ambiguous scale and function.

194

19. Cemetery of San Cataldo, roof plan, axonometric view.

Cemetery of San Cataldo

Cemetery of San Cataldo

20. Cemetery of San Cataldo, ossuaries and columbaria, axonometric view.

195

196

21. Cemetery of San Cataldo, columbaria and ossuary entry sequence.

Cemetery of San Cataldo

Cemetery of San Cataldo

22. Cemetery of San Cataldo, ossuary, axonometric view.

197

198

23. Cemetery of San Cataldo, ascending ossuaries flank the central axis, culminating in the conical monument.

Cemetery of San Cataldo

1. Office for Metropolitan Architecture/Rem Koolhaas, Jussieu Libraries, Paris, 1992–93.

8. Strategies of the Void Rem Koolhaas, Jussieu Libraries, 1992–93 If Walter Benjamin, in his famous and oft-quoted remark, said that architecture is viewed in a state of distraction, then the iconic building so prevalent today may reflect this condition. This prominence of iconic buildings relates to two factors: first, perhaps to a tendency to treat the diagram as an icon; second, a tendency to apply the iconic diagram directly to the problem of generating form. Much of Rem Koolhaas’s earliest work explores the diagram as a symbolic form; for example, the New York Athletic Club becomes symbolic of a discontinuous formal diagram. However, much of his recent work, such as the Seattle Public Library or the Casa da Musica in Porto, privileges the idea of an iconic diagram in that the realized form of the building has a visual similitude to its diagram of functions. It can be argued that Koolhaas’s 1992 project for the Jussieu Libraries takes a position between these two types of diagrams—that is, that it marks an inflection point in Koolhaas’s shift from a symbolic to an iconic diagram, and that this movement is registered through a critique of the diagrams of Le Corbusier and Mies van der Rohe. Le Corbusier’s Palais des CongrèsStrasbourg will be seen to be an important precedent for Jussieu, because in section the project implies a continuity between the ground and the roof. This sectional continuity denies the ground as a datum by suggesting that the ground is conceived as a malleable fabric, capable of being pulled up to meet the roof. In seeking to produce a diagram that is differentiated from that of Le Corbusier without resorting to classical poché, Koolhaas uses the void, which is conceived as an inversion of poché, as a conceptual armature in a series of projects leading up to the Jussieu Libraries. While architects such as Luigi Moretti sought to solidify the void, Koolhaas instead seeks to capture its energy by conceptualizing

202

Jussieu Libraries

2. OMA, Mission Grande Axe, La Defense, Paris, 1991.

3. Le Corbusier, Unité d’Habitation, Marseilles, 1946.

the void as a latent force contained between layers of solid floors. “Strategy of the Void” is the title of Koolhaas’s statement for his Très Grande Bibliothèque competition entry (1989), in which he describes the library as a solid stack from which volumes are carved: “The major public spaces are defined as absences of building, voids carved out of the information solid.” Void thus becomes poché, as a cutting into both the building and urban fabric. In adapting Koolhaas’s title to “Strategies of the Void” for an analysis of the Jussieu Libraries, this chapter argues that the project conceives the void not only as a critique of modernist precedents, but also as a means to rethink the relationship between the subject and the object of architecture, and ultimately to suggest another form of close reading. Koolhaas frames the Jussieu project in the context of its Paris university campus, the expansion of which had been initially truncated by the student uprisings of May 1968. Koolhaas belongs to a generation of architects who describe being powerfully affected by the events of 1968 and the literary and cultural theories generated in its wake. These challenged the humanistic notion of the subject and the modernist notion of the object as a part of a rational whole. While the work of a

prior generation, including Venturi, Rossi, and Stirling, used strategies of fragmentation and materiality to critique the modernist idea of the whole, the fragment cannot help but to recall an absent “whole,” and in that sense maintains, at least in concept, a traditional part-to-whole relationship. The post-1968 generation engaged in a different idea of the part-to-whole dialectic, which reflected alternative ways to view the subject, drawing on structuralist and poststructuralist theory. For example, Jacques Lacan’s notion of the split subject and the development of the conscious subject as a function of its projected/ reflected image suggests a different idea of partto-whole in the context of the subject/object relationship. The split subject is no longer considered a fragmentary part of a whole, and the concept of “wholeness” becomes increasingly untenable. A second aspect of Koolhaas’s strategy of the void involves creating situations which introduce a voyeuristic gaze and in which the voided space both blocks direct vision and reveals supposedly hidden elements. Koolhaas, like Le Corbusier before him, presents a didactic image of a hand lifting up a corner of the ground. For Koolhaas, this image suggests that the surface is malleable and pliable, that it is no longer specifically related

Jussieu Libraries

203

4. New York Athletic Club, section, 1978.

5. OMA, Parc de La Villette model, Paris, 1982.

to the ground, and that it partakes in a vertical continuum. More importantly, in lifting up the city’s fabric, a hidden aspect of its infrastructure is revealed as an underlying object. With his first book, Delirious New York (1978), Koolhaas presents a radical conception of an architecture using the New York Athletic Club as a model. It is not a traditional diagram of function, but is rather a diagram symbolic of the dismantling of the traditional physical contiguity of part-to-whole relationships. For Koolhaas, the New York Athletic Club diagram proposes an idea of discontinuity in its questioning of traditional functional adjacency relationships. Koolhaas suggests that the presence of the elevator denies the need for contiguous functional relationships in a skyscraper such as the New York Athletic Club, whose stacked functional layers happen to be physically contiguous, yet there is neither functional need nor meaning in their physical contiguity. Such a diagram, one that presents what can be called contiguous discontinuity, also appears in Koolhaas’s entry for the 1982 Parc de La Villette competition in Paris. One of Koolhaas’s most diagrammatic projects, the park is depicted as a series of horizontal programmatic strips. La Villette’s didactic plan places a strip of theme parks next

to a strip of playgrounds, and a strip of discovery gardens beside a strip of museums. These functional conditions do not require such contiguous spatial relationships. In envisioning the ground plane as a series of strips, Koolhaas’s La Villette proposal breaks with a figure/ground urbanism to propose a montage of programmatic lateral bands linked by the strong vertical of a proposed promenade. This is a clear echo of the New York Athletic Club’s discrete programmatic layers linked by the elevator. This leads Koolhaas to suggest that not only the isolated building but also the field of the city can be rethought, beginning with the very distribution of the ground itself. This denial of the ground as a datum begins to appear in Koolhaas’s project for the Très Grande Bibliothèque in 1993, which further develops Koolhaas’s diagram of contiguous discontinuity in the section of the building. The library is conceived in section, not as a series of free plans, but as a vertical stacking of differentiated horizontal planes that do not share a contiguity of program from one level to the other. Only the structural grid links the floors, but in pragmatic and not theoretical terms. While it could be argued that the Très Grande Bibliothèque’s nine main columns are organizing elements, these columns are not

204

Jussieu Libraries

6. OMA, Très Grande Bibliothèque, competition model, Paris, 1998.

thematic of a free plan in that they do not provide a regular backdrop for the figuration of free forms. Koolhaas denies the thematic of the columns, which disappear and reappear randomly, depending on how they intersect with the walls. This structural system also could be seen as a critique of Mies’s umbrella diagram. Large Xshaped columns at the base of the Très Grande Bibliothèque create a giant truss that supports the building. Whereas Mies hung his buildings off of a truss, as in Crown Hall and the National Gallery in Berlin, Koolhaas places the truss under the building, reversing Mies’s umbrella diagram. The spaces for the library stacks of the Très Grande Bibliothèque are organized around the structure with a discontinuity reminiscent of La Villette, while the circulation is treated as a series of figured objects, much like Le Corbusier’s ramp at Strasbourg. Rather than the floors, it is the elements of circulation which are given form—and as cuts in the floors, or holes that punch through walls, these voids capture the figured energy of the project. However, there remains a formal perimeter condition that is maintained in a classi-

cal frontispiece, which distinguishes the project’s back from the front and differentiates the front from the sides. The figured energies of the project’s diagonals, horizontal cuts, and plans never disrupt the clear geometric boundary of the building’s edge. This condition is one not so much of floor levels, but rather becomes a matrix of volumes that do not obey any horizontal datum. The 1990 project for a convention center at Agadir, Morocco, presents a critique of the horizontal extension of space proposed by Le Corbusier with his Maison Dom-ino diagram. While the Maison Dom-ino located the individual in relationship to a larger context, Koolhaas establishes the interrelationship of individuals as a kind of free play of the individual in space and time. Agadir presents an intermediary step between the Très Grande Bibliothèque and the Jussieu Libraries. Instead of a horizontal continuum there are three possible sectional interpretations: one, the ground becomes modulated; two, the sectional space becomes modulated where the ground rises substantially in the section; and three, the horizontal is no longer a continuum. In

Jussieu Libraries

205

7. OMA, Palm Bay Seafront Convention Center, Agadir, model, Morocco, 1990.

the first, the ground is no longer a datum, and the roof is no longer part of a ground/floor/roof continuum, but becomes something else. This is neither the umbrella diagram of Mies nor the horizontal continuum of Le Corbusier. Rather, it manifests an energy that causes an undulation in section. Agadir is no longer the didactic horizontal continuum of Cartesian x- and y-axes but rather proposes a figured section where the horizontal is a figured void cut out of an equally figured solid poché. Koolhaas introduces a horizontal disturbance in section as a dominant mode of discourse rather than the vertical extrusion typical of classical architecture and present in the work of postwar American architects such as Louis Kahn. The modulation of section creates conditions in which space can be occupied by a subject who may become a voyeur while hidden from the view of another subject and vice versa. The resultant coup d’oeil and peripheral views shift the focus of opticality from the physical object to the subject, who looks through, around, beneath, above, and at spaces, becoming part of

a different kind of spatial relationship between subject and object. This imitation of a voyeuristic space is what Jeffrey Kipnis calls a performative discourse. While the Jussieu Libraries project of 1992–93 is essentially a vertical project, the warped floors signal an evolution from Agadir. Yet unlike Agadir, whose warped floor remains horizontal, the floors at Jussieu are warped in section to the extent that floors rise to touch the floor of the next level. The floors become a series of continuous surfaces which tilt from the horizontal. Conceptually the circulation and the floor levels become a continuous surface, yet the project retains a discontinuous relationship in terms of a Cartesian axiality. No longer is the subject in a one-to-one relationship with another subject but, because of the inclined planes, it views other subjects—and is viewed by them— as objects. There are spaces in Jussieu which allow a voyeuristic tendency to take place. This manipulation of the visual field begins with the Jussieu Libraries and continues, for example, in the Seattle Library.

206

Jussieu Libraries

8. OMA, Jussieu Libraries, section drawing.

The unfolded section of the Jussieu Libraries competition entry becomes a critique of Koolhaas’s earlier New York Athletic Club diagram, and produces an entirely new diagram that focuses on the internal continuity of surfaces. The diagrammatic icon of circulation now becomes literal circulation. The sections also suggest that the only real volumes in the building are the interstitial spaces between floors. These are bounded as figures by the bow and bend of the floor planes, and in that sense can be viewed as residual, while the linking of circulation with the floor planes suggests that a diagram of circulation is its governing form. At Jussieu, as in the Très Grande Bibliothèque, Koolhaas retains aspects of classical architectural notation, such as the differentiation of the building’s front, back, and sides. The differential between the building’s sides, one of which appears almost eaten away by interior voids while others are mostly intact, also maintains the classical legibility of front and back. Finally, the open-

ing provided by a type of porte-cochère, a porticolike element marking the transition from exterior to interior at ground level, twists up through the project, seemingly autonomous from the building’s formal organization. The cross section of the Jussieu Libraries expresses the discontiguous relationships of program, and the continuity of the ramped floors produces an entirely other section, no longer the stacked layers of the New York Athletic Club or La Villette. The section reveals that the only volumes in the buildings are trapped voids, interstitial spaces between floors, which are not continuous. The drawings reflect a new ethos of perception, and suggest a departure from architecture as a product of close attention. Koolhaas’s strategies of the void are important, linking the issue of the void—present in the postwar work of Moretti and Venturi—to new methods of working that confront questions about part-to-whole relationships, inattention versus close attention, disjunction versus fragmentation.

Jussieu Libraries

These voided elements, which are rendered as gray solids in the large competition model, cannot be read as purely poché, nor as purely figure, because they are residual spaces created by the continuity of the floor planes. These darkened segments emphasize the connections between levels, allowing the horizontal to be read as a continuous flow which is only interrupted by the gray elements. The energy, whether it is registered in different shapes, horizontal cuts, spirals, or ramping floors, does not disturb the edge. This contrasts with the edge stress of cubist painting and the centrifugal stress of Venturi’s and Moretti’s work. At Jussieu, there is neither a centralized stress, nor a stress at the edge. Rather, the stress is diffused; it appears in the different layers of the object, particularly when the only centralizing figure is clearly thrown off center, but not enough to be seen as moving toward the edge. The energies which figure the voids are equidistant from the edge and from the center, carefully balanced in a kind of dynamic

207

equilibrium. The void produces a space of unresolved tension between center and edge. It is this irresolution that introduces what has been called here the idea of undecidability. Similarly, while there appears to be little continuity to the circulation, it remains contained within its cubic framework: the floor and the circulation are joined as a single element. The Jussieu Libraries become the model for many of the later projects, including the Seattle Library, the Casa da Musica in Porto, and the Dutch Embassy in Berlin, all by Koolhaas. It becomes clear that the influences of the diagrams of Le Corbusier and Mies are inverted, and are transformed into figured, volumetric, objectlike elements as “strategies of the void.” The critique that begins with Strasbourg and the lessons learned in the Très Grande Bibliothèque and at Agadir coalesce in the warped sections and figured voids of Jussieu. Yet the critical and theoretical arguments condensed in Jussieu’s model, drawings, and section give way to an increasingly iconic use of the dia-

208

gram by Koolhaas at the Seattle Public Library and at the Casa da Musica. These buildings adopt a visual similitude with the ideas posited in their diagrams. Because the diagram of discontinuous layers of program retains a visual similitude to the architectural form, Koolhaas’s work begins to suggest another attitude toward close reading. When the reading of the diagram approaches the reading of the building, close reading is no longer necessary. While this may not be the case in Jussieu, certainly the projects that follow, Seattle and Porto, seem to belie these earlier directions. They give up close reading for the immediacy of shape and a more popular appeal: the diagram as logo and branding.

Jussieu Libraries

Jussieu Libraries

9. Jussieu Libraries, third sublevel, plan and sections.

10. Jussieu Libraries, second sublevel, plan and sections.

11. Jussieu Libraries, first sublevel, plan and sections.

209

210

Jussieu Libraries

12. Jussieu Libraries, mezzanine-entry level, plan and sections.

13. Jussieu Libraries, first floor, plan and sections.

14. Jussieu Libraries, second floor, plan and sections.

Jussieu Libraries

15. Jussieu Libraries, fourth floor, plan and sections.

16. Jussieu Libraries, roof level, plan and sections.

211

212

17. The Jussieu Libraries project responds to several of Le Corbusier’s diagrams: the Maison Dom-ino, squared and stacked, can be considered one such precedent. Le Corbusier’s Dom-ino diagram located the individual clearly in relationship to a context which was elevated from the ground. The stacking function maintains layers as discontinuous elements.

Jussieu Libraries

18. A stacked parking-garage diagram demonstrates the possibility of a continuous relation between the stacked floor levels. This registers as a disturbance in the horizontal section.

Jussieu Libraries

19. An analysis of the horizontal disturbances in the stacked layers of the Jussieu Libraries demonstrates the shifts, ramping, and discontinuities that serve as a critique of the Dom-ino section. It no longer suggests that there is an extension of Cartesian space into a horizon.

213

20. At Jussieu, the distinction of floor and ramp is effaced. A little noted aspect of Koolhaas’s discourse is the cant of the roof plane, which indicates that this no longer represents a flat extension of a horizontal continuum of space, but instead produces a perspectival twisting in space. This false perspectival twisting continues through to the roof, marking the spiraling energies upward.

214

21. The discontinuous circulation and warping floor levels create a condition in which the only discernible volumes are the interstitial spaces between floors. Certain of these interstitial spaces are highlighted in the competition model (gray areas). These zones are articulated as part of the relationship between one floor and the other, yet these figured voids are not continuous. These spaces could also be considered as partial figures. These figures disrupt the reading of a continuum from floor to floor in terms of function. The discontinuous vertical circulation, horizontal circulation, and condition of objects which are not located on a ground can be seen as a critique of the notion of ground as a datum.

Jussieu Libraries

22. (Right) In an exploded axonometric diagram, the figured voids can be characterized as cuts (A), tears (B), and holes (C). The horizontal layers that constitute the floors are opened by these cuts, tears, and holes, suggesting that the floor has become a fabric joining the roof and ground.

Jussieu Libraries

215

A

C B C

B

A

C

A A C

B A

A

B C A

A

B

B C A A

B B A A

216

23. Lifting the horizontal plane, as if cutting away the surface, produces the idea that the surface is malleable and pliable. This also implies that the surface is no longer necessarily only related to the ground, but that it becomes part of the vertical continuum.

Jussieu Libraries

24. Jussieu is framed as a square. If envisaged as a sequence of folded planes in a soft and pliable material, it is apparent that the distinct edges of the folded plane maintain the geometrical form of the square.

Jussieu Libraries

25. The folded condition of a single sheet of pliable material produces spaces whose variable section registers the deformation of the plane. The folded plane is a diagram for the Jussieu Libraries.

217

26. The fabric of the folded plane registers a series of cuts and shears. Certain cuts are clearly ramps; others are void figures unto themselves, lacking clear function. These voids do not seem to follow a clear organization, but rather disrupt continuity and the idea of a single thematized reading.

218

27. While the Jussieu project has a regular structural grid, certain grid elements tilt and bend, demonstrating that the regularity of the grid is not thematic. Yet the structural grid indicates that there are narrow bays flanked by a series of wider bays on the inboard side, which take on a different organizing structure than those at Le Corbusier’s Palais des Congrès– Strasbourg.

Jussieu Libraries

28. In addition to Jussieu’s large-scale ramping circulation, there are smaller orthogonal figures created by stair cores and giant columns, similar to those in the Très Grande Bibliothèque. These forms emphasize vertical circulation; the conception of Jussieu as a folded, lifted, and cut plane presents a different spatial idea. Thus, the circulation is not treated at either the large scale of the ramp or the smaller scale of inner circulation as a figured object.

Jussieu Libraries

29. Jussieu Libraries, circulation diagram.

219

220

30. Jussieu Libraries, third sublevel, axonometric view.

Jussieu Libraries

Jussieu Libraries

31. Jussieu Libraries, second sublevel, axonometric view.

221

222

32. Jussieu Libraries, first sublevel, axonometric view.

Jussieu Libraries

Jussieu Libraries

33. Jussieu Libraries, mezzanine-entry level, axonometric view.

223

224

34. Jussieu Libraries, first floor, axonometric view.

Jussieu Libraries

Jussieu Libraries

35. Jussieu Libraries, second floor, axonometric view.

225

226

36. Jussieu Libraries, third floor, axonometric view.

Jussieu Libraries

Jussieu Libraries

37. Jussieu Libraries, fourth floor, axonometric view.

227

228

38. Jussieu Libraries, roof level, axonometric view.

Jussieu Libraries

1. Studio Daniel Libeskind, Jewish Museum, Berlin, Germany, 2000.

Libeskind Layout 02 08.indd 230

4/9/2008 11:44:29 AM

9. The Deconstruction of the Axis Daniel Libeskind, Jewish Museum, 1989–1999 In the 1970s, Rosalind Krauss gave two important lectures titled “Notes on the Index” at the Institute for Architecture and Urban Studies in New York, which subsequently appeared as essays in the journal October, numbers 3 and 4, published in the spring and fall of 1977. Krauss’s discussion of the index drew on the distinctions between icon, symbol, and index first put forward by C. S. Peirce. If, for Peirce, an icon had a visual likeness to its object and a symbol had an agreed upon or conventional meaning, then an index was a trace or record of an actual event or a process. The index displaces the movement outward of both the icon and the symbol to a signified, while referring inwardly to its own processes. But most important for architecture, the index is also closely tied to the issue of presence and absence. For example, Robinson Crusoe’s discovery of footprints in the sand gave him the idea that there was life on the island without having seen the living being itself. The index, Krauss describes, “establishes its meaning along the axis of a physical relationship to its referent.” The footprints are the trace of a previous presence, yet also record the current absence of that presence. The footprints in the sand imply several registers of an index as both imprint and trace. When the foot lifts up from the sand it leaves the imprint of human presence in the depressed sand, yet at the same time a layer of sand clings to the bottom of the foot. Thus, the trace remains on the object—the foot—while the beach registers the imprint of the foot, or the human presence. Another component of the footprint as index is its notation of time; the footprint registers the span between the moment of human presence in making the imprint and that of human absence. The idea of presence and absence suggests a significant difference between an idea of an index or trace in a

Libeskind Layout 02 08.indd 231

4/9/2008 11:44:48 AM

232

Jewish Museum

2. Daniel Libeskind, Micromegas, Time Sections, 1978.

linguistic or photographic context, and an index in a physical context like architecture. Krauss suggests that language presents us with an historical framework which preexists its own being, and therefore joins language to a metaphysics. The idea of an architectural language becomes problematic when it assumes that any historical context is a stable entity. Because architectural representation is presumed to be a stable relationship between a sign and its object, the idea of the index in architecture seeks to undermine the idea that its language is a decidable physical presence with a one-to-one relationship to a signified. Krauss has suggested that the importance of the index counters the overwhelming physical presence of an object, in that it is a trace of some other object, and not a sign or a representation of the thing itself. Thus an index in architecture attempts to deny pure presence by presenting a condition of absence in presence. If a metaphysical presence presupposes a notion of fullness because it is present in physical terms,

Libeskind Layout 02 08.indd 232

an index undercuts such a metaphysical fullness because its referent is to a prior condition—or in other words, a condition of absence. In the second part of “Notes on the Index,” Krauss considers the photograph as another example of an index, as it introduces a set of abstractions which include both process and absence. It is the absence of the actual event, or the photograph’s actual relation to the past event, that is signified. The photograph is considered the index or trace of some condition of fact or reality, and while it is a physical object in itself, it also reproduces signs of former presences, and therefore undercuts fullness by introducing these absences. Krauss describes the index as the mute presence of an uncoded event, one that operates without conventions. She cites Gordon Matta-Clark, whose cutting of holes in floors and facades of buildings creates the ultimate icon for an indexical architecture. Krauss describes these cuts as akin to the linguistic shifter, which she describes in October 3 as a term in linguistics

4/9/2008 11:44:48 AM

Jewish Museum

233

3. Line of Fire, plan, 1988.

which is filled with signification only because it is empty. A word like this, as in this table or this chair, lends signification to its referent but remains empty itself, or in Krauss’s terms, is “an empty pronominal sign.” The cut in Matta-Clark’s work becomes an empty sign of an event, a trace of someone’s having cut into the building. The cut also empties the metaphysical content of the house because the house is no longer functioning as a house. Once its enclosure is breached it can no longer shelter: its content and functions are emptied. If the form of a house with a pitched roof harbors metaphysical and meaningful implications related to the image and function of shelter, then these meanings are shattered by any kind of cut. Not only is the cut in itself a trace of the cutting, but in the act of cutting the house reduces its metaphysical content. The cuts in the work of MattaClark become the index of the absence, displacing a metaphysics of presence with what could

Libeskind Layout 02 08.indd 233

be considered a more literal presence. The index thus traces the movement from metaphysical presence to pure presence itself. The logic of such indexical signs seeks to undermine the iconic and symbolic, yet the index can easily be transformed into an icon of its own indexicality. Daniel Libeskind’s Jewish Museum in Berlin is just such a project. It is one of the important realized buildings of what can be called the indexical project in architecture. This building engages the index as a critique of architectural persistencies, in particular that of linear axiality, which can be considered fundamental to Cartesian and classical space. Libeskind’s indexical work begins with his 1978 Micromegas drawings, a series of lines that attempt to question Cartesian space. The Micromegas drawings in fact were not merely drawn lines but tectonic architectural lines, indexical markings of conditions in space and time on a virtual object; the series is an index of the denial of any Cartesian coordinates or picture plane.

4/9/2008 11:44:49 AM

234

Jewish Museum

4. City Edge competition, model, 1987.

With lines that shatter space and undermine iconic and symbolic references, the Micromegas drawings initiate this breakdown of axial space. If all sites contain axes, and if site specificity is an idea relating a particular building to a site, then all buildings contain axes which relate building to the subject’s movement through it. Libeskind, in presenting the axis as a trace, an inaccessible void, and a series of discontinuous segments, offers a critique of axiality, site specificity, and ultimately the classical subject/object relationship. One of the dominant persistencies of architecture is the traditional movement of the subject from the entry of a building through its major spaces, which are typically perceived through symmetrical sequences. Whether this path is called a promenade, a marche, or merely a symmetrical x-axis is of little concern here, but to deny the idea of the subject’s understanding of space through Cartesian coordinates is to challenge one of the persistencies of architecture. Libeskind’s project for the City Edge competition of 1987 continues this exploration of the axis with a linear project sited to slice across the grain of the divided city of Berlin: in this case it is a

Libeskind Layout 02 08.indd 234

political divide that initiates the disruption of the axis, but it is a physical gesture which ultimately destabilizes the continuity of the part-to-whole relationship. The indexical project of Libeskind’s Jewish Museum becomes more apparent in its direct relationship to his 1988 work Line of Fire, an installation in Le Corbusier’s Unité d’Habitation in Briey-en-Forêt. The ground level of Le Corbusier’s Unité provided an axial space defined by pairs of massive pilotis. In this and other Corbusian buildings, these symmetries identify the path of the subject in much the same way as in a Palladian villa, the symmetry of columns providing a simple geometric means of recognition. These symmetrical pairings in architecture make the time of the subject’s movement and the time of the object (or its physical axis) the same. When the path of the subject does not seem to correspond with the form of the space, then the time of the subject’s movement and the time of the object become differentiated. Libeskind’s Line of Fire project does just that by disrupting the possibility of axial movement around and through the Unité’s pilotis. The zigzagging form

4/9/2008 11:44:50 AM

Jewish Museum

235

5. Line of Fire, plan 1988.

6. Line of Fire, installation, Briey-en-Forêt, 1988.

of his installation denies the idea of an axis, both as a real pathway and as a concept of an axis of symmetry. When the installation was placed in an architectural context it disrupted the metaphysical idea of an axis, that of a single movement through space and its engagement with an idea of axis. This disruption questions the classical notion of the continuity and symmetry of x-axes, while creating a disjunction in time and a dislocation in space. It is that distance in time that causes the indexical quality to become part of the project in the disruption of the axis. Line of Fire establishes a series of different axes which counter the subject’s path and call attention to the discrepancy between the experience and comprehension of space. The subject’s mental tracing of a zigzag route is disorienting, even though an implied axis remains present. Libeskind’s installation suggests that this axis is not a pure and continuous vector, but one that may be modified by historical circumstance—in this case referring to the destinations of deported Jews in Nazi Germany. Libeskind, like Henri Bergson before him, questions the relationship of the time of the object to the time of the subject. Libeskind

suggests that the time of the experience can no longer be assumed to be calibrated with the time of the object, because the time of the object will not reveal itself with the path that the subject is taking. This is the central issue of Libeskind’s Jewish Museum, which is one of the first real evocations of an attempt to deny the continuity of the axial path to the object of architecture. The Jewish Museum in Berlin in one sense is itself a repetition, a trace and an index of the Line of Fire exhibition. In fact, mirroring Line of Fire about a horizontal axis produces the identical form of the Berlin project. It could be argued then that the axiality challenged by Line of Fire is displaced again, this time rotated in its context to produce the Berlin project. Libeskind’s own argument that the Berlin project represents the fragmentation of a Jewish star, or is an index of the points in Berlin where the Jews were transported out of the city, has little to do with the argument here and its relation to Line of Fire. While Libeskind will always claim that the Berlin Museum comes from connecting the lines from the points of Jewish embarkation from Berlin to the death camps and that the intersection of these

Libeskind Layout 02 08.indd 235

4/9/2008 11:44:52 AM

236

Jewish Museum

7. Jewish Museum, preliminary working model, 1990.

lines produced this form, the correspondence in form between the Jewish Museum and the Line of Fire seems to suggest other interpretations. The original idea for the building emerged from a competition for an extension to the neoclassical Berlin Museum for German history. The competition model and some early drawings suggest that Libeskind originally connected the Line of Fire to the museum as an extension, not as an entirely new Jewish Museum. The first working model of the scheme for the Jewish Museum had canted and battered exterior walls. In addition to the zigzagging form of the museum disrupting the x-axis, the walls were canted at various angles so that the vertical y-axis was similarly challenged. But in both the models of the project and in the realized building there remains a trace in the roof of the continuous axial path that is disjuncted, first by the angled body of the museum, and next by the actual inaccessible void which extends down to the floors from the roof trace, denying any continuity along the x-axis. The facades are already marked by indexical cuts similar to Matta-Clark’s, which articulate apertures in a building in a radically different

Libeskind Layout 02 08.indd 236

manner from the conventional relationship of windows to their respective interiors. Windows typically reveal the scale of rooms and reflect the scale relationship of interior to exterior. Such a relationship is fragmented by these cuts on the Jewish Museum facade: certain of these cuts are small, others large, but they bear no relationship to the interior spaces. In Libeskind’s museum, the apertures are divorced from function, registering the conflict between interior and exterior scales, as well as between light source and the possibility of functional exhibitions. These cuts recall the lines in Micromegas, and shift the role of the window from function to indexical marker. The cuts work in a similar way to the openings in the sloping wall of Le Corbusier’s chapel at Ronchamp in that they relate to an implied vertical datum. Ronchamp is successful in that its sloped wall and its cut-outs play against an implied but nonexistent vertical plane. There is nothing more indexical in plan and elevation than the random, arbitrary cuts throughout the museum. This is similar to MattaClark’s cuts as the traces of that cutting, resembling a photographic plate of a series of arbitrary

4/9/2008 11:44:56 AM

Jewish Museum

237

8. Jewish Museum, zinc model, 1990.

9. Jewish Museum, site model.

gestures. It could be argued that these gestures relate to the arbitrary and random execution of Jews by the German state under Hitler. Yet the dominant mode of discourse is one of traces of some event that internalizes “no meaning,” which in itself takes on a symbolically meaningful relationship to the willful, i.e. “meaningless,” killing of people. In one sense, the cuts establish the condition of the arbitrary, which then relates to a real meaning, which is the ultimate condition of the arbitrary. Libeskind’s Jewish Museum oscillates between the indexical and the symbolic, as the indexical register triggers a symbolic key, which then returns the symbolic to the arbitrary. In Libeskind’s final rendition of the Jewish Museum, the conceptual resonance for the cuts in the facade as indexing the play between y-axis and canted wall no longer exists, as the realized walls are no longer canted. However, the cuts still challenge the traditional use of windows for orientation, for the narrow cuts of light produce a strong contrast with the museum’s dark walls; the cuts create what in painting is called a halation, which causes the light to produce an afterimage on the retina—in other words, an index of

perception on the eye itself. Thus, again the cuts become indexical, not so much of the political/historical narratives articulated by Libeskind as of the act of cutting itself. If both Line of Fire and the Jewish Museum present a disruption of the x-axis, the scale of the Jewish Museum allows Libeskind to articulate this disruption in more precise ways. Circulation plays a key role in a critique of the need to understand space through movement. The stairs of the museum do not provide connection, but in one sense function to interrupt continuous movement. The location of the staircases further denies any continuity of movement. It is important to understand that one cannot follow a horizontal route, nor remain at a horizontal level when moving through the museum. The subject’s movement along a Cartesian conceptual axis is interrupted, as is the ability to remain on a single comprehensible horizontal datum. The length of the building represented by spaces on the same floor cannot be experienced as the typical horizontal datum provided by a floor plane. Rather, the horizontal axis must be traversed through a sequence of interrupted levels, as stairs and

Libeskind Layout 02 08.indd 237

4/9/2008 11:44:57 AM

238

Jewish Museum

ramps move the subject across the series of voids enclosed in the museum. Such disruptions frustrate programmatic and formal expectations, but more importantly separate the time of the experience of space from the comprehension of its organization. The traditional subject/object relationship depended on a continuous horizontal datum that could be traversed, but the Jewish Museum denies that possibility. The sequence of inaccessible voids at the center of the museum, while variously described in terms of poetic resonance, can also be interpreted as a continuation of Libeskind’s critique of Cartesian axiality. These voided zones slice through the center of the zigzagging form of the museum, so that a void zone, which might be comprehended by a legible axis, is never experienced as such and instead becomes one of the devices to impede the subject’s movement. The visual parameters of what is being seen do not produce an overall image or gestalt, but produce something that is difficult to extrapolate from the experience of the building. Not only are the axes and floor levels denied as parts that ultimately relate to a whole, but the overall impression of a functioning whole is denied by the parts. Ultimately Libeskind’s museum is a struggle between the indexicality of the building and the symbolic resonance of the rhetoric. Perhaps it was the specific context of the earlier projects that provided the necessary rhetoric implied in the work. Later projects suggest that the symbolic predominates over the indexical and diagrammatic nature of the earlier work. The new work becomes more open to expressionist gestures that move it closer to an iconic project, no longer requiring the close reading of indexical traces. As such, the Jewish Museum represents the cusp of a relationship between the indexical and the iconic.

Libeskind Layout 02 08.indd 238

4/9/2008 11:44:59 AM

Jewish Museum

239

10. Jewish Museum, underground level plan.

11., Jewish Museum, first-floor plan.

Libeskind Layout 02 08.indd 239

4/9/2008 11:44:59 AM

240

12–13. Jewish Museum, envelope and circulation diagrams. The void spaces (dark gray) prevent movement

Libeskind Layout 02 08.indd 240

Jewish Museum

along a level horizontal axis, while ramps and stairs further disrupt the continuous line of movement.

4/9/2008 11:45:00 AM

Jewish Museum

Libeskind Layout 02 08.indd 241

241

4/9/2008 11:45:00 AM

242

14-15. Line of Fire, plan and projected elevations, based on the 1988 plan drawings. An incised pattern of crossing lines, resembling an uneven grid, is applied to

Libeskind Layout 02 08.indd 242

Jewish Museum

the envelope of the installation piece. Highlighted in red are rectangular elements which resemble the markings on the envelope of the Jewish Museum.

4/9/2008 11:45:01 AM

Jewish Museum

16. Jewish Museum, diagram of envelope highlighting rectangular motifs that appear to be traces based on surface markings in Line of Fire. These suggest that the

Libeskind Layout 02 08.indd 243

243

indexical project registers on the museum’s envelope as well as its form.

4/9/2008 11:45:01 AM

244

Jewish Museum

17. Jewish Museum, roof plan, including underground level, axonometric view.

Libeskind Layout 02 08.indd 244

4/9/2008 11:45:01 AM

Jewish Museum

245

18. Jewish Museum, underground level, axonometric view.

Libeskind Layout 02 08.indd 245

4/9/2008 11:45:02 AM

246

Jewish Museum

19. Jewish Museum, ground-floor plan, axonometric view. The highlighted void forms the horizontal axis.

Libeskind Layout 02 08.indd 246

4/9/2008 11:45:02 AM

Jewish Museum

247

20. Jewish Museum, first-floor plan, axonometric view. The highlighted void forms an inaccessible volume on this floor.

Libeskind Layout 02 08.indd 247

4/9/2008 11:45:02 AM

248

Jewish Museum

21. Jewish Museum, second-floor plan, axonometric view. The highlighted void forms an inaccessible volume on this floor.

Libeskind Layout 02 08.indd 248

4/9/2008 11:45:04 AM

Jewish Museum

249

22. Jewish Museum, third-floor plan, axonometric view showing the volumetric form of the void.

Libeskind Layout 02 08.indd 249

4/9/2008 11:45:04 AM

250

Jewish Museum

23. Jewish Museum, exploded sectional axonometric through void axis, revealing the figured voids.

Libeskind Layout 02 08.indd 250

4/9/2008 11:45:06 AM

Jewish Museum

251

24. Jewish Museum, section through void axis, axonometric view. The circulation around the void spaces is highlighted.

Libeskind Layout 02 08.indd 251

4/9/2008 11:45:08 AM

252

Jewish Museum

25. Jewish Museum, unfolded. The trace of the voids, highlighted in red, retains the zigzagging form of the actual plan.

Libeskind Layout 02 08.indd 252

4/9/2008 11:45:08 AM

Jewish Museum

Libeskind Layout 02 08.indd 253

253

4/9/2008 11:45:09 AM

254

Jewish Museum

26. Jewish Museum, roof plan, axonometric view.

Libeskind Layout 02 08.indd 254

4/9/2008 11:45:11 AM

Libeskind Layout 02 08.indd 255

4/9/2008 11:45:11 AM

1. Frank O. Gehry & Partners, Peter B. Lewis Building, Weatherhead School of Management, Case Western Reserve University. Cleveland, Ohio, 2002.

10. The Soft Umbrella Diagram Frank O. Gehry, Peter B. Lewis Building, 1997–2002 Most of the diagrams discussed in this book, whether they are iconic, symbolic, or indexical, secure their importance by displacing an original and preceding condition. Greg Lynn, in his Embryological House and in several recent projects, proposes another kind of diagram, one that has no originary condition. Lynn suggests that form harbors, as an integral aspect of its being, conditions which he calls form’s own diagrammatic necessity. This internal logic renders it possible to produce diagrams that refer not to an external transcendental signified, but to their own operations. Such diagrams may not depend on any of the a priori notions that could be assumed to determine architecture, such as site or program. If that is the case, for example, if the problematic of part-to-whole—the relation of a building to its site, of its inside to its outside, or of the building to the city—is no longer necessarily an a priori truth, then the necessity of a part-to-whole relationship is in fact undermined, as is the type of close reading that searches for this relationship. Rather than the part, Lynn’s work deals with the component as an infinitely repeatable entity. He suggests that it is possible to work on components—whether they are components of a building or components of the city—which have no necessary relationship to the whole, nor to a precedent, but result from a set of internal or computational logics. Lynn argues that a computer algorithm operates both in the Peircian sense of the symbol and the index, in that its meaning is legible as a representation of such processes and that these operations take place over time, which is recorded in an indexical manner. These digital processes, Lynn suggests, do not depend upon an external relationship of site, program, or some prior architectural necessity.

258

Peter B. Lewis Building

2. Peter B. Lewis Building, study model, June 1997.

3. Peter B. Lewis Building, study model, June 1997.

Architecture may always look like architecture because it shelters, encloses, resists the forces of gravity, and is sited, yet these thematic operations do not necessarily need to refer to previous disciplinary architectural conditions— in other words, to precedents or what are called here disciplinary persistencies. Lynn’s argument implies that these prior conditions of architecture’s own disciplinary precedents are not necessarily relevant to those of the future, given that these algorithmic processes are in fact unfamiliar to architecture. This effectively suggests that it may not be necessary to study the history of architecture or the history of the twentieth century in order to be able to work diagrammatically using digital processes. Lynn’s argument is probably the most relevant summation to date on the condition of the digital as a critique of precedents. This argument about the role of the digital in undermining architectural precedents is useful in considering the relationship of digital and analogic processes in Frank Gehry’s Peter B. Lewis Building for the Weatherhead School of Management at Case Western Reserve. It is necessary to distinguish first Lynn’s idea of the digital from that of Gehry’s, and second, between the conceptual in Lynn and the phenomenal in Gehry. There is little question that there

is a significant separation between Lynn’s and Gehry’s invocations of the digital. While Gehry might argue that his work is the result of computation, it could similarly be argued that Gehry occupies a terrain that is not as clearly defined, being sited between personal expression—or analogic processes—and digital processes. Gehry’s diagrams originate in analogic methods, and the subsequent digital work is one of reproduction of these forms. Perhaps it would be more productive to say that the diagram in Gehry’s work is iconic and, more importantly, it situates his work in the realm of the phenomenal. The crucial difference between the conceptual and the phenomenal lies in the domain of close reading, with the nexus of attention shifting from the eye to the mind in the conceptual, and from the mind to the eye in the phenomenal. Little of Gehry’s previous work can be called diagrammatic, yet certain projects suggest that Gehry has always had an implied diagram, which bears some relationship to Mies’s concept of the umbrella diagram. Gehry’s diagram could be called a “soft umbrella,” resembling a dropped parachute or napkin, which settles in various ways over an internal organization of spaces and structure. This type of diagram depends on the articulation of the roof and the roof’s impact on the

Peter B. Lewis Building

259

4. Peter B. Lewis Building, model, September 1997.

5. Peter B. Lewis Building, sketch, October 1997.

section; the plan becomes residual to the process. The dropped napkin or soft umbrella diagram is subsequently translated into a digital format. While the digital processes are those from which the precise form is generated, the conceptual diagram remains analogic. In addition to engaging Gehry’s soft umbrella diagram, another of the originary conditions for Gehry’s Lewis Building for the Weatherhead School of Management is a classical precedent, more precisely, Karl Friedrich Schinkel’s Altes Museum. Gehry’s use of a plan as precedent departs radically from the top-down approach implied by the soft umbrella diagram. If the Lewis Building begins with a classical plan, this precedent is increasingly eroded and corrupted in section. Gehry uses the classical plan as an a priori ideal that evolves vertically, and at the same time challenges the idea of sectional extrusion implicit in the classical plan. The Altes Museum, which can be considered the historical prototype for the Lewis Building, has an orthogonal rectilinear plan with a central drum; the drum is extruded, so that its condition is the same in section from the plan to the roof. Most classical buildings are vertical extrusions from a plan and Schinkel continues this tradition, which is also apparent in the postwar work of Louis Kahn, among others. The Lewis Building thus makes a critique of

precedents, which may be of limited use today if, in Lynn’s terms, the precedent is understood as enforcing a part-to-whole relationship. In one sense, the development of this building operates against the top-down system of the soft umbrella diagram. The result resembles a classic Gehry expression, but the building requires the digital processes of the computer to erode the section, which begins as an orthogonal condition, in a way that would not have been possible with analogic methods. This invocation of the digital is crucial to understanding the evolution of the Lewis Building, and its conceptual differences in engaging precedents, from Lynn’s work, which undermines the role of precedents. Tracing the evolution of the Lewis Building through study models and early sketches in some sense reveals the undecidability of any origins. The earliest study models of June 1997 reveal a tension between orthogonal organizations with clear historical precedents and biomorphic forms related to Gehry’s exploration of digital modeling. A two-color model, reminiscent of a Richard Neutra or Rudolf Schindler project of the 1920s and 1930s in terms of its blocky massing, has a base on which the smaller blocks of its upper level sit. This U-shaped organization of blocks is frontalized like any classical building with a distinct propylaea or frontispiece. There is a clearly

260

Peter B. Lewis Building

6. Peter B. Lewis Building, study model, October 1997.

articulated, voided space in the center, in which is housed a bi-nuclear element formed by two colored cubes bound together by a smaller rectangular element. A vertical cut emphasizes a strong central axis. This axis, the frontispiece, and the U-shaped body recall both classical and neoclassical precedents. A second model from June 1997 shares the blocky forms, U-shape, and frontispiece of the first model, but in this case its voided center is a wellspring for curving and biomorphic forms in metal and plastic. It is clear that the energy is not coming from above, as would be the case in a soft umbrella diagram, but from below, as if the blocky organization of the model were being overcome from within. The next model, from September 1997, returns to a building of boxlike units, yet introduces a distinct pinwheeling character. The tension between the biomorphic and orthogonal forms is poignantly captured in a sketch for the Lewis Building from October 1997, which appears at first glance to be little

more than a doodle. However, this sketch provokes several interesting interpretations. First, despite the loose hand, a base condition which is more or less orthogonal is visible, as is the bipartite relationship of volumes around a depressed center. The sketch contains a series of biomorphic, nonorthogonal forms that seem either to grow out of, or are being pulled down into, a central vortex; the section implied by the drawing reveals this force that could be either centrifugal or centripetal. The model of October 1997, seemingly based on this sketch, suggests the integration of a U-shaped and corner-towered palazzo with a diagram of biomorphic forms exploding from a voided center. The color scheme of the model marks an intention to show a difference between the biomorphic and the biotechnic, the base and the superstructure, the center and the edge. The digital model produced in April 1998 manifests the coexistence of these two types of organizations, maintaining their distinction in its two-color

Peter B. Lewis Building

261

7. Peter B. Lewis Building, digital model, April 1998.

scheme. This is not a top-down strategy, nor is it a monochromatic or monolithic material strategy, but one that remains dialectical in its nature and bi-nuclear around a voided center. The two study models of May 1998 and March 1999 are sectional models revealing the presence of the base and corner towers, which are articulated in a different material. This model has a distinct base, an external wrapper that is the Ushaped body of the building, a voided center, and, within that void, a bi-nuclear central element— which itself seems to wrap around yet another element, creating an interior and exterior wrapper. The section suggests that the biomorphic forms are lifted off the base, creating a sectional deployment of the voided center. These two components—base and biomorphic forms—share a dialectical relationship, but the question remains whether the biomorphic forms are coming up from the base, being pulled down to the base, or, alternatively, are suspended between the base and the roof. More interesting is that the sec-

tional energies in the May 1998 and March 1999 models present a section in which the biomorphic form becomes a wrapper for an internal volume, a form within a form. While this sectional model maintains some of the earlier ideas, this introduction of the shell and solid forms adds another dimension to the evolution of the section. The section produces a dialogue between container and contained, figure and ground, vertical and horizontal, and forces of erosion and stability. All of these dialectical characteristics are apparent in the model. While supposedly an expressionist artist, Gehry adopts a process, as evidenced in these study models, that combines intuition with the understanding of the less-thanconscious influence of historical precedents. The corkscrewlike energy of the section differs significantly from that at Le Corbusier’s Palais des Congrès-Strasbourg or from the ramp at Poissy. The section at the Lewis Building is not the product of a vertical extrusion; rather it evolves from a classical plan as an initial whole, which

262

Peter B. Lewis Building

8. Peter B. Lewis Building, study model, May 1998.

is corrupted as it moves vertically to the point where the parts seem to have no relationship to their origin in that classical plan. The section, for example, is reminiscent of James Stirling’s Staatsgalerie in Stuttgart, suggesting that a critical evaluation of the relationship between the Altes Museum, the Staatsgalerie, and the Lewis Building could prove interesting. A digital rendering from 2000 of the Lewis Building can be analyzed as if the rectilinear forms of the underlying elevation were removed from the drawing. The remaining image is exclusively one of biomorphic forms seemingly exfoliated out of the plane of the paper. The two towers are now seen as the ground of the paper, and the vertical explosion is a series of cutouts coming out of the literal ground of the paper. These are important conceptual images reiterat-

ing a concept of a quasi-invisible “ground” which is rooted in historical precedents, such as corner towers, set against the energy of emergent biomorphic forms. The compositional trajectories are apparent in this project in a way that differs from other Gehry projects. This is one of the few Gehry projects that could be considered an unintentional subliminal critique of historical precedent. The resonance of Schinkel’s Altes Museum in Gehry’s scheme for the Lewis Building is first and foremost visible in the plan. While there is a literal section in the Altes Museum, it is not thematic as in the Lewis Building. This is not to say that the influence of Schinkel’s plan reflects a wholly conscious decision; such influences can penetrate one’s unconscious, particularly when a diagrammatic operation allows such unconscious projec-

Peter B. Lewis Building

263

9. Peter B. Lewis Building, study model, March 1999.

tions to surface. The diagram provides a vehicle for the unconscious expression of figural warping to emerge: diagrams often activate the unconscious memory, which in this case could reflect Gehry’s prior work on a 1995 competition in Berlin for a museum that was to be added to the Museum Island. The project for the Lewis Building falls between the conscious and the unconscious, between the analogic and the digital, and as such is different from Gehry’s other projects. That is what makes the Lewis Building a fulcrum project between Gehry’s past work and the projects that follow. The Lewis Building engages the combination of the analogic and the digital, and in particular how the digital may impact the notion of section in architecture. The traditional or analogic section is produced from the plan and extends vertically to

a roof. Digital modeling provides the possibility of an extension of space that is no longer necessarily Cartesian, yet is different from Koolhaas’s Agadir section or Libeskind’s erosion of the x-axis at the Jewish Museum. This technology enables the modeling of new forces of vertical extension, such as erosion and warping. The vertical erosion developed at the Lewis Building is of value in relationship to a series of plan and sectional precedents that do not have such an erosion in section. The lateral and continuous extension of space as a horizontal datum seen in the Maison Dom-ino can now be modulated in a more nuanced manner, as is the case in Agadir, or in Foreign Office Architects’ project for Yokohama, each of which focuses on the disturbance of the horizontal section as their thematic. In Gehry’s building for the Weatherhead School, section is not merely a horizontal datum

264

Peter B. Lewis Building

10. Peter B. Lewis Building, side elevation, digital model, 2000.

extension of space, but rather becomes a modulation of space in the vertical: the section warps and spirals as it evolves vertically. It is the rethinking of section—differently from Koolhaas, Libeskind, and Le Corbusier—that makes the Lewis Building again a fulcrum between past and future ideas of section. Gehry simultaneously denies both the idea of vertical extrusion from a plan and his own soft umbrella diagram. That the warped section in the Lewis Building is masked by a facade further denies the part-to-whole relationship of the section to facade. Gehry’s Lewis Building is a pivotal project in that it raises the question of the transgression of architectural precedents. It marks a shift in conceptualizing the diagram as an analogic device and in differentiating between analogic and digital processes. While Lynn’s design relies on computation, and while Koolhaas and Libeskind work in analogic terms, there are aspects of the Lewis Building that could only be developed in the digital. As the final project in this book, the Lewis Building provides a frame for the other nine projects to be seen within the evolution of architectural critiques of modern-

ism and the critical uses of the diagram in its various forms. The Lewis Building is a cusp project between the past as present and the present as future, and broaches the underlying paradigm shift that occurs in questioning the precedence of the unity of the classical part-to-whole relationship. Gehry’s Lewis Building for the Weatherhead School addresses the problematic of part-to-whole differently than the other works in this book on the question of precedent. It can be argued that all of the ten buildings discussed here depend on the possibility of processes which relate to some form of precedent seen as originary, truthful, or ideal. Each of the ten buildings also refers diagrammatically to some precedent. Historically, any paradigm shift begins with the denial of precedent as a necessary agent. In this sense, the analysis here may be a work of sublime yet necessary uselessness in the face of the evolving ability to produce conditions internal to component relationships that have no necessary analogic relationship to any prior, or precedent, condition. If anything in architecture has changed as a result of these ten buildings, it is primarily the

Peter B. Lewis Building

11. Peter B. Lewis Building, section, 2000.

subtle change in the relationship of subject to object. This occurs in two senses: first, the change in close reading necessitated by the emergence of figural forces produced through digital processes; second, the change in the subject’s physical relationship to the object, with the subject himself becoming an object of the gaze. There is no unifying theme in these works. If anything, these readings reveal the breakdown of close reading predicated on part-to-whole relationships and the concommitant idea of decidable meanings. When narrative—as an altered sense of time in the subject/ object relationship—is diminished, close reading cannot help but be affected, which produces an idea of undecidability. Text is thus the engine of the undecidable. It is these changes in close reading which ultimately suggest a rethinking of the canon which has always been underpinned by a received idea of close reading. To question these received ideas is perhaps what marks a canon of this moment in time.

265

266

12. Peter B. Lewis Building, sublevel and ground-floor plans.

Peter B. Lewis Building

Peter B. Lewis Building

13. Peter B. Lewis Building, first- and second-floor plans.

267

268

14. Peter B. Lewis Building, third- and fourth-floor plans.

Peter B. Lewis Building

Peter B. Lewis Building

15. Peter B. Lewis Building, roof plans.

269

270

Peter B. Lewis Building

B

A

A

B

C

B

16. The floor plan of the Altes Museum can be seen as a precedent for Gehry’s Lewis Building for the Weatherhead School. Schinkel’s Altes Museum seems at first glance to be the prototypical neoclassical palazzo, with four corner blocks. It is an ABCBA parti, whose frontality is emphasized by a frontispiece running the entire length of the facade.

C

B

A

A

17. The ground-floor plan of the Lewis Building also retains distinct traces of the classical plan in its U-shaped organization, frontispiece and ABCBA organization. The equivalent of the frontispiece stoa in the Schinkel plan is isolated in the Lewis Building as a distinct component of the building.

Peter B. Lewis Building

18. In the Altes Museum, a drum establishes the vertical axis. Symmetry is established by a vertical axis that cuts through the central volume, which is flanked by two square spaces and two rectangular blocks.

271

19. The ground-floor plan of the Lewis Building is similarly a bipartite organization around a central void. There is a paradoxical play between the small size of the central nucleus and what can be interpreted as its impact on the surrounding blocks. It seems to exert a force that presses the rear blocks backward and carves into its neighboring blocks.

272

20. In the Altes Museum, the central drum is flanked by two identical volumes. The drum is also compressed by the staircases which are pushed into the central space. Such compression can also be read in the width of the niches carved into the poché around the drum. The broad niches to the rear of the building become compressed at the front of the drum, as if registering the impact of the stairs.

Peter B. Lewis Building

21. The central nucleus of the Lewis Building is framed by what can be read as two major volumes forming a bi-nuclear element. These volumes reinstate a rough symmetry around the vertical axis, which is subsequently obscured by the asymmetrical frontispiece as it exerts a similar push toward the center, like the frontispiece stoa of the Altes Museum.

Peter B. Lewis Building

22. There is an anomalous zone (in red) in the Altes Museum which animates the plan.

273

23. There is also an anomalous zone in the Lewis Building shifting the central axis of the actual plan.

274

Peter B. Lewis Building

A

B

C

B

A

A

B

C

B

A

24. The plan of Schinkel’s Altes Museum produces symmetrical and asymmetrical readings. While the plan is symmetrical about the vertical axis, no single horizontal axis is dominant.

25. Essentially, an ABCBA organization can be can be read across the Lewis Building, from top to bottom. The central axis of the C zone does not correspond to that which is established by the center of the drum. The drum in the C zone intrudes into the rear B zone to press against the rear A zone. Such compressive forces act against the stability of any symmetry.

Peter B. Lewis Building

26. The neoclassical view of the Altes Museum is approached from the corner, resulting in its being seen perspectivally rather than frontally. In this corner view, it is the relationship of the side to the front that is stressed, making the corner thematically important.

275

27. The Lewis Building adopts certain of these neoclassical tropes, not only in plan but also in perspective. At the corner, a difference between a vertical stacking of windows on the left side and a pyramidal stepping on the right side can be noted. The corner tower is articulated in such a way that each side presents different information yet frames the corner as a central element.

276

28. Peter B. Lewis Building, vertical warp from ground-floor plan to roof.

Peter B. Lewis Building

Peter B. Lewis Building

29. Peter B. Lewis Building, column grid with tilted columns.

277

278

30. Peter B. Lewis Building, vertical Cartesian extrusion of plan.

Peter B. Lewis Building

Peter B. Lewis Building

31. Peter B. Lewis Building, contrast between vertical extrusion and sectionally warped elements.

279

280

32. Peter B. Lewis Building, ground floor, axonometric view.

Peter B. Lewis Building

Peter B. Lewis Building

33. Peter B. Lewis Building, second floor, axonometric view.

281

282

34. Peter B. Lewis Building, third floor, axonometric view.

Peter B. Lewis Building

Peter B. Lewis Building

35. Peter B. Lewis Building, fourth floor, axonometric view.

283

284

36. Peter B. Lewis Building, fifth floor, axonometric view.

Peter B. Lewis Building

Peter B. Lewis Building

37. Peter B. Lewis Building, roof plan, southeast axonometric view.

285

286

38. Peter B. Lewis Building, roof plan, northwest axonometric view.

Peter B. Lewis Building

Bibliography

Introduction Barthes, Roland. “From Work to Text.” In Art After Modernism: Rethinking Representation, edited by Brian Wallis, 168-174. New York: The New Museum of Contemporary Art, 1984. Bloom, Harold. The Western Canon: The Books and School of the Ages. New York: Harcourt Brace, 1994. Derrida, Jacques. Of Grammatology. Translated by Gayatri Chakravorty Spivak. Baltimore: Johns Hopkins University Press, 1976. Eisenman, Peter. “Aspects of Modernism: Maison Dom-ino and the Self-Referential Sign.” Oppositions 15/16 (Winter/Spring 1980): 119-128. Eisenman, Peter. “Post-functionalism.” Oppositions 6 (Fall 1976): i-iii. Rowe, Colin. “The Mathematics of the Ideal Villa.” In The Mathematics of the Ideal Villa and Other Essays. Cambridge, MA: MIT Press, 1982. Tafuri, Manfredo. Architecture and Utopia: Design and Capitalist Development. Translated by Barbara Luigi La Penta. Cambridge, MA: MIT Press, 1976. Tafuri, Manfredo. Progetto e utopia: Architettura e sviluppo capitalistico. Bari, Laterza, 1973. Tafuri, Manfredo. Teorie e Storia dell’architettura. Bari: Laterza, 1967. Tafuri, Manfredo. Theories and History of Architecture. Translated by Giorgio Verrecchia. New York: Harper & Row, 1980.

Colophon Layout 02 08 FINAL.indd289 289

4/9/2008 11:51:20 AM

290

Bibliography

1. Profiles of Text Luigi Moretti, Casa “Il Girasole.” Rome, Italy, 1947–1950. Banham, Reyner. “Casa del Girasole: Rationalism and Eclecticism in Italian Architecture.” Architectural Review 113 (February 1953): 73-77. Bucci, Federico and Marco Mulazzani. Luigi Moretti: Works and Writings. Translated by Marina deConciliis. New York: Princeton Architectural Press, 2002. Eisenman, Peter. “Luigi Moretti and the Culture of Fragments.” Area 74 (May/June 2004): 170-181. Eisenman, Peter. “La Casa ‘Il Girasole,’ Moretti visto da Moretti. Rome: Palombi, 2007. Finelli, Luciana. Luigi Moretti, la promessa e il debito: architetture 1926-1973. Rome: Officina, 1989, 2005. Moretti, Luigi. “Valori della Modanatura.” Spazio 6 (1951-2). Translated by Thomas Stevens as “The Values of Profiles.” Oppositions 4 (October 1974): 109-139. Moretti, Luigi. “Strutture e sequenze di spazi,” Spazio 7 (1952-3). Translated by Thomas Stevens as “The Structures and Sequences of Space.” Oppositions 4 (October 1974): 109-139. Stirling, James. “‘The Functional Tradition’ and Expression,” Perspecta 6 (1960): 88-97. Venturi, Robert. Complexity and Contradiction in Architecture. New York: Museum of Modern Art, 1966.

Colophon Layout 02 08 FINAL.indd290 290

4/9/2008 11:51:25 AM

Bibliography

291

2. The Umbrella Diagram Ludwig Mies van der Rohe, Farnsworth House, Plano, Illinois, 1946–1951. Blaser, Werner. “Farnsworth-Haus in Plano, Illinois von Mies van der Rohe, 1945-50.” Detail 26 (November/December 1986): 526-529. Blaser, Werner. Mies van der Rohe: Farnsworth House: weekend house. Basel: Birkhauser, 1999. Eisenman, Peter. “miMISes READING: does not mean A THING.” In Mies Reconsidered, organized by John Zukowsky. New York: Rizzoli, 1986. Eisenman, Peter. “Mies and the Figuring of Absence.” In Mies in America, edited by Phyllis Lambert, 706-715. New York: H.N. Abrams, 2001. Guisado, Jesus Maria Aparicio. “The Dematerialisation of the wall, an evolution of tectonics: Gottfried Semper, Mies van der Rohe and the Farnsworth House.” Arquitectura 310 (1997): 16-21, 116-119. Hartoonian, Gevork. “Mies van der Rohe: The Genealogy of Column and Wall.” Journal of Architectural Education 42 (Winter 1989): 43-50. Lambert, Phyllis. “Mies Immersion.” In Mies in America, edited by Phyllis Lambert, 192-589. New York: H.N. Abrams, 2001. Lohan, Dirk. Farnsworth House, Plano, Illinois, 1945-50. Tokyo, Japan: A.D.A. EDITA Tokyo, 1976. Schulze, Franz. The Farnsworth House. Chicago: Lohan Associates, 1997. Schulze, Franz. Mies van der Rohe : A Critical Biography. Chicago: University of Chicago Press, 1985.

Colophon Layout 02 08 FINAL.indd291 291

4/9/2008 11:51:26 AM

292

Bibliography

3. Textual Heresies Le Corbusier, Palais des Congrès-Strasbourg, France 1962–64. Deleuze, Gilles. Francis Bacon: The Logic of Sensation. Translated by Daniel Smith. Minneapolis: University of Minnesota Press, 2002. Eisenman, Peter. “Aspects of Modernism: Maison Dom-ino and the Self-Referential Sign.” Oppositions 15/16 (Winter/Spring 1980): 119-128. Frampton, Kenneth. “Le Corbusier and ‘l’Esprit Nouveau.” Oppositions 15-16 (Winter-Spring 1979): 12-59. Frampton, Kenneth and Roberto Schezen. Le Corbusier: Architect of the Twentieth Century. New York: H.N. Abrams, 2002. Frampton, Kenneth. Le Corbusier 1933-1960. Cambridge, MA: MIT Press, 1980. Le Corbusier, Oeuvres Completes. Zurich: Les Editions d’Architecture, 1970. Le Corbusier, Towards an Architecture. Translated by John Goodman. Los Angeles: Getty Trust Publications, 2007. Le Corbusier, Vers une Architecture. Paris: Les Editions G. Crès et Cie, 1923. Rowe, Colin. “Mannerism and Modern Architecture.” In The Mathematics of the Ideal Villa and Other Essays, 29-58. Cambridge, MA: MIT Press, 1976. Sarkis, Hashim. “Constants in motion: Le Corbusier’s “rule of movement” at the Carpenter Center.” Perspecta 33 (2002): 114-125. Smet, Catherine de. Le Corbusier, Architect of Books. Baden: Lars Müller Publishers, 2005. Von Moos, Stanislaus and Arthur Rüegg. Le Corbusier before Le Corbusier: Applied Arts, Architecture, Painting, Photography, 1907-1922. New Haven: Yale University Press, 2002.

Colophon Layout 02 08 FINAL.indd292 292

4/9/2008 11:51:26 AM

Bibliography

293

4. From Plaid Grid to Diachronic Space Louis Kahn, Adler and DeVore Houses, Philadelphia, Pennsylvania, 1954–55. Bonnefoi, Christian. “Louis Kahn and Minimalism.” Oppositions 24 (Spring 1981): 2-25. Brownlee, David B. Louis I. Kahn: In the Realm of Architecture. Introduction by Vincent Scully. New York: Rizzoli, 1991. Fitch, James Marston. “A Building of Rugged Fundamentals.” Architectural Forum 113 (July 1960): 82–87. Frampton, Kenneth. “Louis Kahn and the French Connection.” Oppositions 22 (Fall 1980): 20-53. Jordy, William H. “Criticism, medical research building for Pennsylvania University, Phila.: Louis I. Kahn, architect.” Architectural Review 129 (February 1961): 99-106. Kahn, Louis. “Two Houses.” Perspecta 3 (1955): 60-61. Kahn, Louis. “Monumentality.” In Architecture Culture 1943-1968, edited by Joan Ockman with Edward Eigen. New York: Columbia Books of Architecture/Rizzoli, 1993. “Louis I. Kahn, architect, Alfred Newton Richards Medical Research Building, University of Pennsylvania, Philadelphia, 1958-1960.” Museum of Modern Art Bulletin 28 (1961): 3–23. Maniaque, Caroline. “Louis Kahn: La Maison comme Laboratoire d’éxperimentation.” Moniteur architecture AMC 108 (June/July 2000): 94-100. Norberg-Schulz, Christian. “Kahn, Heidegger, and the Language of Architecture.” Oppositions 18 (Fall 1979): 28-47.

Colophon Layout 02 08 FINAL.indd293 293

4/9/2008 11:51:26 AM

294

Bibliography

5. The Nine-Square Grid and Its Contradictions Robert Venturi, Vanna Venturi House, Chestnut Hill, Pennsylvania, 1959–64. “Interview: Robert Venturi and Peter Eisenman.” Skyline (July 1982): 12-15. Raggatt, Howard. “A Zone of the Blur.” Transition 41 (1993): 7-13. Rowe, Colin. “Robert Venturi and the Yale Mathematics Building.” Oppositions 6 (Fall 1976): 1-23. Mother’s House: The Evolution of Vanna Venturi’s House in Chestnut Hill. Edited by Frederic Schwartz. New York: Rizzoli, 1992. Steele, James et al. “Special Issue: Venturi Scott Brown & Associates on Houses and Housing.” Architectural Monographs 21 (1992): 24-30. Venturi, Robert. Complexity and Contradiction in Architecture. New York: Museum of Modern Art, 1966. Venturi, Robert and Denise Scott Brown, “Some Houses of Ill-Repute.” Perspecta 13 (1971): 259-267 Venturi, Robert, Denise Scott Brown, and Steven Izenour. Learning from Las Vegas: The Forgotten Symbolism of Architectural Form. Cambridge, MA: MIT Press, 1972. Von Moos, Stanislaus. Venturi, Rauch and Scott Brown, Buildings and Projects. New York: Rizzoli, 1987. Wrede, Stuart. “Complexity and Contradiction Twenty-five Years Later: An Interview with Robert Venturi.” In American Art of the 1960s, 143-163. New York: The Museum of Modern Art, 1991.

Colophon Layout 02 08 FINAL.indd294 294

4/9/2008 11:51:26 AM

Bibliography

295

6. Material Inversions James Stirling, Leicester Engineering Building, Leicester, England, 1959–63. Banham, Reyner. “The New Brutalism.” Architectural Review 118 (December 1955): 354-361. Cook, Peter. “Stirling and Hollein.” Architectural Review 172 (December 1982): 52-54. Eisenman, Peter. “Real and English: Destruction of the Box. I.” Oppositions 4 (October 1974): 5-34. Frampton, Kenneth. “Leicester University Engineering Laboratory.” Architectural Digest 34, (February 1964): 61. James Stirling, Buildings and Projects, 1950-1974. London: Thames and Hudson, 1974. James Stirling, Buildings and Projects, 1950-1980. Edited by Peter Arnell and Ted Bickford. New York: Rizzoli, 1993. Maxwell, Robert. James Stirling/Michael Wilford. Basel, Boston: Birkhauser Verlag, 1998. McKean, John. Leicester University Engineering Building. London: Phaidon, 1994. Rowe, Colin and Robert Slutzky, “Transparency: Literal and Phenomenal.” Perspecta 8 (1963): 45-54. Rowe, Colin and Robert Slutzky, “Transparency: Literal and Phenomenal. II.” Perspecta 13 (1971): 287-301. Scalbert, Irne. “Cerebral Functionalism: The Design of the Leicester University Engineering Building.” Archis 5 (May 1994): 70-80. Stirling, James. “Garches to Jaoul: Le Corbusier as a Domestic Architect in 1927 and 1953.” Architectural Review (September 1955): 145-151. Stirling, James. “‘The Functional Tradition’ and Expression.” Perspecta 6 (1960): 88-97. Stirling, James. “Regionalism and Modern Architecture.” Architects’ Yearbook 8 (1957): 62-68. Tafuri, Manfredo. “L’Architecture dans le Boudoir.” Translated by Victor Caliandro. Oppositions 3 (May 1974): 37-62.

Colophon Layout 02 08 FINAL.indd295 295

4/9/2008 11:51:26 AM

296

Bibliography

7. Texts of Analogy Aldo Rossi, Cemetery of San Cataldo, Modena, Italy, 1971–78. Aldo Rossi, Buildings and Projects. Edited by Peter Arnell and Ted Bickford, essays by Vincent Scully and Rafael Moneo. New York: Rizzoli, 1985. Aureli, Pier Vittorio. “The Difficult Whole.” Log 9 (Spring 2007): 39-61. Bataille, Georges. Le Bleu du Ciel. Paris: J.Pauvert, 1967. Bataille, Georges. Blue of Noon. Translated by Harry Mathews. New York: Unizen, 1978. Eisenman, Peter. “The House of the Dead as the City of Survival.” In Aldo Rossi in America: 19761979, 4-15. New York: Institute for Architecture and Urban Studies, 1979. Johnson, Eugene J. “What Remains of Man – Aldo Rossi’s Modena Cemetery.” Journal of the Society of Architectural Historians 41 (March 1982): 38-54. Moneo, Rafael. “Aldo Rossi: The Idea of Architecture and the Modena Cemetery.” Oppositions 5 (1976): 1-30. Rowe, Colin and Fred Koetter. Collage City. Cambridge, MA: The MIT Press, 1978. Rossi, Aldo. The Architecture of the City. Translated by Diane Ghirardo and Joan Ockman. Cambridge, MA: MIT Press, 1984. Rossi, Aldo. “The Blue of the Sky.” Oppositions 5 (1976): 31-34. Savi, Vittorio. “Il cimitero aldorossiano: traccia di racconto critico [the Aldorossian Cemetery: outline of a critical account].” Lotus International 38 (1983): 30-43. Tafuri, Manfredo. “The ‘Case’ of Aldo Rossi.” In History of Italian Architecture, 1944-1985, translated by Jessica Levine. Cambridge, MA: MIT Press, 1989. Thom, Deborah. “The City of the Dead as the City of the Living: Aldo Rossi’s Modena Cemetery.” Dimensions 4 (Spring 1990): 14-17.

Colophon Layout 02 08 FINAL.indd296 296

4/9/2008 11:51:26 AM

Bibliography

297

8. Strategies of the Void Rem Koolhaas, Jussieu Libraries, Paris, France, 1992–93. Aureli, Pier Vittorio and Gabriele Mastrigli. “Postmodern Oppositions: Eisenman contro Koolhaas.” Arch’It files (28 January 2004). Cornubert, Christophe. “Ein offenes Fenster: die Fassade der Bibliotheken von Jussieu in Paris.” Deutsche Bauzeitung 128 (1994): 152-156. Fernández-Galiano, Luis. “The Butterfly’s Fate.” Log 2 (Spring 2004): 15-23. Koolhaas, Rem. “Junk Space.” In Content, 162-171. Cologne: Taschen, 2003. Lucan, Jacques. “Le concours de Jussieu.” Moniteur architecture AMC 38 (February 1993): 22-25. “Office for Metropolitan Architecture: two libraries for Jussieu University, Paris.” AA files 26 (Autumn 1993): 36-44. Somol, R.E. “12 Reasons to Get Back into Shape.” In Content, 86-7. Cologne: Taschen, 2003. Somol, R. E. and Sarah Whiting. “Notes Around the Doppler Effect.” Perspecta 33 (2002): 72-77. Vidler, Anthony. “Books in Space: Tradition and Transparency in the Bibliothèque de France.” Representations 42 (Spring 1993): 115-134. Zaera Polo, Alejandro, Rem Koolhaas, et al. “OMA/Rem Koolhaas 1992-1996.” El Croquis 79 (1996).

Colophon Layout 02 08 FINAL.indd297 297

4/9/2008 11:51:26 AM

298

Bibliography

9. The Deconstruction of the Axis Daniel Libeskind, Jewish Museum, Berlin, Germany, 1989–1999.

Anderson, Stanford et al. “Memoria.” Diadalos 58 (December 1995): 122-125. Evans, Robin. “In Front of Lines that Leave Nothing Behind.” AA Files 6 (1984): 89-96. Gonzales Cobelo, J. L., Donald L. Bates, and Daniel Libeskind. “Daniel Libeskind 1987-1996.” El Croquis 80 (1996). Huyssen, Andreas. “The Voids of Berlin.” Critical Inquiry 24 (Autumn 1997): 57-81. Krauss, Rosalind. “Notes on the Index: Seventies Art in America.” October 3 (Spring 1977): 68-81. Krauss, Rosalind. “Notes on the Index: Seventies Art in America, Part II.” October 4 (Autumn 1977): 58-67. Libeskind, Daniel. “Between the lines: extension to the Berlin Museum, with the Jewish Museum.” Assemblage 12 (August 1990): 18-57. Libeskind, Daniel. Chamberworks: Architectural Meditations on Themes from Heraclitus. London: Architectural Association, 1983. Libeskind, Daniel. Daniel Libeskind: Jewish Museum Berlin: Between the Lines. Munich and London: Prestel, 1999. Libeskind, Daniel. Jewish Museum Berlin. Berlin: G + B Arts International, 1999. Libeskind, Daniel. A Passage Through Silence and Light. London: Black Dog Publishers, 1997. Rambert, Francis. “Presence de l’absence: Musée juif de Berlin.” Connaissance des arts 561 (May 1999): 98-105. Ullman, Gerhard. “El rayo del entendimiento: Libeskind, Museo Judo en Berlin.” Arquitectura Viva 11 (March-April 1990): 14-19.

Colophon Layout 02 08 FINAL.indd298 298

4/9/2008 11:51:26 AM

10. The Soft Umbrella Diagram Frank O. Gehry, Peter B. Lewis Building, Weatherhead School of Management, Case Western Reserve University, Cleveland, Ohio, 1997–2002. Dal Co, Francesco and Kurt W. Forster. Frank O. Gehry: The Complete Works. New York: The Monacelli Press, 1998. Frank Gehry, Buildings and Projects. Edited by Peter Arnell and Ted Bickford. New York: Rizzoli, 1985. Foster, Hal. “Re: Post.” In Art After Modernism: Rethinking Representation, edited by Brian Wallis, 188-201. New York: The New Museum of Contemporary Art, 1984. Krauss, Rosalind. “The Originality of the Avant-Garde: A Postmodern Repetition.” In Art After Modernism: Rethinking Representation, edited by Brian Wallis, 12-29. New York: The New Museum of Contemporary Art, 1984. Litt, Steven. “Business unusual [Peter B. Lewis Building, Cleveland].” Architecture 91 (October 2002): 68-73. Martin, Jean-Marie. “Frank O. Gehry.” Casabella 63 (September 1999): 12-21. Polano, Sergio et al. “Forms and methods of deconstruction [Forme e modi della deconstruzione].” Casabella 63 (September 1999): 12-20, 88-89. Owens, Craig. “The Allegorical Impulse: Toward a Theory of Postmodernism.” In Art After Modernism: Rethinking Representation, edited by Brian Wallis, 202-235. New York: The New Museum of Contemporary Art, 1984.

Colophon Layout 02 08 FINAL.indd299 299

4/9/2008 11:51:27 AM

300

Index Numbers in italics denote pages upon which illustrations appear. 50 by 50 House, 57, 57 Adler House, 12, 20, 23, 102–119, 102, 104, 106, 108, 112–119, 131 Alberti, Leon Battista, 29, 53 Altes Museum, 259, 262, 270–275, 270–275 Alumni Memorial Hall (IIT), 56, 57 American Academy in Rome, 130 Arch of Titus, 29 Archigram, 157 Assembly Hall at Chandigarh, 76, 77 Aureli, Pier Vittorio, 180 Bacon, Francis, 73 Banham, Peter Reyner, 27, 157 Barcelona Pavilion, 53, 53, 54 Barnes, Edward, 130 Barthes, Roland, 51 Bataille, Georges, 179 Benjamin, Walter, 201 Bergson, Henri, 235 Berlin Museum 235, 236 Blake, Peter, 157 Blanchot, Maurice, 103–104, 107 Bloom, Harold, 12, 15 Borromini, Francesco, 21 Boullée, Etienne-Louis, 185, 190 Brick Country House, 52 Breuer, Marcel, 52, 107 Bunshaft, Gordon, 130 Cambridge History Faculty Library, 159 Campo Marzio, 182 Carpenter Center, 76–78

Colophon Layout 02 08 FINAL.indd300 300

Casa da Musica in Porto, 201, 207–208 Casa “Il Girasole,” 18, 22, 26–48, 26–48, 130, 136 Casa Giuliani-Frigerio, 28 Case Western Reserve, 11; and Weatherhead School of Management, 19 Cemetery of San Cataldo in Modena, 22, 23, 178– 180, 184–198, 185–198 Central Business District Proposal, Turin, 180 Chandigarh, 75, 80: Assembly Hall, 76, 77; Parliament Building, 76 Chapel at Ronchamp, 236 CIAM (Congrès International d’Architecture Moderne), 129, 157 City Edge competition, 234, 234 Cobb, Harry, 130 Concrete Country House, 52 Corbusier. See Le Corbusier Costa Cemetery, 185–186, 191 Crown Hall (IIT), 56, 57, 204 Cullen, Gordon, 157 De Chirico, Giorgio, 179, 184 Debord, Guy, 129 Deleuze, Gilles, 73, 129 Derrida, Jacques, 17, 27–28, 129 De Stijl, 51–52 DeVore House, 12, 19, 20, 23, 103–111, 120-126, 104–107, 109, 120–126, 131 Doesberg, Theo van, 29 Dom-ino. See Maison Dom-ino Durand, J.N.L., 183 Dutch Embassy in Berlin, 207

4/9/2008 11:51:27 AM

301

El Lissitzky, 29 Empson, William, 17 Esposizione Universale di Roma, 179 Exeter Library, 20, 110 Farnsworth House, 12, 18, 22, 50–71, 50–52, 55, 57, 60–71, 60–71 Federal Housing Authority, 130 Florey Building at Queens College, Oxford, 159 Franzen, Ulrich, 130 Frick Museum, New York, 21 Foreign Office Architects, 263 Foucault, Michel, 9 Gallaratese housing complex, 180, 181, 184, 187 Gehry, Frank, 11, 18–21, 23–24, 256–286 Glass House, 56, 58 Graves, Michael, 130 Guggenheim Museum Bilbao, 18–21 Giedion, Sigfried, 129 Gowan, James, 154, 159 Gropius, Walter, 52, 107 Half-House, 136 Hamilton, Richard, 156–157 Heidegger, Martin, 52 Hejduk, John, 106-109, 130–131, 133, 136, 188 Henderson, Nigel, 156 Hilberseimer, Ludwig, 192 Illinois Institute of Technology (IIT), 55–57 Independent Group, 156–157 Indian School of Management, Ahmedabad, 20 Institute for Architecture and Urban Studies in New York, 231

Colophon Layout 02 08 FINAL.indd301 301

Jewish Museum in Berlin, 23–24, 23, 230–231, 230, 233–254, 236–254, 263 Johansen, John, 130 Johnson, Philip, 51, 56, 58, 130 Jung, Carl, 188 Jussieu Libraries, 11–12, 21–22, 22, 24, 79–80, 200–202, 205–228, 200, 206–228 Kahn, Louis, 12, 19, 20, 23, 102–127, 131, 161, 181, 205, 259 Kipnis, Jeffrey, 24, 205 Koolhaas, Rem, 11–12, 21–22, 24, 79–80, 200–228, 263, 264 Krauss, Rosalind, 231–232 Labatut, Jean, 130 Lacan, Jacques, 202 Le Corbusier, 11–12, 19, 21, 23, 29, 51, 53, 54–56, 62–63, 72–100, 106, 111, 129–130, 138, 155–156, 158, 181, 186, 201–202, 204–205, 207, 212, 218, 234, 236, 261, 264 Lapadula, Ernesto, 179 Leicester Engineering Building, 11–12, 20, 23, 154–176, 154, 156, 158–176 Libeskind, Daniel, 23–24, 23, 188, 230–254, 263– 264 Line of Fire installation, 234–237, 235, 242–243, 242–243 Liverpool School of Architecture, 156 Loos, Adolf, 33, 53, 138, 181 Louvre, Paris, 21 Lyndon, John, 130 Lynn, Greg, 257–259, 264

4/9/2008 11:51:27 AM

302

Maison Citrohan, 51 Maison Dom-ino, 10, 51, 54, 63, 204, 212, 263 Maison Jaoul, 106, 158 Mannheim Theater, 57 Martello towers, 157 Matta-Clark, Gordon, 232–233, 236 Melnikov, Konstantin, 157, 162 Mies van der Rohe, Ludwig, 12, 18, 22, 29, 50–71, 110, 138, 181, 201, 204–205, 207, 258 Mission Grande Axe, La Defense, Paris, 202 Modena Cemetery, 186–188, 190 Moore, Charles, 130 Moretti, Luigi, 12, 18, 22, 26, 26–48, 110, 130, 136, 157, 201, 206–207 Mundaneum project, 186 Museum of Modern Art, 24, 130 Mussolini, Benito, 155 National Football Foundation Hall of Fame, 137 National Gallery in Berlin, 57, 58, 204 Neutra, Richard, 259 New York Athletic Club, 201, 203, 203, 206 Notre Dame du Haut, Ronchamp, 75 OMA. See Rem Koolhaas Ozenfant, Amédée, 74 Palais des Congrès-Strasbourg, 12, 19, 20, 23, 72–101, 72, 75, 77–101, 201, 207, 218, 261 Palazzo della Civiltà Italiana, 179 Palazzo del Te, Mantua, 34 Palladio, Andrea, 19, 138, 188 Palm Bay Seafront Convention Center, Agadir, Morocco, 204, 205, 207, 263

Colophon Layout 02 08 FINAL.indd302 302

Index

Paolozzi, Eduardo, 156–157 Parliament Building at Chandigarh, 76 Parc de La Villette, 203, 203, 204, 206 Parthenon, 73, 74 Pei, I.M., 130 Peirce, Charles Sanders, 22, 53, 130, 231 Peter B. Lewis Building, 19, 20, 23 24, 256–286, 256, 258–286 Philips Pavilion, 75 Piazza San Marco, 103 Piranesi, Giovanni Battista, 182 Price, Cedric, 157 Proust, Marcel, 103–104, 107 Quarter-House, 136 Rainaldi, Carlo, 32, 136 Resor House, 56 Richards Medical Center, 20, 105, 110, 110, 161 Rietveld, Gerrit, 51 Robertson, Jaquelin, 130 Rodgers, Ernesto, 181 Romano, Giulio, 34 Rossi, Aldo, 22, 23, 129, 178–198, 202 Rowe, Colin, 10, 11, 16, 76, 79, 156, 159, 182 Russakov Worker’s Club, 157, 162 Saint Andrew’s Dormitory, Scotland, 159 San Cataldo Cemetery. See Cemetery of San Cataldo Sant’Andrea, 29 Santa Maria in Campitelli, 32, 136 Schindler, Rudolf, 259 Schinkel, Karl Friedrich, 20, 259, 262, 270

4/9/2008 11:51:27 AM

Index

Schroeder House, 51 Scully, Vincent, 130 Seagram Building, 56, 181 Seattle Public Library, 201, 205, 207–208 Segrate Monument, 180, 181 Slutzky, Robert, 159 Smithson, Peter & Alison, 156–157 Staatsgalerie in Stuttgart, 262 Stella, Frank, 10 Stirling, James, 11–12, 20, 23, 154-176, 202, 262 Sydney Opera House, 18 Tafuri, Manfredo, 129 Team Ten, 129, 157 Terragni, Giuseppe, 28 Texas House, 106, 107, 130, 131, 133, 136 Tourette, La, 76–80, 181 Trenton Bathhouse, 20, 23, 105–107, 105 Très Grande Bibliothèque, 79, 202–203, 204, 204, 205, 218 Tugendhat House, Brno, 52 Unité d’Habitation, Marseilles, 202, 234 Utzon, Jørn, 18 Vanna Venturi House, 19, 23, 27, 108, 128–152, 128, 131–152 Villa Malcontenta, 19 Villa Radieuse, 130 Villa Rotunda, 19 Villa Savoye in Poissy, 74, 76 Villa Stein at Garches, 19, 74-75 Vitruvius, 53, 74

Colophon Layout 02 08 FINAL.indd303 303

303

Venturi, Robert, 11, 20, 23, 27, 51, 128–152, 130, 181, 202, 206–207 Von Erlach, Fischer, 185, 190 Vreeland, Tim, 130 Wall House, 107, 108–109 Weatherhead School of Management, Case Western Reserve, 19, 24, 258, 263, 270. See also Peter B. Lewis Building Yale University Art Gallery, 20 Zumthor, Peter, 33

4/9/2008 11:51:27 AM

Illustration Credits

Every effort has been made to identify the provenance of each image. If inaccuracies have inadvertently occurred, they will be corrected in subsequent printings.

Ajay Manthripragada: cover, 20 right, 22 left, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 190, 191, 192, 193, 194, 195, 196, 197, 198.

John Bassett: 18, 21, 23, 36 right, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 62, 63, 64 left, 66, 67, 68, 69, 70, 105, 120, 121, 122, 123.

Mies van der Rohe Archive; digital images © The Museum of Modern Art / Licensed by SCALA / Art Resource, NY. © 2007 Artists Rights Society (ARS), New York/VG Bild-Kunst, Bonn: 50, 52, 53, 54 right, 55, 56 left, 57, 58.

Collection Centre Canadien d’Architecture/Canadian Centre for Architecture. Montréal: 157 left.

Luigi Moretti, Archivio Centrale Dello Stato: 26, 28, 29, 30, 31, 32, 33, 34, 35.

Collection Centre d’Art Contemporain Genève: 233, 235 left. Chicago History Museum © Hedrich-Blessing: 56 right.

Office for Metropolitan Architecture: 200, 202 left, 203, 204, 205, 206, 207, 209, 210, 211.

Conway Library, Courtauld Institute of Art, London: 275.

Matthew Roman: 89, 99.

Le Corbusier, Licensed by SCALA/Art Resource, NY. © 2007 Artists Rights Society (ARS), New York/ADAGP, Paris/FLC: 54 left, 72, 74, 75, 76, 77, 78, 79, 157 right, 202 right.

Aldo Rossi Fonds, Collection Centre Canadien d’Architecture/ Canadian Centre for Architecture. Montréal: 180 left, 185, 189.

Peter Eisenman collection: 158, 159, 160 left, 161 right, 183 left.

Fondazione Aldo Rossi © Eredi Aldo Rossi, Photographs: © Alessandro Zambianchi, Simply.it: 178, 180 right, 182, 183 right.

Gehry Partners, LLP: 256, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269. Archivio Ghirri© Eredi di Luigi Ghirri: 181, 186, 187.

James Stirling / Michael Wilford Fonds, Collection Centre Canadien d’Architecture/Canadian Centre for Architecture. Montréal: 154, 156, 160 right, 161 left, 162, 163.

Andrew Heid: 22 right, 23 right, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285, 286.

Venturi Scott Brown and Associates, Inc. and Rollin La France: 128, 131, 132, 133, 134, 135, 136, 137.

John Hejduk Archive, Collection Centre Canadien d’Architecture/Canadian Centre for Architecture. Montréal: 106 right, 107 right, 130. Udo Hesse: 236, 237 left.

Michael Wang: 23 left, 240, 241, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254. Carolyn Yerkes: 19 right, 20 left, 81, 82, 83, 84, 85, 86, 87, 88, 90, 91, 92, 93, 94, 95, 96, 97, 98, 100, 112, 113, 114, 115, 116, 117, 118, 119, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152.

Louis I. Kahn Collection, University of Pennsylvania and the Pennsylvania Historical and Museum Commission: 102, 104, 105 left, 106 left, 107 left, 108, 109, 110. Studio Daniel Libeskind: 230, 234, 235 right, 237 right, 239. Libeskind/Licensed by SCALA/Art Resource, NY: 232. Ariane Lourie: 19 right, 36 left, 60, 61, 64 right, 65, 117, 124, 125, 126, 242, 243.

Colophon Layout 02 08 FINAL.indd304 304

4/9/2008 11:51:28 AM