Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We use cookies to personalise your experience; learn more in our Privacy and Cookie Policy. You can opt out of some cookies by adjusting your browser settings; see the cookie policy for details. By using this site, you agree to our use of cookies.

‘Pompidou cannot be perceived as anything but a monument’


Reyner Banham discusses the roles of Megastructure, Archigram and modern technology in Pompidou’s design

This article was republished online in March 2012

Reyner Banham is surely right to interpret the Centre Pompidou in terms of recent architectural history; and to see it as the apotheosis of Archigram and of the Megastructure idea. For this is the only door through which to reach an understanding of how and why this remarkable building is as it is.

Unfortunately - and this accounts for the rough reception, the Centre has received from Parisians - ‘recent architectural history’ is a small door through which only architects customarily pass; and though it must seem almost blasphemous to say this, not many outside the charmed circle of modern architecture have even heard of Archigram and of its apocalyptic struggles in an unresponsive society.

Centre Pompidou

Centre Pompidou

This is substantially true of England, but it is altogether true of France; for the French have always held modern architecture at arm’s length: they have done as much modern building as anybody else; but they regard it, not as part of la culture francaise but as unavoidable evil, to be kept as far as possible to the peripheries of their historic cities or to selected sites, such as La Defense or the New Towns. That is why you don’t go to Paris to look at post-Corbusian modern architecture. Why then was the Centre Pompidou built to this sort of design?

It is of course true that Gaullist France (like Francoist Spain) has undergone a last-minute change of heart in this matter; but we suspect nonetheless that this was primarily an art collector’s decision made in the name of the universality of French culture. Someone said: ‘Look, the National Buildings Collection is weak on High Modern Movement: let us commission a resounding example before it is too late’.

The public square and west facade with its external escalators connecting galleries acting as a vertical extension of the square. The facade is equipped to take clipped-on tv screens, mobile signs and a variety of panels, making it a potential information

The public square and west facade with its external escalators connecting galleries acting as a vertical extension of the square. The facade is equipped to take clipped-on tv screens, mobile signs and a variety of panels, making it a potential information bulletin for everyone.

Paris is a homogeneous city; and it is cloistered in the sense that the high unbroken eaves box in the view. The Centre Pompidou, when you first catch a glimpse of it, is startling indeed, but it registers as a relief, not as an incongruous interruption. Its neighbours (like footballers shielding a player who is changing his pants) hug it round, so that you can only see small patches at a time-a patch of painted tank work above the roofs from the Ile de la Cite, a rack of pipes framed by a narrow street from the Marais.

But to come into its full physical presence is a daunting experience. For it is a menacing building which stands like a man in full armour in a room full of civilians - indeed the glittering, rounded form of the outboard escalators gives a suggestion of greaves. Even so, the menace lies, not in a chance reference of this kind, but in the concept of society, which this celebration of high technology supposes. One Centre Pompidou (like one Faber Dumas in Ipswich) is an exhilarating sight; but only contemplate what the centres of our cities would be like if they were chiefly composed of buildings of this kind and you see at once what a repellant fix we would be in.

The Centre reflects the supreme moment of technological euphoria in Western society: the moment when we genuinely believed that ‘freedom’ was to be got by providing ourselves with endless power-supplied facility: with servicing which would be so elaborate and so heavily duplicated that you could do anything you want, any where, at any time. We are wiser now; for we know that even if our resources allowed this sort of indulgence, the political machinery we would have to forget to operate it would be so offensive that it would remove true freedom from the face of the earth.

This is why there is no reason to expect a multiplicity of Pompidou-like buildings; and why there is no call to agonise too much about the political implications of such a structure. There will only be one. The one in the Paris Collection. Whatever doubts we may entertain about the working of this building there can be no question that it; is the most stunning new ‘go-to’ to be seen in any city. To the architects and to the French state we offer awed congratulations.

The public reference library occupies parts of the first, second and third floors which are connected internally by escalators. The view provides a good example of the uninterrupted floor space achieved by means of the large-span structure and the externa

The public reference library occupies parts of the first, second and third floors which are connected internally by escalators. The view provides a good example of the uninterrupted floor space achieved by means of the large-span structure and the external distribution of vertical services.


Somewhere, wherever he is now, Sigfried Giedion must be smiling his famous myopic smile behind those enigmatically thick-lensed spectacles. Centre Pompidou uncannily realizes the programme for the monument of the future that he, along with Fernand Leger and Jose Lluis Sert, mapped out in the depths of the Second World War, and called Nine Points on Monumentality. It is not a view of monumentality that has been much entertained of late, and was unknown to the architects of the center-for more than a quarter of a century, Le Corbusier’s increasingly geriatric understanding of monumentality as mere mass and impenetrable substance has dominated Modern Movement thinking and deflected that movement away from its own basic beliefs in high-performance tenuous constructionas the proper architectural mode for public statements about matters of social import.

This is not to say that Giedion’s vision of lightweight, highly coloured mobile elements is necessarily any more valid or valuable than Corb’s latter-day Bismarckisms, but it is a vision that is quintessential to the Modern Movement. And it is a vision that Pompidou delivers from ‘re-planning on a large scale … will create vast open spaces in the now decaying areas of our cities. In these open spaces, monumental architecture will find its appropriate setting …’ to ‘light elements like ceilings can be suspended from big trusses covering practically unlimited spans’. Furthermore, Pompidou cannot be perceived on the ground in Paris as anything but a monument; the competition brief may have said nothing about monuments, but the hidden hand of presidential power behind that brief sought a monument. Just how Piano+Rogers, who are audibly wary of ideas like monumentality, could have come to design so convincing a monument is not the least of the enigmas that surrounds this enigmatic ‘facility’ (building is not the word for it) on the Rue du Renard.

The spirit of the Modern Movement, which Giedion so often interpreted with perspicuous accuracy while imagining that he was doing something else seems to have taken its revenge on two generations of academic doubters, intellectual Luddites and all those energetic breast beaters who nowadays contribute to the comic pages at the beginning of the AR. Centre Pompidou is the only public monument of international quality the ’70s have produced, and old Sigfried appears posthumously to have done it again.

John Partridge

Piano and Rogers' celebrated Centre Pompidou, seen at night from the Rue Rambuteau with the Church of St Merri in the background. The area on the right is being turned into a public square.

Piano and Rogers’ celebrated Centre Pompidou, seen at night from the Rue Rambuteau with the Church of St Merri in the background. The area on the right is being turned into a public square.


la question du Marais

Admittedly, it is almost as difficult to judge the Centre on its merits as to see it in a historical context that goes back as far as 1943, when Giedion wrote his essay. More it immediate controversies have pressed close around the site, and have questioned the nature of the facility itself. In 1971, the idea of a centralised, national, official, high technology arts-facility could hardly have been more thoroughly discounted; the idea that part of historic Paris, near Les Hallesand on the fringe of Le Marais, should be razed by Presidential intervention to create the site for it was equally intolerable to the survivors of the Evenements of 1968.

In spite of the 681 entries the competition received from all over the world, educated opinion at the time was that the resources involved should have been spread more evenly over the cultural life of the whole of France, and in so far as any construction working Paris was required it should have been in the form of piecemeal, small-is-beautiful rehabilitations in Le Marais, directly serving the community in the zone hiatorique, together with enterprising re-cyclings of the cast-iron of Baltard’s super-sheds in Les HaIles.

Part of the Musee d'Art Moderne is double height.

Part of the Musee d’Art Moderne is double height.

In the upshot, the cause of Les Halles has been lost, utterly, and Paris has been left with two cataclysmically embarrassing holes in the ground, but the case of Le Marais has been surprisingly different. Its rehabilitation has proceeded in the same mode as when the area was first discovered two decades ago, but the whole district has gained a new definition from the building of Centre Pompidou.

The manner in which this has happened derives from decisions made by the architects about matters that have almost nothing to do with laquestion du Marais. Purely local and operational considerations seem to have driven them to locate the building on the extreme eastern edge of the site, hard against the Ruedu Renard, and to put most of its public and ceremonial entrances on the side facing the piazza to the west. To anyone approaching from this side—that is, from ‘cultural’ Paris the face of the Centre will be seen ‘across the open space as a wall, terminating that part of Paris, and Le Marais will be seen as what comes next.

The effects of the location of the Centre on the eastern edge are equally intriguing in townscape close-up. The decision to locate the building so was really a fairly safe one in current English site planning practice. It is directly comparable to the Smithsons’ decision to put the housing blocks at Robin Hood Gardens on the noisy perimeter of the site, in order to create a quiet zone within. It is even more directly comparable to the Erskine team’s decision to back up their Byker brick ‘wall’ in Newcastle against the threat of traffic noise.

But unlike Byker, Pompidou’s blind side is a wall or rather a zone, of services-a 200-metre rack of coloured vertical pipes, ducts and conduitry The effect is sensational as one sees it from the Rue du Renard, but equally sensational is that it does not destroy the street in any way. The reasons are more optical than topographica1, the main Olle being that its true scale is almost unreadable. It looks nothing like a stall as it really is; the casual promeneur in the Rue du Renard probably supposes it to be not much taller than the conventional buildings facing it across the street.

Could he be taken up to the catwalk directly below the Centre’s air-conditioning coolers, however, he would dizzily realise that it is twice as high as leatoits de Paris. Furthermore, the obsessive verticality of this facade’s 90-odd service-risers and structural elements does not optically increase the apparent height, rather, it seems to echo the close-spaced verticals of the conventional Parisian constructions on the other side, and draw the two sides of the street into a closer unity.

Like a Jersey tank-farm

Admittedly, again, the form, function and what is fashionably termed ‘meaning’ of Pompidou’s blue, green and orange tubage is utterly unlike that of the reassuringly boring Parisian constructions on the other side. Not for nothing does local humor call it the Pompidolium-it does look like an oil refinery. In built fact, it looks less like an oil refinery than the famous coloured photographs of the great model suggested, but this is mostly because the form of the narrow Ruedu Renard prevents one from anywhere seeing the full panorama of that facade.

View south from the Musee d'Art Moderne, with Notre Dame seen thought the diagonal bracing of the structure. The sculpture is by Hans Arp.

View south from the Musee d’Art Moderne, with Notre Dame seen thought the diagonal bracing of the structure. The sculpture is by Hans Arp.

It brings to mind Denise Scott-Brown’s disdainful putdown of Archigram’s mid-’60s visions: ‘Many of the cities do look familiar. They look like the industrial outskirts of American cities, like Jersey tank-farms a d cracking towers, or the oil-derricks, pumps and cranes of San Pedro harbor ‘They thus revive the other immediate controversies that surrounded the design when it was young: how far was it, crudely put, a rip-off of the megastructure movement?

It is very difficult nowadays to see it as anything other than a kind of terminal monument to that movement. Even that colour scheme seems to say ‘Archigram’ (if not ‘Yellow Submarine’!), the concept of a stack of clear floors that can be adapted to a variety of cultural and recreational functions seems to recall the ‘Neo-Babylon’ of Constant Niewenhuis, or the Fun Palace of Cedric Price and Joan Littlewood, even if the project was never as radical as the floorless Fun Palace or as casually innovatory as Price’s Interaction Centre now completing for Ed Berman.

Even so, the historians must surely find it a place in the megastructure category, the last of that rare class of completed megastructures that begins with Cumbernauld Town Centre. More than that, they will have to concede that it shares with Cumbernauld the honour of being the most complete realisation of the megastructure dream-too many of its potential competitors, like Ontario Place (which looks even more like Archigram) lack at kind of central-city location that lends authority to megastructure’s urban pretensions.

What sets it apart from Cumbernauld, however, is its aesthetic-Geoffrey Copcutt’s design, however brilliant, was lumbered (willingly, no doubt) with that ponderous Corbusian monumentality of mere mass, whereas Pompidou’s Giedion/Archigram transparency and colour seem even truer nowadays to the departed aspirations of ‘the swinging ’60s’. Seen against low raking winter sunlight in the fresh snow of the last day of 1976, the west facade flashed with those ‘explosions of fire, ice and light’ that we were bidden to observe with our ‘third eyes of the soul’ a psychedelic decade earlier.

‘Had anyone set out deliberately to memorialize the Megastructure Age, the rest must surely look like this’

Had anyone set out deliberately to memorialize the Megastructure Age, the rest must surely look like this. What remains enigmatic is whether Piano + Rogers ever set out to do anything of the sort. Indeed it is already difficult to reconstruct whether they ever set out to do anything at all-even entering the competition is something that they admit they only did after massive persuasion from Ted Rappold of Arup’s (who is now Frei Otto’s demiurge as well, of course). Beyond that, however, the architects will admit to no more than the resolution of the architectural and technical problems posed by the brief.

This, notoriously, is all that most architects will ever claim and what makes it even more provoking in the case of Piano + Rogers is that what is claimed to be similarly modest attention to the business in hand in their earlier works had led to results that were equally striking and strikingly similar. Indeed, even if they had not claimed these prototypes themselves, a moderately observant person with some sense of styles would probably have identified their jointed-panel yellow plastic attic on the DRU building in London, the ARAM mobile medical unit project, and the B + B Italia building at Novedrate as the likely progenitors of Pompidou. It seems unavoidable that the critic must accept their view that the technology and the style of Pompidou derive from the internal logic of the development of their own previous works, whatever else the evidence of his eyes may suggest.

Nota Simple complex

Something else the evidence of the eyes will suggest, for certain, is that this facility, for all it is a single ‘building’, is still a complex. Functionally, it is a large simple loft building, but it is not simple in any other way, since the determined resolution of one, purely professional problem of considerable rarefaction and marginal relevance, has complicated its construction beyond belief.

This problem, as put to me by one of the engineers involved, concerns the relationship of the enclosing skin to the supporting columns. To the pure of mind (that is, ‘real architects’) it could not be acceptable to put the skin between the columns, nor flush with them, nor in front of them. The fact that all three of these solutions have proven mechanically serviceable in buildings of acclaimed architectural quality (including a masterpiece or two by Mies van del’ Rohe) does not avoid the objection that such solutions involve compromises either in the dimensional regularity of the glass skin itself, or in the modular regularity of the movable internal partitions that have to meet that skin at a column or elsewhere.

Therefore - and this decision affects everything else about the whole design-the glass skin runs through clear and uncompromised behind the columns, and offers an unvarying dimensional and mechanical condition for internal partitioning and other equipment to meet. Normally, however, when this kind of solution is employed, there is an intellectual, if not visual, botch wherever a beam from an external column has to engage with a floor slab’s edge. To look convincing, the column needs to stand well beyond the building skin, and the beam’s entry should disturb the external surfaces as little as possible and these two considerations together usually result in some unacceptable bending moments and other mechanical inefficiencies.

Pompidou avoids all by avoiding bending moments, at the cost of great visual complexity and an expenditure only thinkable in a privileged commission of this kind. The ends of the deep floor trusses are brought up to a pin-joint in the plane of the external skin, and are there supported by the short ends of giant ‘rocker arms’ (the so-called gerberettes), which pivot about pin-joints on the main supporting columns, what time the longer, outward cantilevers of the rocker-arms are literally tied down by the external tracery of vertical and diagonal braces.

Let it be said at once that however much one may question the necessity of doing things this way, what has been done has been done with immense skill and conviction. The massive steel elements have the blunt authority of high quality civil engineering construction, which is what they are, without pretentious playacting at machine aesthetics; the surfaces of the columns were fettled after casting by being struck off in sweeps of a hand power grinder, and each sweep can still be seen.

The use of this balanced-cantilever solution also generated a useful zone of space between the structural columns and the external bracery- a zone, which accommodates the services on the street side, and horizontal pedestrian circulation ducts on the piazza side. Diagonal pedestrian circulation, however, via the escalators, is slung clear of the structure, outside the bracery not only for visual effect but also because it reduces the impact of local fire regulations on the design. ‘Impact of fire regulations …’ there’s a phrase that’s been on the lips of every detail factor of the design, but over the months the tone of envy has changed to bewildered admiration. ‘How have they got away with it?’ In fact, they haven’t got away with much.

They have argued hard for their own preferred way of handling fire problems, and they have still made some major concessions. For instance, there is almost no straight-run glass wall on the street side-there wouldn’t be much of a view through the massed services and ‘all that stuff could be dangerous’. On the other three sides, the glass has remained-but that is a statement that requires serious qualification when applied to the side toward the piazza. Already on the enormous model of the final design, some opaque panels are shown near the columns, but hardly what has now accumulated there.

The final score, all to avoid radiant heat damage to the columns, now adds up to a full-height blank panel immediately behind every column at each floor blanked-off sections at the top of each glass panel on either side; water-spray heads, trained on the column, projecting from each of these blanking panels; and snap-down steel roller shutters to cover the glazed remainder of the flanking panels. And it is such a travesty of the idea of the clear and uncompromised glass wall that it calls the whole concept in question, and I find myself stunned that a design team that had poured so much inventiveness into setting up that very pure concept in the first place, should allow it to be ad-hocked to pieces in the last resort.

Plus (to coin a phrase)  ça change

Ad-hocism, of course, is somewhat in fashion at present, and will always have a big place in the art of design, but its visible presence at Pompidou is curiously disturbing. It is conspicuously inappropriate to a design, which otherwise has avoided manifest compromise. The existence of the great model of the final design shows how final that design already was by late 1972; the earlier models show how little has changed conceptually from the first competition design. In a period when far less complex designs have had to expect almost total transformation through ‘reasonable’ compromise, almost nothing has changed on this project. And this, surely, is its ultimate enigma.

Here is a response to a commission that called fur adaptability, interpreted by the architects in an extreme form that drove the idea of perfect adaptability almost as far as it can bed driven; therefore there was, presumably, a bigger potential for changes of direction and interpretation than normal. But the finished building simply looks like the model multiplied. The architects could claim that they got the design right in the first place. This must be a remarkable claim to make about a facility intended to be changed and adapted over along period of time, and it raises two very worrying questions; one operational, the other symbolic.

‘Centre Pompidou is clearly a monument, a very permanent monument presenting what is already a fixed image to outward view’

The operational question is one that is already concerning the architects much as it concerned Archigram as the design of their Monaco entertainment centre progressed: how will the place be managed? What is the point of producing a machine of perfect adaptability if it will not be imaginatively adapted-remembering that, the more nearly perfect the adaptability of the design, the fewer the clues the design will give on how adaptation should be wrought upon it.

The question of management will qualify all attempts to judge whether the Centre is functionally adequate; a fair number of critics will be sure to pan the gallery spaces as unsympathetic to whatever is exhibited there-but how many will ask whether they would be’ more sympathetic if someone other than Pontus Hulten was running the Musee d’Art Moderne and using them differently.

The symbolic problem is less scrutable. Centre Pompidou is clearly a monument, a very permanent monument presenting what is already a fixed image to outward view, and few of the routine modifications that might be adapted to its services and other externals is likely to have much effect on that fixed image of transparency and tracery, bright colour and mechanical equipment.

But can one have a permanent image of change? The problem is one that has fascinated philosophers and poets as long as language has been able to distinguish the two concepts, but it has troubled architectural theory only since the Futurists at the beginning of the present century decided to celebrate the impermanence of technology. On the whole, the response has been to design permanent statues to ideas of impermanence, but Piano+Rogers have gone further beyond that-impermanence, in the sense of adaptability, applies to everything except the most massive members of the structure.

Maybe it was necessary to generate a fixed image at an early point, in order to keep the rest of the design process under control, but if that fixed image can retain its present power until, say, the century’s end, Centre Pompidou will prove to have succeeded in one of the most teasing but central tasks that were in the unwritten (pace Giedion) programme of the Modern Movement.


Architects: Piano + Rogers.Engineers,cost controllers and project planne s ave Arup & Partners, Paris and London.

Client: The Ministries for Cultural Affairs,Finance and Education

Brief: A cultural centre to consist of four major specialist activities: Museum of Modern Art; a reference library; a centre of industrial design; and a centre for music and acoustic research, together with supporting services such as car park, restaurant etc., totaling 1 million sq ft for an approximate cost (in 1973) of 280 million French francs, to be completed by December 1976. An international competition, set up under President Pompidou’s direction in the summer of 1970, attracted 681 entries.The jury consisted of Jean Prouve (president), Gaetan Picon(vice-president), Emile Aillaud, Sir Frank Francis, Philip Johnson, Michel Laclotte, Oscar Niemeyer, Wilhelm Sandberg, Herman Liebaers and Henri Pierre Maillard.

Site: In the center of Paris within 1 km of Notre Dame, the Louvre, and on the edge of the densely populated historic Marais quarter. At the time of the competition there were no sizeable spaces for open-air activities in central Paris. The neighbouring Les HaIles, one of the world’s great markets and a dynamic centre of popular activity, were being demolished to be replaced above ground by a commercial centre, with a major public transport interchange below. The competition site had been mainly cleared in the ’30s. To the north a large new housing development is proposed.The buildings around the site are mainly of stone, seventeenth to nineteenth century, between 24 m and 34 m high.

The Rue du Renard is the only major road bordering the site. The new building leaves over half the site as an open public square for both specialist and amateur markets, exhibitions and visual happenings, circus, meetings, creches, games, cafes etc. The roads have been closed over a large area, so that shops can spill out over the pedestrianised area. Half the building is below ground; the rest is si tedon a north-south axis along the heavily congested Rue du Renard, both retaining its street quality and shielding the square on the west side of the building from noise and fumes.

Concept and organization: The client specified a number of specialist cultural activities, but the architects wished to broaden the brief while not forgetting specialist needs: tocreate a centre for both tourists and people who live in the neighborhood; not a centre divided into four water tight departments, but a true dynamic meeting place where activities overlapping flexible, well-serviced spaces. The greater the public involvement,the greater the success. There are four major zones: the environment and the square; the sub-structure; the superstructure; and IRCAM (not discussed here).

The multi-purpose auditorium at the south end of basement floor.

The multi-purpose auditorium at the south end of basement floor.

The superstructure is further subdivided into four zones: the five large open-plan floors; the west 6 m wide structural zones facing the square; the east 6 m wide structural zone facing the Rue du Renard; and the roof. The environment and the square have already been discussed under Site. The sub-structure at street and square level contains large public areas: forum, theatre, cinema, shops, reception, cafe, children’s areas, current events areas, exhibition areas etc. Below this are technical and storage areas, such as computer, photographic security control and mechanical services control.

Under the square are the bus, truck and car arrival and parking areas, benefiting from the view and giving constant life to the building: restaurant, experimental cinema and temporary exhibitions. Departments are open from 10 am to 10 pm. The west structural zone facing the square is for vertical and horizontal movement benefiting from a wonderful view over Paris. Escalators, lifts, escape stairs, glazed and open galleries or corridors, audio visual screens, announcements, exhibitions, animate and continue the activities of the square below. The east facade structural zone on the road contains all the mechanical services, goods lifts and fire stairs, with continuous steel galleries for ease of maintenance and change. The roof zones contain plant rooms, cooling towers etc.

Client organization

Contrary to the usual practice in competitions, this was carefully structured, with senior representatives from the ministries of cuIture, education and finance, under the control of an experienced senior civil servant. This body has defended the building and supported the architects in a number of critical situations. Unfortunately it lacked the finance, power or interest to change the official brief to cater for the numerous activities beyond the confines of the four specific ones, which would prevent the project from becoming an elitist cultural centre and make it a true university of the street. Much will depend on the new organization, which took over last March.

The terrace of the Musee d'Art Modern at the south -west corner. The view also shows the joint between beam and gerberette and between gerberette and column.

The terrace of the Musee d’Art Modern at the south -west corner. The view also shows the joint between beam and gerberette and between gerberette and column.

Growth, change and scale

The architects believe in buildings, which are able to change and adapt in answer to technical and/or client needs, not only in plan but in section and elevation. They believe in a framework which allows people freedom to do their own thing, the order, scale and ‘grain’ coming from a clear understanding and expression of the process of building; in the optimization of each individual element, its system of manufacture, storage, transportation, erection and connection, all within a clearly defined and rational framework; in a giant meccano set rather than a traditional static transparent or solid doll’s house.

Construction of the Centre began at sketch design stage, continued at an average of some £1.5 million per month and was subjected throughout to a continuously changing brief. There were two types of change: due to technical needs which happen mainly during construction, such as unforeseen fire, security, loading, cost, delivery, quality and political problems; and due to the development of the brief. The building has had to cope, and will continue to cope, with both types of change. Not until a few months before the opening, for example, was it decided areas the museum would occupy; and a late technical change was the substitution of solid for r-glass panels on the greater part of the Rue du Renard facade.

‘The building is inside out, making it both technically easier to adapt, clip on or remove components, and giving the building scale, transparency and movement’

It has been possible to adapt the deep plan floors and open-ended detailing to take these changes into account. Each of the major floors are166 m by 48 m by 7 m high with no fixed vertical interruptions of either structure, mechanical services or movement to limit the users’ freedom. These large open loft spaces are serviced both from the ceiling, where all ducts and conduits are exposed for ease of change and maintenance, or from the floor where there is a 14 cm high computer floor.

All vertical connections are run along the east and west sides of the building. Corridors, ducts, fire stairs, escalators, lifts, columns and bracing, which normally interrupt floor spaces, are exposed on the outside. The building is inside out, making it both technically easier to adapt, clip on or remove components, and giving the building scale, transparency and movement. All partitions in the superstructure are of dry construction and movable.


The design called for six clear uninterrupted floors 48 m by 166 m. in addition, 6 m on either side of the 48 m main span had to be reserved for circulation-people on one side and services on the other. These circulations routes had to be supported and enclosed by the structure. The problem was solved by providing a central main truss beam supporting the floors, and suspending it off tied cantilevers supporting the galleries (the gerberettes-a system pioneered by Gerber, the nineteenth century German bridge engineer).

Having decided to set the facade 7.6 m back from the outer structural line and 1.6 m back from the main column line, it was relatively easy to settle on a suspended span of 44.8 m between facades, and cantilever spans of 7.6 m, held down by vertical tension bars on the two outer structural faces. For clear detailing, each junction had to be simple and solve one problem at a time.

Thus the joint between the horizontal structure (gerberette) and the vertical structure (column) is separated from the place where the structure passes through the facade, which both defines the transition in the structure, and is in a position where the joint detailing becomes relatively straightforward. It was necessary to find a way of expressing the form and detailing of the structure, and at the same time to establish a method of maintaining a strict control of the design as it developed. Cast steel, molded to follow precisely the form required for its specific function and used here for all the joints, is ideal for this.

‘It was necessary to find a way of expressing the form and detailing of the structure, and at the same time to establish strict control’

The other two structural components used were solid round bars for tension and hollow tubes for compression. The method, therefore, of expressing the structural performance of the elements and controlling the design was in the choice of the components making up the main elements. The main steel structure is made from a small number of major elements: beams, gerberettes, columns, ties, bracing and floor panels.

Except for the floor, these main elements are made up from the three basic components. Columns are centrifugally cast-steel tubes (850 mm diameter). Ties are solid round bars (200 mm diameter), screwed into the end of the gerberette. Gerberettes are single cast-steel pieces, 8’2 m long and weighing 10 tones. Beams consist of a continuous double tube as top compression boom and a continuous double solid round for the bottom tension boom. Single members, alternately tube and solid round, makeup the diagonals and are welded on to the cast-steel nodes which are fitted and welded into the booms.

The bracing, which supports both compression and tension forces, is mainly tubes, but on the two outer structural faces of the long facades it is solid bar, pre-tensioned against compression. Bracing elements are screwed into cast components attached to the main elements, with the exception of the floor, which is made up from I sections bolted or pinned to the nodes of the top boom of the beams. These I sections support a composite 110mm concrete slab.

On plan the building is divided into 13 bays, each 12’ 8m wide and each supported by a standard frame. This consists of beams spanning on to the gerberettes, which rest on the columns and are restrained by the ties. Each frame supports six floors. Joints between beam and gerberette and between gerberette and column are pinned. Stability is provided by the bracing. Along the short end facades the beams in the frame are interconnected to provide transverse stability, the floors acting as horizontal beams which span 166mbetween the braced end frames.

All the frames are thus joined to provide lateral support for stability, wind and temperature. Longitudinal stability is provided by pre-tensioned vertical bracing on the two outer faces of the long facades. Horizontal bracing on every other level connects floor and column to this vertical bracing. The design was conceived as a sort of giant meccano, each element simply connected to its neighbor. All the elements were fabricated offsite, and the largest of the elements; the beams (46 m long, 3 m high, and weighing 75tonnes) were brought by rail from Germany to Porte de la Chapelle in northern Paris and then by road to the site. Three beams arrived each week. A single 500t crane was used to handle and erect the beams and other elements. It took 10 days to erect one bay, and the whole erection, including concreting all the floors, took eight months.

The terrace of the Musee d'Art Modern at the south -west corner. The view also shows the joint between beam and gerberette and between gerberette and column.

The terrace of the Musee d’Art Modern at the south -west corner. The view also shows the joint between beam and gerberette and between gerberette and column.


The services distribution systems can be divided into two major elements: 1 Infrastructure or concrete area of the building, which contains the major heavy equipment, refrigeration and boiler plant, water storage with pressurisation systems, main hv/mv electrical supply. Distribution in these zones follows conventional practice, being located predominantly at high-level in the major circulation routes.2 Superstructure or steel areas of the building where, to meet the requirement that the floor area of the building remains unimpeded, all services distribution, both horizontal and vertical, is contained in the 7 m wide zone on the Rue du Renard facade.

Only secondary ‘room’ distribution is run inside the building at high level. The building is ‘all electric’, the heating and cooling being provided from the following major equipment: 1 Centrifugal refrigeration machinery, part of which is filled with double-bundle condensers which recuperate heat from the air-conditioning exhaust systems and transfer this heat to the air-conditioning supply systems; 2 Direct-resistance electric boilers which ‘top-up’ the double-bundle condenser systems when the recuperated heat is insufficient for the demands of the building; 3 Hot-water heating storage vessels(holding two hours’ supply) which are fed by the heating boilers, a system which is used during periods of ‘peak’ electric tariff when it is not economical to use the boiler plant: 4 Direct electrical resistance heaters located round the perimeter of the floors, used during periods of extreme cold, and in pre-heating the building. The building is completely air-conditioned, and comprises 25 independent ‘all-air’ systems.

The connecting galleries on the west facade are transparent tunnels of steel framing and glass

The connecting galleries on the west facade are transparent tunnels of steel framing and glass

The principal installation is at roof level, which serves the five superstructure floors, and comprises 14 dual-duct, variable-volume, high-velocity systems, the variable-volume feature being chosen to allow for the anticipated large variation in lighting levels for any specific zone, and also to enable the air-conditioning to any floor, or part of a floor, to be shut down when that zone of the building is not in use. Plant rooms in the infrastructure located near the conditioned zone contain the remainder of the air-conditioning plant.

The building is fed from three independent 20kV cables, each 10MVA, one of these coming from the securite grid and acting as standby. Transformers are located throughout the building, near centers of electrical power use, and have a total capacity of 42MVA, the estimated instantaneous power consumption being 2OMVA.Except in the circulation areas, stores, etc, lighting tracks on a modular basis have been installed. These tracks are designed to permit an average lighting equivalent to 50 watts/m2, but allowing lighting peaks of 80 watts/m2 in local areas, with a corresponding reduction elsewhere in the zone.

A major element in the services design and construction has been the requirements of the fire authorities. These include: that all areas of the building have mechanical smoke exhaust systems to give 12 air changes per hour; that the building is equipped throughout with sprinklers, except in ‘high electrics’ areas where halogen gas systems are used; that in addition certain parts of the facade structure have water spraying; that rise-of-temperature detection systems are installed, linked back to a central fire control station, and that these systems operate automatically the smoke exhaust systems; that all services support elements have t-hour, 2-hour or3-hour fire ratings, depending on the service they support; that in addition special precautions are taken to avoid the hour services collapsing on to 2-hourservices in the event of fire; and that fire dampers and fans in air-conditioning systems, in addition to the normal fusible links, are motorized and controlled by the fire detection systems. All mechanical and electrical installations are linked to a central computerized supervisory system.

The main entrance hall and reception area. The escalators lead to the north mezzanine which houses the Centre de la Creation Industrielle.

The main entrance hall and reception area. The escalators lead to the north mezzanine which houses the Centre de la Creation Industrielle.

Cost control

The necessity of a contract cost control team became clear when the architects were commissioned to take the drawings to an advance sketch plan state and to submit an estimate. The client had declared his intention of completing the building within five years. A government decree, requiring public authorities to pay professional fees based on a percentage of the estimate prepared by them as the realistic cost of the work (and not as before, based on a percentage of the final contract sum) was to become effective early in 1973.

The building had been designed for a competition without a budget. The client’s initial budget (not apparently based on proper calculation) proved to be considerably less than the estimate arrived at by the design team, so new guide lines had to be set and the estimates studied in more detail to establish an overall cost compatible with the ideas of all parties. The revised estimate (January 1972), waste be developed within two months into the official Avant Project Sommaire(APS) and Cout Objectif Provisoire(COP). Under the APS and COP (approved August 1972) the design team was responsible for keeping to the estimate ± 20 per cent. Variations in either direction would result in the application of a penalty clause.

Exposed services on the east facade showing the air-handing units on the roof and the blue air-conditioning ducts from them.

Exposed services on the east facade showing the air-handing units on the roof and the blue air-conditioning ducts from them.

An additional penal ty for failing to complete the building on time would be shared with the contractor.  The design team’s next task was to establish within a further six months the Cout Objectif Definitif(COD), which involved bringing the project to a detailed design stage, or Avant Projet Detaille (APD). The tolerance margin was reduced to ± 12 percent, and the design team would be held responsible for economic changes in market conditions (supply and demand) but not for national fluctuations. Because no suitable programme could be found to put the cost control on computer, most of the operation was manually executed.

Every month cost control statements were submitted to the client, showing the COD breakdown by contracts, and the updated actual cost of the latest estimate check and of works contracts actually placed. In the latter case the value of any approved variation order and estimates of future variations were added. The submission of all these figures built up to a final account which was expressed in January1973 figures and compared with the COD.To comply with the five-year programme the design team proposed, and the client accepted, the engagement, for a fixed fee, of a management contractor.