Originally published in AJ 13.05.87, this piece by Peter Blundell Jones and Waldemar Januszczak debates the merits of Venturi, Scott Brown & Associates’ designs for the Sainsbury Wing
Peter Blundell Jones
Venturi’s National Gallery extension is very much to do with inbetween-ness; between monument and commercial street, between Trafalgar Square and Pall Mall, between differently oriented parts of the urban grid. This surely is appropriate, for in function it must be subservient to the mother institution to which it is bound by an umbilical link, while the site conditions could scarcely be more pressingly asymmetrical. Response to these conditions inevitably produces disharmonies and disjunctions relished by Venturi, although disdained by his critics, but perhaps tranquil harmony is unattainable here.
Certainly the more pavilion-like proposals such as that of Stirling seem unconvincing in comparison with the winner just because they fail adequately to engage the site. Venturi accepted asymmetry from the start. Only the gallery floor has a central axis, while the all important main staircase is placed not centrally by asymmetrically outside, between old and new. This is his crucial strategic move, for it deals with two of the major peculiarities of the task: that the piano nobile is two floors up and that it is connected at that level to next door. Venturi’s great running stair both shows what the building is about and clarifies the point of transition, for everybody enters the new galleries at the same place. It also sets the staircase in a transitional role between galleries and Jubliee Walk which promised, in some of the other schemes, to be somewhat dead. The side galleries in turn have windows on to the staircase, making it an inside/outside space like the stair at Lutyens’ Castle Drogo, then they look further across Jubilee Walk to remind one of the mother building beyond. Stairs in Jubilee Walk itself rise in the same direction as those within the glass wall, suggesting interaction between the life of the city flowing through and life within the gallery.
The top floor axis, along which the largest and hierarchically most important exhibition rooms are aligned, gives some sense of clarity and completeness to the extension, finding its termination in a large end window.
Interestingly, it is shifted 43/4 degrees with respect to the orientation of the mother building, precisely one third of the 141/4 degree angle change between it and Pall Mall, an acknowledgment of inbetween-ness with appropriate hierarchical loading.
On the large scale figure-ground plan where it compares with the axial shift of St Martin-in-the-Fields, it looks convincing. This shift also allows the trick perspective in the main stair, which will seem longer from the top, shorter from the bottom, inverting the precedent of the Scala Regia. Finally, it produces the axial disjunction where the route from the main gallery penetrates the extension, making this transition felt.
The flanking galleries are differently oriented on the two sides, in keeping with the accepted asymmetry, and reveal their subordinate nature by taking up the angles of the site boundary. The irregularities – which incidentally are no more drastic than one finds in many an old palazzo now serving as a gallery - should be less obvious in practice than they are on plan, and will not really change one’s relationship with the paintings: after all, one does not stand resolutely on axis when one is looking at them.
Top down design
Venturi evidently organised his building from the top down, which was hierarchically appropriate, although it leads to some inelegant planning at the lowest levels, particularly in getting the stair to work with the lift. His asymmetrical strategy allowed an appropriately hierarchical use of the edges, so the south-east corner looking towards Trafalgar Square is taken up with a generous foyer at ground level, and the restaurant - an increasingly important and often undervalued element - takes pride of place at first floor. Functions slightly lower on the agenda such as the shop and conference rooms get the best end of Whitcomb Street, while the servicing of the building is all at the back. This inevitably leaves the streets behind rather dead but, as the extent of public face required by the brief was less than the available perimeter, it could have been avoided only by bringing in new public elements, for example, by letting out the ground floor back edges as shops.
Despite all the foregoing, it is the facade on to Trafalgar Square which is provoking the most comment, and was perhaps crucial in gaining Venturi the commission. After the carbuncle incident nobody dared do anything uncompromisingly modern, but a return to tradition proved less easy than the heir to the throne seemed to suggest.
What can one add to a well loved face that is not a wart or a tumour, or something equally monstrous? A second nose, an extra eye or ear, no matter how perfect in itself, is still a freak to be excised surgically at the earliest opportunity. The metaphor is a little extreme, for faces do not have to be as symmetrical as faces, but even so the National Gallery is symmetrical and there is no way of adding more of the same without throwing it out of balance.
Gunnar Asplund explored this problem half a century ago in a series of elevation drawings for extending Gothenburg Town Hall demonstrating that what might seem at first sight the obvious solution merely throws the main building off balance and devalues it. Building a second smaller version also proved unconvincing for in that case, as in this, the new building was an extension whose dependency needed to be visible: thus it could not be allowed too much autonomy.
The problem in our case is further exacerbated by the change of angle and of street line, for the National Gallery is set back and the line of Pall Mall does not meet its south-west corner, with inadequate site space simply to pull the new building back.
Venturi’s strategy is cleverer than it looks, because behind the games with columns lies a crucial shift of the troublesome corner. He moves it to a position where it can echo the
south-west corner of the old building, giving it a convincing raison d’être framing the now symmetrical entrance to Jubilee Walk, enhancing the identity of this pedestrian street which runs through to Leicester Square. This symmetrical figure centred on a void, the street, is of course in direct conflict with the symmetry of the elevation of the National Gallery as a whole, a solid; but Venturi is bargaining on the fact that the elevation is so long that it is not normally seen as a whole. Jubilee Walk has a pair of gates and the strong sense of threshold produced at this point gets Venturi around the corner allowing him to switch to a modern glass facade, since it belongs to a different space.
Having established this corner he has to be very careful not to create another one, but to join up with Pall Mall in one go. Hence the double angle change at 40 degrees and 60 degrees with respect to the new axis, which allows the facade to be read as a continuous wall between Pall Mall and Jubilee Walk.
The controversy over the columns, both within and without, and over the classical details, which dissolve away as they get further from the mother building, will doubtless be with us for a long time. But it is clear that without them Venturi would never have won, and after Rogers’ reception in the first competition it seemed unlikely that anyone proposing a modernist solution would remain in the running, regrettable though this undoubtedly was.
If columns are obligatory - and I personally would rather not see this currency debased by stretching it even further - then it may be better to have them in Venturi’s form than otherwise.
Borrowed from the past for purely iconic value, they are stuck on without reference to the tectonic reality of the building - very much ‘decorated shed’, to put it in the architect’s own terms.
Knowing historic references
Even so, many of the virtues of the winning scheme lie, as described above, in its asymmetrical organisation, its relation to the site, its hierarchy and disposition of spaces. These things could have been provided without the nodding and winking, the knowing historical references, arched thresholds and Latin inscriptions which may seem in retrospect a rather cheap way of achieving cultural legitimacy. In the current architectural climate this represents only a more sophisticated and subtle version of the gaudy columning and pedimenting indulged in by so many fashionable practices, and thus may be seen by posterity as expressing a current zeitgeist. Perhaps the trustees and the wider public have what they deserved.
Ancient and modern
In conclusion, we might ponder over the example of Asplund’s town hall, a historical lesson we seem to have missed. Trained in neo-Classicism; he found a way to engage certain themes in the mother building without slavishly adopting its decorative elements, producing an extension which respected its context and belonged to its own time and technology.
When compared with the ‘obvious solution’ extending it in the same vein, his modern version is not only better in itself, but better for the mother, which is allowed in turn to belong to her time, ageing gracefully without the violence of being ‘reinterpreted’ by a face lift signifying a youth she never lived. It is time we realised that tradition and modernity are not necessarily incompatible.
Prominently positioned in the display outlining the proposals for the National Gallery extension is a messy drawing done on the front of menu from the Grill Room of the Savoy by Venturi himself. The drawing is a swirl of orientation lines and arrows radiating from a central felt-tipped whirlpool.
The position of this drawing, first on the left as the enter the main display, at the symbolic beginning of the exhibition route, leads me to suppose that it contains a terribly important idea in a rough state, perhaps the crucial idea for the entire scheme. Certainly the drawing above it, this one scrawled on a Pan Am clipper class menu, unmistakably predicts the eventual distribution of volumes on the main elevation. Venturi obviously thinks well when he eats well.
The Savoy drawing traces in crude diagrammatic form the relationship of the extension to the points of the compass and the position of Trafalgar Square. The chronic orientation problems of the Venturi scheme can traced back to the ambition revealed here to give the new wing a face that faces Nelson’s Column.
It makes very clear why Venturi ended up cutting a corner off his building. The convolutions became necessary because this entire building is the architectural equivalent of a growing plan twisting itself around to face a window. This uneasy elevation is like a skirt that moved round until the zip is at the front when it should be at the side. A facade that ought to be sealing off an end is trying to straddle a corner. And so Venturi has done exactly what the painter does when he is unable to two edges meet convincingly: he smudges the join with his finger blurs the divide.
Approached from Pall Mall the building looks at its worst, its lack internal coherence reflected in an unfocused blur of converging edges.
A peripheral temple
The interiors have, after all, been like a permanent shrine to early Renaissance art. The banal Victorian device of carving the names of Masaccio, Piero, Mantegna on the walls of th ascending stairs brand the building as a mausoleum, so do the pretentious inscriptions in large Roman letters suggested for the vestibule. (Since this is supposed to be a populist building I hope that if the inscriptions are in Latin they will have English subtitles!) All these dodgy ageing devices cast the building in the role of a permanent temple to the early sources of Western art. But the exterior would have you believe its role was much more peripheral than that.
Nevertheless, Venturi’s main top-lit galleries look as if they will serve their purpose well enough. Heavily reliant on his beloved Soane’s example at Dulwich they form a series of open rooms with doorways large and tall enough to command some of those long gallery vistas that art-gallery architects adore.
Banging the nail in
Those critics who have commented that the main interiors look like a stage set for an art gallery have a point. Venturi is absurdly fond of the theatrical gesture masquerading as the witty conceit, witness the Borromini-style illusionistic arcade that will greet visitors crossing over from the old building to the new. There are many ugly and nonsensical qualities to this scheme but none is uglier or more nonsensical than this retreating perspective penetrating the main galleries from the side like a nail banged crudely into a plank. Having swivelled the orientation of Wilkins’ building Venturi is forces to resort to this silly gimmick to lead visitors on. This is architecture by problem-solving and it is most unsatisfactory.
In arranging the seemingly well proportioned galleries in aisles and making the nave of central galleries taller and wider than the flanking ones the architect is dealing in an ecclesiastical symbolism which some will deem appropriate - providing they can spot the top-floor church in a messy symbolic mix that also includes classical, Egyptian, Gothic, Baroque, art nouveau and Victorian elements.
Given the famously difficult conditions he was working under perhaps this is the best we could have expected of him or any architect. But I do not think so. I am not at all surprised that the proposed extension is a set of compromises and exotic details searching for a building. But the extent to which this is clearly the fault of Venturi rather than of the commission amazes me.