17. THE SPREAD OF THE ART NOUVEAU: THE WORK OF C. R. MACKINTOSH AND ANTONI GAUDÍ…

17. THE SPREAD OF THE ART NOUVEAU: THE WORK OF C. R. MACKINTOSH AND ANTONI GAUDÍ

THE initiation of the Art Nouveau by Horta in 1892 was sudden and its spread extremely rapid. Almost concurrently forms very similar to those he had invented began to appear in other European countries. Rarely has a new idea in the visual arts been taken up internationally with so little lag. Advanced artistic circles at this time were evidently thoroughly prepared to accept major innovations and new periodicals, starting up almost one a year, provided vehicles for their transmission: Pan in 1895, for example, Jugend in 1896, Dekorative Kunst in 1897, and Die Kunst in 1899, to mention only German magazines. Had the Art Nouveau not already been invented by Horta the year before, three works of art dated 1893, Aubrey Beardsley’s ‘Cello Player’, an illustration in black and white, Toorop’s picture ‘Three Brides’, and Munch’s ‘The Cry’, first a painting but widely available as a colour-lithograph the following year, might well have supplied the impetus for other designers to do so; doubtless such inspiration did encourage rivalry rather than direct imitation of Horta. In Germany a Munch exhibition in Berlin in 1892 and a Toorop exhibition in Munich in 1893 called attention to the long waving curves and the general linearity of style of these artists. In 1893, moreover, the Studio began to bring to designers and architects everywhere well-chosen illustrations of current English decorative work.

England itself was least responsive to the new Continental mode. It is, indeed, improper to call the Bishopsgate Institute in Bishopsgate in the City of London, built in 1893-4 by C. Harrison Townsend (1850-1928), Art Nouveau. Yet, despite its evident dependence on Webb, the way in which Townsend took the characteristically stylized but basically naturalistic patterns of contemporary English wallpapers and chintzes and used them in relief at architectural scale is as drastic an innovation as are the bits and pieces of more abstract stone carving that Horta used on his Brussels houses of these years. Townsend remained a ‘fellow-traveller’ rather than a member of the international Art Nouveau group for a decade. For example, the façade of his Whitechapel Art Gallery in the Whitechapel Road in the East End of London, designed in 1895 and built in 1897-9, is an improved version of that of the Bishopsgate Institute (Plate 134B). The broad and almost Richardsonian arch is placed off centre, the ornament is freer and bolder, and the few windows are organized in a continuous band below the plain wall of the upper portion.

Less successful, though perhaps more advanced, is Townsend’s Horniman Museum of 1900-1, a free-standing edifice in London Road, Forest Hill, south of London. This has less external ornamentation, except for the façade mosaic by Anning Bell, but there is a very plastically conceived tower with rounded corners placed at one side of the front façade. His church of St Mary the Virgin, consecrated in 1904, at Great Warley in Essex, is very simple, indeed rather Voysey-like as regards the buttressed and roughcast exterior. However, the elaborate decorations inside by Sir William Reynolds-Stephens (1862-1943) offer the most virtuoso example of Art Nouveau in England—at least they are about as close to the Continental mode as the English came.[380] No other English architect came nearer the Art Nouveau than Townsend; in quality, moreover, his work excels most of that done on the Continent by the various imitators and emulators of Horta, even if it lacks the humble integrity of Voysey’s best houses of these years.

The earliest and, later, the most versatile Art Nouveau architect of France[381] was Hector Guimard (1867-1942). But his first work of consequence, the complex block of flats in Paris called the Castel Béranger[382] at 16 Rue La Fontaine, which was completed after several years of construction in 1897, still represents a very ambiguous exploitation of the new ideas coming from Brussels. It must be remembered, however, that the original design almost certainly antedates by a year or two all other Art Nouveau work outside Belgium. Also notable is the fact that the façade of the Castel Béranger was premiated by the City of Paris in 1898, since this indicates the rapidity with which the new mode won approval in France.

In 1896, while the Castel Béranger was building, Siegfried Bing, a Hamburg art-dealer whose wares included Japanese prints—now even more in demand than at any time since their introduction to Europe in the late fifties—and also the new English decorative products, decided to open a shop in Paris. Bing’s Maison de l’Art Nouveau at 22 Rue de Provence was designed for him by L.-B. Bonnier (1856-1946) in the Belgian mode, which thereby acquired its familiar name. This shop was of no great architectural interest, however, except that it was the first of the multitude that were produced in the next ten or fifteen years. Not only in Paris but in most Continental cities large and small, and even in England and in America, where the Art Nouveau otherwise hardly penetrated, these shop-fronts can still be noted; one of the finest has even been transferred from Paris to the Philadelphia Museum of Art in America.

Bing also enlisted the services of Van de Velde, still quite immature as a designer compared to Horta, but very articulate as a critic. Influenced more intellectually than visually by the English, Van de Velde’s personal development as a decorator now proceeded very rapidly. The lounge he designed for the Dresden Exhibition of 1897, for example, was an accomplished if somewhat heavily scaled example of an Art Nouveau interior and much more elaborate than those completed in his house at Uccle the year before.

By the time the Maison du Peuple in Brussels opened three years later in 1899 and Horta’s early career reached its apex of achievement, the Art Nouveau was already a favourite mode with young French designers and generally in rising favour in fin de siècle Paris. As a result even established architects were not averse to introducing its curves in interior decoration and for the detailing of exposed metal structural elements, although most of them had little understanding of its real possibilities. The giant stone colonnades of the Grand Palais in Paris, designed in 1897 and built in 1898-9 for the Exhibition of 1900, were presumably intended to rival those of the plaster palaces of the Chicago World’s Fair of 1893; but behind them the architectural team of H.-A.-A. Deglane (1855-1931), L.-A. Louvet (1860-1936), both pupils of Richardson’s master, André, and A.-F.-T. Thomas (1847-1907) provided a vast iron-and-glass interior detailed in a coarse sort of Art Nouveau way that is quite unrelated to the academic treatment of the exterior.[383]

The entrance feature, designed by René Binet (1866-1911), and the Pavilion Bleu by E.-A.-R. Dulong (1860-?), the principal exhibition restaurant in the Champ de Mars, were even more whole-heartedly à la mode. One can hardly regret, however, that these gaudy structures, unlike the Grand Palais, were only temporary. A much superior example of Art Nouveau decoration, Maxim’s Restaurant in the Rue Royale, remains intact as it was redecorated in 1899 by Louis Marney. This is full of period flavour and still splendidly maintained, but it has no real existence as interior architecture. Soon the Art Nouveau would be vulgarized in dozens of cafés, large and small, all over Europe. Of these the Brasserie Universelle in the Avenue de l’Opéra in Paris by Niermans, carried out two or three years after Maxim’s and lately demolished, was perhaps the most sumptuous; there, however, the new mode was eclectically combined with a lush Neo-Rococo.[384]

The architect Charles Plumet (1861-1925), working with the decorator Tony Selmersheim (b. 1871), built in 1898 at 67 Avenue Malakoff the first of a series of houses in which Art Nouveau decoration was grafted on to a general scheme of design that was more or less Late Gothic. This has also been demolished. Such eclecticism, based more usually on eighteenth-century models, is characteristic of the rapid Parisian dilution of the Art Nouveau and doubtless played a great part in its early descent into the obsolescence of the démodé. Yet Auguste Perret (1874-1954), in a large block of flats built in 1902 at 119 Avenue de Wagram, exploited in masonry a heavier and richer sort of Art Nouveau than Plumet’s with considerable success (Plate 134A). This edifice is in curious contrast to the flats of ferro-concrete at 25 bis Rue Franklin, designed by Perret in 1902 also, with which his career is generally considered to begin. Even the latter, moreover, have considerably more Art Nouveau feeling in their panels of faience mosaic than is usually recognized (see Chapter 18). The block in the Avenue Wagram is quite typical of French production in these years but of much higher than average quality.

The most accomplished French Art Nouveau designer remained Guimard, the first to take up the mode. His most conspicuous works, however, the Paris Métro entrances of 1898-1901, lie outside the normal realm of architecture (Plate 137B). These are executed entirely in metal of the most sinuous and vegetable-like character, and their extreme virtuosity is the more surprising in that they consist of metal castings produced in series. His no longer extant Humbert de Romans Building of 1902 in the Rue Saint-Didier in Paris, on the other hand, illustrated the usual difficulties of Art Nouveau architects when working with masonry. The exterior was neither Neo-Rococo nor Neo-Flamboyant but curiously crude and gawky in its originality, like his Castel Béranger, with none of the Art Nouveau grace that even Plumet sometimes evoked with success, or the rather lush ornamentation of Perret’s block of flats in the Avenue Wagram. The auditorium inside, however, employed curved structural members even more boldly than Horta had done in that of the Maison du Peuple. Here Guimard succeeded in giving a masculine vigour to the rather feminine forms of a mode already passing its brief prime.

As late as 1911, however, Guimard remained faithful to the Art Nouveau in an extensive range of contiguous blocks of flats that he built at 17-21 Rue La Fontaine near the Castel Béranger. For his own flat there he designed ironwork as boldly abstract as advanced mid twentieth-century sculpture in metal, but also as suavely elegant as comparable Rococo detail of the eighteenth century. The exteriors, moreover, which are entirely of stone, have a great deal of the refinement and restraint of Horta’s Hallet house of 1906 in Brussels. They are, however, more plastically treated with boldly moulded bay windows and attic storeys. Except for Perret’s, few Parisian blocks of flats of the period rival these in interest or in quality of design and execution.

Three Paris department stores of the early years of the century continued to use the metal-and-glass interior structure of Boileau and Eiffel’s Bon Marché, with notable success. In presumable emulation of Horta’s Innovation in Brussels, moreover, the architects of two of these extended considerably the external use of exposed metal introduced by Sédille at the Printemps in the eighties. These two stores remain, with Guimard’s Métro entrances, the most prominent Parisian examples of the Art Nouveau. The main branch of the Samaritaine[385] in the Rue de la Monnaie near the Pont Neuf was built in 1905 by C.-R.-F.-M. Jourdain (1847-1935). This has several fine galleried courts inside in the tradition of the Galeries du Commerce et de l’Industrie of the 1830s, but it is even more distinguished for the sturdy scale and the straightforward design of the external metal frame (Plate 133). The actual structural members are hardly bent at all by the exigencies of the mode; but they were characteristically ornamented not only with decorative metalwork but also with inset panels of polychrome faience, now painted over. On the north front, however, other panels, here of faience mosaic, remain visible; these are of even greater delicacy and elegance than Perret’s foliate panels in his block of flats of 1902-3 in the Rue Franklin.

The contemporary Grand Bazar de la Rue de Rennes, now the Magasins Réunis, at 134-136 Rue de Rennes by H.-B. Gutton (b. 1874) is generally fussier in design than the Samaritaine. Gutton achieved, however, a more completely volumetric expression, emphasizing the lightness and the thinness of metal-and-glass construction somewhat as the early monuments of the 1840s and 1850s in England had done. New shop-windows below and the removal of the open grillework that once rose against the sky have now much diminished its effectiveness. Binet’s earlier galleried court of 1900 at the Printemps was burned out in 1923, unfortunately. With the lifts rising in the corners and the staircases swooping down in great splashing curves, this court was altogether superior to his Entrance to the Exhibition of 1900 and even to Frantz Jourdain’s small later courts in the Samaritaine. It seemed somehow to epitomize what a great metropolitan department store ought to look like somewhat as Garnier’s Opéra epitomizes what later generations came to expect of an opera-house. If Prince Danilo supped with the ‘damen’ of Maxim’s, we can be sure the ‘Merry Widow’ and the ‘Pink Lady’ did their shopping here.

It was the Art Nouveau structures at the Exhibition of 1900 which first focused public attention on the new mode, occasioning also that rapid Parisian vulgarization which brought its early end. At the exhibition, besides the crude but conspicuous things designed by Binet and Dulong that have been mentioned, there was the Pavillon Art Nouveau Bing by Georges de Feure (1868-1928), a designer rather than an architect, which had rooms by Edward Colonna, back from working for Tiffany in America, and others of the best artists and craftsmen employed by Bing; but their exhibits represented decoration, not interior architecture properly speaking. However, by 1900 the Art Nouveau was not at all the strictly Parisian manifestation that it must have seemed to most of those who visited the exhibition. The Germans, notably, had already taken it up with great enthusiasm, beginning about 1897.

The Studio Elvira of 1897-8 in Munich by August Endell (1871-1925) had a plain stucco façade cut by a few strategically placed windows of varied shape; but this façade was splashed across the centre with a very large abstract relief of orientalizing character resembling something half-way between a dragon and a cloud. Endell’s studio, if not the first manifestation of the Art Nouveau in Germany, was certainly the most striking; moreover, it followed immediately upon the showing of Van de Velde’s Lounge at the Dresden Exhibition of 1897. Already, however, in that portion of the Wertheim Department Store in Berlin in the Leipzigerstrasse which was begun in 1896, Alfred Messel (1853-1909) had used a great deal of exposed metal and glass and even perhaps modified the detail a bit towards the Art Nouveau. This was five years before Horta designed the Innovation Department Store in Brussels and ten years earlier than Jourdain’s Samaritaine in Paris. Messel made the spacing of his heavily moulded masonry piers quite wide and opened up completely the bays between. The result was at least as close to Sullivan’s Gage Building of 1898-9 as to the Paris department stores of a decade later. In those portions of this department store that Messel added in 1900-4, however, the façades, although highly stylized, were of rather Late Gothic character and certainly quite remote from the Art Nouveau.

In 1899 Van de Velde moved from Paris to Berlin. There he designed the Hohenzollern Kunstgewerbehaus, a shop parallel to Bing’s Maison de l’Art Nouveau in Paris in its interests and its activities. In the next year he carried out the Haby Barber Shop and the Havana Cigar Store, two of the most extravagant of all Art Nouveau shop interiors. With the opening of the new century, however, in his full-scale architecture Van de Velde moved almost as rapidly away from the Art Nouveau as did Messel, although in a different direction (see Chapter 20). By this time strong counter-influences were reaching Germany from Glasgow and Vienna.

Although not disdaining the Art Nouveau as completely as did the English and the Americans, the Austrians showed little of the enthusiasm of the French and the Germans. There is in Vienna one block of flats[386] of about 1900 so completely Art Nouveau that it might well have been designed by Horta himself. But the leading Austrian architects, old and young, reflected the new Belgian mode only with considerable diffidence and restraint. Otto Wagner (1841-1918), long a well-established academic architect and indeed Professor of Architecture at the Akademie, introduced more and more Art Nouveau detail in the Stadtbahn stations that he built over the years 1894-1901, most notably in the one at the Karlsplatz with its curved metal frame and inset floral panels. However, even this seems tentative and hardly rivals in interest Guimard’s contemporary Métro stations in Paris.

Wagner’s so-called Majolika Haus, a block of flats at 40 Linke Wienzeile designed about 1898, is far more distinguished and original (Plate 138A). Although the ironwork of the balconies is here and there curvilinear in detail and the faience plaques that completely cover the wall are decorated with great swooping patterns of highly colourful flowers, the architectonic elements of the façade are nevertheless very crisp, flat, and rectangular. That Vienna would very shortly become the focus of a reaction against the Art Nouveau does not seem surprising in the light of this façade. Moreover, on an office building erected in the Ungargasse for the firm of Portois & Fix in 1897 by Max Fabiani (b. 1865), who had been Wagner’s assistant in 1894-6, the coloured faience slabs which sheathe its surface are arranged in a purely geometrical chequer-board pattern; only the ironwork has a slightly Art Nouveau flavour. In the late nineties it would be hard to say whether Art Nouveau influence was arriving or departing but for the projects other Viennese architects were publishing in the review Ver Sacrum started in 1898.

The design of the art gallery built in the Friedrichstrasse in Vienna in 1898-9 for the Sezession, a newly founded society of artists in revolt against the Academy, by J. M. Olbrich (1867-1908) seems more influenced, however, by the façade of Townsend’s Whitechapel Art Gallery—only just begun but already published as a project in the Studio in 1895—than by the work of the Belgians or the French, which had affected him strongly in the immediately preceding years. The pierced dome of floral metalwork alone vies in virtuosity with Horta or Guimard, and the pattern of this is actually quite English in character. The bronze doors are by Gustav Klimt, an Austrian Post-Impressionist who can be grouped, up to a point, with the Dutch, Belgian, Norwegian, and Swiss Post-Impressionists mentioned earlier (see Chapter 16). Olbrich was called to Darmstadt in Germany to work at the artists’ colony sponsored there by the Grand Duke Ernst Ludwig in 1899 and Darmstadt, like Vienna, soon became a centre of reaction against the Art Nouveau under his leadership (see Chapter 20).

Both in Vienna and in Darmstadt the influence of the Scottish designer Mackintosh helped most to crystallize an alternative mode. Mackintosh first exhibited a room on the Continent at Munich in 1898, the same year that Baillie Scott was called by the Grand Duke to decorate an interior in the palace at Darmstadt. In 1900 Mackintosh was invited to design a room in the Sezession Exhibition in Vienna. That exhibit undoubtedly encouraged Viennese architects, already diffident towards the Art Nouveau, to turn very sharply away from it. This Adolf Loos (1870-1933) had already done in designing a completely rectilinear shop interior in Vienna in 1898. Loos, Wagner after about 1901, and Wagner’s pupil Josef Hoffmann (1870-1956) were all leaders in the international reaction against the Art Nouveau (see Chapter 20). The position of Mackintosh, however, is rather hard to state so categorically and must be considered here in more detail.

At home in Scotland Mackintosh’s early decorative work of the mid nineties approached Continental Art Nouveau more closely than that of any other Briton, not excluding Townsend. Indeed, he was castigated by his compatriots and his English contemporaries for participating in so exotic a movement. But Mackintosh also came nearer to possessing genius than most of the men of his generation associated with the Art Nouveau, not even excluding Horta. That genius, all the same, was of so ambivalent a nature that he could seem for a few years to go along with the general stream of Continental fashion and yet, almost at the very same time, provide also a real protest against its excesses and its superficialities by the craftsmanlike integrity and the almost ascetic restraint of his best work. That protest the Austrians and the Germans were not slow to heed.

Mackintosh made his first mark in Glasgow, which had earlier been the home of the highly original ‘Greek’ Thomson (see Chapter 4). By the nineties, moreover, interest in contemporary French painting was probably livelier there than it was in London. But Glasgow was also as notorious as Chicago, that major focus of architectural achievement in the America of the nineties, for its presumed philistinism. Touches of Mackintosh’s hand can be distinguished in work of the office of John Honeyman (1831-1914) and his partner Keppie, where the young architect was employed at the start of his career, notably in the Martyrs’ Public School in Glasgow of 1895. But it was in the decoration of the first of a series of Miss Cranston’s ‘tea-rooms’ (scottice, restaurants), the one in Buchanan Street remodelled by him in 1897-8, that Mackintosh’s personal talents were first effectively exploited. His very earliest decorative compositions and the murals that he and his wife provided here, full of heavy and presumably Gaelic symbolism, are parallel to, rather than derivative from, the work of the Belgians. They are, in fact, much closer to the drawings of Beardsley and the paintings of Toorop and Munch than to the plant-like ironwork and almost Neo-Rococo carved stone ornament characteristic of Horta. But the same long swinging curves are present, the same linearity, and the same rejection of all stylistic influence from the past.

In this same year 1897 Mackintosh’s firm had the good fortune to win the limited competition for the Glasgow School of Art with a project that was entirely their young designer’s (Plate 132A). Thus he very soon had an opportunity to prove himself architect as well as decorator in a way that only two or three of the Europeans associated with the Art Nouveau had been able to do up to this point. The school was built during the next two years, just as Horta was finishing his Maison du Peuple in Brussels. The only element in the design that relates to the contemporary Art Nouveau of the Continent is the ironwork. This is quite incidental to the major architectonic qualities of the building, moreover, since it is purely decorative, not structural. It is also extremely restrained in its abstract curves, like Fabiani’s of this date in Vienna, and almost totally devoid of vegetable or floral reminiscence.

The entrance to the Glasgow Art School seems to derive from Webb, but, like that of Townsend’s contemporary art gallery in London, it is rather less traditional in character than Webb’s work of this period. The somewhat wilful asymmetry and the plastic elaboration of the central part of the façade contrast nevertheless with the straightforwardness of the general treatment. There are two ranges of very wide studio windows—reputedly derived from a Voysey project—like ‘Chicago windows’ but larger, with the reinforced-concrete lintels above them frankly exposed, and little else in the whole composition. To later eyes this façade, expressing so clearly the uncomplicated plan that it fronts, tends to appear deceptively simple and obvious. But Mackintosh’s very sensitive proportions and the delicate touches of linear detail provided by the ironwork create a design at once very direct and very subtle.

The north end of the building is a tall plain wall of rather small-scaled random ashlar broken only by a few strategically spotted windows of various shapes. At once medievally dramatic and quite abstract, this façade makes one appreciate all the more the almost classical serenity and horizontality of the main front. The Art School is clearly the manifesto of an architectural talent of broad range and great assurance—very different indeed from that of Voysey.

Mackintosh was not alone in Glasgow in these years. A real ‘school’ existed, chiefly in the field of decoration, of which George Walton was another notable exponent.[387] Like Baillie Scott and Ashbee, Walton had some success as an architect in England (see Chapter 15) as Mackintosh did not, even though he executed a few interiors below the Border. But local support was not what it should have been for any of them in either Scotland or England. While the Art School was in construction, however, Mackintosh was asked in 1898 to provide the already-mentioned room in Munich, first of many that he showed at various exhibitions in Germany and Austria. This interior was very different indeed, both in the basic rectangularity of the forms and in the delicacy of the membering, from Van de Velde’s Art Nouveau Lounge at the Dresden Exhibition of the previous year. Thus, even before Van de Velde reached Berlin in 1899, a new line of influence from Glasgow into Germany—and soon into Austria also—was established whose general tendency was in sharp opposition to the lusher currents flowing from Brussels and Paris.

When Olbrich settled in Darmstadt—just before Mackintosh’s room was shown at the Sezession—he also rejected almost completely in the work he carried out at the Grand Duke’s Art Colony the still slightly Art Nouveau leanings—in any case already closer to the English Townsend than to Horta or Van de Velde—of his newly completed Sezession Building (see Chapter 20). Only his Pavilion of the Plastic Arts of 1901 at Darmstadt retained curved elements, and those were structural rather than merely decorative. The general rectangularity and the broad horizontal windows of the Ernst Ludwig Haus, a block of artists’ studios also completed by Olbrich in 1901, suggest comparison with Mackintosh’s Glasgow School of Art. Whether or not, in fact, Olbrich knew Mackintosh’s building—he may well have seen drawings if not photographs of it—his approach here was certainly very similar.

Mackintosh had a good many further opportunities as a decorator, both at home and abroad, but only too few commissions to design whole buildings. However, his two houses near Glasgow, Windy Hill at Kilmacolm of 1899-1901 and Hill House at Helensburgh of 1902-3, are both very notable. Externally they have a certain generic similarity to Voysey’s, with their moderate pitched roofs of dark slate, roughcast walls, and plain stone trim. His prototypes are not English but Scottish, however—the simple seventeenth-century houses of the minor lairds. As one would expect from his interiors, moreover, the façades of Mackintosh’s houses are much more carefully and abstractly composed than Voysey’s; they even include some simple geometrical features that are not at all reminiscent of the past in their design. Like Voysey’s houses, Mackintosh’s show no real novelties in planning, although the disposition of the rooms is always straightforward and commodious. The interiors are very original and rather less forced than those he was producing for exhibitions on the Continent.

Mackintosh built very little after 1903 except the Scotland Street School of 1904 in Glasgow, the north wing of the Glasgow Art School in 1907-8, and the finest of the various tea-rooms that he remodelled for Miss Cranston. This was the Willow Tea Room in Sauchiehall Street of 1904, for which he remade the façade as well as reorganizing the interior. Internally this tea-room was arranged on several interrelated levels subdivided by ingenious screenwork; the exterior was a flat surface of white stucco cut by broad horizontal openings, one to a storey. The Scotland Street School is equally straightforward in design, the rather plain façade with its ranges of horizontal windows being flanked by rounded stair-towers articulated into continuous stone grids by mullions and transoms, like the bay windows of Voysey’s Broadleys but much taller.

The north wing of the Glasgow Art School is more remarkable, quite worthy of the original front but much more stylized (Plate 135A). Where the front is strongly horizontal the new end façade, like that on the south, is markedly vertical, in part because of the way the ground falls off. But the tall oriels, glazed at the outer plane of the stonework, are striking features, and the whole composition is tense and dramatic. The library inside is a tour de force of spatial subdivision somewhat like the Willow Tea Room. Most notable is the way the rectangular stick-work makes manifest the complex articulation of the total volume. This sort of handling of interior space was unique up to this time as a product of conscious design, although already present inside Paxton’s Crystal Palace in the mid nineteenth century. Certainly there is no evidence here of a decline in Mackintosh’s creative powers; indeed, quite the contrary. Yet this library proved to be his swan song; for want of further commissions Mackintosh’s career all but closed at much the same time that the Art Nouveau was coming to an end on the Continent. Not since Ledoux perhaps had so great a talent been thus thwarted by circumstances, although just what the thwarting circumstances were, other than Mackintosh’s own temperament, is not so evident as in the case of the revolutionary French architect.

The Art Nouveau, so extensively propagated by exhibitions, is often thought to have terminated with an exhibition, that held at Turin in 1902. This is more than a slight exaggeration, as various already mentioned buildings executed as late as 1911 will have made evident. Yet after the early years of the century the decline of the Art Nouveau was almost universal except in provincial places and in outlying countries such as those of Latin America and eastern Europe. At Turin the Belgian section had characteristic Art Nouveau interiors by Horta. Mackintosh, wholly detached by now from the Art Nouveau, contributed a Rose Boudoir, typically light in colour and delicate in line with the predominant verticals and horizontals relieved by little abstract knots, so to say, of curvilinear decoration. Raimondo D’Aronco (1857-1932), the Italian architect responsible for the principal pavilions, wavered between a rather plastic, somewhat Neo-Baroque, version of the Art Nouveau, not unrelated to the seventeenth-century work of the great local architect Guarino Guarini, and a crisper mode much influenced by Mackintosh and the Viennese.

D’Aronco’s finest building, however, was not at Turin but the Pavilion of Fine Arts that he designed for the Udine Exhibition the next year. Moving sharply away from the turgidity of much of his work at the earlier exhibition, he produced for Udine a façade that was unified in design, frankly impermanent in its materials, and at once festive in spirit and dignified in tone. This was a most distinguished piece of exhibition architecture in a period when leading designers gave a great part of their attention to such rather ephemeral things—largely, doubtless, because so few opportunities to build permanent structures came their way. In Istanbul, D’Aronco built a small mosque in 1903, prominently located by the Galata Bridge, and also several blocks of flats that signally fail to maintain the promise of his Italian exhibition buildings. The very awkwardly sited mosque, raised on top of an existing structure, is as Viennese in character as the Udine pavilion.

Other Italian architects, however, remained faithful for a few years to the stile floreale, their version of the Art Nouveau. In Milan the Casa Castiglione, a palazzo or mansion-like block of flats at 47 Corso Venezia built by Giuseppe Sommaruga (1867-1917) in 1903, is a very large and ponderous example. The detail is extremely bold, inside and out, the materials rich, and a very large part of the interior is given up to a monumental stair-hall of almost Piranesian spatial complexity. A Milanese hotel at 15 Corso Vittorio Emmanuele of 1904-5 by A. Cattaneo and G. Santamaria is of a comparable extravagance. Finer perhaps, certainly simpler, is the Casa Tosi of 1910 at 28 Via Senato in Milan by Alfredo Campanini (1873-1926).[388]

To judge from the rather stile floreale character of some work of this period in Latin America, Italians as well as Iberians may well have carried the Art Nouveau there. In Cuba and Brazil, especially, memories of Colonial exuberance encouraged a profusion of carved or moulded ornament beyond even the excesses of the French around 1900. The most prominent example, but not the most characteristic, is the Palacio de Bellas Artes in Mexico City begun for President Diaz by Adamo Boari after 1903 and completed in 1933 by Federico Mariscal; this is ‘Beaux-Arts’—not inappropriately, perhaps—in all except its detailing; in the latest portions this reflects the Paris of the Exposition des Arts Décoratifs of 1925 rather than the Art Nouveau Paris of 1900.

In Spain itself the international current of the Art Nouveau was not very influential outside Barcelona. Gaudí, whose earlier work of the seventies and eighties has already been described (see Chapter 11), continued to be as much apart from the contemporary Spanish architectural scene as he was from the international Art Nouveau. His finest late works, moreover, all but post-date the demise of the Art Nouveau in the major European capitals. Nor is there any such close, if ambivalent, linkage between Gaudí’s career and the general rise and fall of the mode as in the case of Mackintosh. One can only say that his personal style is more closely related to the Art Nouveau than to the new stage of modern architecture that was already succeeding it by the time he produced his final masterpieces. The premonitory character of his early ironwork has been discussed and illustrated already (Plate 96B).

Gaudí’s work on the church of the Sagrada Familia[389] in Barcelona went on more or less continuously from 1884 to 1914 and began again in 1919 after the First World War. The most conspicuous portion that has so far been executed, one of the transept façades, was designed and largely built in the nineties. Dominating Barcelona with its four extraordinary towers—not finally completed until after Gaudí’s death in 1926—this façade, begun in 1891, breaks quite sharply with the Neo-Gothic of Villar’s crypt and his own chevet. The portals with their steep gables have a generically Gothic ordonnance; but the extraordinary profusion of sculpture, mostly executed after 1903, gives a highly novel flavour. While conventional enough as regards the figures, this is otherwise either naturalistically floral or else meltingly abstract. It resembles the Art Nouveau in many minor details, but is generally bolder in scale, more fully three-dimensional, and, in places, somewhat nightmarish.

Although only about two-thirds as tall as the cluster of towers intended by Gaudí to rise over the crossing, the four openwork spires above this façade—with the two in the centre taller than those on the sides—reach a wholly disproportionate height in relation to the roof that should ultimately cover the still unbuilt transept. At the top they break out into fantastically plastic finials whose multi-planar surfaces are covered with a mosaic of broken tiling in brilliant colours. The prototypes for these finials are the chimney-pots of the Palau Güell, but here their note of free fantasy is raised to monumental scale. The inspiration of the towers, so remote in character from anything that the Art Nouveau ever produced, came from certain native buildings which Gaudí had seen in Africa: these strange primitive[390] forms he first exploited in a project of 1892-3 for the Spanish Franciscan Mission in Tangier which was never executed.

In posse the Sagrada Familia is perhaps the greatest ecclesiastical monument of the last hundred years; beside it such a suave late example of monumental Neo-Gothic in England as Liverpool Cathedral, begun by Sir Giles Gilbert Scott in 1903, lacks both vitality and originality of expression, if not nobility of scale. However, Gaudí’s church still remains a fragment, and a very incoherent one at that, even though he prepared in 1925, the year before his death, a brilliant new project for the nave. Gaudí really stands or falls by the few secular buildings that he was able to carry to completion, beginning with the Palau Güell of 1886-9 (Plate 96B), and not, as many compatriots assume, by the unrealized—perhaps unrealizable—plans for the Sagrada Familia. (Construction has gone slowly forward, however, on the other transept for a decade now.)

Gaudí’s next Barcelona mansion after the Palau Güell, that built at 48 Carrer de Casp for the heirs of Pedro Mártir Calvet in 1898-1904, is much less impressive. Baroque rather than medieval in its antecedents, this is interesting chiefly for the detailing of the ironwork; but even that is no more remarkable here than that at the Palau Güell of a decade earlier. It is of interest, however, as illustrating the support which Gaudí received all along from his fellow citizens, that the Casa Calvet was awarded a prize in 1901 as the best new façade in Barcelona, quite as Guimard’s Castel Béranger was premiated three years earlier in Paris.

A wholly new spirit, quite comparable in its total originality to the Art Nouveau, first appears in the work that Gaudí did for Don Eusebio Güell at the Park Güell (now the Municipal Park of Barcelona), carried out over the years 1900-14, and in the walls and the gate he built in 1901-2 for the suburban estate of Don Hermenegildo Miralles in Las Corts de Sarriá. In the latter all the forms are curved and no stylistic reminiscence whatsoever remains, but it is a production of minor importance compared to the park. The park is mostly landscaping, but partly architecture in that it includes several small buildings and much subsidiary construction. A sort of Neo-Romantic naturalism, exceeding in fantasy that of the most exotic landscape gardening of the eighteenth century, controls the whole conception. Sinuous and megalomaniac near-Doric colonnades of concrete support a sort of flat vault that is of great interest technically;[391] yet these colonnades also suggest artificial ruins of the eighteenth-century sort raised to giant scale. The other porticoes and grottoes, however, recall no architecture of the past. Their rubble columns seem rather to emulate slanting tree-trunks, but in fact their profiles were worked out statically with the most careful study of the forces involved.

The ranges of curving benches surrounding the great open terrace over the Doric hypostyle, although covered with a mosaic of the most heterogeneous bits and pieces of broken faience, seem like congelations of the waves of the sea; the roofs of the lodges, also tile-covered, toss in the air like cockscombs. A strange biological plasticity, rather like that of the small-scale carved detail of Horta’s or Guimard’s buildings very much enlarged, turns whole structures into malleable masses as in some Gulliverian dream of vegetable or animal elements grown to monumental size. Everything but the ironwork is moulded in three dimensions, and even the ironwork tends towards a heavy scale more comparable to that of the structural members of metal used in Belgian or French work of the day than to the delicacy of Art Nouveau decorative detail.

Gaudí’s major secular works belong to the same years as the execution of the park. It is hard to believe that the Casa Batlló at 43 Passeig de Gracia in Barcelona, a small block of flats, is not a completely new structure but a remodelling carried out in 1905-7. This fact perhaps explains the relative flatness of the façade. Yet Gaudí made the lower storeys extraordinarily plastic and open, using a bony articulation of curvilinear stone members, and the high roof in front that masks the roof terrace is of even more cockscomb-like character than those on his park lodges (Plate 136). The upper storeys of the façade glitter with a fantastic plaquage of broken coloured glass considerably more subtle in tonality than his usual mosaic of faience fragments.[392] But architecturally the façade is handled more like Horta’s, with most of the windows nearly rectangular even though bulging balconettes of metal project at their bases. The effect, as with Horta, is slightly Neo-Rococo. But the sort of Rococo which this façade recalls is not circumspect French eighteenth-century work but the lusher mode that was exploited in Bavaria and Austria—and still more appositely in Portugal and Spain. The entire wall surface seems to be in motion, and all its edges waver and wind in a way that even interior panelling did rarely in eighteenth-century France. This effect of total motion is even more notable in the interiors, which seem to have been hollowed out by the waves of the sea.

The rear façade of the Casa Batlló is remarkable for its openness. The wide window-walls in the paired flats open on to sinuous balconies extending all the way across. Above, there is a simpler plastic cresting than on the front; over this the curious forms of the chimney-pots provide a range of abstract sculptural features covered with polychrome tiling, always a favourite terminal theme of Gaudí’s.

Much larger than the Casa Batlló is the edifice built for Roser Segimon de Milá in 1905-7 at 92 Passeig de Gracia, appropriately known in Barcelona as ‘La Pedrera’ (the quarry). Surrounding two more or less circular courts, this large block of flats occupies an obtuse corner site, and the entire plan is worked out in curves as well as all the elements of the exterior (Figure 35). The façade of the Casa Milá is not a thin plane, curling like paper at the edges and pierced with squarish holes like that of the Casa Batlló; instead ranges of balconies heavier than those on the rear of the Casa Batlló sway in and out like the waves of the sea beneath the foamlike crest of the roof, making the whole edifice a very complex plastic entity (Plate 137A). From a distance La Pedrera looks as if it were all freely modelled in clay; in fact, it is executed in cut stone with boldly hammered surfaces that appear to result from natural erosion.

There is no external polychromy of glass or tile here, and the frescoed colour used on the court walls has suffered such serious deterioration that it is difficult to know what it was like originally. On the other hand, Gaudí’s detail was never more carefully studied nor more consistent; there are no straight lines at all, and in the forms of the piers rising from the ground to support the balconies of the first storey he suggested natural formations with real success (Plate 135B). These elements look as if they had been produced by the action of sea and weather rather than by the chisel, quite as does much of the mid-twentieth-century sculpture of Henry Moore.

The marine note is seen at its strongest and most naturalistic in the ironwork however. Strewn over the balcony parapets and across various openings, like seaweed over the rocks and sand of the seashore, the railings and grilles are full of intense organic vitality with none of the graceful droopiness of Guimard’s Métro entrances. Gaudí’s metalwork frequently suggests the work of various mid-twentieth-century sculptors in welded metal, quite as his handling of masonry does later sculpture in stone. Indeed, his iron grilles often exceed such sculptors’ metalwork in richness and variety of form, as also in the fine hand-craftsmanship of the execution.

The detailing on the Casa Milá, whether of the masonry or the ironwork, avoids the nightmarish overscaling of the somewhat similar elements at the Parc Güell, and also the coarseness of the broken faience mosaic surfaces that he used so much there and elsewhere but here restricted to the roof-tops. As regards the masonry, moreover, it is really wrong to speak of detailing, for the very fabric of the structure, not just its edges and its trimmings as on the Casa Batlló, has been completely moulded to the architect’s plastic will. Whether or not it be correct to consider the Casa Milá an example of the Art Nouveau—and technically it is not—La Pedrera remains one of the greatest masterpieces of the curvilinear mode of 1900, rivalled in quality only by the finest of Sullivan’s skyscrapers (Plate 119), which it does not, of course, resemble visually at all.

Despite the esteem in which his work has always been held by his fellow-citizens of Barcelona, Gaudí had few local imitators of consequence. However, such detailing on early twentieth-century buildings there as may appear at first to be conventionally Art Nouveau is often in fact a bit Gaudian. Only his assistants Francisc Berenguer (1866-1914) and J. M. Jujol Gibert (1879-1949) seem to have understood Gaudí’s mature style. At least the house by Jujol at 335 Diagonal in Barcelona, though quite small and simple, and the Bodega Güell at Garraf of 1913 by Berenguer are of a quality worthy of comparison with Gaudí’s own best work.[393] The big Palau de la Musica Catalana, built by Luis Domenech Montaner (1850-1923) in 1908, is a very extravagant example of the architecture of the period, bold and coarse and rich, but with none of Gaudí’s personal flair and integrity.

In Glasgow Mackintosh after 1908 was a prophet with far less honour than ‘Greek’ Thomson had received there in an earlier day. But the countercurrent that he had helped to set going on the Continent was in full swing, particularly in Austria and in Germany (see Chapters 20 and 21). Even in Horta’s own Brussels, Josef Hoffmann had been called from Vienna as early as 1905 to build the suburban Stoclet mansion (Plate 154A) at 373 Avenue de Tervueren (see Chapter 21).

Despite the ephemeral nature of much of its production and the completeness with which it was ultimately rejected everywhere, the Art Nouveau has very great historical importance. The Art Nouveau offered the first international programme for a basic renewal of architecture that the nineteenth century actually set out to realize. Most earlier programmes, moreover, even if not primarily revivalistic, aimed chiefly at the reform of architecture; this was still true of Voysey and his English contemporaries in these very years, though not, of course, of Sullivan and Wright, working in isolation in the American Middle West. Thus the Art Nouveau was actually the first stage of modern architecture in Europe, if modern architecture be understood as implying, before anything else, the total rejection of historicism.

The proto-modernity of earlier stages of nineteenth-century architectural development is almost always ambiguous, since the leaders of the various successive movements rarely intended to break with the past entirely. The characteristic ideal of nineteenth-century architects, as of their late eighteenth-century predecessors, had been to react against what they considered the decadence of the building arts current in their day by returning to the principles of some earlier and supposedly purer or more vital age. The very considerable amount of innovation that many European architects before Horta introduced in their work was not exactly unconscious; but it was rather a matter of achieving personal expression by adapting old forms to new needs, new materials, and new methods of construction than of creating a wholly original modern style.

Well before the nineties a very few men had consciously sought absolute originality and total freedom from the disciplines of the past. But such architects found little or no public support for their programmes of architectural revolution nor even fellow-artists to share in their highly individualistic campaigns. After the relatively universal acceptance of the doctrines of Romantic Classicism there had followed chiefly a succession and a multiplication of divergences; now, in the nineties, a real pattern of convergence appeared. But this convergence was premature. The renewal of ornament and of the accessories of architecture outran the renewal of the more basic elements of the art of building towards which the technical developments of the nineteenth century had been so inevitably leading.

Thus the Art Nouveau stands apart both from the architecture of the preceding hundred years and from the modern architecture of the following sixty which extends down to the present. It did not bring the one to an end, as the profusion of so-called ‘traditional’ buildings of the early twentieth century makes very evident (see Chapter 24), nor did it provide much more than a preface to the major new developments that mark the early decades of the present century (see Chapters 18-21). That the Art Nouveau was completely rejected on principle by ‘traditionalists’ is not surprising: it was the first serious attack on the position they continued to maintain. But the very rapidity with which the Art Nouveau rose to popularity and descended to vulgarization encouraged its denigration in the name of ‘taste’ by almost all other architects soon after it reached its climax around 1900. In recompense, interest in the Art Nouveau began to revive early, by the early thirties, after a much shorter period of neglect than other phases of nineteenth-century architectural development have undergone and are still undergoing.

The place of the Art Nouveau in the story of modern architecture, if only as an episode of youthful wild-oat-sowing, is now well established. Most of its exponents actually lived long enough to receive in their later years embarrassing praise for youthful work they had quite disowned if not forgotten. It is a curious paradox that although most of the leaders of the Art Nouveau survived for decades—and Van de Velde died only in 1957—not one except Gaudí[394] maintained after 1910 the position of relative pre-eminence that had been his in 1900. A wholly new cast of characters, many of them no younger, came to the fore in the first decade of the twentieth century; they constitute the first generation of modern architects, properly speaking.

------------------------------------------------------------------------

18. MODERN ARCHITECTS OF THE FIRST GENERATION IN FRANCE: AUGUSTE PERRET AND TONY GARNIER

NO better name than ‘modern’ has yet been found for what has come to be the characteristic architecture of the twentieth century throughout the western world, well beyond its confines also in Japan, India, and Africa, and increasingly in most of the Communist countries. Alternative adjectives such as ‘rational’, ‘functional’, ‘international’, or ‘organic’ all have the disadvantage of being either vaguer or more tendentious. Whether the Art Nouveau or such things as Sullivan’s skyscrapers and Voysey’s houses all truly belong, in their rather sharply differing ways, to a first stage of modern architecture or are transitional and prefatory may still be debated; but from the earliest years of this century several continuous lines of development can certainly be traced. These lines were in the main convergent through the twenties, if increasingly divergent in the middle decades of the century. By stressing generic changes rather than specific achievements the development can be presented almost anonymously, somewhat as the nineteenth-century development of commercial architecture was outlined earlier in this book (see Chapter 14). But it is more humanistic, and at least as true to the detailed facts, to consider modern architecture as deriving from the individual activities of a few leaders rather than from some Hegelian historic necessity. Of those leaders one group, born in the late 1860s, constitutes the first generation; a group born some twenty years later forms a second generation; since the 1930s still another generation has come to the fore.

A somewhat similar succession of three generations could be distinguished in the case of Romantic Classicism, the last universal style in architecture. What sets the twentieth-century situation apart from that of the earlier period has been the marked prolongation of the activity of the first generation, two of whose leading members, Wright and Perret, lived on and remained active well beyond 1950. Wright continued in vigorous production down to his death in 1959. The leaders of the second generation, who first moved towards the centre of the stage in the early twenties, are mostly still alive; two of them at least, Le Corbusier and Mies van der Rohe, have been rather more productive since 1946 than they were earlier in their careers (see Chapter 21).

While some influence from their juniors can be noted in the later work of the modern architects of the first generation, a real difference between their approach to architecture and that of the second generation has continued. Those who have come forward since the mid thirties owe much to the first generation as well as to the second, yet they have also manifested some significant characteristics that are their own. The modern architecture of the last sixty years may well be presented historically in terms of the work of two generations of leaders (see Chapters 18-23), and then of the production of the decade following the Second World War (see Chapter 25). But modern architecture, even very broadly interpreted, includes only a small fraction of all building production down to the war; the work of those supporters of the ‘tradition’ in the twentieth century bulked much larger in quantity, even if it very rarely rivalled the modern work in interest or quality (see Chapter 24). An Epilogue will touch on the current scene in the early sixties.

The leaders of the first generation of modern architects remained great individualists to the last. It is therefore not easy to draw any general stylistic picture from their production, even for the years before the twenties when they were the only modern architects. The leaders of the second generation drew their inspiration, in most cases, not from one but from several of the older men; yet their work was so convergent that by the mid twenties a body of doctrine had come to exist deriving partly from their theories and partly from their few executed buildings and their many projects. With the increasingly wide acceptance of this body of doctrine critics were soon ready to recognize the existence of a new style as coherent, as consistent, and almost as universally employed by younger architects everywhere as the Romantic Classical style had been at the opening of the nineteenth century (see Chapter 22).

Towards the constitution of this new style each of the great architects of the first generation had made notable contributions; yet their executed work, and even more their theories, remained independent of it. To appreciate that work only in the light of what they had in common with their juniors is to miss much of the richness and all of the idiosyncrasy of their achievement. In considering the work of these older architects for its own sake, what sets it apart from the Art Nouveau, whose protagonists were in many cases their exact contemporaries, must first be indicated and evaluated. For example, their rejection of ornament, at most but relative, provides only a minor and negative point of differentiation. In their positive preoccupation with structure and its direct architectonic expression, and also their reform and revitalization of planning concepts, however, they went much further than most of the Art Nouveau designers of 1900. It is true that such architects as Horta and Jourdain, when working with metal and glass, were concerned with the expression of structure, but that expression was usually more decorative than architectonic (Plates 132B and 133). Traditional materials, such as stone and brick, in the hands of Art Nouveau architects and their spiritual brothers often lost all their natural character, being treated like so much clay. The sense of materials, both new and old, and the determination of their proper use preoccupied all the leading architects of the first generation, something for which only the English and the Americans prepared the way in the nineteenth century.

The new importance of structure and its expression, the preoccupation with a particular building material, is nowhere more evident than in the work of Auguste Perret (1874-1954), the only great French architect of this generation. Associated as he was with the family contracting firm of A. & G. Perret, which specialized early in the use of reinforced concrete, he saw as his principal task the development of formulas of design for concrete as valid as those so long established in France for building with stone. The other architects of his generation came more gradually and less whole-heartedly to the exploitation of new materials—it is paradoxical, for example, that the characteristic Art Nouveau interest in exposed metal construction came generally to an end about 1905—and their work as a result is more various and less doctrinaire. Because of Perret’s clear definition of his goal and his single-minded advance along a predetermined line, his somewhat limited architectural achievement may well be considered before the protean many-sidedness of Wright’s in America and the ambiguity of Peter Behrens’s in Germany, not to speak of the important contributions of Wagner and Loos in Austria, and of Berlage and de Klerk in Holland (see Chapters 19, 20, and 21).

Auguste Perret came of Burgundian stock, but by the accident of his father’s exile from France after the Commune he was born in Brussels. His education was entirely French. He left the École des Beaux-Arts to enter the family’s building firm without waiting to receive the Government’s diploma, somewhat as Wright went out into the practical world with but two years of engineering school behind him. His career began almost at once, for he built his first house at Berneval in 1890. Several blocks of flats and an office building in Paris followed in the next eight years; the Municipal Casino at St-Malo, built in 1899, was the first work of any real consequence. There he and his brother Gustave (1876-?) used reinforced concrete for an unsupported slab floor of 54-foot span. Executed otherwise in local granite and wood, this building has a certain bold simplicity as remote from ‘Beaux-Arts’ as from Art Nouveau work of the period.

Reinforced concrete,[395] that is concrete strengthened by internal reinforcing rods of metal, seems to have been invented by a French gardener named Joseph Monnier in 1849, but he used it only for flower pots and outdoor furniture. In 1847 François Coignet (1814-88) built some houses of poured concrete without reinforcement; in 1852 for a house at 72 Rue Charles Michel in St-Denis, Seine, Coignet first employed his own system of béton armé, to use his term. That term has since remained current in French—the German term is Eisenbeton, the Italian cimento armato. During the next four decades ferro-concrete, to give it its simplest English name, was developed very gradually by Coignet and by François Hennebique (1842-1921) with no very notable architectural results. Detailed research is gradually revealing many instances of its early use by various men in different countries; but neither in the scale of its employment nor in the achievement of new and characteristic modes of expression does its history in these decades rival that of iron in the first half of the nineteenth century (see Chapter 7).

In 1894, just as the Art Nouveau was reaching France, ferro-concrete was used for the first time in a structure of some modest architectural pretension by J.-E.-A. de Baudot[396] (1836-1915) for a school in the Rue de Sévigné in Paris. This is overshadowed in interest, however, by the church he began to build in 1897. Saint-Jean-de-Montmartre at 2 Place des Abbesses in Paris has very little connexion with the Art Nouveau except for its drastic novelty. On the contrary, de Baudot employed for his structural skeleton very much simplified Gothic forms. Actually, it is incorrect to call the material used here béton armé; it is more properly ciment armé since there is no coarse aggregate as in concrete. Like his master Viollet-le-Duc’s projects, Saint-Jean is curious rather than impressive and not at all to be compared in intrinsic interest with Gaudí’s Sagrada Familia. Worth noting, however, is the use of faience mosaic to decorate the concrete structural members, something de Baudot had already tried out on his earlier school. The authorities were dubious of the strength of de Baudot’s structure, as well they might have been considering the iron-like delicacy of the membering, and a hiatus of several years held up the construction after 1899, the church being completed only in 1902-4. As has been mentioned already, the contractor was Contamin working with Soubaux, his partner of the period.

Before Saint-Jean-de-Montmartre was finally finished in 1904, Perret had already demonstrated the architectural possibilities of the new material rather more effectively in the block of flats that he built in 1902-3 at 25 bis Rue Franklin in Paris. Despite the echo of the Art Nouveau already noted in the foliage patterns of faience mosaic filling the wall-panels on the exterior, most of the interest of the building resides in its structure and its planning. Like that of Anatole de Baudot’s church, the structure is visibly a discrete framework, but made up entirely of vertical and horizontal elements with no curved members of either Gothic or Art Nouveau inspiration. However, the concrete is nowhere exposed but always covered with glazed tile sheathing. Within the wall-panels the windows are crisply outlined by plain projecting bands of tile; this provides an early instance of that encadrement, or framing, on which Perret came to insist in all his work after the mid twenties.

The skeletal structure of 25 bis Rue Franklin allowed great freedom in planning (Figure 36). Around a small court, sunk into the front of the building, the principal living areas of each flat all open into one another, somewhat as in Wright’s Hickox house of 1900 but with less spatial unification (Figure 31); the result is closer to Horta’s treatment of the main floor of his Aubecq house of 1900 in Brussels (Figure 34).

The next year Perret built another block of flats at 83 Avenue Niel in Paris with an internal skeleton not of concrete but of metal, and façades of stone treated somewhat like those of his Art Nouveau flats of the previous year in the Avenue Wagram (see Chapter 17). He returned, however, at once to the use of ferro-concrete and rarely deserted it again.

The Garage Ponthieu, which was built in 1905-6 in the Rue de Ponthieu in Paris, is a much more striking example of the possibilities of the new material than the earlier blocks of flats; moreover, the concrete is here exposed (Plate 139A). Inside, galleries carried along both sides of the L-shaped space provide a second level for parking motor cars and the whole interior is almost as light and open as if it were built of metal, thus recalling a little de Baudot’s church. The façade, likewise, is as skeletal as if executed with a metal frame. But Perret’s determination, somewhat comparable to Sullivan’s in the Wainwright Building in St Louis of fifteen years before, to organize the expression of a new type of construction along basically Classical lines is as evident as the maximal fenestration. The thin slab which projects at the top provides a sort of cornice and the range of small windows underneath it a sort of frieze, while the arrangement of the elements of the façade below is very formal indeed. The rose-window-like glazing of the big central panel is somewhat rudimentary and rather less Classical in feeling than the rest, but the essentials of Perret’s concrete aesthetic are all adumbrated here as they were not in the more tentative block of flats in the Rue Franklin.

In the solid, marble-sheathed façade of the Théâtre des Champs Élysées in the Avenue Montaigne in Paris, Perret’s largest and most conspicuous early work, his classicizing intentions are even more evident, but the expression of concrete-skeleton structure is much less complete; these intentions are underlined, moreover, by the large stylized reliefs by Antoine Bourdelle that provide the only external decoration. Originally, in late 1910, the commission for this theatre was given to Van de Velde. He at once proposed that it should be built of ferro-concrete with the Perret firm as contractors. During the course of the following year Perret proposed various changes in the plan to make more practical its construction with a concrete skeleton. When he later offered an alternative design for the façade this was preferred by Van de Velde because it seemed then so expressive of the underlying structure, as it hardly does to posterity. By September Van de Velde made a final report as consulting architect and withdrew completely. Needless to say, there has been controversy ever since as to the degree of Perret’s responsibility for this major monument of twentieth-century Paris; as built, however, there can be little question that it is very largely of his design. How different a theatre by Van de Velde would have been is at least suggested by the one that he erected in 1914 for the Werkbund Exhibition in Cologne (see Chapter 20).

The foyer of the Théâtre des Champs Élysées expresses the possibilities of ferro-concrete in a more architectural way than do the interiors of the earlier block of flats and the garage. The actual structural members of the skeleton are visible in the free-standing columns, as are also the beams that they support; the walls are very evidently only thin panels between the piers. A few simple mouldings are used to assimilate the new expression to the conventions of academic design—too few to satisfy contemporaries, though too many for later taste.

There is less clarity of expression in the great auditorium because of the profusion of murals contributed by various Symbolists and Neo-Impressionists—Maurice Denis and K.-X. Roussel most notably—and by the over-all gilding of the principal structural members, which are also elaborated by semi-Classical detailing. Even so, the fact that the dome is carried on the four pairs of tall slender columns is very evident, and the swinging curves of the successive balconies give early evidence of the ease with which ferro-concrete lends itself to bold cantilevering.

The presumed necessity of achieving monumentality undoubtedly compromised the purity of Perret’s expression of structure throughout the Théâtre des Champs Élysées. During the War, which followed so soon after the inauguration of the theatre in 1913, an important industrial commission of Perret’s produced what would be for the next generation of architects a more exemplary work. The warehouses built at Casablanca in North Africa in 1915-16—there are also others there of 1919—required no representational display; they are almost ‘pure’ engineering in concrete. But the lightness of their walls, pierced with abstract patterns formed by ventilating holes, and the elegance of their thin shell vaults of segmental section displayed the potentialities of a quite new structural aesthetic, at once delicate and precise, with no echoes at all of the massive masonry buildings of the past.

The interior of the Esders Clothing Factory at 78 Avenue Philippe-Auguste in Paris, erected just after the War in 1919, and several smaller industrial buildings for the metal-working firm of Wallut & Grange at Montataire, Oise, of 1919-21 were more readily studied by younger architects and, in the case of the Esders factory, much grander in scale than the North African warehouses. Even more elegant than the warehouses, and equally ‘pure’, was the atelier of the decorator Durand built in Paris in the Rue Olivier-Métra in 1922. This has a shell vault rising from the floor broken, along one side only, by a long skylight over widely spaced ribs that continue the curve of the vault.

By this time, of course, ferro-concrete was in general use for industrial building throughout most of the western world. In France the vast parabolic-vaulted aircraft hangar at Orly, Seine, designed by the engineer Eugène Freyssinet (1879-1962) in 1916, overshadowed in size and boldness anything built by Perret. This very exceptional utilitarian construction, magnificent in form yet quite without architectural pretension, was destroyed during the Second World War. To Tony Garnier’s work in Lyons we shall turn later.

In America Frank Lloyd Wright used ferro-concrete for his modest E.Z. Polish Factory in Chicago in 1905, just as Ernest L. Ransome was completing the first mature example of a large plant of ferro-concrete frame construction, the United Shoe Machinery Plant in Beverly, Mass., begun in 1903.[397] All over the Middle West, moreover, grain elevators[398] were rising in the form of gigantic linked cylinders. In Switzerland the great engineer Robert Maillart (1872-1940) in his factories and bridges was using concrete in several new ways as different from the elevators as from the usual timber-like frames of the French and the Americans or the shell vaults of Perret and Freyssinet. Everywhere the importance of ferro-concrete as the prime building material of the twentieth century was receiving increasing recognition; for it was a material more universally available than structural steel and also so elastic in its potentialities that these have hardly even yet been adequately explored.[399] In the early twenties, when a younger generation of architects all over Europe turned their major attention to ferro-concrete as the most modern of building materials, Perret was the architect who had the most to offer them—how limited had been Wright’s exploitation of concrete up to this time we shall shortly see (see Chapter 19). When Perret erected the church of Notre-Dame at Le Raincy, S.-et-O., near Paris in 1922-3 concrete came of age as a building material in somewhat the same way that cast iron had done in a series of major English and French edifices of the 1840s (see Chapter 7).

The Le Raincy church is not revolutionary in plan, being a basilica with aisles and an apse; unlike de Baudot’s church, however, it has no specific elements of Gothic reminiscence in the interior (Plate 141). Instead it provides what the medieval builders of Saint-Urbain at Troyes or King’s College Chapel in Cambridge had obviously sought to achieve, a complete cage of glass supported by a minimal skeleton of solid elements. The broad segmental shell vault of the nave, with smaller vaults running crosswise over the aisle bays in the Cistercian way, is carried on no walls at all but only on the slightest of free-standing columns reeded vertically by the forms in which they were cast (Figure 37). Quite separate from this supporting skeleton is the continuous enclosing screen of pre-cast concrete units, pierced and filled with coloured glass designed by Maurice Denis. This is carried round the entire rectangle of interior space and bowed out at the east end in a segmental curve to form a shallow apse behind the altar. Only at the front is the clarity of the conception compromised by the awkward impingement of the clusters of columns that shoot up to form the tower.

Deserting the dilute Classicism that was his natural bent, Perret allowed the clustered piers of his tower to rise into the sky, supporting nothing at the top, in order to approximate the outline of a Gothic spire. Even more than in the interior, where one is aware only of the lowest stage, the verticalism and the medieval suggestion of this feature, so over-ingeniously composed of standard ferro-concrete elements, seems quite at odds with the severe concrete-and-glass box that provides the body of the church. Few other ferro-concrete churches[400] of the twenties, least of all Perret’s own Sainte-Thérèse at Montmagny, S.-et-O., of 1925-6 and other French ones by his imitators, rival Notre-Dame at Le Raincy. The largest and boldest, Sankt Antonius at Basel in Switzerland, built by Karl Moser (1860-1936) in 1926-7, seems somewhat heavy and factory-like. Its plain rectangular tower, however, rising free at one corner of the church, is much simpler and more original than Perret’s spire and has been frequently and successfully emulated by other architects. Of quite a different order are the Expressionist churches of the German Dominikus Böhm, which have, in the long run, had at least as wide an influence (see Chapters 20 and 25).

Two remodelled Paris banks, one of 1922 for the Société Marseillaise de Crédit in the Rue Auber and another of 1925 for the Crédit National Hôtelier, gave evidence of Perret’s capacity to extend the implications of ferro-concrete design to more conventional problems. These interiors are almost wholly devoid of ornament, and they largely depend for their effectiveness, like the foyer of the Théâtre des Champs Élysées, upon the careful proportioning of the exposed elements of the skeleton construction. In 1924 the Palais de Bois, a temporary exhibition building at the Porte Maillot in Paris, showed how this sense of direct structural expression could be exploited in a building all of timber. This was much more successful than the theatre that Perret built in 1924-5 for the Paris Exposition des Arts Décoratifs. Of a quite different order was the Tour d’Orientation at Grenoble, also of 1924-5. Here Perret was far happier in achieving something comparable to the richness of medieval spires with standard structural elements and pre-cast panels than in the tower of his church at Le Raincy, for this is much more structurally conceived and quite devoid of Gothic reminiscence in the outline.

The mid twenties also brought to Perret, by this time widely recognized in advanced circles as the leading French architect, several commissions for houses, chiefly for artists, in France and even as far afield as Egypt. Characteristically French in his preoccupation with large, not to say monumental, problems, house-design was not Perret’s forte in the way it was that of his American and Austrian contemporaries Wright and Loos. Moreover by this date certain younger architects, particularly Le Corbusier and two or three others in Paris, had set under way a revolution in domestic architecture as drastic as Wright’s of twenty-five years earlier (see Chapter 22).

Perret’s best houses, such as the Mouron house at Versailles of 1926 or the Nubar house in the Rue du 19 Janvier at Garches of 1930, have an almost eighteenth-century dignity and serenity. The ‘stripped-Classical’ apparatus of terminal cornices, encadrements around the openings, and occasional free-standing columns is doubtless logical as an expression of the construction, but it is also very conservative in effect. Yet the ferro-concrete construction encouraged Perret to introduce very wide openings leading out on to surrounding terraces and to open up the main living areas even more than he had done in the flats of 1902-3 in the Rue Franklin. Such treatments were still rather advanced for Europe, however common they may have been in America for a quarter of a century and more. The characteristic quality of Perret’s domestic work is seen at its best in a small block of flats at 9 Place de la Porte de Passy in Paris facing the Bois de Boulogne that he built in 1930 (Plate 139B). This has a façade towards the park so superbly proportioned that it might almost be by Schinkel and a flow of space inside the individual flats that is worthy of Wright, although much more formal in organization.

Now Perret began to receive the official commissions that are generally given in France only to men well on in years. The building designed in 1929 that he erected for the technical services of the Ministry of Marine in the Boulevard Victor in Paris is one of the largest and most typical of his later works (Plate 140B). The complex rhythms and subtle three-dimensional play of this façade are entirely produced by the actual structural elements. The skeleton divides the long façades into a series of horizontal panels within which are set the vertical frames of the windows separated by pre-cast slabs; in one storey the windows even extend the full width of the bays.

To a considerable extent Perret had succeeded in achieving what he had long consciously sought, that is, a vocabulary of design in concrete as direct, as expressive, and as ordered as the masonry vocabulary of the seventeenth and eighteenth centuries—a style Louis XX, so to say—still very French in a quite traditional way, yet unmistakably of this century. In the Garde Meuble or National Furniture Storehouse in the Rue Croulebarbe in Paris, begun the next year, the vocabulary is—from principle—all but identical; yet fewer windows and more solid panels were necessary here so that the general effect is flatter and blanker. The curved colonnade across the open side of the court is almost archaeologically reminiscent of the eighteenth century, despite the breadth of its spans and the ingenuity of its detailing. The small concert hall of 1929 in the Rue Cardinet for the École Normale de Musique is less pretentious but also less impressive.

Concrete to Perret, after all these years of employing it, was not a crude or a substitute material. By the use of coloured aggregates which he found various means of exposing he was able to vary the texture and colour of his poured and pre-cast elements with considerable subtlety and elegance. In the later buildings the workmanship is usually of the highest quality—it was by no means so in the early twenties—with arrises brought to a sharp edge in pure cement and such classicizing details as the flute-like facets on piers and the capital-like treatment of their tops carried to a finish comparable to that of chisel-cut freestone.

Thus Perret was eventually able to avoid the industrial brutality of much work in concrete where the material is left as it comes from rough timber forms with crumbling arrises and pockmarked surfaces. Such lack of finish is acceptable in large-scale engineering work but certainly awkward when seen close to as in Notre-Dame at Le Raincy. On the other hand, Perret kept well away also from that slickness of surface—especially popular with younger architects in the twenties—that is produced when concrete is covered with a smooth stucco rendering and painted.[401] Such slickness is, of course, generally very soon lost as the original surface grows cracked and stained; only too rarely is it properly maintained by frequent patching and repainting. Concrete was to Perret a worthy material, like stone, and therefore deserved the effort and the cost required to give it an expressive finish requiring little or no maintenance.

The reticulated wall system of the big government buildings was also used for a block of flats at 51-55 Rue Raynouard, built in 1932, where Perret himself lived and also maintained his atelier. The necessary adaptation of his formalized open planning to a trapezoidal site produced suites of interior space of considerable complexity yet perfect orderliness. Though Perret was still without a governmental diploma, the atelier[402] he ran here was associated with the École des Beaux-Arts. It almost seemed now as if he wished to demonstrate how much truer a representative he was of real French tradition than those who were its official, though unworthy, custodians. Thus the older he grew the farther his work drew away from that of the more revolutionary modern architects of the second generation. By 1930 it had definitely begun to date; yet it was only in the last twenty-five years of his life that there came to him the greatest opportunities of realizing his ambitions for French twentieth-century architecture.

In comparison with Perret’s own pioneering of 1902-22 his late work seems to lack vitality. For all the thought that went into its finish, for all the virtuosity of certain features—such as the self-supporting curve of the broad stair that spirals down into his atelier in the Rue Raynouard—his very ambition to create a new French tradition gave his later buildings something of the banality of those designed by the more conventionally ‘traditional’ architects of his generation. This applies in particular to his principal work of the thirties in Paris, the still unfinished Musée des Travaux-Publics in the Avenue du Président-Wilson which he began in 1937. Here the ingeniously pseudo-Classical—yet also truly structural—apparatus of external engaged columns and the intricate plan spreading out from a circular auditorium at the apex of the site are quite in the Beaux-Arts manner. But the grandeur of scale in the interiors and the exciting upward sweep of the boldly curving stairs lend value, and even novelty, to a scheme that is in many ways extremely conservative.

After the Second World War Perret was asked to provide plans for the rebuilding of several bombed cities: Le Havre in 1945; Amiens in 1947; and the Vieux-Port district of Marseilles in 1951. For Amiens he designed a skyscraper, long physically complete but still unoccupied, that derives more from his decorative Tour d’Orientation at Grenoble than from the skyscrapers of the New World. This is one of his few complete failures, if for no other reason than the competition its tall and awkward silhouette offers to the cathedral, whose towers had so long dominated the city’s skyline. The executed Marseilles buildings are not of his design any more than are most of those at Amiens.

At Le Havre, however, his control of the rebuilding was more complete. The Place de l’Hôtel de Ville, or at least the three sides completed between 1948 and 1950 by his associates, outweighs by a great deal the failure of the Amiens skyscraper (Plate 140A). Ranges of four-storey buildings, all carried out in the reticulated vocabulary of his Government buildings of the early thirties in Paris, surround a large sunken plaza; the Hôtel de Ville in the near-Beaux-Arts manner of his Musée des Travaux Publics occupies the fourth side. Shops open towards the square under a continuous colonnade. Behind, rising out of small courts, are taller towers occupied by flats; these lend great three-dimensional interest to the formal and absolutely symmetrical layout of this section of the rebuilt quarter. Since his death similar ranges of buildings have been carried out along the quais to the south. On the whole the extensive work of the team[403] is superior to the public monuments by their captain, the Hôtel de Ville and the church of St Joseph, both designed in 1950 and completed before Perret’s death in 1954.

Impressive as is Perret’s Le Havre in the international roster of post-war urban rebuilding, it seems curiously out of date today, a mere realization in the 1940s and 1950s, one might almost say, of the aspirations of the early decades of the century. Since that period had few such opportunities as was Perret’s here to realize urbanism on this scale, however, what he accomplished there is a welcome addition to the city-building achievements of this century.

Until the second generation appeared on the scene in the twenties France produced little modern architecture of much interest besides Perret’s work. The department stores of the early years of the century, still strongly under the influence of the Art Nouveau, have already been mentioned (see Chapter 17). After Perret the most important architect was Tony Garnier (1867-1948), and he is of more significance for a vast project that he prepared in his youth than for the executed work of his maturity. In the later decades of the eighteenth century, when the Romantic Classical revolution in architecture was getting under way, projects were often of more interest than executed buildings for their premonitions of what was to come, and this was particularly true in France. It was true again in the early decades of the twentieth century, down at least to Le Corbusier’s project for the Palace of the League of Nations of 1927-8.

Ledoux’s ‘Ville Idéale’ summarized his own aspirations and also provided a wealth of ideas from which later generations of Romantic Classical architects could draw inspiration. So, at the opening of the twentieth century, Garnier’s very complete scheme for a ‘Cité Industrielle’[404] contained a wealth of ideas on which architects drew well into the 1920s. Like that of the ‘Ville Idéale’, the interest of the ‘Cité Industrielle’ is threefold: sociological, urbanistic, and architectural. Henceforth the industrial city would be more and more accepted as normal and not exceptional. Its needs both general and specific—so notably recognized by Garnier, all the way from the provision of adequate workers’ housing to various sorts of industrial plants—would become more and more important preoccupations of most modern architects. In coping generally with the manifold needs of an industrial community Garnier also faced in detail many very different individual architectural problems with considerable ingenuity.

Garnier’s solutions in the main were very simple and direct, but they often had a merely negative character, as of buildings of academic design scraped of all surface paraphernalia, rather than displaying any fresh and creative approach. But an important part of the main architectural development for some twenty years was to be such a purging of inherited excess. Garnier reduced architecture to basic, if not particularly unfamiliar, terms; on his foundations the next generation began, in the twenties, to build something much more positive; thus his influence was parallel to that of Loos (see Chapters 20 and 21). His contribution to the twentieth century’s repertory of forms was less than Ledoux’s had been to that of the nineteenth a hundred years earlier; notably inferior in quality to Ledoux’s was his own actual production, moreover.

Garnier’s appointment as Architect of the City of Lyons in 1905, a position which he retained until 1919, might seem to have provided the perfect opportunity to realize his dreams as, but for the Revolution, should Ledoux’s appointment by Louis XV to build the Royal Saltworks at Arc-et-Senans. But neither the Municipal Slaughterhouse of Lyons at La Mouche, executed in 1909-13, the Herriot Hospital at Grange-Blanche, designed in 1911 and begun in 1915, nor the Olympic Stadium of 1913-16 at Lyons realize much more than the obvious practical implications of the detailed projects for various buildings in his ‘Cité Industrielle’.[405] The slaughterhouse is bold structurally but clumsily industrial in its handling, with none of the refinement of Perret’s factories; the more highly finished stadium has irrelevant Classical touches in the detailing, simple though it is, of the concrete elements.

Garnier’s work after the First World War began with the hospital, which was completed only in 1930, and included a large low-cost housing project in the États-Unis quarter of Lyons designed as early as 1920 but executed only in 1928-30. Both are quite overshadowed by the comparable work of the next generation in these years—that in other countries at least, if not that in France. The Moncey Telephone Office at Lyons of 1927, the Textile School at La Croix-Rousse of 1930, and the Hôtel de Ville of the Paris suburb of Boulogne-Billancourt of 1931-4, on which another architect, J.-H.-E. Debat-Ponsan (b. 1882), a pupil of Victor Laloux, collaborated, differ very little from the scraped academicism of most French public architecture of this period. The houses Garnier built in 1909 at St-Rambert and in 1910 at St-Cyr (Mont d’Or) are among his best executed works; all the same, except for their early date, they are hardly very notable.

Two blocks of flats built by Henri Sauvage (1873-1932) in 1925 in the Rue des Amiraux and in the Rue Vavin in Paris, faced with glazed white brick and stepped back in section to provide terraces for the upper floors, are well above the level of quality of Garnier’s later work without approaching that of Perret’s. That in the Rue des Amiraux, being for working-class occupancy, is more significant of the international aspirations of the period. Although less drastically novel than the low-cost housing of the twenties in Holland and Germany, this has survived very well because of its permanent grime-proof surfacing. It has been rather unjustly forgotten, largely because it lies off the main line of international development (see Chapter 21).

Most French production in the twenties remained completely subject to academic discipline although it was often tricked out with the sort of modish decoration that flourished particularly at the Paris Exposition des Arts Décoratifs of 1925. Yet at the same time Paris, as the world capital of modern art, was one of the three great foci of architectural advance. The linkage between advanced painting and the Art Nouveau in the nineties was discussed earlier (see Chapter 16). Perret employed Symbolist and Neo-Impressionist painters as collaborators, beginning with the Théâtre des Champs Élysées before the First World War. But there is no real parallel between his architecture and that of Garnier or Sauvage on the one hand and the art of the great twentieth-century masters of the École de Paris on the other. Picasso, Gris, Braque, Matisse, and Derain had no effective influence on architecture. Characteristically Perret employed Bourdelle, not Maillol, when he needed sculpture. With the next generation the situation entirely changed; but the new architects of the twenties, not only in France but everywhere, for all their greater sophistication and their close association with advanced painters and sculptors, still owed at least as much to Perret and to Garnier if not to Sauvage.

To the most creative new architects who appeared around 1920 Garnier’s project for the ‘Cité Industrielle’ offered both a challenge and an inspiration, but Perret was by far the more important influence. Somewhat later, towards 1930, that influence became almost ubiquitous in France, and its effect grew increasingly banal as the ferro-concrete Classicism of Perret’s later work gradually replaced the official and inherited tradition of the École des Beaux-Arts, by that time nearly obsolete even in France.[406] As has so often happened in France before, a youthful rebel, after being accepted late in life by the academic authorities, was only too ready to support a new discipline that had itself already become academic. Thus is cultural continuity maintained in France at the expense of variety and recurrent new growth. The situation was rather different in America, as we shall soon see.

------------------------------------------------------------------------

19. FRANK LLOYD WRIGHT AND HIS CALIFORNIA CONTEMPORARIES

WRIGHT in America found himself, in his seventies, as generally accepted a master as did Perret in France, but his influence never became at all academic in the way of Perret’s after 1930. There could hardly be a greater contrast between the careers of two contemporaries in the same field. Both were very productive over a length of time that is more than a third of the whole period covered by this book, but this is about all that they did have in common. Perret’s career progressed gradually over several decades to general and even official acceptance. Wright’s career, on the other hand, had very notable ups and downs, and he only once received a governmental commission.

After the years of preparation discussed earlier (see Chapter 15) there followed some ten years of great success. But this success was largely restricted to a particular region, the Middle West, and to a particular field, the building of good-sized suburban houses. Following that, in a decade interrupted by the First World War, Wright’s influence rapidly increased, not at home but abroad, although he had considerably fewer, if much larger, commissions. Then, paradoxically, in the twenties, while the United States swung into the biggest building boom in history, there began a decade in which Wright’s production all but ceased. Many assumed that his career had closed and that his work had passed into history as had Voysey’s and Mackintosh’s by that time. This, of course, was not at all true. In the mid thirties Wright’s activity revived, and his production continued at a rising rate until his death. Moreover, there was little sign of any decline into personal academicism such as marked the late work of Perret in the same decades.

Where Perret had, in effect, only a double architectural career, being largely occupied on the one hand with industrial commissions close to the dividing line between architecture and engineering, and on the other hand with public buildings, Wright’s career was increasingly multifarious. Beginning chiefly as a domestic architect, he never ceased to build houses; but by the 1950s there were few fields, including that of urbanism, which he had not entered, if only to present challenging projects and announce controversial theses. Disciple of a great skyscraper architect, author of a succession of skyscraper projects, Wright had to wait a full half century after Sullivan completed his last skyscraper in Chicago before he built his first, the Price Tower in Bartlesville, Oklahoma, in 1953-5. Some of his planning projects may yet come to posthumous execution, and his work at Florida Southern College at least was of urbanistic scope.

Perret consciously summarized and continued earlier French tradition; but Wright wished to initiate a new tradition, one which he preferred to call ‘Usonian’ rather than American. Perret’s disciples, emulators, and imitators in his later years were able to take control of French architecture to a quite considerable extent. Wright’s disciples, despite the fifty years during which he maintained offices that were also training ateliers in Oak Park, in Chicago, in Tokyo, in Wisconsin, and in Arizona, have only rarely made any significant mark of their own; nor has his influence had much more specific effect on the character of modern architecture in America than it has had generically on that of the world outside. Where Perret’s influence, particularly outside France, has been largely restricted to architects working with ferro-concrete, the material that he was the first to master architecturally—and even in concrete construction this influence has inhibited as often as it has liberated—Wright’s influence has been protean on the international scene. From the day when the German publisher Wasmuth first made Wright’s work available to Europeans at the opening of the second decade of the century this has been true, down to the time, a decade ago, when the Italian architect, critic, and historian Bruno Zevi (b. 1918) tried to invert chronology so that Wright’s ‘architettura organica’[407] might seem to succeed rather than to precede the ‘funzionalismo’ or ‘International Style’ of the second generation of modern architects.

Before turning to a more detailed consideration of Wright’s work after 1900 one further comparison with the œuvre of Perret may be made. Although Wright never confined himself to one material or to one method of construction—indeed, his versatility in this respect continued to increase right down to his death—he was from the first especially interested in the possibilities of concrete. He published in The Brickbuilder for August 1901 a project for a small village bank, still very Sullivanian in its rich detailing, that was intended to be executed entirely in concrete. This was only two years after Perret had first used the material with little or no attempt to develop its architectural possibilities and a year before his block of flats in the Rue Franklin was designed. His E.-Z. Polish Factory of 1905 at 3005-17 West Carroll Avenue in Chicago has already been mentioned. The Unity Church in Oak Park of 1906 (Plate 143B), entirely of concrete surfaced with a special pebble aggregate and decorated with integral ornament, precedes by many years Perret’s church at Le Raincy (Plate 141). Perret’s ultimate development of various refined finishes for exposed concrete came still later. Admittedly, however, the Oak Park church is a much smaller and less striking edifice than Perret’s; and the work of Kahn and other industrial architects soon overshadowed Wright’s modest factory. Moreover, it was only with the twenties that Wright, like the Europeans, really gave major attention to building in concrete.

Wright’s creative powers in the first decade of this century were largely concentrated on his ‘Prairie Houses’. Their essentials were already present in the two Kankakee houses of 1900 (Plate 142A) and the first house designed for the Ladies Home Journal (see Chapter 15). But these essentials received more masterly—one might well say more classic—expression two years later. The large W. W. Willitts house at 715 South Sheridan Road in Highland Park, Ill., of 1902 is of wooden-stud construction, but covered like the Kankakee houses with stucco (Plate 142B). The C. S. Ross house off the South Shore Road on Lake Delavan in Wisconsin, also of 1902, has the rough board-and-batten sheathing of the River Forest Golf Club (Plates 143A and 128B). Both offer versions of the cruciform plan (Figure 38) with the interior space ‘flowing’ round a central chimney core and also extended outward on to covered verandas and open terraces quite as in Price’s Tuxedo Park houses of fifteen years earlier (Figure 28).

Another major work of 1902 is the Arthur Heurtley house at 318 Forest Avenue in Oak Park, Ill. There the principal living areas, which are on the upper floor as in the Husser house of 1899, form an articulated L-shaped within the basic square that is defined by the overhanging roof. The brick walls of the lower storey have broad projecting horizontal bands and the wide, low entrance arch remains quite Richardsonian. The upper storey consists largely of continuous ranges of wooden-mullioned casement windows.

No notable progression is observable in the series of suburban houses built during the remainder of this decade before Wright went to Europe in 1909; but he produced many other brilliant illustrations of both the cruciform and the square plan as well as a more elongated sort extending along a single axis. Of the many fine examples of the Willitts or Ross type around Chicago, the small house for Isabel Roberts at 603 Edgewood Place in River Forest of 1908 is one of the best; there the living room in the front wing is carried up two storeys, as was proposed for one version of the Ladies Home Journal house. The larger F. J. Baker house at 507 Lake Avenue in Wilmette of 1909 also has a two-storeyed living room; but here the tall cross element of the plan which this feature provides was moved to one end of the house so that the plan is of a T or L shape rather than cruciform.

The E. H. Cheney house at 520 North East Avenue in Oak Park of 1904 is square like the Heurtley house near by. It is raised off the ground on a sort of extended square stylobate so that the living area, which runs all across the front as at the Hickox house, can open freely through french doors on to the walled terrace in front. In the T. P. Hardy house at 1319 South Main Street in Racine, Wis., of 1905 a declivitous lakeside site encouraged a vertical rather than a horizontal organization of the interior with a two-storey living room as the spatial core.

A very different feeling pervades the small, squarish house at 6 Elizabeth Court in Oak Park that Wright built for Mrs Thomas Gale in 1909. Here flat slabs—which had been proposed as early as 1902 in a project (perhaps for execution in concrete) for the Yahara Boat Club in Madison, Wis.—replace the low-pitched hip or gable roofs of the characteristic Prairie Houses. Moreover, parapeted balconies and other simple rectangular features elaborate plastically the composition in a fashion that suggests the abstract sculpture of a decade later in Europe (see Chapter 21).

The W. A. Glasner house of 1905 at 850 Sheridan Road in Glencoe, Ill., on the contrary was extended longitudinally and the living area for the first time not at all articulated but completely unified (Figure 39). Something of the same longitudinal extension marks the much larger F. C. Robie house at 5757 Woodlawn Avenue in Chicago of 1909. But there the living room and dining room are separated by the chimney core and raised above the ground level. Built of fine Roman brick, this is the most monumental of these early houses. The long horizontal lines of the balcony below and the roof above dominate the composition; yet a cross element comes forward in the upper storeys to provide, less symmetrically than in his houses of cruciform plan, something of the abstract plasticity of the Gale house.

Another large house of the end of the decade, the Avery Coonley house at 300 Scottswood Road in Riverside, Ill., of 1908, offers a quite different and much more extended plan. The square block containing the living room rises above a terrace and a reflecting pool as the main element of the design, but from this block two long wings project. That to the left includes a large dining room and also very extensive service facilities at the rear; in the one to the right are the master’s suite and other bedrooms. Thus the house is, in a later phrase of Wright’s, ‘zoned’ according to function. The upper walls of this house are covered with a geometrical pattern produced by setting coloured tiles into the stucco. Wright never did quite the same thing again, but this led the way to his use of patterned concrete blocks a few years later.

Two of Wright’s non-domestic works of this period are of considerable importance. Unity Church in Oak Park has already been mentioned; the other was the Larkin Administration Building in Buffalo, N.Y., of 1904. Massive and even sculptural externally, particularly at the ends, this had a tall glass-roofed court running down the centre, around which the upper ranges of offices extended on galleries carried by somewhat Sullivanian piers. All the fittings of the offices, including the steel furniture—probably the first to be designed by an architect—were Wright’s. Thus he set here a wholly new standard of elegance, consistency, and coherence in semi-industrial building.

Within the massive slab-roofed block of the Unity Temple (Plate 143B), which is echoed beyond a low entrance link by the smaller block of the Sunday School, Wright achieved even more notably than inside the Larkin Building a new sort of monumental space-composition such as even his biggest houses hardly provided room for. The square auditorium with incut corners has double galleries on three sides and a pulpit platform on the fourth, behind which rises the organ. The multiple spatial elements seem to cross one another at different levels in a sort of three-dimensional plaid. Moreover, this theme is echoed in all the minor features, such as the wood stripping of the sand-finished plaster walls and the prominent lighting fixtures. Of this spatial development there had been some premonition in the auditorium block at one end of the Hillside House School that he built for his aunts outside Spring Green, Wis., in 1902; but there the masonry of the exterior walls and piers was still rather Richardsonian and the internal gallery consisted of a square set lozenge-wise.

Wright’s work down to 1910 was made available to Europeans by two publications of Wasmuth, the Berlin publisher; and the end of the first decade of the century does, coincidentally, mark a real turning point in his career. He would not be so prolific again before the forties; and henceforth, although he never ceased to build houses, these would no longer constitute the bulk of his production.

The production of the next decade, after his return from Europe in 1911, opens with two houses, however. Taliesin, which he built outside Spring Green for his mother in 1911, was soon much enlarged when he moved there himself and it always remained his principal residence. As a result of the growing needs of his family and of his school—not to speak of two major fires in 1914 and 1925—the Taliesin of today is very different, above all in its endless ramification, from what he planned in 1911; but the vocabulary of materials and design stayed more or less constant through all the years. Where the Prairie Houses echoed in their horizontal lines the flat Illinois terrain on which most of them were set, Taliesin is wrapped around a hill-top just below the crest. The use of various levels in the interior and a landscape-like elaboration of the low-pitched roofs represent his response to this more interesting site; after that the ‘Prairie’ master avoided flat sites for houses whenever he could!

Taliesin, combining a house, drawing-office, living accommodation for apprentices, and even farm buildings, had from almost the first a complex plan not readily definable as square, cruciform, or unilinear. But in a project of the same year 1911 in which Taliesin was originally built, that for the S. M. Booth house at Glencoe, Ill.—never executed, unfortunately, according to these plans—a new sort of organization appeared, related to the elaborated cube of the Gale house and also to the ‘zoned’ scheme of the Coonley house. A two-storey living-room was to provide both the spatial and the plastic core; from this wings serving different purposes would shoot out swastika-like.

The relative homogeneity of Wright’s production in the first decade of the century, following after the gradual convergence of his early work during the nineties, is explained by the nearly identical problems and sites that he faced in designing the houses mentioned so far. This homogeneity now gave way to an increasing variety that makes it difficult to summarize the work of these years. The Coonley Playhouse, built on the Coonley estate at Riverside in 1912, bears little resemblance to the original house of four years earlier. The plan is cruciform and symmetrical; but what is new here is the way the slab roofs, set at two different levels and pierced through their wide projections in order to let light reach the windows below, were used to achieve an even more boldly sculptural quality than in the project of 1902 for the Yahara Boat Club or the Gale house of 1909. Wright’s mastery of abstract decoration was wholly mature by this time. From the first he had used leaded glass in simple geometrical patterns in his windows,[408] but the windows in this playhouse are the finest of all. Moreover, these festive compositions of circles of coloured glass arranged asymmetrically resemble quite closely the abstract paintings that such artists as Kupka, Delaunay, and the Constructivists would shortly be producing in Europe.

Northome, the F. W. Little house at Wayzata, Minn., of 1913, is also quite different from all the earlier houses, yet not at all similar to the Coonley Playhouse. Raised on a ridge above the southern shore of Lake Minnetonka, this house consists of a series of pavilions—some open, some closed—strung along a single axis parallel to the water’s edge. That containing the living room, which is of almost monumental size and scale, dominates the whole. Wright seemed able now to invent a new mode almost with every individual commission, each one with potentialities as great as those of the Prairie Houses he had so thoroughly exploited in the decade before 1910.

The major work of the immediate pre-war years, the Midway Gardens of 1913-14 on the Midway south of Chicago, is rather hard to define precisely. Not quite a beer or Heuriger garden, nor yet a music-hall or cabaret in the ordinary European sense, the establishment consisted of a large outdoor dining and entertainment area with raised terraces on two sides, a stage and orchestra shed at the far end, and a closed restaurant block towards the street. Here Wright’s ambitions as a decorative artist could have free play. Abstract compositions of coloured circles like those in the windows of the Coonley Playhouse appeared here as wall-high murals at the ends of the covered restaurant. Moreover, the sculptural implications of the general composition of the playhouse were carried farther in the openwork ‘constructions’ that he set on the tops of the towers. At the same time he introduced a great deal of figurative sculpture stylized in a rather Cubist way. Thus several different aspects of the abstract and near-abstract art which was just coming into independent existence in Europe were closely paralleled in the adjuncts to Wright’s architecture here.

More architectonic patterns produced by simple geometrical means also ran riot at the Midway Gardens. Notable and significant was the use of extensive areas of patterned concrete blocks; these were somewhat like the patterned upper walls of the Coonley house of 1908 but all monochrome. The early demolition of the Midway Gardens makes it difficult to know whether this tremendous elaboration of the decorative aspects of Wright’s architecture was symphonic or cacophonous in total effect. Whatever the degree of their success or their failure, however, they opened a sort of ‘Mannerist’ or ‘Baroque’[409] period in his career that was destined to last for more than a decade.

During the First World War, in 1915, Wright was approached by emissaries of the Japanese Imperial Household to design and build the Imperial Hotel in Tokyo. Proceeding to Japan, Wright was largely concerned with this commission for the next seven years, finally bringing it to completion in 1922. This is the principal production of his ‘Baroque’ phase. It was also a notable engineering triumph, for his ingenious use of concrete slabs carried on a multitude of concrete piles brought it safely through the earthquake of 1923. Paul Mueller, the engineer of the old Adler & Sullivan office, was his collaborator here.

Abstract ornament proliferated on the hotel; some of it, carved in greenish lava, elaborates the garden courts of the vast H-shaped plan; still more is painted in gold and colour on the ceilings of the principal interiors. Moreover, the massive proportions of the masonry walls produce an effect of castle-like solidity wholly inexpressive of the method of their support and very far removed from the light and floating character of the Prairie Houses. On the whole this hotel represents, far more than the Midway Gardens, a cul-de-sac in Wright’s development.

Overlapping the period of construction of the Imperial Hotel came a series of houses in southern California in which the ‘Baroque’ element was gradually restrained. The earliest of these, Hollyhock House in Los Angeles and two smaller houses near by, were built for Aline Barnsdall in 1920 on a large estate bounded by Sunset and Hollywood Boulevards, Edgemont Street, and Vermont Avenue. These are of poured concrete very massively handled and carry considerable abstract sculptural ornamentation. For a slightly later series of four houses around Los Angeles, beginning with the house of 1923 for Mrs G. M. Millard at 645 Prospect Crescent in Pasadena, Wright developed a type of concrete-block construction with reinforcement in the joints that was of considerable technical interest and also offered special decorative possibilities. The idea of using concrete blocks cast with relief patterns of geometrical character goes back to the Midway Gardens, however, and walls covered with repeating ornamental units had first appeared at the Coonley house.

In the Millard house, particularly, the scale of the moulded blocks and the ingenious inclusion of pierced units—very similar to the pre-cast elements that Perret was using for the screen walls of his Le Raincy church at just this time—produced a masterpiece (Plate 144). This house, however, is not solely of interest for its construction and its decoration. In contrast to the horizontal composition of almost all his earlier houses except that in Racine for the Hardys, this is a tall vertical block, entered at the middle level, with the dining room and kitchen below and the two-storey living room opening out to a balcony at the front (Figure 40). The main bedroom is reached from a gallery overhanging the rear of the living-room. Both organizationally and visually this represents a surprising change, and the result closely resembled what a leading architect of the second generation had just then been proposing in Europe (Figure 45). There are, for instance, no hovering eaves here; instead a parapet continues the wall plane upwards and confines a roof terrace. This is as close as Wright ever came to building a ‘box-on-stilts’, his term of abuse for the advanced European houses of the twenties. It was as if, after the expansiveness of his work from the Midway Gardens to Hollyhock House, Wright wished to prove here his capacity to produce a house modest in scale and compact in section as well as in plan.

In the next decade, from 1924 to 1934, Wright’s actual production declined almost to zero although he was working on a series of important projects, some of which later provided the basis for executed buildings. Taliesin was rebuilt after a fire in 1925, however—it had already been rebuilt once before after an earlier fire in 1914—and a large house of concrete blocks, with almost no use of pattern except for occasional pierced grilles, was erected for his cousin Richard Lloyd Jones in 1929 at 3700 Birmingham Road in Tulsa, Okla. That is about all.

The small M. C. Willey house of 1934 at 255 Bedford Street, S.E., in Minneapolis marked the beginning of what proved to be almost a second career for Wright. Low and L-shaped, with practically no ornament whatsoever, this modest brick house introduced a major change in domestic planning. Not only are the living room and the dining room completely unified, as was first done at the Glasner house in 1905, but the kitchen—now re-christened ‘work-space’—opens into the main living area behind a range of glazed shelves (Figure 41). Thirty years later the full implications of this development are still not quite digested in America or even fully apprehended abroad; on the contrary, a reaction from open planning has perhaps begun.

It was not the Willey house, however, modest in size and very quiet in expression for all its revolutionary plan, that signalized the renewal of Wright’s activity. That he could take up his career again at the highest level of creativity became apparent to everyone with the construction of two much larger buildings both designed in 1936. Falling Water, a large house in the Pennsylvania woods, is cantilevered over a waterfall with a sense of drama even Wright had never hitherto approached. The Administration Building for the S. C. Johnson Wax Company at 1525 Howe Street in Racine, Wis., his first semi-industrial commission since the Larkin Building of 1904, was built in 1937-9. Both are as remarkable for the technical boldness of their use of concrete—totally different in the two cases—as for their design.

Falling Water has a rear section built of rough stone which rises like a tower from the native rock on the banks of Bear Run. From this solid vertical core are cantilevered out a series of concrete slabs bounded by plain parapets at their edges. This produces a very complex horizontal composition related to, but infinitely elaborated from, that of the Gale house of 1909 (Plate 145A). The completely unified living space is closed in by stone walls on the inner or dining side. It also extends out over the waterfall; the all-glass walls on that side, with their thin metal mullions, hardly seem to separate the interior space at all from that of the open terraces outside. A similar relationship exists between the bedrooms and their terraces on the upper floors.

Never before had Wright exploited the structural possibilities of concrete so boldly. In this amazingly plastic composition—if ‘plastic’ be the word for anything so light and suspended in appearance—it seems as if he had determined to outbid the European architects of the second modern generation at their own games (see Chapter 22). His early work has, in the clarity and axial character of the organization and the serenity of its expression, a classic if hardly a Classical quality; his work of 1914-24 shows a Baroque exuberance in the proliferation of the ornament. Now that he was approaching seventy his Romantic or anti-Classical tendencies—call them what you will—reached an intensity of purely architectonic expression comparable to the musical intensity of the late quartets of Beethoven that Wright so much admired. Falling Water, which might easily have been the swan song of Wright’s career, soon to be halted again by a second World War, proved in fact but the opening allegro in a new period of innovation and experiment.

The Johnson Building is very different from Falling Water. In it the curve rather than the cantilever provides the principal theme, and enclosure rather than interpenetration of exterior and interior space controls both the planning and the design (Plate 146A). The main office area is tall and unified, but it is filled with a forest of inverse-tapered concrete piers rising from tiny bronze shoes to carry circular slabs of concrete whose edges all but touch. The spaces between these lilypad-like disks were filled with tubes of Pyrex glass, and bands of similar tubes are carried around the building below the balcony and at the top of the plain red brick walls to provide additional natural light. In the more specialized adjuncts to the general office area curved and diagonal plan-elements lend a machine-like elegance to the shape of the building as a whole. Additional bands of glass tubing interrupt the smooth and continuous masonry surfaces at intervals, thus clearly indicating that these portions are of several storeys.

Falling Water and the Johnson Building were large and expensive structures; so also was Wingspread, the H. F. Johnson house that Wright built in Racine at the same time. This is zoned in the manner of the Booth project of 1911 around a tall central core. But in 1937 Wright also erected the first of what he called his ‘Usonian’ houses, the Herbert Jacobs house at Westmorland, near Madison, Wis. This modest L-shaped dwelling, with wooden ‘sandwich’ walls and a flat wooden slab roof, carried farther than the Willey house the integration of the ‘work-space’ or kitchen with the main living area. Here this rises in a masonry tower and is lighted by a clerestory, yet it is closely related to the space of the interior as a whole. A very considerable range of Wright’s later houses are variants of the Usonian model. Some were built before the War, even more in the last decade; some are of modest dimensions like the Jacobs house, others much larger. They exist in all parts of the United States, including the East, where he had hardly worked at all before this time.

The earlier Usonian houses were designed on a square module. This is true, for example, of the version that he prepared for Life magazine in 1938,[410] which thereby received the same sort of national circulation that the Ladies Home Journal gave to three of his projects more than a generation earlier.[411] But Wright was now interested also in developing the hexagon and the triangle as basic units. Beginning with the Hanna house of 1937 at 737 Coronado Street in Palo Alto, Cal., he continued in many others to explore the possibilities of planning based on 60-30-degree angles.

In the most extraordinary house that he built in these pre-war years, his own winter residence, Taliesin West, begun in 1938 in the desert outside Phoenix, Ariz., 45-degree diagonals are used in the planning and almost all the structural elements are battered or canted. However, it is the materials which give this edifice—like Taliesin itself at once a house, a working place, and a school—its unique qualities. The substructure is of ‘desert concrete’, that is great rough blocks of tawny local stone placed in forms and loosely stuck together, so to say, with concrete; the superstructure is of dark-stained timber frames mostly filled only with canvas to allow a maximum flow of air. As at the original Taliesin in Wisconsin, Wright kept on enlarging Taliesin West, not always to its advantage. Another example of ‘desert-concrete’ construction, the Rose Pauson house of 1940 in Phoenix, was destroyed by fire. It was, in its very sculptural way, a masterpiece of this period unlike anything else he ever built and is still an impressive ruin.

It was characteristic of Wright’s activity in his ‘second’ career that the versatility of his invention knew no bounds. Many earlier ideas that had existed only in projects could come to fruition now that his services were in such demand. At the same time it is hard to believe that in the plain white stucco walls, extensive window bands, and thin roof slab of the E. J. Kaufmann guest house, built just above Falling Water in 1939, or in the G. D. Sturges house of the same year at 449 Skyway Road in Brentwood Heights near Los Angeles, cantilevered out from a hill-slope, Wright was not consciously rivalling the effects of the European architects of the second generation whom he professed to scorn—rivalling them, but also making very much his own such of their effects as he cared to emulate.

Wright did not drop the novel methods of construction that he had developed earlier as he tried out new ones. In his most extensive late commission, the layout of a new campus for Florida Southern College at Lakeland in Florida, begun in 1938, the plan is highly formal at the same time that it is markedly asymmetrical. It thus elaborates upon the angular themes of his project of 1927 for a desert resort at Chandler, Arizona—incidentally the point at which his interest in 60-30-degree angles began. The buildings at Florida Southern, starting with the Ann Pfeiffer Chapel of 1940 to which many more were later added, are mostly of concrete-block construction, but with much less use of patterned elements than in the executed work and projects of the twenties.

The Second World War interrupted Wright’s career less than the First. Various projects initiated in the war years came to fruition soon after the war was over and gave evidence of the continuing vitality of his powers of invention. The second house for Herbert Jacobs at Middleton in the country west of Madison, Wis., was very different from the Usonian one of 1937. Ever since an unexecuted house project of 1938 Wright had been fascinated by the possibilities of using the circle in planning. While he had tried out the form in the Florida Southern Library before the war, the Jacobs house of 1948 was the first of a series of houses that he built with curved plans. Its two-storey living area bends around a circular sunken garden court with the bedrooms opening off a balcony above (Figure 42). On the other side the house is half buried in the hill-top, above which rise its walls of coursed rubble. A tower-like circular core near one end of the convex side provides a strong vertical accent.

Another house of the post-war years, also based on the circle, is quite different in character. The Sol Friedman house in Pleasantville, N.Y., is roofed with mushroom-like concrete slabs; the two intersecting closed circles of the actual dwelling are balanced at the end of a straight terrace parapet by the open circle of the carport (Plate 145B). This was completed in 1949 with battered walls of almost Richardsonian random ashlar masonry below a strip of metal-framed windows. A still later ‘house of circles’ for his son David J. Wright was built near Phoenix, Ariz., in 1952. This is of concrete blocks and raised off the ground, with the approach up a gently sloping helical ramp to the various curved rooms on the first storey. The circle and the helix appear also in an urban building of these years, the shop for V. C. Morris in Maiden Lane, San Francisco, completed in 1949. Here the street façade is a sheer plane of yellow brick broken only by the entrance, which is a Sullivanian—or Richardsonian—arch like that of the Heurtley house of 1902. Inside, a helical ramp rises around the central circular area beneath a ceiling made of bubble-like elements executed in plastics.

A major work of these years, the extension of the Johnson Administration Building in Racine, Wis., also completed in 1949, makes much use of circles also (Plate 146A). North of the existing office building Wright surrounded a square court with open carports whose outer walls of solid brickwork shut out the surrounding city; inside these walls are ranged short concrete columns with lily-pad tops like those in the section that he built ten years earlier. In the centre of the ‘piazza’ thus defined rises a laboratory tower of tree-like structure. The upper floors of this, alternately square with rounded corners and circular, are all cantilevered out from a central cylindrical core which contains the lift and the vertical canalizations. Alternate bands of brickwork and Pyrex tubing, such as were used on the original building, enclose the tower except at ground level; there the space of the court continues under the cantilevered floors above as far as the solid central core.

This relatively modest tower prepared the way for Wright’s skyscraper in Bartlesville, Okla., of 1953-5, which has been mentioned earlier. Actually, however, this Price Tower,[412] which is partly occupied by offices and partly by flats, is the final realization of a project originally prepared in 1929 for a block of flats for St Mark’s Church in New York. This he had elaborated in the intervening years in projects for blocks of flats in Chicago and for a hotel in Washington.

While Wright was continuing to employ in his houses of the late forties and early fifties a variety of modes of design that go back to the thirties, and also developing at Florida Southern and in Bartlesville ideas dating from his inactive period in the late twenties, he continued to strike out in other directions too. The Neils house at 2801 Burnham Boulevard on Cedar Lake in Minneapolis, Minn., completed in 1951, is all of coloured marble rubble provided by the client; the Walker house at Carmel, Cal., completed in 1952, is a glazed polygonal pavilion overhanging the sea. Where the Prairie Houses of the first decade of Wright’s mature career may all seem in retrospect to have come out of the same, or nearly identical, moulds, the many houses designed in his seventies and eighties are notable for the great variety of their siting, their materials, and the geometrical themes of their planning.

Nor was the domestic field anything like the sole area of his activity. In addition to the college buildings, the shop, the skyscraper, and the laboratory that have been mentioned, Wright built during the years 1947-52 a Unitarian church in Madison, Wis., of very original character. The products of his multifarious activity in these years include, moreover, many projects for all sorts of structures, some of which have been completed—notably the Solomon R. Guggenheim Museum in New York (Plate 188). A decade and more of designing and redesigning preceded the initiation of this remarkable helical concrete building in 1956. Of three other late projects, those for an opera-house in Baghdad and for an Arizona state capitol in Phoenix, dating from 1957, are unlikely to be built; but the county buildings for Marin County, Cal., are now well advanced.

In spite of so much late activity, greater than that of his early maturity, in spite (or perhaps, in part, because) of its kaleidoscopic variety, Wright’s actual influence was less significant than forty years before; at least it was of a very different order. He still outpaced his juniors both of the next generation and the one after; but few if any were able to follow with any success along the intensely personal paths he opened.[413] Like Perret to the end of his life, Wright continued at ninety to offer an inspiration to all architects, but there has risen no school of imitators to vulgarize his manner as there was long a school of imitators of Perret in France.

In creative power, in productivity, and, over the forty years and more since 1910, in influence, Wright overshadowed all the other American architects of his generation. Inspired by Wright as well as by Sullivan, there flourished for a while a sort of ‘Second Chicago School’ to which Purcell & Elmslie; George W. Maher (1864-1926); Schmidt, Garden & Martin, and several other architects who were active in the Middle West before the First World War may be considered to belong.[414] But this school flickered out in the twenties as most of its members succumbed to the dominant ‘traditionalism’ of the day or else ceased to find clients.[415] Four rather more vital and original architects appeared shortly after 1900 in California: the brothers Greene (Charles S., 1868-1957, and Henry M., 1867-1954), Irving Gill (1870-1936), and Bernard R. Maybeck (1862-1957).[416] But the productive careers of the Greenes, of Gill, and, to a lesser extent, that of Maybeck came pretty much to a close, like those of the Chicagoans, around 1915 with the resounding success of the ‘traditional’ buildings designed by Bertram G. Goodhue (1869-1924) for the San Diego Exhibition of that year.[417] These were in the most ornate sort of Spanish Baroque, quite archaeologically handled; and the emulation of them, which at once became endemic in California, turned most local architects away from innovation for almost twenty years.

Maybeck, who had been a graduate of the École des Beaux-Arts in the eighties, contributed to the San Francisco Exhibition[418] of the same year the still extant Fine Arts Building in an equally ‘traditional’ but more Classical vein. Partly ruined today, his tawny stucco columns and entablatures have the air of a painting by Pannini or Hubert Robert. For all its charm, this was a surprising work to come from a man who had earlier shown himself, in the Christian Science Church of 1910 in Berkeley, Cal., almost as bold an innovator as Wright even though he employed for that a fantastically eclectic vocabulary of reminiscent forms (Plate 146B). Many Berkeley houses, moreover, ranging over several decades in date, also prove Maybeck to have been an architect of great originality and surprising versatility.

In Berkeley also are several houses by John Galen Howard (1864-1931) as well as his building for the University of California’s School of Architecture, of which he was for long the Dean. His building at the University (which has in addition a Faculty Club and one or two other things by Maybeck), the Gregory house of about 1904, and the architect’s own house of 1912 are also notable examples of free design dating from the first decades of the century. Howard’s informal work is more directly related than are Wright’s houses to the Shingle Style of the preceding period, though not specifically to that of Richardson, for whom, however, Howard had actually worked in the mid eighties before he came to California. Most of his work at the University, in fact, is in an Italianate vein, and the campus is dominated by his tall, campanile-like clock tower.

The production of the Greene brothers in this period, entirely domestic and largely in Pasadena, offers a more coherent corpus than that of any modern American architect of their generation except Wright. Related, like the work of Howard, to the Shingle Style, which had been brought to Pasadena and Los Angeles by Eastern architects in the eighties and nineties, the Greenes’ houses are most interesting for their successful assimilation of oriental influences. The best example is the Gamble house at 4 Westmorland Street in Pasadena of 1908-9 (Plate 147A). But the Pitcairn house of 1906 and the Blacker house of 1907, at 289 West State Street and at 1157 Hillcrest respectively, as well as the later Thorsen house of 1909, at 2307 Piedmont Avenue in Berkeley, now a fraternity house, are also excellent.

Shingled walls, low-pitched and wide-spreading gables, and extensive porte-cochères and verandas of stick-work surpassing in virtuosity those of the Stick Style, were combined by the Greenes in rather loosely organized compositions. Less formal and regular than Wright’s Prairie Houses, theirs are executed throughout with a craftsmanship in wood rivalling that of the Japanese, whom they, like Wright, so much admired. The Greenes’ plans are less open than Wright’s, but they made more use of verandas and balconies than he. Superb woodwork and fine stained glass combine with the specially designed furniture in the interiors to produce ensembles of a sturdy elegance hardly matched by any of Wright’s. Those in the Blacker and Thorsen houses, whose clients were both in the lumber business, are especially rich.

Moreover, a ‘California Bungalow’ mode[419]—at worst but a parody at small scale of the Greenes’ expensive mansions, at best sharing many of their virtues of directness and simplicity if not of imaginative craftsmanship—became widely popular thanks to national magazines, pattern-books, and the activities of many builders. This was true not alone in the West but throughout the country in the very years after 1910 when ‘traditionalism’, usually in Neo-Colonial guise, closed in most completely on American domestic architecture.

The reputation of the Greenes today is less than that of the more articulate but less consistent Maybeck. But when modern architecture revived in California in the thirties the new men were fully aware of what the Greenes had accomplished. Thus their work provided, together with that of Maybeck and Howard, a background and a tradition for the local development of a largely autochthonous domestic architecture in the San Francisco Bay area. This was a truly living tradition[420] quite unlike the abortive revival of the architecture of the Spanish Missions, which it has now almost completely displaced. But the Mission influence was not altogether a negative one in early twentieth-century California, as the work of Irving Gill illustrates.

Gill was less prolific than the Greene brothers, and most of what he built is less striking. Like Voysey, he was in principle a reformer not a revolutionary, finding his inspiration consciously in the local structural tradition of the early Spanish Missions and haciendas. As a result some of his buildings, such as the First Church of Christ Scientist of 1904-7 in San Diego or in Los Angeles the Laughlan house of 1907 and the Banning house of 1911, at 666 West 28th Street and 503 South Commonwealth Avenue respectively, with their elliptically arched loggias and their grilles of ornamental ironwork, are almost as ‘Spanish Colonial’ as the work of the outright traditionalists around him.

Gill’s most interesting and mature houses, thanks to their smooth stucco walls, large window areas, and avoidance of stylistic detail, can also have a deceptive air of being European rather than American and of a period some years later than that in which they were actually built. In his best work, such as the Dodge house (Plate 147B) of 1915-16 at 950 North Kings Road in Los Angeles or the Scripps house at La Jolla of 1917, now the Art Centre, the asymmetrically organized blocks, crisply cut by large windows of various sizes carefully sashed and disposed, with roof terraces or flat roofs above, more than rival the contemporary houses of the Austrian architect Adolf Loos (Plate 155A) in the abstract distinction of the composition. They even approach rather closely the most advanced European houses of the next decade (see Chapters 21 and 22).

Gill’s interiors are especially fine and also quite like Loos’s. Very different from the rich orientalizing rooms designed by the Greenes, they are in fact more similar to real Japanese interiors in their severe elegance. The walls of fine smooth cabinet woods, with no mouldings at all, are warm in colour, and Voysey-like wooden grilles of plain square spindles give human scale. The whole effect, in its clarity of form and simplicity of means, is certainly more premonitory of the next stage of modern architecture than any other American work of its period.

Gill continued to practise intermittently down into the thirties, but his finest work was done in the second decade of the century. He had little influence locally and still less nationally, yet his best houses extend very notably the range of achievement of the first generation of modern architects in America, even though his later production declined sadly in quantity and even in quality. Wright alone was able to renew his career successfully after the reaction against modern architecture that dominated America from coast to coast during the twenty years from the First World War to the mid thirties finally came to an end.

------------------------------------------------------------------------

20. PETER BEHRENS AND OTHER GERMAN ARCHITECTS

THE pattern of architectural development in Germany in the early decades of this century was rather different from that in either France or the United States. No academy, native or foreign, no influences from the École des Beaux-Arts discouraged innovation; yet there was an early and general reaction against the whimsicality and the decorative excesses of the Art Nouveau at which most of the younger men had tried their hands before 1900. After the First World War, however, the example of Expressionism in painting and sculpture led many architects to excesses of another sort. Expressionism in architecture,[421] or something very close to it, is not restricted to Germany. The most extreme example of any consequence, and probably the earliest, is Dutch, the Scheepvaarthuis in Amsterdam of 1912-13 by van der Meij (see Chapter 21). In Germany around 1920 various architects who had earlier been predominantly ‘traditional’ in their approach were influenced by Expressionism, as well as others who were already programmatically modern; nor was that influence restricted to the modern architects of the first generation (see Chapter 22).

The boundary line between what, in retrospect, still seems definitely modern and what now seems very similar to the ‘traditional’ work of these decades in other countries is much less sharp than in America. And no German architect of their own generation had the continuously creative achievement of a Perret or a Wright to his credit. Nevertheless Peter Behrens stands out among his contemporaries because of the vigorous boldness of his industrial buildings. Moreover, the influence of his factories of around 1910 was crucial on the next generation, and several of the later leaders actually worked in his office at that relevant period. Yet all but Behrens’s finest work can be matched in the production of other German architects; while his own vitality as an innovator was rather strictly limited to a few years and to what he did for one corporate client. That client was the A.E.G. (German General Electric Company), which had already employed Messel down to his death in 1909.

Messel and Ludwig Hoffmann (1851-1932) dominated the architectural scene in Berlin, where the latter was appointed City Architect in 1896 on the strength of his vast academic Imperial Law Courts of 1886-95 in Leipzig. In the early years of the century they both developed a formal mode that was more ‘traditional’ than modern. Despite Messel’s and Hoffmann’s usual preference for conventional sixteenth- or eighteenth-century models, Behrens was certainly not uninfluenced by their mode of design, even though his more positive sources of inspiration were of a less conservative order. Yet, in so far as one can sort out the different architectural camps in Germany in these years, Behrens must be considered well to the artistic ‘left’ of Messel and Hoffmann.

Germany was certainly very receptive to new ideas in decoration when Behrens’s architectural career began at the turn of the century—receptive rather than creative. There were other Germans who handled the Art Nouveau with considerable originality besides August Endell, notably Bernard Pankok (b. 1872) and Richard Riemerschmid (1868-1957); but two foreigners, neither of them very prolific builders, seem to have been the most influential figures on the German architectural scene at the opening of the new century. The Belgian Van de Velde had moved from Paris to Berlin in 1899; the Austrian Olbrich was called to Darmstadt by the Grand Duke in the same year. Olbrich stayed at Darmstadt until his early death in 1908; Van de Velde, however, left Berlin in 1902 when he was invited to Weimar to head the School of Arts and Crafts there which later became the Bauhaus. Van de Velde’s finest Art Nouveau furniture dates from his Berlin years around 1900. As late as 1906,[422] the Central Hall which he designed in the Dresden Exhibition showed him still a competent if rather heavy-handed decorator in the Art Nouveau tradition.

Van de Velde’s remodelling of the Folkwang Museum at Hagen of 1900-2, quite Art Nouveau in its details, his Esche house at Chemnitz of 1903, and his Leuring house at Scheveningen in Holland of the next year, both very massive and heavily mansarded though unornamented externally like his own house of 1895-6 at Uccle, hardly require particular mention. However, for the school that he headed in Weimar he completed in 1906 a building even more devoid of Art Nouveau elements and notably straightforward in character. The plain white stucco walls below his usual heavy mansards were very frankly fenestrated with ranges of wide studio windows, perhaps in emulation of Mackintosh’s Glasgow Art School. Indeed, the general effect is even simpler and more rectilinear than that of its possible Scottish prototype. The problem of his responsibility or lack of responsibility for the design of the Théâtre des Champs Élysées in Paris of 1911-13 has already been discussed (see Chapter 17).

Van de Velde continued to build occasionally throughout all his long life—some portions of his Kröller-Müller Museum near Otterlo in Holland were only completed in 1953—but his last pre-war work was the theatre that he designed and executed in 1913-14 for the Werkbund Exhibition at Cologne. Some trace of the massively plastic quality of his Dresden hall of 1906—so different from the delicacy and grace of the Art Nouveau in its best period—remained in the curved walls and roof of this edifice, but the whole effect was lighter and plainer, more abstract one might almost say.

The resemblance of Olbrich’s Ernst Ludwig Haus of 1901 at the Darmstadt Artists’ Colony to Mackintosh’s Art School has already been noted (see Chapter 17). At Darmstadt he also continued to build houses for some years, and his work there culminated in the Exhibition Gallery and the Wedding Tower on the Matildenhöhe, erected in 1907. The former was blocky and somewhat classicizing in character, at once very plain and very formal. The latter, of brick, had a more Hanseatic flavour because of its arched and panelled gable; but it also included a novel motif, bands of windows that seem to carry round a corner, that was destined to be very influential everywhere in the twenties.

In the next and last year of Olbrich’s life—he died, it will be recalled, at the early age of forty-one—two important commissions came to him away from Darmstadt. The Feinhals house at Marienburg near Cologne repeats the blocky symmetrical composition of the Exhibition Building, the walls being articulated only with flat oblong panels. The loggia between, however, has a range of Greek Doric columns, clear evidence of the influence of Romantic Classicism that was growing stronger in Germany all through this decade. But Olbrich had little real appreciation of the subtle elegance of the work of Schinkel and his contemporaries, or so it would appear from this house.

The buildings of the East Cemetery in Munich, designed by Hans Grässel (1860-?) in 1894 and completed in 1900, are perhaps the first examples of this sort of ‘Neo-Neo-Classicism’. Yet beside the contemporary Neo-Baroque of the Munich Palace of Justice built in 1897 by Grässel’s master, Friedrich von Thiersch (1852-1921), nearly as over-scaled and aggressive as Wallot’s Reichstag in Berlin, the rather Schinkelesque work at the cemetery appears, in its crispness and its relative simplicity, almost as ‘modern’ as anything by Olbrich. As has been noted earlier, Schinkel remained a major inspiration to such a leader of the second generation of modern architects as Mies van der Rohe, so this influence has a continuing significance.

A much larger building by Olbrich than the house at Marienburg, also completed in the year of his death, the Tietz (now Kaufhof) Department Store in Düsseldorf, repeats the reiterative verticalism of those portions of Messel’s Wertheim store in Berlin that were built in 1900-4, though Olbrich’s detailing is not medievalizing like Messel’s but rather semi-Classical. Neither of these later things maintains the promise of his Ernst Ludwig Haus; they rather illustrate that general recession from bold innovation which characterized the architecture of this decade in Germany, a recession corresponding more or less closely to the general resurgence of ‘traditionalism’ in England and America that came a few years later (see Chapter 24).

Peter Behrens (1868-1940), only a year younger than Olbrich, began his career as an architect at Darmstadt. From 1896 on, before being called there, he had only done decorative work of a markedly Art Nouveau sort. In his own house in the Artists’ Colony of 1900-1—the only one not built by Olbrich—the interiors are still quite Art Nouveau, but the clumsy exterior has little interest except as a document of revolt. Yet the plan is quite like that of Wright’s own house of 1889 in Oak Park, allowing a real flow of space through wide openings between entrance hall, living-room, and dining-room. By 1902 the ‘Hessian’ interior that he contributed to the Turin Exhibition was wholly rectilinear, presumably under the influence of Olbrich and Mackintosh. A similar severity characterized the work that he did, much of it merely open pergolas, for the Düsseldorf Garden and Art Exhibition of 1904.

By this time Behrens’s personal style was maturing, although his debt to Olbrich remained very evident. The Art Pavilion for the North-West German Art Exhibition held in Oldenburg in 1904 was a symmetrical composition of cubical masses, the flatness of their surfaces even more emphasized by linear panelling than in Olbrich’s work. The Obenauer house of 1905-6 at Sankt Johann near Saarbrücken is rather more loosely composed; indeed, its white stucco walls, slated roofs, and grouped windows distinctly recall Voysey’s houses, which were by this time very well known in Germany thanks to the Studio and Muthesius’s book. The garden front, however, is symmetrical and the plan not as open as that of his own house of four years earlier.

In Behrens’s next two buildings, the small Concert Hall in the Flora Garden at Cologne of 1906 and the large Crematorium at Delstern near Hagen completed the following year, the geometrical panelling in black and white, used both inside and out, recalls a little San Miniato in Florence. But the blocky geometry of the Oldenburg pavilion and its smooth flat surfaces were also repeated, so that both these buildings have a curiously model-like look as if they were made of sheets of cardboard.

Behrens’s two finest works up to this time, the Schröder house of 1908-9—no longer extant—and the Cuno house of 1909-10 in the Hassleyerstrasse at Eppenhausen near Hagen, have a much more solid appearance, with quarry-faced masonry below and roughcast walls above (Plate 148B). The symmetrical façades, which correspond to completely symmetrical plans, are at once more tightly and more subtly composed. Here English influence seems to have been superseded by an attempt, rather more successful than Olbrich’s at Marienburg, to emulate Schinkel. A third early house by Behrens, the Goedecke house at Oppenhausen of 1911-12, is equally formal but not symmetrical, recalling thus a little Schinkel’s Schloss Glienecke near Potsdam.

Somewhat similar to Behrens’s work of this period in its evident derivation from German Romantic Classicism, but more delicate in scale, was the work of Heinrich Tessenow (1876-1950), notably his Festival Theatre of 1910-13 and the other buildings he designed and erected for the Art Colony at Hellerau near Dresden. But such German work, of which a great deal was produced in the decade before the First World War, corresponds rather closely, despite the frequent stylization of detail and the serious concern with geometrical clarity in composition, to the Neo-Georgian of England and America in the early twentieth century, and also to much parallel work in the Scandinavian countries that is usually of rather higher quality (see Chapter 24).

Moreover, those Frenchmen who castigated the Théâtre des Champs Élysées as ‘Boche’ during the First World War because of the presumption that it was designed by Van de Velde, born a Belgian but head of a German art school, were not altogether wrong. In its scraped Classicism and rigidly geometrical ordonnance Perret’s façade was not at all remote from one of the most characteristic German modes of the years just before 1914. Perret’s industrial work was, of course, much more significant for the future.

So also with Behrens it was the challenge that his position as architect of the A.E.G. brought of working in the industrial field that made him briefly a rival of Wright, and even more particularly of Perret, as a major architectural innovator. Behrens’s first work for the A.E.G., the Turbine Factory at the corner of the Hussitenstrasse and the Berlichingenstrasse in Moabit, an industrial suburb of Berlin, was erected in 1909 immediately upon his appointment as successor to Messel. This broke new ground in several ways. It was built partly of poured concrete, partly of exposed steel, with both materials very directly expressed (Plate 149A). The side wall of glass and steel more than rivals in its openness those of the department stores designed by Art Nouveau architects (Plates 131B and 133). But Behrens’s façade, in contradistinction to theirs, has no applied ornament whatsoever. Moreover, he ordered the whole composition as carefully as Schinkel might have done if either large factories or metal-and-glass construction had come within his purview.

The end façade of the Turbine Factory is slightly less frank in design. The concrete corners on either side of the central window-wall of metal and glass are battered and striated horizontally as if to suggest rusticated masonry. The gable of the multi-faceted roof is brought forward to shelter the window-wall; this projects slightly in front of the concrete corners, almost like a Shavian bay-window raised to industrial scale. The treatment of the window-bands of the lower concrete block to the left resembles that of Schinkel’s articulated walls on the Berlin Schauspielhaus, but with all the Greek mouldings omitted. Thus the functional elements of a factory executed throughout in new materials were here for the first time in Germany architectonically ordered with no dependence on decoration of any sort. Wright had done much the same four years earlier in his little-known E.-Z. Polish Factory in Chicago, but the scale of that is modest and its walls are not extensively fenestrated. Perret had come closer to it in his Garage Ponthieu in Paris, also built in 1905. There can be little question, however, that Behrens’s is the finest building of the three.

In two more factories built in 1910 for the A.E.G., both much larger but neither of them quite so striking, Behrens broadened his range as an industrial architect. The High Tension Factory in the Humboldthain is of brick, not concrete or steel. Except for a few minor elements somewhat suggesting pedimented temple-fronts translated into an industrial vocabulary, he handled the vast façades here with the same directness as the side elevation of metal and glass at the Turbine Factory. The Small Motors Factory in the Voltastrasse is similar but much finer (Plate 148A). There the brick piers have rounded corners and rise unbroken almost the full height of the building. The effect is somewhat like that portion of Messel’s Wertheim Store which was built in the late nineties, but the scale is larger, and there is none of Messel’s rich, half-traditional, half-Art-Nouveau detailing. Instead, the careful proportioning and the suave but extremely straightforward treatment of the structural elements again suggests Schinkel’s sort of ‘rationalism’ yet succeeds in doing so, as at the earlier Turbine Factory, with almost no reminiscence of actual Romantic Classical forms.

Thanks to the widening range of responsibility that German industry was now ready to give architects, Behrens not only built these big factories for the A.E.G. and also redecorated their retail shops all over Berlin, but he was soon asked in addition to provide some blocks of flats for the company’s workmen at Hennigsdorf outside Berlin. This was a social challenge which neither Wright nor Perret had to meet. (In fact, however, Wright did in 1904 design terrace-houses that were never executed for Larkin Company workers in Buffalo; while in France low-cost housing had a very important place in Garnier’s projects for a ‘Cité Industrielle’.) Henceforth, such housing would be a major preoccupation of most modern architects. This is true not only in Germany but all over the western world, and especially in Holland and Scandinavia. The origins of low-cost housing go back to the 1840s in England when Henry Roberts, whose Fishmongers’ Hall in London has been mentioned, became the first architect to specialize in this field. But the early history of housing[423] is of more sociological than architectural interest. Moreover, what the nineteenth century esteemed to be ‘model’ low-cost dwellings have too often had to be demolished as ‘sub-standard’ in the twentieth. Even the interest and activities of present-day architects may not spare the twentieth century the shame of building again as a public service what posterity will consider slums.

Various small A.E.G. factories for making porcelain, lacquer, and other specialized products were also erected by Behrens in 1910 and 1911, none of particular interest. In 1911-12, however, there followed the Large Machine Assembly Hall at the corner of the Voltastrasse and the Hussitenstrasse near the Small Motors Factory. This rivals in quality the Turbine Factory of 1909. Once more a great rectangular volume is covered with a multi-faceted steel-framed roof, the structure below being in this instance also of steel with no use of concrete. The metal frame is largely filled with glass, but brick was introduced at the base and on the ends. The scale of this unit is less monumental than that of the Turbine Factory, though the size is much greater. The general effect, particularly that of the interior with its travelling cranes, is at once light and dramatic. A big A.E.G. plant was also built by Behrens at Riga in Russia in 1913.

Three large non-industrial commissions of 1911-12 show how this work for the A.E.G. affected Behrens’s approach to design. Although it is built of stone not brick, the German Embassy (Plate 27A) opposite Monferran’s St Isaac’s Cathedral in Petersburg is, at first sight, deceptively like the Small Motors Factory. Actually, the façade has a range of engaged Doric columns, but by their tall slim proportions and their lack of entasis these were, so to say, ‘industrially’ stylized. The great scale, the absolute regularity, and a certain coldness surely derived in part from the factories of the previous two years; but these also recall Romantic Classical monuments of Alexander I’s time in Petersburg.

Behrens’s enormous office building for the Mannesmann Steel Works on the Rhine at Düsseldorf was less successful, as was also that for the Continental Rubber Company in Hanover. The latter was designed in 1911 and begun in 1913, but not completed until after the First World War, in 1920; it was destroyed in the Second World War. The heavily reiterative sort of scraped Classicism Behrens used for these overpowering masonry blocks lacked the subtlety of composition of the Hagen houses yet retained something of the directness of expression of the A.E.G. factories. They were not untypical, however, of much large-scale German building of the second and third decades of the century. This mode developed fairly directly out of the Berlin work of Messel and Ludwig Hoffmann, although it was usually much less specifically ‘traditional’ in its detailing and even more aggressive in scale; a not dissimilar mode returned to official favour under Hitler in the mid thirties, usually with very coarse detailing.

With these big office buildings by Behrens and others one may compare the work of this period by various other German architects who preferred less classicizing modes. Early buildings by Fritz Schumacher (1869-1947), such as his crematorium in Dresden of 1908, also illustrate the megalomaniac tendencies of the period that seem so expressive of the expansive ambitions of William II’s Second Reich. The many schools that Schumacher built in Hamburg just before the First World War are simpler, although still rather heavily scaled, and more comparable in quality to Behrens’s work. One in particular, built in 1914 in the Ahrensburgerstrasse, almost echoes the elongated colonnade of Behrens’s Petersburg Embassy, but the ‘columns’ are plain piers executed in dark red brick[424] and strung along a front that is concave not flat. The bath-house at Eppenhausen, also of 1914, is very like the schools; while in the Kunstgewerbe Haus of the previous year on the Holstenwall in Hamburg a similar mode was employed for what is, in effect, a large office building. This seems to have initiated a local tradition of design for commercial buildings which was maintained in the twenties with little change, not only by Schumacher but by several other Hamburg architects. Schumacher’s cemetery chapel, built as late as 1923, follows much the same line.

In Stuttgart the railway station by Paul Bonatz (1877-1951) and F. E. Scholer (b. 1874) is the finest though not the largest of several built in Germany in these years. Designed in 1911, it was started only in 1914, just as the enormous and much less interesting one at Leipzig with its six parallel sheds, begun by Wilhelm Lossow (1852-1914) and M. H. Kühne in 1907, was reaching completion. That at Stuttgart was not finished until 1927 because of the interruption caused by the First World War. This structure has a rather Richardsonian flavour in its extensive unbroken wall surfaces of rock-faced ashlar and its plain round arches (Plate 152). But the influence here came rather from the Munich architect Theodor Fischer (1862-1938). Fischer’s Romanesquoid churches, such as that of the Redeemer in Munich of 1899-1901 and the Garrison Church of 1908-11 in Ulm, were among the largest and most strikingly novel built in the opening years of the century in Germany; in the latter he even used ferro-concrete principals to carry the roof of the nave. Fischer’s Art Gallery of 1911 in Stuttgart was both more delicate in scale and rather more archaeological in its detailing; Bonatz’s Stuttgart work is bolder, simpler, and quite as admirably expressive of the traditional materials used.

With the Stuttgart Station may be contrasted the rather earlier one at Hamburg that Heinrich Reinhardt (1868-?) and Georg Süssenguth (1862-?) built in 1903-6. There the major sections—shed, concourse, etc.—designed by the engineer Medling resemble rather closely Contamin and Dutert’s Galerie des Machines at the Paris Exhibition of 1889. These great constructions of iron and glass fortunately quite overshadow the low ranges of accessory elements in masonry, with ornament still in the Meistersinger mode of the eighties, contributed by the architects. The differences between these two notable stations well illustrate that reaction towards masonry construction and a more or less traditional approach to design that was developing strength in the decade preceding the First World War. In the history of the railroad station as a type the Hamburg Hauptbahnhof represents, not a new beginning, but the end of a line descending from the great shed-dominated stations of the mid nineteenth century.

Intermediate in date between the Hamburg and Stuttgart stations was that at Karlsruhe built by August Stürzenacker in 1908-13. Although masonry construction and masonry forms dominate here as at Stuttgart, the simplification of mass and space composition throughout, and above all the elegant detailing, give evidence of the continuing leadership of Olbrich at the time of his death. Olbrich never built a station himself, but he won third place in the 1903 competition for that at Basel and second place in the 1907 competition for Darmstadt.

In other specialized fields of building a forward line of development is more evident. Two big circular halls, one in Frankfort built by Thiersch in 1907-8, the other in Breslau built by Max Berg (b. 1870) in 1910-12 (Plate 149B), are more notable than the contemporary railway stations at Stuttgart and Karlsruhe. Like Behrens’s industrial work for the A.E.G., these structures illustrate the vital stimulus that German architects were obtaining in these generally somewhat reactionary years from the use of engineering solutions and materials other than masonry—steel at Frankfort, ferro-concrete at Breslau—to cover and enclose space. In the case of Thiersch this is the more remarkable when one remembers the ponderous traditionalism of his Neo-Baroque Palace of Justice in Munich built ten years before. While Berg on the exterior of his vast hall approaches the attenuated Classicism of Perret’s work of the next decade, the superb interior reminds one at once of Piranesi and of the much later structures of Nervi.

German architects of this generation were rarely able to carry over into the designing of more conventional structures the boldness and freshness of approach of their large-scale work. They seem to have felt no such call to regenerate architecture as Wright had imbibed from Sullivan; nor did they, like Perret, attempt to use the new materials and the new structural methods consistently for all sorts of buildings whatever their particular purpose. German production before and after the First World War, as represented in the œuvre of such then highly esteemed figures[425] as Oskar Kaufmann (b. 1873), German Bestelmeyer (1874-1942), and Wilhelm Kreis (b. 1873), to mention but three of the best known, shades over almost imperceptibly from industrial and semi-industrial buildings of bold and original character to a range of structures in various tasteful modes that are, in retrospect, hardly distinguishable from the traditional work of this period in other countries. This has already been noted as regards Tessenow. Characteristic examples of these men’s work were Bestelmeyer’s extensions of the University and the Technical High School in Munich, of 1906-10 and 1922 respectively, both in the local tradition of Theodor Fischer’s work. The Museum of Prehistory in Halle that Kreis built in 1916 with K. A. Jüngst was more traditional even than Bestelmeyer’s work, although Provincial-Roman rather than Romanesque in inspiration.

As in England in the late nineteenth century, individual idiosyncrasies were much cultivated, and architects tended to specialize in particular types of buildings. Kaufmann, for example, had a very personal Neo-Rococo manner, delicate and frivolous, that he employed with real appropriateness in various Berlin theatres, notably the remodelling of the Kroll Oper and the Komödie, both carried out in 1924. But Behrens remains on the whole the most interesting and accomplished architect of this generation, whose opportunities for building were often to be even greater under the Weimar Republic in the early twenties than they had been under the Kaiser.

No very great change is observable in Behrens’s work after the First World War. The terrace-houses that he built in 1918 for A.E.G. workers at Hennigsdorf, and the semi-detached dwellings of a low-cost housing estate for which he was responsible at Othmarschen near Altona in 1920 are simple and solid in construction, quite like those of before the war but more conservative in design. However, at this point comes a characteristic, though brief, change of phase that illustrates his ready response to influences from the new painting and sculpture of the day. In the big complex erected for the I. G. Farben Company in 1920-4 at Höchst Behrens gave up the direct expression of new industrial building methods characteristic of his A.E.G. factories of 1909-11. The exterior was massive and almost medievalizing, even though the ranges of arches were of the unconventional parabolic form that seems to have appealed especially to Expressionist taste. In the tall glass-roofed court inside the angular forms of Expressionism were most strikingly evident; but he also introduced wholly abstract wall paintings and a few rather Constructivist lighting fixtures elsewhere in the reception rooms and offices. The result was, to say the least, ambiguous and incoherent, although the exterior was not unimpressive in its general effect.

Expressionist influence had first appeared a little earlier than this in the work of other German architects, but it reached a peak in these years of the early twenties. In his pre-war industrial work Hans Poelzig (1869-1936) was not yet Expressionist. The chemical works that he built at Luban near Posen in 1911-12 rivalled in size and even in directness of expression—though not in distinction—Behrens’s factories for the A.E.G. After the war, however, Poelzig became a principal exponent of Expressionism in architecture. One of the earliest and most striking examples of Expressionist design on a large scale was his remodelling of the Grosses Schauspielhaus in Berlin in 1919. Here the cavernous, stalactite-ceilinged interior round the central circular stage was itself like an Expressionist stage-set and the planning implied a major revolution in dramatic presentation that never, in fact, quite came off. Yet his industrial work of the early twenties soon became much more straightforward again, and he later reverted to something very comparable to the stripped monumentality of Behrens’s Düsseldorf and Hanover office buildings. The most prominent extant example of this is the enormous I.G. Farben Company headquarters that he built in 1930 in Frankfort.

One can hardly leave the subject of Expressionism in German architecture, largely confined though its more extreme manifestations were to a very short post-war period of three or four years, without mentioning two more names, those of Fritz Höger (1877-1949) and Dominikus Böhm (1880-1955).

The twenties saw a few skyscrapers erected in Germany, none of them of the great height then current in America, but sometimes as conspicuous above the existing skyline as the first skyscrapers in New York had been in the seventies. The largest, though not the tallest, and certainly the most impressive was the Chilehaus, built by Höger in Hamburg in 1923, with its Schumacher-like piers of patterned brickwork and its upper three storeys receding behind narrow terraces (Plate 153A). A large and irregular site encouraged the employment of a long double curve on the right-hand side of the hollow block, and the sharp angle at that end produced automatically a silhouette of the shrillest Expressionist order. Actually, however, Höger like other German architects was already returning by this time from earlier and wilder Expressionist adventures. To what extent he was aware of the skyscrapers of Sullivan is uncertain. The emphatically vertical scheme of design he used here, with arches linking the brick piers together below slab cornices, certainly suggests some knowledge of them, even though they were by this time all but forgotten in America.

Considerably taller than the Chilehaus, but not otherwise very distinguished, were two other German skyscrapers of the twenties. Kreis’s Wilhelm Marx Haus of 1924 in Düsseldorf, a thirteen-storey tower crowned with curious openwork tracery of inter-laced brick, is still a conspicuous feature of the local skyline; but the Planetarium and associated buildings that he erected at the Gesolei there two years later are better examples of the fairly restrained mode that he and others usually employed in these years. The plainer and better proportioned seventeen-storey Hochhaus am Hansaring in Cologne was built in 1925 by Jacob Koerfer (b. 1875).

Although only a few skyscrapers actually rose in European cities in the twenties, the theme nevertheless fascinated the younger architects. Many bold designs for them were projected, some of them of real significance for later developments in both the Old World and the New (see Chapter 22). The international competition for the Chicago Tribune Tower held in 1922, which many Europeans entered and the Finn Eliel Saarinen all but won, signally focused attention on a type of building hitherto considered unsuitable for the Old World, and generally accepted in Europe only in the 1950s (see Chapters 21 and 25).

The churches of Böhm, all of them Catholic, have a suavity that Höger’s work lacks, but at least equal forcefulness. The Suabian War Memorial Church of 1923 at Neu-Ulm is like an imaginative film-set of the period, being a sort of free fantasia on Gothic themes with little feeling of structural reality. But the boldest of Böhm’s churches, that he built at Bischofsheim in 1926, seems almost to take off from the engineer Freyssinet’s hangars at Orly. The paraboloid forms are here very frankly used; yet the concrete ‘barrel’ vault of the nave, intersected by lower cross-vaults over the bays of the aisles, creates a strong emotional effect that is both Gothic and Expressionist in tone. The finest of his churches, however, may be Sankt Engelbert at Cologne-Riehl of 1931-3. This is circular in plan and very ingeniously roofed, not with a dome,[426] but with lobes of paraboloid barrel-vaulting.

However, in a church built in 1929, Sankt Josef at Hindenburg in Upper Silesia, Böhm had already turned away from the emotionalism of his earlier work towards simple rectangular forms.[427] This simplicity he has maintained in his post-war churches, with the result that his last work, Maria Königin,[428] built at Marienburg outside Cologne in 1954, with its squarish plan, very slender metal supports, and side wall of glass, has very little churchly flavour left. Yet some of Böhm’s very late projects indicated that many of his ambitions of thirty years ago still remained with him to the end; they may well some day find effective expression at the hands of his son or of Rudolf Schwarz now that a more emotional approach to church-design has been revived internationally.

Compared to such a French church of the twenties as Perret’s Notre-Dame at Le Raincy or such a Swiss church as Moser’s Sankt Antonius in Basel, both using concrete in the rectangular and skeletal mode usually preferred at that time, Böhm’s churches of the twenties once seemed semi-traditional rather than modern. One can now see, however, that there is a different and more emotive line of development in modern church architecture to which, for example, Gaudí’s unfinished churches at San Coloma and Barcelona belong, as do also such later Latin American examples in ferro-concrete as the Purísima at Monterrey in Mexico by Enrique de la Mora (b. 1907) of 1939-47, São Francisco at Pampulha in Brazil, built by Oscar Niemeyer (b. 1901) in 1943, Nuestra Señora de los Milagros in Mexico City by Felix Candela (b. 1910), completed in 1955, and several completed in the mid fifties by Juvenal Moya at Bogotá in Colombia[429] (see Chapters 23 and 25). Expressionism may have been less of a cul de sac than its brief impingement on Behrens might lead one to suppose. Certainly it was a potent force for a few years after the First World War, and played then a significant role in breaking down the rule of ‘tasteful’ traditionalism inherited from the preceding decade.[430]

As the twenties progressed, however, and extreme Expressionist influence generally receded, Behrens gave evidence of his awareness of the quite different direction that modern architecture had just taken in the hands of certain younger men, several of whom had actually been his own pupils or at least his employees. In 1925-6 he built New Ways, a house in Northampton, England, for S. J. Bassett-Lowke, earlier a client of Mackintosh’s. With its smooth white stucco walls, horizontally grouped windows, and flat roof, this is of considerable historical interest, although of very little intrinsic merit.[431] No such advanced work had yet been done in England by local architects, and at this time only a very few houses of a comparably advanced character had been executed anywhere (see Chapter 22).

Despite his unusual openness of mind, which led Behrens in his fifties to attempt to rival juniors barely started on their careers—or, quite as probably, because of the lack of strong personal conviction of which this gives evidence—Behrens did not, like Perret and Wright in later life, continue to be very creative beyond this date. In Vienna, where he was called in the mid twenties to be professor of architecture at the Akademie, he settled into a sort of compromise mode. The low-cost housing blocks that he built in Vienna in 1924-5 on the Margaretengürtel, in the Stromstrasse, and in the Konstanziastrasse illustrate his characteristic uncertainty of direction in these years. If considerably sounder, they are also much less adventurous than the Bassett-Lowke house designed at almost the same time. This can be seen still more clearly at the Weissenhof in Stuttgart where many of the buildings of the German Werkbund’s housing exhibition held in 1927 remain in use today. There Behrens’s block of flats stands very near one designed by the director of this exhibition, his former assistant Mies van der Rohe (Plate 162B), and not far from houses by such other leaders of the new generation as Gropius, Le Corbusier—who had both worked in his office also—and Oud (see Chapter 22). The contrast between his massive block and their light and open structures is the more striking because Behrens here so evidently set out to meet his juniors more than half-way.

Behrens’s very latest work, the factory for the Austrian Tobacco Administration at Linz built in 1930 in association with Alexander Popp (b. 1891), was rather less conservative because of the nature of the commission. It is less mechanistic than the industrial work done so much earlier for the A.E.G., yet nonetheless impressive for its consistency of treatment and also for its human scale. The Linz factory provides a not unworthy concluding note to Behrens’s ambiguous career.

The vast productivity of the German architects of Behrens’s generation, both before and after the First World War, building in a boom which only came to a close around 1930 with the world-wide depression, makes it difficult to choose specific examples worth the emphasis of even brief mention. The situation is made no easier by the considerable versatility of most of the leading figures. Those few buildings that have been specifically mentioned—even most of Behrens’s own work except for his A.E.G. factories—should be considered typical of the upper level of German achievement in these decades rather than monuments of unique distinction like the best things done by Perret and by Wright in the same decades. Yet, it is worth noting, for a long time neither Wright nor Perret had much effect on the general scene in their own countries, for all the seminal effect of their influence on younger architects everywhere; while the Germans achieved a tremendous volume of what can be called ‘half-modern’ work that notably changed the whole character of several large cities. Thus the way was prepared for a very early and widespread acceptance of the next stage of modern architecture, an acceptance so premature that it induced in the thirties a sharp reaction.

In 1933 a regime rose to power in Germany with doctrinaire objections to the latest phase of modern architecture, ironically castigated as Kultur-Bolschevismus immediately after the Bolsheviks had rejected it as unacceptably bourgeois! As a result, the leaders of the younger generation almost all emigrated (see Chapter 23); while with few exceptions those German architects who remained at home turned backwards in their tracks, though not very far backwards. Most German production in the Nazi period is all but indistinguishable, indeed, from what was considered most advanced before the First World War and even for some years thereafter. Very little of it deserves specific mention. As was the case around 1910, the more nearly the structures were of an engineering order—as for instance Bonatz’s bridges for the Autobahn built over the years 1935-41—the less they were likely to be stylized along the heavy near-Classical or semi-medieval lines the later Imperial period had established as conventional a generation before. Even the housing that Bonatz built after the War in 1945-6 at Ankara in Turkey and his Opera House there of 1947-8 are hardly as advanced as his Zeppelinbau office building of 1929-31 opposite the station in Stuttgart. Like Behrens at the same time, he had attempted there—with a certain amount of real success—to follow the ascetic principles of the younger generation that had just been so well illustrated at Stuttgart in the Werkbund Exhibition of 1927 on the Weissenhof (see Chapter 22).

Immediately after the Second World War there was for several years some continuing use of the modes of 1910, so to call them. This was natural because of the prolonged absence of most of the leaders of the intervening generation from the country—Gropius, Mies, and Mendelsohn never returned—and the renewed activity of so many of the older generation who had made their reputation in the period 1905-25 with which this chapter has chiefly dealt. Today it is as if Germany had lived through the stylistic developments of the twenties a second time, and now the newer sort of architecture is once again as ubiquitous there as it was in 1930.

These tidal waves of changing taste in Germany, each representing a sharp reaction against its predecessor, make difficult such a focusing of attention on a few creative and insurgent figures as gives dramatic pungency to the history of these decades in America and France. Jugendstil, Expressionismus, Neue Sachlichkeit,[432] these general movements, more than even so distinguished an individual as Behrens, are the real protagonists of the German story from 1900 to 1933; but in the international frame of reference they must be subordinated to the broader currents that dominated the architecture of the western world in the period. In that frame of reference the contribution of a few Austrians more than equalled that of the more prolific Germans, down at least to the First World War.

------------------------------------------------------------------------