7. The American Prefab as a Crisis Response

The Depression: Housing the “Other Half”

In 1932, Kenneth Kingsley Stowell, in an article entitled “Housing the Other Half,”1 sought to draw the architectural profession’s attention to the critical dimensions of the housing problem then being experienced in the United States. This was at the climax of the depression, when millions were jobless, and housing production had declined precipitately by 84 percent from the 1922–28 average.2 To cope with this crisis, Stowell advocated, among other steps, the radical reorganization of the building industry, in order to exploit the potential of industrially produced housing; the technical problems involved, he believed, were on the way to being solved by the pioneer prefabricators. Stowell here highlights the two main themes which are constantly to be reiterated by protagonists of the factory-made house: the theme of the housing crisis and the potential solution of that crisis through the prefabrication of dwellings. Now we have pointed to the fact that the history of the development of prefabrication has very largely been a story of crisis and response, and it is true that the advances in prefabrication in Germany in the 1920s, in which Gropius and Wachsmann had been involved, were similarly made in response to the social and economic stresses of that era. But the problems they were addressing, with Hirsch and Christoph and Unmack, were not the basic housing problems of the working class, or—despite the “back-to-the-land” movement—of the unemployed. The prefabrication of the one-family house in Germany, although it was seen as a possible solution to certain middle-class housing problems in an age of eroding bourgeois standards, was nevertheless marginal, in quantitative terms, to the housing crisis of the poor. Gropius of course fully realized this. The proper response to the challenge of housing the urban poor, that is, of responding to the housing crisis in its most urgent sense, was seen in the social housing of the Siedlungen in such cities as Berlin and Frankfurt. These mass-housing projects were large-scale and comprehensively planned urban areas, usually on open land on the periphery of the larger cities. They involved a synthesis of advanced concepts of architecture and urban and social planning. In construction they were technically sophisticated and generated their own significant experiments in industrialization. However, although some of these, such as Gropius’ rationalization of the building process at Toerten-Dessau or Ernst May’s ambitious program of standardization and prefabrication at Frankfurt, advanced the cause of industrialized building, they were exceptions and not the norm. The considerable achievements of German social housing, indeed of European housing in general, were based not on total systems of prefabrication but rather on highly mechanized and efficient uses of conventional building methods, together with the extensive application of factory-made components.3 We have the anomalous situation therefore of the technologically most advanced systems in Germany, the fully prefabricated systems in steel, timber, and copper, being directed to what was, in that social environment, a retrogressive market, while the socially radical housing of the Siedlungen was built largely by the traditional (but efficiently organized) building industry.

Now, in the United States, the concept of social housing was not conceived in these radical terms. Except for a few notable, and relatively small-scale experiments,4 there was to be no American experience in the 1930s to parallel that of Germany. This we say notwithstanding the impetus to social housing given by the programs of the Reconstruction Finance Agency and the Housing Division of the Public Works Administration, in the early years of the decade, as a response to the crisis of the depression. Nor did the admiration for German models (and especially for the work of Gropius and May) expressed by those architects5 involved in some of the pioneer efforts in the United States generate a wide movement to emulate the European achievement. Catherine Bauer summed up the housing situation in 1934, in these depressing terms:

Throughout the country, then, there are not more than twenty thousand dwellings erected since the war on a permanently nonspeculative basis, and with any pretensions to large-scale planning or fundamental change in the quality of house production and neighbourhood environment. Twenty thousand to set against 4,500,000 in a section of Europe with only slightly more population than that of the United States. Moreover, not more than half of the twenty thousand really achieve a degree of permanent amenity and freedom from congestion which is the minimum working standard for “modern housing” in Europe. And of the remaining ten thousand few or none were available to the lower-paid half of the population who need the houses the most.6

Housing the “other half”7 was clearly not being achieved, least of all in terms of the European notion of comprehensive planning on a mass scale. The European concept of social housing, so alien to American perceptions and expectations, was thus not seriously proposed as a viable alternative, neither by the administrators—except on a petty scale—nor by the architects and builders responsible for solving the housing problem. Mass housing was perceived in the negative sense of “tenement building”; and the true dimensions of the housing problem were not adequately sensed. On the other hand, although the crisis generated no radical proposals in a social sense, it served sharply to focus attention on the accelerated production of the normal housing product, the one-family dwelling. Technical ingenuity, in conception and manufacture, became a preferred direction of development. Prefabrication seemed an answer to the problem, once it was conceived in technological rather than social terms, and as it was advocated in the early 1930s, in the United States, this was seen as the industrial production of the individual house as a product—to a greater or lesser degree of technical sophistication—analogous to the automobile. This, after all, was the universally accepted paradigm of the industrial process, a success story of ever-increasing production at ever-lower costs. The car was the obvious model for such advanced concepts as Buckminster Fuller’s Dymaxion house of 1927, but it was also the underlying, if not apparent, presupposition of a large number of less radical, technically more conservative, systems of prefabrication, with which the pioneers of the twenties and thirties experimented. With the individual house the preferred solution to the need for a dwelling, the prefabrication of dwelling units seemed not only appropriate, but standing a real chance—as it had not done in Germany—of making a significant contribution to the housing problem. Mass production, rather than mass housing, seemed to be the answer to a desperate need. There was one basic if untested assumption held in common by all advocates of this housing strategy: the factory-made house would significantly lower housing costs. But the economic justification, however essential to the viability of the concept of prefabrication, was perhaps not its prime motivation. In the American context, this was, somewhat surprisingly, philosophical. In discussing the use of technology to “mass-produce tracts of one-family houses,” Catherine Bauer claimed in retrospect, “we chose individualism rather than collectivism.”8 In the light of this philosophy it is not surprising that the initial impetus to the development of prefabrication in the United States came not from the constructors of large-scale buildings but from those sectors of the building industry much more directly concerned with constructing and marketing the individual house, or supplying the components and materials necessary for its manufacture. The role of big industry—the steel companies and the large manufacturing companies—was, as we shall see, indirect, ambiguous, and in the long run indecisive, in developing and supporting the movement for the prefabrication of the dwelling. Even the most ambitious proposals thought only of the production of the one-family dwelling or its components.

Growth of an Industry

By the early 1930s, within the parameters of this approach, there was some considerable ferment of ideas in the technological field, resulting in a fair amount of practical exploration. Spurred by the depression and its “confluence of factors, economic, social, and technical,” this activity led to prefabrication becoming, in Burnham Kelly’s phrase, “a widely recognized movement,” gaining the interest of many beyond “the handful of inventors and small companies which had previously been concerned.”9 The dimensions of this new movement are indicated by the fact that, by 1935, some 33 prefabricated systems were being offered on the market.10 It was, however, nearly a decade behind Europe, and on a much more tentative scale to begin with.

Although the conditions prevailing after the first world war had helped to stimulate the development of prefabrication first in Britain and eventually in Germany, that war had little direct effect on housing in the United States. The traditional building industry was flexible, and could expand its resources to deal adequately with all housing needs, even the accumulated shortage generated by the war. By 1921 construction activity in the United States was well above the prewar index, and by its peak year, 1925, it was 2.6 times as great.11 The demand for less substantial houses—temporary dwellings, farm houses, housing for oil-field workers—was catered for not only by the traditional construction industry but by several long- and well-established manufacturing firms producing precut and sectional houses.12 Involved in this business were mail-order houses like Montgomery Ward, and especially Sears Roebuck, which, through vigorous promotion and a determined sales policy, coupled with liberal financing and loan facilities, managed also to invade the traditional housing market. Sears’ “Modern Homes” department flourished with considerable sales of precut homes “packages” in the boom years of the midtwenties, only to decline dramatically and ultimately collapse, in the general disaster following the crash of twenty-nine.13

Precut and sectional houses helped introduce industrialized methods into house construction but were not prefabrication in the full meaning of the term. Neither could construction in precast concrete be so considered, although it could reduce the site-labor content, sometimes significantly. There had been a continuing interest in precast concrete systems, since the days of the pioneers: since Ransome produced his “Unit System,” utilized mainly for industrial buildings, or Conzelman, with a system similarly named, developed between 1910 and 1916, constructed a group of 300 houses of reinforced concrete precast elements at the Industrial Housing Colony in Youngstown, Ohio.14 That this was a concept acceptable to constructors was noted in Engineering News in 1916: “There is nothing new about the so-called Unit System of concrete construction. It is of interest to record that this system of construction appears to be coming into favor among engineers.”15 Large-scale, story-height precast panels, such as Grosvenor Atterbury had used for the Russell Sage Foundation at Long Island, or May at Stuttgart, were the logical form in which to prefabricate such elements, but they demanded the use of cranes and other mechanical hoisting devices, such as could be economically deployed only for extensive housing schemes on the European model. For the dispersed American market, with its emphasis on the individual house, this was clearly not appropriate. Many ingenious systems were designed employing smaller, lighter elements,16 but these inevitably increased the amount of site work, especially in the placing of components and their jointing, usually by grouting. In addition in most precast concrete systems there was considerable on-site labor involved with internal partitions and finishes.

The movement for full prefabrication in the United States was reinforced in some measure through the European experience in the mid-twenties. The experiments carried out in Britain and Germany with steel housing, with its apparent use of high technology, and a sophisticated employment of materials superficially analogous to those used in the making of automobiles, attracted the interest of certain sectors of the American steel industry. A series of articles in the journal of the industry, Iron Age,17 during 1926, drew attention to the British experience with the Weir, Telford, and Atholl steel houses, argued the technical and economic advantages of steel construction, and foresaw the potential for industrialization: “The development of standard, unit-type, interchangeable metal parts adaptable to any ordinary house plan will permit the use of large scale factory production.”18 The author’s views on prefabrication were close to those of Gropius, when he went on to advocate “the development of standardized units for construction, rather than . . . uniform design or duplicate dwellings.”19 And we may note the congruence between the American philosophy of the individual, and Gropius’ emphasis on the need for standardization and diversity. The unit steel house was conceptually attractive to the American, both technically and ideologically, and when the U.S. Department of Commerce issued its report on the production of steel houses in Germany—a report to which we have previously drawn attention—Iron Age noted “considerable interest” on the part of domestic steel producers.20

One development in particular that drew the attention of the metal industry, but evoked no immediate response in the architectural journals,21 was the copper houses produced by Hirsch. Data on these houses have been collected by the Copper and Brass Research Association of New York, and descriptions of the prototype house, and many illustrations of it during erection and after completion, were published in the trade journals Iron Age, and the Copper and Brass Bulletin.22 Attention was drawn, in both articles, to the advances made over the past decade in the design of what was called “knockdown” houses and pointed to Europe’s lead over the United States in the design of substantial and permanent dwellings, that could be mass produced. That these could be expected to penetrate the American market was evident in the granting of an application for United States patents to Frigyes Förster. The message to Americans concerned with prefabrication was clear: metal construction was the technology of the future, and they were lagging far behind Europe in its development. Steel framing for houses of course had a long history in the United States, and the use of steel as a replacement for timber framing and studs goes back at least to a venerable example reported in Iron Age of 1907.23 In fact Germany itself looked with some admiration to the American experience in this field, as we have already noted, and Spiegel, in his Stahlhausbau, devoted an entire chapter to the study of the framed structures pioneered by Robert Tappan, Weltcrete, Broderick and McKay,24 generally in the mid-twenties.

Buckminster Fuller, in 1927, designed his revolutionary Dymaxion house, a lightweight metal structure suspended from a mast, which he exhibited in model form.25 In its exploitation of advanced technology, and its departure from the constraints of traditional house forms, it was conceptually of an entirely new order, as compared with its European and contemporaries. In its original version it was obviously not ready for translation into practice, but later, development brought it closer to the stage of production. However, despite industrial backing,26 optimistic prognoses in the press,27 and Fuller’s dedication and fervent advocacy, it never really got going: the Dymaxion house remained no more than a tantalizing promise, a highly innovative prototype of what an industrially produced metal house might be.28

It was only in 1932 that the first experimental steel houses were produced in the United States, using steel not only for framing but as the enclosing membrane in a total integrated building system, on the British and German models. “In recent years,” it was pointed out in the Architectural Record, “vitreous enameled sheet steel has been used for the exterior surface of a number of residences in Europe. In the United States, however, the year 1932 appears to have seen the first attempt to utilize enameled sheet steel for this purpose. . . . The first house so covered was designed by Charles Bacon Rowley and Associates in Cleveland, Ohio.”29 There is a nice historical irony here. In 1932, as American architects and industrialists moved to produce steel houses, inspired largely by German models, Germany itself ceased their manufacture, to all intents and purposes. At that showcase of advanced prefabrication methods, the Wachsende Haus exhibition of 1932 in Berlin, there was not a single system utilizing metal-faced panels, other than the Hirsch houses designed by Gropius: and Hirsch themselves went out of business in 1932. Even Böhler, a pioneer of steel systems, had by 1932 moved to steel frame construction, with lightweight block or asbestos infill. A strange cycle of events was being played out. When, in 1926, prompted by over-capacity of production in a time of reduced demand, the German steel industry had decided to enter the steel house business, they were in part motivated by the British Weir house; they had failed, however, to study those problems to which the Weir house was about to succumb. Now, in the depths of the depression, American industry of diverse kinds, was similarly motivated to increase its sadly reduced production by entering the vast, unsatisfied housing market. And they too turned to the steel house as their paradigm, its virtues having been much more widely publicized than its faults. As Santayana once said, He who does not learn from history is doomed to repeat it.

By 1932 a frameless steel house using Armco steel panels30 had been erected, on lines very similar to those of the pioneer house of Rowley, by the American Rolling Mills, “the first big industrial concern to eye the new market with serious intent.”31 In the same year two major firms were founded: Howard T. Fisher’s General Houses Inc.,32 which produced an all-steel house assembled out of components made by a large number of contributing firms, but undertaking no manufacturing on its own account; and architects Holden and McLaughlin’s American Houses, Inc.,33 which manufactured a steel-framed, asbestos-clad product. These two corporations were, for a short time, managed by a holding company, Houses, Inc., formed by Foster Gunnison, in collaboration with the American Radiator Co. (a prime backer of the important research institute for prefabrication, the Pierce Foundation) and the General Electric Company, until Gunnison set up his own firm to manufacture stressed skin plywood houses, in 1935.34 These experiments with steel houses were described in the Architectural Forum with interest, but not with undue commitment or faith in their future: “The makers of prefabricated steel houses have gone ahead preparing a product which the public certainly does not want right now, may not want for many years.”35 Despite this professional caution, and “the competition of subdividers and the deeply entrenched likes and dislikes of the American Public,”36 the newspapers seized upon the new development and, through widespread publicity, generated much interest, and perhaps an inflation of anticipation for what the new industry was likely to achieve. When General Houses was incorporated, it was probably reasonable—after all, the name of the company was a conscious analogy of General Motors—for the New York Times to announce “General Houses, Inc. Formed to Market Ready-Made Steel Homes Like Automobiles” (although the heading “‘Fordized’ Housing Plan of New Group” stressed the wrong make of carl). However, it was certainly over-optimistic to write of “Mass Production of Homes in View,” and to suggest that prefabrication could lift the country back to prosperity, as did the Times in 1933.37 However, by 1933, the mood of the country was slowly moving back to optimism: in March 1933 Franklin D. Roosevelt was inaugurated as president, and he proceeded vigorously to heal the economy, and restore confidence in the nation.

The Century of Progress Exposition held in Chicago in 1933 took place at a turning point in American history. The exhibition could therefore not only look back with pride to the achievements of the past but forward to the future with a confidence, shaken it is true by the depression but restored to a degree by the groundswell of hope kindled by Roosevelt’s election. At the exhibition about fifteen modern houses were on show. Of these, several steel houses, by General Houses, Armco, and Stran-Steel, and Architect John Moore’s wooden house of bolted panel construction, together with the exotic steel-frame and phenoloid board “House of Tomorrow” by George Fred Keck, were “the first of the prefabricated or mass-production types which the public has had the opportunity of investigating.”38 The architectural profession at large also began to learn about the new systems, both directly through the exhibition, or, in their offices, through the widespread and detailed coverage given to prefabrication in the professional journals.39 Public and professional interest helped to give the expanding industry a firmer base; and it was also underpinned, in a technical sense, by the active programs of research into new methods being undertaken by the Pierce and Bemis Foundations, the Forests Products research laboratory, and Purdue University.40 Also to be mentioned was the innovative program41 of the Tennessee Valley Authority—part of the Roosevelt New Deal—which provided an early spark for what probably was the most successful (from a quantitative point of view) branch of the prefabrication industry, the mobile home. During 1935 and 1936 several new firms of prefabricators were established, and it is instructive to note that by now the preponderance of interest was no longer in steel but in the more prosaic, and better understood, field of timber construction, which could be operated with less capital investment. By the time Walter Gropius came to the United States, in 1937, there was not only significant activity but a climate of interest in prefabrication which was supportive of further work in this field.

The underlying factors of the housing situation were naturally not static. Economic revival was underway in what was becoming the postdepression era. In 1937, in his second-term inaugural address, President Roosevelt made housing his own policy concern, with the poignant phrase, “I see one-third of the nation ill-housed, ill-clad, ill-nourished”; and Senator Wagner at last achieved the enactment of the United States Housing Act.42 Positive motivation, a new legislative framework, and decisive action achieved results. “By 1939, starts of new dwellings had recovered to about the 1929 level but with the help of 56,000 publicly financed units which were unheard of in 1929.”43 Despite these achievements, the housing problem, particularly of the low-income group, remained severe. Prefabrication of dwellings continued to be proposed as a possible solution.

Gropius drew attention to this in an interview published in the Boston Herald,44 “There is something wrong if nothing on the housing market can be delivered to the ordinary man,” he is quoted as saying. After referring to the high costs of land, he went on: “Mechanization of industry has brought down the costs of manufactured goods, but houses are still largely constructed by hand. One solution is the house made of parts constructed beforehand, the pre-fabricated house. These parts would come out of mass production—and there is no limit on the size of the building into which they would go. Of course, there are more than 50 American manufacturers dealing in prefabricated houses, but their enthusiasm has been slowed down because of the financial market and problems of land costs.” He then concluded optimistically: “Changes for the better will come. Remember, food and clothing prices have been adapted to the needs of the ordinary man, but not rentals. Some day they will be.”

Gropius here compares housing with other commodities, food and clothing. The more usual analogy, which Gropius himself had cited many times and refers to again in the interview, was the automobile. Gropius saw the automobile as an example of industry’s potential, but not as a model, for he was interested in producing components rather than the house as a complete product.

As Gropius was painting this somewhat hopeful picture of industrial effectiveness, the shadows lengthened on the international scene. By the end of the year Europe was once more at war, and although the United States remained, as yet, outside the conflict, the effect on its economic situation, and especially on its housing needs, was immediately felt. “The transition to a wartime economy in the United States began in 1940, even though under the pseudonym of ‘defense’ . . . . On June 28, 1940, the United States Housing Act of 1937 was amended to authorize the use of its loan and subsidy provisions for housing defense workers during the emergency . . . . On July 21, 1940, an Office of Defense Housing Co-ordinator was established in the Council of National Defense to plan and carry out defense housing programs. On September 9, 1940, $100 million was appropriated for the erection of defense housing by the War and Navy Departments. On October 14, 1940, the so-called Lanham Act, the basic defense housing law, was passed, involving direct Federal financing and construction.”45

Walter Gropius and Martin Wagner

In this rapidly evolving new situation Walter Gropius played only a limited part. He continued to speak out for prefabrication, presenting the arguments, both pro and con, rationally. He spoke, not as a crusader for a cause but pragmatically. His measured terms of advocacy derived no doubt from his first-hand experience, in the Hirsch saga, with the enormous difficulties that stood in the way of realizing the dream of prefabrication. This caution provoked his old friend of Berlin days, fellow émigré and present Harvard colleague, Martin Wagner, to take him to task for his negative attitude, for stressing “the vast amount of problems concerned” rather than the great opportunities offered by prefabrication.46 Wagner had an impressive record, in the city planning office of Berlin, of housing the masses. In 1931–32 he had demonstrated his interest in prefabrication by the guiding hand he gave to the organization of the Wachsende Haus competition and exhibition, and his book on the subject showed a firm grasp of the social and economic questions, as well as the technological aspects involved.47 But his views on prefabrication were untempered by the flame of experience, especially in the new socioeconomic environment in which he now found himself transplanted, and they remained essentially academic.

Wagner’s theoretical position on prefabrication remained ideologically consistent. He spelled it out in the manuscript of a proposed book, Prefabricated Housing,48 on which he had been working for two years, since 1939. It owes much to the introductory chapters which he wrote for Das Wachsende Haus. His case for prefabrication grew out of a continuing and broad-based study of the housing problem (once again acute, because of the war situation). It is probably the most comprehensive study by an architect of its day, dealing with the socioeconomic dimensions of the housing question; the need for a new approach to construction management; and the technology of structure, enclosure, and environmental control. It is based on a clearly articulated thesis: that the proper module of industrialized housing, from both the sociological and technical points of view, is not the whole dwelling (vide Fuller), nor the component panel (vide Gropius or Bartning),49 but the individual room. These rooms he sees as standardized cells linked together in free combinations to form an infinite variety of house plans. This inherent variability is an essential aspect of his philosophy of flexibility in house design, for which he makes an impressive case. That he argues convincingly for the “growing and shrinking house” comes as no surprise, when we recall his key role in the episode of the Wachsende Haus in Berlin. However, he is equally effective in dealing with other facets of dwelling flexibility, such as mobility and adaptability in the face of changing site and programmatic demands. After the breadth of this analytic presentation, his own solution to the problem (the “igloo house”)50 is idiosyncratic, but it should not be allowed to detract from the importance of his contribution as a whole.

Gropius and Wagner had frequently been associated in the past in various ventures: architectural competitions, housing schemes in Berlin, the inner circle of Der Ring. There was mutual respect, because between the two men’s philosophy—despite the temperamental friction—there existed a basic congruence. Consequently for a while they were able to continue to work together and produced several studies in association.51 Perhaps the most important outcome of this collaboration was a joint paper, “How to Bring Forth an Ideal Solution of the Defense Housing Problem?” which was presented as evidence to the House Select Committee investigating National Migration, of the 77th Congress, in 1941.52 In this paper they advocated a program of prefabricated housing within the context of a long-term regional plan of decentralization. They also stressed the need for continuing governmental support of systematic research. Their specific recommendation for prefabrication reads as follows:

To create a new type of low-cost dwelling of high quality, with up-to-date amenities and composed of standardized parts which should be interchangeable for use in different types of houses of varying sizes. These dwellings should be demountable for reerection but simultaneously to be qualified for permanent use when desired.53

We notice that the generic phrase “standardized parts” leaves open the disputed question, whether they are Wagner’s room units or the panel components favored by Gropius.

In the published version of the proceedings of the Select Committee, the following item is noted in relation to the Lanham Act (passed to expedite the provision of housing in connection with national defense, and for other purposes): “At the time this hearing goes to press, a bill authorizing a further appropriation of $300,000,000 is proposed, making a total of $600,000,000 which will be available under the provisions of the act . . . ,”54 The effect of such massive underwriting of the housing program upon the prefabrication industry was noted in the Architectural Forum. “THE FORUM in December 1940 reasoned that ‘national defense may do for Prefabrication what World War I did for the aircraft industry—raise it from infancy to adolescence in no time.’ Since then Prefabrication has grown in stature, has grown from a group of experiments into a fledgling industry-within-industry which has been assigned some 17,500 houses for the Government defense program.”55 Gropius and Wagner, in the American context, were ideologues of prefabrication, and not practitioners (although Wagner’s igloo house brought him into the realm of the “experimenter”). A month after this exuberant notice was published, Konrad Wachsmann arrived in the United States. With his arrival, he and Gropius took their first steps (with which Wagner was not associated) to enter the fledgling industry of prefabricators.