Elementhus: Sweden’s ultra modern & totally forgotten, awesome, factory-built house
There was a time in the early twentieth century when a group of progressive architects dreamt of changing the world by revolutionizing the ordinary house. A central theme of modern architecture was the belief that houses should be made in factories, just like cars. The world’s greatest names in architecture produced many theoretical designs and built many prototypes, which are faithfully recorded in architectural history books. But the truth is that none ever went into production. In commercial and industrial terms, they were failures
— Colin Davies 2005
History has ignored one of the most interesting attempts ever made to move the production of houses into a factory. This is the story of what happened, but also an example for how architecture and industry might better relate in order to solve for productivity in construction.
In his 2005 book, “The Prefabricated Home”, Colin Davies makes the case that there is a tendency to celebrate the serial failure of well-known architects who have attempted to design buildings for factory production, but failed to realize commercial success.
While Davies is right about the pattern, Lennart Bergvall and Eric Dahlberg, working in Sweden, at exactly the same time as Gropius and Wachsmann were attempting the same thing in California, invented and patented a housing system and built one of the most technically impressive and expensive factories in history. It began full-scale production production in 1952, as “AB Elementhus” in the town of Mockfjärd, in the central part of Dalarna county.
Over the next 25 years, this novel system, which used wood in sophisticated, elegantly engineered and economical ways, operated as a completely vertically integrated, highly automated, house factory; producing about 2.75 houses per day, for an estimated total of 18,000 houses, many of which are still inhabited and make up part of the architectural heritage found mostly in working class Swedish towns. These homes cost less to build and were more economical to operate than what existed before in Sweden.
They were then ignored by their profession, and forgotten by history
While the names, Bergvall and Dalhberg, haven’t been included in any history of modernism, their Elementhus enterprise was by no means obscure, as it was financed by some of Sweden’s most famous industrial firms and stands still as one of the most ambitious attempts to industrialize the production of houses. They are perhaps the most accomplished modernist architects of the mid century that no one has ever heard of … this is the first time their story has been told in English.
Not merely a historical oversight, a professional blind spot
The fact that students today are not taught the example set by these two architects is not an obscure historical oversight, it reflects and illustrates a tragic lack of perception by a profession, i.e. architecture, which is of vital importance to humankind. This story should be known today, not merely to correct the historical record, but to hold up a model of practice and thinking that avoids the serial failure of architecture to release itself from a prison locked from the inside, that cultivates an inability to engage with the industrialized production of housing.
The challenge of housing productivity & industrialization is one that lies ahead, not behind us. Increasing construction productivity has recently been defined by McKinsey as at $1.6 Trillion economic opportunity , and industrial engagement identified as one of the few paths to productivity. Of course, since the 1940’s the powers of industry have only magnified, what has not changed are the habits of thinking that are required to harness industry’s potential for good — and unless architects are given better role models, they stand little chance of contributing to the future, more industrialized building industry. Today, architects drawings, are, more than ever, actual manufacturing data — but unless architects act, and think, like manufacturers, they have little hope of success.
In 2018 so far, venture capitalists have already poured more than $900 million into the industry in 16 construction technology deals, on pace to top last year’s total of 40 as a new decade-high. If architects could do one thing to prepare to be part of this future, it would be to model the way of thinking and engagement exemplified by Lennart Bergvall and Erik Dalhberg, whose use of the term “Elementhus” has entered the Swedish vocabulary for any building produced of factory built components, somewhat in the same fashion as “Kleenex” became the popular name for tissues in American English. The story of this truly remarkable building concept and the effort behind it, have never been shared outside of the Swedish language (which also has failed) to give it context or the respect it deserves as an astounding part of the story of the modern movement.
Modular coordination: an American idea, implemented in Sweden
The story of Lennart Bergvall and Erik Dahlberg begins with the the idea of “Modular Coordination” an idea the men encountered during their studies in architecture at Kungliga Tekniska Högskolan (the Royal Institute of Technology) in the early 1930’s. This meaning of this term has largely been replaced by its use today to mean “buildings made by stacking boxes” but in the mid-century it was a central part of modernist thinking toimprove housing though rationalization, standardization, and industrialization.
The idea of a modular coordinating principle, spread to other fields outside buildings, but originated in the architectural proposals of Albert Farwell Bemis, an American industrialist who was vastly wealthy and spent his last ten years advocating and studying the problem of improving housing, publishing his “cubical modular method” as a patent in the 1930’s as the last volume in “The Evolving House”. The oxford dictionary cites him as originating the 20th century meaning of the term “modular”.
Bergvall and Dahlberg emerge into professional life during the initial proliferation modernist architectural idealism in Sweden. As recent graduates, Bergvall and Dahlberg were involved in various architectural offices. Bergvall worked for Albin Stark in 1936, Gunnar Asplund in 1937–1938, Gustaf Clason 1938–1939, and then was the director of the newly formed “Svensk Byggtjänst” or “Swedish Building Centre” an information clearing house cooperatively owned by members of the Swedish building industry association.
When the Swedish building industry decided to engage in research into “modular dimensional coordination” by creating “Kommittén för Byggstandardiseringen” (Construction Standardization Committee), Bergvall’s role in Byggtjänst, and his previous work analyzing the effect of modular coordination on construction, made him a natural choice to play the role of coordinator. Bergvall stipulated that Erik Dahlberg was to be his co-chair, as the two men had formed a commitment to work together, since their friendship began in their school days based on a shared promise to work on improving and modernizing the way buildings were built.
The committee’s effort was financed by a 35,000 kroner fund (worth approximately $250,000 in 2015 US dollars) put up by 18 major Swedish building materials manufacturers to create a system for dimensioning rules for the building industry and was chaired by Sven Markelius, one of the founders of Congrès internationaux d’architecture moderne (CIAM) in 1928, and an author of the book-length manifesto Acceptera!, the key Swedish text on modernism. While Markelius’ influence made him a figurehead of the architecture profession’s engagement with the Swedish building industry, Bergvall and Dahlberg were the ones doing the detail work. They were recognized as the authors of the committee’s 1946 report, “Modulutrendning”, which translates to “Modular Investigation”, which implies being of a serious and official nature in the Swedish use of the term. This fact that the effort was official and serious, is worth noting because the Swedish work was parallel with an initiative in America, published the same year, of “Project A62 Guide” on modular coordination, published by the Modular Service Association, and sponsored by the American National Standards Institute (ANSI), the American Institute of Architects (AIA), and The Producers Council.
While these two initiatives could not have been more similar in purpose or inspiration, and trace common origins in the advocacy of Bemis, it is interesting to compare their outcomes, as it illustrates the advantage that a small country like Sweden had in promulgating standards and in securing industry wide cooperation and agreement, and it also provides some useful context to understand why the Elementhus concept, based on rigid a strict modular concept was embraced by Swedish industry.
While “Modulutrendning”, has a quarter of the pages contained in the 275 page A62 Guide, it deals with a wider range of industry categories, is far more complete, and represents a final agreement by a “who’s who” of Swedish building product manufacturers. This influential group was not only voluntarily participating in the effort, but also paying for it. Theirs was no top down exercise with Government making rules, nor was it a wishful academic proposal, but a wide cross section of decision makers resolving a problem they agreed was in their interest to fix. This example nicely fits with a pattern described in the best-selling 1936 book “Sweden: The Middle Way”, that argued Sweden’s economy was a compromise between the extremes of Capitalism and Communism.
In Sweden, industry therefore quickly and decisively undertook a cooperative effort, generated agreed standards which they self funded and designed. The American effort on the other hand, may not have even happened were it not for philanthropy of the heirs of Albert Falwell Bemis, who established a charitable foundation to further Bemis’ desire to change the building industry for the better, and created and funded the evangelistic, nonprofit, Modular Service Association.
The lack of American industry’s ability to cooperate around shared standards, was confirmed a year later, in 1947, when the Bemis foundation ceased funding the Modular Service Association. Attempts to interest the Office of Technology Services in Washington to take on the work failed, the sponsor organizations, including the AIA, declined to fund further work, a move that signaled indifference by America’s architects and building material manufacturers. In a 1949 report, by Arthur D Little, paid for by the Housing and Home Finance Agency, cites a lack of industry cooperation and financial support as the reason modular coordination wasn’t likely to deliver results in lowering costs for better homes, despite a general agreement that it could.
So, while modular coordination in the building industry was an American idea, and both the Americans and Swedes agreed that the size of a unit of coordination, or module (M), should follow Bemis’ thinking, that arrived at 4 inches (100 mm in Sweden), American industry was nowhere near as willing to put the ideas into practice.
While the post war housing crunch in America was going to produce epic attempts at redesigning the house for factory production, including the infamous Lustron house, which ironically was a bootstrap vision by a Swedish American, Carl Strandlund, Sweden’s industrialists, had gotten behind a plan to back a system based on the countries forest production, Elementhus, which by 1944, had been prototyped and patented by Bergvall and Dahlberg, who has also served as the coordinators for industry wide agreement on dimensional standards.
From Bemis to AB Bostadsforskning: Inventing Elementhus
Comparing American engagement in modular coordination with those of Swedish industry, illustrates the expeditious way Swedish industry cooperated to embrace these ideas and to invest in building industry rationalization. For the legacy of Albert Farwell Bemis, the acknowledged father of “modular coordination” it was truly a case that supported a concept present in all four Gospels: “Nemo propheta in patria” — no man is a prophet in his own land, as it was large Swedish industries that ended up implementing what his own research and philanthropy could only encourage his fellow Americans to follow.
The eagerness within Sweden’s industry and architectural community to embrace and put Bemis’ ideas into practice, also helps understand the first part of the commercial phase of the Elementhus story. Concurrently with Bergvall and Dahlberg’s selection to lead the Kommittén för Byggstandardiseringen, another architect, Jöran Curman who was responsible for the housing portfolio of Uddeholm AB, a large Swedish steel conglomerate, came to them seeking a way to apply industrialized ideas to the firm’s desire to improve and expand worker housing and to support the planned growth of its steelworks. Curman arranged meetings between Bergvall and Dahlberg and Ernst Wehtje the CEO of Skånska Cement AB and Nils Danielsen, who was appointed CEO of Uddeholm AB, that same year.
What emerged from this meeting was an agreement to create a venture called AB Bostadsforskning, which translates as, “The Home Building Research Corporation” the letters “AB” stand for “Aktiebolag”, Swedish for corporation. The idea behind this company was, as its name suggests, to study how to apply industrial technology, modular coordination and rationalization to the question of how to improve housing. This was the same exact problem that progressive architects and industrialists were playing with everywhere, only this effort was being backed by Swedish industrial capitalists and Swedish architects, who were operating under the assumptions of the “middle way” thinking, responsible for Sweden’s emergence as an exceptional societal model that existed, between competing and indeed, now, warring political extremes.
It is unclear just who sold who on the venture, the architects or the industrialists, but it seems likely that there was not so big a distinction between their views of the world. Bergvall and Dahlberg were not speaking a different language from Danielson or Wehtje, either culturally or professionally, and the record shows that Bergvall and Dahlberg emerged charged with a substantial mission and an impressive commitment to oversee a five year, R&D project with an annual budget of something like 200,000 Kronor (the equivalent of something like $1.5M US when adjusted to 2015 using a GDP implicit price deflator).
Ultimately what emerged from the work of AB Bostadsforskning was the intellectual property and business strategy to build Elementhus AB, which incorporated as its own venture in 1949 — but research itself was involved a particularly Swedish way of breaking down the question of how industry could be applied to improving housing. Given that waves of enthusiasm about the potential for prefabrication to improve housing have reliably yielded some fantastic economic disasters in many different places, Sweden’s building industry has been particularly good at integrating manufacturing ideas with the building industry and today has the world’s highest fraction of what is now commonly spoken of as “offsite construction”.
It merits looking at the way Bergvall and Dahlberg thought about the problem of industrialized building, before introducing the ideas they patented, and before looking into the machine they designed for building houses, and Elementhus concept itself.
Interpreting the dream: redesigning the house
Bergvall and Dahlberg’s research program did not start with an architectural or even a structural idea. Its sponsors, steel and cement producers, were not pushing for a house solution to use their existing factories, or to consume materials they would benefit from selling. In fact, the research didn’t start with a commitment to do anything other than find a better answer to the status quo, so the first thing Bergvall and Dahlberg did was study the existing industry.
Sweden already had a noteworthy and growing prefabricated house industry, and they also had an industrial landscape that was ripe with innovation. Other countries may be more associated with the industrial revolution, but Sweden punches far above its weight, and especially so where it concerned the creative use of wood and machines. Bergvall and Dahlberg were also in contact with the entire spectrum of different building material manufacturers, and many of their ideas involved creative ways to adopt other innovation to be useful in their house system. Above all they did something that is essential to good design: clearly understanding the problems they were working to solve.
In their view, industrialization had to be applied to improve quality and productivity, but they also understood it as part of a bigger process, that all had to work together. Looking at their design solutions, one sees things that are commonplace in housing now, but were not generally used then, and especially not used in the building industry. Swedish uses a different word than English does for a “prefabricated house” the Swedish is monteringsfärdiga trähus” which means “finished assembly wood house” — a meaningfully different way to speak about the concept. Bergvall and Dahlberg were not so much trying to come up with the perfect system — as they were seeking the reasons that the process was inefficient and slow. Their analysis shows that they were viewing the problem as one of involved concepts of waste elimination and value adding, ideas that are hugely relevant today, and have come to be called “lean” thinking.
Looking at house construction, they ascertained that the factory value added content of existing Swedish wooden houses, was normally about 30% of the total construction cost, and of this, approximately 20% corresponded to the industrially value-added components, they also understood how that inefficiency or waste derived from various kinds of problems, like waiting, or not having the parts needed to keep working. They didn’t try to move the entire construction process to the factory, indeed some aspects of building, such as the house foundation, were viewed as impossible to improve upon. Also, some of the things that are common now, such as small flexible crane technology, that supports the use of large wall sized elements did not yet exist, and cranes were limited to very large installations, where they were expensive to use — it was assumed therefore that the solution would have to be packaged for shipping but then completely assembled by hand.
AB Bostadsforskning’s research contains so many interesting component and industrial design concepts that it is hard to appreciate just how innovative and well detailed the system was. It was truly a complete house, with every detail, sanitary, heat, ventilation, thermal performance, cleanliness, light, and aesthetics; as well as manufacture, delivery and construction, completely ironed out, along with patented machinery to automate the key processes, which thinking that considered all aspects of the factory’s organization, logistics and purchasing processes as part of the solution.
The research identified the following design recommendations:
a) All the walls and floors constructed from completely standardized elements which were simple to install. This yielded a system based very close to the 100 mm modular dimension and created a system that allowed for flexibly in design.
b) Thermal insulation installed inside the box-shaped elements, which were a byproduct of the wood planning needed to create the boards.
c) A prefabricated chimney which is lighter than the traditional masonry chimney
d) All hot water radiators installed by fitters and connected in series as opposed to the usual system, where each radiator has a flow line and a return line.
e) All pipes and fittings (pre-cut lengths of the module) fitted as sanitary installation in the bathroom.
f) The electrical installation pre-wired in the factory for direct installation on site
g) cabinets and wardrobes and kitchen units fully manufactured in the factory and transported as elements for on-site mounting.
h) A completed bathroom floor ready-made for direct mounting — which eliminated one of the more problem prone areas of the house.
AB Bostadsforskning estimated that the design itself reduced wood consumption in the system by about about 50% when compared to the existing way similar houses were built using the Housing Research building systems. It was only in the case of foundations, where the research felt there were not ways to improve the current system. Bostadsforskning also worked out designs for the production machines that the system required. The planned production system included a sawmill, radiator factory, roof factory, carpentry shop for cabinets, a kiln for lumber, sheet metal and plumbing shop and workshop for electrical installations.
It is not clear when the first Elementhus prototype building was created, but the basic idea of the “box beam” was based on the design of a 3 ply laminated board developed in the Limhamn parquet flooring factory — the 2nd and 3rd test houses were reportedly made in the AB Träkol Fabriker in Vansbro and set up in Uddeholm because of the interest the steelworks had in housing for expanding and modernizing their plants. If there is a record of the earliest experiments they remain hidden.
Between the initial investment in AB Bostadsforskning, by Skånska and Uddelholms in 1944, and 1948, when the Bostadsforskning system was presented to outside investors as a proposal for a new house production company known as “Elementhus”, Bergvall and Dahlberg had managed to devise a completely brand new system for manufacturing wood houses that was unlike anything that had ever been proposed, and importantly the homes were built to a higher standard than the prevailing norm, at a fraction of the price. The mandate: to lower cost, without also lowering quality, was not just a slogan for the architects. Elementhus, is a case study in how manufacturing technology could be deployed to build a better quality house at a lower price, than the prevailing norm.
Among the largest house factory investments ever made:
The investment intended to produce houses for the workers of the sponsors’ expanding companies cost 1.5 Million Swedish Kronor to plan, and 6 million kronor evenly split between buying land and developing a former sawmill in Mockfjärd, and producing the machines and technology, making it among the most automated and ambitious house factories ever created. An investment of $7.5 million Swedish Kronor in 1944 would equate to $385 million Kronor today, or approximately a $56 million US dollar sum. Elementhus, therefore was a project in league with any industrial investment in house production ever attempted, before or since.
1949 AB Elementhus: radical redesign of the house
The intellectual challenge of appreciating the modernist value of their design, is compounded by the fact that their finished product did not contain any stylistic or modernist design cues to communicate either its novelty or the way in which highly automated the factory production technology was used to make it.
The key innovation that Bergvall and Dahlberg came up with was a structural unit that could be continuously produced, it also involved invented the machine tools to accomplish this production. Either of these accomplishments alone would be impressive but to successfully design not just the house with the machines to make the houses, truly elevates their accomplishment to something rare in the history of architectural engagement with house production.
Bergvall and Dahlberg’s design was far more sophisticated and precise than either a plank or frame wall and it has never been executed in any comparable fashion since. Theirs was a hollow box beam element that was designed to be manufactured in 20 cm. x 20 cm. units that fit together for use as either as a load bearing outer wall or as structural weight bearing floor component. Internal wall elements, were built identically, but half as thick 20 cm x 10 cm. All surfaces of the dwelling were made of 20 mm cross laminated, three ply spruce panels to make them dimensionally stable to within a few hundredths of an inch. Across their thickness, they were connected by a sheet of fiberboard glued into a groove mailed into each of the laminated faces.
Each unit therefore was rigid, but made of very little material. The void formed in these box beams was filled with compressed wood shavings and sawdust which provided insulation (include u value). The material economy of this design is truly elegant and a real departure from anything done before. The design showed real mastery of material, creating an insulated wall that could be assembled quickly, by hand, and perform above the prevailing standard. What is so jolting about this design is the light weight of each element and the extreme tolerances that the factory could maintain. The shop drawings, unusual for house building, are presented with tolerances of “+/- .2 mm.
The exterior surface of the exposed panels was deeply grooved to relieve the tendency of the exposed surface to cup, or check. The appearance of this system was very similar to the traditional vertical cladding which was traditionally used on Swedish houses.
The connections between these units were accomplished by a combination of tongue and groove connection of the interior and exterior faces and dowels which were designed to be inserted into holes bored at the factory through the interior surface of the elements and the sill and header plates. Both windows and doors were designed to be installed using similar dowel connections. Corrugated cardboard strips were glued in between the box beam elements, which compressed to create airtightness between the tongue and groove connections of the interior and exterior wood faces. The system relied on both precision and simplicity to deal with the problems of tolerance and accumulated errors that were a central issue for all systems built structures.
This truly is a tour de force of using wood in clever and highly nuanced ways to create something of particularly elevated engineering. The house was essentially “friction fit” in that it relied on tolerances and dowel pins to hold itself together.
To making these structural units, wood was fed from a sawmill and kiln, continuously into a lamination process and then into a continuous box section joined together with sheets of hard fiberboard that were glued to form an endless box. The resulting box moved in a completely automatic process and was cut and end machined into specified lengths after being planned level and square. Again, the tolerances of this process were measured in tenths of millimeters. It is important to note that this was true, continuous production, which required completely new machinery, it was not a wood shop repurposed into producing building parts but a truly automatic processing line of completely novel design. The entire process from the point where the boards were fed, to the point where the box elements were cut to length was operated by means of one switch. This truly was a machine that made houses, and not a large workshop space. The rate of production was over 2 houses per day.
The continuous gluing operation was among the more remarkable aspects of this system as normally parts are glued by being camped or loaded into a press and then removed. In the Elementhus system Bergvall and Dahlberg, designed and patented, a continuous press that operated by means of carriages that moved on an oval track. This circular clamp allowed the boards to advance continuously at a rate of 6 meters per minute, under fifteen PSI of clamp pressure (10 kilos / sq. cm) and also heated the glue to 100°C via high frequency radio waves that heated the resin glue but did not heat the wood.
The shavings form the continuous planning and machining operation was collected into six silos by means of a cyclonic collector and then used to fill each box section with densely packed shavings. A lid is inserted into one of each box, then a screw feeder would fill each box using counter weight to pack the shavings, and then the filled box elements would be capped at the other end. Each box was then bored for dowels which prevented the assembled units from sliding past each other under wracking forces. It was not until a decade later that mineral fiber insulation became industrially available in Sweden, 20 centimeters of compressed wood shavings was, for its time, was phenomenally good insulation.
After being milled to sub millimeter tolerances, the parts were then racked and sent to an automatic spray booth to be painted.
To the untrained eye, it is impossible to detect that what Bergvall and Dahlberg had designed was a radical departure from anything that had ever been attempted in an industrially produced house. Because what they produced looked almost exactly like what the untrained eye thought of as “a house”, unlike many of the designs celebrated by modernist architects. Elementhus was presented as something close to a traditional Swedish house.
While Bergvall and Dahlberg completely reinvented the manufacture of the house, in a way that far surpassed any other effort, they did not make any architectural statement, and left the house’s traditional shape unmolested.
This remarkable technical achievement therefore, had nothing to excite the ideology that modernism demanded by way of a new form. If you are busy selling visions of “a new Architecture” as modernist architects certainly were, or if your real interest was in the forms or shapes of buildings, as opposed to the questions of how you manufacture them, Elementhus was a major letdown.
It is deeply inconvenient to admit that Elementhus exposes modernism's most cherished claims as total nonsense — while simultaneously proving that houses could be delivered, exactly as modernist thinkers imagined: as highly automated production packages, delivered directly from a factory.
The most successful modernist attempt at a factory produced houses did not even remotely imply a new aesthetic for buildings. Despite this house being built completely by modern machines, the elegant use of industrially modified materials, the great speed and precision of its production, there was nothing “modern” look at in the finished house.
The seeds of their own un-noticing
Bergvall and Dahlberg themselves knew that the traditional appearance of Elementhus risked making it invisible, and in describing their system in a company publication for the first time they wrote:
“The houses now being delivered resemble those of the traditional type, except for notable differences in detail. There is nothing, however, to prevent the use of the building units for houses of less traditional style. For instance, the roofs could be made flat instead of sloping, thus permitting much greater freedom in design and shape. So far this system of building has not been finally developed architecturally and is now planned to elaborate the idea in cooperation with independent architects”.
An example of this anticipated “architectural engagement” was soon to follow, but also destined to be ignored. Clearly meant made to look modern, Elementhus, debuted at the Helsingborg Exhibition 1955 in a design concept called “Bo med Bil” (living with car) which featured a flat roofed form by the firm AOS (Ahlgren Olsson Silow Arkitektkontor AB), and sported a Volvo PV444 parked under an integrated carport.
The promotion talked of the central place of the car in the family’s life, but if there was any excitement or interest in of the building technology that had been created at Mockfjärd, it is not evident beyond the fact that Elementhus was noted as the manufacturer. It is all but impossible to find mention of this exhibit in the historical accounts, however, the history of this exposition, invariably does mention the architectural works of acclaimed designer Le Corbusier, and the packaged aluminum house designed by Pierre Vivien and Jean Prouvé which were featured next door in the French pavilion.
While Prouvé’s ambition, like Gropius, to contribute to mass produced housing are widely known, and occur during the exact same time as Bergvall and Dahlberg’s work, and, at least at H55, were erected side by side, observers did not seem to celebrate, let alone understand, the work of the Swedes.
Prouvé’s efforts, which consist of stamped and folded sheet metal, were never produced beyond a few prototypes, but have none the less attracted a kind of cult art object status. Bergvall and Dahlberg’s genuine mastery of industrial production was never noted, or promoted for its genius, and its application as “architecture” despite the attempt to present it as a modernist box, and not a traditional cottage, was essentially ignored.
A deep confusion
It is worth considering for a second the deep confusion that exists in modernist discourse around the idea that changing the way something is made, necessarily implies that it should look different. It is perhaps on these shoals that thinking about how to apply machines to house production, has most struggled, specifically because there is no relationship between being machine made and looking machine made, at least not at the level of style or form. Elementhus provides an undeniable case, as no other wooden house has ever been built around structural units that are so rigidly standardized that they could be turned out through fully automatic methods of manufacture, yet had such a traditional and non-confrontational aesthetic, appearance.
While the design for manufacturing demonstrates a highly attuned appreciation about how machines could be used to do things that radically improved output, Elementhus was presented in a way that appeared traditional despite its radical production system, and while it seem likely that there was a hope that the system could be executed in a way that would be respected as architecture, this never happened.
Before moving on past the question of design, is worth pointing out that, especially in a cold and snowy climate of Sweden, the traditional house, with a cubic volume and a sloping roof is an incredibly functional and smart form if you are concerned with being warm and dry, and if the economy of materials applies to the design brief. Executing the flat roofed box, which modernism demanded, must have represented a maintenance headache for the box beam technology as the lightweight design and wood chip filling would be quickly deteriorated if exposed to regular wetting.
Failure in America but success in Sweden:
In addition to the contrast between Bergvall and Dalhberg and other famous architectus like Prouvé, and especially Walter Gropius and Konrad Wachsmann, is worth dwelling on. The much celebrated “General Panel Corporation”, set out, at the same time, to redesign the way wooden houses could be built by using manufacturing thinking, absolute standardization and dimensional control based on a modular system of parts, and a construction process that was to be unfettered by building traditions, trade customs or union rules.
Despite this shared origin in modernist thinking in how to transform house production, Gropius and Wachsmann’s efforts were plagued with problems and quickly abandoned, while Bergvall and Dahlberg’s Elementhus, did what it promised, cutting the cost of housing production and producing complete houses delivered out of a single factory location that could provide good living conditions for working class people. Furthermore, their work was deeply influential on the evolution of Sweden’s high quality factory building technology, and its founders participated in the un-glamorous work of establishing standards, which are in everyday use and are employed in all housing built in Scandinavia (of not all of Europe).
Bergvall’s ongoing influence on construction
Bergvall who retired in 1977, was an unsung hero of the building industry though his continued engagement with core questions of the modernist movement. From 1960 to 1973, he was Chairman of the International Modular Group (IMG), which was responsible for Committee W.24 of the International Council for Building Research Studies and Documentation (CIB), and he was a major contributor on the Technical Committee (TC) 59 of Building Construction for the International Organization for Standardization (ISO). Furthermore, Elementhus spawned direct competition in Sweden, and can be said to have fired the imagination of people working in the wood products industry across Sweden.
Before this system, sawmills in Sweden, had developed industrially produced houses, which were built by layering planks of wood in alternating directions into wall sized sections. These walls were initially sold as “Plankhus” (literally plank houses), but this idea was hardly much of an advancement on the ancient craft of log building which was standard across Scandinavia where there is an abundance of straight softwood trees and there was no need to conserve wood by using it as a frame, like Germany’s traditional “Fachwerkhaus houses” or “Post and Beam” structures common in most of Europe where wood is scarce relative to the demand for buildings. Log buildings in Scandinavia were usually improved squaring the timbers and adding vertical boarding over the wall. Often this involved adding layers of insulation or asphalt paper to make the wall more wind resistant-but it consumed excessive quantities of wood, something frame solve through engineering.
The most startling aspect of Elementhus was how completely realistic it was, while also being almost insanely complicated. Eespecially when evaluated with the aid of hindsight, and from the perspective of measuring what was they actually produced, no other modernist era architect has come close to creating both such a practical and technically ambitious proposal to design a building for manufacture. Few building firms anywhere, anywhere, architect led or not, can claim to have produced homes in on the scale of Elementhus.
Bergvall and Dalhberg had essentially embedded themselves inside of industry and had built consensus working within a mandate, from sponsors to produce housing whereas General Panel Corporation, was a design idea, from above, in search of backers who would accept its commercial potential.
The instinct that architects have for working in this mode is illustrated by , Gilbert Herbert’s “Dream of the Factory Built House”, which characterizes the work of Gropius and Wachsmann as “technically impeccable”, and recounts their failure as a grave shock to the whole movement for industrially-produced housing, without blaming the men behind the effort, for its failure. The adoration that led Herbert to praise a failed system, is the flip side of the reason that Bergvall and Dahlberg were ignored for their success, and the reason that architecture feeds its students with heroic tails of unappreciated genius, while ignoring actual genius.
Architects must learn new ideas about engaging with industry, and Bergvall and Dahlberg offer an rare example of everything that has to happen to build an acceptable house in the context of industry.
The problems that brought ruin, almost without exception, to every modernist architect who ever tried to solve the problems of factory production in housing were addressed by these men, in part by their humility and the collaborative nature of their engagement. Ironically it is likely the same traits that have prevented their own profession from appreciating them. Because the houses were intended as working class homes, and not presented as bold architectural ideas, it may have been hard for the architectural press to see that Elementhus was as innovative a modernist masterwork, that anyone had ever produced.
Wood in Focus
The Elementhus story is also worth disseminating as an example of using wood. There is a resurgent interest in wood as an engineering and building material, partly because wood’s environmental benefits as a renewable material and ability to store carbon makes it attractive when compared to concrete and steel. Elementhus is a case study in the complexity of using and mitigating the the difficult natural properties of wood as an engineering material. Because Elementhus was designed around certain economic assumptions that are no longer true, by today’s standards, it is obsolete, however, from a design perspective, and because of its absolute mastery of the material, Elementhus was, and still is, an amazing thing for a wood technologist to study.
Wikihouse losing ground
Over the last decades, CNC machines and sophisticated computer software have come down in price such that they are common in all architecture schools. This has led to a trend where architects create shapes directly from computer files out of sheets of plywood, and then assemble them as “prefab” housing solutions or mix them ideas from the field of software development to create concepts like “open source” or “wiki” houses. While these projects get noticed because of novelty, they invariably resemble the plywood dinosaur skeletons kits that museums sell to children.
When held up to the kind of thinking and realistic industrial engagement that Lennart Bergvall and Eric Dahlberg brought to creating a real housing solution, the next young architect who wants to make a house with a CNC router should be encouraged to consider reality. More architects should be encouraged to model themselves on the example set by these Swedish designers, and to first study the problems they are trying to solve, engaged deeply with industry and work collaboratively, with a common set of references, values and vocabulary.
About the Author:
Scott Hedges is the founder of Bygghouse, a building technology company based on working with modernizing construction, and he writes, speaks and lead tours to connect people with Swedish building industry’s ideas around industrialization.
He is part of the facebook group, “Elementhus på Skansen” which seeks to interpret the story of Elementhus in Sweden’s oldest open air history museum.
A Swedish version of this essay can be located here.