The Bosch Vault: Reinterpretation and Exploration of the Limits of the Traditional Thin-Tile Vault in the Post-War Context

Abstract

After the Spanish Civil War, the shortage of building materials in the country and the restrictions imposed by the Dirección General de Arquitectura limited the use of steel in construction, encouraging solutions that reduced the consumption of this material. In this context, the thin-tile vault gained new relevance due to its low cost, speed of execution and good structural and fire performance. Among the architects who revisited this system, Ignasi Bosch Reitg (1910–1985) developed an innovative procedure for the construction of continuous ceilings, based on double-curved vaults with a single layer of brick. His cousin, Josep Maria Bosch Aymerich (1917–2015), an industrial engineer and architect trained in the United States, brought a business vision to the table when he discovered the potential of this system. This paper proposes an in-depth study of the patents requested on this system by the two architects, questioning the reasons for their success or failure in different countries, both in terms of dissemination and exploitation, in regard to the historical context in which it was developed. The analysis, based on original documents from the Bosch Aymerich Archive, uncovers the tensions that the reinterpretation and global projection of a traditional technique can generate.

1. Introduction

Ceramic vaulting is a construction technique with a long history in the field of building. Recent studies have pointed to its use in constructions dating back to the 4th and 2nd centuries BC [1]. Thin-tile vaulting, the subject of this paper, originated in Spain [2] and is unique in that it is self-supporting and does not require falsework in its construction.

The technique, examples of which have been found dating back to the 15th century in Spain [3], where it was used regularly since then [4], also underwent notable development in France in the 18th century [5]. It also reached the United States thanks to Rafael Guastavino (1842–1908), who devoted himself to studying the system and obtained up to 24 patents for it, extending its use in the United States through the works carried out by his company, Guastavino Fireproof Construction Company [6,7]. The technique also underwent significant development in 19th-century Catalonia, where it was even used by Antoni Gaudí [8]. In the early 20th century, Spanish architects such as Jaime Bayó and Jeroni Martorell i Terrats wrote treatises on the technique, studying its structural properties [9,10]. Also in this century, the technique would undergo notable development in Latin America [11]. It even aroused the interest of Le Corbusier, who used it in the Maisons Jaoul (1954–1956) and other projects [12].

In the post-war context, specifically in Spain, which had been embroiled in a civil war between 1936 and 1939, ceramic block structures of all kinds experienced a resurgence. This study focuses on the thin-tile vault, more specifically on the structural system for ceilings and roofs used by Ignasi Bosch Reitg (Gerona, 1910–1985) and later patented, together with his cousin Josep Maria Bosch Aymerich (Girona, 1917–Barcelona, 2015). This system, which Bosch Reitg initially experimented with, and which would be implemented by both architects in their work, presented a variation on the traditional technique, allowing structures to be supported by pillars or even thickened partitions, without the need to use more iron than a few braces to counteract the horizontal thrust generated by the vault. In 1950, both architects applied for a patent in Spain, drawing on the experience Ignasi had gained with a previous patent in the same year and Josep María’s experience in other fields. Later, the architects jointly applied for patents for the same structural system abroad, with varying degrees of success in different countries.

While Bosch Reitg’s involvement in the Spanish patent has been mentioned in several studies [1,13,14] (p. 12), it has been dealt with from the point of view of its constructive foundations and structural calculation. However, the patenting and development process has not been studied in detail, nor has the attempt by both architects to export the system they had invented to the international market, something unprecedented in post-war Spanish architecture [15]. In addition to the novelty of the system itself, which has also not been dealt with, the patenting process clearly reflects the economic and social situation of the country, as well as the possibility of adapting a traditional construction system to its contemporary use and the “tensions” that this generates. That is why this article proposes a detailed study of the patenting process and later internationalisation of the “Volta Bosch”—as they themselves called the system [16] (p. 519)—with the aim of showing its relevance and novelty in the postwar context and even today.

The aim is to bring this unprecedented case to light and analyse its success or failure in terms of dissemination and exploitation.

Therefore, the purpose of this study is not to review “The Bosch Vault” from the point of view of structural calculation or its typological classification within thin- tile vaults, but rather to study the reasons that made its approval possible: as a patent it was approved in more than half a dozen countries at a time when Spain was a country that did not “export” architecture, and to assess that achievement. The structural calculation and typological classification of thin-tile vaults have been addressed by authors such as Santiago Huerta Fernández [17], David López López [18] or Arturo Zaragozá Catalán and Rafael Soler Verdú [19], among others. However, analysing the place of the Bosch Vault within these categories would lead to a different study.

2. Materials and Methods

The main material used for this research was the archive of Josep Maria Bosch Aymerich. Specifically, the documentation relating to the patents for the thin-tile vault structural system found in folder no. 2012, the contents of which, after digitization with a view to the imminent public exhibition of the archive through the Bosch Aymerich Foundation (FBA) database and the Coeli system (Barcelona, Spain), were compiled in the 959-page 2012 archive–cited as [16] in the reference list.

To introduce the scope of the research, the bibliography on the historical moment of the patent has been reviewed, especially with regard to the use of thin-tile vaults and particularly in the Spanish context in which it was developed, as well as the specialized bibliography on the architects Bosch Reitg and Bosch Aymerich, supplemented with information found in the Bosch Reitg archive in Girona and the Bosch Aymerich Foundation archive in Barcelona. A review of the general bibliography on “Thin-tile vaults” has also been carried out, as it is considered relevant to the contextualization of this work.

The methodology followed for the analysis of the documentation in Pdf 2012 consisted, first, of a quantitative classification in which a total of 204 differentiated documents were extracted from the original file. In a second place, a qualitative classification was undertaken, separating items according to the type of document (letter, patent report or graphic documentation). Finally, documents were grouped according to the country they were directed to or received from (Spain, Italy, France, Belgium, Portugal, UK, Germany, Canada, Mexico, Brazil, Uruguay, EE. UU. and Argentina); within this classification, they were ordered chronologically (between 1950 and 1956).

Once classified according to these criteria, the documents were analysed in search of information about the patenting process and the possible reasons that could have led to the decline or approval of the patent in each of the countries. This information was extracted and summarised in

Table 1. Summary table of patents by country.

Country	Apply Date	Approval Date	Quitting Date			Objections	Interferences		Projects *
				Vulgarized	Patent	Geometry	Structure	Material
Spain	24/7/50	7/11/50	-						X
Italy	25/3/52	2/12/53	-
France	18/3/52	30/9/53	-						X
Belgium	21/2/52	15/4/52	-
Portugal	15/3/52	22/4/52	-
UK	25/3/52	3/6/54	-		X	X	X
Germany	24/3/52	-	21/8/56		X	X	X		X
Canada	18/3/52	8/12/53	-			X
Mexico	19/3/52	-	30/4/54						X
Brazil	26/3/52	-	30/4/54	X		X
Uruguay	1/4/52	-	2/6/54	X
EE. UU.	17/3/52	-	10/12/56		X	X	X	X
Argentina	-	-	29/3/55						X

* Of all the preliminary designs and studies carried out by Bosch Aymerich, only those in Spain came to fruition.

, allowing for a comparison on the trajectory of the patent in different countries, and also the different reasons why the patent was rejected. This table was used to structure the Section 4.

Section 4 also brings forward the data that highlights the innovations that the promoters of the Bosch Vault offered in respect to other contemporary structural systems of the time. In other words, although this is not an article focused on the structural analysis of the vault, the calculation methods documented in the sources are presented in order to compare them with the systems in use at the time.

Moreover, the discussion cross-references historical context and the results in order to determine the success or failure of the patenting process of the system in terms of dissemination and exploitation and the reasons that could have led to this. Finally, the answers to the questions raised in the discussion are summarized in an orderly manner, by way of conclusions.

3. Historical Context

After its success in the United States since the second half of the 19th century with Guastavino, the thin-tile vault experienced a decline in popularity with the emergence of the modern movement—although, as recent studies have shown, it was not completely absent from it [20]. However, it would experience a resurgence during and after World War II. The shortage of materials in Europe would encourage the development of this technique in some areas such as France, Germany, and Spain [21]. The thin-tile vault would also be used in the 1940s in Latin America, with architects such as the Spanish emigrant Antonio Bonet and the Argentine Eduardo Sacristé, who would use it assiduously in his homes [11,22].

Spain and Spanish architects played an important role in this resurgence of the thin-tile vault as a construction method. Plunged into a civil war that lasted from 1936 to 1939, the country experienced an early version of the post-war situation that would affect Europe from 1945 onwards. Thus, the experimentation carried out by Spanish architects would not be insignificant for some European experiences. The work of Luis Moya and Ignasi Bosch would have been known, for example, by the German engineer Max Rank, who would carry out several projects using this technique in Germany [22] (p. 170).

In Spain, in 1941, with Francisco Franco’s new regime already in place, the Ministry of the Interior published a decree restricting the use of steel in both public and private construction projects [23,24], although there were exceptions. A year later, the Dirección General de Arquitectura (General Directorate of Architecture) of the same ministry—a body created in 1939 with the functions of “national planning of architecture,” “directing the intervention of architects in public services that require it,” and “directing professional activities of this order” [25]—published Special Floor Systems for Building. Types approved and reviewed by the Research and Standards Section [26]. This was a guide to construction solutions with reduced use of steel for Spanish architects, which did not include thin-tile vaults, but did include numerous solutions for flat ceramic slabs.

Architect Luis Moya (1904–1990) was perhaps the first to use this construction solution in his works in the context of post-war Spain. In 1942, he had already completed the construction of the Houses in the Usera Neighbourhood, two-story row houses whose floors were formed by Thin-tile vaults supported by a set of parallel walls. From then on, the architect experimented with different types of Thin-tile vaults in various types of projects (housing, church restoration, new construction). In 1947, he wrote a short treatise entitled Bóvedas Tabicadas (Thin-tile Vaults), which would have a great influence on the subject. However, this did not mark the end of the use of this technique, as he would carry out projects with different types of thin-tile vaults and domes until the late 1950s, as in the case of the Laboral University in Gijón (1957) [27].

However, alongside Luis Moya, other architects also used this technique in both public and private projects. This was the case, for example, with Rafael Aburto, Francisco Asís Cabrero, Secundino de Zuazo, Ángel Truñó, and even Carlos de Miguel [28]. Rafael Aburto, for example, proposed the use of cable-stayed vaults in his project for five-story apartment blocks in Toledo, promoted by the Obra Sindical del Hogar (1943) [29]. The project was not approved, but it is an interesting example of the application of this construction system, even proposing a five-story apartment block based on it (Figure 1). He also designed a Farm School in Toledo (1947) [30], which again used parallel load-bearing walls to support the brick vaults.

For his part, architect Ángel Truñó (1895–1979) wrote another treatise on Thin-tile vaults while teaching construction at the Barcelona School of Architecture between 1948 and 1966. This text, unpublished until 2004, was written by the architect around 1951 [4] (p. XLII), after Moya’s treatise, analysing the different types of vaults that existed. Although it was not published at the time, Bonaventura Bassegoda i Musté (Barcelona, 1896–1987) mentioned this text in a speech in 1952, which is a clear indication of the popularity that the construction system was gaining, at least in some architectural circles [4] (p. XLII).

Bonaventura Bassegoda also studied the “Catalan vault” in depth, giving a lecture at the School of Architecture in 1946 [31], which had a great influence on the architects of the time, based on the numerous references they make to him in their writings. In this speech, the architect not only gave a historical overview of the “Catalan vault”, but also discussed its different formal variations and the different structural properties they presented.

Based on the above, it can be said that the thin-tile vault, present in Spanish building tradition, experienced a resurgence after the civil war. In this context, Ignasi Bosch Reitg (1910–1985), an architect born in Girona who had already been practicing for several years and had various works to his name, submitted a manuscript to the “Biennial Competition of the Official Association of Architects of Catalonia and the Balearic Islands” in 1947 [32]. In it, he mentioned Bassegoda’s discourse, as well as Luis Moya’s treatise, and even the works of other early 20th-century Catalan architects such as Jerónimo Martorell and Jaime Bayó. The subject of the work was the “Bóveda vaída tabicada” (double-curved thin-tile vault) with a lowered arch, a vault formed by a layer of hollow bricks held together with quick-setting cement. His interest in it was due to its lightness and economy of execution and the possibility of supporting it on pillars, without the need for load-bearing walls. It is also a floor slab without beams, requiring only steel braces to counteract the thrust of the diagonals. The work was illustrated with images of projects on which Bosch Reitg himself had worked (Figure 2), perhaps even with his father, who mentions that he had extensive experience in their implementation [32].

This work would later be published in Revista Nacional de Arquitectura [33], in which, with a structure similar to that submitted to the College of Architects, Bosch Reitg once again presented the results of his experiments and calculations. This time, the article is illustrated with plans and photographs of some of his works executed using this technique. Plans are shown for the Trade Union House in Gerona, the Farmers’ Home, and the Church of the Group of Sant Narcís, as well as several two-storey houses. (Figure 3).

A few months later, in July 1949, Professor Bassegoda published an article in the same magazine in which he gave his former student an “applause with reservations.” In this text, Bassegoda criticized some of the data presented by Ignasi in his previous article, attributing a lack of rigor to his calculations, although the review ended with words of encouragement:

“My friend Bosch Reitg: There will always be those who, like me now, play the role of the long-legged curlew, who gives advice to everyone but none to himself. But I am pleased, as is fortune, to lift up the bold, and therefore I welcome with joy your desire for novelty, even if it is not tempered by healthy restraint and a deep mathematical and experimental study of the structure you advocate.

The attempt represents a positive value; if it is not recognized today, do not be discouraged, for tomorrow God will bring the dawn, and we shall prosper”

[34].

Translation by the author from original:

“Amigo Bosch Reitg: Nunca ha de faltar quien, cual ahora yo, desempeñe el papel de alcaraván zancudo, que da a todos consejos, más para sí ninguno. Pero me place, como a la fortuna, aupar a los audaces, y, por ello, saludo con alborozo su afán de no-vedad, aunque no vaya arrendado por un sano comedimiento y por profundo estudio matemático y experimental de la estructura que propugna.

La tentativa representa un valor positivo; si hoy no se reconoce, nada de alebro-narse, que mañana amanecerá Dios y medraremos”

[34].

Thus, Professor Bassegoda recognized the initiative of the young Bosch Reitg, who would indeed continue to use the double-curved thin-tile vault. Proof of this would be the fact that in 1950 he filed his application for a patent for “A method of constructing double-curved vaults,” which was accepted that same year (Patent ES194095A1 in Espacenet). This patent, as its name indicates, would register the construction process that included that of the double-curved vault, supported on pillars and the bracing of the same. Reitg, for his part, already had some experience with the patenting process, as in 1942 he had also registered “A method for constructing continuous roofs” (Patent ES-150629-A1 in Espacenet).

Ignasi’s experience took a new turn when his cousin, Josep Maria Bosch Aymerich (1917–2015), began his professional career in 1948. Trained in architecture (1940–1944, 1947–1948) and industrial engineering (1939–1944) in Barcelona and business management at the Massachusetts Institute of Technology (1945–1946) [35], Bosch Aymerich brought a complementary vision to that of his cousin, characterized by a business and patent management approach. The engineer had developed activities linked to industry and management. By the age of 31, he had accumulated experience as a delegate of the National Institute of Industry in the United States, as a consultant in New York, and as the founder of American Standard—Constant Card Co., a company dedicated to managing Spanish patents abroad. In Spain, he had also applied for some patents individually, such as “Improvements in the manufacture of tapes, bands, sheets, or similar items with adhesive properties,” granted in 1949 (ES-185534-A3).

Josep Maria’s hybrid profile explains his initiative to set up, together with his cousin, the company “Estructuras Bosch”, with the aim of legally protecting the vault system and promoting its dissemination in Spain and abroad, taking advantage of the context of restrictions on the use of steel. Together, they would apply for a new patent in Spain, not only for the construction process, as Ignasi had done previously, but for the entire structural system. After its approval and with the company established, they would embark on the path of patenting abroad, which, together with the Spanish patent, is the subject of this study.

Josep Maria’s knowledge of the system used by his cousin was consolidated early on when he visited the construction site in the San Narcís neighbourhood of Girona (1944–1952), where this construction system was being tested [36]. That same year, in his first major project—a block of flats on a block in Sarriá (Barcelona)—he already resorted to using it [37]. This was only the beginning of a series of projects in which he incorporated the technique, such as the Remei Clinic on Carrer Escorial (1948) and the residential complexes for SEAT workers (1953) [38]. The continued use of the “Volta Bosch” demonstrates not only technical confidence in the system but also a deliberate strategy of professional validation in different architectural programs on the part of both cousins.

The situation of scarcity that the nation had faced, and which had favoured the use of brick, was coming to an end. In the mid-1950s, the improvement in the country’s situation led to its economic recovery, and brick was no longer seen as a strictly necessary material in construction [28].

In 1954, a “Critical Architecture Session” was held—a meeting of architects from across the country to discuss hot topics in the profession [39]—with the theme “Defense of Brick.” The aim was not so much to discuss the material necessity of brick as to consider the formal implications of its use. In this context, however, reflections such as those of Fisac emerged, who proposed that, although the use of brick in walls continued to be relevant due to the impossibility of obtaining other materials, it was not so necessary for structural elements such as vaults [39] (p. 25).

In 1962, Law M.V. 101/1962 was approved, ending any restrictions on the use of steel in construction [28] (p. 332). Although the use of brick did not cease, by this time few people continued to use Thin-tile vaults. This was the case with Luis Moya, with the churches of Santa Maria del Pilar (1963–1965) and Santa María Madre de la Iglesia (1966–1969) [27]. Josep Maria Bosch Aymerich would use it in his works at Cap Sa Sal (1955–1963) [14] (p. 636). However, on the one hand, the country’s economic improvement and, on the other, the resurgence of modern architecture, which, after having been strongly rejected in the early years of Franco’s regime, took place in the 1950s [40,41,42], would lead to a new decline in the use of the thin-tile vault, which had proved so efficient in the reconstruction of Spain.

4. Results

From this point onwards, the article is based mainly on material found in folder 2012 of the FBA archive, as explained in Section 2.

4.1. The System and the Patent

The initial Spanish patent owned by Ignasi Bosch Reitg and Josep Maria Bosch Aymerich describes a double-curved vaulted ceiling made with just a single layer of 3 cm hollow bricks and Portland cement as a binder (Figure 4(4)). In order to create a floor slab on top of this shell (Figure 4(6)), diagonal ribs are added to support the bricks that will bear the weight of the flooring (Figure 4(20)). If loads are to be borne by a partition wall, a rectangular crown would be added to transfer the loads to the structure (Figure 4(4,19)).

One of the advantages of this design is that it can be supported entirely on pillars. Thus, the savings in materials achieved by the horizontal structure do not require a greater investment in the vertical structure, eliminating the need for load-bearing walls. However, this approach has a well-known drawback: the lateral thrust produced by the shell on the pillars. The patent includes a calculation method for bracing the pillars with a steel tie consisting of a bar between 16 and 20 mm in diameter. The tie rods could be located at the base of the vaults or flush below the floor slab to conceal them within the system, which required reinforcing the pillars to compensate for the moment produced by this displacement (Figure 5 and Figure 6).

To calculate the amount of needed steel, the following formula is used in the first place:

$$\mathrm{H}={\displaystyle \frac{\mathrm{P}\times \mathrm{A}\times \mathrm{B}\times \mathrm{L}}{2\times \mathrm{F}\times 8}}$$

where H = horizontal thrust along the diagonal direction; P = unit load of the vault; A and B = sides of the vault; L = equal span of the diagonal arch (equal to the length of the diagonal in plan); F = deflection of the diagonal arch (sum of the deflections of the directrix and generatrix arches, i.e., double that of the generatrix) [16] (p. 505).

This formula can be used when the arrows of the guide formwork and the generatrix are equal, a necessary condition for constructing an isolated vault supported only by the four corners [43] (p. 38).

Bosch Reigt applies the following calculation example for bracing assemblies in his report for “The Bosch Vault”:

“We assume that the surface area to be covered by the vault is 5 × 3 m, or a total of 15 square metres. The diagonal arch span will be 5.40 metres. The vault will be constructed using medium-sized hollow bricks 4 cm thick, of which we will only consider 3 centimetres for the purposes of the resistant section. We will give the diagonal arch a deflection of 1/10 × L = 54 cm. Therefore, the formwork—generatrix and directrix—will have a deflection of 27 cm. The permanent load will be 100 kg per square metre; and the overload, 150 kg per square metre, i.e., total load per unit, 250 kg per square metre.

To find the value of the thrust at the angle, we will use Formula (3).

$$\mathrm{H}={\displaystyle \frac{250\times 3\times 5\times 5^{\prime }40}{2\times 8\times 0.54}}=2^{\prime }36 \mathrm{tonnes}.$$

To calculate the iron section, as we indicated at the beginning, we will break down this thrust, according to the diagonal, into two components, according to the sides of the vault; which, when calculated using vectors, gives us: Ha = 1.22 tonnes; Hb = 2.04 tonnes. The iron section to be placed on the shorter side will be:

$$S_a={\displaystyle \frac{1^{\prime }22}{1^{\prime }20}}=1.02 {\mathrm{cm}}^2 ; S_b={\displaystyle \frac{2^{\prime }04}{1^{\prime }20}}=1.70 {\mathrm{cm}}^2$$

In other words, for bracing, we will place a 12 mm round bar on the shorter sides and a 15 mm round bar on the longer sides.

The total surface area of the vault will be = 15 m².

The total weight of iron with fork = 20 kg.

The iron per square metre of surface area = 1.33 kg/m².”

[16] (p. 506).

Translation by the author from original:

“La superficie a cubrir por la bóveda suponemos que es de 5 × 3 m, o sea en total 15 metros cuadrados. La luz del arco diagonal corresponderá a 5′40 metros. La bóveda la ejecutaremos con ladrillo hueco mediano de 4 cm de espesor, del que solo consideramos 3 centímetros a los efectos de sección resistente. Daremos una flecha al arco diagonal de 1/10 × L = 54 cm. Por tanto, las cimbras—generatriz y directriz—tendrán una flecha de 27 cm. La carga permanente será de 100 kilogramos por metro cuadrado; y la sobrecarga, 150 kilos por metro cuadrado, o sea, carga total por unidad, 250 kilos por metro cuadrado.

Para conocer elvalordel empuje en el ángulo haremos uso de la Fórmula (3).

$$\mathrm{H}={\displaystyle \frac{250\times 3\times 5\times 5^{\prime }40}{2\times 8\times 0.54}}=2^{\prime }36 \mathrm{toneladas}$$

Para calcular la sección de hierro, como ya hemos indicado al principio, descompondremos este empuje, según la diagonal, en dos componentes, según los lados de la bóveda; lo que efectuado por vectores nos da: Ha = 1.22 toneladas; Hb = 2′04 toneladas. La sección de hierro a colocar sobre el lado menor será:

$$S_a={\displaystyle \frac{1^{\prime }22}{1^{\prime }20}}=1.02 {\mathrm{cm}}^2 ; S_b={\displaystyle \frac{2^{\prime }04}{1^{\prime }20}}=1.70 {\mathrm{cm}}^2$$

O sea, para arriostramiento colocaremos un redondo de 12 mm en los lados menores, y un redondo de 15 mm en los lados mayores.

La superficie total de la bóveda será = 15 m².

El peso total de hierro con horquilla = 20 kg.

El hierro por metro cuadrado de superficie = 1.33 kg/m².”

[16] (p. 506).

The savings compared to flat ceramic slabs described in Special Slab Systems for Building Construction. Types approved and reviewed by the Research and Standards Section of the Ministry of the Interior are significant, as this publication estimates a weight of 4 kg/m² [26] (p. 96).

According to Ignasi Bosch Reitg’s practice, this structural assembly resulted in savings of 35–40% in the overall construction cost of a low-class dwelling and 25% in a middle-class dwelling [16] (p. 691). On several occasions, they cite an example in which they used the soil from the foundations and from leveling the ground to make bricks on site and then use them to build the vaults [16] (p. 535), increasing the savings by eliminating the need to transport the material. The list of buildings constructed using this structural system by 1950 was extensive, but mainly located in the province of Girona and the city of Barcelona, as cited on page 582 (text translated from Spanish by the author):

“Barcelona.

-

Block of 270 homes for Ecisa, S.A., worth 34 million pesetas. (Figure 7)

-

Clínica del Remedio, on Calle Escorial.

Verges.

-

Industrial warehouse with a span of 19 m.

Gerona.

-

San Narcis suburb: church, social center, schools, and 527 homes with 65,000 m² of vaulted ceilings. Seven-story union building. Conciliar seminary. Apartment buildings for Messrs. Guerra y Esofet, Bosch, Picamal, and Colubret, etc. Nuestra Señora de los Angeles children’s home. Ultonia Theater, with two amphitheaters and capacity for 1500 spectators.

Figueras.

-

Warehouse with a span of 16 m.

Mataro.

-

Industrial building on La Rambla.

To avoid making this list endless, we will only mention the construction of working-class neighborhoods in Huelva, Jaén, Valladolid, La Coruña, etc., and a suburb built in Olot consisting of 389 homes, a church, and a social center, with 44,000 m² of vaulted ceilings”

[16] (p. 582).

Figure 7. Housing for Ecisa, S.A. in Sarriá, Barcelona. FBA Archive, C1_4.21.

Figure 7. Housing for Ecisa, S.A. in Sarriá, Barcelona. FBA Archive, C1_4.21.

This list shows works of various types that require very different geometries. Another advantage offered by the system is the elasticity of the geometry. The double-curved vault could be adapted to any perimeter (Figure 8), whether it was the roof of a theater, the dome of a church, or a simple rectangular slab.

4.2. National Diffusion

With the Spanish patent (applied for on 24 July 1950, and published in June 1951 with the number ES197134A1) and an extensive portfolio of completed projects, Bosch Aymerich was actively involved in promoting the system, corresponding with construction companies in different cities. The preserved documentation provides different cases of its dissemination throughout the peninsula, as well as constant monitoring of the works carried out. For example, they received a letter requesting information to build 500 subsidised homes in Madrid [16] (p. 612). Bosch Aymerich responded by putting them in touch with a construction company in Madrid that “has carried out a considerable amount of work using this construction system and will be able to provide you with more detailed verbal references” [16] (p. 613). In some cases, this defence took on a clearly administrative tone: a letter (13 June 1955) addressed to the municipal architect of Valladolid denounced the construction of projects using the technique without the intervention of the patent holders, requesting information on the contractors involved with a view to collecting exploitation fees [16] (p. 568). On 26 January 1955, a direct complaint to a builder, Gaudencio Fraguell, along the same lines is also documented [16] (p. 475). On the back of a letter, various projects underway to disseminate the patent are noted: the IMOSA factory in Vitoria and a preliminary project in Germany [16] (p. 584). These episodes reflect both a certain spread of the system and the conflicts inherent in the management of exploitation rights.

The scope of application of the Bosch vault in Spain is outlined in an informative letter dated 24 January 1953 [16] (p. 612) and in the notes from a meeting [16] (p. 614). These documents list various complexes that have already been built, which are added to those in the previous list:

-

Complex of 300 dwellings in Olot [16] (p. 612, 7).

-

Complex of 123 dwellings in Huelva [16] (p. 612, 7).

-

Complex of 722 dwellings in Valladolid [16] (p. 612, 7).

-

Other 6- to 8-storey buildings in Barcelona, adding 650 additional homes [16] (p. 612, 5) (Given the date and scale, it is understood that these are homes for SEAT workers in the Zona Franca).

-

Groups of 500 and 150 homes in other locations [16] (p. 614, 6 and 7).

Added to this list are two cases (mentioned in the captions of two missing images [16] (p. 916)) of the construction of an experimental 4 × 21 m vault and the foundations of another planned 40 × 21 m vault. These data highlight the versatility of the system in both residential and industrial programmes and the ambition to test large spans throughout the peninsula.

In fact, both the 270-home project in Barcelona and the Sant Narcís neighbourhood in Girona (Figure 9) were used by the Bosch cousins as showcases for architects from Spain and beyond. Bosch Aymerich’s business training allowed him to look beyond the Pyrenees and the Atlantic: architects and businesspeople from France [16] (p. 700), Argentina [16] (p. 534) and Mexico [16] (p. 864) visited these projects and played important roles in his international dissemination strategy.

4.3. International Diffusion

The journey of the patent held by Ignasi Bosch Reitg and Josep Maria Bosch Aymerich, which began with its approval in Spain in 1951, clearly shows the tensions that arise when traditional knowledge attempts to legitimize itself as technical innovation in the post-war domestic market. Analysis of the documentation reveals a dynamic marked by ambivalence: in some contexts, the Bosch vault was recognized as an applicable and novel invention; in others, however, it was interpreted as a mere variation on existing systems and therefore lacking in patentable value. A horizontal reading of the various reviews and responses highlights both the specificity of the system and the limits of its recognition outside the Spanish context.

Through correspondence with Clarke Modet & Co., an international patent firm, it is possible to reconstruct the history from the selection of countries to the procedure carried out in each of them. In a list of countries with their respective patent rates, the following are highlighted: Portugal, France, Germany, Italy, England, Belgium, Uruguay, Brazil, Mexico, the United States, and Canada [16] (p. 542). In the aforementioned minutes [16] (p. 584), there is a handwritten note by someone named Martorell: “France, no! Discredited” and on the next line, “USA, too far away and expensive.” After discussion, both countries were included in the international dissemination plan, as can be seen in the list.

In European countries, they were approved with hardly any revisions. The cases of Portugal (applied for on 15 March 1952, granted on 22 April 1952) [16] (p. 363) and Belgium (applied for on 21 March 1952, granted on 15 April 1952) [16] (p. 121) are the fastest: in just one month, they received news of their grant, marking a promising start. By the following year, patents in Italy (applied for on 25 March 1952, granted on 2 December 1953) [16] (p. 373) and France (applied for on 18 March 1952, granted on 30 September 1953) [16] (p. 436) would also be granted without receiving any requests from the patent offices. The reviewer in England (applied for on 25 March 1952, granted on 3 June 1954) [16] (p. 70) found, however, interference with two patents: No. 540,925 407 (Evelyn Hurden) and No. 113,268 (Karl Pauli Billner, 1882–1965). Bosch Aymerich points out that there is no interference as they are different systems: the first is a slab of asbestos cement arches and the second is a system of elliptical domes that requires beams and walls [16] (p. 23). The English patent was approved, even though the examiner found another patent with possible interference, No. 669,806. Even so, it was decided to include a note referring to that patent so as not to prolong the process.

The German patent was the only European patent not granted (applied for on 24 March 1952) [16] (p. 310). It was unsuccessful due to interference with patent no. 605,838 (Franz Dischinger, 1887—1953), known as the Zeiss-Dywidag system (Figure 10). Even though it was made of reinforced concrete and required perimeter beams, the short response times and slow receipt of translated notifications made communication with the patent office costly. Among the extensive documentation contained in the German patent [16] (pp. 264–344), it is difficult to find reviews that do not mention the Zeiss-Dywidag system, even though communications continued until 29 September 1956. Ultimately, the patent was not granted.

In the Americas, the patent was not as well-received as in Europe. It was only approved in Canada (applied for on 18 March 1952, granted on 8 December 1953) [16] (p. 183) (Figure 10). In Mexico (submitted on 19 March 1952) [16] (p. 78), Brazil (filed on 26 March 1952) [16] (p. 116), Uruguay (filed on 1 April 1952) [16] (p. 179) and the United States (filed on 17 March 1952) [16] (p. 896) it was not granted. Despite receiving requirements from the Canadian office, Bosch responded by clarifying once again the basic points of his invention [16] (pp. 229–245): the self-supporting nature of the vault constructed with a single layer of brick, without the need for hoops, beams or reinforcement mesh; the possibility of direct support on isolated pillars, and not only on load-bearing walls or arches; and the elasticity of the geometry, capable of adapting to different configurations without losing stability. These three points entail economic and management advantages: the system allowed for significant savings in steel, reaching up to 40% of the building’s construction cost.

Despite the similarities between the reviews from different countries, there are other formal differences. In Canada, for example, the reviewer demanded that the analogy used by Bosch Aymerich—comparing the vault to a handkerchief blown by the wind—be developed in detail, analysing how the shape and tensions varied depending on the direction of the wind or the thickness of the fabric [16] (p. 260) (Figure 11).

Even so, the debate did not always resolve in his favour. The case of the United States illustrates the limits of his strategy. For four years, Bosch Aymerich maintained extensive correspondence with the examiners, providing photographs, revised reports, and experimental examples. In his writings, he went so far as to formulate sophisticated analogies to demonstrate that innovation can lie in structural differences invisible to the naked eye, just as a tubeless tire represents an invention compared to a conventional tire [16] (pp. 710–714). However, the US office remained firm in its view that the system did not offer sufficient novelty, mainly in comparison with patent no. 173,953 (Ralph Hills, 1809–1875) (Figure 12). Bosch Aymerich managed to refute alleged interferences with other patents using the arguments used in other countries. Even so, it is not surprising that the reviewer makes no reference to Guastavino’s patents. Before filing the US patent application and taking advantage of a trip to the US in October 1951, Bosch Aymerich took data from three Guastavino patents (Nos. 430,122, No. 460,562, and No. 471,173) to prepare the defence of his invention in that country [16] (p. 948). Despite all attempts to convince the examiner of the novelty of the patent, the office ultimately issued a final rejection. Even though he had the possibility of appealing the final rejection, Bosch Aymerich wrote to Clarke Modet & Co., communicating his desire to cancel the project: “I am sorry to inform you that I have decided to abandon this matter, without incurring any further expenses in the United States” [16] (p. 703). The defeatist tone of the communication shows, on the one hand, Bosch Aymerich’s affection for the country and, on the other, the amount of work and hours devoted to the project, which is the one with the most documentation in the archive [16] (pp. 703–959).

The other countries have less ground to cover. Brazil argues that architectural history offers us a catalogue of solutions for converting a curved surface into a square floor plan and that, therefore, it does not deserve to be patented [16] (pp. 100–101). Uruguay presents a damning report by the Chamber of Industries, the Faculty of Engineering, and the Industrial Property Directorate stating that “the proposed procedure does not deserve to be patented because it is a method already widely used in the construction industry and does not constitute any novelty” [16] (p. 153) (Figure 13). In contrast, the proceeding in Mexico never received a response [16] (p. 541).

Although Bosch’s dissemination strategy was not limited to patenting the system, he understood that he himself had to seek out projects abroad to convince local technicians of its advantages. It has already been mentioned above that there was a preliminary project in Germany [16] (p. 584), the development of which is unknown. There is a series of five letters with a technical and professional institution in the French ceramics sector [16] (pp. 639–641, 645–646, 700). The correspondence arose following a visit by the Centre de Documentation de la Terre Cuite to the works in Sant Narcís neighbourhood of Girona, where they were received by both cousins. Bosch Aymerich took advantage of a subsequent meeting in Paris to propose the construction of some vaults for display at the Marseille Fair. Unfortunately, despite having the plans for the necessary foundations (Figure 14) and the measurements, a bricklayer and a carpenter were unable to build the vault upon their arrival. The tight deadlines did not allow for its construction, and the project was abandoned.

Taking advantage of his relationship with a Mexican colleague at the Massachusetts Institute of Technology, he saw an opportunity to build a sulphuric acid production plant in Mexico, as recounted in correspondence between April and May 1955 [16] (pp. 856, 864, 866–867). But this project also failed to materialise.

In Argentina, despite not having attempted to patent the system, Bosch Aymerich tested the possibilities to open another front. In 1954, after an Argentinian friend visited his home in Barcelona, Bosch Aymerich came into contact with Ricardo Fernández Vallespín (1910–1988), a Spanish architect who, after being ordained as a priest, went to live in Argentina.

The correspondence focuses on studying the feasibility of its construction in the country [16] (pp. 516–536). Vallespín, after commenting on some administrative difficulties in building houses with this system—only houses with reinforced concrete slabs were allowed to be built—points out that, even considering the many advantages of the system, the success of its implementation does not depend solely on them. It requires what is available in Spain: a team of quantity surveyors and site managers familiar with this method of construction, as well as someone to set up and manage the company necessary for its organisation. The only possible way forward, he says, is to collaborate with a foreign company on low-demand projects, such as industrial warehouses, thus avoiding the administrative problem. Unfortunately, the letters end with the promise of a more detailed study of these industrial building projects, which, after a reminder from Bosch Aymerich, never reached him.

Table 1 summarises all the cases in the different countries where attempts were made to expand the system.

5. Discussion

As seen in the historical context, the period following the Spanish Civil War was a time of scarcity and rationing of reinforced concrete and, in particular, steel. This situation favoured the use of traditional techniques that did not require these materials for their execution, such as thin-tile vaults, specifically in the period known as “autarky”, which we can place around the 1940s in Spain. These needs will change as the effects of the development plans implemented in the 1950s by the so-called “technocrats” [44] become apparent. The Bosch cousins deduced that a similar situation would arise in the international context as a result of World War II. However, the circumstances varied greatly from country to country, depending on how the war had impacted their industry and their territory. Another influencing factor would be the aid they could receive from those territories where the war industry had favoured the development of certain materials [45].

However, the economic/industrial context is not the only factor influencing the application of one technique or another. Artistic trends, which are certainly related to technological developments but not entirely determined by them, also influence the choice of one construction solution over another. In post-Civil War Spain, a more historicist style of architecture was promoted, exalting national identity and favouring traditional techniques [40]. However, as modernity was gradually reinstated during the years of developmentalism and the opening up of the regime, historicism was less and less promoted by public administrations [41]. As for the rest of Europe and even the world, the post-war period (in this case, the world war) did not generally coincide with a promotion of historicism, but rather the opposite, a trend which, as we have seen, even Franco’s regime eventually joined.

This is the environment in which the patents are carried out by architects Bosch Reitg and Bosch Aymerich. In these circumstances, one objection that could have led to the patent being rejected would be that it is a traditional technique applied without major apparent modifications. In fact, this was the criticism that led the patent owners to reject its approval in Brazil and Uruguay (Table 1). However, this criticism does not seem to hold water when, as we have mentioned in the results, the inventors of the patent themselves have to recommend specialized construction companies for its implementation in Madrid. Or, as also mentioned in the previous chapter, Ricardo Fernández Vallespín advises against the implementation of the patent in Argentina, as he considers that expert knowledge of its construction process is necessary in order to carry it out (Figure 15).

The question at this point would be whether there were any previous patents equivalent to those that Ignasi and Josep Maria Bosch intended to register. In fact, this was the initial response they received in England, Germany, and the United States (Table 1). In the first case, it was possible to demonstrate that the patents held by Evelyn Hurden and Karl Pauli Billner were not exactly equivalent to the one the cousins were proposing. However, this was the reason that frustrated the patent in the other two countries, as the reviewers did not accept that the patents of Franz Dischinger in Germany and Ralph Hills in the United States were different cases from those proposed by the Catalan architects. In the first case, it was a reinforced concrete vault. In the second, although the vault was ceramic, the function of this material was fire protection, with the load-bearing function being performed by an internal reinforcement together with metal tubular sections protected inside the ceramic vault.

Paradoxically, in Spain, where there was a greater tradition of ceramic vaults, there was no objection from the patent examiners when it came to approving it. In fact, there were patents for ceramic systems in Spain for floor slabs, but the novelty of the Bosch vault was clearly understood.

Finally, the commercial success of the patent remained to be discussed. On this point, it must be acknowledged that, in light of the results obtained, it cannot be said with certainty that profits were made from the exploitation of these patents, despite their approval in up to seven countries.

As mentioned in the previous section, there is evidence of several works carried out in Spain based on the patent, but of these, all those that have been identified had been developed by the inventors of the system themselves (two-thirds of the total) (Figure 16). There are records of two claims for fraudulent use of the system without prior payment of exploitation rights, but there is no outcome of these claims. There are also records of requests for information about the patented system, but there is no evidence that these contacts resulted in payments for the use of the “Bosch Vault.”

Preliminary studies were carried out abroad on projects involving the implementation of the vault in France, Germany, Mexico, and Argentina. Of these, attempts were made to obtain patents in all but the last, and only the first was successful. However, all these designs were projects by Josep Maria Bosch Aymerich, which means that even if they had been implemented, they would not have required income from the exploitation rights of a possible patent. The fact is that in none of the six foreign countries where the patent was approved has any evidence been found of income from the exploitation of these patents (Table 1).

In the context of the internationalisation of post-war Spanish architecture [46] until the late 1950s, except for Jose Antonio Coderch y de Sentmenat (1913–1984), the greatest success stories were among exiles who, once naturalised, were no longer considered foreigners in their new host countries. Such is the case of Antoni Bonet (1913–1989) in Argentina, Josep Lluís Sert (1902–1983) in the United States and Félix Candela in Mexico (1910–1997). The latter’s example is related to patents for laminated vaults, in his case made of concrete, but which, unlike the Bosch vault, could be developed by their inventor himself, directly through his own company established in the country. A similar precedent, but in this case involving ceramic vaults, would be that of Rafael Guastavino (1842–1908) with his company Guastavino Fireproof Construction in the United States.

6. Conclusions

The Bosch vault is a system based on traditional techniques but with a degree of innovation that allows it to dispense with load-bearing walls. Its originality is evidenced by the fact that it has been patented in up to seven countries. Moreover, as stated in the results, the system enabled noticeable reductions in the weight of the structure and quantity of steel used (1.33 kg/m² versus 4 kg/m², which was the weight of other slab systems recommended by the government for the reduced use of steel in construction). However, it is not in the scope of this article to carry out an in-depth analysis of the structural behaviour of the Bosch Vault, other than verifying the results of the calculation methods used.

The system proved useful in post-war Spain, with more than 3000 homes built (270 in Sarrià, Barcelona; 527 in Gerona; 389 in Olot; 123 in Huelva; 722 en Valladolid; 650 inn Zona Franca, Barcelona; 650 in other areas of Spain), and aroused interest among developers and builders, both nationally and internationally (from France [16] p. 700; from Argentina [16] p. 534; from Mexico [16] p. 864). Thus, after the first year of exploitation of the patent in Spain, procedures began to export it to a dozen countries (Italy, France, Belgium, Portugal, United Kingdom, Germany, Canada, Mexico, Brazil, Uruguay, United States and Argentina) (Table 1).

However, it was precisely the degree of innovation of the system, which made its registration as a patent plausible, that ultimately required specific knowledge for its implementation, which prevented its successful exploitation in those places where it was approved. In Spain, at least two-thirds of the works carried out using the system were designed and directed by the inventors of the Bosch vault themselves (those built in Barcelona, Gerona, Olot and Valladolid), and abroad there is no record of any work that has been completed using the patent (Table 1).

This criticism coincides with that made by Ricardo Fernández Vallespín during the study to export the system to Argentina:

“The possibility of constructing all those buildings that have been built in Spain using your system is due to the fact that you ran a perfectly organized construction company, which had reliable technical staff (quantity surveyors and site managers). In order to introduce your construction method in Argentina, I believe it would be necessary to seek the collaboration of a construction company. It would be reckless to set up a new company to exploit a new system without the resources you have had at your disposal in Spain. I attribute much of your success to your personal qualities as an organizer and technician, and believe that your system, if applied by others who do not meet your standards, could have failed. Not because of the system itself, but because of the difficulties of organization”

[16] (p. 520).

Translation by the author from original:

“La posibilidad de realizar, todos esos edificios que se han construido en España con su sistema es debido a que Vd. dirigía una empresa constructora perfectamente organizada, que contaba con personal técnico de confianza (aparejadores y encargados de obra). Para introducir en Argentina su método de construcción, opino que sería preciso recurrir a la colaboración de alguna empresa constructora. Pues sería una temeridad constituir una empresa nueva para explotar un sistema nuevo, sin contar con los elementos de los que Vd. ha dispuesto en España. Porque atribuyo en gran parte el éxito que Vd. ha tenido a sus condiciones personales de organizador y de técnico, y que su sistema aplicado por otras personas que no reunieran sus condiciones podría haber fracasado. No por el sistema en sí, sino por las dificultades de organización”

[16] (p. 520).

Finally, changes in both economic and industrial circumstances and in historical and artistic circumstances meant that a system based on tradition was no longer favored, to the extent that there is no documentation on patents beyond 1956 (Table 1). This objection was expressed by Miguel Fisac in the “Critical Architecture Session” of 1954:

“Regarding brick vaults. It was wonderful that when our war ended, and despite the tremendous difficulties we suffered, all kinds of attempts were made to keep construction going and, therefore, to find whatever solutions were necessary. It is a case similar to that of gas generators in cars. Gas generators were installed because we were not being supplied with gasoline; but once we have gasoline, no one would think of continuing to use gas generators. It seems to me that we are in a similar situation with these vaults, which are absurd in the current circumstances because they have terrible acoustic conditions and because they leave wasted space that weighs down the building”

[39].

Translated by the author from original:

“Respecto a las bóvedas de ladrillo. Fué [sic] estupendo que al terminarse nuestra guerra, y existiendo aquellas dificultades tan tremendas que padecimos, se hicieran tentativas de todo género para no parar la construcción, y, por tanto, ir a las soluciones que fueran necesarias. Es un caso análogo al del gasógeno en los coches. Se pusieron gasógenos porque no nos mandaban gasolina; pero una vez que tenemos gasolina, el ir con gasógeno detrás no se le ocurre a nadie. A mi me parece que en un caso semejante estamos con esto de las bóvedas, que en las condiciones actuales son absurdas porque tienen unas malísimas condiciones acústicas y porque dejan unos espacios perdidos que gravan el edificio”

[39].

It is a happy coincidence that the two architects mentioned are teacher and disciple, respectively [47].