In today’s debates on sustainable architecture, building materials occupy a particularly dominant position. Since they are what buildings are made of, they generally represent a building’s energy and resource consumption. Beyond this, however, is the need to minimise, as much as possible, the energy consumed in a building’s operation. Both these aspects of consumption are ostensibly technical matters, and so in recent years engineers have focused, above all, on making advances in material and energy efficiency.

However, the lifespan of buildings is usually left out of the discussion on sustainability, as is the far too strong a link between the design of a building and its intended use. In times of increasing resource scarcity, one finds simultaneously both a demand for buildings and an abundance of vacant buildings, often in the same place. The globally absurd problem of the astonishingly short lifespan of many buildings is a result of the very limited sustainability of the architecture itself, not that of its materials. For a robust architecture to establish itself in the form of a sustainable building, architecture’s relationship to use has to change. A building has to be able to demand a more intelligent contribution from the load-bearing structure which enables its architectural spaces, so that it might also enable other possible architectures in future. In this, the interaction between architects and engineers will be enlarged, oriented towards the joint planning of a more complex determination of a building’s paths of possibility. The structure, as the most permanent part of the building, represents an interlocking of space and time, with cycles of use and changing layers of different degrees of permanence, rooted in the site yet detached from the specific building, in the form of a raw meta-landscape.

Constrained skeletons 

When we talk about buildings, we often speak of a ‘roof over one’s head’, habitats and places of refuge, provision and infrastructures. Underlying this image of refuge are functional aspects connected with people’s basic needs, i.e. the provision of water and heat, light and air, but above all a basic stability. The most solid component of the structural envelope around the human being, the load-bearing structure, is not always a visible or special part of the envelope. In stone buildings and log buildings, structure is wholly integrated within the overall fabric. In skeleton-framed buildings, we see the load-bearing structure most clearly: in timber frame buildings, steel and early concrete buildings, structures of linear elements clearly dominates. While the role of the envelope for representation and weather protection became far more important, that of the skeleton did not change: it was only supposed to reliably hold up. With modernism’s programmatic detachment of the building envelope from its supporting structure, the skeleton as a space-forming factor moved to the centre of architectural production. The term ‘structure’ reveals its original connection with the process of building: struere, meaning “to build”, but also to make an “overall layout, inner structure, [or] arrangement”. The central role of the skeleton becomes visible when we are dealing with the alteration of existing buildings or ruins. Through alteration, the skeleton is uncovered, and this unchangeable part is then operated on, built upon. The uncovering can be compared to the work of archaeologists, the intervention to that of surgeons.  

Load-bearing structures have often been compared to bones, indeed the frequent reference to skeletons clearly indicates this.1 The bones inside the body support and give stability. And they are what survives, when all other parts of the body are gone. It is the bones that hold the living together and upright, while also symbolising death. Despite their significance, by themselves bones form only a very loose and unconnected collection of stable individual objects. Only in the functional connection with the body parts which surround them do they form a stable framework. The bone supports, while muscles and ligaments hold them together and in position: a complex pattern of interaction between one another and upon one another.A

Separate structures 

In this context, the bones which serve as an inner load-bearing structure can also be understood as part of an infrastructure, i.e., the underlying, organising system. The term, which was first used in France of the mid-18th century, vividly illustrates the intertwining of far-reaching mechanisation and administration. The French state organised the structure for transport and important facilities centrally, and for the first time it began to train the specialists to plan such things. The École nationale des ponts et chaussées in Paris had been training engineers since 1747, and the École polytechnique since 1794. This was the beginning of an intensive specialisation, caused on the one hand by the development of the modern state, and on the other hand by the increasing role of science and technology in construction. In the course of this professionalisation, the art of bridge building, like that of other kinds of infrastructure, was separated from the art of building to become a core subject of the formative discipline of civil engineering. Technical construction was increasingly able to develop without an architecture that embraced and contextualised it. The splitting off of fields of knowledge and their deepening within the sciences also repeatedly involved the separation of areas within these fields that were otherwise firmly connected. When, in the 18th century, the encyclopaedists compiled their treatise of contemporary knowledge, they also depicted construction knowledge, drawing dividing lines between crafts and the latest professions.2

Functional and temporal layers 

Just like bones, load-bearing structures only have a meaning in a functional context. Only when they are embedded in an effective structure do they become meaningful in what for which they were intended. This can be seen in recurring discussions in building conservationH on the survival of unused cultural monuments or the prevalence of vacant office buildings in many city centres. But not only do structures and their buildings need a function, they also need a flexibility for functional change. Stewart Brand concluded in his influential study of how buildings change over time: “All buildings are predictions, and all predictions are wrong”.3 The narrow, functional-spatial planning does not stand up to the reality of use that actually follows. In their constructive entanglement, facade, building services, load-bearing structure and fittings very often cause problems when they later need to be adapted. Depending on the frequency of adaptations due to changes in use or maintenance, these functional layers can also be regarded as knotted temporal layers, which are each subject to their own rhythms of existence. Such layers have been called ‘shearing layers’B – a term coined by the architect Frank Duffy and elaborated by Brand. The more tightly the layers are connected, the greater the problem: “Because of the different rates of change of its components,” says Brand, “a building is always tearing itself apart.” Ideally, therefore, components of different functions within the building should remain independently manageable, which means, for example, that the load-bearing structure remains separate from the building services. In the case of fire, structure is of course only one of the many functional layers. In its permanence, it must, for instance, allow for rapid change around it through spatial generosity. Detached disciplinary thinking can therefore even be helpful here. It can make it possible to think of structure as a space of possibilities for various specific architectures, to each of which it offers particular qualities, but always outlasts these architectural constellations. In this way, structure and architecture never decouple spatially and sensually, but only temporally.  

What do the possibilities for change mean in concrete terms for load-bearing structures? Since they are so much firmer than the softer layers of building, it has to be considered what might be expected of them in a later context. How does everything around the structural framework change and how does the skeleton determine this process? When dealing with an existing building – in its reuse and transformation – one can see that the significance of its internal structure is about far more than the stable support of the spaces it holds. Beyond this initial firm and stable support of the created spaces and their uses, is its eventual appropriation when there is a need or desire for the building to change. In its layout and design, the skeleton would therefore have to accommodate all likely contextual transformations within itself. This does not at all mean that load-bearing elements and the structure as a whole must thereby be as neutral and abstract as possible. These imprints of possible future changes and transformations can even be particularly generative of form. Their forms, as a legible totality, can be a superimposition of contingencies that the structure provides initially, later, or possibly never. This kind of structural redundancy was common practice in design until the establishment of the technical–scientific disciplines surrounding construction. In this reading, the structure represents what is currently happening (what it is used for), but also what may have had to be dealt with earlier or what is yet to come. As a circular strategy, this type of structural design means that the skeleton remains in place, so that it continues to be available as a framework for every subsequent use. At most, only lighter elements of the envelope and interior spatial layers would be removed and replaced. The movement of the most energy-intensive functional layer – that of the load-bearing structure – is thus kept as little as possible.

Change as load case 

This simultaneity of the non-simultaneous is also today fundamentally the practice of scientific, indeed modern, structural design. Unfortunately, it is usually limited to the modelling of loads. We define requirements that arise from uses, the effects of winds or earthquakes and equip buildings with a structure that then has the appropriate capacities for precisely these requirements. This concerns the shape of the parts and the whole as well as their materials. But it is precisely in this purely technical view, which is a consequence of the isolated disciplinary way of thinking already mentioned, that the banality of a structure often lies. So while we project the external impacts onto our present buildings as future possibilities, we accept their internal arrangements for use as narrow corridors of possibility that hardly allow for any alternatives. Structural design unfortunately still mostly designs for robustness with generously dimensioned building components or resistant building materials instead of finally getting involved in spatial – and especially temporal – variation strategies of buildings. This is because structural design is technically made overwhelmingly in reference to the origin of a building’s history. The load-bearing structure developed from an overarching architectural concept is a direct translation, often even a diagram, of the requirements for the (initial) use of the building.  

This is precisely where the higher-level, actually creative approach of the engineer in the structural design of an adaptable building becomes apparent: instead of translating uses into loads and dimensions, spatial fields of function should determine the component arrangements and dimensions and allow possible later breakthroughs to be strategically made in parts of the building. It is then also a question of determining what these mean in terms of costs and effort and where these contingencies make sense in the fit-out. How would a permanently meaningful load-bearing structure that emerges from a stable architecture and use look like? How could a functionally stable structure be made that would function for as many conceivable demands of the building and probable conversions as possible without costly adaptations? And how can it be durable beyond its material qualities? Strong structures are so valuable architecturally that they significantly shape the identity of spaces and buildings. This means that we have to understand a structure not only as a technical apparatus that silently solves problems, but also as a carrier of meaning as the most permanent aspect of a building over generations of uses.

Building skeleton as a porous mass 

Load-bearing structures – a self-evident aspect of architecture – embrace and permeate all parts of a building. It thus frames the layers of meaning of the building, in that it not only proverbially holds the building together, but also acts with its necessary permanence through all the episodes of a building’s history. In its pervasiveness and homogeneous presence, but above all in its non-negotiable physical existence (while many values change), the structure itself becomes the rigid body with which and within which architecture happens. The arrangement of building elements and voids resembles the modelling of a porous mass. In this sense, the skeleton is what remains when the mass is dissolved as far as possible. 

To describe how a building can be reconfigured and adapted – ideally by a variety of different users – it must also be understood as a permeable structure. This approach follows Richard Sennett’s thinking on the porosity of the city,4 describing the permeability of buildings and their role in separating private and public space within the city. Sennett uses the Nolli Plan to describe blocks of buildings as closed or permeable bodies that reveal the boundaries of the built environment. In this sense, a building can also be thought of as a mass hollowed out by perforations,C which, depending on its configuration, creates different interior spaces and connections between them.

For Herman Hertzberger, it is public space that holds the city together – yet he also saw that public space was subordinated to the logic of the street network that represents the most enduring part of the urban fabric.5 If we thus understand such corridors on the one hand as dividing lines of private and less private spaces in buildings, and, on the other hand, as the most permanent zones in which access is mediated, then it is the interplay between access and structure – in conjunction with their associated spectrums of use – which becomes the possibility space for building configurations.  

Like urban blocks, street facades and buildings themselves, the walls and slabs of a structure aren’t ever entirely closed, but have numerous openings for a variety of reasons, e.g. for utility shafts or doors. But more than that, these structural surfaces (especially those from the first half of the 20th century) often have an internal substructure, as they consist of a skeleton filled with further elements. As a porous structure, the skeleton – originally chosen for economy – allows for much greater flexibility in later conversions, as openings can be made more easily. The hierarchical structure of the ribbed slab thus becomes an architectural membrane that allows for specific forms of permeability as needed. This generic, non-specific space-formation is the real intelligence of the building component, which does not respond to rigid load ideas, but creates windows of possibility that are interwoven with the structural idea and the building process. These internal possibilities for change made by hollowing out must be included in the discussion of the porous structure, without, of course, compromising the overall integrity of the structure. Consequently, there are three ways in which a building can be made permeable: 1) towards its surroundings, through entrances and exits but also through open or closed facades in general, 2) in the traversing of the building, with corridors to connect uses, and 3) as the possible penetration of space-enclosing building components such as ceilings or walls in order to connect units of use horizontally and vertically. The concept of porosity can help to identify what should be permanent structural elements and how to translate them into an appropriate material. The load-bearing skeleton itself then does not necessarily have to consist of demountable, re-usable components, because it remains in place. All other softer layers, on the other hand, should be made suitable for future re-use.

Intelligent ruin 

But, structure can also be conceptually transposed onto a larger scale the other way around. Due to its permanence, in its overarching meaning the structure then quite naturally detaches itself from the momentary building. As the most rigid part of the building, it belongs more to the outside, to the urban environment, rather than to the inside and its current use.G From a macro perspective, the load-bearing structure becomes a building block for neighbourhood and urban development, in which spectrums of use are created and appropriations are made possible. These respond to those of the neighbours, allow continuity between buildings and guide them through their own internal order. The orientation of the building and thus the order of accesses can change, users and user combinations fluctuate and floor plans shift with new fixtures or cross-connections. The structures of buildings could accordingly be read as a meta-landscape; one which determines usability and relationships, but at the same time can be redesigned at any time, and so established has an effect both inwards on its spaces and outwards towards its neighbourhood. Following Brand’s model of shearing layers, the almost permanent structure thus merges with the permanent site; it becomes an extended building site, with all its preconditions for an ongoing process of building.D  

The Belgian architect Bob Van Reeth has characterised such changeable structures as intelligent ruins. “A building is a possibility, is conducive, preferably taciturn, silent, is willing, liberates space, mediates. Building as intelligent ruins. Suitable for use, fit for purpose, as Charles Voysey expressed it. Good buildings hide daily use, they are stable and stubborn, obstinately distributive and (following Kant) ‘purposive without
 a purpose’. Therein lies the quality of their durability, of their cultural durability, which yields dignity. Expediency asks for the right scale, an utmost precision that leaves everything open that cannot be predicted.”6 In this sense, the design of a structure is not only the conceptual frame of space, time and scale, but also the basic tactile order that transcends the various episodes of a place: uses, materials, people flow through it, condense, grow, and find themselves anew.  

We know of many successful conversions, especially of old industrial buildings whose robust structures carry a much higher value than the architects attributed to them at the time. Just think of Lina Bo Bardi’s cultural centre SESC PompéiaE in São Paulo. When Cedric Price presented his vision of architecture for society in the 1960s, he showed lattice structures in which various uses were indicated, envelopes of space, large and small, that seemed to grow into each other. There, people appropriated the Fun Palace for their own purposes.F But we should not automatically think of large halls when we think of transformable structures, because the necessary transformation affects every building. Price believed that no building should be eternal, as a building only ever serves particular needs and will, at some point, naturally reach the end of its life. Today, with the efforts towards sustainable architectures (and structures), we have somewhat different requirements, namely to meaningfully extend the use-cycles of buildings and their components. In addition to the urgent discussion about sustainable materials and the obvious focus on reusable components, we thus also urgently need to rethink the currently far too narrow conception of the durability of structures and functions. The structural engineer can make a far more profound contribution than simply making structures leaner and connections more flexible; but rather, they can also differentiate structures, i.e. link them to spectrums of use and changing contexts and thus understand them as easily appropriable, intelligent ruins. In this, they must be supported by others with design agency over the building and those who determine the trajectories of the city and the neighbourhood. If we were to interweave the vision of changeable buildings with Aldo Rossi’s idea of the city growing over time, with its elements from different periods, we start to get a rough vision of potential strategies for living buildings and cities and the role of technology. Structures are an intrinsic part of architecture. One cannot make sense without the other, nor could they endure.