Introduction

Architecture is often discussed as a finished object: a building photographed, visited, praised or criticised. Yet every building is the end point of a long chain of decisions, drawings, models, negotiations and revisions. Architectural theory, if it is to be more than a catalogue of styles, must therefore explain how architecture is thought into existence. This essay approaches the theory of architecture from that angle. It treats the design process not as a neutral technical procedure but as the place where theory is tested, where ideas about beauty, function, society and technology become concrete choices.

The argument unfolds in five movements. First, it traces how the classical treatise tradition linked theory to the act of making. Second, it examines the twentieth-century attempt to turn design into a rational, scientific method. Third, it considers the critique of that attempt through the concepts of wicked problems and reflective practice. Fourth, it looks at the typical stages of the process (analysis, concept, development and representation) as sites of theoretical thinking. Finally, it discusses innovation, in the form of computational design, material research and socially engaged practice, and asks what these developments mean for the discipline. The central claim is that architectural theory is best understood as a theory of design thinking: an evolving account of how architects frame problems, generate possibilities and judge outcomes.

The Classical Foundations: Theory as Guidance for Making

The oldest surviving architectural treatise, Vitruvius’ Ten Books on Architecture, already presents theory and practice as inseparable. For Vitruvius, the architect needs both fabrica (the work of the hand) and ratiocinatio (the reasoning that explains why a thing is made as it is). His well-known triad of firmitas, utilitas and venustas, usually rendered as firmness, commodity and delight, functions less as a definition of beauty than as a set of criteria against which a design can be evaluated (Vitruvius, 1999). A design must stand up, serve its purpose and please the senses, and the architect must hold these demands in balance.

The Renaissance deepened this connection. Leon Battista Alberti, writing in the fifteenth century, separated the idea of a building from its material execution. He argued that the architect’s task is to devise, through lineaments and structure, a coherent whole in the mind and on paper before any stone is laid (Alberti, 1988). This separation of conception from construction is arguably the founding moment of the modern design process. It established the drawing as the architect’s primary instrument and made the project, rather than the finished building, the architect’s real product. Much subsequent theory, from Palladio’s proportional systems to the academic composition taught at the École des Beaux-Arts, can be read as an effort to systematise how that mental and graphic construction should proceed.

What these early sources share is a conviction that architecture rests on transmissible principles such as proportion, order and appropriateness. The process of design was a matter of applying and adapting such principles to a particular site and patron. This confidence in stable rules would be challenged only when industrialisation, new materials and new social demands made inherited rules appear insufficient.

The Quest for a Scientific Method

The twentieth century produced a different ambition: to make the design process itself an object of rigorous study. Modernist architects had already argued that form should follow from function and from the logic of industrial production. After the Second World War, this faith in rationality merged with advances in operations research, systems theory and early computing to produce what became known as the design methods movement.

Christopher Alexander’s Notes on the Synthesis of Form (1964) is a landmark of this period. Alexander proposed that a design problem could be understood as the effort to achieve a good fit between a form and its context. By listing the requirements that bear on a problem, mapping their interactions and decomposing them into clusters, a designer could in principle arrive at a form that resolves the misfits systematically. J. Christopher Jones (1970) similarly catalogued methods, from brainstorming to checklists, intended to make design more transparent and less dependent on inspiration. Herbert Simon (1969), writing from the perspective of the sciences of the artificial, defined design in strikingly general terms as the devising of courses of action aimed at changing existing situations into preferred ones. On this view, architecture was one instance of a universal problem-solving activity.

The significance of this movement for theory is considerable. It insisted that design knowledge could be articulated, taught and criticised, rather than being treated as the private mystery of the gifted artist. It also introduced vocabulary that remains in use, such as constraints, criteria, alternatives and optimisation. However, the movement’s confidence soon met resistance, including from some of its own founders. Alexander himself later distanced himself from rigid decomposition, turning toward the pattern language and the idea of living, incremental growth. The difficulty lay in the assumption that a problem could be fully specified before a solution was sought.

Wicked Problems and the Reflective Practitioner

In 1973, Horst Rittel and Melvin Webber offered a decisive critique. They argued that problems in planning and design are “wicked” rather than “tame”. A wicked problem has no definitive formulation, no stopping rule that tells the designer the work is finished, and no solution that is simply true or false, only better or worse. Each attempt at a solution changes the problem, and every such problem is in some respects unique (Rittel & Webber, 1973). Anyone who has seen a brief change after the first sketches appear, or a site reveal unexpected conditions halfway through a project, will recognise the description. Architecture is the paradigmatic wicked discipline, because it combines technical, social, economic, aesthetic and political questions that cannot be separated neatly.

Donald Schön gave this insight a positive account. In The Reflective Practitioner (1983), he contrasted the model of technical rationality, in which professionals apply scientific knowledge to well-defined problems, with the actual behaviour of skilled practitioners in fields such as architecture. Observing design studio teaching, he described designing as a “conversation with the situation”. The designer makes a move, such as drawing a line or placing a volume, and the drawing talks back by revealing consequences that were not foreseen. The designer then reflects on this response and adjusts. Schön distinguished reflection-in-action, which happens during the making, from reflection-on-action, which takes place afterwards, and he argued that the tacit knowledge involved cannot be reduced to rules.

Empirical studies support this picture. Bryan Lawson’s (1980) comparisons of architects and scientists found that scientists tended to analyse a problem in order to find the rule governing it, while architects tended to propose solutions in order to learn about the problem. Nigel Cross (2006) summarised decades of research under the heading of “designerly ways of knowing”, emphasising that designers work with ill-defined problems, use solution-focused strategies, and rely on non-verbal media such as sketches. Peter Rowe’s Design Thinking (1987) examined the working methods of well-known architects and identified recurring heuristics, among them the use of precedent, organising ideas and typological reasoning, which designers draw on to bring order to open-ended situations.

Taken together, these contributions shift the theoretical centre of gravity. Theory is no longer a body of rules that precedes design but a way of understanding the inquiry that design itself performs. The question becomes how designers frame a situation, what repertoire of examples and ideas they bring to it, and how they evaluate what emerges.

Stages of the Process as Sites of Theory

Although real design is iterative rather than linear, it is helpful to examine the stages that architects conventionally recognise, because each embodies theoretical assumptions.

The first is analysis and programming. Here the architect studies the site, the client’s needs, the regulations and the cultural setting. Theory enters through the choice of what counts as relevant. Kevin Lynch’s work on the image of the city and Christian Norberg-Schulz’s writing on the genius loci, for instance, direct attention to perception, memory and the character of place rather than to measurable quantities alone. The act of framing the problem is already a theoretical act: a brief that emphasises energy performance will generate a different building from one that emphasises civic symbolism, even for the same site.

The second stage is concept generation. Architects often rely on a guiding idea, which may be a metaphor, a diagram, a spatial parti or a reference to a precedent, to hold the project together. Rowe (1987) shows how such organising ideas allow designers to make early commitments without full information. Gabriela Goldschmidt (1991) described the “dialectics of sketching” as an interplay between seeing-as and seeing-that: the designer sees an ambiguous line as a wall, a path or a courtyard, then tests that interpretation against requirements. The concept is therefore a working hypothesis, open to revision, and not a final answer.

The third stage is development, in which the idea is tested against structure, materials, environment, cost and construction. This is where Vitruvian firmness becomes technical detail and where the tension between ideal form and material reality is most visible. Frei Otto’s form-finding experiments with soap films, hanging chain models and tensile membranes illustrate a design method in which the form is discovered through the behaviour of materials rather than imposed upon them (Otto & Rasch, 1995). His lightweight roof for the 1972 Munich Olympic Park shows how a process based on physical experiment can generate an architecture that is both efficient and expressive.

The fourth stage concerns representation. Plans, sections, perspectives, physical models and digital simulations are not neutral records; they shape what can be imagined. Alberti’s drawing, Schön’s sketching conversation and the contemporary parametric model are all media that make certain thoughts possible and others difficult. For this reason, a change in representational technology is never only a change of tools. It alters the design process and, with it, the theory that describes the process.

Innovation I: Computation and the Transformation of the Process

The most visible innovation of recent decades is the integration of computation into design. William Mitchell (1990) anticipated this development in The Logic of Architecture, arguing that computers could be used to represent and manipulate the rules underlying architectural form, extending the formal-analytic tradition of earlier theorists. In the 1990s, designers such as Greg Lynn (1999) explored continuous, animated geometries made possible by software, while Frank Gehry’s office adopted CATIA, a program developed for aerospace engineering, to translate sculptural physical models into precise, buildable data. The Guggenheim Museum Bilbao (1997) became the emblem of this shift. Its titanium-clad curves could be documented, fabricated and assembled within a conventional budget only because the digital workflow connected the designer’s gestural models to the production process.

A further step distinguishes using computers to draw from using them to think. Kostas Terzidis (2006) separates computerisation, in which existing procedures are automated, from computation, in which the designer writes rules and algorithms that generate formal possibilities. In such work, the designer’s creativity lies partly in devising the system rather than shaping a single outcome. Patrik Schumacher (2011) built on this logic to propose parametricism as a style and a theory, in which buildings are conceived as fields of interrelated, adjustable variables. Whatever one thinks of the polemical claims attached to the term, the underlying idea has altered practice. A parametric model allows hundreds of variants to be produced and compared, turning the design process into the exploration of a space of options.

Mario Carpo (2017) offers a historical interpretation. He suggests that digital fabrication and data-driven design are overturning assumptions that have governed architecture since the Renaissance, namely that a design is a fixed, authored notation which is then reproduced faithfully. In the first digital turn, smooth curves and mass customisation became affordable. In the second, the emphasis moves to large datasets, machine learning and search, in which the designer curates and selects among outputs that no human could have drawn by hand. Carpo’s account is useful because it relates technical change to the deeper theoretical question of authorship. If Alberti’s separation of design from execution defined the modern architect, then a workflow in which design and fabrication are continuous, and in which software proposes alternatives, forces a reconsideration of that role.

Research-led practice has extended these ideas into material and structural experimentation. At the University of Stuttgart, the Institute for Computational Design and Construction and the Institute of Building Structures and Structural Design have developed research pavilions whose geometries are derived from the biomechanics of natural structures such as sea urchin plates and beetle shells, and are built through robotic fabrication (Menges, 2015). Here the process is a closed loop of simulation, prototyping, making and measurement, and the building functions as an experiment that generates new knowledge. This model recalls Otto’s physical form-finding, now amplified by computation, and it shows that innovation often occurs when the design process is itself redesigned.

Innovation II: Participation, Incremental Design and Critical Regionalism

Innovation is not limited to technology. A second line of development concerns who participates in design and what values guide it. The Chilean office Elemental, led by Alejandro Aravena, offers an instructive example in its housing project at Quinta Monroy in Iquique (2004). Faced with a very limited budget per family, the architects designed only half of each house, the part that families could not build for themselves, and left the remainder to be completed incrementally by the residents (Aravena & Iacobelli, 2012). The key innovation lies in the framing of the problem. Rather than attempting to deliver a small complete dwelling, the team redefined the task as providing a good starting point that would gain value over time. This is a clear instance of Rittel and Webber’s insight that the way a problem is formulated determines the range of possible solutions.

Another strand concerns place and culture. Kenneth Frampton’s (1983) essay on critical regionalism argued for an architecture that uses modern technique while resisting both the homogenising force of universal civilisation and the sentimental imitation of local tradition. In process terms, this requires a design method that listens closely to climate, topography, craft and material availability, and that treats these as generators of form instead of obstacles. Contemporary concerns about sustainability reinforce the point. Environmental simulation, life-cycle assessment and low-carbon material research now enter the process early, no longer as late verification, so that performance becomes a driver of concept rather than a constraint added afterwards.

A Critical Perspective: Technique and Human Experience

Not all commentators welcome the growing technical sophistication of the design process without reservation. Juhani Pallasmaa (1996), in The Eyes of the Skin, argued that modern architecture has been dominated by vision and has neglected touch, sound, smell and the bodily experience of space. The danger, in his view, is an architecture of striking images that does not support the inhabitant’s sense of being in the world. Powerful rendering and modelling tools can amplify this danger, because what is easy to depict on a screen may differ from what is rewarding to occupy.

This criticism does not require rejecting technology. It supports Schön’s insistence that the design conversation must stay anchored in judgement. Tools can generate options, but deciding which options deserve to be built depends on values: what a community needs, what a place deserves, and how a space will feel in use. Post-occupancy evaluation, in which completed buildings are studied in use, provides one way of feeding such experiential knowledge back into the process, closing the loop between theory, practice and life.

Conclusion

Viewing architectural theory through the design process reveals a continuous but changing relationship between thinking and making. The classical treatises offered principles and criteria to guide the architect. The design methods movement attempted to convert those principles into a systematic procedure. The critique of wicked problems and the theory of reflective practice recognised that architectural problems resist full specification and that designers learn through the act of designing. Studies of design cognition have since confirmed that architects work by proposing, testing and reframing, and the stages of analysis, concept, development and representation can be understood as moments within this cycle of inquiry.

Innovation, in turn, can be understood as a change in the conditions of that cycle. Computation expands the number of options that can be generated and tested, and collapses the distance between design and fabrication. Material research turns buildings into experiments. Participatory and incremental strategies redefine the problem and the roles of those involved, while critical regionalism and environmental thinking widen the values that guide decisions. Against these developments, phenomenological critique reminds designers that the ultimate measure of architecture is lived experience.

The lesson for students and practitioners alike is that theory and process are not separate domains. A theory of architecture is, in the end, an account of how to design well under uncertainty, and the most meaningful innovations are those that deepen the architect’s capacity to frame problems, to learn from the evolving work, and to remain responsible to the people who will inhabit it.

References

Alberti, L. B. (1988). On the art of building in ten books (J. Rykwert, N. Leach, & R. Tavernor, Trans.). MIT Press. (Original work published 1485)

Alexander, C. (1964). Notes on the synthesis of form. Harvard University Press.

Aravena, A., & Iacobelli, A. (2012). Elemental: Incremental housing and participatory design manual. Hatje Cantz.

Carpo, M. (2017). The second digital turn: Design beyond intelligence. MIT Press.

Cross, N. (2006). Designerly ways of knowing. Springer.

Frampton, K. (1983). Towards a critical regionalism: Six points for an architecture of resistance. In H. Foster (Ed.), The anti-aesthetic: Essays on postmodern culture (pp. 16–30). Bay Press.

Goldschmidt, G. (1991). The dialectics of sketching. Creativity Research Journal, 4(2), 123–143.

Jones, J. C. (1970). Design methods: Seeds of human futures. Wiley.

Lawson, B. (1980). How designers think. Architectural Press.

Lynn, G. (1999). Animate form. Princeton Architectural Press.

Lynch, K. (1960). The image of the city. MIT Press.

Menges, A. (2015). The new cyber-physical making in architecture: Computational construction. Architectural Design, 85(5), 28–33.

Mitchell, W. J. (1990). The logic of architecture: Design, computation, and cognition. MIT Press.

Norberg-Schulz, C. (1980). Genius loci: Towards a phenomenology of architecture. Rizzoli.

Otto, F., & Rasch, B. (1995). Finding form: Towards an architecture of the minimal. Axel Menges.

Pallasmaa, J. (1996). The eyes of the skin: Architecture and the senses. Academy Editions.

Rittel, H. W. J., & Webber, M. M. (1973). Dilemmas in a general theory of planning. Policy Sciences, 4(2), 155–169.

Rowe, P. G. (1987). Design thinking. MIT Press.

Schön, D. A. (1983). The reflective practitioner: How professionals think in action. Basic Books.

Schumacher, P. (2011). The autopoiesis of architecture, Vol. 1: A new framework for architecture. Wiley.

Simon, H. A. (1969). The sciences of the artificial. MIT Press.

Terzidis, K. (2006). Algorithmic architecture. Architectural Press.

Vitruvius. (1999). Ten books on architecture (I. D. Rowland, Trans.). Cambridge University Press.