Jeremy Till's paper "Architectural Research: Three Myths and One Model" was originally commissioned by the Royal Institute of British Architects (RIBA) Research Committee, and published in 2007. In the past decade, however, it has grown in popularity not just in the UK, but around the world to become a canonical paper on architectural research. In order to help the paper reach new audiences, here Till presents an edited version of the original. The original was previously published on RIBA's research portal and on Jeremy Till's own website.
There is still, amazingly, debate as to what constitutes research in architecture. In the UK at least there should not be much confusion about the issue. The RIBA sets the ground very clearly in its founding charter, which states that the role of the Institute is:
The advancement of architecture and the promotion of the acquirement of the knowledge of the various arts and sciences connected therewith.
The charter thus links the advancement of architecture to the acquirement of knowledge. When one places this against the definition of research given for the UK Research Assessment Exercise (RAE), “research is to be understood as original investigation undertaken in order to gain knowledge and understanding”, one could argue that research should be at the core of RIBA’s activities. This essay is based on the premise that architecture is a form of knowledge that can and should be developed through research, and that good research can be identified by applying the triple test of originality, significance and rigor. However, to develop this argument, it is first necessary to abandon three myths that have evolved around architectural research, and which have held back the development of research in our field.
Myth One: Architecture is just architecture
The first myth is that architecture is so different as a discipline and form of knowledge, that normal research definitions or processes cannot be applied to it. “We are so unlike you,” the argument goes, “that you cannot understand how we work.” This myth has for too long been used as an excuse for the avoidance of research and the concomitant reliance on unspecified but supposedly powerful forces of creativity and professional authority. This myth looks to the muse of genius for succor, with the impulsive gestures of the individual architect seen to exceed the dry channels of research as the catalyst for architectural production. The problem is that these impulses are, almost by definition, beyond explanation and so the production of architecture is left mythologized rather than subjected to clear analysis. Architecture is limited to a form of semi-mystical activity, with the architect, as heroic genius, acting as the lightning rod for the storm of forces that goes into the making of buildings. This first myth also treats architecture as an autonomous discipline, beyond the reaches or control of outside influences, including those of normative research methodologies. This leads to the separation of architecture from other disciplines and their criteria for rigor. Self-referential arguments, be they theories of type, aesthetics or technique, are allowed to evolve beyond the remit or influence of accepted standards, and research into these arguments is conducted on architecture’s own terms.
The myth that architecture is just architecture, founded on the twin notions of genius and autonomy, leads eventually to the marginalization of architecture. A knowledge base is developed only fitfully and so architecture becomes increasingly irrelevant and, ultimately, irresponsible.
Myth Two: Architecture is not architecture
The second myth works in opposition to the first and argues that in order to establish itself as a credible and ‘strong’ epistemology, architecture must turn to other disciplines for authority. Architecture is stretched along a line from the arts to the sciences and then sliced into discrete chunks, each of which is subjected to the methods and values of another intellectual area. For example, the highly influential 1960s Oxford Conference on architectural education looked to scientific research as the means of establishing architecture within the academy. More recently architectural theory has immersed itself in the further reaches of critical theory in an effort to legitimize itself on the back of other discourses. In both these cases, and others that also rely on specific intellectual paradigms, architecture’s particularity is placed within a methodological straightjacket. In turning to others, architecture forgets what it might be in itself. The second myth, that architecture is not architecture, in editing the complexity of architecture thus describes it as something that it may not be. It is a myth fuelled by the funding mechanisms for research, with the various research councils defining acceptable areas through particular research paradigms, which simply do not fit the breadth of architecture.
Interestingly Myth One and Myth Two can and do operate in parallel, often within the same institution. Thus it is common to find the design core of a School of Architecture – playing out myth one - physically and intellectually separate from the ‘research’ core, with mutual antipathy between the two.
Myth Three: Building a building is research
The third myth is that designing a building is a form of research in its own right. It is a myth that allows architects and architectural academics to eschew the norms of research (and also to complain when those norms are used to critique buildings as research proposals). The argument to support this myth goes something like this:
- Architectural knowledge ultimately resides in the built object
- Every building is by definition unique and thus original
- The production of buildings can thus be defined as the production of original knowledge
- This is a definition of research
It is compelling enough an argument to allow generations of architects (as well as designers and artists) to feel confident in saying that the very act of making is sufficient in terms of conducting research, and then to argue that the evidence is in front of all our eyes if we would just choose to look. However, it is also an argument that leads to denial of the real benefits of research, and so it is worth unpicking.
- Architectural knowledge may lie to some extent in the building, but it also lies elsewhere: in the processes that lead to the building, in the representation of the building, in its use, in the theories beyond the building, in the multiple interpretations of the building and so on. Architecture exceeds the building as object, just as art exceeds the painting as object. Architectural research must therefore address this expanded field.
- A ‘good’ building is not necessarily good research. Architecture is often described as ‘good’ because it fits into known and tested canons of taste, type or tectonics. But this ‘goodness’ does not necessarily constitute good research, in so much as it is not particularly original or significant. A ‘good’ building, far from pushing towards new forms of knowledge, merely establishes or incrementally shifts the status quo. The various architectural award systems consolidate and perpetuate such attitudes.
- Good research may lead to ‘bad’ buildings. For example the technologies and construction procedures of food distribution centers are pioneering in many ways and based on systematic research into the various options, but the resulting buildings clearly do not fit received values of what constitutes good aesthetics or tectonics. Of course ‘good’ buildings dominate architectural culture, which means that the research lessons from the ‘bad’ buildings are hardly ever transferred across.
- If we take Bruce Archer’s definition of research (that it is “systematic inquiry whose goal is communicable knowledge”), then the building as building fails the test. While architects may believe that knowledge is there in the building to appropriated by critics, users or other architects, they very rarely explicitly communicate the knowledge. The knowledge remains tacit, thereby failing Archer’s second test of communicability.
Designing a building is thus not necessarily research. The building as building reduces architecture to mute objects. These in themselves are not sufficient as the stuff of research inquiry. In order to move things on, to add to the store of knowledge, we need to understand the processes that led to the object and to interrogate the life of the object after its completion.
Making Architecture Speak
Against these myths, one has to understand that architecture has its own particular knowledge base and procedures. This particularity does not mean that one should avoid the normal expectations of research, but in fact demands us to define clearly the context, scope and modes of research appropriate to architecture, whilst at the same time employing the generic definitions of research in terms of originality, significance and rigor.
The normal stretching of the field of architecture along the arts to science line (with the social sciences somewhere in the middle) results in each place along the line being researched according to a particular paradigm and methodology from the research spectrum. This ignores design, which is clearly an essential feature of architectural production; design cannot be so easily categorized as a qualitative or quantitative activity, but should be seen as one that synthesizes a range of intellectual approaches. Architectural research is better described by Christopher Frayling’s oft-cited triad of research ‘into’, ‘for’ and ‘through.’ Frayling developed this approach for design research in order to address the specific relationship between design and research. In this model, research ‘into’ takes architecture as its subject matter, for instance in historical research, or explanatory studies of building performance. Research ‘for’ refers to specifically aimed at future applications, including the development of new materials, typologies and technologies; it is often driven by the perceived needs of the sector. Research ‘through’ uses architectural design and production as a part of the research methodology itself.
Architectural research may be seen to have two main contexts for its production, the academy and practice. Research ‘in’ is traditionally the domain of the academy and research ‘through’ that of practice, with research ‘for’ somewhere in the middle. Research ‘in’ has the most clearly defined methodologies and research outcomes, but at the same time is probably the most hermetic. Research ‘through’ is probably the least defined and often the most tacit but at the same time a key defining aspect of architectural research. It is this area that needs developing most of all.
It is vital that neither academic or practice-based is privileged over the other as a superior form of research, and equally vital that neither is dismissed by the other for being irrelevant. (“You are all out of touch with reality,” says the practitioner. “You are muddied by the market and philistinism,” says the academic). There is an unnecessary antipathy of one camp to the other, which means that in the end the worth of research in developing a sustainable knowledge base is devalued.
The key to overcoming this problem lies in communication. Both the academy and practice often do not meet this central test for research: the academy because of its inward looking processes, practice because of its lack of rigorous dissemination. Whilst academic research is subjected to stringent peer review and assessment procedures, it has been argued that this had led to inward-looking results produced more for the self-sustaining benefit of the academic community and less for the wider public and professional good. On the practice side, much of the most innovative research in design and, particularly, technology is founded in practice. However, much of this research remains tacit; it is either, for commercial reasons, not shared with the rest of the community or else in its dissemination through the press is not communicated with the rigor it deserves. For the leading practices intellectual property is what defines them and sustains them, and they understandably are loath to give it away. Research goes on, but silently. The development of architectural knowledge happens but fitfully, and so the long-term sustainability of the profession is threatened. To avoid this, we need to make architecture speak.
A New Model for Architectural Research
As we have seen, the stretching of architecture across separate areas of knowledge does not address the particular need for architectural knowledge and practice to be integrative across epistemological boundaries. Buildings as physical products function in a number of independent but interactive ways – they are structural entities, they act as environmental modifiers, they function socially, culturally and economically. Each of these types of function can be analyzed separately but the built form itself unifies and brings them together in such a way that they interact. Research into architecture thus has to be conscious of these interactions across traditionally separate intellectual fields.
In order to give some clarity to the scope of architectural research, these interactions can be divided into three stages:
- Architectural processes
- Architectural products
- Architectural performance
The first stage, process, refers to research into processes involved in the design and construction of buildings, and thus might include for example issues of representation, theories of design, modeling of the environment, and so on. The second, product, refers to research into buildings as projected or completed objects and systems and might include for example issues of aesthetics, materials, constructional techniques and so on. The third stage, performance, refers to research into buildings once completed and might for example include issues of social occupation, environmental performance, cultural assimilation, and so on. The advantage of this model is that it avoids the science/art and qualitative/quantitative splits, and allows interdisciplinary research into any of three stages. The model thus breaks the hold of research method and allows instead thematic approaches to emerge. It is possible for scientist and historian, academic and practitioner, to contribute to the research into each of the three stages.
Most importantly the model also describes architecture temporally (as opposed to a set of static fragments), with one stage leading to another and, crucially, creating an iterative loop in which one stage is informed by another. For research to be most effective, and thus for architectural knowledge to develop, it has to feed this loop. For example:
- Research into performance in use informing the processes of design.
- Research into the products of design looking backwards to knowledge about the processes of design.
- Research into the performance of buildings being critically informed by knowledge of the processes of architecture.
A dynamic system thus emerges from this tripartite model, but it will only operate if academia and practice collaborate in order that the loop is continually fed with both data and analysis. However, this will open happen once we have cleared the three myths out of the way, and accept that architecture can, and should, be a research discipline in its own right, which both accords to the accepted criteria of research, but at the same time applies them in a manner appropriate to the issues at hand. There is some urgency in this, because as long as architecture fiddles around at the margins of the research debate, it will be confined to the margins of the development of knowledge. The present state of architecture, increasingly used to provide a velvet glove of aesthetics for the iron fist of the instrumental production of the capitalist built environment, is perhaps indicative that the state of marginality has been reached. The establishment of the discipline founded on research-led knowledge in the manner outlined above may be one small way of claiming a bit more of the center ground.
A version of this paper was first written as a position paper for the RIBA Research Committee, and subsequently published on the RIBA Research Wiki. This is the reason that it starts with the RIBA Charter.
- In an exhaustive research project carried by Edinburgh College of Art in 2004, a significant minority of responders still clung to this belief. See Jenkins. P, L.Forsyth and H. Smith., "Research in UK Architecture Schools – an institutional perspective," Architectural Research Quarterly, Vol. 9, Issue 1, March 2005, pp33-43
- Frayling, C., Research in art and design, (London: Royal College of Art: Research Paper, 1993)