Following Patrick Schumacher’s manifesto unveiling at the Venice Biennale in 2008 much has been written and debated about ‘parametricism’, ‘parametric design’ and as part of those ‘parametric urbanism’. This advent-garde agenda has quietly begun to make its way into architectural discourse over the past 15 years, especially within progressive schools of architecture in Western Europe and the United States and with the coming to prominence of Zaha Hadid Architects. At the biennale Schumacher, director at Zaha Hadid Architects and leading contemporary writer, architect and theorist, enunciated Parametricism as the ‘next great style after modernism’. This movement has been made possible by the development and advancements in digital design tools which have existed in other industries for some time but have found their way into offices of architecture only recently. Software such as Bentley’s Generative Components and McNeel’s Grasshopper for Rhino have brought associative and generative modelling tools to the realm of architecture and urban design, and, whilst associative modelling is often considered as the main tool in a parametric tool set, other advancements in scripting, simulation and agent based modelling have also contributed to the research and development of the movement. However, despite these tools now making parametric and generative design possible, this essay would like to explore whether they have a place in urban design and whether we have in fact been designing parametrically intuitively for decades. The design of cities have to take into account many different variables which all have to be juggled and be resolved if it is to be a successful urban space. To explore this notion of urban design being a parameter driven process, this text will examine and contrast parametric design to what is seen as possibly the least progressive movements in contemporary urbanism, New Urbanism. New Urbanism has had its many critics due to its neo-traditional approach to urban design which looks to recreate the towns and villages of an older generation through the replication of historical street patterns and vernacular architecture. In spite of this traditionalist approach, it is the aim of this text to see if New Urbanism can be read in terms of a parametric system and explore whether these two seemingly different movements share some ideological, philosophical and design ideals. 

This discussion is important as there is current debate within the profession about in which direction it should be heading. It has been argued that as a society we haven’t had an overarching design movement to guide progress since the decline in modernism during the 1970s and 1980s. Patrick Schumacher describes this as ‘the crisis of modernism’ and argues that since then we have only had a series of transitional styles such as postmodernism, deconstructivism and minimalism. Schumacher looks to utilize this ‘crisis’ by pushing parametricism as the successor to modernism but it could also be argued equally that exponents of New Urbanism have looked to exploit the crisis by extolling the negative impact contemporary architecture and urban design has had on society and through exemplifying traditional forms of urbanisation as a model for progression. This debate has been seen recently in the development of Chelsea Barracks in London where Lord Rogers’ contemporary design was criticised by Prince Charles, a prominent New Urbanism supporter and critic of modern architecture and urbanism, who favours a neo-traditional design by Quinlan Terry. The Qatari royal family, who owned the investment company heading the project, subsequently withdrew the planning application for the Rogers scheme. The whole saga has played out and been debated not only in the trade press but also national newspapers also, bringing the debate into the public consciousness.

To evaluate and attempt to read the neo-traditional New Urbanist movement in terms of the contemporary techniques of parametric design, it is important at this point to make two distinctions. First, this essay will examine a range of parametric design tools, which shouldn’t be confused with ‘parametricism’. Parametricism is a movement and style that has a strong aesthetic agenda and whilst it is true that parametricism cannot exist without parametric tools and software, parametric software can, and does, exist without parametricism. Secondly, parametricism will be considered as a frame of reference when discussing New Urbanism as a movement and set of theoretical and philosophical ideals but whilst both parametricism and New Urbanism are both movements only Parametricism claims to be an aesthetic style as well. 

What is parametric design?

Parametric design (sometimes known also as generative design) is an established concept in computer aided design (CAD) and originates in the aeronautical industry. In its simplest form, parametric design involves setting up a complex set of relationships between various elements within a 3D model. Lars Hesselgren, practising architect and co-founder of the Smart Geometry group explains the process of designing using these tools as:

“We’re modelling relations, and building intelligence into those relations. This is not about modelling new geometries – this is a specific skill. Generative Design is not about designing a building it is about designing the system that designs a building.” (Littlefield, 2005)

This process is a paradigm shift from the traditional method of designing buildings where the control and manipulation of geometry was the main preserve of the designer, with these tools geometry is the result of the relationships between all the smaller elements that make up the composite whole. The process can equally be used for designing urban spaces as it can for designing individual buildings. In terms of architecture and urbanism, a more refined set of tools such as agent-based modelling, scripting and simulation tools have developed but all aim for the creation of geometry and order through the setting up of variables and relationships between the different elements within the system. The advantages of such systems are obvious, once the rules have been established different design options can be explored through the manipulation of parameters which proliferate through the system making it easy, quick and productive to test and explore varying options, iterations and scenarios. 

Are these tools compatible with traditional urban design techniques?

 On the surface of things such progressive and cutting-edge tools seem to point towards the future rather than the past but haven’t we been designing parametrically for years? Since the onset and formalisation of a planning system, a case could be made for the fact that we have been designing to a set of parameters both on a conscious and subconscious level. It is these pieces of legislation, frameworks, guidance and policy that have shaped our urban environments throughout the last century. Although there hasn’t been a tool where the values can be input and a layout produced, these documents have had a significant impact on the outcome of urban design, more so than in any other discipline. One example of this is design codes that are applied to individual developments or areas and although the use of design codes is not a new thing, they have come to prominence again recently. The Office of the Deputy Prime Minister define design codes as:

“a set of specific rules or requirements to guide the physical development of a site or place. The aim of design coding is to provide clarity as to what constitutes acceptable design quality and thereby a level of certainty for developers and the local community alike that can help to facilitate the delivery of good quality new development.” (ODPM, 2005).

When looked at in this light it starts to become clearer how designing using design codes starts to resemble the process of parametric design and if we look into some of the specifics of a design code we can begin to see what types of variables could theoretically be used to drive a generative design. Using the design code for Upton in Northamptonshire as an example variables that are controlled or ‘guided’ by the codes include, urban hierarchy, character zones, boundary treatments, building use and density, building height, street width and materials. It could be argued that all of these things could be used as the basis of a parametric system that could generate a series of designs for this area. 

To take this a step further, it could be conceived that this process could be applied not just to one particular document but to a system of design values and principles. New Urbanism is one such system, as it is a movement that is governed by its own set of values, principles and ideals. New Urbanism is a particularly good example as the whole movement works towards a standardized set of goals that are laid out in its Charter of the New Urbanism. The 28 principles set out in the charter aim to ‘guide public-policy, development practice, urban planning, and design’ (Congress for the New Urbanism, 2001) and when looked at in the context of a parametric system it is easy to pick out over-riding principles (the system rules), design intentions (the variables or parameters within the system) and a third set of principles that aim to guide external policy.

(i) The system rules - these are the prevailing principles that define the entire system. In terms of the New Urbanism charter these could include principles 1, 4, 22, 25 and 27 (see Appendix A) that would define the system in terms of its geographical boundary, its strategy for development patterns (in this case brown in fill development will take priority over peripheral expansion), pedestrians take priority over vehicles, civic buildings occupy the prime locations within the development and conservation takes priority over new build. All of these rules could be programmed into a tool that could begin to generate a design. However, within this set of rules there are different variables that can be tweaked to produce the different scenarios or iterations. In terms of the more conventional example of a building design these system rules could include the structural stability of the building or its maximum height. 

(ii) The parameters – these do not define or govern the system and in that respect they are less important but the balance of these elements can be played with to produce different iterations or to refine a design. In the case of applying this to the New Urbanism ideal, these variables could control an individual buildings relationship to its neighbour, walking distances to public transport or local amenities, density of dwellings, mixture of uses or housing type, zoning, range and placement of green spaces and the relationship and networking of street and paths. Again, return to the conventional example these variables might include floor to ceiling height, level of glazing on the facade or the size and positioning of the core.

The third set of principles are not concerned with design and in this respect they cannot be included within the system as these principles seek to influence external policy. This could be influencing how funding is distributed at a municipality level, planning policy and land-use designation at a regional level or the selection of contractors and designers/architects. It is interesting to note this third set of principles as although they are not connected to design they seek to influence it indirectly as well as the politics of urbanisation. In fact one of the biggest criticisms of New Urbanism is that it produces architectural mediocrity and whilst Cliff Ellis tries to defend this by claiming “New Urbanism is not itself an architectural style, and can make use of virtually all extant architectural styles, depending on the context” (Ellis, 2002) it is quite clear from the Charter of the New Urbanism that the only style acceptable to it is neo-traditional or vernacular styles. In this respect New Urbanism has an aesthetic agenda much like Parametricism and this appears where the two movements differ the most. Parametricism rejects all forms of repetition where as New Urbanism supports it through design codes and principles. Patrick Schumacher has condemned the vernacularism of New Urbanism by claiming there can ‘no longer be national or regional architectures’ (Schumacher, 2009) instead the architecture should have the inherent capacity for local adaptation with the complexity this brings represented in the aesthetic of the design. The ‘soft’ forms and constant differentiation called for by parametricism are inherently modern in their design and the aesthetic is in complete contradiction to those prescribed by the New Urbanism charter. However, whilst the two movements differ aesthetically there are certain fundamental aspects where they are in agreement. Both movements believe that social order requires a spatial order and that currently the visual chaos of current urban spaces, with buildings that have no relationship to each other and urban quarters that lack consistency in style, are inhibitive to creating coherent and legible urban environments. At their very basic both movements attempt to do this and the creation of coherent community and city spaces could be seen as their ultimate aim.

At this point it’s beneficial to introduce some other parametric design tools, ones that are not necessarily associated with parametricism but which still have a relevance to this discussion. One such tool is agent-based modelling that aims to model and replicate the intelligence and emergent behaviour of people as they populate a city. This notion of emergence has become popular in recent architectural discourse and Neil Leach argues that in terms of design it can be seen most clearly in urban design. He argues that is viewed “most clearly in traditional urban formations” for “it is precisely the less self-conscious forms of urban aggregation that characterise the development of traditional settlements, from medieval villages to Chinese hutongs or Brazilian favelas.” (Leach, 2009). If New Urbanism is looking to the past for a technique for creating true communities they could possibly learn from these tools, after all, despite the abject poverty in some of these slums, the communities built are incredibly strong. It could be argued that the haphazard and random nature of these developments is impossible to replicate but it is exactly these patterns that multi-agent based modelling systems aim to find. As Leach describes it “these forms of urbanism constitute a relatively homogenous field of operations, where individual components do not stand out, but conform to the pervasive logic of their surrounding environment.” (Leach, 2009). In brief, cities can be seen as operating through a form of swarm or collective intelligence and if this intelligence can be built in to a multi-agent system where intelligent agents interact with each other and are capable of spatial mobility a design can be generated. A similar tool that has been around for longer and which is based on similar anthropological theory is Space Syntax which is a tool designed to test existing urban configurations for different performance criteria based on how people move and interact with space and their urban environment. 

Final Thoughts

Although the end result can be very different, parametricism, parametric design and New Urbanism can be seen to have similar goals and it is my conjecture that the principles and the different design techniques are not mutually exclusive. In this respect I believe that it is possible to build New Urbanist logic into a parametric system that could create an urban environment that performs well on several levels. However, where I see that there are irreconcilable differences are in their fundamental ideologies. In all cases the vision for urban design in terms of the final outcome can be read as similar (even if the aesthetics differ) but how each movement places society within a theoretical framework is entirely different. Parametricism aims to ‘organise and articulate the social complexity of a post-fordist network society’ (Schumacher, 2009) and in this respect New Urbanism hasn’t recognised or acknowledges that society has moved on; in fact it could be argued that with the political principles of the Charter of the New Urbanism, they are actively trying to retain or recover an ideology of a past time. Rather than trying to shape politics parametricism attempts to work within the system that it finds itself; Schumacher believes that the movement ‘cannot substitute itself for politics or the politics of urbanisation’ and for this reason I believe parametricism is more adaptable to changing social conditions. It just gets on with it if you will. Schumacher argues that parametric urbanisms recent success in large competitions being won by Zaha Hadid Architects is due to no other style or system being able to successfully integrate itself within an existing city fabric but at the same time offer the type of architecture he sees as being needed for his vision of modern society. Of course, we must recognise that Schumacher is trying to promote the parametricism movement with the agenda that it will bring more commissions into the office and also raise his profile and standing, and in this respect his intentions aren’t entirely altruistic, but I believe the tools and techniques being developed to support his vision have a place within urban design and can be beneficial no matter what the style or aesthetic is produced at the end. That is not to say that other parametric design tools cannot have a role also. It rather seems that Schumacher is unconvinced with agent-based modelling techniques despite working on this research with students at the Architectural Association Design Research Lab (AA DRL) for several years. I strongly suspect that whilst these techniques remain amorphous in the design in that their influence isn’t detected in the all important parametric aesthetic of the design then he won’t accept the techniques as part of the movement; “if you can’t see it, I’m not interested” if you will. Whilst I understand this position, I also think that these other techniques have something to offer that isn’t necessarily about aesthetics but about organisation. Urban design has to be much more systematic than other design disciplines due to the quantity and complexity of the information and issues that it deals with. Parametric design tools can help us deal with this complexity and for that reason they should have a place in contemporary practice.