Michael Chen & Jason J. Lee

[Drawing by: Jose Blanco, Andres Correa, and Ivan Delgado]

According to most estimates, we have entered an era when for the first time, the majority of the world's population reside within urban areas. An explosion in urban growth is underway, most notably in the developing world where the urban population is expected to double by 2030.¹ The urban world produces its own forms of diversity, innovation, and intelligence, but the unprecedented expansion of cities worldwide is also not without challenges. The extraordinary scale and speed of urban growth already exceeds the capacity of the systems that would service, support, and manage it. Much of this growth will take place beyond the legal and administrative boundary of the city proper, uninhibited by centralized planning or management, and in absence of an adequate set of municipal infrastructures.

The decentralized growth of the city - that which takes place in patches, on the edges, self- regulated, and often remote from central administrative control - is the site of extraordinary volatility and undergoes a constant process of transformation. In many regions, this type of development, which often takes place on the margins of socioeconomic scales, is the most intense and fastest growing. The city is increasingly the site of extremes that challenge to a large degree, conventional notions of center and periphery. The urban footprint concept, for instance is another factor linking the city and its development to terrains and scales outside of and often remote from the city proper, linked via networks of consumption and capital, and paralleling the unbundling of preexisting orders of sovereignty, boundary, and hierarchy that is a feature of globalization.²

These are logics and relationships that resist conventional forms of representation and to operate within them requires the development of tools of analysis and design that account for the multiplicities of scale and complex logics that are found within the contemporary city. In an era when the land mass required to support a city like London exceeds the area of Spain³ for instance, it is inevitable that a certain ambiguity and perhaps a degree of confusion surrounds the issue of what is and what is not the City. New ways of engaging and mapping the city will be required.

It is worthwhile to refer to Fredric Jameson's analysis of Kevin Lynch's work The Image of the City (1960) and his suggestion that spatial and societal confusion might be somehow alleviated with the advent of more radical forms of mapping and representation that enable a heightened sense of individuals and their relation to both local and larger global systems. Following Jameson, the cognitive map might be understood to extend the instrumentality of maps from their use to account for the city's form and to construct itineraries through it to "enabl(ing) a situational representation on the part of an individual subject to that vaster and properly unrepresentable totality which is the ensemble of society's structures as a whole".⁴

This is an intriguing notion that might also be extended to forms of representation and analysis developed for the cartography of emergent phenomena, instances of collective intelligence, and other forms of dynamic organization that arise within the city in addition to accounting for scales of relation and organization that are found within the city but are operative within contexts that exceed the city's boundaries. It suggests direct and quantifiable linkages between the form of the city and the attributes of its architecture and infrastructure and a set of emergent social and economic phenomena that arise. Mapping may be employed not only to provide and promote an individual's understanding of his or her relationship to larger structures and institutions, but also to render visible and available many of the highly novel and innovative ways that people develop on their own to negotiate the city, and to speculate about the future growth and transformation of the city as it pertains to them. In other words, the cognitive map concept might be augmented for use at some level as a design tool, where representation and visualization are latent with potential for intervention and constructed to anticipate it. This would involve looking closer at the logics and protocols of the city and the strategies and emergent forms of intelligence that arise there, as well as projecting the growth and activity of the city against their near and long term consequences. The development of such tools for design and analysis offers new forms of agency and praxis and new opportunities for innovation.

Cognitive maps might be thought to privilege logics over form and many scales and intensities over single scales. They are representational tools that span directly between the physical, material, and organizational expressions of the city and its social structures, economies, and power dynamics, between scales that pertain to form and those that pertain to politics and policy. As such, they also suggest techniques and approaches for design that directly and simultaneously engage multiple agendas and where mechanisms of feedback may be constructed to ensure the continuous and reciprocal development of each. The cognitive map might be viewed as an intelligent tool for analysis and design that supports an understanding of a set of relationships that might otherwise remain invisible, and it might also be thought of as an independent form of intelligence, a sensory, or regulating, or otherwise cognitive entity that is installed within the city's physical fabric or its infrastructure.

The use of mapping as a design tool is not especially unique in the context of urban design and planning, nor is it unfamiliar within architecture. For instance, the ability to correlate statistical and other data against points in space is available in many robust tools such as GIS (Geographic Information Systems). These are systems that are used to great effect to identify patterns and logics that become visible once data is reconciled against geography. These techniques certainly speak to the production new forms of vision that afford understandings of one's relationship to larger systems that regulate and order societies. However powerful their potential to outline new contexts and relationships within the city though, their use as design tools for architecture is more limited. Robust systems for feedback between data and intervention, as well as a desire for contexts that support logics and forms of intelligence that are conducive to architectural design lead us to other digital platforms.

[Drawing by: Katie Adee, James Baldauf, and Natasha Harper]

Over the past two years we have been working with degree project students at Pratt Institute in an ongoing inquiry into contemporary global crises that suggest new demands and agendas for architecture, and the potential afforded by parametric and generative digital design tools to engage them. These are methodologies that are recognized to be particularly well suited to producing tremendous degrees of precision and complexity. Working through the authoring of scripts in Rhinoceros and Processing as well as through data set tools in Microsoft Excel, a broad range of analysis and design scenarios can be generated that manage complexity without reduction or generalization through the establishment and exploration of multiple simultaneous relational logics.

These are extremely productive tools that afford tremendous levels of precision with respect to the logics of input, processing, output, and feedback. Data sets are populated initially with information obtained through research and the data is cross-referenced with coordinate and other spatial data within a base map. Processing modules and algorithms are developed that reference data against other data for evaluation or to establish ratios and output parameters in the form of map notation: line work, coloration, area, etc. As the complexity and intelligence of the base algorithms is increased they repopulate the dataset, returning new relational data and establishing systems of recursive feedback. Recursion is a defining characteristic of generative design methodologies. It plays an important role in this context where analytical tools and design tools have the possibility of influencing each other and where the emergence of patterns and organization that pertain to behavior and those that pertain to formal and material attributes are both relevant to the other.

At present this is a prototype methodology that relies on some degrees of simulation to reproduce the characteristics of dynamic data sets. As more sophisticated dynamic data input and acquisition become available (GPS coordinates, SMS reporting, etc) the generative potential of iteration to account for and develop intrinsic logics of growth, differentiation, and transformation within design work will continue to be important. The direct association of generative design logics with dynamic data sets creates a scale of intelligent feedback that is useful within the design process, and simultaneously highly suggestive of conditions that might be installed as operational systems for more intelligent modes of infrastructure, monitoring and regulatory regimes, aid scenarios, and social movements. All are predicated on the creation of tools for identifying emergent patterns and behaviors arise from the iterative evaluation of a large quantity of data.

This follows a shift in contemporary methodology in the sciences and contemporary culture, and in particular in the computer sciences with respect to the generation and analysis of very large sets of data. Perhaps more than any other recent development, the advent of very large data sets has challenged the traditional research and knowledge production paradigm, moving towards a model where tools are designed to monitor and identify emerging patterns and behaviors that arise from the interrelation of more entities and bodies than can be controlled or observed individually. This is a principle behind the Google search algorithm as well as a research model pioneered by scientists like the geneticist Craig Venter who has identified several hundred new patterns in gene sequences that are sufficiently distinct as to constitute new species despite having never actually observed these species in person.

In both instances, developing applications that extract and track behavioral patterns from massive dynamic data sets produces new knowledge and agency. The aim of this work is to articulate a line of inquiry that is distinct from other related methodologies of computation and design including those that privilege the purely representational visualization of data and those that are concerned primarily with generative logics. We are interested in developing works of design that have a high potential for intelligence precisely because they develop in the context of data rich mapping environments where the underlying organizational and relational factors of the city are accounted for and where strategic opportunities that might arise in the intermingling of multiple logics and organizations can be explored. Because these mappings are parametric in nature, they can continually be updated and their resolution increased, and they provide a rich environment for feedback between the logics of a system and the effects of any intervention, thus enabling the incremental engagement between the proposal and its context. The cognitive map is as much a milieu itself as it is a tool for analysis, and maintains direct correlation between the development of design scenarios and their attendant effects with respect to other systems and logics tabulated in the map.

This is an approach that addresses two important concerns. The first is simply the desire to find opportunities for design innovation where architecture engages the most urgent crises of the day in their full extent and complexity. This is not an optimization or representation-based strategy, but one that seeks out strategic and selective methods of direct engagement. Second, these maps are able to articulate contexts and address scales of effect that exceed those of architecture. These are the scales of urban practices, spatial tactics, navigation, and collective intelligence. These are understood to be present within complex systems but vehicles for architecture to negotiate these scales may not always be apparent. This is a move to decentralized and distributed models for performance, and the organization of multiple sites within larger networks of affiliation that support certain practices across them. These are territories of extreme complexity and also great opportunity, and they also speak to a moment of agency for design where one was not previously thought to exist and architectures that anticipate new protocols and practices of rapidly urbanizing world.

continue