Designers of cities and systems must deal with complexity. Divide-and-conquer, abstraction and modelling, negotiation are all workable tactics. The way a designer explicitly or tacitly chooses to deal with complexity says a lot about what they believe the problem to be. The time-honoured approach to complexity, espoused by ‘design science’, is to decompose on the basis of function. In civic planning, this leads to zones; in software engineering, functional decomposition; in sociology, category dilemmas. In the spirit of ‘Form follows Function’, functionalist designers argue that classification and decomposition are effective design strategies that result in highly efficient designs. For some kinds of systems this is undoubtedly true. But behaviourist designers counter with the claim that functionalism works only for design problems of low complexity and in well-understood, stable, closed systems. Cities, like societies, organisations and cultures, need other kinds of thinking about design.
♦ ♦ ♦
Design theorist Christopher Alexander was the first to contrast functional (structural) versus emergent design in ‘A City is not a Tree’ (1965). Alexander’s paper was selected as one of the winners of the 1965 Kaufmann International Design Awards, in the esteemed company of Ada Louise Huxtable, Lewis Mumford, and prior winners Charles and Ray Eames and Walter Gropius. ‘A City is not a Tree’ opens with the clarification that ‘the tree of my title is not green with leaves… it is an abstract structure’. Alexander’s tree is not the one that underlies hierarchical plans, or any manifestation of functional decomposition. It is a metaphor for the structuring and bureaucratic control of the commercial, social and domestic functions of a city. ‘A City is not a Tree’ proposes an alternative, emergent behavioural paradigm for planning and design that attempts to replace established processes of decomposition (the kinds of process that produce a tree) with those of devolved evolution (alternate processes that produce a network, lattice or organic form). Alexander’s approach to complex, multi-dimensional planning problems commences with a cameo of habitation:
In Berkeley, at the corner of Hearst and Euclid, there is a drugstore, and outside the drugstore a traffic light. In the entrance to the drugstore there is a newsrack where the day’s papers are displayed. When the light is red, people who are waiting to cross the street stand idly by the light; and since they have nothing to do, they look at the papers displayed on the newsrack which they can see from where they stand. Some of them just read the headlines, others actually buy a paper while they wait. This effect makes the newsrack and the traffic light interactive; the newsrack, the newspapers on it, the money going from people’s pockets to the dime slot, the people who stop at the light and read papers, the traffic light, the electric impulses which make the lights change, and the sidewalk which the people stand on form a system — they all work together. (Alexander 1988, p. 68)
This vignette reveals static and dynamic elements. The static part is the physical arrangement of newsrack, lights and sidewalk—a structural pattern within the system which Alexander calls a ‘unit’ of a city. This particular unit is inhabited by a dynamic flow of people and objects (newspapers, coins, ‘electrical impulses’). A unit is bounded by the dynamic system it supports, and contains only those physical objects that are required to make the dynamic system work. Units can exist on a much larger scale than the one involving a street corner, such as the arrangements of facilities that support dynamic systems for shopping, eating, socialising, working and entertaining.
In highly evolved cities (Alexander calls these ‘natural’ cities), units and the dynamic systems they support overlap, Alexander claimed. The news rack may be a part of an adjacent unit that supports the system of people waiting at a nearby bus stop. Alexander argued that this kind of overlap is emergent, rich, and complex—often too complex to be designed a priori, and certainly too difficult for conventional town planning methods to anticipate. When units overlap, they form what Alexander calls a sub-lattice structure. When units are non-overlapping (disjoint), they form a tree. Alexander’s network (sub-lattice) model of emergent design is behavioural rather than structural. Similar emphases (that account for the emergence of system structure based upon dynamic behaviours or scenarios) have found their way into software, user experience, and open systems design methods. To avoid a tree of units in which the interactions between places, objects and people are artificially simplified or inappropriately constrained, network behaviours must be recognised and harnessed.
♦ ♦ ♦
Alexander encourages his readers to think of a city’s complex social dimension when structuring the physical space. Rather than viewing a city planning problem as one of simply structuring space to create pieces that can individually be administered, built, purchased or outsourced, Alexander advocates viewing the system as a sub-lattice of information and communication flows between people. Some of the most effective design in an online networked society is effective because it recognises established flows and supplements or augments them, rather than forcing an ill-fitting structure that disrupts established flows.
Divide-and-conquer might work for closed systems but it fails miserably in city planning and in a networked society. Alexander spent much of his academic career pondering the question of why his experience of inhabiting the artificial [master-planned] city differed from his experience of the natural [evolved] city. His answer—‘that the natural city has the organisation of a semi-lattice but when we organise a city artificially, we organise it as a tree’ applies equally to other open, sociological systems.