Pattern language context theory

Christopher Alexander's pattern language and landscape architecture 

The pattern language Alexander's pattern language provides a way of creating functional relationships between development projects and their contexts. The preceding theories of context have dealt with the ecological and aesthetic aspects of the relationship between projects and contexts. The pattern language, as proposed by Christopher Alexander and his colleagues, provides an approach to the social dimension: to contextual relationships between people and places. It can assist in planning projects to have a favourable impact on the social environment (see page.. x-ref to preceding chapter). Alexander, having studied math ema tics before architecture, was initially attracted by the scientific-deductive approach to design and planning. This method was pushed to its limit, in Notes on the synthesis of form, at which point it failed. Alexander then turned back and published a famous essay with the title 'A city is not a tree'. He argued that it is a fundamental error to think of cities as hierarchical tree-structures, as he himself had done. Human settlements, Alexander proclaimed, are semi-lattices rather than trees. Lattice theory is an abstruse branch of math ema tics which deals with the relations between different parts of the same whole. This led to the development of what became known as The Pattern Language (Alexander et al 1977). The full pattern language comprises 253 patterns, with an open invitation to readers to contribute additions. The patterns range from large to small. Pattern 1, deals with the independence of city regions. Pattern 253 deals with the display of personal bric-a-brac in your house. Each is an archetype. Some, which need not concern us here, are for the interior planning and construction of buildings. Others are for the planning and design of good outdoor space, urban and rural. When embarking on a design or planning project, users are invited to choose from the list of 253 patterns that one which most clearly describes the project. Alexander gives the example of a project to make a front porch. Pattern 140 is chosen as the closest. It is for a PRIVATE TERRACE ON THE STREET. The pattern is related to its context by using patterns from higher on the list, such as Pattern 120, PATHS AND GOALS. The porch also creates a context for patterns lower on the list, such as Pattern 167 SIX-FOOT BALCONY and Pattern 242, FRONT DOOR BENCH. When completed, 'the character of the porch is given by the ten patterns in this short language' (Alexander et al 1977: xxxvii). Instead of the porch being a pre-fabricated object, slotted into position like a shower-cubicle, it becomes a unique structure enjoying an intimate contextual relationship with its surroundings. Alexander speaks of a pattern language to remind us that the patterns fit together like words in a sentence. The social environment is seen to have a structure which stands comparison with a written language, or with a machine. If a car is disassembled and all the parts laid out on the ground, it no longer has a structure. If the parts are bolted together in the wrong combinations, it may have a structure but it will not be a good structure. It will resemble a badly planned city. To make a good car, the parts must be grouped in the correct clusters, to make a body, an engine, a transmission and a passenger compartment. There is a hierarchy of assemblies and the assembly rules can be described as a language, with paragraphs, sentences and words. 'Language' is, again, a useful analogy for the environment because there are so many different things to be said and so many types of place to be made. The strengths and weaknesses of the pattern language approach may be summarised as follows: Strengths: 1) The language is incrementalist. It is 'piecemeal planning', not 'master planning' or 'blueprint planning'. 2) The language makes use of archetypes. 3) The language offers a way of inter-relating development projects, by responding to what has gone before and giving thought to what may come after. Weaknesses: 1) It is claimed, unnecessarily, that each of the patterns is an objective truth. This gives a totalitarian flavour to the language. 2) Some of the patterns are eccentric and at the far left of the political spectrum. In view of the preceding point, this gives the language a menacing character to those of different political persuasions. 3) It is claimed, wrongly in my opinion, that the patterns are independent of culture and climate. 4) The language appears to use a hierarchical structure, of the type which Alexander criticised. 5) The language takes no account of natural or aesthetic patterns and does not provide a basis for commenting upon the Australian problem, as discussed above. The weaknesses of the theory are important if the language is to be used in guiding contextual design. From the viewpoint of context theory, the most significant weakness of the language is the way in which the man-made environment is privileged over the natural environment. Some of the patterns do refer to natural site characteristics (eg Pattern 161 SUNNY PLACE). But surely natural characteristics are as important, or more important than, man-made characteristics? McHarg's overlay approach showed how natural factors could and should be incorporated into the planning process. A pattern-based planning model has the potential to inter-relate the patterns of the existing landscape with Alexander's patterns of human use and aspiration. When preparing a design for a particular location, designers should know what Alexander patterns are present on the site and what provision could be made for the establishment of other patterns.