Mercado Municipal de São Paulo by FlaviaC (http://creativecommons.org/licenses/by-sa/3.0) via Wikimedia Commons
It’s time for another installment from the late T. William (Bill) Kleinsasser’s self-published textbook Synthesis. I’m always happy to feature excerpts from Synthesis here on my blog. Bill’s legacy is too important for me to allow it to become forgotten.
As I’ve mentioned previously, Bill would never see Synthesis widely published. I attribute this largely to Bill’s penchant for incessantly overhauling the manuscript. He never appeared happy with the book, believing it could always be improved. Ironically, I find its earlier incarnations far richer than the later editions, albeit less cohesive and ordered.
A recurring theme in Bill’s writings is that buildings should enjoy long and varied lives—that we should design them to be adaptable and resilient over time. For example, imprintability and changeability—the notion that the best places gracefully accept constant refinement and reshaping by their occupants. Or precise ambiguity—the power of spaces to evoke, but not dictate; to help, but not limit; or to be particular, but not closed. He also challenged the common practice of designing precisely to suit immediate needs without regard for how those needs might and are likely to change in the future. In the following passage from my earliest edition of Synthesis, Bill speaks more pointedly to the need for us to approach the design of buildings from this mindset.
Activity Families and Families of Use
- Because people are unpredictable, a space intended for a particular use will sooner or later be used in ways not anticipated.
- Because people seek perfection, a space intended for general use can be successful only if the uses to which it is put are sufficiently related in regard to need for equipment, physical characteristics, experiential impact, and connections to other places and things.
- While human activities are hard to predict, they can be expected to form patterns or come in sets; and while accommodation of particular activities tends to invite environmental obsolescence, it is possible to organize space in anticipation of changes forecast by any one activity with the situational family of activities.
The third statement above implies that the unplanned uses will probably all be related, or come in a family or set. For example, the bathroom described above probably never will be used for food preparation or sleeping. Spaces for particular uses tend to be used for more, but within limits, because each original uses establishes a family of uses.
The second statement above implies that a space for unspecified uses must to some extent be changeable; and the several uses to which it is put to use must to some extent be adaptable. If complete changeability of space were possible, any use could be perfectly accommodated. On the other hand, if the character of any use could be changed to fit the limitations of a space changeability would not be necessary. Since it is unlikely that either could ever happen, the success of a space for unspecified use depends upon the developed interface between the changeableness of the space and the adaptableness of its use.
For example, a space with a sloping floor is not as changeable as one with a flat floor. Similarly, a single space cannot successfully accommodate both a basketball game and a string quartet. The fixed-or-flat floor example demonstrates relative changeableness. The single-space example demonstrates the incapability of specialized or spatially demanding activities to adapt to spaces which were not made particularly for them.
It would seem then that a space can be successful only for determined families of uses. Since some of those families might be very small and exclusive, some spaces must be highly particular and designated. Conversely, those spaces whose families of use are broad can be loosely fitting and relatively undesignated.
To repeat, if we are concerned with the quality of human accommodations, particular-use spaces tend to want to accommodate more, and unspecified-use spaces tend to want to accommodate less.
On the other hand, if we do not consider the quality of human accommodation, particular-use spaces can become particular to the point of being machines and unspecified-use spaces can become general to the point of being inexact, improperly equipped, or chaotic.
And since spaces for particular uses and spaces for unspecified uses both accommodate facilities of related uses, they are alike.
In the design of spaces for people it would seem important to always consider the nature of the activity—families to be accommodated rather than the single use or several uses which may be first and most easily identified. This simple expansion of responsiveness would cause the making of new kinds of space which are not either-or accommodations but instead balanced combinations of support for families of use, some known, some anticipated. We could then have spaces that be closer to what we need.