ANT 417S

 Updated 20 January 2004
 Sources of Variation in the Built Environment  

Course Schedule Evaluation Readings FAQ ARH Prog Anthro Events


As noted in the first lecture, human beings managed for more than a million years without more than the flimsiest sorts of shelter. Even though architecture, like clothing, helped humans to colonize relatively inhospitable climates, it still did not need to be very substantial.

What is less obvious, perhaps, is that differences in climate only explain a small portion of the differences in architecture that occur among societies. Rapoport (1969), in fact, describes climate as only a modifying factor.

In this lecture I review some of the sources of variation in domestic architecture that Rapoport discusses and return to the issue that Flannery (1972) tends to emphasize (last week), namely the factors that may favour curvilinear or rectilinear architecture in different contexts.

  Houses as Containers

Hunter-Anderson (1977) makes an interesting functional analogy between houses as facilities, and containers as implements. Facilites, like containers, constrain the movement of material. The walls of houses protect their contents from the environment, while the walls of granaries or pots prevent or limit the dispersal of their contents into their environment.She notes that the properties of walls, roofs, floors and other structural elements depend on both the kinds of environmental interference their builders anticipate and the properties of the things to be contained or housed.

She goes on to make the analogy more specific by comparing round houses with bins (or pots) and rectangular houses with such things as tool boxes.

Pots and bins, most of which tend to be round, have as their primary function preventing the dispersal of their contents, sometimes during their transport, while still allowing access to those contents. The contents vary along a continuum from homogeneous to heterogeneous, but round containers tend to be used for relatively homogeneous contents, such as water or bulk grain. Such containers Hunter-Anderson terms "bins." She notes that all these contents have an equal "target value." For wheat, for example, one grain is as good as any other, so when somebody is removing wheat from the bin it is not important to distinguish particular grains. Consequently, there is no need for the bin to be structured into compartments that separate the grains from each other. Furthermore, once you remove material from a bin, you usually consume it, rather than putting it back. Hunter-Anderson argues that the ideal shape for bins is a sphere.

On the other hand, what she calls "wareboxes" tend to be rectangular. They are ideal for storing heterogeneous materials because you can organize them in compartments, or subdivisions, within the box. Furthermore, wareboxes tend to be used for items you would remove and then return to their places. Suitcases, tool boxes and tackle boxes, to cite three familiar examples, are used to organize clothes, tools and fishing tackle in ways that make it easy for us both to transport and to find particular items, and we can return items, after use, to their proper place. Their contents are not kept in bulk (as with bins), but different kinds of contents are kept separately, and have different target values. When you're looking for a screwdriver you don't want to retrieve a hammer. If there are internal partitions (including drawers in chests or desks), these prevent intermingling of different kinds of contents. In some cases, as with folded clothes, the shapes of the contained items facilitate organization within the warebox even without internal subdivisions. Much as Flannery (1972) argues that rectangular buildings facilitate agglomeration, Hunter-Anderson argues that rectangular shape is ideal for wareboxes, and allows efficient use of partitioned space.

She goes on to compare bins with "binhouses" or round/oval houses, and wareboxes with "warehouses" or rectilinear houses. She explores the possibility that the high frequency of rectangular storehouses is related to the ease with which "chunks" of material of standardized shape can be stacked.

When bulk materials, such as grain, are stored, the timing of introduction and retrieval of contents is generally not important.

Heterogeneous materials in warehouses are accessible (usually by some corridor) in a way that reflects the timing, or frequency, with which we would like to access them. They can be organized so that items needed more frequently or more urgently can be reached most quickly and easily, while other items accessed only occasionally can be less accessible (as at the far end of the corridor). Consequently, it is very common for storehouses to be long rectangles, with central corridor, and long, relatively narrow rooms opening off the corridor.

But the houses of greatest interest in this course are not storehouses for passive materials, but domestic housing for active people. Hunter-Anderson considers residential houses a category of "activity houses." Some of the contents of houses are active individuals who move within (and in and out of) the house and in relation to each other. Their activities have both "living aspects," concerning their biological needs, such as sleeping, and "role aspects," concerning individuals' social identities.

Where there are many living and role aspects, one solution to organizing them is to have separate sleeping houses, cook-houses, dining houses, work houses, and so on. This is essentially what Flannery (1972) was describing in the context of compounds. Another solution is to pack these into a single structure but use timing to prevent mixing of and interference between the different activities. In other words, the same structure can be used for cooking at one time, eating another, entertaining another, and sleeping yet another time. A third solution is to pack them into one structure but to partition the space, much as in tool boxes, to prevent the activities from interfering with one another. Where there are simultaneous activities, it is not possible to resolve interference by timing, and so it is necessary to use partitions of space.

She summarizes the factors affecting house form in three dimensions. One is the Heterogeneity and Simultaneity of Living and Role Aspects. Another is the Number of Living and Role Aspects, and the third is the Volume of Associated Materials and Facilities. Where values along all three dimenstions are low, we expect to find round or oval houses. Where they are all relatively high, we expect to find rectangular or rectilinear ones.


Design Theory Approach to House Form

McGuire and Schiffer (1983) attempt to explain variation in vernacular architecture (i.e., architecture built by its users, rather than by professional architects) as the product of design decisions that represent compromises between competing goals, including the goals of minimizing the costs of construction, use and maintenance.

However, the various goals conflict, and the design process represents an attempt to resolve these conflicts. For example, using materials that would keep maintenance costs as low as possible (for example, by making the house or its parts last longer), but use of these materials might significantly increase construction cost.

Goals of Use

Use functions can be utilitarian or symbolic. Architecture's utilitarian functions include mediating between people, their artifacts, and the environment. Such mediation includes insulating the house's contents from heat or cold and, as Hunter-Anderson (1977) also notes in some contexts, partitioning space to facilitate performance of different activities. Architecture's symbolic functions facilitate communication and maintenance of ideological and social concepts. The symbolic function is not necessarily costly or at odds with utilitarian function, but sometimes involves expensive materials or grand scale, both of which increase construction cost, as with cathedrals, palaces, and banks.

Goals of Production and Maintenance

A common goal of production (or construction) is to minimize its cost. A goal of maintenance is to minimize the cost of keeping the structure in usable condition. We can measure both these costs in terms of energy expended in their accomplishment, in materials, and in the expertise of the people who carry them out.

An important observation is that these two goals often are in direct competition. Cheap construction materials or hurried construction can lead to higher maintenance costs. Obtaining a very reliable building that does not need maintenance for a long time, or whose maintenance does not need to be frequent, often involves investing more expensive materials or more care and expertise in the initial construction.

McGuire and Schiffer argue that whether production or maintenance is given greater priority in builders' decisions is related to social inequality. In societies with large degrees of social inequality, poor people cannot afford the initial investments required to ensure lower maintenance costs. Wealthy builders can afford the initial investments that will lower maintenance costs, increase the use-life of structures, and control the symbolic content of the structures in ways that they view as positive (e.g., promoting prestige).

Other factors that affect the weighting of these goals is the degree of residential mobility or settlement longevity. How long the builders anticipate using a structure clearly affects their decisions about how much initial investment to make, and how important maintenance will be. If they anticipate abandonning the structure after less than a year, they are unlikely to invest too much in it unless it is critical to achieve important use goals. This recall's Whiting and Ayres's (1968) observation that round or oval structures are more common among more nomadic people, who anticipate occupying a particular site for only a short time.

McGuire and Schiffer note a number of utilitarian goals that domed structures accomplish very well where anticipated mobility is high. Domes require less material to enclose a given volume than do boxes, can be made fairly easily from flexible, light materials, such as sticks and hides, and are thermally efficient.

However, domes have disadvantages in other contexts. Although domes enclose more volume with less material, much of that volume is unusable, for lack of head-room at the periphery or taking up space in high ceilings. They are more difficult to partition efficiently (as Hunter-Anderson 1977 notes), and building them from materials, such as stone or brick, that last longer and require less maintenance also requires more expertise and higher construction cost. In sedentary settlements, it is more important to decrease maintenance costs and to partition activities and roles (use goals). Most importantly, longer use-lives often require occasional renovations, such as addition of rooms to accommodate growing households. This is the "domestic developmental cycle," which will be the focus of a later lecture. As Flannery (1972) already noted, adding rooms is easier with rectangular architecture. In McGuire and Schiffer's terms, this reduces maintenance costs because neighbouring rooms can share walls. Agglomerating circular or domed rooms requires building entirely new walls and wastes space at the interstices of circles.

 Readings   References

Amos Rapoport (1969), Alternative theories of house form, pp. 18-45, and Socio-cultural factors and house form, pp. 46-82 in House Form and Culture (Prentice-Hall).

Randall H. McGuire and Michael B. Schiffer (1983), A theory of architectural design, Journal of Anthropological Archaeology 2: 277-303.

Rosalind L. Hunter-Anderson (1977), A theoretical approach to the study of house form, pp. 287-315 in For Theory Building in Archaeology, edited by Lewis R. Binford. Academic Press.

Other references:

Michael C. Robbins (1966), House types and settlement patterns: An application of ethnology to archaeological interpretation, Minnesota Archaeologist 28: 3-26.

John Whiting and Barbara Ayres (1968), Inferences from the shape of dwellings, pp. 117-133 in Settlement Archaeology, edited by K.-C. Chang. National Press Books.

Last Update 27 December 2003

Contents and design copyright E. B. Banning 2001-2004