ORIGINAL ARTICLE

Theory, Culture and Housing

Amos Rapoport
The School of Architecture and Urban Planning, University of Wisconsin Milwaukee, Milwaukee, WI, USA

Housing, Theory and Society 2001; 17: 145165. In this article the study of housing (and other built environments) is located within the context of an explanatory theory of environment-behavior relations. The need for operational definitions and dismantling as general approaches is emphasized. After defining housing, it is suggested that it is impossible to relate culture (or society) to housing (or any built environment), because these concepts are too broad, general and abstract. More specific and concrete variables derived by dismantling culture are then shown to be related easily to housing, housing choices and preferences, group differences, etc. The value of using popular and mass media, advertisements and the like in this process is emphasized. Key words: theory, environment-behavior relations, built environments, housing, systems of settings, culture, operational definitions, dismantling, expressions of culture. Amos Rapoport, The School of Architecture and Urban Planning, University of Wisconsin Milwaukee, PO Box 413, Milwaukee, WI 53201, USA. Tel: 414 229 4014. Fax: 414 229 6976.

INTRODUCTION This article takes a particular conceptual and methodological approach to the three terms of the title of this journal, by considering them hierarchically, starting with a general, explanatory theory of environmentbehavior relations (EBR), through a framework based on culture, to a series of highly speci c cultural and social variables, useful both for the analysis and design. The article builds on two recent publications (Rapoport, 1998, 2000). The rst concerns explanatory theory in EBR. The second addresses the use of culture in housing. First each is summarized. The rst article (Rapoport, 2000) proposes a scienti c approach to EBR, and provides a metatheoretical sketch of an explanatory theory of EBR within which environment-behavior studies (EBS) needs to be seen. Housing is a particularly striking example of the need for theory. There is too much information, numerous disconnected pieces of empirical research, which, in effect, become counterproductive. An important role of theory is to subsume much data in easily remembered and used formats (known as compressibility). Even a conceptual framework can help by organizing material, although not as much as theory. The second (Rapoport, 1998) suggests a general process dismantling . Since I believe that holism (in the non-scienti c sense) is impossible,
1

Note that society, like culture (and many other concepts) needs to be dismantled in order to make it usable. As it stands, it seems just as unusable as culture, and for similar reasons discussed later in this article.
2001 Taylor & Francis. ISSN 14036096

dismantling is a constant , a standard technique or approach and, for scienti c holism, re-synthesis at higher levels of abstraction is necessary. Applied to culture, it makes it possible to relate culture (and hence society 1) to housing. A series of examples is used to show how easily components and expressions of culture, unlike culture can be related to housing, although that also needs to be de ned explicitly. Although this will be discussed later in somewhat more detail, I brie y discuss the nature of housing and of one of its attributes environmental quality. All de nitions and dismantlings of environment clearly also apply to housing. The most useful here, because it is the most concrete, considers housing as a system of settings within which a certain system of activities takes place (Rapoport, 1990a, 1998). The environmental quality of housing is described as a set of attributes, obtained by dismantling and can be represented as a pro le (Rapoport, 1995d (1990); Khattab, 1993). Housing choice is of a particular system of settings and its associated environmental quality (Rapoport, 1985, 1995c, (1985), 1995a, (1990a)). Environmental quality is evaluated through several of the components of culture to be discussed later (speci cally ideals, images, schemata and meanings, and norms, standards, expectations and rules part of a general model of preference involved in choice and design). Speci c lters and evaluative criteria de ne variable environmental quality pro les, helping to answer the question of what is a good or better environment, better for whom, why and how one knows it is better (Fig. 1).
Housing, Theory and Society

146

A. Rapoport

Fig. 1. Model of evaluative process (which, with minor modi cations, can also be applied to other subdomains).

EXPLANATORY THEORY AS A GENERAL CONTEXT I approach housing in the context of EBS, understood as the science of EBR a context even broader than explanatory theory, although the latter is possibly the most important attribute of science. One begins by de ning the domain of EBS, through both subject matter and important questions about it (Shapere, 1977; Rapoport, 1990b, Ch. 1). The metatheoretical sketch of an explanatory theory of EBR is derived directly from that de nition with just one epistemological assumption: that science provides the only reliable way of acquiring knowledge. Explanatory theory helps us to understand the world, and can be contrasted with so-called architectural theory, which is a (usually unfounded) normative position about how buildings should be. Explanatory theory is not normative, but tries to understand (and thus explain) the patterns, linkages and forms of interaction between environment and behavior, including their mechanisms. These are investigated empirically since empirical testing is central to explanatory theory. Although such theory must be general , able to explain general phenomena, it should be applicable to speci c cases by introducing speci cs, as the examples given later will show. However, nding patterns is a crucial step in the process of theory development. The search for patterns requires the largest and most diverse body of evidence which, in turn, requires expansion of the body of evidence in four ways: to include all environments, all cultures, the whole environment and the full span of
2

history (Rapoport, 1990b:1119). This helps with generalization and provides model systems (corresponding to those used in, for example, the biomedical sciences) (c.f. Rapoport, 2000). The starting point is the three basic questions of EBR which offer the simplest, briefest, most fundamental, and thus most useful, de nition of the domain.
*

What biosocial, psychological and cultural characteristics of human beings (as individuals, member of a species and of various groupings) influence (and, in design, should influence) which characteristics of the built environment? What effects do which aspects of which environments have on which groups of people, under what circumstances and why? Given these two-way interactions between people and environments, there must be mechanisms linking them; what are these mechanisms?2

These questions describe the subject matter of EBS, subsuming much speci c material, and can be dismantled to any degree of speci city and precision. As questions they are too general as posed, but are capable of being dismantled. This formulation is also openended and new developments in EBS or other disciplines can be incorporated. One clearly needs an operational de nition of environment. Since holism in its non-scienti c sense is not an option, this de nition involves explicit dismantling; thus, this aspect of ontology emerges naturally. I have found four complementary conceptualizations of environment to be most useful, (and their utility varies with the topic or question being studied):
*

The possible mechanisms seem few in number; a rst list includes eight plus an as yet undetermined number of the components of culture to be discussed later (Rapoport 2000).

the organization of space, time, meaning and communication;

Housing, Theory and Society

Theory, culture and housing

147

Fig. 3. Relationships between housing, de ned as a system of settings, and the larger systems of neighborhood and settlement. Fig. 2. The two buildings above cannot be compared as dwellings. The two systems of ten settings each de ne the dwelling and form the units suitable for cross-cultural comparison. (Based on Rapoport 1990c, Fig. 2.5, p. 16; 1994a, Fig. 2, p. 464, 1998, Fig. 2.)
*

* *

a system of settings within which systems of activities (including their latent aspects) take place; the cultural landscape; composed of fixed, semi-fixed and non-fixed elements.

The rst is the most abstract, followed by systems of settings and cultural landscapes; xed, semi- xed and non- xed elements are the most concrete. This also ranks them from the most complex to the simplest, because the rst needs much further dismantling (for example, many varieties of space can be listed easily (Rapoport, 1977:1213)). The second was used to de ne housing so as to enable cross-cultural studies helping in theory development (Fig. 2). This system, in turn, is embedded, in different ways, into larger systems of settings (e.g. blocks, compounds, neighborhoods, settlements and sometimes even regions). Housing must then be considered in relation to streets, open spaces, other settings and neighborhoods. It becomes important to discover the extent of the relevant systems rather than assume them (Rapoport, 1977, 1990a; c.f. a useful general approach is progressive contextualization (Vayda, 1983)). The number of specialized settings grows as societies become more complex (Rapoport, 1990a; Kent, 1991; Yellen, 1985; Sancar and Koop, 1995). Also involved are lifestyles and rules about what behavior is appropriate in which setting (Baumgartner, 1988), which in uences acceptability and de nitions of crowding,

privacy, etc. and thus environmental quality (Fig. 3). This is important, because what happens, or does not happen, in some settings in uences what happens, or does not happen, in others (e.g. Rapoport, 1986, 1990a), explaining some of the differences in housing. Such differences seem to be mainly in the transitions between the dwelling and related out of dwelling settings, including stoops, stairs, workplaces, shops, trees, streets and so on (e.g. Rapoport, 1977, 1983; Rybczynski and Bhatt, 1986; Lawrence, 1986). It also helps to explain the role and nature of neighborhoods (Rapoport, 1986, 1997b), settlement patterns and transportation routes. Via the other conceptualizations of environment it relates housing to time and communication, ambience and its components, meaning (and, hence, latent functions, status, identity, etc.) and the role of xed, semi- xed and non- xed elements in communicating such meanings. One nds a major and natural uni cation, and increased understanding so that, for example, since meanings such as identity, will vary with culture they will be communicated, through different xed, semi- xed and non- xed cues; if only the latter are used buildings or furnishings play no role (Rapoport, 1981, 1990d). CULTURE AS A SPECIFIC CONTEXT Culture plays a role in all three questions. Regarding housing an important question is why, when considered historically and cross-culturally there are so many different forms of housing in such different settlement types (e.g. Oliver, 1987, 1997), although activities in dwellings are much less variable and fewer in number. The general answer has been cultural differences, particularly since housing as the primary setting par
Housing, Theory and Society

148

A. Rapoport advances were impossible, because culture is not a thing but a concept (or de nition). One cannot deal with culture-environment relations at that level of generality and abstractness. I have, therefore, gradually developed ways of dismantling this concept, the latest version of which is Figure 4. The rst step is to try to de ne culture. There are three useful, complementary classes of de nitions that ask, what is culture and three that ask, what does culture do (Rapoport, 1995a (1986)). How important ` -vis other human characteristics, and its culture is vis-a variability versus possible pan-human constancies or constraints (such as human nature or epigenetic rules) are empirical questions. These are in a state of ux due to ongoing research in evolutionary science, behavioral genetics, molecular biology, evolutionary psychology, sociobiology and the like; the answer is not self-evident (Rapoport, 1997a, 1998, 2000). One can discuss in detail, and give reasons for, each step of the dismantling process. This cannot be repeated here (see Rapoport, 1998); also, some components of culture, such as lifestyle and activity systems need to be dismantled in turn. The variables in the model enable us to begin to analyze and synthesize the many existing housing studies, much information in the popular press and media, and also to relate material from many places, groups and periods (Rapoport, 1997a) as well as changes. One can begin to develop and clarify these variables and establish patterns and numerous lateral linkages on the way toward theory development. Although the nature of groups is important (and the usual starting point) it might prove more useful to begin with the variables identi ed. This I do later, using examples, mainly from newspapers, but also some from advertisements and novels 3 (Rapoport, 1990c, d, 1995c (1985)). Before discussing speci c variables, however, some more general remarks about applying the model are in order. APPLYING THE MODEL In this section I address two topics: the nature of groups involved in culture-speci c housing, and the notion of environmental quality, to suggest how the variables identi ed in Figure 4 and used in the examples can begin to be related. Human groups Among de nitions of culture is one that considers culture as that which de nes and distinguishes among groups (which have been called pseudo-species). It is their existence that leads to culture housing relationships. Moreover, such groups can be (and often tend to be) rather small, (although countries and cultures are

excellence , has traditionally been the most typical product of vernacular design and, therefore, most directly related to culture. Vernacular environments, and spontaneous settlements in developing countries, i.e. most housing, is the result of selectionism, an evolutionary process whereby environments gradually become congruent with activity systems, lifestyles, meanings, etc. (Rapoport, 1995a (1986) by applying rules which are often unwritten, as in most cultural landscapes (and housing may comprise large portions of some such landscapes). The recognizable nature of cultural landscapes and style in buildings both result from the systematic and consistent application of systems of rules. Vernacular design was believed to use unwritten rules, but it appears that in some contexts rules may be written, or formalized in other ways, through Feng Shui masters and manuals in China and Korea (Nemeth, 1987) or the legal rules (and local traditions) in Moslem cities (Hakim, 1986, 1994). High-style and, above all, contemporary environments typically involve codi ed, legalistic, formal, written rules and design is instructionist (Rapoport, 1995a (1986)). This is due partly to the greater involvement of professional designers, partly the greater heterogeneity of human groups, the existence of many more specialized settings and also the greater mobility of populations, rapid culture-change, etc. (Rapoport, 1983, 1985). Since culture-change continues, and there is also ecological succession in housing, changing age structures of populations, etc., open-ended design is needed, as it is for other reasons, some discussed later (Rapoport, 1995b (1990)). The replacement of selectionism by instructionism is a major reason for the need for research-based knowledge that EBS and EBR theory can provide, especially in the case of housing. The close relation of housing and culture also implies that housing often, although not always, communicates identity (e.g. Duncan, 1981; Rapoport, 1981). It follows that latent functions (meanings) and images are more important than instrumental functions (Rapoport, 1994) which, however, set limits and constraints. The importance of latent functions, especially of meaning (which, itself needs to be dismantled (Rapoport, 1988, 1990c)) helps to explain the variability of housing. Some of the variability may be due to constraints , the effect of which are sometimes dif cult to distinguish from the effects of wants. Also, culture seems to be more important in wants than in needs, and also in how both are satis ed. The need for culture-speci c housing (or design) has increasingly been recognized. However, further
3

I could also have used lm, TV, etc.

Housing, Theory and Society

Theory, culture and housing

149

Fig. 4. Dismantling of culture and relating its expressions to the built environment (the width of arrows corresponds approximately to the hypothetical feasibility and ease of relating the various elements). (Based on Rapoport, 1977, Fig. 1.9, p. 20; 1993a, Fig. 1, p. 16; 1994a, Fig. 3, p.476, 1995e; Fig. 2, p.16, 1998, Fig. 6; 2000, Fig. 2, p. 127.)

still confounded (e.g. Newell, 1997)); it is also an under-researched topic in EBS. The relatively small size of groups is not suf ciently emphasized in discussions of culture or society. For example, in India, 4635 distinct human communities, such as castes, tribes and the like, have been identi ed, including 75 endangered tribal groups, and 324 functioning languages, using 25 different scripts (Bagla, 1999). In connection with housing, these can probably be subdivided further by degree of modernization, education, occupation, place of residence (e.g. size, type and location of settlement), sex, age, etc. Thus Devlin (1994) found no socioeconomic or regional differences in housing preferences in the USA, but there were sex and age differences. Such further subdivision might well result in many more lifestyle groups, as discussed later (c.f. Rapoport, 1985, 1998). In Guatemala 60% of the population of 11,000,000, identify themselves as Maya, but speak 22 different languages (El Futuro Maya 1999). The average size of one of these groups is, therefore, 300,000. Variables like those suggested for India would lead to smaller lifestyle groups.

One result of the natural emergence of culture from the three basic questions obviates the need to introduce special user groups. One just needs to refer to groups (part of question one) which have sets of attributes. The typical categories used, such as children, youth, elderly, women, disabled, homeless, cognitively impaired, urban poor of the developing countries, and the like, are not needed, except possibly in applied research and research applications . They may also be useful as model systems, corresponding to those used in biomedical research, in which EBR are seen more clearly, being more extreme and at higher criticality. Also, these aggregates and/or groups are then de ned not by age, or economic status or other imposed etic categories, but through lists of emic or derived etic attributes that follow from the three basic questions and the dismantling of culture shown in Figure 4. As the small size of groups has become clearer, the number of lifestyle groups used in marketing and advertising generally has gone up. In (Rapoport, 1977) 4 were used, but now more than 50 lifestyle groups relevant to housing are used in the USA (Rapoport, 1985). These only include potential buyers of new
Housing, Theory and Society

150

A. Rapoport Another component of environmental quality is the preference not only for suburbs (over cities) but, increasingly, rural areas. This change is based on quality of life concerns, and is made possible by new technology, among other things. The cultural landscape of rambling porches, picket fences, Victorian-style houses, a town square shaded by towering trees and safety from crime attracts people with children (Brown, 1996a; Johnson, 1996). Such attributes re ect ideal images (c.f. Hummon, 1990) and, except for house and lot size (i.e. density), are almost identical to those used in the New Urbanism which, in addition appeals to speci c lifestyle groups and is also promoted in terms of environmental quality (as in Celebration, Florida). Such areas also have characteristics that some people might regard as negative, for example, little privacy and a lack of anonymity (e.g. Baumgartner, 1988). However, the search for community has become an aspect of environmental quality for certain people, and such characteristics also lead to safety. In addition to larger houses on larger lots other components of the good quality of life include lower taxes, good schools and greenery and quiet and peace away from the noise of the city4 (Holly, 1997). Since, as already pointed out, housing is more than the dwelling, the neighborhood and its environmental quality pro le become important. In a 1995 survey of 5000 prospective house buyers in several US states (Urban Land Magazine cited in Wiechman, 1997) the amenities desired could clearly be expressed as a pro le.5 Of respondents, 77% wanted natural open spaces, 74% walking and biking paths, 56% gardens; 18 other variables ranged from 55% to 18%. Similarly, in a special supplement for rst-time house buyers in Winnipeg, Canada (Winnipeg Free Press, 1992) evaluation moves from the neighborhood to the yard, the outside of the house and, nally, to the house interior, with desirable attributes listed for each scale, resulting in what I call an environmental quality pro le. An English novel describes a desirable dwelling environment as a suburb that should look like a stretch of woodland with no houses visible, no gates, and the only sign being mailboxes placed discreetly in gaps of hedges. Green lawns terminate in these hedges or low walls at the sidewalk, and because you knew it must be there you caught the faintest glimpse of mellow brick [see later materials ] between the great gray beech trees, the delicate silver birches and the branches of a majestic cedar (Rendell, 1995:116). Changes such as modernization of dwellings may also be evaluated differently depending on the variables shown in Figure 4, and such changes have been evaluated positively in Japan and negatively in India (e.g. Grenell, 1972; Grimaud, 1986; Bauhain, 1986);

housing. Considering buyers of existing houses, renters, residents of mobile homes, the homeless, etc., more such groups are likely. Also, in the USA multiple and overlapping group memberships are common (e.g. Gordon, 1978), further increasing the potential number of groups (and hence the need for open-ended design). The small size of groups is also shown by the presence of several in rather small urban areas (e.g. Suttles, 1968; Suchar and Rotenberg, 1994). Such groups are de ned by geographics, demographics and psychographics and are used to segment markets (e.g. Barmash, 1988; on housing see Sternlieb and Hughes, 1986:16ff; Gilbert, 1989). Not only is different housing often required for different groups, but so are different sales techniques (Anders, 1991). Differences among groups also in uence the nature and design (and location) of shops, including supermarkets, in the USA (Stevenson, 1992), the goods carried, displays, etc. In US politics, one consultant divides voters into 62 lifestyle clusters, whereas another uses 120 ethnic categories (Milbank, 1999:26). Once again, these could be subdivided further, or differently, for purposes of housing and EBR. These are empirical questions on which research is urgently needed. Environmental quality As already mentioned, group differences in the evaluation of environmental quality are the result of values, ideals, images, etc., and norms, expectations, etc. (often related to status). There are two major meanings of environmental quality; we are concerned with that emphasizing social and psychological aspects. It is made up of many components, which require dismantling and can be expressed as a pro le (Rapoport, 1977, 1985, 1995d (1990); Khattab, 1993). One such component is space, so that while household size in the USA has decreased, dwelling size has been going up (e.g. Brown, 1996b) as a result of af uence, i.e. a reduction of constraints and related to status and changes in lifestyle. Some houses for families with three children are up to 7000 sq. ft in area on 1.7 acres of land, with 6 bedrooms and 7 bathrooms. In 1996, the latest year for which gures are available, 14% of new houses in the USA had an area of 3000 sq. ft or more (in 1980 2400 sq. ft). Built mainly in the suburbs of large cities, such houses comprise 18% of new houses in the northeast, 15% in Texas and Florida and 12% in the west; all have features once found only in true estates (West, 1998).
4 5

This is in Milwaukee, USA, an unusually quiet city, implying an ideal image rather than reality. It will be seen later that similar amenities are provided in a new housing development in Yogyakarta, Indonesia (Merapi View).

Housing, Theory and Society

Theory, culture and housing they have worked in the former but not the latter. More generally, the same environments are evaluated very differently by different groups, including designers and users as a whole, and insiders versus outsiders (Rapoport, 1977, 1990c; Brower, 1988, 1989). Environmental quality pro les not only provide invaluable information for analysis and design, but also make possible the transfer of ndings from one scale to another. The suggestion that successful cities are distinguished from those that will merely survive by their quality of life (Urban Age, 1999:11) also applies to housing, since it is de ned partly through its environmental quality pro le and linked to cities through the de nition of housing as a system of settings. 6 (Plafker, 1990; c.f. Rapoport, 1985). 7 Signi cantly the word dream is often used in connection with housing, linking housing to ideal images and there are striking overlaps in the environmental quality components and evaluative images of cities and housing (e.g. Nasar, 1998). EXAMPLES Very few examples are discussed, and these are intended to support my claim that whereas it is impossible to link culture and housing (or any other environment), it is easy to use the variables shown in Figure 4, especially those with wide arrows. As they also interact this helps with synthesis. Most of the examples are from newspapers (and some novels) to show that much data are outside the research literature. This makes it much easier to develop lateral connection and conceptual frameworks and to integrate diverse material on the way to theory development. Starting with the horizontal axis (breadth and generality to speci city), I discuss values, ideals etc.; norms, expectations, etc.; lifestyle and activity systems. Along the vertical axis (excessive abstractness to concreteness), I consider kinship, family structure, roles, social networks, institutions, status and identity. Horizontal axis in Figure 4 Values. These describe how people value various goods (and play a role in economics). This leads to differences in resource allocation, which is also the operational de nition of lifestyle (Michelson and Reed, 1970; c.f. later). Values are studied in marketing, including of housing (e.g. Hall, 1988; Rapoport, 1985,
6

151

Of course, the bulk of any settlement is composed of housing, especially as de ned here. 7 In this article the word dream appears, linking values to images, ideals, etc., therefore, also to standards and norms. 8 This has implications for the importance of open-ended design in housing.

1995c (1985), 1998). Typically, tradeoffs must always be made; in housing they are re ected in the relative value attached to the dwelling, the neighborhood, location within the settlement, taxes, microclimate, school quality and other environmental quality components, as well as other goods (e.g. Kaitilla, 1992; Rapoport, 1985, 1995d (1990)). This helps to explain the choices made. Values also help de ne groups and make housing particularly important, because dwellings play an important role in acculturation and, hence, the survival of groups through the transmission of values, linking values to family (Rapoport, 1990c:6570, Sebba, 1991). Major differences in values distinguish designers and users as a whole (and can also be understood as differences between outsiders and insiders (Brower, 1988, 1989)). This is in addition to differences among groups of users I have emphasized, and their small size, both because users values are more important and because most housing is not professionally designed. For example, in gentri cation, initiators have very different values from later arrivals who, depending partly on the success of gentri cation, may move out when they have children, so that school quality, playspace and safety become more important (e.g. Kleiman, 1988 8; c.f. Suchar and Rotenberg, 1994). In addition, religious values of three groups lead to different problems in standardized housing, the more severe problems being linked to activities rather than symbols (Chua, 1988). Values help explain many preferences and choices. For example, private open spaces, such as backyards, are valued more than public open spaces (which are also valued) (Peiser and Schwann, 1993). Urban images in Africa are positive, whereas those of the bush are negative (Olofson, 1975). This in uences the location of professionals at the macroscale (e.g. Gould and White, 1974) and preferences for housing types and materials (e.g. Beckman, 1976). This is very different from the suburban, exurban and rural values of many developed countries. The relation among values and landscaping (native vegetation, lawns, trees, etc.) is well known and commonly found in newspapers. Thus, in Germantown, Wisconsin, USA, the city, under police protection, compulsorily mowed a natural lawn regarded as weeds and as inappropriate for an urban community, because the grass was over 12 inches tall and there were unapproved plants; the battle prior to the compulsory mowing lasted for 2 years and the residents have since decided to leave (Cioni, 1993a; c.f. Rapoport, 1990c). Neighbors supported them and put up pink amingos and spinning plastic owers which, apparently were acceptable (Cioni, 1993b). That values
Housing, Theory and Society

152

A. Rapoport adjoining lots, different siding, garage doors, etc. (Zavis, 1998, although excessive differences are also rejected (e.g. Rapoport, 1990c). Similar con icts concern house styles (Rapoport, 1990c) such as con ict about a modern house in a small New Hampshire town (White, 1988) and a con ict between locals and Laotian immigrants in rural Minnesota regarding some temporary shelters on vegetable plots; some locals described these shacks as an embarrassment (New York Times, 1987). Values, leading to norms and standards are clearly involved, as they are in con icts in Queens, New York, in an area described as consisting of nice, private houses; the complaint was that the suburbs are becoming the city (Bruni and Sontag, 1996; c.f. Rapoport, 1990c, Ch. 1.). Multiple occupancy was communicated by doorbells, mailboxes and utility meters, many cars lining the streets, many people sitting on stoops, more garbage and garbage containers, for rental signs, basement doors, converted garages and backyard shanties i.e. many semi- xed and non- xed cues. Ideals, images, schemata, meanings, etc. Values are often expressed through such ideals, images etc, linking these topics and making the use of advertising, novels, TV, lm, etc., so useful and important (Rapoport, 1977, 1985, 1990c, d, 1995c (1985)). Thus, an advertisement for the Historic Third Ward (note the name) (Milwaukee Journal Sentinel , 1998) emphasizes a prestigious lifestyle (linking images to lifestyle and status); Milwaukee at your doorstep; art galleries; theaters; Lake Michigan; fantastic views; shopping; chic shops; entertainment; stylish restaurants, etc. An image is built up, describing a particular lifestyle and environmental quality pro le. Similarly, an advertisement for Unitex (Architecture Australia 1998) shows a fac ade with roof pediment, window pediments and columns (which Unitex sells), captioned turn your home into a palace. The media (e.g. TV) shape not only fads and fashions but environmental images and body images, as in Fiji it has lead to eating disorders (Goode, 1999). The importance of images implies latent functions (meanings) which should be expected and sought. The ideal image is more important than instrumental utility, although the latter imposes constraints on choice (Shang, 1999). Since ideal images change, openended design is important (Rapoport, 1995b (1990/ 91)), as it also is for aging, so that dwellings can be modi ed without leaving familiar surroundings (Van der Voordt, 1990; Teltsch, 1993). Instrumental utility also in uences transformations, reinforcing the importance of open-ended design (Rapoport, 1995b (1990/ 91); Salama, 1995; Brown, 1996a). A striking example of the power of images is the

are involved is made even clearer by new subdivisions that feature natural landscaping, such as prairie grasses and wild owers the residents of which clearly share those values (e.g. Brown, 1998a). Some outsiders still dig out such plants re ecting values leading, through ideal images, to acceptable or unacceptable standards, norms and rules; these change over time and can be studied (Rapoport, 1977, 1990c; c.f. Pavlides, 1985). The high value given to natural landscaping in these cases becomes an ideology, which unites the individuals and groups in question. Taken further, it leads to the development of ecologically sensitive co-housing. This is run by many committees; house costs are low and resources (e.g. snow blowers and washing machines) are shared, gardening is organic and public transport tends to be used (Brown, 1998b). 9 This sense of ideological community is clearly based on shared values. A recent exhibition in Montreal (The American lawn: surface of everyday life) (Muschamp,1998) dealt with the importance of lawns in Anglo-American culture. There are now 32,000,000 acres of lawn in North America, more than is used for the cultivation of wheat, corn or tobacco. $750 million is spent annually on seed, and $25 billion for various tools. Clearly much more is involved than an area of grass there are deeply held values and meanings, expressed in ideal images seen as a symbol of paradise (a term used both in the exhibition and by Muschamp (1998; c.f. Rapoport, 1990c)). Values are also involved in house colors. In my own neighborhood social pressure forced a resident to subdue bright house colors, and there have been con icts about house colors between Portuguese immigrants and locals in Toronto and Montreal, Canada. In Redmond, Washington, USA, a house, in an area of gray and brown houses, was painted mauve, purple and teal. A court case resulted; under threat of imprisonment and heavy nes the house was repainted, but the owner plans to leave (Egan, 1993). Con icts such as these are growing because, as of 1993, 32,000,000 people in the USA lived in communities governed by covenants and restrictions enforced by homeowner associations (Egan, 1993). There are also laws to prevent monotony, addressing the problem of excessive sameness (a negative value and image), so that anti-monotony ordinances have been spreading across the USA since the 1970s. These prescribe variations in color, roo ine, window design, placement of houses on lots, different heights and facades on
9

In 1998, 30 co-housing schemes had been built in the USA, and 100 were in progress (Brown, 1998). Participants have different values to other groups in the USA, that seek large, expensive houses (as discussed earlier; c.f. also Baumgartner, 1988). Also, this number is still rather small compared with Europe and to suburban developments.

Housing, Theory and Society

Theory, culture and housing suburban house. Thus, in Bangkok, Thailand, English Tudor houses were the most popular, followed by Mediterranean, Bavarian half-timbered; Roman-villa styles were also popular (Milwaukee Journal , 1988; H. Ross, personal communications in the mid-1990s). Europeans tended to favor Thai houses, in the same way that at a reception during a conference in Japan, I noticed Japanese participants drinking Scotch, whereas Americans drank sake. Also in Japan, western houses have become popular. European (e.g. Swedish) or US houses are imported, as are house images (Andrews, 1995). Victorian houses are also popular, with prefabricated Victorian houses imported from England, with minor modi cations for Japanese lifestyles (Lawson, 1993). Exteriors and interiors of Tokyo suburban houses are often completely Western (New York Times, 1986). I have observed the popularity of California Spanish suburban houses in Indonesia, even in spontaneous (squatter) settlements. In new developments in Jakarta, If you squint. youd think it was another Irvine [California] (Vrana, 1994). Provided are biking paths, gated entrances, etc., as in southern California housing developments. Attempts to design more culture-sensitive houses were rejected in favor of Western houses (Vrana, 1994). A sales pamphlet from a new development in Yogyakarta, Indonesia (Merapi View: Exclusive and Nostalgic; note the use of an English name) projects a suburban image and, except for the Indonesian text, the houses illustrated are dif cult to distinguish from those elsewhere, even when described as Indish or Vernacular. The facilities advertised (in English) include: tennis court, tness center, clubhouse, restaurant, sauna, minimarket, drug store, jogging track, playground, swimming pool, 24-h security, satellite TV dishes, underground wiring, telephone and air conditioning. This is similar to the environmental quality pro le of USA subdivisions described earlier (and to the very upscale town of Kochav Yair in Israel). Thus, ideals, images, etc. help explain not only cultural differences, but also convergence, probably due to the impact of images in the mass media. Suburbs of Moscow are also similar, where the detached houses are described as American, with wall-to-wall carpeting, two-car garages, cul-de-sacs and manicured lawns dotted with barbecue grills and picnic tables; there is a six month waiting list (Myre, 1997). Similarly, a 1997 photograph of the rebuilt South Bronx, New York, (compared with a 1980 photograph) shows suburban houses with lawns and fences (New York Times, 1997; see Rapoport, 1990c:160163 for a Milwaukee example.) The elements in the image of good housing used in advertisements, novels, TV, lms and newspapers, help explain choice, and can be used in design. Thus,

153

redesign of public housing in Boston involved replacing components of negative environmental quality (a desolate institutional style, monotony, a sea of asphalt surrounded by a high inhospitable chain-link fence) by positive images of quality private housing pitched roofs, bay windows, front yards, front doors with private addresses, low wrought-iron fences, backyards, landscaping and irregular con gurations and different colors to communicate individuality (Deitz, 1984; c.f. Krolls redesign at Alenc on, France (Architecture Interieure , 1981: esp. 6263)). Norms, standards, expectations, rules, etc. Values, images, etc. lead to norms, standards, rules, etc.; their application results in environments that approach the ideals as closely as possible, given various constraints. These variables also help explain the evaluation of environments (Fig. 1). Thus, the discussion above about Laotian shacks (New York Times, 1987) involves not only values, but the resulting norms, expectations and standards, as do con icts about brightly colored wooden shanties (Casitas) in Puerto Rican community vegetable gardens in New York (Gonzales, 1990). New Yorkers evaluate them negatively; they also violate building codes, ownership laws, etc., but to their builders they are attractive, they become gathering places and centers of community life, music, crafts and acculturation; children learn traditional dances and music, etc., [they become institutions related to identity , both to be discussed later] (Gonzales, 1990). Such con icts between the norms, standards, etc., of different groups, are common and may concern the use of streets or the very existence of garden plots (Blackwell, 1981). In a Los Angeles neighborhood, the types of people (Russian immigrants), their dress, use of streets and other spaces, levels of house maintenance and other cues, at odds with Anglo-Americans, resulted in the area being de ned as a slum (Spalding, 1992). Such cues in uence how areas in USA urban areas are evaluated (e.g. Halle, 1984; Weiss, 1988; Rapoport, 1990c). Such phenomena begin to explain the variability of standards (e.g. Choldin, 1976). Space use within dwellings also re ects standards and norms. Different groups may continue to use space differently even after acculturation, with consequent con icts (e.g. between generations), although eventually space use will conform to the new expectations, norms, standards and rules (and thus values), especially among offspring of the original population (Kent, 1984, 1990; Pader, 1993; Cooper and Rodman, 1995). The application of rules also leads to styles and the identity of cultural landscapes, (e.g. the choice model of design; Rapoport, 1977, 1990c). As already mentioned, rules can be unwritten or written, informal or
Housing, Theory and Society

154

A. Rapoport architecture I saw a very small house, half of which was a formal living room, hardly ever used. Currently, while houses in the USA are becoming larger, living rooms are shrinking or even disappearing (Iovine, 1999), having gone from being central to the dwelling to being of no importance, replaced by family rooms, linked to the kitchen; many new houses have none at all. Thus, in Celebration, Florida, behind the traditional facades, there are modern, single space interiors with, at most, vestigial living rooms for display or nostalgia (Krause, 1997). New values and norms result in new lifestyles affecting housing. Earlier I discussed the almost universal positive image of the suburban house. Some groups, as their lifestyles change, move back downtown (Brooke, 1998), although the numbers are still very small compared with suburban growth. Converting warehouses and the like, or building new downtown apartments illustrate this lifestyle shift. Finally, in gentrifying urban areas, the small size of groups if de ned by lifestyle reappears. In a small area in Chicago (Lincoln Park) three different lifestyle groups were found, with very different evaluations of housing, use of housing and hence furnishings and decorations. For one group the dwelling provides a stage for social performance; for the second it needs to function as a setting for expressing ones unique individuality; for the third it needs to sustain an atmosphere of private family life and domesticity (Suchar and Rotenberg, 1994). These are not only related to lifestyle, but to values, images, norms and activity systems, to status and identity and to the other variables in Figure 4. Activity systems. Activities, the most speci c expression of culture, need further dismantling. A four-fold division is useful (Rapoport, 1990a, 1998, Fig. 3, 2000, Fig. 1), starting with the activity itself (its manifest or instrumental aspects), then how it is carried out, how associated with others into a system and, nally, the meaning of activities (their latent aspects). Variability increases and latent aspects are most variable and the most important (e.g. Esber, 1972; Zeisel, 1973). They help to explain the variability of dwelling and settlement forms and also suggests that the distinction between function and meaning is misconceived meaning may well be the most important function of built environments, especially housing. It is useful to ask who does what, where, when, including/excluding whom (and why). The answer determines how activity systems are distributed in systems of settings and varies among groups (Rapoport, 1977, 1990a) and may change over time. The separation of work and dwelling during the industrial

formal (legalistic). Rules are thus not only central in design, guiding choice among alternatives, but also guide appropriate behavior in settings (Rapoport, 1990c). In a US suburb, for example, unwritten rules make most behaviors generally inappropriate in public settings and also in uence behavior within houses (Baumgartner, 1988). Lifestyle. I have argued for some time that lifestyle is the most useful criterion for de ning groups, especially today. It is more general than activity systems but more speci c than the variables discussed so far. Also, all other variables in Figure 4 can be said to become relevant only when they in uence lifestyle and thus choice, whether among city, suburb or small town/rural (Hummon, 1990) or of housing (including design and use). Other aspects of culture not explicitly listed (e.g. education, class, race, ethnicity, religion, etc.) only relate to housing if they affect lifestyle; otherwise they become relatively insigni cant for our purposes. Lifestyle itself needs further dismantling. A useful rst step is an operational de nition (Michelson and Reed, 1970), which I have long used. One can also identify the speci c components involved and to describe lifestyle as a pro le which can be matched against environmental quality pro les (Rapoport, 1985; 1995c (1990) esp. 484486). For Hasidic Jews, a homogeneous neighborhood on the basis of religion and the resulting lifestyle is critical (c.f. Rapoport, 1997b). The neighborhood needs to be compact, since driving is not allowed on the Sabbath and certain holidays. This, in turn, leads to the mingling of different income levels (unusual in the USA). Within the dwelling, kitchen design re ects the need to separate dairy and meat among observant Jews (Sontag, 1998). In a rural Hasidic community (Berger, 1997) con icts with other residents concern lowdensity and rural quality. Hasidim convert houses to religious schools (Yeshivas) with crowding when 53 boys occupy two houses, and families with 12 children living in a single apartment raise densities. This is exacerbated by the introduction of multiple dwellings into an area of single family houses, which together with synagogues and Yeshivas bring outside people into quiet lanes [c.f. Baumgartner, 1988], and exacerbated further by Hasidim turning cul-de-sacs into shtetls (Eastern European small towns) (Berger, 1997). There are also con icts with the law concerning re codes, zoning and other regulations. Essentially, there is a clash between cultures (lifestyles, norms, etc.) that has split the town apart (Berger, 1997), suggesting the need for homogeneity, i.e. clustering by lifestyle (c.f. Rabkin, 1994 on Kiryas Joel, New York). In the Danish outdoor museum of vernacular
Housing, Theory and Society

Theory, culture and housing revolution changed not only work settings, but dwellings, settlements, roles, activity systems and so on (e.g. Halle, 1984; Jackson, 1985; Fishman, 1987). The current growth of dwelling-based work is having major effects on both manifest and latent aspects of activities and on settlements, neighborhoods and dwellings. Within the dwelling it in uences activity systems, space and time use and organization, furnishings and so on. Visits by clients, customers or colleagues make important the environmental quality of the neighborhood, communicating professionalism, status and identity at least in the USA (Ahrentzen, 1989, 1990). Changes in roles, family structure, etc., discussed below, change lifestyles and, hence, activities so that among career families house cleaning either becomes less important (e.g. Belkin, 1985) or cleaning help is used. Cooking may become less important (except as a hobby), replaced by eating prepared foods or eating out. These in uence the use of parts of dwellings and the larger system of settings. The fear of crime (or rising crime) often changes activities. People do not go out at night, do certain things or visit certain areas. As crime (at least in the USA) is currently going down, predictions are possible about changes in activity systems (and hence lifestyles). Activity systems are also in uenced by changes or differences in mobility, due to aging, handicaps, income, car ownership, the availability of public transport, car-oriented milieus that discourage walking, and so on. Activity analysis helped Sun Rhodes (1993) arrive at a culture-speci c American Indian dwelling, linking activities with identity (as discussed below). Activities such as religious processions and the like clarify space use and settlement form as in temple cities such as Madurai in south India, village form in Sri Lanka (Bechhofer, 1989) and urban space use in Mexico (Brown, 1995; Rapoport, 1990a, c). In India, national and regional space can be organized through pilgrimages. Since activity systems are related to lifestyle, values norms, ideals, etc., they also in uence evaluation of environmental quality, as shown by work on Australian Aboriginal housing a useful model system (e.g. Heppell, 1979; Ross, 1987, 1991; Morel and Ross, 1993). Group membership in uences the extent of activity systems, such as home ranges, often resulting from rule systems internal or external and thus peoples knowledge of environments, their mental maps (Wheatley, 1976); so do judgements about safety (e.g. Gaster, 1991). Since housing is part of a system of settings, what happens (or does not happen) in some settings in uences what happens (or does not happen) in others. Variations in home ranges thus greatly

155

in uence what happen in dwellings. In addition, where activities occur depends on the perceived appropriateness of settings for such activities (Kamau, 1978/79; Baumgartner, 1988; Rapoport, 1990a, c). Vertical axis in Figure 4 Kinship. In traditional societies in particular, identifying kinship patterns is indispensable to understand housing (e.g. Shokoohy and Shokoohy, 1994; Shrestha et al. 1997; Chandhoke, 1990; Bourdier and Trinh, 1996; Leigh and Asojo, 1999; Oliver, 1997), e.g. through clustering, since kinship is an important form of homogeneity at the neighborhood level. Kinship helps explain settlement form, neighborhoods, street patterns, house clusters, and house forms, such as communal dwellings, densely clustered courtyard housing relying on direct inter-house access (e.g. Bonine, 1980) or compounds of multiple dwelling (e.g. Schwerdtfeger, 1982; Stahl, 1991) all were made possible by kinship relationships. As for other forms of homogeneity it is the subjective (emic) de nition of kinship that is crucial. This can also affect the perception of crowding (at given physical densities) which may occur when strangers are involved but not kin (as in China and Hong Kong). Also traditional kinship clusters may persist, albeit in new forms such as the family circle in Scandinavia (Gaunt, 1991). Kinship is also important in developing countries, among traditional groups, but is disappearing among modernized groups, among whom other forms of homogeneity (income, class, lifestyle, etc.) become more important. For example in Kirtipur, Nepal (Shokoohy and Shokoohy, 1994; Shrestha et al. 1997), the importance of the extended family (kinship group) meant that ancestral dwellings were subdivided and/or extensions added, increasing density, as rear vegetable gardens were built over. In addition, kin clustering resulted in domestic and social activities in streets and squares. With kinship less central, life now is much more private and activities are (nuclear) family centered in the interior of houses. Yet as late as the 1950s, kinship could still play an important role in residence patterns in the matri-local clusters in Bethnal Green (in the East End of London, UK) (Young and Wilmott, 1962; Wilmott, 1963) which were disrupted with resettlement in Dagenham, Essex, UK. The lesser contemporary role of kinship may, however, be reversed, either naturally or by design. The former may be due to the aging of many populations and consequent use of granny- ats (and their equivalents) as in Australia, Japan and elsewhere, various forms of co-housing (Hinds, 1985) or houses shared by young and old. The growth of single-parent families may result in communal houses
Housing, Theory and Society

156

A. Rapoport ence family life positively or negatively (Rosser and Harris, 1965; Gasparini, 1973; Pruchno et al. 1993). As always, family housing interaction is two-way: family type and organization in uence dwelling which, in turn, affect the family, particularly when time-lags occur another reason for open-ended design. In general, the household and family have been the same. There are exceptions, although even in extreme cases, such as communal dwelling (which are also settlements), the family, however de ned, is still the basic unit of the dwelling community. In addition, many current changes in housing (and even settlement) form and use are intimately related to changes in the nature of the family unit, which also affect lifestyle, roles, activity systems, etc. (e.g. Despre s, 1991; Franck and Ahrentzen, 1991). Changes (or differences) in family structure (single parent, working couples, extended families, co-housing groups, young and elderly, large families, polygamous families, etc.) not only in uence dwelling, (e.g. size, kitchens, living rooms, etc. (c.f. Baumgartner, 1988)). Household size is basic to housing demand (Sternlieb and Hughes, 1986:14). In general, household size in the USA has been shrinking, with major implications for housing. Through systems of settings, this affects urban forms. Smaller households increase the area of housing, because kitchens, bathrooms, parking, etc., are still needed, reducing density. Large families, tend to increase densities through additions, with a need for openendedness. In the case of a Hasidic community in Brooklyn, USA (Sontag, 1998) special zoning followed, allowing coverage of 65% of the lot, and reduced setbacks and backyards. Family size link otherwise very different groups, such as Hasidic Jews and polygamous Mormons (Sontag, 1998; Berger, 1997; Williams, 1997). Polygamous Mormons require numerous bedrooms (in one case 10), bathrooms (7), kitchens (2), nurseries (2), laundry rooms (2) and a school room (Williams, 1997). In both cases, family size demands space, and ease of expanding houses is critical. However, the different family structure in the two cases leads to different housing forms and use, and different settings and institutions at larger scales. Family is also important because it is a most important intermediate institution that helps people cope. It therefore becomes particularly important in the case of both urban migrants (e.g. Rapoport, 1983) and immigrants (e.g. Yans-McLaughlin, 1977). Roles. Changes in family structure have an impact on roles, but so do other forces. The different de nition of roles among different groups (i.e. cultures) has major effect on the organization and use of built environments. Roles probably have both a constant and variable component, so that when considered historically and

(Despre s, 1991). These can be seen as based on ctitious kinship as in some traditional societies, or as a link between kinship and family structure, not always easily separable. The development of multigenerational households can also cause con icts and problems, especially for men and teenagers (Pruchno et al. 1993), related to values, ideals, expectations, lifestyles and activity systems. An example of the renewal of kinship links by design may be the deliberate revival or retention of still existing patterns, in the design of supportive environments in developing countries (Morgan and Hiroshima, 1983; Rapoport, 1983). One can also deliberately use the natural clustering of extended families and kinship groups among recent migrants. Often place of origin, language, religion, etc., may replace kinship (Rapoport, 1977, 1997b; Belluck, 1996), but clustering is discouraged , as in Wisconsin, USA, where a proposed Hmong development was rejected on constitutional grounds (Lisheron, 1994; c.f. Rabkin, 1994). The importance of kinship, however, may persist, so that among Hispanic groups in the USA a rent initiative was rejected to retain the possibility of extended-family co-residence (Los Angeles Times, 1983). Related are recent ndings about the non-use of child-care facilities by this group, because of a cultural preference for family-like care, based on a warm, family atmosphere with personal relationships, either by relatives or in family-like daycare, with one person in charge of a number of children in a house (Chira, 1994). There are links here with family structure, social networks, values, lifestyle, etc., which together result in different preferences of this group for house styles and the spatial arrangement of houses on lots (Wheeler, 1977), and also color, decoration, landscaping etc. (Arreola, 1981, 1988); when clustering can occur distinct cultural landscapes result (Rapoport, 1990c, esp. Fig. 21, p. 138). The need for a large and diverse sample, already discussed, means that traditional societies throughout history must be studied and in those cases must be considered, especially because kinship and family are not always easily distinguished. Thus, in Mexico family includes what in Anglo-American (and possibly other Western developed) countries would be considered distant kin (e.g. Stea, 1995). Family structure. Family structure is useful in relating culture to housing (e.g. Rapoport, 1999) and it in uences the form and spatial organization of housing (e.g. Ross, 1991; Stahl, 1991; Bourdier and Trinh, 1996; Cornell, 1997). Clearly, the congruence between housing and households is important (Netting et al. 1984; Wilk and Ashmore 1988; Wilk, 1989; Blanton, 1994), particularly since dwellings can in uHousing, Theory and Society

Theory, culture and housing cross-culturally, certain patterns will be found, although variations and changes clearly exist (possibly following epigenetic rules). At the settlement scale, changing womens roles (e.g. employment) means that womens driving increased by 50% and was largely responsible for the growth of traf c in London (Hawkes, 1998); the reduced use of public transport and the perceived bene ts of cars10 were also signi cant, so that it is easier to take children to and from school, and to shop by car particularly if one also works. Role changes in uence the system of settings at various scales: neighborhoods (e.g. Wheatley, 1976), dwellings (Stea and Prussin, 1978; Franck and Ahrentzen, 1991; Madigan and Munro, 1991) and parts of dwellings e.g. kitchens (Hassell et al. 1993). Thus, changes in sex roles in uence house types and space use both inside and outside the dwelling in Portugal (Lawrence, 1988). The increase in two-job families leads to changes in household roles. Men may take on new tasks or, as is often the case, women maintain traditional household roles while working, although this was beginning to change by the mid-1980s in the USA (e.g. Hood, 1985) with clear links to family/housing relationships. Another set of responses (already mentioned) involves changing standards of cleanliness (Belkin, 1985) or using outside cleaners. The details are less important than the fact that these various changes have an impact on dwellings and other settings in the system, such as the use of restaurants; prepared foods (shops), childcare facilities and so on. Even more important is the fact that not only is it relatively easy to trace relationships between roles (and changes in roles) and housing, but that these are also related to values, lifestyle, activity systems, family structure and all the other aspects (or expressions) of culture. Social networks. Kinship and family structure are types of social networks. Others have been studied e.g. with neighbors, friends, ctive kin, etc. (e.g. Frick, 1986, part B). They provide a useful way to relate housing and culture, and some aspects have already been discussed (e.g. Young and Wilmott, 1962; Wilmott, 1963; Shokoohy and Shokoohy, 1994; Stea, 1995; Shrestha et al. 1997; Chira, 1994; Sontag, 1998). Not yet discussed is their spatial extent, i.e. whether they are intensive, local, based mainly on neighboring 11 (loca10

157

The negative perceptions of public transport include inconvenience (journey awkwardness ), perceived lack of security, especially after dark and walking home from transport (related to crime), and cost; the perceived advantages of cars include convenience, security, cost, time savings, feelings of control of personal space and of the journey as a whole (Hawkes, 1998). 11 As already pointed out, neighboring itself can also be based on lifestyle, religion, language, caste, place of origin, kinship, ideology and other forms of perceived homogeneity.

lites) or extensive, based on shared interests, lifestyles, activities, hobbies, etc. (cosmopolites) (Rapoport, 1977, Ch. 5). It has been suggested that the former are being replaced by the latter, but it is far from proven (e.g. Rapoport, 1997b). Technology helps increase the spatial extent of networks, e.g. cars among Samoans in Los Angeles (Ablon, 1971), or telephones (Martin, 1967; Gaunt, 1991). Current discussion concerns the possible effects of computers and computer networks. In traditional Kirtipur, Nepal, religion and caste were important, and Buddhist and Hindu populations clustered around their respective religious institutions. Traditionally, caste and religion rigidly xed social position and ritual status and, through them religious rituals, social organization, marriage, personal relations, etc., and hence, social networks. The last were neighborhood centered, based on strong ties due to their homogeneity. Caste (as elsewhere in South Asia) was more important in Hindu than Buddhist areas and led to ner grain clustering. Currently such social networks are replaced by nuclear family-centered, dwelling-oriented, private relations due both to social changes and new dwelling forms; again, there is a twoway relationship (Shokoohy and Shokoohy, 1994; Shrestha et al. 1997; for a Turkish example, see Erman, 1996). Social networks may also change with circumstances, so that among immigrants religion or caste may be replaced by place of origin, enabling a small group to establish solidarity and identity. Social networks also vary with lifestyle, as in the case of two groups in a single community. For one group, it is the absence of networks that is typical, not only at the neighborhood level but even within dwellings. For the other group networks are still very important (Baumgartner, 1988). The rst group can also be contrasted with those for whom social networks are primary, e.g. Australian Aborigines (Heppell, 1979; Ross, 1979; Morel and Ross, 1987); Navajo (Kent, 1984), Mexicans (Wheeler, 1977; Pader, 1993; Stea, 1995) and so on. An important role of social networks is supportiveness (Rapoport, 1983), so that as criticality goes up, whether due to migrant status, age, health, resources, prejudice, etc., such networks become more important (e.g. Abu-Lughod, 1969; Badura, 1986; Margulis, 1992). Thus, one can not only explain various residential patterns, but begin to predict their occurrence and importance, with implications for policy and design. Social networks, whether intensive or extensive, although usually centered on the dwelling, may also be anchored by various group-speci c institutions, linking social networks, institutions and identity, to which certain institutions may be related. Institutions. Religious institutions have already
Housing, Theory and Society

158

A. Rapoport neatly into social tiers, like something a sociologist had built to prove a theory (Macdonald, 1971:67). Similarly self-evident is an English novel (Aird, 1981:1213), listing a series of cues that communicate respectability and status. Among these are: A nice, quiet neighborhood; decent houses; trees on the footpath; no through traf c and just local vehicles; grass verges; good gardens; near the tennis club and so on. 12 This also describes an environmental quality pro le. Although I often emphasize the importance of semi- xed elements, xed-features also communicated status and identity and can do so over surprisingly long periods (e.g. Cherulnik and Wilderman, 1986). Status may also be communicated partly through modern forms, styles and materials (such as concrete, metal window frames, etc.), and by semi- xed elements, such as cars, motorcycles, TVs, satellite dishes and the like (Shokoohy and Shokoohy, 1994; Shrestha et al., 1997: 61), so that status is not always expressed through built environments (Duncan, 1981, 1985; Rapoport, 1981). The differential investment in housing by groups for whom housing does or does not, in uences norms, standards and the like, can cause con icts (Rapoport, 1990c) and in uence the adaptation of urban migrants, e.g. in Colombia (Ashton, 1972; Kellett, 1999). How the expectations, values, ideals, images and norms of a correspondent in uence judgements of status is shown in an area of Beijing in which Chinese communist leaders live. Surprise is expressed that important, high-status people would live in an area which is a a maze of dusty lanes and grubby back alleys [with blank walls] punctuated by simple red doorways be tting ramshackle homes. The point is made that behind these doors lie elegant and spacious courtyard dwelling and a contrast noted between the limousines (visiting Dengs family after his death) and the dirt-and-cobblestone lane (Faison, 1997a; c.f. Hedges, 1994 later). This is a result of the expectation that dwellings should communicate status, and the role of new standards, materials, etc. These latter topics recur in the description of a New Town in China [Zhangjagang] seen as the nations ideal (Faison, 1997b). The photograph resembles a Scandinavian housing area, with two-story, outwardlooking buildings with pitched roofs. The emphasis on clean living clean streets and lawns and shrubbery (rare in China) re ect a new set of values which communicate prosperity, show obedience to the law and keep the area clean, orderly and fully modern. Sidewalks are of spotted red tile, there is a pedestrian shopping street, parking is controlled and there are strict rules about garbage. New residents are given a handbook about how to act, and they describe the town as a nice place to live (Faison, 1997b).

been mentioned, are generally important and differ among groups. In the case of the Orthodox Jewish communities discussed earlier, religious schools (Yeshivot), synagogues, Kosher shops and ritual baths are most important (Sontag, 1998; Berger, 1997; Rabkin, 1994). They are highly group-speci c and are often the reason for clustering, which needs to be tight and networks intensive, because of religious rules against driving on the Sabbath. Extensive networks may also be based on religious institutions, such as churches, e.g. among Samoans in Los Angeles (Ablon, 1971) and other groups there (e.g. Serbians). Institutions can also be related to identity and, hence, group speci c (e.g. cultural centers among American Indian groups or churches among Blacks (Leigh and Asojo, 1999)). Important institutions may include tea-shops in Korea, coffee houses in Turkey, Bodegas among Puerto Ricans in New York (Hoffman and Fishman, 1971, c.f. Casitas discussed earlier (Gonzales, 1990)), outdoor markets, laundromats, stables, etc. (Rapoport, 1977, 1986, 1990c). Some may be counter-intuitive, based on latent functions. Thus pubs or taverns are central for men in certain groups (Young and Wilmott, 1962; Wilmott, 1963; Suttles, 1968; c.f. Rapoport, 1977, Ch. 5). The role of various settings as important institutions needs to be discovered rather than assumed. Particularly striking is the case of Nantucket, Massachusetts, USA, where for 60 years the town dump had been an important institution and central setting community center, trading post and party hall (Crossen, 1990). Its forthcoming closure has serious social implications. Similarly book shops and museums play a latent role as settings for meeting single people with compatible values, lifestyles and interests who would not use singles bars, which attract very different groups (New York Times, 1985). Signi cantly, the groups are identi ed by the particular settings they use. Status. Built environments, especially housing, play an important role in communicating status, especially in contemporary societies (such as the USA, Australia, etc.) with no rigid status categories (Rapoport, 1988, 1990c). Its increasing importance in communicating status can be seen in India with the change from traditional ( xed status) to modern groups (Duncan, 1981, 1985). When housing does play that role, it is striking how apparently clear and self-evident such communication is (Rapoport, 1990c esp. Ch. 6). For example, a novel describes a settlement in where the location and types of houses (and other built environments) communicate status. It is said to be divided
12

Note that nice, decent and good themselves consist of, possibly, large sets of cues and need further dismantling.

Housing, Theory and Society

Theory, culture and housing Hassan Fathys well-know example of New Gurnia failed, at least partly because of the use of mud-brick and a form derived from Nubia, a low status area of Egypt. It is signi cant that his private houses for wealthy clients, with the same characteristics, have proved highly successful. The signi cance of mudbrick is shown by another unselfconscious use of a correspondents values, standards and images, in a story about an Egyptian village. The photograph caption refers to a village so poor that houses are built with mud bricks, and in the text the point is made that unable even to afford cement, the 50,000 villagers in the Gharb al Banawaan area still live in mud-andwattle homes with dirt- oors (Hedges, 1994). The role of materials has recently been studied empirically and con rms theoretical expectations. Thus, Sadalla and Sheets (1993) show that materials have meaning, both intrinsic and culturally speci c (i.e. constancy and variability both play a role). The social meanings of materials also help de ne social identity (as discussed below). Thus, the rejection of bamboo as a building material in Ecuador is due to its meaning as a traditional, and hence poor, substandard material (Livingston, 1992; c.f. Beckman, 1972 on Liberia and Kaitilla, 1991, 1994 on Papua New Guinea). There can also be status differences between timber and stone (as in Scotland (G. Slaven, personal communication 1999)) and brick. Hence, the common use of brick veneer over what are timber houses or apartments (e.g. Australia and the USA). In villages in developing countries one nds new materials and forms and manufactured paint with its new colors used rst by high-status individuals. Color can also be used for identity, as in identifying Moslem houses in parts of India (e.g. at Jamshedpur), or as in the case of con icts between Portuguese immigrants (who use very bright colors) and locals in Toronto and Montreal (c.f. the case of Mexicans in the USA (Arreola, 1988)). The use of the full spectrum of cues new materials, new styles and forms, height, glazed windows, color, plants and landscaping, furniture, etc. tends to be used by high-status individuals and groups, and then trickles down over (sometimes long) periods of time, as was the case in India (King, 1984, esp. Ch. 1) where it is still not complete (e.g. Grenell, 1972; Grimaud, 1986). Similarly with the use of a western (Paladian) style by e lites in 19th century Cairo, Egypt (Asfour 1993). Although here the traditional organization and use of space was kept so that the new forms connote social prestige and status, but traditional ideas about family privacy, family structure
13

159

and guest reception persist. This shows syncretism, which is still to be found (and to be encouraged) in developing countries (Rapoport, 1983). Identity. The many different groups discussed earlier, as well as individuals, all need to establish and maintain identity one of the roles of culture and dwellings and other built environments often communicate identity (e.g. Nasar, 1998; Nasar and Kang, 1989; Leigh and Asojo, 1999). It is also signi cant that in some developing countries efforts (usually unsuccessful) are made to retain (or create) identity through physical environments (e.g. Barnard, 1984). We have already seen that currently identity is a much more complex matter than in the past, sometimes with many overlapping group memberships (e.g. Gordon, 1978), with more choice and a greater emphasis on individual identity, which can arise even from ethnic identity (Goldberg, 1987). This makes personalization of housing (and hence open-endedness) more important. Given the importance of identity almost any element can be used (e.g. Leigh and Asojo, 1999); the con ict in Minnesota discussed earlier is also related to identity. The Laotian shacks were said to make the neighborhood look like a Southeast Asian village, whereas locals wanted something Americanized (New York Times, 1987). It thus seems relatively easy to relate identity to housing, built environments and material culture generally. These may be landscape settlements, neighborhoods, buildings and interiors, materials, space organization and use, and various semi- xed elements (Kamau, 1978/79; Greenbaum and Greenbaum, 1981a, b; Kron, 1983; Hamilton, 1984; Environment and Behavior , 1987). Clothing, beards, transport, places of worship, etc., may also be used (e.g. the Amish), or among Hasidim, clothing, language, hairstyles (beards and earlocks for men, wigs for women), behavior and the specialized institutions already discussed (Sontag, 1998). These are also used by many other groups, as are the many components of ambience (Pred, 1963; Chudacoff, 1973; Ginsberg, 1975). Almost any aspect of material culture can be used, so that body decoration, facial perforation and even food can be used to establish identity. Although the same foods are available, different Indian tribes in the Amazon rainforest have different diets, which are used by anthropologists to identify the tribes. The diets differ, because the choice made from an extensive repertoire is used to distinguish, or isolate oneself from ones neighbor (Gibbons, 1992). What is eaten or avoided is used to mark cultural boundaries. Even in the USA, Thanksgiving foods differ among ethnic groups, maintaining their identity (Gibbons, 1992). 13 We have already seen that Casitas in New York become identi ed with Puerto Ricans (Gonzales,
Housing, Theory and Society

A student of mine, Paul Maas, analyzing housing advertisements in Chicago, titled his term paper You are where you live, which echoes Gibbons (1992) You are what you eat.

160

A. Rapoport
Samoan community, SW Journal of Anthropology 27 (1) Spring: 7596. Abu-Lughod, J. (1969) Migrant adjustment to city life: the Egyptian case, pp. 376388. in Breese, G. (ed.), The City in Newly Developed Countries. Princeton: Princeton University Press. Ahrentzen, S. B. (1987) Blurring Boundaries: Socio-spatial Consequences of Working at Home. Research Report. Milwaukee: University of Wisconsin-Milwaukee, Center for Architectural and Urban Planning Research. Ahrentzen, S. B. (1989) A place of peace, prospect and a P.C.: the home as of ce, Journal of Architecture and Planning Research 6 (4) Winter: 271288. Ahrentzen, S. B. (1990) Managing con ict by managing boundaries: how professional homeworkers cope with multiple roles at home, Environment and Behavior 22 (6) November: 723752. Aird, C. (1981) Some Die Eloquent . New York: Bantam Books. Anders, C. M. (1991) Demographics in uence home sales techniques, Honolulu Sunday Star-Bulletin and Advertiser July 14. Anderson, E. N. (1972) On the folk art of landscaping, Western Folklore 31 July: 179188. Anderson, J. R. and C. K. More (1972) A study of object language in residential areas, Mimeo. Department of Architecture, University of Illinois, Champaign Urbana, Andrews, E. L. (1995) Raise the roof beams Japan (A boom in imported housing built the Western way), New York Times September 14. Architecture Australia (1998) 87 (5) September/October. Architecture Interieure (1981) Kroll apres la me me [Kroll in his own words], (181) February/March: 5467. Arreola, D. D. (1981) Fences as landscape taste: Tucsons barrios, Journal of Cultural Geography 2 (1): 96105. Arreola, D. D. (1988) Mexican American housescapes, Geographical Review 78 (3) July: 299315. Asfour, K. (1993) Cairene traditions inside Palladian villas, Traditional Dwellings and Settlements Review 4 (2) Spring: 3950. Ashton, G. T. (1972) The differential adaptation of two slum subcultures to a Columbian [sic] housing project, Urban Anthropology 1 (2) Fall: 176194. Badura, B. (1986) Social networks and the quality of life, pp. 5560 in Frick D. (ed.), The Quality of Urban Life . Berlin: de Gruyter. Bagla, P. (1999) News of the week: India prepares research policy analysis, Science 283 (5400) January 15: 309. Barmash, I. (1988) Splintering of markets: a ne art, New York Times August 22. Barnard, B. (1984) Cultural Fac ade: Ethnic Architecture in Malaysia. USFI Reports (1984/5, Asia). Hanover, New Hampshire: Universities Field Staff International. Bauhain, C. (1986) Lhabitation Japanaise: tradition et modernite [The Japanese dwelling: tradition and modernity], Architecture and Behavior 2 (3/4): 229259. Baumgartner, M. P. (1988) The Moral Order of the Suburb . New York: Oxford University Press. Bechhofer, W. (1989) Procession and urban form in a Sri Lankan village, Traditional Dwellings and Settlements Review 1 (1) Fall: 3948. Beckman, J. (1976) A decision model for estimating concurrent school attendance among tribal people of Liberia together with an application regarding differential cognition toward traditional housing, pp. 275285 in

1990); as are specialized shops (Bodegas) (Hoffman and Fishman, 1971). Also used is a speci c aesthetic complex, a set of elements arranged in speci c ways, in this case in public housing in Boston (Jopling, 1974). Also used for identity generally are plant materials and their organization (Kimber, 1966, 1971, 1973; Anderson, 1972), fences (Anderson and Moore, 1972; Arreola, 1981) and many other items (Rapoport, 1981, 1990d). Particular housescapes, comprising front fences, brilliant colors and occasional religious shrines (Arreola, 1988) identify Mexican-Americans and changes toward the Anglo norm depend on the length of residence in the USA It seems that complaints about the lack of identity in suburbia are mistaken and identity actually ourishes (in Phoenix subdivisions). Identity varies with location, age of developments, the population groups involved and speci c names of the developments [c.f. Rapoport, 1977, Ch. 2]. A variety of cues is also used and personalization, is more common in lower-priced areas, whereas institutional design is more common in higher-priced subdivisions (Blake and Arreola, 1996). Similar distinctions are found in Milwaukee, USA, between lower status areas with higher degrees of personalization than in higher status areas (Janz, 1992). CONCLUSION These few examples show that housing needs to be explicitly de ned, and culture, environmental quality, etc., dismantled and made operational. Whereas culture (or society) is not useful, the speci c, more concrete results of dismantling are easily related to housing seen as systems of settings with speci c environmental quality pro les. These also link a large number of elds other than EBS, and make possible the use of a wide range of material, including popular media. In addition, as was to have been expected, the variables are related. Since holism is impossible, scienti c holism requires: (i) that all the variables identi ed be studied; (ii) that they then be reassembled or synthesized; and (iii) that the relative contribution of all the variables be studied, both in general and for any given case or situation. The interplay of constancy and change, uniformity or variability and generality or speci city have not been considered, but are essential. It is to be hoped that the approach discussed might stimulate these further steps that are needed in order to relate housing to various groups within a scienti c framework of an explanatory theory of EBR. REFERENCES
Ablon, J. (1971) The social organization of an urban
Housing, Theory and Society

Theory, culture and housing

Rapoport A. (ed.), The Mutual Interaction of People and their Built Environment . Chicago: Aldine. Belkin, L. (1985) Cleaning no longer gets top priority, New York Times April 11. Belluck, P. (1996) Turkish Mosaic grows in New York, New York Times April 2. Berger, J. (1997) Growing pains for a rural Hasidic enclave, New York Times January 13. Blackwell, M. I. (1981) Co-op gardens: eyesore or source of pride?, Milwaukee Journal November 5. Blake, K. S. and D. D. Arreola (1996) Residential subdivision identity in metropolitan Phoenix, Landscape Journal 15 (1) Spring: 2335. Blanton, R. E. (1994) Houses and Households: A Comparative Study . New York: Plenum. Bonine, M. E. (1980) Aridity and structure: Adaptations of indigenous housing in Iran, pp. 193219 in K. N. Clark and P. Paylore (eds), Desert Housing, Tucson, University of Arizona Press. Bourdier, J-P. and Trinh T. M-H. (1996) Drawn from African Dwellings. Bloomington Indiana: Indiana University Press. Brooke, J. (1998) Denver stands out in trend toward living in downtown, New York Times December 29. Brower, S. (1988) Design in Familiar Place. New York: Praeger. Brower, S. (1989) Residents and outsiders perceptions of the environment, pp. 189202 in Low S. M. and E. Chambers (eds), Housing, Culture and Design. Philadelphia: University of Pennsylvania Press. Brown, D. F. (1995) Un espacio construido e mero: Las alformbras de Assern en Humantla, Tlaxcala [A shortlived constructed space: The Alfombras de Assern] , pp. 157168 in Amerlinck M-J. (ed.), Hacia una Antropologia Arquitectonica [Toward an architectural anthropology]. Guadalajara Mexico: University of Guadalajara. Brown, P. L. (1996a) A house that bends to a familys needs, New York Times September 19. Brown, P. L. (1996b) In a clash of decades a house surrenders, New York Times November 14. Brown, P. L. (1998a) It takes a pioneer to save a prairie, New York Times September 10. Brown P. L. (1998b) For leftist guinea pigs, the ecological good life, New York Times April 16. Bruni, F. and D. Sontag (1996) Behind a suburban fac ade in Queens, a teeming, angry, urban arithmetic, New York Times October 8. Chandhoke, S. K. (1990) Nature and Structure of Rural Habitations . New Delhi: Concept Publishing Co. Cherulnik, P. D. and S. K. Wilderman (1986) Symbols of status in urban neighborhoods: contemporary perceptions of nineteenth-century Boston, Environment and Behavior 18 (5) September: 604622. Chira, S. (1994) Hispanic families avoid using daycare, study says, New York Times April 6. Choldin, H. M. (1976) Housing standards versus ecological forces: Regulating population density in Bombay, pp. 287231 in Rapoport A. (ed.), The Mutual Interaction of People and their Built Environment . Chicago: Aldine. Chua, B. H. (1988) Adjusting religions practices to different house forms in Singapore, Architecture and Behavior 14 (1): 325. Chudacoff, H. P. (1973) A new look: Residential dispersion and the concept of visibility in a medium-sized city, Journal of American History 60 June: 7791.

161

Cioni, T. (1993a) Of cials make hay of natural lawn, Milwaukee Journal August 19. Cioni, T. (1993b) Unkindest cut of all, Milwaukee Journal August 25. Cooper, M. and M. Rodman (1995) Culture and spatial boundaries: Cooperative and non-pro t housing in Canada, Architecture and Behavior 11 (2): 123138. Cornell, L. L. (1997) House architecture and family form: On the origin of vernacular traditions in early modern Japan, Traditional Dwellings and Settlements Review 8 (2) Spring: 2131. Crossen, C. (1990) A sad pall hangs over the social hub of Nantucket island, The Wall Street Journal August 22. Deitz, P. (1984) A gentle redesign for a 1949 project, New York Times June 21. Despre s, C. (1991) The form, experience and meaning of home in shared housing. Milwaukee: University of Wisconsin-Milwaukee, Department of Architecture PhD Dissertation. Unpublished. Devlin, A. S. (1994) Childrens housing preferences: Regional, socioeconomic, sex and adult comparisons, Environment and Behavior 26 (4) July: 527559. Duncan, J. S. (1981) From container of women to status symbol: The impact of social structure on the meaning of the house, pp. 3659 in Duncan J. S. (ed.), Housing and Identity: Cross-cultural perspectives. London: CroomHelm. Duncan, J. S. (1985) The house as symbol of social structure: Notes on the language of objects among collectivistic groups, pp. 133151 in Altman I. and C. M. Werner (eds), Home Environments (volume 8 of Human Behavior and Environment) . New York: Plenum. Egan, T. (1993) House of different color is shunned, New York Times June 7. El Futuro Maya: Voces del Presente, (1999) Video, shown at the Center for Latin America, University of WisconsinMilwaukee, February 26. Environment and Behavior (1987) Special Issue Home Interiors: a European perspective, 19 (2) March. Erman, T. (1996) Women and the housing environment: The experiences of Turkish migrant women in squatter (Gecekondu) and apartment housing, Environment and Behavior 28 (6) November: 764798. Esber, G. S. (1972) Indian housing for Indians, The Kiva 37 Spring: 141147. Faison, S. (1997a) Condolence calls put rare light on Dengs family, New York Times February 22. Faison, S. (1997b) A Chinese city is a model of good behavior, New York Times March 21. Fishman, R. (1987) Bourgeois Utopias: The Rise and Fall of Suburbia . New York: Basic Books. Franck, K. A. and S. Ahrentzen (eds) (1991) New Households, New Housing . New York: Van Nostrand Reinhold. Frick, D. (ed.) (1986) The Quality of Urban Life (Social, Psychological and Physical Conditions) . Berlin: de Gruyter. Gasparini, A. (1973) In uence of the dwelling on family life, Ekistics 36 (216) November: 344348. Gaster, S. (1991) Urban childrens access to their neighborhood: Changes over three generations, Environment and Behavior 23 (1) January: 7985. Gaunt, L. N. (1991) The family circle: A challenge for planning and design, Journal of Architectural and Planning Research 8 (2) Summer: 147163.
Housing, Theory and Society

162

A. Rapoport
Hood, J. C. (1985) Becoming a Two-job Family . New York: Praeger. Hummon, D. M. (1990) Common places Community Ideology and Identity in American Culture . Albany: SUNY Press. Iovine, J. V. (1999) The last gasp for the American living room, New York Times January 28. Jackson, K. T. (1985) Crabgrass Frontier: The suburbanization of the United States. New York: Oxford University Press. Janz, W. R. (1992) The extension of identity into home fronts: Two Milwaukee Wisconsin neighborhoods , Journal of Architectural and Planning Research 9 (1) Spring: 4863. Johnson, D. (1996) In ux of newcomers part of rural upturn, New York Times September 23. Jopling, C. F. (1974) Aesthetic behavior as an adaptive strategy, presented at the XL1 Congreso Internacional de Americanistas . Mexico D. F. September 27, Mimeo. Kaitilla, S. (1991) The in uence of environmental variables on building material choice: the role of low-cost materials on housing improvement, Architecture and Behavior 7 (3) September: 205221. Kaitilla, S. (1992) The in uence of overseas trade on housing quality among the Saissi of Papua New Guinea, Architecture and Behavior 8 (4): 367378. Kaitilla, S. (1994) Urban residence and housing improvement in a Lae squatter settlement, Papua New Guinea, Environment and Behavior 26 (5) September: 640668. Kamau, L. J. (197879) Semipublic, private and hidden rooms: Symbolic aspects of domestic space in urban Kenya, African Urban Studies 3 Winter: 105115. Kellett, P. (1999) Cultural values and housing behavior in spontaneous settlements, Journal of Architectural and Planning Research, 16 (3) Autumn: 205224. Kent, S. (1984) Analyzing Activity Areas (An Ethnoarchaeological Study of the Use of Space). Albuquerque: University of New Mexico Press. Kent, S. (ed.) (1990) Domestic Architecture and the Use of Space . Cambridge: Cambridge University Press. Kent, S. (1991) Partitioning space: Cross-cultural factors in uencing domestic spatial segmentation, Environment and Behavior 3 (4) July: 438473. Khattab, O. (1993) Environmental quality assessment: An attempt to evaluate government housing projects, Open House International 18 (4): 4147. Kimber, C. (1966) Dooryard gardens of Martinique, Yearbook, Association of Paci c Coast Geographers 28: 97118. Kimber, C. (1971) Interpreting the use of space in dooryard gardens: a Puerto Rican example, Mimeo. Texas A and M University College Station, Texas. Kimber, C. (1973) Spatial patterning in the dooryard gardens of Puerto Rico, Geographical Review , 63 January: 626. King, A. D. (1984) The Bungalow (The Production of a Global Culture). London: Routledge and Kegan Paul. Kleiman, D. (1988) A city where young families middleclass dream fell at, New York Times March 9. Krause, J. (1997) The incredible shrinking living room, Milwaukee Journal Sentinel October 12. Kron, J. (1983) Home-Psych. New York: Potter. Lawrence, D. L. (1988) Suburbanization of house form and gender relations in a rural Portuguese agrotown, Architecture and Behavior 14 (3): 197212.

Gibbons, A. (1992) Rain forest diet: You are what you eat, Science 255 (5041) January 10: 163. Gilbert, S. (1989) Demographics show different housing needs, Honolulu Sunday Star-Bulletin and Advertiser, September 3. Ginsberg, Y. (1975) Jews in a Changing Neighborhood . New York: Macmillan. Goldberg, H. (1987) The changing meaning of ethnic af liation, The Jerusalem Quarterly 44 Fall: 3950. Gonzales, D. (1990) Casitas: Urban oases or illegal shacks?, New York Times September 20. Goode, E. (1999) Study nds TV trims Fiji girls body image and eating habits, New York Times May 20. Gordon, M. M. (1978) Human Nature, Class and Ethnicity . New York: Oxford University Press. Gould, P. and R. White (1974) Mental Maps . Harmondsworth, UK: Penguin Books. Greenbaum, S. D. and P. E. Greenbaum (1981a) The spatial ecology of social networks in four urban neighborhoods , pp. 443 in Osterberg A. E. et al. (eds), Design Research Interactions (EDRA 12). Washington, DC EDRA (Abstract only). Greenbaum P. E and S. D. Greenbaum (1981b) Territorial personalization: group identity and social interaction a Slavic-American neighborhood , Environment and Behavior 13 (5) September: 574589. Grenell, P. (1972) Planning for invisible people: Some consequences of bureaucratic values and practices, pp. 95121 in Turner J. F. C. and R. Fichter (eds), Freedom to Build . New York: Macmillan. Grimaud, V. (1986) Societe hierarchique et habitat en lnde, Architecture and Behavior 2 (3/4): 207227. Hakim, B. S. (1986) Arabic-Islamic Cities: Building and Planning Principles . London: KPI. Hakim, B. S. (1994) The urf and its role in diversifying the architecture of traditional Islamic cities, Journal of Architectural and Planning Research 11 (2) Summer: 108127. Hall, T. (1988) Changing U.S. values, tinged with caution, show up in spending, New York Times October 26. Halle, D. (1984) Americas Working Man (Work, Home and Politics Among Blue-collar Property Owners). Chicago: University of Chicago Press. Hamilton, L. (1984) The evolution of neighborhood identities: Competing images of a Chicago community, Urban Resources 1 (4) Spring: 38. Hassell, M. J. et al. (1993) Gender choice and domestic space: Preferences for kitchens in married households, Journal of Architectural and Planning Research 10 (1) Spring: 122. Hawkes, N. (1998) Rise in women drivers adds to congestion, The Times (London) July 2. Hedges, C. (1994) Promised land taken from Egypt farmers, New York Times September 28. Heppell, M. (ed.) (1979) A Black Reality (Aboriginal Camps and Housing in Remote Australia) . Canberra: Australian Institute of Aboriginal Studies. Hinds, M. dc (1985) Communal living for elderly people, New York Times January 31. Hoffman, G. and J. A. Fishman (1971) Life in the neighborhood , International Journal of Comparative Sociology 12 (2) June: 85100. Holly, L. (1997) Waukesha County building boom leaves other counties in the dust, Milwaukee Journal Sentinel December 14.
Housing, Theory and Society

Theory, culture and housing

Lawrence, R. J. (1986) Le Seuil Franchi [The threshold being crossed]. Geneva: Georg Editeur. Lawson, D. (1993) Take-away Victorian, Japanese style, The Independent (London) September 29. Leigh, K. E. and A. Asojo (1999) A spiritual celebration of cultural heritage, Traditional Dwellings and Settlements Review 10 (2) Spring: 4759. Lisheron, M. (1994) 1000-home subdivision proposed to help preserve Hmong heritage, Milwaukee Journal April 25. Livingston, M. (1992) Bamboo housing and the mangrove of Guayaquil, Traditional Dwellings and Settlements Review 4 (I) Fall: 13 (Abstract only). Los Angeles Times (1983) Proposition M: Minorities helped defeat measure, November 25. Macdonald, R. (1971) The Way Some People Die. New York: Bantam Books. Madigan, R. and M. Munro (1991) Gender, house and home: social meanings and domestic architecture in Britain, Journal of Architectural and Planning Research 8 (2) Summer: 116132. Margulis, H. L. (1992) Asian villages: Downtown sanctuaries, immigrant reception areas, and festival marketplaces, Journal of Architectural Education 45 (3): 150 160. Martin, J. I. (1967) Extended kinship ties: An Adelaide study, Australia and New Zealand Journal of Sociology 3 (1): 4463. Michelson, W. and P. Reed (1970) The Theoretical Status and Operational Usage of Lifestyle in Environmental Research , Research paper No. 36. Toronto: Center for Urban and Community studies, University of Toronto, September. Milbank, D. (1999) Virtual Politics, The New Republic 4, 407: 2227. Milwaukee Journal (1988) English-castle look wins favor in Bangkok, June 17. Milwaukee Journal Sentinel (1988) March 22. Morel, P. and H. Ross (1993) Housing Design Assessment for Bush Communities. Alice Springs: Tangentyere Council. Morgan, S. P. and K. Hiroshima (1983) The persistence of extended family residence in Japan: Anachronism or alternative strategy, American Sociological Review 48 (2) April: 269281. Muschamp, H. (1998) Looking at the lawn and below the surface, New York Times. July 5. Myre, G. (1997) Russia gets a taste of suburban life: USStyle developments increasingly in demand, Milwaukee Journal Sentinel November 20. Nasar, J. L. (1989) Symbolic meanings of house styles, Environment and Behavior 21 (3) May: 235257. Nasar, J. L. (1998) The Evaluative Image of the City. Thousand Oaks: Sage. Nasar, J. L. and J. Kang (1989) Symbolic meanings of buildings style in small suburban of ces, pp. 165172 in Hardie, G. et al. (eds), Changing Paradigms (EDRA 20). Oklahoma City. Nemeth, D. J. (1987) The Architecture of Ideology: NeoConfucian Imprinting on Cheju Island, Korea. Berkley: University of California Publications in Geography, No. 26. Netting, R. McC., R. R. Wilk and E. J. Arnould (1984) (eds) Households: Comparative and Historical Studies of the Domestic Group. Berkeley: University of California Press.

163

Newell, P. B. (1997) A cross-cultural examination of favorite places, Environment and Behavior , 29 (4): 495514. New York Times (1985) Merits of the cultural gambit in the mating game, October 18. New York Times (1986) July 29. New York Times (1987) June 9. New York Times (1997) December 11. Oliver, P. (1987) Dwellings: The House Across the World. Austin: University of Texas Press. Oliver, P. (ed.) (1997) Encyclopedia of Vernacular Architecture of the World (3 Volumes). Cambridge: Cambridge University Press. Olofson, H. (1975) Cultural values, communication and urban image in Hausaland, Urban Anthropology 4 (2) Summer: 145160. Pader, E. J. (1993) Spatiality and social change: Domestic space use in Mexico and the United States, American Ethnologist 20 (1): 114137. Pavlides, E. (1985) Vernacular Architecture in Its Social Context: A Case Study of Eressos, Greece. Philadelphia: University of Pennsylvania PhD Dissertation. Peiser, R. B. and G. M. Schwann (1993) The private value of public open space within subdivisions, Journal of Architectural and Planning Research 10 (2) Summer: 91104. Plafker, T. (1990) The new Chinese dream, Urban Age 6 (4) Spring: 69. Pred, A. (1963) Business thoroughfares as expressions of Negro urban culture, Economic Geography 39 July: 217233. Pruchno, R. A. et al. (1993) Multigenerational households of caregiving families: Negotiating shared space, Environment and Behavior 25 (3) May: 349366. Rabkin, J. (1994) The curious case of Kiryas Joel, Commentary 98 (5) November: 5861. Rapoport, A. (1977) Human Aspects of Urban Form . Oxford: Pergamon Press. Rapoport, A. (1980) Towards a cross-culturally valid de nition of housing, pp. 310316 in Stough, R. R. and A. Wandersman, (eds), Optimizing Environments (Research, Practice and Theory) (EDRA 11). Washington, D.C.: EDRA. Rapoport, A. (1980/81) Neighborhood homogeneity or heterogeneity, Architecture and Behavior 1 (1): 6777. Rapoport, A. (1981) Identity and environment: a crosscultural perspective, pp. 635 in Duncan, J. S. (ed.), Housing and Identity: Cross-Cultural Perspectives. London: Croom-Helm. Rapoport, A. (1983) Development, culture change and supportive design, Habitat International 7 (5/6): 249 268. Rapoport, A. (1985) Thinking about home environments: a conceptual framework, pp. 255286 in Altman, I. and W. C. Werner (eds), Home Environments (Vol. 8 of Human Behavior and Environment) . New York: Plenum. Rapoport, A. (1986) The use and design of open spaces in urban neighborhoods , pp. 159175 in Frick, D. (ed.), The Quality of Urban Life: Social, Psychological and Physical Conditions . Berlin: de Gruyter. Rapoport, A. (1987) Learning about settlements and energy from historical precedents, Ekistics 54 (325327) JulyDecember: 262268. Rapoport, A. (1988) Levels of meaning in the built environment, pp. 317326 in Poyatos, F. (ed.), CrossHousing, Theory and Society

164

A. Rapoport
Rybczynski, W. and V. Bhatt (1986) Understanding slums [sic]: The use of public spaces, pp. 144149 in Quinn, P. and R. Benson (eds), The Spirit of Home (Proceedings of 74th Annual Meeting of ACSA, New Orleans). Washington, DC: ACSA. Sadalla, E. K. and V. L. Sheets (1993) Symbolism in building materials: Self-presentational and cognitive components, Environment and Behavior 25 (2) March: 155180. Salama, R. (1995) User transformation of government housing projects: Case study, Egypt. Montreal: McGill University, M.Arch Thesis (unpublished) March. Sancar, F. H. and T. T. Koop (1995) Proposing a behavioral de nition of the vernacular based on a Comparative analysis of the Behavior Settings in three settlements in Turkey and Greece, Journal of Architectural and Planning Research 12 (2) Summer: 141165. Schwerdtfeger, F. W. (1982) Traditional Housing in African Cities (A Comparative Study of Houses in Zaria, Ibadan and Marrakech). Chichester, UK: Wiley. Sebba, R. (1991) The role of the home environment in cultural transmission, Architecture and Behavior 7 (3): 223241. Shang, S. (1999) Image is more important than experience: A case study of hi-tech home building in response to shifting home identities, pp. 5361 in Mann, T. (ed.), The Power of Imagination (EDRA 30). Edmond: OK: EDRA. Shapere, D. (1977) Scienti c theories and their domains, pp. 518573 in Suppe F. (ed.), The Structure of Scienti c Theories . Urbana, University of Illinois Press (2nd edition). Shokoohy, M. and N. H. Shokoohy (eds) (1994) Kirtipur (An Urban Community in Nepal its People, Town Planning, Architecture and Arts). London: Araxus. Shrestha, U. S. et al. (1997) Social effects of land use changes in Kirtipur, Nepal, Urban Design Studies 3: 51 73. Sontag, D. (1998) Devout community burgeons in Brooklyn, New York Times. January 7. Spalding, S. (1992) The myth of the classic slum: Contradictory perceptions of Boyle Heights Flats 19001991, Journal of Architectural Education 45 (2) February: 107 119. Stahl, P. H. (1991) Maison et groupe domestique e tandu (examples Europe ens), [The house and the extended domestic group (European examples)], ARMOS (Greece) 3: 16671692. Stea, D. (1995) House and home: Identity, dichotomy or dialectic? (with special reference to Mexico), pp. 181 201 in Benjamin, D. N. (eds), The Home: Words, Interpretations, Meanings and Environments . Aldershot: Avebury. Stea, D. and L. Prussin (1978) Society, sex roles and vernacular architecture, paper presented at Congress of Anthropological and Ethnographic Sciences. New Delhi, December, Mimeo. Sternlieb, G. and J. W. Hughes (1986) Demographics and Housing in America, Population Bulletin 41 (1). Washington, D.C: Population Reference Bureau. Stevenson, R. W. (1992) Catering to consumers ethnic needs, New York Times. January 23. Suchar, C. S. and R. Rotenberg (1994) Judging the adequacy of shelter: A case from Lincoln Park, Journal

Cultural Perspectives in Non-Verbal Communication . Toronto: C. J. Hogrefe. Rapoport, A. (1990a) Systems of activities and systems of settings, pp. 920 in Kent, S. (ed.), Domestic Architecture and Use of Space (An Interdisciplinary Crosscultural Study) . Cambridge: Cambridge University Press. Rapoport, A. (1990b) History and Precedent in Environmental Design. New York: Plenum. Rapoport, A. (1990c) The Meaning of the Built Environment . Tucson: University of Arizona Press (revised edition). Rapoport, A. (1990d) Indirect approaches to environmentbehavior research, The National Geographical Journal of India 36 (Part 1 and 2) (March/June): 3046. Rapoport, A. (1994) Sustainability, Meaning and Traditional Environments . Berkeley: Center for Environmental Design Research, University of California-Berkeley, Traditional Dwellings and Settlements Working Paper Series, Vol. 75, IASTE: 7594. Rapoport, A. (1995a) Culture and built forma reconsideration, pp. 399436 in Rapoport, A., Thirty-Three Papers in Environment-Behavior Research . Newcastle (U.K.): Urban International Press. Rapoport, A. (1995b) Flexibility, open-endedness and design, pp. 529562 in Rapoport, A., Thirty-Three Papers in Environment-Behavior Research. Newcastle: Urban International Press. Rapoport, A. (1995c) On diversity and Designing for diversity, pp. 373397 in Rapoport, A., Thirty-Three Papers in Environment-Behavior Research. Newcastle: Urban International Press. Rapoport, A. (1995d) Environmental quality and environmental quality pro les, pp. 471488 in Rapoport, A., Thirty-Three Papers in Environment-Behavior Research. Newcastle: Urban International Press. Rapoport, A. (1997a) Theory in environment-behavior studies: Transcending times, settings and groups, pp. 399421 in Wapner, S. et al. (eds), Handbook of JapanU.S. Environment-Behavior Research (Toward a Transactional Approach) . New York: Plenum. Rapoport, A. (1997b) The nature and role of neighborhoods, Urban Design Studies 3: 93118. Rapoport, A. (1998) Using culture in housing design, Housing and Society 25 (1 and 2): 120. Rapoport, A. (1999) On the relationship between family and housing, pp. 136 in Awotona, A. (ed.), Housing Provision and Bottom-up Approaches (Family Case Studies from Africa, Asia and South America. Aldershot: Ashgate. Rapoport, A. (2000) Science, explanatory theory and environment-behavior studies, pp. 107140 in Wapner, S. et al. (eds), Theoretical Perspectives in EnvironmentBehavior Research: Underlying Assumptions, Research Problems and Methodologies . New York: Kluwer Academic/Plenum Publishing. Rendell, R. (1995) Simisola . New York: Crown Publishers. Ross, H. (1987) Just for Living (Aboriginal Perceptions of Housing in Northwest Australia) . Canberra: Aboriginal Studies Press. Ross, H. (1991) Household composition and use of space, pp. 6472 in Cook, P. (ed.), Social Structures for Sustainability , Fundamental Questions Paper No. 11. Canberra: Australian National University, CRES. Rosser, C. and C. Harris (1965) The Family and Social Change: A Study of Family and Kinship in a South Wales Town . London: Routledge and Kegan Paul.
Housing, Theory and Society

Theory, culture and housing

of Architectural and Planning Research , 11 (2) Summer: 149165. Sun Rhodes, D. (1993) My home: A communal American Indian home, Native Peoples . Spring: 4043. Suttles, G. D. (1968) The Social Order of the Slum . Chicago: Chicago University Press. Teltsch, K. (1993) Stable neighborhoods help aging New Yorkers, New York Times. November 6. Urban Age (1999) 6 (4) Spring: 11. Van der Voordt, T. (1990) Building adaptable housing from theory to practice, current developments in the Netherlands, Architecture and Behavior 6 (1): 1737. Vayda, A. P. (1983) Progressive Contextualization, Human Ecology 11 (3) September: 265282. Weiss, M. J. (1988) The Clustering of America. New York: Harper and Row. West, D. (1998) Mass-market mansions for the suburbs, New York Times. March 18. Wheatley, P. (1976) Levels of space awareness in the traditional Islamic city, Ekistics, 42 (253) December: 354366. Wheeler, L. (1977) Behavioral and social aspects of the Santa Cruz riverside project, Man Environment Systems 7 (4): 203205. White, V. (1988) House divides New Hampshire town, New York Times. September 26. Wiechman, L. (1997) More than just a neighborhood: New homeowners are demanding big-ticket amenities from developers, Milwaukee Journal Sentinel .

165

Wilk, R. R. (1989) (ed.) The Household Economy: Reconsidering the Domestic Mode of Consumption . Boulder, Colorado: Westview Press. Wilk, R.W. and W. Ashmore (eds) (1988) Household and Community in the Meso-American Past, Albuquerque: University of New Mexico Press. Williams, F. (1997) A house, 10 wives: Polygamy in suburbia, New York Times. December 11. Wilmott, P. (1963) The Evolution of a Community . London: Routledge and Kegan Paul. Winnipeg Free Press (1992) Close-up (Special supplement for rst time house buyers), September 20. Vrana, D. (1994) Exporting Irvine: Orange Countys architects of success bringing Western-style comforts to Indonesian communities, Los Angles Times. July 26. Yans-McLaughlin, V. (1977) The Italian family, New York Times. July 6. Yellen, J. (1985) Bushmen, Science 85. May: 4148. Young, M. and P. Wilmott (1962) Family and Kinship in East London . Harmondsworth, UK: Penguin. Zavis, A. (1998) Little boxes made of ticky-tacky: Subdivision sameness prompts anti-monotony laws, Milwaukee Journal Sentinel . November 29. Zeisel, J. (1973) Symbolic meaning of space and the physical dimension of social relations, pp. 252263 in Walton, J. and D. E. Carns (eds), Cities in Change: Studies on the Urban Condition . Boston: Allyn and Bacon.

Housing, Theory and Society