Author's note.Martha Cooper took all of the photographs except numbers 8 and 14. Numbers 8 and 14 are by the author. I wish to thank Hyman Korman and Frank Romo for assistance with some of the statistical analysis.

1. Recently, a number of writers have stressed, and critically examined, the role that the public museum has played, since its relatively recent origin, in deciding what will count as art. See, for example, Foucault, Michel, “Fantasia of the Library,” in Language, Counter-Memory, Practise, trans. Bouchard, D. F. and Simon, S. (Ithaca: Cornell University Press, 1977), pp. 87–109Google Scholar; Lebensztejn, Jean-Claude, Zigzag (Paris: Flammarion, 1981)Google Scholar; Krauss, Rosalind, “Photography's Discursive Spaces: Landscape/View,” Art Journal 42 (Winter 1982): 311–319CrossRefGoogle Scholar; DiMaggio, Paul, “Cultural Entrepeneurship in Nineteenth-century Boston: The Classification and Framing of American Art,” Media, Culture and Society 4 (1982): 303–22CrossRefGoogle Scholar; the various essays in Stocking, George, ed., Objects and Others (Madison: University of Wisconsin Press, 1985)Google Scholar; and Clifford, James, The Predicament of Culture (Cambridge: Harvard University Press, 1988), pp. 189–251Google Scholar; Jonaitis, Aldona, “Franz Boas, John Swanton, and the New Haida Sculpture at the American Museum of Natural History,” Institute for Social Analysis at S.U.N.Y. Stony Brook, Paper Series (1989).Google Scholar

2. On the tendency today for works to find their way into a museum only after a long period in the private house, see Simpson, Charles, Soho: The Artist in the City (Chicago: University of Chicago Press, 1981), ch. 2.Google Scholar

3. Although there are few systematic studies of the subject, there are numerous suggestions of the importance of the house in understanding art and cultural items. For example, in a study of the emergence of impressionism, Harrison and Cynthia White pointed to the fact that most paintings in France in the 1830s and 1840s were now being purchased for the homes of the bourgeoisie, rather than the aristocracy. This affected both the form of the art (huge ceiling paintings were too large for most bougeois homes) and its content (panoramic battle scenes were unsuitable for bourgeois homes; genre paintings and landscapes were more suitable). See Harrison, and White, Cynthia, Canvases and Careers: Institutional Change in the French Painting World (New York: John Wiley, 1965).Google Scholar

The link between 20th-century art and the house is, in fact, implicit in many studies. For example, a preface to a biography of Piet Mondrian comments: “A Mondrian painting hung in a house and room designed entirely in the spirit of Mondrian … is a most sublime expression of a spiritual idea.…” See Schmidt, George, “Piet Mondrian Today,” preface to Michel Seuphor, Piet Mondrian: Life and Work (New York: Harry Abrams, 1957) pp. 11–12.Google Scholar

Picasso, in a famous outburst in 1945, protested: “… painting is not done to decorate apartments. It is an instrument of war for attack and defence against the enemy.” But this protest implies what Picasso knew — that regardless of the artist's intention, most of his paintings and those of his artist friends are purchased to hang in homes. Anyway, that is where Picasso too hung his prized works, in his apartment and in several houses. As one account put it: “When a residence filled up with paintings, sculptures, furniture … Picasso left and bought a new one” (Art News [12 1986]: 85)Google Scholar. (Picasso's outburst is recorded in an interview conducted by Simone Tery published in Lettres Frangaises, 03, 1945.)Google Scholar

4. There are a few well-known “living-room” studies by sociologists. Such studies make an inventory of items in the living room — furniture, drapes, type of floor, ornaments, and so on — and then attempt to predict a person's social class from these items. But these studies seldom ask people about the meaning of their art and cultural items, nor do they analyze the context of these items. For example, they rarely consider a picture's subject matter or ask what it means to people. Further, they confine the study to the living room. For these studies, see Chapin, F. Stuart, Contemporary American Institutions: A Sociological Analysis (New York: Harper and Bros., 1935), ch. 19Google Scholar; and Laumann, Edward O. and House, James S., “Living Room Styles and Social Attributes: The Patterning of Material Artefacts in a Modern Urban Community,” Sociology and Social Research 54 (1970):321–42.Google Scholar

There is an excellent study of the responses of a sample of residents in a Chicago suburb to the question “What are the things in your home which are special to you?” The study discusses the meaning that people attach to items such as furniture, visual art, photographs, books, and stereo sets. See Csikszentmihalyi, Mihaly and Rochberg-Halton, Eugene, The Meaning of Things (Cambridge: Cambridge University Press, 1981).CrossRefGoogle Scholar

5. Gans, Herbert, “American Popular Culture and High Culture in a Changing Class Structure,” Prospects 10 (1986): 17–38.CrossRefGoogle Scholar

6. For Bourdieu, see La Distinction: Critique Sociale de jugement (Paris: Les Editions de Minuit, 1979)Google Scholar, translated as Distinction: A Social Critique of the Judgement of Taste (Cambridge: Harvard University Press, 1984)Google Scholar; and “Outline of a Sociological Theory of Art Perception,” International Social Science Journal 20, no. 4 (1968)Google Scholar. For DiMaggio, see (with Michael Useem) “Social Class and Arts Consumption: The Origins and Consequences of Class Differences in Exposure to the Arts in America,” Theory and Society 5, no. 2 (03 1978): 141–61Google Scholar; (with Michael Useem) “The Arts in Class Reproduction,” in Apple, Michael, ed., Cultural and Economic Reproduction in Education (Boston: Routledge and Kegan Paul, 1981)Google Scholar; and “Classification in Art,” American Sociological Review 52 (08 1987): 440–55.Google Scholar

Perhaps the main difference between Bourdieu and DiMaggio is that Bourdieu holds that separate tastes are associated with each social class. DiMaggio, at least in his recent work, holds that the taste for popular culture is common among all social classes, but that only members of the higher social classes also have a taste for high culture.

7. Bourdieu, , Distinction, pp. 2–3.Google Scholar

8. See Kandinsky, , Concerning the Spiritual in Art, trans. Sadler, Michael (New York: Margin Press, 1972)Google Scholar, first published as Uber das Geistige in der Kunst (1912)Google Scholar; y Gasset, Ortega, The Dehumanization of Art (1925)Google Scholar, reprinted in Velasquez, Goya and the Dehumanization of Art, trans. Brown, Alexis (New York: Norton, 1972)Google Scholar; Greenberg, Clement, “Avant-Garde and Kitsch” (1939)Google Scholar, reprinted in Art and Culture: Critical Essays (Boston: Beacon Press, 1961)Google Scholar; Steinberg, Leo, “Contemporary Art and its Public,” reprinted in Other Criteria (New York: Oxford University Press, 1972).Google Scholar

9. For the figures cited see Ford Foundation, The Finances of the Performing Arts: A Survey of the Characteristics and Attitudes of Audiences for Theater, Opera, Symphony, and Ballet in 12 U.S. Cities, 2 vols. (New York: Ford Foundation, 1974)Google Scholar. The same pattern emerges from other surveys of involvement in culture and high culture in the United States. Involvement with high culture does vary with socioeconomic level, above all with level of education. Those with college degrees are more likely than those without to have some interest in high culture. Yet only a minority of even the college-educated population spend much of their leisure time in these ways. For example, a study of men in a variety of occupations in Detroit found that, though only 1 percent of blue-collar workers read a “quality” newspaper every day, only 11 percent of engineers did so either. Even among professors and lawyers, a highly educated group, only a minority read a “quality” newspaper (42 percent of all professors and 36 percent of all lawyers). See Wilensky, Harold, “Mass Society and Mass Culture,” American Sociological Review 29 (04 1964): 173–97CrossRefGoogle Scholar. See also Robinson, John, How Americans Use Time (New York: Praeger, 1979), p. 159.Google Scholar

10. See Kanter, Rosabeth Moss, Men and Women of the Corporation (New York: Basic Books, 1979).Google Scholar

11. The response rates for each area are as follows: Greenpoint - 60 percent; Manhattan - 53 percent; Manhasset - 67 percent. In Greenpoint, the response in the predominantly Polish section was 74 percent, while that in the predominantly Italian section was 39 percent.

12. Among employed male heads of household in the Greenpoint sample, 54 percent had blue-collar occupations (“blue-collar” here includes doormen and first-line supervisors), 23 percent had lower-white-collar occupations (“lowerwhite-collar” is defined as clerical, secretarial, or retail sales jobs), 15 percent had upper-white-collar occupations (“upper-white-collar” is defined as managerial or professional occupations), and 8 percent owned small businesses (defined as employing five persons or less). Among female heads of household (here a female spouse, like her husband, is considered a head of household), 33 percent had lower-white-collar occupations, 22 percent had blue-collar occupations, and 44 percent were housewives (or homemakers).

13. Among male heads of household in the Manhattan sample who were in the labor force, 47 percent had upper-white-collar occupations, 18 percent were capitalists, and 35 percent were owners of small businesses. Among female heads of household in Manhattan 67 percent had upper-white-collar occupations, 11 percent had small businesses, and 22 percent were housewives (or homemakers).

Among the heads of household in the Manhasset sample who were in the labor force, 37 percent had upper-white-collar occupations, 25 percent were capitalists, 25 percent own small businesses, and 12 percent were blue-collar workers. Among female heads of household in Manhasset, 57 percent were homemakers, 29 percent had upper-white-collar occupations, and 14 percent had lower-white-collar occupations.

14. On the connection between country villas and interest in landscape pictures, see Turner, A. Richard, The Vison of Landscape in Renaissance Italy (Princeton: Princeton University Press, 1966), ch. 10Google Scholar. For a detailed account of the growth of the suburban villa among wealthy Romans from 1420 to 1585, see Coffin, David, The Villa in the Life of Renaissance Rome (Princeton: Princeton University Press, 1979)Google Scholar. On the emergence of landscape painting as a distinct genre, see Gombrich, E. H., “The Renaissance Theory of Art and the Rise of Landscape,” in Norm and Form in the Art of the Renaissance (London: Phaidon Press, 1966)Google Scholar, and Friedlander, M. J., Landscape, Portrait, Still-Life: Their Origin and Development, trans. Hull, R. C. (New York: Philosophical Library, 1949).Google Scholar

15. On this 19th-century suburbanization, see Warner, Sam Bass, Streetcar Suburbs (Cambridge: Harvard University Press, 1962)Google Scholar; and Jackson, Kenneth, Crabgrass Frontier: The Suburbanization of American (New York: Oxford University Press, 1985).Google Scholar

16. On the growing popularity of landscape paintings in 19th-century America, see Novak, Barbara, Nature and Culture: American Landscape and Painting, 1825–1875 (New York: Oxford University Press, 1980).Google Scholar

17. On blue-collar works in post-World War II suburbs, see Gans, Herbert, The Levittowners (New York: Vintage, 1967)Google Scholar; Berger, Bennett, Working-Class Suburb (Berkeley: University of California Press, 1968)Google Scholar; and Halle, David, America's Working Man (Chicago: University of Chicago Press, 1984).Google Scholar

18. Gombrich, Ernest, “Tradition and Expression in Western Still Life,” in Meditations on a Hobby Horse, and Other Essays on the Theory of Art, 4th ed., (London: Phaidon, 1985).Google Scholar

19. See Venturi, Robert, Brown, Denise Scott, and Izenour, Steven, Learning from Las Vegas: The Forgotten Symbolism of Architectural Form (Cambridge, Mass.: M.I.T. Press, 1980).Google Scholar

20. J. B. Jackson traces the rise of tourism in the West back four centuries. (He defines tourism as the urge, for nonreligious and noncommercial reasons, to see famous monuments and to explore remote regions and observe strange customs.) See “Learning About Landscape” in The Necessity for Ruins (Amherst: University of Massachusetts Press, 1980).Google Scholar

21. On the connection between religious beliefs and landscape in 19th century America, see Novak, , Nature and Culture.Google Scholar

22. Turner, , The Vision of Landscape in Renaissance Italy, ch. 10.Google Scholar

23. Figure 10 is based on the statistical technique of logistic regression, using maximum likelihood estimates. The probabilities in the chart were calculated from the results of a single-equation logit model. In the model the log odds of a landscape picture being peopled were considered as a function of the type of society depicted in the landscape (a dichotomous variable), and the type of neighborhood of the house where the landscape is hung (a trichotomous variable). The parameter estimates for the type of society depicted were highly significant (at the 001 level). By contrast, the parameter estimates for the type of neighborhood of the house where the landscape is hung were not significant at even the 05 level. In other words, whether a landscape picture is peopled is affected by the type of society that the landscape depicts, but not by the social class of the owners of the picture.

On a more technical level, the parameter estimates were actually the mean of thirty separate estimates. For each estimate the computer was programmed to choose at random one of the prominently displayed pictures from each house, and the logistic regression was performed on this subset of landscape pictures. This procedure was repeated 30 times, with a new random selection each time. The coefficients of the final equation were thus the mean of the coefficients calculated on each of the thirty separate regressions. The reason for taking the mean of many separate estimates was that, since the number of landscape pictures prominently displayed varies from house to house, a statistical analysis could not be based simply on the total number of landscape pictures prominently displayed. Also, the coefficients were coded on an effect-coded design matrix. For more on these techniques, see Swafford, Michael, “Three Parametric Techniques for Contingency Table Analysis: A Nontechnical Commentary,” American Sociological Review 45 (08 1980): 664–90.CrossRefGoogle Scholar

24. See Adhemar, M. Jean, “Les Lithographes de Paysage en France à L'Epoque Romantique,” in Archives de I'Art Français: Nouvelle Piriode (Paris: Armand Colin, 1938), vol. 19, pp. 230–32Google Scholar. Victor Adam was the best known of the artists who specialized in adding the figures to landscapes.

25. Schapiro, Meyer, Paul Cezanne, 2nd ed. (New York: Harry Abrams, 1962), p. 14.Google Scholar

26. Monet's haystacks, for example, enthused young Kandinsky in 1889. See Kandinsky, Wassily, “Reminiscences,”Google Scholar in Herbert, Robert, Modern Artists on Art (1913; rept. Englewood Cliffs, N.J.: Prentice-Hall, 1964).Google Scholar

27. Rewald, John, Paul Gauguin (New York: Harry Abrams, 1954).Google Scholar

28. See Halle, David, “The Family Photograph,” Art Journal 46 (Fall 1987): 217–25.CrossRefGoogle Scholar

29. For a study that argues that the difference between high and popular culture has been exaggerated, see Gans, Herbert, Popular Culture and High Culture: An Analysis and Evaluation of Taste (New York: Basic, 1974).Google Scholar

30. On the importance, for individual advancement, of appearing committed to the goals of the corporation, see Kanter, , Men and Women of the Corporation.Google Scholar

31. I deal with all of these issues in a book currently under contract with the University of Chicago Press.