The Subcultural Theory of Urbanism: A Twentieth-Year Assessment

by Claude S. Fischer
Citation
Title:
The Subcultural Theory of Urbanism: A Twentieth-Year Assessment
Author:
Claude S. Fischer
Year: 
1995
Publication: 
The American Journal of Sociology
Volume: 
101
Issue: 
3
Start Page: 
543
End Page: 
577
Publisher: 
Language: 
English
URL: 
Select license: 
Select License
DOI: 
PMID: 
ISSN: 
Abstract:

CENTENNIAL ESSAY The Subcultural Theory of Urbanism: A Twentieth-Year Assessment1

Claude S. Fischer

University of California, Berkeley

Over the last 20 years, some urban sociologists have placed Fi- scher's 1975 article, "Subcultural Theory of Urbanism," on equal footing with Wirth's 1938 classic, "Urbanism as a Way of Life," as an explanation of urban-rural differences. But ambiguities in Fischer's subcultural theory require clarification before its validity can be thoroughly judged. Ironically, a few of these conceptual problems may have led to underestimating the theory's plausibility. Recent empirical research is typically consistent with subcultural theory but includes important failures and significant gaps. Clearer theoretical specification should enable researchers to better test sub- cultural theory and decide among different explanations of urban- ism's social implications.

Almost 20 years after Louis Wirth (1938)published his seminal article, "Urbanism as a Way of Life," in this Journal he complained that sociolo- gists had accepted his theory as true instead of testing it (Wirth [I9561 1969). Twenty years ago, I presented in this Journal "Toward a Subcul- tural Theory of Urbanism" (Fischer 1975a).' Since then, some sociolo- gists have accepted that model as being equally valid as or more valid than Wirth's3 Although "Urbanism as a Way of Life" has often been di~sected,~

subcultural theory has not. This article assesses first the theo- retical and, then, the empirical status of subcultural theory after 20 years.

'This work was supported in part by the Committee on Research, University of California, Berkeley. Shana Cohen provided some research assistance. Paul Burstein, Ann Swidler, and several AJS reviewers contributed important comments. Address correspondence to Claude S. Fischer, Department of Sociology, University of Califor- nia, 410 Barrows Hall, Berkeley, California 94720-1980.

Although I speak here in first person, this paper was submitted for review in the third-person voice.

See, e.g., LaGory and Pipkin (1981, pp. 39-40), Choldin (1985, pp. 50-53), Spates and Macionis (1987, pp. 295-97), Schwab (1992, pp. 19-20, 355-62), and Flanagan (1990, pp. 73-76).

See, e.g., Dewey (1960), Gans (1962), P. Hauser (1965), Morris (1968), Castells ([I9681 1976), Fischer (1972), Salerno (1988), and Miller (1992).

O 1995 by The University of Chicago. All rights reserved.

0002-9602196110103-0001$01.50

AJS Volume 101 Number 3 (November 1995): 543-77 543

For practical reasons, I do not cover all of the relevant research but focus on select, recent studies most germane to the theory. I build upon my earlier summary of the literature (Fischer 1976, 1984), while noting book reviewers' cautions that I had "stretche[d] available data extremely far" in The Urban Experience to support the theory (Michelson 1977, p. 805; Ward 1978). Before appraising the empirical status of subcultural theory, several conceptual problems must be addressed. For example, there is confusion about the object of study, urbanism, about the place of external factors in the model, and about the causal connections be- tween cities and concentrations of particular groups. These problems limit the conclusiveness of the existing research. And, while much of the research on how urbanism affects subcultural diversity and unconven- tionality is consistent with subcultural theory, some is not and, more important, key empirical pieces are missing.

SUBCULTURAL THEORY SUMMARIZED

Subcultural theory (Fischer 1975a, 1982, 1984) contends, as many analy- ses have, that city life is relatively unconventional-a term that covers a range of behavior from artistic innovation, to expressions of dissenting values, to serious criminality, but all deviate from societal norms (Fischer 1975a, pp. 132 1-24; see esp. n. 5 for an explanation of the term "uncon- ventional"). Wirth (1938) explains this association by "social breakdown": city life undermines traditional communities, which in turn frees, isolates, and alienates people and ultimately produces individuation and normlessness. But evidence of urban social breakdown, other than the unconventionality itself, has been weak. Others, whom I call composi- tionalists (esp. Gans 1962; see also Reiss 1955), explain the correlation between urban life and unconventionality as largely incidental: for vari- ous reasons, particular types of people-ethnic minorities, the artistic avant-garde, professionals, and so on-come to live in cities and, conse- quently, their lifestyles then typify cities. City lifeways differ from life- ways in smaller places because, and only because, their residents differ.

According to subcultural theory, cities do stimulate unconventionality, but not through breakdown. Population concentration generates a variety of subcultures, or "a set of interconnected social networks . . . and the . . . norms and habits common [to it] . . . ; it is loosely synonymous with 'social world'" (Fischer 1984, p. 297, n. 26). More explicitly, a subcul- ture is a large set of people who share a defining trait, associate with one another, are members of institutions associated with their defining trait, adhere to a distinct set of values, share a set of cultural tools (Swidler 1986), and take part in a common way of life. These attributes are mat- ters of degree, and subcultures' boundaries may be vague and overlap-

ping (see Fischer 1982, pp. 194-96). Because many of these emergent

subcultures are unconventional, city life is relatively unconventional. Put

more fully:

PROPOSITION1.-Larger places develop more and more specialized sub- cultures than do less populous ones, and are therefore more culturally heterogeneous, because of two processes: (a) larger places attract migrants from a wider hinterland, each bringing along their cultures; and (b) they generate greater diversity through economic, spatial, institutional, and cultural specialization. (This second process draws on Durkheim's [(1893) 19331 model of "dynamic density.") So, big cities are likelier than small towns to have more and more differentiated subcultures organized around ethnic, occupational, leisure, and other defining traits.

PROPOSITION populous places develop not only more distinct

2.-More subcultures but also more intense subcultures than less populous places do. ("Intensity" refers to how institutionalized, solidary, and socially controlling subcultures are.) This is so for two reasons. (a) Larger places tend to have larger subcultures, which in turn more easily sustain institu- tions, resist outside influences, and envelop members' social networks than do small ones. Size is especially important for narrowly defined populations, like specialized-taste publics. These populations, although they may be identified in small places, typically reach the critical mass needed for a subculture only in larger places. (b) Because cities are rela- tively diverse, their residents more often encounter members of different subcultures than do rural residents. Such encounters often lead to tension and conflict, which in turn reinforce group boundaries. This second pro- cess is especially likely to affect broader subcultures.'

PROPOSITION3.-At the same time, between-group contact leads to mutual influence. When larger groups influence smaller ones (e.g., Anglo family values spread to immigrant Vietnamese youths), unconventional- ity declines. When smaller subcultures affect larger ones (e.g., bohemian ideas spread to the middle class), unconventionality grows-at least until the offbeat itself becomes the norm. Diffusion from more unusual to more typical groups is likelier the larger the atypical subculture and is therefore likelier in urban places. Moreover, subcultural fusions also become urban innovations, or unconventionalities. The net result from these contrary processes of intergroup repulsion and of intergroup borrowing, subcul- tural assimilation versus articulation, depends on a variety of contingen- cies (see Fischer 1975a). Nevertheless, in net:

PROPOSITION4.-The more urban the place, the higher the rates of

'This is so because members of larger subcultures have many fewer encounters with distinctive others outside urban places (while members of smaller subcultures more commonly meet members of large groups almost everywhere).

unconventionality relative to the wider society, because (a) larger places generate more diverse and more specialized subcultures; (b) critical mass and intergroup friction are likelier in larger places, which produces more intense subcultures, especially for atypical populations; and (c) the greater intergroup contact in larger places diffuses cultural elements of atypical subcultures to others in the area. Empirically, then, we should observe that residents of larger towns vary more in social behavior than do those in smaller towns and also observe that the distribution of social behavior in larger towns is more often skewed in the atypical, unconventional direction.

I have qualified the argument in a few ways, for example, acknowledg- ing that at each step in the causal chain other influences intrude, so that the net observed correlation between urbanism and unconventionality is typically modest, and stipulating various circumstances that condition the process. Subcultural theory does not preclude additional explanations of how urban life could cause unconventionality (e.g., compositional ones), except perhaps those theories contending that urbanism weakens group solidarity. In sum, urbanism is correlated with unconventionality, in part because it stimulates development of subcultures.

THEORETICAL AND CONCEPTUAL PROBLEMS Key conceptual problems beset the theory and its applications. There are ambiguities in the details. For example, how does a researcher bound one subculture from another or even identify when one comes into exis- tence? How large must a subculture be to attain critical mass? (See Gol- denberg 1987, pp. 169-82.) When does intergroup contact lead to greater contrast and when to assimilation? And how does the theory handle suburbia? But space limitations require that I focus here on the funda- mental theoretical problems.

What Is the Object of Study? Some have challenged the theory's premise that towns and cities are distinctive, meaningful objects distributed along a rural-urban contin- uum (e.g., Sauders 1985; Gottdiener and Feagin 1988; earlier statements include Castells [I9761 and Mellor [1977]). Urban places are but depen- dent parts of a larger structure, they argue. Perhaps city and country once were separable units, but any residual differences today result from their functional locations in the economic structure, not from places' internal properties such as population size. I have answered (Fischer 1978a; 1978b; 1980; 1982, pp. 255-57; 1984) by arguing that popular opinion still recognizes rural-urban contrasts and that research still shows behavioral differences between urban and rural people. These responses are not sufficient. Popular opinion can be woefully mistaken; behavioral differences might be spurious. (Are residents of major cities especially liberal, e.g., because those places are large or because their residents are more often college-educated, administrative workers?) This issue proba- bly can'not be resolved on the basis of first principles, but in research practice. How useful it is to classify places by size depends on the theoret- ical question at hand. Defenders of the rural-urban construct can point to ways in which urban-nonurban differences continue to make a difference, economic traits notwithstanding. A key instance is the correlation today between community size and crime. In the long run, we will need to see empirically whether internal traits, such as size or social composition, or relational traits, such as location in a functional hierarchy, do a better job of explaining place differences.

Variable Place A Place B
1. Population .............................................. 1,000 5,000
2. Proportion who are potentially members of    
subculture X (e.g., ethnicity) ....................... .10 .10
3. Number who are potentially members of    
subculture X-(1) X (2) ............................. 100 500
4. Residents active in subculture X .................. 67 200
5. Number of formal institutions of X .............. 1 3
6. Solidarity, power, influence of X ................. Low High

What Are the Units of Analysis?

What is the proper level of analysis? Subcultural theory is about places, but much of the empirical research-my own as much as anyone else's-is about individuals, how people differ by size of place. This slip is understandable. The competing theories invoke individual-level processes, such as "psychological overload" for Wirth (1938) and Simmel ([I9051 1969; see Fischer 1975b) and cultural discomfort in subcultural theory. Also, researchers often measure attributes of places by aggregat- ing attributes of residents-for example, by measuring local unconven- tionality as residents' mean responses to survey questions. Still, subcul- tural theory is, at core, an ecological theory, not a theory about persons. The translation between levels is neither simple nor to be taken for granted.

Consider the comparison, presented in table 1, between two hypotheti- cal places with arbitrary but plausible attributes. In this illustration, the larger the city and the larger the group, the less the individuals' involve- ment in subculture X. The probability of individual participation in larger place B is 2001500 = .40 and in smaller place A is 671100 = .67 (line 4). At the ecological level, however, the larger the city and the group, the greater subculture X's institutionalization and solidarity (B > A; lines 5 and 6). To the question, Does size lead to subcultural intensifi- cation? the answer is another question: At what level of analysis? Ecologi- cally, size intensifies subculture X; individually, size reduces likelihood of involvement.

Certain implications follow. First, researchers conveniently use indi- vidual-level data to characterize places-for example, urbanism corre- lates with liberalism to the extent that urban residents express more liberal opinions than do rural residents. But, in using these data, re- searchers underestimate emergent processes. For example, while one town may have only a modestly higher proportion of liberals than the town down the road, that small edge can translate-through subcultural intensification-into a dominating climate of opinion. (Put another way, the association between individual traits and collective effects is probably not linear.)6 Second, researchers therefore probably underestimate differ- ences among places when they simply correlate the size of individuals' towns with other individual attributes. Ecological correlations usually exceed individual-level ones. Third, ideal measures should be ecological rather than simply summated individual assessments. Finally, even eco- logical analyses leave open the issue of which ecological level is appro- priate-neighborhood, municipality, or region.

The level of analysis problem is not particular to urban theory. How to deal with contextual effects and the "micro-macro link" is a problem for much of sociology (see, e.g., Hauser 1970; Blau 1981; Blalock 1984; Lieberson 1985; Ragin and Becker 1992). Subcultural theory specifies certain cross-level processes that are open to empirical test, for example, the association between place size and personal networks, but its ultimate tests require place-level analyses and, ideally, emergent, place-level mea- sures. For instance, a subculture's intensity might be indicated by the presence and nature of its local institutions, businesses, public displays, collective events, and so forth. The values typical of a local commu- nity-to answer the question, for instance, of whether urban communi- ties more commonly tolerate unconventionality-might be measured by local laws, police practices, political controversies, and the like. Do

Even a statistical majority need not determine a community trait. A powerful minor- ity may shape the community culture. Mere summation of individual data may incor- porate a "democratic fallacy."

smaller communities more often enforce marijuana laws or more often witness the success of politicians who run on "Christian" platforms? And, if so, do such differences persist when comparing places of differing size that are compositionally similar? These sorts of empirical approaches may not finally resolve levels-of-analysis conundrums, but they better fit the theoretical claims of subcultural theory.

To argue that subcultural theory is about places rather than people is not to dismiss the research on individuals. We can recognize the problem without endorsing Lieberson's extreme conclusion that "application of lower-level data to higher-level issues can make no contribution to knowl- edge" (Lieberson 1985, p. 110; emphasis in original). Individual-level evidence, such as survey respondents' answers, is useful; more often than not a translation between levels can be made. But such translations should be cautiously interpreted and ecological-level analyses ought to be preferred.

Mediating Processes: A Theory of Access?

The correlation between city size and subcultural processes is, even as posited in the original theory, modest. I have noted (Fischer 1975a, pp. 1325-26, n. 7) the example of the college town which, despite its small size, nonetheless hosts a student subculture. The problem of modest cor- relations between size and subcultures, however, goes beyond exceptions to a rule.

For the most part, the theory predicts subcultural emergence and inten- sification where particular people amass. When distinctive populations concentrate outside the urban centers-for example, when Sikh immi- grants settle in the California hinterland, or when "hippies" settle in the mountains-then the causal chain of

Urbanism +Group Concentration +Subcultures; Subcultural Intensity

is broken at the first arrow. Thus, subcultural theory seems really to be a theory of group concentration. The first arrow is also weakened by the expansion of dense central cities into multinucleated metropolitan regions (Berry 1976; Frey 1987). Distinctive subcultures can be found throughout these urbanized areas. Finally, as modern technologies allow interaction without proximity (perhaps even "community without propinquity" [Webber 19701-more on this later), subcultural processes are revealed to be fundamentally about intragroup accessibility. Spatial agglomeration is only one way group members gain access to one another. In the end, subcultural theory is largely about the ability of subculture members to communicate, to create "moral density" (Durkheim 1933), and it is not necessarily about cities per se. (This critique does not apply as well to the other dynamic in the theory, the one involving intergroup contact- proposition 2b, above. Cities remain distinctive in the number and vari- ety of groups that reside therein, and therefore cities are distinctive in their intergroup encounters. But, even this process could be assimilated within a model of cross-group communication.)

Empirically, one implication of this conclusion is that, in our still- spatial world, larger places will be more unconventional than smaller ones only to the degree that group members have greater access to one another in larger than in smaller places. Where and when spatial concen- tration is less necessary for accessibility-perhaps mid-2lst-century America-then the model would fit less well. More broadly, this is a subcultural theory of urbanism only to the degree that urbanism deter- mines particular population concentrations or, still more fundamentally, groups' internal accessibility. Extrapolating further, a subcultural theory of accessibility might suggest principles that govern the formation and functioning of subcultures across space. For example, specialized and unconventional aspatial subcultures should emerge for those populations with access to modern means of communication and transportation. The subcultural theory of urbanism is, by this analysis, a special instance of a subcultural theory of communication.

Historical and Cultural Contexts

The Chicago school's portrayal of urbanism has been described as pecu- liar to 19th- and 20th-century American society. I, however, mean sub- cultural theory to be widely applicable across societies and eras. That claim incurs additional conceptual and empirical burdens.

For one, unconventionality becomes a relational construct. Wirth could have claimed that America's urban centers were notable for, say, sexual licentiousness, as they seemed to be in the 1920s. But subcul- tural theory implies that, where a society is sexually liberal, its most met- ropolitan places will be instead relatively puritanical (because urban- ism promotes unconventionality). What is unconventional depends on the particular context. If so, systematic research in non-American, non- contemporary contexts is necessary.' But the research currently available is overwhelmingly from late 20th-century America, a distinctive society.

To some observers, America today is particularly distinguished by space-erasing technologies that may permit spatially "liberated commu-

'Yet another difficulty arises because structural correlates of urbanism may vary across societies and eras. In one, e.g., poverty may be especially rural and in another not. When one then proceeds to simulate ceteris paribus conditions statistically, what is to be controlled will vary in different contexts.

nities" (Wellman 1979; Webber 1963, 1970; Abler and Falk 1975). Wil- son (19856) suggests, for example, that small-town Americans now "meet" members of unfamiliar subcultures through the mass media. If proximity no longer matters, then perhaps cities' cultural distinctiveness disappears. Gottdiener (1994, p. 182) puts it this way: "Fischer (1976) made his claims about citylrural differences when that distinction still mattered." (Such technological changes are less likely to make the alter- native theories of urbanism obs~lete.)~

I have insisted, however, that urban-rural differences endure today and do so because space continues to matter (e.g., Fischer 1978b, 1980). Perhaps these technologies are not as pervasive or as effective as imagined; perhaps people still need to meet face to face. Yet, space-transcending technologies must, even according to subcultural theory, shrink city-country differences. If so, research on contemporary American communities provides atypical-probably lower-bound-estimates for the presumed universal effects of urbanism. The need for historical and comparative research is evident.

Ceteris Paribus Conditions

The typical study, be it at the individual or ecological level, examines the correlation between place size and an outcome, holding constant other relevant attributes. Any residual association presumably demonstrates a direct effect of size, of urbanism, on that cultural trait. But subcultural theory does not posit direct effects of urbanism, only effects mediated through subcultural processes. Neither I nor others have fully explored the role of third factors as exogenous influences or as mediating factors.

Begin with this general causal model: urbanism, U, causes some out- come, Y, via a set of subcultural processes, {S), so that U + {S) -+ Y. Holding constant a third factor, Z, is clearly appropriate in two general cases: (1) Z may cause both urbanism and the outcome; that is, Z + U and Z + Y. Then, the researcher needs to test the urbanism-outcome (U-Y) correlation for spuriousness. For example, coastal locations in- crease towns' population sizes and coastal location increases adoption of foreign fashions. One should control for coastal location to see whether size itself correlates with adopting such fashions. Or, (2) Z mediates between urbanism and the outcome: U + [{S), Z] -+ Y. Here, the re- searcher needs to hold Z constant to assess the {S) + Y effect. For example: larger places are wealthier than smaller ones and wealth subsi-

To the extent that Wirthian theory, following Simmel(1969), attributes consequences to the sensory experience of city life, cities will be alienating. And to the degree that there are still social and economic processes that bring distinctive types of people to the city, the compositional approach expects continuing city-country differences.

dizes innovative art forms. To assess how much size increases art innova- tions solely through subcultural processes (e. g., through the formation of "art worlds"), the researcher needs to control for wealth. There are other, more complex cases that also call for controlling third factors.' The most problematic instances, however, are those in which subcultural processes are themselves implicated in these third factors.

Self-selection is an important instance. If people choose cities or avoid cities because of a subcultural phenomenon, is it appropriate to control for that phenomenon in estimating the effect of cities? If, for instance, some young singles move to cities because of the "singles scene," should age (either individuals' ages or mean ages of residents of places) be held constant? If fundamentalist Christians avoid cities in part because cities seem satanic, should religion be held constant when explaining urban- rural differences in values? Even though these traits (age, religion) appear to be, and for the most part are, causally prior to place of residence and should for that reason be held constant, the causal arrow runs both ways. A community's subcultures attract some people to that community, reinforcing its subcultures. The urban "gay" world, for example, appar- ently attracts some homosexuals from smaller communities and their ar- rival bolsters that world. Should preexisting sexual orientation (either as an aggregate or individual trait) be held constant when a researcher asks whether urbanism promotes gay subcultures in a community or individ- ual adherence to a gay subculture? Probably not. (Other processes of self-selection also operate, particularly economic ones. Well-educated people, e.g., may disproportionately move to large cities because that is where the financial returns to their education are greatest. But, since this self-selection is not a response to subcultural phenomena, it would be advisable to control for educational level in assessing the effects of com- munity size.)

Several implications follow. The routine procedure of holding virtually everything constant probably overcontrols third factors and thus underes- timates the effects of urbanism. Conservative testing is preferable to laxness, but may lead to false negative findings. Also, third factors vary in how much they involve self-selection by subculture. Moving to cities to participate in a subculture probably accounts for little, if any, of the covariation between population concentration and, say, gender, income, and education. But moving for subcultural reasons may account for some of the covariation between urbanism and marital status, race, ethnicity,

If Z +U+{S}+Y,controlling for Z reduces the variance in U and thus underesti- mates U's "true effect" on Y. If Z + U + {S}+ Y, and Z + Y, the researcher would need to decompose the direct and indirect effects of Z. One must also consider the many cases where Z +{S},as when subcultural processes, such as attainment of critical mass, result from something other than general urbanization.

religion, occupation, or avocation. Thus, researchers need to consider carefully what is controlled in any particular analysis, instead of simply using the whole laundry list. Perhaps, also, researchers ought to consider models more complex than simple linear ones, and instead test models that employ feedback. This, in turn, counsels longitudinal studies that might disentangle the causal chain.'' More broadly, the theory itself might benefit by explicitly including such feedback effects. Finally, sub- cultural processes may themselves determine levels of a third factor, self-selection aside. The urban "singles scene," to use that example, might lead long-time residents to stay or to become unmarried (symboli- cally: U +{S)+Z, where Z is the proportion unmarried). To control for marital status, Z, would, again, underestimate the total effects of the subcultural processes and urbanism on some outcome, Y, in U + {S)+ Y.

In various ways, then, the standard procedure to control for spurious- ness, while it is appropriately conservative, may overcontrol and under- estimate the model's parameters. Research would improve if more careful attention was paid to the causal status of third factors and if some com- plex linkages were actually modeled.

Connections to Other Debates

The plausibility of subcultural theory rests in part on resolutions of other debates in sociology, at least three of which are especially germane. One concerns crime: Are variations in crime across spatial or social categories explainable by-among several suggested models-socioeconomic conditions, values, or social disorganization? (Recent contributions include Felson et al. 1994; Sampson and Groves 1989; Matsueda et al. 1991; Braithwaite 1989; Bursik and Grasmick 1993; see Gibbs and Erickson 1975.)" Subcultural theory would gain plausibility if value differences

lo A researcher might collect data for a sample of communities over decades to assess the reinforcing effects, if any, of self-selection. One might test the following model, e.g., where U = urbanism, S = singles, and SS = the "singles scene" (as measured by, e.g., fern bars, dating services):

That is, the number of singles and the "scene" stimulate one another, the process

being set in motion and reinforced by urbanism.
" The classic statement of the "subculture of violence" is Wolfgang and Ferracutti
(1967). Researchers have applied the theory especially to understand racial and re-
gional differences in American violence (e.g., Parker 1989). Theories of "differential
association" (e.g., Cloward and Ohlin 1961) are consistent with subcultural theory.
"Routine activities theory" (e.g., Cohen and Land 1987), which stresses victim char-
acteristics (e.g., living in neighborhoods that are "guarded" vs. those that are not),
can be added to either side of the debate.

partly accounted for variations in crime. A second debate concerns reli- gious diversity and religious adherence. Does religious heterogeneity un- dermine religious groups' solidarity? Or, does a great "marketplace" of religious options produce intense sects and churches? (See, e.g., Bruce 1992; Warner 1993.)12 Subcultural theory seems most consistent with the "market" model: Because of subcultural processes, cities ought to gener- ate a variety of religious options, each with strong adherents. Cities should also generate skepticism and atheism, so that the variation in religious position should be greater there than in smaller places. For both reasons, urbanism ought to be associated with unconventional but stronger religious expressions. A third debate addresses ethnicity. Ac- cording to the Chicago school (e.g., Zorbaugh 1929), the modern city erases ethnic differences and divisions. More recent scholars (notably Glazer and Moynihan 1970) argued that ethnic groups maintain their distinctiveness; they may even accentuate their differences in cities (see, e.g., Horowitz 1977). Still newer scholarship, however, claims that as- similationist influences erode and trivialize ethnic differences (e. g., Alba 1990; Waters 1990; Yinger 1985; see Hirschman 1983), a view that under- mines subcultural theory. I will return to these three debates in reviewing the empirical work. At this point, I simply note that subcultural theory rests, at least in part, on other theories of cultural change.

Conclusion

I have so far identified several difficulties in the formulation of the subcul- tural theory of urbanism. A few are fundamental, such as ambiguity about the specific object of study (is it, e.g., cities or particular population concentrations), the level of analysis (communities or individuals), the specification of third factors (exogenous or mediating). These theoretical issues, in turn, raise questions about the foundations of the empirical research. Ironically, that research may in some ways (in focusing on individuals, in overcontrolling third factors, in concentrating on contem- porary America) underestimate the power of the theory.

THE EMPIRICAL LITERATURE

The following review focuses on the key claim that subcultural processes explain how urbanism fosters unconventionality. Many aspects of the theory might be and have been tested-for example, claims regarding

l2 The strongest champions of the "market" model are Finke and Stark (Finke 1990; Finke and Stark 1992; see also Iannaconne 1991). The empirical literature is addressed below.

social relationships and psychological states of mind. In an article, how- ever, I can concentrate on only one main line of argumentation. For the same reason, I also focus on the recent literature.13

Urbanism and Subcultural Heterogeneity

On the average, cities are more diverse than smaller places. For Wirth (1938), heterogeneity partly defined urbanism. The more populous the community, the larger the proportion of ethnic and racial minorities (al- though there are often exceptions, e.g., Native Americans in the United States). Cities tend also to have a greater variety of occupational groups, a more differentiated class structure, and a wider array of special-interest groups organized around activities such as hobbies, albeit with exceptions here, too (see Fischer 1984, pp. 80-86, 126-28). Evidence of urban heterogeneity also includes the cross-sectional correlation between town size and the presence of specialized institutions. For example, population size accounts for the number of high art institutions in a metropolitan area (Blau 1986; see also Fischer 1982, chap. 15).14 In a different ap- proach, Wilson (1986) found that the larger their communities, the greater the variation in residents' social attitudes.15

While the proposition that urbanism is associated with sociocultural heterogeneity appears firm, the critical tests apparently are yet to be done: to correlate measures of internal diversity with population size across a sample of places (and to do so in various societies and eras).

Urbanism and Subcultural Intensity

Central to subcultural theory is the claim that urbanism is not only associ- ated with more distinctive subcultures, but also with greater subcultural intensity, because groups more frequently attain critical mass and more frequently encounter distinctive others in cities. The earlier discussion noted that urbanism should be associated with intensity largely to the

l3 The short time frame does not arise from any "sociological amnesia" (Gans 1992), but from the need to be brief and the intention to build, where reasonable, upon my earlier review (Fischer 1984).

l4 Blau (1986) found that for some institutions city size and the interaction of city size

with city diversity best explained the variance in institutions. l5 Wilson (1986) showed that the variation in attitudes toward subjects like politics and sexuality was greater for residents of metropolitan places than of smaller places-but variation on other sorts of topics was not. His study, however, illustrates a units-of- analysis problem. Since Wilson grouped residents of many different places together into place-size categories, we do not know whether the variability is within particular places, which is what subcultural theory suggests, or among them. (Is it that New Yorkers disagree with one another, or that they disagree with Los Angelenos?)

degree to which specific groups concentrate in cities, that is, to the degree to which urbanism and heterogeneity are correlated. That discussion also alerted us that we are interested in the intensity of subcultures, not the average level of individuals' involvements, although the two are related.

Ethnic subcultures.-In my 1975 paper, I called ethnicity "the most difficult test case" for subcultural theory (1975a, p. 1334). Unlike urban subcultures that emerge around jobs, lifestyles, and the like, ethnic sub- cultures have usually emerged from a rural base. Also, cities expose ethnic groups to novel influences. Whatever the long-term trends, how- ever, at any given moment minority cultures should be stronger in cities than in rural areas. In 1982, I amended the argument. First, I stated that the key factor in generating ethnic solidarity is group concentration, not urbanism per se. This amendment was made to account for finding that Mexican-Americans in outlying towns were more involved in a Chi- cano subculture than were those in the urban core. (This argument may, by the way, be the outcome of confusing individual and ecological levels of analysis.) Second, I distinguished among dimensions of subcultural involvement: social immersion in an ethnic world results from concentra- tion, but formal ties to the ethnicity may not; indeed the latter may be more important to minority group members who are isolated in small communities. Later, I further modified the argument: the urban environ- ment both strengthened and weakened ethnic cultures, with the net out- come depending on a host of factors (Fischer 1984, p. 145). Moreover, the strengthening effect is greatest for the smaller ethnic groups and with respect to "central" cultural elements, such as marriage and "ethos," rather than "peripheral" cultural features such as language and dress. I explained the erasure of ethnic differences by subcultural processes, too:

(1) cities foster new subcultures that compete with ethnicity and (2) some cultural elements diffuse from majorities to minorities.16 My admittedly tepid conclusion from all this was that the "tide of history moves most [ethnic] groups toward assimilation, but that tide is resisted at least some- what in the larger cities" (Fischer 1984, p. 155). For all these hedges, the distinctive claim of subcultural theory remains that, all else equal, cities increase rather than diminish ethnic distinctiveness.

Despite its problematic nature, studies of ethnicity constitute most of the research on urban subcultures. Minority cultures generally, albeit

l6 Peil (1984, p. 195) makes a similar claim about tribal groups in African cities: "Competition and conflict often increase communalism: self-conscious groups whose members share a common origin associate more with each other than with strangers and work to further the interests of their groups. However, this must not be overem- phasized. Urban life provides many opportunities for the development of new identi- ties based on income, education, occupation, or neighborhood. These cross-cut ethnic identity and provide a more complex set of social relations than ethnic boundaries."

with interesting exceptions, dissolve into the urban majority (see, e.g., Model1 and Lees 1988; Zunz 1982; Breton et al. 1990; Kalmijn 1991). Still, community size-mediated by the sizes of specific groups-is positively correlated with what Breton (1964) labeled the "institutional com- pleteness" of ethnic subcultures (formal and informal organizations that sustain ethnic solidarity; see also Darroch and Marston 1984; Morawska 1994; Hirabayashi 1986; Olzak and West 1991; Venturelli 1982). Most of the research, however, has examined not group properties like institu- tionalization but individuals' commitments to ethnic groups, the wrong level of analysis.

Take the African-American case. It is in many ways special,17 but nevertheless contributes a test of the theory. At the ecological level, con- sistent with theory, the larger the city or the black population, the more "developed" black communities and the more assertive black politics are (e.g., Karnig 1979; Olzak and West 1991). For example, McAdam (1982, pp. 98-102) attributes the rise of the civil rights movement in great part to the urbanization of the African-American population. In particular, black churches in cities were larger, more professional, and more effective than rural ones. At the individual level, however, the findings are different. A few survey studies show that African-Americans in large cities are no more or are slightly less likely than those in smaller places, all else equal, to express racial solidarity or to report having predominantly black friends (Broman, Neighbors, and Jackson 1988; Ellison and London 1992; Deng and Bonacich 1991).

Conceding that the bulk of the research on ethnicity deals with individ- uals, what is the connection between urban residence and individual involvement? (The comparison here is between people living in large places to those in small ones within the host society, not to villagers in the native country.) The answer is complex, varying by group and type of the attachment. North American evidence suggests that, in some ways such as ethnic identification and organizational membership, urban peo- ple express equivalent or less ethnic commitment as do rural people, but that urbanites are more involved socially with fellow ethnics, for exam- ple, in-marrying more often (Fischer 1982; Alba 1990; Bernard 1980).

Are individual members of large ethnic groups-group size being a mediating variable in the subcultural model-more highly committed than members of small ones? Yes, but again the association seems to vary by type of commitment. Alba (1990) found, for example, that the concentration of a European ethnic population in a neighborhood did

l7 One way is that, unlike most American minorities, African-Americans were, until only about two generations ago, a heavily rural and small-town population within the United States itself.

not correlate with respondents' ethnic identification, but did correlate with their social involvement. Rabinowitz et al. (1992) found that the larger a city's Jewish population, the less individual Jews religiously and formally engaged in the Jewish community, but the more Jewish friends they had (see also Goldscheider [I9861 for data at the neighborhood level). Other studies report some association between group size, at either the town or neighborhood level, and individuals' ethnic involvement (e.g., Hwang and Murdock 1988; Lewis 1988; Bernard 1980; Jankowski 1985; but not Breton et al. 1990; see Morawska 1994).18 However, a method- ological problem qualifies the conclusion that group size fosters individual commitment. The correlation may be due to ethnically committed people choosing to live in ethnically attractive places. Whether such self-selection undermines subcultural theory or actually reaffirms it depends on how one theoretically models self-selection (see above).

In all, population concentration-urbanism-appears to sustain ethnic institutions, but not necessarily to sustain individuals' ethnic attach- ments. Particular ethnic concentrations, generally but not uniformly greater in cities, more clearly sustains group and individual ethnicity. Even then, the patterns vary by type of activity. It may be, for example, that people living in the midst of fellow ethnic group members are behav- iorally involved in the subculture, but not self-consciously so.

Religious groups.-Urbanism, I argued, strengthens religious involve- ment-for minority religions and for people to whom religion is impor- tant (Fischer 1982, chap. 16). I found that urbanism increased individu- als' communal affiliation with coreligionists, but apparently undermined their endorsement of conventional religious ideas and practices (part of the nontraditionalism effects of urbanism; see also Rabinowitz, Kim, and Lazerwitz 1992; Stinner et al. 1990). More recent studies typically show that the larger the town, the less Americans' fidelity to conventional religious doctrine and practice (e.g., Petersen and Takayama 1984; El- lison 1993; Harley 1993; cf. Chalfant and Heller 1991).

In the earlier theoretical discussion, I noted the debate between the "sa- cred canopy" and the "religious market" schools concerning the correla- tion of local religious diversity with individual religious belief and commit- ment. The jury is out on this debate. It has revolved to a great extent around the empirical question of whether population size, or indices of religious heterogeneity, or both correlate with church membership. It remains unre-

l8 Bernard (1980) found a positive correlation between group size and in-marriage at the city level; Tomaskovic-Devey and Tomaskovic-Devey (1988) found one at the state level; and Alba and Golden (1986) found one at the national level. Abma and Krivo (1991) failed to find a connection with the tendency of Mexican-Americans to have group-typical (i.e., high) birth rates, a behavior which might can be seen as commitment to group norms.

solved in part because of difficult problems in data analysis. lg Subcultural theory would seem most consistent with the market model.

Although some studies imply that, at least today, city dwellers are less often conventionally religious than are country people, there is little systematic evidence on urbanism and religious subcultures at the ecologi- cal level. Historically, cities have nurtured the rise and institutionaliza- tion of major religious traditions (see, e.g., .Weber 1963, pp. 468ff.), but we apparently lack systematic evidence to establish causal claims.

Other subcultures.-The theory claims that larger places have more in- tense non-descent-based subcultures, such as those focused around profes- sions (e. g., medicine), lifestyles (e.g., bohemians), pastimes and avocations (e.g., the arts), exceptional traits (e.g., the blind), and so on. This associa- tion appears to be taken for granted. For example, the journal, Urban Life (later, Urban Life and Culture) was largely devoted to ethnographic studies of such subcultures. But there is surprisingly little comparative research on this point. In the earlier section on subcultural heterogeneity, Inoted prima facie evidence for such a correlation: larger places more often have institu- tions that indicate the presence of a subculture, for example "gay" newspa- pers. Also, historical evidence points to a connection between urban growth and class cultures (e.g., Nash 1979, 1987; Blumin 1989; Meyero- witz 1988). Still, direct, ecological assessments of how intense such subcul- tures are across places of different sizes is still rare.

In support of subcultural theory, researchers have found that living in cities and in especially poor neighborhoods encourages individual involvement in a welfare "underclass" (Rank and Hirschl 1988; Hirschl and Rank 1991; Crane 1991; Rosenbaum and Popkin 1990; Furstenberg et al. 1987). I found that urbanism encouraged individuals' involvements in hobby-based subcultures, but failed to show that urbanism increased involvement in job-based subcultures (Fischer 1982, chap. 17). And re- cently, Laumann et al. (1994, pp. 302-9; see also Rogers and Turner 1991) found that the proportion of Americans who reported homosexual experiences increased greatly with the urbanism of their current residence and of their adolescent hometowns, a pattern consistent with the impres- sion that gay subcultures are larger in bigger cities.

Logic and illustration (e.g., bohemian cafes, science fiction bookstores) point to an urban intensification of specialized subcultures, but these are

l9 See Finke and Stark (1988), Breault (1989), Bainbridge (1990), Iannaconne (1991), Land, Deane, and Blau (1991), Blau, Land, and Redding (1992), Finke and Stark (1992, esp. pp. 203-7, 294, n.4), Bruce (1992), Warner (1993), and Chaves and Caven- dish (1994). Much of the empirical work has used city- or county-level data that are highly collinear, ratio variables, and complex innovations like Land et a1.k (1991) use of a spatial "adherence potential" measure.

hardly proof.'O The theory requires more systematic evidence, at the ecological level of analysis, that size leads to "institutional completeness" and other indicators of intensity for small, minority, and unconventional populations.

Conclusion.-Ethnicity and religion, "traditional" bases of affiliation, are indeed difficult cases for subcultural theory. This is conceded in the formulation of the theory, because such affiliations are presumably chal- lenged in cities by the emergence of innovative subcultures. (Consider, e.g., the classic immigrant fiction depicting tension between the tradi- tional home and the "street.") What limited empirical evidence exists also poses a problem for the theory. It seems that urban places may-to the degree to which they house specific critical masses-sustain subcultural institutionalization, but simultaneously undermine individuals'for- ma1 involvement in those subcultures, and yet still reinforce their social involvement in those subcultures. For subcultures based on other affilia- tions-occupation, pastime, politics, lifestyle, and so on-we expect to see clear evidence of urban intensification. But there is simply little com- parative research on them.

Urbanism and "Unconventionality"

Subcultural theory claims to explain an association between urbanism and "unconventional" behavior. The theory posits that the larger the place, the higher its rates of unconventionality, and that subcultural processes-the emergence of innovative subcultures and the diffusion of their culture to others in the city-mediate that association. In contrast, Wirthian theory explains that association by the "breakdown" of social control and moral order. The most frequently studied form of unconven- tionality is major crime.

"Street" crime.-All the relevant theories, including the more recent "routine activity" theory (e.g., Cohen and Felson 1979; Cohen and Land 1987; see review in Bursik and Grasmick 1993, pp. 62-89),*' imply that population size is correlated with crime even after compositional traits of the communities-for example, income levels-are held constant. Re- searchers often correlate size with crime rates, in part because the data (notably FBI statistics) are so easily available. Unlike most of the research so far reviewed, the level of analysis here is typically the locality, at some

20 One could counterargue, for instance, that after a certain size, the largest cities dissipate subcultures by offering distractions to potential members (see, e.g., Peil 1984, p. 56).

Cities may have more crime than rural places because they especially lack "guard- ianship." In the city, e.g., people are less often home during the day.

level of aggregation. The subject is almost always "street" crime and most often it is violent crime, especially homicide. Despite all the work, technical problems confound many of these studies and it is no wonder that researchers disagree about whether key place attributes, such as percentage nonwhite or percentage poor, determine crime rates.'* Also, the recent literature overwhelmingly employs contemporary American criminal data, which, given the particularities of both the nation and the epoch, raises questions about generalizability.

Nevertheless, the research consistently shows that in contemporary America, the larger the community, the higher the homicide rate (see, e.g., Land, McCall, and Cohen 1990; Williams 1984; Sampson 1985b; Messner and Sampson 1991; cf. Wilkinson 1984). Some studies suggest that larger places are relatively high only in homicides connected to prop- erty crime but not in fatal crimes of passion (e.g., Loftin and Parker 1984; Parker 1989; Williams and Flewelling 1988; Peterson and Krivo 1993). As to the question of whether urbanism correlates with homicide rates once compositional differences among places are held constant, the answers are yes (Land et al. 1990; Sampson 1987; Messner and Sampson 1991; Jackson 1984; Kowalski and Duffield 1990); no (Sampson 1985a, 1985b; Balkwell 1990; Harer and Steffensmeier 1992; Miethe et al. 1991); and it depends (on the type of homicide or the covariates-Williams 1984; Parker 1989; Williams and Flewelling 1988; Peterson and Krivo 1993).

Studies of other countries and eras, however, typically show that larger communities have equal or lower rates of personal violence, including homicide, than smaller communities (e.g., Gillis 1989; Johnson 1992; Diederiks 1993; Kennedy, Silverman, and Forde 1991; vs. Van Poppel 1989; Archer and Gartner 1984, chap. 5; see also review and citations in Fischer [1984, pp. 100-109, 226-30]).'~ Perhaps the strong link between

22 For example, results often differ depending on whether the data come from reports to police, arrest data, or victimization surveys; crime definitions, crime recording, and crime reporting procedures fluctuate across jurisdictions; units of analysis vary, ranging from blocks to metropolitan areas; multivariate analyses often include collin- ear variables, such as population size and population density, which make estimates unstable; and statistical models invariably include several ratio variables, which often raise problems of interpretation (see, e.g., O'Brien 1983; Byrne and Sampson 1986; Bursik and Grasmick 1993).

23 Archer and Gartner (1984, chap. 5) found that, in 18 of 24 comparisons, primary cities' homicide rates exceeded those of their nations. However, six of the 18 were Anglo societies (e.g., Scotland, the United States, New Zealand) and six were other Western European nations, so that, in a count of distinctive culture areas, the balance is about even. The authors acknowledge, too, that often rural areas' homicide rates exceed those of nonprimary urban places. Archer and Gartner also found that homi- cide rates were not associated with the growth of population in the primary cities. (See also Fischer 1984, p. 303, n.33.)

cities and murder is particular to contemporary America.24 Since urban- ism has not been shown to stimulate homicide-especially outside America today-none of the explanations are needed. (There is less, and less consistent, evidence on other forms of physical violence.)

Urbanism, on the other hand, does consistently stimulate property crime and vice crime in contemporary America (e.g., Byrne 1986; Samp- son 1986a; Messner and Sampson 1991) and elsewhere (e.g., Block 1987; Osterberg 1992; Johnson 1992; Shelly 1981).~' How might we explain this? Subcultural theory contends that community size fosters critical masses of victims, criminals, and criminals' clients and servicers (e.g., Johns, fences), and in particular sustains a criminal subculture. Descrip- tions of organized crime (e.g., Egmond 1993) are compatible with this explanation, but most of the evidence is indirect. Criminological research shows that delinquent behavior is tied to peer groups and criminal associ- ates (e.g., Sampson and Groves 1989; Warr 1993; McCarthy and Hagan, in press; Felson et al. 1994), that criminals and victims often know one another (e.g., Irwin 1987), and that hard-core criminals seem to share points of view (see, e.g., Matsueda et al. 1991). The key evidence cited against general subcultural explanations for crime is that criminals and delinquents seem to endorse the same end-values as the wider public (e.g., Bursik and Grasmick 1993, pp. 138-42). But that critique employs a too-simple concept of cultural difference^.^^ As Braithwaite (1989) ar- gues, criminal subcultures are real and need to be to play some role in any complete explanation of criminal patterns.

Charles Tittle (1989a, 1989b; Tittle and Stafford 1992) provided a rare direct test of the hypothesis that subcultures mediate the urbanism-crime correlation. In a three-state telephone survey, the more populous respon- dents' communities, the likelier they were to say that they would gamble illegally, would smoke marijuana, or would cheat on their taxes (but not that they would commit an assault; Tittle 1989~). The correlation be- tween urban residence and these answers remained significant, although

24 Land et al. (1990) report that size was associated with homicide in the United States from 1950 through 1980, but, it is interesting that the unstandardized regression coefficients for urbanism increased sharply over that period. The lower rates of urban relative to rural violence before the 20th century might be explained by the greater force of state power in cities then and by the urban location of the "civilizing process" described by Norbert Elias (see Johnson and Monkkonen 1993).

25 I treat robbery here as a property crime, although it is often categorized with

assault, rape, and homicide as violent crime, because its purpose is to gain property. 26 The key to cultural differences among subgroups probably lies less in the different explicit end values than in the different cultural "tool kits," or repertoires, that groups may have (Swidler 1986). So, e.g., gang youths may envision a conventional "Ozzie-and-Harriet" lifestyle in their adult years, but lack command of the cultural skills needed to attain it in a conventional manner. Also, group influence may operate independently of individuals' values (Felson et al. 1994).

substantively small, after controlling for many personal traits. (Urban respondents were also likelier, all else equal, to report actually having gambled, smoked marijuana, cheated, and committed petty theft; see Tittle 1989b.) Tittle (1989~) then constructed measures of how involved respondents were in specific deviant subcultures. Controlling for these measures eliminated the residual correlation between community size and forecast deviance, implying that membership in such subcultures interpreted the correlation. The basic results support subcultural the- ~ry,~'

although the study's conclusions must be q~alified.~'

The major alternative interpretation of the urbanism-crime correlation is that social disorganization mediates between the two. In an important series of studies, Robert J. Sampson (see, esp., Sampson 1986b, 1987; Sampson and Groves 1989) has shown that indicators of local disorganiza- tion-most notably, the local proportion of female-headed households- predict local rates of criminal behavior. In one study (Sampson 1987), the percentage of female-headed households in cities correlated with cities' robbery rates. The distribution of female-headed households, however, did not account for the association between robbery rates and city size.29 In a study of 238 British localities (Sampson and Groves 1989), center-city rates

''Tittle (1989a) derived another prediction from subcultural theory: the effects of being in a deviant subculture on forecasted deviance should be greater in more-urban areas than in less-urban ones (because the urban subcultures are more intense than small-town ones). This interaction effect did not appear in his data; the measure of deviant subculture membership was an equally strong predictor of forecasted criminal- ity across place sizes. Tittle sees this as a key strike against the theory. The interaction effect derivation is plausible but, given the complexity of the analysis and given the main effects, far from conclusive.

"The most serious qualification concerns the subcultural involvement measures. Tit- tle constructed factor scores for each one-say, involvement in an illegal gambling subculture-based on several questionnaire items, including whether respondents knew people who engaged in the activity, whether they dismissed moral qualms about the activity, and, most important, whether they had previously engaged in the behavior. These reports of past behavior, which were included in the subculture measures, correlate with predicted future deviance at from r = .31 to r = .58 (Tittle

1989a, p. 285; see also Tittle 1989b, p. 281). Thus, the association of urbanism with past deviance may largely explain why the composite subculture measures interpret the correlation of urbanism with predicted deviance. If so, little in the urbanism- predicted deviance correlation is left for the simply subcultural processes-e.g., social relations-to explain. Other concerns include the skewness of the dependent variables that may lead to underestimating effects; the few cases at the higher ends of urbanism; and the atypicality of the urban measure (urbanism in this sample, for instance, is uncorrelated with education). Nevertheless, this remains one of the most thorough tests of the theory, at least at the individual level, to date.

29 The concentration of female-headed households did not explain away the urbanism- crime correlation, because, for black offenders city size did not correlate at all with robbery rates and for white offenders city size correlated with robbery rates even after controlling for female-headed households.

of victimization for property crimes (but not violent ones) exceeded those

elsewhere. Two indicators of "neighborhood disorganizationn-whether

respondents reported friends in the neighborhood and whether respondents

reported unruly groups of teenagers on the street-explained part of the

center cities' excess crime. But center-city property crime rates exceeded

those elsewhere even when these and other variables were held constant

(Sampson and Groves 1989, tables 4,5,6). In sum, these indicators of disor-

ganization account for some of the variations in crime among localities, but

do not suffice to explain why urban centers have higher rates.

Sampson and Groves's (1989) British study raises a general interpretive issue, too. Their strongest explanatory variable was residents' reports of how commonly teenagers made public nuisances of themselves. Sampson and Groves sensibly interpret this as an index of low social control in the neighborhood. But it could also be read as a measure of local alternative subcultures, the availability of gangs with which to "hang." The more general issue is that some measures can be seen both as indicators of breakdown and of subcultures. Are, for example, social worlds where welfare is "a way of life" the outgrowth of social breakdown, of subcul- tural formation, or of both? How might one distinguish between them?

The connection of urbanism to property and vice crime might also be explained in terms of opportunities: cities contain more wealthy people, more potential targets, easier access, easier escape routes, and so on. Two kinds of evidence lend weight to this explanation: historical studies (e.g., Osterberg 1992; Johnson 1992) find that urban property crime esca- lated rapidly after about 1900, presumably because these sorts of oppor- tunities multiplied. And some neighborhood studies (e.g., Sampson 1983; but not, e.g., Smith and Jarjoura 1989) point out architectural features such as high-rise buildings that apparently increase theft and robbery rates because they make predation easier.30 Property crime may be higher in cities, in part, for opportunity reasons, but there is still room for additional explanations, be they subcultural or breakd~wn.~'

30 Some have used measures of "guardianship" as predictors of crime, but their use is ambiguous. Measures indicating that people are away from home-rates of working wives and of nonfamily households-can also be labeled measures of breakdown.

31 Subcultural theories of crime have also appeared in a different context, in the idea of a "subculture of violence" (Wolfgang and Ferracuti 1967). American researchers have often located such a subculture among African-Americans and Southerners. Most studies have found that the proportion of black residents in a community correlates strongly and independently with local rates of violence (e.g., Williams 1984; Loftin and Parker 1984; Wilkinson 1984; Balkwell 1990; Miethe et al. 1991; vs. Harer and Steffensmeier 1992; Ellison 1991; Peterson and Krivo 1993). But Sampson (1987) has argued that the effect is fully accounted for by structural features of black communi-

To sum up, rates of property and vice crime are consistently higher in larger communities. Since crime may be a consequence of several condi- tions, such as targets of opportunity, customers, laxity of control, and residents with criminal propensities, more than one factor may account for the urbanism-crime correlation. Breakdown explanations earn some support in the research literature, except for a major anomaly: in broad historical and cross-cultural perspective, urbanism does not usually stim- ulate violent crime. Breakdown presumably should "release" all sorts of criminal impulses. Also, the measures of breakdown are somewhat ambiguous: Does a concentration of single-parent households and teenage loiterers indicate breakdown or alternative social orders? Or both? (One referee suggested that Wirth and I may have observed similar phenom- ena, but what he described as breakdown in the 1930s I depicted as "alternative community" in the 1970s-a reflection of a different histori- cal context.) Subcultural explanations of the urbanism-crime association also gain some support in the research, but several studies cast doubt on the importance of group processes in explaining individual criminality. Given the great complexity of the topic (I have not even touched on issues such as differential labeling or law enforcement) and the technical problems in data analysis, we still await more definitive studies. These should be conducted at the ecological level with attention to the modeling issues raised in the first section of this paper, such as statistical over- control.

Other forms of unconventionality.-The idea that large cities nurture various forms of avant-garde culture-innovative art, music, sexual, and political social worlds, for example-seems part of common wisdom, but there is little systematic evidence on the point. There is, instead, research on "problem" behaviors and groups.

Tittle (1989a, 1989b), in the study discussed above, found that, all else equal, urban residents were likelier than rural residents to report having smoked marijuana, cheated on taxes, and lied to a spouse. As noted earlier, a few studies found that welfare-eligible Americans were likelier to take stigmatizing welfare if they lived in more urban places, probably because of cities' more supportive climates of opinion (Rank and Hirschl 1993; Hirschl and Rank 1991; Rank and Hirschl 1988; Rosenbaum and

ties, that no appeal to a distinctive culture is necessary. Similarly, studies differ on whether Southern residence or "Southernness" is independently associated with violence or violent attitudes (see, e.g., Huff-Corzine, Corzine, and Moore 1986, 1991; Williams and Flewelling 1988; Land et al. 1990; Ellison 1991; Sampson 1985a). The subcultural theory of violence shares premises with the subcultural theory of urban- ism. Doubts about the first should apply to the latter. But we should recall that urbanism is not usually associated with higher rates of violence.

Popkin 1990). Also as noted earlier, admitted homosexuality-still considered deviant by most-is much more common in larger places (Lau- mann et al. 1994; Rogers and Turner 1991).~' The connection between drug use and urbanism is complex, but one large-scale study of high school students (Wallace and Backman 1991) is instructive: big-city youths more often used marijuana and cocaine, but youths in smaller places more often used cigarettes and alcohol. Urbanism would seem unrelated to substance use per se, but correlated with using "unconven- tional" as opposed to "traditional" substances. A few studies describe "concentration effects": where local rates of problem behaviors, such as dropping out of school or early pregnancy, are especially high, individual youths are likelier, all else equal, to engage in that behavior (Crane 1991; Furstenberg et al. 1987; Anderson 1991; but not Ku et al. 1993). Concentration effects suggest a cultural influence.

The discussion above on religious subcultures noted that individuals' adherence to conventional religiosity is lower in large cities. At the same time, it appears that cities stimulate new, alternative religious forms, such as sects, heterodoxies, and innovative practices.

In contrast to these examples of social deviance, urbanism is not consis- tently associated with psychological disorders (Fischer 1984, pp. 189-90; also, e.g., Amato and Zuo 1992; Romans-Clarkson et al. 1990; Lewis et al. 1992; Takei et al. 1992). It seems uncorrelated with crimes of passion (see above). Similarly, urbanism is not associated with suicide (apologies to Durkheim) in contemporary America (Breault 1986; Kowalski, Faupel, and Starr 1987; Pescosolido 1990)-although it is so elsewhere.33 This contrast implies that city life largely encourages group-based, socialized unconventionality rather than individualistic unconventionality. The point deserves emphasis: breakdown theory posits that all sorts of uncon- ventional impulses are "released" in cities; subcultural theory argues that only socially constructed unconventionality emerges in cities. The fact that solitary deviance (such as mental disorder, homicides of passion) is not generally or consistently greater in larger places, but group devi- ance (such as organized crime, political dissent, and the cultural avant- garde) is, supports subcultural theory.

Unconventional attitudes.-In lieu of data on noncriminal unconven- tional behavior, many researchers have examined survey respondents'

32 In Rogers and Turner (1991), having had male-male sexual encounters is associated with the size of current place of residence, but not place of origin. In Laumann et al. (1994), it is associated with both-and urbanism is the strongest correlate of reported homosexual encounters or desires.

33 In 19th century data (e.g., Durkheim [I8971 1951; Van Poppel 1989) and in other parts of the world (Simpson and Conklin 1989), urban suicide rates exceed those of rural areas. The methodological question arises in these cases, however, that suicides may be and have been better recorded in urban places.

attitudes toward unconventional and presumably "problem" behavior such as marijuana smoking, premarital sex, political radicalism, and homosexuality. Over the years, the research has usually shown that city dwellers are more "tolerant" of such behavior than are residents of smaller places and often that the attitudinal differences could not be explained away by compositional differences, such as levels of education (Fischer 1984, chap. 8). Typically, the differences are small but consis- tent; hardly ever are rural or small-town respondents found to be more tolerant. Recent research also typically finds that urban residence corre- lates modestly with such tolerance when other factors are held constant (see, e.g., Wilson 1985a, 1991; Tittle and Stafford 1992; Tuch 1987; Jang and Alba 1992;34 McCloskey and Zaller 1984; but not Bobo and Kluegel 1993; Ellison and Musick 1993).

The research on urbanism and verbal acceptance of unconventionality raises several issues. Some are particular to this genre of research: What is the connection between attitudes and behavior? Does urban tolerance reflect acceptance of deviance generally, or of only a leftist political stance?35 Other issues are more general: What are the appropriate ceteris paribus conditions? Is, for example (see Ellison and Musick 1993), theo- logical conservatism a spurious covariate to be controlled or a mediating variable between urbanism and social attitudes? What happens when the unconventional becomes the conventional, as in the tolerance of premari- tal sex?36

Yet broader issues reemerge that were mentioned earlier in this paper. First, the attitudinal literature does not (except perhaps for Tittle 1989a, 1989b) examine subcultural processes that presumably explain urban ef- fects. Second, the attitudinal literature deals with individual differences rather than place differences. This may work against subcultural theory, in that modest differences in individual opinions can aggregate to sub- stantial differences in local climates of opinion. Third, in subcultural theory urbanism is ultimately important because cities are where specific populations cluster (clustering, in turn, that increases access). For some issues, however, the key population may not concentrate in precisely the

34 Although Jang and Alba (1992) found that rural respondents were least tolerant in their study, they also found that their tolerance measures peaked in the suburbs of large cities (see also Wilson 1992).

35 For example, Wilson (1985) found that urbanites, ceteris paribus, were more willing than ruralites to extend freedom of speech to all sorts of deviant groups, left- or right-wing. But Ellison and Musick (1993) found that city dwellers were distinctive in being tolerant only toward homosexuals. See also Marcus et al. (1980).

36 I argued (Fischer 19786) that when dissident views become orthodox, urbanites are the new dissenters to those views. But except for suggestive data in that paper, we have no evidence for the claim.

largest places. If, for example, the ideology of civil liberties and political tolerance is largely a possession of the well educated, we might expect that tolerant climates of opinion would form where those highly educated people live-say, in the suburban ring of metropolitan centers (cf. Jang and Alba 1992; Wilson 1992).

In conclusion, far less systematic research has been done on noncrimi- nal deviance, especially at the aggregate level, than on crime. The best guess is that urban places are more often the sites of unconventionality. But, even then, evidence about why-breakdown, subcultural processes, other intervening variables-is notably missing.

CONCLUSION

Several extended treatments have dissected Wirth's theory, "Urbanism as a Way of Life." The same has not been true of "Toward a Subcultural Theory of Urbanism," its coequal rating in several textbooks notwith- standing. Nor has a summation (aside from Fischer [1976, 19841) been delivered on the latter's empirical value. In this essay I have tried to fill those two gaps.

How plausible is the subcultural theory of urbanism? Where there is evidence, it usually supports the theory and only occasionally contradicts it. But most of the critical tests have not been done. Urbanism probably promotes the emergence of numerous and diverse subcultures within a community. Whether urbanism promotes more intense subcultures is still uncertain. The research on ethnoreligious groups yields mixed and con- fusing results depending on level of analysis (place vs. individual) and the specific dimension of intensity (social, organizational, ideological). Systematic research on other kinds of subcultures is sparse. Urbanism is consistently associated with unconventionality, including property and vice crime. How much of that association is explained by subcultural processes, rather than by breakdown or other mechanisms, is still open. The distinction between group-based and solitary deviance points toward subcultural theory. The best summation is that subcultural theory is perhaps the most promising theory of urbanism, but is far from estab- lished (or rejected) as the theory of urbanism.

While this uncertain conclusion arises from a dearth of critical tests, it also results from basic ambiguities in the theory's formulation. These ambiguities include a confusion about levels of analysis. Although research on individuals can be informative, the test of the theory rests on place-level analyses. There needs to be better formal modeling of the theory, distinguishing in particular variables that are exogenous and to be held constant from those which are endogenous. Especially important among the latter are those that might indicate self-selection by subculture.

Earlier research may have erred through overcontrolling. Ironically, these problems probably led to underestimating the theory's validity. Most generally perhaps, urbanism needs to be better connected to the concentration and internal accessibility of specific populations: How do cities-particularly in the modern era-facilitate people's access to one another in ways that produce critical mass and intergroup dynamics?

For future research to answer the question of whether this theory is more plausible than others, interested researchers must reformulate the theory and take certain other steps. Ideally, the latter include using places as the units of analysis; directly measuring intervening subcultural pro- cesses, such as group concentration, access, critical mass, intergroup fric- tion, and self-selection; controlling for truly spurious factors; examining a range of subcultures (not just ethnic groups) and a range of unconven- tional outcomes (not just crimes); and widening the sites of study to other nations and times. Few, if any, pieces of research will meet all these criteria, but moves in these directions will yield more confident conclu- sions than have been possible here.

One might imagine that 20 years would be sufficient to determine the validity of a theory. But sociologists have learned to be patient. Complex and multiply determined phenomena in natural settings rarely permit a critical experiment, so we must wait for observations to accumulate. In the meantime, we can refine our theories and hone our tools. Perhaps that is why I entitled the 1975 paper "Toward a Subcultural Theory."

REFERENCES

Abler, Ronald, and Thomas Falk. 1975. "Effects of Space-Adjusting Technologies on the Human Geography of the Future." Pp. 35-56 in Human Geography in a Shrinking World, edited by R. Abler, D. Janelle, A. Philbrick, and J. Sommer. Belmont, Calif. : Duxbury Press.

Abma, Joyce C., and Lauren J. Krivo. 1991. "The Ethnic Context of Mexican Ameri- can Fertility." Sociological Perspectives 34 (2): 145-64. Alba, Richard. 1990. Ethnic Identity: The Transfownation of White America. New Haven, Conn.: Yale University Press. Alba, Richard D., and Reid M. Golden. 1986. "Patterns of Ethnic Marriage in the United States." Social Forces 65 (September): 202-23. Amato, Paul R., and Jiping Zuo. 1992. "Rural Poverty, Urban Poverty, and Psycho- logical Well-Being." Sociological Quarterly 33 (Summer): 229-40.

Anderson, Elijah. 1991. "Neighborhood Effects on Teenage Pregnancy." Pp. 375-98 in The Urban Underclass, edited by Christopher Jencks and Paul Peterson. Wash- ington: Brookings Institution.

Archer, Dane, and Rosemary Gartner. 1984. Violence in Crass-National Perspective. New Haven, Conn.: Yale University Press. Bainbridge, William Sims. 1990. "Explaining the Church Member Rate." Social Forces 68 (June): 1287-96. Balkwell, James W. 1990. "Ethnic Inequality and the Rate of Homicide." Social Forces 69 (September): 53-70.

Bernard, R. M. 1980. The Melting Pot and the Altar. Minneapolis: University of Minnesota Press. Berry, Brian J. L., ed. 1976. Urbanization and Counter-Urbanization. Beverly Hills, Calif.: Sage. Blalock, Hubert M. 1984. "Contextual-Effects Models: Theoretical and Methodologi- cal Issues." Annual Review of Sociology 10:353-72. Blau, Judith R. 1986. "The Elite Arts, More or Less de Riguer: A Comparative Analysis of Metropolitan Culture." Social Forces 64 (June): 875-905.

Blau, Judith R., Kenneth C. Land, and Kent Redding. 1992. "The Expansion of Religious Affiliation: An Explanation of the Growth of Church Membership in the United States, 1850-1930." Social Science Research 21 (December): 329-52.

Blau, Peter. 1981. "Behavioral Sociology or Behavioral Sociology?"American Sociol- ogist 16 (August): 170-71.

Block, Richard. 1987. "A Comparison of Neighborhood Crime in EnglandIWales, Scotland, the Netherlands, and the United States." Paper presented to the Ameri- can Sociological Association, Chicago.

Blumin, Stuart 1989. The Emergence of the Middle Class: Social Experience in the American City, 1760-1900. New York: Cambridge University Press.

Bobo, Lawrence, and James R. Kluegel. 1993. "Opposition to Race-Targeting: Self- Interest, Stratification Ideology, or Racial Attitudes?" American Sociological Re- view 58 (August): 443-64.

Braithwaite, John. 1989. Crime, Shame and Reintegration. New York: Cambridge University Press.

Breault, Kevin D. 1986. "Suicide in America: A Test of Durkheim's Theory of Reli- gious and Family Integration." American Journal of Sociology 92 (November): 628-56.

. 1989. "New Evidence on Religious Pluralism, Urbanism, and Religious Par- ticipation." American Sociological Review 54 (December): 1048-53. Breton, Raymond. 1964. "Institutional Completeness of Ethnic Communities and the Personal Relations of Immigrants." American Journal of Sociology 70: 193-205.

Breton, Raymond, Wsevolod W. Isajiw, Warren E. Kalbach, and Jeffrey G. Reitz.1990. Ethnic Identity and Equality: Varieties of Ethnic Experience in a Canadian City. Toronto: University of Toronto Press.

Broman, Clifford, Harold W. Neighbors, and James S. Jackson. 1988. "Racial Group Identification among Black Adults." Social Forces 67 (September): 146-58.

Bruce, Steve. 1992. "Pluralism and Religious Vitality." Pp. 170-94 in Religion and Modernization: Sociologists and Historians Debate the Secularization Thesis, edited by Steve Bruce. Oxford: Clarendon Press.

Bursik, Robert J., Jr., and Harold G. Grasmick. 1993. Neighborhoods and Crime. New York: Lexington Books. Byrne, James N. 1986. "Cities, Citizens, and Crime: The Ecological/Nonecological Debate Reconsidered." Pp. 77-101 in The Social Ecology of Crime, edited by James

M. Byrne and Robert J. Sampson. New York: Springer-Verlag.

Byrne, James M., and Robert J. Sampson. 1986. "Key Issues in the Social Ecology of Crime." Pp. 1-24 in The Social Ecology of Crime, edited by James M. Byrne and Robert J. Sampson. New York: Springer-Verlag.

Castells, Manuel. (1968) 1976. "Is There an Urban Sociology?" Pp. 33-59 in Urban Sociology: Critical Essays, edited by C. G. Pickvance. London: Tavistock.

Chalfant, H. Paul, and Peter L. Heller. 1991. "RuralIUrban versus Regional Differ- ences in Religiosity." Review of Religious Research 33 (September): 76-86.

Chaves, Mark, and James Cavendish. 1994. "More Evidence on U.S. Catholic Church Attendence: Deflated Rates and Religious Economies." Paper presented to the American Sociological Association, Los Angeles, August.

Choldin, Harvey M. 1985. Cities and Suburbs. New York: McGraw-Hill. Cloward, R. A,, and L. A. Ohlin. 1961. Delinquency and Opportunity. Glencoe, Ill.: Free Press. Cohen, Lawrence E., and Marcus Felson. 1979. "Social Change and Crime Rate Trends." American Sociological Review 44: 588-607.

Cohen, Lawrence E., and Kenneth C. Land. 1987. "Age Structure and Crime: Sym- metry versus Asymmetry and the Projection of Crime Rates through the 1990s." American Sociological Review 52 (April): 170-83.

Crane, Jonathan. 1991. "The Epidemic Theory of Ghettos and Neighborhood Effects of Dropping Out and Teenage Childbearing." American Journal of Sociology 96 (March): 1126-59.

Darroch, A. Gordon, and Wilfred G. Marston. 1984. "Patterns of Urban Ethnicity." Pp. 127-59 in Urbanism and Urbanization, edited by Noel Iverson. Leiden: E. J. Brill.

Deng, Zhong, and Phillip Bonacich. 1990. "Some Effects of Urbanism on Black Social Networks." Social Networks 13:35-50. Dewey, R. 1960. "The Rural-Urban Continuum: Real but Relatively Unimportant." American Journal of Sociology 66 (July): 60-66.

Diederiks, Herman. 1993. "Urban and Rural Criminal Justice and Criminality in the Netherlands since the Late Middle Ages: Some Observations." Chap. 7 in manuscript, "Violent Crime in Town and Country since the Middle Ages," edited by Eric Johnson and Eric Monkkonen. University of California, Los Angeles.

Durkheim, mile. (1893) 1933. The Division of Labor in Society, translated by G. Simpson. Glencoe, Ill. : Free Press. . (1897) 1951. Suicide, translated by G. Simpson. Glencoe, Ill.: Free Press.

Egmond, Florike. 1993. "Between Town and Countryside: Organized Crime in the Dutch Republic." Chap. 6 in manuscript, "Violent Crime in Town and Country since the Middle Ages," edited by Eric Johnson and Eric Monkkonen. University of California, Los Angeles.

Ellison, Christopher G. 1991. "An Eye for an Eye? A Note on the Southern Subcul- ture of Violence." Social Forces 69 (June): 1223-39. . 1993. "Religious Involvement and Self-Perception among Black Americans." Social Forces 71 (June): 1027-55. Ellison, Christopher G., and Bruce London. 1992. "The Social and Political Partici- pation of Black Americans." Social Forces 70 (March): 681-701. Ellison, Christopher G., and Marc A. Musick. 1993. "Southern Intolerance: A Funda- mentalist Effect?" Social Forces 72 (December): 379-98.

Felson, Richard B., Allen E. Liska, Scott J. South, and Thomas L. McNulty. 1994. "The Subculture of Violence and Delinquency: Individual vs. School Effects." Social Forces 73 (September): 155-73.

Finke, Roger. 1990. "Religious Deregulation: Origins and Consequences." Journal of Church and State 32 (Summer): 609-26.

Finke, Roger, and Rodney Stark. 1988. "Religious Economies and Sacred Canopies: Religious Mobilization in American Cities." American Sociological Review 53 (Feb- ruary): 41-49.

. 1992. The Churching of America, 1776-1990: Winners and Losers in Our Religious Economy. New Brunswick, N.J.: Rutgers University Press. Fischer, Claude S. 1972. "'Urbanism as a Way of Life"': A Review and an Agenda." Sociological Methods and Research 1 (November): 187-242. . 1975a. "Toward a Subcultural Theory of Urbanism." American Journal of Sociology 80 (May): 1319-41. , 19756. "The Study of Urban Community and Personality." Annual Review of Sociology 1:67-89.

. 1976. The Urban Experience, 1st ed. San Diego: Harcourt Brace Jovanovich. . 1978a. "On the Marxian Challenge to Urban Sociology." Comparative Urban Research 6 (2-3): 10-19. . 1978b. "Urban-to-Rural Diffusion of Opinions in Contemporary America." American Journal of Sociology 84 (July): 151-59. . 1980. "The Spread of Violent Crime from City to Countryside, 1955 to 1975." Rural Sociology 45 (Fall): 416-34. . 1982. To Dwell among Friends: Personal Networks in Town and City. Chicago: University of Chicago Press. . 1984. The Urban Experience, 2d ed. San Diego, Calif.: Harcourt Brace Jova-

novich. Flanagan, W. 1990. Urban Sociology. Boston: Allyn & Bacon. Frey, William H. 1987. "Migration and Depopulation of the Countryside." American

Sociological Review 52 (April): 240-57.

Furstenberg, Frank F., Jr., S. Philip Morgan, Kristin A. Moore, and James L. Pe- terson. 1987. "Race Differences in the Timing of Adolescent Intercourse." American Sociological Review 52 (August): 511-18.

Gans, Herbert. 1962. "Urbanism and Suburbanism as Ways of Life: A Reevaluation of Definitions." Pp. 625-48 in Human Behavior and Social Processes, edited by

A. M. Rose. Boston: Houghton-Mifflin. . 1992. "Sociological Amnesia: The Noncumulation of Normal Social Science." Sociological Forum 7 (4): 701-10. Gibbs, Jack R., and Maynard L. Erickson. 1975. "Major Developments in the Socio- logical Study of Deviance." Annual Review of Sociology 1:21-42. Gillis, A. R. 1989. "Crime and State Surveillance in Nineteenth-Century France." American Journal of Sociology 95 (September): 307-41. Glazer, Nathan, and Daniel P. Moynihan. 1970. Beyond the Melting Pot, rev. ed.

Cambridge, Mass.: MIT Press. Goldenberg, Stephen. 1987. Thinking Sociologically. Belmont, Calif.: Wadsworth. Goldscheider, Calvin. 1986. Jewish Continuity and Change. Bloomington: Indiana

University Press. Gottdiener, Mark. 1994. The New Urban Sociology. New York: McGraw-Hill. Gottdiener, Mark, and Joe R. Feagin. 1988. "The Paradigm Shift in Urban Sociol-

ogy." Urban Affairs Quarterly 24 (December): 163-87.

Harer, Miles D., and Darrell Steffensmeier. 1992. "The Differing Effects of Economic Inequality and Black and White Rates of Violence." Social Forces 70 (June): 1035-54.

Harley, Brian. 1993. "Urbanism and Religiosity, 1972-1991: How Childhood, Cur- rent, and Constant Residence Predict Church Attendance." Paper delivered to the American Sociological Association Meetings, Miami.

Hauser, Philip H. 1965. "Observations on the Urban-Folk and Urban-Rural Dichoto- mies as Forms of Western Ethnocentrism." Pp. 503-17 in The Study of Urbaniza- tion, edited by P. H. Hauser and Leo F. Schnore. New York: John Wiley.

Hauser, Robert M. 1970. "Context and Consex: A Cautionary Tale." American Jour- nal of Sociology 75 (January): 645-64. Hirabayashi, Lane Ryo. 1986. "The Migrant Village Association in Latin America." Latin American Research Review 21 (3): 7-29. Hirschl, Thomas A,, and Mark R. Rank. 1991. "The Effect of Population Density on Welfare Participation." Social Forces 70 (September): 225-35. Hirschman, Charles. 1983. "America's Melting Pot Reconsidered." Annual Review of Sociology 9: 397-423. Horowitz, D. L. 1977. "Cultural Movements and Ethnic Change." Annals of the American Academy 433:6-18. Huff-Corzine, Lin, Jay Corzine, and David C. Moore. 1986. "Southern Exposure:

Deciphering the South's Influence on Homicide Rates." Social Forces 64 (June): 906-24.

Huff-Corzine, Lin, Jay Corzine, and David C. Moore. 1991. "Deadly Connections: Culture, Poverty, and the Direction of Lethal Violence." Social Forces 69 (March): 715-32.

Hwang, Sean-Shong, and Steve H. Murdock. 1988. "Residential Segregation and Ethnic Identification among Hispanics in Texas." Urban Affairs Quarterly 23 (March): 329-45.

Iannaccone, Laurence R. 1991. "The Consequences of Religious Market Regulation." Rationality and Society 3 (April): 156-77. Irwin, Don. 1987. "Half of Crime Victims Knew Perpetrators." Los Angeles Times, January 19. Jackson, Pamela Irving. 1984. "Opportunity and Crime: A Function of City Size." Social Science Research 68 (2): 172-93.

Jang, Sung Joon, and Richard D. Alba. 1992. "Urbanism and Nontraditional Opin- ion: A Test of Fischer's Subcultural Theory." Social Science Quarterly 73 (Septem- ber): 596-609.

Jankowski, Martin Sanchez. 1985. City Bound. Albuquerque: University of New Mexico Press.

Johnson, Eric A. 1992. "Cities Don't Cause Crime: Urban-Rural Differences in Late Nineteenth- and Early Twentieth-Century German Criminality." Social Science History 16 (Spring): 129-76.

Johnson, Eric, and Eric Monkkonen. 1993. Introduction to the manuscript, "Violent Crime in Town and Country since the Middle Ages," edited by E. Johnson and

E. Monkkonen. University of California, Los Angeles. Kalmijn, Matthijs. 1991. "Shifting Boundaries: Trends in Religious and Educational Homogamy." American Sociological Review 56 (December): 786-800. Karnig, A. A. 1979. "Black Economic, Political and Cultural Development: Does City Size Make a Difference?" Social Forces 57: 1194-1211. Kennedy, Leslie W., Robert A. Silverman, and David R. Forde. 1991. "Homicide in Urban Canada." Canadian Journal of Sociology 16 (4): 397-410.

Kowalski, Gregory S., and Don Duffield. 1990. "The Impact of the Rural Population Component on Homicide Rates in the United States: A County-Level Analysis." Rural Sociology 55 (Spring): 76-90.

Kowalski, Gregory S., Charles E. Faupel, and Paul D. Starr. 1987. "Urbanism and Suicide: A Study of American Counties." Social Forces 66 (Fall): 85-101.

Ku, Leighton, Freya L. Sonenstein, and Joseph H. Pleck. 1993. "Neighborhood, Family, and Work: Influences on Premarital Behaviors of Adolescent Males." Social Forces 72 (December): 479-503.

LaGory, Mark, and J. Pipkin. 1981. Urban Social Space. Belmont, Calif.: Wads- worth.

Land, Kenneth C., Patricia L. McCall, and Lawrence E. Cohen. 1990. "Structural Covariates of Homicide Rates: Are There Any Invariances across Time and Social Space?" American Journal of Sociology 95 (January): 922-63.

Land, Kenneth C., Glenn Deane, and Judith Blau. 1991. "Religious Pluralism and Church Membership: A Spatial Diffusion Model." American Sociological Review 56 (April): 237-49.

Laumann, Edward O., John H. Gagnon, Robert T. Michael, and Stuart Michaels. 1994. The Social Organization of Sexuality: Sexual Practices in the United States. Chicago: University of Chicago Press.

Lewis, Glyn, Anthony David, Sven Andrasson, and Peter Allebeck. 1992. "Schizo- phrenia and Mental Life." Lancet 340 (18 July): 137-40. Lieberson, Stanley. 1985. Making It Count: The Improvement of Social Research and Theory. Berkeley and Los Angeles: University of California Press.

Loftin, Colin, and Robert Nash Parker. 1984. "An Errors-in-Variable Model of the Effect of Poverty on Urban Homicide Rates." Paper Presented to the American Sociological Association, San Antonio.

Marcus, George E., James Piereson, and John Sullivan. 1980. "Rural-Urban Differ- ences in Tolerance: Confounding Problems of Conceptualization and Measure- ment." Rural Sociology 45 (4): 731-37.

Matsueda, Ross L., Rosemary Gartner, Irving Piliavin, and Michael Polakowski. 1991. "The Prestige of Criminal and Conventional Occupations: A Subcultural Model of Criminal Activity." American Sociological Review 57 (December): 752-70.

McAdam, Doug. 1982. Political Process and the Development of Black Insurgency, 1930-1970. Chicago: University of Chicago Press.

McCarthy, Bill, and John Hagan. In press. "Getting into Street Crime: The Structure and Process of Criminal Embeddedness." Paper presented to the American Socio- logical Association, Los Angeles.

McCloskey, Herbert, and John Zaller. 1984. The American Ethos. Cambridge, Mass.: Harvard University Press. Mellor, J. Rosemary. 1977. Urban Sociology in an Urbanized Society. London: Routledge & Kegan Paul.

Messner, Steven F., and Robert J. Sampson. 1991. "The Sex Ratio, Family Disrup- tion, and Rates of Family Crime: The Paradox of Demographic Structure." Social Forces 69 (March): 693-713.

Meyerowitz, Joanne J. 1988. Women Adrqt: Independent Wage Earners in Chicago, 1880-1930. Chicago: University of Chicago Press. Michelson, William. 1977. Review of The Urban Experience in American Journal of Sociology 83 (November): 803-6.

Miethe, Terance D., Michael Hughes, and David McDowall. 1991. "Social Change and Crime Rates: An Evaluation of Alternative Theoretical Approaches." Social Forces 70 (September 1991): 165-85.

Miller, Zane. 1992. "Pluralism, Chicago School Style: Louis Wirth, the Ghetto, the City, and 'Integration.' " Journal of Urban History 18 (May): 25 1-79.

Modell, John, and Lynn H. Lees. 1981. "The Irish Countryman Urbanized." Pp. 311-42 in Philadelphia: Work, Space, Family, and Group Experience in the Nine- teenth Century, edited by Theodore Hershberg. New York: Oxford University Press.

Morawska, Eva. 1994. "In Defense of the Assimilation Model." Journal of American

Ethnic History 13 (Winter): 75-87. Morris, R. N. 1968. Urban Sociology. New York: Praeger. Nash, Gary B. 1979. The Urban Crucible: Social Change, Political Consciousness,

and the Origins of the American Revolution. Cambridge, Mass: Harvard University Press. . 1987. "The Social Evolution of Preindustrial American Cities, 1700-1820: Reflections and New Directions." Journal of Urban History 13 (February): 115-45. O'Brien, R. M. 1983. "Metropolitan Structure and Violent Crime: Which Measure of Crime?" American Sociological Review 48 (June): 434-37. Olzak, Susan, and E. West. 1991. "Ethnic Conflict and the Rise and Fall of Ethnic Newspapers." American Sociological Review 56 (August): 458-74.

Osterberg, Eva. 1992. "Criminality, Social Control, and the Early Modern State: Evidence and Interpretations in Scandinavian History." Social Science History 16 (Spring): 67-98.

Parker, Robert Nash. 1989. "Poverty, Subculture of Violence, and Type of Homi- cide." Social Forces 67 (June): 983-1007.

Peil, Margaret, with Pius 0. Sada. 1984. African Urban Society. New York: Wiley.

Pescosolido, Bernice. 1990. "The Social Context of Religious Integration and Suicide: Pursuing the Network Explanation." Sociological Quarterly 3 1 (3): 337-5 7.

Petersen, Larry R., and K. Peter Takayama. 1984. "Community and Commitment among Catholics." Sociological Quarterly 25 (Winter): 97-112. Peterson, Ruth D., and Lauren J. Krivo. 1993. "Racial Segregation and Black Urban Homicide." Social Forces 71 (June): 1001-26.

Rabinowitz, Jonathan, Israel Kim, and Bernard Lazerwitz. 1992. "Metropolitan Size and Participation on Religio-Ethnic Communities." Journal for the Scienti$c Study of Religion 31 (3): 339-45.

Ragin, Charles C., and Howard S. Becker, eds. 1992. What Is a Case? Exploring the Foundations of Social Inquiry. New York: Cambridge University Press.

Rank, Mark R., and Thomas A. Hirschl. 1988. "A Rural-Urban Comparison of Welfare Exits: The Importance of Population Density." Rural Sociology 53 (2): 190-206.

. 1993. "The Link between Population Density and Welfare Participation." Demography 30 (November): 607-22. Reiss, Albert J. 1955. "An Analysis of Urban Phenomena." Pp. 41-51 in The Metrop- olis in Modern L$e, edited by R. M. Fisher. New York: Doubleday.

Rogers, Susan M., and Charles F. Turner. 1991. "Male-Male Sexual Contact in the USA: Findings from Five Sample Surveys, 1970-1990." Journal of Sex Research 28 (November):491-519.

Romans-Clarkson, Sarah E., Valerie A. Walton, G. Peter Herbison, and Paul E. Mullen. 1990. "Psychiatric Morbidity among Women in Urban and Rural New Zealand." British Journal of Psychiatry 156 (January): 84-91.

Rosenbaum, James E., and Susan J. Popkin. 1991. "Employment and Earnings of Low-Income Blacks Who Move to Middle-Class Suburbs." Pp. 342-56 in The Urban Underclass, edited by Christopher Jencks and Paul E. Peterson. Washington, D.C.: Brookings Institution.

Salerno, Roger A. 1988. "Fifty Years Later: An Historical and Contextual Appraisal of Wirth's Urbanism Thesis." Paper Presented to the American Sociological Associ- ation, Atlanta.

Sampson, Robert J. 1983. "Structural Density and Criminal Victimization." Criminology 21 (May): 276-93. . 1985a. "Race and Criminal Violence: A Demographically Disaggregated Analysis of Urban Homicide." Crime and Delinquency 31 (January): 47-82. , 19856. "Structural Sources of Variation in Race-Age-Specific Rates of Offending across Major U.S. Cities." Criminology 23 (November): 647-73.

. 1986a. "The Effects of Urbanization and Neighborhood Characteristics on Criminal Victimization." Pp. 3-25 in Metropolitan Crime Patterns, edited by Rob- ert Figlio, Simon Hakim, and George Rengert. Monsey, N.Y.: Criminal Justice Press.

. 1986b. "Neighborhood Family Structure and the Risk of Personal Victimiza- tion." Pp. 25-46 in The Social Ecology of Crime, edited by James M. Byrne and Robert J. Sampson. New York: Springer-Verlag.

. 1987. "Urban Black Violence: The Effect of Male Joblessness and Family Disruption." American Journal of Sociology 93 (September): 348-82.

Sampson, Robert J., and W. Byron Groves. 1989. "Community Structure and Crime: Testing Social-Disorganization Theory." American Journal of Sociology 94 (January): 774-802.

Sauders, Peter. 1985. "Space, the City, and Urban Sociology." In Social Relations and Spatial Structures, edited by Derek Gregory and John Urry. London: Mac- millan.

Schwab, William A. 1992. The Sociology of Cities. New York: Prentice-Hall.

Shelley, Louise I. 1981. Crime and Modernization. Carbondale: Southern Illinois University Press.

Simmel, Georg. (1905) 1969. "The Metropolis and Mental Life." Pp. 47-60 in Classic Essays on the Culture of Cities, edited by Richard Sennett. New York: Appleton- Century-Crofts.

Simpson, Miles E., and George H. Conklin. 1989. "Socioeconomic Development, Suicide and Religion: A Test of Durkheim's Theory of Religion and Suicide." Social Forces 67 (June): 945-64.

Smith, Douglas A., and G. Roger Jarjoura. 1989. "Household Characteristics, Neigh- borhood Composition and Victimization Risk." Social Forces 68 (December): 62 1-40.

Spates, J. L., and J. J. Macionis. 1987. The Sociology of Cities, 2d ed. Belmont, Calif.: Wadsworth.

Stinner, William M., Mollie Van Loon, Seh-Woon Chung, and Yongchan Byun. 1990. "Community Size, Individual Social Position, and Community Attachment." Rural Sociology 55 (4): 494-521.

Swidler, Ann. 1986. "Culture in Action: Symbols and Strategies." American Sociolog- ical Review 5 1:273-86.

Takei, N., E. O'Callaghan, P. Sham, G. Glover, and R. M. Murray. 1992. "Winter Birth Excess in Schizophrenia: Its Relationship to Place of Birth." Schizophrenia Research 6:102.

Tittle, Charles R. 1989a. "Urbanness and Unconventional Behavior: A Partial Test of Claude Fischer's Subcultural Theory." Criminology 27 (May): 273-306. . 1989b. "Influences of Urbanism: A Test of Predictions from Three Perspec- tives." Social Problems 36 (June): 270-88. Tittle, Charles R., and Mark C. Stafford. 1992. "Urban Theory, Urbanism, and Suburban Residence." Social Forces 70 (March): 725-44.

Tomaskovic-Devey, Barbara, and Donald Tomaskovic-Devey. 1988. "The Social Structural Determinants of Ethnic Group Behavior: Single Ancestry Rates among Four White American Ethnic Groups." American Sociological Review 53 (August): 650-59.

Tuch, Steven. 1987. "Urbanism, Region, and Tolerance Revisited: The Case of Racial Prejudice." American Sociological Review 52 (August): 504-10.

Van Poppel, Frans. 1989. "Urban-Rural versus Regional Differences in Demographic Behavior: The Netherlands, 1850-1960." Journal of Urban History 15 (August): 363-98.

Venturelli, Peter J. 1982. "Institutions in an Ethnic District." Human Organization 41 (1): 26-35. Wallace, John, Jr., and Jerald G. Backman. 1991. "Explaining RacialIEthnic Differ- ences in Adolescent Drug Use." Social Problems 38 (August): 333-57. Warr, Mark. 1993. "Parents, Peers, and Delinquency." Social Forces 72 (September): 247-64. Ward, Sally K. 1978. Review of The Urban Experience, 1st ed. Contemporary Sociol- ogy 7 (January): 104-05.

Warner, R. Stephen. 1993. "Work in Progress Toward a New Paradigm for the Sociological Study of Religion in the United States." American Journal of Sociology 98 (March): 1044-93.

Waters, Mary. 1990. Ethnic Options. Berkeley and Los Angeles: University of Califor- nia Press.

Webber, Melvin. 1963. "The Urban Place and the Nonplace Urban Realm." Pp. 79-153 in Explorations into Urban Structure, edited by M. M. Webber et al. Philadelphia: University of Pennsylvania Press.

Webber, Melvin. 1970. "Order in Diversity: Community without Propinquity." Pp. 792-811 in Neighborhood, City, Metropolis, edited by R. Gutman and D. Popenoe. New York: Random House.

Weber, Max. (1963) 1978. "The Sociology of Religion." 1:399-634 in Max Weber: Economy and Society, edited by Guenther Roth and Claus Wittich. Berkeley: Uni- versity of California Press.

Wellman, Barry. 1979. "The Community Question." American Journal of Sociology 84 (March): 1201-31. Wilkinson, Kenneth P. 1984. "A Research Note on Homicide and Rurality." Social Forces 63 (December): 445-52. Williams, Kirk R. 1984. "Economic Sources of Homicide: Reestimating the Effects of Poverty and Inequality." American Sociological Review 49 (April): 283-89.

Williams, Kirk R., and Robert L. Flewelling. 1988. "The Social Production of Crimi- nal Homicide: A Comparative Study of Disaggregated Rates in American Cities." American Sociological Review 53 (June): 421-31.

Wilson, Thomas C. 1985a. "Urbanism and Tolerance: A Test of Some Hypotheses Drawn from Wirth and Stouffer." American Sociological Review 50 (February): 117-23.

. 19856. "Urbanism, Misanthropy and Subcultural Processes." Social Science Journal 22 ( July): 89-101. . 1986. "Community Population Size and Social Heterogeneity: An Empirical Test." American Journal of Sociology 91 (March): 1154-70. . 1991. "Urbanism, Migration, and Tolerance: A Reassessment." American Sociological Review 56 (February): 117-23. . 1992. "Urbanism and Nontraditional Opinion: Another Look at the Data." Social Science Quarterly 73:610-12. Wirth, Louis. 1938. "Urbanism as a Way of Life." American Journal of Sociology 44 (July): 3-24.

. (1956) 1969. "Rural-Urban Differences." Pp. 165-70 in Classic Essays on the Culture of Cities, edited by Richard Sennett. New York: Appleton-Century-Crofts. Wolfgang, Marvin, and F. Ferracuti. 1967. The Subculture of Violence. London: Tavistock.

Yinger, Milton J. 1985. "Ethnicity." Annual Review of Sociology 11:151-80.

Zorbaugh, H. W. 1929. The Gold Coast and the Slum. Chicago: University of Chicago Press.

Zunz, Olivier. 1982. The Changing Face of Inequality. Chicago: University of Chicago Press.

Comments
  • Recommend Us