CAUSAL INFERENCE MODELS
X b c Y e
a f d
X1 X2 Y1 Y2
g
W
Figure 5. Simple Measurement-Error Model, with Two Unmeasured Variables X and Y, Two Indicators of Each, and
Source of Bias W
Here, k = 3, so that at least two variables must guilt. Perhaps school performance can be intro-
be left out of each equation, meaning that their duced as Z3 by making the assumption that Z3
respective coefficients have been set equal to zero. directly affects both self-esteem and delinquency
One can rather simply check on the necessary but not guilt level. A check of this revised model
condition by counting arrowheads coming to each indicates that all equations are properly identified,
variable. In this instance there can be no more and one may proceed to estimation. Although
than two arrows into each variable, whereas in the space does not permit a discussion of alternative
case of guilt X1 there are three. The equation for estimation methods that enable one to get around
X1 is referred to as being ‘‘underidentified,’’ mean- the violated assumption required by ordinary least
ing that the situation is empirically hopeless. The squares, there are various computer programs
coefficients simply cannot be estimated by any available to accomplish this task. The simplest
empirical means. There are exactly two arrows such alternative, two-stage least squares (2SLS),
coming into delinquency X3, and one refers to this will ordinarily be adequate for nearly all sociologi-
as a situation in which the equation is ‘‘exactly cal applications and turns out to be less sensitive to
identified.’’ With only a single arrow coming into other kinds of specification errors than many of
self-esteem X2, one has an ‘‘overidentified’’ equa- the more sophisticated alternatives that have been
tion for which one actually has an excess of empirical proposed.
information compared to the number of unknowns
to be estimated. It turns out that overidentified CAUSAL APPROACH TO
equations provide criteria for evaluating goodness MEASUREMENT ERRORS
of fit, or a test of the model, in much the same way
that, for recursive models, one obtains an empiri- Finally, brief mention should be made of a grow-
cal test of a null hypothesis for each causal arrow ing body of literature—closely linked to factor
that has been deleted. analysis—that has been developed in order to
attach measurement-error models to structural-
Since the equation for X1 is underidentified, equation approaches that presume perfect meas-
one must either remove one of the arrows, on a urement. The fundamental philosophical starting
priori grounds, or search for at least one more point of such models involves the assumption that
predetermined variable that does not belong in in many if not most instances, measurement errors
this equation, that is, a predetermined variable can be conceived in causal terms. Most often, the
that is assumed not to be a direct cause of level of indicator or measured variables are taken as effects
264
CAUSAL INFERENCE MODELS
of underlying or ‘‘true’’ variables, plus additional If one labels the path coefficients (which all
factors that may produce combinations of random connect measured variables to unmeasured ones)
measurement errors, which are unrelated to all by the simple letters a, b, c, d, and, e, then with 4(3)/
other variables in the theoretical system, and sys- 2 = 6 correlations among the four indicators, there
tematic biases that are explainable in causal terms. will be six pieces of empirical information (equa-
Thus, measures of ‘‘true guilt’’ or ‘‘true self-es- tion system 5) with which to estimate the five
teem’’ will consist of responses, usually to paper- unknown path coefficients.
and-pencil tests, that may be subject to distortions
produced by other variables, including some of rx1x2 = ab rx1y2 = ace (6)
the variables in the causal system. Perhaps distor- ry1y2 = de rx2y1 = bcd
tions in the guilt measure may be a function of rx1y1 = acd rx2y2 = bce
amount of delinquent behavior or parental educa-
tion. Similarly, measures of behaviors are likely to One may now estimate the correlation or path
overestimate or underestimate true frequencies, coefficient c between X and Y by an equation
with biases dependent on qualities of the observer, derived from equation system 6.
inaccuracies in official records, or perhaps the
ability of the actor to evade detection. c2 = (–aa–bb–)c–(2d–dee–) = –rrx–x11–yx–12r–rxx–21–yy2–2 = –rr–xx1–1yx–22–rrx–y21–yy–12 (7)
In all such instances, we may be able to con- Also notice that there is an excess equation that
struct an ‘‘auxiliary measurement theory’’ (Blalock may be used to check on the consistency of the
1968; Costner 1969) that is itself a causal model model with the data, namely the prediction that rx
that contains a mixture of measured and unmeasured 1y 1 rx2y2 = rx1 y2 rx2y1 = abc2de.
variables, the latter of which constitute the ‘‘true’’
or underlying variables of theoretical interest. The Suppose next that there is a source of meas-
existence of such unmeasured variables, however, urement error bias W that is a common cause of
may introduce identification problems by using one of X’s indicators (namely X2) and one of Y’s
more unknowns than can be estimated from one’s (namely Y1). Perhaps these two items have similar
data. If so, the situation will once more be hopeless wordings based on a social survey, whereas the
empirically. But if one has available several indica- remaining two indicators involve very different
tors of each of the imperfectly measured con- kinds of measures. There is now a different expres-
structs, and if one is willing to make a sufficient sion for the correlation between X2 and Y1, namely
number of simplifying assumptions strategically r x2y1 = bcd + fg. If one were to use this particular
placed within the overall model, estimates may be correlation in the estimate of c2, without being
obtainable. aware of the impact of W, one would obtain a
biased estimate. In this instance one would be able
Consider the model of Figure 5 (borrowed to detect this particular kind of departure from
from Costner 1969), which contains only two theo- randomness because the consistency criterion
retical variables of interest, namely, the unmeasured would no longer be met. That is, (acd)(bce) ≠ (ace)(bcd
variables X and Y. Suppose one has two indicators + fg). Had W been a common cause of the two
each for both X and Y and that one is willing to indicators of either X or Y alone, however, it can be
make the simplifying assumption that X does not seen that one would have been unable to detect
affect either of Y’s indicators, Y1 and Y2, and that Y the bias even though it would have been present.
does not affect either of X’s indicators, X1 and X2.
For the time being ignore the variable W as well as Obviously, most of one’s measurement-error
the two dashed arrows drawn from it to the indica- models will be far more complex than this, with
tors X2 and Y1. Without W, the nonexistence of several (usually unmeasured) sources of bias, pos-
other arrows implies that the remaining causes of sible nonlinearities, and linkages between some of
the four indicators are assumed to be uncorrelated the important variables and indicators of other
with all other variables in the system, so that one variables in the substantive theory. Also, some
may assume measurement errors to be strict- indicators may be taken as causes of the conceptual
ly random. variables, as for example often occurs when one is
265
CAUSAL INFERENCE MODELS
attempting to get at experience variables (e.g., mistake to throw out portions of one’s theory
exposure to discrimination) by using simple objec- merely because data to test it are not currently
tive indicators such as race, sex, or age. Further- available. Indeed, without a theory as to how miss-
more, one’s substantive models may involve feed- ing variables are assumed to operate, it will be
back relationships so that simultaneous equation impossible to justify one’s assumptions regarding
systems must be joined to one’s measurement- the behavior of disturbance terms that will contain
error models. such variables, whether explicitly recognized or
not. Causal modeling may thus be an important
In all such instances, there will undoubtedly tool for guiding future research and for providing
be numerous specification errors in one’s models, guidelines as to what kinds of neglected variables
so that it becomes necessary to evaluate alternative need to be measured.
models in terms of their goodness of fit to the data.
Simple path-analytic methods, although heuristi- (SEE ALSO: Correlation and Regression Analysis; Episte-
cally helpful, will no longer be adequate. Fortu- mology; Multiple Indicator Models; Scientific Explana-
nately, there are several highly sophisticated com- tion; Tabular Analysis)
puter programs, such as LISREL, that enable social
scientists to carry out sophisticated data analyses REFERENCES
designed to evaluate these more complex models
and to estimate their parameters once it has been Allison, Paul D. 1995 ‘‘Exact Variance of Indirect Effects
decided that the fit to reality is reasonably close. in Recursive Linear Models.’’ Sociological Methodology
(See Joreskog and Sorbom 1981; Long 1983; and 25:253–266.
Herting 1985.)
Arminger, Gehard 1995 ‘‘Specification and Estimation
In closing, what needs to be stressed is that of Mean Structure: Regression Models.’’ In G. Arminger,
causal modeling tools are highly flexible. They C. C. Clogg, and M. E. Sobel, eds., Handbook of
may be modified to handle additional complica- Statistical Modeling for the Social and Behavioral Sci-
tions such as interactions and nonlinearities. Caus- ences. New York: Plenum Press.
al modeling in terms of attribute data has been
given a firm theoretical underpinning by Suppes Blalock, Hubert M. 1968 ‘‘The Measurement Problem:
(1970), and even ordinal data may be used in an A Gap Between the Languages of Theory and Re-
exploratory fashion, provided that one is willing to search.’’ In H. M. Blalock and A. B. Blalock, eds.,
assume that dichotomization or categorization has Methodology in Social Research. New York: McGraw-Hill.
not introduced substantial measurement errors
that cannot be modeled. ———1985 ‘‘Inadvertent Manipulations of Dependent
Variables in Research Designs.’’ In H. M. Blalock,
Like all other approaches, however, causal ed., Causal Models in Panel and Experimental Designs.
modeling is heavily dependent on the assumptions New York: Aldine.
one is willing to make. Such assumptions need to
be made as explicit as possible—a procedure that Blau, Peter M., and Otis Dudley Duncan 1967 The
is unfortunately often not taken sufficiently seri- American Occupational Structure. New York: Wiley.
ously in the empirical literature. In short, this set
of tools, properly used, has been designed to Blossfeld, Hans-Peter, and Gotz Rohwer 1997 ‘‘Causal
provide precise meaning to the assertion that nei- Inference, Time and Observation Plans in the Social
ther theory nor data can stand alone and that any Sciences.’’ Quality and Quantity 31:361–384.
interpretations of research findings one wishes to
provide must inevitably also be based on a set of Bollen, Kenneth A. 1989 Structural Equations with Latent
assumptions, many of which cannot be tested with Variables. New York: Wiley.
the data in hand.
Costner, Herbert L. 1969 ‘‘Theory, Deduction, and
Finally, it should be stressed that causal mod- Rules of Correspondence.’’ American Journal of Soci-
eling can be very useful in the process of theory ology 75:245–263.
construction, even in instances where many of the
variables contained in the model will remain Herting, Jerald R. 1985 ‘‘Multiple Indicator Models
unmeasured in any given study. It is certainly a Using LISREL.’’ In H. M. Blalock, ed., Causal Models
in the Social Sciences, 2nd ed. New York: Aldine.
Joreskog, Karl G., and Dag Sorbom 1981 LISREL V:
Analysis of Linear Structural Relationships by the Method
of Maximum Likelihood; User’s Guide. Uppsala, Swe-
den: University of Uppsala Press.
266
CENSORSHIP AND THE REGULATION OF EXPRESSION
Long, J. Scott 1983 Confirmatory Factor Analysis: A Preface EARLY SYSTEMS OF GOVERNMENT
to LISREL. Beverly Hills, Calif.: Sage.
CENSORSHIP
Pearl, Judea 1998 ‘‘Graphs, Causality, and Structural
Equation Models.’’ Sociological Methods and Research Traditional conceptions of censorship are rooted
27:226–284. in rigid systems of ecclesiastic and governmental
control over discourse and printing. The term
Sobel, Michael E. 1995 ‘‘Causal Inference in the Social itself derives from the office of the census in early
and Behavioral Sciences.’’ In G. Arminger, C. C. Rome, where the censor served as both census
Clogg, and M. E. Sobel, eds., Handbook of Statistical taker and as supervisor of public conduct and
Modeling for the Social and Behavioral Sciences. New morals. Before the advent of the printing press in
York: Plenum Press. the fifteenth century, most manuscripts in Europe
were produced in monasteries, which controlled
Sobel, Michael E. 1996 ‘‘An Introduction to Causal their production. Centralized systems of control
Inference.’’ Sociological Methods and Research 24:353–379. over books developed largely in response to the
invention of the printing press, which both church
Spirtes, Peter, Thomas Richardson, Christopher Meek, and state perceived as threats to their authority. In
Richard Scheines, and Clark Glymour 1998 ‘‘Using the mid-sixteenth century the Catholic Church
Path Diagrams as a Structural Equation Modeling issued the Index of Forbidden Books, which was
Tool.’’ Sociological Methods and Research 27:182–225. enforced through compliance of the faithful, pre-
publication screening of books, the burning of
Strotz, Robert H., and Herman O. A. Wold 1960 ‘‘Re- heretical tracts, and the persecution of heretics. In
cursive Versus Nonrecursive Systems.’’ Econometrica Protestant countries, the State generally assumed
28:417–427. control over the publication of books. The English
monarchy published its first list of prohibited
Suppes, Patrick 1970 A Probabilistic Theory of Causality. books in 1529 and exercised its control through a
Amsterdam: North-Holland. contractual arrangement with the Stationers’ Com-
pany, which, in 1557, was granted a monopoly on
Von Eye, Alexander, and Clifford C. Clogg (eds.) 1994 the production and distribution of printed materi-
Latent Variables Analysis: Applications for Developmen- als. This charter remained in effect until 1694
tal Research. Thousand Oaks, Calif.: Sage Publications. when it was allowed to expire, primarily because of
difficulties in its administration. These centralized
Wright, Sewall 1934 ‘‘The Method of Path Coefficients.’’ mechanisms of control were replaced by less sys-
Annals of Mathematical Statistics 5:161–215. tematic methods, such as laws against seditious
libel, through which speech that merely criticized
———1960 ‘‘Path Coefficients and Path Regressions: government policies could be punished.
Alternative or Complementary Concepts?’’ Biometrics
16:189–202. Although systems similar to these proliferated
worldwide and continue to exist in modern-day
HUBERT M. BLALOCK, JR. authoritarian regimes, they are consensually viewed
as incompatible with the practice of democracy.
CENSORSHIP AND THE The history of free speech principles in the West
REGULATION OF coincides with the rise of democratic thought, as
EXPRESSION expressed in the writings of the eighteenth-centu-
ry Enlightenment philosophers in France and in
Modern discussions of censorship center on the the influential political philosophies of John Locke,
legitimacy of the regulatory structures and actions John Milton, and John Stuart Mill in England. The
through which expression and communication turn of mind that gripped Europe during this
are governed, and the extent to which these struc- period is reflected in Locke’s dictum that govern-
tures meet the requirements of democratic socie- ments are the servants of the people, not the
ties. In this entry, we first survey prominent his- reverse, and in the insistence by each of these
torical examples of centralized censorship systems philosophers that self-governance cannot function
in the West. This history provides a context for a under regimes where the circulation of ideas is
discussion of modern conceptions of censorship dependent upon the whims of rulers. In modern
and issues associated with the term itself. We then
turn to legal structures regulating speech and
press in the United States, to government control
of political speech, and to some modern-day con-
troversies over the regulation of expression.
267
CENSORSHIP AND THE REGULATION OF EXPRESSION
political theory, Habermas (1989) considers the refer to nongovernmental restrictions on speech.
demise of systematic statecensorship to be a pre- For example, in one of the few instances in which
condition of the rise of a ‘‘public sphere,’’ an the Supreme Court has applied the term to private
admittedly idealized realm of discourse—indepen- concerns (Red Lion v. FCC, 395 U.S. 367, 1969), the
dent of both state and market—in which public Court stated that ‘‘The First Amendment does not
issues can be deliberated in an environment where protect private censorship by broadcasters who
reason, not the status of speakers is honored. are licensed by the Government to use a scarce
resource which is denied to others.’’
THE RHETORIC OF ‘‘CENSORSHIP’’
In each of these three descriptive domains the
The use of language plays a critical role in framing term ‘‘censorship’’ almost invariably carries its
thought and discussion about the legitimate con- pejorative inflection. In its broadest sense, censor-
trol of speech. The term ‘‘censorship’’ most fre- ship signifies control over the means of expres-
quently arises in debates over the desirability of sion; the determination of what content will not be
restricting access to one or another form of com- communicated. This sense of the term is found in
municative content. However, the term is fraught the work of Bourdieu (1991), for whom censor-
with the complexities of multiple meanings, uses, ship is located not only in explicit prohibitions, but
and understandings. These complexities arise from also in the everyday practices and power relation-
the historical baggage it carries, the multiple con- ships that determine what is and is not said. It
texts—popular, legal, and scholarly—in which it is includes not only the sixteenth-century Inquisitor
used, and the highly contested nature of the de- who decides which books will be burned, but also
bates in which it is invoked. Although systematic the twentieth-century film editor who leaves a
empirical analyses of the term have yet to be scene on the cutting room floor. It includes both
conducted, its meanings appear to differ along the intentional actions of individuals and, as im-
two dimensions, connotative and descriptive. Un- portantly, the de facto results of impersonal forces
derlying the connotative force of the term is the that lead some ideas not to be expressed. This
strong conviction that suppression of speech is at expansive meaning, which has yet to make signifi-
best a necessary evil. As such, the term typically cant inroads into popular discourse, is incompat-
carries with it a highly pejorative connotation and ible with the pejorative connotation found in eve-
a strong air of illegitimacy. Although First Amend- ryday usage: If censorship is an integral part of
ment scholar Smolla (1991) informs us that ‘‘cen- everyday life, it cannot always, or even typical-
sorship was not always a dirty word,’’ this facet of ly, be evil.
the term is now found in Webster’s Dictionary,
which states that it refers especially to control that The prototypical understanding of ‘‘censor-
is ‘‘exercised repressively.’’ ship’’ is firmly anchored at the point where the
term’s pejorative connotation intersects with gov-
Descriptive uses of the term differ according ernmental restrictions on speech. This can be seen
to the breadth of the domain covered. Under strict in how the term is and is not used in legal and
uses of the term, censorship means the prior re- popular discourse. First, there are numerous forms
straint of information by government. It is this of government restrictions that tend not to be
meaning that enables the Federal Communica- thought of as censorship at all. Examples include
tions Commission (FCC) to enact regulations that punishment for perjury, threats, and libel, and
impose post hoc penalties for some forms of speech prohibitions on misleading advertising. The core
while at the same time declaring that ‘‘nothing in meaning of the term ‘‘censorship’’ tends to be
the Act [that governs the FCC] shall be . . . con- applied to restrictions that are consensually deemed
strued to give the Commission the power of cen- to be illegitimate; it tends not to be applied to
sorship.’’ In a second form of use, censorship restrictions that are consensually deemed to be
refers to any form of government regulations that legitimate, although they fall within the descrip-
restrict or disable speech. Fines imposed by the tive scope of the term. As such, the term typically
FCC for ‘‘indecent’’ speech on radio fit this use, as serves to label those policies and actions that have
the intent is to deter further indecency. In a third, been deemed undesirable rather than to describe
less conventional usage, the term is modified to a set of activities whose legitimacy could then be
judged. A thorough sociological understanding of
268
CENSORSHIP AND THE REGULATION OF EXPRESSION
‘‘censorship’’ would require a mapping of the like Baker (1998) use the term ‘‘structural regula-
shifting boundaries around which this legitimacy tion’’ to refer to government interventions de-
is withheld or conferred. For example, punish- signed to increase viewpoint diversity in the public
ment for blasphemy was once considered legiti- forum. The goal of such rhetorical interventions is
mate, whereas penalties for inciting hatred against to reframe the discussion of ‘‘free speech’’ in a way
ethnic groups may not be thought of as ‘‘censor- that is meticulously sensitive to the threats of
ship’’ in the future. Second, in areas where the government incursions on speech, yet, that at the
legitimacy of a restriction is under dispute, those same time allow this discussion to incorporate
who oppose the restriction tend to label it as both the speech-restricting characteristics of pri-
censorship, whereas those who support the restric- vate action and the speech-expanding possibilities
tion attempt to distance themselves from the term. of government.
This core understanding of the term—as ille- The second consequence of our core under-
gitimate government action—leads to a number of standing of ‘‘censorship’’ is linked primarily to the
consequences, two of which will be mentioned distaste the term evokes. This pejorative connota-
here. First, discussions of the legitimate control of tion renders the term useful in singling out illegiti-
expression are typically framed as battles between mate expressive restrictions, and as a persuasive
censorship, on the one hand, and Free speech, on device in popular and legal debate. However, the
the other: Free speech is the absence of govern- pejorative nature of the term also disables it as a
mental control. This framing of the communica- useful construct in discussions that aim to present
tive needs of a democracy—codified in the First a balanced account of contemporary debates over
Amendment—places a wholesome burden on gov- the control of expression. For example, to frame a
ernments to justify regulatory action. On the other controversy concerning the legitimacy of restric-
hand, this framing excludes from discussion the tions on hate speech as a debate over whether such
positive role that governments can (and do) play in speech should be censored is to ask, in effect,
supporting these needs. It also excludes considera- whether the illegitimate suppression of hate speech
tion of the instances in which private concerns and is or is not legitimate. This framing, though a
impersonal market forces can produce deleterious popular one, clearly biases the discussion toward
effects; ones that exclude particular viewpoints one side of the debate.
from the marketplace of ideas. The restrictions
placed by health maintenance organizations (HMOs) In this entry we attempt to use less loaded
upon what physicians can say to their patients (i.e., terms to describe the concerns that the term ‘‘cen-
‘‘gag rules’’) are but one of countless areas in sorship’’ evokes and the underlying controversies
which private institutions ‘‘censor’’ valuable speech. in which the term is typically used. Corresponding-
ly, we strive to reserve the term ‘‘censorship’’ for
Research is needed to uncover the role that those uses in which it is integral to the viewpoint
the rhetoric of ‘‘censorship’’ plays in maintaining being expressed.
the conception of free speech as the absence of
government regulation. It is apparently with the LEGAL PRINCIPLES GOVERNING THE
goal of disabling this entrenched dichotomy that REGULATION OF EXPRESSION IN THE
some theorists have used the term—e.g., ‘‘soft
censorship,’’ ‘‘de facto censorship’’ ‘‘private cen- UNITED STATES
sorship’’—to refer to those private and imperson-
al forces that lead some forms of expression to be The legal principles governing freedom of expres-
systematically excluded from the marketplace of sion in the United States are based largely on
ideas (e.g., Barbur 1996; cf. Post 1998). Barbur, for interpretations of the First Amendment to the
example, argues that ‘‘monopoly is a polite word Constitution, whose ‘‘speech’’ and ‘‘press’’ clauses
for uniformity, which is a polite word for virtual are combined in the statement that ‘‘Congress
censorship—censorship not as a consequence of shall make no law . . . abridging the freedom of
political choices, but as a consequence of ineleastic speech, or of the press.’’ The amendment refers
markets, imperfect competition, and economies only to actions taken by the U.S. Congress, and it
of scale . . .’’ (1996, pp. 137–138). Similarly, theorists was not held to apply to laws made by the individu-
al states until 1925 (Gitlow v. New York, 2688 U.S.
269
CENSORSHIP AND THE REGULATION OF EXPRESSION
652, 1925) when the Court ruled that the four- the First Amendment. Although no theory is uni-
teenth Amendment required that state laws not be versally accepted as dominant, most agree with the
in conflict with federal law. In addition to state and judgment of constitutional scholar Emerson (1973)
constitutional law, other major sources of legal that the value of free expression lies in its ability to
regulation originate in common law (e.g., sedition, promote participation in the decision-making proc-
privacy) and administrative law (e.g., rulemaking ess by all members of society; to advance knowl-
of the FCC). edge and the pursuit of truth; to promote individu-
al self-fulfillment; and, by allowing dissent to be
Despite the apparent clarity of the First Amend- publicly vented, to promote a balance between
ment language, both Congress and the states have stability and change.
crafted many laws that regulate freedoms of speech
and press. These regulations stem primarily from Disputes over the role of government in regu-
three sources. First, the First Amendment is sub- lating speech turn, in part, on which of these
ject to widely varying interpretations. Some consti- values is given precedence. Those who emphasize
tutional scholars conclude that at time the Bill of the role of speech in promoting deliberative de-
Rights was drafted, conceptions of freedom of mocracy often favor an affirmative role of govern-
expression referred only to prior restraints on ment to promote viewpoint diversity or to ensure
speech. Others note ambiguities in the words that the distribution of viewpoints in the public
‘‘abridge’’ and ‘‘speech,’’ and further question the sphere is somewhat reflective of their distribution
intent of the framers of the Constitution. For in the community. In practical terms, this may lead
example, do regulation of defamatory tracts, false to restrictions on the amount of money that corpo-
commercial advertising, perjury, threats, violent rations can contribute to political campaigns, to
obscenity, or nude dancing necessarily count as the enforcement of rights of access to media by
abridgments of speech (Sunstein 1993)? Some citizens, to the strategic placement of ‘‘public fo-
have also argued that the speech clause governs rum’’ spaces on the Internet, to stricter enforce-
only political speech or that the press clause should ment of the antitrust laws as they pertain to media,
not be interpreted to apply to tabloid entertain- or to government subsidies for valued forms of
ment that masquerades as news. speech (Baker 1998). In many instances, the pro-
motion of viewpoint diversity and balance may
Second, the modern world differs from the require the government to regulate or infringe
world of the late eighteenth century. The framers upon the speech or property rights of some in
could not have foreseen the technological advanc- order to advance the needs of deliberative democracy.
es that have brought us telephones, television, and
the Internet. Neither could they have had in mind DIFFERENTIAL JUDICIAL TREATMENT OF
the concentrated corporate ownership or the ad- DIFFERENT FORMS OF EXPRESSION
vertising-driven programming that characterizes
today’s media. Some argue that government regu- The corpus of First Amendment decisions shows
lation is required in order to achieve the principles that all speech is not equal in the protections it is
embodied in the Constitution. afforded. Political discourse—broadly defined to
include speech about social, cultural, and religious
Third, speech rights do not stand alone. For issues—is considered speech of the highest value.
example, the rights of some to speak may conflict As a result, the government must demonstrate a
with the rights of others to be let alone or to be ‘‘compelling’’ interest to warrant its restriction.
treated with dignity and equality. Also, the speech Particularly, political speech cannot be directly
rights asserted by some inevitably conflict with the restricted unless it is ‘‘directed to inciting or pro-
speech rights claimed by others. ducing imminent lawless action and is likely to
incite or produce such action.’’ Advertising was
In light of such issues, the clarity and simplici- long considered outside the purview of First
ty that at first seem to characterize the First Amend- Amendment protections and was subject to strict
ment turn out to be illusory. As a result, scholars regulation. Commercial speech has experienced
have developed theories of the underlying values tremendous leaps in its status under the Court as
that they believe free speech should promote in a
democracy and should govern interpretation of
270
CENSORSHIP AND THE REGULATION OF EXPRESSION
many local restrictions have been declared uncon- property through which television cables are dis-
stitutional. Obscenity lies at the lowest rung of tributed. Telephone companies are classified as
speech and receives no protection under the First ‘‘common carriers,’’ which interdicts their editori-
Amendment. Although the requirements for dem- al control over the information that passes through
onstrating that speech is obscene have become their wires. The newest form of communication,
progressively more demanding, Congress and the the Internet, has so far been granted the highest
states are permitted to ban the public dissemina- rung of protection from government regulation.
tion of speech that meets these criteria (see be-
low). Depending upon how they are classified in The willingness of the Court to allow govern-
terms of value to society, other forms of speech ment regulation of broadcast media lies largely
receive greater or lesser protection. For example, on three rationales: public ownership of the air-
unlicensed medical advice and misleading adver- waves, scarcity of the broadcast spectrum, and
tising can be restricted through a ‘‘balancing test’’ ‘‘pervasiveness’’ of the broadcast signal. First, the
which shows that the harm that stems from sup- airwaves through which broadcast signals are trans-
pressing them are fewer than the harm they cause. mitted are owned by the public and licensed on a
A ‘‘balancing test’’ which shows that the harm that renewable basis to radio and television broadcast-
stems from suppressing them are less than the ers. Second, only a limited number of broadcast
harm they cause. signals can coexist in any given segment of air-
space (scarcity principle). In other words, the broad-
Current First Amendment doctrine also re- cast spectrum is a scarce resource, one that has
flects a second, cross-cutting mode of classification historically required some regulatory body to de-
according to which proposed regulations are cate- cide which of the many interested broadcasters
gorized as content neutral, content based, or view- will be allotted the frequencies that exist. The
point based. Of the three, content neutral regula- management of these tasks is the function of the
tions must pass the lowest constitutional hurdle FCC, which was authorized to regulate broadcast-
whereas viewpoint-based regulations must pass ers in the ‘‘public convenience, interest, or necessi-
the highest. An example of content-neutral regula- ty.’’ In interpreting the ‘‘public-interest’’ clause,
tions might include rules that designate sound- the FCC has issued a number of requirements,
level restrictions on expressive activities, such as including, for example, a modicum of public-inter-
music in Central Park. Examples of viewpoint- est programming such as local news, rights of reply
based restrictions might include a ban on a gay to those attacked in political editorials, and re-
parade or on either Democrat or Republican bill- quirements to air political advertisements. Regula-
boards. Regulations on expression that are con- tions governing the latter are governed by a ‘‘no
tent based, but viewpoint neutral face a high, but censorship’’ clause, whereby the FCC deprives
not insurmountable, constitutional hurdle. For stations of editorial rights over political advertise-
example, in their contractual relations with cable ments. Although most of these regulations are
companies, municipalities have been allowed to aimed to increase viewpoint diversity, broadcast-
require the companies to provide local news and ers often argue that FCC rules infringe upon their
sports, and the federal government has been al- rights to free speech. Finally, the Supreme Court’s
lowed to prohibit all partisan political campaign- judgment that radio and television broadcasts are
ing on army bases. ‘‘pervasive’’ refers to the assumed inability of view-
ers and listeners to fully control their own access,
The application of First Amendment law is or that of their children, to unexpected program
also a function of the technological environment, content. The susceptibility of audience members
or medium, in which expression is conveyed. Ex- to be caught unawares by programming that of-
cept through the sporadic use of antitrust laws, the fends them, or that they deem harmful to their
Court has been least likely to permit regulations of children, has led the court to characterize such
print-based news and most likely to allow regula- broadcasts as an ‘‘uninvited intruder’’ into the
tions of broadcast media. Cable television has privacy of a viewer’s abode. Primarily to provide a
fallen somewhere in between, with regulations ‘‘safe haven,’’ for children, the Court has allowed
reflecting the monopolistic control that compa- time restrictions on the broadcast of ‘‘indecent’’ or
nies exert over cable access to individual house- ‘‘patently offensive’’ programming.
holds and the fact that municipalities own the
271
CENSORSHIP AND THE REGULATION OF EXPRESSION
Given the different levels of protection afford- that employees send over company e-mail systems.
ed by the First Amendment for different forms of An evolving legal controversy surrounds the grow-
expression, the determination of which category ing replacement of public forums—sidewalks,
of speech, which kind of regulation, and which streets, and public parks—by private ones, such as
form of media a given case will be deemed to shopping malls, condominiums, and gated com-
embody, is of critical importance. Levels of protec- munities. Should the political protesters who once
tion for different forms of speech have changed convened in front of the local store be barred from
over time and there is no reason to think that they entry to the shopping mall where the store now
will not continue to do so. Movies were long stands? Should those gathering signatures for po-
considered a form of crass entertainment, outside litical initiatives be excluded from gated communi-
of First Amendment protections, and birth con- ties? These issues pit the property rights of owners
trol information was once classified as obscene. In against the expressive needs of communities. Su-
the not-distant future, the plethora of program- preme Court decisions have given individual states
ming opportunities on the Internet is likely to some leeway in deciding which and when one of
result in Supreme Court challenges of the scarcity these interests is the more compelling. Related
principle upon which public-interest broadcast regu- issues arise in the domain of media, where some
lations are based. Particularly in the area of new point to the growing concentration of media own-
media, the difficulties inherent in determining ership (Bagdikian 1997), the presence of only one
how to categorize a particular speech situation daily newspaper in most U.S. cities, and the
often leads the categorical approach to cede to an ‘‘skewing’’ of media content away from the inter-
approach in which the harms of competing out- ests of the poor (Baker 1998) as evidence that the
comes are less formally ‘‘balanced’’ against each marketplace of ideas is not fully served by unregulated
other. Projects for sociology might include exami- economic markets (Sunstein 1993).
nation of the social underpinnings of the origin of
these categories, of the application of these catego- A strict reading of the First Amendment sug-
ries to different forms of speech, and of the gests that the government has no right to inter-
evidentiary criteria used to assess both the harms cede to regulate communication between a corpo-
of speech and the harms of regulation. For in- ration and its employees (harassment rules provide
stance, a near consensus has been reached in the an exception) or to intervene when privately owned
scientific community that media violence leads to media fail to meet the needs of a community.
social aggression and violence (e.g., Donnerstein Many argue that this is as it should be: employees
and Smith 1997), whereas evidence of harms done can always seek employment elsewhere and those
by ‘‘indecent’’ broadcasts is anecdotal. What then whose speech is barred in one forum can always
are the social and political processes that lead one seek another. Others insist that differences be-
and not the other to be regulated? tween the conditions of modern society and those
at the time the Constitution was drafted warrant
PUBLIC SPEECH, PRIVATE SPACES the extension of communicative rights beyond
those provided in the First Amendment.
As can be seen in the language of the First Amend-
ment, the central concern of the framers of the THE LEGAL SUPPRESSION OF SPEECH
Constitution was to protect the private realm from DEEMED TO THREATEN THE
domination by the state. The First Amendment ESTABLISHED ORDER
itself has little to say about the control that private
organizations exert over the expressive rights of A central rationale for laws restricting the expres-
individuals, or the control that private media com- sive activities of people and press is to preserve
panies exert over the communication of public national security. The chief mechanisms of en-
issues. Examples of private-domain controversies forcement under democratic governments have
include disputes over the ‘‘gag rules’’ that prohibit been restrictions on governmentally controlled
physicians from discussing alternate, more expen- information, laws prohibiting seditious libel, and,
sive treatments; corporate contracts that prevent in times of war, systems of prepublication clear-
employees from publicizing questionable employ- ance of press dispatches. Most agree that there are
er practices; and control over the correspondence some circumstances, particularly in times of war,
272
CENSORSHIP AND THE REGULATION OF EXPRESSION
in which some national security restrictions are War statute prosecuting speech that sowed ‘‘dis-
necessary. However, a strong proclivity exists to content among the free population or insubordi-
abuse these laws to protect policies or govern- nation among the slaves’’ provides just one exam-
ments that are losing, or have lost, their base of ple of how law could be enlisted in the service of
popular support. majoritarian predjudices. When the federal gov-
ernment returned to the sedition business with
The prototypical conception of censorship is passage of the Espionage and Sedition Acts, the
firmly rooted in governmental restrictions on country was characterized by a climate of isolation-
speech that is critical of government, yet poses ism and a receptive ear to the tenets of Socialists,
significant threats to security. The law of seditious who opposed the entry of the United States into
libel was imported to the United States from Eng- World War I. The Sedition Act rendered it illegal
land, where, during the eighteenth century, it to speak against the draft or to advocate strikes
served as a principal vehicle through which gov- that might hinder wartime production. It resulted
ernments attempted to stave off criticism and to in more than 800 convictions, including that of
control public opinion. The power of seditious Socialist presidential candidate Eugene Debs. Prose-
libel laws lay largely in their breadth—advocacy of cutions for political speech continued after World
an ‘‘ill opinion’’ of government was considered War II, when provisions of the 1940 Smith Act
actionable—and from the selectivity with which were used as the legal arm of an extensive and
they could be enforced. During most of the eight- popularly backed campaign to suppress Commu-
eenth century, these cases were decided by judges nist viewpoints in the press, the workplace, and the
rather than juries, and the truth of a charge was entertainment media.
not a defense until the nineteenth century.
A significant judicial outcome of cases stem-
The climate in which the Constitution’s press ming from the enforcement of these acts was the
clause was drafted was one in which a central evolution of the criteria used to determine wheth-
threat to viewpoint diversity was posed by govern- er political speech warrants conviction. Whereas it
ments who could use their monopolies on legiti- had previously sufficed that expression result in
mized force to control the marketplace of ideas. only a ‘‘bad tendency’’ to cause harm, these cases
The most noted illustration of this in colonial eventually produced today’s criterion of ‘‘imma-
America was the case of John Peter Zenger, whose nent lawlessness.’’ Had the expressive restrictions
newspaper, the New York Weekly was launched in of the first part of the century been in place during
1734 as the only voice of opposition to the widely the civil rights movement and the Vietnam War,
reviled colonial Governor, William Cosby. When much of what was said and written during these
the newspaper’s opening salvo attacked Cosby’s periods would have resulted in federally sanc-
arbitrary exercise of power, Zenger was promptly tioned jail sentences.
charged with seditious libel. After spending eight
months in jail, Zenger was acquitted by a jury National security interests have also been in-
(reluctantly allowed in this case), who ignored the voked to justify secrecy classification systems, the
judge’s instructions that the truth of a charge was labeling of foreign ‘‘propaganda,’’ and the use of
irrelevant. Although many thought the First Amend- contract law by agencies including the Central
ment would change such abuse, Congress soon Intelligence Agency and the Voice of America to
passed the Sedition Act, which was used by Presi- require prepublication clearance of communica-
dent John Adams, a Federalist, to silence criticism tions by both current and former employees.
launched by Republican supporters of Thomas
Jefferson. WARTIME RESTRICTIONS
The Sedition Act was allowed to lapse when Compulsory systems of prepublication clearance
Jefferson took the Presidency and would not re- during wartime were first used during the Civil
turn until World War I. However, in the years War and, with the notable exception of Vietnam,
during which the Constitution did not apply to the have continued to be used in all major military
states, statutory attacks on specific viewpoints could conflicts. The exception of Vietnam was prompted
be fierce at the state level. Louisiana’s pre-Civil
273
CENSORSHIP AND THE REGULATION OF EXPRESSION
by the government’s fear that prior review systems well as changes in the progress and interpretation
would alienate the press and by the fact that of research on the effects of pornography. The
readily available nonmilitary air transportation in uses of the conclusions drawn in these and other
the area would have rendered enforcement diffi- social-scientific reports are of particular sociologi-
cult. However, there is credible evidence that mili- cal interest. For example, the Nixon Administra-
tary officials systematically misrepresented prog- tion was quick to dismiss the conclusions of a
ress by the United States during the war and that report (which it had itself commissioned) that
media routines and practices rendered them sus- were contrary to its political goals. Here, as in
ceptible to this manipulation (Hallin 1986). other contested areas of expression, the political
climate in which research findings are interpreted
Despite the consensus on the need for some is often the deciding factor in how research is used.
form of press management system to protect the
lives of soldiers during wartime, controversies arise Regarding the related question of obscenity,
over the scope and mechanisms of enforcement. in 1973 the Supreme Court (Miller v. California,
Of particular concern to press and public is the 413, U.S. 15, 1973) established three criteria for
fear that controls ostensibly designed to protect determining whether a given work was obscene:
the lives of soldiers are used instead to manage An average person, applying contemporary local
public opinion at home. community standards, finds that the work, taken
as a whole, appeals to prurient interest; the work
PORNOGRAPHY depicts in a patently offensive way sexual conduct
specifically defined by applicable state law; and the
The term pornography comes from the Greek work in question lacks serious literary, artistic,
words for ‘‘prostitute’’ and ‘‘write,’’ and originally political, or scientific value.
referred to writings about prostitutes and their
activities. Today, pornography is broadly used to Political struggles over the regulation of por-
mean material with explicit sexual content. Like nography have been especially intriguing because
the word censorship itself, however, the definition they have brought together conservatives and femi-
of pornographic or obscene material is often con- nists, groups traditionally on opposite sides of the
tested. For example, feminist writers often draw a ideological spectrum. Although both sides may
distinction between pornography, which combines support banning pornography, they have different
sexuality with abuse or degradation, and erotica, reasons for doing so, and the scope of the material
which is sexually arousing material that respects they wish to have regulated differs as well. Femi-
the human dignity of the participants. A distinc- nists focus on the degrading character of pornog-
tion is also sometimes drawn between hard-core raphy, whereas conservatives view pornography as
pornography, which shows actual sexual inter- morally corrosive. Antipornography activists such
course or penetration, and soft-core pornography, as Catherine MacKinnon have defined pornogra-
which may be only suggestive of these activities. phy as a civil rights issue, arguing that pornogra-
Child pornography is prohibited in most nations, phy itself is a form of sexual discrimination; by
and restrictions on its production and distribution presenting women in dehumanizing ways, pornog-
tend to be noncontroversial. raphy subordinates them.
In the United States, the Commission on Ob- Although there is wide variation in porno-
scenity and Pornography (1970) and the Attorney graphic content, pornography often presents what
General’s Commission on Pornography (1986) many consider to be an unrealistic view of sexual
have provided recommendations for government relations. Encounters take place most often be-
action regarding pornography. The first report tween strangers, not in the context of enduring
suggested that pornography should not be regulat- relationships; participants are sex objects rather
ed by law, while the second rejected the claims that than complete individuals. Sexual activity always
pornographic material was harmless and urged results in ecstasy, and consequences (such as preg-
prosecution especially for materials that contained nancy or disease) are nonexistent.
violence or degradation. These differing conclu-
sions highlight the ongoing tensions between indi- Empirical research on the effects of pornog-
vidual rights and perceived community needs, as raphy has shown a range of negative effects. In
274
CENSORSHIP AND THE REGULATION OF EXPRESSION
typical experimental studies, participants are ran- conducted in-depth ethnographic style interviews
domly assigned to pornographic stimuli or with producers, performers, and other employees
nonpornographic stimuli, and their attitudes, be- in the pornography industry. In this work, the
haviors, or physiological reactions are then as- researchers allowed the people who create por-
sessed. There are also paradigms assessing pro- nography to speak in their own words, providing
longed exposure, including ones in which participants both defenses of pornography and insight into
return to the laboratory for multiple sessions of why individuals choose, for better or for worse, to
exposure to pornography, to better simulate real- work in the pornography industry.
life consumption patterns. Among other effects,
studies have shown that after exposure to pornog- HATE SPEECH
raphy, participants viewed rape as a less serious
crime, overestimated the popularity of less com- From the advent of the printing press onward,
mon sexual practices, and showed greater callous- most efforts to legally regulate expression have
ness toward women. Furthermore, pornography focused on various forms of mass communication.
consumers have shown weaker beliefs in the de- Efforts to restrict speech can also target communi-
sirability of marriage and having children, and cation in interpersonal settings. One such exam-
stronger beliefs in the normality of sexual promis- ple can be found in attempts to place legal limits
cuity. Ironically, pornography has also been shown on hate speech; defined as harassing or intimidat-
to reduce viewers’ satisfaction with their own sex ing remarks that derogate the hearer’s race, gen-
lives and partners. der, religion, or sexual orientation.
Although these outcomes have been found in Those who support regulating hate speech
a variety of studies using different research proce- compare it to ‘‘fighting words,’’ a category of
dures, not all studies have confirmed these find- speech not protected by the First Amendment, or
ings. Some critics of this research argue that rape argue that it creates a ‘‘hostile environment,’’ which
and other antisocial sexual behavior existed even violates provisions of the Civil Rights Acts of 1964
before pornography became widely available, and and 1990. Champions of hate speech regulations
that forces besides pornography play a greater cite the harms its victims suffer: these may range
causal role in antisocial behavior. Others argue from feelings of exclusion from a community, to
that the increased levels of aggression, hostility, or the experience of debasement that leads students
bias often found in experimental studies of por- to skip classes or distress that leads them to leave
nography might be traced to differences between school. In short, the liberty of a speaker to harass
the experimental and naturalistic environments in may deny the hearer’s right to equality. Critics of
which pornography is viewed: particularly, it is hate speech codes cite the administrative excesses
argued that these effects may be due to the lack of they allow. For example, the student guide to the
opportunity for men to ejaculate in the experi- University of Michigan code stated that students
mental setting. They also caution that the political could be punished for making a comment ‘‘in a
biases of investigators may influence the interpre- derogatory way about someone’s appearance.’’
tation of results. Others believe that even if the
effects research is accurate, the costs of suppress- Following an increase in reported incidents of
ing pornographic material—as measured in state hate speech in the late 1980s and early 1990s,
encroachment on individual autonomy—outweigh many universities in the United States adopted
any benefits that might be gained. policies forbidding discriminatory verbal harass-
ment. However, as they affect public institutions,
An argument sometimes raised for control- these rules have been declared unconstitutional in
ling or eliminating pornography in the form of the courts, in part for being either too broad—
images (rather than words) is that the individuals affecting too many forms of speech—or for being
pictured in the pornographic photos and films viewpoint based (see above).
suffered harm or coercion during the creation of
the materials. A counterpoint to this argument is The problems inherent in drafting laws that
provided by Stoller and Levine (1993), who have affect only the speech these rules target can be
275
CENSORSHIP AND THE REGULATION OF EXPRESSION
seen in comparing the provisions of the University BATTLES FOR CONTROL OVER SCHOOL
of Michigan code with provisions in the bill ad-
vanced by Senator Jesse Helms to limit what kinds CURRICULA
of works could receive funding from the National
Endowment for the Arts. Among the Helms bill’s Schools are the arena in which the values of open
other planks (aimed at homoerotic art, for exam- deliberation and the constructive exchange of ide-
ple), it sought to prohibit funding of ‘‘material as are most prized in democratic societies. It is
which denigrates . . . a person, group, or class of with some irony, then, that schools are also the
citizens on the basis of race, creed, sex, age . . .’’ domain in which some of the most concerted
Similarly, the Michigan code sought to bar speech efforts to limit the scope of deliberation have been
that ‘‘stigmatizes or victimizes an individual on the focused. The content of textbooks used by most
basis of race, ethnicity, religion, sex . . .’’ and also students in the United States has been selectively
‘‘creates a hostile environment.’’ The projects tailored to the ideological viewpoints of organized
targeted by the Helms bill included one with a pressure groups, and at least one-third of high
crucifix immersed in a bottle of urine. This work school students are not exposed to books and
could also be a violation of the Michigan code if it films that parents and pressure groups have suc-
were to be hung in an area frequented by offended cessfully purged from their educational experi-
Christians. Difficulties of this sort have prompted ence (Davis 1979).
those supporting hate speech codes to argue that
this is one area in which viewpoint-based speech From the vantage point of most teachers, pa-
regulations should be allowed: that, similar to laws rental and school board mandates to omit works
in Germany that selectively ban Holocaust denial, of literature from the curriculum are viewed as
speech laws in the United States should be allowed illegitimate challenges to their expertise and in-
to account for the history of discrimination experi- fringements on their rights as professionals. In
enced by some groups, but not others. Critics say effect, teachers believe that they are in the best
that viewpoint discrimination of this sort would positions to judge the educational needs of the
open the floodgates for future laws that selectively students they teach. However, the tradition of
target viewpoints that happen to be deemed academic freedom that characterizes both public
undesirable. and private universities is rarely present in elemen-
tary and secondary education. Public school cur-
The hate speech debates highlight a funda- ricula are generally approved by the state or by
mental dilemma faced by modern democracies: to school boards that actively regulate what is taught
what extent are we willing to tolerate speech whose and what students read. School board members
very goal is to silence the speech of others, or to tend to be elected to their positions and are par-
deny their rights to equal education or employ- ticularly responsive to what teachers view as unrea-
ment? Efforts continue to craft codes that are sonable censorial demands.
narrow in scope and that meet constitutional mus-
ter, with even staunch civil libertarians favoring Parents, on the other hand, see their interven-
prosecution when hate speech poses a clear and tions in the school curricula as a legitimate exer-
present danger of violence (e.g., Smolla 1992). cise of control over what their children read in
The hate speech debates also mark a significant school. They justify their decision, in part, with
turn in the rhetoric of ‘‘censorship,’’ one in which respect to the compulsory nature of early school-
advocates of speech codes have attempted to enlist ing, where they see their children as captive audi-
the power of this term in their favor. This rhetori- ences. A number of well-funded organizations
cal strategy is exemplified by the argument of have exerted strong influences on both school
Catharine MacKinnon, one of the staunchest code boards and textbook adoption processes to influ-
supporters, that ‘‘the operative definition of cen- ence what does and does not gain entry to the
sorship . . . shifts from government silencing what schoolroom.
powerless people say, to powerful people violating
powerless people into silence and hiding behind In money terms, the largest impact of pressure
state power to do it’’ (1993, p. 10). groups has been on textbook publication and
adoptions. Most states place orders for textbooks
for schools in the entire state. Because publishers
cannot provide multiple versions for different re-
gions or states, texts are geared to the largest
276
CENSORSHIP AND THE REGULATION OF EXPRESSION
markets, usually Texas and California (Del Fattore they have been difficult to establish. Often teach-
1992). As a result, publishers provide their writers ers simply capitulate to perceptions of the most
with guidelines that govern topics and viewpoints conservative or protective community elements
that are currently deemed objectionable (or desir- and use readings that are sure to be ‘‘safe’’ (Da-
able) in the largest states that engage in statewide vis 1979).
adoptions. In this way, pressure groups in one
state, such as Texas, often exercise veto power In 1990s most lawsuits were mounted by fun-
over the entire country’s textbook market. Local damentalists who condemned many books on
communities in small and mid-size states exercise a school lists as anti-Christian, antiparent,
relatively small influence on content. antigovernment, immoral, and obscene. The chal-
lenged books change from year to year, but fre-
Community protest and lawsuits are the prin- quently banned books have included Of Mice and
cipal vehicles used to challenge textbooks and Men (Steinbeck), Catch 22 (Heller), and Catcher in
reading lists in literature, social studies, and health the Rye (Salinger). Fundamentalists are effective
education. An illustrative controversy—at McClintock out of all proportion to their numbers because of
High School in Tempe, Arizona in 1996—high- the intense dedication they bring to their cause.
lights the susceptibility of teachers and schools to
parental and community demands. In this case, a Research has yet to provide a sound basis for
parent of a McClintock High student protested the confidently assessing the effects of reading on
assignment of Mark Twain’s classic The Adventures school children’s beliefs. When a story contains a
of Huckleberry Finn because the word ‘‘nigger’’ is character’s arguments against the existence of God,
included in its dialogue. In the ensuing controver- might students question their faith? Qualitative
sy, some held that if schools were forced to shield research, particularly, has shown that students
students from exposure to the term, such interven- take widely divergent readings from stories and
tion would undermine the legitimate authority of that these understandings are frequently critical of
teachers and would endorse ignorance of Ameri- the viewpoints adopted by both characters and
can history and the practice of censorship. writers.
We may ask whether the students grasp the Can we maintain that literature gives readers
distinctions among Twain’s stance on racism, the insights into life? That it can change lives? Par-
use of the term ‘‘nigger’’ by a character in the ents—former pupils, after all—have been encour-
novel, and societal endorsement of the term then aged to believe that great works contain great
and today. Of course the crux of such conflicts also truths. It is not surprising that some of those
rests upon the educational and social contexts in parents who discern arguments against the social
which sensitive, taboo, or potentially affronting order in their children’s readings seek to eliminate
topics that appear in the world of fiction are the threat from children’s lives. Eventually, these
discussed. For instance, if The Adventures of Huckle- ongoing and recurring controversies may be illu-
berry Finn or any other literature is misused to minated by scientifically grounded understanding
communicate racism, the problem is larger than of the social and cognitive foundations of narra-
the choice of which books are read in the classroom. tive impact (Green, Strange, and Brock 2000).
Unfortunately, the social context of this con- One of the strongest arguments for keeping
troversy added complexity. School benches at the school curricula open to ideas to which we current-
same high school were etched with the words ‘‘I ly object or to heinous artifacts of historical preju-
hate niggers.’’ Of course, there is an important dices is that one of the important functions of
distinction between eliminating the word ‘‘nig- education is the development of critical inquiry.
ger’’ from The Adventures of Huckleberry Finn and From this perspective, to turn the schoolhouse
removing ‘‘I hate niggers’’ from a school bench. into an environment where only ‘‘safe’’ ideas are
The etching creates a hostile educational and so- encountered is to do students and society a dis-
cial environment in a way that the assignment of service. This position is rendered more compell-
book should not do, as long as teachers and schools ing in the context of the current media landscape
are doing their jobs well. In controversy after which makes it certain that students will also en-
controversy, these distinctions are as important as counter these ideas outside the context of a critical
learning environment. This argument, of course,
277
CENSORSHIP AND THE REGULATION OF EXPRESSION
rests on the assumption that schools are, in fact, year old, a film that was prohibited to persons
places where critical inquiry is fostered. under eighteen years of age would be more desir-
able than a film that was prohibited to all persons.
THE SOCIAL PSYCHOLOGY OF In the former case, the censorship has more im-
‘‘CENSORSHIP’’ pact because it is viewed as violating a personal
freedom; freedom to view a film is more threat-
When information (e.g. a book, a film, a speech, ened if there are many other people (all persons
etc.) is explicitly ‘‘censored’’ do people desire that over eighteen years, say) who do have that freedom.
information much less—in accordance with the
presumed intention of the censoring agents? Are The experimentally grounded insights from
people less likely to be influenced by a point of social psychology—that explicit censorship usually
view when its expression has been prohibited by doesn’t work as planned, and that it often back-
those in authority? To answer such questions, the fires—is often not factored into the planning and
impact of the label ‘‘censored’’ can be studied interventions of censoring agents. In some in-
experimentally by exposing people to a communi- stances this may be good, as it reduces the effec-
cation (or a ‘‘taste’’ of a communication) to which tiveness of those who attempt to limit information
‘‘censored’’ is affixed and by then observing the that is vital to public deliberation. In some instanc-
consequences of such a juxtaposition of label and es, public information officers appear to have
communication on their values and beliefs. When become aware that perceptions of censorship can
this careful experimental work has been done, the hamper their efforts. In the 1970s, the Depart-
answer to both of these questions is ‘‘no.’’ Numer- ment of Defense directed that the term be com-
ous experiments have shown that explicit censor- pletely dropped. This seemingly cosmetic change
ship often backfires: the prohibition of communica- affects the way news coverage labels the sources of
tions leads people to covet them and to change news reports: what was once attributed to the
their attitudes in the directions the censored mate- Office of Censorship, became, during the Gulf
rial advocates. War, a report from the Joint Information Bureau.
A social-commodity theory (Brock 1968) ac- Research on related forms of content advisory
counts for the backfire effect of censorship in labels, such as ‘‘parental discretion advised,’’ ‘‘viewer
terms of the psychology of supply and demand. All discretion advised,’’ and the more specific codes
public communications are perceived to have some signaling levels of violence, nudity, sex, and pro-
limits on their distribution, and message recipi- fanity on television, has begun to reveal the extent
ents usually have some dim awareness of the ex- to which different kinds of labels affect the selec-
tent of these limits. However, if the distribution is tion choices of boys and girls of different ages
explicitly limited by government statute, or other (Cantor and Harrison, 1994–1995). In several cases,
overt interventions, awareness of scarcity is sharp- these studies have revealed a ‘‘backfire’’ or ‘‘for-
ly heightened and the desirability and impact of bidden fruit’’ effect similar to the pattern found in
the communication are consequently increased. reactions to the censorship label. The finding that
When some information becomes less available, restricted materials become more attracted—which
this unavailability augments the information’s val- depended on age, gender, and the kinds of labels
ue. A decrease in supply causes an increase in used—have been extended to video game and
demand, and perceived censorship invariably en- music advisories.
tails a perceived decrease in supply.
Advocates of requiring such labels argue that
Experimental social psychology has illuminat- their goal is not to limit communication at its
ed the impact of censorship by showing that (a) an source, but rather to provide parents and children
individual can be influenced by a censored com- a greater degree of control over their communica-
munication, even without actually receiving it, (b) tion choices. Some critics argue that, regardless of
the backfire effect increases as the penalties for intent, these labels lead producers to alter their
violating censorship are increased; and, somewhat productions in order to avoid undesirable label-
paradoxically, (c) the backfire effect is greater ing. As with the ‘‘censored’’ label, the effects on
when the prohibition applies selectively to one consumption of the targeted content will depend
group of people. For example, for a seventeen not only on how the label affects desirability of the
278
CENSORSHIP AND THE REGULATION OF EXPRESSION
materials, but also on the ultimate availability of advertising contracts, the licensing of journalists,
the labeled content. If parents use advisory labels controls on distribution of newsprint, and, more
to preselect their children’s television viewing generally, government ownership and control of
choices, these labels will effectively place greater broadcast media.
selective control in the hands of parents. The
probability that parents will do so is heightened by Controls at the site of reception provide an-
technologies like the ‘‘V-chip,’’ through which pa- other means of suppressing viewpoints that lack
rental choices can be automated. Similar labels the official stamp of approval. For example, a
may have a much different effect on the selection number of countries pose outright bans on owner-
of music and video games, where children are ship of satellite receiving dishes. However, new
more likely to make their own media selections. technologies, particularly the Internet, pose seri-
ous threats to the control of political information
Research has also begun to examine the social by authoritarian regimes. The Internet’s impend-
and psychological origins of calls to restrict deemed ing threat is linked to the growing dependency
harmful to children or to society at large. Studies upon computer-based communications in the mar-
of the ‘‘third person effect’’ suggest that support ketplace. Authoritarian governments appear to be
for restrictions on such content may sometimes be faced with a Hobson’s choice: allow connections to
grounded in systematic overestimates of the ef- the vast and diverse information resources of the
fects of media on other persons. Sociological treat- Internet or face economic stagnation.
ments of the rise of public alarm are particularly
relevant here. Nicola Beisel’s (1992) account of the Particularly when looking beyond suppressive
late-ninteenth-century attack on obscenity that re- controls on political speech, it is important to
sulted in the highly suppressive Comstock Laws is recognize the distinction between government-
one example of a fruitful direction that research controlled media and government-funded media.
might take. The BBC is but one example of the government-
funded, public television stations that have devel-
INTERNATIONAL ISSUES oped reputations for independent news coverage
that gives voice to diverse perspectives of both
Whereas most democratic nations have codified majority and minority interests. Publicly-subsidized
and enforced protections against the most blatant television in Holland instantiates an intriguing
forms of government suppression of sociopolitical model of ‘‘free speech’’ through a system that
speech, governments around the world continue allots television channels and channel space on a
to engage in harsh repression. The bounty placed proportional basis to key institutions and sociocul-
on the life of Salman Rushdie in retribution for his tural groups. Public television stations in many
book, The Satanic Verses, is a well-publicized in- European nations seek independence from the
stance of the most egregious form of government- demands of both government and marketplace
sanctioned repression. However, a variety of both through nongovernmental boards of trustees and
legal and extra-legal tactics continue to be used through stable funding mechanisms not subject to
worldwide to protect those in power. To cite just annual tampering by politically driven legislatures.
two typical examples from 1999: A court in
Kazakhstan suspended publication of an opposi- INTERNATIONAL INFORMATION FLOW
tion newspaper for the three months preceding
presidential elections on the pretext that it failed A handful of news agencies, led by the Associated
to indicate the place of printing on its masthead; Press (AP), Reuters, and Agence-France Presse
and, shortly after the director of the Cameroon dominate the news that people around the world
newspaper Le Messager-Populi was released from a read about issues and events originating outside
ten-month prison sentence for spreading ‘‘false their borders. The dominance of both print and
information,’’ the paper’s founder was driven from broadcast news by a small group of companies is
the country upon threats to his life (see generally, particularly felt in developing nations, ones lack-
Webb and Bell 1998). Typical extra-legal means of ing the capital to compete with the economies of
suppressing speech critical of the government in- scale enjoyed by entrenched foreign corporations.
clude awarding and withdrawing governmental The concern generated in developing nations fo-
cuses not only on the news they receive from
279
CENSORSHIP AND THE REGULATION OF EXPRESSION
abroad, but also on the images that news agencies as ‘‘civic’’ journalism, its detractors suggest that
project about them. This concern is compounded the values of developmental journalism cater to
by news agency values that typically stress large- the needs of local authority. And whereas NWIO
scale violence and disruption—’’coups and earth- critics argue for a ‘‘free flow’’ of information across
quakes’’ as the saying goes—and ignore the cultur- international boundaries, NWIO supporters advo-
al achievements and developmental efforts for cate a ‘‘balanced flow.’’ Although these issues re-
which these countries would like to be known. main unresolved, promising developments include
the rise of regional news agencies, programs such
Those who control international news flows as CNN’s World Report, in which locally produced
view their position of dominance as a desirable broadcasts gain worldwide exposure; and pros-
and natural outcome of free-market practice. Those pects that the Internet will facilitate the interna-
in developing nations view the imbalances as a tional communication of locally produced news.
form of de facto neo-imperialism. This view finds
its historical roots in the late-nineteenth and early- The increasing global dominance of U.S. mov-
twentieth centuries, when three European news ies and television programming have prompted
agencies, Havas (France), Wolff (Germany), and the European Economic Community and Canada,
Reuters (England) carved the world into three among other regional and national entities, to
areas of monopoly control, ones that built upon seek ceilings on imported cultural goods as excep-
their respective colonial empires. Some insight tions to free trade principles in international trade
into the present-day perspective of developing agreements. These demands reflect the belief that
nations can be gained by examining how news when applied to the area of expression, free mar-
about the United States was communicated in the ket principles support forms of ‘‘soft’’ or ‘‘de facto
early part of the century. During those years, the censorship’’ in which locally produced speech is
‘‘Ring Cartel’’ prevented AP from sending news driven from the marketplace of ideas (cf. Frederik
about the United States beyond its borders. Ac- 1993). More generally, these demands reflect the
cording to Kent Cooper, the AP manager who growing tension between the ongoing course of
fought to dismantle it, the cartel decided ‘‘what the globalization and the threats to cultural sovereign-
people of each nation would be allowed to know of ty it entails.
the people of other nations and in what shade of
meaning the news was to be presented.’’ It was REFERENCES
Reuters who ‘‘told the world about Indians on the
war path in the West, lynchings in the South, and Bagdikian, B. 1997 The Media Monopoly. Boston: Bea-
bizarre crimes in the North. . . . The charge for con Press.
years was that nothing credible to America was
ever sent’’ (cited in Frederik 1993, p.39). To those Baird, R.M., and S.E. Rosenbaum 1991 Pornography:
in developing nations, it is the AP who now plays Private Right or Public Menace? Buffalo, N.Y.:
the role of cartel. Prometheus Books.
Several proposals have been advanced to cor- Baker, C.E. 1998 ‘‘The Media that Citizens Need.’’
rect the imbalance in international news flows. University of Pennsylvania Law Review 147, 317.
Notable are those advanced through the United
Nations Educational, Scientific, and Cultural Or- Barbur, B. R. 1996 Jihad vs. McWorld: How Globalism
ganization (UNESCO) during the late 1970s and and Tribalism are Reshaping the World. New York:
early 1980s. These calls for a ‘‘New World Infor- Ballantine Books.
mation Order’’ (NWIO) generated immediate and
prolonged controversy, and provoked the United Beisel, Nicola. 1992 ‘‘Constructing a Shifting Moral
States to withdraw from UNESCO in 1985. The Boundary: Literature and Obscenity in Nineteenth-
contours of these debates are as intriguing as they Century America.’’ In M. Lamont and M. Fournier,
are complex, reflecting sharp differences in news Eds., Cultivating Differences: Symbolic Boundaries and
values, international economic principles, and the the Making of Inequality. Chicago: University of Chica-
very definition of ‘‘free speech.’’ Whereas NWIO go Press.
supporters liken the values of developmental jour-
nalism to current trends in the United States, such Bourdieu, P. 1991. ‘‘Censorship and the Imposition of
Form.’’ In J.B. Thompson, Ed. Language and Symbolic
Power. Cambridge, Mass.: Harvard University Press.
Brock, T.C. 1968. ‘‘Implications of Commodity Theory
for Value Change.’’ In A.G. Greenwald, T.C.Brock,
280
CENSUS
and T.M.Ostrom, eds. Psychological Foundations of Jensen, C. 1997 20 Years of Censored News. New York:
Attitudes. New York: Academic Press. Seven Stories Press.
Cantor, J., and K. Harrison 1994–1995 Ratings and MacKinnon, C. A. 1993 Only Words. Cambridge, Mass.:
Advisories for Television Programming.’’ In National Harvard University Press.
Television Violence Study: Scientific Papers. Los Ange-
les: Mediascope Inc. Post, R.C. (ed.) 1998 Censorship and Silencing: Practices of
Cultural Regulation. Los Angeles: The Getty Research
Christenson, Peter G., and Donald F. Roberts 1998. ‘‘It’s Institute for the History of Art and the Humanities.
Not Only Rock and Roll: Popular Music in the Lives of
Adolescents.’’ Cresswell, N.J.: Hampton Press. Shoemaker, P.J., and S.D. Reese 1996 Mediating the
Message: Theories of Influences on Mass Media Content.
Court Case Citation, Gitlow vs. New York, 268 U.S. New York: Longman.
652, 1925.
Smolla, R.A. 1993 Free Speech in an Open Society. New
Court Case Citation, Red Lion Broadcasting vs. FCC, York: Vintage Books.
395 U.S. 367, 1969.
Stoller, R.J, and I.S. Levine 1993 Coming Attractions: The
Court Case Citation, Miller vs. California, 413 U.S. Making of an X-Rated Video. New Haven, Conn.: Yale
15, 1973. University Press.
Curran, J., and M. Gurevitch 1996 Mass Media and Sunstein, C. 1993. Democracy and the Problem of Free
Society. London: Edward Arnold. Speech. New York: Free Press.
Davis, J. Ed. 1979. Dealing with Censorship. Urbana, Ill.: Webb, W.L, and R. Bell (eds.) 1998 An Embarrassment of
National Council of Teachers of English. Tyrannies: Twenty-Five Years of Index on Censorship.
New York: George Braziller.
Deetz, S. A. 1992. Democracy in an Age of Corporate
Colonization: Developments in Communication and the Zillmann, D., and J. Bryant (eds.) 1989 Pornography:
Politics of Everyday Life. New York: State University of Research Advances and Policy Considerations. Hillsdale,
New York Press. NJ: Erlbaum.
Del Fattore, J. 1992. What Johnny Shouldn’t Read: Text- JEFFREY J. STRANGE
book Censorship in America. New Haven, Conn.: Yale MELANIE C. GREEN
University Press. TIMOTHY C. BROCK
Dennis, E. E., et al. 1991. The Media at War: The Press and CENSUS
the Persian Gulf Conflict. New York: Gannett Founda-
tion Media Center. A national census of population is ‘‘the total proc-
ess of collecting, compiling, evaluating, analysing
Donnerstein, E., and S.L. Smith 1997. ‘‘Impact of Media and publishing or otherwise disseminating demo-
Violence on Children, Adolescents, and Adults.’’ In graphic, economic, and social data pertaining, at a
S. Kirschner and D. A. Kirschner, Eds., Perspectives on specified time, to all persons in a country’’ (United
Psychology and the Media. Washington, D.C.: Ameri- Nations 1988, p. 3). The United Nations encour-
can Psychological Association. ages its members to take regular censuses and
provides technical assistance. It also publishes an
Emerson, T. I. 1973 The System of Free Expression. New annual demographic yearbook and maintains a
York: Random House. web site with selected census data and other popu-
lation information for most countries.
Foerstel, H.N. 1997. Free Expression and Censorship in
America: An Encyclopedia. Westport, Conn.: Green- A nation may conduct censuses on other sub-
wood Press. jects such as housing, business firms, agriculture,
and local governments. Because of the close link
Frederick, H.H. 1993 Global Communication and Interna- between families and housing units, censuses of
tional Relations. Belmont, Calif.: Wadsworth. population and housing are usually combined.
Green, J. 1990 The Encyclopedia of Censorship. New York: The actual enumeration for a national popula-
Facts on file. tion census is usually spread over a period of weeks
Green, M.C., J.J. Strange, and T.C. Brock (eds.) 2000.
Narrative Impact: Social and Cognitive Foundations.
Mahwah, N.J.: Erlbaum Associates.
Habermas, J. 1989 The Structural Transformation of the
Public Sphere. Cambridge Mass.: Polity.
Hallin, D. C. 1986. The ‘‘Uncensored War’’: The Media and
Vietnam. New York: Oxford University Press.
281
CENSUS
or months, but an attempt is made to record on many characteristics, such as age, sex, and
circumstances as of a designated census day (April relationship to others living in the same house-
1 for the 2000 U.S. census). In some censuses, hold, educational level, and occupation. The cen-
persons are reported in their legal or usual place sus then reports on population size and character-
of residence (a de jure enumeration). According to istics for provinces, states, counties, cities, villages,
the Bible, a census decree issued by the govern- and other administrative units. To obtain all this
ment in Rome ordered that persons be counted information, the 2000 U.S. census (U.S. Census
and taxed in their home towns. In response, Jo- Bureau 1999) asked every household to complete
seph and his pregnant wife, Mary, traveled from a ‘‘short form’’ with six questions about each
Nazareth to Bethlehem, where Mary gave birth to person and one question about each housing unit.
Jesus. Other censuses record people where they A carefully selected sample of one of every six
are on the census day (a de facto enumeration). households received a ‘‘long form’’ that included
Whatever residency rule is followed, all conceiv- additional questions on fourteen topics for each
able special circumstances must be anticipated, person and thirteen topics for each housing unit.
questionnaires well designed, and census staff thor-
oughly trained. Reporting information for the entire country
and for the thousands of cities and other subareas
The U.S. census uses the concept of ‘‘usual results in many volumes of printed reports. Many
residence,’’ where the person lives and sleeps most nations issue additional data on computer-read-
of the time, but some exceptions are made. A few able files designed for convenient use by national
examples from the 2000 U.S. census illustrate the and local governments, organizations, and indi-
range of circumstances. Some persons staying in viduals. Increasing quantities of data are available
hotels and rooming houses are transients tempo- to anyone with access to the Internet.
rarily absent from their usual homes; they com-
pleted special census forms that were later com- In the United States, the idea for a regular
pared to the census forms received from their census emerged during debates about the prob-
home addresses to confirm that they were proper- lems of creating a representative form of govern-
ly reported by other household members. An op- ment. The U.S. Constitution (Article I, Section 2)
eration called ‘‘Service-Based Enumeration’’ was directs that membership in the House of Repre-
designed to locate homeless persons and others sentatives be based on population: ‘‘The actual
with no usual residence. College students were enumeration shall be made within three years
presumed to be residents at the place they lived after the first meeting of the Congress of the
while attending college, even if they were at home United States, and within every subsequent term
on spring break on census day. Efforts were made of ten years, in such manner as they shall by law
to avoid having an away-at-college child erroneous- direct.’’
ly double-counted as a member of the home house-
hold. Pre-college students at away-from-home The questions asked in a census reflect politi-
schools were assigned to the parental household. cal and social issues of the day and often provoke
Citizens working in another country for the U.S. spirited debates. For example, slavery was a recur-
government, and their families, were enumerated rent divisive issue for the Constitutional Conven-
(primarily through the use of administrative rec- tion. One of the two newly created legislative
ords), but other citizens living abroad at the census bodies, the House of Representatives, was to have
date were not counted. Short-term tourists, how- the number of legislators from each state propor-
ever, were supposed to be reported by other mem- tional to the population of the state. Including
bers of their households, or by late enumeration slaves in the population count would increase the
upon their return home. Persons not legally resi- legislative power of southern states. A compro-
dent in the United States were supposed to be mise (Article I, Section 2) provided that: ‘‘Repre-
counted, but many feared all contact with authori- sentatives. . . shall be apportioned among the sev-
ties and took steps to avoid detection by census eral states. . . according to their respective numbers,
procedures. which shall be determined by adding to the whole
number of free persons, excluding Indians not
The total population count is only one result taxed, three-fifths of all other persons.’’ The dis-
of a modern census. Information is also obtained tinction between ‘‘free persons’’ and ‘‘all other
282
CENSUS
persons’’ (slaves) was eliminated after the Civil Governmental statistical agencies are often set
War (Amendment 14, Section 2). in their ways, concerned with continuity rather
than innovation and removed from the higher
The 1790 census was modest in scope. Assis- levels of policymaking. For several decades begin-
tants to federal marshals made lists of households, ning in the late 1930s, the U.S. Census Bureau was
recording for each household the number of per- extraordinarily creative. Social scientists and stat-
sons in five categories: free white males over six- isticians employed by the Census Bureau were
teen and under sixteen, free white females, other active in research and scholarly publication and
free persons, and slaves (Anderson 1988, p. 13). played leadership roles in professional organiza-
tions. Census Bureau personnel pioneered in de-
Censuses from 1790 through 1840 were con- velopment of the theory and practice of popula-
ducted with little central organization or statistical tion sampling. To accommodate large increases in
expertise. A temporary federal census office was content and geographic scope, the 1940 census
established to conduct the 1850 census. The indi- questionnaire was divided into two parts. A set of
vidual rather than the household became the focus basic questions was asked of everyone, while a set
of the enumeration, and the content of the census of supplementary questions was asked of a one-in-
was expanded to include occupation, country or twenty sample. Subsequent U.S. censuses have
state of birth, and other items. Experienced statis- continued to use sampling, as with the ‘‘long-
ticians were consulted, and the United States par- form’’ versions of the 1990 and 2000 question-
ticipated in the first International Statistical Con- naires that went to one of every six households.
gress in 1853. Sampling theory was also applied to the develop-
ment of periodic sample surveys to provide timely
The temporary office for the 1890 census information between censuses and to provide in-
became one of the largest federal agencies, em- formation on additional topics.
ploying 47,000 enumerators and 3,000 clerical
workers. To help with the enormous task of tally- For the 1950 census, the Bureau participated
ing data, census officials supported the develop- actively in the development and utilization of an-
ment by a young inventor, Herman Hollerith, of other new technology, the computer. In later cen-
an electrical tabulating machine. His punched- suses, the schedules were microfilmed and optical-
card system proved effective in census operations ly scanned for direct input of information (without
and later contributed to the growth of the IBM names and addresses) into a computer for error-
corporation. checking, coding, and tabulating.
A permanent census office was established in Through 1950, census enumeration in the
1902. An increasingly professional staff assumed United States was accomplished almost entirely by
responsibility for the population censuses and a having an enumerator conduct a personal inter-
broad range of other statistical activities. Deficien- view with one or more members of each house-
cies in the federal statistical system became appar- hold and write the information on a special form.
ent during the 1930s as the nation tried to assess By 1990, most census questionnaires were distrib-
the effects of the Great Depression and to analyze uted and returned to the Census Bureau by mail,
an array of new programs and policies. Several with telephone follow-up replacing many personal
federal agencies successfully advocated expansion visits. Large-scale use of the Internet and other
of the social and economic content of the 1940 innovative response technologies may be practica-
census. New questions in 1940 asked about partici- ble for widespread use in the 2010 census.
pation in the labor force, earnings, education,
migration (place of residence in 1935), and fertili- National censuses have so far been the best
ty. A housing census was paired with the popula- method for obtaining detailed information about
tion census to provide information about housing the entire population. Only censuses provide counts
values and rents, mortgages, condition of dwell- and characteristics for small administrative and
ings, water supply, and other property issues. The statistical areas, such as villages, voting districts,
practice of providing population and housing in- city blocks, and census tracts. Only censuses pro-
formation for subareas of large cities was greatly vide reliable information on small population
expanded.
283
CENSUS
groups, such as the income distribution of female Congress must approve the final census schedule.
plumbers. Only censuses provide the large num- A question on pet ownership has been regularly
bers of cases needed for complex multivariate proposed but rejected because there is no com-
analyses. But censuses are so large and difficult to pelling governmental interest in such a question
process that by the time the data are released they and private sample surveys can provide the de-
are, for some purposes, out of date, and this sired information. A third reason the content of
problem increases throughout the interval, com- the questionnaire changes little is that proponents
monly ten years, until the next census. of new topics and questions must compete for
questionnaire space, while questions previously
To address the timeliness problem, most na- asked tend to already have a network of users who
tional statistical agencies conduct a series of large have a vested interest in retaining the topic. This
sample surveys that provide select information on competition spurs extensive lobbying and mobili-
key topics at timely intervals. In the United States, zation of support from federal agencies, congres-
for example, the latest unemployment rate, re- sional committees, and interest groups.
leased on the first Friday of each month, is head-
line news. It is based on the Current Population An innovative approach to providing timely
Survey, a monthly sample survey that provides data is ‘‘continuous measurement’’ using a ‘‘roll-
current information on many social and economic ing-sample’’ survey. The American Community Sur-
characteristics. vey has been implemented by the U.S. Census
Bureau (1999), with plans to expand to three
A few countries keep population registers, million households a year in 2003. Because the
continuous records of where people live along survey uses a small, permanent, well-trained and
with some of their characteristics. If reasonably supervised staff (as compared to the large, tempo-
complete and accurate, these may be used to rary, briefly trained staff utilized for a census), data
supplement and update some census information. quality may be higher than in a census. Another
In many nations, administrative records such as advantage of a monthly survey over a decennial
social security, national health, tax, and utility census is the ease of introducing new topics and
company files, may be adapted to provide periodic testing the effects of changes in the wording of
estimates of population size and a few characteris- questions.
tics, but difficult questions arise about data quality
and comparability with census- and survey-based American Community survey results will be cu-
information. Citizens of many nations are wary mulated over a year to provide estimates of num-
about letting the government and corporations bers and characteristics for the nation, states, and
compile extensive personal data. Legislative re- places or groups of 65,000 or more people. For
strictions are increasingly being placed on what smaller places and groups, or for more reliable
information may be gathered and stored and how and detailed data for larger places, data will be
it may be used. cumulated for periods of up to five years. The five-
year cumulated sample is designed to be approxi-
The questions asked in successive censuses mately equivalent to the census ‘‘long-form’’ sam-
typically change more slowly than the procedures. ple and may eliminate the need for a ‘‘long form’’
Keeping the same topics and the same wordings of in the 2010 census. Moving averages could provide
questions help a government measure change from annual updates and be used for time series. There
one census to the next. This appeals to researchers are many issues to be worked through in develop-
and policy analysts, but policymakers and adminis- ing and evaluating this type of survey and in
trators, whose attention is focused mainly on cur- determining how well information averaged over
rent programs and next year’s budget, often plead a long period compares to standard statistical
for new wording to better serve current concepts. measures based on time-specific censuses and
Another reason tending to stifle innovation is that surveys.
the census is an expensive and visible tool. A
lengthy review process confronts any agency that In many nations, questions on race and eth-
seeks to add, delete, or alter a question. In the nicity are a sensitive and contentious topic. In the
United States, both the executive branch and the United States, the groups recognized have changed
284
CENSUS
from each census to the next. For example, special who were anxious for the government to enumer-
tallies were made from the 1950 and 1960 census- ate and classify their group appropriately.
es of persons with Spanish surnames, but only for
five southwestern states. Beginning in 1970, and In the United Kingdom, controversy about a
elaborated in later censuses, persons were asked if proposed question on ethnic identity led to the
they were of Spanish/Hispanic origin. Many re- question’s omission from the 1981 census sched-
sponses to this question have seemed to be incon- ule, thus hindering analyses of an increasingly
sistent with responses to the separate question on diverse population.
race. The Bureau has conducted sample surveys to
test various question wordings, and has sponsored The most politically intense and litigious con-
field research and in-depth interviews to provide troversy about recent U.S. censuses is not that of
insights into how people interpret and respond to content but of accuracy. States, localities, and
questions about ethnicity. For the 2000 census the many interest groups have a stake in the many
basic concepts were retained, but the wording of billions of dollars of government funds that are
each was adjusted, provision was made for indi- distributed annually based in part on census
viduals to report more than one race, and the numbers.
question on Hispanic ethnicity was placed before
the question on race. President George Washington commented
about the first census that ‘‘our real numbers will
In recent decades, extensive social science exceed, greatly, the official returns of them; be-
research has been conducted in many parts of the cause the religious scruples of some would not
world on issues of racial and ethnic identity. These allow them to give in their lists; the fears of others
identities are almost always more flexible and that it was intended as the foundation of a tax
more complex than can be captured with simple induced them to conceal or diminish theirs; and
questions. A further complication is that many through the indolence of the people and the negli-
persons have ancestral or personal links to two or gence of many of the Officers, numbers are omit-
more racial and ethnic groups; how they respond ted’’ (quoted in Scott 1968, p. 20).
depends on how they perceive the legitimacy and
purposes of the census or survey. A perfect census of a school, church, or other
local organization is sometimes possible, if mem-
In the 1990s, the U.S. Office of Management bership is clearly defined and the organization has
and Budget conducted an extensive review of its up-to-date and accurate records, or if a quick and
policy statement that specifies the standard race easily monitored enumeration is feasible. A na-
and ethnic classifications to be used by all govern- tional census, however, is a large-scale social proc-
ment agencies, including the Census Bureau ess that utilizes many organizations and depends
(Edmonston and Schulze 1995). Serious methodo- on cooperation from masses of individuals. Plan-
logical problems have arisen as a result of inconsis- ning, execution, and tabulation of the 2000 U.S.
tencies among classifications of individuals on Census of Population extended over more than
repeated interviews and in different records. A ten years. Hundreds of thousands of people were
dramatic example occurs with infant mortality employed at a cost of several billion dollars. A
rates for race and ethnic groups, which are based discrepancy between the results of a national cen-
on the ratio of counts based on birth registration sus and ‘‘our real numbers’’ is inevitable.
to counts based on death registration of infants
less than one year old. Comparing birth and death Statisticians, recognizing that error is ubiqui-
certificates for the same person, recorded less tous, have developed many models for identifying,
than one year apart, revealed large numbers for measuring, and adjusting or compensating for
whom the race/ethnic classification reported at error. Census statisticians and demographers
birth differed from that reported at death. The around the world have participated in these devel-
high-level government review also heard from many opments and in trying to put professional insights
interest groups, such as Arab-Americans, multiracial to practical use.
persons, and the indigenous people of Hawaii,
Following the 1940 U.S. census, studies com-
paring birth certificates and selective service rec-
ords to census results documented a sizable net
285
CENSUS
undercount. To provide more information on Barabba 1983). More litigation occurred with re-
census coverage and accuracy, a post-enumeration spect to the 1990 census. In neither case were the
survey was conducted following the 1950 census initially reported census counts adjusted to reflect
using specially selected and closely supervised enu- estimated errors in enumeration, although in the
merators. One finding from the survey was that case of the 1990 census the Census Bureau techni-
the undercount of infants in the census did not cal staff and director thought their methodology
arise, as first thought, from a tendency for new would improve the accuracy of the counts.
parents to forget to mention a new baby to the
census enumerator. The problem, rather, was that The Census Bureau developed innovative plans
many young couples and single parents had ir- for the 2000 census to integrate an evaluation
regular and difficult-to-find living arrangements, survey with the census, to use sampling to in-
and entire households were missed by census enu- crease quality and reduce costs of enumerating
merators. Based on this and related findings, much nonresponsive households, and to produce offi-
effort was given in subsequent censuses to precensus cial counts that already incorporate the best avail-
and postcensus review of lists of dwelling units. able procedures for minimizing error.
Special surveys preceding, accompanying, and These Census Bureau plans engendered ma-
following censuses are increasingly used to pro- jor political and budgetary battles between the
vide evidence on magnitude of error, location of Republican-controlled Congress and the Demo-
error in specific groups or places, and procedural cratic administration. Many of the persons hardest
means for reducing error. Techniques of ‘‘demo- to reach using traditional census methods are
graphic analysis’’ have also been developed and poor, often minorities. Many members of Con-
continually refined to provide evidence of census gress assumed that a census with near-zero
error. The basic technique is to analyze the num- undercount would increase the population report-
bers of persons of each age, sex, and race in ed for states and localities that tend to vote more
successive censuses. For example, the number of for Democrats, and hence reduce, relatively, the
twenty-eight year-old white women in the 1990 representation for states and localities that tend to
census should be consistent with the number of vote more for Republicans. At the national level,
eighteen-year-olds in 1980 and eight-year-olds in this could alter the reapportionment among states
1970. Information on births, deaths, immigration, of seats in the House of Representatives. Legisla-
and emigration is taken into account. Based on tures in states, counties, and cities are also subject
demographic analysis, the estimated net undercount to decennial redistricting according to the ‘‘one-
for the total U.S. population was about 5.6 percent person, one-vote’’ rule, and similar shifts in rela-
in 1940, 1.4 percent in 1980 (Fay, Passel, and tive political power could occur.
Robinson 1988, Table 3.2) and 1.8 percent in
1990. Estimates of net undercount have also been The controversy between legislative and ad-
made for specific age, sex, and race groupings. ministrative branches over methods for the 2000
census was taken to the Supreme Court, with
If the net undercount were uniform for all plaintiffs seeking a ruling that the Bureau’s plans
population groups and geographic regions, it would to use sample-based estimates would be unconsti-
not affect equity in the distribution of seats in tutional. The Supreme Court avoided the constitu-
Congress or public funds. But census errors are tional argument, but ruled that current law did not
not uniform. The estimates for 1990 show net permit such use of sampling. The Democrats lacked
undercounts exceeding 10 percent of adult black the power to rewrite that part of the basic census
males and small net overcounts for some age and law. The two parties compromised by appropriat-
race groups. ing additional dollars to cover the extra costs of
complete enumeration while retaining funds for
In 1980, the mayor of Detroit, the City of many of the methodological innovations that per-
Detroit, New York City, and others sued the feder- mit preparation of improved estimates. Reappor-
al government, alleging violation of their constitu- tionment among states of seats in Congress could
tional rights to equal representation and fair distri- therefore use the results of traditional complete
bution of federal funds (Mitroff, Mason, and enumeration. Most scholars, analysts, and other
286
CENSUS
users of census data are likely to consider the (SEE ALSO: Demography; Population)
adjusted estimates more accurate and hence
more useful. REFERENCES
The percentage net undercount understates Anderson, Margo J. 1988 The American Census: A Social
the total coverage error in the census. If one History. New Haven: Yale University Press.
person is omitted and another person counted
twice, the total count is correct but there are two ——— and Stephen E. Fienberg 1999 Who Counts? The
‘‘gross’’ errors that may affect the counts for spe- Politics of Census-Taking in Contemporary America. New
cific places and characteristics. Identification of York: Russell Sage Foundation.
gross errors is receiving increased attention, to
improve information on the nature of census er- Edmonston, Barry, and Charles Schultze 1995 Moderniz-
ror and potential error-reducing methods, and to ing the U.S. Census. Washington, D.C.: National Acade-
facilitate development of techniques of data analy- my Press.
sis that compensate for known and unknown errors.
Mitroff, Ian I., Richard O. Mason, and Vincent P. Barabba
One of the difficulties in debates about 1983 The 1980 Census: Policymaking and Turbulence.
undercount and other census error arises from Lexington, Mass.: Lexington Books.
confusion between the idea of a knowable true
count and the reality that there is no feasible Scott, Ann H. 1968 Census, USA: Fact Finding for the
method for determining with perfect accuracy the American People, 1790–1970. New York: Seabury Press.
size and characteristics of any large population. A
national census is a set of procedures adopted in a Starr, Paul 1987 ‘‘The Sociology of Official Statistics.’’ In
political, economic, and social context to produce W. Alonso and P. Starr, eds., The Politics of Numbers.
population estimates. It is politically convenient if New York: Russell Sage Foundation.
all parties accept these results as the common basis
for further action, much as sports contests use Steffey, Duane, and Norman Bradburn (eds.) 1994 Count-
decisions by umpires or referees. The more that ing People in the Information Age. Washington, D.C.:
policy and budgets depend on census results and National Academy Press.
the more aware that politicians and citizens be-
come of the importance of the census, the more United Nations 1998 Principles and Recommendations for
contentious census taking will be. Population and Housing Censuses. ST/ESA/STAT/
SER.M/67/Rev.1. New York: United Nations.
Every nation confronts political and social
problems with its censuses. Regularly scheduled United States Census Bureau. Census Bureau Home Page.
censuses are often postponed or abandoned be- On the Internet at http.//www.census.gov. See links
cause of international conflict. The United States to Census 2000 and American Community Survey.
and the United Kingdom canceled plans for mid-
decade censuses because of national budget KARL TAEUBER
constraints. West Germany was unable to take a
census for several years because of citizen fears CHANGE MEASUREMENT
about invasions of privacy. Ethnic conflict has
interfered with census taking in India, Lebanon, See Experiments; Longitudinal Research; Quasi-
and other nations. Experimental Research Design; Measurement.
The processes by which census procedures are CHILD ABUSE
determined, the ways in which census figures are
used, and the conflicts that occur about these See Childhood Sexual Abuse; Family Violence;
procedures and numbers are not merely ‘‘techni- Incest; Sexual Violence.
cal’’ but are embedded in a broader social process.
The character of a nation’s census and the con- CHILDBEARING
flicts that surround it are core topics for the ‘‘soci-
ology of official statistics’’ (Starr 1987). See Family Planning; Family Size; Fertility
Determinants; Pregnancy and Pregnancy
Termination.
287
CHILDHOOD SEXUAL ABUSE
CHILDHOOD SEXUAL ABUSE DEFINITION OF THE PROBLEM
Although child abuse is probably as old as child- Before delving into the field of childhood sexual
hood itself, serious research into child abuse ar- abuse, one must first understand what is meant by
guably began in 1962 with the publication of Kempe the term. Indeed, the lack of a standard definition
and colleagues’ seminal paper, ‘‘The Battered Child has been a major criticism of the field and contro-
Syndrome’’ (Kempe, Silverman, Steele, versies abound (Finkelhor 1994a; Gibbons 1996;
Droegemuller, and Silver 1962). Not long after- Gough 1996a; Green 1993). In general, the legal
wards, health care professionals began to direct and practical research definitions of child sexual
their attention to the specific problem of child- abuse require the following two elements: (1) sexu-
hood sexual abuse (Cosentino and Collins 1996). al activities involving a child (sometimes construed
In the past few decades, numerous instances have as including adolescence), and (2) the existence of
been documented in detailed case histories, and an ‘‘abuse condition’’ indicating lack of consensuality
important research into the causes and conse- (Faller 1993; Finkelhor 1994a). ‘‘Sexual activities’’
quences of childhood sexual abuse has been refer to behaviors intended for sexual stimulation;
initiated. such activities need not involve physical contact,
however, leading to separate definitions for ‘‘con-
Numerous extensive reviews have been pub- tact sexual abuse’’ and ‘‘noncontact sexual abuse.’’
lished that summarize what is presently known Contact sexual abuse includes both penetrative
about childhood sexual abuse, focusing on the (e.g., insertion of penis or other object into the
following domains. vagina or anus) and nonpenetrative (e.g., unwant-
ed touching of genitals) acts. Noncontact sexual
• Short-term effects (Beitchman et al. 1991; abuse refers to activities such as exhibitionism,
Beitchman et al. 1992; Briere and Elli- voyeurism, and child pornography (see Faller 1993
ott 1994; Browne and Finkelhor 1986; and Finkelhor 1994a for reviews on definitions of
Finkelhor 1990; Gomes-Schwartz et al. childhood sexual abuse).
1990; Green 1993; Kelley 1995; Kendall-
Tackett et al. 1993; Trickett and McBride- An abuse condition is said to exist when there
Chang, 1995). is reason to believe that the child either did not, or
was incapable of, consenting to sexual activity.
• Long-term consequences (Beitchman et al. Three main conditions can be distinguished: (1)
1991; Briere 1988; Briere and Elliott 1994; the perpetrator has a large age or maturational
Briere and Runtz 1991; Cahill et al.1991a; advantage over the child; or (2) the perpetrator is
Collings 1995; Ferguson 1997; Finkelhor in a position of authority or in a caretaking rela-
1987; Gibbons 1996; Glod 1993; Green tionship with the child; or (3) activities are carried
1993; Murray 1993; Polusny and Follette out against the child’s will using force or trickery.
1995; Trickett and McBride-Chang 1995; As evident from this brief summary, ‘‘Childhood
Wolfe and Birt, 1995). Sexual Abuse’’ covers a wide range of acts and
situations, and therefore is open to considerable
• Prevention of abuse (Adler and McCain subjective interpretation.
1994; Berrick and Barth 1992; MacMillan
et al. 1994; Olsen and Widom 1993; Wolfe INCIDENCE AND PREVALENCE OF
et al. 1995). CHILDHOOD SEXUAL ABUSE
• Treatment of both survivors and abusers Increasing attention has been directed toward
(Cahill et al. 1991b; Cosentino and Collins childhood sexual abuse not only because of the
1996; Faller 1993; Finkelhor and Berliner psychosocial sequelae associated with its occur-
1995; O’Donohue and Elliot 1993). rence, but also because it now appears to be more
widespread than previously thought (Adler and
Consequently, this chapter is not intended to McCain 1994). However, accurate estimates of the
provide a detailed analysis and review of the litera- occurrence of childhood sexual abuse are difficult
ture on childhood sexual abuse. Rather, it is meant
to serve as a brief overview of, and introduction to,
this area of inquiry.
288
CHILDHOOD SEXUAL ABUSE
to obtain because many cases are never reported. their lives (Ferguson 1997; Finkelhor 1994a; Gib-
Thus, available statistics represent only those cases bons 1996; Green 1993; Gorey and Leslie 1997).
reported to child protection agencies or to law Among the general adult female population, preva-
enforcement, and therefore underestimate the lence rates range from 2 percent to 62 percent
true magnitude of the problem. (Finkelhor 1987; Finkelhor et al. 1990; Saunders et
al. 1992). Rates are even higher within various
There are two official sources of incidence clinical populations, with 35 percent to 75 percent
data for cases in the United States: The National of female clients reporting a history of some form
Incidence Study of Child Abuse and Neglect (NIS), a of sexual abuse during childhood (Gibbons 1996;
federally funded research project; and statistics Wurr and Partridge 1996).
from state child protection agencies (Finkelhor
1994a). The NIS conducted in 1993 documented The number of males who have been sexually
over 300,000 cases of sexual abuse among children abused is difficult to estimate because it has been
known to professionals in the course of a year, or a the subject of fewer high-quality studies (Holmes
rate of approximately forty-five cases per 10,000 and Slap 1998; Watkins and Bentovim 1992). None-
children (Sedlak and Broadhurst 1996). Child pro- theless, a few studies have examined this issue
tective services data suggest that approximately among men and estimate that approximately 3
140,000 cases of childhood sexual abuse occur percent to 16 percent of all men in the United
annually, or twenty-one cases per 10,000 children States were sexually abused during childhood
(Leventhal 1998). An editorial by Finkelhor (1998) (Finkelhor et al. 1990; Gibbons 1996; Holmes and
suggested that the incidence may be declining. Slap 1998). As is the case for women, rates are
However, this remains to be substantiated. In higher when studying clinical populations, with
general, reports suggest that the rate of childhood estimates ranging from 13 percent to 23 percent
sexual abuse has increased substantially over the (Holmes and Slap 1998; Metcalfe et al. 1990).
past decades (Sedlak and Broadhurst 1996).
In sum, based on the evidence available in the
Retrospective surveys provide a second source literature, Finkelhor (1994a) estimates that 20 per-
of information on the occurrence of childhood cent of adult women and 5 percent to 10 percent
sexual abuse. Reported rates depend upon how it of adult men are survivors of childhood sexual
is defined and operationalized in any given survey. abuse. Similarly, a synthesis of sixteen cross-sec-
When childhood sexual abuse has been assessed tional surveys on the prevalence of it among
with a single item that narrowly defines it as rape nonclinical populations reported unadjusted esti-
or sexual intercourse, reported prevalence rates mates of 22.3 percent for women and 8.5 percent
tend to be low (e.g., less than 12 percent). Con- for men (Gorey and Leslie 1997).
versely, when it is defined more broadly (e.g.,
touching genitalia) and assessed using multiple SEQUELAE
items, prevalence rates tend to be much higher,
but with a wide range. The discrepancies observed Research conducted over the past decade indi-
in the estimated prevalence of childhood sexual cates that a wide range of psychological and inter-
abuse points to the need for increased standardiza- personal problems are more prevalent among those
tion and the use of better assessment instruments people who have been sexually abused than among
in this research. those who have not been sexually abused. Al-
though a definitive casual relationship between
Because rates of childhood sexual abuse are such problems and sexual abuse cannot be estab-
substantially greater among females than males lished using retrospective or cross-sectional re-
(e.g., Cosentino and Collins 1996; Finkelhor et al. search methodologies (Briere 1992a; Plunkett and
1990; Sedlak and Broadhurst 1996), the majority Oates 1990), the aggregate of consistent findings
of this research has focused on women. Surveys in this literature has led many investigators and
suggest that anywhere from 18 percent to 30 per- health care providers to conclude that childhood
cent of college women and 8 percent to 33 percent sexual abuse is a major risk factor for a variety of
of male and female high school students report problems, both in the short-term and in later
having experienced this abuse at some point in
289
CHILDHOOD SEXUAL ABUSE
adulthood (Briere and Elliott 1994; Faller 1993). history of childhood sexual abuse (Beitchman et
The various problems and symptoms described in al. 1992; Browne and Finkelhor 1986; Cahill et al.
the literature can be categorized as follows: (1) 1991a; Polusny and Follette 1995). Greater depres-
posttraumatic stress; (2) cognitive distortions; (3) sive symptomatology is also found among children
emotional distress; (4) impaired sense of self; (5) who have been abused (Lipovsky et al.1989; Yama
avoidance phenomena; (6) personality disorders; et al.1993).
and (7) interpersonal difficulties. In the remainder
of this section, each of these categories is defined Similarly, elevated anxiety levels have been
and illustrated using examples from the research documented in child victims of sexual abuse and
literature. adults who have a history of it (Gomes-Schwartz et
al. 1990; Mancini et al. 1995; Yama et al. 1993).
Posttraumatic stress refers to certain endur- Adults with a history of childhood sexual abuse are
ing psychological symptoms that occur in reaction more likely than their nonabused counterparts to
to a highly distressing, psychiatric disruptive event. meet the criteria for generalized anxiety disorder,
To be diagnosed with posttraumatic stress disor- phobias, panic disorder, obsessive compulsive dis-
der (PTSD) an individual must experience not order, or all of these (Mancini et al. 1995; Mulder
only a traumatic event, but also the following et al. 1998; Saunders et al. 1992).
problems: (1) frequent re-experiencing of the event
through nightmares or intrusive thoughts; (2) a Another common emotional sequel of child-
numbing of general responsiveness to, or avoid- hood sexual abuse is anger; specifically, chronic
ance of, current events; and (3) persistent symp- irritability, unexpected or uncontrollable feelings
toms of increased arousal, such as jumpiness, sleep of anger, difficulties associated with expressing
disturbances, or poor concentration (American anger, or all of these (Briere and Elliott 1994; Van
Psychiatric Association (APA) 1994). In general, der Kolk et al. 1994). During childhood and ado-
researchers have found that children and adults lescence, anger is most likely to be reflected in
who have been sexually abused are significantly behavioral problems such as fighting, bullying, or
more likely to receive a PTSD diagnosis than their attacking other children (Chaffin et al. 1997;
nonabused peers (McLeer et al. 1992; Murray Garnefski and Diekstra 1997).
1993; Rowan and Foy 1993; Saunders et al. 1992;
Wolfe and Birt 1995). Sexual abuse also can damage a child’s devel-
oping sense of self, with adverse long-term conse-
Cognitive distortions are negative perceptions quences. The development of a sense of self is
and beliefs held with respect to oneself, others, the thought to be one of the earliest developmental
environment, and the future. In the abused indi- tasks of the infant and young child, typically un-
vidual, this type of thought process is reflected in folding in the context of early relationships (Alex-
his or her tendency to overestimate the amount of ander 1992). Thus, how a child is treated early in
danger or adversity in the world and to underesti- life critically influences his or her growing self-
mate his or her worth (Briere and Elliott 1994; awareness. Childhood sexual abuse can interfere
Dutton et al. 1994; Janoff-Bulman 1992). Numer- with this process, preventing the child from estab-
ous studies document such feelings and percep- lishing a strong self image (Cole and Putnam
tions such as helplessness and hopelessness, im- 1992). Without a healthy sense of self, the person
paired trust, self-blame, and low self-esteem among is unable to soothe or comfort himself or herself
children who have been sexually abused (Oates et adequately, which may lead to overreactions to
al. 1985). Moreover, these cognitive distortions stressful or painful situations, and to an increased
often continue on into adolescence and adulthood likelihood of re-victimization (Messman and Long
(Gold 1986; Shapiro and Dominiak 1990). 1996). Indeed, numerous studies have found high
rates of sexual re-victimization (e.g., rape, coercive
Emotional distress or pain, which typically sexual experience) among individuals reporting a
manifests itself as depression, anxiety, anger, or all history of childhood sexual abuse (Fergusson et al.
of these, is reported by many survivors of child- 1997; Urquiza and Goodlin-Jones 1994; Wyatt et
hood sexual abuse. Depression is the most com- al. 1992)
monly reported symptom among adults with a
Avoidance is another major response to hav-
ing been sexually abused. Avoiding activities among
290
CHILDHOOD SEXUAL ABUSE
victims may be viewed as attempts to cope with the accommodation responses to ongoing abuse (e.g.,
chronic trauma and dysphoria induced by the passivity). Often, victims of childhood sexual abuse
abuse. Among the dysfunctional activities associat- know the perpetrator, who might be a family
ed with efforts to avoid recalling or reliving specif- member, clergy, or friend of the family. The abuse
ic memories are: dissociative phenomena, such as is thus a violation and betrayal of both personal
losing time, repression of unpleasant memories, and interpersonal (relationship) boundaries. Hence,
detachment, or body numbing (APA 1994; Cloitre it is not surprising that many children and adults
et al. 1997; Mulder et al. 1998; Nash et al. 1993); with a history of childhood sexual abuse are found
substance abuse and addiction (Arellano 1996; to be less socially competent, more aggressive, and
Briere 1988; Wilsnack et al. 1997); suicide or suici- more socially withdrawn than their nonabused
dal ideation (Briere and Runtz 1991; Saunders et peers (Cloitre et al. 1997; Mannarino et al. 1991;
al. 1992; Van der Kolk et al. 1991); inappropriate, Mullen et al. 1994).
indiscriminate, and/or compulsive sexual behav-
ior (McClellan et al. 1996; Widom and Kuhns, Among adults, interpersonal difficulties are
1996), (for reviews see Friedrich 1993; Kendall- manifested in difficulties in establishing and main-
Tackett et al. 1993; Tharinger 1990); eating disor- taining relationships (Finkelhor et al. 1989; Liem
ders (Conners and Morse 1993; Douzinas et al. et al. 1996) and in achieving sexual intimacy
1994; Schwartz and Cohn 1996; Wonderlich et al. (Browne and Finkelhor 1986; Mullen et al. 1994).
1997); and self-mutilation (Briere 1988; Briere and Children who have been sexually abused are more
Elliott 1994). Each of these behaviors serve to likely to exhibit increased or precocious sexual
prevent the individual from experiencing the con- behavior, such as kissing and inappropriate genital
siderable pain of abuse-specific awareness and touching (Cosentino et al. 1995).
thus reduces the distress associated with remem-
brance. However, avoidance and self-destructive The consequences of childhood sexual abuse
methods of coping with the abuse may ultimately can be quite severe and harmful. However, it is
lead to higher levels of symptomatology, lower important to note that an estimated 10 to 28
self-esteem, and greater feelings of guilt and anger percent of persons with a history of it report no
(Leitenberg et al. 1992). psychological distress (Briere and Elliott 1994;
Kendall-Tackett et al. 1993). This raises the ques-
Numerous investigations have found a greater tion of why some persons exhibit difficulties while
rate of borderline personality and dissociative iden- others do not. Research addressing this question
tity (formerly multiple personality disorder) disor- has found the following to be important mediators
ders among adult female survivors of chilhood of individual reactions to childhood sexual abuse:
sexual abuse (Green 1993; Polusny and Follette
1995; Silk et al. 1997). Borderline personality dis- • Age at onset of abuse. Although further
order includes symptoms of impulsiveness associ- elucidation is needed, the available evi-
ated with intense anger or suicidal, self-mutilating dence suggests that postpubertal abuse
behavior, and affective instability with depression is associated with greater trauma and
which are typical sequelae in sexually abused child- more severe adverse sequelae than is
ren and in adult survivors of sexual abuse (APA prepubertal abuse (Beitchman et al. 1992;
1994). Childhood trauma, especially continued Browne and Finkelhor 1986; McClellan et
sexual abuse, is an important etiological factor in al. 1996; Nash et al. 1993);
many cases of dissociative identity disorder, the
most extreme type of dissociative reaction (APA 1994). • Gender of the victim. One of the main
findings in this area is that male victims
Finally, interpersonal difficulties and deficient appear to show greater disturbances of
social functioning often are observed among indi- adult sexual functioning (Beitchman et al.
viduals who have been sexually abused (Briere 1992; Dube and Herbert 1988; Garnefski
1992b; Cloitre et al. 1997). Such difficulties stem and Diekstra 1997);
from the immediate cognitive and conditioned
responses to victimization that extend into the • Relationship to perpetrator. Abuse involving
longer term (e.g., distrust of others), as well as the a father or father figure (e.g., stepfather),
which accounts for an estimated 25 per-
cent of all cases (Sedlak and Broadhurst
291
CHILDHOOD SEXUAL ABUSE
1996), is associated with greater long- Nonetheless, cross-cultural studies can shed
term harm (Browne and Finkelhor 1986; new light on the origins and impact of sexual
Gold 1986); abuse (Leventhal 1998; Runyan 1998). Such inves-
tigations may enable us to understand better the
• Duration and frequency of abuse. In general, relative importance of different factors that influ-
the greater the frequency and duration of ence the occurrence of abuse and teach us about
abuse, the greater the impact on later societies that have been successful at protecting
psychological and social functioning (Nash children. Intercultural comparative investigations
et al. 1993); can also help us appreciate the range of sexual
behaviors that are, or can be considered ‘‘nor-
• Use of force. The use of force or threat of mal,’’ and thereby contribute to a better under-
force is associated with more negative standing of abnormality in childhood sexual be-
outcomes (Kendall-Tackett et al. 1993); havior and adult behaviors toward children.
• Penetration or invasiveness. Penetrative It is not surprising that definitions of abuse
abuse is generally associated with greater not only vary within a culture, but also between
long-term harm than are most other forms cultures. The meaning of ‘‘abuse,’’ especially, de-
of abuse (Kendall-Tackett et al. 1993); and pends upon ideas of individual rights and roles
and responsibilities between people and groups
• Family characteristics and response to abuse within society (Gough 1996a). How a child is viewed
disclosure. Individuals who have been will influence what is evaluated as abuse. Defini-
abused are more likely to originate from tions of child abuse would be quite different, for
single-parent families, families with a high example, in a feudal state in which children are
level of marital conflict, and families with considered to be their parents’ possessions. Cul-
pathology (e.g., parent is an alcoholic, tural expectations about sexual interactions among
violence between parents, maternal disbe- adolescents and between different age groups will
lief, and lack of support); all of which are also affect whether particular practices are defined
associated with a poorer outcome and as abuse (Abramson and Pinkerton 1995). Among
greater levels of distress (Beitchman et al. the Sambia people of the highlands of Papua, New
1992; Draucker 1996; Green 1996; Ro- Guinea, for example, all young boys are expected
mans et al. 1995). to participate in ritualized fellatio with older boys
as part of their initiation into manhood (Abramson
CROSS-CULTURAL ISSUES and Pinkerton 1995). In essence, the younger boys
are forced to submit; thus, this practice fulfills the
This review has focused exclusively on studies two main criteria for childhood sexual abuse as
done in the United States. This is important to defined above (coercive sex with a child or adoles-
keep in mind because the nature of abuse varies cent). Nevertheless, the Sambia consider ritual-
according to one’s cultural belief system (Gough ized fellatio to be critical for survival (they believe
1996). As a result, attempts to compare societies that semen is the source of manly strength and that
on the basis of their care of children or the extent it must be obtained through ingestion).
of violence in family relations are fraught with
problems (Gough 1996b; Levinson 1989). In addi- Definitions of abuse can also be expected to
tion, methodological factors, such as the questions change over time to reflect societal changes. For
asked and how childhood sexual abuse is defined instance, anecdotal evidence suggests that in Vic-
and measured, hamper the ability to make direct torian England it was considered acceptable for a
comparisons among the rates across different coun- nurse or nanny to quiet a male infant by putting his
tries. All that can be surmised, to date, is that it is penis in her mouth (Abramson and Pinkerton
not a phenomena just of the United States, but is 1995). Today, this practice would clearly be con-
an international problem (Finkelhor 1994b). sidered childhood sexual abuse. The recent broad-
Finkelhor’s synthesis indicates that most countries ening of definitions of abuse has been accompa-
have rates similar to those found in the United nied by a greater sensitivity to signs of abuse that
States and that females are abused at a greater fit these changing definitions, and a greater will-
rater than males. ingness for professionals and others to intervene
292
CHILDHOOD SEXUAL ABUSE
into the private family lives of others, or beyond defined. Since it was first brought to the public’s
the walls of institutional life, and so to offer great- attention in the 1960s, a vast amount of research
er visibility of children’s experiences (Gough 1996a). has been conducted in an effort to understand its
impact on victims. From the various reviews on the
DIRECTIONS FOR FUTURE RESEARCH topic, several conclusions can be drawn. First,
despite the considerable heterogeneity among sexu-
Perusal of the literature in this domain suggests al abuse victims, as a group, sexually abused child-
several avenues for future studies. First and fore- ren and adolescents tend to display significant-
most, large-scale, longitudinal investigations of child ly higher levels of symptomatology than their
victims are needed that examine global function- nonabused, nonclinic-referred peers. Second, com-
ing, abuse-specific functioning, and attributional pared to other clinic-referred children, two prob-
and coping strategies along with abuse, child, family, lem areas appear to differentiate sexually abused
and community factors. Such studies could pro- children and adolescents: post-traumatic stress dis-
vide information about both the initial effects of order symptomatology and sexuality problems.
sexual abuse and the factors that influence adjust- Third, the type and severity of sequelae experi-
ment at later developmental stages and into adult- enced by victims depends on the specific charac-
hood. Because not all victims of childhood sexual teristics of the abuse situation and the perpetrator.
abuse develop adjustment problems, a better un- Fourth, research on adult survivors suggest that
derstanding of who is likely to experience such abuse-related problems tend to persist into adult-
problems is needed. hood. Indeed, childhood sexual abuse histories
are common among several clinical populations,
Second, further research is needed regarding including patients initially diagnosed as depressed
the efficacy of various approaches for the treat- or as having a borderline personality disorder.
ment of related problems (see Cosentino and Col-
lins 1996; O’Donohue and Elliot 1993 for reviews). Taken together, these findings suggest that
The utility or effectiveness of any particular treat- clinicians and health care providers should screen
ment modality has yet to be demonstrated using a for sexual abuse among children, adolescents, and
large-scale, randomized study in which treatment adults. For instance, questions about childhood
outcomes are measured using standardized instru- sexual abuse could become part of routine intake
ments and untreated control groups. Consequent- procedures. Moreover, it is important that a wide
ly, treatment decisions often are made by clini- range of service providers are sensitive to, and
cians without empirically tested guidelines. have staff trained to deal with, issues of childhood
sexual abuse when it emerges. Service providers’
Third, greater attention needs to be paid to policy and practice guidelines should explicitly
methodological rigor in the conduct of childhood acknowledge the prevalence and impact of it on
sexual abuse studies (Green 1993; Plunkett and women, men, and children.
Oates 1990; Trickett and McBride-Change 1995).
In particular, investigators need to: (1) define it Schools also need to have greater involvement
clearly and consistently; (2) use instruments with in the prevention and diagnosis of childhood sexu-
documented psychometric properties; (3) use con- al abuse, especially since it generally occurs be-
trol or comparison groups, when appropriate; and tween the ages of six and twelve (Finkelhor 1994a;
(4) employ large sample sizes whenever this is Sedlak and Broadhurst 1996). School-based edu-
feasible. Studies of treatment modalities should cation programs may be a useful vehicle for inter-
also conduct multiple follow-up assessments to vention and prevention. Teachers need to be edu-
determine whether intervention effects are sus- cated and trained about their role in recognizing
tained over time. Finally, studies are needed that and reporting suspected cases.
clearly disentangle the effects of sexual abuse from
other forms of abuse or maltreatment. Finally, further community awareness is need-
ed to help prevent it from occurring. Greater
SUMMARY community efforts toward providing treatment
services for persons with a history of childhood
Childhood sexual abuse is a significant and wide- sexual abuse are needed as are programs targeting
spread problem in our society, no matter how it is
293
CHILDHOOD SEXUAL ABUSE
potential perpetrators. Society must work togeth- ———, and M. Runtz 1991 ‘‘The Long-term Effects of
er to stamp out this abhorrence and to assist its Sexual Abuse: A Review and Synthesis.’’ New Direc-
survivors. tions for Mental Health Services 51:3–13.
Preparation of this article was supported in part by Browne, A., and D. Finkelhor 1986 ‘‘Impact of Child
a training grant from the Maternal and Child Health Sexual Abuse: A Review of the Research.’’ Psychologi-
Bureau, U.S. Department of Health and Human Serv- cal Bulletin 99(1):66–77.
ices (MCJ 9040). The author thanks Steven D. Pinkerton,
Ph.D. for his helpful comments on this manuscript. Cahill, C., S.P. Llewelyn, and C. Pearson 1991a ‘‘Long-
term effects of Sexual Abuse Which Occurred in
REFERENCES Childhood: A Review.’’ British Journal of Clinical Psy-
chology 30(2):117–130.
Abramson, P.R., and S.D. Pinkerton 1995 With pleasure:
Thoughts on the Nature of Human Sexuality. New York: ——— 1991b ‘‘Treatment of sexual abuse which oc-
Oxford University Press. curred in childhood: A review.’’ British Journal of
Clinical Psychology 30(1):1–12.
Adler, N.A., and J.L. McCain 1994 ‘‘Prevention of Child
Abuse. Issues for the Mental Health Practitioner.’’ Chaffin, M., J.N. Wherry, and R. Dykman 1997 ‘‘School
Child and Adolescent Psychiatric Clinics of North America Age Children’s Coping with Sexual Abuse: Abuse
3(4):679–693. Stresses and Symptoms Associated with Four Coping
Strategies.’’ Child Abuse and Neglect 21(2):227–240.
Alexander, P.C. 1992 ‘‘Application of Attachment Theory
to the Study of Sexual Abuse.’’ Journal of Consulting Cloitre, M., P. Scarvalone, and J.A. Difede 1997
and Clinical Psychology 60:185–195. ‘‘Posttraumatic Stress Disorder, Self- and Interper-
sonal Dysfunction Among Sexually Retraumatized
American Psychiatric Association 1994 Diagnostic and Women.’’ Journal of Traumatic Stress 10(3):437–452.
Statistical Manual of Mental Disorders, 4th ed. Wash-
ington, D.C.: Author. Cole, P.M., and F.W. Putnam 1992 ‘‘Effect of Incest of
Self and Social Functioning: A Developmental Psy-
Arellano, C.M. 1996 ‘‘Child Maltreatment and Sub- chopathology Perspective.’’ Journal of Consulting and
stance use: A Review of the Literature.’’ Substance Use Clinical Psychology 60:174–184.
and Misuse 31(7):927–935.
Collings, S.J. 1995 ‘‘The Long-term Effects of Contact
Beitchman, J.H., K.J. Zucker, J.E. Hood, G.A. DaCosta, and Noncontact Forms of Child Sexual Abuse in a
and D. Akman 1991 ‘‘A Review of the Short-term Sample of University Men.’’ Child Abuse and Neglect
Effects of Childhood Sexual Abuse.’’ Child Abuse and 19(1):1–6.
Neglect 15:537–556.
Connors, M.E., and W. Morse 1993 ‘‘Sexual Abuse and
———, and E. Cassavia 1992 ‘‘A review of the long-term Eating Disorders: A review.’’ International Journal of
effects of child sexual abuse.’’ Child Abuse and Neglect Eating Disorders 13(1):1–11.
16:101–118.
Cosentino, C.E., and M. Collins 1996 ‘‘Sexual Abuse of
Berrick, J.D., and R.P. Barth 1992 ‘‘Child Sexual Abuse Children: Prevalence, Effects, and Treatment.’’ An-
Prevention: Research Review and Recommendations.’’ nals of the New York Academy of Sciences 789:45–65.
Social Work Research and Abstracts 28(4):6–15.
Cosentino, C.E., H.F. Meyer-Bahlburg, J.L. Alpert, S.L.
Briere, J. 1988 ‘‘The Long-term Clinical Correlates of Weinberg, and R. Gaines 1995 ‘‘Sexual Behavior
Childhood Sexual Victimization.’’ Annals of the New Problems and Psychopathology Symptoms in Sexual-
York Academy of Sciences 528:327–334. ly Abused Girls.’’ Journal of the American Academy of
Child and Adolescent Psychiatry 34(8):1033–1042.
——— 1992a ‘‘Methodological Issues in the Study of
Sexual Abuse Effects.’’ Journal of Consulting and Clini- Douzinas, N., V. Fornari, B. Goodman, T. and Sitnick
cal Psychology 60:196–203. 1994 ‘‘Eating Disorders and Abuse.’’ Child and Ado-
lescent Psychiatric Clinics of North America 3(4):777–796.
——— 1992b Child Abuse Trauma: Theory and Treatment
of the Lasting Effects. Newbury Park, Calif.: Sage Draucker, C.B. 1996 ‘‘Family-of-Origin Variables and
Publications. Adult Female Survivors of Childhood Sexual Abuse:
A review of the Research.’’ Journal of Child Sexual
———, and D.M. Elliott 1994 ‘‘Immediate and Long- Abuse 5(4):35–63.
term Impacts of Child Sexual Abuse.’’ Sexual Abuse of
Children 4(2):54–69. Dube, R., and M. Hebert 1988 ‘‘Sexual Abuse of Child-
ren Under 12 Years of Age: A Review of 511 Cases.’’
Child Abuse and Neglect 12(3):321–330.
294
CHILDHOOD SEXUAL ABUSE
Dutton, M.A., K.J. Burghardt, S.G. Perrin, K.R. Chrestman, Gibbons, J. 1996 ‘‘Services for Adults who Have Experi-
and P.M. Halle 1994 ‘‘Battered Women’s Cognitive enced Child Sexual Assault: Improving Agency Re-
Schemata.’’ Journal of Traumatic Stress 7(2):237–255. sponse.’’ Social Science and Medicine 43(12):1755–1763.
Faller, K.C. 1993 Child Sexual Abuse: Intervention and Glod, C.A. 1993 ‘‘Long-term Consequences of Child-
Treatment Issues. Washington, D.C.: U.S. Dept. of hood Physical and Sexual Abuse.’’ Archives of Psychiat-
Health and Human Services, Administration for Child- ric Nursing 7(3):163–173.
ren and Families, Administration on Children, Youth,
and Families, National Center on Child Abuse and Gold, E.R. 1986 ‘‘Long-term Effects of Sexual Victimiza-
Neglect. tion in Childhood: An Attributional Approach.’’ Jour-
nal of Consulting and Clinical Psychology 54:471–471.
Ferguson, A.G. 1997 ‘‘How Good is the Evidence Relat-
ing to the Frequency of Childhood Sexual Abuse and Gomes-Schwartz, B., J.M. Horowitz, and A.P. Cardarelli
the Impact Such Abuse has on the Lives of Adult 1990 Child Sexual Abuse: The Initial Effects. Newbury
Survivors?’’ Public Health 111(6):387–391. Park, Calif.: Sage Publications.
Fergusson, D.M., L.J. Horwood, and M.T. Lynskey 1997 Gorey, K.M., and D.R. Leslie 1997 ‘‘The Prevalence of
‘‘Childhood Sexual Abuse, Adolescent Sexual Behav- Child Sexual Abuse: Integrative Review Adjustment
iors and Sexual Revictimization.’’ Child Abuse and for Potential Response and Measurement Biases.’’
Neglect 21(8):789–803. Child Abuse and Neglect 21(4):391–398.
Finkelhor, D. 1987 ‘‘The Sexual Abuse of Children: Gough, D. 1996a ‘‘Defining the Problem: Comment.’’
Current Research Reviewed.’’ Psychiatric Annals Child Abuse and Neglect 20(11):993–1002.
17:233–241.
——— 1996b ‘‘Child Abuse in Japan.’’ Child Psychology
——— 1990 ‘‘Early and Long-term Effects of Child and Psychiatry Review 1:12–18.
Sexual Abuse: An Update.’’ Professional Psychology
21(5):325–330. Green, A.H. 1993 ‘‘Child Sexual Abuse: Immediate and
long-term Effects and Intervention.’’ Journal of the
——— 1994a ‘‘Current Information on the Scope and American Academy of Child and Adolescent Psychiatry
Nature of Child Sexual Abuse.’’ Future of Children 32(5):890–902.
4(2):31–53.
——— 1996 ‘‘Overview of Child Sexual Abuse.’’ In S. J.
——— 1994b ‘‘The International Epidemiology of Child Kaplan, ed., Family Violence: A Clinical and Legal
Sexual Abuse.’’ Child Abuse and Neglect 18(5):409–417. Guide. Washington, D.C.: American Psychiatric
Press, Inc.
——— 1998 ‘‘Improving Research, Policy, and Practice
to Understand Child Sexual Abuse.’’ Journal of the Holmes, W.C., and G.B. Slap 1998 ‘‘Sexual Abuse of
American Medical Association 280(21):1864–1865. Boys. Definition, Prevalence, Correlates, Sequelae,
and Management.’’ Journal of the American Medical
———, and L. Berliner 1995 ‘‘Research on the Treat- Association 280(21):1855–1862.
ment of Sexually Abused Children: A Review and
Recommendations.’’ Journal of the American Academy Janoff-Bulman, R. 1992 Shattered Assumptions: Towards a
of Child and Adolescent Psychiatry 34(11):1408–1423. new Psychology of Trauma. New York: Free Press.
Finkelhor, D., G. Hotaling, I.A. Lewis, and C. Smith Kelley, S.J. 1995 ‘‘Child Sexual Abuse: Initial Effects.’’
1990 ‘‘Sexual Abuse in a National Survey of Adult Annual Review of Nursing Research 13:63–85.
Men and Women: Prevalence, Characteristics, and
Risk Factors.’’ Child Abuse and Neglect 14(1):19–28. Kempe, C.H., F.N. Silverman, B.F. Steele, W. Droegemuller,
and A.K. Silver 1962 ‘‘The Battered Child Syndrome.’’
——— 1989 ‘‘Sexual Abuse and its Relationship to Journal of the American Medical Association 187:17–24.
Later Sexual Satisfaction, Marital Status, Religion,
and Attitudes.’’ Journal of Interpersonal Violence 4:279–399. Kendall-Tackett, K.A., L.M. Williams, and D. Finkelhor
1993 ‘‘Impact of Sexual Abuse on Children: A Re-
Friedrich, W.N. 1993 ‘‘Sexual Victimization and Sexual view and Synthesis of Recent Empirical Studies.’’
Behavior in Children: A Review of Recent Litera- Psychological Bulletin 113:164–180.
ture.’’ Child Abuse and Neglect 17(1):59–66.
Leitenberg, H., E. Greenwald, and S. Cado 1992 ‘‘A
Garnefski, N., R.F. and Diekstra 1997 ‘‘Child Sexu- Retrospective Study of Long-term Methods of Cop-
al Abuse and Emotional and Behavioral Problems ing with Having Been Sexually Abused During Child-
in Adolescence: Gender Differences.’’ Journal of the hood.’’ Child Abuse and Neglect 16(3):399–407.
American Academy of Child and Adolescent Psychiatry
36(3):323–329. Leventhal, J.M. 1998 ‘‘Epidemiology of Sexual Abuse of
Children: Old Problems, New Directions.’’ Child Abuse
and Neglect 22(6):481–491.
295
CHILDHOOD SEXUAL ABUSE
Levinson, D. 1989 Family Violence in Cross Cultural Per- Psychological Impairment Among Children.’’ Child
spective. Newbury Park, Calif.: Sage. Abuse and Neglect 17(3):401–408.
Liem, J.H., J.G. O’Toole, and J.B. James 1996 ‘‘Themes Oates, R.K., D. Forest, and A. Peacock 1985 ‘‘Self-
of Power and Betrayal in Sexual Abuse Survivors’ esteem of Abused Children.’’ Child Abuse and Neglect
Characterizations of Interpersonal Relationships.’’ 9:159–163.
Journal of Traumatic Stress 9(4):745–761.
O’Donohue, W., A. and Elliot 1993 ‘‘Treatment of the
Lipovsky, J.A., B.E. Saunders, and S.M. Murphy 1989 Sexually Abused Child: A review.’’ Journal of Clinical
‘‘Depression, Anxiety, and Behavior Problems Among Child Psychology 3:218–228.
Victims of Father-child Sexual Assault and Nonabused
Siblings.’’ Journal of Interpersonal Violence 4:452–468. Olsen, J.L., and C.S. Widom 1993 ‘‘Prevention of Child
Abuse and Neglect.’’ Applied and Preventive Psychology
MacMillan, H.L., J.H. MacMillan, D.R. Offord, L. Grif- 2(4):217–229.
fith, and A. MacMillan 1994 ‘‘Primary Prevention of
Child Sexual Abuse: A Critical Review: II.’’ Journal of Plunkett, A., and R.K. Oates 1990 ‘‘Methodological
Child Psychology and Psychiatry and Allied Disciplines Considerations in Research on Child Sexual Abuse.’’
35(5):857–876. Pediatric and Perinatal Epidemiology 4(3):351–360.
Mancini, C., M. Van Ameringen, and H. MacMillan Polusny, M.A., and V.M. Follette 1995 ‘‘Long-term Cor-
1995 ‘‘Relationship of Childhood Sexual and Physi- relates of Child Sexual Abuse: Theory and Review of
cal Abuse to Anxiety Disorders.’’ Journal of Nervous the Empirical Literature.’’ Applied and Preventive Psy-
and Mental Disease 183(5):309–314. chology 4:143–166.
Mannarino, A.P., J.A. Cohen, J.A. Smith, and S. Moore- Romans, S.E., J.L. Martin, J.C. Anderson, M.L. O’Shea,
Motily 1991 ‘‘Six- and Twelve-month Follow-up of and P.E. Mullen 1995 ‘‘Factors that Mediate Between
Sexually Abused Girls.’’ Journal of Interpersonal Vio- Child Sexual Abuse and Adult Psychological Out-
lence 6:494–511. come.’’ Psychological Medicine 25(1):127–142.
McClellan, J., C. McCurry, M. Ronnei, J. Adams, A. Rowan, A.B., and D.W. Foy 1993 ‘‘Post-traumatic Stress
Eisner, and M. Storck 1996 ‘‘Age of Onset of Sexual Disorder in Child Sexual Abuse Survivors: A Litera-
Abuse: Relationship to Sexually Inappropriate Be- ture Review.’’ Journal of Traumatic Stress 6(1):3–20.
haviors.’’ Journal of the American Academy of Child and
Adolescent Psychiatry 35(10):1375–1383. Runyan, D.K. 1998 ‘‘Prevalence, Risk, Sensitivity, and
Specificity: A Commentary on the Epidemiology of
McLeer, S.V., E. Deblinger, D. Henry, and H. Orvaschel Child Sexual Abuse and the Development of a Re-
1992 ‘‘Sexually Abused Children at High Risk for search Agenda.’’ Child Abuse and Neglect 22(6):493–498.
Post-traumatic Stress Disorder.’’ Journal of the Ameri-
can Academy of Child and Adolescent Psychiatry 31:875–879. Saunders, B.E., L.A. Villeponteaux, J.A. Lipovsky, D.G.
Kilpatrick, and J.L. Veronen 1992 ‘‘Child Sexual
Messman, T.L., and P.J. Long 1996 ‘‘Child Sexual Abuse Assault as a Risk Factor for Mental Disorders Among
and its Relationship to Revictimization in Adult Wom- Women: A community Survey.’’ Journal of Interper-
en: A Review.’’ Clinical Psychology Review 16(5):397–420. sonal Violence 7:189–204.
Mulder, R.T., A.L. Beautrais, P.R. Joyce, and D.M. Schwartz, M.F., and L. Cohn 1996 Sexual Abuse and
Fergusson 1998 ‘‘Relationship Between Dissociation, Eating Disorders: A Clinical Overview. New York:
Childhood Sexual Abuse, Childhood Physical Abuse, Brunner/Mazel.
and Mental Illness in a General Population Sample.’’
American Journal of Psychiatry 155(6):806–811. Sedlak, A.J., and D.D. Broadhurst 1996 Third national
incidence study of child abuse and neglect. Washington,
Mullen, P.E., J.L. Martin, J.C. Anderson, S.E. Romans, D.C.: U.S. Dept. of Health and Human Services,
and G.P. Herbison 1994 ‘‘The Effect of Child Sexual Administration for Children and Families, Adminis-
Abuse on Social, Interpersonal and Sexual Function tration on Children, Youth, and Families, National
in Adult life.’’ British Journal of Psychiatry 165(2):35–47. Center on Child Abuse and Neglect.
Murray, J.B. 1993 ‘‘Relationship of Childhood Sexual Shapiro, S., and G. Dominiak 1990 ‘‘Common Psycho-
Abuse to Borderline Personality Disorder, Posttraumatic logical Defenses Seen in the Treatment of Sexually
Stress Disorder, and Multiple Personality Disorder.’’ Abused Adolescents.’’ American Journal of Psychotherapy
Journal of Psychology 127(6):657–676. 44(1):68–74.
Nash, M.R., O.A. Zivney, and T. Hulsey 1993 ‘‘Charac- Silk, K.R., J.T. Nigg, D. Esten, and N.E. Lohr 1997
teristics of Sexual Abuse Associated with Greater ‘‘Severity of Childhood Sexual Abuse, Borderline
Symptoms, and Familial Environment.’’ In M.C.
296
CHINA STUDIES
Zanarini, ed., Role of Sexual Abuse in the Etiology of Yama, M.F., S.L. Tovey, and B.S. Fogas 1993 ‘‘Child-
Borderline Personality Disorder Washington, D.C.: hood Family Environment and Sexual Abuse as Pre-
American Psychiatric Press, Inc. dictors of Anxiety and Depression in Adult Wom-
en.’’ American Journal of Orthopsychiatry 63(1):136–141.
Tharinger, D. 1990 ‘‘Impact of Child Sexual Abuse on
Developing Sexuality.’’ Professional Psychology: Research HEATHER CECIL
and Practice 21(5):331–337.
CHILDREARING
Trickett, P.K., and C. McBride-Chang 1995 ‘‘The Devel-
opmental Impact of Different Forms of Child Abuse See Parental Roles; Socialization.
and Neglect.’’ Developmental Review 15(3):311–337.
CHINA STUDIES
Urquiza, A.J., and B.L. Goodlin-Jones 1994 ‘‘Child Sexu-
al Abuse and Adult Revictimization with Women of The sociological study of China has a complex
Color.’’ Violence and Victims 9(3):223–232. history. It could be argued that Confucian thought
embodied a native tradition of sociological think-
Van der Kolk, B.A., J.C. Perry, and J.L. Herman 1991 ing about such things as the family, bureaucracy,
‘‘Childhood Origins of Self-destructive Behavior.’’ and deviant behavior. However, modern Chinese
American Journal of Psychiatry 148(12):1665–1671. sociology initially had foreign origins and inspira-
tion. The appearance of the field in China might
Van der Kolk, B.A., A. Hostetler, N. Herron, and R.E. be dated from the translation of parts of Herbert
Fisler 1994 ‘‘Trauma and the Development of Bor- Spencer’s The Study of Sociology into Chinese in
derline Personality Disorder.’’ Psychiatric Clinics of 1897. The earliest sociology courses and depart-
North America 17(4):715–730. ments in China were established in private, mis-
sionary colleges, and Western sociologists such as
Watkins, B., and A. Bentovim 1992 ‘‘The Sexual Abuse D. H. Kulp, J. S. Burgess, and Sidney Gamble
of Male Children and Adolescents: A Review of played central roles in initiating sociology courses
Current Research.’’ Journal of Child Psychology and and research programs within China (see Wong 1979).
Psychiatry and Allied Disciplines 33(1):197–248.
The Chinese Sociological Association was es-
Widom, C.S., and J.B. Kuhns 1996 ‘‘Childhood Victimi- tablished in 1930, and in the following two dec-
zation and Subsequent Risk for Promiscuity, Prosti- ades a process of Sinification progressed. Chinese
tution, and Teenage Pregnancy: A Prospective Study.’’ sociologists, trained both at home and in the West,
American Journal of Public Health 86(11):1607–1612. emerged. Sociology courses and departments be-
gan to proliferate in government-run, public col-
Wilsnack, S.C., N.D. Vogeltanz, A.D. Klassen, and T.R. leges. Increasingly, texts using material from Chi-
Harris 1997 ‘‘Childhood Sexual Abuse and Women’s nese towns and villages displaced ones that focused
Substance Abuse: National survey findings.’’ Journal on Chicago gangs and the assimilation of Polish
of Studies on Alcohol 58(3):264–271. immigrants into America. As this process of set-
ting down domestic roots continued, Chinese soci-
Wolfe, D.A., N.D. Reppucci, and S. Hart 1995 ‘‘Child ology developed some distinctive contours. Inspi-
Abuse Prevention: Knowledge and Priorities.’’ Jour- ration for the field came as much from British
nal of Clinical Child Psychology 24(Suppl):5–22. social anthropology as from sociology. The crisis
atmosphere of the 1930s and 1940s fostered a
Wolfe, V.V., and J.A. Birt 1995 ‘‘The Psychological strong emphasis on applied, problem-solving re-
Sequelae of Child Sexual Abuse.’’ Advances in Clinical search. As a result of these developments, Chinese
Child Psychology 17:233–263. sociology in those years came to include what in
the United States might be considered social an-
Wonderlich, S.A., T.D. Brewerton, Z. Jocic, B.S. Dansky, thropology and social work. Within these broad
and D.W. Abbott 1997 ‘‘Relationship of Childhood contours, Chinese sociologists/anthropologists pro-
Sexual Abuse and Eating Disorders.’’ Journal of the duced some of the finest early ethnographies of
American Academy of Child and Adolescent Psychiatry
36(8):1107–1115.
Wurr, C.J., and I.M. Partridge 1996 ‘‘The Prevalence of
a History of Childhood Sexual Abuse in an Acute
Adult Inpatient Population.’’ Child Abuse and Neglect
20(9):867–872.
Wyatt, G.E., D. Guthrie, and C.M. Notgrass 1992 ‘‘Dif-
ferential Effects of Women’s Child Sexual Abuse and
Subsequent Sexual Revictimization.’’ Journal of Con-
sulting and Clinical Psychology 60(2):167–173.
297
CHINA STUDIES
‘‘nonprimitive’’ communities—Chinese peasant vil- 1957 to get the ban lifted. In the more relaxed
lages (see Fei 1939; Yang 1944; Lin 1948). political atmosphere prevailing at that time, ush-
ered in by the CCP’s slogan ‘‘let 100 flowers bloom
Although many of the earliest leaders in estab- and 100 schools of thought contend’’ and encour-
lishing Chinese sociology were foreigners, by the aged by the revival of sociology in the Soviet
time of the Chinese communist victory in 1949, its Union, leading figures such as Fei Xiaotong, Pan
leading practitioners were Chinese. There were Guangdan, Wu Jingchao, and Chen Da wrote arti-
only a small number of Western sociologists who cles and made speeches arguing that sociology
concentrated their research on Chinese society could be useful to socialist China in the study of
(see Lang 1946; Levy 1949), and their methods, social change and social problems— precisely the
problems, and data were not distinctive from those same rationale used prior to 1952. When the
being employed by their Chinese counterparts. political atmosphere turned harsh again in the
Originally a foreign transplant, Chinese sociology latter part of 1957, with the launching of the ‘‘anti-
had become a thriving enterprise with increasingly rightist campaign,’’ Fei and many of the other
strong domestic roots. leaders of this revival effort were branded ‘‘rightist
elements’’ and disappeared from view.
When the Chinese Communist Party (CCP)
swept to national power in 1949, some Chinese From 1952 to 1979 some of those trained in
sociologists left the country but most remained. sociology were assigned to work that had some
(The development of sociology in the Republic of links to their abolished field. Throughout this
China on Taiwan after 1949 will not be dealt with period ethnology remained an established disci-
here.) Initially, those who remained were optimis- pline, devoted to the study of China’s various
tic that their skills would be useful to the new ethnic minorities. When Fei Xiaotong emerged
government. Experience in community fieldwork from his political purgatory in 1972, it was as a
and an orientation toward studying social prob- leading figure in ethnology. But Fei and others
lems seemed to make Chinese sociologists natural could not do research on the 94 percent of the
allies of those constructing a planned social order. population that is Han Chinese, and they also had
These hopes were dashed in 1952 when the CCP to renounce publicly all their former work and
abolished the field of sociology. That decision was ideas as ‘‘bourgeois.’’ Psychology survived better
motivated by the CCP’s desire to follow the Soviet than sociology by becoming defined as a natural,
model. Joseph Stalin had earlier denounced soci- rather than a social, science. Some areas of social
ology as a ‘‘bourgeois pseudo-science’’ and banned psychology (for example, educational psychology)
the field from Soviet academe. More to the point, remained at least somewhat active through the
there was no room in the People’s Republic of 1960s and 1970s (see Chin and Chin 1969). Other
China (PRC) for two rival sciences of society, components of sociology disappeared in 1952 but
Marxism-Leninism-Maoism and sociology. The CCP reappeared prior to 1979. Demography was estab-
argued that it had developed its own methods of lished as a separate field in the early 1970s and
‘‘social investigations’’ during the revolutionary remains a separate discipline in Chinese academe
process, with Mao Zedong playing a leading role in today. With these partial exceptions, however,
this development (see Mao [1927] 1971; Thomp- Chinese sociology ceased to exist for these twenty-
son 1990). This approach stressed grass roots in- seven years.
vestigations designed to further official revolu-
tionary or economic goals of the CCP rather than While sociology was banned in China, the
any sort of attempted value-free search for objec- sociological study of that society developed gradu-
tive truth. Chinese sociologists, trained in a differ- ally in the West. Starting in the late 1950s, Ameri-
ent tradition and able to use professional claims to can foundations and government agencies, and to
raise questions about CCP policies, were seen as a some extent their European and Japanese coun-
threat to the ideological hegemony of the new regime. terparts, provided funds for the development of
the study of contemporary China—for fellowships,
The ban on sociology in China was to last research centers, journals, language training facili-
twenty-seven years, until 1979. One major attempt ties, and other basic infrastructure for the field.
was made by Chinese sociologists in 1956 and Within this developmental effort, sociology was
298
CHINA STUDIES
targeted as an underdeveloped discipline (com- an unrealistic impression of uniformity, thus rein-
pared to, say, history, literature, and political sci- forcing the prevailing stereotype of China as a
ence) deserving of special priority. Stimulated by totalitarian society.
this developmental effort, during the 1960s and
1970s the number of Western sociologists active in Partly in reaction to these limitations, a sec-
the study of China increased from a handful to ond primary method of research on China from
perhaps two dozen. the outside developed—the refugee interview.
Scholars in sociology and related fields spent ex-
Prior to the early 1970s, with a few minor tended periods of time in Hong Kong interviewing
exceptions (e.g., Geddes 1963), it was impossible individuals who had left the PRC for the British
for foreign sociologists to travel to China, much colony. Typically, these individuals were used not
less to do research there. Even when such travel as respondents in a survey but as ethnographic
became possible during the 1970s, it was initially informants ‘‘at a distance’’ about the corners of
restricted to very superficial ‘‘academic tourism.’’ the Chinese social world from which they had
In this highly constrained situation, foreign soci- come. Although the individuals interviewed obvi-
ologists studying China developed distinctive re- ously were not typical of the population remaining
search methods. in the PRC, for some topics these refugees provid-
ed a rich and textured picture of social reality that
Much research was based on detailed culling contrasted with the unidimensional view conveyed
of Chinese mass media reports. Special translation by Chinese mass media. Elaborate techniques were
series of Chinese newspapers and magazines and developed by researchers to cope with the prob-
of monitored radio broadcasts developed in the lems of atypicality and potential bias among refu-
1950s and became the mainstays of such research. gee informants (see Whyte 1983).
The Chinese media provided fairly clear state-
ments about official policy and social change cam- By combining these methods with brief visits
paigns. By conveying details about local ‘‘positive to China, sociologists and others produced a large
and negative models’’ in the implementation of number of useful studies of contemporary main-
official goals, the media also provided some clues land society (Whyte, Vogel, and Parish 1977). Even
about the difficulties and resistance being encoun- though the arcane methods they used often seemed
tered in introducing social change. In a few in- to nonspecialists akin to Chinese tea-leaf reading,
stances the media even yielded apparently ‘‘hard’’ the best of these studies have stood the test of time
statistical data about social trends. A minor indus- and have been confirmed by new research made
try developed around scanning local press and possible by changes in China since 1979. However,
radio reports for population totals, and these were the exclusion from meaningful field research in
pieced together to yield national population esti- China did have unfortunate effects on the intellec-
mates (typically plus or minus 100 million). tual agendas of researchers. Much energy was
focused on examining the ‘‘Maoist model’’ of de-
Some researchers developed elaborate schemes velopment and whether or not China was succeed-
to code and subject to content analysis such media ing in becoming a revolutionary new type of social
reports (see, for example, Andors 1977; Cell 1977). order. Since the reality of Chinese social organiza-
However, this reliance on media reports to study tions in the Maoist period was quite different from
Chinese social trends had severe limits. Such re- the slogans and ideals upon which such construc-
ports revealed much more about official goals tions were based, this effort appears to have been a
than they did about social reality, and the research- waste of scarce intellectual resources. Both this
er might erroneously conclude that an effort to unusual intellectual agenda and the distinctive
produce major changes was actually producing research methods sociologists studying China were
them. The sorts of unanticipated consequences forced to use reinforced the isolation of these
that sociologists love to uncover were for the most researchers from ‘‘mainstream’’ work in sociology.
part invisible in the highly didactic Chinese media.
It was also very difficult to use media reports to The year 1979 was significant in two ways for
analyze variations across individuals and commu- the sociological study of China. First, as already
nities, the staple of sociological analysis elsewhere. noted, that was the year in which China’s leaders
Studies based on media reports tended to convey
299
CHINA STUDIES
gave approval for the ‘‘rehabilitation’’ of sociolo- rural development he had championed a genera-
gy. After the death of Mao Zedong in 1976, and tion earlier.) A number of departments of sociolo-
given the authority crisis that followed in the wake gy were established in leading universities, and
of the Cultural Revolution, the post-Mao leader- new sociological journals and a large number of
ship reopened many issues. China faced a number monographs and textbooks began to appear. Insti-
of serious social problems that had been ignored tutes of sociology were established within the Chi-
previously, and in this situation long-standing ar- nese Academy of Social Science in Peking as well
guments about the utility of sociology for the study as in many provincial and city academies. Appar-
of social problems could be raised once again. ently, the novelty of the field, the chance to make a
place for oneself, and perhaps even the chance to
The second major change that occurred in study abroad without having to face competition
1979 was the ‘‘normalization’’ of diplomatic rela- from layers of senior people more than compen-
tions between the United States and China. That sated for the checkered political past of sociology.
development led to approval for American and In any case, the speed and vigor with which the
other Western researchers to conduct extended field revived surprised the skeptics (see Whyte and
field research in China for the first time since Pasternak 1980).
1949. The days of frustrating confinement to re-
search from outside and academic tourism were The intellectual focus of the revived sociology
over. It also became possible for U.S. sociologists was fairly broad. Social problems and the sociolo-
to meet Chinese counterparts and to begin to gy of development were two major foci. Within
discuss common research interests and plans for these and other realms, two sets of issues loomed
collaboration. Westerners began coming to China large. One was the study of the after-effects of the
to teach sociology courses, Chinese began to be Cultural Revolution and the radical policies pur-
sent abroad for training in sociology, and collabo- sued under Mao Zedong in the period 1966 to
rative conferences and publications were initiated. 1976. The other major issue concerned the impact
Leading figures in the West, including Peter Blau, of the opening to the outside world, decollectivization
Alex Inkeles, Nan Lin, W. J. Goode, Hubert Blalock, of agriculture, and the market-oriented reforms
and Jeffrey Alexander, lectured in China and re- launched in 1978. A large number of studies began
cruited Chinese courterparts for collaborative re- on these and other issues, and by the end of the
search and for training in their home institutions. 1980s the problem facing foreign sociologists study-
All of the kinds of intellectual interchange be- ing China was not so much the scarcity of data on
tween Chinese sociologists and their foreign coun- their object of study but its abundance. The publi-
terparts that had been impossible for a generation cation of large amounts of statistical material, the
were resumed. availability of census and survey data dealing with
demographic and other matters, and the mush-
The dangerous political aura surrounding so- rooming of social science publications threatened
ciology in the Mao era and the lesson of the to overwhelm those who tried to keep track of
abortive 1957 attempt to revive the field might Chinese social trends.
have been expected to make it difficult to attract
talented people to Chinese sociology after 1979. In pursuing their new intellectual agendas,
However, the field revived very rapidly and showed Chinese sociologists easily found common ground
surprising intellectual vigor. A number of surviv- with Western specialists. The latter were no longer
ing pre-1949 sociologists reappeared to play lead- so entranced with the Maoist model and shared a
ing roles in the revival, considerable numbers of fascination with the social impact of the post-Mao
middle-aged individuals trained in related fields changes. Many Western specialists readily adapted
were recruited (or ordered) to become sociolo- to the chance to conduct research in China and to
gists, and many young people expressed eagerness collaborate with Chinese colleagues. Hong Kong
to be selected for training in the field. The Chinese did not die out as a research site and remained
Sociological Association was formally revived, with vital for some topics, but work from outside China
Fei Xiaotong as its head. (Fei ‘‘unrenounced’’ his was increasingly seen as supplementary to research
pre-1949 works, weakened his ties with ethnology, conducted within the country. (After 1997 and
and resumed advocacy of the program of Chinese Hong Kong’s reversion to China this was still the
300
CHINA STUDIES
case, as the special autonomy enjoyed by the for- crackdown following the massacre to attack soci-
mer British colony produced a freer research at- ology and other newly revived social sciences in
mosphere than in the rest of the country.) China terms reminiscent of 1952 and 1957. The field was
specialists now had to share the stage with many once again accused of being an ideological Trojan
sociologists who did not have area studies training. horse designed to spread doubts about socialist
With data more readily available and collaboration orthodoxy. Some sociology departments were not
with Chinese counterparts possible, area studies allowed to enroll new students, while others had
training and familiarity with the arcane techniques their enrollment targets cut. While these restric-
developed in the 1960s and 1970s were no longer tions eased after a year or two, new and serious
so important. Many non-China specialists found threats to collaborative research projects contin-
the opportunity to study the world’s most popu- ued through the 1990s. Edicts designed to restrict
lous society increasingly attractive. These changes and maintain strict political control over survey
have been beneficial for China specialists within projects involving foreign collaboration were passed
sociology. Given their small numbers and the com- in secret in 1990 and 1995, resulting in the confis-
plexity of Chinese society, there is more than cation of survey data already collected in one
enough ‘‘turf’’ to be shared with sociologists not instance and delays and torpedoing of projects in
trained in China studies. And these changes have others. (The data confiscated after the 1990 secret
reduced the intellectual isolation of the specialists, edict were eventually released to the researchers
from both colleagues in China and those in their involved, and other stalled projects resumed, but
own departments. only after several years of tense negotiations.) In
1998 an international collaborative survey project
As a result of these changes, Chinese sociolo- in social gerontology was publicly attacked within
gists ended their isolation from world sociology, China for allegedly endangering the health of
and Western sociologists conducting research on China’s elderly and having a concealed aim of
China escaped from a highly constrained environ- enabling foreigners to profit from the ‘‘genetic
ment and began to produce work that was less secrets’’ of the Chinese people. Only after a vigor-
idiosyncratic. It should be noted, however, that ous public defense from both foreign and Chinese
sociological research in China is not without lim- researchers was the disputed project saved. These
its, either for foreign researchers or for Chinese. developments mean that researchers who embark
Even in the most liberal phases of the initial re- on sociological research within China on even
form decade from 1979 to 1989, some topics quite nonpolitical topics face some risk that they
remained taboo, some locales were off limits, and will not be able to complete their projects due to
bureaucratic obstruction and interference were political interference.
constant facts of life. Nonetheless, even for for-
eigners it became possible to conduct extended Despite the unpredictable political environ-
ethnographic studies, carry out probability sample ment, Chinese and foreign sociologists pressed
surveys, and gain access for secondary analysis to ahead with a wide-ranging variety of research ac-
the raw data from census and survey studies con- tivities during the 1990s. The success of this effort
ducted by Chinese, experiences that those who is a testament to the increasingly robust ties link-
were studying the Soviet Union before its collapse ing Chinese and foreign sociologists. During the
could only envy. The result was a new spurt of 1980s large numbers of aspiring young Chinese
research activity and publication that enriched our sociologists were sent abroad for doctoral train-
knowledge of Chinese social life immensely. (For a ing, primarily to the United States. By the latter
review of Western research during the decade part of that decade and into the 1990s, a majority
after 1979, see Walder 1989. For an overview of of the Western sociology doctoral theses dealing
developments within Chinese sociology during with China were being produced by students from
that period, see Lu 1989.) China. After the 1989 Tiananmen massacre most
of these newly minted specialists on Chinese socie-
However, the political fallout from the 1989 ty did not return to China, but instead embarked
Tiananmen massacre raised new problems for the on research careers in the West. However, mindful
sociological study of China. At least some conser- of the desirability of leaving the door open for
vatives within the Chinese leadership used the their eventual return, China did not treat this
301
CHINA STUDIES
decision as disloyalty, but instead in most instanc- from a comparison with developments in the for-
es encouraged and invited these foreign-based mer Soviet Union and its satellites. Sociologists
Chinese sociologists to visit China to assist in the have joined others in trying to understand why to
development of the discipline there and to under- date the dismantling of central planning and the
take collaborative research projects. These young, introduction of market reforms have generally
foreign-trained sociologists are able to come and been much more successful in China than in East-
go and to navigate the complex political and bu- ern Europe (see, for example, Rozman 1992;
reaucratic terrain surrounding sociological research Walder 1996a). The bulk of this research, howev-
within China much more easily than their non- er, looks not at the causes, but at the social conse-
Chinese counterparts. Furthermore, the relatively quences of the reforms, and here the concerns are
smaller number of Chinese sociologists trained varied and wide-ranging. For example, spirited
abroad who have returned to the PRC usually have debates have arisen about whether or in what
found their new skills and contacts valued and respects the reforms are contributing to increased
rewarded, rather than treated with envy and suspi- inequality within China, and about whether prereform
cion. Many of these individuals are moving up political elites are largely monopolizing the gains
rapidly into positions of responsibility and leader- produced by the reforms or are being displaced
ship within Chinese research institutes and univer- increasingly by new elites (see, for example, Nee
sities, where they are generally able to keep re- 1996; Xie and Hannum 1996; Walder 1996b; Bian
search projects moving ahead despite bureaucratic and Logan 1996; Zhou, Tuma, and Moen 1996;
obstacles. As a result of these trends, an increasing Buckley 1997). Sociologists and others are also
share of the new sociological research initiatives examining the new entrepreneurial class arising
and publications that are having an impact on within China as a result of market reforms to
China studies in the West involve sociologists from determine what sorts of social origins they have,
China, whether currently based abroad or with- and researchers on this topic debate whether these
in China. entrepreneurs are a force for political change or
are, instead, co-opted and controlled as a result of
The changing sources of funding for research their dependence upon the state (see, for exam-
on Chinese society also are having an impact. ple, Unger 1996; Nevitt 1999). Research on the
Whereas in the past most research by Chinese latter theme is part of a larger exploration of
social scientists was funded by the Chinese govern- whether China’s reforms are leading to the devel-
ment and was responsive to state priorities and opment of institutions of a ‘‘civil society’’ that
concerns, in the reform era Chinese sociologists might eventually lead to the democratization of
and others in China face increasing constraints on China’s political system (see Whyte 1992; White
government funding of research. New competitive 1993). Within research on Chinese stratification
peer review procedures are being introduced to there is also a growing body of work considering
determine who gets scarce research funds, and whether, on balance, the reforms have been harm-
Chinese authorities encourage research institutes ful or beneficial in terms of the quest for gender
and researchers to find ways to raise or obtain equality (see Bauer et al. 1992; Croll 1994; Lee
funds on their own to support their research. In 1998; Entwisle and Henderson forthcoming). An-
this altered environment a variety of Western and other major focus of research concerns changes in
international sources of funding—from United Chinese family life, with the investigation of topics
Nations agencies, the World Bank, the Ford Foun- such as rising divorce and premarital sex rates and
dation, foreign nongovernmental organizations, whether such trends indicate that Chinese family
etc.—have become important in financing research patterns are increasingly ‘‘converging’’ toward the
that Chinese sociologists otherwise could not car- patterns found in other societies (e.g. Xu and
ry out. This changed funding picture thus rein- Whyte 1990; Davis and Harrell 1993; Logan, Bian,
forces other trends that are strengthening the and Bian forthcoming). Still another major focus
links between sociology within China and abroad. is on tracing the impact of particular reform
changes—for example, the relaxation of migra-
While the topics pursued within the sociologi- tion restrictions that have allowed tens of mil-
cal study of China in the 1990s were varied, the lions of rural migrants to flood into China’s cities,
dominant focus was still on the process of market the attempt to revive labor markets as a way of
reforms. Some of the impetus for this focus came
302
CHINA STUDIES
distributing jobs, and the layoffs and unemploy- 1980s and 1990s have provided many new re-
ment being generated by reforms of China’s search questions for sociologists to explore, and a
inefficient state-owned enterprises (e.g. Bian 1994; maturing network of Chinese sociologists and for-
Guthrie 1997; Matthews 1998)Solinger 1995; eign China specialists (and nonspecialist sociolo-
Sensenbrenner 1996. gists interested in studying the world’s most popu-
lous and most rapidly developing society) have
Another primary research focus of China soci- taken advantage of new research opportunities. As
ologists, in addition to the causes and consequences a result of these changes the sociological study of
of the post-1978 economic reforms, involves Chi- China at the end of the 1990s was much better
na’s dramatic fertility decline and its social conse- established than it was two decades earlier, with a
quences. The transformation of China from a wide variety of social processes and trends better
society with moderately high fertility (total fertility studied and more fully understood.
rates of 5+) to extraordinarily low rates (total
fertility rates under 2) has occurred in roughly the REFERENCES
same time period as China’s economic reforms
(from the 1970s through the 1990s), but these two Andors, Stephen 1977 China’s Industrial Revolution: Poli-
fundamental changes have been produced by con- tics, Planning, and Management, 1949 to the Present.
tradictory processes. Whereas in economic affairs New York: Pantheon.
the state has increasingly withdrawn from bureau-
cratic management and has allowed markets to Banister, Judith 1987 China’s Changing Population. Stan-
govern distribution, in regard to population the ford, Calif.: Stanford University Press.
state has shifted away from indirect management
and persuasion in favor of direct bureaucratic Bauer, J., Wang Feng, N. Riley, and Zhao Xiaohua 1992
mandating of fertility quotas. The high national ‘‘Gender Inequality in Urban China.’’ Modern China
priority of this fertility reduction program has 18:333–369.
complex implications, in terms of sociological re-
search. High priority, combined with recurring Bian, Yanjie 1994 Work and Inequality in Urban China.
foreign criticism of China’s human rights record Albany, N.Y.: SUNY Press.
in the realm of family planning (see Whyte 1998),
produce political sensitivities that often obstruct ———, and John Logan 1996 ‘‘Market Transition and
and constrain research on this topic. However, the Persistence of Power: The Changing Stratifica-
that same priority has made research on topics tion System in Urban China.’’ American Sociological
related to fertility so important that foreign as well Review 61:739–758.
as Chinese demographers and others have been
recruited to help plan and carry out a large num- Buckley, Christopher 1997 Social mobility and stratifi-
ber of high-quality surveys of China’s demograph- cation in urban China from Maoism through reform.
ic dynamics. These data have also in most instanc- Ph.D. diss. Australian National University.
es been made available for examination by foreign
as well as Chinese researchers. On balance the Cell, Charles P. 1977 Revolution at Work. New York:
result has been a boom in research on China’s Academic Press.
population and a much better understanding of
current trends in fertility, mortality, and other Chin, R., and A. Chin 1969 Psychological Research in
demographic domains than existed prior to the Communist China 1949–1966. Cambridge, Mass.:
1980s (see, for example, Banister 1987; Coale MIT Press.
1984; Feeney and Yuan 1994; Riley and Riley 1997).
Coale, Ansley 1984 Rapid Population Change in China,
In sum, recurring political turbulence in Chi- 1952–1982. Washington, D.C.: National Academy Press.
na has in some cases obstructed and delayed socio-
logical research, but both the field and its practi- Croll, Elisabeth 1994 From Heaven to Earth: Images and
tioners have continued to forge ahead despite the Experiences of Development in China. London: Routledge.
uncertainties involved. The rapid and complex
social changes that have occurred in China in the Davis, Deborah, and Stevan Harrell (eds.) 1993 Chinese
Families in the Post-Mao Era. Berkeley: University of
California Press.
Entwisle, Barbara, and Gail Henderson (eds.) forthcom-
ing Redrawing Boundaries: Work, Household, and Gen-
der in China. Berkeley: University of California Press.
Feeney, Griffith, and Yuan Jianhua 1994 ‘‘Below Re-
placement Fertility in China? A Close Look at Recent
Evidence.’’ Population Studies 48:381–394.
303
CHINA STUDIES
Fei Xiaotong [Fei Hsiao-t’ung] 1939 Peasant Life in Thompson, Roger 1990 Mao Zedong: Report from Xunwu.
China. London: G. Routledge. Stanford, Calif.: Stanford University Press.
Geddes, W. R. 1963 Peasant Life in Communist China. Unger, Jonathan 1996 ‘‘‘Bridges’: Private Business, the
Ithaca, N.Y.: Society for Applied Anthropology. Chinese Government, and the Rise of New Associa-
tions.’’ The China Quarterly 147:795–819.
Guthrie, Douglas 1997 Strategy and structure in Chi-
nese firms: Organizational action and institutional Walder, Andrew (ed.) 1996a China’s Transitional Econo-
change in Shanghai. Ph.D. diss. University of Califor- my. London: Oxford University Press.
nia, Berkeley.
——— 1996b ‘‘Markets and Inequality in Transitional
Lang, Olga 1946 Chinese Family and Society. New Haven, Economies: Toward Testable Theories.’’ American
Conn.: Yale University Press. Journal of Sociology 101:1060–1073.
Lee, Ching Kuan 1998 Gender and the South China Mira- ——— 1989 ‘‘Social Change in Post-Revolution China.’’
cle. Berkeley: University of California Press. Annual Review of Sociology 15:405–424.
Levy, Marion J., Jr. 1949 The Family Revolution in Modern White, Gordon 1993 ‘‘Prospects for Civil Society in
China. Cambridge, Mass.: Harvard University Press. China: A Case Study of Xiaoshan City.’’ Australian
Journal of Chinese Affairs 29:63–87.
Lin Yaohua [Lin Yüeh-hwa] 1948 The Golden Wing: A
Sociological Study of Chinese Familism. New York: Ox- Whyte, Martin King 1998 ‘‘Human Rights Trends and
ford University Press. Coercive Family Planning in the PRC.’’ Issues and
Studies 34:1–29.
Logan, John, Bian Fuqin, and Bian Yanjie forthcoming
‘‘Tradition and Change in the Urban Chinese Family: ——— 1992 ‘‘Urban China: A Civil Society in the Mak-
The Case of Living Arrangements.’’ Social Forces. ing?’’ In A. Rosenbaum, ed., State and Society in China.
Boulder, Colo.: Westview Press.
Lu Xueyi 1989 China Yearbook of Sociology, 1979–1989 (in
Chinese). Peking: Chinese Encyclopedia Press. ——— 1983 ‘‘On Studying China at a Distance.’’ In
Anne Thurston and Burton Pasternak, eds., The
Mao Zedong (1927) 1971 ‘‘Report on an Investigation of Social Sciences and Fieldwork in China. Boulder, Colo.:
the Peasant Movement in Hunan.’’ In Mao Tsetung, Westview.
Selected Readings from the Works of Mao Tsetung. Pe-
king: Foreign Languages Press. ———, and Burton Pasternak 1980 ‘‘Sociology and
Anthropology.’’ In Anne Thurston and Jason Parker,
Matthews, Rebecca 1998 Where do labor markets come eds., Humanistic and Social Science Research in China.
from? The emergence of urban labor markets in the New York: Social Science Research Council.
People’s Republic of China. Ph.D. diss. Cornell Uni-
versity, Ithaca, N.Y. Whyte, Martin K., Ezra F. Vogel, and William L. Parish,
Jr. 1977 ‘‘Social Structure of World Regions: Main-
Nee, Victor 1996 ‘‘The Emergence of a Market Society: land China.’’ Annual Review of Sociology 3:179–207.
Changing Mechanisms of Stratification in China.’’
American Journal of Sociology 101:908–949. Wong, Siu-lun 1979 Sociology and Socialism in Contempo-
rary China. London: Routledge and Kegan Paul.
Parris, Kristen 1999 ‘‘Entrepreneurs and Citizenship in
China.’’ Problems of Post-Communism 46:43–61. Xie, Yu, and Emily Hannum 1996 ‘‘Regional Variation
in Earnings Inequality in Reform-Era Urban China.’’
Riley, Nancy, and Robert Riley 1997 China’s Population: American Journal of Sociology 101: 950–992.
A Review of the Literature. Liege: International Union
for the Scientific Study of Population. Xu, Xiaohe, and Martin King Whyte 1990 ‘‘Love Matches
and Arranged Marriages: A Chinese Replication.’’
Rozman, Gilbert (ed.) 1992 Dismantling Communism: Journal of Marriage and the Family 52:709–722.
Common Causes and Regional Variations. Washington,
D.C.: Woodrow Wilson Center Press. Yang, C. K. 1944 A North China Local Market Economy.
New York: Institute of Pacific Relations.
Sensenbrenner, Julia 1996 Rust in the iron rice bowl:
Labor reforms in Shanghai state enterprises, 1992– Zhou, Xueguang, Nancy Tuma, and Phyllis Moen 1996
93. Ph.D. diss. Johns Hopkins University. ‘‘Stratification Dynamics under State Socialism: The
Case of Urban China, 1949–1993.’’ Social Forces
Solinger, Dorothy 1995 ‘‘China’s Urban Transients in 74:759–796.
the Transition from Socialism and the Collapse of
the Urban Public Goods Regime.’’ Comparative Poli- MARTIN KING WHYTE
tics 27:127–147.
304
CITIES
CITIES living and eventually to industrial production
(Childe 1950). A second view is that some towns
A city is a relatively large, dense, permanent, het- and cities first developed as trade centers, and
erogeneous, and politically autonomous settlement were then nourished by agricultural activity in
whose population engages in a range of nonagri- their hinterlands (Jacobs 1970).
cultural occupations. Definitions of cities and their
associated phenomena vary by time and place, and Increasing complexity of social organization,
by population size, area, and function (Shryock, environmental adaptation, and technology led to
Siegel, and associates 1976, pp. 85–104). The city is the emergence of cities (Child 1950; Duncan 1964).
often defined in terms of administrative area, Excluding preagricultural settlements, towns and
which may be larger than, smaller than, or equal to then cities were first established in the fourth to
the area of relatively dense settlement that com- third millennium B.C. in ancient Mesopotamia
prises what is otherwise known as the city proper. within the Tigris and Euphrates river valleys, in the
The suburb is a less dense but permanent settle- Harappa civilization in the valleys of the Indus
ment that is located outside the city proper and River and its tributaries, and in the Egyptian Old
contains populations that usually have social and Kingdom in the lower Nile valley. Other centers
economic ties to the city. appeared in the Huang Ho basin on the east coast
of China and the Peruvian Andes in the second to
Definitions of urban vary by nation; in the first millennium B.C., and in Mesoamerica in the
United States the term refers to populations of first millennium B.C. (Phillips 1997, pp. 82–85).
2,500 or more living in towns or cities and to
populations living in urbanized areas, including Small agricultural surpluses and limits on trans-
suburbs. In other nations, the lower limits for portation meant that the first towns were small
settlements defined as urban vary between 200 and few in number, and contained only a small
and 50,000 persons. United Nations definitions of proportion of the populations of their regions.
urban areas emphasize a population of 20,000 or Economic activities of the earliest towns were tied
more, and cities a population of 100,000 or more. largely to their surrounding areas. After the rise of
Urbanization refers to the economic and social towns in the Middle East, trading centers appeared
changes that accompany population concentra- on the shores and islands of the Mediterranean;
tion in urban areas and the growth of cities and some, such as Athens, became city-states. After
their surrounding areas. developing more effective communication and
social organization, some Western city-states ex-
Cities reflect other areas with which they are panded and acquired empires, such as those of
linked and the civilizations of which they are a Alexander the Great and of Rome. Following the
part. Cities are centers of markets, governments, decline of Rome, complex city life continued in
religion, and culture (Weber 1958, pp. 65–89). A the East and in the West in the Byzantine and
community is a population sharing a physical envi- Muslim empires, while the population of Europe
ronment and leading a common and interdepen- declined and reverted, for the most part, to subsis-
dent life. The size, density, and heterogeneity of tence agriculture and organized into small territo-
the urban community have been described as ries held together by the Catholic Church (Hawley
leading to ‘‘urbanism as a way of life,’’ which 1981, pp. 1–35).
includes organizational, attitudinal, and ecologi-
cal components different from those of rural areas Sjoberg (1960) has described preindustrial cit-
(Wirth 1938). ies as feudal in nature and sharing social, ecologi-
cal, economic, family, class, political, religious,
THE CITY IN HISTORY and educational characteristics different from those
in modern industrial cities. In the former, the city
Town and city development has been described as center, with its government and religious and
tied to a technological revolution in agriculture economic activities, dominated the remainder of
that increased food production, thereby freeing the city and was the locale of the upper social
agriculturalists to engage in nonagricultural occu- classes. Homogeneous residential areas were found
pations. This resulted in an evolution to urban throughout the city, but nonresidential activities
305
CITIES
were not confined to distinct neighborhoods The population classified as urban by the U.S.
(Sjoberg 1960). Census Bureau (based on aggregations of 2,500 or
more and including the surrounding densely popu-
Beginning in the tenth century, further town lated territories) increased from 5 percent in 1790
development in the West was facilitated by increas- to 75 percent in 1990. In 1790 the largest urban
es in agricultural technology, population, trade, place in the United States had fewer than 50,000
and communication; the rise of an entrepreneurial inhabitants. As late as 1840, not a single urban
class; and an expanding web of social norms re- place in the United States had more than half a
garding economic activity. Communication and million inhabitants. There were four such places in
manufacturing were revived, which led to the 1890, fourteen in 1940, twenty-two in 1980, and
growth of towns with local autonomy and public twenty-three in 1990 (U.S. Bureau of the Census
administration, and eventually to networks of cit- 1997, p. 44).
ies. Surplus rural populations migrated to towns
and cities, which grew because of their specializa- City growth in the United States during the
tion and larger markets, becoming focal points of nineteenth century was driven by migration, since
European societies (Hawley 1981, pp. 37–83). there were sometimes excesses of urban deaths
over births in the early part of that century and
The emergence of a global economy struc- lower birthrates in the last part of that century.
tured city development (Lo and Yeung 1996, pp. Population concentration facilitated greater divi-
1–13). In the thirteenth and fourteenth centuries, sions of labor within and among families and
some European central place and port cities, such individuals, as well as increasing numbers of vol-
as those of the Hanseatic League of Northern untary associations centered on new urban inter-
Europe, established commercial links with others. ests and problems. At first, cities were compact,
Meanwhile, other cities in the Eastern Hemisphere growth was vertical, and workers resided near
and some cities in the Western Hemisphere were their workplaces. The outward expansion of the
linked together by long-distance trade routes tran- residential population was facilitated in the last
scending the boundaries of empires and states part of the century by steam and electric railways,
(Wolfe 1982, p. 250). Military-commercial alliances the outward expansion of industry, and the in-
facilitated the incorporation of territories into creasing role of the central business district in
states. Commercial ties expanded during the late- integrating economic activities (Hawley 1981,
fifteenth and the sixteenth centuries, as Europe- pp. 61–145).
ans developed merchant capitalism, and traded
with and colonized peoples on other continents. The increasing scale of production in the Unit-
European and North American states developed ed States led to the development of a system of
more complex economies during the next three cities that organized activities in their hinterlands.
centuries, facilitating development of a ‘‘Europe- Cities were differentiated from each other by their
an world economy’’ with a global market, a global degree of dominance over or subordination to
division of labor, and a global system of cities other cities, some of which were engaged in cen-
(Wallerstein 1974). tralized manufacturing, others of which depended
on transportation, commercial, administrative, or
WESTERN CITIES SINCE THE INDUSTRIAL other functions. According to Hawley (1986), cit-
REVOLUTION ies may have certain key functions that dominate
other cities, that is, integrating, controlling, or
In most regions of the more developed world, coordinating activities with these cities. Examples
urban growth and urbanization have occurred at of key functions are administration, commerce,
an increasing rate since the beginning of the In- finance, transportation, and communication (Dun-
dustrial Revolution. After 1820, the numbers and can et al. 1960). Since each city exists within its
sizes of urban areas and cities in the United States own organizational environment, the expansion
increased as a result of employment concentration of linkages of urban organizations has accompa-
in construction and manufacturing, so urban are- nied the development of a system of cities (Turk 1977).
as began to grow more rapidly than rural ones.
In the United States, the nature of key func-
tions in city systems changed with the expansion of
306
CITIES
settlements westward, as colonial seaports, river freeway and mass transit networks, facilitate inter-
ports, Midwestern railway towns, central places on action among ‘‘social circles’’ of people who are
the Great Plains, extractive centers, and govern- not neighborhood-based. Meanwhile, urban neigh-
ment centers were integrated into an urban sys- borhoods organized around such factors as status,
tem. A nationwide manufacturing base was estab- ethnicity, or lifestyle, also persist. As the city ages,
lished by 1900, as were commercial and financial so does suburban as well as centrally located hous-
centers. By 1960 a fully developed system of differ- ing—a delayed consequence of the spread of ur-
entiated urban centers existed within the United ban settlement.
States (Duncan and Lieberson 1970).
Kasarda (1995, p. 239) states that major cities
METROPOLITAN AREAS in the United States have been transformed from
centers of manufacturing and wholesale and retail
In the United States, metropolitan areas are de- trade, to centers of finance, administration, and
fined as being of two kinds. Metropolitan statistical information processing. Frey (1995, p.272) indi-
areas (MSAs) are areas including one or more cates that during the 1970s and 1980s populations
central cities with a population of 50,000 or more, concentrated in metropolitan areas with diverse
or areas with a less densely populated central city economies that emphasized service and knowl-
but with a combined urban population of at least edge industries, as well as in recreation and retire-
100,000 (75,000 in New England), including sur- ment areas. In the 1970s there was a metropolitan
rounding counties or towns. Consolidated metropoli- and nonmetropolitan population turnaround.
tan statistical areas (CMSAs) contain at least one Populations in smaller metropolitan areas grew
million population and may have subareas called more rapidly than did those in larger metropolitan
primary metropolitan statistical areas (PMSAs) (Frey areas and population in developing countries grew
1990, p. 6). In 1990 a majority of the population of more rapidly than in developed countries. In the
the United States resided in the thirty-nine major 1980s, metropolitan population shifts, enhanced
metropolitan areas: those with more than one by international and internal migration, led to an
million population (Frey 1995, p. 276). urban revival, increasing regional racial, skill, and
age divisions, the suburban dominance of growth
New urban-population-density patterns ap- and employment and more isolation of the urban
peared in the United States with the development core (Frey 1995, pp. 272–275).
of metropolitan areas. Growth was characterized
by increases in population density in central cities, Residential moves—the individual counterpart
followed by increases in density in the metropoli- of population redistribution processes—are af-
tan ring. Transport and communication technolo- fected by population characteristics and other fac-
gies facilitated linkages of diverse neighborhoods tors. Elements that influence mobility decisions
into metropolitan communities dominated by more are socioeconomic and psychological factors per-
densely populated central cities. These linkages in taining to the family and the family life cycle,
turn organized relationships between central cit- housing and the local environment, and occupa-
ies and less densely populated hinterlands and tional and social mobility (Sabagh et al. 1969). For
subcenters. Older metropolitan areas became the each of these factors, conditions may restrain per-
centers of CMSAs, while newer areas were the sons from moving, or push or pull them to new
frontiers of expansion, growing by natural increase locations. Information concerning new housing
and especially by net migration. Metropolitan opportunities, the state of the housing market,
sprawl extended beyond many former nonurban and the availability of resources may impede or
functions, as well as more centrally located older facilitate moves; subsequent moves may stem from
features of the cityscape. recent migration or residential turnover in the
metropolitan areas (Long 1988, pp. 219–224).
The interior of large Western metropolitan
areas represents a merging of urban neighbor- TRADITIONAL EXPLANATIONS OF CITIES
hoods into complex overlapping spatial patterns,
which reflect to some extent the dates when urban During the nineteenth century the character of the
neighborhoods were settled and built-up. The blur- Western city was seen as different from that of
ring of neighborhood distinctions, facilitated by
307
CITIES
noncity areas. Beginning with the public-health Theodorson (1982) and Michaelson (1976,
movement and concerns with urban housing, so- pp. 3–32) have described research on American
cial scientists documented ‘‘pathologies’’ of urban cities as reflecting neo-orthodox, social-area analy-
life through the use of social surveys, the purpose sis, and sociocultural approaches, each with its
of which was to provide policy-relevant findings. own frame of reference and methods. Neo-ortho-
But since ‘‘urbanism as a way of life’’ has permeat- dox approaches have emphasized the interdepend-
ed the United States, many of the social and eco- ence of components of an ecological system, in-
nomic problems of cities have spread into smaller cluding population, organization, technology, and
towns and rural areas, thus rendering the notion environment (Duncan 1964). This view has been
of unique urban pathologies less valid than when it applied to larger ecological systems that can ex-
was formulated. tend beyond the urban community. Sustenance
organization is an important focus of study (Hawley
In the first six decades of the twentieth century, 1986). While neo-orthodox ecologists did not inte-
explanations of city development were based largely grate the notion of Social Darwinism into their
on Western human ecology perspectives, and disa- work, the economic aspects of their approach have
greements have arisen (Sjoberg 1968, p. 455). helped to guide studies of population phenome-
Theories based on Social Darwinism (Park [1916– na, including population shifts and urban differ-
1939] 1952) and economics (Burgess 1925, pp. 47– entiation (Frey 1995, pp. 271–336; White 1987),
52) were first used to explain the internal struc- residence changes (Long 1988, pp. 189–251), ra-
tures of cities. ‘‘Subsocial’’ aspects of social and cial and ethnic diversity (Harrison and Bennett
economic organization—which did not involve 1995, pp. 141–210), segregation (Farley and Allen
direct interpersonal interaction—were viewed as 1987, pp. 103–159; Lieberson and Waters 1988,
generating population aggregation and expansion. pp. 51–93), housing status (Myers and Wolch 1995,
Market related competitive-cooperative process- pp. 269–334), and industrial restructuring and the
es—including aggregation-thinning out, expansion- changing locations of jobs (Kasarda 1995, pp.
contraction, centralization, displacement, segre- 215–268).
gation, migration, and mobility—were believed to
determine the structures and patterns of urban Social-area analysis, in contrast, regards mod-
neighborhoods (Quinn 1950). The competition of ern industrial society as based on increasing scale,
differing urban populations and activities for opti- which represents ‘‘increased rates and intensities
mal locations was described as creating relatively of social relation,’’ ‘‘increased functional differen-
homogeneous communities, labeled ‘‘natural are- tiation,’’ and ‘‘increased complexity of social or-
as,’’ which display gradient patterns of decreasing ganization’’ (Shevky and Bell 1955). These con-
densities of social and economic activities and cepts are related to neighborhood dimensions of
problems with increasing distance from the city center. social rank, urbanization, and segregation, which
are delimited by the factor analysis of neighbor-
On the basis of competitive-cooperative proc- hood or census tract measures. ‘‘Factorial ecolo-
esses, city growth was assumed to result in charac- gy’’ has been widely used for classifying census
teristic urban shapes. The Burgess hypothesis speci- tracts, sometimes for planning purposes. There
fies that in the absence of countervailing factors, are disagreements concerning whether factor-analy-
the American city takes the form of a series of sis studies of urban neighborhood social structure,
concentric zones, ranging from the organizing which are associated with social-area analysis and
central business district to a commuters’ zone similar approaches, result in theoretically useful
(Burgess 1925, pp. 47–52). Other scholars empha- generalizations (Janson 1980, p. 454; White 1987,
sized star-shaped, multiple-nuclei, or cluster pat- pp. 64–66)
terns of development. These views were descrip-
tive rather than theoretical; assumed a Western Sociocultural ecology has used social values
capitalist commercial-industrial city; were distort- such as sentiment and symbolism to explain land
ed by topography, and street and transportation use in central Boston (Firey 1947) and other as-
networks; and generally failed to take into account pects of city life. While values and culture are
use of land for industrial purposes, which is found relevant to explanations of city phenomena, this
in most city zones. perspective has not led to a fully developed line of
investigation.
308
CITIES
Michaelson (1976, pp. 17–32) has argued that emphasize the role of the state, sometimes gov-
none of these aforementioned ecological approach- erned by urban elites, which relys on urban re-
es to the city explicitly study the relationship be- sources, and favors urban over rural development
tween the physical and the social environment, for (Lipton 1977). Kasarda and Crenshaw (1991) re-
the following reasons: (1) their incomplete view of gard these perspectives as underspecified and as
the environment; (2) their focus on population lacking empirical confirmation.
aggregates; (3) their failure to consider contribu-
tions of other fields of study; and (4) the newness Political economy perspectives explain city
of the field. Since Michaelson wrote his critique, growth in more developed as well as less devel-
sociologists have given more attention to the ur- oped regions as a product of globalization, includ-
ban environment. ing capital accumulation and nation-state forma-
tion (Tilly 1975). Europe and the United States
NEW EXPLANATIONS OF CITIES colonized non-European peoples and obtained
raw materials from the colonies which they proc-
Traditional explanations of cities did not ade- essed and exchanged with their colonies as manu-
quately account for the economic, political, racial factured products. Colonial areas gained political
and social upheavals in U.S. cities during the 1960s. independence after World War II. Economic re-
Further, urbanization in less developed regions structuring then resulted in the globalization of
does not necessarily follow the Western pattern. economic and cultural life, and changes in the
City growth may absorb national population incre- international division of labor between cities.
ments and reflect a lack of rural employment, a
migration of the rural unemployed to the city, and Sassen (1991) indicates that following World
a lack of urban industrial development. Increasing War II communications technology facilitated the
concentration of population in larger cities in less dispersion of manufacturing by multinational cor-
developed regions may be followed by the emer- porations to low-wage cities, some in former colo-
gence of more ‘‘Western-style’’ hierarchies of cit- nial areas. A globally integrated organization of
ies, functions, and interorganizational relation- economic activity then supplanted world econom-
ships. Many cities in less developed regions experience ic domination by the United States. Economic
the environmental hazards of Western cities, a activities were integrated beyond national urban
compartmentalization of life, persistent poverty hierarchies into a small number of key global
and unemployment, a rapidly worsening housing cities, which are now international banking cen-
situation, and other symptoms of Western social ters, with transnational corporate headquarters,
disorganization. The juxtaposition of local urban- sophisticated business services, information proc-
ism and some degree of Western urbanization essing, and telecommunications. These cities (Lon-
may vitiate a number of traditional Western solu- don, New York, and to a lesser degree Tokyo)
tions to urban problems. command the global economy and are supported
by worldwide hierarchies of decentralized special-
Kasarda and Crenshaw (1991, pp. 467–500) ized cities (Sassen 1991). International networks
have summarized and compared theoretical per- of cities sometimes provide opportunities for mul-
spectives concerning how urbanization occurs in tinational corporations from less developed re-
less developed regions. Modernization-human gions to penetrate more developed regions (Lo
ecology perspectives portray city building as the and Yeoung 1998, p. 2).
result of changes in social organization and appli-
cations of technology. Rural-urban migration, re- Feagan and Smith (1987, pp. 3–33) describe
sulting from urban industrialization and excess economic restructuring in U.S. cities in the 1980’s
rural fertility, can eventually lead to the reduction as including plant closures and start-ups, the devel-
of rural-urban economic and social differences opment and expansion of corporate centers, and
(Hawley 1981). Dependency-world-system perspec- corporate movement to outlying areas, all of which
tives describe the capitalist world-system as guid- impact different groupings of cities. They portray
ing change in the less developed regions so as to economic restructuring as a product of interac-
maintain the dominance of the more developed tions between governmental components of na-
core (Wallerstein 1974). Urban bias perspectives tion-states; multinational, national, and local cor-
porations and businesses; and nongovernment
309
CITIES
organizations. These interactions guide policies 38 percent of the populations in the less devel-
affecting local taxation, regulation, implementa- oped regions were urban dwellers, compared with
tion, and public-private partnerships. Some city 75 percent of the populations in the more devel-
neighborhoods have become expendable locations oped regions. Urban areas in less developed areas
for rapidly shifting economic activity. Policies to are expected to account for approximately 90
accommodate to shifting sites of economic activity percent (2.4 billion) of the total of 2.7 billion
influence international and intranational migra- persons that the United Nations projects will be
tion, family life, as well as the use of urban, subur- added to the earth’s population from 1995 to 2030
ban, and rural space. (United Nations 1998). (Cities in more developed
regions are expected to account for only 140
Aspects of political-economy perspectives have million of the population increase in the same
been taken into account in more traditional stud- period.)
ies of cities (Frey 1995, pp. 271–336; Kasarda 1995,
pp. 215–268). Walton (1993, pp. 301–320) main- There will continue to be significant differ-
tains that political-economy perspectives have made ences in urbanization between the less developed
the following contributions to the study of cities: regions. About three fourths of Latin-American
(1) showing that urbanization and urbanism are population was urban in 1995, roughly the average
contingent upon the development of social and level of industrialized regions in Europe, Japan,
economic systems; (2) generating comparative stud- Australia, New Zealand, the United States and
ies, particularly in developing countries; (3) eluci- Canada. The most extensive future urban growth
dating the operation of the informal economy in will be in Asia and Africa, which are now only
cities; (4) outlining a political economy of place; about one-third urban (O’Mera 1999; United Na-
(5) showing how globalization relates to ethnicity tions 1998).
and community; and (6) relating urban political
movements to changes in the global economy. An increasing portion of urban population is
residing in giant urban agglomerations. An urban
RETROSPECT AND PROSPECT agglomeration, according to the United Nations
(1998), is the population within a contiguous terri-
Globalization of economic and cultural life is asso- tory inhabited at urban levels without regard to
ciated with new urban trends. While cities in less administered boundaries. In 1995 the fifteen larg-
developed regions lack resources when compared est ranged in size from 27 million (Tokyo) to 9.9
with those in more developed regions, cities in million (Delhi). Less developed countries are some-
both types of regions are becoming more respon- times characterized by primate cities, that is, the
sive to changes in worldwide conditions. largest city in a country dominates other cities and
are larger than would be expected on the basis of a
During the twenty-first century the majority of ‘‘rank-size’’ rule, which indicates that the rank of a
human population is expected to be urban resi- population aggregation times its size equals a con-
dents (United Nations 1998). In 1995 approxi- stant (Shryock, Siegel, and associates 1976), thus
mately 46 percent (2.6 billion) of the world’s popu- resulting in an inadequate supporting hierarchy of
lation lived in urban areas. By the year 2005, smaller cities. Primate cities often appear in small
approximately half of the population is projected countries, and in countries with a dual economy,
to be urban, compared with fewer than one of but are not as apparent in large countries or those
every three persons in 1950. By 2030 approximate- with long urban histories (Berry 1964).
ly six-tenths (5.1 billion) of the population is pro-
jected to be urban. Megacities, defined by the United Nations (1998)
as those cities with 10 million or more inhabitants,
Urban areas in less developed areas are pro- are increasing in number and are concentrated in
jected to dominate the growth of the less devel- less developed regions. There were fourteen
oped regions and of the world during the twenty- megacities in 1995, including ten in less developed
first century. Reasons are the high population regions. Twenty-six megacities are projected in
growth rates and high numerical population growth 2015, including twenty-two in less developed re-
of less developed regions, and high rural-urban gions, of which sixteen will be in Asia. Megacities
migration to cities in these regions. In 1995, only have both assets and liabilities. Brockerhoff and
310
CITIES
Brennan (1998, pp. 75–114) have suggested that urban poor. Such trends may be related to a
cities’ size and rates of population growth are spread of ghetto-underclass neighborhoods and to
inversely related to welfare. Kasarda and Crenshaw increasingly polarized intracity neighborhood dif-
(1991, pp. 471–474) note that megacities, which ferences of poverty and affluence (Morenoff and
may have serious social and economic problems, Tienda 1997, pp. 59–72).
can be driving engines for industrial production in
developing regions, and contribute disproportion- Trans-border cites are new urban forms that
ately to their countries’ economies (Kasarda and transcend the boundaries of nation-states and re-
Crenshaw 1991, pp. 467–501). flect increasingly borderless economies. They re-
sult from globalization trends, including the eco-
Berry (1978) indicates that urbanization may nomic integration of regions, reductions of trade
be accompanied by ‘‘counter-urbanization’’ or de- barriers, and the establishment of multinational
creasing city size and density. City growth some- free trade zones. They link cities adjacent to a
times occurs in cycles, which begin with rapid border, and sometimes metamorphose into trans-
growth in the urban core, followed by rapid growth border systems, complete with specialized func-
in the suburban ring, a decline in growth in both tions and populations, and extensive cross-border
the core and the ring, and then rapid growth in the social and economic ties (Rubin-Kurtzman et al.
core (United Nations 1998). These cycles—of ur- 1993). Examples are Southern California (U.S.)-
banization, suburbanization, counter-urbanization, Baja California (Mexico), the Singapore-Johore
and reurbanization—appear to be associated with (Malysia)-Riau (Indonesia) region, and the Beijing
concentrations of services in city centers, followed (China)-Pyongyang (North Korea)-Seoul (South
by improvements in commuting by the labor force Korea)-Tokyo (Japan) urban corridor. Trans-bor-
and increased suburban home ownership by ur- der cities pose questions regarding the limits of
ban labor forces. The 1970s deurbanization (met- national sovereignty, but can also integrate human
ropolitan turnaround) in the United States was and economic resources and thus enhance inter-
followed by reurbanization in the 1980s, and the national stability.
start of another period of deurbanization in the
1990s (United Nations 1998). During the 1970s Cities will continue to exhibit extremes of
and 1980s counter-urbanization occurred in other affluence and poverty, but the extent and conse-
more developed regions and in less developed quences of these extremes are unclear. Massey
regions, including a slow-down of population (1996, pp. 395–412) has argued that both the
growth rates in some megacities in developing affluent and the poor are concentrating in cities in
regions, particularly in Latin America (United Na- the United States; consequences may include in-
tions 1998). creased densities of crime, addictions, diseases,
and environmental degradation, the emergence of
Changes in the global economy and in the oppositional subcultures, and enhanced violence.
aforementioned city growth cycles are associated Massey assumes these trends apply to less devel-
with new forms of urban land use in the United oped regions as well as to the United States. Farley
States, which are laid over preexisting urban pat- (1996, pp. 417–420) has advanced a counterargument
terns. Businesses, which provide the economic that while economic inequality is increasing in the
base for cities, move between optimum locations United States, as may be the geographic segrega-
in different cities (Wilson 1997, p. 8). Globalization tion of the poor, the continuing (1996) rise in
has enhanced the growth of suburbs and ‘‘edge prosperity increases welfare at all income levels.
cities’’ organized around outlying business and Firebaugh and Beck (1994, pp. 631–653) maintain
high-technology centers linked by telecommunica- that economic growth, even in dependent less
tions networks to other cities (Castells, 1989; Mul- developed countries, may eventually increase wel-
ler 1997). Unregulated informal sectors of the fare. Hout et al. (1998) state that political institu-
economy develop in a variety of intraurban loca- tions, which are partially responsible for growing
tions. High-income native and immigrant popula- inequality, can provide appropriate remedies.
tions, which profit from the new global economy,
sometimes cluster in protected enclaves. Low-in- As cities account for increasing shares of the
come immigrant and ethnic populations often earth’s population, they will consume increasing
occupy areas inhabited by earlier cohorts of the shares of the earth’s resources, produce increasing
311
CITIES
shares of pollution, and their populations will be Mexico City, while not commanding the same
more subject to negative feedbacks from human global economic stature as Los Angeles, is the
impacts on the environment. O’Mera (1999, p. primate city of Mexico. Mexico City shares many
137) estimates that populations of cities, while of Los Angeles’s problems, including environmen-
occupying only two percent of the earth’s surface, tal hazards (earthquakes, and air pollution), water
consume 75 percent of the earth’s resources. Cit- shortages, lack of housing and services, expansion
ies are often sited on areas that contain prime of low-income settlements, dependence on the
agricultural land; urban expansion then inhibits private automobile, lack of sufficient transporta-
and degrades crop production. Conversion of ru- tion, earthquakes, crime, and rising ethnic vio-
ral land to urban use intensifies natural hazards lence. Effects of policy responses to these prob-
including floods, forest and brush fires, and earth lems have been dampened by economic reversals
slides. Further development concentrates and then (Ward 1998).
disburses to outlying locations such artificial haz-
ards as air pollution, land and water pollution, and These illustrations suggest that cities at some-
motor vehicle and air traffic noise. Cities are also what similar levels of influence within their respec-
sited on the shorelines of oceans, thus increasing tive countries share similar characteristics, wheth-
numbers of city residents are subject to the im- er in the more developed or in the less developed
pacts of storm surges and erosion. Lowry (1992) regions. This would support a view that determi-
argues that environmental impacts attributed to nants of city development are rooted in the global
cities reflect population growth, industrialization economy and are influenced by similar trends, but
and prosperity rather than city growth itself. vary according to the city’s place in the system.
Some of the aforementioned trends are appar- REFERENCES
ent, for example, in Los Angeles and Mexico City,
which ranked seventh and second in size, respec- Berry, Brian J. L. 1978 ‘‘The Counter-Urbanization
tively, among the world’s largest megacities in Process: How General?’’ In Niles M. Hansen, ed.,
1995. After World War II and until the 1970s, Los Human Settlement Systems: International Perspectives on
Angeles’s industrial and population growth and Structure, Change and Public Policy. Cambridge, Mass.:
suburban sprawl made it a prototype for urban Ballinger.
development that brought with it many inner-city,
energy, suburban, environmental, state-manage- ——— 1964 ‘‘Cities as Systems Within Systems of Cit-
ment, ethnic, and capital-accumulation problems. ies.’’ In J. Friedman and W. Alonso, eds., Regional
Development and Planning. Cambridge, Mass.: MIT Press.
Since the early 1970s, Los Angeles has become
a radically changed global megacity based on accu- Brockerhoff, Martin, and Ellen Brennan 1998 ‘‘The
mulation of global capital, economic restructur- Poverty of Cities in Developing Regions.’’ Population
ing, communications, and access to new interna- and Development Review 241:75–114.
tional markets. The reorganization of Los Angeles
has affected labor-force demands and the charac- Burgess, Ernest W. (1924) 1925 ‘‘The Growth of the
ter of both native and immigrant populations. City, an Introduction to a Research Project.’’ In
Immigration transformed Los Angeles into a Robert E. Park, Ernest W. Burgess, and R. D.
multicultural metropolis (Waldinger and Bozorgmehr McKenzie, eds., The City. Chicago: University of Chi-
1996, pp. 3–37). The area’s economy emphasizes cago Press.
high-income service-sector jobs and low-income
service-sector and industrial-sector jobs. Los Ange- Castells, Manual 1989 The Informational City: Information
les has a great concentration of industrial tech- Technology, Economic Restructuring, and the Urban-Re-
nology and great international cultural impor- gional Process. Cambridge, Mass.: Basil Blackwell, Ltd.
tance, is dependent on the private automobile, has
environmental hazards (earthquakes, brush fires, Childe, V. Gordon 1950 ‘‘The Urban Revolution.’’ Town
and air pollution), limited water supplies, housing Planning Review 21:3–17.
shortages, crime, and has experienced uncontrolled
intergroup conflicts in 1965 and 1992. Duncan, Beverly, and Stanley Lieberson 1970 Metropolis
and Region in Transition. Beverly Hills, Calif.: Sage.
Duncan, Otis D. 1964 ‘‘Social Organization and the
Ecosystem.’’ In Robert E. L. Faris, ed., Handbook of
Modern Sociology. Chicago: Rand McNally.
312
CITIES
———, William R. Scott, Stanley Lieberson, Beverly ———, and Edward M. Crenshaw 1991 ‘‘Third World
Duncan, and Haliman H. Winsborough 1960 Me- Urbanization: Dimensions, Theories, and Determi-
tropolis and Region. Baltimore: Johns Hopkins Uni- nants.’’ In W. Richard Scott, ed., Annual Review of
versity Press. Sociology 17:467–500.
Farley, Reynolds 1996 ‘‘The Age of Extremes: A Revi- Lieberson, Stanley, and Mary C. Waters 1988 From Many
sionist Perspective.’’ Demography 37:417–420. Strands: Ethnic and Racial Groups in Contemporary
America. New York: Russell Sage Foundation.
———, and Walter R. Allen 1987 The Color Line and the
Quality of Life in America. New York: Russell Sage Lipton M. 1977 Why Poor People Stay Poor: Urban Bias in
Foundation. World Development. Cambridge, Mass.: Harvard Uni-
versity Press.
Feagin, Joe R., and Michael Peter Smith 1987 ‘‘Cities
and the New International Division of Labor: An Lo, Fu-chen, and Yoe-Man Yeung 1996 ‘‘Introduction.’’
Overview.’’ In Michael Peter Smith and Joe R. Feagin, In Fu-chen Lo and Yoe-Man Yeung, eds., Emerging
eds., The Capitalist City. New York: Basil Blackwell, Ltd. World Cities in Pacific Asia. Tokyo: United Nations
University Press.
Firebaugh, Glen, and Frank D. Beck 1994 ‘‘Does Eco-
nomic Growth Benefit the Masses? Growth Depend- Long, Larry 1988 Migration and Residential Mobility in the
ence and Welfare in the Third World.’’ American United States. New York: Russell Sage Foundation.
Sociological Review 59:631–653.
Lowry, Ira S. 1991 ‘‘World Urbanization in Perspec-
Firey, Walter 1947 Land Use in Central Boston. Cam- tive.’’ In Kingsley Davis and Mikhail S. Bernstram,
bridge, Mass.: Harvard University Press. eds., Resources, Environment, and Population: Present
Knowledge: Future Options, Population and Development
Frey, William 1995 ‘‘The New Geography of Population Review, a supplement to Volume 16, 1990.
Shifts.’’ In Reynolds Farley, ed., State of the Union:
America in the 1990s, Volume Two: Social Trends. New Massey, Douglas S. 1996 ‘‘The Age of Extremes: Con-
York: Russell Sage Foundation. centrated Affluence and Poverty in the Twenty-first
Century.’’ Demography 33:395–412.
——— 1990 ‘‘Metropolitan America: Beyond the Tran-
sition.’’ Population Bulletin 45, 2:1–53. Michaelson, William 1976 Man and His Urban Environ-
ment: A Sociological Approach. Reading, Mass.: Addi-
——— 1987 ‘‘Migration and Depopulation of the Me- son-Wesley.
tropolis: Regional Restructuring or Rural Renais-
sance?’’ American Sociological Review 52:240–257. Morenoff, Jeffrey, and Marta Tienda 1997 ‘‘Underclass
Neighborhoods in Temporal and Ecological Per-
Harrison, Roderick, and Claudette E. Bennett 1995 spective.’’ In Alan W. Heston, ed., David Wilson,
‘‘Racial and Ethnic Diversity.’’ In Reynolds Farley, special ed., The Annals of the American Academy of
ed., State of the Union: America in the 1990s, Volume Political and Social Science Globalization and the Ameri-
Two: Social Trends. New York: Russell Sage Foundation. can City 551:59–72.
Hawley, Amos H 1986 Human Ecology: A Theoretical Muller, Peter O. 1997 ‘‘The Suburban Transformation
Essay. Chicago: University of Chicago Press. of the Globalizing American City.’’ In Alan W. Heston,
ed., David Wilson, special ed., The Annals of the
——— 1981 Urban Society: An Ecological Approach, 2nd American Academy of Political and Social Science:
ed. New York: Wiley. Globalization and the Changing American City 551: 44–50.
Hout, Michael, Richard Arons, and Kim Voss 1996 ‘‘The Myers, Dowell, and Jennifer R. Wolch 1995 ‘‘The Polari-
Political Economy of Inequality in the ‘Age of Ex- zation of Housing Status.’’ In Reynolds Farley, ed.,
tremes.’’’ Demography 33:421–425. State of the Union: America in the 1990s, Volume Two:
Social Trends. New York: Russell Sage Foundation.
Jacobs, Jane 1970 The Economy of Cities. New York:
Vintage. O’Mera, Molly 1999 ‘‘Exploring a New Vision for Cit-
ies.’’ In Linda Stack, ed., State of the World 1999. New
Janson, Carl-Gunnar 1980 ‘‘Factorial Social Ecology: An York: W.W. Norton and Company.
Attempt at Summary and Evaluation.’’ In A. Inkeles,
N. J. Smelser, and R. H. Turner, eds., Annual Review Park, R. E. (1916–1939) 1952 ‘‘Human Communities:
of Sociology, vol. 6. Palo Alto, Calif.: Annual Reviews. The City and Human Ecology.’’ Collected Papers, vol.
2. Glencoe, Ill.: Free Press.
Kasarda, John 1995 ‘‘Industrial Restructuring and the
Changing Nature of Jobs.’’ In Reynolds Farley, ed., Phillips, E. Barbara 1996 City Lights: Urban-Suburban Life
State of the Union: America in the 1990s, Volume One: in the Global Society. New York: Oxford Universi-
Economic Trends. New York: Russell Sage Foundation. ty Press.
313
The words you are searching are inside this book. To get more targeted content, please make full-text search by clicking here.
Encyclopedia of sociology by Edgar F. Borgatta, Rhonda J. V. Montgomery (z-lib.org)
Discover the best professional documents and content resources in AnyFlip Document Base.
Search
Encyclopedia of sociology by Edgar F. Borgatta, Rhonda J. V. Montgomery (z-lib.org)
- 1 - 50
- 51 - 100
- 101 - 150
- 151 - 200
- 201 - 250
- 251 - 300
- 301 - 350
- 351 - 400
- 401 - 450
- 451 - 500
- 501 - 550
- 551 - 600
- 601 - 650
- 651 - 700
- 701 - 750
- 751 - 756
Pages: