Abstract
*Interpersonal violence has become one of the
main public health issues in Latin American
cities. This article presents a framework for sociological
interpretation that operates on three
levels, expressed in the factors that originate,
foment, or facilitate violence. Macro-social
factors include: social inequality due to the increase
in wealth versus poverty; the paradox of
more schooling with fewer employment opportunities;
increasing expectations and the
impossibility of meeting them; changes in family
structure; and loss of importance of religion
in daily life. At the meso-social level the analysis
highlights: increased density in poor areas
and urban segregation; masculinity cult; and
changes in the local drug market. The microsocial
level includes: an increase in the number
of firearms; alcohol consumption; and difficulties
in verbal expression of feelings. The
article concludes with an analysis of how violence
is leading to the breakdown not only of
urban life but also of citizenship as a whole in
Latin America.
—-
Violence; Sociology; Urban Health; Citizenship
Latin American cities have become the stage
for a silent and undeclared war. According to
the World Health Organization (WHO), there
are some 140,000 homicides a year in Latin
America 1. Most of these deaths occur in cities
and result from interpersonal violence, not
wars or armed conflicts. They represent everyday
violence, in which people meet death on
their own street corners.
For several decades, Latin American families,
institutions, and governments have made
enormous efforts to improve the population’s
health conditions, from prenatal care and immunization
to hospital care. As a result of such
efforts, life expectancy in Latin America increased
from 50 to 70 years in the second half
of the 20th century. Meanwhile, an entire generation
of parents moved to the cities in pursuit
of a better future with greater possibilities
for citizenship and better quality of life built on
citizens’ rights. These same parents raised their
families in the cities and offered them care and
education. They thus watched their children
grow healthier, stronger, taller, and with more
years of schooling than themselves, and full of
aspirations, only to see them murdered.
What has happened to the cities of Latin
America, the home to hopes and dreams, that
they have become a threat to the majority of
their inhabitants?
Briceño-León R 1630
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
Accelerated urbanization
Latin America has witnessed a process of breakneck
urbanization, with people not only living
in cities, but in increasingly larger cities, and
this has meant a major change in daily living
and public health conditions. In 1950 fewer
than one-half (41%) of the population lived in
the cities, but by 2000 the figure had increased
to three-fourths of the total population. The absolute
figures are even more impressive: while
in 1950 the total urban population of Latin
America and the Caribbean was 69 million, by
2000 it had increased to 391 million, that is, 332
million more people in the cities. In 1950, the
urban population was 40 million in South
America, 15 million in Central America, and 6
million in the Caribbean. Fifty years later there
were 228 million more in South America, 76
million more in Central America, and 18 million
more in the Caribbean (Population Reference
Bureau. World Population Data Sheet. Washington
DC; 2004). The figures are overwhelming
(Table 1).
In 1950, Buenos Aires had slightly more than
5 million inhabitants, and the other large Argentine
cities combined had fewer than 3 million.
In Brazil, Sao Paulo had 2.4 million and
Rio de Janeiro 2.8 million. Mexico City had 2.8
million. In fifty years the urban population increased
by more than twofold (Buenos Aries,
12.5 million), threefold (Rio de Janeiro, 10.4
million), six-fold (Mexico City, 18 million), or
seven-fold (Sao Paulo, 17.7 million). In 1950
there was only one city with more than 5 million
inhabitants, but by 2000 there were seven, because
in addition to the ones mentioned above
(Buenos Aires, Rio de Janeiro, Mexico City, and
Sao Paulo), other cities had past the five million
mark: Santiago, Chile (5.5 million), Bogotá,
Colombia (6.2 million), and Lima, Peru (7.4 million).
Combined, these seven cities were home
to 78 million people 2.
At the dawn of the 21st century, 60% of the
population live in cities with more than twenty
thousand inhabitants, and half of these 60%
(that is, one out of three Latin Americans) live
in the 50 cities with a million or more inhabitants
each 2,3. The main problem of violence in
Latin America is located in these 50 cities.
Increasing violence
According to the WHO 1, there are 520,000
homicides a year worldwide, or 8.8 murders
per 100,000 inhabitants, in addition to some
310,000 deaths classified as wartime casualties,
or 5.2 per 100,000 inhabitants. Homicides are
definitely a serious public health problem, surpassing
that of war, as seen from these data.
In the WHO regions, the highest homicide
rate per 100,000 is in Africa (22), followed by
the Americas with 19 and Europe with 8 1.
Even so, the 19 per 100,000 homicide rate in
the Americas is an average that conceals huge
differences in the region: first, between the
wealthy countries (the United States and Canada)
and Latin America, and then among the
Latin American and Caribbean countries themselves.
The United States and Canada have a
completely distinct social and economic reality
and must thus be seen as separate from the
other countries of the Americas (even though
the differences between the two countries is
large, since the United States has a historical
homicide rate of 8 per 100,000, several times
higher than Canada, with fewer than 2 per
100,000). The homicide rate in the United States
is also higher than that of several Latin American
countries, including Chile and Costa Rica.
Secondly, there are major differences between
Latin American countries themselves,
which we propose to classify in four groups, as
follows: 1) fewer than 10 homicides per 100,00
inhabitants: Argentina, Chile, Uruguay, and
Paraguay; 2) 11 to 20 homicides per 100,000 inhabitants:
Peru, Nicaragua, Ecuador, Dominican
Republic, Panama, and Honduras; 3) extensive
violence, with 21 to 30 homicides per
100,000: Brazil, Mexico, and Venezuela; and
4) extremely high homicide rates, which for
methodological reasons are classified as more
than 31 per 100 thousand inhabitants, but
which in practice can reach much higher levels:
Columbia and El Salvador 4.
Table 2 shows the extremes, with low homicide
rates in countries located in the Southern
Cone of the continent, plus some Central American
countries including Costa Rica. These
countries have had homicide rates between 3
Table 1
Urban population of the Americas, 1950-2000.
1950 2000
Million % Million %
South America 48 42.8 279 79.8
Caribbean 6 35.4 24 63.0
Central America 15 39.8 91 67.2
North America 110 63.9 239 77.2
Source: United Nations 2.
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
and 5 per 100,000, and while they experienced
an important increase in the 1990s (such as in
Uruguay, where homicides doubled), the levels
are quite low in comparison to the other Latin
American countries.
At the other extreme, with high violence,
are countries with intense social and political
conflicts and which have suffered from war,
such as El Salvador (from 1979 to the signing of
the Chapultepec Agreement in 1992) or Colombia,
where an armed conflict is still raging between
four factions disputing for control of the
territory: two guerrilla forces, one paramilitary,
and the country’s official army. The homicide
rates can exceed 60 or 100 per 100,000 inhabitants.
Still, most homicides in Colombia are
not the direct consequence of warfare, but of
daily violence. However it is impossible to determine
how many of the victims are the secondary
effect of wartime situation in El Salvador
5 or of covert or low-profile military action
in Colombia. Colombia also has the most
kidnappings in the world, estimated in 2004 at
more than three thousand persons in captivity.
Many kidnappings end in the victim’s death,
and such actions are a combination of guerrilla
warfare and common crime, often difficult to
differentiate 6,7.
Between the two extremes are the other
countries. Brazil, Mexico, and Venezuela have
experienced an increasing wave of violence
and homicides. There are no armed political
conflicts, so the violence is “routine” and associated
either with common crime and drug
trafficking or emotional conflicts in which hatred
and pain culminates in the use of firearms.
Such violence is primarily urban: in 1997 São
Paulo and Rio de Janeiro had an average of 600
homicides per month 8. Homicide rates in some
cities can more than double the national average:
for example, Rio de Janeiro with 102, San
Salvador 139, and Caracas 52. In Cali, Colombia,
where an effort to reduce violence has frequent-
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1631
ly been cited as successful, a homicide rate of 91/
100,000 persisted from 2002 to 2004 9,10,11,12,13.
Some pockets of rural violence exist, but
they are few. The Latin American rural population
has remained stable at around 128 million,
and rural poverty (while more dramatic) has
not increased as rapidly as urban poverty 14.
Traditional rural violence and some political
conflicts continue in peasant areas, including
the guerrilla warfare in Colombia and Mexico,
but the number of homicides they produce is
insignificant as compared to deaths in cities.
Cities of law, cities of fear
Cities should be the locus for rights and safety,
for life rather than death. Citizens as bearers of
rights originated in the city, and for centuries
people saw the city as a refuge against the insecurity
of the countryside, as well as a source of
rights. Greek tradition made a synonym of the
citizen as a person with the right to live in the
city and the right to decide the political future.
To be citizen requires both living in the city and
having rights to participate in political life.
The city was the place for exchange (the
market), but also for order and norms. Weights
and measures regulating relations between
buyers and sellers were created in the city. But
the city was also a place of domination, since
much of the order had to be imposed, and this
meant forcing submission.
The city was the place of safety: people were
controlled, areas were lit, and the police protected
the citizenry. Citizenship originated in
the city, in the connection between equals,
subject to the law, not to others.
The city was also the place of civility: good
manners, courtesy and hypocrisy, all that was
called urbanity 15,16.
Latin American cities represented the refuge
of an elite that had (and exercised) rights, but
Table 2
Classification of Latin American countries by intensity of violence (2000).
Level of violence Homicide rate Countries
(per 100,000 inhabitants)
Low Fewer than 10 Argentina, Chile, Costa Rica, Uruguay, Paraguay
Medium 11 to 20 Peru, Nicaragua, Ecuador, Dominican Republic,
Panama, Honduras
High 21 to 30 Brazil, Mexico, Venezuela
Very high More than 31 Colombia, El Salvador,
Source: 1,102.
Briceño-León R 1632
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
from which the vast majority of the rural population
was excluded, subject to semi-feudal
working conditions and political and social
power, often controlled by private armies enforcing
their lords’ authoritarian and personalized
will. This was the history of the rural
“bosses” or “colonels”, where the law was enforced
with little impartiality. Migration to the
city represented the dream not only for a better
material life, but also a life with rights, where
one could live under the rule of law, not in submission
to individual power.
Despite multiple urban problems in countries
like Colombia, thousands of persons have
moved from rural areas to the cities, fleeing the
arbitrary rule and the horrors of guerrilla and
paramilitary violence. These “displaced” persons
as they are called in Colombia total nearly
three million who have abandoned their homes
and belongings to seek refuge in the large cities,
where they expect peace, security, and rule of
law.
Latin America cites were a place of hope for
security and law, hence the great rural-to-urban
exodus in the 1940s and 50s.
However, beginning in the 1980s, violence
changed significantly, and homicide rates doubled
in most countries 17,18. Worldwide, the estimated
homicide rate (per 100,000) increased
from 5.47 in 1975-1979 to 8.86 in 1990-1994 19.
In Venezuela, it increased from 8 in the early
1980s to nearly 25 by the mid-1990s 20. In Mexico,
in the early 1980s the homicide rate was
10.2/100,000, and by 1995 it had reached 19.6
(an increase of 90%) 21,22,23,24. In Colombia the
rate increased from 20-40 per 100,000 in the
1970s to 70-90 in the 1990s 25. There was an increase
in all Latin American countries, regardless
of the previous conditions (slightly or
highly violent).
This situation led to a feeling of great insecurity
in the cities 26,27. Fear is distributed on a
more egalitarian basis than that of the population’s
real security, because the role of the mass
media, vicarious victimization, and rumor lead
to similar feelings in victimized and non-victimized
groups 28,29,30. Fear displays a subjective
sensation, but that has practical consequences
because people act as if it were true 31.
In a multi-center study sponsored by PAHO in
1996, a group of questions addressed the sense
of insecurity that persons felt in different areas
of the city: home, street, and public transportation.
The results (Table 3) showed a strong sense
of insecurity in the city centers, including
Madrid, which was used as a control for comparisons
because of its low crime rate. In order
of importance, the fear of becoming a victim of
violence in public transportation was next. In
Bahia, Brazil, people felt the greatest insecurity
in public transportation, but in nearly all Latin
American cities buses have become a trap where
passengers and drivers can easily be assaulted
while the vehicle is in movement. The most routine
acts become a source of fear and threat.
Cities are no longer the source of security
once dreamed of. Mothers of tens of thousands
of youths murdered every year and the fearful
inhabitants of cities ask themselves the same
question: why has violence increased so much
in the last twenty years?
A sociological framework
for explaining urban violence
To understand violence, we present a set of hypotheses
on the different dimensions of the
phenomenon. However, since variables of a different
order are concerned, we have grouped
them into a sociological model we developed
under the Social Sciences Laboratory (LACSO)
to provide an explanation for current violence
in Latin America. This framework is not intend-
Table 3
Feeling of insecurity (somewhat insecure, very insecure) in different areas of the city, 1996-1997.
Bahia, Caracas, San Jose, Santiago, Madrid,
Brazil (%) Venezuela (%) Costa Rica (%) Chile (%) Spain (%)
At home or in apartment 64.5 74.8 11.4 12.0 4.7
On street, daytime – 74.6 29.0 18.3 12.1
On street at night – 83.9 51.0 41.6 47.7
In public transportation 91.9 89.3 45.3 65.7 37.1
In city center – 91.1 81.3 71.3 47.2
Source: Activa Survey, PAHO, LACSO (Social Sciences Laboratory).
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1633
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
ed to be exhaustive. Rather, it situates the social
and psychosocial conditions that we consider
relevant for comprehensive sociology,
thereby differentiating it from explanatory proposals
that are individual per se, such as the
Bandura theory of aggression 32,33, ecological
theories 34, economic theories 25, and of course
classical criminology 35 and the theory of deviation
36,37, even while recognizing their contribution
and sharing some aspects with all of
them.
The current proposal is not a model for universal
explanations, because we believe it is
impossible to interpret violence in the same
way for ancient Greece, World War II, crimes
committed by stilted lovers, or serial murders
38,39. Each of these phenomena requires a unique
explanation, because science can offer only conjectures
on specific matters that will never be
universal 40,41.
The sociological framework seeks to work
with various spheres of social life: 1) the situational,
which refers both to general conditions
of society and specific circumstances, such as
physical circumstances in the environment
and social circumstance of other actors, which
are imposed on the individual as binding references
at the time decisions are made; and 2)
the cultural, which is external to the situation,
precedes it in time, and is imposed on individuals
in social learning and marks the way they
interpret signals sent to them by the situation
(the medium or the other actors) and how they
may decide their course of action 42,43,44.
The purpose of this framework is to formulate
hypotheses 45 on the two above-mentioned
dimensions of the social (i.e., situational and
cultural) so that as hypotheses of truth they allow
one to understand 46 what takes place in a
specific social reality. And since we are optimists,
this framework may allow us to propose
an appropriate intervention.
The framework has three dimensions, representing
three distinct levels of explanation
(Figure 1), as follows:
1) The structural level refers to a social
process of a macro nature and with a genesis
and persistence over a longer period of time. At
this level we refer to factors that originate violence,
since their structural character has an
inevitable imprint on society as a whole, giving
it a generalized and diffuse effect. It is thus not
easy to identify immediate associations with
variables at this level, but it determines a transformation
in society that creates the basis for
violent behavior but does not determine what
necessarily occurs. Due to their characteristics,
these circumstances are the most difficult to
change, but perhaps for this very reason they
are the most relevant as first causes.
2) The second level contains meso-social
aspects, with less structural roots and thus representing
the area where the situation and culture
have a more immediate effect on behavior.
Figure 1
The sociological framework of violence levels.
Factors that originate violence
Factors that foment violence
Factors that facilitate violence
Briceño-León R 1634
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
At this level we thus refer to factors that foment
violence. Factors at this level may encourage
and facilitate violence, but modifying this level
is simpler than the previous one. The level of
freedom of individuals in relation to these factors
is much greater than in the previous case.
3) The third level includes micro-social factors
which we also call facilitators because they
have a more individual nature and cannot be
considered causes, but rather accompanying
factors and facilitators for the passage to the violent
act or as responsible for an action’s lethalness.
The connections here are more immediate
and the associations are easier to establish,
but they also indicate more association than
causality.
The macro-social level:
factors that originate violence
At the macro-social level we postulate five
types of factors. Two are of a situational nature:
increased urban inequality and increased education
and unemployment. Two others are of a
contingent nature, namely increased aspirations
and the impossibility of satisfying them,
and changes in family structure. One is cultural,
namely the decreasing capacity of the Catholic
Church to exert social control.
The city has both more wealth
and more poverty
In Latin America, the distance between the poor
and the rich is the greatest in the world. On
other continents, such as Africa, there is more
poverty, but there is also not as much wealth. In
Europe, on the other hand, there is more wealth
but not as much poverty. What is unique in
Latin America is the presence of both components:
more poverty and more wealth. There is
thus more inequality than if everyone were rich
or poor.
When data on distribution of wealth between
the poorest and the wealthiest groups in
the developed countries and Latin America are
reviewed (Table 4), the portion that the poor
detain is quite similar, while that detained by
the wealthiest 10.0% in Latin America is much
greater than what the same group detains in the
developed countries. Inequality is determined
by what the wealthiest detain, not the poorest.
And this is true in all Latin American countries.
Of course there are differences between the
countries. In Brazil, the wealthiest 10.0% detain
45.0% of the wealth, as compared to 27.3%
in Uruguay; but although the percentage is
lower in Uruguay, it is still very high. Moreover,
in all Latin American countries the poorest
10% detain less than 2% of the wealth 47,48.
This situation has changed over time, but
unfortunately not for the better, based on either
the extreme polarization of income (the poorest
1.0% and the wealthiest 1.0%) or the poorest
and wealthiest 25.0%. Londoño & Szekely 49
calculated the income distribution ratio between
the wealthiest fourth (25.0%) and the
poorest fourth using the Lorenz curve and found
important modifications. In 1970 the ratio was
22.9, but there was an improvement in the income
distribution during that decade that led
to a decline in the ratio by 1982 (when it was
18.0). From then on the situation of inequality
deteriorated, pushing the ratio back to the same
level in 1990 as 20 years earlier (22.9), after
which it worsened until reaching 24.4 in 1995.
In 1970, the average income of the poorest 1%
of the population was US$112, reaching US$159
in 1995. That is, the poor improved their real
income as measured in 1985 dollars. But during
the same period, the wealthiest 1% increased
their average income from US$40,711
to US$66,363. The income of the 1% wealthiest
in 1970 was 363 times that of the poorest 1%,
but in 1995 the ratio had risen to 417. What
changed was not poverty, but inequality.
Many of the changes in violence can be related
to transformations in Latin America. The
1980s were marked not only by stagnation, but
also by greater inequality, reflected both in
people’s living standards and fundamentally as
an increase in urban poverty 50. While in rural
areas there was a greater percentage of moderate
and extreme poverty than in the cities (62.0%
vs. 38.0% and 38.0% vs. 13.5%, respectively), this
did not represent major changes because between
1980 and 2002 the percentages did not
increase as much in the countryside as in the
cities, nor was the increase in the absolute
number of persons as great as in the cities, because
the rural population remained relatively
stable during this period.
As shown in Table 5, the percentage of the
poor or indigent population increased from
1980 to 1990 and then dropped until 2002 for
all rural and urban groups. Nevertheless, in
1980 there were 73 million poor in the rural areas,
increasing to 74.8 million in 2002, that is,
two million more poor in the countryside.
Meanwhile, in the cities there were 136 million
poor in 1980 and 221 million in 2002, or 85
million new poor in the cities. The same was
true for extreme poverty, with increased by 6
million in the countryside and 29 million in
the cities 51.
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1635
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
The important points in relation to violence
are urban poverty and indigence, due to the
magnitudes they involve and because unemployment
is a particularly urban issue. Unemployment
is low in rural areas, where people
can always participate in farming. As Table 6
shows, there is an important difference between
three groups of Latin American countries
in terms of poverty, urbanization, and violence.
In groups 1 and 3, violence as measured
by the homicide rate is low, but the factors producing
this situation are different in the two
cases.
In the first group, we contend, there is little
violence because of a low poverty level and
high urbanization. The exception is Costa Rica,
which is not highly urban but is unique in Central
America since it has had some singular social
control mechanisms and is the only Latin
America country that eliminated its army several
decades ago.
The third group has Latin America’s highest
poverty levels, but the poverty is rural because
these countries have low urbanization. Honduras,
with the most poverty among the countries
in this group, has less than half of its population
living in cities.
We thus hypothesize that violence is concentrated
in the countries of Table 6 where
there is high poverty and high urbanization,
that is, with urban poverty. An exception is Argentina,
with a low homicide rate 52. Historically,
Argentina has had low poverty and a large
middle class, but the data in the chart reflect
the country’s crisis, with a 132 billion dollar
debt and recession since 1998, reaching its peak
with the freeze on savings deposits, protests by
the middle class on December 20, 2001, and
the suspension of year-end bonuses. We believe
the violence in Argentina will tend to appear
increasingly like that of Brazil or Mexico rather
than Chile, and we thus chose to keep Argentina
in this group. There is no reliable information
on homicides in Guatemala or Bolivia.
More education, less employment
Cities have offered greater access to education for
broad segments of the Latin American population.
Education in the rural areas has always been
difficult, both because of child labor in agriculture
and the scarcity of (and distance to) schools.
In urban areas, education has been different and
despite numerous multiple limitations, by the
end of the 20th century 86.0% of youth ages 15 to
29 years had managed to conclude their primary
studies while 26.0% from 20 to 24 years of age had
completed secondary education.
However, this improvement in education
has not represented better opportunities for
young people to obtain employment or rise socially.
According to the International Labor Organization
53, the unemployment rate among
youth worldwide is two to three times that of
adults. In Latin America, the unemployment
rate among adults dropped in the late 1990s,
and in 2003 it was estimated at 6.7%. Among
youth the situation was quite different, since it
increased to 15.7% (more than double that of
adults) by that same year.
This youth unemployment has some singularities,
behaving like a Gaussian curve, but in-
Table 4
Distribution of wealth in Latin America and the developed countries.
Latin America (%) Developed countries (%)
wealthiest 10% 48.0 29.0
poorest 10% 1.6 2.5
Source: Ferranti et al. 47.
Table 5
Latin America: moderate and extreme poverty in urban and rural areas.
Moderate poverty Extreme poverty
Urban Rural Urban Rural
million % million % million % million %
1980 135.9 29.8 73.0 59.9 22.5 10.6 39.9 32.7
1990 200.2 41.4 78.5 65.4 45.0 15.3 48.4 40.4
2002 221.4 38.4 74.8 61.8 51.6 13.5 45.8 37.9
Increase 1980/2002 85.5 8.6 1.8 1.9 29.1 2.9 5.9 5.2
Source: Assembled from Economic Commission for Latin America and the Caribbean 51, charts I.2 and I.3.
Briceño-León R 1636
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
verted, that is, with greater employment opportunities
for groups with either very little or
a lot of education: for those with more schooling
because they are more skilled and reached
the university, and for those with less schooling
because they live in the countryside (where
there is less unemployment as a social category)
or are employed in heavy labor with low
wages, which individuals with some education
tend to refuse. Meanwhile, the group from 15
to 24 years of age and with seven to twelve
years of schooling suffers the highest unemployment
rate in the region. It is also the group
that both suffers from and perpetrates more violence.
An estimated 565 young people (10-29
years) are murdered per day worldwide. In
2000, 199,000 young people died of violence,
with a homicide rate of 9.2/100,000. Latin
America has a major share in that figure. The
overall world rate varies greatly among regions,
because there is less than 1 homicide per
100,000 in Europe, 11 per 100,000 in the United
States, 17.6 per 100,000 in Africa, and 34.6 per
100,000 in Latin America. Juvenile violence is
clearly a Latin American problem. The highest
rates in the world occur in Latin America: Colombia
with 84.4, El Salvador with 50.2, Brazil
32.5, Venezuela 25, and Mexico 15.3 homicides
per 100,000 1,54,55,56.
Violence is a youth issue. An estimated
28.7% of all homicide victims in Latin America
are from 10 to 19 years of age (Inter-American
Development Bank. Youth Violence Prevention.
Technical Note 10. Washington DC; 2002). Why
is this age group affected so heavily? There are
several reasons, but we emphasize the difficult
age of adolescence, complicated by social conditions.
People used to be classified in three
ages: childhood, adulthood, and old age. Adolescence
did not exist. Only recently has this
category designated a moment of change in
the individual’s biology, but also to represent
changing roles assigned to the adolescent by
society. Adolescents are neither children nor
adults. They have the physical conditions to
work, but laws prohibit them from doing so until
adulthood; they have the physical conditions
for reproduction but are prohibited from
exercising their sexuality. Presumably they
must study until they reach working age, but
they either have no schools or are expelled
from the educational system. Imprecise, inadequate
social insertion of adolescents is one an
Table 6
Latin America: households in situation of poverty, urban population, and homicides.
Level of violence Countries Poverty rate Urban Homicide rates
(households) population (%) per 100,000
Group 1: low Uruguay 9.3 93.0 4.4
Chile 15.4 87.0 5.4
Costa Rica 18.6 59.0 9.3
Group 2: high or very high Brazil 29.9 81.0 23.0
Argentina 31.6 89.0 9.9
Mexico 31.8 75.0 19.6
Peru 42.3 72.0 11.5
Ecuador 42.6 61.0 15.3
El Salvador 42.9 58.0 55.6
Venezuela 43.3 87.0 35.0
Colombia 48.7 71.0 61.6
Group 3: low Paraguay 52.0 54.0 12.6
Guatemala 52.3 39.0 –
Bolivia 55.5 63.0 –
Nicaragua 62.9 58.0 8.4
Honduras 70.9 47.0 9.4
Source: Assembled from WHO 1, Londoño & Guerrero 10, Buvinic et al. 19, Fundación Mexicana
de la Salud 21, Population Reference Bureau 103, Economic Commission for Latin America and
the Caribbean 51, Instituto Apoyo 104, Lederman 105.
Data on poverty Peru 1999; Brazil, El Salvador, Paraguay, and Nicaragua 2001; Chile 2003,
other countries 2002. Data on population from 2000. Homicide data from 1994 to 1999.
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1637
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
important source of violence, with the incapacity
to make the prescribed and proscribed roles
coincide.
Juvenile violence begins at a prime moment
in adolescence at around 13 years of age, when
youth begin demonstrating the pretensions of
being adults but limited social capacity to behave
as such. At around 13, boys begin to take
interest in women, but young girls of the same
age are more interested in adult men, and
younger girls are still children. At this age they
begin having problems with their schoolwork,
in the 7th or 8th grades, and many drop out of
school, but they are not of age and lack technical
skills to work. This is a group of youth who
do not study or work and who are at great risk
of getting trapped in violence.
As shown in Table 7, 12.0% of all Latin American
youth have no work and are not attending
school. And what can young people from 15 to
18 years of age do if they are not working or
studying? The vast majority do not live in wellto-
do families that can support them and supply
them with resources to meet their needs.
Although this is scarcely one-tenth of the population,
there are some 58 million poor Latin
American young people, 21 million in extreme
poverty. Even if one assumed that 90.0% are
male saints with extremely positive value socialization
and never tempted by evil, 10.0%
of youth are still at risk of crime. At best, this
percentage represents 580,000 poor youth or
210,000 youth in extreme poverty and susceptible
to crime and violence, more than the current
Latin America prison population.
More aspirations, but less capacity
to meet them
Youth outside the labor market and school system
do not have fewer expectations or dreams
than others. Their aspirations are the same as
those of other young people who are in school
and have good incomes, because the mass culture
transmits the same ambitions.
During the 1940s, the sociology of modernization
emphasized the “revolution of expectations”.
According to this theory, when rural
Latin Americans living in a traditional society
made contact with the city and modernity, they
were bound to change their expectations and
dream of a better life, represented by greater
and better consumption, and the fetters on society
would be broken and social forces would
awaken in a development “take-off” 57.
The process actually occurred in Latin America.
The rural exodus changed the expectations
of millions, placing them on equal ground with
the rest of society in terms of expectations.
Paradoxically, the same society denied them
the means to satisfy these expectations. Latin
America shows an asymmetry between aspirations
and the capacity to satisfy them. We are
terribly equal in our desires and frightfully unequal
in our actual possibilities to achieve
them.
The first generation that reached the cities
realized a major share of their dreams: access
to hospitals, schools nearby for their children,
electricity, refrigerators, and TV sets. They had
nothing of the kind in the countryside, and this
was an important change in their lives. Yet their
children were born in a world that already had
hospitals, schools, refrigerators, and TV sets.
For the new generation, these achievements
mean nothing. The children were born into a
world where mass culture imposed new and often
more superficial consumer goals. A young
person from a middle-class family, preparing
to enter the university, and an unemployed
youth from a poor family have the same tastes
and the same aspirations. Urbanization and
television democratized expectations. In 1980
there were 98 TV sets for every thousand Latin
Americans, and by 1997 the number of TV sets
had doubled, at 205 per thousand (UNESCO.
Radio and Television Receivers. Toronto: Institute
of Statistics; 1999). But the latest Nike or
Reebok shoes advertised on TV are out of reach
for youth from the favela in Rio de Janeiro, the
Table 7
Latin America: youth employment and school enrollment.
Works and Only works (%) Only studies (%) Does not work Domestic
studies (%) or study (%) activities (%)
Males 52.7 10.9 22.2 12.3 1.9
Females 28.3 7.8 24.3 14.1 25.6
Source: Assembled from Economic Commission for Latin America and the Caribbean 51.
Briceño-León R 1638
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
comuna in Medillìn, or the colonia in Mexico
City. The 150 dollars that some brand name
shoes cost are more than most young Latin
Americans earn in a month.
Youth have problems not only finding employment;
when they to get jobs they earn less
than adults. According to the International Labor
Organization 53, 93.0% of the jobs available
to young people in the world are in the informal
sector, where they earn 44% of what they
would earn in the formal sector. According to
the Economic Commission for Latin America
and the Caribbean 51, the average income of
young people (15-19 years) is one-third that of
adults, and those from 20 to 24 years of age
earn a little more than half that of adults.
This asymmetry between expectations and
achievements raises a classic sociological drama
58, because as the means prescribed by society
(employment, effort, and savings) do not
allow achieving the ends, many youth take the
proscribed routes of violence as a means to
seize what they cannot attain formally 59,60,61,62.
A youth who sells drugs in Caracas said proudly
during an interview in a juvenile detention
center that he earns on a single Friday night
more than his neighbors do carrying packages
for a month. And he added petulantly that he
was not born to be poor, because as the Mexican
song says, he likes the good life (“Yo no
nací pa’ pobre, me gusta todo lo bueno…”).
Less social control by the family
One of the greatest forces for containing violence
is the family, because it incorporates the
person into a world governed by norms and
with limits. The family teaches children the difference
between what is permitted and prohibited,
initiates them in regard to rewards and
punishments, and introduces them (starting
with the first rule of prohibition of incest) in
the symbolic pact that is the law 63,64. The family’s
influence is as much original and past as
situational and present. Past because it is the
basis for the formation of the individual and
present because it is the context of close social
interaction that can regulate and modulate behavior.
Changes in the Latin American family
thus have an important impact on the violent
individual conduct.
The family has lost force in its two functions
of social control because of the transformations
it has experienced in recent decades.
As shown in Table 8, the traditional family, in
which the father works and the mother stays
home in charge of the housework and caring
for the children, decreased by 10.0% in the
1990s. The change is not only because women
who have finished school and want a satisfying
professional career with financial independence
leave home to work, but also because
the man’s income drop has forced wives to seek
a second income for the household. Under
such conditions, socialization and social control
of children have been seriously jeopardized,
especially because urban life has limited
the presence of the grandparents in many families,
and children are thus left alone for long
periods of the day.
In addition, as shown in Table 8, there are
more single-parent households as a result of
broken relations, whether because of more divorces
in legally constituted couples (as in all
Latin American countries) or due to the dissolution
or non-consolidation of common law
marriages. In 2002, 16.0% of Latin American
families were headed by women, and 37.0% of
these were poor. The percentage varies among
countries. In Colombia, 46.0% of single-parent
families headed by women are poor, 48.0% in
Ecuador, 44.0% in Argentina, 32.0% in Brazil,
and 27.0% in Mexico. For every three singleparent
families, two have a father or a mother
who leaves home to work, so if these families
have no grandmother or other relative available
to care for the children, the latter are left
alone at home or on the streets and in many
cases under the care of an “older” sister who (as
we have observed many times) is rarely more
than 10 years old but is already responsible for
cooking meals and caring for her younger siblings.
This precarious social control by the family
has multiple consequences. One of the most
immediate is to place youth in the street at the
disposition of professional criminals. Families
have little power to control criminal or violent
acts by youth. We asked a 17-year-old (who was
in jail in Caracas for robbery and two murders)
what his mother said when he began giving her
money two years earlier: “At first she asked me
where I got it, but later she didn’t say anything.
What was she going to do? She needed the money.
Other times she cried, but she took it”.
Less force of religion
Religions have always been an important instrument
for regulating individual behavior.
The Fifth Commandment of the Old Testament
is an irrefutable test: “Though shall not kill” is
an explicit order. How many such orders are
obeyed (even by the most religious) is a matter
of time and the moment, that is, of the history
in which decisions are made and norms ap-
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1639
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
plied, since even the Vatican instituted capital
punishment.
In Latin America, the Catholic religion has
lost considerable capacity to affect daily individual
lives. Religion remains, and at least
70.0% of the population is still Catholic and follows
the fundamental rites of human existence:
baptism, religious marriage, and death rites.
But little can be said of its impact on the daily
behavior of persons beyond the broad guidelines
over life. The story is quite different with
Protestants, including Evangelists, who have
experienced major growth throughout the region,
since they exert great control over individual
behavior: daily worship and Bible reading,
tithing, and bans on alcohol and tobacco
consumption, vulgar language, and violent behavior.
The types of control and level of compliance
vary from one denomination to another
and from one country to the next, but they
generally wield more force than the Catholic
Church.
In the Activa survey sponsored by PAHO in
seven Latin American cities, we attempted to
identify a correlation between type of religion
and violent attitudes, and the only significant
association was found in Rio de Janeiro, where
Protestants manifested a stronger rejection of violence
than Catholics. In Caracas, non-Catholic
Christians rejected extra-judicial police activity
more than Catholics, and the non-religious
were those who significantly approved of it 65.
Church attendance functioned, because those
who worshipped regularly were less violent
than those who never did. A recent study by
LACSO in Caracas used the “data mining” technique
to analyze data from a survey on violent
attitudes in 2004 and showed that the main distinguishing
factor was religion, i.e., whether
the individual practiced a religion (regardless
of which one).
When we asked violent youths about religion
and if they considered their actions “sinful”,
they responded resignedly that they were.
Yet religious morals and real action appear to
be two distinct registries. To kill someone is not
good, but it is done, and the justifications vary.
One powerful excuse is self-defense and the
belief that if they do not kill others, they themselves
are condemned to die. For many, religion
no longer inhibits violence, but has not
been replaced by lay morals, which could be
supported by the rule of law to dissuade homicidal
behavior.
Factors that foment violence
Meso-social factors are a second group referring
to specific situations that increase violence
by fomenting a type of exacerbating behavior.
There are three most important factors of this
type. Two are situational: urban segregation,
producing divided cities, and the local drug
market. One is cultural: the masculinity cult.
Segregation and urban density
Latin American cities experienced slow growth
during much of the early 20th century, and
their expansion added new territories to the
edge of cities. Land on the urban outskirts was
less valuable and lacked services. Urban workers
and the recently arrived poor built their
housing there, congregating people with less
schooling and characterizing a type of social
behavior known in many countries by the Spanish
term orillero, a synonym for coarse, uncivilized
behavior. Yet migration to the cities turned
secondary growth of the outskirts into a key
factor of urban life. Favelas in Brazil, comunas
in Colombia, barrios in Venezuela, and pueblos
jóvenes in Peru became essential urban components,
sometimes larger than the formal city
itself, although the urban authorities resisted
recognizing them as such.
The presence of these new social groups
occupying urban territory has been interpreted
in various ways and with dissimilar theoretical
frameworks, but all identify the union and separation
between the formal and informal city,
between the legal and illegal, the planned and
the unplanned. In the mid-1950s some authors
Table 8
Types of urban nuclear families.
Type of family 1990 2002
Two-parent family
Spouse does not work 46.2 36.2
Spouse works 27.0 32.9
Total two-parent 73.2 69.1
Single-parent family
Female head who works 8.0 10.3
Female head who does not work 5.4 5.7
Male head 2.1 2.5
Total single-parent 15.5 18.5
Source: Assembled from the Latin American and Caribbean Center
for Demography 106.
Briceño-León R 1640
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
using the theory of modernization contended
that these areas were a leftover from the rural
countryside and a tradition that had been installed
in the modern cities 57,66,67. Others,
based on Marxist categories, felt that an “industrial
reserve army” or a “relative super-population”
lived on the outskirts 68,69. According
to others, there were two urban circuits that
functioned differently but were integrated at
some points 70. Still, all attempted to explain a
phenomenon that still calls attention, since it
was a novel urbanization process, different from
that of Europe and not preceded or accompanied
by the city’s industrialization. It was urbanization
without industrialization 71,72. This
meant that many new urban residents experienced
difficulties finding employment and
housing. Since society could not offer an answer,
they found one themselves in the interstices
of land ownership, a space that formal
urbanization had overlooked and where they
proceeded to build their homes. And they found
work in what we now call the informal sector.
For several decades Latin American cities
grew like this, and 30 to 80% of the urban population
came to live this way. Urban inhabitants
made great efforts to improve the city, as
they tell in beautiful stories in all the countries
73. Until the 1980s, such growth always meant
systematic consolidation of housing and the
physical surroundings. Visitors may have been
confused when they found precarious housing
and classified it as deteriorated (as one school
of North American sociology contended), but
this was not true, because the conditions improved
year by year; the makeshift walls and
roofing were replaced with solid masonry walls
and tile roofs, running water and electricity
were installed, and streets and stairways were
built. The time for this transformation varied
from one country to another. In the 1960s the
transformation of a house could take five years
in Caracas and ten years in Lima. Family incomes
in the cities were very different, but there
was a general feeling of living better each day.
That situation changed. In the early 1970s,
36.0% of the Latin American urban population
lived in poverty. This figure increased to 60% by
the early 1990s, even though the urbanization
rate had decelerated significantly. This increase
in urban poverty has had three important impacts
as factors for violence. First, houses that
were continuously improving have begun to
deteriorate, and the feeling of increasing improvement
has been replaced by a bitter emotion
of being increasingly worse off. This relates
to the aging of the housing’s physical structure,
because after 30 or 40 years it requires maintenance
that the inhabitants can scarcely afford
because of their shrinking income. Second, the
population density has increased in low-income
areas due to natural demographic growth,
because children and grandchildren of the
original occupants continue to live on the same
lots and the formerly empty spaces are now occupied
(the backyard and the area between the
house and the neighbors). When there is no
more ground space, a second and up to seventh
floor are built on the land originally occupied
by a makeshift single-family dwelling 74.
In Caracas, in some low-income barrios (like
Los Erasos), there is a higher density than in the
area with the city’s tallest residential buildings
(Parque Central). From an ecological perspective,
high density is a reason for permanent
conflicts between people, both because of aggression
that appears with many people and
few effective norms for cohabitation, but also
(and this is the third factor) because unplanned
urban growth and subsequent densification
produce tortuous territories that are easily
controlled by criminal groups and refractory to
efficient and secure action by the police 75. A
similar situation occurred in medieval cities,
thus the large avenues and diagonals in Paris,
built by Baron Haussman after the Revolt of the
Commune for the Prussian army to be able to
move and take control of the different areas,
avoiding irregular alleys with spontaneous urbanization.
The culture of masculinity
Violence is a men’s issue. Men both practice
and suffer from it. Worldwide, the male homicide
rate is three to five times that of women.
Up to 14 years of age, there is no difference between
the sexes, but from 15 (when conduct is
defined by sex) up to 44 years of age the difference
is abysmal, because the male rate is five
times greater: 19 homicides per 100,000 inhabitants
in males as compared to 4 per 100,000 in
women 1.
In Latin America the situation is even worse.
In Colombia, El Salvador, and Venezuela, the
probability of men being murdered is 12 times
that of women, in Ecuador 11, and in Brazil 10.
The situation is similar in countries with low
homicide rates, like Chile and Costa Rica, where
the probability is 6 times greater in men than
women (Table 9).
Why is there such a marked difference between
the sexes? We believe that there is a culture
of masculinity that favors violent actions
and exposure to the risk of violence. This culture
exists, as in all relations defined by sex, as mark-
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1641
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
ing difference in what is considered feminine
culture vis-à-vis risk and violence. Men act in a
way that differentiates them from women and
thus become victims of violence 76,77,78.
Female culture basically involves a conduct
of avoidance. Women avoid conflict, fighting,
and risk and do not care if they are labeled as
afraid. This is important for men, because to
avoid fighting and risk is equated with behaving
like a woman, a dangerous situation for a
15-year-old in a male chauvinist culture, because
it means losing face and becoming the
object of ridicule and social discredit. Assuming
a conduct stigmatized as feminine means
to submit to other men.
The culture of masculinity acquires special
dimensions during adolescence, when males
seek the construction of their own identity. It is
a difficult moment for males and females under
any circumstances, but in relation to violence
it is much worse for men, who are obliged
to reaffirm themselves in a culture of masculinity
that exposes them to risk 79. The culture
of “respect” as recognition of their identity
and virility by their peers acquires much greater
force at this time 80,81. Respect is an important
component of masculinity in different societies
and ages, but among youths it is more relevant
because of their own lack of identity. Becoming
a man in a low-income context is hard for youth,
and violence is a way of growing. The research
thus showed the ostentatious exercise of violence
as taken for granted among younger adolescents,
before they are consolidated as outlaws
respected on their turf. Once they are recognized
as such and begin a stable sexual life
with a partner, they reduce their excess violence
and begin to administer it with a rationality
in keeping with their goals 82,83.
The local drug market and impunity
Drug use itself does not appear to be a major
factor for violence, but the drug market is. Users
may adopt violent behavior while under the influence,
but this is not common. It tends to happen
more frequently during prolonged abstinence
by addicts or when they commit crimes
to buy drugs, but not while they are under the
influence 84.
The main problem is the drug market and
its transformation beginning in the 1980s. For
some time the commercial arrangement between
wholesalers and retailers was cash payment
of a commission on the sale of a specific
drug. A dealer sold a kilo and was paid a thousand
dollars, for example. This situation changed
in many places in the 1980s, and cash payment
was replaced by payment in kind, that is with
drugs. The retailers earned more money, because
the drugs received in payment were
worth much more than the previous cash payment
and wholesalers were now able to eliminate
the problems with employee payments by
converting them into businessmen through a
kind of outsourcing.
The problem now shifted to the retailers,
who had to sell more drugs to make a profit.
There were two possibilities: either the regular
customers bought more, or they expanded the
market to include new buyers. This was not so
simple, thus the easiest way to expand one’s
market was by eliminating other sellers. And
this is the story of gang war, or the turf war to
control local drug markets.
A study on homicides in 1995-1996 in Cali,
Colombia 85 showed that 15.0% of homicides
were connected to drug trafficking, but on observing
the murders in detail the figure increased
to 46.0%, because it included 20 double
homicides, 5 triple homicides, and one quadruple
and one sextuple homicide in 1996, in addition
to 14 victims of professional assassins.
This drug-related violence involves high
levels of impunity. Punishment for homicide is
rare in Latin America, but the connection to
drugs has aggravated this reality and fomented
the perception that drug crime goes unpunished.
Drug earnings are so high that they have
corrupted the police in various countries, but
when something goes wrong with bribes or
death threats, they have also controlled judiciary
officials (judges and attorneys become
victims of violence). And if something still goes
wrong and traffickers go to jail, the drug lords
provide them with protection and comfort in
prison. In Latin American jails, where everything
costs money, drug-related inmates have
Table 9
Latin America: homicide rates according to gender in selected countries.
Men Women
Colombia, 1995 116.8 9.0
El Salvador, 1993 108.4 8.4
Brazil, 1995 42.5 4.1
Venezuela, 1994 29.7 2.3
Mexico, 1997 29.6 3.1
Ecuador, 1996 28.2 2.5
Costa Rica, 1995 9.3 1.4
Chile, 1994 5.4 0.8
Source: Assembled from WHO 1.
Briceño-León R 1642
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
separate sections or special rooms, appliances,
cell phones, and even bodyguards hired within
the prison institution, which they can afford
with money from the organization.
With prospects of such high earnings as
those from the drug business and such limited
risk of being arrested and punished, drugs become
an alternative for many people and an
important fomenter of violence in Latin America.
An even greater effect is the tremendous
damage from drug business on penal institutions,
which is not restricted to impunity for a
specific trafficker, but rather the deterioration
of the system’s overall operation. The penal
system requires institutions and an ideology
that sustains it and legitimizes its sanctions.
When fear and money grip officials, the crisis is
not of this or that judge, but of the institution
as a whole. A Colombian judge who was trying
a drug case reported having received offers of
millions to absolve a trafficker charged with
several crimes, which he had systematically rejected.
One day a gift arrived at his office. Fearing
that it was a bomb, the security guards inspected
the package carefully, found no obvious
danger, and delivered it to the judge. The
judge opened it to find a simple album of family
photos: his daughter playing in the schoolyard,
his son entering the movies with some
friends, his wife shopping at the market… the
message was clear and he understood it. With
great regret and shame, he abandoned the case.
Factors that facilitate violence
There is a third kind of factor that is not the origin
of violence and thus cannot be causally incriminated,
but which facilitates violent behavior
or makes it more damaging and more
lethal, because it enables and stimulates such
behavior. Such factors are not in the social
structure, but in the individual.
Increase in firearms among the population
The world has witnessed a major increase in
the possession of light firearms: revolvers, pistols,
mini-machine guns. Firearms are not consumed
while they are used, unlike many other
individual or wartime products. Arms remain
and are reused and resold in secondary or tertiary
markets. Such arms produce more than
200,000 deaths every year in non-warlike events
(in wars they add another 300,000 deaths) 86.
More than one thousand companies produce
firearms in 98 countries around the world,
but 70% of the world market is supplied by
American and Russian companies. Firearms legislation
varies greatly between countries, from
the strictest prohibition in the United Kingdom
to the most lenient in the United States, where
bearing arms is a constitutional right. Although
the existence of firearms in a society is not necessarily
a direct efficient indicator of violence,
it is true that the existence of firearms in the
population facilitates lethal violence, because
interpersonal conflict, street fights, and unrequited
passion can end in blows or death, and
the substantive difference can result from
firearms, not the rage, hatred, or pain involved.
The same hate can produce either a bruised
face or a fatal victim. A study in 25 high-income
countries showed that homicides suffered
by women were significantly associated
with the availability of firearms.
The availability of arms among citizens
makes crime more violent because perpetrators
know they may meet armed resistance and
thus prepare themselves and act with greater
violence than what they will presumably find
with their victims. In societies without firearms,
delinquents can dominate victims with knives
or simply with physical force, because they
know that others will not have weapons to defend
themselves.
Latin America has the most firearms-related
homicides in the world. Table 10 shows the
world’s regions and how Latin America triples
the African homicide rate and quintuples the
North American and Central and Eastern European
rates, while it is a staggering 48 times
higher than in Western Europe.
In Latin America, an estimated 45 to 89 million
firearms are in the hands of civilians. Since
the vast majority are illegal, there are no precise
data, and the experts base these estimates
on a lower and upper threshold and believe
that there are 20 to 30 million in Brazil, 2.5 to
16.5 million in Mexico, 4 to 6 million in Argentina,
1.4 to 2 million in Chile, and 200,000
to 500,000 in Ecuador. Yet the lethalness of such
firearms varies greatly between countries. In
Ecuador, with fewer arms, more homicides are
committed (one for every 150-380 available
guns). In Chile, with many more arms, there is
one homicide for every 17-24 firearms 87,88.
Firearms are not directly responsible for violence,
but under conditions of social and individual
conflict they facilitate serious and mortal
aggression.
Alcohol consumption
Excessive alcohol consumption is associated
with violent behavior and casualties. The AC-
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1643
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
TIVA study showed an association between victimization
and consumption of more than five
drinks per occasion several times a week. This
association appeared in Rio de Janeiro, Caracas,
San José de Costa Rica, and Madrid 89. Likewise,
a study on violence in couples in Caracas
showed that a factor associated with serious
aggression between spouses was excessive alcohol
consumption by one or both 90.
Alcohol itself does not have to be a cause of
violence, since (like other drugs) it can produce
a sleepy and tranquilizing effect in some persons.
But alcohol consumption also releases
inhibitions and reduces barriers and repressions
that culture has planted in the individual.
Internalized norms and the Freudian superego
are undermined by drinking, and people can
become more expressive, more sincere, and
more aggressive.
Many fights and homicides result from a
banal but murderous combination of intoxication
and possession of weapons. Banal, because
many of the people (both men and women) we
interviewed in jail reported that they would
have avoided such acts if they had been able to
consider the consequences of their actions.
One of the most successful policies adopted in
cities with high rates of violence has thus been
a Dry Law decreed on what are considered critical
dates. In Cali, Colombia, this prohibition
was tested on days of important soccer matches,
and it reduced the homicides involving elated
(victorious) versus revolted (defeated) fans.
The inability to verbally express feelings
Finally, an individual circumstance that facilitates
the passage to a violent act is the extreme
difficulty some people experience in expressing
their inner feelings of rage or disgust in
words. Our hypothesis is that people who cannot
communicate their outrage in words express
it with acts such as slaps, kicks, fistfights,
or use of weapons. Words can be a substitute
for the violent act and in this sense are also violence,
but with far fewer consequences and
much less physical damage than violence itself.
Violence is always an act of communication, a
language perverted by feeling or ratified by
functional reason. Words can exorcise rage and
make others receive the aggression, but without
physically injuring them 91.
Some researchers of gender-related violence
contend that verbal and physical violence
must be equally condemned. From a
moral perspective, one can agree and consider
both incorrect (and even contend that words
can occasionally cause more damage than a
slap). Silence and indifference can also hurt
more than a slap. But in terms of the violence
that damages the body or kills, words are a great
help for conflict-resolution in the field of the
symbolic.
The issue is why some people turn their impulses
into acts while others do not. Why do
some people say they really want to punch you,
while others simply do it? There are two factors:
the moral controls over the passage to the
act, and the substitute realization of the desire.
We have observed that some people who fail to
construct the verbal substitution express their
feelings and desires in the passage to the act.
Psychoanalysis has worked through these substitute
mechanisms. Dreams are one example,
and Freud 92 thus wrote that healthy men
dream what perverse men do. One could paraphrase
Freud by saying that peaceful men say
what violent men do.
The inverse association some researchers
have found between more schooling and a decrease
in violent behavior or serious victimization
has various explanations. Education offers
more opportunities for employment, and individuals
incorporate more social norms. On the
other hand, years of study provide more verbal
skills and allow people to express feelings and
manage conflicts through negotiation and
agreement, i.e., with words rather than violence.
City, citizenship, and violence
These three levels of factors allow a multi-factorial
approach to urban violence in Latin
America whereby we can grasp both the particular
aspects of the social structure of Latin
Table 10
Estimates (low and high) of homicides involving firearms, by region.
Homicide rate per 100,000
Low threshold High threshold
Africa 3.83 5.90
Latin America and the Caribbean 12.80 15.47
North America 3.17 3.50
Middle East 0.52 1.8
Central and Eastern Europe 1.63 3.09
Western Europe 0.32 0.35
Southeast Asia 1.04 1.45
Asia and Pacific 0.51 0.54
World Total 2.85 3.96
Source: Assembled from Small Arms Survey 86.
Briceño-León R 1644
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
America and its cities and the dynamics of people’s
behavior. People’s action synthesizes the
social determinants and singularities of individual
freedom, which in the ultimate analysis
always makes each event of daily violence
unique and unrepeatable.
Each of these levels covers the next, encompasses
it, and contributes to its comprehension.
We should not be satisfied with overarching
explanations that claim to solve the problem
by attributing violence to neoliberalism.
The social process leading to violence in Latin
America is highly complex, and we must avoid
simplifications, because the goal is not to reduce
the phenomenon’s multifaceted nature,
but to formulate scientific hypotheses to transform
a complex and incomprehensible phenomenon
into one that is equally complex, but
somewhat more comprehensible.
We can thus better comprehend the social
transformations in violence in Latin America
and their public health consequences. The impact
on the population’s health relates not only
to the alarming figures on mortality presented
here or in the scarcely reliable statistics on
morbidity, with thousands of wounded and
maimed (indeed, we have not even taken issue
with such statistics). The impact also deals with
the millions of indirect victims who share the
pain of their relatives, neighbors, and friends,
and the population at large, living a terrified
life, in which the city and citizens’ rights lose
out 93,94.
The fear of becoming a victim of violence
produces different responses by members of
society. There is both an increase in private security
and a demand for greater public crackdown
on crime. Enhanced private security seeks
to reduce individual risk exposure and create
conditions to avoid being victimized, by limiting
going out in public or circulating in certain
parts of the city or at certain times, increasing
security in the home, privatizing public spaces,
and increasing private protection. The demand
for cracking down on violence (often referred
to as the “war on crime”) calls for greater police
presence on the streets and in public areas,
more aggressive treatment of delinquents by
the police, including support for extra-judicial
police actions (e.g., arrest without warrant and
even torture and summary executions), and
more severe punishment 65,95,96,97,98,99.
The city is transformed to adapt reactively
to insecurity. The divided city tends to reinforce
(intentionally or not) the segregation of
territories occupied by different social groups.
The middle class begins closing off streets in its
neighborhoods and hiring private guards. Then
the poor do the same with pedestrian walkways,
and since they can not afford private police,
they exercise vigilance on their own. The
street as an open air market is increasingly replaced
by shopping centers, with private boulevards
and security, because they have few doors
and private security systems. Shopping centers
began as a luxury for the middle class, but have
gradually become the favorite place for all social
strata. Not only have malls been created for
low-income strata, but all shopping centers
have become the favorite place for social outings
by the urban poor. One of the key reasons
is security.
Violence represents a loss for both the city
and citizenship (social rights represented by
the modern city as both illusion and reality).
Violence is a permanent threat to the most fundamental
right, i.e., the right to live. The city
used to be the place for a more protected life,
but the urban environment has become a
threat. The right to free circulation is curtailed
when streets are closed off and people abandon
many areas of the city because they fear
being victimized. People have the right to a secure
home, but families of all social strata feel
unsafe even in their own homes. The middle
class places fences, electrical wiring, and alarms,
and the poor build their houses without windows
to protect themselves. The house becomes
a refuge where people isolate and enclose themselves
as a defense. The city also represents the
right to work and recreation, and workers now
refuse to work overtime at night (which could
provide additional income) and stop going to
parties because they fear coming home late at
night. Violence also threatens the same rights,
because fear and distress induce many citizens
to support extra-judicial action that violates individual
human rights (including but not limited
to those of delinquents). People on the street
ask why society should defend the human
rights of outlaws. What about the human rights
of law-abiding citizens? All of this represents a
terrible loss of citizenship.
Achieving peace and overcoming violence
means making cities a place of freedom and
citizenship. The city should be the place of inclusion,
where different and unequal people
meet. The city is a space for negotiation between
diverse individuals and groups. A homogeneous
city is bland and boring. Large cities
(Rome, Istanbul, Paris, and New York) have always
been meeting places for diverse social
groups, ideologies, and religions. Although on
a more modest scale, the world’s cities of today
tend to repeat a pattern that increases with
globalization 100,101. Cities are places for agree-
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1645
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
ment and cohabitation among the unequal. Although
we have no reason to claim equal living
conditions, exclusion need not exist either.
Cities are the place to include everyone in equal
citizens’ rights, with the possibility of achieving
a decent and healthy urban life, without either
ostentatious wealth or dire poverty. Cities
are the place to generate agreements requiring
progress in social life and social rights, where
the unequal can meet (whether on a friendly or
conflictive basis) and mutually advance the
construction of an urban space, since cities are
the privileged place for individual rights and
collective cohabitation.
There is an old German saying, “Stad Luft
mach frei”: the air of the city makes you free.
The expression originated in medieval times,
when a serf who succeeded in proving that he
had lived in the city for a year and a day was
entitled to his freedom and the right to remain
in the metropolis. Latin American cities will
only reclaim the widespread dream of freedom
they represented in the 20th century if they
succeed in overcoming the epidemic of violence.
Thus, cities need social transformations
that grant more freedom rather than restricting
it. Healthy cities that aspire to public health
must also be safe cities.
Resumen
La violencia interpersonal se ha convertido en uno de
los principales problemas de salud pública de las ciudades
de América Latina. El artículo presenta una interpretación
sociológica de la violencia en tres niveles:
(a) macro-sociales – la desigualdad social debida al
incremento de la riqueza y la pobreza; la paradoja del
mayor nivel educativo de las personas, pero las menores
oportunidades de empleo, el incremento de las
expectativas y de la imposibilidad de satisfacerlas; los
cambios en la familia y la pérdida de importancia de
la religión en la vida cotidiana de las personas; (b)
meso-sociales – el incremento de la densidad en las
zonas pobres y la segregación urbana, la cultura de la
masculinidad y los cambios en el mercado local de la
droga; (c) micro-sociales – el incremento de las armas
de fuego, el consumo de alcohol y las dificultades de
expresión verbal de los sentimientos por las personas.
El artículo concluye con un análisis sobre cómo la violencia
está llevando no sólo a la pérdida de las ciudades,
sino a la ciudadanía en América Latina.
Violencia; Sociología; Salud Urbana; Ciudadanía
References
1. World Health Organization. World report on violence
and health. Geneva: World Health Organization;
2004.
2. United Nations. World population prospects. The
1999 revision. New York: Population Division,
United Nations; 2001.
3. United Nations. The state of world population.
New York: United Nations; 2004.
4. Briceño-León R. La nueva violencia urbana en
América Latina. In: Briceño-León R, editor. Violencia,
sociedad y justicia en América Latina.
Buenos Aires: Consejo Latinoamericano de Ciencias
Sociales; 2002. p. 13-26.
5. Cruz JM. Violencia, democracia y cultura política.
Nueva Sociedad 2000; 167:132-46.
6. Franco S. El quinto: no matar. Contextos explicativos
de la violencia en Colombia. Santafé de Bogotá:
TM Editores; 1999.
7. Camacho A, Guzmán A. Ciudad y violencia. Revista
Foro 1990; (12):22-32.
8. Pinheiro PS. Violência, crime e sistemas policiais
em países de novas democracias. Tempo Social –
Revista de Sociologia da USP 1997; 9:43-52.
9. Instituto de Estudos da Religião/ Banco Interamericano
de Desenvolvimento. Magnitude, custos
econômico e políticas de controle da violência.
Washington DC: Banco Interamericano de
Desenvolvimento; 1997.
Briceño-León R 1646
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
10. Londoño JL, Guerero R. Violencia en América
Latina. Epidemiología y costos. Washington DC:
Banco Interamericano de Desarrollo; 1999. (Documento
de Trabajo R-375).
11. Souza ER. Homicídios no Brasil: o grande vilão de
saúde pública na década de 80. Cad Saúde Pública
1994, 10 Suppl 1:45-60.
12. Reichenheim ME, Werneck GL. Anos potenciais
de vida perdidos no Rio de Janeiro, 1990. As mortes
violentas em questão. Cad Saúde Pública
1994; 10 Suppl 1:188-98.
13. Guerrero R. Epidemiología de la violencia: el caso
Cali, Colombia. In: Banco Interamericano de Desarrollo,
editor. Hacia un enfoque integrado del
desarrollo: ética, violencia y seguridad ciudadana.
Washington DC: Banco Interamericano de
Desarrollo; 1996.
14. Baker J, Schuler N. Analyzing urban poverty: a
summary of methods and approaches. Washington
DC: World Bank; 2004. (World Bank Policy Research
Working Paper 3399).
15. Elias N. El proceso de la civilización, investigaciones
sociogenéticas y psicogenéticas, México
DC: Fondo de Cultura Económica; 1987
16. Remy J, Voye L. Ville, ordre et violence. Espace et liberté.
Paris: Presses Universitaires de France; 1981.
17. Del Olmo R. Ciudades duras y violencia urbana.
Nueva Sociedad 2000; 167:74-86.
18. Rotker S. Ciudades escritas por la violencia. (A
modo de introducción) In: Rotker S, editor. Ciudadanías
del miedo. Caracas: Nueva Sociedad;
2000. p. 7-22.
19. Buvinic M, Morrison A, Shifter M. La violencia en
América Latina y el Caribe. Un marco de referencia
para la acción. Washington DC: Banco Interamericano
de Desarrollo; 2000.
20. Pérez-Perdomo P. Contar los cuerpos, lamer las
heridas: la tarea de cuantificar la violencia delictiva
en Morir en Caracas. Violencia y Ciudadanía
en Venezuela. Caracas: Facultad de Ciencias Jurídicas
y Políticas, Universidad Central de Venezuela;
2002.
21. Fundación Mexicana de la Salud. Tendencias y
causas de delito violento en el Distrito Federal.
Informe final. México DF: Fundación Mexicana
de la Salud; 1999.
22. Lozano R. La violencia en ciudad de México: Análisis
de la magnitud y su repercusión económica.
México DC: Centro de Economía y Salud, Fundación
Mexicana para la Salud; 1997.
23. Lozano R, Hijar M, Zurita B, Hernández P, Avila L,
Bravo ML, et al. Capital lesionada: violencia en
ciudad de México. In: Londoño JL, Gaviria A,
Guerrero R, editores. Asalto al desarrollo: violencia
en América Latina. Washington DC: Banco Interamericano
del Desarrollo; 2000. p. 205-32
24. Hernádez-Bringas H. Las muertes violentas en
México. México DF: Universidad México Americana
del Norte; 1989.
25. Rubio M. Crimen e impunidad. Precisiones sobre
la violencia. Bogotá: TM Editores; 1999.
26. Martín-Barbero J. La ciudad entre medios y miedos.
In: Rotker S, editor. Ciudadanías del Miedo.
Caracas: Nueva Sociedad; 2000. p. 29-35.
27. Pegoraro J. Violencia delictiva, inseguridad urbana.
La construcción social de la inseguridad
ciudadana. Nueva Sociedad 2000; 167:114-132.
28. Ferraro KF. Fear of crime. Interpreting victimization
risk. New York: State University of New York;
1995.
29. Cisneros A, Zubillaga V. La violencia desde la perspectiva
de la víctima: La construcción social del
miedo. Espacio Abierto 1997; 6:71-97
30. Navarro JC, Pérez-Perdomo R. Seguridad personal
un asalto al tema. Caracas: Ediciones Instituto
de Estudios Superiores de Administración; 1991.
31. Soarez LE. O inominável, nosso medo. In: Violência
e política no Rio de Janeiro. Rio de Janeiro: Relume
Dumará/ Instituto de Estudos da Religião;
1996. p. 59-64.
32. Bandura A. Aggression: a social learning theory.
Englewood: Prentice-Hall; 1973
33. Bandura A. Social foundations of thought and action.
A social cognitive theory. New Jersey: Prentice
Hall; 1986.
34. Moser C, Shrader E. Crimen, violencia y pobreza
urbana en América Latina: hacia un marco de referencia
integrado. Washington DC: World Bank;
1998.
35. Cerqueira D, Lobão W. Determinantes de Criminalidades:
Arcabouços teóricos e Resultados Empíricos.
Dados 2004; 47:233-69.
36. Carranza E, editor. Delito y seguridad de los habitantes.
México DC: Siglo XXI Editores; 1997.
37. Reiss Jr. JA, Roth A. Understanding and preventing
violence. Washington DC: National Academy
Press; 1993.
38. Arendt H. On violence. San Diego: Harcourt Brace
& Company; 1970.
39. Wievorka M. Violence en France. Paris: Seuil; 1999.
40. Popper K. La lógica de la investigación científica.
Madrid: Editorial Tecnos; 1977.
41. Boudon R. La logique du social. Paris: Les Èditions
Hachette Litérature; 1979.
42. Briceño-León R. Buscando explicaciones a la violencia.
Espacio Abierto 1997; 6:45-69.
43. Minayo MC. Relaciones entre procesos sociales,
violencia y calidad de vida. Ciência e Saúde Coletiva
2005; 1:69-78.
44. Ferrel J. Cultural criminology. Annu Rev Sociol
1999; 25:395-418.
45. Popper KR. Conjectures and refutations. The
growth of scientific knowledge. London: Routledge/
Kegan Paul; 1972.
46. Weber M. Economía y Sociedad. México DC: Fondo
de Cultura Económica; 1977.
47. Ferranti D, Perry G, Ferreira FHG, Walton M. Inequality
in Latin America & Caribbean. Breaking
with History? Washington DC: World Bank; 2004.
48. Gasparini L. Income inequality in Latin America
and the Caribbean: evidence from household
surveys. La Plata: Centro de Estudios Distributivos,
Laborales y Sociales; 2003. (Documento de
Trabajo 2).
49. Londoño JL, Szekely M. Persistent poverty and
excess inequality: Latin America, 1970-1995. Washington
DC: Inter American Development Bank;
1997. (Working paper 357).
50. Fassin D. Exclusion, underclass, marginalidad.
Figures contemporaines de la pauvreté urbaine
en France, aux Etats-Unis et en Amerique Latine.
Rev Fr Sociol 1996; 37:37-76
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1647
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
51. Comisión Económica para América Latina y el
Caribe. Panorama social de América Latina. Santiago
de Chile: Comisión Económica para América
Latina y el Caribe; 2004.
52. Spinelli H, Alazraqui M, Macías G. Muertes violentas
en la Ciudad Autónoma de Buenos Aires.
Una mirada desde el sector salud. Buenos Aires:
Organización Panamericana de la Salud; 2005.
53. Internacional Labor Organization. Información
sobre el empleo de los Jóvenes. Geneve: Internacional
Labor Organization; 2004.
54. Organización Panamericana de la Salud. Taller
sobre la violencia de los adolescentes y las pandillas
(“Maras”) Juveniles. San Salvador: Organización
Panamericana de la Salud, Organización
Mundial de la Salud; 1997.
55. Yunes J, Rajs D. Tendencia de la mortalidad por
causas violentas en la población general y entre
los adolescentes y jóvenes de la región de las
Américas. Cad Saúde Pública 1994; 10 Suppl 1:88-
125.
56. Londoño JL, Gaviria A, Guerrero R. Asalto al desarrollo.
Violencia en América Latina. Washington
DC: Banco Interamericano de Desarrollo; 2000.
57. Germani G. Política y sociedad en una época de
transición. Buenos Aires: Paidós; 1961.
58. Merton RK. Teoría y estructura social. México DC:
Fondo de Cultura Económica; 1965.
59. Cruz JM, Portillo-Peña N. Solidaridad y violencia
en las pandillas del Gran Salvador. Más allá de la
Vida Loca. Colección Debate. v. 9. El Salvador:
UCA Editores; 1998.
60. Zaluar A. Violence, argent facile et justice au Brésil:
1980-1995. Revue Internationale des Sciences Sociales
2001; 169:477-84.
61. Collison M. In search of the high life: drugs, crime,
masculinities and consumption. Br J Criminol
1996; 36:429.
62. Dubet F. La galere: jeunes en survie. Paris: Points
Actuels; 1987.
63. Levi-Strauss C. Las estructuras elementales del
parentesco. Buenos Aires: Paidós; 1964.
64. Lacan J. Ecrits. Paris: Seuil; 1976.
65. Briceño-León R, Camardiel A, Ávila O. El derecho
a matar en América Latina. In: Briceño-León R,
editor. Violencia, sociedad y justicia en América
Latina. Buenos Aires: Consejo Latinoamericano
de Ciencias Sociales; 2002. p. 383-40.
66. Veckmans R, Venegas R. Marginalidad, incorporación
e integración. Santiago de Chile: Centro
para el Desarrollo Económico y Social de América
Latina; 1966.
67. Desarrollo Económico y Social de América Latina.
La marginalidad urbana: origen, proceso y
modo. Buenos Aires: Editorial Troquel; 1970.
68. Nun J. Superpoblación relativa, ejercito industrial
de reserva y masa marginal. Revista Latinoamericana
de Sociología 1969; 2:178-236.
69. Murmis M. Tipos de marginalidad y posición en
el proceso productivo. Revista Latinoamericana
de Sociología 1969; (2):413-21.
70. Santos M. Geografía y economía urbana en los
países subdesarrollados. Barcelona: Oikos-Tau;
1973
71. Quijano A. El proceso de urbanización en América
Latina. Santiago de Chile: Comisión Económica
para América Latina y el Caribe; 1996.
72. Quijano A. Dependencia, urbanización y cambio
social en Latinoamérica. Lima: Mosca Azul; 1977.
73. Bolivar T. Urbanizadores, constructores y ciudadanos
Rev Mex Sociol 1995; (1):71-87.
74. Bolivar T, Guerrero YM, Rosas I, Ontiveros T, Freitas
J. Densificación y vivienda en los barrios caraqueños.
Contribución a la determinación de
problemas y soluciones. Caracas: Mindur-Conavi;
1994.
75. Bell D. Los portuarios controlados por el racket:
la telaraña de la economía y la política en el fin
de las ideologías. Madrid: Tecnos; 1964.
76. Messerschmidt J. Masculinities and crime. Lanham:
Rowman and Littlefield Publishers; 1993.
77. Jefferson T. Introduction: masculinities, social relations
and crime. Br J Criminol 1996; 36:337-47.
78. Spierenburg P. Masculinity, violence and honor,
men and violence. Ohio: The Ohio State University;
1998.
79. Zubillaga V, Briceño-León R. Exclusión, masculinidad
y respeto: algunas claves para entender
la violencia entre adolescentes en barrios. Nueva
Sociedad 2001; 173:34-78.
80. Vidal D. Le respect: catégorie du social, catégorie
du politique dans une favela de Recife. Cultures
et Conflits 1999; (35):95-124.
81. Nisbett RE, Cohen D. Culture of honor: the psychology
of violence in the South. Oxford: Westview
Press; 1997.
82. Castillo A. Menores transgresores: en búsqueda
de adaptación social. Caracas: Universidad Central
de Venezuela; 1997.
83. Marquez P. The street is my home. Stanford: Stanford
University Press; 1999.
84. Palmer DS. Drogas y seguridad en el Mercosur.
Buenos Aires: Instituto Nacional de Administración
Pública; 1998.
85. Velez LF, Espitia VE, Banguero H, Méndez F, Muñoz
E, Rotawinsky W, et al. Violencia en Colombia
un análisis exploratorio del caso de la ciudad de
Cali. Cali: Universidad de Valle/World Bank; 1999.
86. Small Arms Survey. Small Arms Survey 2004:
rights at risk. Oxford: Oxford University; 2004.
87. Small Arms Survey. Small Arms Survey 2003: development
denied. Oxford: Oxford University; 2003.
88. Peres-Tourinho MF. Violência por armas de fogo
no Brasil. São Paulo: Núcleo de Estudos da Violência,
Universidade de São Paulo; 2004.
89. Cruz JM. La victimización por violencia urbana:
niveles y factores asociados en ciudades seleccionadas
de América Latina y España. Washington
DC: Organización Panamericana de la Salud;
1999.
90. Briceño-León R, Camardiel A, Avila O, De Armas
E. ¿Quiénes son las víctimas de la violencia en
Caracas? Un análisis social del riesgo de la violencia
no-fatal. Trib Invest 1998; 5:5-19.
91. Safouan M. La parole ou la Mort. Comment une
societe humaine est-elle possible? Paris: Éditions
du Seuil; 1993.
92. Freud S. La interpretación de los sueños. Obras
completas. Tomo I. Madrid: Biblioteca Nueva;
1973.
93. Minayo M.C, Souza ER. Violência para todos. Cad
Saúde Pública 1993; 9:65-78.
Briceño-León R 1648
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
94. Sen A. Health in development. Bull World Health
Organ 1999; 77:619-23.
95. Briceño-León R, Piquet L, Cruz JM. O apoio dos
cidadãos à ação extrajudicial da policia no Brasil,
em El Salvador e na Venezuela”. In: Pandolfi DC,
Carvalho JM, Carneiro LP, Grynzpan M, organizadores.
Cidadania, justiça e violência. Rio de Janeiro:
Editora da Fundação Getulio Vargas; 1999.
p. 117-27.
96. Cano I. The use of lethal force by police in Rio de
Janeiro. Rio de Janeiro: Instituto de Estudos da
Religião; 1997.
97. Cruz-Neto O, Minayo MCS. Extermínio: violentação
e banalização da vida. Cad Saúde Pública
1994; 10 Suppl 1:199-212.
98. Pinheiro PS. El Crimen, la violencia y la pobreza
Urbana. In: Seminar on Urban Poverty. Rio de
Janeiro: Asociación Latinoamericana de Organizaciones
de Promoción/World Bank; 1998.
99. Mateus-Guerrero S. Limpiezas sociales. La guerra
a la indigencia. Santa Fe de Bogotá: Ediciones
Tiempo de Hoy; 1995
100. Briceño-León R, Zubillaga V. Violence and globalization
in Latin America. Current Sociology 2002;
50:19-37.
101. Tavares-dos-Santos T. The worldization of violence
and injustice. Current Sociology 2002; 50:
123-134.
102. Pan American Health Organization. Repercusiones
de la violencia en la salud de las poblaciones
Americanas. Washington, DC: Pan Americana
Health Organization; 2003.
103. Population Reference Bureau. Population data
sheet. New York; 2004.
104. Instituto APOYO. La violencia en el Perú: dimensionamiento
y políticas de control. Informe preliminar.
Lima: Red de Centros de Investigación,
Banco Interamericano de Desarrollo; 1997.
105. Lederman D. Crime in Argentina: a preliminary
assessment. Washington DC: World Bank; 1999.
106. Centro Latinoamericano y Caribeño de Demografía,
División de Población. Transformaciones
demográficas en América Latina y el Caribe y
consecuencias para las políticas públicas, en panorama
social de América Latina. Santiago de
Chile: Comisión Económica para América Latina
y el Caribe; 2004.
Submitted on 30/May/2005
Approved on 03/Jun/2005
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1649
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
Juan Díez-Nicolás
Debate on the paper by Roberto Briceño-León
Debate sobre el artículo de Roberto Briceño-León
Universidad Complutense
Madrid, Madrid, España.
Análisis Sociológicos
Económicos Políticos,
Madrid, España.
[email protected]
The article analyzes one of the most important
and fastest-growing problems in many Latin
American countries: urban violence, a problem
that (as Briceño-León indicates) results directly
or indirectly from rapid urban population
growth rates, producing huge urban agglomerations,
and from increasing social and economic
inequalities between the “haves” and
the “have-nots”. The article’s main focus is thus
of great theoretical importance and of even
greater political significance.
The author’s classification of Latin American
countries in four categories according to
the degree of violence (based on the number of
homicides per 100,000 inhabitants/per year) is
especially pertinent and reinforces his research
strategy, taking countries as his units of analysis,
a decision he maintains throughout the article,
even when analyzing feelings of insecurity
(since they are country averages obtained
through surveys in which individuals are the
primary units of analysis).
The theoretical framework adopted by Briceño-
León to interpret the data and develop
his explanatory argument analyzes three types
of factors affecting urban violence: factors that
facilitate violence, factors that promote or stimulate
violence, and factors that originate (or
produce) violence. This is the article’s real core
and thus provides the author’s main theoretical
contribution to sociological theory. Among the
“situational” factors he mentions as “originating”
urban violence are poverty, and especially
Latin America’s increasing social and economic
inequalities. In this sense, some changes
could be made in Table 6 to facilitate its interpretation.
The fact that two groups of countries
are labeled “Group 1: Low violence” introduces
some confusion. This could be avoided by replacing
the chart with two four-fold tables, one
cross-tabulating the degree of violence with
the poverty rate, and the other cross-tabulating
the degree of violence by the urban population
percentage. A second solution would provide
the correlation coefficients between the homicide
rate and the poverty rate and urban population
percentage, respectively. Other factors
he mentions as originating violence are the contrast
between populations with more schooling
and greater unemployment. This argument is
well presented, and the author might benefit
from the four-fold classification developed by
Huaco 1 based on society’s demand for skills
versus their availability. Many current societies
(probably including many Latin American societies)
show a supply of skilled labor that greatly
exceeds the social demands for it (mainly because
of a deficient organization of society that
is only “functional” for the power elites’ interests).
Regarding the contrast between growing
aspirations and limited possibilities to satisfy
them, Briseño-León might also benefit from
analyzing the difference between the “objective
level of living” and the “subjective living standard”
2, but the author’s argument is well presented
nevertheless. The importance of the declining
influence of family and religion is also
well explained and underlines the importance
of the “cultural factor” in the Weberian sense,
thus aligning with a growing trend towards research
on values (e.g., World Values Survey, European
Values Study) that emphasizes the explanatory
importance of cultural values for human
behavior and institutions.
Among the factors that promote or stimulate
violence, Briceño-León mentions segregation
and urban density (Durkheim’s social density?),
the culture of masculinity (most violence
is practiced by men, not women), and the impunity
of the urban drug traffic. It might be argued
that the drug traffic’s extraordinary importance
is usually underestimated in sociological
research, although it has deeply changed
the lifestyle and social organization of current
cities, especially among the younger generations.
Among the factors that facilitate violence,
Briceño-León mentions the increased
availability of firearms, growing alcohol consumption,
and other individual characteristics.
The article is certainly very provocative and
(as mentioned above) makes a significant contribution
both to theory and political action
programs. In a sense it links with an important
sociological tradition, that of social problems
research (and particularly the old Chicago school
of Park and Burgess), but also with the Weberian
tradition of emphasizing the key significance
of the “cultural factor”. “Culture counts”
could be the motto for this interesting and
provocative article.
1. Huaco GA. The functionalist theory of stratification:
two decades of controversy. Inquirí 1966; 9:215-40.
2. Díez-Nicolás J. Social position and orientation towards
domestic issues in Spain. v. 3. Amsterdam:
Polls; 1968.
Briceño-León R 1650
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
Social Science Academy
of Nigeria, Abuja, Nigeria.
[email protected]
Layi Erinosho The main challenge facing humankind today is
violence or terror, to use the cliché that has followed
the September 11 attacks in the United
States. Violence or terror in whatever forms are
transcendent in all human societies, whether
in the North or South, East or West. That is why
this paper’s subject matter is extremely important,
timely, and worthwhile.
Briceño-León explores the interplay between
urban violence and public health in Latin
America from a sociological standpoint. The
paper addresses two hypotheses. The first is
that violence can be closely linked to urbanization
and its underlying social processes, while
the second is about the public health consequences
of urban violence.
The paper begins by suggesting that the
rapid pace of urbanization in Latin America
has been accompanied by violence. In other
words, Briceño-León seems to suggest that
there is a positive correlation between urbanization
and urban violence. As the Latin American
cities have grown in size, so has the level of
violence. An attempt is made to support the assertion
with data. Further attempt is made in
the paper to invoke a paradigm for the phenomenon
of urban violence in Latin America.
The rest of the paper is about the classification
of factors accounting for urban violence, such
as poverty, unemployment, and high levels of
illiteracy, family breakdown, and the culture of
masculinity, drug/abuse trafficking, the decline
of religion, and access to firearms in urban
centers. Finally, the paper explores the implications
of urban violence for public health.
The paper’s subject matter reminds one of
Emile Durkheim’s classic study on suicide 1.
Durkheim showed how suicide could be linked
to and/or explained by the pervading economic
climate in society. While Durkheim successfully
made this case, the same cannot be said
of the paper under debate here. The reasons for
the failure to produce such an explanation are
not farfetched and can be found in the following
contexts.
First, there is a lack of clarity on the independent
and dependent variables in the paper.
Violence is a broad concept which includes
suicide, homicides, armed conflict(s), assault,
violent strikes and demonstrations resulting in
loss of lives and property, guerrilla warfare, etc.
Violence can also be collective (e.g., violence
from political unrest) or domestic (e.g., violence
within family units like wife-beating, physical
abuse of children, etc.). In any case, Briceño-
León fails to define urban violence in the paper.
Indeed, the author seems to suggest that
urban violence is coterminous with homicides.
The need to define urban violence and its scope
and to specify which of them is being addressed
in the context of the paper is desirable. The paper
is therefore not about urban violence in
Latin America but about one of its forms,
namely, homicide. I hasten to add that the title
should reflect the author’s concern.
The second observation is about the weak
data on the link between urbanization and its
underlying social processes. For example, the
main cause of homicide in Columbia, currently
fraught by guerilla war, is political conflict
rather than urbanization. Although Briceño-
León is quick to make this clarification, this example
undermines the guiding hypothesis about
the positive correlation between urban violence
(defined as homicide) and urbanization.
The third observation relates to the idyllic
picture of urban centers painted by Briceño-
León. According to him, urban centers are supposedly
a haven for rights, safety, and good
manners. This is far from the characterization
of urban centers as compared to rural settings
in sociological classics and/or in what is generally
known about cities. Once more Emile
Durkheim and Ferdinand Toennies (to name
two) argued that urban centers are more socially
differentiated than rural, while the Durkheim
noted that they (urban centers) are more susceptible
to anomie and alienation than the rural.
There is no doubt that urban centers are developed,
but they are also not necessarily peaceful
environments as argued in various sections
of the paper, most especially in the penultimate
paragraph.
The fourth remark is about the tables used
to justify the paper’s main hypothesis. The tables
in their present form do not provide definitive
evidence, because they are descriptive. A
rigorous analysis using multivariate statistical
methods might indicate the predictors of homicide
(see Table 5). Moreover, there is no evidence
in the paper to support the following assertion:
“In Latin America, the distance between
the poor and the rich is the greatest in the world.
On other continents, such as Africa, there is more
poverty, but there is also not as much wealth. In
Europe, on the other hand, there is more wealth
but not as much poverty” (see section entitled
The City has More Wealth and More Poverty).
The fifth and final remark is about the theory
invoked to explain urban violence. It is difficult
to discern the differences among the factors
that are grouped under the macro, meso,
and micro. Table 11 below shows an attempt to
produce a matrix of factors.
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1651
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
It is self-evident from the matrix that some
of the macro factors are not sufficiently overarching
and could easily be placed in the meso
and micro. A macro factor should be an overarching
one that stands out and is therefore fundamental.
The macro provides the explanation
because it transcends the society. Thus, there
should not be more than one macro. It would
have been urbanization and its underlying social
processes if Briceño-León had settled for
one, which according to the guiding hypothesis
provides the clue to the high homicide rates
in Latin American cities. However, the matrix
does not include urbanization, which is proposed
in the paper as the macro factor.
Finally, the paper’s real challenge is to define
urban violence and provide convincing data
on the interplay between urbanization and
whatever Briseño-León defines as violence.
1. Durkheim E. Le suicide. Étude de sociologie. 2nd
Ed. Paris: PUF; 1967.
Table 11
Proposed matrix of macro, meso, and micro factors for violence.
S/N Macro Meso Micro
1. More wealth than poverty in the city Segregation and urban poverty More firearms in the population
2. More education, but less employment Culture of masculinity Alcohol consumption
3. More aspirations, but less capacity to meet them Local drug market and impunity Inability to verbally express feelings
4. Less social control by the family
5. Less force of religion
De La Salle University,
Manila, Philippines.
[email protected]
Pilar Ramos-
Jimenez
When I began reading Dr. Briceno-Leon’s paper
on urban violence in Latin America, I thought
he was also describing the Philippine urban
condition. Had it not been for its geographical
location, I would have added the Philippines to
the list of Latin American countries with medium
to high rates of urban violence. The Philippines
have numerous similarities with a number
of Latin American countries, due mainly to
more than three centuries of Spanish colonial
rule. It is the only Catholic country in Asia.
Violence in Metropolitan Manila where I reside
is regularly depicted by the media, particularly
in the daily newspapers, radio, and TV
evening news. Often reported are cases of murder,
manslaughter, rape, and aggravated assault
committed mainly by male adolescents
and young adults who are reportedly poor, with
little education, jobless or underemployed, and
under the influence of illicit drugs. The young
men tend to hail from slum and squatter communities,
a segment comprising over a third of
Manila’s population. These cases of violence
are reported to cover about a third of all crime
or the crimes reported to the police in the past
year. The constant presence of security guards
in virtually every establishment and location
reflect the pervasive fear among residents, not
only in Metropolitan Manila but also in large
and small cities throughout the country.
The public at large seeks plausible explanations
from the media, the police, and other social
institutions regarding the escalating urban
violence. The various sources of information
often attribute this situation to weak leadership
by government, an inadequate or poor security
system, corrupt politicians and the police,
terrorists, drug pushers, mass poverty, and
declining moral standards in society.
While many of the above factors may play
roles in the rapidly growing violence in Philippine
cities, there has been no comprehensive
explanation for this situation. Briceño-Leon
and his research institution, the Social Sciences
Laboratory (LACSO), offer a sociological
framework that is useful for explaining urban
violence, not only in Latin America but also in
other regions of the world. This framework
considers the contributions of other explanatory
models on violence, but it appears to be
more comprehensive because it considers the
situation in society and the cultural dimension
which can affect individual decisions and participation
in violence. It does not “pretend to
be exhaustive”, nor is it a “model for universal
explanations”, and it recognizes urban violence
as a complex phenomenon. Thus it posits “three
distinct levels of explanation”: the structural or
Briceño-León R 1652
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
macro-social, the meso-social, and the microsocial.
The specific factors within each level of
the proposed framework appear appropriate to
demonstrate their roles in fomenting violence
in many cities of the developing world.
The structural factors proposed by Briceño-
Leon are often utilized by social theorists to explain
various development issues. However,
the evidence and arguments he presents for
the widening gap between the small rich and
large poor populations in a social setting that
showcases wealth and power, the reduction of
work opportunities among the educated youth,
youth’s rising expectations, and the inability of
cities to meet their aspirations, as well as declining
social control by the family and religion
are convincing to demonstrate their relevant
association with violence in cities. It would be
worthwhile to include the declining ability of the
government to control urban violence because
of burgeoning population, corruption, and inadequate
resources to meet basic needs, including
the health and security of city-dwellers.
On the other hand, the author contends that
meso-social factors have “less structural roots”
and thus instigate violence, because they are
easier to modify than macro-social factors.
While it is understandable that segregation of a
large segment of the urban population in slum
and squatter colonies is a factor that foments
violence because these settings harbor the
poor and dysfunctional, I find the masculinity
factor at this level a bit underplayed. As a gender
issue, the patriarchy has structural dimensions
and has persisted longer. Men across
generations and cultures have instigated civil
and global wars, so that to box in masculinity
as one of many factors that foment urban violence
is rather limiting. Men are the ones who
are heavily involved in the entry, marketing,
and use of drugs, firearms, and alcohol as well
the use of force to persuade others to engage in
unlawful behavior. They are generally less verbally
expressive of their feelings and are more
likely than women to act aggressively. They also
play a key role in religion, the family, and
government because they dominate these social
institutions. I believe that this gender perspective
should cut across the three levels of
the sociological framework.
Briceño-Leon’s description of the resilience
of segmented populations to respond to conditions
of insecurity is a phenomenon that is taking
place in many cities the world over. The
presence of security guards in wealthy and
middle-class housing areas as well as neighborhood
vigilantes to maintain peace and order
in poor urban communities, the use of various
gadgets to ward off intruders in homes
and establishments, the use of malls as “excursion”
sites to provide a temporary breath of freedom
and security, and the avoidance of neighborhoods
and occasions that hasten the victimization
of law-abiding citizens are the present-
day city-dwellers’ coping strategies to survive
in the cities. Will the city as a former haven
of “freedom and citizenship” ever be restored
in this complex world?
Faculdade de Saúde
Pública, Universidade
de São Paulo,
São Paulo, Brasil.
[email protected]
Luciana
Scarlazzari Costa
The article follows a line of reasoning that is
pleasant to read from beginning to end and
demonstrates the theme’s scope.
The proposal is to explain urban violence in
Latin America. The author uses a sociological
theoretical model with three levels of explanation:
factors that originate violence, structural
factors or those referring to long-lasting social
processes that allow violent behaviors to be
created; factors that foment violence, of cultural
or situational factors with an immediate effect
on behavior; and finally factors that facilitate
violence, with a more individual nature
and which facilitate the violent act.
As the basis for the theoretical model, Briceño-
León uses both theoretical justifications
and data from surveys. Since he presents some
quantitative data, from the scientific point of
view it would be interesting if some statistical
tests were performed to demonstrate the relations
that are discussed, such as the relationship
between the urban population percentages
and the homicide rates.
All use of indicators is known to involve
some limitations, since they are summary measures,
it is possible to conceive of studies with
a multilevel methodology, which take into account
the social, economic, cultural, and other
indicators pertaining to the various levels or
dimensions in order to test the hypotheses
raised in the present article, as well as to verify
the factors with the greatest weight or which
make the greatest contribution to urban violence
(expressed here as the homicide rate) in
order to orient actions.
According to the article, individual factors
or those that facilitate violence involve greater
flexibility of action, but when acting on individuals
there are forces representing the contextual
factors of the society to which they be-
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1653
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
long. The question then is how much the violence
rates are modified with measures like the
public disarmament campaign now under way
in Brazil.
An ecological-type study by Gawryszewski
& Costa 1 in the city of São Paulo used a multiple
logistic regression analysis to show that the
independent factors related to the homicide
rate were income (with a negative correlation)
and the proportion of adolescents from 15 to
17 years of age not attending school (with a
positive correlation). These findings illustrate
what the current author states, i.e., that violence
is a problem that occurs with youth, and according
to Gawryszewski & Costa 1, with young
people who are not in school!
Organized society needs to invest in citizens
with social programs, education, medical
care, and vocational training, but it also needs
to look further. How can society provide education
when the citizens’ family structure is
jeopardized, when citizens live in segregated
urban areas with a weak state presence, lack of
public lighting, recreation, and culture, and to
top it off a limited job supply.
In order to modify society’s structural factors,
it is necessary to study their contribution
to urban violence, or else one runs the risk of
merely implementing isolated measures. The
main challenge is identifying where and how to
act to attempt to modify this panorama of urban
violence in Latin America.
1. Gawryszewski VP, Costa LS. Homicídios e desigualdades
sociais no Município de São Paulo. Rev
Saúde Pública 2005; 39:191-7.
Escola Nacional de Saúde
Pública Sergio Arouca,
Fundação Oswaldo Cruz,
Rio de Janeiro, Brasil.
[email protected]
Maria Cecília de
Souza Minayo
The dynamics of social violence
in Latin America
The paper by Roberto Briceño-León is an undeniably
profound, mature, and sociologically
and socially committed effort by the author to
the process of transformation in Latin America,
highlighting how the dynamics of violence
corrupt the region’s entrails. His erudite, methodical,
and complex reflection on the convergence
of factors that contribute to generate
this phenomenon is certainly the most complete
document of its kind produced in recent
years, dealing with the issue in an explicative,
contemporary, and comprehensive way. His paper
especially reveals an experienced sociologist,
with vast capacity to read the international
and Latin American classics, in dialogue with
contemporary data on the region’s demographic,
epidemiological, educational, socioeconomic,
and cultural transformations. Thus, without
a shadow of a doubt, Briceño-León’s paper is a
milestone in the reflection on violence, due
both to its original proposal and the wealth of
theoretical and informational material provided
to readers as the basis for his arguments.
In my role as discussant, I will attempt to
raise a few questions to complement the reflection,
in the understanding that a study of such
breadth will leave out a few questions that need
to be approached. In this debate I will mention
three aspects: I will approach conceptual aspects;
I will delve into some necessary distinctions
in the Brazilian case; and I will contextualize
the issue of contemporary crime (a term
used throughout the paper as an implicit quasi-
synonym for violence) in the world and Brazil,
creating an ethos differentiated from the traditional
forms of social transgression.
Conceptual aspect
The first point I wish to highlight is that Briceño-
León does not refer exactly to violence as
a multifaceted phenomenon ranging all the
way from cruel and fatal aspects (e.g., homicides)
to those that are taken for granted (like
traffic violence) and covert, subtle forms (as in
the case of intra-family violence, moral harassment,
and others). Rather, he uses the phenomenon
that is most visible and easiest to
count, to treat statistically, and to compare,
namely homicide. As a backdrop, his concept
of violence runs up against the notion of crime,
especially crime perpetrated and experienced
by poor youth. In fact, this is an appropriate
approach, as long as it is made explicit and
viewed in a relative light, since homicide rates
are the most reliable and sensitive indicator to
reflect on social violence and its trends. However,
this category does not encompass the entire
phenomenon of violence which, from the
public health intervention perspective, needs
to be unveiled, understood, and explained in
its multiple dimensions.
Further within the conceptual sense, and
even referring only to the homicide phenomenon,
Briceño-León leaves a conceptual gap
when he omits the weight of subjectivity (and
what kind of contemporary subjectivity is being
created?). This theoretical and practical parameter
is crucial for dealing with processes of
Briceño-León R 1654
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
violence and victimization if we are not to believe
in a historical trend moved only by economic
and political macro-forces, responsible
for the maintenance of secular inequalities and
poverty. The presence and complicity of the
historical subject, as Sartre 1 reminds us, occurs
even when he accepts all the determinations.
Proof of the need for “subjects’ adherence
to the crime” is the fact that, as Briceño-
León recalls in his paper, the majority of poor
and destitute young people, although victimized
by inequality, behave peacefully, opting
for the established order.
I hope not to be parochial if I make some
distinctions about the case of Brazil, a country
of continental dimensions which, in addition
to its metropolitan areas, has prosperous and
dynamic large and medium-sized cities as well
as very small county seats, the latter often formally
classified as urban, but where the urban
and the rural coexist indistinctively. First, in
Brazil, despite high urbanization along with
perennially poor and even destitute remote areas,
especially in the Northeast, it is not true
that the countryside is stagnated. On the contrary,
in recent decades there has been an increase
in cultivated land, agricultural and cattle-
raising areas, and family farming, in an unprecedented
process of agribusiness modernization,
to the point where the country has become
the world’s largest grain producer. However,
the rural world in Brazil, which is undeniably
changed and incomparable to various
other Latin American countries in this regard,
has failed to become less unequal or unjust.
In the distribution of violence, we have a
quite diversified situation, and while population
density is certainly accompanied by extensive
poverty and lack of decent housing conditions
and unemployment have become a predisposing
factor for entry into the illegal work
market where the labor force feels more valued,
recognized, and better remunerated, there
are many other intervening factors in localities
with high homicide rates. According to a stillunpublished
study by Souza (personal communication;
2005) in which the author works
with compound indicators capable of orienting
the interpretation of the main causes of violence
in Brazil, taking 2000 as the base year, of
5,507 Brazilian municipalities (or counties),
48% did not record a single homicide. Of those
with the majority of the country’s homicides,
the first ten places were occupied by some capitals
and metropolitan areas, in the following
order: (1) São Paulo; (2) Rio de Janeiro; (3) Recife,
Pernambuco; (4) Guarulhos, São Paulo; (5)
Diadema, São Paulo; (6) Jaboatão dos Guararapes,
Pernambuco; (7) Olinda, Pernambuco; (8)
Nova Iguaçu; Rio de Janeiro; (9) Brasília, Federal
District; and (10) Osasco, São Paulo. The
twenty municipalities with the highest violence
and accident rates ? which include seven capitals
(out of the total of 27), 12 metropolitan areas,
and a large city from the interior ? cover
54% of all these deaths. In short, this remark is
meant to say that the highest violence and
crime rates occur in large urban concentrations.
However, not all urban concentrations in
Brazil (and perhaps not all those in Latin America
as a whole) have high homicide rates and
other rates of violence, and it is important to
distinguish between different social dynamics.
Finally, besides the elements proposed by
Briceño-León, I wish to add the specificity of
contemporary violence, where the revolution
in microelectronics and communications media
and formats play a special role in mediation.
The French social historian Wieviorka 2,
also cited by Briceño-León, Castell 3, and Minayo
4, shows the specificity of this global crime
(operating in networks) linked by the most violent
and despicable interests in the shadow of
national states’ weakness, constituting an infra-
politics in itself, linking legal and illegal apparatuses,
international and national players,
and perversely encompassing the poor, who
are left with the warfront and the alternative of
eking out employment in an increasingly restricted
and exclusionary market. It is clear that
current crime that has drastically increased the
homicide rates in Latin America does not consist
only of activity by organized groups acting
in collusion. It reveals an exacerbation of growing
social conflicts, without civilizing mediation
and added to the traditional and culturally
manifest threshold of violence present in the
history of Latin American countries, merging
all these dynamic relations that are expressed
in the current homicides rates, where the various
expressions feed on each other.
Is there hope? Yes. The study by Chesnais 5
on two hundred years of violence in European
history shows the beneficial role of processes
of social inclusion and rights, decreasing inequalities,
and access to formal education as
significant elements for a downward curve in
homicide rates which in the 19th century were
higher than in the most conflict-ridden countries
of Latin America.
1. Sartre JP. Sartre. São Paulo: Abril Cultural; 1980.
(Coleção Os Pensadores.)
2. Wieviorka MO. O novo paradigma da violência.
Tempo Social 1997; 9:5-42.
3. Castells M. Fim do milênio: a era da informação.
Rio de Janeiro: Editora Paz e Terra; 1998.
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1655
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
4. Minayo MCS. A violência dramatiza causas. In: Minayo
MCS, Souza ER, organizadores. Violência
sob o olhar da saúde. Rio de Janeiro: Editora Fiocruz;
2003. p. 23-47.
5. Chesnais JC. Histoire de la violence en Occident de
1800 à nos jours. Paris: Robert Laffont Éditeur; 1981.
Instituto de Filosofia
e Ciências Humanas,
Universidade Federal
do Rio Grande do Sul,
Porto Alegre, Brasil.
[email protected]
Juan Mario Fandino-
Marino
The article provides a wealth of statistical information
on homicides in Latin America and on
some of the correlates traditionally discussed in
the criminological literature. This information
originally appears rather chaotically in different
sources, and the author presents them here in
an organized way, which has great merit for the
debate on the issue and represents a huge research
effort. Among the correlates are poverty,
level of schooling, employment, family structure,
and urbanization. Other traditional correlates,
especially youth aspirations and religion,
are treated at a more conceptual level. Even so,
their treatment systematizes ideas which are
not necessarily original, but which form a valuable
theoretical framework.
Another highly traditional quality, yet still
noteworthy in the overall context of Latin
American sociology, is the fact that the article
is based on a dialogue between theory and data,
which naturally launches a disciplined debate.
It is thus a source of satisfaction to encounter
such a paper for these reasons. Following
these remarks, we now approach our critical
sociological position towards the study.
As for the simple information on homicides,
a very important recent trend, already recognized
in the literature, escapes the battery of
data provided by the author, namely a highly
significant drop in homicides in some key
cities, like São Paulo (37% in five years), Bogotá
(from 80 to 23 per 100 thousand in seven years),
and Cali (25% in nine years) (Kahn T, Zanetic A.
Personal communication; 2005). These decreases
are relevant for the issue, since they may result
both from successful public policies (Kahn
T, Zanetic A. Personal communication; 2005),
as well as cyclical historical trends 1,2, or both,
and they find no echo in the argument presented
by Briceño-León.
The discussion on cities and urbanization
provides theoretical elements that are not mistaken,
but which are far from constituting a basis
for what the author calls a “sociological
framework for the explanation of violence”. The
discussion at stake is limited to the assertion
(as evident as it is un-analytical) that the cities
are not what we wish they were, and are what
we wish they were not. The affirmation that
“Latin America cites were a place of hope for security
and law, hence the great rural-to-urban
exodus in the 1940s and 50s” simply fails to
agree with the data 3. In the latter study we find
that in Colombia the attraction exerted by
cities is equivalent to only 10% of the explanatory
power of expulsion by demographic pressure
and agricultural technology, the most important
factors at the time. The etiological literature
on migration generally did in fact contemplate
the security issue, but in a localized
way and in specific cases, and it has generally
never been considered a determinant factor in
Latin America.
The problem with the author’s structuring
of a sociological framework in three causally
different dimensions (macro-factors that “originate”,
meso-factors that “foment”, and microfactors
that “facilitate” violence), begins when
he abandons the causal epistemological status
proper to each of the three traditional levels of
analysis for crime: micro, meso, and macro.
The traditional effort in the literature working
with causal links between these levels is very
clear, beginning with Sutherland in 1924 4,
moving on to Cloward & Olhin in 1960 5, and
reaching the “integrated models” in recent
years (where the explicit objective of analysis is
to work with this issue). This is what is at stake
in terms of an epistemological framework, and
not the three ad hoc dimensions of Bricenõ-
León. It is true that the “violentological” and
criminological specificity of Latin America
should be understood at the macro level, which
would justify placing this specificity as having
originated at this level of aggregation. But contemporary
urban violence in Latin America is
not such a unique legacy for our continent, as
shown by Gómez-Buendia 6. The links are much
more complex.
As for the author’s empirical analysis, it is
generally limited to presenting data on each
aspect or variable separately, after which,
based only on their approximate contemporaneity,
(!?), he derives conclusions on their
causal nexus with homicides. In other words,
the author’s empirical analysis contains practically
no relational factual evidence where we
might observe some types of co-variations or
associations. The case in which the author approaches
a relational empirical methodology is
his analysis on the relationship between poverty,
urbanization, and homicide rates. We do not
refute that in a broad and generic sense, these
Briceño-León R 1656
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
phenomena and their interactivity (correctly
emphasized by Briceño-León) may be associated
in some way. However, at least based on
the Brazilian studies on the subject, we know
that the relationship between poverty and homicides
is not linear, and that extreme poverty levels,
including those in urban areas, are not the
ones that stand out as factors for homicides.
Following the level of spatial-temporal aggregation
that Briceño-León intends to adopt,
rightfully and pertinently, the treatment of the
hypotheses raised is virtually worthless, methodologically
speaking. In fact, based on a simple
visual inspection of Table 6, invoked by the author,
one cannot conclude in favor of the rigor
of his hypothesis. Based on a superficial test
using the author’s own Table 6, these relationships,
as I will illustrate next, prove to have very
little explanatory power, even though they may
be interesting. Based on the data from Table 6,
we calculated the following multiple regression
equation:
H = a+b
1
X
1
+b
2
X
2
+b
3
X
1
X
2
+E,
were H represent the homicide rates and X1
and X2 represent poverty and urbanization, respectively.
Despite the calculating problems
based on n = 16 (very small), the results are: a
precarious adjusted R2 of 0.074; the betas, even
with the model’s precarious overall adjustment,
are interesting and lend some credit to
Bricenõ-León’s theory: -1.67 for poverty, -0.76
for urbanization (both negative!), and finally a
positive beta of +1.59 for the multiplicative interactivity
term. None of the coefficients is significant
at 0.05.
1. Fandino-Marino JM. A violência na América Latina
e seus ciclos altruísta e egoísta/anômico. Revista
do Direito 2000; (14):1-22.
2. Fandino-Marino JM. The moral cycle of egoistic
and altruistic violence: a century of bloodshed in
Colombia. In: Anderson M, editor. Cultural shaping
of violence. West Lafayette: Purdue University
Press; 2004. p. 268-83.
3. Fandino-Marino JM. Determinantes econômicos
e sociológicos da migração rural-urbana. Revista
de Economia Rural 1979; 6:124-42.
4. Sutherland E, Cressey D. Principles of criminology.
Chicago: J.B. Lippincott; 1955.
5. Cloward R, Ohlin LE. Delinquency and opportunity.
New York: Free Press; 1960.
6. Gomez-Buendia H. Urban crime: global trends
and policies. Tokyo: United Nations University;
1989.
Grupo de Investigación
en Violencia Urbana,
Universidad de Antioquia,
Medellín, Colombia.
higarcia@quimbaya.
udea.edu.co
Carlos Alberto
Giraldo
Héctor Iván García
From various social spaces, members of society
have called attention to the gap between (1)
the magnitude and omnipresence of violence
in the lives of Latin American peoples and (2)
the level of output of systematic knowledge
and capacity for social and political response
to such a disturbing phenomenon, which blocks
the potential of individual and societal projects.
This situation calls for a renewed social
and political commitment by researchers and
society at large.
The combination of the categories violence,
the urban, public health, and Latin America
and their interrelations constitute a highly suggestive
set. This combination appears to be in
the minds of many researchers, but few have
dared (like the author) to propose an alternative
that links (within a single view) the question
concerning the threatening nature of the
Latin American city and the growth of violence.
The author’s distinction between two levels
of social life as the point of departure for establishing
an explanatory framework for social reality
represents an important methodological
wager. To a major extent it corresponds to the
need to relate the structural and situational dimensions
in order to propose explanatory connections
for violence. The author tackles the
trends that propose explanatory theories focused
exclusively on the social structure and
others that rely on the situational to establish
linear causal equations.
What really stands out is the proposed explanatory
structure between the three levels of
social event, i.e., macro, meso, and micro-social,
with a differential explanatory potential
between that originating violence, that which
foments it, and that which facilitates it. The
limitation is that these categories are treated
with a high level of generalization and with
such a nonspecific empirical reference that it
would be difficult to reach agreement among
researchers on the pertinence of the proposed
levels of determination and about which factors
belong to one category or another.
Focusing on the basic concepts, we emphasize
the field’s complexity and the need to establish
Latin American consensuses on the basic
concepts, in order to spawn rapid and productive
exchange among the researchers. The
perspective that violence does not represent
merely a pathological event produced by various
factors that are exogenous or alien to the
development of societies and their collective
existence, but on the contrary, that it is a phenomenon
that accompanies the development
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1657
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
of the widest range of relations 1 in both the
private and public spheres, stems from the article’s
approach (unless we have misread it) to
violence as a kind of disorder that parasitizes a
society, that acts as foreign body in it and is
therefore something to be extirpated. This point
of view has consequences for the final focus of
research and the localization of violence in
specific geographic areas of cities, and in certain
human groups, like youth and the poor, in
certain regions of the country, placing the rest
of society in the position of victims, which at
least in the Colombian case has generated an
infernal circular perpetuation of violence.
The other concept refers to the urban. The
author’s reflection on the distinction between
“city, citizenship, and violence” is highly interesting.
What is out of sync is that he has introduced
something as a footnote which in our
opinion should link the overall argument together,
due to its huge explanatory potential.
In our opinion 2, urban violence does not relate
necessarily to the topography where it occurs,
but to the violations of various types of rights
and freedoms that occur in interactions among
citizens, and between the latter and the state
or other organizations (all of whom are actors
in our contemporary urban society); to the logics
and dynamics woven into the construction
of the urban and the city and its characteristic
as a horizon for conflicts that gives rise to violence
as a multifaceted and ubiquitous phenomenon.
1. Uribe MT. Nación, ciudadano y soberano. Medellín:
Corporación Región; 2001.
2. Gómez JA, Agudelo LM, Álvarez T, Cardona M, De
Los Ríos A, García HI, et al. Estado del conocimiento
sobre la violencia urbana en Antioquia en
la década de los noventa. In: Angarita P, editor.
Balance de estudios sobre violencia en Antioquia.
Medellín: Editorial Universidad de Antioquia;
2001. p. 163-92.
Universidad Nacional
de Lanas, Lanas, Argentina.
[email protected]
Hugo Spinelli In his article, Briceño-León develops the analysis
of what he calls “the stage for a silent and
undeclared war”. The eloquence of the data
and their relevance to almost any city in Latin
America show the recurrent horror portrayed
in the majority of the articles dealing with the
issue of violence at the population level. But
what should be done with such horror? How
can the silence of this undeclared war be broken?
By speaking, generating one, two, a thousand,
a million conversations to break this
tragic muteness, this paralysis resulting from
the horror. Such conversations must change
the sense and meaning of what is said when
talking about violence, proving that things can
also be done with words. Violence is not a state,
it is a process. Enough of cowardice!
We agree that in our countries the city of
law has become the city of fear. That urbanization
and television have democratized expectations,
but that the result is inequality and exclusion,
the connected and the disconnected.
That violence and its consequences are reproduced
numerically in the statistical reports,
and that the subjective level increases the perception
of becoming possible victims of violent
acts. In his attempt to analyze the “object”,
Briceño-León develops a structuring proposal
that he assumes as a non-universal model. This
proposal has been used in the health sector,
ranging from the Situational Strategic Planning
logic of Carlos Matus 1 to the work of Pedro
Luis Castellanos 2, who links Matus’ logic to the
study of the health-disease process under a
structure of the general, the particular, and the
unique. These interpretative processes are highly
useful to approach the logic of actors and
scenarios, so as to avoid crystallizations or simplifications
that justify technocratic norms, or
on the other hand the kind of inaction that results
from economic over-determination.
Models, structures, and classifications, but
what purpose do they serve? Yes, fine, if they hierarchically
organize the interpretations of citizenship
on the problem. No, not if they are to
achieve “scientific explanations” that crystallize
such a complex and dynamic process as violence.
The risks of medicalizing violence are still
present. By classifying the problem, to what extent
do I accept it as part of my field of knowledge?
If I accept it, to what extent do I problematize
it at the social level? Or do I include it
as an object of investigation in such a way as to
ensure my reproduction as investigator? Is this
a valid dilemma? Is this always the situation?
We should not simplify the process of violence.
To avoid the temptations of graphs and
to tackle complexity is part of being honest as
researchers. Of course complexity should not
be measured merely by speeches, but by acts
(which include words) and better still, by their
impact. Words and acts are nothing more than
actions by subjects. Individual and/or collective
subjects. Subjects of language.
It is difficult to take a step back as the discussant
of a theme that affects us as deeply as
Briceño-León R 1658
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
violence, an issue that we have experienced personally
and which we fear experiencing again,
since we do not know when, where, or with
whom we will face another violent situation.
We physicians, health workers, who imagine
ourselves vanquishing disease, now face an
“epidemic” or rather “pandemic” problem, the
main cause of death among young adults in the
majority of our countries. We find ourselves
immersed in a process where we shift from
subjects of knowledge to “objects” of violence.
We must not be indifferent to this process; may
it at least serve for us to review our ways of understanding
the problems. We citizens of Latin
America, men and women, must understand
that we should not take violence for granted.
We need to reclaim public space and build social
citizenship that turns the city of fear into
the city of law, of rights, and of social citizenship.
I believe this is Roberto Briceño-León’s
spirit when he ends his paper by quoting the
old German saying “Stad Luft mach frei”.
1. Matus C. Política planificación y gobierno. Caracas:
Fundación Altadir; 1987.
2. Castellanos P. Sobre el concepto de salud-enfermedad,
un punto de vista epidemiológico. Cuad
Med Soc 1987; (42):15-24.
Institute of Political Science,
Free University of Berlin,
Berlin, Germany.
[email protected]
Ruth Stanley Nasty, brutish and short: violence
and the unrule of law in Latin America
Urban violence in Latin America has become
so ubiquitous that it can be seen not only as a
problem of individual security but also of public
health, as the WHO data cited in the first paragraph
of Briceño-León’s article amply demonstrate.
Any attempt to grasp the underlying
causes of violence is to be welcomed, as an understanding
of the causes is a necessary first
step towards overcoming them and hence reducing
violence. It is particularly to be welcomed
when, as here, a broad-ranging analysis
is presented that distinguishes the impact of
various factors. Thus, Briceño-León differentiates
between originating factors of violence,
factors that foment, and factors that facilitate
violence. This approach avoids the pitfalls of
more simplistic explanatory theses, by distinguishing
between causes in the strict sense of
the term, and factors that tend to exacerbate
the propensity to use violence. Thus, the widespread
availability of firearms is seen here as a
facilitating condition, not an underlying cause.
Canada, a country where firearms possession
is fairly common, is at the same time a society
with very low levels of violence because the
originating factors – essentially Briceño-León
identifies these as being related, not to poverty,
but to inequality – are absent.
In the space allocated to me for this commentary,
it would be impossible to do justice
to the scope of this rich, wide-ranging, and
thought-provoking article. Instead, I propose
to point to two factors that I believe could usefully
be given more weight in the analysis. I
then turn to the role of the state as a factor of
violence and end by linking this with the question
of citizenship.
My first point relates to those factors that
Briceño-León identifies as fomenting violence,
i.e., they are not its originating causes, but tend
to stimulate or encourage the use of violence.
Briceño-León identifies three such factors: urban
density, the culture of masculinity, and the
drugs market. Two additional factors could, I
believe, be usefully added. The first is the culture
of hedonistic consumerism that has become
pervasive throughout much of the world.
Probably never before has the acquisition of
material goods mattered so much to individual
self-esteem and peer group recognition – especially
among the adolescents that Briceño-
León identifies as being particularly violenceprone
because of their imprecise insertion into
society in the transitional stage between childhood
and adulthood. He argues, compellingly,
in my view, that there is a huge discrepancy between
the democratization of cultural aspirations,
in which television plays a crucial role,
and the actual possibilities of fulfilling such aspirations
in an increasingly unequal society:
“We are terribly equal in what we desire and
frightfully unequal in our real possibilities to
achieve it.” So the culture of consumerism is
present in his analysis; I suggest that it deserves
to be treated in its own right as a factor that foments
violence and its impact analyzed in the
same way as he proposes for the culture of masculinity.
A further factor that foments violence and
that, in my view, deserves rather more weight,
is the discursive treatment of the phenomenon
in the mass media. Discourse not only reflects,
but also creates, perceptions of reality, and as
Briceño-León points out, a subjective feeling of
insecurity has real consequences for behavior,
from the acquisition and hasty use of firearms
to various forms of violent and sometimes anticipatory
self-help. It is noteworthy that the
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1659
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
The author replies
O autor responde
Roberto Briceño-
León
I wish to begin by expressing my appreciation
for the generous comments and sharp criticism
sparked by my article. The opportunity
offered by the Debates section of Cadernos de
Saúde Pública confirms the key importance
that the scientific community ascribes to expanding
our knowledge and scientific practice.
It behooves me to humbly and joyfully accept
such comments and criticism, because the precise
remarks on the aspects that are missing or
that merit greater substantiation reveal our limitations.
The commentary is valid and demonstrates
our inevitably limited way of reconstructing
the world, and the intelligent and
knowledgeable debate is so varied culturally
and professionally that it allows me to move
forward and improve the sociological framework
through which I propose to understand
violence, refining its favorable aspects and correcting
its flaws.
At the end of a beautiful text on the empirical
basis of knowledge, Popper 1 proposes a
metaphor according to which we do not build
our knowledge on a rock, but rather on muddy
ground. That epistemological premise always
holds true, because a theory’s role is to blaze
subjective sensation of insecurity has grown
considerably in recent years, including in cities
that can be counted as among the safest in the
world. Within Latin America, Briceño-León
identifies Argentina as one of the region’s least
violence-prone countries (Table 2), yet recent
media discourses in that country could lead
one to imagine that a war without quarter is
raging between society and violent criminals,
and private security provision has become one
of the country’s few growth industries. If we accept
the proposition that discourse can impact
on behavior, I believe it is a logical second step
to integrate a critical analysis of discourse into
this explanatory model as a factor that foments
violence.
On a more general level, and from a West
European perspective, the question arises:
What is the role of the state in all this? The notable
weakness of the Latin American state, especially
in its legal dimension (as argued above
all by Guillermo O’Donnell) has a huge impact
on citizenship, and the state fails in its most
fundamental and essential task, that of guaranteeing
a basic level of physical security for its
citizens. Thus, weak state performance is reflected
not only in an inability to fulfill welfare
functions, but also in non-fulfillment of the
central function of providing security. Given
this, it is perhaps not surprising that life in
many of Latin America’s cities is “solitary, poor,
nasty, brutish, and short”, just as Hobbes imagined
life before the social contract. Of course
there are exceptions within the region, states
such as Chile and Costa Rica, where the state is
comparatively strong in its legal dimension and
the level of interpersonal violence is low (lower
indeed than in an old-established democracy
like the United States, as Briceño-León points
out), as well as important gradations that
should make us cautious about sweeping generalities.
In one respect, the situation is worse
than Hobbes’ imagined state of nature, for not
only does the state fail to provide a minimum
level of security, but it also invests its agents
with an authoritative power that is frequently
used arbitrarily and with no regard for legality.
Citizens are thus doubly insecure, threatened
by high levels of interpersonal violence and also
by state agents whose supposed function is
to protect them. Pre-state forms of political organization
no longer exist, and the state, through
its agents, is effectively an additional source of
violence through its acts of omission and commission.
This argument does not (necessarily)
imply the need for more security agents, but
rather points to the need for more rule of law.
The weakness of the state in its legal dimension
is also directly relevant to the meaning of citizenship.
If we define citizenship with Tilly 1 (p.
253) as “a set of mutually enforceable claims relating
categories of persons to agencies of government”,
we can readily perceive that the concept
has little meaning where formal claims,
foremost among them the right to a basic level
of physical security, are effectively non-enforceable.
The ideal of urbanity thus gives way
to the truncated citizenship described by Briceño-
León.
1. Tilly C. Conclusion: why worry about citizenship?
In: Hanagan M, Tilly C, editors. Extending citizenship,
reconfiguring states. Lanham: Rowman &
Littlefield; 1999. p. 247-59.
Briceño-León R 1660
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
trails; it does not pretend to cover everything,
but rather some essential aspects; it does not
propose to eliminate complexity, but to make it
somewhat more comprehensible and manageable
for researchers and policy-makers.
I will answer the comments and criticisms
with the purpose of intensifying and continuing
the debate, not to close it or to treat it as
concluded.
The definition of violence
and homicides
Several commentators have insisted on the definition
of violence I used and why the term is
restricted to homicides when there are many
other expressions of violence.
We at LACSO use a simple but restricted definition
of violence that has a strong operational
value: “the use or credible threat of use of
physical force against others or oneself” 2 (p. 196).
This definition is very similar to those used by
the WHO in its World Report on Violence and
Health 3, the National Research Council of the
National Academy of Sciences 4, and the American
Sociological Association 5, but we prefer
to limit our definition to physical violence (we
do not consider psychological or structural
forms or their consequences, which may be
physical or psychological). We do not distinguish
whether the act is committed against an
individual or a social group, or if it results from
legal action (capital punishment in some countries)
or illegal. Some definitions associate violence
with any serious offense: speaking in a
loud voice and with a certain tone, using offensive
language in an argument, or even using silence
as an instrument can entail major psychological
aggression. We see a basis to this
and agree with Bandura 6 that this type of behavioral
response is intended to harm others,
and is thus similar to violence; however, we
think it is inadequate to place insult and homicide
on the same level of harm. Therefore, like
Bandura, we prefer to call all such offenses aggression,
but use violence exclusively for acts
with a physical component and the use of force.
Still, physical aggression varies widely, from
slapping one’s spouse in the face to passionate
homicide, from a fistfight at school to a shootout
between gangs. Our studies consider all these
forms of violence 7,8, but I have limited myself
to its most dramatic and irreversible form:
homicide. There are several reasons for this
choice. The statistics are much more reliable
for mortality than violence rates in general. We
can reasonably trust the figures on consummated
homicides, unlike those on failed attempts,
where victims are only injured. Statistics on
overall violence are not highly reliable because
crimes are rarely reported in Latin America,
since victims and their relatives fear reprisal by
aggressors and have little trust in the police to
offer effective protection (or in the criminal
justice system to punish the perpetrators). Filing
complaints is thus risky and has little benefit,
and this has a negative impact on statistics.
We also intend to show an important change
in the criminal perspective: the new violence is
striking not so much because of the increase in
crime as a whole, but because of the increase
in violent crime and its lethalness. This was described
in the United States in a landmark study
9, but a similar trend has occurred in Latin
America. In the 1950s and 60s, in the golden
era of capitalist growth, juvenile gangs fought
with knives and chains, while they now wield
high-powered weapons. The social conflict is
the same, but the lethalness of firearms is different,
and homicides are thus a relevant indicator
of the new violence. The current violence
also gives us a different perspective from that
of criminology, considering the phenomenon
not from an individual perspective, but from a
social one, as so-called critical criminology has
done 10. Individual motives are not important
to us, nor are serial killers or those with an individual
pathology. The social fact of violence
is important because of its magnitude. Most
killers in Latin America are not mentally ill, but
ordinary young people with the drama of a
senseless life devoid of a future, in which they
describe with astonishing tranquility and with
no remorse how they have killed two, three, or
more people. In their restricted social milieu,
violence has become normal in the Durkheimian
sense 11, and they are only actors of that
normality.
The city and crime
The history of cities is linked to state control of
violence and the constitution of individual
rights. Of course cities have not been a haven
for saints. I do not see the city as paradise.
Rather, I attempt to retrieve the notion of citizenship
connected to the city as a privileged
place for individual rights.
Early theoreticians of urban sociology like
Robert Ezra Park 12 viewed the city as the place
where natural instincts were controlled and the
cultural order was constructed. In a metaphor
with categories from the second topic of Freud,
the city is the place where the Uber-Ich (super-
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1661
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
ego) takes precedence over the rebel Es (id) 13,
or in more contemporary terms, where social
control based on the rule of law is imposed.
In a rural society controlled by landowners
and their private armies, with a limited institutional
base or state power, the cities represented
the place where rights could be pursued. In
rural society, order was based on customs and
the traditional social division of power. There
might be tranquility, but with no individual
rights or rule of law. In Latin America, the city
represented a place to which peasants could
migrate in search of welfare and freedom. If it
is difficult today to organize rural unions or
peasants’ movements, we can only imagine the
situation in the 1930s and 40s, when people
had less schooling and there was no mass media
to report abuses. I have not meant to say
that people left the countryside because of insecurity
or violence, but because they were
searching for the welfare and rights offered by
urban capitalist society. I do not claim that
cities were a paradise, but that despite their deficiencies
and difficulties, they offered better
conditions than the abandonment people experienced
in the countryside. Interestingly, in
Colombia some 2 to 3 million displaced persons
have fled to the cities, terrified by the rural
guerrilla warfare. Although violence in Colombian
cities is extremely serious, people feel
they have some protection. In the insecurity of
the city, they can aspire to more security. They
can enjoy something of the citizenship that is
denied to them in the countryside.
Cities represented modernity, with rights
and benefits and the possibility of a normative
order and opportunities. That dynamic is found
in studies on urbanization 14. The 20th-century
Latin American city does not correspond to the
urban industrial model of Europe or the United
States. Rather, it resembles the pre-capitalist
city that was the seat of the landowners who
spent the rural surplus there. However, the magnitude
of the phenomenon far exceeded the
city’s possibilities, and that is what I argue, that
the phenomenon was rapid and widespread
and raised hopes that were eventually frustrated.
This does not involve the question of whether
cities are producers of violence. Quite to the
contrary, they tended to reduce violence, as occurred
with the first generation to arrive in the
cities. However, when many more migrants arrived
and the process became overwhelming, a
new generation was born and the city became
a hostile environment where one had to resort
to violence to survive.
Minayo’s data on Brazil clearly show that the
problem is shifting from the countryside to the
urban areas and from small towns to large cities,
and that urbanization in Latin America tends to
concentrate power, wealth, and violence.
My analysis on violence and the city acknowledges
the influence I have received from
the Chicago school of urban sociology (also
called human ecology) 15. Contemporary sociology
has neglected how the population’s spatial
location and distribution influences people’s
behavior and societal organization, one of
the great lessons from studies by Burgués &
Lucke 16 on the family and how this changed
according to the concentric zones they identified
in American cities, or the analysis by Bell
17 on the relationship between behavior by the
mafia in the Port of New York and the possibilities
arising from intricate territorial organization
of the docks. My proposed model reclaims
the importance of situational analysis, both in
its strictly physical dimension of the city and
areas, and in the social and political conditions
that actors identify as possibilities for their
course of violent action.
The specificity of violence
in Latin America
Violence is becoming increasingly homogeneous
worldwide, but two different processes
occur in parallel. First, the mechanisms of violence
are increasingly similar because the operational
modes of violence and crime have
spread, and delinquents learn and repeat them
in different countries and regions. Second, there
are local specificities whereby these mechanisms
are adapted to the circumstances in each
city or country.
This occurs as a consequence of the globalization
that makes societies and possibilities
for crime more similar. For example, the widespread
use of ATMs offers an opportunity for
crime. Criminals first attempted to break the
machines to remove the cash, but they soon realized
it was easier to threaten card holders at
gunpoint and force them to withdraw the money.
Thus was how the so-called “flash kidnappings”
spread so quickly around the world, as
fast as ATMs have been installed. Violence has
also gone global as a cultural process. Perhaps
the most relevant example is that of the “maras”,
or Central American youth gangs, stemming
from cultural exchange through the merger between
gangs from Los Angeles and the dynamics
of poor youth in El Salvador. Thus the term
“mara salvatrucha” (mara = group; salva = El
Salvador; trucha = friends). The phenomenon
spread to other neighboring countries and oth-
Briceño-León R 1662
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
er cities in the United States and Spain, in places
where Central American migrants have reached,
reproducing an apparently local phenomenon
elsewhere.
But violence is not identical worldwide.
WHO statistics show an important difference
between regions in the proportion of suicides
and homicides (Table 12). In Africa and the
Americas there are two or three times more
homicides than suicides, while the opposite
occurs in Europe and the Western Pacific.
Africa and Latin America are not completely
similar. The figures for Latin America are far
higher, because the rate shown in Table 12 for
the Americas includes the WHO classification
with the United States and Canada, so that the
Latin America rates are really higher than
shown, surpassing those of Africa. The African
situation is also different because it includes a
component of war that does not exist in Latin
America. The violence in South Africa is probably
most similar to that of Latin America, possibly
because as a nation it is also more similar
to the Latin American countries.
Despite Africa’s great poverty, its inequality
has been historically lower than in Latin America,
although these differences tended to decrease
in the 1990s. Milonevic 18 explains this
situation based on a hypothesis by Kutznets,
whereby African countries are characterized by
agriculture and widespread poverty, and in
such cases the differences between lower and
higher income strata tend to be smaller. Table
13 shows the Gini coefficients and per capita
GDP for three regions, where Latin America has
the greatest inequality.
The relationship between poverty
and violence
A topic that has received great interest and criticism
is shown in Table 6, where I report on
poverty, urbanization, and homicides. The commentators’
observations and suggestions are
appropriate, and these data certainly require
more detailed analysis and better presentation.
The idea was to identify an important research
path through aggregate data showing
differences between countries and suggesting
that in countries with high urbanization and
high poverty there are more homicides than in
others with only one of these conditions. The
purpose was to formulate a hypothesis in this
direction. Based on this and the very limitations
of a text (already quite long) that attempted
to touch on multiple dimensions rather than
delving deeply into any of them, we did not
perform the statistical tests that the discussants
are fairly requesting.
However, I do wish to clear up a possible
misunderstanding. I do not believe that poverty
is a direct cause of violence. When there is
high urbanization and high poverty what really
exists is extensive social inequality, yet the
most urbanized areas of Latin America are also
those with the greatest wealth. We agree (and
have written) that poverty itself is not a cause
of violence, but the majority of both victims
and perpetrators are poor. The worst violence
is not in Haiti, Northeast Brazil, or the poorest
areas of Venezuela or Mexico. It occurs where
there has been heavy, rapid urbanization and
poverty coexisting with wealth, as in São Paulo,
Rio de Janeiro, Caracas, and Mexico City. My
hypothesis takes this direction.
The cultural dynamic and violence
Violence is a cultural act in its origins and consequences.
I thus agree that we should pay
more attention to the hedonistic culture and
consumerism that dominates the aspirations
of citizens and is converted into frustration
and a motive for violence. But it is interesting
that these processes occur under circumstances
in which it is not possible to seize the wealth
from others. A study by Nisbett & Cohen 19 on
the culture of honor argues that this distinguishes
the violence in agricultural as compared to
pastoral economies, because in the latter it is
possible to seize wealth (cattle) more easily than
robbing crops. In a culture of consumerism,
objects form part of the individual’s “being”, so
to kill or be killed for a pair of shoes means a
struggle for the essence, for identity, and violence
is the means that allows conquering the
objects that permit one to be a person. We thus
argue that violence gives sense to some senseless
lives.
Table 12
Homicide and suicide rates by regions of the World
Health Organization (per 100,000 inhabitants).
Regions, WHO Homicides Suicides
Africa 22 7
America 19 8
Europe 8 19
Western Pacific 4 21|
Source: based on data from WHO 3.
URBAN VIOLENCE AND PUBLIC HEALTH IN LATIN AMERICA 1663
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
Few studies approach violence from a gender
perspective, which should really be considered
in a much broader dimension. Most studies
on violence and gender focus on violence
against women. We did not want to emphasize
the importance of the culture of masculinity as
a factor in the gender dimension of social relations
and ways to solve conflicts, and we thus
limited it to a factor that foments violence.
However, we agree that this issue must be extended
to other factors, because for example
the meaning of weapons as phallic symbols is
relevant to the construction of masculinity, in
children’s games, and in power management
among men.
The role of the mass media (especially television)
in violence is a true challenge for the
social sciences. Managing this issue is very difficult,
since it covers highly diverse topics. We
believe it is necessary to avoid excesses such as
classifying pornography on television as violence
20. However, it is relevant for violent
movies, where the hero decapitates dozens of
people in a few seconds, and violent cartoons
like Tom and Jerry. The issue also relates to the
treatment of news and to what extent it responds
to the legitimate right of citizens to be
informed or to scandalize them and to take delight
in violence for marketing purposes. The
effects are highly diverse, and the importance
delinquents assign to the media and their pride
in saving and flaunting newspaper clippings
showing them in photographs is surprising.
They display not shame, but pride in being recognized
as someone on the “social pages” of
crime.
Much further research is needed on such
effects and on a theoretical matrix that imposes
limits but respects individual rights and freedom.
The media serve as a powerful weapon,
and as with all power regulations, a counterbalance
must be imposed on them. Still, the
easy resort to censorship and control must be
avoided as highly risky in societies with such a
fragile institutional base and with authoritarian
temptations such as those existing in Latin
America. The question posed by Norberto Bobbio
thus acquires great force: Who controls the
controllers?
Rule of law and the state
In Latin America the government may be weak
or strong, but the state is always weak. There is
a fragile institutional base that is able to pass
many laws, but is unable to enforce norms or
punish violators. The discussants are right when
they say that this aspect is not fully developed
in the paper and should be part of the sociological
framework I propose.
The lack of a strong institutional base (a
solid state and the rule of law) contributes to
the increase in violence, both by inducing delinquency
through impunity and propitiating violent
responses by citizens (attempts at social
cleansing, etc.), who feel unprotected, and by
the police, who justify their extrajudicial action
on the grounds that they must apply direct
punishment since the criminal justice system
fails to do so.
This same institutional weakness fuels a
sense of insecurity among citizens, who feel
they receive inadequate protection from the
state (or in some cases even worse, that state
agencies pose a threat). The resulting dynamic
is highly complex. In many cities the police
force has been cleaned up to expel or punish
delinquent members, and while citizens applaud
such measures, their fears are confirmed,
namely that the border is fuzzy between police
and criminal violence.
Just as the standard of living is subjective,
so is security. Security is a subjective sensation
based on both experience (personal or vicarious)
and information obtained about reality. In
the development of this sensation, people’s expectation
of security in the social context plays
a key role: in safer cities, even minor violence
is portrayed as serious by the media, while in
others a dramatic event is necessary to impress
the public. Thus, disarmament campaigns in
some countries have used drawings with crossedout
revolvers, while in Colombia mothers have
painted silhouettes of sub-machine guns on
their posters protesting against violence.
This dynamic of reality and perception of
state policies and society has its effects. I am
unfamiliar with the situation in São Paulo, but
according to some analysts the decrease in
homicides in Brazil is related to the program to
restrict possession of firearms and limitations
on alcohol consumption (Diadema). In Colom-
Table 13
Inequality in three different regions, 1998.
Africa Latin America Asia
Mean Gini 47.1 50.5 35.6
Gini standard deviation 7.9 6.2 7.7
Average GDP per capita 1670 5825 6177
Source: based on Milonevic 18.
Briceño-León R 1664
Cad. Saúde Pública, Rio de Janeiro, 21(6):1629-1664, nov-dez, 2005
bia, Bogotá and Cali have taken to measures
against the factors that facilitate violence, and
public policies have been implemented to restrict
alcohol consumption on certain dates to
limit possession of guns. Meanwhile, in Cali a
youth employment program was developed
that would not affect the macro social processes
of unemployment that we describe as an
originating factor, but which could have an impact
on the final expression and provide opportunities
for many youths. Some poor areas
now have multi-purpose public buildings with
combined services (police, care for battered
women, primary health care) demonstrating
the state’s presence close to the people. Campaigns
have been conducted in Bogotá to motivate
and promote peaceful conflict resolution
and civic behavior (including clowns as animators),
while changing people’s perception of the
state and helping improve the city’s life and
economy. This is surprising in a country with
high levels of conflict and violence, but this
may also be where people have felt the greatest
impact of violence and have decided they need
to seek alternatives.
Our infinite ignorance
I have left many questions unanswered. Neither
my abilities nor the common sense that
prevents us from abusing the space provided
by this fine journal would allow otherwise. But
that is how science is, making advances by correcting
our errors. Violence is multidimensional,
and my contribution to the debate is not intended
to be exhaustive. Rather, it is proposed
as a theoretical framework allowing us to place
some order in the myriad of variables and factors,
making them more comprehensible and
orienting future research to identify the real
scope of public policies, depending on the types
of factors which may become targets for intervention.
Our knowledge is neither definitive nor useless.
Our disagreement keeps the desire to search
alive and our doubts serve as a stimulus, all of
which confirms the beautiful expression by
Popper 21 (p. 29): “Our knowledge can only be
finite, while our ignorance must necessarily be
infinite.”
References
1. Popper K. La lógica de la investigación científica.
Madrid: Tecnos; 1962.
2. Briceño-León R., Camardiel A, Avila O, Armas E,
Zubillaga V. La cultura emergente de la violencia
en Caracas. Revista Venezolana de Economía y
Ciencias Sociales 1997; 3:195-214.
3. World Health Organization. World Report on Violence
and Health. Geneva: World Health Organization;
2002.
4. Reiss A, Roth J. Understanding and Preventing Violence.
Washington DC: National Academy Press;
1993.
5. Levine F, Rosich K. Social Causes of Violence:
crafting a science agenda. Washington DC: American
Sociological Association; 1995.
6. Bandura A. Aggression: a social learning theory.
Englewood: Prentice-Hall; 1973.
7. Briceño-León R, Camardiel A, Avila O, Armas E.
La violencia doméstica: predoctores sociales y
culturales. Acta Científica Venezolana 1998; 49:
248-59.
8. Zubillaga V, Briceño-León R. Exclusión, masculinidad
y respeto: algunas claves para entender la
violencia entre adolescentes en barrios. Nueva
Sociedad 2001; 173:34-78.
9. Zimring FE, Hawkins G. Crime is not the problem.
Lethal violence in America. New York: Oxford
University Press; 1977.
10. Del Olmo R. América Latina y su criminología.
México DF: Siglo XXI; 1981.
11. Durkheim E. Les règles de la méthode sociologique.
Paris: Presses Universitaires de France; 1999.
12. Park RE. Human communities: the city and human
ecology. Glencoe: The Free Press; 1952.
13. Freud S. El yo y el ello. Obras completas, v. III.
Madrid: Biblioteca Nueva; 1973.
14. Germani G. Sociología de la modernización. Buenos
Aires: Paidós; 1971.
15. Briceño-León R. La naturalización de lo social. In:
Acosta M, Briceño-León R, editors. Ciudad y capitalismo.
Caracas: Ediciones de la Biblioteca de la
Universidad Central; 1987. p. 79-96.
16. Burgués E, Lucke F. The Family: from institution
to companionship. New York: American Book
Company; 1953.
17. Bell D. El fin de las ideologías. Madrid: Tecnos;
1964.
18. Milonevic B. Is inequality in Africa really different?
Washington DC: World Bank; 2003.
19. Nisbett RE, Cohen D. Culture of Honor: the psychology
of violence in the South. Oxford: Westview
Press; 1996.
20. Kriegel B. La violence á la televisión. Rapport de
la Misión d’evaluation, d’analyse et de propositions
relatives aux représentations violentes á la
televisión. Paris: PUF; 2003.
21. Popper K. Conjectures and refutations. The growth
of scientific knowledge. London: Routlegdge/
Kegan Paul; 1972.
