Tendencias historicas y diferencias transnacionales en los...
Transcript of Tendencias historicas y diferencias transnacionales en los...
Tendencias historicas y diferencias transnacionales en los homicidios: Como cambia la violencia letal y sus posibles determinantes
Manuel Eisner
Director, Violence Research Centre (VRC)
Institute of Criminology
University of Cambridge
Structure
How much variation in homicide rates is there across time and between societies?
Where does it come from?
Variation in Homicide Rates in the
Contemporary World
-
10,0
20,0
30,0
40,0
50,0
60,0
70,0
Icel
and
Norw
ay
Sw
itze
rlan
d
Mo
rocc
o
Slo
ven
ia
Un
ited
Kin
gdo
m
Aust
rali
a
New
Zea
lan
d
Iran
(Is
lam
ic R
epu
bli
c of)
Slo
vak
ia
Cze
ch R
epubli
c
Port
ugal
Bel
giu
m
Ser
bia
Mal
aysi
a
Bulg
aria
Tu
rkey
Pak
ista
n
Mongoli
a
Isra
el
Hai
ti
Un
ited
Sta
tes
of
Am
eric
a
Jord
an
Suri
nam
e
Rep
ub
lic
of
Mold
ova
Mad
agas
car
Sen
egal
Mal
i
Nic
arag
ua
Nig
eria
Sie
rra
Leo
ne
Bah
amas
Mex
ico
Ph
ilip
pin
es
Ken
ya
Pan
ama
Les
oth
o
So
uth
Afr
ica
Mal
awi
Sw
azil
and
Jam
aica
Guat
emal
a
Homicide Rate per 100,000
Note: WHO homicide statistics 2008.
Variation in Homicide Rates in the
Contemporary World
-
10,0
20,0
30,0
40,0
50,0
60,0
70,0
Icel
and
Norw
ay
Sw
itze
rlan
d
Mo
rocc
o
Slo
ven
ia
Un
ited
Kin
gdo
m
Aust
rali
a
New
Zea
lan
d
Iran
(Is
lam
ic R
epu
bli
c of)
Slo
vak
ia
Cze
ch R
epubli
c
Port
ugal
Bel
giu
m
Ser
bia
Mal
aysi
a
Bulg
aria
Tu
rkey
Pak
ista
n
Mongoli
a
Isra
el
Hai
ti
Un
ited
Sta
tes
of
Am
eric
a
Jord
an
Suri
nam
e
Rep
ub
lic
of
Mold
ova
Mad
agas
car
Sen
egal
Mal
i
Nic
arag
ua
Nig
eria
Sie
rra
Leo
ne
Bah
amas
Mex
ico
Ph
ilip
pin
es
Ken
ya
Pan
ama
Les
oth
o
So
uth
Afr
ica
Mal
awi
Sw
azil
and
Jam
aica
Guat
emal
a
Homicide Rate per 100,000
Note: WHO homicide statistics 2008.
Argentina 2012: 7.2 per 100,000
(Fleitas, 2014)
Variation in Homicide Rates in the
Contemporary World
-
10,0
20,0
30,0
40,0
50,0
60,0
70,0
Icel
and
Norw
ay
Sw
itze
rlan
d
Mo
rocc
o
Slo
ven
ia
Un
ited
Kin
gdo
m
Aust
rali
a
New
Zea
lan
d
Iran
(Is
lam
ic R
epu
bli
c of)
Slo
vak
ia
Cze
ch R
epubli
c
Port
ugal
Bel
giu
m
Ser
bia
Mal
aysi
a
Bulg
aria
Tu
rkey
Pak
ista
n
Mongoli
a
Isra
el
Hai
ti
Un
ited
Sta
tes
of
Am
eric
a
Jord
an
Suri
nam
e
Rep
ub
lic
of
Mold
ova
Mad
agas
car
Sen
egal
Mal
i
Nic
arag
ua
Nig
eria
Sie
rra
Leo
ne
Bah
amas
Mex
ico
Ph
ilip
pin
es
Ken
ya
Pan
ama
Les
oth
o
So
uth
Afr
ica
Mal
awi
Sw
azil
and
Jam
aica
Guat
emal
a
Homicide Rate per 100,000
Contemporary Rates vary between
about 0.5 (Japan, Austria, Norway)
and 50 (Jamaica, El Salvador,
Guatemala) per 100,000
inhabitants
Note: WHO homicide statistics 2008.
Map of the World by Homicides
(WHO recorded homicides 2002, www.worldmapper.org)
Map of the World by Research Output
(Published papers in 2001 www.worldmapper.org)
0,1
1
10
100
1400 1600 1800 2000
Homicide Scale
> 100 Civil War
10-100 Violent Societies
1-10 Semi-Pacified Societies
< 1 Pacified Societies
Ho
mic
ide
Rat
e p
er
10
0,0
00
0,1
1
10
100
1400 1600 1800 2000
Homicide Scale
> 100 Civil War
10-100 Violent Societies
1-10 Semi-Pacified Societies
< 1 Pacified Societies
• Unrelated men in public space • Organized and sectarian • Violent enterpreneurs
• Large % female victims • Individual pathologies • Marginal groups
Homicide as Politics
Homicide as Pathology
Ho
mic
ide
Rat
e p
er
10
0,0
00
0,1
1
10
100
1400 1600 1800 2000
Homicide Scale
> 100 Civil War
10-100 Violent Societies
1-10 Semi-Pacified Societies
< 1 Pacified Societies
Current Global HR = 6.4 (430,000 deaths p.a.)
Ho
mic
ide
Rat
e p
er
10
0,0
00
Source: UNODC (2014) Global Study on Homicide
0,1
1
10
100
1400 1600 1800 2000
Homicide Scale
> 100 Civil War
10-100 Violent Societies
1-10 Semi-Pacified Societies
< 1 Pacified Societies
Singapore
Uganda
USA
South Africa
Switzerland
Mexico
Canada Ho
mic
ide
Rat
e p
er
10
0,0
00
Honduras
Source: UNODC (2014) Global Study on Homicide
0,1
1
10
100
1400 1600 1800 2000
Homicide Scale
> 100 Civil War
10-100 Violent Societies
1-10 Semi-Pacified Societies
< 1 Pacified Societies
Singapore
Uganda
USA
South Africa
Switzerland
Mexico
Canada Ho
mic
ide
Rat
e p
er
10
0,0
00
Honduras
• Low HDI • High ethnic fractionalization • High inequality • High corruption • Low democracy • Low governance • Low social policy and health exp. • Low social trust • Low legitimacy
Lappi-Sappälä & Lehti (2014) Global Homicide Trends
Nivette(2011) Hom Studies
Nivette & Eisner (2012) Hom Studies
Sources:
Correlates of Homicide
0,1
1
10
100
1400 1600 1800 2000
Homicide Scale
> 100 Civil War
10-100 Violent Societies
1-10 Semi-Pacified Societies
< 1 Pacified Societies
Ho
mic
ide
Rat
e p
er
10
0,0
00
10 % world population 45% homicides
45 % world population 9% homicides
0
1
10
100
1400 1600 1800 2000
Homicide Scale
> 100 Civil War
10-100 Violent Societies
1-10 Semi-Pacified Societies
< 1 Pacified Societies
Ho
mic
ide
Rat
e p
er
10
0,0
00
0
1
10
100
1400 1600 1800 2000
Homicide Scale
> 100 Civil War
10-100 Violent Societies
1-10 Semi-Pacified Societies
< 1 Pacified Societies
Ho
mic
ide
Rat
e p
er
10
0,0
00
0
1
10
100
1400 1500 1600 1700 1800 1900 2000 2100
Homicide Scale
> 100 Civil War
10-100 Violent Societies
1-10 Semi-Pacified Societies
< 1 Pacified Societies
Ho
mic
ide
Rat
e p
er
10
0,0
00
Eisner (2014) Swords to Words, C & J
Source:
Homicide Trends in 8 European Regions, 15th to 21st c.
Why did homicide decline in Europe?
Why did homicide decline in Europe?
Four hypotheses
1. The long-term decline in homicide in the general population was
preceded by a pacification of the political elites.
2. The Early Modern (15th -18th century) decline in homicide was
associated with a transition from a punitive regime anchored in the
spectacle of suffering to a punitive regime focused on regulation, discipline, and imprisonment.
3. The Early Modern (15th -18th century) decline in homicide was partly
due to the frontal lobe training in self-control resulting from the
diffusion of the book and literacy.
4. The up and down of homicide across the Western World since the 1950s
is the result of cultural change related to self-control and discipline.
The Hypothesis
Any decline in homicide in the general population requires a prior
pacification of the political elites.
The Data
1. Regicide Database: Coding of violent deaths of all 1513 rulers in 45
monarchies of Europe, 600-1800. See Eisner, M. (2011). Killing Kings…, British
Journal of Criminology.
2. VIP Violent Death Database: Coding of all male entries in the Oxford
Dictionary of National Biography, 600-2000 for homicide and execution.
Elite Violence and the Decline in Homicide, 600-2013
0
10
20
30
40
50
60
70
0
500
1000
1500
2000
2500
3000
600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000
Ho
mic
ide R
ate
per
100,0
00
Eli
te K
illi
ng
Rate
per
100,0
00
Year
Regicide Elite Murder Pre-Modern Homicide Modern Homicide
Source: Eisner, M. (2014). From Swords to Words – Does Macro-Level Change in Self-Control predict Long-Term Variation in Homicide Rates.
Crime and Justice (in press).
The Hypothesis
The decline in homicide was associated with a transition
from a punitive regime anchored in the spectacle of suffering
to a punitive regime focused on regulation, discipline, and imprisonment.
The Data
Series of capital punishment in 17 European cities, 1200 – 1800, covering Italy,
Switzerland, Germany, Spain, Netherlands, England.
Retrieved from various historical publications.
17 Series of Capital Punishment and European Trend,
1200-2000
0
5
10
15
20
25
30
35
40
1200 1300 1400 1500 1600 1700 1800 1900 2000
Exe
cuti
on
s p
er 1
00,0
00
Chester Frankfurt2
Nürnberg Augsburg
Zurich Luzern
Breslau Sweden
London Danzig
Mechelen2 Antwerp
Doornik Bologna
Venice Navarra
Rome Mean Trend
Source: Eisner, M. (2014). From Swords to Words – Does Macro-Level Change in Self-Control predict Long-Term Variation in Homicide Rates.
Crime and Justice (in press).
Decline in Capital Punishment vs. Decline in Homicide
0
5
10
15
20
25
30
35
40
45
50
0
5
10
15
20
25
1200 1300 1400 1500 1600 1700 1800 1900 2000
Ho
mic
ide
Rat
per
100,0
00
Exe
cuti
on
Rat
e p
er 1
00,0
00
Executions Pre-Modern Homicide Modern Homicide Rate
Source: Eisner, M. (2014). From Swords to Words – Does Macro-Level Change in Self-Control predict Long-Term Variation in Homicide Rates.
Crime and Justice (in press).
The Hypothesis
The Early Modern (15th -18th century) decline in homicide
was partly due to the frontal lobe training in self-control
resulting from the diffusion of the book and literacy.
The Data
Number of printed books per 1000 inhabitants for 13 European regions between
1450 and 1800, by 50-year periods.
Buringh, E. and J. L. Van Zanden (2009). "Charting the “Rise of the West”:
Manuscripts and Printed Books in Europe, a long-term Perspective from the Sixth
through Eighteenth Centuries." The Journal of Economic History, 69(02): 409-445.
Jan Steen (1665) The Schoolmaster.
Book Production and Homicide Rate, 1450-1800
1
10
100
0,1 1 10 100 1000
Ho
mic
ide R
ate
Books per 1000 Inhabitants
1450-99
1500-49
1550-99
1600-49
1650-99
1700-49
1750-99
Book Production and Homicide Rate, 1450-1800
1
10
100
0,1 1 10 100 1000
Ho
mic
ide R
ate
Books per 1000 Inhabitants
1450-99
1500-49
1550-99
1600-49
1650-99
1700-49
1750-99
Log-Log –Relationship:
Homicide rates went down as book production
went up
R2 = 54%
The Hypothesis
The late 19th / early 20th century decline in
homicide was partly a result of self-control training
associated with a Victorian cultural emphasis on
domesticity, temperance, and rational recreation.
The Data
Annual time-series of alcohol consumption per capita for England (1840-2012),
Norway (1851-2002), Sweden (1861-2005 and Denmark (1892-2002) .
Victorian Christmas Card
Alcohol Consumption versus Homicide in Northern Europe
Source: Eisner, M. (2014). From Swords to Words – Does Macro-Level Change in Self-Control predict Long-Term Variation in Homicide Rates.
Crime and Justice (in press).
0
50
100
150
200
250
1840 1860 1880 1900 1920 1940 1960 1980 2000
Alcohol-Denmark
Alcohol-Norway
Alcohol-Sweden
Alcohol-England
Homicide-Denmark
Homicide-Norway
Homicide-Sweden
Homicide-England
The Hypothesis
The increase in violent crime from the early 1960s to the
90s and the decline since then reflect a move away from
and a return to cultural values of self-control and discipline.
The Data
NGRAM frequencies of nine thematic fields related to loss of self-control (sex, drugs,
narcissicm), inner control (shame, politeness, honesty), and outer control (CCTV, anger
management, zero tolerance).
Corpus of all English books in google.books, no smoothing.
Al series standardized to Mean (1950-2008) = 100.
http://chrismaverick.deviantart.com
Trends in Cultural Themes related to Discipline and
Western World Homicide Rates
0
50
100
150
200
250
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010
Ind
ex 1
950-2
008 =
100
Sex
Drugs
Narcissism
Shame
Politeness
Honesty
CCTV
Anger Management
Zero Tolerance
Homicide rates
What Drives Homicide Down? Governance and Civil Society
• Control over protection entrepreneurs
• Enforcement of ‘rule of law’, compliance with authority
• ‘Civil service ethos’
• Legitimacy (“fairness”) of social order
• Inclusive state services Rotberg (2004) When States Fail; Acemoglu and Robinson (2013)
Why Nations Fail.
Bundles of Social Control Technologies • Monitoring technologies (CCTV, Mobile) • Control over disorderly conduct (ASBOs,
Curfews, alcohol sale, hooliganism) • Early screening, victim protection
services, home visits, CD treatment • Probation & parole Finkelhor (2013), Farrell et al. (2011)
Morality and Self Control Coalitions of ‘moral entrepreneurs’ who promote health, child development, civility, self-control, respect, morality.
Proposal 1 Any sustained violence reduction will require better governance and a movement towards inclusive civil societies
Level Examples
Universal Rule of law Civil servants Anti-corruption policies Control over private protection markets Compliance with human rights State legitimacy
Selective Effective policing Victim protection services Effective complaints and internal affairs procedures
Indicated Absence of torture Respect for rights of defendant Fairness of judiciary Availability of offender treatment
Theme Legitimacy of Social Order
Background
Violence is not just an issue of individual pathology, it has systemic features influenced by, amongst others, benefits of violence, costs of violence, and beliefs about the benefits of cooperative behaviors, e.g. Eisner (2011).
What makes a social order legitimate? – E.g. Tankebe and Bottoms , Tankebe (2012)
Why do people comply with moral rules? – E.g. moral socialization, e.g. Malti et al (2009)
– developmental antecedents of legal cynicism (e.g. Nivette et al 2014)
• What systemic features of families, kindergartens, schools, care-homes, prisons, states make people more likely to follow rules?
Example 1 Salience of Risk Factors in Zurich and Montevideo
Police recorded crime Montevideo Zurich
Homicide 12.5 0.8 x 15
Robbery 1050 192 x 5
Eisner and Trajtenberg (2014) Towards a more effective prevention policy in Uruguay
Example 1 Salience of Risk Factors in Zurich and Montevideo
-1,2
-0,8
-0,4
0
0,4
0,8
1,2
Zurich higher
Montevideo higher
Police recorded crime Montevideo Zurich
Homicide 12.5 0.8 x 15
Robbery 1050 192 x 5
Standard Deviations Cohen’s d
Eisner and Trajtenberg (2014) Towards a more effective prevention policy in Uruguay
Example 1 Salience of Risk Factors in Zurich and Montevideo
-1,2
-0,8
-0,4
0
0,4
0,8
1,2
Zurich higher
Montevideo higher
Police recorded crime Montevideo Zurich
Homicide 12.5 0.8 x 15
Robbery 1050 192 x 5
Standard Deviations Cohen’s d
Eisner and Trajtenberg (2014) Towards a more effective prevention policy in Uruguay
Universal Effective supervision of children Urban Planning Elimination of physical money CCTV monitoring
Selective Control over truancy Enforcement of alcohol regulations Effective gun control
Indicated Higher likelihood of punishment Faster punishment Electronic tagging
Proposal 2 Any sustained violence reduction will rest on more intensive and effective formal and informal social control mechanisms
Example 2 Punishment for Homicide in Honduras Background Punishment (costs for norm transgressions) is important. Punishment is here to stay for the next 30 years. If the state doesn’t punish somebody else will.
There is quite some knowledge about what characteristics of punishment are more or less conducive to deterrence. E.g. Nagin (2013).
What is the average ‘tariff ’ for a murder in Honduras? - Imprisonment rate 160 per 100,000, Daniel Ortega (2014)
- Assume 50% in prison for murder
- Homicide rate 80 per 100,000
-> Average tariff is 1 year per murder committed
Example 2 Punishment for Homicide in Honduras Background Punishment (costs for norm transgressions) is important. Punishment is here to stay for the next 30 years. If the state doesn’t punish somebody else will.
There is quite some knowledge about what characteristics of punishment are more or less conducive to deterrence. E.g. Nagin (2013)
What is the average ‘tariff ’ for a murder in Honduras? - Imprisonment rate 160 per 100,000.
- Assume 50% in prison for murder
- Homicide rate 80 per 100,000
-> Average tariff is 1 year per murder committed
Today Future
Slow Fast
Severe Lenient/moderate
Unfair Fair
Rare Certain
0
1
10
100
1400 1500 1600 1700 1800 1900 2000 2100
Homicide Scale
> 100 Civil War
10-100 Violent Societies
1-10 Semi-Pacified Societies
< 1 Pacified Societies
Ho
mic
ide
Rat
e p
er
10
0,0
00
Eisner (2014) Swords to Words, C & J
Source:
Homicide Trends in 8 European Regions, 15th to 21st c.
0
1
10
100
1400 1500 1600 1700 1800 1900 2000 2100
Homicide Scale
> 100 Civil War
10-100 Violent Societies
1-10 Semi-Pacified Societies
< 1 Pacified Societies
5-7% Conviction Rate for Murder
Ho
mic
ide
Rat
e p
er
10
0,0
00
Eisner (2014) Swords to Words, C & J
Source:
Homicide Trends in 8 European Regions, 15th to 21st c.
5-7% Conviction Rate for Murder
Proposal 3 A major decline in violence will require substantial efforts by coalitions of moral entrepreneurs to change moral beliefs and sensitivities.
What is morally right for adolescents in Amman?
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Boys
Girls
OK to kill if dishonored family OK for husband to beat if….
Source: Eisner and Ghuneim (2013). Honor Killing
Background: Moral Neutralisations of harm-doing are powerful facilitators of violence, especially if supported by religious and political elites.
N = 840
What is morally right for adolescents in Amman?
39% 38% 39%
49%
77%
57% 55% 58%
87%
18% 18% 20% 25%
67%
32% 29%
13%
78%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Boys
Girls
OK to kill if dishonored family OK for husband to beat if….
Source: Eisner and Ghuneim (2013). Honor Killing
Background: Moral Neutralisations of harm-doing are powerful facilitators of violence, especially if supported by religious and political elites.
N = 840
Finally…..
Public Health • Child development • Early prevention • Social skills
Justice and Governance • Better policing • Fight corruption • Better justice • Offender treatment
Civil Society • Sectarian violence • Minorities, ethnic tensions • Transitional justice • Political violence
Public Health • Child development • Early Prevention • Social Skills
Justice and Governance • Better policing • Fight corruption • Better justice • Offender treatment
Civil Society • Sectarian violence • Minorities, ethnic tensions • Transitional justice • Political violence
Six Papers
Historical trends and patterns in homicide Eisner, Manuel (2003). Long-term trends in violent crime. Crime and Justice: A Review of Research, 30, 83-142.
Eisner, Manuel (2014). From Swords to Words. Crime and Justice, Vol 38.
Historical trends and patterns in elite violence Eisner, Manuel (2011). Killing kings; patterns of regicide in Europe, 600-1800. British Journal of Criminology,
General theory of violence Eisner, Manuel (2010). The uses of violence: An examination of some cross-cutting issues, International Journal of Conflict and Violence, 3, 40-59.
Cross-national comparison of homicide rates Nivette, Amy and Manuel Eisner (2014). Political Legitimacy and Homicide: A cross-national analysis. Homicide Studies.
Macro-level prevention of homicide Eisner, Manuel and Amy Nivette (2012). How to reduce the global homicide rate to 2 per 100,000 by 2060. in: Welsh, Brandon and Rolf Loeber (eds.). The Future of Criminology. Oxford: Oxford University Press.
• end