I would never condone kids being taken away from their family because they spanked them. That would harm a lot of children. That would be silly and achieve nothing.
Then what would be the legal consequences for spanking? Small fine similar to a parking ticket? How many tickets is too many?
The goal of such a law is to prevent it from happening rather than introducing petty fines with the expectation that the law will be ignored.
It seems that the countries who have already taken steps to ban all forms of corporal punishment have had desirable effects. The focus being on education more than prosecution.
This article is a little dated, (2002) but does give an idea of how the policy has been working in places like Sweden.
http://www.nurseryworld.co.uk/news/725230/
Laws against physically punishing children have been used to educate parents rather than prosecute them. Jackie Cosh reports
Aban on smacking children was implemented in Sweden in 1979. Since then only four children have been killed after being assaulted by an adult, and only one of these was at the hands of a parent - one child in 23 years, compared with Britain's one child a week.
I'd like to know Sweden's baseline, though. It's a powerful statistic, but it's backed up by comparing apples and oranges.
Bodie, did you ever address this? I scanned for you replying to me but didn't find it. I would legitly like to know.
Sorry, missed this.
The biggest study carried out (that i can find) is found here
http://snifferdogonline.com/reports/Corporal%20Punishment/Evaluating%20the%20Success%20of%20Sweden%27s%20Corporal%20Punishment%20Ban.pdfSuccess has been measured on five variables - public support for corporal
punishment, reporting of child physical assault, child abuse mortality, prosecution rates, and intervention by the social
authorities. Lines of best fit were generated and Cox and Stuart tests for trend were conducted.
Article pasted for Jack, to save her the trouble of asking
INTRODUCTION
IN 1979, SWEDEN became the first nation to abolish all types of corporal punishment of children
by all caretakers. This law represents the end of a series of legislative reforms spanning 50 years
which were aimed at making the rejection of corporal punishment increasingly explicit in law.
While the history of this law has been described in detail elsewhere (Durrant, 1996; Durrant &
Olsen, 1997; Newell, 1989; Ziegert, 1983, 1987), it will be summarized here to provide a context
for the analyses to follow.
HISTORY OF THE SWEDISH CORPORAL PUNISHMENT BAN
At the beginning of the twentieth century, severe corporal punishment was common in Sweden
(Sverne, 1993). However, concerns about the welfare of children began to be expressed early in the
century (see Durrant & Olsen, 1997); the first legislative reform took place in 1928 when corporal
punishment was abolished from Swedish secondary schools (
gymnasiums
).
Continued concern about the levels of violence directed toward children during the ensuing
The data collection for this study was partially supported by the University of Manitoba Research Grants Committee.
Received for publication June 8, 1998; final revision received September 23, 1998; accepted September 28, 1998.
Requests for reprints should be sent to Joan E. Durrant, Ph.D., Department of Family Studies, University of Manitoba,
Winnipeg, MB R3T 2N2, Canada.
Pergamon
Child Abuse & Neglect, Vol. 23, No. 5, pp. 435– 448, 1999
Copyright © 1999 Elsevier Science Ltd
Printed in the USA. All rights reserved
0145-2134/99/$–see front matter
435
decades led to the repeal in 1957 of the Penal Code defence for caretakers using corporal
punishment. It was intended that children would then receive the same degree of protection from
assault that adults receive. A second goal of the repeal was to clarify the grounds for criminal
prosecution of parents who physically harmed their children. In 1966, the paragraph permitting
mild forms of corporal punishment was also removed from the civil Parents’ Code
(Fo
̈ra
̈ldrabalken).
It was expected that these legislative changes, together with bans on corporal punishment in child
care institutions and reformatory schools, would convey that the use of physical force with children
was no longer acceptable. However, in the 1970s, a case of serious child physical abuse occurred
and the father who had beaten the child was acquitted by the court. Public outcry over this case,
and over child abuse in general, contributed to a decision by the Minister of Justice to appoint a
Commission on Children’s Rights with the objective of reviewing the Parents’ Code.
The Commission concluded that the guidelines for parents and legal authorities were not
sufficiently clear with regard to corporal punishment and proposed unanimously to add a paragraph
to the Parents’ Code stating explicitly that this practice was not permitted. The proposal was
supported by 28 of the 30 experts who reviewed it, all political parties, and 98% of Parliamentary
members. The proposal was put into force on July 1, 1979. The relevant paragraph in the Parents’
Code states:
Children are entitled to care, security and a good upbringing. Children are to be treated with respect for their person and
individuality and may not be subjected to physical punishment or other injurious or humiliating treatment. (Chapter 6,
Section 1)
Therefore, it is clear that the criminal law on assault applies equally to assaults of adults and
children. However, the law was not intended as a means of criminalizing caretakers; it was written
into the Parents’ Code, which carries no criminal penalties, rather than the Penal Code for this
reason. Its primary purpose was to educate, not to coerce.
Should physical chastisement meted out to a child cause bodily injury or pain which is more than of very temporary
duration it is classified as assault and is an offence punishable under the Criminal Cod
e...
although as before trivial
offences will remain unpunished, either because they cannot be classified as assault or because an action is not brought.
(Ministry of Justice, 1979)
The passage of the ban was accompanied by national distribution of a 16-page public education
brochure that was translated into all major immigrant languages. In addition, information about the
law was printed on milk cartons for 2 months in order to have information about the law present
at mealtimes so that families could discuss the issue. For further information on these and other
supportive measures (see Durrant & Olsen, 1997).
Objectives of the Swedish Corporal Punishment Ban
The ban had three primary objectives. First, it was intended to alter attitudes toward the use of
physical force with children as a first step toward eliminating its use. It was expected that the law
would produce a shift in social pressure such that a “good” parent would be seen as one who does
not use corporal punishment.
Second, the ban was intended to set a clear guideline for parents and professionals. Professionals
now could state clearly to families that physical force was not permitted and could identify at-risk
parents early in the cycle of abuse. It was also expected that Swedes would now act promptly upon
witnessing or hearing disclosures of physical harm.
Third, earlier identification was expected to result in earlier intervention. It was intended that
436 J. E. Durrant
more supportive and less coercive measures would be used than is the case where intervention
occurs only after a child has been harmed.
THE PRESENT STUDY
Purpose
Given the international significance of the Swedish corporal punishment ban, it is surprising that
no systematic studies have been conducted to evaluate social changes following its passage. The
primary purpose of the present study is to examine such trends in an attempt to determine whether
the goals of the Children’s Rights Commission have been met, that is, whether attitudes have
shifted, early identification has increased, and intervention has become less intrusive.
Hypotheses
Public attitudes toward corporal punishment.
Given the evolutionary nature of Swedish legislation
aimed at reducing the prevalence of corporal punishment, it was expected that public support for
this method would have gradually declined over recent decades.
Identification.
The corporal punishment ban was intended to increase awareness and early identi-
fication of child physical abuse. Therefore, it was expected that child assault reports would be
found to have increased following its passage and that the majority of assaults reported after 1979
would not be of a serious nature. It also was expected that the ages of adult assault suspects would
increase as greater numbers of younger Swedish parents were raised in a post-ban culture.
Corporal punishment bans now have been passed in all Nordic countries. At the same time, the
annual number of immigrants to Sweden more than doubled between 1980 and 1993 (in 1993, 90%
were from non-Nordic countries) and the annual number of refugees seeking asylum in Sweden
was 19 times greater in 1993 than it was in the early 1980s (Svenska Institutet, 1994). Such
individuals are likely to be experiencing high levels of stress and to have been raised in nations
without corporal punishment bans. It was expected that the proportion of adult assault suspects who
were raised in nations without such bans would increase over time, while the proportion of assault
suspects who were raised in Sweden or other Nordic nations would decrease over time as the bans
had their long-term effects.
Finally, it was hypothesized that earlier identification would be reflected in a decreasing rate of
fatal child physical abuse.
Intervention.
Early identification was intended to lead to earlier, more supportive intervention.
Therefore, it was hypothesized that, following the passage of the ban, prosecution rates would not
increase and that child welfare measures would become increasingly preventive.
METHOD
Data Collection
The research findings presented here are based on data collected by three methods. First, a series
of interviews was conducted in Sweden to provide an understanding of the history and implemen-
tation of the corporal punishment ban. The individuals interviewed included: (1) the former Chair
and a former member of the Children’s Rights Commission which proposed the ban; (2) the
Administrative Director of and the Legal Advisor to the Office of the Children’s Ombudsman; (3)
two prosecuting attorneys and a legal scholar; (4) the First Secretary of the Ministry of Health and
Sweden’s corporal punishment ban 437
Social Affairs; (5) the Director of the Department of Pediatrics at a University Hospital; and (6)
several child psychiatric social workers, well-baby clinic nurses, and academics working in the
areas of violence against children and children in the care of Social Services.
Second, to permit examination of trends in the variables of interest, primary data were collected
from: (1) Statistics Sweden (Statistiska Centralbyrån: SCB); (2) the National Board of Health and
Welfare (Socialstyrelsen); and (3) the National Crime Prevention Council (Brottsfo
̈rebyggande
rådet: BRÅ). The data were collected in Sweden during the spring of 1997. For each variable, data
were obtained from before or shortly after the ban’s passage until the mid-1990s. The time frames
for which reliable data were available were as follows: (1) 1965 to 1994 for public attitudes toward
corporal punishment; (2) 1984 to 1994 for identification and characteristics of assault suspects; (3)
1971 to 1996 for rates of fatal child abuse; (4) 1984 to 1994 for prosecution rates; and (5) 1982 to
1995 for child welfare measures. To ensure the accuracy of translations and interpretations of the
data, personal interviews were conducted with the Principal Research Officer at BRÅ and the
researcher responsible for the collection of child welfare statistics at SCB.
Third, interviews were conducted with several Swedish experts to obtain background informa-
tion on sociocultural shifts, legislative reforms, and criminal procedures which contextualized
trends found in the data. These individuals were: (1) the Vice-Prosecutor-General; (2) a Detective
Inspector with the National Criminal Investigation Department who investigates child abuse
reports; (3) a District Court Judge; (4) a criminologist at Stockholm University; and (5) a social
worker and child abuse expert at the National Board of Health and Welfare.
Data Analysis
Overall trends across time were examined with regard to the following variables: (1) public
attitudes toward corporal punishment; (2) reporting of assaults against children; (3) characteristics
of suspects in cases of reported assaults against children; (4) estimates of child maltreatment rates;
(5) prosecution of reported assaults against children; and (6) measures taken by the social
authorities. A line of best fit was generated for each of these variables; one-tailed Cox and Stuart
tests for trends (Conover, 1980) were conducted to determine whether their slopes were statistically
significant.
RESULTS
Public Attitudes Toward Corporal Punishment
National opinion polls have been conducted intermittently in Sweden since the mid-1960s by the
Swedish Opinion Research Institute (Svenska Institutet fo
̈r Opinionsunderso
̈kningar: SIFO) and
SCB to chart trends in support for corporal punishment. As Table 1indicates, support for corporal
punishment has declined markedly over the past 30 years (
n
5
2,
t
5
2,
p
5
.0000;
range
5
11%
Table 1. Percentage of Swedish Public Supportive of Corporal
Punishment
Year Percent Supportive
1965 53
1968 42
1971 35
1981 26
1994 11
Sources: SCB, 1996c; SIFO, 1981.
438 J. E. Durrant
to 53%,
M
5
33.4%). While in 1965, 8 years after the repeal of the Criminal Code defence,
one-half of the Swedish population believed that corporal punishment was necessary in childrear-
ing, by 1981 that proportion had decreased by 50% (SIFO, 1981). By 1994, the proportion of
Swedes supportive of corporal punishment, even in its mildest forms, had been halved once again
(SCB, 1996c).
Among the current cohorts of adults, those who support corporal punishment are three times as
likely to be over 54 years of age (18%) than under 35 years of age (6%) (SCB, 1996c). Those who
still support its use are almost three times as likely to be men (16%) as women (6%) and more than
four times as likely to have an elementary school (18%) than University level (4%) education.
Despite these group differences, only a minority of any particular group is positively inclined
toward the use of corporal punishment.
Among middle school pupils (grades 6, 7, and
, support for corporal punishment is related to
immigrant status. Those pupils who were born in Sweden or who immigrated to Sweden before
1985 are less likely to have positive attitudes (5% and 6%, respectively) than are those who
immigrated to Sweden in 1985 or later (12.5%). Therefore, the “the longer the pupil has lived in
Sweden, the less he/she accepts physical punishment of children” (SCB, 1996c, p. 9). However,
even among children who immigrated to Sweden after 1990, support is relatively low (13%).
IDENTIFICATION: REPORTING OF ASSAULTS AGAINST CHILDREN
Data are available on the rate of assaults against children that were reported to the police (crimes)
between 1981 and 1996, as well as on the proportion of children who were allegedly assaulted
(cases) between 1984 and 1996 (BRÅ, 1997; SCB 1995a). It should be noted that any reports that
are made to the police must be recorded in the official statistics; the police have no discretionary
power to not register reported crimes, no matter how minor. Also, these figures include not only
alleged assaults by parents, but by all individuals from the ages of less than 15 years to more than
50 years. These rates, therefore, refer to all assaults reported to the police prior to investigation or
prosecution.
As expected, rates (per 1,000 population) of reported crimes (
n
5
8,
t
5
8,
p
5
.0000;
range
5
.6 to 3.1,
M
5
1.5) and cases (
n
5
6,
t
5
6,
p
5
.0000;
range
5
.7 to 2.8,
M
5
1.5) have increased.
However, even at its peak, the annual proportion of children under the age of 15 who were
allegedly assaulted has not exceeded 2.8 per 1,000 population.
Types of Assaults Reported
Assault is categorized into three levels in Sweden. Aggravated assaults are serious offenses
warranting prison sentences of one to 10 years. Common assaults are of a moderate nature, carrying
a maximum prison term of 2 years. Petty assaults constitute the mildest level and are punishable
by fines.
Between 1981 and 1996, the vast majority of reported assaults against children (averaging 92%)
were petty or common offenses, suggesting that identification has been occurring before serious
injury is sustained (BRÅ, 1997; SCB 1995a). The proportion of reports that concern aggravated
assaults did not increase between 1981 and 1996 (
n
5
8,
t
5
4,
p
5
.3633;
range
5
3% to 15%,
M
5
7.50%), indicating that the seriousness of reported assaults has not increased.
Characteristics of Suspects
Limited data are available on the characteristics of suspects (SCB, 1995a); that is, alleged
perpetrators in reported cases that are legally pursued. It is important to note that the data on adult
Sweden’s corporal punishment ban 439
suspects presented here are not limited to parents. They may include relatives, strangers, baby-
sitters, or others who have contact with the child.
Ages of suspects.
It was expected that, as the proportion of the youngest adult age group who were
children when the ban was passed increased, the proportion of total suspects that they would
constitute would decrease. As expected, the proportion of adults suspected of assault against
children who were in the youngest age group (20 to 29 years) at the time of the offense declined
between 1984 and 1994 (
n
5
5,
t
5
4,
p
5
.0312;
range
5
15% to 27%,
M
5
21.3%) (SCB,
1995a). This was the case whether the alleged victim was 0 to 6 years of age (
n
5
5,
t
5
5,
p
5
.0000;
range
5
28% to 52%,
M
5
38.5%) or 7 to 14 years of age (
n
5
5,
t
5
4,
p
5
.0312;
range
5
11% to 25%,
M
5
16.5%).
In contrast, the proportion of adult suspects charged with assaults against young children who
were in their 30s (
range
5
27% to 47%,
M
5
37.4%) or 40s (
range
5
13% to 25%,
M
5
17.3%)
at the time of the alleged assault increased during the same period (
n
5
5,
t
5
4,
p
5
.0312, in both
cases) (SCB, 1995a). In the case of alleged assaults against older children, the proportion of adult
suspects who were in their 30s remained steady (
n
5
5,
t
5
2,
p
5
.5000;
range
5
30% to 41%,
M
5
36.3%), while the proportion who were in their 40s increased (
n
5
5,
t
5
4,
p
5
.0312;
range
5
25% to 38%,
M
5
31.2%).
The proportion of suspects charged with assault against young children who were 50 years of age
or older at the time of the alleged assault remained steady between 1984 and 1994 (
n
5
5,
t
5
2,
p
5
.5000;
range
5
3% to 12%,
M
5
6.9%). However, the proportion of suspects charged with
assault against older children who were in this age group declined over the same time period (
n
5
5,
t
5
4,
p
5
.0312;
range
5
11% to 23%,
M
5
15.7%).
The proportion of suspects aged 15 to 19 in cases of alleged assaults against young children
(aged 0 to 6) showed a declining, but nonsignificant, trend between 1984 and 1994 (
n
5
5,
t
5
2,
p
5
.5000;
range
5
2% to 7%,
M
5
4.2%) (SCB, 1995a). It should be noted that the annual mean
number of youth suspected of assaults against young children was 3, with a range of 0 to 5 over
the decade.
The number of youth aged 15 to 19 years who were suspected of assault against older children
increased between 1984 and 1994 (
n
5
4,
t
5
4,
p
5
.0312;
range
5
88 to 354,
M
5
166.3).
However, as a proportion of total suspects, youth involvement did not increase significantly during
that decade (
n
5
5,
t
5
2,
p
5
.5000;
range
5
34% to 48%,
M
5
40.4%).
Cultural socialization of suspects.
To test the hypothesis that individuals raised in Sweden or other
Nordic countries with corporal punishment bans would comprise a decreasing proportion of child
assault suspects, Nordic and nonNordic suspects were compared. The proportion of assault suspects
who were Swedish or Nordic declined between 1984 and 1994 (
n
5
5,
t
5
5,
p
5
.0000), while
the proportion of suspects who were nonNordic citizens increased (SCB, 1995a).
CHILD PHYSICAL ABUSE
Clearly, reporting rates are not equivalent to rates of actual assaults against children. As public
awareness of child physical abuse increases, reporting rates will increase accordingly— even if
rates of actual abuse remain steady or decline. Therefore, additional measures are needed to
determine whether the recent trends in reporting rates reflect increased public awareness of the
problem or actual increases in child physical assault.
Although the actual child physical abuse rate can never be known, it can be estimated through
an examination of child criminal deaths, which are unequivocal as indicators of violence. Criminal
child deaths in Sweden did not increase between 1974 and 1996 (
n
5
11,
t
5
3,
p
5
.8867;
range
5
440 J. E. Durrant
3to14,
M
5
8.0) and have never exceeded .009 per 1,000 children (SCB, 1976a, 1978a, 1979a,
1980a, 1981a, 1982a, 1983a, 1984a, 1985a, 1986a, 1987a, 1988a, 1990a, 1991a, 1992a, 1992b,
1993a, 1994a, 1994b, 1996a, 1997a, 1998).
It is important to note, however, that not all criminal deaths of children result from child physical
abuse. They can occur, for example, within the context of maternal postnatal depression, suicide of
the perpetrator, neglect or neonaticide (Somander & Rammer, 1991). It is those child deaths that
result specifically from physical abuse which the World Health Organization considers to most
reliably reflect the rate of child abuse in a nation (UNICEF, 1994).
Between 1971 and 1975, five children died in Sweden as a result of physical abuse during
incidents in which the caretaker’s motive was “a disciplinary measure to eliminate a disturbing
behaviour of a child without the intention to kill” (Somander & Rammer, 1991, p. 53). However,
during the ensuing 15 years (1976 to 1990), no children died in Sweden as a result of abuse (SCB,
1978a, 1979a, 1980a, 1981a, 1982a, 1983a, 1984a, 1985a, 1986a, 1987a, 1988a, 1990a, 1991a,
1992a, 1992b). Between 1990 and 1996, four children died from the effects of physical abuse; only
one of these children was killed by a parent (SCB, 1992a, 1993a, 1994a, 1994b, 1996a, 1997a,
1998) and this rate does not represent a significant increase since 1971 (
n
5
13,
t
5
3,
p
5
.9539).
INTERVENTION: CRIMINAL PROSECUTION
Prior to 1982, in the case of petty or common assaults committed on private property, legal
proceedings could only be initiated by the victim or the victim’s caretaker. (The prosecutor had the
power to prosecute in the case of serious assaults.) Thus, the prosecutor could not proceed without
the victim’s consent, which had to be explicitly given to the police investigator. Since 1982, the
prosecutor has been obliged to prosecute where there is sufficient evidence to proceed.
Decisions to Pursue Reported Assaults
When an assault report is made to the police, a preliminary investigation is carried out. The
decision to pursue it further depends primarily on whether: (1) the alleged perpetrator is at least 15
years of age; (2) adequate evidence exists; and (3) the alleged act constitutes a crime. If these
criteria are not met, the investigation is not pursued.
Between 1984 and 1994, the largest proportion of reports were not pursued and this proportion
increased during that period (
n
5
5,
t
5
4,
p
5
.0312;
range
5
22% to 34%,
M
5
24.5%) (SCB,
1995a). This increase appears to be largely due to an increasing trend in the number of reported
assaults by youth under the age of 15 (
n
5
5,
t
5
4,
p
5
.0312;
range
5
126 to 733,
M
5
335.3),
although the proportion of total reports which they comprise did not increase significantly (
n
5
5,
t
5
3,
p
5
.1875;
range
5
10% to 15%,
M
5
12.5%). When reports involving alleged perpetrators
under the age of 15 years were removed from the analysis, the proportion of reports that were not
pursued remained steady between 1984 and 1994 (
n
5
5,
t
5
3,
p
5
.1875;
range
5
13% to 24%,
M
5
15.9%).
Decisions to Prosecute Reported Assaults
When a report is pursued, four outcomes are possible: prosecution in court, summary punishment
(generally a fine levied by the prosecuting attorney), a waiver of prosecution, or no measures taken.
Summary punishments consist of day fines issued by the prosecutor, the number and size of which
are determined by the seriousness of the crime and the suspect’s income, respectively. In order to
issue a summary punishment or a waiver of prosecution, the crime must be of a minor nature and
the suspect must have confessed to its commission (Sarnecki, 1991).
The prosecution rate shows a declining, but nonsignificant, trend between 1984 and 1994 (SCB,
Sweden’s corporal punishment ban 441
1995a) (
n
5
5,
t
5
2,
p
5
.5000;
range
5
17% to 34%;
M
5
20.3%). It is possible that this trend
line is confounded by the number of reported assaults allegedly committed by individuals who are
not old enough to be prosecuted (i.e., less than 15 years of age). Therefore, an analysis of
prosecution rates was conducted on those reported crimes allegedly committed by individuals 15
years of age and older. Prosecution rates still show a declining, but nonsignificant, trend when only
those reported assaults allegedly committed by individuals old enough to be prosecuted are
included in the analysis (
n
5
5,
t
5
2,
p
5
.5000;
range
5
19% to 38%,
M
5
23.6%).
The proportion of reports ending in summary punishments remained steady between 1984 and
1994 (
n
5
5,
t
5
3,
p
5
.1875;
range
5
3% to 5%,
M
5
3.4%), while the rate of prosecution
waivers decreased (
n
5
5,
t
5
5,
p
5
.0000;
range
5
2% to 4%,
M
5
2.6%) (SCB, 1995a). This
is also the case when only those alleged offenders old enough to face prosecution are considered
(for summary punishments
n
5
5,
t
5
2,
p
5
.5000;
range
5
3% to 6%,
M
5
4.0%; for prosecution
waivers
n
5
5,
t
5
4,
p
5
.0312;
range
5
2% to 5%,
M
5
3%).
The proportion of reports which were pursued but for which no measures were taken shows an
increasing, but nonsignificant, trend between 1984 and 1994 (
n
5
5,
t
5
3,
p
5
.1875;
range
5
1% to 4%,
M
5
2.5%). This finding still obtains when only those offenders old enough to be
prosecuted are considered (
n
5
5,
t
5
3,
p
5
.1875;
range
5
1% to 5%,
M
5
2.9%). Note that
the percentages of reports that are cleared up in various ways do not sum to 100%, as many crimes
are not cleared up in the year in which the report was made (Olsson, personal communication,
February 13, 1998).
INTERVENTION: MEASURES TAKEN BY THE SOCIAL AUTHORITIES
Numbers of Children Receiving Support and Care Measures
Two indicators can be examined of the numbers of children receiving support and care measures
in a given year: (1) the number of children subject to measures at any time during that year, that
is, all those still receiving measures first implemented in previous years plus all those who received
measures for the first time during that year; and (2) the number of children who first received
measures in that year (“debutants”—these children were not subject to measures within the
previous 5 years).
As would be expected on the basis of the increase seen in rates of reported assaults against and
by children, the overall trend in the number of children subject to one or more support or care
measures in any given year showed an increasing trend between 1982 and 1995 (Socialstyrelsen,
1995, 1996). However, this increase was not statistically significant (
n
5
7,
t
5
5,
p
5
.0625;
range
5
22,707 to 28,702,
M
5
24,705.
. The mean rate of children receiving support or care
measures in a given year was 13.1 per 1,000 population in the 0- to 17-year-old age group (
range
5
12.2 to 14.6).
The overall trend for debutant children also was an increasing one between 1982 and 1995
(Socialstyrelsen, 1995, 1996). However, this increase was not statistically significant (
n
5
7,
t
5
5,
p
5
.0625). The annual average number of debutant children receiving support or care measures
was 5,692.2 (
range
5
4,528 to 6,974), yielding a mean rate of 3.0 per 1,000 population in the 0-
to 17-year-old age group (
range
5
2.4 to 3.6).
The following findings refer to the former indicator. An appropriate interpretation of these
figures requires an understanding of the types of measures that are possible in Sweden.
Types of Measures
Social services professionals are guided by two acts: (1) the Social Services Act (Socialtja
̈nst-
lagen: SoL); and (2) the Special Provisions for Care of Young People Act (Lagen med sa
̈rskilda
442 J. E. Durrant
besta
̈mmelser om vård av unga: LVU). SoL measures are voluntary, that is, they are carried out
with parental consent—while LVU measures are compulsory. Either of these types of measures can
be implemented if the home is deficient (i.e., in the case of abuse, neglect, parental substance abuse,
parental physical or mental illness) or if the child’s health and development is jeopardized by his
or her own behaviour.
It is important to note that children under the age of 15 alleged to have committed a criminal act
are always referred to social services; although they are registered in the police statistics, the
responsibility for dealing with their crimes lies with the social, not judicial, authorities. Response
to crimes committed by youth between the ages of 15 and 18 (or, in some cases, 20) is shared
between social services and the criminal justice system (Sarnecki, 1991). For those young offenders
who are successfully prosecuted, court-imposed sanctions can include care in accordance with SoL.
Therefore, it should be remembered that Swedish support and care statistics include young
offenders.
It should also be noted that the figures reported below refer to numbers of measures, not numbers
of children. Therefore, children who receive more than one measure in a given year will be counted
more than once.
Measures administered under SoL.
SoL, passed in 1982, substantially altered policies regarding
children receiving support and care measures. This Act was intended to reduce compulsion and
increase voluntary and preventive measures. For example, it eliminated supervision and introduced
the “contact person” (or “contact family”), whose role is to provide friendship and support to the
family (Gould, 1988; Sarnecki, 1991).
Together with the intended preventive function of the corporal punishment ban, it could be
expected that this Act would lead to an increase in voluntary measures and a decrease in
compulsory measures since 1982. This has been the case. Between 1982 and 1995, the proportion
of measures taken with parental consent (SoL) increased significantly (
n
5
7,
t
5
6,
p
5
.0078;
range
5
65% to 83%,
M
5
77.7%), while the proportion of measures implemented on a
compulsory basis (LVU) decreased steadily (
n
5
7,
t
5
7,
p
5
.0000;
range
5
18% to 35%,
M
5
22.4%) (Socialstyrelsen, 1995, 1996). In every year between 1982 and 1995, the majority of
measures have been of a voluntary nature; in 1995, fewer than 20% of measures were implemented
without parental consent.
SoL measures are of two types: (1) assignment of a contact person; or (2) placement of the child
outside the parental home (generally in a foster home, a children’s home operated by a county
council, a special supervisory home, a privately operated care home, or a children’s psychiatric
clinic). While out-of-home placements exceeded provision of contact persons in 1982, the reverse
was true by 1989 and this trend has continued through the mid-1990s (Socialstyrelsen, 1995, 1996).
Assignment of contact persons more than doubled between 1982 and 1995 (
n
5
7,
t
5
6,
p
5
.0078;
range
5
27% to 60%,
M
5
47.0%), while out-of-home placements were reduced by
one-third (
n
5
7,
t
5
7,
p
5
.0000;
range
5
39% to 65%,
M
5
52.1%).
Measures administered under LVU.
Under LVU, three measures are possible: (1) short-term
out-of-home care, primarily during investigations; (2) long-term or permanent out-of-home care;
and (3) assignment of a contact person, a measure available since 1985. These measures are
undertaken when the well-being of the child is seriously jeopardized.
Long-term out-of-home care declined steadily between 1982 and 1995 (
n
5
7,
t
5
7,
p
5
.0000;
range
5
75% to 90%,
M
5
82.0%) in relation to short-term out-of-home care (
n
5
7,
t
5
7,
p
5
.0000;
range
5
18% to 35%,
M
5
22.4%) (Socialstyrelsen, 1995, 1996). The assignment of contact
persons did not increase between 1985 and 1995 (
n
5
5,
t
5
2,
p
5
.5000;
range
5
0.3 to 1.1,
M
5
.8%).
Sweden’s corporal punishment ban 443
DISCUSSION
The primary purpose of the present study was to examine trends in social variables relevant to
the 1979 Swedish corporal punishment ban. The goals of the ban were to: (1) reduce public support
for corporal punishment; (2) encourage earlier identification of children at risk for physical abuse;
and (3) facilitate earlier, more supportive intervention. Data collected from a range of official
sources indicate that these objectives are being met.
Support for Corporal Punishment
Since 1965, public support for corporal punishment has declined dramatically in Sweden, from
53% to 11%. The decrease seen in support for corporal punishment may have contributed to a
decline in its use. Although no longitudinal studies of use exist, findings of a series of cross-
sectional studies suggest that corporal punishment has become less prevalent in Sweden over the
past 40 years. In the late 1950s and early 1960s, Stattin, Janson, Klackenberg-Larsson, and
Magnusson (1995) found that more than 90% of mothers in their Swedish sample had struck their
preschool-aged children. However, by 1994, only one-half of adults (aged 18 and older) reported
having received physical punishment as children from a mother or father (SCB, 1996c). Moreover,
only one-third of middle school pupils reported having received physical punishment from a
mother or father and, of these, most had experienced only its mildest forms (e.g., arm grabbing and
mild slaps). Of the children surveyed, only 3% had received harsh slaps and 1% had been hit with
objects.
In another recent study, Durrant, Broberg, and Rose-Krasnor (in press) asked Swedish mothers
of preschool-aged children how they would respond to various child misbehaviors. Only 16% of
these mothers said that they would apply physical force and, of these, none said that they would
strike the child; rather, all said that they would grab or tightly hold the child.
Together, these findings suggest that support for physical punishment has decreased dramatically
in Sweden over recent decades, that this decline has been accompanied by a reduction in its use,
and that its forms have become increasingly mild. Indeed, it appears that the very definition of mild
physical punishment has come to exclude hard slaps (SCB, 1996c), which would be considered
normative in many other countries. Therefore, the corporal punishment ban and ongoing public
education campaigns appear to have been extremely effective in altering the social climate with
regard to corporal punishment. While parental striking of children was a common occurrence 40
years ago, today this practice clearly can be termed nonnormative.
Early Identification
One of the goals of the corporal punishment ban was to increase awareness of child physical
abuse and to encourage the public to protect children at risk. Together with the public awareness
campaigns conducted in Sweden through the 1970s and 1980s, this legal change was expected to
produce an increase in child assault reports through the 1980s.
Indeed, reporting of assaults against children has increased since 1981. However, the proportions
of these reports that are pursued and prosecuted has remained steady since 1984, even when
reported assaults by alleged offenders too young to be prosecuted are excluded from the analysis
and despite a 1982 policy change making prosecution substantially easier to pursue. Together, these
findings clearly indicate that although Swedes have become increasingly aware of child physical
abuse and increasingly willing to report it, the corporal punishment ban has not resulted in greater
criminalization of caretakers.
When the characteristics of adult suspects in child assault cases are examined, further support is
found for the conclusion that the corporal punishment ban has had its desired effects. The
proportion of adult suspects that is comprised of individuals reared after the passage of the ban has
444 J. E. Durrant
decreased since 1984, as has the proportion who are citizens of Sweden or other Nordic countries.
Therefore, individuals raised within a social climate rejecting of corporal punishment are decreas-
ingly likely to become suspects in child assault cases.
Interestingly, the proportion of suspects charged with assault against older children that was
comprised of individuals aged 50 or older was also found to have decreased. This finding may
indicate that grandparents have changed their disciplinary methods in the wake of the corporal
punishment ban. However, conclusions about this group must be drawn cautiously, as it is highly
heterogeneous.
While the number of reports of youth assaults against peers has increased, the proportion of
suspects in assault cases that is comprised of youth has not increased. These findings raise the
question of whether youth are becoming increasingly violent or whether, as in the case of adults,
others are becoming increasingly willing to report assault incidents.
Several sources of data can help to answer this question. First, victimization studies, based on
the reports of young people regarding their experience of “street violence,” which are not subject
to fluctuations in enforcement, indicate that the victimization rate of young people aged 16 to 20
remained steady between 1976 and 1994 (von Hofer, 1995). These findings suggest that the
increase seen in reports of youth violence against peers may be more a function of enforcement
than of change in the behaviour of young people (Junger-Tas, 1996).
Second, the rate of homicides against young people (aged 15 to 19) remained consistently low
from 1975 to 1994 (
range
5
.2 to 1.6 per 100,000), as did the number of young people suspected
of homicide (
range
5
.8 to 3.2 per 100,000) (von Hofer, 1995). Further, the proportion of
individuals convicted of homicide who are aged 15 to 17 years did not increase between 1975 and
1996 (
n
5
11,
t
5
7,
p
5
.1133) (BRÅ, 1994, 1996, 1997, 1998; SCB, 1976b, 1977, 1978b, 1979a,
1980b, 1981b, 1982b, 1983b, 1984b, 1985b, 1986b, 1987b, 1988b, 1989, 1990b, 1991b, 1992c,
1993b, 1994c, 1995b, 1996b, 1997b). (It should be noted that the annual average number of 15- to
17-year-olds convicted of homicide was 3.3, a rate of 1 per 100,000, between 1975 and 1996.)
Finally, the proportion of individuals convicted of rape who are between the ages of 15 and 17
years decreased during the same period (
n
5
11,
t
5
10,
p
5
.0005) (BRÅ, 1994, 1996, 1997, 1998;
SCB, 1976b, 1977, 1978b, 1979a, 1980b, 1981b, 1982b, 1983b, 1984b, 1985b, 1986b, 1987b,
1988b, 1989, 1990b, 1991b, 1992c, 1993b, 1994c, 1995b, 1996b, 1997b). These findings suggest
that major violence by and toward young people has not increased since the mid-1970s.
Together, these findings suggest that the increasing trend in the number of youth assault suspects
is at least partially attributable to increased enforcement. Indeed, a recent campaign against
bullying in Sweden has resulted in school bans on all forms of aggressive behaviour; principals
now routinely report to the police any instances brought to their attention, including threats and
minor assaults (von Hofer, 1995). The police, in turn, have no discretionary power in registering
such reports; all are entered into the criminal statistics.
An analysis of seasonal trends in youth violence suggests that anti-bullying policies do contrib-
ute to the increase seen in reporting of youth assaults (Olsson, 1995). These data indicate a clear
pattern of increased reporting of assault against older children (aged 7 to 14) during the school
months and decreased reporting during vacation months. Further, approximately 60% of reported
assaults against 7- to 14-year-olds take place on weekdays; in contrast, only 20 to 30% of reported
assaults against adult men take place on weekdays (Olsson, 1995).
Therefore, as in the case of reported assaults by adults against children, increased reporting of
youth assault against peers appears to be largely due to increased awareness of the problem and is
congruent with other social changes in Sweden that have occurred over the past 150 years which
have defined virtually all forms of private and public use of force as illegitimate violence (von
Hofer, 1995). The recent campaign aimed at eliminating bullying has re-defined as assault what
was once considered to be common, even expected, behaviour among youth. And while increasing
Sweden’s corporal punishment ban 445
cultural rejection of violence has resulted in increased reporting of assaults by adults and youth,
their relative proportions have not changed since the mid-1980s.
It is possible, however, that youth assault rates have truly increased to an unknown extent. If this
is the case, could such an increase be directly attributed to the effects of the corporal punishment
ban? Clearly, numerous major social changes occur in any society over a period of almost 20 years.
In Sweden, such changes include the importation of violent videos and television programs, an
increase in the youth unemployment rate, and the erosion of social welfare programs (see Olsen,
1996). Such factors would have to be considered in any interpretation of data demonstrating an
increase in rates of youth assault in Sweden since 1979.
Estimates of Child Physical Abuse
Reports of assaults against children cannot be assumed to reflect rates of child physical abuse,
as they are highly vulnerable to shifts in public awareness and definitions of violence. Child
criminal death rates, which have remained at a constantly low rate since 1974, suggest that child
physical abuse has not increased in the wake of the corporal punishment ban. In fact, by the late
1980s, the Swedish rate of infant (under 1-year-of-age) homicide was among the lowest in the
world; this rate (.009 per 1,000 live births) reflected the criminal death of only one infant between
1985 and 1990 (Belsey, 1993).
Swedish child abuse mortality rates, a more specific indicator of the extent of physical abuse in
a nation, are internationally very low; for 15 years, no children died in Sweden as a result of abuse.
Recent fluctuations of the rate from zero reflect the death of one child in each of four years between
1990 and 1996. These findings suggest that recent trends in reporting rates reflect increased public
awareness of the problem of child abuse, rather than an actual increase in violence.
Measures Taken by the Social Authorities
Through early identification, the corporal punishment ban was intended to contribute to a
stronger emphasis on prevention of abuse. The data presented here demonstrate that since 1982,
support and care interventions have become increasingly preventive, voluntary, and supportive of
families. The proportion of measures carried out with parental consent has increased. Of these, the
proportion involving out-of-home placements has been reduced by one-third, while the proportion
involving the assignment of support persons has doubled. Of compulsory interventions, which
constitute fewer than 20% of measures, the proportion comprised of long-term measures has
decreased steadily while the proportion comprised of short-term measures has increased. The
assignment of LVU contact persons did not increase, most likely because of the serious nature of
these families’ difficulties which would warrant more drastic intervention.
Therefore, child apprehensions and other coercive measures have decreased since the passage of
the ban, while preventive and voluntary measures have increased. Of course, such a dramatic
change cannot be attributed solely to the effects of the corporal punishment ban. A radical change
in legislation, the Social Services Act (SoL), was implemented in 1982 with the explicit goal of
reducing compulsory measures. However, the corporal punishment ban needs to be viewed as one
part of a package of proactive legislative changes aimed at prevention (see Durrant & Olsen, 1997).
CONCLUSION
It is important to note that direct causal relationships between the passage of the corporal
punishment ban and the trends reported here cannot be drawn. Many social changes have occurred
in Sweden over the past 25 years, including ongoing legislative reform, demographic shifts, and
modifications to social policies. These forces have likely interacted with the attitudinal shifts
446 J. E. Durrant
engendered by the corporal punishment ban to produce the trends reported here. However direct or
indirect the route from the ban to these outcomes, it is clear that the original goals of the Swedish
corporal punishment ban have been met
ponds from now on.