The current issue and full text archive of this journal is available at

http://www.emerald-library.com

PIJPSM
23,4

416

Deviant organizational
messages among suspect
police officers in Israel
Sergio Herzog

Department of Sociology and Anthropology, University of Haifa,

Haifa, Israel
Keywords Police, Israel, Discipline
Abstract In light of criticism against the Israel National Police in recent years for excessive use
of physical force, the personal views of Israeli police officers were elicited regarding the illegal use
of force. Personal questionnaires were administered to a sample of police officers who had been
investigated for using illegal force against citizens between 1989 and 1997. Informal messages
contrary to the organization's formal messages regarding the use of force, and justifying it in
certain circumstances and for certain types of offense, seemed highly prevalent, especially among
middle-rank police officers. The results obtained provide support for the existence of a deviant
organizational subculture.

Despite the social expectation that the police force should set an example to the
public at large, police officers are often suspected and sometimes even
convicted for violation of civil rights, abuse of their powers, corruption,
manipulation of the criminal process, and violence. These offenses are defined
as ``occupational crimes'' in that they are committed by ordinary people in the
course of their routine legitimate occupational activity, either as a group or as
individuals (Robin, 1974). Among them, and basically due to both its
seriousness and sometimes irreversible consequences for the victims, the illegal
use of force by police officers has acquired special interest.
Although numerous theories have attempted to isolate variables and factors
that might account for the phenomenon (see Friedrich, 1980), to date no theory
has succeeded in explaining or predicting its occurrence in terms of one single
model. Researchers concur that the violent police officer's personality makeup
includes a personal predisposition towards aggressiveness. However, for this
potential to become manifest, interaction with external factors (situational,
organizational, or environmental) has to take place (see Worden, 1996). The
literature suggests that the main external factor influencing police violence is
organizational (Skolnick and Fyfe, 1993; Uildriks and Van Mastrigt, 1991).
According to the organizational approach, police officers normally do not
become violent all at once: police violence is best understood as a group
behavior and a developmental process. This is directed by a variety of formal
and informal occupational norms, both of the police force in general and of the
Policing: An International Journal of
Police Strategies & Management,
Vol. 23 No. 4, 2000, pp. 416-438.
# MCB University Press, 1363-951X

This study was supported by a grant from the Human Rights Center, Faculty of Law, The
Hebrew University of Jerusalem. The author would like to thank Simcha F. Landau and
Menachem Amir for their suggestions and help.

individual police officer's work group (Barker and Roebuck, 1974). Many
studies suggest that the majority of violent police officers are quite ordinary
people working under stressful occupational conditions conducive to
manifestations of violence (Cain, 1973; Chevigny, 1969; Skolnick, 1975; Uildriks
and Van Mastrigt, 1991; Westley, 1970). Notable among these conditions are
the organizational socialization of police officers, and particularly the
organizational subculture arising from informal on-the-job socialization that
sometimes encourages deviant organizational behavior, and consequently
becomes a deviant organizational subculture (Barker, 1977; Fyfe, 1993;
Gardiner, 1974; Garner, 1977; Street, 1990; Uildriks and Van Mastrigt, 1991).
The occupational socialization process and the police subculture
Beyond childhood and adolescent socialization, and like every new worker in a
working organization, the police officer undergoes a process of occupational
socialization. Through it, he/she becomes identified as an officer, and begins to
adopt a police worldview and to share perspectives relevant to the police role
(see Berg, 1990; Klofas et al., 1990; Strecher, 1967). In general, occupational
socialization is a complex process by which an organizational participant learns
the required behaviors and supportive attitudes necessary for organizational
membership. Its influence on shaping attitudes and beliefs imparts to it the
definition as a process of changing, learning, and adopting a new occupational
identity (Fielding, 1986; 1988). This process consists mainly of every worker in
every work framework learning and internalizing the behavioral norms,
required skills, and the organizational system of expectations. It usually
advances on two parallel, sometimes conflicting, levels: the formal or explicit
level and the informal level (see Reuss-Ianni, 1984). The formal level involves
the transfer of the information required by the worker to fulfill his or her
functions in a manner reflecting the formal spirit and norms of the
organization.
Essentially, the formal world is defined in abstract and general language
because its purpose is not to solve problems but to present types of problems.
Moreover, the formal level cannot cover every possible behavioral pattern that
may arise in the organizational reality. In practice, due to these limitations,
individual workers enjoy some leeway in exercising discretion on the job.
Hence an informal level develops, on which the newcomer learns the
application of the formal rules, through interaction with more experienced
veteran workers. For the new, low-rank worker, this level of socialization
represents realistic (as opposed to abstract) training for actual job performance
(Fielding, 1988; Reuss-Ianni, 1984; Van Maanen, 1973; 1975; Westley, 1970).
The police academy represents the first overt process of socializing, in which
the formal messages of police work are learned[1]. There, new officers interact
and become reinforced by others, among them more experienced officers,
eventually developing attitudes and behaviors generally consistent with those
of other police employees (Fielding, 1986; Fielding and Fielding, 1991).
Moreover, some police training programs are intentionally designed to break

Deviant
organizational
messages
417

PIJPSM
23,4

418

down the recruit, in order to build him/her up as a police officer (Klofas et al.,
1990). This stage is followed by practical training in the field under the
supervision of a veteran, middle-ranking police officer, whose main
responsibility is to teach the new officer how to put the academy's classroom
training into practice. The new officer will tend to drive and interact with
citizens in a similar manner to his/her supervisor, and even eat at the same
places. On completion of the initial training period, the rookies move on to a
probation stage as active policemen on the street. At this stage too the new
officer will modify his/her behavior and attitudes to conform with those of the
supervisor (Radelet and Carter, 1994, p. 179, for an account of the socialization
stages in the police; see also Bennett, 1984; Van Maanen, 1973).
The informal influence of both the supervisor and the other more
experienced police officers on the rookie, right from the start of his/her formal
training at the police academy, constitutes a natural and necessary process. In
its course, the trainee learns important aspects of police work that will help
him/her once on the street or during routine work, as well as informal on-thejob norms. In these stages he/she will internalize the norms of the
organizational subculture, namely the values, meanings, and behavioral
patterns that are unique to his or her new occupational group (Bahn, 1975;
Morgan, 1986; Westley, 1970). Skolnick (1966, p. 42) phrased this as the fitting
of ``cognitive lenses through which to see situations and events.'' Within the
confines of the overall or predominant culture of a society, this distinct police
subculture is seen as very important in shaping the views, attitudes, and
behaviors of police officers (Westley, 1970). Manifestations of this informal
influence will include the development of idiosyncratic language patterns
(argot) and ideologies according to which the world, people, and situations
acquire unique meanings, common only to the group's occupational members
(Beigel and Beigel, 1977; Skolnick and Fyfe, 1993).
According to Barker and Roebuck (1974), Putti et al. (1988), Savitz (1971),
and Westley (1970; 1973), as the new police officers progress in acquiring their
role, and continue their informal socialization (and consequently become
socially isolated from the community), they come closer to the values of the
police subculture at large and particularly to those of their more experienced
counterparts. This process is most salient in police work. Manning (1989, p.
363) states that as a result of the uncertain nature of police work, the police
officer is highly dependent on other officers for assistance, advice, training,
working knowledge, protection in case of threats from internal or external
sources, and insulation against the public and periodic danger. Yet while this
process is generally an essential part of the officer's socialization, it may also
produce serious ethical problems. It is in these stages that the rookie is
occasionally exposed to the message that much of his or her formal training in
the police academy is irrelevant, inefficient, and disconnected from reality
(Bayley and Bittner, 1989, p. 87; Hunt, 1985), and that his or her loyalty lies first
and foremost with the peer group and its values (Sparrow et al., 1990; Van
Maanen, 1975). Accordingly, it becomes a rigid norm in policing that one officer

does not ``rat'' on another. In practice, this means that they are increasingly
expected to tolerate, although not necessarily condone, misbehavior (violence,
corruption) by other officers, and the principle of mutual non-denunciation (the
``Blue Curtain''). Generally, messages of this type are prevalent in deviant
organizational subcultures that tolerate and even encourage a degree of deviant
behavior (Christopher Commission, 1991; Radelet and Carter, 1994; Uildriks
and Van Mastrigt, 1991). Hence, the deviant behaviors of police officers are not
merely a consequence of personality; on the contrary, many of them are likely
to be based on wide organizational subcultural perceptions that significantly
contradict the declarative formal messages of the organization (Goldstein, 1986;
Klockars, 1996; Uildriks and Van Mastrigt, 1991; Westley, 1970[2].
The Israeli experience
Unlike the American model, Israel has only one national police force, the Israel
National Police (INP), a highly centralized force, commanded and directed,
operationally and organizationally, by its Commissioner who is appointed by
the Minister for Internal Security. This national force is organized
geographically into six regions or districts and 13 sub-districts, and delivers its
services to the public from 70 local police stations, each headed by a local
commander. In 1999 the INP had 25,700 serving police officers, the great
majority (about 90 per cent) Jewish, with women constituting about 20 per cent
of the overall police workforce.
To fulfill its functions, the INP enjoys wide and extraordinary powers,
anchored in the law and police regulations. A central feature of such powers is
the authority to use a reasonable degree of physical force against citizens if
necessary. Nevertheless, recent years have seen a marked increase in media
coverage of alleged police brutality by Israeli police officers. It is claimed that
illegal physical force is employed quite routinely during arrest, in criminal
investigations, in dispersing demonstrations, and even in traffic-control
situations. Moreover, some of the allegations in no way refer to acts that might
be deemed necessary or justified police action in that they are arbitrary, and
usually performed against minorities or groups identified as ``typical criminal
offenders.'' Such cases usually occur after arrest and are likely to involve
humiliations, beatings, maltreatment, and even direct torture resulting in
physical injury. These deeds are usually defined by those in command as
isolated incidents, expressing only the violent predisposition of the individual
police officers involved.
As a result of this coverage, the subject of police violence has featured
repeatedly on the Israeli public agenda over the past two decades. Social and
political pressure has caused the attention of public committees to be turned
with great intensity to police violence (Shimron Committee, 1978; Sirota-Eitan
Committee, 1980; Kremnitzer Committee, 1994), as well as the State Comptroller
(1986), the Ministry of Police Comptroller (1993), and government and
parliamentary committees.

Deviant
organizational
messages
419

PIJPSM
23,4

420

Like the Christopher Commission, Israel's civilian comptrollers mostly

focused on the double-edged organizational messages conveyed to low-rank
police officers regarding the use of physical force in certain situations or
against specific population groups. On the one hand, the formal messages
issued by upper-rank commanders emphasized adherence to the rule of law,
impeccable conduct, and condemnation of violence. On the other, informal
messages were conveyed, especially by experienced and veteran police officers
(particularly in the middle ranks), that did not always adhere to declared policy
and prohibition of violence. Informal messages of this type defined violence as
an effective way of achieving success in the war against crime; therefore, police
officers who resorted to violence should be shielded from disciplinary or
judicial action against them. Protection of such police officers was achieved by
the development of a behavioral pattern of conscious avoidance of, or noninterest in violent incidents ``turning a blind eye'' (Sirota-Eitan Committee,
1980). This form of dissociation was reinforced by the omission of reference to
the use of physical force in written reports by police officers. When cases of
violence did come to light, internal police treatment (both on the investigative
and the judicial level) tended to be non-professional, partial, and forgiving. The
procedure tended to be prolonged, complaints were refuted, investigations were
defective, police investigators covered up for and protected investigated police
officers. Judicial verdicts were lenient, and punitive measures were ridiculously
light considering the gravity of the charges (Sirota-Eitan Committee, 1980;
State Comptroller, 1986; Ministry of Police Comptroller, 1993).
Following publication of the findings of the above reports, and the
consequent public debate, the management level of the INP adopted measures
aimed at dealing with the problem: dismissal of violent police officers,
improved procedures in the internal discipline courts, and the establishment, in
1992, of an external civilian board of officers (subordinate to the Ministry of
Justice) to investigate public complaints against police officers suspected of
using illegal force and/or criminal offenses.
The study
As mentioned above, the findings of the civilian comptrollers suggested that
deviant organizational messages of violence prevail in the Israeli police force.
These conclusions were based mainly on individual testimonies of police
officers and/or internal police treatment of suspect police officers investigated
for using illegal force. However, unlike the situation in the USA (see Weisburd
et al., 1998 as a recent example), the existence of such deviant informal
messages had yet to be confirmed by empirical examination of the personal
attitudes and perceptions of Israeli police officers regarding this and other
faulty behaviors (such as the principle of mutual non-denunciation) in the
course of their routine work. Police officers' attitudes to these patterns of work
constitute an important area of study, mainly because these attitudes usually
play a role in determining the actual behavior of police officers. Accordingly,
the study reported here was conducted, examining the issue for the first time in

Israel. Operationally, its aim was to address the following questions: Do police
officers suspected of committing offenses involving the illegal use of force hold
generally deviant attitudes about their routine use of it in police work? If
deviant attitudes exist, what are the significant independent variables
predicting them? Are they basically organizational? What are their specific
trends?
The chosen sample
To check the possible existence of deviant attitudes among police officers in
Israel, this study focused on personal perceptions of a representative random
sample of police officers personally suspected of, and investigated for,
employing violence against citizens. Theoretically, one could argue that for
valid conclusions, this sample's attitudes should be compared with those of a
control group, mainly a representative random sample of police officers against
whom illegal use-of-force files were not opened in the years covered by the
study (see Geller, 1985, p. 164). Logically, the author wholly agreed with this
methodological position, and submitted a request to the INP to draw such a
control sample. Unfortunately, the police management formally allowed
research only of the first group, and not a control. The police argument was
that the frame of this research offered the police no organizational interest to
sanction such a request. Note that because the INP is a national police force, no
study on the police can be conducted without its formal permission.
Still, despite the methodological limitations, the chosen sample did not
consist of a small proportion of violent and deviant police officers (according to
the ``rotten apple'' theory). It included police officers against whom files were
opened, not only those against whom the illegal use of force was substantiated.
Indeed the overall official proportion of violence files culminating in the
recommendation to take measures against the offending police officers between
1985 and 1997 was only about 14 per cent.
Method
In the investigation of issues of excessive force, survey methods can be used to
scan the police, as well as the public (Adams, 1996). The suspect police officers'
individual views were collected by means of an anonymous personal
questionnaire. Due to the larger size of this population (about 1,000 police
officers per year, some of them obviously appearing more than once during the
years), they were administered to a random sample of police officers
investigated for the illegal use of force (N = 600) between 1989 and 1997. The
questionnaires were sent to the respondents by mail due to their wide dispersal
throughout the country[3]. They included both closed and open-ended questions
about the issues mentioned above. The closed questions were formulated as
multiple-choice, namely respondents were asked to choose the statement nearest
to their views (quantitative findings). The open-ended questions offered the
respondents blank lines to justify/explain their previous choices (qualitative
findings).

Deviant
organizational
messages
421

PIJPSM
23,4

422

A total of 255 questionnaires were completed and returned (a response rate

of 44 per cent)[4], which is hardly surprising. First, the police officers were
under no moral or occupational obligation to complete the questionnaire.
Moreover, Israeli police officers (generally including senior commanders)
mostly perceive external researchers and academics as working against the
police interest, so they tend to avoid collaboration with them. Second, the
sampling was on a negative personal basis in that all the respondents had been
investigated for employing illegal force. Finally, the fact that the questionnaire
dealt with extremely sensitive issues may have appeared threatening to many
of them. Nevertheless, almost one half of them did respond fully.
The research sample was shown to be a fairly homogeneous group with
regard to sex and marital status: almost all the police officers were married men
(only two were women). Despite the fact that over 80 per cent of the respondents
had 12-15 years of education, less uniformity was obtained for age, length of
service, rank, and function. Mean age was 36 (range 24-52), mean length of
service was 13 years (range 1-26 years), and although almost all possible ranks
and functions in the INP fulfilling traditional police functions were represented,
about a third were concentrated in the middle ranks (senior sergeants) and about
a third in the higher ranks (officers). Regarding their specific police functions,
about a half of them performed many kinds of operational police work
(especially patrol functions), about 30 per cent investigative functions
(especially detectives), and almost 15 per cent served in the Israel Border Police.
Regarding these characteristics, the research sample seems to be
representative of the population of Israeli police officers investigated for illegal
use of force, which consequently seems to be representative of the entire Israeli
police population. As shown in the Appendix, the adequacy of the statistical
representation of some of these variables of our research sample was confirmed
by comparison with data for recent years. These had been obtained from the
civilian board and covered a wide sample of use-of-force files (over 600) (see
Herzog, 1998), and from the INP, covering the characteristics of the whole
police population in Israel[5].
Variables
Main variables
These were the respondents' answers to several questions regarding:
.
The necessity of using violence for increasing effectiveness (agree;
disagree).
.
The influence of the wish to avoid complaints about their behavior (to a
large/some degree; it is impossible to avoid complaints).
.
The identification of violent situations (possible; not possible).
.
The influence of training programs on avoiding complaints (possible;
not possible).
.
The existence of a ``code of silence'' (exists; does not exist).

.
.

Support for this code (support; opposed).

Characterization of the complainants against them (normative citizens;
offenders).

Explanatory variables
These were variables (correct for the time of completing the questionnaire)
emerging from the respondents' questionnaires, in three groups:
(1) Personal variables: three variables:
.
age;
.
education;
.
family status.
(2) Organizational variables: five variables:
.
length of service;
.
rank;
.
geographic district;
.
function;
.
branch.
(3) Features of the use-of-force complaint against him/her: four variables:
.
circumstances of the event (arrest; abuse of authority; public
disturbance; traffic offense);
.
number of accumulated complaints (including those for which the
respondent was included in the sample);
.
investigative outcomes (sustained/not sustained);
.
investigative body (police internal affairs unit; external civilian
board).
Results
Table I presents the responses to the seven questions defined as dependent
variables. As noted, the findings are presented according to the quantitative
and qualitative answers of the police officers in the questionnaires.
According to Table I, only about a half of the police officers rejected violent
means under any circumstances (almost 60 per cent if we count the police
officers who gave a conclusive answer). Justifications for the rejection of
violence focused mainly on personal experience, which had taught them that it
is possible to do a good and even effective job within the legal limitations of
reasonable force. In their opinion, resort to violence by police officers
constituted a serious breach of citizens' rights.
On the other hand, one half of all the respondents did not totally reject
violent means; about 15 per cent did not express any position on the use of
force, and more than a third supported the use of illegal force under certain

Deviant
organizational
messages
423

PIJPSM
23,4

424

Question

Answers

1. Necessity of violence for

police effectiveness

1. Agree
2. Disagree
3. Don't know
Total [N]

%
35.2
50.2
14.6
100.0

(41.2)
(58.7)
[247]

2. Influences behavior to avoid 1. To a large/some degree

complaints
2. Not possible
Total [N]

58.0
42.0
100.0

[247]

3. Ability to avoid complaints

by improving training
programs

1. To a large/some degree
2. Not possible
Total [N]

60.7
39.3
100.0

[244]

4. Ability to identify
situations in which police
violence is likely

1. Possible
2. Not possible
Total [N]

44.9
55.1
100.0

[245]

5. Perceived existence of a
code of silence among
police officers

1. Exists
2. Does not exist
3. Don't know
Total [N]

38.7
43.9
17.3
100.0

(46.8)
(53.2)

6. Support for code of silence

among police officers

1. Support
2. Opposed
3. Don't know
Total [N]

34.3
37.5
28.2
100.0

(47.7)
(52.3)

1. Pro-police normative citizens

2. Anti-police normative citizens
3. Offenders who suffered violence
4. Offenders out to slander police officers
Total [N]

2.4
16.1
13.3
68.1
100.0

7. Characterization of
complainants against police
officers

Table I.
Respondents' attitudes
to illegal use of
physical force by police
Note:
officersa

[248]

[248]

[248]

Percentages in the parentheses do not contain the ``don't know'' responses

circumstances. Moreover, these last respondents tended also to reject the

possibility of performing certain police actions without exceeding the limits of
reasonable or legal force. Three main areas of police activity were mentioned in
this context: apprehending drug offenders, dispersing violent demonstrations,
and actions (arrest, search) related to individuals defined by the police as
``known offenders.'' In such instances, these police officers claimed that ``the
street'' dictated reality; violent behavior by suspects dictated a violent reaction
by them. It was virtually impossible to successfully perform the above police
actions using ``silk gloves or polite language,'' to cite one respondent. To cite
another:
My dear sir, have you ever tried to ask a group of people for example, a group of
demonstrators to unblock a street or vacate a public park voluntarily using a megaphone?
Have you ever asked a violent criminal to accompany you to a police car, to be met by refusal
all this in front of the public, citizens, neighbors who expect you to be capable of dealing
with criminals?

The police officers were also asked whether, and to what extent their behavior
in the street was influenced by the wish to avoid complaints, especially in
situations in which this possibility was likely. Wide variability on this issue
emerged. Almost 60 per cent of the respondents testified to considerations of
this type (to a high or to some degree). The remaining 40 per cent, however,
reported that it was impossible to avoid complaints if the goal was to produce
effective results.
About 60 per cent of the respondents thought that citizens' complaints could
be avoided by improved training. Suggestions to this end focused mainly on
operational preparation and discussion of sensitive police situations liable to
lead to the submission of complaints, and on refresher courses on the law,
police regulations, and the powers vested in police officers.
Table I also shows that opinions varied with regard to the ability to identify
situations in which illegal force was likely to be exercised. More than half of all
respondents claimed that it was not possible to identify such situations. Among
the others (close to half of the sample), demonstration dispersal clearly headed
the list of such situations (about two-thirds of all these responses).
The table shows a more or less equal division of responses with regard to the
existence of a code of silence among police officers forbidding mutual
denunciation or confession in investigations against colleagues; about 17 per
cent did not express any position on the existence of such a code. Support for
such a code, to the extent that it exists, was also divided: about a third
supported it, a similar proportion did not, and more than a quarter of all
respondents failed to express a clear position.
The sampled police officers were finally asked to characterize the specific
complainants against them. Respondents generally characterized them in
extremely negative terms: almost 70 per cent stated that they were simply
``criminals out to slander them,'' and the others described them as normative
citizens with negative attitudes to the police (16.1 per cent) or as criminals who
truly believed that illegal force had been used against them (13.3 per cent).
With the seven main variables set out as dichotomized, dependent variables,
Table II presents the different logistic regression analysis, indicating the
significant logit coefficients of the explanatory variables, serving as prediction
variables, in each model.
Some pinpointed and isolated significant coefficients were found among the
different predictor variables, with regard to the dependent variables.
Nevertheless, and as predicted before, the most notable finding in Table II is
the large number of statistically significant coefficients between the
independent rank variable and the respondents' answers. Accordingly, Table
III gives the details of the statistical differences between the answers and the
three central values of the rank variable.
According to Table III, significant differences were found by rank regarding
the support of violent means for increasing police effectiveness: in the middle
ranks, only fewer than 40 per cent rejected resort to violence, as against about
three-quarters in the lower ranks and over half in the higher ranks. If the ``don't

Deviant
organizational
messages
425

PIJPSM
23,4

Question/kind
of variable
Personal

426

Organizational

Complaint

Number variables

(1) Age
(2) Education
(3) Family status

**
**

*
*

(4) Length of service

(5) Rank groupa
. Low rank
. Middle rank
(6) Geographic districtb
. Center
. Tel Aviv
. Other
(7) Functionc
. Detective
. Other
(8) Branchd
. Investigative
. Other

**
**
**
**

**
**
**
*

*
**
*
**

(9)
(10)
(11)
(12)

Circumstances
Number of complaintse
Investigative outcomes
Investigative body

**
*

*
*
**
**
**

*
**
**
**

**
**
**
**

*
*
**
*
*
**
**
*

*
*

*
*

*
*

*
**

*
*

**
*

**

Table II.
Notes:
Logistic regression for
* p < 0.05; ** p < 0.01
respondents' attitudes
a
Dummy variable: ``High rank'' is comparison (n = 92)
to illegal use of
physical force by police b Dummy variable: ``North district'' is comparison (n = 53)
c
Dummy variable: ``Patrol functions'' is comparison (n = 99)
officers (statistical
d
Dummy variable: ``Operational section'' is comparison (n = 126)
significant logit
e
Including those on account of which they were included in the sample
coefficients)

know'' answers are excluded, support for illegal physical force was especially
marked in the middle ranks (more than 60 per cent) and among high-ranking
officers (about 40 per cent).
There was considerable variability by rank on the possibility of respondents
changing their behavior to avoid complaints. More than 80 per cent of the
younger, novice, police officers stated that they attempted to modify their
behavior for this purpose. About two-thirds of the middle-ranking police
officers were of the opinion that preemptive measures of this type were not
possible, and less than half of the high-ranking officers reported modifying
their conduct to preempt complaints. From the qualitative data, we learn that
the middle-ranking officers especially considered it reasonable and even
natural that ``professional and good'' police officers involved in arrests and
other confrontational actions were likely to be exposed to public complaints.
According to this approach, an accumulation of complaints does not
necessarily indicate that the suspect police officer is violent but that he or she

Rank/question

Ranking groupsb answers

Necessity of violence
for police
effectiveness

(1) Agree
(2) Disagree
(3) Don't know
Total [N]
Statistical tests (X2)c

Influences behavior
to avoid complaints

1. Low

2. Middle

3. High

18.5 (22.0)
65.4 (77.9)
16.0
100.0 [81]
15.22**

47.3 (62.0)
28.9 (37.9)
23.7
100.0 [76]
14.33**

40.0 (42.3)
54.5 (57.6)
5.6
100.0 [90]
1.01

(1) To a large/some degree 84.8

(2) Not possible
15.2
Total [N]
100.0 [79]
34.49**
Statistical tests (X2)c

33.3
66.7
100.0 [75]
26.90**

54.9
54.1
100.0 [91]
0.54

Ability to avoid
complaints by
improving training
programs

(1) To a large/some degree 76.2

(2) Not possible
23.8
Total [N]
100.0 [80]
12.12**
Statistical tests (X2)c

41.1
58.6
100.0 [75]
16.94**

63.0
37.0
100.0 [89]
4.87*

Identifiability of
situations in which
police violence is
likely

(1) Possible
(2) Not possible
Total [N]
Statistical tests (X2)c

51.2
48.7
100.0 [80]
1.93

53.3
46.6
100.0 [75]
3.10

32.2
56.6
100.0 [90]
9.23**

Code of silence among

police officers

(1) Exists
(2) Does not exist
(3) Don't know
Total [N]
Statistical tests (X2)c
Statistical tests (X2)d

44.4 (49.3)
45.6 (50.6)
9.8
100.0 [81]
0.28
4.67*

50.0 (50.6)
48.6 (49.3)
1.3
100.0 [76]
0.69
19.62**

24.1 (38.6)
38.4 (61.4)
37.4
100.0 [91]
2.14
40.21**

Support for code of

silence among police
officers

(1) Support
(2) Opposed
(3) Don't know
Total [N]
Statistical tests (X2)c

32.1 (47.2)
35.8 (52.8)
32.1
100.0 [81]
0.25

60.5 (75.4)
19.8 (24.6)
19.7
100.0 [76]
28.45**

14.3 (20.9)
53.9 (79.1)
31.9
100.0 [91]
27.35**

Characterization of
complainants against
police officers

(1) Pro-police normative

(2) Anti-police normative
(3) Offenders who suffered
(4) Offenders out to slander
Total [N]e

2.5
11.2
17.5
68.7
100.0 [80]

2.6
21.3
12.0
64.0
100.0 [75]

2.1
16.1
10.7
71.0
100.0 [93]

Notes:
* p < 0.05; ** p < 0.01
a
Percentages in the parentheses do not contain the ``don't know'' responses
b
The three ranking groups are: low ranks (from ``raw police officer'' to ``first sergeant''
those who have not yet passed all the sergeant's ranking exams); middle ranks high
sergeants (from ``high sergeant'' to ``senior high sergeant'' ranks those who have passed the
first stage but have not yet passed the high officer's ranking exams); and high ranks (from
``sub-inspector'' and up those who have passed the second stage)
c
The statistical tests checked the difference between each column (low, middle or high rank)
vs. the others, by the two clear answers (excluding ``don't know''), for example, low vs
middle and high rank by agree/disagree (df = 1)
d
In this case, the statistical tests check the difference between each column (low, middle or
high rank) vs the others, by the two clear answers together (exist + does not exist)
vs. ``don't know''
e
No significant differences by rank were found on this item

Deviant
organizational
messages
427

Table III.
Respondents' attitudes
to illegal use of
physical force by police
officers by their ranks
(in per cent)a

PIJPSM
23,4

428

works hard and is highly motivated in the police struggle against crime. This
stance is often to be found in the personal files of these police officers: they
usually receive high evaluations, are recommended for promotion, and they
enjoy vigorous protection by their direct commanders against punishment or
dismissal.
Concerning the ability to avoid complaints by improving training programs,
middle-rank police officers were significantly more pessimistic (only about 40
per cent thought that it was possible) than the lower ranks (about threequarters) and the higher ranks (about two-thirds). In addition, among the lower
and middle ranks, more than half claimed that situations in which police
violence was likely were identifiable. Among the higher ranks only about a
third expressed this opinion.
Regarding the existence of a code of silence among Israeli police officers,
significant relationships were also found. With omission of the ``don't know''
answers, more than 60 per cent of the high-ranking police officers disclaimed
the existence of such a code. In the low and middle ranks, the proportions of
those who claimed that such a code did exist and those who disclaimed any
such notion were similar (close to 50 per cent). The most important and
significant difference was found among those who simply did not know of its
existence: only 5 per cent among the low and middle ranks, and more than a
third among high-ranking police officers. Moreover, a significant relationship
was found between the responses supporting and opposing such a code and
rank: about three-quarters of the middle ranks supported it whereas close to 80
per cent of the high-ranking officers opposed it. Among the low ranks, support
for and opposition to this code were similar.
Unlike the other examples, no significant difference was found with regard
to the relationship between the rank variable and the characterization of
complainants. Most of the police officers in every ranking group (from twothirds among the middle-ranking officers to three-quarters among the highranking officers) characterized the complainants as offenders out to slander the
respondents.
Discussion
General deviant attitudes: a deviant subculture?
As mentioned earlier, an important part of the criticism made by the Israeli
civilian comptrollers about the INP in recent years was linked to an apparent
deviant organizational subculture, which tolerated and even encouraged
violence in certain police contexts. However, unlike the research in other
countries, no empirical findings regarding personal views and opinions of
Israeli suspect police officers on these issues supported this view. Many of the
views evinced by the findings presented here are clearly not compatible with
explicit formal police regulations (Table I): the sampled police officers
expressed permissive attitudes towards illegal force in certain situations; they
were able to identify events in which illegal force was used; they acknowledged
and even supported a code of silence to protect violent police offenders and to

cover up their offenses; they expressed pessimism regarding the ability to

avoid complaints; and they characterized the complainants in negative terms.
Note here that deviant police attitudes, generally or specifically justifying
illegal use of force, identifying frequent violent events, and supporting the code
of silence among police officers are not unique to Israel. Similar findings
(although mostly in lower percentages than in the present study) were also
found in surveys conducted among police officers in other countries, mainly the
USA (see Carter, 1976; Christopher Commission, 1991; Corbett et al., 1979; Lester
and Ten Brink, 1985; Waegel, 1984; Weisburd et al., 1998; Westley, 1970).
Arguably, these deviant attitudes are the direct consequence of a small
group of deviant and violent police officers (``rotten apples''), sampled for the
study. Various counter-arguments to this view may be offered, despite the
absence of a control group for a comparison of these findings. Recall that
respondents were not police officers convicted of the use of violent means
against citizens, but suspected police officers against whom citizens had filed
complaints. Accordingly, pending formal investigation, complaints against
police officers represent only subjective claims by citizens who feel they have
suffered injustice or injury at the hands of a police officer. For most of them, the
commission of the alleged criminal acts has not been proven. In addition, as
shown in the Appendix, the sampled respondents in this research tend to
represent the large population of suspect police officers in Israel. Moreover,
findings from a sample of more than 600 use-of-force files opened in recent
years (Herzog, 1998) showed that public complaints were submitted against a
very broad group of police officers, not a small proportion of deviant police
officers who happened to have accumulated the majority of public complaints.
No uniform common denominator was identifiable in the files, and complaints
about illegal use of force seemed to occur in all areas of police activity. As in
this sample, the police officers who appeared in that larger sample formed a
heterogeneous, geographically dispersed group, across a wide range of
functions and ranks. This may indicate that the police officers personally
suspected of, and investigated for, employing violence against citizens to some
extent reflect the entire population of Israeli police officers. It may be concluded
with a high degree of certainty that the attitudes expressed in these
questionnaires do not originate in individual perceptions or failure to
internalize formal regulations by a small deviant group of police officers.
The contrary seems to be the case. The findings set out in Tables II and III
about the importance of organizational variables in these attitudes, especially
police ranks, indicate that the problem of police violence in Israel seems to be
strongly rooted in organizational factors. A deviant organizational subculture,
developed in a different manner in different ranking groups, apparently
permits deviant behaviors in certain contexts in order to achieve effective job
performance.
According to both the quantitative and qualitative data of the research, the
problem of police violence in Israel is likely to arise when the demand for
effective results, usually in the form of arrests, evidence, confessions, and

Deviant
organizational
messages
429

PIJPSM
23,4

430

convictions (for example, due to rising crime rates or an unstable internal

security situation) stands in conflict with, or even overrides, other values of
police work such as the legality of police actions. Under such conditions,
officers feel pressured to evade or bend the rules: it is usually assumed that the
police force will be more successful in its struggle against crime and
maintaining social order without legal constraints on the deployment of its
power and authority originally legislated in order to safeguard civil rights
(Morris and Heal, 1981; Sparrow et al., 1990). The pressure to obtain ``results''
may well cause police officers to decide that ``always following the rules'' is not
compatible with ``getting the job done'' (Weisburd et al., 1998, p. 54), and
therefore, to choose ``dirty'' but effective police methods, such as use of violence
(Skolnick, 1966; Skolnick and Fyfe, 1993). The dilemma presented by
effectiveness as an organizational goal on the one hand, and legality of the
means to reach the goal on the other, has been called the ``Dirty Harry
Dilemma'' (Klockars, 1985).
The importance of the rank: several police subcultures?
With regard to the variables predicting respondents' attitudes (Table II),
logistic regression analysis seemed to indicate the organizational variables as
the most significant, among them very clearly their ranking group. The
findings are evident concerning the ranks in which these deviant perceptions
seem to be more prevalent, hence more internalized (Table III). Although some
of these perceptions were expressed by relatively new police officers, and few
of them even by higher-ranking officers serving as supervisors and
commanders, they were significantly more prevalent in the study among
middle-rank police officers.
According to the present findings, and compatible with the common
conclusion of many civilian comptrollers of the police in Israel, the problem of
police violence in this country seems to stem from the double message to which
police officers are exposed. An overt formal message is disseminated,
especially by high-ranking managerial commanders, opposing the use of
excessive force under any circumstances; and a covert informal message,
emanating mainly from the middle ranks (and adopted in part by lower-rank
police officers in the field), condones resort to illegal force in certain contexts.
This hidden message is also expressed by the forgiving attitude of middle-rank
commanders to manifestations of such violence among their subordinates. This
phenomenon often stems from the belief that due process must occasionally be
sacrificed in face of the need to use aggressive and illegal methods in
maintaining social order and apprehending criminals, which are the main goals
of effective police work (Brown, 1981).
On the theoretical level, these research findings are compatible with ReussIanni's (1983; 1984) main conclusion about two different competing
organizational police subcultures (see also Bittner, 1984; Reiner, 1985). On the
one hand, the formal management-cop culture, mainly adopted by the highest
ranking police commanders, ``seeks to maximize the bureaucratic benefits of

efficient organization, rational decision making, cost-effective procedures, and

objective accountability at all levels of policing'' (Reuss-Ianni, 1984, p. 6). On the
other hand, the more informal street-cop culture, mainly adopted by middleranking police officers working in the streets, takes the ``good old days''
conception as the organizing ethos of policing. Then,

Deviant
organizational
messages

. . . being a policeman was something special: a cop put his life on the line and people
appreciated and respected his willingness to do so. As a result, policemen were allowed to do
their jobs without too many questions or too much interference from outside the department
(Reuss-Ianni, 1984, pp. 1-2).

431

Accordingly, in large police organizations (such as the Israel Police), police

subculture does not seem to be monolithic. Several ``sub-subcultures,'' so to
speak, probably exist: one at the line level, another at the supervisory and
middle management level, and yet another at the upper administrative level.
Some attitudes and values will be common to all levels, some will differ and
even conflict (Radelet and Carter, 1994, pp. 189-90). From the significant
differences in respondents' attitudes by their ranking group (Table III), and
also from the stages of the socialization process described above, it may be
concluded that the extent to which police officers conform to the occupational
subculture (i.e. legitimate and deviant organizational messages) changes in
relation to their progress in the rank hierarchy. These views (and even this
subculture) are expressed differently by different police ranking groups.
According to Lefkowitz (1974), Perrott and Taylor (1995), and Weisburd et
al. (1998), attitudinal differences among police officers can be understood by
their ranking groups: as they ascend the rank hierarchy and become
supervisors, they tend to identify more with the high commanders'
organizational attitudes and formal regulations, and consequently, less with
the informal subculture. As they reduce their daily and frustrating contact with
the most problematic groups of the public, and increase their direct contact
with the media, politicians, and community leaders, they become more
circumspect about expressing deviant attitudes, in order to protect the force.
However, the findings of the present study seem to indicate that this
``supervisors vs non-supervisors'' dichotomy is not enough to explain these
attitudinal differences among police officers. Although the various stages of the
socialization process should not be interpreted as a rigid linear model (Fielding,
1986, p. 328), through them the significant differences among the three ranking
groups may be explained, as shown in Table III, in an alternative manner (for a
similar interpretation based on police officers' age, see Ferdinand, 1980). Most
of the rookies, who are in the initial stages of their careers and still under the
influence of the formal messages, tend to express perceptions that are closer to
the formal regulations learned at the police academy (see Fielding and Fielding,
1991). However, the first period on the job may constitute a rude awakening for
many new officers, what Westley describes as ``reality shock'' (1970, pp. 159-60;
see also Fielding, 1986). If a deviant street subculture does exist in the police
force, the findings may indicate that a few of them have already adopted some
of its deviant messages. By contrast, most of the middle-ranking police officers
(specifically sergeants, some of them supervisors), who have accumulated

PIJPSM
23,4

432

expertise, length of service, and also wear, and have learned about the gap
between formal learning and reality, tend to express more ``instrumental''
perceptions of the job (see Cook, 1977), that are opposed to the formal
regulations and consequently closer to the deviant street subculture. Finally,
there are high-ranking police officers, who have not yet reached managerial
positions (but aspire to them), and have continued fulfilling traditional police
duties as commanders and supervisors. They have also undergone a tough
selection process to get to their position. They gradually identify more with the
high commanders' views, hence with the formal regulations, especially
compared with the middle ranks. A conclusion of this study is that although
different variables may be used as predictors of personal police attitudes (most
of them on the organizational level), individual views seem to be mainly
influenced by the stages of the socialization process they have gone through.
These stages are expressed in their ranking group and they also display
different intensities of identification with these messages[6].
The complainants as criminals
An additional aspect of the research is the characterization of the complainant
population in highly negative terms, the majority of them being defined as
criminals out to slander police officers personally (Table I). Unlike Butler and
Cochrane (1977), who found that experienced officers have more unfavorable
perceptions of citizens than do new recruits, this conception seems to be general
in this study, and no significant difference was found among the different
ranking groups (Table III).
Complainants against police officers are generally defined by the latter as
young aggressive males, of low socioeconomic status, without a permanent job
or unemployed, with previous convictions, and holding anti-police attitudes[7].
Evidence of this characterization is provided by two observations made by the
police. One refers to a large number of false complaints against police officers,
and the other to the simplicity of the complaint process, which allows criminals
to submit malicious complaints with the sole aim of turning the investigative
authorities' attention away from the suspect and onto the police officers (see
Lustgarten, 1986).
Regarding the situation in Israel, the accuracy of these police officers'
characterizations of the complainants was not confirmed empirically (see
Herzog, 1998). Complainants there formed an extremely heterogeneous group
with regard to variables such as age, education, family status, and occupation.
Moreover, for more than 90 per cent of them, the listed complaint was their first
ever submitted against a police officer. Moreover, reports submitted by the
Ministry of Police Comptroller (1993) and the Kremnitzer Committee (1994)
contain evidence that not only did police officers tend to dismiss the credibility
of complainants, they even submitted false complaints against suspects in the
face of actual or threatened complaints against them. In this context, based on
Sykes and Matza's (1957) ``techniques of neutralization,'' Box and Russell (1975)
and Waegel (1984) suggested that a police suspect, like any other suspect, tends

to use several techniques in order to avoid the ``deviant'' label and punishment.
Among them are ``condemnation of the condemners'' (deflection of the focus of
attention away from the offense and onto the reasons that led them to act in an
illegal manner) and ``denial of the victim'' (the denial of the existence of a victim
by his /her transformation into someone who deserves to be hurt due to his/her
affiliation with a negative group in society, for example, the criminal group).
Similarly to the general findings, this negative police characterization of
complainants was seen also as a part of deviant police subcultures, and
consequently it was directed not only against them. In many cases a great part
of the general public was seen as hostile, untrustworthy, and potentially violent
(Christopher Commission, 1991). According to Westley (1970), this subcultural
symptom mainly results from the police never meeting private citizens on
particularly good terms; most meetings are selective, with citizens who would
prefer no contact with the police. Skolnick (1966) observes that the police
officer's attitude toward work is much like that of a combat soldier: life is
combat; people are either good or bad; the situation is safe or unsafe. It is a ``us
vs them'' attitude, the former consisting of other police officers, the latter
encompassing almost everybody else. Accordingly, the subculture requires
secrecy, mutual support, and unity on the part of the police.
Conclusions
The existence of an organizational subculture is a normal and inevitable
phenomenon in all organizations. It is influenced directly by work conditions
and its messages sometimes deviate from the formal organizational messages.
In the police context, a deviant organizational subculture usually includes the
exercise of violence against certain citizens and in certain situations. Similarly
to findings for other police forces, this seems to be the source of the anomalous
attitudes expressed by our research sample. These findings, together with
evidence provided by external factors and public criticism, support the
conclusion that the illegal use of force by Israeli police officers in general, and
cover-up for such offenses by colleagues in the police department, are related to
a deviant organizational subculture of the organization.
However, these findings also show that this police subculture seems not to
be homogeneous and unique along the rank lines; on the contrary, this study
indicated that police officers greatly vary in their attitudes (and consequently
in their behavior). We should be careful not to confuse the general police
subculture common to all police officers with various police sub-subcultures
(see Worden, 1996, pp. 29-30). Like the relationship between the dominant
culture of a society and various smaller subcultures (such as that of the police),
members of these sub-subgroupings, while sharing in many of the dominant
police subculture's values, norms, and other social ideas and patterns, differ
distinctly in some manner. Because the organizational (normative and deviant)
subculture is acquired through informal socialization, and because it is
decisively influenced by the combination of proximity to the field, length of

Deviant
organizational
messages
433

PIJPSM
23,4

434

service, experience, and wear, its patterns are more internalized among middleranking police officers.
Similarly to police situations in other countries, the existence of illegitimate
organizational messages in Israel can be seen as expressions of a conflict
between two main values of the organization. These are adherence to the rule of
law, which emphasizes the safeguarding of human rights and legal constraints
on the powers vested in the police, and ``organizational effectiveness,'' which
emphasizes preservation of social order and prevention of crime by means of
professional and assertive police action. In some police departments, in which
this conflict is solved by giving preference to ``producing results'' at the expense
of upholding the principle of basic civil rights, ``effective'' means achieving the
goal in a shorter time (Klockars, 1985).
To make changes in a police subculture that permits and even encourages
resort to violence and other illegitimate means in its routine work, it is
recommended that a series of preventive organizational/general measures be
adopted that might, directly or indirectly, influence the informal organizational
subculture. Among the suggestions contained in the literature are the
following: gradual change towards community models of policing;
improvement of police recruitment criteria; reinforcement of internal
mechanisms for identifying potentially violent police officers; changes in police
training programs to incorporate the study of situations likely to provoke
police violence and avoidance of the use of illegal force; changes in
organizational evaluation criteria by emphasizing and rewarding appropriate
means rather than goal attainment per se; prevention or avoidance of situations
known to induce police violence; and application of data from the complaints
systems to identify ``problematic'' situations, functions, and units.
Formal rules and regulations, no matter how sophisticated, cannot constitute
a control mechanism merely by virtue of the threat of negative sanction that
they contain. The efficacy of a rule is measured by its capacity to be
internalized by the people whose conduct it tries to regulate. They have to be
convinced as to its justification and rationality (O'Leary and Duffee, 1971;
Stone, 1975, p. 228). It is suggested that normative conduct by police officers
can be achieved more readily if it is not merely attained through fear of
punishment (Smith and Gray, 1983). A recommended way to reach this goal is
to involve middle-rank police officers in the process of formal legislation and
organizational enforcement of the rules and regulations, and in this way to try
to reduce the perceived gap between the formal world and the frustrating and
complicated reality in the field, which produces the conditions for the
development of a deviant organizational subculture. Traditionally, as a result
of the adoption of a quasi-military model, the determination of policy and
regulations has always been perceived as a privilege of the high command,
without any participation of lower-rank commanders. In the spirit of the wide
organizational change towards new models of community policing, taking into
account the views of the latter regarding the form and content of formal
regulations is likely to increase the probability of their compliance with them
(Goldsmith, 1990).

Notes
1. It is also argued that the occupational socialization process begins before an individual
becomes a regular member of the occupation, based on his/her demands and expectations
of it (see Brim and Wheeler, 1966; Klofas et al., 1990).
2. A negative direction taken by an entire police department is exemplified in the Christopher
Commission report (1991) on the Los Angeles Police Force after the media exposure of the
Rodney King case.
3. Anonymity was especially important in this study because of the sensitive nature of the
issues examined. To enhance motivation, the questionnaires were accompanied by a
personal letter detailing the research goals, emphasizing the importance of completing the
questionnaire, assuring anonymity and that no personal identifying details would be
revealed to any judicial, academic, or police body, that the sampling method was random,
and that the findings would be used for statistical purposes only. Moreover, the
questionnaires were not designed as self-reports for respondents to describe their deviant
(and even criminal) behavior, but for the elicitation of personal views and attitudes mainly
around general issues and other police officers' behavior (see Adams, 1996, pp. 74-5).
Comprehensibility of the questionnaire was verified by a pre-test administered to a small
number of subjects.
4. About 3.5 per cent came back in the mail because the addresses to which they were sent
were incorrect.
5. The civilian board, and especially the INP, were not able to provide updating data
regarding all the relevant variables.
6. As Perrott and Taylor pointed out (1995, p. 338), a considerable difference in the police
rank hierarchy exists also between Israel and other Western countries, so any attempt to
draw exact rank equivalencies between countries is problematic.
7. Studies conducted in England (Maguire and Corbett, 1991) and in Scotland (Uildriks and
Van Mastrigt, 1991) found that this description of complainants in the case of suspected
illegal use of force conformed to the above stereotype.
References
Adams, K. (1996), ``Measuring the prevalence of police abuse of force'', in Geller, W. and Toch, H.
(Eds), Police Violence: Understanding and Controlling Police Abuse of Force, Yale
University Press, New Haven, CT.
Bahn, C. (1975), ``The psychology of police corruption: socialization of the corrupt'', Police Journal,
Vol. 48 No. 1, pp. 30-6.
Barker, T. (1977), ``Peer group support for police occupational deviance'', Criminology, Vol. 15
No. 3, pp. 353-66.
Barker, T. and Roebuck, J. (1974), ``Police corruption as a form of organizational deviance'', in
MacNamara, D. and Riedel, M. (Eds), Police: Perspectives, Problems, Prospects, Praeger,
New York, NY.
Bayley, D. and Bittner, E. (1989), ``Learning the skills of policing'', Law and Contemporary
Problems, Vol. 4, pp. 35-59.
Beigel, H. and Beigel, A. (1977), Beneath the Badge. A Story of Police Corruption, Harper and
Row, New York, NY.
Bennett, R. (1984), ``Becoming blue: a longitudinal study of police recruit occupational
socialization'', Journal of Police Science and Administration, Vol. 12, pp. 47-58.
Berg, B. (1990), ``First day at the police academy: stress reaction training as a screening out
technique'', Journal of Contemporary Criminal Justice, Vol. 6 No. 2, pp. 89-105.
Bittner, E. (1984), ``The broken badge: Reuss-Ianni and the culture of policing'', Book review of
Two Cultures of Policing, by E. Reuss-Ianni, in American Bar Foundation Research
Journal, Vol. 1, pp. 206-13.

Deviant
organizational
messages
435

PIJPSM
23,4

436

Box, S. and Russell, K. (1975), ``The politics of discredibility: disarming complaints against the
police'', Sociological Review, Vol. 23 No. 2, pp. 315-46.
Brim, O. and Wheeler, S. (1966), Socialization after Childhood, John Wiley, New York, NY.
Brown, M. (1981), Working the Street, Russell Sage, New York, NY.
Butler, A. and Cochrane, R. (1977), ``An examination of some elements of the personality of police
officers'', Journal of Police Science and Administration, Vol. 5, pp. 441-50.
Cain, M. (1973), Society and the Policeman's Role, Routledge and Kegan Paul, London.
Carter, D. (1976), ``Police brutality'', in Blumberg, A. and Niederhoffer, E. (Eds), The Ambivalent
Force, Holt, Rinehart and Winston, New York, NY.
Chevigny, P. (1969), Police Power: Police Abuse in New York City, Vintage Press, New York, NY.
Christopher Commission (Independent Commission) on the Los Angeles Police Department
(1991), Report, California Public Management Institute, Los Angeles, CA.
Cook, P. (1977), ``Facing the future'', Police Review, Vol. 11, pp. 1128-60.
Corbett, M., Meyer, F., Baker, R. and Rudoni, D. (1979), ``Conflict, consensus, and perceptions of
support among the police'', Journal of Police Science and Administration, Vol. 7, pp. 449-58.
Ferdinand, T. (1980), ``Police attitudes and police organization: some interdepartmental and
cross-cultural comparison'', Police Studies, Vol. 3 No. 3, pp. 46-60.
Fielding, N. (1986), ``Evaluating the role of training in police socialization: a British example'',
Journal of Community Psychology, Vol. 14, pp. 319-30.
Fielding, N. (1988), Joining Forces: Police Training, Socialization and Occupational Competence,
Routledge, London.
Fielding, N. and Fielding, J. (1991), ``Police attitudes to crime and punishment'', British Journal of
Criminology, Vol. 31 No. 1, pp. 39-53.
Friedrich, R. (1980), ``Police use of force: individuals, situations and organizations'', Annals of the
American Academy of Political and Social Sciences, Vol. 452, pp. 82-97.
Fyfe, J. (1993), ``The split-second syndrome and other determinants of police violence'', in
Dunham, R. and Alpert, G. (Eds), Critical Issues in Policing: Contemporary Readings,
Waveland, Prospect Heights, IL.
Gardiner, J. (1974), ``Law enforcement corruption: explanations and recommendations'', in
Sherman, L. (Ed.), Police Corruption, a Sociological Perspective, Anchor Press, New York,
NY.
Garner, G. (1977), ``Peer pressure as an influential factor in police behavior'', Law Order, Vol. 25
No. 3, pp. 40-4.
Geller, W. (1985), ``Officer restraint in the use of deadly force: the next frontier in police shooting
research'', Journal of Police Science and Administration, Vol. 13 No. 2, pp. 153-71.
Goldsmith, A. (1990), ``Taking police culture seriously: Police discretion and the limits of law'',
Policing and Society, Vol. 1 No. 2, pp. 91-114.
Goldstein, H. (1986), ``Controlling and reviewing police-citizen contacts'', in Barker, T. and
Carter, D. (Eds), Police Deviance, Anderson, CT.
Herzog, S. (1998), Police Violence in Israel: The Police Complaints System on Use of Force
Complaints, PhD dissertation, the Hebrew University of Jerusalem (in Hebrew).
Hunt, J. (1985), ``Police accounts of normal force'', Urban Life: A Journal of Ethnographic
Research, Vol. 13 No. 4, pp. 315-41.
Klockars, C. (1985), ``The Dirty Harry problem'', in Elliston, F. and Feldberg, M. (Eds), Moral
Issues in Police Work, Rowman & Allenheld, New York, NY.
Klockars, C. (1996), ``A theory of excessive force and its control'', in Geller, W. and Toch, H. (Eds),
Police Violence: Understanding and Controlling Police Abuse of Force, Yale University
Press, New Haven, CT.

Klofas, J., Stojkovic, S. and Kalinich, D. (1990), Criminal Justice Organizations: Administration
and Management, Brooks/Cole, Pacific Grove, CA.
Kremnitzer Committee (1994), Treatment of Police Violence by the System, Ministry of Police,
Jerusalem (in Hebrew).
Lefkowitz, J. (1974), ``Job attitudes of police: overall description and demographic correlates'',
Journal of Vocational Behavior, Vol. 5, pp. 221-30.
Lester, D. and Ten Brink, W. (1985), ``Police solidarity and tolerance for police misbehavior'',
Psychological Reports, Vol. 57, p. 326.
Lustgarten, L. (1986), The Governance of Police, Sweet and Maxwell, London.
Maguire, M. and Corbett, C. (1991), A Study of the Police Complaints System, HMSO, London.
Manning, P. (1989), ``The police occupational culture in Anglo-American societies'', in Bailey, W.
(Ed.), Encyclopedia of Police Science, Garland, Dallas, TX.
Ministry of Police Comptroller (1993), A Critical Report on the Systems Treatment Police Violence,
Ministry of Police, Jerusalem (in Hebrew).
Morgan, G. (1986), Images of Organization, Sage, New York, NY.
Morris, P. and Heal, K. (1981), Crime Control and the Police: A Review of Research, Home Office
Research Studies, London.
O'Leary, V. and Duffee, D. (1971), ``Correctional policy: classification of goals designed for
change'', Crime and Delinquency, Vol. 17, pp. 373-86.
Perrott, S. and Taylor, D. (1995), ``Attitudinal differences between police constables and their
superiors'', Criminal Justice and Behavior, Vol. 22 No. 3, pp. 326-39.
Putti, J., Aryee, S. and Kang, T. (1988), ``Personal values of recruits and officers in a law
enforcement agency: an exploratory study'', Journal of Police Science and Administration,
Vol. 16 No. 4, pp. 249-54.
Radelet, L. and Carter, D. (1994), The Police and the Community, Macmillan, New York, NY.
Reiner, R. (1985), Politics and the Police, St. Martin's Press, New York, NY.
Reuss-Ianni, E. (1983), Street Cops and Management Cops, Transaction Books, New Brunswick,
NJ.
Reuss-Ianni, E. (1984), Two Cultures of Policing, Transaction Books, New Brunswick, NJ.
Robin, G. (1974), ``White collar crime and employee theft'', Crime and Delinquency, Vol. 20,
pp. 251-63.
Savitz, L. (1971), ``The dimensions of police loyalty'', in Hahn, H. (Ed.), Police in Urban Society,
Sage, Beverly Hills, CA.
Shimron Committee (1978), Report of the Commission to Examine the Topic of Crime in Israel,
Ministry of Justice, Jerusalem (in Hebrew).
Sirota-Eitan Committee (1980), Treatment Measures of Police Officers for the Problem of Police
Violence, Ministry of Police, Jerusalem (in Hebrew).
Skolnick, J. (1966), Justice Without Trial, Wiley, New York, NY.
Skolnick, J. (1975), ``Why police behave the way they do'', in Skolnick, J. and Gray, T. (Eds), Police
in America, Little Brown, Boston, MA.
Skolnick, J. and Fyfe, J. (1993), Above the Law: Police and the Excessive Use of Force, Free Press,
New York, NY.
Smith, D. and Gray, J. (1983), Police and People in London, Part IV: The Police in Action, Police
Studies Institute, London.
Sparrow, M., Moore, M. and Kennedy, D. (1990), Beyond 911, A New Era for Policing, Basic
Books, New York, NY.

Deviant
organizational
messages
437

PIJPSM
23,4

438

State Comptroller (1986), State Comptroller Report Number 36, Office of the State Comptroller,
Jerusalem, pp. 544-62 (in Hebrew).
Stone, C. (1975), Where the Law Ends: The Social Control of Corporate Behaviour, Harper and
Row, New York, NY.
Strecher, V. (1967), ``When subcultures meet: police-negro relations'', in Yefsky, S. (Ed.), Science
and Technology in Law Enforcement, Thompson, Chicago, IL.
Street, R. (1990), ``A quiet revolution'', Criminal Justice, Vol. 8, pp. 1-10.
Sykes, G. and Matza, D. (1957), ``Techniques of neutralization: a theory of delinquency'',
American Sociological Review, Vol. 22 No. 6, pp. 664-70.
Uildriks, N. and Van Mastrigt, H. (1991), Policing Police Violence, Aberdeen University Press,
Aberdeen.
Van Maanen, J. (1973), ``Observations on the making of policemen'', Human Organizations,
Vol. 32 No. 4, pp. 407-18.
Van Maanen, J. (1975), ``Police socialization: a longitudinal examination of job attitudes in an
urban police department'', Administration Science Quarterly, Vol. 20, pp. 207-28.
Waegel, W. (1984), ``How police justify the use of deadly force'', Social Problems, Vol. 32, pp. 14455.
Weisburd, D., Greenspan, R., Bryant, K., Hamilton, E., Williams, H. and Olson, D. (1998), Abuse of
Police Authority in the Age of Community Policing: A Preliminary Study of Issues and
Attitudes, Police Foundation, Washington, DC.
Westley, W. (1970), Violence and the Police, MIT Press, Cambridge, MA..
Westley, W. (1973), ``Secrecy and the police'', in Neiderhoffer, A. and Blumberg, A. (Eds), The
Ambivalent Force: Perspectives on the Police, Ginn, London.
Worden, R. (1996), ``The `causes' of police brutality: theory and evidence on police use of force'', in
Geller, W. and Toch, H. (Eds), Police Violence: Understanding and Controlling Police Abuse
of Force, Yale University Press, New Haven, CT.
Appendix

Variable

Values

Mean police The suspect The present

population
sample
sample
(%)
(%)
(%)

District

Jerusalem
Northern
Southern
Central
Tel Aviv
Border police
Other
Total (N)
Operational
Investigative
Assistance-security
Border police
Administration
Other
Total (N)

8.0
16.6
10.1
9.9
14.6
19.9
20.9
100 (21,276)
31.7
21.3
7.6
19.9
12.2
7.2
100 (21,276)

Branch
Table AI.
Geographic service
district of sampled
respondents (1989-1997),
suspect police officers
and entire police
population (1993-1997)

8.4
17.8
14.8
16.9
13.8
17.6
10.6
100 (813)
41.8
29.3
9.1
17.6
1.5
0.7
100 (788)

6.3
21.0
10.3
15.2
17.0
13.2
17.0
100 (252)
50.1
28.8
5.9
13.2
1.3
0.7
100 (251)