Not all victims are equal
Assessing how characteristics of victims have an effect on the
decision to prosecute police homicides
Author: R. Bergsma
Student number: 1896237 Supervisor: dr. M. Liem Second supervisor: dr. J. Vüller Date: 13 January 2019 Word count: 17.676
1
Index
1 Introduction ... 2
2 Literature review ... 7
2.1 Theoretical framework ... 7
2.2 Legal framework on use of police force in the Netherlands ... 11
2.3 Previous research in the Netherlands on deadly incidents by police force ... 14
2.3.1 Deadly incidents ... 14
2.3.2 Prosecutions ... 17
3 Methodology ... 20
3.1 Data collection ... 20
3.2 Conceptualization of victim’s characteristics ... 24
4 Analysis ... 26
4.1 Overall findings ... 26
4.2 Victim’s characteristics ... 27
4.2.1 Physical characteristics ... 27
4.1.2 Situational characteristics ... 35
4.1.3 Prosecution’s decision and article 12 procedure ... 43
5 Conclusions and discussion ... 45
5.1 Conclusions ... 45
5.1.1 Physical characteristics ... 45
5.1.2 Situational characteristics ... 50
5.1.3 Victim’s characteristics and Black’s Theory of Law ... 52
5.2 Discussion ... 52
5.2.1 Future research ... 54
5.3 Law proposal for a specific criminal offence for police officers... 55
6 Bibliography ... 57
Appendix ... 66
1 Overall statistics ... 66
2 Statistics on police related traffic accidents ... 70
2
1 Introduction
States own a monopoly on violence. Therefore, the police are allowed to use force against civilians if a certain situation requires this.1 Each state deals with this monopoly on violence in a different manner. Police officers are trained differently since they face different types of crime. Since, for example, the police in the United States will face more violence including firearms, they use lethal force more easily than the police in the European Union.2 On the other hand, the judicial system differs globally on how to deal with police officers that use deathly force. Whereas self-defence is a common ground to use lethal force, in the Netherlands the police are allowed to even use their firearm to arrest a perpetrator.3 The use of deadly force causes public attention and leads to public debates on how to deal with these types of killings. In the United States for example, the debate is heavily affected by discrimination, since more Afro-American people are the victim of deadly use of force than so-called ‘white’ Americans. The ‘Black Lives Matter’-demonstrations took off in 2016 after a neighbourhood watch was acquitted of the murder of an unarmed 17-year old boy.4 The demonstrations got worldwide attention after the killing of Michael Brown by an American police officer.5
The debate on how to deal with deadly use of force by police officers is globally spread. Much attention is given to this topic in the Netherlands too. For example the death of Mitch Henriques in June 2015 led to protests in The Hague and a large criminal trial.6 He died after the police tried to get him under control in order to escort him away. Henriquez was on a festival and screamed that he had a gun. When he did not stop screaming, while the police asked him to, they used force against him. His death caused a public debate on the use of the so-called neck-clamp and that multiple police officers used excessive force to control Henriquez.7 Another example that caused a lot of public attention is the death of the 17-year old Rishi Chandrikasing, who was killed by a police bullet. The policemen thought Rishi had a gun, but in the end it turned out that he was unarmed.8 In these two cases causes of death differed; whereas Rishi was killed by a police bullet, Henriques died after the police placed an arm around his neck to restrain him.
1 ‘State monopoly on violence’, Encyclopaedia Brittanica Inc. 6 March 2013. 2 Page, J. (30 October 2018).
3 https://www.politie.nl/themas/geweldsinzet.html#alinea-title-wanneer-mag-een-politieagent-schieten. 4 Grierson, Swaine & Laughland (8 July 2016).
5 Van Outeren, E. (4 November 2016).
6 Kievit, A. (30 June 2015); Klumpenaar, S. (4 July 2015).
7 Rb. Den Haag 21 december 2017, ECLI:NL:RBDHA:2017:15095.
3
The public debates on deaths by police force repeatedly focus on the question whether the police use more violence against minorities or people of a lower social status and furthermore about whether the public prosecutor dismisses those cases more easily.9 In the Netherlands, more than ninety percent of cases in which the police used deadly force are dismissed.10 Yet it is still unknown what factors contribute to the decision to prosecute. Therefore, this thesis will look into the last question to find out whether indeed the characteristics of the victim affect the decision for prosecution.
Victim’s characteristics could affect the decision for prosecution, as the Theory of Law by Black explains.11 Black argues that law is a quantitative variable. By counting certain characteristics of the victim, such as wealth, culture and social control, it is possible to see which cases are more likely to end up in the criminal justice system and which cases will more likely be dismissed. In other words: which cases have more law and which have less law.12 More or less law is a relative measurement, decided by comparing cases to each other. For example, an arrest warrant is more law than no warrant and a lawsuit is more law than an arrest warrant.
Previous research looked into the numbers of deadly shootings by police officers and victims that were under the responsibility of the police.13 However, they did not look into the prosecution’s decision and certainly not at the effect victim characteristics could have on the aftermath of a deadly incident. Earlier Dutch research on police use of force recommends looking into the factors that could affect the prosecution’s decision.14 This is relevant since all deadly incidents are investigated by the National Police Internal Investigations Department [Rijksrecherche], but only a small percentage leads to a criminal trial. In order to overcome any form of bias it is important to look at whether such bias exists and how it relates to the decisions of public prosecutors.
Additionally, this research could attribute to the discourse on the new proposed law on the criminal justice response to the excessive use of police force.15 This proposal of the Minister of Justice and Security aims to create a new legal framework to assess the use of force by police
9 D. Smithuijsen (24 September 2017); Kievit, A. (30 June 2015); Klumpenaar, S. (4 July 2015). 10 Van der Steeg, M. (2016), p. 2-3; Naeyé, J. (2005), p. 426-427.
11 Black, D. (1976). 12 Black, D. (1976).
13 Blaauw, E., Kerkhof, A. D., & Vermunt, R. (1997); Timmer, J.S. (2005); Thoonen, E. & Duijst, W. (2015); Van
der Steeg, M. (2016).
14 Timmer, J.S. (2005).
4
officers. The minister proposes three new regulations in order to create a special status for police officers within the Dutch Criminal Code. This special status is necessary, according to the Minister of Justice and Security, because police officers are often obliged to use force when civilians would step aside. Police officers are therefore more vulnerable in the criminal justice system. They easily fulfil all requirements of a punishable offence, but according to the minster it would be unfair to treat them like suspects every time an officer uses force against a civilian. This would disavow their special position in which they are part of the state monopoly of violence.16 Harmon endorses this special position as she explains that police officers act under threat when they use force, so it is not a normal, prepared form of state coercion or punishment. On the other hand, police officers are trained in repressing violence, so it is not a normal or pure form of self-defence.17
The new proposal entails (1) a proposal for a new punishment exclusion ground. This new ground would explicitly say that police officers are not punishable if they legitimately use force in the course of their profession. It will be applicable to prosecutions based on the articles of manslaughter or (aggravated) assault.18 Additionally he proposes (2) the creation of a new criminal offence, targeting civil servants. The criminal offence would be the culpable violation of the Instructions on violence [Geweldsinstructie] placed in new article 372 Dutch Criminal Code. These Instructions on violence are the general binding regulations on the use of force. The offence includes a causal relation, since there has to be causality between the violation of the instructions and the occurred consequences.19 The last proposal (3) creates a new legal framework for research on the incident. The main object of this last proposal is to only treat a police officer as a suspect when it is immediately clear that he acted in violation with the Instructions on violence.20
It is striking that this proposal it does not look at the possibility of getting acts of violence in a disciplinary system for the police, as proposed by Van der Steeg and Timmer.21 Such a disciplinary system could form an alternative for a criminal procedure. Criminal law is an
ultimum remedium and should therefore be the last option. A police officer who uses force
against a civilian needs to have a special place within the judicial system since he are part of
16 Kamerstukken II 2016-17, 34641, nr. 3, p. 1-5. 17 Harmon, R.A. (2008), p. 3. 18 Kamerstukken II 2016-17, 34641, nr. 3, p. 6. 19 Kamerstukken II 2016-17, 34641, nr. 3, p. 14. 20 Kamerstukken II 2016-17, 34641, nr. 3, p. 7-9. 21 Van der Steeg, M. & Timmer, J.S. (2016).
5
the state monopoly on violence. This research forms a contribution to the understanding of deadly police force and the judicial response.
By examining whether the characteristics of the victim have any effect on the use of force and especially on the question of how to deal with those violent acts it is possible to overcome any form of bias in the process. If a correlation between those characteristics and the prosecution decision exists, it will be necessary to investigate whether abovementioned proposal of law will form an addition to the current legal system, like the minister of Justice and Security proposed or that the bias should be removed from the system in a different manner.
This thesis will be part of the research project into the flow of homicide through the system, as conducted by Leiden University. The overall research focuses on the gradient of the homicide flow, which contains six steps and seven stages. These stages are: suspicious death, investigation, clearance, arrest, prosecution, conviction and imprisonment. Each step knows a ‘flow out’ of the system, which means that the total amount of cases decreases with each step. Some cases will flow out because they are solved or do not meet the standards of the step they are in. However, some cases will flow out for unknown reasons. The question is whether this loss in every step is the effect of any form of bias in the process. By examining this bias, the loss can be explained and further research could possibly prevent the loss.
This research particularly looks at the step between investigation and prosecution, and it will focus solely to the cases of deadly police force. The research gains more insight whether victim’s characteristics affect the prosecution’s decision. In addition to the other research related to the flow of homicide, this study will give a perspective on legal killings and the prosecution’s decision regarding those cases. More information can be retrieved on how certain decisions are formulated and what role a victim has in the decision-making process. Additionally, this research supplements the Dutch Homicide Monitor, since most police killings are not included in this database.
All violent acts of police officers are investigated by either the chief of police or the National Police Internal Investigations Department however, only a small percentage leads to criminal prosecution. Both the small percentage leading to criminal prosecution as well as the role of victim’s characteristics need further examination. To answer this question, a statistical analysis of the cases is required.
6
This leads to the following research question:
How can Black’s Theory of Law explain which deadly incidents - as a result of the use of force by the police - will be addressed in the criminal justice system based on the period between 1992 and 2016, when focusing on victim’s characteristics?
Since three people get killed on average by a police officer on duty annually, all deaths caused by police officer’s actions between 1992 and 2016 are examined.22 This wide timeframe is needed to establish enough cases that are valid for proper analysis, which results in relevant conclusions. A registration system for homicides was not available in the Netherlands before 1992, which is reason why the Dutch Homicide Monitor does not contain any homicide information prior to that date.23 The database is up to date until 2016. Therefore, the research of this thesis is restricted to the period between 1992 and 2016.
Although, not all cases will be included in the Dutch Homicide Monitor since most research into deadly use of force by police ends in a dismissal of the case.24 Therefore additional open source research is needed in publications of the public prosecutor, the National Police Internal Investigations Department and newspaper articles.
It is important to see that in the period between 1992 and 2016 there seems to be an increase in cases that get a criminal justice response.25 The question is whether and how this increase can be explained by Black’s Theory of Law and whether the characteristics of the victim affect the decision-making of the public prosecutor whether to prosecute or dismiss a case.
This thesis will first, in chapter 2, address the existing literature and academic knowledge on Black’s Theory of Law and research into police killings. Chapter 3 will elaborate on the methodology and explain the use of the Dutch Homicide Monitor. Chapter 4 contains the analysis of the data, and the conclusions thereof can be found in chapter 5, which also contains the discussion on the possible defects in this research and recommendations for further investigation.
22 Timmer, J.S. & Cozijn, V.M. (2016), p. 308.
23 https://www.universiteitleiden.nl/en/research/research-projects/law/homicide (latest checked on
29-07-2018).
24 Van der Steeg, M. (2016), p. 2-3; Naeyé, J. (2005), p. 426-427. 25 Van der Steeg, M. (2016).
7
2 Literature review
2.1 Theoretical framework
In order to examine killings by police force properly, Black’s Theory of Law will be the focal point of this thesis.26 Black’s Theory of Law relates to the assumption in which Black generalizes the behaviour of governmental social control, which he explains to be law.27 Black argues that law is a quantitative variable, which can be measured in several ways and increases and decreases in various situations. That means that law contains an ‘amount’, which lies in a continuum from zero (no law) to infinite positive values. For example an arrest warrant is more law than no arrest warrant, likewise, a lawsuit is more law than the arrest warrant. When a perpetrator gets arrested law increases for the person concerned. On the other hand it decreases if a defendant wins a case in court. The severity of a prosecution of a major crime or its adjudication is more law than that of a minor crime. Each step in the judicial process shows, in principle, an increase in law. Black states that the quantity of law is constructed by obligations, prohibitions, legislation, litigation and adjudication. Different factors of social life therefore affect the existence and extent of law. Black categorized these factors as stratification,
morphology, culture, organization and social control.28 Each will be described below.
Stratification sees differences in wealth and the way it is divided. Black calls this
vertical distance. He argues that stratification and law vary directly from one another. Therefore, a society has more law when more inequality in wealth exists. Black explains that people of the same rank within society are more likely to settle a conflict together without law than people from different ranks. Taking this into account, downward law, which means a complaint from someone of a higher rank than the offender, is more law than the other way around.29 The police investigate more crimes committed by people without an income or who receive social benefits than wealthier people that do earn an income.30 This research will therefore look into the social status of the victims of deadly police force.
Additionally, stratification addresses various styles of law; downward law is more likely to be penal, whilst upward law is more likely to be compensatory or therapeutic. If an offender is of lower rank than the victim it is more likely that he will be punished for his crime. If the
26 Black, D. (1976). 27 Black, D. (1976), p. 2. 28 Black, D. (1976), p. 3-4. 29 Black, D. (1976), p. 11-25. 30 https://www.cbs.nl/nl-nl/nieuws/2018/14/bij-drugsmisdrijf-vaker-verdachte-zonder-inkomen.
8
offender is of higher rank it is more likely that he will compensate the damage or will be treated as if he is in need of medical treatment. Black says that a rich man is less likely to be punished as a criminal than a poor man. However, when the offender and the victim are of equal rank they will most likely aim for a compromise.31 Since the offenders in this research are all police officers it is important to look at the characteristics of the victim, e.g. was the victim wealthy or not and does this significantly relate to the prosecution’s decision.
Research shows that police officers are more likely to arrest a person of lower status who was in conflict with someone of different status, than someone of lower status who was in conflict with someone of the same status. Therefore the police are more likely to use coercive power against those lower status persons than against higher placed ones. Minorities seem to be more susceptible to police related violence.32 The higher the inequality in American states, the higher the use of deadly force by police, which confirms the above mentioned conclusion.33
This explanation of Black’s Theory of Law asserts that cases of younger victims receive less law than cases of older victims since younger people have a lower social rank than older people.34 It also assumes that cases of younger victims are less likely to lead to a prosecution than cases of older victims. In 2002 Litwin examined whether cases of younger victims were less likely to be cleared. He did not find any support for this hypothesis that was retracted from Black’s Theory of Law. His study showed the exact opposite relation between the age of the victim and the clearance rate of homicide cases.35
Following the aforementioned approach, the behaviour of the citizen is less important than its characteristics such as wealth, age and social background in police encounters. This hypothesis cannot be supported since the nature and circumstances of the offences seem as important as the characteristics of the victim and the offender.36 Behaviour of the citizen is of decisive importance, as is shown in Norris, Birkbeck and Gálbadon’s research on the Mexican, American and Venezuelan police.37 If a citizen’s behaviour was aggressive, the police was more likely to use force than with a cooperative citizen. Since Norris et al. have been unable to explain stratification in these cases they suggest looking into the concept of organisation, which will be explained below. They found differences in the use of force between police stations in Mexico,
31 Black, D. (1976), p. 29.
32 Worden, R.E. (2015), p. 150-151. 33 Jacobs, D. & Britt, D. (1979).
34 Chappell, A. T., & Maggard, S. R. (2007), p. 267. 35 Litwin, K. J. (2002).
36 Schulenberg, J.L. (2010), p. 110.
9
America and Venezuela.38 On the other hand, numbers show that large communities with a high level of inequality point towards a higher amount of deaths by the use of police force.39 Since the results of research into the effect of stratification on the use of force and the prosecution’s decision show differences, investigation is needed whether stratification affects the use of deadly police force in the Netherlands.
Whereas stratification is a vertical concept, morphology is a horizontal one. Morphology is the so-called distribution of relations between people, more specifically the concept of differentiation. More differentiated societies experience more law, whereas more close knitted societies know less law. The differentiation does not only look at differences between people, but also at the distance between people. Morphology has a curvilinear relationship with law. Law increases as the distance between people increases until it reaches the point at which people live in separate worlds. Living in separate worlds means that people will never meet each other randomly. Different worlds are likely to have different cultures and conflicts will therefore be in a social vacuum. Law will decrease from this point, while morphology still increases.40
Research from 1995 shows that the closer the relation between offender and victim, the less likely it is that the law intervenes. Buzawa examined the discrepancy in arrests in assault cases in which the offender and victim were spouses and in cases in which the offender and victim were strangers. Fewer arrests were made in domestic cases than in assault cases with strangers.41 Similar results were found by Fyfe, which seems to endorse Black’s hypothesis on morphology.42 Since the relation between the police officers and citizens will be the same in all cases in this study, the aspect of morphology will not be part of this research.
Culture is a symbolic aspect according to Black. It directly varies with law, therefore it
grows with it. Culture includes science, religion, technology and the more aesthetic things like art, poetry, painting and clothing. It is a wide concept that includes all ideas about the nature of reality. If the culture is rich, there is more law. Rich culture contains for example lots of folklore, many traditions and a variety in languages. Black explains sparse law by referring to tribal people. They only have one language, all look alike and eat the same food almost every day. There is not a lot of variety in sparse law and therefore, less culture and less law. Societies that have several subcultures, different food habits and technological advancement have a rich
38 Norris, M., Birkbeck, C., & Gabaldón, L. G. (2006), p. 325. 39 Jennings, J.T. & Rubado, M.E. (2017), p. 232.
40 Black, D. (1976), p. 37-39.
41 Buzawa, E., Austin, T. and Buzawa C.G. (1995), p. 460. 42 Fyfe, J.J., Klinger, D.A. & Flavin, J.M. (1997).
10
culture and therefore more law. Elaborating on this further, law varies with literacy and education. Black argues that conflicts between more cultured people are more serious than conflicts between less cultured people. He explains this regarding two types of cultural direction. Law is greater when it goes towards less conventionality than towards more. The distance in culture is constructed by the quantity of culture and by the frequency or conventionality.43 Cultural variables are part of this study.
Organization on the other hand is a more corporate aspect of social life. It looks at the
presence of administration, rules and the centralization of decision making. Law increases when the organization of society increases, therefore it varies directly with each other. If people are forced to act individually there is less organization and law decreases. Organizational distance is important, which is constructed by members of the organization. Groups have more law than individuals that are not part of any organization. An assault on a group is therefore more important than on an individual and a complaint of a group against an individual is more serious than vice versa.44
Smith examined whether the organizational factors of police stations have any effect on the use of deadly force by its police officers he did, however, not find concrete indications to support this. Actually, a greater amount of rules does not seem to affect the amount of the use of deadly force by police officers. On the other hand, measures of accountability did show an effect. Bureaucratic control decreased the overall number of killings by the police.45 The number of deaths by use of force increase when a police department lacks organization and control. As a result of lack of command, control and clear communication officers are less accountable in a certain situation.46 Since the Netherlands has a national police, the organization will be similar in the analysed cases. This factor will therefore not be part of the research.
Another important concept is that of social control. Black explains that in places with less social control by for example family or neighbours, it is more likely that law intervenes more often in social life.47 Schulenberg endorses this hypothesis when looking into the acts of police against youths. When the informal social control of parental responses are high the police are less likely to interfere in comparison to situations that lack parental involvement.48
43 Black, D. (1976), p. 61-79. 44 Black, D. (1976), p. 85-103. 45 Smith, B.W. (2004).
46 Lee, H. & Vaughn, M.S. (2010); Jennings, J.T. & Rubado, M.E. (2017). 47 Black, D. (1976), p. 5-6.
11
Black distinguishes four types of social control, which could either be exercised by the police or other forms of social control: penal, compensatory, therapeutic and conciliatory. When a group or society acts against an offender, they will most likely punish him for these acts. This punishment is the penal form of social control. Compensatory on the other hand is based on the victim’s initiative. The offender owns the victim a debt. Therapeutic and conciliatory are both remedial styles of control in which the needs of the offender are central. Therapeutic control tries to effectuate normality in order to help the offender to get re-enter society. Conciliatory on the other hand, tries to effectuate social harmony between the victim and the offender.
Black acknowledges that those types regularly intertwine and social control deviates from the pure form. The police can choose which type or types are most optimal to use in different situations.49 According to Klinger and Schulenberg, those forms of social control have to be viewed as if they act in a continuum. Police officers do not choose one of the styles; it rather is a fluent mechanism in which they operate.50
This research will focus mostly on the question whether the aforementioned characteristics have any effect on the decision of the public prosecutor to prosecute or dismiss an incident. Since the perpetrators will be the same in all cases, namely police officers, the characteristics of the victim are leading. The study focuses on stratification and culture. To understand the process of how deadly incidents by police are treated, it is important to know the legal framework in the Netherlands. This will be explained in the next paragraph.
2.2 Legal framework on use of police force in the Netherlands
The state is the only actor that can legitimately use violence.51 In the Netherlands policemen are authorized to use force whenever the goal justifies this use and it cannot be achieved in any non-violent way. The use of force should at all times be proportionate.52 This authorization on the lawful use of force is placed in article 7 of the Dutch Police Act 2012 [Politiewet].53 According to article 7 of the Dutch Police Act 2012 the police are allowed to use force in cases in which it is clear that the suspect will use a weapon against the officers or when the suspect tries to withdraw himself from the arrest, when the perpetrator is suspected of a
49 Black, D. (1976), p. 5-6.
50 Klinger, D.A. (1996); Schulenberg, J.L. (2010), p. 126.
51 Kamerstukken II 2016-17, 34641, nr. 3, p. 5; Munro, A. (2013). 52 Timmer, J.S. & Cozijn, V.M. (2016), p. 298-302.
12
crime with a sanction of imprisonment of at least four years and when the suspect is a threat to society or threatens serious violation of physical integrity.
In addition to the Police Act, an official instruction for police officers [Ambtsinstructie] is given.54 This official instruction entails conditions under which the use of force is legitimate. It indicates which means of violence the police officers are allowed to use in a certain situation and under what conditions.55 Those means of violence for example are pepper spray, the use of a firearm, a weapon stick, use of the police dog or physical violence.
When police officers use force against a civilian they are required to report the incident to their manager, based on article 17 of the official instruction [Ambtsinstructie]. The manager files a report about the incident.56 If the civilian has no injury or a minor one, the case is handled internally by the police. The manager is responsible for assessing whether the police officer acted legitimately or not. If an officer did not act legitimately according to the manager he will inform the police chief and sends the case to the National Police Internal Investigations Department [Rijksrecherche].57 Incidents in which the use of force led to severe injury or death will always be sent.58 The police chief informs the public prosecutor within 48 hours.59 The police chief will also inform the public prosecutor if a firearm is used or when the incident could cause a big impact on society.60
The report and incident are investigated by the National Police Internal Investigations Department. This department is an independent detective department that is directly accountable to the attorney generals of the Public Prosecution Service.61 Proper investigation of the deadly incident is required based on article 2 of the European Convention for Human Rights, regarding judgements of the European Court for Human Rights (ECHR).62 The ECtHR acknowledges that every case is unique and that it is impossible to construct a list of actions that need to be performed when investigating the incident. However, they determined minimum conditions under which the investigation should take place. Those conditions must be met in
54 Ambtsinstructie voor de politie, de Koninklijke marechaussee en andere opsporingsambtenaren.
55 https://www.politie.nl/themas/geweldsinzet.html#alinea-title-wanneer-mag-de-politie-geweld-gebruiken. 56 https://www.politie.nl/themas/geweldsinzet.html#alinea-title-wanneer-mag-de-politie-geweld-gebruiken. 57 Timmer, J.S. & Cozijn, V.M. (2016).
58 Art. 4.1 Aanwijzing handelwijze geweldsaanwending (politie) ambtenaar, Stcrt. 2006, 143. 59 Art. 17 par. 3 Ambtsinstructie.
60 https://www.politie.nl/themas/geweldsinzet.html#alinea-title-wanneer-mag-de-politie-geweld-gebruiken. 61 https://www.politie.nl/themas/geweldsinzet.html#alinea-title-wanneer-mag-de-politie-geweld-gebruiken. 62 EHRM 15 mei 2007, nr. 52391/99, par. 323-325 (Ramsahai e.a./Nederland); Thoonen, E. & Duijst, W. (2015),
13
case of a deadly incident. The investigation should be adequate and effective. Therefore, it should be independent and impartial.63
Within 24 hours after the incident the police officer needs to testify. This testimony can be used during a criminal trial following the incident. Note the fact that the police officer will initially be treated like a witness, contrary to non-police citizens who would be treated like a suspect in a case that caused severe injury or death.64 The discrepancy is caused by the fact that police officers have the legitimacy to use force in the course of their profession. Therefore, they are presumed to act according to official orders.65 The problem is that the obligation to give a testimony seems to be contrary to article 6 ECHR, which entails the nemo-tenetur principle.66 This principle entails that a suspect cannot be forced to incriminate himself.
Based on the outcomes of the research as conducted by the National Police Internal Investigations Department the public prosecutor decides about the further process. He has two options; either he can dismiss the case because it is clear that the judge will accept an appeal on a punishment exclusion ground, or he can initiate a prosecution. Based on the proportionality principle he is free to make this choice.67 The decision is based on the principles of subsidiarity and proportionality.68 If the decision of the public prosecutor relates to the use of a firearm he first needs to present his point of view to the Advisory committee of police use of firearms [Adviescommissie politieel vuurwapengebruik].69
If the public prosecutor dismisses the case, it is not necessarily finalized. In the Netherlands the victim, or a closely related person, can instigate a prosecution via a so-called article 12 procedure. Which is based on the Dutch Code of Criminal Procedure. If the applicant has enough reasonable grounds to prove that a prosecution is feasible, he can ask the higher court to force the public prosecutor to conduct more research and initiate the prosecution.
On the other hand, relatives of the victim can initiate a civil case in order to receive a compensatory payment. Those cases will not be part of this research and therefore the civil procedure will not be explained further.
63 EHRM 4 mei 2001, nr. 24746/94, par. 105-109 (Hugh Jordan/Verenigd Koninkrijk).
64 Art. 4.3 Aanwijzing handelwijze geweldsaanwending (politie) ambtenaar, Stcrt. 2006, 143. 65 Art. 7 Politiewet.
66 Groenhuijsen, M.S. & Jansen, R. (2018).
67 Art. 5.3 Aanwijzing handelwijze geweldsaanwending (politie) ambtenaar, Stcrt. 2006, 143;
https://www.politie.nl/themas/geweldsinzet.html#alinea-title-wanneer-mag-de-politie-geweld-gebruiken.
68 Naeyé, J. (2005), p. 44-46.
14
Practice shows that prosecution will be primarily based on the accusation of manslaughter or (aggravated) assault with a deadly ending.70 The police officer can appeal on the punishment exclusion grounds. Exclusion grounds make the act liable or they remove the culpability of the suspect since he or she acted accordingly to a statutory regulation. Acts of police officers are liable if they are exercised under law or official order according to articles 42 and 43 of the Dutch Criminal Code. These exclusion grounds are the most commonly referred to concerning police officers’ use of force. Article 41 of the Dutch Criminal Code can remove the culpability of the suspect, which entails the ground of self-defence. Note that exclusion grounds have different meanings for police officers than they do for civilians. Which is the result of the so-called Garantenstellung. This Garantenstellung entails that a specific individual in a public role, such as police officers, have more responsibilities than civilians.71
Next to disposal in the criminal justice system, acts of violence could be addressed through a disciplinary system or through a complaint procedure at the National Ombudsman. The legal framework for a prosecution within the disciplinary system is arranged in the Decision on general legal status of police [Besluit algemene rechtspositite politie]. The Minister of Justice and Security proposed a new law to amplify the existing legal framework. The proposed law aims to create a new legal framework to assess police officers’ use of force.72
2.3 Previous research in the Netherlands on deadly incidents by police force
2.3.1 Deadly incidents
Blaauw, Kerkhof and Vermut examined deaths under responsibility of the police in the Netherlands in the period between 1983 to 1993. During this period fifty-nine people died at a police station. Suicide and intoxication were the most important causes. Twelve incidents were placed in the category of ‘other/unknown cause of death’. Blaauw et al. do not pay special attention to this category, so it is unknown whether this category entails deaths caused by violence of the police.73
Research by the National Police Internal Investigations Department shows that in the period between 2000 and 2004 intoxication was the number one cause of death of civilians
70 Art. 287 and art. 300 par 1 and 3 and art. 302 Dutch Criminal Code; Kamerstukken II 2016-17, 34641, nr. 3, p.
5.
71 Timmer, J.S. (2005), p. 202-203. 72 Kamerstukken II 2016-17, 34641, nr. 3.
15
under police responsibility. The other causes were mostly suicides and natural deaths.74 Timmer examined the effects of intoxication as a cause of death. Intoxication could lead to exited delirium, which means that people hallucinate, are disoriented, have thinking disorders and suffer from an increasing heartbeat. Due to the nuisance such people create, police officers need to act. As Timmer says death is often unavoidable since the intoxicated person will get more excited upon action of the police. Therefore, their heartbeat will increase even further, which could become fatal.75
Thoonen and Duijst examined in 2015 whether there was room for improvement in the investigation of deaths of people who were arrested and taken into custody by the police. They examined incidents between 2005 and 2010 in which civilians died while under police responsibility. The cases include suicides of people that were taken into custody, natural deaths, medical conditions, deaths after car incidents or fall of great height and unknown reasons.76 Between 2005 and 2010 seventy-eight deaths were counted while under police responsibility. A police bullet caused fifteen of these deaths and five were suicides. A total of seventy percent of the people that died were mildly or heavily intoxicated.77
Researches show that the causes of death could vary and not always have a clear cause since multiple factors could contribute. This research will only look into the deaths caused by violent acts of the police.
Between 2000 and 2004 only two of the thirty-seven incidents led to a criminal investigation while three of them led to recommendations for a settlement within the disciplinary system.78 Of those two cases that led to a criminal investigation, one started as a criminal investigation and the police officers in this case were appointed as suspects. The civilian died after a violent arrest and suspicions arose that he died as the result of a positional asphyxia.79 In the end, the medical report concluded that the arrested died of cocaine. The other case led to a criminal investigation after the National Police Internal Investigations Department did a factual research.80
74 Rijksrecherche (2006), p. 7 and 11. 75 Timmer, J.S. (2005), p. 182-183.
76 Thoonen, E. & Duijst, W. (2015), p. 91-115. 77 Thoonen, E. & Duijst, W. (2014).
78 Rijksrecherche (2006), p. 25.
79 Positional asphyxia is caused because the person is held in a position that blocks or limits the flow of oxygen
to the brain. The person is not able to breath normally. This could lead to, for example, cardiac arrest. Positional asphyxia is mostly used by police, military or other forms officers, in order to control a person.
16
The problem the researchers of the National Police Internal Investigations Department discovered was the fact that not all deaths were known by the minister due to the fact that police officers officially only need to report incidents that took place at the police station and not during the arrest or transport. Another reason is the use of different definitions of the violence used.81 Another finding of the National Police Internal Investigations Department concerns the availability of data on further procedures. In a majority of the cases, twenty-six out of thirty-seven, no final letter of the public prosecutor was available in the dossier, because all police officers involved gave a testimony as a witness and not as a suspect according to the researchers.82
Research of Timmer shows that between 1978 and 2002 sixty-nine people died by a police bullet. Eleven of these shots were unwanted, which means that a firearm went off unintentionally. Timmer categorizes the unwanted shots in shots fired as a reflex, clumsy shots and shots fired during strangulation.83 Some people died of the so-called ‘suicide by cop’, which means that civilians try to get killed by police officers during their arrest. At least four of these deadly incidents, including the use of firearms, fall in this category.84
Timmer pays special attention to another category being the one of arrested people that die due to positional asphyxia. Police officers in the Netherlands are well aware that the way in which they clasp a perpetrator by the neck in order to restrain that person, could lead to this suffocation. Although Timmer argues that the phenomenon is still not always recognized soon enough.85 81 Rijksrecherche (2006), p. 11. 82 Rijksrecherche (2006), p. 25. 83 Timmer, J.S. (2005), p. 169-170. 84 Timmer, J.S. (2005), p. 177-182. 85 Timmer, J.S. (2005), p. 177-182.
17 Table 1 Deaths under responsibility of the police
Research Period Deaths Type of deaths investigated Cause: police bullet Blaauw, Kerkhof & Vermut (1997)86 1983 - 1993 59 While in police custody, including suicide, drug
overdose and natural deaths
-
Timmer (2005)87 1978- 2002 - Police shootings 69
Thoonen & Duijst (2015)88
2005 - 2010 78 While in police custody, including suicide, drug
overdose and natural deaths
15
2.3.2 Prosecutions
Not much is known yet about the decision making process of the public prosecutor regarding acts of violence by the police. Previous research focussed primarily on incidents including the use of a firearm or regular acts of violence, not necessarily with a deadly ending. No Dutch research gives an overview of deadly incidents by the police including both the use of a firearm and other forms of violence and the reason why some of them lead to a prosecution and some do not.
Between 1978 and 2000 the public prosecutor dismissed ninety percent of the cases in which a civilian was hit by a police bullet. Ten cases led to an article 12 procedure of which the outcomes are not always known, since the court sessions are private. Twenty-four police officers were prosecuted between 1978 and 1995, of which eight were convicted. Between 1995
86 Blaauw, E., Kerkhof, A. D., & Vermunt, R. (1997). 87 Timmer, J.S. (2005)..
18
and 2000 seven police officers were prosecuted, of which five were convicted.89 The second period shows an increase of incidents that were prosecuted and an increase of convictions.90
In 2000 a total of eighteen cases out of 2663 reported as acts of violence by police officers (0,7%) led to a criminal justice response. The criminal justice response varies from a conviction or acquittal by the criminal judge to a transaction by the public prosecutor. Naeyé’s research does not explicitly explain the construction of this low percentage.91
The discrepancy between Naeyé’s research and that of Timmer and Van der Steeg can partly be explained because Naeyé examined all forms of criminal justice convictions. This includes transactions imposed by the public prosecutor. Timmer and Van der Steeg’s cases are all resolved in the presence of a judge. Of the eighteen cases Naeyé presented only ten of them led to a criminal trial including a judge.92 Another difference between the researches is that Naeyé looked into all acts of violence, while Timmer and Van der Steeg only looked into incidents in which injury was caused by a police bullet.
Between 2000 and 2015, twenty-one incidents in which police officers used their firearm led to a criminal trial. However, twenty-three police officers faced criminal prosecution, since two of these incidents led to more than one prosecutions. Fourteen of the twenty-one prosecutions took place between 2013 and 2015. Although the number of prosecutions increased a majority of the reported acts of violence is dismissed by the public prosecutor. According to Van der Steeg and Timmer this number cannot only be explained due to an increased use of article 12 procedures. The public prosecutor also instigated more prosecutions in this period than before.93 Investigation is needed into the cause of the increase in cases that appear before trial and into the public prosecutor’s considerations to instigate prosecution. This research will look into these questions.
Only eight of the twenty-one cases led to a conviction. It is striking is the fact that these eight convictions were all declared after 2013. Between 2013 and 2015 there has not only been an increase in the amount of prosecutions, but also in the amount of prosecutions that led to a conviction.94 This coincides with Timmer’s findings between 1978 and 2000. It is important to
89 Timmer, J.S. (2005), p. 221-222. 90 Van der Steeg, M. (2016), p. 2-3. 91 Naeyé, J. (2005), p. 426-427. 92 Naeyé, J. (2005), p. 426-427.
93 Van der Steeg, M. (2016), p. 2; Timmer, J. (2015). 94 Van der Steeg, M. (2016), p. 2-3.
19
keep in mind that those cases do not only look at deadly incidents, but also at the use of a firearm by police officers with different outcomes.
The legitimacy of the use of force, like for example clasping a perpetrator by the neck, is in the first place decided by the manager of the police corps and the public prosecutor (see “legal framework”). The national ombudsman Brenninkmeijer investigated the use of force by the police and examined which acts are found to be legitimate and which are not between 2011 and 2012. Only a maximum of 5 percent of the reported incidents in which the police used force against a civilian actually led to a (criminal) complaint.95
Table 2 Prosecutions and convictions of police officers using firearms
Research Period Prosecutions Convictions Timmer (2015)96 1978 – 1995 24 8
1995 – 2000 7 5
Van der Steeg (2016)97
2000 – 2013 7 0
2013 – 2015 14 8
95 Brenninkmeijer, A.F.M. (e.a.), p. 20-21. 96 Timmer, J. (2015).
20
3 Methodology
The research goal of this thesis is to find out whether and to what extent victim’s characteristics relate to the prosecution’s decision, regarding deadly police force. It will explain and predict where possible which deadly incidents will end up in the criminal justice system and which will not. The cases will be analysed in both a qualitative and a quantitative manner. According to Black’s Theory of Law, victim’s characteristics affect decisions such as the prosecution’s decision. It approaches law as a quantitative variable. In order to substantiate the qualitative analysis the data will be analysed in a statistical manner. A quantitative analysis will be conducted to show a correlation between the victim’s characteristics and the prosecution’s decision. The prosecution’s decision is the dependent variable and the victim’s characteristics are the independent variables.
3.1 Data collection
The research is part of a broader research into the Dutch Homicide Monitor of Leiden University. This monitor is provided by Leiden University and is a database including all homicide cases between 1990 and 2016. The data is collected from six sources: ‘all newspaper articles related to homicide as generated by the Netherlands National News Agency (ANP); Elsevier annual report; files from the National Bureau of Investigation (NRI); files from the Public Prosecution Service; files from the Judicial Information Service and the Ministry of Justice; additional data stemming from regional police; files from the Criminal Justice Knowledge Centre (WODC)’.98 The information in the database entails, among others, details
on the homicide incident and several offender’s and victim’s characteristics.99
The database’s problem is that it only contains cases that are actually designated as a homicide, which means manslaughter, murder or assault resulting in death. However, most of the police cases are dismissed by the public prosecutor and deaths are not registered as homicide. An analysis of the database leads to the conclusion that an open source research is necessary. The database did not provide enough information to be of use in this research. The open sources are mostly newspaper articles, press releases of the public prosecutor and information from the police or National Police Internal Investigations Department.
98 https://www.universiteitleiden.nl/en/research/research-projects/law/homicide (latest checked on
29-07-2018).
21
Each year the public prosecutors publish on their website an overview of deadly incidents as the result of firearm use by the police. This overview only shows the number of deaths and does not show any details of the cases. It is available for the period between 2008 and 2017.100 The National Police Internal Investigations Department presented a report on what
grounds they start investigating an incident. In this report they give an overview of the deadly incidents as the result of firearm use by the police between 1978 and 2002.101 Therefore, the
total amount of deaths by police shootings is known for two periods: between 1992 to 2002 and between 2008 to 2016. This contributes to the data collection, since it shows the total amount of cases to be collected.
Because of the repeated judicial discussion about the cause of death and the police’s involvement this results in a lack of official numbers of these deaths. Therefore it is important to search open sources as thoroughly as possible. The National Police Internal Investigations Department might could for example conclude that someone died of a natural cause, whereas relatives might argue that the death was caused by police strangulation during an arrest. See for example the case of Corné Pieterse, who died in 2007 after a wrestle with the police. His daughters filed an article 12 complaint following the court decision in the Mitch Henriquez case.102
Deaths by police force are ordinarily mentioned in the news and on authority’s websites. The public prosecutor regularly publishes press releases about deadly incidents involving police officers. Those press releases do not contain enough victim’s characteristics for this research, but they form a starting point in the collecting of information about the incident. This research therefore started with searching for press releases on the website of the public prosecutor. Since all deadly incidents of police force are investigated by the National Police Internal Investigations Department that department will be used as a search term. Figure 1 shows the flowchart of research on the website of the public prosecutor.
100 https://www.om.nl/vaste-onderdelen/zoeken/@31569/rijksrecherche-doet/;
https://www.om.nl/@101653/rijksrecherche/.
101 Rijksrecherche (2002). 102 Quekel, S. (1 December 2016).
22 Figure 1 Flow-chart of searching ‘Rijksrecherche schietincidenten’ on website of the public prosecutor
Figure 2 Flow-chart of searching ‘Rijksrecherche overleden’ on website of the public prosecutor
The next step was collecting more detailed information from news articles. In order to collect valid data, the LexisNexis Academic database was used. This database contains more than 17.000 sources including newspapers.103 In this research, both national and local
newspapers are used to collect data on the characteristics of victims by police force. Whereas national newspapers contain more general information on the incident and the victim, like gender, age and residence, local newspapers present more detailed characteristics of the victims and its relatives, such as information on their civil status, wealth, psychological health and social life. Which local newspapers can be used, differs per area. Examples of local newspapers are Brabants Dagblad and de Gelderlander.
Local newspapers complement victim information, which made it possible to search more closely for relevant articles. First more general searching terms were used. The searching always included the terms politie (police) and the cause of death, for example doodgeschoten (killed). Additionally a victim characteristic was added, such as the age, gender, nationality or residence of the victim. If those terms led to more details, e.g. the name of the victim, this led to another round of searching in the LexisNexis database. The next search would be based on
103 https://www.lexisnexis.nl/producten-en-diensten/producten/lexisnexis-academic. Term: 'Rijksrecherche schietindicenten' Database: www.om.nl 612 results
Selected all cases including: 'schietincidenten' or 'vuurwapengebruik politie' or 'Rijksrecherche onderzoek' Term: 'Rijksrecherche overleden' Database: www.om.nl 2083 results
Selected all cases including: 'Rijksrecherche onderzoek' or 'arrestant overleden' or 'schietincident' or 'neergeschoten' or 'ongeval agent'.
23
the type of death and the name of the victim. Figure 3 shows a flow-chart of the searching process on one of the cases investigated, namely the case of Samir Laifa who died after he was shot by the police. The last step of the flow-chart shows the categories of information that were retrieved by this search.
Figure 3 Flow-chart terms 'Doodgeschoten Algerijn' and 'Samir Laifa' on LexisNexis
The third database that was used is rechtspraak.nl. This database contains all published court rulings from 1989 until present. Not all court rulings are published.104 In most court cases
police officers are addressed as politieambtenaar, which was therefore used as the search term in this database. To minimize the amount of results, the term overleden (deceased) was added. This filtered the cases in which a citizen was suspected of committing a crime against a police officer and in which the detainee did not die. See figure 4.
104 Besluit selectiecriteria uitsprakendatabank Rechtspraak.nl. Term: 'doodgeschoten Algerijn' Database: LexisNexis 377 results Selected timeframe: 1998 17 results, which revealed the name of the victim
Term: 'Samir Laifa' Database: LexisNexis 5 results Information: name, age, gender, citizenship, residence, psychological history, criminal history, type of violence, crime scene, prosecution’s decision, article 12 procedure, public protests.
24 Figure 4 Flow-chart term 'politieambtenaar overleden' on rechtspraak.nl
3.2 Conceptualization of victim’s characteristics
In order to conduct both a qualitative as well as a quantitative analysis, a plethora of variables is needed. The cases are all related to the police force, in which the police officer is presented as the offender. In all selected cases the police officers were on duty. Cases in which a person murders someone, the offender being a police officer off duty, are not included in this research since the relation between the perpetrator and victim in those cases differs from the relation between a victim and police officer on duty. This difference is prompted by the fact that police can legitimately use force, which is not true when they are off duty.
Use of force, in this research, contains all physical acts of violence by the police. This entails shootings with a police weapon, use of a police dog, use of a weapon stick, pepper spray and physical violence without the use of equipment.105 An example of this solely physical
violence is a neck clamp in which an officer tries to restrain a person or a wrestling during the arrest.
Additionally cases which a victim died after another confrontation with the police are included. This includes victims who were chased by the police and jumped of a roof or in a river during that process. Those incidents are categorized as ‘other’ in the variable modus operandi.
A special category, next to the cases in which police used force, are cases in which victims died after they got in a car incident following a chase by the police. This category is supplemented with traffic incidents including a police vehicle.
105 Timmer, J.S. & Cozijn, V.M. (2016), p. 306. Term: 'politieamb tenaar' Database: rechtspraak .nl 5305 results Additional term: 'overleden' 435 results Selected all cases including: 'artikel 12' or 'arrestant overleden' or 'nabestaan-den arrestant' or 'vervolging politie-ambtenaar'
25
The independent variables in this research are the victim’s characteristics. In all cases in which the civilian died due to police force, they are considered a victim even though initially they may have been perpetrators. In order to find out whether victim’s characteristics affect the dependent variable, the prosecution’s decision, it is necessary to work with the important variables. The characteristics relate to the two factors of Black’s Theory of Law that are part of this research, as explained in Chapter 2 Literature review. Those two factors are stratification and culture. The following victim’s characteristics will be analysed related to stratification: gender, age, civil status, housing situation, professional status, psychological disorder or a history of mental illness and criminal history. The following victim’s characteristics that will be analysed related to culture are: ethnicity, crime scene and protests following the incident.
Denno and Cramer studied judicial decisions in order to categorize victim-related effects.106 They indicated that there are two categories of victim’s characteristic that influence
judicial decisions: physical and situational characteristics. The physical category entails characteristics such as age, gender, sex and race, whilst the situational category entails characteristics such as relationships between offender and victim, other relationships around the victim and conditions of the incident. Denno and Cramer found that the severity of the offender’s punishment directly relates to the judges perception of the victim. The more favourable the perception of the victim, the more severe the punishment for the offender. Provocation and being part of a racial minority were victim’s characteristics that negatively affected the judges perception.107 The analysis of the data are arranged into physical and
situational factors. First the physical victim’s characteristics will be analysed followed by the situational characteristics.
106 Denno, D. & Cramer, J. (1976); Bynum, T. S., Cordner, G. W., & Greene, J. R. (1982). 107 Denno, D. & Cramer, J. (1976), p. 303.
26
4 Analysis
4.1 Overall findings
The dataset contains 83 cases in which a civilian died after confrontation with the police. Fourteen of those cases are traffic accidents in which either a police car or motorcycle crashed with a civilian or in which a civilian died as the result of a pursuit. This study focuses on both deaths as the result of police force and police related traffic accidents, as explained in Chapter 3 on methodology. In the following analysis both types of deaths will be analysed separately.
Graph 1 shows the number of deadly victims due to police force between 1992 and 2016. The deaths of traffic incidents are not included. Previous studies showed that an average of three people per year died by a police bullet.108 Graph 1 does not only show victims of police shootings, but also the victims of other forms of force such as wrestling during the arrest or the use of pepper spray. Therefore it is notable that the average number of killings based on this dataset is 2,76 per year. This number is slightly lower than the aforementioned three deaths per year, whilst the dataset focussed on various uses of force.
The graph also shows a slight increase in the number of victims over the whole period. This could possibly be related to the fact that it is harder to find data on the cases between 1992 and 2005, see chapter 5 for a further explanation. According to the numbers of deadly police shootings, published by the public prosecutor, the average of victims did not change over the years.109
Graph 1 Dispersion of victims of use of force between 1992 – 2018, including prosecutions of police officers (orange dots)
108 Timmer, J.S. & Cozijn, V.M. (2016), p. 308. 109 Timmer, J.S. & Cozijn, V.M. (2016), p. 308.
0 1 2 3 4 5 6 7 8 1990 1995 2000 2005 2010 2015 2020 N U MBE R OF VICT IMS
27
In a majority of the cases (68,1%) in which the police used deadly force the modus operandi was the use of a firearm. Table 3 shows the dispersion in modus operandi of the police.
Table 3 Modus operandi police use of force
Victims N % Dog 1 1,4 Firearm 47 68,1 Neck clamp 4 5,8 Other 3 4,3 Pepper spray 2 2,9 Wrestle 10 14,5 Unknown 2 2,9 Total 69 100
4.2 Victim’s characteristics
4.2.1 Physical characteristics
The physical characteristics that are analysed in this study are gender, age, ethnicity and mental health.
Gender
In only six out of 83 cases, the victim was a woman. This is 7,2% of the total. The other 77 victims were males. When splitting the dataset in two categories, namely use of force victims and victims of police related traffic accidents, it becomes clear that only one woman was killed by police force out of 69 cases. That means that 98,6% of all victims of deadly police force were male.
Given the small number of female victims it is not possible to do a significance test to see whether gender affects the prosecution’s decision. Furthermore, the prosecution’s decision in the case of the woman is unknown.
28
Age
Out of 83 cases in the dataset of this study, the age of six victims is unknown. The average age of all deaths in this study is 33 (M = 33,84; SD = 11,986). The average age of victims that were killed by police use of force is 34 (M = 34,61; SD = 11,347) and the average age of victims that died in a police related traffic accident is 30 (M = 30,08; SD = 14,677).
To understand whether the prosecution’s decision on the use of force against civilians is affected by the age of the victim, it is first important to check the distribution. The age of victims of police force is normally distributed (skewness = 0,641; kurtosis = 0,475), see graph 3. Therefore a parametric T-test is performed. Only three cases led to prosecution and the Levenes test for equality of variances shows that there is no significant relation between the age of the victim and the prosecution’s decision (t (49) = -0,579; p = 0,913). The victims in which cases prosecution was initiated were 17, 32 and 42 years old.
Graph 2 Age of victims of use of force, including a normality curve
In police related traffic accidents, the age of the victims is not normally distributed as shown in graph 4 (skewness = 1,510; kurtosis = 2,156). Out of fourteen cases that are part of this dataset, the public prosecutor initiated prosecution in five cases. To analyse the relation between the age of the victim and the prosecution’s decision in those cases, a non-parametric Kruskal Wallis test is performed. Also in this type of offence, there is no significant relation between the prosecution’s decision and the age of the victim (H = 0,778; p = 0,378 > 0,05).
0 2 4 6 8 10 12 14 15-20 20-25 25-30 30-35 35-40 40-45 45-50 50-55 55-60 60-65 65-70 70-75
29 Graph 3 Age of victims of police related traffic accidents, including a normality curve
It is remarkable that most victims are relatively young. A majority of 18,8% of the victims of deadly police force were between 20 and 25. The victims of police related traffic incidents were even younger; a majority of 30,8% was between 15 and 20.
Ethnicity
In 44 cases the ethnicity of the victim was known, of which seventeen persons were Dutch, which is 38,6%. Table 4 shows the number of victims of police killings in the Netherlands (both by use of force or traffic accidents) with a non-Dutch ethnicity.
0 1 2 3 4 5 15-20 20-25 25-30 30-35 35-40 40-45 45-50 50-55 55-60 60-65 65-70 70-75
30 Table 4 Citizenship of victims of police
Victims N % Albania 1 1,2 Algeria 1 1,2 Aruba 1 1,2 Bulgaria 1 1,2 Germany 2 2,4 Macedonia 1 1,2 Morocco 3 3,6 Netherlands Antilles 1 1,2 Nigeria 1 1,2 Poland 3 3,6 Russia 1 1,2 Sierra Leone 1 1,2 Suriname 5 6,0 Turkey 4 4,8 USA 1 1,2 Total 27 32,4
Map 1 shows the global distribution of ethnicity of the victims. A majority of the cases involved a victim with a non-Dutch ethnicity or background. The data does not show information on the skin colour of the victims. Based on this information it is not possible to say whether there is a racial bias concerning black skinned people in the approach by police officers. In the current data, only sixteen people were non-Western: they were African, Middle Eastern and Caribbean. Including Dutch victims, nineteen people came from Western countries and eight victims had a Slavic or Eastern European background.
31 Map 1 Distribution citizenship victims of both use of force and traffic accidents involving police.
The victims in the dataset are grouped in Dutch and non-Dutch ethnicity. By performing a chi-square test it is possible to see whether there is a correlation between the use of a firearm by police officers and the ethnicity of the victim. The Fisher exact test shows that there are no significant differences between the use of a firearm against Dutch victims and non-Dutch victims (p = 0,488 > 0,05).
In order to see whether there is a racial bias in the prosecution’s decision, also a chi-square test and Fisher exact test are performed. The independent variable was the ethnicity of the victim, grouped in Dutch and non-Dutch. The dependent variable was the prosecution’s decision, grouped in prosecution and dismissal. Table 5 shows the dispersion of the prosecution’s decision in relation to the ethnicity of the victim. The Fisher exact test shows a p-value > 0,05 (p = 0,261), which means that there is no significant relation between the prosecution’s decision and the ethnicity of the victim. Therefore, it is not possible to indicate differences in the prosecution’s decision regarding Dutch victims and non-Dutch victims.
Table 5 Dispersion of prosecution’s decision related to ethnicity of the victim, including percentages of total
Prosecution Dismissal
N % N %
Dutch - - 13 39,4
Non-Dutch 3 9,1 17 51,5
