Autonomous Weapons Systems in Law Enforcement
Potential threat towards the most fundamental of human rights?
Mattias Leppänen
Human Rights 180 Credits Bachelor Thesis
12 Credits
Spring Semester 2021 Supervisor: Lena Karlbrink
Abstract
This paper focuses on that less researched area of the emergence of autonomous weapons systems. The point of this thesis is to research and map out this relatively under researched area, with the goal of answering research questions, and ending the thesis with an attempt at creating a hypothesis based on the conducted research. This is done by applying an inductive qualitative study, a normative framework, in terms of the doctrine of necessity, as well as analysing primary and some secondary sources by applying a normative, interdisciplinary approach with a hermeneutic analysis. One could most definitely argue that this research area is rather under researched, with a few exemptions, which is why this specific topic was worth choosing. The main findings of the conducted research were that there are many questions surrounding the topic of autonomous weapons systems usage. Although, perhaps the main question that needs to be answered before any other questions can have a substantial meaning, is the question of the level of autonomy that these weapons systems will possess.
Key words: international human rights law, arbitrary, necessity, imminent, autonomous weapons systems, law enforcement
Table of content
1. Introduction……….3
1.1 Introduction to Thesis Topic………3
1.2 Aims of the Thesis………...5
1.3 Research Questions………..5
1.4 Relevance to the Field of Human Rights……….6
1.5 Limitations and Delimitations………..7
1.6 Chapter Outline………....7
2. Previous Research………...8
2.1 Definitions of AWS and a Potential Ban………..8
2.2 AWS in Relation to Armed Conflict………...10
2.3 AWS in Relation to Law Enforcement………....12
3. Normative Framework………..14
3.1 Doctrine of Necessity………..15
4. Methodology………...17
4.1 Qualitative Method………..17
4.2 A Normative Approach with a Hermeneutic Analysis………18
4.3 Material………...19
5. Analysis………...21
5.1 Terminology, Definitions and Possible Characteristics of AWS……….21
5.2 Article 6.1 of the ICCPR, and General Comment No. 36………...22
5.3 United Nations Code of Conduct for Law Enforcement Officials………..24
5.4 Basic Principles on the Use of Force and Firearms by Law Enforcement Officials...27
5.5 Results of the Research/Creating a Hypothesis for Future Research………..30
6. Conclusion………...31
1. Introduction
1.1 introduction to Thesis Topic
One could argue that trust is essential to all human beings, that it entails aspects of human emotions such as how we communicate, how we understand each other, and what we can expect of each other, etc. (Warren., Hillas 2020, p. 824, 825). Moreover, there is also a factor of trust in the relationship between human beings and robots, which in this thesis involves autonomous weapons systems. Including how humans interact with the robot, how the robot acts in terms of human emotions such as comfort and satisfaction, and essentially the robot's level of autonomy (Warren., Hillas 2020, p. 825). The development and potential usage of robots in different sectors of our societies today mostly depends on the level of trust between human beings and robots, especially in terms of the factors mentioned above, and as the development of robots progresses, the need for understanding of the autonomy of robots increases (Warren., Hillas 2020, p. 825, 826). Furthermore, it has been proven that robots possess a substantially higher capacity to solve algorithmic problems than humans, although when it comes to judgment calls involving human factors, a robot lacks the proper
understanding of humans to exceed in that area as well (Warren., Hillas 2020, p. 826). Some argue that these two above-mentioned different qualities between robots and humans could work well together if one could find a way to combine the two. Essentially, a way to combine human emotional intelligence and the robots' exceeding capability to solve problems with the help of carefully programmed algorithms (Warren., Hillas 2020, p. 826). These
above-mentioned features could also be incorporated into the topic of future usage of autonomous weapons systems in both warfare and law enforcement. Autonomous weapons systems is a sensitive area of research and development, to say the least, and the factors of autonomy must work well with other factors; to achieve that cooperation between humans and robots, there is a need for when to apply autonomy and when not to (Warren., Hillas 2020, p. 827). According to some promoters of autonomous weapons systems, there are three important factors that, if fulfilled, autonomous weapons systems will most certainly have a place in warfare and likely in law enforcement as well. Those factors include system factors, individual factors, and situational factors; nevertheless, the most important aspect still comes down to the actual autonomy of the weapons and that there is always a balance of the
is that there wouldn't be any issues involving autonomy if the human in the loop would have the proper tools and understanding of the autonomous weapons system, which once again comes down to trust (Warren., Hillas 2020, p. 828). Discussions regarding preemptive bans on autonomous weapons systems and the seemingly inevitable usage of these types of weapons in situations of armed conflicts have taken place. Some argue that these types of weapons should be carefully introduced into the battlefield. Others have been arguing for a preemptive ban on all autonomous weapon systems. Warren and Hillas, however, argue that even if autonomous weapons systems would be introduced into the battlefield as carefully as possible, and even if humans would stay in control, there are still issues under both
international and national legal frameworks (Warren., Hillas 2020, p. 828). Assuming that autonomous weapons systems would eventually be introduced into the battlefield, and even assuming that the national and international legal and ethical frameworks regulating armed conflicts would be adjusted to fit this new type of warfare, there are still potential issues in other aspects concerning the usage of autonomous weapons systems. As the development of autonomous weapons systems continues to progress, one could argue that it is inevitable that these sorts of weapons will eventually be introduced in law enforcement as well, which entails completely different legal and ethical issues, in contrast to those applicable under armed conflict. As of today, the aspect of the future usage of autonomous weapons systems in law enforcement is relatively under-researched, with a few exceptions, which is why this thesis will focus on the emergence of autonomous weapons systems in law enforcement. Earlier in the introduction of this thesis, the factor of trust was mentioned as one of the arguments in terms of how the future usage of autonomous weapons systems would function in relation to human beings. However, even before human beings can feel trust towards an autonomous robot, the level of autonomy needs to be proven. Even more importantly, the legal and ethical frameworks need to properly regulate these types of weapons in law
enforcement, which in comparison to the laws of armed conflicts, is much more complicated and perhaps even more essential. Autonomous weapons systems in situations of armed conflicts will focus on chosen targets that are deemed a threat, which of course, still could resolve severe human rights violations. Although one could argue that it is even more important to have the proper regulations of autonomous weapons systems within law
enforcement, mainly since domestic law enforcement is supposed to protect the human rights of the people under state authority. Therefore some clear rules and laws must exist to regulate the usage of autonomous weapons systems.
1.2 Aims of the Thesis
This thesis aims to interpret and analyze a few selected sources, as well as provide an attempt to map out the future relationship and potential legal and ethical issues that will arise with the emergence of autonomous weapons systems in domestic law enforcement. In terms of the topic that will be analysed and presented in this thesis, there are several legal and ethical aspects that somewhat overlap each other, hence the decision to have an interdisciplinary approach that focuses on both legal and ethical aspects of this chosen topic. The focus of the interpretational aspect, which is conducted through the hermeneutic analysis section, is placed upon the wording or language of a few selected sources, both legal and ethical, that is directly related to the field of autonomous weapons systems in law enforcement. After the sources have been interpreted and analyzed, the findings of those sources will be further interpreted in relation to the future usage of autonomous weapons systems. Furthermore, due to the inductive aspect of this thesis, a hypothesis will be presented in the conclusion of this thesis. Lastly, since this thesis is of a normative nature, another aim is to answer two
normative research questions that will be presented under the following subheading. If the study has been conducted in a proper way, the answers to the research questions should not be answered with ease. Instead, the research question and its related answer should aim to add knowledge and research to the already under-researched area of autonomous weapons systems in relation to law enforcement.
1.3 Research Questions
The research questions of this thesis were chosen based on the lack of knowledge involving the area of autonomous weapons systems within law enforcement, in contrast to the much more well-researched topic of autonomous weapons systems in relation to armed conflict. Therefore, the two research question of this thesis is:
1. In order for the right to life under IHRL to be adequately protected, should the legal use of force be modified to impose proper regulations on AWS in law enforcement?
2. Should AWS be considered as weapons per se, or as duty-bearers of the state in the shape of law enforcement officials?
1.4 Relevance to the Field of Human Rights
As of now, international humanitarian law seems to be the primary legal framework that will regulate the usage of autonomous weapons systems in situations of armed conflicts and there are for sure questions surrounding that relationship as well. Although, the questions around the relationship between autonomous weapons systems and international human rights law is most likely even more complicated. In 2012, Peter Asaro raised the questions about whether or not the international human rights laws in place would be able to protect fundamental human rights against the potential threat posed by autonomous weapons systems, which is still very relevant today (Asoro 2012, p. 363). The right to life, according to article 6.1 of the ICCPR states, "Every human being has the inherent right to life. This right shall be protected by law. No one shall be arbitrarily deprived of his life" (International Covenant on Civil and Political Rights, 1966). However, as Joseph explains, there are situations that somewhat allow the deprivation of life, without violating the right to life, essentially in a non-arbitrary way (Joseph 2019, p. 349). Joseph continues by explaining the role of law enforcement in the protection of the right to life, and also the role law enforcement can play in the deprivation of life, in both an arbitrary, and non-arbitrary way (Joseph 2019, p. 349). When reading Joseph's article, one can see that the relationship between law enforcement and the right to life is important, and does not exist without problems. The difference between an arbitrary
deprivation of life, and a non-arbitrary deprivation of life, revolves around the characteristics of a situation, and is based on the situational judgement of whether or not there is a direct threat. Since there are apparent present day issues involving the relationship between law enforcement and the right to life, one can only imagine the potential issues that might arise in the future when autonomous weapons systems become available for usage within law
enforcement. Lastly, when a situation in which there is a question of whether or not an arbitrary deprivation of life has taken place, there are a few important questions to be answered. George P. Fletcher explains that these type of questions involves the right-holder of the right to life, the duty-bearer responsible for the either positive or negative obligation to protect the right to life, as well as the circumstances where one can take another's life in a non-arbitrary way are important in regards to this topic (Fletcher 1980, p. 135, 136). Based on the above mentioned details on the right to life, the future usage of autonomous weapons
and its potential impact on fundamental human rights becomes visible, which entails a relevance to the field of human rights.
1.5 Limitations & Delimitations
Limitations: In terms of the limitations of this thesis, perhaps the biggest ones are connected to the choice of a qualitative method. According to Jana Bradley, there are four main issues involving qualitative research. These issues include, first; the interpretation itself since it is carried out from one researcher view; second; that the natural characteristics of the method might be too much decided beforehand; third; the meanings of information differs from person to person; and fourth; the reliability of the research conducted and the sources that were collected as data might be questioned (Bradley 1993, p 433-436). In regards to this thesis, perhaps the first one of the limitations is the most prominent, although one needs to be aware of these possible issues or limitations. Another limitation is that autonomous weapons systems are still years away from being fully developed. Therefore aspects such as questions of human control and other characteristics of autonomous weapons systems have yet to be proven.
Delimitations: One delimitation of this thesis was the decision to focus on the future usage of autonomous weapons systems in relation to law enforcement, which allowed the research to be focused on the perspective of international human rights law instead of focusing on autonomous weapons systems in armed conflicts. If one would choose to focus on the aspect of armed conflicts, the thesis would have been less focused on human rights. It could also be important to add that if the focus would have been on international humanitarian law, instead of international human rights law, the principle of distinction would have been rather
important. However, since the focus is on international human rights law, the principle of distinction under international humanitarian law is not applicable, and the same goes for article 51 of the UN Charter which entails state and collective self defense. Furthermore, another delimitation was the choice to focus on one right, in this case, the right to life according to article 6.1 of the International Covenant on Civil and Political Rights.
The outline of this thesis includes six chapters. Chapter 1, which was just presented, is the introduction. Chapter 2 presents the previous research that has been conducted by other researchers and scholars involving the chosen thesis topic. Chapter 3 presents the chosen normative framework and the motivation to why the framework was chosen, and chapter 4 covers the used methods, research approach, and the selected material. Chapter 5, which is the most significant part of the thesis, is the analysis section where the interpreted and
analyzed material is presented and connected to the previous sections. The last chapter, which is chapter 6, is the conclusion of this thesis, where the findings of the research will be
presented in summary.
2. Previous Research
The following chapter will examine, discuss, and analyze the previously carried out research within the field of AWS, focusing on discussions involving AWS and its presumed impact on the world, in terms of ethical and legal aspects of human rights. Furthermore, in order to make the previous research field as clear and transparent as possible, the relevant literature will be divided into three groups consisting of, The definitions of AWS and a potential ban, AWS in relation to armed conflict, and AWS in relation to law enforcement. By dividing the chapter into the three different groups, it will hopefully allow for an overall understanding of the selected thesis topic, with the first group consisting of more general discussions on the topic of AWS. The second group will provide a slightly more narrowed down discussion of the topic of AWS in relation to armed conflict, and the last group will focus solely on some previous research in regards to AWS in relation to law enforcement, which directly relates to the chosen topic of this thesis.
2.1 The Definitions of AWS and a Potential Ban
When looking at the discussions involving the development of AWS, one can see that there is a great deal of dissenting opinions within this field, mainly due to the fact that AWS
development has yet to reach a point of complete autonomy. Therefore, there are also many different opinions on the topic, and perhaps at the center of the discussions is the question of whether or not there should be a ban on all AWS development, or not. Although first of all,
the definition of AWS, and the numerous subcategories within the definition of AWS, are one thing that the researchers within this field all try but somewhat fail to determine. For
example, Didier Danet in his article Do Not Ban "Killer Robots", defines LAWS (Lethal Autonomous Weapon Systems) as lethal weapon systems that, without human control, carry out programmed lethal targeted killings with full autonomy (Danet 2017, p. 716).
Furthermore, the term LAWS that Danet is using is merely one of the definitions or
subcategories within the topic of AWS, which also includes different levels of autonomy and human control, as well as defensive in contrast to offensive weapon systems (Danet 2017, p. 718, 719). Danet is a firm believer that banning AWS at an early stage, two to three decades before it has been adequately developed and used, would be rather difficult, if not impossible (Danet 2017, p. 720).
Edward Moore Geist continues on the same track as Danet. However, he furthers the
argument by saying that a preemptive ban would not only be difficult but also a mistake since the usage of AWS could potentially be beneficial (Moore Geist 2016, p. 319). Moore Geist might agree with Danet in terms of their stance on a ban on AWS. However in contrast to Danet when it comes to the question of how to define AWS, Moore Geist places emphasis on the difficulties of definition, which he argues is the main problem with a preemptive ban (Moore Geist 2016, p. 319). Frank Sauer is another researcher within the field of AWS, although it needs to be stated that Sauer has the opposite approach in comparison to Danet and Moore Geist when it comes to the question of a ban on AWS. Furthermore, in terms of providing a definition, Sauer uses a definition that Pentagon brought forth, which states that an AWS is defined as a weapon system that "once activated, they can seek, select and engage targets without intervention by a human operator" (Sauer 2016, p. 8). The definition that Sauer mentions, provided by the Pentagon, seems to be one of the more useful definitions within the field. However, Denise Garcia, in the article, Killer Robots: Why the US Should Lead the Ban, mentions another definition that is similar to the one provided by the Pentagon, yet it is not quite the same. According to Garcia's article, an AWS is a "system that, once activated, can choose targets without further human intervention" (Garcia 2015, p. 57). As mentioned earlier, the definition that Garcia mentions is quite similar to the one of Sauer, although one could argue that since the definitions do not use the same wordings, they should be treated as different from one another, especially since the wording is vital when
interpreting laws, which would be the case if an attempt at a ban would be pursued. The last example of literature that will be examined in this group will be the Human Rights Watch
report named Losing Humanity: The Case Against Killer Robots. Based on the title of the report, it becomes rather clear what the intention and approach of the Human Rights Watch are, which most likely need no further explanation. What is interesting, on the other hand, is the definition that the Human Rights Watch brings forth. According to the Human Rights Watch, an AWS, or killer robot, which is the used term in the report, is "machines that have the power to sense and act based on how they are programmed", and "they all possess some degree of autonomy, which means the ability of a machine to operate without human supervision" (Human Rights Watch 2012). Human Rights Watch also divides and defines three different subcategories of killer robots. The first subcategory is the human-in-the-Loop which entails "robots that can select targets and deliver force only with a human command, the second is the Human-on-the Loop which points to "robots that can select targets and deliver force under the oversight of a human operator who can override the robots' actions", and lastly the Human-out-of-the-Loop which includes "robots that are capable of selecting targets and delivering force without any human input or interaction" (Human Rights Watch 2012). It is important to acknowledge that the report from the Human Rights Watch was published earlier than the other literature that has so far been mentioned. However, since AWS development is relatively fresh and still quite many years from being ready to be used, it is still interesting to see how the discussion has been developed over the years. When looking at the definitions that, for example, Danet provides, one can see that the issues of definition have at least been further developed, and there seems to be a development within the understanding of the field. The next group that will be examined in order for this chapter of the thesis to be narrowed and eventually leading up to the last group of literature, AWS in relation to law enforcement, is the existing research on AWS and its presumed impact on the relevant laws which exist to regulate armed conflicts.
2.2 AWS in Relation to Armed Conflict
Moving on to the second group, AWS, in relation to armed conflict, the lens shifts from merely the general discussions involving AWS, such as the question of definitions, to a somewhat more focused area of the literature. Questions and issues involving definitions give a good introduction to the topic of AWS, although to further narrow the research field, it is important to look at the discussions revolving around AWS as military means in armed conflict. For example, in their article, Steven Umbrello, Phil Torres, and Angelo F. De Bellis
attempt to analyze the situation involving autonomous robots and their relation to the laws of war. First, like most researchers within this field, the authors provide a definition of LAWs as well as a description of the two main categories of LAWs, which are defensive and offensive autonomous weapon systems (Umbrello., Torres., De Bellis 2019, p. 273, 274). Umbrello, Torres, and De Bellis goal with the article is to assess the possible impact on the laws of war and armed conflict by the development and future usage of AWS, or LAWs, which is the preferred term of the authors (Umbrello., Torres., De Bellis 2019, p. 273, 274). The authors focus on legal issues that might occur when LAWS eventually reach the battlefield, with the Geneva and Hague conventions being at the center of the evaluation, and the authors also take the different opinions, in terms of the discussions on a possible ban, into account
(Umbrello., Torres., De Bellis 2019, p. 274). Another author who has researched and focused on AWS and its relationship with the law of war and armed conflicts is Elliot Winter. When reading Winter's article The Compatibility of Autonomous Weapons with the Principle of Distinction in the Law of Armed Conflict, one can see that there is one aspect within the law of armed conflicts which Winter seems to place a greater emphasis on. According to Winter, the principle of distinction is the single most valuable aspect in terms of regulating armed conflicts, as well as for the protection of civilians (Winter 2020, p. 845). Furthermore, Winter explains that, essentially, the purpose of the principle of distinction is to distinguish between civilians and combatants, as well as civilian properties and military properties, etc., which in some instances could be rather difficult to distinguish since a person or property can
interchange between "targetable and unforgettable" (Winter 2020, p. 845). As Winter continues the assessment of AWS and the law of armed conflict, it becomes clear that the principle of distinction, even though it is one of the strongest and most valued principles within the legal regimes that regulate armed conflicts, might be tested in new ways with the emergence of autonomous weapon systems (Winter 2020, p. 848, 849). Another researcher that has been researching and evaluating the emergence of AWS in relation to the principle of discrimination (another word for distinction) is Ariel Guersenzvaig, who argues that AWS, as understood today, would most likely fail to meet the requirements of the principle of
distinction, (Guersenzvaig 2020, p. 61). On the other hand, Guersenzvaig also recognizes that even if AWS would be able to meet the principle of distinction under international
humanitarian law, there might be other possible issues in regards to situations where proper judgment is required (Guersenzvaig 2020, p. 61). However, not all researchers agree on the issues involving AWS and the laws of war; some would even argue that the laws of war could be the solution to the possible issues. Two of those authors are Kenneth Anderson and
Matthew Waxman which, in their research paper Law and Ethics for Autonomous Weapon Systems: Why a Ban Won't Work and How the Laws of War Can, argues the opposite of the previously mentioned Guersenzvaig. Instead of portraying the current legal regime as incapable of handling the potential issues in regards to the usage of AWS on the battlefield, Anderson and Waxman argue that the current laws of war are more than capable of handling these potential issues, and according to them, the principle of distinction will most definitely be able to regulate AWS in the future (Anderson., Waxman 2013, p. 11, 12, 27). Based on the previously mentioned literature in this group, it seems as if the topic of AWS, especially in relation to armed conflict, is a quite grueling area of research. One author named Sigrid Redse Johansen, who is also the last author that will be mentioned in this group, has attempted to dissect the core issues of AWS in relation to the Law of Armed Conflict or LOAC in short. First, Reduce Johansen discusses what entails autonomy and whether or not weapons can be treated as autonomous in the same way we treat human autonomy, or if autonomous weapons still should be labeled as weapons, thus being regulated by LOAC (Redse Johansen 2018, p. 90, 91). Secondly, Reduce Johansen makes an important point in regards to the question of the definition of AWS. Reduce Johansen argues that "definitions are crucial for legal regulation", and due to the fact that a universal definition of what constitutes autonomous weapon systems does not currently exist, the LOAC does not have anything to regulate (Redse Johansen 2018, p. 92). Furthermore, Redse Johansen argues that the main distress that the development and future usage of autonomous weapon systems causes in those involved in this area of knowledge could be found in the autonomy, the possibility of a weapon that could act without human control (Redse Johansen 2018, p. 96).
2.3 AWS in Relation to Law Enforcement
Since it is the topic of which this thesis revolves around, AWS in relation to domestic law enforcement will be the last group to be looked at in this chapter of the thesis. When looking at the general discussions involving AWS, one can see that the focus tends to always be placed on the presumed impact that AWS could have on armed conflict and international humanitarian laws. However, according to Cristof Heyns, there is a lack of discussions around the possible impact that AWS could have on domestic law enforcement and domestic laws, as well as on international human rights law, which Heyns argues, should not be
of the relationship between AWS, law enforcement, and international human rights law, which could be seen in his article, Human Rights and the use of Autonomous Weapons Systems (AWS) During Domestic Law Enforcement. The article aims to explore the situation and possible issues that will arise if or when, like Heyns argues, AWS will be used in policing under domestic law enforcement, and also the role of international human rights law in this future way of policing (Heyns 2016, p. 253). First of all, Heyns states the much more restrictive nature of international human rights law in contrast to international humanitarian law, especially in terms of the limited use of force, and also how law enforcement officers main obligation is to make sure that the populations right are respected and protected (Heyns 2016, p. 253). Under international human rights law, there are no clear restrictions on what weapons to develop, which is why a prohibition on AWS under international humanitarian law becomes much more important since limitations of weapons under international humanitarian law also affect the limitations under international human rights law (Heyns 2016, p. 253, 254). Thus would a prohibition on AWS under IHL automatically mean a prohibition on AWS under IHRL. Although according to Heyns, no matter what happens in terms of prohibitions on AWS in relation to IHL, AWS in law enforcement should be strictly prohibited (Heyns 2016, p. 254). Heyns recognizes that the terminology when discussing this topic is important and that the conceptualization must be able to include human rights bodies and not only IHL, which caused Heyns to decide upon using the most general term available, autonomous weapons systems, AWS in short (Heyns 2016, p. 255, 256). Furthermore, Heyns proceeds to outline the possible impact of autonomous weapons systems on human rights. According to Heyns, increased autonomy also increases the likelihood of unpredictable behavior and situations, which do not necessarily have to include a negative outcome. However, with increased autonomy comes an increased chance of violations of fundamental human rights, such as the right to life and the right to human dignity (Heyns 2016, p. 356, 362). Other important aspects in regards to this research area are the United Nations Code of Conduct for Law Enforcement Officials, as well as the principles of necessity and
proportionality under international human rights law, which Heyns chooses to dissect in his article (Heyns 2016, p. 263, 264). Literature, or previous research, specifically in regards to autonomous weapons systems within domestic law enforcement, is rare, especially in
comparison to the discussions involving these types of weapons in relation to armed conflicts. Thus, the research of Christof Heyns, and others like him, becomes crucial. Another
researcher within the field of autonomous weapons systems, domestic law enforcement, and human rights is Andrea Spagnolo. While Spagnolo acknowledges the good work of Heyns,
Spagnolo also recognizes the problem in terms of the lack of research in regards to
autonomous weapon systems and their relation to international human rights law. Spagnolo somewhat follows the same patterns as Heyns, focusing on the government's obligations under international human rights law, with a special focus on fundamental rights, such as the right to life, liberty, and security of the person (Spagnolo 2017, p. 46). Arbitrarily is an important wording in human rights treaties, according to Spagnolo, since the term gives guidelines in situations where it needs to be decided if deprivation of life was carried out within the legal framework or if the life, in fact, was arbitrarily taken (Spagnolo 2017, p. 46). To Spagnolo, much like to Heyns, the principle of necessity is most likely one of the most important principle in terms of the protection against arbitrary deprivation of the right to life, and Spagnolo mentions that, even if deprivation of life was not clearly arbitrary in legal terms, it could still be deemed arbitrary if the use of force exceeds the necessity of the situation (Spagnolo 2017, p. 47). The research area of AWS is a tricky one, which of course, is understandable since it consists mostly of predictions of a future that is yet unclear when it becomes the present. Nevertheless, many of the researchers/authors mentioned in this chapter have attempted to dissect the topic from different angles, which at least gives a sense of the common problems and issues. Although, not all researchers have the same view on what is a problem and what is not. There are definitely common themes throughout the literature, with many of the researchers making use of qualitative data by interpreting primary sources such as legal frameworks such as treaties and conventions, and also some secondary sources such as previous literature, scholarly work, and reports from international organizations. Issues involving a universal definition of AWS remain, much like the question of how international humanitarian law can regulate the usage of AWS in warfare. However, one of the more unknown aspects of future AWS usage involves domestic law enforcement, the use of force, and autonomous weapons systems impact, as well as limitations, under international human rights law.
3. Normative Framework
This chapter will present the framework that, hopefully, can help understand the possible issues involving the emergence of autonomous weapons systems in domestic law
enforcement, especially in terms of the principle of necessity under international human rights law. When conducting legal research, it is common to answer normative questions,
which creates the need for a normative framework rather than a theoretical framework, essentially a framework that focuses on the interpretations of the law in terms of good or bad, instead of trying to explain the characteristics of a law (Taekema 2018, p. 6). Furthermore, as Taekema explains, "In order to answer normative questions, a framework is needed that provides a yardstick, a set of standards or values that can serve to support a judgment" (Taekema 2018, p. 7), hence the decision to make use of the doctrine of necessity.
3.1 Doctrine of Necessity
The doctrine of necessity could be applied in different ways, although, due to the nature of this thesis, it will be applied in situations where human life is taken. Essentially, the doctrine of necessity entails the circumstances where necessity might play a role in determining whether or not taking human life was a criminal act, or in fact, a necessity (Bohlander 2006, p. 147, 148). Bohlander explains how "necessity must be clearly distinguished from the duress of circumstances", meaning that, while duress of circumstances involves outside pressure that forces a decision, necessity, on the other hand, entails a situation where moral values create choices and eventually a decision that might lead to the action of taking a human life (Bohlander 2006, p. 150, 151). States, and the law enforcement of states, can also be introduced in this doctrine according to Bohlander, and to prove that, he uses the situation involving 9/11 as an example. As the hijacked planes flew towards their targets, the
possibility to shoot down the planes did not exist, and even if there would have existed a possibility to legally shoot down the hijackers, the questions of whether or not it could be deemed necessary to kill the innocent passengers of the planes in the process (Bohlander 2006, p. 157). Now, after explaining the fundamentals of the doctrine of necessity, what remains is to further explain one of the problems that remain in terms of the application of legal and moral necessity. Once again, 9/11 will be used as an example. If the laws would have allowed for the planes to be shot down, the question remains of who would deem which action necessary and whose life should outweigh the others (Bohlander 2006, p. 158)? Bohlander mentions that some would argue that the destiny of passengers on the planes was already decided, that they would not come out of this situation alive. Thus the planes should have been taken out. However, what if the planes would have been shot down before reaching their targets and ended up killing people on the ground due to the crash? (Bohlander 2006, p. 158). If the presumed death toll in the towers would have been the same as if the plane would
have been shot down and crashed on the ground, the situation becomes even trickier, since the lives of the passengers, as well as the lives of the people in the towers contra the lives on the ground, needs to be calculated (Bohlander 2006, p. 158). Furthermore, according to Bohlander, it is also possible to make the question of necessity even more difficult. If the hijackers communicate to the authorities that they will safely release the passengers of the planes after they have dropped bombs on a building, the decision to shoot down the planes would, in this case, be the only outcome where the passengers of the planes lose their lives, and a clear question of necessity (Bohlander 2006, p. 158). Bohlander then continues to argue that these types of issues could potentially occur on an individual level as well. For example, in a situation where there is a threat towards others, and an individual might know of the possible outcome, although the individual itself is not threatened at all, is he allowed to terminate the threat if he deems it necessary, even if he is not under direct threat? (Bohlander 2006, p. 159). Besides Bohlander, others had argued and explained the doctrine of necessity, even before Bohlander shared his take on the doctrine. Two of those authors, whose work in regards to the defense of necessity in criminal law, is Edward B. Arnolds and Norman F. Garland. Arnolds and Garland produced their work in regards to necessity in 1975, 40 years before Bohlander, and one could imagine that Bohlander might have been basing some of his work on them. For example, Arnolds and Garland explain necessity as "By necessity is meant the assertion that conduct promotes some value higher than the value of literal compliance with the law" (Arnolds., Garland 1975, p. 289). Furthermore, one could imagine that Bohlander based some of his work on the works of Arnolds and Garland, for example, in terms of the shared emphasis on the important differences between necessity and excuse. According to Arnolds and Garland, the important distinguishment between the two terms is that necessity is an otherwise criminal act that becomes justifiable due to the fact that "his act of breaking the law prevents more evil than it causes", and excuses, on the other hand, entails either a mistake or a situation under duress (Arnolds., Garland 1975, p. 289, 290). Before moving on to the methodology chapter, perhaps an example of the legal use of the doctrine of necessity is of relevance. In Pakistan 1977, a coup d’etat was carried out by the army with General Zia as their leader, in order to replace the sitting prime minister Zulfikar Ali Bhutto (Stavsky 1983, p. 377-379). The question of whether or not the coup d’etat had been carried out lawfully was debated, although, the Pakistani Supreme Court eventually provided their judgement, in which they declared that the coup d’etat was perhaps not lawfully carried out, but still necessary and in the best interest of Pakistan (Stavsky 1983, p. 379-382). The court based their judgment of this situation on the doctrine of state necessity which, in order for it
to be relevant, needed to fulfill four major requirements, entailing the circumstances, last resort, proportionality, and that is was carried out due to the “exceptional circumstances” (Stavsky 1983, p. 383). This is merely one example of the legal usage of the doctrine of necessity, and there are questions whether or not this doctrine can be applied when involving state level, especially in situations involving dictatorships (Stavsky 1983, p. 393). However, the application of necessity, as explained under this section, definitely needs to be evaluated in situations of the use of force, especially involving law enforcement.
4. Methodology
According to Mark Van Hoecke, legal doctrines can take several different approaches, which have been discussed and argued by many scholars. However, one should accept that legal research can take different shapes and that the different types of legal doctrines and disciplines can coexist (Van Hoecke 2011, p. viii). As mentioned briefly in section 1.2 detailing the aims of the thesis, the research approach is not of a mere legal nature, instead, the approach will be interdisciplinary, focusing on both legal and ethical elements. The decision to make use of an interdisciplinary approach, is based on the nature of both legal and ethical disciplines which, as mentioned under section 1.2, has a tendency to overlap and work very well together. Furthermore, legal research and ethics both contain elements of normative and hermeneutic characteristics, which is yet another reason why these two disciplines are interdisciplinary and can coexist in research (van der Burg 2010, p. 2). This chapter will outline the chosen research method, as well as the chosen approach, discipline, and analysis of the selected topic in the most valid way possible.
4.1 Qualitative Method
First of all, this thesis will use a qualitative method, which essentially entails that the interpretation of data is the most important aspect of this thesis, and that the quantity of the data collected does not necessarily have to be that high. Instead, the ones that have been selected need to be interpreted and analysed thoroughly (Merriam., Tisdell 2016, p. 2). There are several ways that qualitative research can be conducted, although, one of the common aspects of most qualitative research is that the goal of this type of research method is to fully
understand the nature of a topic in depth by interpreting as well as analyzing wordings in text, documents, or even interviews and surveys (Merriam., Tisdell 2016, p. 2, 6). Another
definition of what qualitative method is was brought forth by Patton in 1985, that said, "Qualitative research is an effort to understand situations in their uniqueness as part of a particular context and the interactions there," meaning that the research method carried out was intended to be as carefully analyzed as possible (Patton 1985, Cited in Merriam., Tisdell 2016, p. 15). By choosing a qualitative research method, this thesis intends to evaluate, interpret, and analyze the laws and principles within both international and national legal systems as well as the wordings involving AWS in order to investigate the possible situation that will emerge with the introduction of AWS in law enforcement. Thus, a qualitative research method is the best-suited method. Nonetheless, there are gaps in this way of
conducting research, even if it is the best suited for this thesis. For example, as Merriam and Tisdell explain, the researcher conducting the research is the most important instrument in the process of collecting the data to analyze, mainly due to the possibility to obtain the relevant data in different ways, although, the issue of the inevitable human biases remains (Merriam., Tisdell 2016, p. 16). Furthermore, by committing to a qualitative research method, one also needs to look at the inductive process that accompanies the qualitative research method, which essentially entails that the researcher collects relevant data to create hypotheses, themes, or concepts due to the lack of established research involving the chosen topic (Merriam., Tisdell 2016, p. 17). In this thesis, the topic revolves around a future relationship between laws, principles, and the introduction of AWS in law enforcement. Due to the nature of this topic being one that has yet to occur since AWS has not yet reached the level of full autonomy, the process of this qualitative research method is inductive, with the intention of creating a hypothesis for the future relationship between human rights, and AWS.
4.2 A Normative, Interdisciplinary Approach with a Hermeneutic Analysis
This thesis takes a normative, interdisciplinary approach, focusing on different legal and ethical frameworks in relation to international human rights law, with the goal of comparing and evaluating these laws, ethics and norms in relation to the future usage of autonomous weapon systems. Van Hoecke argues that the foundation of a normative discipline focuses on what ought to be, evaluating and making decisions in regards to values, as well as
(Van Hoecke 2011, p 10). Van Hoecke mentions that the interpretations of the law in normative research can make use of external elements, from outside of the field of law, in order for the interpretations to be understood from another context, although not all
researchers support that idea of normative interpretation (Van Hoecke 2011, p. 10). To show one of the dissenting approaches to normative research, Van Hoecke uses another researcher, Hans Kelsen, as an example of a researcher within the discipline of law who "underestimated the importance of interpretation in law and the influence and the influence of non-legal elements through such interpretation", which, according to Van Hoecke, disregards the important hermeneutic feature to normative research (Van Hoecke 2011, p. 10, 11). The analysis of primary normative sources, i.e., treaty law, conventions, customary law, and general principles of law, will be the main focus, although secondary sources such as
scholarly literature on the law will also be examined and analyzed (Van Hoecke 2011, p. 11). Van Hoecke continues the discussions involving the interpretational aspect of normative sources, especially in terms of validity, essentially if the law is still relevant in the current legal regime, as well as the weight of different laws, which involves determining which law that outweighs the other (Van Hoecke 2011, p. 12). Interpretation and the weighing of one normative source against another will be at the center of this thesis. Therefore the analysis will also have elements of a hermeneutic discipline, mainly since the interpretation of texts and documents is one of the distinctive features of legal research, according to Van Hoecke (Van Hoecke 2011, p. 4). Another author that discusses the approach of normative
hermeneutics is Michal Beth Dinker, who, in the article Beyond the Normative/Descriptive Divide: Hermeneutics and Narrativity, mentions that normative hermeneutics focuses on "what ought to be", and has its basis in interpretations (Dinkler 2020, p. 128).
4.3 Material
The material collected throughout this thesis focuses on different systems of law, as well as AWS-related materials, such as documents of development and predicted future usage of AWS in law enforcement. First, the material used in the literature review all focuses on the topic of AWS since that is the topic of the thesis. The research involving the literature review was structured and divided into three groups, with the first group (The definitions of AWS and a potential ban) starting out with more general discussions involving autonomous
on AWS. The sources chosen in the first group were chosen with the intention of giving an introduction to the previous scholarly work involving AWS, and also the general discussions around the topic. The first group, and the used sources in that group, was an important step since it gives a basic understanding of the discussion around AWS development and future usage, which one could argue is crucial to this thesis topic, especially since autonomous weapons systems technology has yet to reach a sufficient status in its development and usage. The second group (AWS in relation to armed conflict) allowed for the topic to be narrowed down a bit, with the intention of creating a deeper understanding of the preemptive
discussions involving the future usage of AWS in armed conflict under international
humanitarian law and arguments to why a ban is necessary, or why some researchers find it impossible to ban something which does not have a universal definition. The sources in the second group focus mostly on the presumed relationship between AWS usage and its impact on international humanitarian law, whether or not they could coexist, or if principles under international humanitarian law, such as the principle of distinction, would be impossible to uphold when using AWS on the battlefield. Some scholars argue that the relationship between AWS and IHL is doomed to fail, and others argue that the current legal regime in the form of international humanitarian law is well suited to deal with the potential issues that might arise with the introduction of AWS in warfare. The third, and last group presented in the literature review (AWS in relation to domestic law enforcement), is as narrowed down as possible within the theme of this thesis and will focus on the emergence of AWS in domestic law enforcement and its impact on international human rights law. In the second group, when focusing on armed conflict, international humanitarian law is in focus, although the intention of this thesis is to focus on the international human rights law, which led up to the third group, focusing on the more under-researched area of AWS in relation to domestic law enforcement.
In terms of the framework chosen for this thesis, there was a question of whether a theoretical framework or a normative framework would be the best-suited way to help this thesis move in the right direction, as mentioned in the short introduction under chapter three (Normative Framework/Theory, Sanne Taekema explains "In order to answer normative questions, a framework is needed that provides a yardstick, a set of standards or values that can serve to support a judgment" (Taekema 2018, p. 7). When taking Taekemas statement regarding normative frameworks into account, the decision to choose the doctrine of necessity as the normative framework seemed like a valid option, mainly since the doctrine of necessity is
connected to the principle of necessity in some ways, but also due to the doctrine not giving any answers, but simply providing those standards of a normative framework that Taekema mentioned in the above-mentioned quotation. Furthermore, the examples that Bohlander used to explain how the doctrine of necessity could be used and what the term necessity could entail, especially in life and death situations, could be beneficial when looking at AWS in relation to domestic law enforcement and the impact on the principle of necessity under international human rights law. Lastly, the sources used for the main section of the thesis, the analysis chapter, involves legal instruments, such as international human rights law,
especially the right to life explained in article 6.1 of the International Covenant on Civil and Political Rights, the United Nations Code of Conduct for Law Enforcement Officials (1979), and the Basic Principles on the Use of Force and Firearms by Law Enforcement Officials (1990), as well as other relevant legal and ethical material provided by the UN.
5. Analysis
In this chapter, the findings of the research that has been interpreted, evaluated, and analyzed with the assistance of the selected normative framework, as well as the chosen methodology, will be presented. Furthermore, this chapter is also where the hermeneutic element of this thesis will show, in terms of the interpretations of legal frameworks in the context of AWS in law enforcement that has been carried out. All of the data that has been collected, as well as the interpretations and analysis of said data, will be presented in its entirety.
5.1 Terminology, Definition, and Possible Characteristics of AWS
Currently, there is no universal definition of autonomous weapons systems. Since there are different kinds of autonomous weapons systems with different qualities, the research field is somewhat scattered, to say the least. Nevertheless, in order to give some context to the legal frameworks that will be presented in this chapter, the first thing one has to do is to decide upon one definition to use. The decision on which definition to use is based on the context one is trying to display. Therefore, one has to decide upon the terminology first. Since this thesis is focusing on the potential use of autonomous weapons systems in domestic law enforcement, the best terminology to use is the most general term AWS, especially since
weapons within law enforcement are classified as less-lethal weapons (Heyns 2016, p. 355). Furthermore, based on the chosen terminology, the best-suited definition of AWS in terms of this thesis is also provided by Heyns, and it states, "AWS may be defined as robotic weapons that, once activated, can select and engage targets without further human intervention" (Heyns 2016, Chapter 1 in Autonomous Weapons Systems: Law, Ethics, Policy, p. 4).
Another important aspect of understanding AWS is to understand the characteristics of AWS, which is something that Christof Heyns brings forth. According to Heyns, if what he calls meaningful human control gets exceeded by fully autonomous weapons systems, there is a great threat towards fundamental human rights (Heyns 2016, p. 374). Heyns continues by arguing that human dignity and respect for human rights under international human rights law is impossible to achieve if these types of weapons reach a state of full autonomy, essentially where meaningful human control becomes irrelevant (Heyns 2016, p. 375. Meaningful human control entails that there is a human decision in the process of whom the lethal use of force should be used on and how it should be used, even though it does not necessarily mean that the human is taking the absolute final decision (Heyns 2016, p. 376).
5.2 Article 6.1 of the ICCPR and General Comment No. 36
The first relevant legal source to present in this chapter is article 6.1 of the International Covenant on Civil and Political Rights that states as follows, "Every human being has the right to inherent right to life. This right shall be protected by law. No one shall be arbitrarily deprived of his life" (International Covenant on Civil and Political Rights, 1966). Article 6.1 of the ICCPR is treated as the by far most important and fundamental of human rights, and it might not seem as there are any interpretational issues, nor many discussions around that fact. Although, one could make a case in regards to that assumption and argue that it is not
necessarily true. For example, the word arbitrarily, found in the last sentence under article 6.1, is a word that regulates the action of taking a human life. The intentions behind article 6.1 are rather clear, although, according to General Comment No. 36 (2018) on article 6 of the ICCPR, one also has to recognize that this article does not leave a huge room for
interpretation since it is the most fundamental right in order for humans to enjoy other rights (General Comment No.36 2018, para 1-3). Even though this article should be interpreted carefully, the word, arbitrarily, is one word that could be analyzed and somewhat interpreted. In fact the General Comment No. 36 places emphasis on the said word. In paragraphs 10
through 13 of the General Comment, it can be read that the right to life is not always absolute, hence the wording "arbitrarily deprived", which includes not only self-defense by private individuals, but also the use of force carried out by law enforcement officials (General Comment No.36 2018, para 10-13). Furthermore, there are several factors that are essential when trying to understand the meaning of arbitrarily, especially in terms of law enforcement, which is the focus of this thesis. One of those important factors, perhaps even the most important one, is the question of necessity. In order for a law enforcement officer to take a human life in a non-arbitrarily way, the action needs to comply with the factor of necessity, among other factors, under international human rights law as well as with other international legal frameworks (General Comment No.36 2018, para 13, 14), of which a few will be mentioned later in this chapter. When reading article 6.1 and then General Comment No. 36, one definitely gets a sense of what arbitrarily entails, mostly based on the language of General Comment No.36, since the interpretation of the article itself can only be interpreted to a certain limited extent. Looking even further into the meaning of the word, arbitrarily, especially in the context of the use of force by law enforcement officials, it needs to be stated that even in a situation where the law might allow the killing of a human being, for example in terms of self-defense, the act might still be carried out in an arbitrary way if it, for
example, does not fall under the factor of necessity (General Comment No.36 2018, para 11-13). Furthermore, the connection between arbitrary deprivation of life and the factor of necessity becomes apparent when reading two sentences under paragraph 12 of General Comment No. 36. The first sentence states that "The use of potentially lethal force for law enforcement purposes is an extreme measure, which should be resorted to only when strictly necessary in order to protect life or prevent serious injury from an imminent threat", and the second sentence that, "The intentional taking of life by any means is permissible only if it is strictly necessary in order to protect life from an imminent threat" (General Comment No.36 2018, para 12). Based on these two aforementioned sentences, one could argue that one possible interpretation would be that only when, either the law enforcement officer itself or if the law enforcement officer has knowledge of someone else being under an imminent threat, can the last resort of potential lethal use of force be resorted to. If one would accept these interpretations as valid, one would also have to agree upon the related factors that could play a part in a situation where lethal use of force might occur. For example, one could argue that, in order for lethal use of force to be allowed in a situation, besides the legal aspect, the law enforcement officer also needs to be able to properly evaluate and judge the situation,
human feelings into account and also be able to understand when the threat is imminent and when it's not, which of course, falls under the category of necessity. In general, the language in regards to the right to live according to article 6.1 of the ICCPR, accompanied by the General Comment No. 36 provided by the Human Rights Committee, is of a rather harsh nature, which one could argue is crucial since the right to life is the most fundamental human right. According to Rhona Smith, the right to life is, without question, the single most
important and fundamental human right, especially since without the right to life, no other right could be enjoyed, as mentioned earlier in this chapter (Smith 2014, p. 259). Smith also voices that the right to life creates an obligation on states to respect and protect the right to life and that the protection must be adequate, in line with international human rights (Smith 2014, p. 259). One could find similar statements in General Comment No. 36, more
specifically under paragraph 19, where the duty, or in other words, obligation, to protect the right to life with all necessary measures in terms of legal regulations, as well as limitations to the interpretational aspect to ensure that there is no arbitrary deprivation of human life under state jurisdiction (General Comment No.36 2018, para 19).
5.3 United Nations Code of Conduct for Law Enforcement Official
Under subheading 5.2 of this thesis, one could see that the right to live according to article 6.1 of the ICCPR, accompanied by General Comment No. 36, places limitations on law enforcement officials in terms of the allowed use of force that is allowed, both in law and in principles. Although General Comment No. 36 does not cover all aspects of the relationship between law enforcement officials and the right to life, therefore, the UN Code of Conduct for Law enforcement Officials, which was adopted in 1979 by the General Assembly, has an important role to play in terms of the allowed use of force carried out by law enforcement. One feature of the Code of Conduct for Law Enforcement Officials that becomes apparent when reading is that, after every article, there is a section of commentary of the forth
mentioned article, which definitely could be interpreted as a sign that the General Assembly, when drafting the document, there was a need for a clear, and structured language (Code of Conduct for Law Enforcement Officials 1979). In article 1, the term, law enforcement officials are mentioned, and in the following commentary on article 1, it is stated who is considered a law enforcement official, which essentially includes everyone who, through the law, can exercise the designated powers appointed to policing (Code of Conduct for Law
Enforcement Officials 1979). Moreover, article 2 of the Code of Conduct entails the duties and obligations for law enforcement officials to uphold human rights for all individuals under their protection, and the following commentary simply exhibits some of the national and international laws that need to be protected and respected by the law enforcement officials (Code of Conduct for Law Enforcement Officials 1979). The first two mentioned articles, and the related commentaries, are essential and provide a thorough guideline on what a law enforcement official is allowed to do within the limits of their power, as well as which human rights that needs to be abided by. However, in regards to the aims of this thesis, article 3 might be the one article that is the single most important one. Article 3 states, "Law
enforcement officials may use force only when strictly necessary and to the extent required for the performance of their duty" (Code of Conduct for Law Enforcement Officials 1979). The word 'necessity, in regards to the allowed use of force by law enforcement officials, as mentioned in this article, is a word that also was found to be essential earlier in this chapter, more specifically under subheading 5.2. Language is crucial when talking about human rights, which becomes obvious when reading and analyzing legal frameworks, and in the case of restrictions on the use of force that is allowed by law enforcement officials, the word 'necessary', seems to be one of the more important ones. When one continues to read the commentaries that follow article 3 of the Code of Conduct, one can see further linguistic explanations involving the aspect of necessity. For example, it is mentioned that the use of force is in no way prohibited, although it is strictly limited, and that the use of force may never exceed the necessary means due to the nature of a situation (Code of Conduct for Law Enforcement Officials 1979). Legal frameworks detailing the regulations involving the conduct, especially in terms of the limited use of force, are important. However, to give more context to the Code of Conduct, it needs to be said that there are flaws within these systems, as Tyler Cawthray, Tim Prenzler, and Louise E. Porter explain in their article Updating International Law Enforcement Ethics: International Codes of Conduct. For example, even though the laws exist, issues and problems in conduct by law enforcement officials continue to take place, including the use of force which exceeds the necessary limits that are regulated in legal frameworks (Cawthray., Prenzler., Porter 2013, p. 188, 189). As the above-mentioned authors mention in their article, the necessary use of force in relation to human rights is one of the most fundamental issues in connection to law enforcement, and while there are several legal frameworks that focus on regulating these issues, there might be a need for
modernization of these frameworks (Cawthray., Prenzler., Porter 2013, p. 196). Furthermore, as the climate of law enforcement continues to develop, the weapons available to the law
enforcement officials develop with it, and the authors argue that the legal frameworks as well as the ethical aspects, might need to be updated in accordance with the development of the law enforcement environment (Cawthray., Prenzler., Porter 2013, p. 197). Cawthray, Prenzler, and Porter might not mention AWS, although to exclude autonomous weapons systems from such a discussion might have dangerous implications. The article Updating International Law Enforcement Ethics: International Codes of Conduct does not touch upon autonomous
weapons systems, perhaps because it was not as relevant in 2013 as it is now, or maybe due to AWS not commonly being introduced in discussions involving its possible future usage in law enforcement. Necessary use of force has been a red thread throughout the analysis chapter, and by using the doctrine of necessity, one gets a guideline on how to interpret the word necessary, especially in relation to the use of force. For example, as explained in chapter 3, the doctrine of necessity is a framework that focuses on judgment when force is necessary and how much force is necessary, as well as the circumstantial factors that can play a role in terms of necessary use of force (Bohlander 2006, 147, 148, 150, 151). Perhaps the main aspect of the doctrine of necessity that is important to apply to this research is that situations that involve a judgment of the necessary means to protect human life rely heavily on the factor of human emotions, on human judgment. Therefore so far in this analysis section, based on the requirements involving evaluation and judgment of human emotions that is crucial in situations where the necessary use of force needs to be calculated, one could answer the first research question that was posed in the introduction of this thesis. The question states, 'In order for the right to life under IHRL to be adequately protected, should the legal use of force be modified to impose proper regulations on AWS in law enforcement?'. With the guidelines of the doctrine of necessity as the normative framework, in combination with the red thread of the implications of necessity that can be found in the legal frameworks and related commentaries that have been evaluated so far, the answer to the research question is, most likely, a yes. Based on the presumed nature and characteristics of fully autonomous weapons systems explained under section 5.1 in this chapter, such a weapon is highly unlikely to be able to properly judge a situation involving human emotions. When
interpreting the legal sources and commentaries presented so far in this chapter, in terms of the regulations, one can see that, only if there is an imminent threat towards the law
enforcement official itself, or possibly if the official knows of an imminent threat towards another person, is lethal use of force considered necessary. In a situation where this type of human judgment is crucial, an AWS would arguably be at a high risk of failing. Furthermore, in terms of the imminent threat, as long as an AWS is treated as a weapon, it could not be
under an imminent threat in the same way a human law enforcement official would be. Therefore, the answer to the first research question would be yes; one could most definitely argue that the legal use of necessary force under international legal frameworks needs to be modified in order to protect fundamental human rights with the possible emergence of AWS in law enforcement. However, that is not the only question posed in this thesis. Thus the next subheading in this chapter will deal with the second and last research question.
5.4 Basic Principles on the Use of Force and Firearms by Law Enforcement Officials The last relevant source that will be presented in this thesis is the Basic Principles on the Use of Force and Firearms by Law Enforcement Officials, which was adopted by the Eighth United Nations Congress on the Prevention of Crime and the Treatment of Offenders, Havana, Cuba 27 August to 7 September 1990. First of all, in the preamble of the document, one can see the stated importance of law enforcement officials to protect, respect, and uphold fundamental human rights, which is a clear red thread throughout the documents that mention the obligations and duties by states and law enforcement officials under the direct authority of the state (Basic Principles on the Use of Force and Firearms by Law Enforcement Officials 1990). Another red thread throughout the sources in this thesis could also be found in the preamble of the document, where the Code of Conduct for Law Enforcement Officials is referenced, specifically article 3 of the Code of Conduct which focused on the necessary use of force (Basic Principles on the Use of Force and Firearms by Law Enforcement Officials 1990). Furthermore, another paragraph in the preamble mentions and supports the Seventh United Nations Congress and the discussions held at that meeting, which involved the use of force in terms of weapons or firearms, and the limitations and restrictions of law enforcement in order to ensure that human rights are abided by (Basic Principles on the Use of Force and Firearms by Law Enforcement Officials 1990). When reading the first of the general
provisions stated in the document, one can notice a few words that need some interpretations. Besides the legal obligations brought forth in the provision which is aimed towards the state and the authorities in charge of law enforcement, there is also an obligation upon the ones in charge of the law enforcement departments to make sure that there is constant surveillance and investigations involving the ethical issues that could occur, which potentially could involve questions of necessary use of force (Basic Principles on the Use of Force and Firearms by Law Enforcement Officials 1990). However, the fifth provision is the single
most important one, at least in regards to the aims of this thesis, with special focus on
paragraph b of the said provision, which regulates the conduct of law enforcement officials in terms of respect and the upholding of the most fundamental human right, the right to life (Basic Principles on the Use of Force and Firearms by Law Enforcement Officials 1990). As mentioned, paragraph 'b', under the fifth provision should be treated as essential, especially when putting it into the context of autonomous weapons systems. Based on the previously stated characteristics of autonomous weapons systems, it is not impossible to see some potential issues involving the future usage of AWS and its impact on international human rights law and related principles, such as the fifth provision in question. Paragraph 'b' states in its entirety that law enforcement officials should "minimize damage and injury, and respect and preserve human life" when resorting to the use of force (Basic Principles on the Use of Force and Firearms by Law Enforcement Officials 1990). Necessity is not mentioned under paragraph b; nevertheless, the judgment of the necessary use of force should definitely be connected to the deeper meaning of the paragraph, especially since the wording of the provision entails that, in a situation where the currently presented provision is relevant, the factor of judgment becomes crucial to respect fundamental human rights. Furthermore, under special provisions 9 and 10 of the same document, the situations where a law enforcement official are allowed to use force in terms of firearms is when there is a calculated and judgment involving direct self-defense from an imminent threat, or if it can be judged that another person is under an imminent threat (Basic Principles on the Use of Force and
Firearms by Law Enforcement Officials 1990). Imminent is another word, besides necessity, that can be seen throughout the analysis section of this thesis, and according to Cambridge Dictionary (Online), an imminent threat is defined as a situation that will most likely occur very soon. One can somewhat understand some of the stress and difficulties involving the aspect of judgment in situations where the threat is, in fact, imminent towards oneself or another person, although, in terms of AWS, there are some potential issues that need to be raised. In the previous section, the question, 'In order for the right to life under IHRL to be adequately protected, should the legal use of force be modified to impose proper regulations on AWS in law enforcement?’, was answered, which both related to the word 'necessity, as well as the word 'imminent. Many of the issues surrounding the topic of AWS in relation to its future emergence in law enforcement involve questions of the autonomy of these weapons. For example, which role human control should play in the decision of lethal use of force, and whether or not the current legal and ethical frameworks that exist to regulate the use of force carried out by law enforcement officials, based on judgments of necessity or imminence,
