Etikk i praksis. Nordic Journal of Applied Ethics (2019), 13(1),  21–38

A Values-based Methodology in Policing

Jens Erik Paulsen

Norwegian Police University College

Professional work is currently based on explicit knowledge and evidence to a greater degree than in the past. Standardising professional services in this way requires repetitive (or at least similar) scenarios and might be seen as a challenge to professional autonomy. In the context of policing, officers perform a range of familiar tasks, but they may also encounter novel challenges at any moment. Moreover, police tasks are not well-defined. Therefore, many missions require police officers to rely on common sense, tacit knowledge or gut feeling. In this article, I argue that a values-based methodology may serve as a tool to help evaluate decisions in unfamiliar situations, to learn from experience, as well as be a quality control for established routines.

Keywords:  ethics, policing, decision-making, values, experiential learning


Most professions have made efforts to establish robust knowledge-based or evidence-based ‘best practices’ during the last three decades (Banks 2004). Within policing, for instance, this tendency is apparent in intelligence-led patrolling and in interrogation methods, such as the British PEACE (Clarke, Milne, & Bull 2011) or the Norwegian KREATIV models (Fashing & Rachlew 2009). Further, knowledge-based protocols or ‘cue cards’ (Mazerolle, Antrobus, Bennett, Tyler 2013: 41) provide step-by-step procedures for handling common, well-defined scenarios. However, the police often encounter situations for which no cue cards exist (Bittner 2005; Reiner 2010). In this article, I argue that the practice of policing will benefit from making values-based assessments, not only before and after intervening in unfamiliar situations, but also as a means of evaluating familiar routines. Standardisation of professional interventions may seem advantageous to many stakeholders (Sherman 2015), also in the case of policing. For the police leadership, standardisation enables a higher degree of organisation of street-level officers. For the public, it increases the predictability of police behaviour. For street-level police officers, standardisation provides a safeguard against making mistakes in stressful situations. On the other hand, standardisation is typically contrasted with an artisan view on policing (Gundhus 2013), which emphasises professional autonomy. Standardisation may limit the discretional space for officers – one of the hallmarks of being a professional (Freidson 2007) – by having them follow checklists meticulously. Besides, following procedures written by bureaucrats challenges the natural authority of the police. Written guidelines, even if they are knowledge-based, may be seen to represent generalised knowledge that cannot compete with the intimate, relational, contextual knowledge that only hands-on experience can produce. The expert has, after all, handled common scenarios hundreds of times and therefore readily recognises situations, discerns important nuances, and remembers what seemed to work. Expertise comprises the combination of pattern-recognition and experience-based decision-making (Klein 2009).

Expert know-how cannot simply be transferred to other persons, as is the case with ‘knowing what’ (Polanyi 1966). As it seems, every mission has a unique character, and expert choices must be made based on years of experience. According to a Norwegian study, the internal status of Norwegian expert police officers is connected to ‘experiential professionalism’, which is ‘characterized by gut feelings, hunches, intuition (rather than analysis), loyalty to colleagues, and attitudes aimed at crime control...’ (Gundhus 2013: 186). In a British setting, Loftus (2010) describes this ‘sixth sense’ as follows: ‘[…] the police learn to treat their geographical domain as a “territory of normal appearances”. Their task is to become sensitive to those occasions when background expectancies are in variance.’ The ‘ecological’ (Evans 2017: 84) rationality of experts, sensitive to what is present in and what is missing from a scenario, is intensified by a feeling of rightness (Thompson, Turner, & Pennycook 2011). This ability to discern nuances and its accompanying feeling of rightness may grow strong, almost incorrigible. Says one police officer in a Norwegian study of decision-making: ‘I do what my gut-feeling tells me to do. To this date, it has served me well… At least from my point of view’ (Sjøtrø & Olsen 2013: 54, my translation). This attitude represents a romantic, artisan view on policing. Expert knowledge only appears as ‘models’ for colleagues (Finstad 2000: 162), or in hindsight as tales of caution or success – what may be called ‘paradigm cases’ (Jonsen & Toulmin 1988: 324), providing broad pointers for the less experienced. The authority of expertise must be trusted. However, this romantic view on expertise is difficult to justify publicly.

Both the development of evidence-based practices and the artisan, naturalistic expertise hinge on the occurrence of repetitive scenarios (Kahneman & Klein 2009). Unfortunately, crime and public order disturbance can take on a wide range of forms. Bittner (2005: 161) famously wrote that police work involves ‘something-that-ought-not-to-be-happening-and-about-which-someone-had-better-do-something-now’, underlining both the emergency character and the variety of police tasks. Seemingly familiar scenarios can, without warning, become confusing. At times, even experienced police officers may find themselves in situations that can be described as ‘deceptive’ (Klein 2009: 104), ‘unfriendly’ (Shanteau 2015), or ‘weak’ (Meyer, Dalal, & Hermida 2010). As Bittner (2005) continues, public order maintenance might require the police to find solutions to ‘unknown problem[s] arrived at by unknown means’. Today, globalisation and digitalisation have created new types of crime and new criminal structures, adding new flavours to the covertness and deception that already were key elements to most forms of criminal activity.

How can good professional decisions be made when both the experiential and the scientific framework fall short? According to the literature, police officers often depend on ‘common sense’ under such circumstances (Bittner 2005; Crank 2004; Reiner 2010; Sanders 2010), a pragmatic attitude that basically involves a form of inferencing to the best explanation (Harman 1965) or ‘abductive reasoning’, seeking out the most probable solution given uncertain premises. The problem with this type of reasoning is that one’s biases are difficult to control, and what immediately seems like a ‘useful’ outcome for the involved parties appears justified without further reasoning.1 For the police, further reasoning is not necessarily required.

Authority and reasoning

The police can handle Bittner’s aforementioned emergent and pressing ‘something’ better than people in general because they are authorised to use physical force. They can terminate conflicts on the spot, not necessarily because they are experts in conflict management, but because they are authorised to physically separate and detain the parties involved in the conflict. In fact, it is their duty to create and maintain order – now, rather than later, using the means they find appropriate (OSCE 2001). However, although quick-fix authoritarian solutions may seem effective and efficient in the short-term, the long-terms effects may prove detrimental. According to several studies (Hough, Jackson, Bradford, Myhill, & Quinton 2010; Sunshine & Tyler 2003), the manner in which the officer treats the public influences public perception of the police far more than the efficiency of the police in terms of outcomes.


The claim made in this article is that a values-based methodology can both provide decision-making support in unfamiliar situations, and prove useful in evaluating procedures and routines. In other words, values may serve as a means for the police to form justifiable interventions.

Thinking as a mid-level values-based practice

Several values-based methodologies exist, for instance the ‘Potter Box’ (Potter 1965) or the comprehensive Values-Based Practice (Fulford 2008). The so-called ‘National Decision Model’ of the British police Code of Ethics (2014) is an effort to base police work robustly on values. Certain features from these models play a role in the following, but the main inspiration for the present methodology is the six-step method for providing decision support used by Norwegian Clinical Ethics Committees (CECs) since the 1990s. This methodology consists of six steps, which are: 1) Identify the problem. 2) Identify the facts. 3) Identify all involved parties and their viewpoints and interests. 4) Identify the relevant values, principles and virtues, and experience from similar situations and judicial constraints. 5) Identify possible courses of action. 6) Discuss and then formulate acceptable actions and a conclusion/ summary (Ruyter, Førde, & Solbakk 2014).


Inspired by Grimen & Molander (2008: 183) and Toulmin (1958: chapter iii), the CEC-list is transformed into a mental model, consisting of three domains (Figure 1). In this model, Steps 1 to 3 make up the situational awareness domain, whereas the decision domain consists of Steps 5 and 6.2 These domains rest on the set of relevant values (Step 4), here referred to as the value landscape (the ‘warrants’ in Toulmin’s terminology).

Figure 1

The first step (1) of the situational awareness domain (I) involves formulating one’s reason for addressing the case, that is, one’s initial moral concern (rather than CEC’s ‘identify the problem’). This might simply consist in describing and exploring one’s feeling of discomfort, or one’s moral (affective) reaction (Haidt 2012). Sometimes the concern is distinct: X has been treated unfairly! or There is a conflict between moral norms! (for example, between the obligation to help and the requirement to treat people equally). In any case, the aim of Step 1 is to explicate one’s moral preconception and one’s framing of the case.


In Step 2, factual information is written down, typically in the form of items usually included in formal police reports: location, time, persons present at the scene, and well-documented event sequences. One should also note missing information that would potentially make a difference to the case.


Step 3 comprises a list of the people (or groups of people) that are likely to be involved in, or affected by, the case. In the context of policing, the involved parties will often include offenders, victims, witnesses, next of kin, police officers, and the public. One should also indicate their immediate interests/ideas, concerns and expectations, referred to as ‘ICE’ by Matthys et al (2009). These are based upon communication, observation or, if necessary, educated guesses. Making explicit reasonable assumptions may improve the situational awareness, and strengthen the awareness of one’s own biases.3

After establishing the situational awareness domain, a suitable intervention is often already present in the police officer’s mind, but the values-based method requires that one first attempts to formulate the values that ought to matter in the situation before deciding what to do. But which values, and from whose perspective?


Some ‘mid-level’ methodologies limit the value landscape to a fixed domain of principles, as in bioethics (Beauchamp & Childress 2009) or in the ‘procedural justice approach’ en vogue in contemporary police science (Mazerolle et al. 2014). No doubt, more values, principles or duties might be of significance in the wide variety of cases the police encounter. As there are usually several parties involved in the cases, Step 4 ought to establish a set of values that includes multiple angles on the moral point of view.4

First, one should try to establish a general ‘moral point of view’ (Nagel, 1986; Williams, 1993), asking which values an impartial spectator would emphasise in the case at hand.5 Whereas Step 3 describes the particular interests (ICEs) present in the situation, the value landscape (Step 4) focuses on general long-term values considered important to an intervention. Even though moral views ​​appear to differ among people, certain values still appear to be widely shared in a given society at a given time. In the literature, such values are described as ‘common’ (Bok, 1995), or as part of the ‘common morality’ (Gewirth, 1993).


According to the Norwegian Police Act, the police shall protect some common values, through contributing to ‘promoting and consolidating citizens’ legal protection, security and general welfare’ (Section 1). More specifically, in accordance with a comprehensive empirical study (Haidt, 2012, p. 146) and regardless of cultural background, people all react to the same five moral challenges: harm/care, cheating/fairness, betrayal/loyalty, subversion/authority, and degradation/sacredness.6 Every community – or individuals living in a community – grapple with these moral dimensions, although their relative importance may vary among individuals and communities. In this manner, the central values (care, fairness, loyalty, authority, and sacredness) all seem to count as prima facie norms within a common morality.


Other values play special roles within specific groups. Professional duties of confidentiality (or a Mafioso code of omertà) are, for instance, usually stricter than everyday notions of secrecy. Some values (and virtues) are of special importance or play a different role within a profession, even though they are referred to by the same terms as their everyday counterparts. Therefore, when outlining a value landscape as a basis for intervention, there are good reasons for stating such special group values separately.


Several special values govern police work (in addition to the abovementioned common values). These values typically characterise the manner in which a police officer interacts with citizens. The Norwegian Police Act states, for instance, that the police shall act in a ‘business-like and impartial manner and with consideration for persons’ integrity […]’ (Section 6). The Police Instruction also stresses the importance of truthfulness, confidentiality, moderation, efficiency and trust, as well as the need to respect human rights and the individual’s dignity. In addition, the police force’s ethical guidelines (Politidirektoratet 2018) identify tenets like openness, solidarity, courage, and holistic orientation as important in policing.7 These are also considered common values, but they take on a special meaning among professionals.


Special values also pertain to other more or less distinct groups of people. Some gain special status due to age or illness, others because they differ from the majority culture in terms of faith, ideology, or customs. Whether or not special values and vulnerabilities ought to influence police intervention is controversial from a justice perspective. Nonetheless, most people readily agree that some groups (e.g. young people) deserve special attention or must be approached differently than people in general. Avoiding stereotyping or prejudice is of course crucial when interpreting group values. Spelling out beliefs about one’s values seems better than just letting prejudice tacitly influence one’s intervention. Special status typically affects the manner of police intervention, not the outcome.


Within the decision domain (III), the first task (Step 5) is to formulate possible interventions based on the value landscape. If a suggested intervention is considered a reasonable expression of the value landscape of the case, this suggestion qualifies as ‘values-based’. In most cases, several possible interventions will pass the value criterion. To determine which choice is morally optimal requires scrutinising both the intention and the consequences of the various suggestions. This is the aim of Step 6. This methodology represents a ‘mid-level’ theory,8 where the theoretical deliberations are condensed into a set of questions covering central tenets of deontology and consequentialism.9 The intention check of an action consists of asking five questions, intended to clarify the intention’s consistency, benevolence, public appeal, possibility, as well as its legality.10 More specifically the questions are:

i)     Is the suggested line of action always appropriate under similar conditions?

ii)    Does the intervention imply that the police officer uses his or her professional authority in the best interest of the clients and not just as a means to fulfil other goals?

iii)   Can the intervention withstand public scrutiny?

iv)  Are there sufficient resources and competence available to carry out the intervention?

v)    Is the intervention legal?

If the answer is ‘no’ to any of these questions, the suggested intervention should be rejected or modified. If it is considered permissible by the deontological criteria, its consequences are estimated. One may protest that the very order of assessment involves a deontological bias, but the police should hardly act inconsistently (i), punish clients (ii), commit shameful choices (iii), attempt the impossible (iv), or perform illegal acts (v) in the first place.


Studying the consequences involves estimating how the various proposed interventions affect the parties involved, both in the short term and in the long run, keeping in mind possible side effects, and taking into account the probabilities of the outcomes. The optimal choice is the one that shows the most positive aggregated outcome. Admittedly, the precision level in these consequentialist assessments may be low. The most important point, however, is probably that the police officer actually tries to consider the different points of view, and conscientiously estimates the different outcomes for each party. By mapping the hypothetical and actual outcomes, favouritism or disregard of the welfare of some of the involved parties may be exposed. Reasoning about possible outcomes is taxing and requires both imagination and estimates of probabilities, but even if imprecise, it provides a good point of departure for further discussion and learning. Besides, common sense assessments often fail to take long-term effects or side effects into account.11

The last step (7)12 consists simply in making a brief summary of the process. The purpose is to secure the coherence of the reasoning. The following format is one way of summarising the argument: In this case where <based on Step 2>, the moral concern was initially considered to be <Step 1>. The most important involved parties were <Step 3>, and the most central values to the case were <Step 4>. Based upon these values, several actions <Step 5> were proposed (A1…An). Ax was deemed permissible <Step 6> according to the intention test and judged to entail the best overall consequences.

Two bottlenecks

My own anecdotal experience with this methodology in clinical ethics committees as well as in police education has revealed two bottlenecks in the methodology. The first occurs upon entering the value landscape. Identifying the values relevant to the case may sometimes feel like guesswork to the practitioner. This feeling is often exacerbated by a lack of familiarity with discussing norms and values. Even seasoned practitioners may find naming the central professional values difficult. Are values just any positive feeling? Are the values simply the interests, concerns and expectations of Step 3? Scheffler’s criteria for valuing (Scheffler 2010: 29) may help differentiate these tenets to some degree. In his view, what we value is not just any idea of something commonly believed to be good – it must also produce a range of ‘merited’ emotions, and be capable of motivating action in relevant deliberative contexts. In this manner, the importance of values seems highly context dependent. For instance, to chess players the idea of loyalty or solidarity may not produce any merited emotions – or motivate professional action – but to police officers or soldiers, loyalty may make the heart beat faster and can make a real difference to the officer’s decision-making. An understanding of the role played by different values in different groups is therefore important, both to the understanding of ICEs in Step 3 and for the long-term values in step 4.


What are the typical moral values in our context? There is of course no universal answer, but often clues are present in the previous steps, indicating where to start looking. For instance, if Step 1 describes a feeling or opinion of something being unjust, then fairness or justice seem like reasonable values to consider. Some of the short-term interests, concerns and expectations of the various involved parties may also provide clues to longer-term values. For example, the offender may be concerned about how to evade the situation (freedom), and the victim about getting even (justice). In matters of crime and disorder, common values such as fairness/justice, compassion/care, and safety are often likely candidates for common values. Considering the special values of the law enforcement profession, these are, as shown above, stated in various Acts, ethical codes and guidelines. Familiarity with the formal and informal values of the profession is indispensable.


After drafting the value landscape, it may prove helpful to link the identified values of the value landscape to the involved parties of the situational awareness domain. Fairness may be of particular relevance to a victim and a bystander, whereas solidarity may seem crucial to the police officer. If you cannot connect a value to any involved party, the value can probably be eliminated. Likewise, if an involved party cannot be linked to any of the values, he or she will probably have little impact on the decision. Linking involved parties and values precludes the set of values from seeming like pure guesswork.


The second bottleneck arises when the practitioner looks for interventions (Step 5) that express the value landscape (Step 4). Several strategies can be applied. In an after-action reflection, one may start by assessing whether the chosen action expresses the value landscape. If the intervention fails to express the selected values criteria, it can be modified until it does so. If the actual intervention proves to be an impossible point of departure (or if the values-based methodology is applied in a planning phase), one might simply envision interventions that reflect the selected values, that is, work from values. For instance, one can start by asking how justice can be expressed in the situation at hand, keeping in mind that values can inform both what is done (i.e. reaching a just solution), as well as the manner in which the action is performed (i.e. letting the offender present his or her side of the story). After a just intervention is described, one adds the next value, until all the values are included. In practice, if the number of values exceeds four or five (which is often the case) the task becomes too hard. Selecting the three or four most crucial values is therefore appropriate.13 Alternatively, one can follow a Recognition-Primed Decision model (Klein 2009: 90) and suggest an intervention from one’s experience with similar situations, and then modify it until it coheres with the value landscape. If coherence proves impossible to establish, one envisions another line of action14.

As we can see, Step 5 typically involves a process of conjecture and refutation informed by experience. This is quite all right, as long as the value landscape is explicit and serves as a first criterion for the approval or disapproval of a suggested intervention.

Example case

What does the values-based methodology look like in practice? The following case, inspired by a police student’s experience during practice, demonstrates values-based after-action reflection. Here, a police officer found herself in a situation where she initially felt there was conflict between benevolence and justice (Step 1). She was unsure whether her solution was optimal and decided to review the case using the values-based methodology. The facts of the case (Step 2) were as follows:

‘I was heading towards the city centre when I noticed a car driving unsteadily in front of me. I flashed my lights and the driver stopped immediately at the side of the road. A quick check of the registration plate came up empty on both the car and the owner. I wanted to check the condition of the driver, so I walked up to the car and asked him how he was doing. The driver, a man about 40, answered: “I’m fine, constable. If you thought my driving was sloppy, I agree, I have not driven a car with manual transmission in ten years. My foot slipped on the clutch pedal during a gear shift and I got a little stressed. I am sorry! You see, it is my brother’s car… I can assure you that the vehicle is in excellent shape – but me, I’m simply not familiar with manual gear boxes.” I asked for vehicle documentation and his driver’s license. He found the vehicle registration certificate, but his driver’s licence had been stolen at the train station when he arrived yesterday, on his annual summer visit to his brother. He claimed that he was actually in the process of going to the police station to report the theft of his wallet (also containing his credit card), and to obtain a temporary driving licence’.

Initially, the police officer saw herself in the role of a ‘law officer’ (Banton 2005) and considered penalising him, but during the encounter, she became increasingly sympathetic to his predicament. In the end, she chose to take a compassionate line. She decided to believe his explanation and to help him. She applied a kind of abductive reasoning (Evans 2017; Harman 1965), commonly used in situations involving uncertainty, and held that her line of action seemed to be to the benefit of all parties.


Step 3 – the involved parties and their ICEs – can be summed up as follows: The driver is obviously involved, and his agenda is purportedly to report the theft of his wallet and to get a temporary driving licence. The police officer’s initial focus was on road safety, but during the encounter, the driver’s licence and his predicament became her focal points. The driver’s brother plays no active role in the story, but counts as an involved party through his apparent ownership of the car. The driver’s erratic driving affects other road users. Other affected parties might exist, but for now we consider the situational awareness domain complete, and move on to the value landscape (II).


First, the relevant common values are considered. In general, road safety seems central. The citizens probably also take an interest in how the police intervene in such situations. Is the driver treated fairly, that is, is the police’s reaction proportionate to his offence, given its context? Fairness seems to include both the element of compassion as well as equality in this case.


In addition, some special professional values matter here. The police officer must consider the seriousness of his offence. Is the case worth pursuing, and if so, how much time to spend on it (efficiency, prioritisation)? Further, car thieves are not necessarily honest with police officers, so to what extent can she trust his story? Alertness or thoroughness therefore seems important. As the police officer chooses to overlook traffic regulations, it seems reasonable that she – in order to maintain solidarity with her fellow officers – provides acceptable reasons for doing so. We should therefore include ‘fellow police officers’ as an involved party in Step 3. The value landscape thus contains at least safety, equality, fairness, compassion/care honesty as common values. The professional values include efficiency, alertness/thoroughness, and solidarity. These form the basis for deciding how to intervene (in this example, the values of ‘other groups’ seem irrelevant).


To avoid the first bottleneck, we link the selected values to the involved parties. Fairness matters for the driver. The police officer can be linked to all the professional values mentioned, as well as compassion/care and safety. We know very little about the brother, but honesty seems at stake both for the driver and his (alleged) brother. The trust of the police officer also plays a role here. For the road users as representatives of the public, safety and equal treatment seem crucial.


All values and involved parties are now connected. Before proceeding to the action domain (III), we consider which values are crucial to the case. Here, road safety can hardly be disregarded. In addition, fairness (which harbours elements of both equality and compassion) seems important to all the involved parties. If the police officer chooses to pardon (or extend good will to) the driver, she ought to be able to substantiate the driver’s story somehow. Therefore, alertness or ‘thoroughness’ seems central to finding a fair solution. The police officer’s reasons for using discretion must also be acceptable to her colleagues. In this sense, solidarity among colleagues seems to be at stake. A good solution must express these four values, which are all fundamental to policing.


The first task in the decision domain (Step 5) consists in finding interventions that express safety, fairness, thoroughness and solidarity. As the present case is an after-action reflection, the original choice can serve as a point of departure. The officer thought her original choice seemed fair, even compassionate, towards the driver. However, she now realises that her decision failed to pay enough attention to safety, solidarity or thoroughness. Can her action be modified so that it expresses the value landscape? If she had checked the driver’s identity more carefully, perhaps by phoning his brother/owner of the car, it would have been a more thorough solution. If she had escorted the driver to the police station, she would have addressed road safety to a greater extent as well. This modification of the original solution would therefore be more in line with the proposed value landscape, and would probably also be more acceptable to her fellow officers.


An alternative intervention can be constructed by taking the selected values as a point of departure. For instance, if equality (equal treatment) matters most, it may imply that the driver ought to be penalised (since other drivers were penalised the week before, during an extensive document control). Further, road safety might be maintained by instructing him to park the car and make his way to the police station by other means. This solution would also express thoroughness, and the police officer does not risk being fooled by some potential tall tale. If the police officer let him hitch a ride with her to the police station, this solution would also express compassion.


A third possible intervention is identical to the second, except for not letting the driver hitch a ride to the police station. Here, the police officer expresses compassion simply by acknowledging the driver’s predicament and by taking the time to explain the reasons for having to maintain a strict line.


We have now (Step 5) identified three interventions that all arguably express the key values. For the sake of brevity, we will limit ourselves here to these three suggestions. Next, the intentions and consequences of these three values-based interventions are scrutinised (Step 6), in order to decide which proposal is the optimal solution.


In the first suggestion, given that the driver’s story is supported by her communication with the car owner, or by the driver’s responses to her control questions – the police officer just escorts the driver to the police station, without any formal reaction. Is this intervention acceptable with regards to intention? (i) We first consider whether this line of action would always be appropriate under similar conditions. A Kantian manner of expressing the ‘line of action’ (O’Neill 1998; Pogge 1998) consists in formulating the type (t) of action, its purpose (p), and its circumstances (c). Here, the line of action can be formulated as ‘(t) ignore traffic misdemeanours, (p) for reasons of fairness, (c) if the offender admits his fault and is in the process of making amends’. However, following this line consistently seems counterintuitive. It would lead to pardoning all speeding drivers that slowed down at the sight of the police, for instance, thus undermining the traffic regulations. However, in the present case, given the on-site databases available to the police, the need for physical documentation is in most cases obsolete. If the driver’s identity and licence to drive a car can be established without much ado, this option might actually pass the first test. From this point of view, the line of action to maintain can be specified as ‘(t) Ignore certain traffic misdemeanours such as a missing driver’s licence, (p) for reasons of fairness, (c) as long as the driver’s identity and driving permission can easily be verified through other means’.


(ii) Next, one considers whether the proposed intervention implies that the police officer uses her professional authority in the best interest of the ‘client’, and not to fulfil some other goal. If the client15 in this case is the driver, the police officer appears to display benevolence (unless her motivation is to avoid paperwork). Third, does this action pass the ‘daylight test’? It probably would, but it might seem unjust to other people that under similar circumstances had been penalised. However, if the technological development (on-site database access) is taken into account, and the practice is followed in the future, this option might pass this test. Concerning the resource question (iv), no extra resources are required, and lastly (v), the action is legal as police officers are allowed use discretion under such circumstances. However, adjustments to the legal framework might be necessary if her benevolent line of action officially became the routine.


The second intervention involved penalising the driver for not carrying his licence, instructing him to park the car, and driving him to the station, so that he could report the theft and apply for a temporary driving licence. As the police station is nearby, and the police officer was en route to the station, driving him there represents just a small favour. Again, we first formulate its ‘rule’, that is ‘(t) rendering small services to offenders after penalising them, (p) in order to express compassion, (c) given that the offence is minor/understandable’. Is this line of action permissible? In other words, can all offenders expect small favours from the police, as long as their offence is minor or understandable? The answer is probably no. A problem with this line of action is that the officer is both extending good will and refusing to give the driver some leeway. This represents a two-faced intention, probably untenable as a general rule. If we hold this view, this option fails the deontological permissibility test.


The third and final intervention resembled the second one, but expressed compassion by explaining why she had to penalise him and in acknowledging his predicament. (i) The rule can here be formulated as: ‘(t) Penalise any driver encountered without a valid licence, in order to (p) maintain law and order, (c) no matter what’. The officer is certainly permitted to adhere to this line consistently. The crucial question is whether this option really expresses the driver’s ‘best interest’ in the sense of question (ii)? The answer is, I think, a qualified ‘yes’. Through acknowledging the driver’s situation, and by arguing that any responsible adult has to abide by the law, i.e. by not making an exception, she emphasises his autonomy. Certainly, there must have been alternative ways to solve his problem. This action probably passes the daylight test (iii), too. After all, police officers are supposed to uphold the law, not challenge it. The resources (iv) are present as long as the officer has the communicative skills necessary. Lastly, it is certainly legal (v) to enforce the law strictly.


After the deontological intention test, the first and the third solutions seem permissible. Next, we estimate the consequences of these two solutions in order to find the optimal solution.


The consequences of the first intervention (no formal reaction) seems largely positive: the driver will probably experience positive effects both in the short and long term, with no noticeable side effects. The police officer may experience a small positive short-term effect from being compassionate, whereas the long-term effect is probably negligible. By ascertaining the driver’s identity, the consequences for the brother (if he exists) are likely to be positive. The public will probably react positively to the flexible attitude of the police, though one might expect that people recently penalised in similar situations are likely to be critical. There are several possible side effects. The driver may get the impression that getting away with misdemeanours is a matter of telling a good story. Some of her colleagues might find this officer lenient – even disloyal – if they learn about her course of action. From a long-term perspective, this option may also generate public expectations and challenge the Traffic Law.


If the third suggested intervention is chosen (the strict response), the driver will probably react negatively in all aspects (though he might learn a lesson). The strict response will probably not have any big impact on the public. The brother, although he might endorse the vigilance of the police, will probably – as a brother – consider the officer’s response unreasonable. From a long-term perspective, the strict approach may be advantageous to the police, not only because it is easier to maintain fairness (equality) in a strict manner, but perhaps also because document controls provide the police with a means to legitimately contact, probe or control drivers.


The comparison indicates that the overall consequences are favourable in the first suggested intervention, despite the possible side effects. Admittedly, the level of precision of the consequentialist assessment is quite low. The main point is to view the suggestions from a wider perspective by assessing the various consequences for the involved parties as conscientiously as possible. Otherwise, it is easy to limit the outlook to how the solution affects oneself (or one’s profession) in the short term.


The last step (7) consists of a report-style summary. The point is to clarify the intervention and its reasons, so that its credibility can be easily checked by others. In the present case, it might read like this:

The case involved a driver who drove unsteadily because he was unfamiliar with the car. Nor could he produce his driver’s licence as his wallet was stolen the day before. The case involved the driver, his brother, the police officer, colleagues, and road users in general. Fairness, safety, solidarity, and thoroughness were considered the most central values. The morally optimal solution consisted in not penalising the driver, but checking his identity thoroughly, and escorting him to the police station afterwards. This option expresses the central values, passes the intention test, and is considered optimal because of its better consequences overall.

The point and scope of the method

The example demonstrates the application of the values-based methodology. Through mid-level theorizing, it provides both decision-making support and quality assurance. As mentioned above, the need for methodical thinking typically arises when there is no ‘neurological programming’ or experience-based response that fits the current task (Evans 2017: 16). This is often the case for students or rookie professionals. Values-based reasoning is also helpful for experienced police officers encountering unfamiliar situations. As a tool for after-action reflection (that is, experiential learning), it is useful, no matter what decision-mechanism was originally used.


The values-based methodology does not produce final answers, rather it generates transparently construed, criticisable suggestions for actions. Therefore, a values-based methodology is suitable as a tool for discerning and explaining differences of professional opinion, as it helps explain the reasons for disagreement, referring to (i) framing, (ii) relevant facts, (iii) involved parties, (iv) values or (v) interventions. Mere feelings, traditions, or individual status do not count. Disagreements are discussed in terms of differing premises and inferences, not as differences in personal preferences or character. The aim of values-based methodology is thus discursive rather than decisive. Regarding the present case, another police officer might well argue that other values are more central than the ones favoured here. He or she may also argue that the second suggested intervention actually passes the intentional test, because the favour is negligible, and so on. This approach contrasts sharply with the tacit, experiential view on knowledge of the ‘romantic’ professionalism outlined above.


Granted, the values-based approach contains some obvious limitations. First, one might question whether a values-based methodology is useful as an instrument for in-action reflection. The methodology probably requires an amount of cognitive capacity that is in short supply during the dynamic part of a mission. The need for reasoning is rarely considered in stressful situations (Evans 2017: 107). Indirectly, however, values-based reasoning may still prove helpful to in-action reflection. If applied systematically, the methodology provides a tool for establishing a library of paradigmatic cases, thereby increasing the possibility for near-instantaneous (pattern-) recognition when encountering similar challenges during a mission.16

Secondly, the decision domain as outlined here – particularly Step 6 – may prove problematic for several reasons. Intentions, for instance, may become difficult to assess when a mission contains several (perhaps) overlapping phases, each demanding a separate assessment. Also, if the case involves a large number of involved parties, consequentialist estimates may prove unmanageable. However, under such circumstances one may still envision interventions, based on the set of values (Step 5). Expert judgment or even a voting process can substitute for Step 6 under such circumstances. The important point is to keep the choice of value-criteria and the selection of intervention separate (Sunstein & Hastie, 2015).


The basis of this methodology – values – differs from evidence-based or knowledge-based practice. Still, the ‘evidentially’ best approach may indeed express the value landscape. Including such best-practice interventions in Step 5 seems prudent, although they are not necessarily the morally optimal solutions. One should also keep in mind that evidence in many cases involves deep considerations of context and of framing, hence of values. Fulford, in his values-based practice approach, actually emphasises the connection between evidence and values-based strategies: ‘The two feet principle is that all decisions, whether overtly value-laden or not, are based on the two feet of values and evidence: [in practice,] this translates into the reminder to “think facts, think values”’ (Fulford 2008).


Experiential knowledge is crucial, too – in assessing the interests, concerns and expectations of the involved parties, in suggesting possible interventions and in assessing consequences. One might say that the methodology is a way to develop experiential knowledge through the prism of common and special values.


To conclude, the methodology as described above seems particularly useful for after-action reflection, or what Schoen (1983) labels ‘on-action’ reflection (experiential learning). Applying the values-based methodology might feel like a challenge to the professional autonomy, or even the authority, of experienced service personnel17. However, if professionalism entails constant pursuit of excellence, this seems like a challenge well worth taking.18


1 See Rasmussen (2017) for medical research on this topic. See also Grimen & Molander (2008) for a more detailed summary of different types of ‘tacit knowledge’.

2 There are also substantial changes made to steps 1, 4 and 6.

3 For instance in the form of stereotyping or conjunction fallacies (Kahneman 2012: 158).

4 The wide scope of the CEC’s step 4, which included ‘experience with similar situations’ and ‘judicial constraints’ is narrowed down here to focussing on values.

5 That is, trying to establishing an ‘outsider’ (or ‘outsight’) point of view (Toulmin, 1961).

6 Haidt later added the liberty/oppression dimension. In a vaguely similar fashion, John Kleinig (2009) highlights ‘vital welfare interests’: physical integrity, mental integrity, autonomy, freedom, dignity, and privacy.

7 Informal or even ‘hidden’ values exist, also in the police culture. To what extent these ought to be included in the value landscape is not clear. In an after-action reflection, such values may have considerable explanatory force in understanding the chosen intervention. One should also be aware that different types of police work (investigation, crime prevention, order maintenance) may require different sets of (special) values.

8 As opposed to mere application of ‘high’ theory on the one hand, and particularism on the other.

9 Aristotle’s words are worth keeping in mind, that ‘our discussion will be adequate if its degree of clarity fits the subject-matter…’ (Aristotle 1985: 1094b).

10 These questions overlap to some degree the ius ad bello questions of the just war tradition (Bellaby, 2012), but their original inspiration is, of course, the various formulations of the Kantian Categorical Imperative (Kant, 1981) and the requirement for public access to maxims from Zum Ewigen Frieden (Kant, 1881).

11 This is inspired by a Benthamite act-utilitarian approach (Troyer, 2003).

12 This step is included in Step 6 in the CEC model.

13 The excluded values do not have to be suppressed from the final consideration of an option, but may provide additional support in Step 6.

14 Also, a more open-ended ‘garbage can’ strategy may also be attempted (Einsiedel Jr. 1983).

15 The public can, of course, also be seen as the true client of the police.

16 Associative or casuist reasoning provides a natural extension of values-based reasoning in emergencies or stressful situations.

17 Much in the same manner that medical doctors are reluctant to adopt a procedural ‘cookbook medicine’ (Knaapen 2014)

18 Thanks to Joshua M. Phelps, Trond Kyrre Simensen, and Pål Frogner for discussions and comments on this manuscript.


Aristotle. (1985). Nicomachean Ethics (T. Irwin, Trans.). Indianapolis: Hackett publishing company.

Banks, S. (2004). Ethics, accountability and the social professions. Basingstoke: Palgrave.

Banton, M. (2005). The police as peace officers. In T. Newburn (Ed.), Policing: key readings (pp. 132-136). Devon: Willan Publishing.

Beauchamp, T. L., & Childress, J. F. (2009). Principles of Biomedical Ethics (6 ed.). New York & London: Oxford University Press.

Bellaby, R. (2012). What’s the Harm? The Ethics of Intelligence Collection. Intelligence and National Security, 27(1), 93-117.

Bittner, E. (2005). Florence Nightingale in pursuit of Willie Sutton: a theory of the police. In T. Newburn (Ed.), Policing: key readings. Devon: Willan Publishing.

Bok, S. (1995). Common Values. Columbia, Missouri: Missouri Press.

Clarke, C., Milne, R., & Bull, R. (2011). Interviewing Suspects of Crime: The Impact of PEACE. Training, Supervision and the Presence of a Legal Advisor. J. Investig. Psych. Offender Profil, 8, 149-162. Crossref

College of Policing (2014). Code of ethics. A Code of Practice for the Principles and Standards of Professional Behaviour for the Policing Profession of England and Wales. Retrieved on 14 May 2019 from

Crank, J. P. (2004). Understanding Police Culture. London & New York: Routledge.

Einsiedel Jr., A. A. (1983). Decision-making and problem solving skills: the rational versus the garbage can model of decision-making. Project Management Quarterly, 14(4), 52-57.

Evans, J. S. B. T. (2017). Thinking and Reasoning: A Very Short Introduction. Oxford: Oxford University Press.

Fashing, I., & Rachlew, A. (2009). Investigative interviewing in the Nordic region. In T. Williamson, B. Milne, & S. P. Savage (Eds.), International Developments in Investigative Interviewing. Cullompton: Willan.

Finstad, L. (2000). Politiblikket. Oslo: Pax Forlag A/S.

Freidson, E. (2007). Professionalism the third logic. Cambridge, UK Malden, MA: Polity Press.

Fulford, K. W. (2008). Values-Based Practice: A New Partner to Evidence-Based Practice and A First for Psychiatry? Mens Sana Monographs, 6(1), 10-21. Crossref

Gewirth, A. (1993). Common Morality and the Community of Rights. In G. Outka & J. P. Reeder (Eds.), Prospects for a Common Morality (pp. 29-52). Princeton, NJ: Princeton University Press.

Grimen, H., & Molander, A. (2008). Profesjon og skjønn. In A. Molander & L. I. Terum (Eds.), Profesjonsstudier (pp. 181-196). Oslo: Universitetsforlaget.

Gundhus, H. I. (2013). Experience or knowledge? Perspectives on new knowledge regimes and control of police professionalism. Policing: a Journal of Policy and Practice, 7(2), s. 178-194. Crossref

Haidt, J. (2012). The righteous mind: why good people are divided by politics and religion (1st ed.). New York: Pantheon Books.

Harman, G. H. (1965). The Inference to the Best Explanation. The Philosophical Review, 74(1), 88-95. Crossref

Hough, M., Jackson, J., Bradford, B., Myhill, A., & Quinton, P. (2010). Procedural Justice, Trust, and Institutional Legitimacy. Policing, 4(3), 203-210.

Jonsen, A. R., & Toulmin, S. (1988). The Abuse of Casuistry: A History of Moral Reasoning. Berkeley: University of California Press.

Kahneman, D. (2012). Thinking, fast and slow. London: Penguin Books.

Kahneman, D., & Klein, G. (2009). Conditions for Intuitive Expertise: A Failure to Disagree. American Psychologist, 64(6), 515-526. Crossref

Kant, I. (1881). Zum Ewigen Frieden. Ein philosophischer Entwurf. Leipzig: Phillipp Reclam jun.

Kant, I. (1981). Grounding for the Metaphysics of Morals. In M. L. Morgan (Ed.), Classics of Moral and Political Theory (pp. 991-1041). Indianapolis/Cambridge: Hackett Publishing Company.

Klein, G. (2009). Streetlights and Shadows: searching for keys to adaptive decision making. Cambridge, MA: The MIT Press.

Kleinig, J. (2009). The Ethical Perils of Knowledge Acquisition. Criminal Justice Ethics, 28(2), 201-222.

Knaapen, L. (2014). EvidenceBased Medicine or Cookbook Medicine? Addressing Concerns over the Standardization of Care. Sociology Compass, 8(6), 823-836. Crossref

Loftus, B. (2010). Police occupational culture: classic themes, altered times. Policing and Society, 20(1). Crossref

Matthys, J., Elwyn, G., Van Nuland, M., Van Maele, G., De Sutter, A., De Meyere, M., & Deveugele, M. (2009). Patients’ ideas, concerns, and expectations (ICE) in general practice: impact on prescribing. Br J Gen Pract, 59(558), 26-36. Crossref

Mazerolle, L. G., Antrobus, E., Bennett, S., & Tyler, T. R. (2013). Shaping Citizen Perceptions of Police Legitimacy: A Randomized Field Trial of Procedural Justice. Criminology, 51(1), 33-63.

Mazerolle, L. G., Sargeant, E., Cherney, A., Bennett, S., Murphy, K., Antrobus, E., & Martin, P. (2014). Procedural justice and legitimacy in policing. Cham: Springer.

Meyer, R. D., Dalal, R. S., & Hermida, R. (2010). A Review and Synthesis of Situational Strength in the Organizational Sciences. Journal of Management, 36(1).

Nagel, T. (1986). The view from nowhere. New York: Oxford University Press.

O’Neill, O. (1998). Consistency in action. In P. Guyer (Ed.), Kant’s Groundwork of the Metaphysics of Morals. Critical essays. Lanham: Rowman & Littlefield Publishers.

OSCE (2001). The European Code of Police Ethics. Council of Europe Publishing.

Pogge, T. W. (1998). The Categorical Imperative. In P. Guyer (Ed.), Kant’s Groundwork of the Metaphysics of Morals. Critical essays. Lanham: Rowman & Littlefield Publishers.

Polanyi, M. (1966). The Tacit Dimension. Chicago and London, The University of Chicago Press.

Politidirektoratet (2018). Etiske retningslinjer for politiet (Ethical Guidelines for the Police). Oslo, Politiet.

Politiinstruksen (The Police Instruction) (1990). Retrieved on 14 May 2019 from

Politiloven (The Police Act). (1995). Lov om politiet (LOV-1995-08-04-53). Retrieved on 14 May 2019 from

Potter, R. B. (1965). The structure of certain American Christian responses to the nuclear dilemma, 1958-1963. (Doctoral thesis), Harvard University.

Rasmussen, E. B. (2017). Balancing medical accuracy and diagnostic consequences: diagnosing medically unexplained symptoms in primary care. Sociol Health Illn., 39(7), 1227-1241. Crossref

Reiner, R. (2010). The Politics of the Police (4 ed.). Oxford: Oxford University Press.

Ruyter, K. W., Førde, R., & Solbakk, J. H. (2014). Medisinsk og helsefaglig etikk (3. utg ed.). Oslo: Gyldendal Akademisk.

Sanders, B. A. (2010). Police chief perceptions of good policing in non-urban departments.