Introduction

Ethnographic methods of data collection such as observations or interviews are well-known and frequently used in the field of Criminology. They are being conducted to study various aspects of deviant behaviour, life in prison, criminogenic places, crime control measures and many more. There are various methods for analysing the data collected this way, some more popular and frequently used, others more specialized and unknown. The goal of this paper is to present two approaches to interpreting qualitative data that are rather new to criminological research and to show how they may complement each other. While ethnographic semantics is a recognized method in the field of ethnography, it is less known in criminology. Next to that, the documentary method, a common German sociological methodology, has rarely been received outside the German-speaking region or in the field of criminology in general. The advantages and practical application of both will be illustrated using data on German Municipal Law Enforcement Services (MLES). Triangulating the two approaches, and that is what is proposed in this article, will allow for a deeper understanding of the cultural practices, everyday routines, and implicit knowledge of MLES.

The police are under constant scrutiny both from civil society and academics from a wide range of research disciplines to such an extent that researchers at times have difficulty gaining access to the coveted police research field. Research interests range from role perception, discrimination, violence by and against police officers, and police error culture to organisational culture, training, and policy, to name a few. This diverse attention is comprehensible when one considers the position of the police, who often carry out their duties in public and hold a monopoly on the use of force. Compared to the publicity the police receive, other actors in the field of urban security are perceived far less. This applies to private security companies as well as law enforcement officers from local authorities, the study of which has not yet been established in the research of policing. The lack of interest in the latter is to some extent surprising, given that MLES have long been present in German cities, operate almost exclusively in public space, and in some cases have duties and rights very similar to those of the police. For this reason, the security forces of municipal authorities, as a measure to maintain order and security in urban areas, are to be put under the microscope. More specifically, the aspect of the MLES’ interactions with the people they encounter in their daily work will serve as an illustration for the practical steps of ethnographic semantics and the documentary method.

In the following, the development and different forms of MLES in Europe and Germany are briefly introduced as a base for the subsequent illustration of the reconstructive approaches. After pointing out the aim and means of data collection, ethnographic semantics and the documentary method are presented in detail and illustrated with examples at appropriate points. Finally, the combination of both is discussed, and the advantages of a complementary analysis are presented.

Municipal Law Enforcement Services

Municipal Law Enforcement Services (MLES) have a long tradition in Germany and are one part of the security infrastructure of most larger cities and many smaller towns. In addition to state police and private security providers, MLES are implemented by the communities and are therefore assigned to and managed by the municipality (hence the name). This is already quite unusual, since many states rely solely on their state police in terms of internal security or security in the public space, and on private policing in terms of e.g., building surveillance.Footnote 1 Social science research on the matter has coined the term plural policing. The taxonomy of public policing on the one hand and private policing on the other does, however, not reflect the full range of plural policing (van Steden, 2017, p. 41) since several governments rely on auxiliary forces of different kinds. While for example the Police Community Support Officers (PCSO) in Great Britain are a well-known and readily mentioned auxiliary police initiative in that context, they are still part of and trained by the police and not the municipality and are therefore not quite comparable to the German model. Of the few countries delegating security responsibilities to the municipalities in continental Europe, Belgium and the Netherlands are two examples worth mentioning.

In Belgium, a police reform in 1998 led to a re-arrangement of police forces into “one integrated police service, structured at two levels” (Devroe & Ponsaers, 2018, p. 300). The two levels are the Federal Police and the Local Police, two separate police forces that are, however, functionally connected with regard to a common regulatory framework, an overarching training programme and a shared database (Devroe & Ponsaers, 2018, p. 300). The Local Police operates in small-scale ‘zones’ consisting of one or a group of municipalities and is assigned to the respective mayor. Two-thirds of the Local Police are funded by the municipality (one-third by the federal police), giving the mayor an autonomous position in providing security and safety within the municipality. This authority is an expression of what applies to many other areas in Belgium: the competences of the municipality are diversified and they bear a large responsibility. This is also reflected in the fact that Belgium has one of the lowest rates of private security employment (with very limited powers) per inhabitant, since an “immense ideological-based resistance […] prevented privatisation growth” (Devroe, 2015, p. 144). Altogether, Belgium diverges from the Anglo-Saxon trend of pluralisation through privatisation and shows a strong focus on municipally managed security measures. However, the Local Police, despite being managed by the municipality, are part of the public police force and therefore have a different status than MLES in Germany.

A more resembling approach can be found in the immediate vicinity: the Netherlands have long relied on Municipal Law Enforcement Officers (MLEO) to maintain order in public spaces and to punish minor violations and misdemeanour (so-called social or physical incivilities) (van Steden, 2017, p. 41). The officers do not belong to the police but are “a type of municipally based ‘quasi-policing’ (Jason-Lloyd, 2003) body, sworn-in to issue fines to people for ‘minor’ nuisance and antisocial behaviour” (van Steden, 2017, p. 41). Implemented as City Wardens with little competences and no uniform in the late 1980s, the MLEO soon developed into a more powerful and professionalised unit within the system of Special Investigation Officers (BOA) (see van Steden, 2017, p. 42). The officers’ tasks range from patrolling the streets (mostly by foot) and issuing fines to helping tourists find their way and socialising with the people in the neighbourhood. The various assignments and roles the employees need to fulfil often lead to an identity crisis since they repeatedly have to switch between the “friendly face” (van Steden, 2017, p. 46) and the tough cop. Unlike the British PCSO and the Belgian Local Police, the MLEO are recruited and trained by the municipality and – just as the MLES in Germany – often have a background in private security companies (such as in public transport), other departments in the municipal administration or from entirely different sectors such as the retail industry.

The structuring of the police forces in Germany has been a matter for the 16 federal states since the occupation by the Allies.Footnote 2 Today, two systems can generally be distinguished (for an illustration see for example Gusy, 2017, p. 30): While some federal states adopting the separation system make a strict distinction between enforcement police (Vollzugspolizei) and regulatory authorities (Ordnungsbehörden), their tasks overlap in other states (unity system) (Knemeyer, 2007, pp. 36–37). MLES have been established in both systems; however, their duties and powers vary significantly. In the unity system the municipality acts as police administrative authority and can deploy its own policing personnel. Both the police and the police administrative authorities operate on a common legal basis with specifically defined (limited) responsibilities for the latter. In contrast, the police authorities and the regulatory authorities (which are part of the municipal government, usually know as Department of Order (Ordnungsamt)) in the separation system both have their own legal bases. The latter are responsible for the general averting of dangers and risk prevention, while the police respond to urgent cases and act as enforcement assistance and within special jurisdictions. However, the regulatory authorities may also deploy their own field staff. Both the field service of the police administrative authorities and of the regulatory authorities are subsumed under the term Municipal Law Enforcement Services in practice and in this paper.

Although the research body is small, few legal scholars and social scientists have been studying the positioning of German MLES in the structure of security stakeholders, their tasks and cooperation in the municipal administration or regional peculiarities in individual federal states. Tuchscherer’s (2017) overview of the emergence, development, and framing of MLES is the first of its kind and provides a structured foundation for further research. However, the analysis remains at a theoretical level, as no empirical data were collected or analysed. The same is true of Balzer’s (2018) work, which provides a comprehensive look at the legal basis, responsibilities, and training of MLES, but does not go beyond the descriptive. Floeting (2014) fills this gap in part through his large-scale survey of executives in 190 regulatory and urban planning authorities in larger cities in Germany. The results show that regulatory authorities and MLES increasingly and more frequently are taking over tasks from the police. Further, the municipal administrations surveyed report that the cooperation between police, city administration and MLES is pragmatic and works just fine, if all stakeholders share a basis of trust. Other than that, the project does not address the practical work of security services in any way. Lastly, Thurn (2021) interviewed and observed one MLES, focusing on procedural justice and the interaction with marginalized groups as well as complaining citizens. The paper touches on various topics related to MLES (such as the communalization of urban security, legitimacy of policing actions, crime hotspots, the feeling of safety as well as legal basics) but misses out on a detailed description of the methods of analysis. The main argument is that procedural justice, even if successfully implemented by MLES, is not sufficient if distributive fairness is not in place.

Furthermore, there is a wide range of scientific literature on overarching topics such as the re-municipalization of police tasks (Behr, 2002) and the custodialization of municipal security (Braun, 2014; van Elsbergen, 2004). In this context, partnerships of regulatory and police authorities, private security services as well as voluntary police forces as actors in the field of communal security are discussed. Yet, specific empirical research on MLES is lacking so far (Lauber & Mühler, 2019, p. 171).

Aim and methodology

The data sequences presented in this paper are part of a dissertation project that explores the tacit knowledge of MLES and how they differentiate themselves from other actors in public space (such as police, social work, or municipal administration). Within this framework, five expert interviews have thus far been conducted with employees and managers in three different MLES in large German cities. The guided interviews took place face-to-face and lasted between 47 and 86min. In addition, 80h of participant observations were carried out. The author accompanied day and night shifts at different times of the year and in different teams. In the process, detailed field protocols were prepared, which, in addition to spoken language, documented facial expressions, gestures and habitus in particular. While these methods of data collection are certainly not new to criminology, the complementary application of two approaches to data analysis that have rarely been used individually in criminological research is.

In principle, both approaches are suitable for interpreting qualitative interviews and observation protocols. In particular, ethnographic semantics examines immanent meaning based on competent language use. The development of a category system of terms and concepts that contain knowledge relevant to action allows for an initial exploration of the under-researched field of MLES. The reconstruction of meaning systems in their practical use serves the understanding of the research object and at the same time as a basis for further analysis. In addition, since the shared experiential knowledge of the MLES is the focus of interest, the documentary method with its methodological stance on conjunctive experiential spaces offers a suitable framework. The praxeological approach searches for implicit regularities of experiences and aims to reconstruct patterns in the data material.

Over the course of the article, it will be argued that by analysing data on MLES using both approaches to data analysis, one may gain a deeper, better-informed and coherent understanding of the tacit knowledge and frames of orientation of MLES. It is assumed that combining the interpretation of actions and language by two different perspectives will reveal different but complementary results. In the following, the methodological schools of both approaches are presented and their practical application is explained in more detail.

Ethnographic semantics

The idea of ethnographic semantics is located in the tradition of anthropological research. First spelled out by Spradley (1970), it was Maeder (1995) who has made the approach popular in German-speaking regions, using it to analyse data from both interviews and participant observation.

Stemming from the field of US-American linguistic cultural anthropology, ethnographic semantics was at the outset developed to examine cultural knowledge in form of identifying names of things in foreign cultures. While since the early stages this included explicit as well as implicit knowledge, cognitive anthropology is nowadays especially concerned with tacit knowledge (Maeder, 2002, p. 2). The approach, also known as ethnoscience or ethnosemantics, examines complex cultural knowledge and meaning focusing on language and linguistic categories in social situations, discourses and milieus (Maeder, 2018, p. 68). It was James P. Spradley who adopted ethnographic semantics for qualitative research in interactionist sociology (Müller, 2015, p. 135), among others with his work about urban nomads (Spradley, 1970) and women working in bars (Spradley & Mann, 1975). Spradley has particularly emphasized participant observation (Spradley, 1980) and ethnographic interviews (Spradley, 1979) as the data basis for his analysis according to ethnographic semantics.

Maeder aptly describes the basic question of ethnographic semantics as follows: “What terms are used by members of a culture, a group, a scene, or the like, to qualify relevant and significant things and events (also: persons and actions, places and times)?” (Maeder 2018, p. 69). This definition already emphasizes the importance of expressions regarding group affiliations and cultural systems of meaning. It is not the factual meaning that is of interest here, but the meaning of the term in context. An analysis in the sense of ethnographic semantics therefore must observe the accepted use of terms in a respective culture. The competent designation of things or social actions indicates (non-) membership in a group or milieu. Spradley argues that meaning is conveyed through a system of symbols (Spradley 1979 , p. 99) but adopts a wider definition of symbols than just in terms of language, particularly when applying ethnographic semantics to participant observations: Culture is then outlined as patterns of behaviour, artifact and knowledge (Spradley, 1980, p. 86) and a complex set of symbols (Spradley, 1980, p. 130), taking into account that social situations in a context carry meaning as much as language does. When employing these assumptions to the study of MLES the focus is on encoding the cultural symbols transported by the language but also the behaviour, artifacts, and knowledge the employees display during their shifts.

Application in previous research

Ethnographic research has a long tradition in the fields of sociology, anthropology as well as criminology, not least since the Chicago School prominently promoted the methodology in the US-American but also the European scientific community. Ethnographic interviews and participant observations have ever since been part of the researcher’s repertoire of data collection in the study of crime, victimization, and crime control, however, the methods of data analysis vary broadly. In the Anglo-American research community, ethnographic semantics (even if not mentioned by this name) has long had a place and it has, among others, been applied to study the use of language (Lakoff, 1987) or actions like ship navigation (Hutchins, 1995). Spradley, who played a key role in coining the approach and who developed the steps of analysis that are still used until today, employed ethnosemantics to analyse his fieldwork with urban nomads and in a bar, but also in prisons (Spradley, 1979). The proliferation of the method in Europe is demonstrated in the research of Holmberg & Åkerblom (2001) on images of leadership in Swedish social media. Few social scientists have been adopting the reconstructive approach in German-speaking countries, with exception of Maeder, who published numerous papers on social milieus (Maeder, 2000), organisations (Maeder & Nadai, 2004), prison (Maeder, 1995, 2002), school children (Maeder, 2008) and market research (Maeder, 2009) using ethnographic semantics. In addition, it has been applied in few studies in the areas of child protection (Franzheld, 2013, 2017) and the transformation of identity (Müller, 2015). Beyond that, ethnographic semantics suffers a rather marginal existence next to more prominent means of analysis in qualitative social research. Maeder attributes this to the fact that the approach originated in a “circle of specialists in American cultural anthropology who primarily wanted to use it to further develop their own theoretical and research foundations” (Maeder, 2009, p. 685).

Likewise, the approach has rarely been adopted in qualitative criminological research. Neither the Routledge Handbook of Qualitative Criminology (Copes & Miller, 2015) nor Advancing Qualitative Methods in Criminology and Criminal Justice (Copes, 2012) list ethnographic semantics (or the documentary method, for that matter). Merely Qualitative Research in Criminology presents one paper analysing prison rape by applying ethnosemantics (Fleisher, 2015). Apart from anthologies on criminological research, Reck (2015) uses the domain analysis (the first of four steps in ethnographic semantics) to investigate police officers’ views of racial profiling, finding that officers semantically relate racial profiling differently across and within communities (p. 32).

Practical approach

The ethnosemantical analysis is carried out in practice by finding relations between symbols and revealing so-called domains. In other words: researchers are attempting to single out relevant symbols, uncovering relations between them and clustering them into domains, while each “symbolic category that includes other categories is a domain” (Spradley, 1979, p. 100). The process of analysis can be roughly divided into four steps: a domain analysis, a taxonomic analysis, a component analysis, and an analysis of cultural themes. However, these do not always occur in that particular order (Maeder, 1995, p. 70) and the process of analysis may be circular just as the qualitative research process in general.

In every language one can find a large number of terms that, again, contain a lot of sub-terms. These generic terms (cover terms) as well as their sub-terms (native terms) are usually connected by what is called a “semantic relationship” (Spradley, 1979, p. 112), such as ‘X is a kind of Y’ (strict inclusion) or ‘X is a characteristic of Y’ (attribution). With the help of a domain analysis worksheet, cover and native terms with a relation can be listed, building up a domain. A domain thus comprises a cover term, two or more native terms that are connected to the cover term by means of a semantic relation, as well as a boundary (Spradley, 1979, p. 112). The latter arises from the common meaning of the sub-terms and their semantic relation to the generic term, which distinguishes them from other domains. Applied to the study of MLES one can identify the cover term ‘people in public space’, listing all different names and native terms the employees articulate for the people they meet during a shift. Looking at the relationship, sub-terms such as “Grattler” (bum), “Junkies”, “Bürger” (citizens) or “Alkoholkranke” (alcoholics) are a kind of people that can be encountered in public space and who the MLES meet, thus altogether building the domain ‘people in public space’.Footnote 3 This type of list can be created not only for people, but also for places, artefacts and actions (Maeder, 2002, p. 3).

The subsequent taxonomic analysis is basically a more detailed structuring of the terms collected in a domain. The analysis further divides the terms into categories and arranges them in subsets as seen by the informants. For the analysis of MLES this means that the people in public space can be further grouped into, e.g., homeless people (such as “Grattler”), citizens (e.g. “Anzugträger” [people in a suit], “Beschwerdeführer” [complainants]), people with an addiction (e.g. “Alkoholkranke”, “Giftler” [people with a drug addiction]), people asking for money (e.g. “Bettler” [beggars]) and people coming from somewhere else (e.g. “Touristen” [tourists], “Migranten” [migrants]) (see Table 1). The taxonomic analysis allows for a deeper insight into the categorization of terms through the informants’ relevance systems.

Table 1 Taxonomy of ‘people in public space’
Full size table

The component analysis involves a shift from looking at similarities to also looking at differences in the terms verbalized by the informants (Spradley, 1979, p. 174). Sub-terms that may have the same relation to a cover term and even belong to the same sub-set of terms might still have different attributes. These attributes – or components of meaning – contain a cultural meaning that cannot be reflected in a taxonomy based on one particular semantic relationship. The componential analysis thus shows how each term has diverse attributes. The people in the category of ‘people from somewhere else’ within the domain ‘people in public space’ can be attributed differently regarding for example where they usually are met (contrast dimension ‘where’) and how to interact with them (contrast dimension ‘what to do’) (cf. Maeder, 1995, p. 77). Migrants for example are often met in a park by the MLES, while tourists are encountered in the inner city and at the train station. The contrast is even more visible when looking at the “what to do” contrast dimension: the MLES are not “doing anything” with refugees without residence permit since they are awaiting the deportation to their home countries and it is rather social work that would be needed to improve their situation:

“There is not really a proper integration work with the people who are actually almost no longer there. What do they do, they are bored, they are in their cultural circles, what happens, there are people who know about drug dealing, you can earn money without paying taxes. So from that perspective, they have nothing to lose, what else can they do. At worst, they will be imprisoned […]. And even if they do, what do they have to lose? Being sent back to Eritrea or Afghanistan. Even if that is certainly not a very attractive perspective, but no one is doing anything with them here. I don’t want to provoke a huge program of social work with refugees who are about to leave the country, because that can’t be it in the end, but there is a need for action somewhere along the line.” (Head of MLES, City B).

In contrast, the MLES are working to make public spaces safe and appealing for tourists and visitors. They take an active part in meeting these groups’ expectations of a busy and supervised public space and advocate a livening up by, for example, street vendors:

“Everything where vendors are, there you cannot lie down, sit down, yes, but then you attract tourists, citizens, other user groups, the commuters and so on and so forth. Or passengers who are just waiting at the stop and get something to eat in between. Or who spend their lunch break there. And with that, I would also enhance the whole thing a bit I would say. So that a bit of life comes in, that is, other life.” (Head of MLES, City B).

“So when do the professionally engaged people arrive here, when do guests of the city arrive, when do trade fair guests and in which time window do they arrive here and in which time window is the largest flow, when do they leave again via the station area? And then at some point it became clear that between eight and nine thirty is the highest traffic in connection with the train. Then we said, well, we will set up a permanent presence here. In the large-capacity vehicle, really, so that you can perceive it directly when you come out of the station. You see a large-capacity vehicle, it is abandoned, a foot patrol is still on the road in the area, when buses pass, the vehicle is seen, possibly the patrol is still seen: The subjective feeling of safety increases.” (Employee, MLES City A).

It becomes apparent that the characteristics of terms can vary, even if they have been taxonomically assigned to the same category. Adding more attributes will eventually lead the researcher to a paradigm of a domain, that “allows you to present a large amount of information in a concise and orderly manner” (Spradley, 1980, S.139). The advantage of this approach is that it “can always make use of information unknown to informants to distinguish a set of terms” since the goal is “to map as accurately as possible the psychological reality of our informant’s cultural knowledge” (Spradley, 1979, p. 176).

Lastly, based on the domains, taxonomies and paradigms developed through componential analysis, the researcher can identify cultural themes within the data. Spradley defines cultural themes as “any cognitive principle, tacit or explicit, recurrent in a number of domains and serving as a relationship among subsystems of cultural meaning” (Spradley, 1979, p. 186). It is revealed once categories or contrasting components appear within several domains and therefore seem to have an overarching cultural value. While a detailed description of the genesis of cultural themes would exceed the scope of this work, the shown excerpts already give an indication of a cultural theme that can be found in various places in the material: responsibility. MLES do express various degrees of responsibility for groups or places through words or actions. In the process, it becomes clear whose expectations they want to fulfil, which goals they are guided by, and for which places and people they feel responsible as security actors and for which they do not. This is an important finding with regard to the research question concerning the distinctions from other actors and the tacit knowledge about their field of operation and the people in it.

The documentary method of interpretation

The documentary method has its methodological roots mainly in the sociology of knowledge (i.e. Mannheim, 1964/2004) and in ethnomethodology (i.e. Garfinkel, 1967). Bohnsack first developed the documentary method in the 1980s with a focus on the interpretation of group discussions. It has since been evolving and is nowadays employed for the analysis of interviews (Nohl, 2017), participant observations (Vogd, 2004; Vogd et al., 2018) as well as video and photography material (Bohnsack et al., 2013).

Karl Mannheim’s sociology of knowledge is led by the question, how the self-evident in actions can be made apparent in a scientifically controlled way. How can implicit, taken-for-granted routines be reconstructed and made tangible by sociologists? Mannheim’s focus lies on everyday actions and he points out that even “pre-scientific everyday experience” (Mannheim, 1964/2004, p. 110) and unnoticed objects and actions contain a theory and meaning. When interpreting the social reality, Mannheim distinguishes three levels of meaning: the factual or objective, the subjective and the documentary meaning (Mannheim, 1923/1970, p. 104). The objective meaning stands for the factual object of investigation itself, representing the unmediated meaning of what is said or observed. The subjective meaning (or intended meaning of expression) unfolds as a certain kind of (self-) representation, uncovering an underlying purpose of an action or speech act. This subjective level of sense can only be captured by knowing contextual information such as the values and norms of the acting person. Lastly, what Mannheim is particularly interested in, is the documentary meaning: in comparison to the other two levels of meaning that are grasping the “what” of an action, the interpretation of the documentary meaning aims at the perception of the “how”. This level of sense is not explicitly conveyed through action or language, but is reflected in facial expressions, gestures, tone of voice, and posture, among other things. It can thus best be identified and interpreted by listening to and observing interactions. The analysis is interested in how reality is constructed more than in factual correctness. Mannheim called this shift of interest “Einklammerung des Geltungscharakters“ (Mannheim, 1980, p. 88), meaning a bracketing of validity allowing for a closer look at what is documented in the practical action about the actors and their orientations.

Regarding MLES, the documentary method of interpretation directs the researcher to take a closer look at the habitus, demeanour, and intonation of the employees in order to find out how they construct social reality. Close listening and observation will then allow to detect the third level of meaning in interactions of the employees at MLES.

In addition to the sociology of knowledge, the documentary method as defined by Bohnsack is influenced by ethnomethodology. Ethnomethodology, considerably coined by sociologist Harold Garfinkel, is similarly engaged with practical everyday actions and knowledge structures. Its focus lies on describing the methods individuals use to manage their daily lives and on reconstructing their approach and tools in (inter-)actions. Just as Mannheim’s work, ethnomethodology aims at analysing the well-known, self-evident and therefore unnoticed characteristics of social situations (Patzelt, 1987, p. 9). Every interaction is assumed to entail unexpressed knowledge and a familiar, unspoken base of things that people know. The question ethnomethodology poses to social situations is how individuals manage to coordinate their actions and meaning in the long term and what role everyday practices and self-evident facts play in this regard (Patzelt, 1987, p. 10). According to ethnomethodology, the emergence and change of the resources – or: methods – of everyday practice only ever receive attention when they are no longer available as a matter of course or when their self-evidence is attacked. Only then do they become conspicuous (Patzelt, 1987, p. 10). This is where the study of MLES comes in: how does the implicit knowledge of the employees show in their practices and habitus, but also in communication? It is the aim of the study of MLES to reconstruct the unsaid and self-evident methods the employees use to achieve their goals at their workplace.

The wide reception of Garfinkel’s ethnomethodology has also called forth critique with regard to the actual research procedure: he is reproached for not having designed its own methodological approach for the investigation of everyday practices and interpretations of meaning (Bohnsack, 2014, p. 59). While referring to Mannheim’s sociology of knowledge, the ethnomethodology did not outline an alternative to the methodology Mannheim proposed with his documentary interpretation. Bohnsack, however, integrated both methodological directions into a praxeological sociology of knowledge and developed the documentary method. Yet the method seems to have found little explicit application in ethnomethodology to date due to the aversion to generalizations, while ethnomethodological data analysis definitely follows the method’s steps of interpretation (Schüttpelz, 2019, pp. 234–235).

In the present project studying MLES, the documentary method is used to examine both interviews and participant observations, thereby allowing insight to a whole bandwidth of implicit knowledge and documentary meaning in everyday practices. Transcripts and field protocols are thoroughly analysed in several stages aiming at the reconstruction of social structures in speech and actions. The following section presents some examples of the application of the approach to other fields of research up to now.

Application in previous research

The documentary method as it was developed by Ralf Bohnsack has mainly been received in the German-speaking region (for a comprehensive overview see Bohnsack 2014) with some exceptions (see e.g. Trautrims et al., 2012; Watson, 2018; Weller & Da Malheiros Silva, 2011; Krzychała & Zamorska, 2014; cf. Bohnsack et al., 2010, pp. 24–25). While several German-speaking researchers conducted studies adopting the documentary method in foreign countries and publishing their results in various languages (cf. Bohnsack, 2014, pp. 284–286), the approach has not yet been much acknowledged by scholars in an international setting.

Emerging out of the field of educational research (Streblow, 2005; Wiezorek, 2005) the documentary method is by now largely recognized in the research areas of childhood and youth (Nentwig-Gesemann, 2002; Bohnsack, 1989; Fritzsche, 2011), culture and migration (Bohnsack & Nohl, 2001; Weller, 2003), media and technology (Schäffer, 2003), organisations and organisational culture (Amling & Vogd, 2017) as well as evaluation (Vogd, 2006). Video and image interpretation in particular has increased in popularity in recent years (Bohnsack et al., 2013). Yet, the documentary method has hardly reached the field of criminology, with notable exception of Mensching’s research on police hierarchies (Mensching, 2008, 2009). She examines police organisational culture with a focus on hierarchies and lived hierarchical relationships. Using the documentary method, she analyses group discussions with German police officers and reconstructs everyday practices of sub- and superordination. The comparative analysis reveals two types of orientation among police officers: the file practice, oriented to written documents and formal organisational structures, and the action practice, oriented to shared experiences, informal organisational practices and spoken words (Mensching, 2008, p. 147). Despite this division, she describes how both orientations are connected to and dependent on each other in everyday police work. This thorough documentary interpretation of group discussions with police officers is one of very few examples where the documentary method is applied for studying institutions of crime control.

Furthermore, Watson’s (2018) analysis of a police-involved shooting is concerned with the analysis of video material. He describes how documentary interpretation in the sense of Garfinkel is used in court when evaluating video evidence, as was done in the case of an on-duty police officer killing a man in a streetcar. Watson is able to show how the “legal significance of [the police officer’s] actions were not depicted in the video itself, rather they became recoverable from the video through interpretations of […] how the actions depicted in those videos might be understood as either reasonable or unreasonable when considered alongside the other evidence” (Watson, 2018, p. 133). Thus, the documentary method in this case was not applied by the researcher to analyse video sequences but is identified as present in the work of the prosecution using video evidence.

Practical approach

To analyse and interpret a situation or action using data from interviews or participant observations, the documentary method in its form according to Bohnsack relies on three consecutive steps: the formulating interpretation, the reflecting interpretation, and the typification. These steps of analysis can within themselves be divided into several smaller steps (see Table 2). While the formulating interpretation is the interpretation of the immanent or literal meaning and builds a basis for the following steps, the reflecting interpretation aims at reconstructing the documentary meaning.

Table 2 Steps of the documentary method. Illustration based on Nohl (2017, p. 30) and Bohnsack et al. (2010)
Full size table

The formulating interpretation begins with a structuring of the themes that are apparent in the interview transcript or field protocol. This step identifies paragraphs that are in connection to the research question or that come up several times in different cases, therefore being suitable for a comparison. A particular emphasis is put on “focusing metaphors” (Bohnsack, 2014, p. 137), which describe situations in an eminently dense way. These sequences can bring up unexpected themes breaking out of the researcher’s sphere of expectations and thus offer space for new orientations and relevance structures introduced by the researched. The subsequent “detailed formulating interpretation” (Nohl, 2010, 116) provides that the text is divided into topics and subtopics. Subtopics are then briefly summarized and this reformulation of the thematic content serves to make the researchers strangers to the text (Nohl, 2010, p. 204). This distance increases the awareness that the content of the interview is not self-evident but demands interpretation.

The following exemplary sequence from an interview with three employees of a MLES (Xm, Yw and Zm; City B) is intended to provide insight into the procedure of analysis following the documentary method.

1

2

Iw:

Yes. What kind of groups do you encounter in your daily work, for example? Or what kind of people do you work with?

3

4

5

Xm:

Well, it’s certainly multi-layered, on the one hand it’s the classic drinking clientele, then we have a lot of punks, so left-wing scene, and then just, () yes, migrants. On the one hand (2), yes, I don’t know how I @should@ describe it-

6

7

Yw:

With residence permit, or with permit for a tolerated stay, who are allowed to stay, but have not yet fully arrived.

8

9

10

11

12

13

14

15

Zm:

So a residence status that is eventually not yet- still in examination, but is not yet completed. So they themselves don’t know yet how it will be. So that’s unsatisfactory for both sides, of course they have obligations that they have to adhere to, but they don’t have a future yet. And that is the uncertainty, they are of course also susceptible to third parties, or more approachable, uh to get into a dependency relationship. One grasps at every straw in any direction to have a future, and whether that is always the straw that brings you back to life is another story. Are then quickly victims for () criminal affairs, I will say now.

16

Iw:

Mhm. Drugs? But also?

17

Zm:

Too. But also exploitation, not only one’s own drug use, but also to be used there directly as a dealer, trafficker whatever.

18

19

Yw:

To commit thefts or or something like that. Or to act as, yeah as a dealer, as a third party.

20

21

Zm:

Or in the desperate situation to carry out acquisitive crime yourself in order to have money for living. Or for his own demands.

The topic of the interview sequence from lines 1–21 is groups encountered in public space. It can be further divided into subtopics and sub-subtopics and Table 3 shows the topical structure of the passage. Since the in-depth presentation of the detailed formulating interpretation, meaning the summary of the subtopics, would exceed the scope of this article, it will be refrained from. Instead, the differentiation of text genres is displayed to the possible extent, which marks the first step of the reflecting interpretation.

The reflecting interpretation includes a formal differentiation of text genres followed by a semantic interpretation with a comparative sequential analysis. Aiming at the interpretation of “how” a topic is treated and becomes clear in its reference framework, the reflecting interpretation first takes a closer look at the text type. Following a classification based on Schütze’s narrative analysis, four main genres can be distinguished: narration, description, argumentation and evaluation (Nohl, 2017, p. 32). This theoretical organisation of the course of discourse allows to reconstruct a collective orientation of communication or interaction (Bohnsack, 2014, p. 138). Especially narrations and descriptions are believed to contain and convey implicit knowledge and atheoretical, self-evident meaning. Therefore, they are the centre of attention of the documentary method.

Table 3 Exemplary text interpretation: topical structuring and differentiation of text genres
Full size table

After structuring the interview or field protocol thematically and according to text genres, the actual interpretation of the documentary meaning comes to pass: by looking for differences or commonalities in text sequences one can perform a comparative analysis. Minimal and maximal contrasts in segments are sought for to reconstruct common orientation frameworks. Text sequences in a single empirical case are analysed “against the background of [the researcher’s] own conceptions of normality, which are the result of experience, thought experiments, (common-sense) theories and/or past empirical research” (Nohl, 2010, p. 210). Different empirical cases are then compared to each other and examined for shared regularities and orientation frames as well as for differences between them. The following paragraphs will illustrate the reflecting interpretation of the previous interview sequences and then compare them to another case.

In the given example, the simple listing of the various groups encountered by the MLES is followed by a detailed depiction of the “migrants”, who are the only group the employees describe in more detail. There appears to be a difficulty in outlining this group (lines 4–5) and Xm’s colleagues offer help by exemplifying the people’s residence status (6). This is noteworthy in that the interviewer did not ask for further details but there seems to be an urge to further elaborate on this group. In the course of this, a residence permit is equated with “arrival” (7), which expresses a bureaucratic understanding of migration: without official certification from the authorities, arrival in the new city and society cannot take place and there is no future for them (10–11). Zm then develops the argument that precarious residency status leads to dissatisfaction and insecurity, which in turn makes people susceptible to dependency on unspecified “third parties” (11). This line of reasoning emphasizes the inability of immigrants to act to improve their situation in other ways, which makes them “victims” (15) of the system. Slipping into criminality is thus by and large not self-inflicted and the relationship of dependence with third parties is a natural consequence (“they are of course also susceptible” (11)) of the uncertain situation. Drug trafficking, theft or acquisitive crime are then not condemned, but understood as the last “straw” (13) to finance one’s own addiction (17) and the means of living (21).

This interpretation reveals an orientation towards a life situation that is externally imposed by the authorities, societal structures or the political treatment of refugees. A similar line of reasoning can be discovered in an interview sequence of another case from a different city: “I mentioned the various groups, some of them methadone patients, alcoholics, and then there are also refugees who have been knocked off course, or thrown off track even more, who are also- They are not yet this huge group, but they are there” (head of the MLES in City A). In terms of minimal contrasts, the comparative sequential analysis indicates a shared frame of orientation regarding people’s (ir)responsibility for their situation (“have been knocked off”) as well as a certain understanding of it in some cases. The author was able to consolidate this orientation with the help of further interview sequences with both minimal and maximal contrasts.

The building of multidimensional typologies across cases marks the final step of the documentary interpretation. Elaborate type formation is often left to comprehensive studies such as doctoral dissertations and the like, while less extensive projects focus on the comparative analysis. Based on the comparative sequential analysis the formation of sense-genetic types within the MLES identifies the essential frames of orientation, which stand out from the case and can also be found in other cases (Nohl, 2017, p. 10). By abstracting orientation frameworks found in one case (one interview transcript or field protocol) and balancing them with other cases, the researcher generates sense-genetic types that show “how different the orientation frameworks are in which research subjects deal with topics and problems that are the focus of research” (Nohl, 2010, p. 212). With regard to the MLES an orientation towards the responsibility of the people in public space for their living situation can be reconstructed. Some groups, such as the migrants in the example before, are perceived as victims of a system (externally imposed). Other groups are attributed an agency and a choice about their stay and their actions (self-determined), as a statement by an employee in City A illustrates:

“When I say ‘people with their centre of life on the street’, it doesn’t mean that they are homeless. […] They’re - I would say it’s a good 85% to almost 90% actually have housing, right? So they don’t live on the street, they have their centre of life on the street. Really, they meet there because they just know, that’s where I can always come back to. They may have lived on the streets at one time, but they just know where to find their people. […] They don’t have the urge to join the mainstream. They could in parts.”

Combination of documentary method and ethnographic semantics

What is the purpose of combining two approaches of analysis? Besides the general trend to gainfully use a mix of different data bases and analysis methods to examine a research subject from different perspectives (see the success of mixed-method studies including qualitative and quantitative data), a consideration of two qualitative, reconstructive methods can be a fruitful endeavour. A triangulation, meaning the adoption of different perspectives on a research subject, isn’t new to qualitative social science (Dörner et al., 2019, p. 7; Flick, 2011, p. 7) – the concept is, however, often misunderstood. A triangulation of qualitative methods shall not be a merging of methods that is simply meant to combine the analysis results, but to produce different results in relation to the object of investigation through different perspectives. These perspectives can be reflected in different types of data, such as interviews and participant observation, or in different theoretical and methodological approaches, both of which are related to each other or should be linked (Flick, 2011, p. 12).

In the given study the theoretical perspectives include the sociology of knowledge, ethnomethodology, and anthropology, whereby ethnomethodology and the common theoretical focus on implicit knowledge represent the main link between the two approaches to interpretation. As a result, the presented analysis excerpts of both methods provide indications of responsibility attributions in the MLES’ conjunctive space of experience. However, they do not confirm each other’s findings, but point to the phenomenon of liability and accountability from different angles. On the one hand, ethnographic semantics reveals the conceptualizations that MLES use to qualify and categorize entities in their environment. Therein, the analysis points to the variances in the MLES’ adoption of responsibility in relation to the people in distinct categories. Through the analysis of the competent use of terms and semantic relations across empirical cases, the adoption of responsibility (in the sense of liability) by the MLES can eventually be identified as a cultural theme within MLES.

On the other hand, the analysis with the documentary method brings forth a typology of responsibilities of the people the MLES meet in public space. In their perception of migrants as victims of a non-functioning integration system or of people spending time on the street as people who make a free decision to do so, an attribution of responsibility (in the sense of accountability) for their respective situation is documented. While we cannot conclude the analysis at this point, it seems reasonable to assume that the different perceptions of these types include aspects that guide the MLES’ actions in interacting with the people. What is self-evident for the employees and a prerequisite for their work, but not for the researcher, was made visible in this inquiry.

The combination of both approaches is twofold productive: the hitherto relatively unexplored research object of the MLES is first captured within the framework of a cultural lexicon, which opens up the central terms that are apparent for the employees, but unfamiliar territory for the ethnographer. The terms for people, places, actions, and artifacts can systematically be categorized and related, then contrasted and examined for deeper cultural themes. Second, once the semantic field has made itself accessible to the researcher, the practical aspects of ‘doing MLES’ can be considered in addition to the more linguistic ones. Along the research questions to what extent the MLES distinguish themselves from other actors in public space and how they interact with them, it can be examined what is documented about them in their everyday routines and how implicit knowledge is expressed in the process.

Both approaches aim at the same subject matter (tacit knowledge), but from different methodological and conceptual perspectives. Fielding and Fielding caution that

“[t]heoretical triangulation does not necessarily reduce bias, nor does methodological triangulation necessarily increase validity. Theories are generally the product of quite different traditions, so when they are combined, one might get a fuller picture, but not a more ‘objective’ one. Similarly different methods have emerged as a product of different theoretical traditions, and therefore combining them can add range and depth, but not accuracy. […] We should combine theories and methods carefully and purposefully with the intention of adding breadth or depth to our analysis but not for the purpose of pursuing ‘objective’ truth”. (Fielding & Fielding, 1986, p. 33)

In view of this criticism, Denzin, who is considered one of the pioneers of triangulation, introduced the concept of sophisticated rigor: Researchers are “committed to making their empirical, interpretive schemes as public as possible” (Denzin 1989, p. 235–236, cited in Flick, 2011, p. 20) in order to minimize threats to the validity of their research. It is the aim of this paper to contribute to this same rigor by detailing the methods and research process and presenting them in a transparent manner. The goal of complementing the documentary method and ethnographic semantics is thus to obtain as comprehensive and differentiated a picture as possible of Municipal Law Enforcement Services in order to establish them as a research subject and to create a basis for further studies. The concept of responsibility as a key cultural theme in the everyday work of the employees depicts only one aspect of many that have yet to be discovered and systematically described.

Conclusion

The purpose of this paper was to expand the canon of qualitative research in criminology by presenting a combination of two lesser-used approaches to analysis and introducing a new research subject. At the outset, Municipal Law Enforcement Services (MLES) were introduced as a stakeholder in the field of crime control and more specifically of security actors. Considerably less discussed in politics and the society than the police, little research has been devoted to MLES, especially not in the context of qualitative empirical studies. Driven by the question of which practices and routines the MLES rely on in their everyday work and how they interact with other actors in the urban environment, two approaches to data analysis aiming at the reconstruction of tacit knowledge from different methodological perspectives were presented: ethnographic semantics and the documentary method.

A linguistically oriented analysis according to the approach of ethnographic semantics took the culturally accepted use of terms in focus. A shared conceptual canon regarding people in public space indicates the employees’ affiliation with a “MLES culture” and allows initial conclusions to be drawn about the extent to which employees attribute responsibility for their own situation to people. In addition to the semantically driven analysis, the evaluation by means of the documentary method gives further insights into the everyday routines and self-evident practices of the MLES. Based on the premise, that every interaction entails unexpressed, tacit knowledge, the documentary method allows for a re-construction of the methods of interaction and how meaning is documented in language and actions. The analysis produced yet another angle on the concept of responsibility, this time focusing on whom MLES take responsibility for on a day-to-day basis.

A systematic analysis of field terminology opened up the research object of MLES for scientists and revealed responsibility as one cultural theme in their logics of action. The documentary method with its focus on the “how” of sense-making was able to show that responsibility is an orientation framework for employees with regard to their counterparts in everyday work. In tandem, the two methods of analysis open up a new perspective on the tacit knowledge of MLES in their everyday routines and complement each other to form a larger picture than any one method can provide individually. The results can inform municipal administration and security policy in municipalities on different levels. An understanding of the employees’ perspective allows for a better adaptation of education and training, for example with regard to intercultural competencies. In addition, a clear allocation of tasks and responsibilities has a positive effect on the employees’ self-image, which in turn has beneficial implications for their work motivation and their interaction with the public.

These findings are of interest to criminology in that the actions of security actors in public spaces and their interactions with the people within them significantly influence the public’s perception of safety. One theoretical approach that would provide a suitable basis for further analysis is that of procedural justice: It can be assumed that the way of interacting with the public has an influence not only on the acceptance of police measures but also on measures by municipal agents. The other way around: what influence does the deployment of MLES have on the image of the police? These questions ought to be addressed in the course of further research in order to learn more about the interplay between security forces.

In addition, the results provide initial indications as to which tasks municipal law enforcement services feel they have to fulfil. In subsequent studies, it would then be worth exploring the extent to which the responsibility for certain places and people corresponds to the tasks envisaged by the municipalities for the MLES. This raises the question of what powers they possess and to what extent these match their own understanding of their role. Moreover, it is important for criminological research, and in particular research on plural policing, to understand how the field of security actors is composed and where they overlap, collaborate or differentiate in their tasks and areas of application. In practice, these insights can influence the allocation of resources among authorities and organisations with security tasks and contribute to better coordination.

It was shown that with the help of ethnographic semantics and the documentary method, qualitative criminological data can be gainfully evaluated. A further dissemination of the methods, whether standing alone or as complementary approaches, would hence be desirable in an international context. Although various challenges remain in conducting qualitative research in criminology, qualitative methods of analysis in criminology continue to evolve. The possibilities of triangulating two or more qualitative approaches to interpretation are not yet exhausted and can potentially make a great contribution to the understanding of crime, victimization, and crime control.