Interpreting or other forms of language support? Experiences and decision-making among response and community police officers in Scotland

This study engages in research of interpreting as a socially-situated practice and explores two main foci: police officers’ experiences of interpreting and the factors shaping decision-making regarding the means of language support they use to communicate with non-native speakers of English. Given its pivotal role in any investigation, most police interpreting research has focused on investigative interviews. The work carried out by police officers and interpreters in police settings involves a wide range of communicative scenarios inside and outside the police station. Drawing on a thematic analysis of data gathered through focus groups, this study explores police officers’ experiences in interacting with non-native speakers of English across the various scenarios that are part of community and response officers’ day-to-day operations, and examines the factors shaping officers’ decision to book an interpreter or to resort to other means to communicate when a language barrier is identified. Officers highlighted the key role of interpreters in enabling communication, issues related to practicalities when booking an interpreter, and reported on the difficulties associated with telephone interpreting. The discussions illustrate the range of means used by police officers while on duty, the impact of linguistic and non-linguistic factors on the decision-making process, such as the urgency of a given situation, and the findings corroborate the complexity of assessing proficiency. Whereas, overall, officers showed a high degree of awareness of language and communication aspects, the need for more informed guidance on the potential risks of different types of solutions emerged as a pattern in the discussions.


Introduction
The aim of this paper is to examine communicative practices used by response and community police officers in their daily operations, the factors shaping their decision-making in relation to language support, and their views of interpreting. The study examines interpreting as a socially-situated practice (Angelelli, 2004;de Pedro-Ricoy, 2017;Hale & Napier, 2016;Inghilleri, 2003;Napier, 2011) aimed at enabling communication that is shaped by contextual conditions, including the participants involved in mediated encounters. This study takes a step back and adopts a lens that examines interpreting not as 'the' form of linguistic assistance but as one of the possible means to communicate when a language barrier exists. It documents the different forms of language support that are used by police officers in a Police Scotland Division, and how they are ranked relative to each other.
The increasing need for collaboration between interpreters and police officers in Scotland was already identified by Perez, Wilson, King, and Pagnier in their review of Public Service Interpreting in Scotland (Perez et al., 2006). In relation to the need to create dialogue and mutual exchange between police officers and interpreters, Perez and Wilson (2007) note that an increased use of interpreting in the police domain "raises a number of questions regarding the impact of interpreting upon the officers' professional role and, conversely, the impact of the police officer on the professional performance of the interpreter" (p. 79). While on duty, police officers respond to a range of situations which pose various constraints and require different actions (Gamal, 2014;Mulayim, Lai, & Norma, 2014). Communicating effectively is of paramount importance across policing situations for effective assessment and decision-making. Communicating with the person or persons involved is part of the information gathering and assessment process, as observed in the Police Scotland Incident Prioritisation and Response Standard Operating Procedure (2017).
When a communicative barrier exists in a policing scenario, officers need to decide on the means to be used to enable communication. As shown by Foulquié-Rubio (2002), institutional end-users tend to be the ones who normally request an interpreter, or otherwise opt for a different form of linguistic support. As Hale puts it, (…) the responsibility for quality interpreting services does not lie solely with interpreters, but with all participants involved in the interaction, as well as with the system that trains, accredits and employs them. (2007, p. 137).
Interpreting users are faced with the sensitive decision of assessing the extent to which a person can communicate or not in the 'institutional' or dominant language(s) and the corresponding communicative needs, in order to identify the type of linguistic assistance required. Assessing language proficiency, especially when the need for an assessment is imminent, is a complex endeavour (English, 2010). The means of checking comprehension may trigger misleading perceptions of proficiency (Pavlenko, 2008) and lead to miscarriages of justice. In addition, the degree of emergency and/or severity of a situation, as well as time constraints, are typical factors that shape actions in policing scenarios (Kredens & Morris, 2010). Little is known, though, about the range of communicative solutions adopted in policing operations and the extraneous and innate factors that shape the decision-making process. This study sheds light on the range of means used by response and community officers, and on their views of interpreting and interpreter-mediated communication.

Literature review
Research on police interpreting has witnessed a rapid growth over the past decade (Monteoliva-García, 2018), and, understandably, the focus lies on the police interview. Investigative interviews with suspects, victim interviews and witness interviews are highly sensitive, high-stakes communicative genres. As evidentiary and/or information-gathering discourse genres, interviews have been the focus of most police interpreting studies and they constitute the communicative event in which most interpreting seems to take place in this domain (Gibbons, 2001). Studies of both monolingual interviews (Gibbons, 2003;Haworth, 2010;Heydon, 2005;Rock, 2007) and interpreter-mediated interviews (Berk-Seligson, 2002;Böser, 2013;Gallai, 2013;Heydon & Lai, 2013;Krouglov, 1999;Mayfield, 2016;Monteoliva-García, 2017;Nakane, 2007Nakane, , 2009Nakane, , 2014Russell, 2002) have unveiled and illustrated a high degree of sophistication, the combination of ordinary and legal language in different parts of the interview and the sensitive nature as features of investigative interviews and investigative interviewing. Those factors probably account for a higher use of interpreting compared to other policing scenarios. In addition, emergency calls and initial contact with the police are often conditioned by time pressure to a larger extent than interviews, making it more likely to resort to solutions other than booking an interpreter.

Studies of end-users' perceptions and decision-making
Research exploring perceptions about interpreting and interpreters through interviews or focus groups includes studies of interpreters' views of their own role in medical and legal settings (Angelelli, 2004;Ortega-Herráez & Martin, 2010), as well as studies of interpreting users. Napier (2011) explored the views of different stakeholders (deaf people, hearing people and interpreters); Hale (2011) conducted a study on working conditions and experiences in the Australian judicial sector, drawing on an assessment of guidelines for working with interpreters and questionnaires with judicial officers and interpreters. Among the findings of studies exploring end-users' perceptions of interpreting in police settings, different viewpoints are observed. In their Final Report 2011-2012 the Improving Police and Legal Interpreting (ImPLI) Partners (2012) discuss how police practitioners tend to expect interpreters to simplify legal texts for lay users and to provide explanations and clarifications (p. 35), thus displaying a high degree of latitude towards interpreters. In his review of the language of police procedure in New South Wales (Australia), Gibbons (2001) reports on a primarily reluctant attitude among police officers to book an interpreter, and mentions several reasons. Some of them are of a practical nature, such as difficulties to obtain an interpreter in a timely manner and the fact that interpreting costs are charged to the local budget. Others, however, are related to the perceived loss of control over interviewing techniques and access to features such as non-verbal cues when the interview is interpreted, and to the belief that interpreters tend to be on the side of interviewees. As Gibbons puts it, whereas the interpreted interview does make it necessary to accommodate turns, space and dynamics, communicating without an interpreter poses serious risks.
In line with Gibbons' findings, Wakefield, Kebbell, Moston, and Westera (2014) also obtained ample evidence that perceptions of interpreting among police personnel in Australia deter interpreter use. The concerns included the cost of interpreting services, and the perceived need to change the way interviews are conducted. Police, military and intelligence personnel involved in high-stakes interviews reported specifically on the impact of interpreters upon rapport-building efforts (Goodman-Delahunty & Howes, 2017). Wakefield et al. (2014) observed that police perceptions of the effectiveness of interpreting were more positive when the face-to-face (rather than telephone) mode was used; that interpreters were used more frequently for more serious crimes and initial observations; and that interpreters were more often resorted to in suspect interviews than in witness or victim interviews. These findings suggest that both the seriousness of the case and the status of the person requiring an interpreter were factors that shaped decision-making. As for interpreters' perceptions, Howes (2019) notes that interpreters view the obstacles that emerge over the course of investigative interviews and the lack of structural organization of the interpreting profession as two factors that can hinder their practice.
Translation and interpreting are just two of the different forms of language support used across various public service settings. The numerous approaches to linguistic support have been more widely documented in healthcare settings, including bilingual staff, communicating in a lingua franca or a shared language, certified interpreters, translated written information and ad hoc interpreters (Meyer, 2012;Roels et al., 2015). In their study of community interpreting provision in the healthcare sector in Flanders, Roels et al. (2015) examined the decision-making process among institutional and non-institutional end-users. The authors explored the underlying decision-making processes within institutions, as well as the perceived effects of interpreting. One of the key findings of the study was a primarily arbitrary reliance on community interpreting among professionals, institutions, and sections within the same institution, as well as at an individual level (institutional end-user). Lack of consistency in the type of language support used was also salient. An absence of explicit guidelines and protocols for needs assessment and resource allocation resulted in reliance on arbitrary forms of language support, which in turn translated into unequal treatment being provided to end-users. The delicacy and sensitivity of a consultation were some of the factors that shaped end-users' decision-making, as well as the severity of the potential consequences for the client and the complexity of the information. Interestingly, it was observed that whereas unmediated forms of linguistic support were preferred by both institutional and non-institutional end-users, the more familiar users became with working with certified interpreters, the more they valued the benefits of communicating via a certified interpreter to ensure effective communication.

Guidelines and language protocols in police departments
Language support mechanisms and decision-making processes are expected to align with, or be somehow informed by, existing policies and language support procedures. Differences among guidelines and trends in language access policies among countries are evident. In the UK, language and interpreting guidelines are contained in policies and standard operating procedures (SOPs) at police departments, such as the London Metropolitan Police Working with Interpreters and Translators SOP (2016) and the Police Scotland Interpreting and Translating Services SOP (2018). These provide guidance to police on when and how to work with an interpreter to different extents. The Police Scotland SOP (PS SOP) opens with a statement acknowledging the right "for all persons in their contact with the Police Service to understand and be understood." It also highlights the limitations of the guidance as an instrument that does not provide coverage for every potential policing exchange, but which is intended to offer guidance on "the use of interpreters in circumstances that imply involvement in the formal judicial process." The PS SOP makes the principle of "fairness to the accused or suspect" the prime consideration if doubts about the need for an interpreter emerge. It also mentions the possibility of using bilingual officers and ad hoc interpreters, such as relatives or friends, to facilitate communication in "informal" settings. Whereas the guidelines on the use of interpreters are detailed and comprehensive, the section on ad hoc interpreting only refers to the "informal" nature of the encounter and to the need to disregard this option when the information is of a confidential or private nature.
In Australia, the Queensland Police Service (QPS) Language Services Strategy (2011) includes a comprehensive description of the protocols in place to meet the communication needs of communities, including the use of NAATIaccredited interpreters and translators, the Queensland Interpreter Card, and appropriate interpreting equipment. The Code of Practice of the New South Wales Police Force also requires officers to use accredited professionals. The QPS Use of Interpreter in Domestic and Family Violence Incidents: Best Practice Guidelines (2016) draw on a flowchart sourced from Australia's National Accreditation Authority for Translators and Interpreters (QPS, 2016, p.4) to map out the initial assessment and steps to be followed when the need for an interpreter is identified. As in the SOPs of the London Metropolitan Police and Police Scotland, the use of 'a non-qualified person' is contemplated (QPS, 2016, p.4), although it is presented as an exceptional measure to obtain assistance when "a police officer encounters a volatile situation and needs to ensure the safety of the persons involved" and "experiences immediate difficulty in communicating (…) and needs to establish details of the occurrence and/or the appropriate language needed to engage a qualified interpreter." The policy states clearly that using children as interpreters is not appropriate and may exacerbate exposure to family violence. In line with the recommendations by Gibbons (2001), the guidelines also contain a statement that a qualified interpreter is required without exception regardless of the officer's views on the individual's level of language proficiency if the person requests one. Finally, the guidelines state that, if judged by the officer, an interpreter must be provided for when an individual has difficulty communicating even if that person shows ability to converse in English.
In the United States, police departments are adopting so-called "language access plans" to address the needs of Limited English Proficiency (LEP) communities. These plans contain various types of language solutions. Police departments seem to favour different solutions depending on the needs and strategic partnerships in each area, but most programs promote in-house bilingual services: using bilingual skills available among officers and personnel and incentivising language learning, hiring civilians to perform language support and interpreting services, and creating partnerships with local organizations. In some departments, such as the one in Las Vegas Metro, interpreters are provided with a dispatch radio and an unmarked police car so that they can be on-site and over the phone autonomously. Both the US Department of Homeland Security (DHS, n.d.) and the Vera Institute of Justice (Shah, Rahman, & Khashu, 2007) have published recommendations from and examples of such language access plans.
At the New York City Police Department (NYPD), bilingual Spanishspeaking operators of the Communications Division respond directly to 911 callers with Spanish. Apart from the Spanish-speaking operators, the NYPD Language Access Plan (NYPD, 2018) envisages three types of language solutions. The first one is Language Line services for telephone interpretation. The NYPD also has a Language Initiative Program with NYPD personnel who have been certified to provide interpreting services in-house, and non-certified interpreters ("bilingual" employees of the NYPD). The Language Access Plan also contemplates the possibility of using members of the public, if necessary. When having to decide whether to use a non-employee interpreter, the nature and importance of police services being provided, the apparent capacity of the interpreter, the apparent impartiality of the interpreter and the sensitivity/ confidentiality of the topic are presented as factors to be considered. The fact that exceptions are contemplated is a positive sign. However, the plan does not make specific reference to how and by whom two highly sensitive aspects are to be assessed, namely the competence and the degree of impartiality of the person appointed to interpret. The NYPD Language Access Plan explicitly states that members of the public or family members should not be used in domestic violence situations, only in life-threatening situations when there is no feasible alternative. It is also stated that "a child should not be used as an interpreter for any kind of police incident, including domestic violence." The guidelines and language access plans adopted in different departments are necessary regulatory measures to provide language support services. Differences among them are obvious, and provisions such as including noncertified interpreters and officers-interpreters raise questions about quality control, training and recruitment, and decision-making.

Methodology
Interviews and focus groups have been used as data-collection methods in other studies exploring the perceptions and views of interpreters or other stakeholders on interpreting (Angelelli, 2004;de Pedro-Ricoy, 2017;Koskinen, 2008;Napier, 2011;Ortega-Herráez & Martin, 2010). As a method aimed at investigating the meaning that individuals or groups derived from lived experiences (Böser, 2016), interviews were selected as the preferred method to gather data about language support, communication and interpreting among police officers in this study. Focus groups were selected due to the focus of the research on the collective views of community and response officers, rather than on individual ones, and because they allow for interaction among group members. Furthermore, they were the method recommended by the Area Commander as the one that would probably yield more participation and richer data.
Following an agreement with the Chief Inspector and Area Commander from Police Scotland with whom the author had collaborated in the past, ethical approval was obtained by both parties. The Area Commander facilitated the recruitment of police officers from three different police stations in Paisley, one of the three command areas and the administrative centre for the Renfrewshire and Inverclyde Council Area. The Renfrewshire and Inverclyde division is one of the 13 divisions of Police Scotland and was identified as an appropriate site for the study given its scope and demographic make-up. Local police officers perform duties within community policing, response policing and divisional road policing teams. The total number of officers in 2018 was 664 (Scottish Government, 2018), who served a population of 264,000 people (Scottish population 5.4 million in 2017). The percentage of population who identify as Scottish only is 65.9% in Renfrewshire and 69.9% in Inverclyde (Scotland average: 62%).
The group size of 4-7 officers was discussed and agreed with the Area Commander, bearing in mind the recommended group size for the study design (Smithson, 2008, p. 358) and the estimated length (30-40min), as it was necessary to accommodate police officers' shifts and availability, as well as to promote active participation of all participants in the discussion. As shown below, six officers were convened per focus group discussion, but, as is foreseeable given the nature of their task forces, some officers were on call when the group discussions were held. As advised by the Area Commander, the discussions took place between shifts to minimise interference with participants' duties and time. The interview prompts, pre-interview questionnaire and information about the project were sent beforehand to all sergeants in charge of the different shifts, and the author was in contact with each of them prior to the focus group discussions to arrange logistical aspects. The author was warned that discussions could be interrupted if officers were called in, and that officer participation numbers could fluctuate depending on who was available at the police station at the time. Only one group (Focus Group 6) had to leave suddenly near the end of the discussion. Even though it was necessary to interrupt an ongoing conversation, all themes and questions had been covered by the time the call came in.
The author facilitated the focus group discussion. Although facilitating the focus group discussions enhanced the degree to which the author was able to moderate conversations and engage in data analysis, the potential impact of the author's dual role as an interpreter and interpreting researcher is acknowledged as one of the limitations of the study.

Focus groups with community and response officers
The six focus group discussions took place over a three-day period, two per day. Upon arrival at Paisley Police Station, the author noticed two aspects that are relevant for this study. The glass doors at the main public entrance feature 'Welcome' in different languages. On entry, a poster on the wall displays flags and a message written in different languages informing the public about the availability of interpreting services. A small-size version of this poster (card) is used by officers and police personnel to interact with members of the public who need it during initial contact. This card helps members of the public to identify their language before booking an interpreter with the appropriate language combination.
Focus groups with response officers, community officers and desk personnel yielded almost three hours of data: Informed consent was obtained in person from all participants, who also filled in an anonymous pre-interview questionnaire aimed at collecting demographic information and quantitative details about their level of experience in working with interpreters. The interviews were audio recorded using the author's smartphone, which was considered to be less intrusive than a video recorder or a tape recorder. The six focus groups started with an introduction of the author and the study, and a discussion of the differences between translation and interpreting as a conversation opener and to clarify how the term "interpreting" was going to be used during the discussions. The outline of the interview was explained, in particular that the first two questions were not related to interpreting but to communication in policing scenarios in general, and that the author was interested in hearing from them, in their own words as experts, what their views on communication were. These introductory remarks were aimed at contextualizing bilingual communication as part of policing communication in general, at avoiding making assumptions about their area of expertise and at building rapport with participants.
The following eight questions were put to the six groups, with rephrasing, prompts and follow-up questions when it was considered necessary and beneficial for the discussion. The questions were adapted depending on the group type (response or community): a) I would like you to think about your daily job as a response/community officer, which I guess involves communicating or interacting with members of the public. What are the main aims of communication when you interact with the public as a response/community officer? b) Which strategies do you use to achieve those aims? c) When you need to communicate with an individual who is deaf or who cannot . When you need to achieve those aims and apply those strategies with the assistance of an interpreter, does it pose any challenges to you? If so, which ones? h) Before we finish, would you like to add or ask anything related to the topics we have discussed?
Questions a-e and h were open-ended; questions f-g comprised a closedended element and a follow-up open-ended question if participants responded affirmatively to the first part.

Participants
Two types of police officers were interviewed in order to obtain a more comprehensive view of language support issues in daily policing operations. Community officers patrol the area, engage with members of the public, respond to calls from the public and gather information about the community to respond to their needs. Response officers deal with emergency calls and incidents requiring immediate action. They patrol a division or sub-division under the police force area and respond to incidents (Police Scotland, 2018). As mentioned above, the officers interviewed belonged to Paisley Police Station, which is the head office of Renfrewshire and Inverclyde Division. The focus groups took place in two of the six police stations in Renfrewshire and Inverclyde Division. Community officers were based in one of them and response officers in the other station.
Six focus groups were conducted: two with community officers, three with response officers, and a sixth group with both response officers (x3) and deskbased officers (x2) who work with response officers on a daily basis. Given the likeliness of response officers having to attend calls, more groups with response officers were arranged. Overall, 28 officers took part in the focus group discussions: 11 community officers, 15 response officers and 2 desk-based officers. They were selected randomly: the officers who were either finishing or about to start their shift on the dates and times agreed with the Area Commander took part in the study. The table below shows gender distribution per group and in the study: The officers age ranges lay between 18-24 (lower bound) and 45-54 (upper bound), with 47% of response officers in the 25-34 age range, and 36% (4) and 36% (4) of community officers in age ranges 35-44 and 45-54 years respectively. Most were native speakers of English (27/28). Only one of them was a native speaker of Polish and had English as a second language. Polish (basic level), Welsh, French (basic level) and Spanish (basic level) were the four languages reported by four other officers as being part of their language repertoires. As for their experience in policing, 36% of participants had between 5 and 10 years of experience in the job. On average, community officers had been on the job for longer, with 36%, 18% and 18% having 5-10 years, 10-15 years and more than 15 years of experience, respectively. Among response officers, 35% had 2-5 years of experience, 35% had been in the job for 5-10 years, and 6% had more than 15 years of experience.
Different degrees of experience in working with interpreters were observed among the participants. Only one of the 28 participants (response officer) stated not having any experience in working with interpreters. The other 27 had different degrees of experience: rarely work with interpreters (55%), occasionally (41%), and frequently (4%). The distribution per task force type shows a higher frequency of interpreting use among response officers (50% rarely, 44% occasionally, 6% frequently) than among community officers (64% rarely, 36% occasionally). Finally, most officers with experience in working with interpreters had worked with interpreters of spoken languages (85%), (94% of response officers and 73% of community officers). The rest (15% overall, 27% community officers and 6% response officers) had experience in working with both sign language and spoken language interpreters.

Results and discussion
The focus groups were transcribed using the transcription software ScribD and the answers of the six groups were analysed thematically. Given that the focus was more collective (input from each group) than individual (input from particular persons), the analysis accordingly treated the data as response-group answers or community-group answers. The "scissor-and-sort technique" (Stewart & Shamdasani, 2006) was used: the sections and excerpts relevant to the research questions were identified and colour-coded with the labels PURPOSE, STRATEGIES, LANGUAGE SUPPORT MEANS, FACTORS, CHALLENGES, OTHER. The results per category are presented below, and differences between response and community officers' answers are highlighted as appropriate.

Purpose of communication in policing scenarios
When asked about the purpose of communication, officers highlighted the variety of purposes across policing scenarios and the central role of communicating effectively in their day-to-day tasks. The main purposes mentioned across groups were gathering information, getting the message across, giving instructions and establishing trust. Gathering information was mentioned in relation to establishing the actions required in each situation, and as the main purpose in interviewing scenarios with witnesses and suspects. Community officers emphasized the importance of engaging in general conversation with community members, and obtaining information about their concerns and what matters to them. Response officers' answers focused more on solving problems. A response officer described their role as "professional problem solvers, that's what I like to think," and highlighted the importance of being able to find out what problems people have and how police can help.
As for giving information, getting the message across was mentioned in relation to different scenarios among community and response officers. Every group mentioned giving information, directions or advice to members of the public. Community officers stressed the relevance of letting people know "who we are, how to get hold of us, what kind of things we do," thereby establishing trust through police presence and informal conversation. Response officers across the four groups mentioned giving instructions and commands in scenarios such as managing traffic in an event or incident or while de-escalating a situation.
Other purposes that were mentioned were calming people down and building trust to establish common ground. A response officer highlighted the need to explain police procedure and to tell people what is going to happen in an environment that is probably not familiar to them. Other response officers also mentioned the educational role of informing members of the publicmaking people aware of laws, of what is allowed or not, and the potential consequences of not complying with the law: "Maybe they haven't reached the point of committing an offence, so you interact with them to make them aware of what the potential of their actions could lead to."

Strategies
With regard to the strategies used to communicate in everyday operations, officers across groups emphasised the need to analyse each situation and adjust the way you speak accordingly. Community officers commented on the importance of being respectful, kind, and empathetic. Response officers also mentioned that trying to find out who each person is, to understand them and to speak the way you would like to be spoken to, is crucial to show respect and empathise. Officers also noted using their tone and their voice to achieve different aims, from calming someone down to giving instructions or deescalate a situation through, for instance, shouting.
Across groups, officers highlighted the importance of non-verbal features. A community officer illustrated this point with a case in which taking their armour off and showing that officers are ordinary people helped them to gain trust from a group of young people in a learning centre. A response officer also stressed the relevance of non-verbal communication and explained the difference it can make. The officer shared a recent experience in which "all it took a very hostile person to calm down in an interview was for me to sit down next to her." Assessing people's capacity to understand was mentioned as pivotal to adjust the way officers interact with them, such as when communicating with children or people with mental health issues, or when someone is intoxicated, aggressive or does not speak the language. Adjustments may involve actions including rephrasing what was said or softening one's accent if the person is not a local.
As for the content and format of interaction, community officers mentioned using humour and finding common interests like football or local matters to initiate a conversation with members of the public, and engaging with the older generations face-to-face rather than per e-mail as a more effective way to engage in interaction with that cohort. Table 3 below presents the different language support mechanisms as reported by officers, and the scenarios and purposes stated in relation to each of them. As shown in the table, the different means are used in different circumstances. Google Translate and "language apps" were mentioned across groups as means used regularly to provide information to the public in informal situations (give directions, answer questions) and/or to establish the initial circumstances while waiting for an interpreter. While language apps are frequently used, officers showed awareness of their limitations and of the fact that they are only adequate when the matters discussed are not of a serious nature. Information requests from members of the public (at airports, on the street, directions) To establish the initial circumstances To establish initial contact while waiting for an interpreter To assess what is going on In "less formal" situations When the urgency of the situation makes it necessary Telephone interpreting

Means to communicate with LEP speakers and factors shaping decisionmaking
To take a statement At charge bar To "sarf" someone (inform of their right to a solicitor, as per the Solicitor Access Recording Form or SARF) Instead of an app, when the matter is not urgent Face-to-face interpreting In formal situations and legal requirements (interview someone under caution, take a statement, if someone has been arrested, someone is getting processed) In serious situations When the urgency makes it necessary At the police station In interviews When availability and time make it possible Bilingual police officers Mentioned as a solution that is not ideal, but a good choice for people who request information or need immediate assistance until an interpreter arrives Family and friends Mentioned in relation to calls about neighbour disputes Not adequate if the situation becomes serious or potentially risky for anyone involved Only if no data protection issues are at stake Communicating in broken English If the person can communicate in broken English, it is a useful resource to establish common ground and try to establish what happened -before calling an interpreter or not Poster with flags and language names and card written in different languages that says "I am an officer at Police Scotland, and I am here to help you" Used to establish which language and dialect they talk and so that people know that assistance is coming In relation to ad hoc interpreters, officers mentioned caution with family members and neighbours, and stated that resorting to family members is only appropriate provided that the matter is not serious and that doing so does not put anyone's rights or interest at risk. A response officer expressed her concern about privacy and consent matters: I'm probably jumping ahead of myself here, but I would imagine that the new data protection rules and laws are going to affect when it's appropriate or when it's not appropriate to use a friend or not to translate a particular situation. It's fine if the person you're speaking to has brought the friend with them and they are consenting to that, but if you are trying to seek a family member to translate a situation you are then giving that family member information. That would be a problem.
Across groups, the dividing line between using a face-to-face interpreter and any other form of language support was determined by the legal nature and the seriousness of the case. A community officer mentioned fairness as a factor and hiring an interpreter as the only way of guaranteeing fairness: I think everything has to be done fairly to everybody and, in that case, the way to do it is via an interpreter. Obviously, the apps make it -we type phrases there, and we get them back in whatever language, but we don't know, don't, if that's exactly what we are-so, in fairness, I think in that case it has to be done via an interpreter.
Officers talked positively about their experiences with face-to-face interpreting and referred to face-to-face interpreting as "the proper way", "the right way" to be used in interviews and in serious cases. Across groups, officers expressed their concerns about telephone interpreting, as discussed in more detail below.

Challenges
Officers' answers in relation to the challenges of communicating via an interpreter in their daily job referred to two main types of challenges: limitations and complications. Under these two broad categories, different problems were mentioned.
In line with the emphasis officers had placed on establishing trust and finding common ground as general communication strategies, it is not surprising that the first challenge raised across groups was the loss of access to emotional aspects and a reduced ability to manage them. Officers felt that communicating via an interpreter makes it hard to convey empathy and emotions, and to build rapport, as illustrated in the following statement by a response officer: We were saying earlier about the different ways in which we speak to people. Empathy, for example. You can't get that regardless of whether it [interpreting] is in person or over the phone. I don't think you can get that emotion and how you are speaking through to the other person.
Other officers described interpreting as "almost robotic" and stressed the difficulties of building rapport. The specific features of interaction in police settings compared to other legal settings is likely to be the reason for these participants' focus on building rapport and empathy. Their views contrast slightly with the preferences among legal professionals in other studies exploring their views of the interpreter's role (Hale, 2007) in that officers seem to give more importance to what is conveyed beyond the words uttered through prosody, non-verbal actions, etc. They felt that, while interpreting makes it possible for them to exchange information, it limits their ability to connect with people, build rapport or engage in the conversation; some officers also felt uncertain of what goes on when interpreted renditions are significantly shorter or longer than what the person said. Officers' misgivings coincide with Jacobsen's (2012) and Mikkelson's (2010) observations on the drawbacks of consecutive interpreting.
Officers also highlighted that interpreting changes the way non-verbal communication works. They were concerned about managing eye contact, and across groups they mentioned how something is lost when communication is mediated. They illustrated this point with two main concerns: difficulties in maintaining eye contact with the person they are interviewing or talking to, and in fully engaging with them. This was also observed by Napier (2011) among hearing participants in her focus group discussions on sign language interpreting, and it is included in the ImPLI Partners Final Report 2011-2012 (2012, pp. 24-25). While officers felt that maintaining eye contact was difficult, they felt unsure about how eye contact with the interpreter should be managed and stated that it felt rude not to look at the interpreter. A response officer phrased it this way "it becomes a relationship triangle that makes it harder to keep eye contact and get to some sort of common ground." This coincides with Wadensjö's observation (1998, p. 236) "one of the 'trouble sources' typical of interpreter-mediated interaction is that mutual feedback between the primary parties is delayed and often non-existent. " Linell, Wadensjö and Jönsson (1992) highlight that primary participants depend upon the interpreter to a large extent to establish rapport, and how that role -which is part of the interpreter's remit -makes her involvement different to that of the primary participants.
In all groups, officers identified limitations in deploying interviewing techniques effectively when interaction is interpreted, including when using inflection, putting questions rapid-fire or the opposite, and employing pauses (according to one illustrative comment by a community officer, "Pause can be used to such a good effect, really. And that's lost."). Officers also mentioned difficulties to assess someone's truthfulness or whether they are unsure or hesitant, and another officer expressed that the degree to which interpreting affects interviewing techniques depends on the type of interview: So, it depends on the type of interview. When it's for CID [Criminal Investigation Division], it's quite crucial that questions have been asked quickly. It may harm it to a certain extent, but, you know, it's just the way it is. You can't do anything about it. Some officers commented on the need to accommodate interpreters' needs when interpreters are not native speakers, for instance by using less slang. They acknowledged that they understand why accommodating is necessary, but felt that something is lost in communication when they adjust the way they express their ideas.
As for complications, the main one mentioned across groups related to the flow of interaction in interpreter-mediated encounters, which implicitly referred to consecutive interpreting. Officers across groups expressed that interpreting is "the proper way" of communicating in formal interviews, criminal cases, and "legal" cases, but they raised three related concerns. Interpreting interrupts the flow and makes using techniques such as pauses or rapid-fire questions difficult or even impossible. Officers also mentioned time as a problem source: the process and the wait can be more protracted and complicated. Interestingly, a response officer in group 3 made an insightful remark about a positive 'side effect' of the time involved in getting an interpreter: Across groups, officers expressed their preference for face-to-face interpreting, which they find easier and more reliable than telephone interpreting. They raised concerns about the logistics of booking an interpreter for over-the-phone services (having to enter a code, then the number, the wait, changes in the number of the agency, getting an interpreter in the middle of the night), and about the extent to which it disrupts the flow. A community officer described communicating via telephone interpreting as "a bit of a palaver," in large part due to having to pass the Airwave radio or mobile phone from one person to another in noisy and confusing situations. Officers also mentioned that "although it has to be done, it prolongs a process as potentially stressful as it is." They expressed what would be an ideal scenario: Ideally it would be great to just have somebody sitting there, but it, it's not always feasible to do that. We don't even have enough police officers to sit and wait, obviously we don't have interpreters to come out.
They also mentioned that interpreting over the phone seems more difficult for interpreters, and that the fact that interpreters can be interpreting from anywhere in the world means that some of them are not familiar with the Scottish accent and struggle to understand. An officer mentioned specifically that he thought interpreters did not have enough context when interpreting over the phone, and how that could be risky. Put together, their misgivings about telephone interpreting seem to make many officers opt for other solutions unless it is a serious case, a witness statement, at charge bar, etc., or the only way of communicating with that person.
Finally, some officers mentioned trust issues, in particular when the person approaches the police because they need assistance. They mentioned that some people feel uneasy with an interpreter and gave the example of a victim of sexual assault who did not want to talk via an interpreter.

Other matters arising
Across groups, officers showed awareness of power and impartiality issues, and became increasingly interested in interpreting matters as the discussions went on. One of the response officers asked the facilitator advice about seating arrangements when working with an interpreter and expressed his view that sitting in the middle, rather than on one side or the other, "although you are impartial, will make you look impartial." In three different groups, officers volunteered comments about the importance of briefing the interpreter by talking to them about the case before the question was asked. A response officer shared how having a quick chat with the interpreter works smoothly but pointed out that a briefing rarely takes place when interpreting over the phone. Another officer stressed the significance of health and safety issues, namely that it is important to care for interpreters' safety because people may be aggressive, or situations escalate.
Other officers mentioned that having to change interpreter after having established some sort of rapport with a person makes things complicated. Sometimes an interpreter can interpret on site, but they cannot go to the police station to interpret in the follow-up interactions, for example. On a similar note, a response officer in Group 2 highlighted the advantage of working with the same interpreters regularly and of knowing them: My experiences in general have been quite positive and, in particular I think because where we are, we often need interpreters for Polish and we know who they are and we have developed rapport with them, so, that makes it a lot easier.
Officers also mentioned that they don't get any training on how to work with interpreters, and that the discussion alone had made them realise that it would be a good thing.

Conclusions
The study aimed at examining the views of response and community officers on communication and interpreting, the means of language support used in their daily operations, and the factors shaping the decision to use one means or another. As shown above, across groups, officers displayed alignment with the Interpreting and Translating Services SOP (Police Scotland, 2018) in their orientation toward using interpreting services in interviews, statement-taking and serious cases. Although in police settings interpreter provision is regulated to a greater extent, this legislative orientation toward using certified interpreters was also observed in studies of prison interpreting (Baixauli, 2013;Martínez-Gómez, 2015. The relative urgency of a situation, the purpose (giving instructions and/or information), low-risk actions, and the focus on building rapport (community officers) were factors favouring options other than formal interpreting, such as language apps, communicating in broken English or ad hoc interpreters.
Telephone interpreting emerged as a problematic service, in particular regarding the practicalities of booking an interpreter, extraneous conditions (not having a dedicated dual handset, sound quality, background noise), the impact of the remote format upon access to non-verbal features and contextual information, and the interactional flow. All those aspects seem to deter telephone interpreting use and favour officers' choice of other solutions when the matter at hand is not serious and other solutions are available. These observations are relevant for interpreting practice and provision, and contribute to the timely discussions surrounding remote interpreting (Braun, 2013). Although across groups officers expressed a clear preference for face-to-face interpreting, remote interpreting is, in most cases, the only interpreting option in emergency situations and to establish initial circumstances. The limitations and complications observed, though, seem to affect end users' experiences and their choices, leading them at times to opt for alternative forms of language support that may pose risks in sensitive situations.
Overall, police officers had a positive view of face-to-face interpreting and valued it as crucial to be able to perform their duties. In line with other studies of police officers' views, though, officers expressed concerns about the limiting impact of interpreting upon interviewing techniques, rapport-building efforts and management of non-verbal actions. Those aspects were mentioned in relation to the main strategies used in policing communication in general. Officers expressed interest in the matters discussed and displayed a proactive attitude in asking for advice on how to work better with interpreters over the course of the discussions.
The findings of this study confirm the need for both more training on how to work with interpreters and further research on how to enhance interviewing dynamics in interpreter-mediated interaction. Recent studies (Hale, Goodman-Delahunty & Martschuck, 2019;Howes, 2019) corroborate the benefits of advanced police interpreting training, as widely advocated by interpreting scholars (Maddux, 2010;Ortega-Herráez, 2011). Training of stakeholders, however, seems just as necessary to increase collaboration and understanding of each other's role (ImPLI Partners, 2012;Perez & Wilson, 2007), and was mentioned as a suggestion by officers in the group discussions. The guidelines in place do serve the combined purpose of regulating the use of a certified interpreter as "the proper way" and raising awareness about the limitations of other forms of language support. However, the existence of these guidelines alone is not sufficient to raise user awareness of how to work with interpreters or about the potential risks of certain solutions. A more comprehensive implementation of awareness-raising measures among interpreting users requires cooperation between the different stakeholders, and a dialogue among interpreting practitioners, police practitioners, interpreting researchers and interpreter educators.