Relational practices in Norwegian students’ e-mail requests in English A focus on openings and closings

The paper explores relational practices in openings and closings in Norwegian students’ e-mail requests in English. It investigates the range of opening and closing sequences and the frequency of their occurrence as well as their variation depending on the level of request imposition and social distance between the e-mail writer and the lecturer. The very high frequency of occurrence of openings and closings, considered to be optional e-mail elements, in a small corpus of 109 authentic e-mails demonstrates a clear orientation to interpersonal aspects of communication. An orientation to familiarity was identiﬁed in the openings while a slight preference for deference was found in the closings. Social distance appeared to affect relational practices to a greater extent than the level of imposition.


Introduction and background
This study investigates relational practices in opening and closing sequences in e-mail requests written by Norwegian students in English.E-mail openings and closings are examined with regard to their range and frequency, as well as their relation to the level of request imposition and social distance between the email sender and the lecturer.As an impersonal and efficient medium, e-mail may be regarded as not creating a need for relational small talk (Döring 2003:134, as cited in Herring, Stein andVirtanen 2013:7-8).However, the central role of interpersonal [1] aspects of computer-mediated communication (CMC) for building and maintaining social relationships has lately been widely recognized (e.g.Bou-Franch 2011; Bunz and Campbell 2004;Herring 1996Herring , 2010;;McKeown and Zhang 2015).
Opening and closing sequences have been chosen for the analysis of relational practices precisely because they represent "mainly phatic, interpersonally loaded structural slots, mostly empty of content regarding the goal or reason for the interaction" (Bou-Franch 2011:1773); they thus provide an insight into relational practices as e-mail writers attempt to establish a social relationship with the lecturer.Opening sequences, including greetings and self-identification, can be regarded as "one means by which the writer constructs his or her social and professional identity and relationship with the addressee(s)" (Waldvogel 2007:457).Closing sequences, comprising, for example, expressions of gratitude, good wishes, leavetaking and signature, "can help consolidate the relationship and establish a relational basis for future encounters" (Waldvogel 2007:457).Hence, rather than simply serving a "window dressing" function (McKeown and Zhang 2015:104), openings and closings serve a range of important socio-pragmatic functions (Kankaanranta 2005;McKeown and Zhang 2015;Waldvogel 2007), including an interpersonal function, in institutional e-mails that are otherwise transactional, i.e. "primarily 'message oriented'" (Brown and Yule 1983:2) in nature.
Institutional e-mail communication has been shown to be challenging for students to navigate in first (L1) and second language (L2) contexts alike for a number of reasons, one of them being fairly heterogeneous views of what constitutes appropriate linguistic behavior in institutional settings (Lewin-Jones and Mason 2014;Savić 2018;Stephens, Houser and Cowan 2009).Lewin-Jones and Mason's (2014) study of student and lecturer attitudes to stylistic features of e-mails has revealed that there is considerable uncertainty about a range of stylistic issues, and that e-mail communication is often a source of frustration and dissatisfaction for both teaching staff and students.It has also been shown that there are mismatching perceptions of what constitutes appropriate academic e-mail communication both among lecturers (Lewin-Jones and Mason 2014;Savić 2018) and between lecturers and students (Economidou-Kogetsidis 2016; Lewin-Jones and Mason 2014).This applies both to contexts where English is the students' first language (Lewin-Jones and Mason 2014) and to English as a foreign language settings (Economidou-Kogetsidis 2016; Savić 2018).Such findings are not surprising since interactional norms in CMC are always negotiated locally, and, compared to the norms in traditional genres, they are "less rigid [and] more open to development and local definition, at least in the early phases of their formation" (Herring, Stein and Virtanen 2013: 11).The socially-situated nature of CMC in general, and e-mail communication in particular, has been investigated in a number of studies that have identified varying computer-mediated discourse practices depending on a number of medium, social and contextual factors (e.g.Androutsopoulos 2006;Bou-Franch 2011;Graham 2007;McKeown and Zhang 2015;Merrison et al. 2012).
A study of openings and closings in the Norwegian university context was conducted by Bjørge (2007).In a corpus of 344 e-mails written by 110 students from 34 national backgrounds, Bjørge examined the influence of national cultures (Hofstede 2001) on opening and closing practices in English as a lingua franca (ELF).Both national and individual factors were found to affect the students' preferred levels of formality.Within the subset of 17 e-mails written by 12 Norwegian students formal openings appeared in 24% and formal closings in 59% of the e-mails.
The present study explores relational practices in e-mail requests written in English by Norwegian university students at an English department, where English is mostly used for student-lecturer communication in and outside the classroom.It seeks to identify the interpersonal resources in e-mail openings and closings that students draw on in initial e-mails, and attempts to unveil potential variability stemming from the level of request imposition and frequency of contact between the student and the lecturer.The study addresses the following research questions: 1. Which opening and closing practices can be identified in Norwegian students' initial e-mail requests in English?Do these practices reveal an orientation to familiarity or deference?2. Do opening and closing practices vary depending on the level of imposition of the e-mail request?
If so, in which ways? 3. Do opening and closing practices in requests differing in the level of imposition vary depending on the social distance between the student and the lecturer?If so, in which ways?
As Norway is regarded as an educational culture in which students and teachers treat each other as equals (Hofstede 2001:101), it was hypothesized that the openings and closings would show evidence of attempts to establish an egalitarian relationship through expressions of familiarity (Bou-Franch 2011).However, expressions of familiarity rest on positive politeness (Brown and Levinson 1987), which, according to Fretheim (2005), tends to be used sparsely in Norwegian.Similarly, Rygg (2016) suggests that phatic talk, including greetings and small talk, and serving a relational function closely linked to positive politeness, is rather uncommon with outgroup members in Norway.Due to the ample evidence of the influence of L1 cultural background on L2 pragmatic practices (e.g.Yates 2010), it was further hypothesized that the relational practices identified would not be very elaborate.
Regarding the second research question, the effects of the level of request imposition on opening and closing sequences in student e-mails have not been examined to the best of my knowledge.Research question 3 was introduced following preliminary data analysis.As there was a significant difference between the number of e-mails written by the students in more frequent contact with the lecturer and by those in less frequent contact, the data set is not ideal for investigating the influence of social distance.However, it was still considered relevant to explore this phenomenon in order to obtain a more nuanced picture of relational practices.Namely, previous studies have revealed the dynamic and socially situated nature of e-mail discourse and the important role played by social distance in constructing openings and closings (Bou-Franch 2011;McKeown and Zhang 2015;Waldvogel 2007).
Although social distance can be viewed as involving a number of components (Spencer-Oatey 1996), in the present study it is operationalized as stemming from the frequency of contact between the student and the lecturer.In line with discursive approaches to politeness (Eelen 2001;Locher 2006;Mills 2011), social distance is regarded as a dynamic variable, continually open for negotiation in interaction.
The study adopts the interpersonal pragmatics perspective in that it focuses on "the interpersonal or relational side of language in use" and "explores in what ways social actors use language to shape and form relationships in situ" (Locher and Graham 2010:1).This perspective is adopted because the ways in which "interactants negotiate the relational aspects of language use in computer-mediated contexts" is still "a subject matter which has received scant attention to date" (Locher 2010:1).(Thomas 1983:99), I discuss them as the e-mail writers' attempts to position themselves in relation to the lecturer in the light of the potential influence of L1 pragmatic norms.
Having provided the background to the study and introduced the research questions, the article will now turn to the methodology employed in the present study.The methodology section, detailing data sources and analysis procedures, is followed by the presentation and discussion of the findings and the limitations.

Data and participants
The data for the current study consisted of 109 authentic e-mails written to the researcher during the spring and autumn semesters of 2014.The original data set comprised 201 e-mail requests (for more information about the corpus see Savić 2018), including both single student e-mails and longer exchanges.As this study aimed to explore relational practices in student-initiated e-mail requests to see how the students positioned themselves when establishing initial contact with the lecturer, only initial emails were selected for this analysis.While Bou-Franch (2011:1773) refers to this as a "structural weakness" in e-mail research, this choice was necessitated by the nature of the research questions.
The selection of the e-mails was made based on the subject line, i.e. a new subject (as opposed to a reply to an e-mail in an ongoing exchange) was considered an initial e-mail request.This resulted in the elimination of 81 e-mails, which constituted parts of longer exchanges.Additionally, since the focus of the present study was on the relational practices of Norwegian users of English, 11 e-mails written by international students were excluded, leaving 109 authentic e-mails in the corpus.
All the e-mail senders were taking BA-or MA-level courses or writing their BA or MA theses in English at a university in Norway.The majority were studying to qualify as English teachers.The e-mails were addressed to the researcher, a female lecturer in her mid-thirties, teaching courses in English pragmatics and English teaching methodology and supervising BA and MA theses at the time of the data collection.
The courses were medium-sized (30-60 students) and I knew each student's name.We were on a firstname basis, which is typical of the Norwegian university context "based on teacher-student equality" (Bjørge 2007:67).
As all the e-mails were addressed to a lecturer, the relationship was asymmetric in terms of Brown and Levinson's (1987) classic social power (P) variable.At the same time, given the egalitarian nature of the Norwegian educational system (Hofstede 2001), the relationship was perceived by the researcher as relatively informal.However, there were differences with regard to the social distance (D) variable, conceptualized as frequency of contact; namely, 28 e-mails were written by three students writing their MA or BA theses under the researcher's supervision.Consequently, they were in contact with the researcher on a more regular basis than the other students.At the time of the data collection, the relationship with the MA student was already well-established as the supervision began in August 2013.
The BA supervision started in January 2014, so the student-lecturer relationships were less firmly established at the outset of the study; however, frequent meetings at the beginning of the semester contributed to a gradual decrease in social distance.
To avoid observer's paradox, i.e. to preclude the influence of the students' awareness that their e-mails might be used for research purposes on their writing style, the students were asked to provide consent for using the e-mails they had sent to the researcher over the past year as research data.The passive consent mechanism was used (Merrison et al. 2012): the students were instructed to respond if they did not want their e-mails to be analyzed.Importantly, consent was requested after the exams, so that the students would not feel that their potential refusal to participate would affect their course grades.Additionally, it was sent after the spring semester had begun and re-sent a month later to ensure the students would read it and have an opportunity to opt out.The project was approved by the Norwegian Center for Research Data.

Data analysis
The data was classified according to the previous classifications of opening and closing sequences (Félix-Brasdefer 2012a; Kankaanranta 2005), as shown in Tables 1 and 2 below.Additional categories were introduced to accommodate for the variation found in the data.Brasdefer (2012b:97), the e-mails were classified as requests for information (RI), requests for validation (RV), requests for feedback (RF), or requests for action (RA).These were defined as follows: (1) requests for information; these requests seek information not already provided in previous discourse […]; (2) requests for validation; these requests seek confirmation or verification of information already provided in the discourse […]; (3) requests for feedback; in which students ask for advice, ask general questions about homework and/or final papers […]; and finally, (4) requests for action, which include utterances that attempt to influence the hearer's behavior in the speech event.
These types of requests represent a continuum in terms of imposition, requests for information being on the low imposition end of the continuum and requests for action being at the high imposition end: RI < RV < RF < RA.
Out of the 109 e-mails, 30 (27.5%) were classified as RI, one (0.9%) as a RV, 35 (32.1%) as RF, 33 (30.3%) as RA, and 10 (9.2%) as a combination of two or more of the above (C), most often of requests for feedback and action.The C requests were regarded as having the highest level of imposition as they always comprised a request for action.All requests were further classified according to the frequency of contact between the e-mail sender and the lecturer.The e-mails written by the students supervised by the researcher were categorized as -D and the ones written by the other students as +D.Table 3 shows the classification of the e-mails according to the request type and social distance.As shown in the table, there is an imbalance between the +D and -D groups, which is an important limitation of the study.However, the potential influence of the D variable is still explored, given its documented influence on openings and closings in institutional e-mail correspondence (Bou-Franch 2011;McKeown and Zhang 2015;Waldvogel 2007).Bou-Franch (2011:1781) relates the discourse practices employed in the greeting and leave-taking sequences to their broader interpersonal meanings, categorizing them as either "expressions of (i) familiarity, involvement and closeness -rapport-building actions (positive politeness) -or (ii) expressions of distance, independence and deference -respect-building actions (negative politeness)".Thus, the classification of opening and closing practices presented below is anchored in Bjørge (2007) and Bou-Franch (2011).
The data analysis included the following steps: identifying the frequency of individual opening and closing sequences in the dataset as a whole, positioning the openings and closings on Bjørge's (2007) formal -informal continuum and classifying them as expressions of familiarity or distance (Bou-Franch 2011), analyzing openings and closings in requests differing in levels of imposition, and analyzing openings and closings in requests differing in the social distance variable.
Only descriptive statistics were used to report on the findings as the number of items in individual categories was too small for inferential statistics.

Relational practices in openings and closings
In order to answer research question 1, regarding the range and orientation of opening and closing practices, the types of openings and their frequency of occurrence were identified first, as presented in Table 4. Eight different types of opening sequences were identified, including a miscellaneous category.The use of a greeting and the lecturer's first name (53.2%) and a greeting word only (31.2%) accounted for nearly 85% of the opening sequences.There was considerable variation within these categories with regard to the choice of the greeting, Hi being by far the most frequently used one in both categories.On Bjørge's (2007:70) formal-to-informal greeting continuum, Hi, whether accompanied by the recipient's first name or not, is regarded as an informal greeting.Within the Greeting + First name category, the second most frequently used greeting was Dear, which, when followed by first name, is also classified as informal, but when followed by the first name and surname is categorized as semi-formal.The latter was found in two instances in our data (Greeting + First name + Surname).Hey or Hei (Norwegian spelling) was a greeting used in 20 e-mails, 8 times on its own and 12 times followed by the recipient's first name.This greeting does not appear on Bjørge's continuum and is interpreted as even more informal than Hi.This interpretation is supported by the lecturers' views in Savić (2018), who investigated lecturer perceptions of im/politeness and in/appropriateness in the selected e-mails from the original corpus.This greeting was evaluated as completely inappropriate and "as rude as you can get" (Savić 2018:63).It could, therefore, be argued that the influence of the students' L1 pragmatic norms in this specific case may have a negative effect on establishing social rapport with lecturers.
In Bou-Franch's (2011) terms, the most frequently used openings identified in the data -greeting word only and greeting word followed by the lecturer's first name -represent expressions of familiarity.Attempting to build rapport through familiarity and closeness appeared to be a strategy favoured by the majority of the e-mail writers, which may be a reflection of the relatively informal educational culture in Norway (Hofstede 2001).However, this strategy may also be an indication that institutional power differences, which may be expected to be acknowledged in e-mail communication, might be superseded by communication norms formed in offline contact in relatively small courses.Such results also mirror the tendencies found in L1 Spanish data (Bou-Franch 2011), where expressions of familiarity in e-mail openings were favoured in unequal encounters in an academic setting.In our data, the use of the first name was found in over 55% of the openings.This seems to reveal the students' awareness of the differences between L1 and L2 communication norms (Dittrich, Johansen and Kulinskaya 2011) as a relatively high percentage of openings containing the lecturer's first name stands in contrast with Fretheim's (2005) and Røkaas' (2000) claims that names are seldom used to form and/or reinforce relationships in spoken Norwegian.Indeed, Savić (2018) demonstrates that the use of the lecturer's first name in the opening contributes to more positive lecturer evaluations of student e-mails.
The use of the lecturer's title was not found in any of the e-mails, which is completely in line with previous research on the use of address terms in Norwegian (Dittrich, Johansen and Kulinskaya 2011;Røkaas 2000), albeit not in e-mail communication.However, the subset of Bjørge's (2007) data written by Norwegian students revealed a slightly different tendency; namely, 24% of greetings were classified as formal, which meant that they contained the lecturer's title and most often the surname.In contrast, in our data only two greetings (1.8%) were classified as formal.Whether and to what extent the use of names and titles in the present study was influenced by the students' L1 pragmatic practices cannot be ascertained with any degree of certainty, especially because L1 English students' opening practices have also been found to vary in this respect to a considerable degree (Félix-Brasdefer 2012a; Lewin-Jones and Mason 2014; Merrison et al. 2012).
While the great majority of the opening sequences were classified as belonging to only two categories, a high degree of variation was identified in the choice of the greeting word within the Greeting word only and Greeting word + First name categories.Additionally, approximately 15% of the openings belonged to the other categories, further adding to the diversity.This high degree of variability reflects the findings of previous studies of openings in student e-mails written in L1 and L2 English (Bjørge 2007

Hi.
Hope you had a good Easter.
The Greeting + Self-identification sequence (Example 1) appeared only twice in the data, which may result from the fact that the lecturer was familiar with all the students taking her courses.Selfidentification was also found in Félix-Brasdefer's (2012a) L1 English and L2 Spanish data; it occurred more frequently in the Spanish data set, but still in only 11% of the e-mails.In contrast, this was a frequent sequence in Bou-Franch's ( 2011) academic e-mails written in L1 Spanish, appearing in over 50% of the e-mails, suggesting that the choice of opening sequences is highly context-sensitive.Emoticons were used as part of the opening in six e-mails.While they can lead to negative evaluations (Savić 2018), the use of emoticons was interpreted as an attempt to express involvement and closeness, in line with Skovholt, Grønning and Kankaanranta (2014), who regard emoticons following signatures as markers of the sender's positive attitude towards the recipient.The students appeared to be using smiley faces to represent paralinguistic features and "adapt the computer medium to their expressive needs" (Herring 2006:617).Finally, the miscellaneous group contained different kinds of comments fulfilling an interpersonal function, referring to previous encounters or e-mail communication (Example 2) and/or expressing positive politeness (Brown and Levinson, 1987), as in Examples 2 and 3.However, such sequences were identified in only four e-mails.In Félix-Brasdefer's (2012a) L1 English data, personal comments were identified in 12.5% of the e-mails, but they appeared more often in closing sequences.Sparse use of positive politeness strategies in the present study, small talk in particular, could also be regarded as an indication of the students resorting to their L1 pragmatic norms.As Rygg (2016) notes, many Norwegians regard small talk as devoid of meaning and prefer to orient to negative politeness and not disturb the interlocutor "with unnecessary talk".However, Rygg's observations regard oral communication and have not been investigated in Norwegian CMC to the best of my knowledge.Metapragmatic data elicited from the students would be necessary to understand their choices regarding positive politeness strategy use.
Turning to closing discourse practices, an even greater variation was found in the data.They were classified into ten categories, as shown in Table 5.Over 90% of the e-mails contained a closing.Three closing sequence categories account for approximately 70% of the closings in the data: leave-taking and name (31.2%), leave-taking, name and surname (23.9%), and signature including only the student's first name (17.4%).Unlike the opening practices, in which the students tended to be informal and orient exclusively to rapport-building, the closings revealed an orientation to deference in a large number of e-mails.Apart from four instances of Mvh and one example of See you soon J, all the other leave-taking expressions were classified as formal (54.2%).Best, Best regards and Kind regards were by far the most frequent leave-taking expressions, belonging to the formal end of the continuum (Bjørge 2007).On the other hand, no closing, signature (first name) without a closing, and thanking and name were regarded as informal closing practices (28.5%).The closings containing the e-mail writer's surname (7.4%) but no leave-taking expression were classified as belonging to the informal end of the continuum.
Similarly to the miscellaneous group in the opening sequences, this one too comprised some personal comments orienting to positive politeness, such as Have a nice day, We hope you have a nice weekend!,but also some personalized closings, such as Optimistic regards or apologies for imposing: Sorry for the bother.As in Félix-Brasdefer's (2012a) data, such comments were more frequent in the closings than in the openings, but they were still fairly infrequent compared to his data.However, contrary to Félix-Brasdefer (2012a), where expressions of gratitude were identified in the majority of the e-mails in both L1 English and L2 Spanish, in the present study, closing sequences including thanking occurred in less than 10% of the e-mails.
The closing discourse practices, therefore, reveal a markedly different orientation from the opening ones.While 98% of the openings are classified as informal and interpreted as showing regard for the receiver's positive face, over 50% of the closings contain expressions of distancing and respect-building.The slight preference for deference in the closings compared to the more conversational openings supports the findings of previous studies in L1 English (Félix-Brasdefer 2012a;McKeown and Zhang 2015), ELF (Bjørge 2007) and L1 Spanish (Bou-Franch 2011).This shift in orientation in over 50% of the e-mails may be a result of a heightened awareness of the importance of laying a solid foundation for future communication.However, qualitative data on students' motivation for using specific relational practices is necessary to support or disprove this interpretation.
The use of opening and closing sequences, which are optional elements primarily meant to personalize emails and foreground sociability, in 97.2% and 91.7% of the e-mails respectively reveals a strong interpersonal orientation.This stands in sharp contrast to the irregular use of greetings in the educational organization e-mails in Waldvogel's (2007) data, which she interprets as a representation of a "business first, people second culture" (2007:471).Contrary to this, other studies examining opening and closing discourse practices have found evidence of the "people first, business second" communication style (Bou-Franch 2011:1783;Félix-Brasdefer 2012a:242), as is the case in the present study.Importantly, research on lecturer perceptions of student e-mails acknowledges the considerable importance assigned to the relational aspect of students' e-mail messages as reflected in openings and closings (Economidou-Kogetsidis 2011; Economidou-Kogetsidis 2016; Savić 2018).The students in this study seem to show an awareness of foregrounding sociability through their consistent use of opening and closing sequences.

Relational practices, level of imposition and social distance
To address the second and third research questions, regarding the opening and closing practices in requests varying in the level of imposition, and the effects of social distance, the openings and closings were classified for each request type: RI, RV, RF, RA and C.Under each heading in Table 6, the number of e-mails written by the students supervised by the researcher is given on the right and by the other students on the left.As can be seen in Table 6, the low frequency of occurrence of many openings does not allow for generalizations about their interrelatedness with request imposition.A closer look at the two categories with the highest frequencies reveals that the use of the greeting word only seems to appear equally commonly in requests with low and high levels of imposition.However, a greeting followed by the lecturer's first name appears more often at the high-imposition end of the continuum, i.e. in requests making higher demands on the lecturer's time and/or workload.The higher degree of imposition seems to have been regarded as creating a need for personalizing the messages to a greater extent.While the students attempted to build rapport through expressions of familiarity and by using largely informal greetings in all the e-mails, a greater tendency for using the lecturer's first name in high-imposition requests may reflect an effort to establish a closer interpersonal relationship before posing a more demanding request.
Regarding the less frequently used openings, the four openings classified as miscellaneous were found in the requests on the higher imposition end of the continuum.Although their frequency of occurrence is very low, this might indicate that these e-mail writers felt the need for more relational work to establish rapport when making requests with a higher degree of imposition, especially requests for action.In these cases, rather than expressing the request immediately after the greeting, four students employed positive politeness strategies (Brown and Levinson 1987) in the opening (Examples 2 and 3).Félix-Brasdefer (2012b) found that the level of imposition affected both the preference for request strategy and the choice of internal modification.As the present study does not offer any conclusive evidence on the influence of imposition on opening practices, further research on larger corpora is necessary to reveal any potential correlations.
When it comes to the influence of social distance on the relational practices in e-mail openings, one interesting finding concerns the use of the lecturer's first name.In the e-mails written by the students writing their theses under the researcher's supervision, it was used in 75% of cases, which is a considerably higher percentage than in the e-mails written by the students in less frequent contact with the lecturer (50.6% [3]).An increased frequency of contact, therefore, appears to have reflected on the frequency of first name use in e-mail openings to acknowledge an ongoing online and offline interpersonal relationship and develop rapport.However, interestingly, the only three e-mails with no greeting were also written by students in more frequent contact with the lecturer.A possible explanation for this could lie in the frequency of online contact itself; namely, in her investigation of the impact of conversational progression on opening practices in L1 Spanish e-mails, Bou-Franch (2011) found that the frequency of openings decreased in non-initial e-mails.In e-mails written in L1 English within a British company, opening salutations were employed less commonly in later e-mails in multi-turn exchanges (McKeown and Zhang 2015).It could be argued that although the e-mails analyzed in this study were initial e-mails, due to regular contact with the lecturer regarding other institutional matters beforehand, the e-mails with no greeting may have been perceived by the students as a continuation of previous exchanges.In the same vein, a number of lecturers interviewed about a selection of e-mails from this corpus (Savić 2018) acknowledged the central importance of the positioning of the e-mail within the longer communication thread for perceptions of im/politeness and/or in/appropriateness.For example, an e-mail without an opening and containing a highly informal request was evaluated as appropriate by many interviewees as it was part of an ongoing conversation with an MA student, which lends further support to the claim that e-mail discourse practices are highly contextualized and firmly grounded in the expectations created in previous communicative exchanges.Although based on a fairly small sample, these results seem to support McKeown and Zhang's (2015) larger-scale quantitative study of openings and closings in British workplace e-mails, in which social distance, operationalized as familiarity, was found to be a socio-pragmatic factor significantly influencing the degree of formality of openings.
The closing practices in the requests differing in the level of imposition and social distance are presented in Table 7. Two out of the three most frequently occurring closing practices, Leave-taking + Name and Signature, name only, seemed to show sensitivity to imposition.These sequences were employed twice as frequently in high-imposition requests (RA) as in low-imposition ones (RI).However, while the former was regarded as a formal closing, exhibiting an orientation to deference, the latter was classified as informal.Contrary to expectations, the presence of thanking (combined with leave-taking, name and/or surname) in the closings did not appear to be affected by the level of imposition.Interestingly, it did seem to be influenced by the social distance variable; namely, it was almost exclusively the e-mail writers in less frequent contact with the lecturer that chose to thank the lecturer in the closing sequence, possibly resulting from a need to establish an interpersonal foundation for future communication.However, expressions of gratitude appeared comparatively infrequently in the data, which makes generalizations impossible.One of the possible explanations of the fairly limited use of thanking could be the fact that expressions of gratitude are open to both positive and negative interpretations, as shown by Economidou-Kogetsidis (2011).Namely, in addition to being evaluated positively, an e-mail-final "thank you/thank you in advance" can also be regarded as "openly presuppos[ing] that the request will be granted" (2011:3208).
In line with the findings regarding opening practices, the -D e-mails lacked a closing more often than the +D ones.This may suggest that the students more frequently engaged in institutional exchanges with the lecturer viewed the relationship as already established and not requiring further work.This interpretation is supported by the students' views of openings and closings in Lewin-Jones and Mason (2014:83), where these elements were regarded as significant in initial e-mails, but as becoming progressively less formal or even superfluous as communication evolved, as a result of "developing informality over time" (2014:83).In the present study, finishing an e-mail without a closing was considered informal.As McKeown and Zhang (2015:104) note in regard to the British workplace community they examined, "greater formality may be expected […] in situations in which there is a degree of social distance".Similarly, Bou-Franch (2011:1783) argues that "as the interaction unfolds, users seem to perceive less need to do complex interpersonal work through openings and closings and, after breaking the ice in the initial contact, negotiate their relationship toward greater informality".While greater frequency of contact may account for greater informality, it is worth pointing out that the percentage of e-mails with no closing was fairly low (8.3%) and that the majority of the students, regardless of their frequency of contact with the lecturer, opted to include relational practices to terminate the exchange in a frictionless manner (Bou-Franch 2011).This may indeed contribute to strengthening professional relationships with lecturers as interpersonal practices in the e-mail closings, although not a major consideration, were found to influence lecturer perceptions of student e-mail requests positively (Savić 2018).

Conclusion
This study explored relational practices in opening and closing sequences in Norwegian students' e-mail requests in English.The high frequency of occurrence of openings and closings, despite the fact that they are optional elements, reveals the students' strong orientation to interpersonal aspects of communication.These relational practices have also been found to play an important role in university lecturers' perceptions of student e-mail requests (Economidou-Kogetsidis 2011; Economidou-Kogetsidis 2016; Savić 2018).A markedly different orientation was identified in the openings and closings.The openings were almost exclusively conversational and oriented towards rapport-building, while the closings displayed, almost equally, deference-and familiarity-orientedness.A preference for solidarity-building in the openings and deference in the closings mirrors the findings of previous studies of academic e-mail discourse (Bou-Franch 2011; Félix-Brasdefer 2012a).
The higher level of request imposition appeared to have created a greater need to personalize e-mails through the use of the lecturer's first name and to employ a wider range of opening sequences, specifically including positive politeness strategies.Social distance, operationalized as the frequency of contact, appeared to exert a greater influence on relational practices.More specifically, the students in more frequent contact with the lecturer tended to use the lecturer's first name in the openings more frequently, but, at the same time, they sometimes omitted the greeting and/or closing altogether and did not employ expressions of gratitude in the closings.These effects of the frequency of contact were interpreted as an indication of the students' view of e-mail communication as dynamic and changing as the offline and online relationship with the lecturer evolved, inherent in the discursive approaches to politeness (Eelen 2001;Locher 2006;Mills 2011).Such views were expressed by a number of British undergraduate students in Lewin-Jones and Mason's (2014:83) study; namely, they felt that openings and closings "could become more informal as an email conversation develop[ed], or be omitted altogether".On the other hand, the lecturers' views on discourse practices in institutional e-mail communication seemed to be more static (Lewin-Jones and Mason, 2014:83).However, the lecturers from a Norwegian university (Savić 2018) were more willing to make allowances for informality reflecting a developing student-lecturer relationship.Arguably, more metapragmatic data on e-mail communication in academic settings is necessary to shed light on the appropriate ways in which evolving student-lecturer relationships can be reflected in the e-mail medium.
The results of the present study need to be interpreted with caution due to certain limitations.Firstly, the relatively small corpus did not allow for inferential statistics to examine the effects of imposition and social distance on linguistic variation in opening and closing practices, so it was only possible to chart some tendencies in the data.Furthermore, the comparatively smaller number of the -D e-mails written by only three students represented less than ideal data for investigating the role of social distance in the choice of relational practices.While examining this variable proved to be relevant for providing a more nuanced view of opening and closing practices, the limitations of the corpus and the exploratory nature of research question 3 need to be kept in mind.Finally, this data does not allow for making firm claims about the students' motivation for employing certain relational practices.A fruitful area for future research would, therefore, be to investigate the students' motivation for adopting specific interpersonal discourse practices and thus determine the extent to which they are aware of the communicative value of their language choices in English and of the way they construct their institutional identities through email communication.

Table 1 :
Opening sequences: Classification categories and examples Greeting + First name + Emoticon Hi Maria ☺ Greeting + Self-identification Hello.I am one of the students in your ENG140 course.

Table 2 :
Closing sequences: Classification categories and examples Each e-mail request was coded in terms of the level of imposition and social distance.Following Félix-

Table 3 :
(Chen 2006;Gains 1999;Gimenez 2000)ings were examined with regard to the degree of formality and the level of familiarity or deference.Drawing on previous research(Chen 2006;Gains 1999;Gimenez 2000),Bjørge  (2007:69-70) established a broad classification of greetings and closings into formal and informal ones.The openings belonging to the formal category include Dear + Honorific/Title + Surname or Dear (Bjørge 2007)cording to the request type and social distance Request type Frequency +D -D Sir/Madam, while, for instance, Hi (+First name) and no greeting belong to the informal category.As for closings, Yours sincerely and Kind/Best/Warm regards, among others, are classified as formal, while Cheers, Mvh[2], Best wishes, See you are placed at the informal end of the continuum(Bjørge 2007).

Table 4 :
Opening sequences and their frequency

Table 5 :
Closing sequences and their frequency

Table 6 .
Frequency of opening sequences in requests differing in levels of imposition and social distance

Table 7 :
Frequency of closing sequences in requests differing in levels of imposition and social distance