Speaker stance and engagement across disciplines in Lithuanian university lectures: the case of mes ‘we’ in medicine and business administration

. To gain a greater understanding of how speakers construct their disciplinary-situated identities and interact with their addressee(s) in Lithuanian spoken academic discourse, this corpus-based exploratory analysis focuses on the use of mes ‘we’ as a marker of stance and engagement in lecturers’ speech in Lithu-anian university lectures on business administration and medicine. The data reveals that the lecturers in business administration used only the inclusive mes ‘we’, which is known to promote student involvement and strengthen lecturer-student rapport. The instructors in medicine frequently employed the exclusive reference to indicate their belonging to professional communities and highlight their level of expertise in the discipline, creating a sense of distance between the lecturer and the student audience.


Introduction
A great deal of research on academic discourse during the last couple of decades has shed light on its interpersonal nature. The construction of academic discourse is no longer seen as being driven solely by the need for the mere representation of external reality, but also as a "persuasive endeavour" encompassing the creation of social interactions (Hyland 2005: 173). Previously regarded as impersonal and detached from the reader, effective academic communication is shaped by the ways authors position themselves in discourse and engage with addressee(s). The interactions between authors and readers as well as the way authors project their stance in the discourse they construct are now regarded among the most critical factors determining the acceptance of one's argument in disciplinary communities (Hyland 2005: 173).
A considerable body of studies has focused on various linguistic devices employed to project authorial self and establish interpersonal relationships with addressee(s) in academic discourse. By far the most systematic account of the linguistic means used in the creation of interaction in academic discourse has been proposed by Hyland (2005). The model of stance and engagement (Hyland 2005) encompasses a vast array of linguistic devices (e.g. hedges, boosters, attitude markers, directives, questions, references to shared knowledge, asides and personal pronouns), which have attracted the attention of many scholars analysing academic discourse (Fortanet-Gómez 2006;Dontcheva-Navratilova 2016;Bellés-Fortuño 2016;Šinkūnienė 2018;Gholi Famian 2020, inter alias).
The use of personal pronouns is a major area of interest in the investigation of academic discourse (Fortanet-Gómez 2006;Okamura 2009;Šinkūnienė 2018;Krammar 2019, inter alias). This category is very useful in researching the interpersonal characteristics of academic communication, as pronouns are regarded as the most explicit markers of stance and engagement in academic discourse (Hyland 2005). According to Brown and Levinson (1978), the use of pronouns reveals how people conceptualize their role and the relationship with addressee(s). Harwood (2005) explains that the inclusive uses of personal pronouns, generally understood as those instances when their reference scope includes addressee(s), are known to convey positive politeness and help establish rapport with the audience, marking that the speaker sees him/herself and the addressee as equals within a particular disciplinary context (Harwood 2005). In the same vein, Hyland (2005: 181-182) demonstrates that the strategic use of the first person pronouns not only helps authors project their stance towards their arguments and, simultaneously, construct their authorial identities but also has immense strength in bringing in the addressee in the discourse. Authors make use of personal pronouns to mark solidarity with their audiences and even navigate addressee(s) through the discourse "towards a preferred interpretation" (Hyland 2005: 183). However, as explained by Biber et al. (1999: 329), the meaning of pronouns is often vague, and this becomes especially relevant for the first person plural 'we', which may be used with various reference scopes. The interpretation of the author's intended messages may thus be a challenging task for the addressee(s) who have to infer who the actual intended referent is (ibid.).
Various studies have provided evidence that personal pronoun preferences in academic discourse differ across cultures, languages and disciplines. In the systematic review of the 22 cross-cultural studies on self-reference across a plethora of European and Asian linguistic and cultural settings, Mur-Dueñas and Šinkūnienė (2016) convincingly demonstrate that the predominant use of personal pronouns is found in research articles authored by Anglophone scholars, contributing to a stronger and more overt authorial presence in their texts. There is an apparent tendency to use personal pronouns in reference to the author to describe procedures, underline key findings and construct arguments in research articles written in English, whereas in other languages and local contexts (e.g. Spanish, French, Russian, Bulgarian, German and Lithuanian), the use of personal pronouns in self-reference is mostly limited to reporting on the structure of the article (ibid.). Mur-Dueñas and Šinkūnienė (2016) also highlight the use of the plural self-reference in single-authored research articles written in Russian, Bulgarian, German, French, Lithuanian and other languages, an opposite tendency than the one observed in academic texts authored by English native speakers, who frequently intervene with I in their academic texts, possibly to explicitly emphasize personal contribution in their fields of research. Cross-disciplinary studies on the use of personal pronouns as stance and engagement devices in written academic discourse across science fields attest to their predominant use in the soft science domains (i.e. humanities and social sciences) (Hyland & Tse 2004). For instance, drawing on a corpus of 240 research papers from 8 disciplines in humanities, natural sciences, engineering and social sciences (1.4 million words), Hyland (2008: 12) demonstrates that English pronouns are more frequently used in humanities and social sciences, with most of the instances detected in philosophy. The same tendency has been reported in studies focusing on Lithuanian written academic discourse (Linkevičienė & Šinkūnienė 2012). The comparison of personal pronouns in Lithuanian research articles across biomedical sciences, humanities, physical sciences, social sciences, and technological sciences also convincingly demonstrates significantly higher use of personal pronouns in humanities and social sciences (ibid.).
The investigation of personal pronouns in spoken academic discourse is predominantly focused on their use in university lectures delivered in English as a native or foreign language. Traditionally, studies identify first and second person pronouns as crucial in the investigation of interactivity in classroom settings due to their implications for the speaker and the addressee (Rounds 1987a;Rounds 1987b;Fortanet-Gómez 2004;Okamura 2009;Yeo & Ting 2014). Scholarly inquiry reveals varying degrees of attempted rapport as well as personal involvement marked by personal pronouns in lecturers' speech. The inclusive personal pronouns (i.e. used in reference to the addressee(s)) are generally linked to significantly higher levels of interactivity in the classroom (Rounds 1987b: 650). They help lecturers signal solidarity with students, guide them throughout the speech event, contextualize information and facilitate comprehension (Fortanet-Gómez 2004). On the other hand, the exclusive use of personal pronouns, when the reference scope does not include the audience, results in certain detachment from addressee(s), i.e. this way lecturers may distance themselves from their student audiences, highlighting, for instance, instructors' superior levels of expertise in the discipline compared with student competence (Fortanet-Gómez 2004).
Ever since the publication of Rounds' pioneering studies (1975;1987a;1987b) on personal pronouns in academic lectures, significant attention has been paid to the use of we in spoken academic discourse.
The most striking part about Rounds' (1987a) quantitative analysis is the supplied evidence on the predominant use of we: it appeared almost three times more frequently than the pronouns I and you in the corpus of 5 university lectures on mathematics at the University of Michigan (26,734 words). However, the latter tendency has been challenged by more recent studies that attest to a higher estimate of you in lecturers' speech than we and other personal pronouns (Fortanet-Gómez 2004;Morell 2004;Okamura 2009, Yeo & Ting 2014. Scholars report on the growing tendency to use I instead of we in written academic communication, too (see Chang & Swales 1999;Hyland 2001). According to Fortanet-Gómez & Bellés-Fortuño (2005: 166), this apparent shift in the use of pronouns in academic spoken and written discourse might mark the ongoing changes in academic discourse worth further scholarly inquiry.
Regardless of the varying accounts of the frequency of personal pronouns across the studies, the first person plural we is highly useful in academic settings, for it can convey both inclusive and exclusive meanings. Whereas the exclusive we is usually associated with lecturers' intentions to signal authority and create a sense of detachment from their audiences, the inclusive we is seen as an effective persuasive means of establishing a close interpersonal relationship with the addressee(s) (Rounds 1987a;1987b;Fortanet-Gómez 2004;Okamura 2019). The pragmatic value of the inclusive we in creating a sense of cooperation, promoting student involvement, directing attention, and maintaining comprehension has been attested in many studies (Rounds 1987a;1987b;Fortanet-Gómez 2004;Fortanet-Gómez & Bellés-Fortuño 2005;Yeo & Ting 2014;Okamura 2019, etc.). Its use has been linked to greater interactivity in the classroom (Morell 2004) and more positive student and supervisor post-lecture feedback (Rounds 1987b).
While many studies have revealed the vital role of personal pronouns in the construction of academic discourse, there is a clear lack of cross-disciplinary inquiry into their use in spoken Lithuanian academic communication. This research will investigate the use of the Lithuanian first person plural pronoun mes 'we' and its grammatical forms, as well as the inflectional realizations of the first person plural in interactive Lithuanian university lectures on medicine and business administration.

Data and methods
This corpus-based research draws on a self-compiled comparable corpus comprised of online interactive lectures on medicine and business administration delivered at Vilnius University in 2020 when due to the Covid-19 pandemic the studies were carried out remotely. The corpus consists of 27,988 words, comprising transcripts of lectures on business administration (14,403 words) and medicine (13,585 words). The corpus was designed to represent the authentic language used in online interactive university lectures delivered at Vilnius University across the disciplines of business administration and medicine and consists of 4 transcriptions of complete university lecture recordings (2 lectures on medicine and 2 lectures on business administration). The recordings of the video lectures were extracted from the Microsoft Stream platform (i.e. the repository of recorded lectures delivered by lecturers affiliated with Vilnius University). Following the national research ethics guidelines and recommendations 1 , all participants of each lecture provided their informed consent regarding the anonymized use of the linguistic data in the process of data compilation, analysis and publication regarding this research.
The selected online lecture recordings had to comply with the following criteria to be included in the corpus. To retain balance, the selected lectures on each discipline were delivered by a male and a female. All lecturers were native speakers of Lithuanian. The selected lectures were delivered in the conversational style of teaching, which is understood as a mode of lecturing which involves speaking without or with minimal notes, rather than reading the texts prepared in advance (Dudley-Evans 1994: 148). Such lecturing style was deliberately chosen as it involves greater spontaneity as well as the interaction between the lecturer and students, which is important for the linguistic investigation of engagement. In addition, studies on pronoun use in classroom discourse attest to a higher number of personal pronouns in interactive lectures than in non-interactive lectures (see Morell 2004). Lastly, 1 "Guidelines for assessing compliance with research ethics", submitted by the Controller of Academic Ethics and Procedures of the Republic of Lithuania in 2020 (December 10) in order no. V-60, available at https://www. vu.lt/site_files/MID/Akademin%C4%97_etika/Atitikties-mokslini%C5%B3-tyrim%C5%B3-etikai-vertinimo-gair%C4%97s.pdf all four samples were collected with the number of participants in each speech event in mind: all the lectures were delivered to groups of no more than 20 students, the primary speaker was the lecturer, with up to 8 actively participating students. In all selected recordings, lecturers were constantly raising questions, inviting students to participate throughout the whole interactive lecture, and answering questions raised by students themselves. A description of each lecture is presented below: Lecture 1: business administration lecture on credit rating procedures; delivered by a female lecturer (duration: 1:20) (7,877 words in total, out of which 7,755 words were uttered by the lecturer). The lecture was delivered to 9 students, with 7 of them actively participating (raising questions and making remarks). Examples extracted from this lecture are labelled 'BA1'.
Lecture 2: business administration lecture on internal audit tasks; delivered by a male lecturer (duration: 1:18) (6,526 words in total, including 6,056 words uttered by the lecturer). This lecture was delivered to 17 students, and 8 of them actively participated by raising questions and making remarks. Examples extracted from this lecture are labelled 'BA2'.
Lecture 3: medicine lecture on newborn care; delivered by a female lecturer (duration: 1:20) (8,904 words in total, out of which 7,451 words were uttered by the lecturer). The lecture was delivered to 6 students, with 3 students participating by raising questions and making remarks. Examples extracted from this lecture are labelled 'M1'.
Lecture 4: medicine lecture on the medical treatment of cancer patients; delivered by a male lecturer (duration: 1:05) (4681 words in total, with 4433 words uttered by the lecturer). This lecture was delivered to 8 students, with 4 students who raised questions and made remarks. Examples extracted from this lecture are labelled 'M2'.
The data was transcribed manually. The corpus is encoded following the Text Encoding Initiative (TEI) Guidelines: all transcripts contain <teiHeader> elements, which detail contextual information, such as the source of the video recordings, the number of participants, as well as the description of a social setting in which the discourse took place. The following <text> elements contain transcribed speech with marked speaker turns and minimal annotation in terms of vocalized and unvocalized pauses, as well as other important contextual information, such as speaker overlaps, interruptions, kinesic (nonverbal, non-lexical) phenomena such as gestures or facial expressions, and other various non-linguistic incidents that influence the course of speech (e.g. coughs or sudden shifts in the recording quality).
A quantitative and qualitative investigation of lexical and morphological realizations of the first person plural was carried out. The quantitative analysis was used to calculate the overall frequencies and distribution of mes 'we' and its grammatical forms (mūsų 'our', mums 'to/for us', mum 'to/for us', mus 'us', mumis 'us', mumyse 'in us') in lecturers' speech across university lectures on medicine and business administration. To estimate the prevalence of mes 'we' and its forms in the dataset, the frequency of mes 'we' was also compared with the numbers of occurrences of the first person singular pronoun aš 'I', the second person singular tu 'you', the second person plural jūs 'you' and all of their grammatical forms. Since the category of person in Lithuanian is realized not only by lexical means (i.e. personal pronouns) but can also be morphologically marked on the inflection of the verb, the inflectional realizations of the category of person were also included in this research. Differently from English, the Lithuanian pronouns in the subject position can be completely omitted as the category of person and number marked on the verb indicates the subject. Occurrences, where the verbal inflection was used with the pronoun, were counted as single instances of pronoun use. Wordsmith Tools (4 th version) (Scott 2004) was used for the automatic frequency analysis of personal pronouns and the grammatical realizations of person in the lecture transcripts of the two disciplines. The raw frequencies of the identified linguistic units were normalized to 1,000 words to allow for comparisons between the two disciplines.
Further analysis focused only on the form mes 'we' and the inflectional realizations of the first person plural in the subject position, as this was the most frequent use detected in the corpus (see Section 3.1).
To establish the linguistic contexts of their use, the automatic detection of collocation clusters and calculation of their frequency in the corpus was performed by using Wordsmith Tools. These findings were further incorporated in the quantitative and qualitative analysis of the referents and discourse functions of mes 'we'.
The most significant part of the study further focused on the referents and discourse functions of mes 'we' and the inflectional realizations of the first person plural in the subject position. The referents and discourse functions were identified manually by investigating their linguistic environments and considering the contextual information of the lectures, such as the current profession of the instructors outside of lecturing. The referents and discourse functions, attributed to instances of mes 'we', were categorized according to the framework of referents and discourse functions of we presented in Fortanet-Gómez (2004: 55-56). Fortanet-Gómez's (2004) model was chosen for the current analysis because it is by far the most comprehensive and systematic account of the referents and discourse functions of we in university lectures. Each reference scope and discourse function detected in my data is thoroughly explained in Section 3. Results and discussion simultaneously with the reported findings. Table 1 provides the total number of occurrences (as well as normalized frequencies per 1,000 words in the brackets) of the first person singular pronoun aš 'I', the second person singular tu 'you', the second person plural jūs 'you' and their grammatical forms in the four lectures on business administration and medicine. All in all, mes 'we' is the most frequent pronoun in the data set compared with the use of the first person singular and the second person plural pronoun forms. The total normalized frequency of mes 'we' (including the inflectional realizations of the first person plural) in the data is 21.6/1,000.

Frequency analysis
Since there are no studies on the use of personal pronouns in Lithuanian university lectures, and most of the scholarly inquiry into the matter has been based on English language data (Rounds 1985;1987a;1987b;Fortanet-Gómez 2004;Morell 2004;Yeo & Ting 2014;Okamura 2019;etc.), the further discussion of the results is mostly limited to the comparison of the findings of the current study and the research on personal pronouns in English university lectures.
Due to the pragmatic strength of these stance and engagement markers in promoting learner involvement and facilitating comprehension, the overall dominant use of mes 'we' in Lithuanian university lectures is not surprising, as the study deliberately focused on interactive, rather than monologic, lectures. The tendency of lecturers to mostly employ the first person plural in university lectures has also been attested in some studies based on English university lecture data (Rounds 1985;1987a). Most research reports that the first person pronoun in university lectures is highly useful and versatile: it can be used to signal inclusive and exclusive meanings (Rounds 1987a;1987b;Fortanet-Gómez 2004;Okamura 2019, etc.). The exclusive we is linked with lecturers' intentions to highlight authority and detachment, whereas the inclusive we aids in establishing a close interpersonal relationship with the student audiences (Okamura 2019: 17) and promoting interactivity in the classroom (Morell 2004). Instructors use the inclusive first person pronoun to create a sense of solidarity with their student audiences, guide them through discourse, as well as maintain and check comprehension (Rounds 1987a;Fortanet-Gómez 2004). Therefore, frequently employing we inclusively has been associated with lecturers being more positively assessed by their students and supervisors (Rounds 1987b).
The rather common use of the first person singular aš 'I' is also apparent: its normalized frequency in the dataset is 16.9/1,000. Male lecturers in both disciplines (BA2 and M2) made more frequent use of the first person singular than the first person plural. However, it would be unreasonable taken such a small sample size to conclude that these patterns of pronoun use are gender-based. The varying frequencies of pronouns across lectures might show that their choice is not always uniform and may depend on each speaker's idiosyncratic language use and/or lecturing style. While the first person plural allows lecturers to engage with their student audiences and create a sense of solidarity, the use of the first person singular is reported to create a certain degree of social distance in the instructor-student relationship (Yeo & Ting 2014: 32). Lecturers' explicit reference only to themselves is generally associated with one's wish to highlight his/her expertise, authority and higher status (ibid.).
The second person pronouns (including the inflectional realizations of the second person) are the least frequent in my data. The complete opposite tendency is usually attested in the studies focused on English lecture data: you is usually found to be the most dominant personal pronoun in lecturers' speech (see Crawford Camiciottoli 2004;Fortanet-Gómez 2004;Fortanet-Gómez & Fortuño-Belles 2005;Morell 2004;Okamura 2009;Yeo & Ting 2014). Explicitly referring to the student audience with the second person plural is mostly associated with instructors' efforts to engage students in a dialogue, activate their prior knowledge and instruct them to act (Okamura 2009;Yeo & Ting 2014).
The results of this study yield a much higher use of the second person plural jūs 'you' than the second person singular tu 'you' in the four lectures. It should be noted that the difference in the meaning of both Lithuanian second person pronouns is reflected not only in the category of number and varying reference scopes but also in the different levels of formality. The ability to mark varying degrees of formality with personal pronouns is characteristic of other languages, such as French, Spanish, German, Russian, Hindi, etc. The singular form tu 'you' in Lithuanian is usually regarded as a less formal way to refer to a single addressee and may signal a greater degree of intimacy and solidarity between the addressor and the addressee (Čubajevaitė 2006: 34). Tu 'you' is also known to be used in a generic way to refer to human beings (Ambrazas 2006: 187). The plural jūs 'you' is used to refer to groups of people, yet may also refer to individual addressees more formally and politely than tu 'you' (Ambrazas 2006: 191;Kamandulytė-Merfeldienė & Vainilavičiūtė 2018: 67). It is common to refer to individual students with the more polite second person plural at the Lithuanian higher education institutions, thus the stark contrast between the frequencies of the second person plural vs. the second person singular pronouns in the data is not surprising.
To further investigate the use of mes 'we' in Lithuanian university lectures, the analysis focused on the syntactic positions of its grammatical forms. Mes 'we' is predominantly used in the subject position in a clause (see Table 2). The second most frequent form in almost all lectures (except for M1) is the possessive mūsų 'our', whereas other tokens, such as those used in the object position, i.e. mums/mum 'to/for us' and mus 'us' are used with lower frequencies in the data set.
Total 213 (27.5) 151 (25) 97 (13) 93 (21) These results are consistent with the findings observed in Okamura (2009: 19), also attesting to the fact that the most frequent syntactic positioning of we in lecturers' talk is the subject position, whereas the second most common occupies the object position and the least frequent form is the possessive.
The least frequent first person plural pronoun forms in both disciplines are those marked with the instrumental case (i.e. mumis 'us'), detected only once in M2. The locative form (i.e. mumyse 'in us') is completely absent in the data set. In a similar vein, the low frequency of the personal pronouns marked with the instrumental and the locative cases has been attested in Lithuanian written academic discourse across numerous disciplines and genres (Linkevičienė & Šinkūnienė 2012: 82-83).
Another interesting point of observation is the use of the dative form mum 'to/for us' in my data. Each lecturer, taken separately, uses mum 'to/for us' at similar rates as the more standard counterpart form mums 'to/for us'. The in-depth linguistic investigations on the morphologic development of the Lithuanian pronouns attest to the fact that a shorter form without the final -s, inherited from the Proto-Baltic language, is a dialectal variant of the now more standard form mums 'to/for us' (Valeckienė 1966: 139;Rosinas 2006: 403;Šeškevičiūtė 2012: 20). In fact, the use of mum 'to/for us' has been attested in the Lithuanian writings dating back to the 19 th century (see Valeckienė 1966;Rosinas 2006;Šeškevičiūtė 2012). Nevertheless, the comprehensive analysis of the grammatical realizations of the first person plural mes 'we' in linguistic and medical research articles published between 2000 -2009 has revealed no such instances (Šinkūnienė 2010). Comparing the results of both studies might suggest that the overall use of the grammatical forms of mes 'we' is more varied in spoken than in Lithuanian written academic discourse. The less standard linguistic forms, such as mum 'to/for us' are perhaps more likely to occur in spoken academic discourse, as it is generally characterized by greater spontaneity, preconditioning fewer possibilities for speakers to observe and maintain the correctness of their linguistic choices.
To further explore the semantic-functional profile of mes 'we', the in-depth analysis of its referents and discourse functions in the four Lithuanian university lectures was further carried out.

Referents and discourse functions
The most important part of the analysis focused on the reference scope and the discourse functions of mes 'we' in the four lectures. The analysis was largely based on the distinction between the inclusive and the exclusive meanings of mes 'we' in the dataset. Following Fortanet-Gómez (2004), inclusive meanings were attributed to those uses of the pronoun mes 'we' and its grammatical variants that include the audience (see example 1), whereas exclusive meanings were assigned to tokens that were used in reference to the speaker but not the addressee (example 2), for instance: (1) <...> šiuos vėlgi rodiklius galim skaičiuoti panašiai taip kaip skaičiavom struktūriniais ir palengvintais modeliais <...> (BA1) 'Again, we can calculate these ratios in a similar way as we calculated according to structural and facilitated models' (2) <...> pas mus pavyzdžiui medicinos personalą mes tai skiepijame <…> (M1) 'At our [hospital], for example, we vaccinate the medical staff' In (1) mes 'we' is used inclusively, as it refers to the lecturer and the student audience. Example (2) clearly illustrates the use of the exclusive mes 'we': the reference is made to the speaker and her colleagues working at her hospital, whereas the student audience falls beyond the scope of this reference.

Referents
The frequency analysis has revealed some interesting tendencies in the use of the exclusive/inclusive reference of mes 'we' in the lectures on business management and medicine worth looking into. Overall, the results yield the predominance of the inclusive mes 'we' in the four lectures (Table 3). The general tendency to mostly employ the inclusive we over the exclusive we by lecturers during university lectures has been attested in previous studies (e.g., Rounds 1985;Fortanet-Gómez 2004;Yeo & Ting 2014, etc.). As further analysis of the discourse functions of mes 'we' reveals, the inclusive meanings of the pronoun are mostly associated with the lecturers' intentions of creating rapport with their student audiences, promoting participant involvement and helping them navigate the lecture/course. Comparing the use of mes 'we' across the disciplines and the lectures reveals stark differences between the inclusive/exclusive reference of mes 'we'. The data shows that the lecturers in business administration use mes 'we' only inclusively with a variety of functions such as promoting student engagement, managing their comprehension, and guiding them through the discourse. On the other hand, both lecturers in medicine also frequently used exclusive reference in their talk (60% and 18% of all mes 'we' in M1 and M2, respectively). The use of the exclusive we in the university lectures in the hard sciences (i.e. medicine, engineering, science, information technology, and the complete absence of it in soft sciences (i.e. social science, economics and business, human resource development, creative arts) is also attested in Yeo & Ting (2014: 31). The fact that lecturers in business administration did not use the first person plural exclusively might signal their wish to establish an interpersonal relationship with the students and project a less authoritative stance (Íñigo-Mora 2004; Yeo & Ting 2014). On the contrary, the higher frequency of the exclusive meanings of mes 'we' in the analysed lectures on medicine is found to be linked to a noticeable tendency of lecturers to indicate their belonging to various restricted groups of people, such as specific research and professional communities (3) or members of the staff of certain institutions (4), which has an effect of the detachment between the speaker and the audience: (3) <...> nu mes skiriam bisfosfonatus nežinau jūs turbūt nelabai ir galit skirti <...> (M2) 'Well we prescribe bisphosphonates I don't know you probably cannot really prescribe it' (4) <…> žindymo specialistai taip <…> mes irgi turim <pause> kitos ligoninės kiek žinau irgi turi (M1) 'Breastfeeding specialists yes <…> we also have them <pause> as far as I know, other hospitals also have' The exclusive mes 'we' in (3) is used to refer to the practices common in the specific professional community, i.e. medical oncologists, which the student audience is not a part of. This results in the use of the exclusive reference that creates the effect of separating the lecturer and his peers (medical oncologists) from the student audience (preparing to become general practitioners) in terms of what both professions can or cannot prescribe to cancer patients. The exclusivity of mes 'we' becomes very clear as the lecturer includes jūs 'you' in the following discourse, clearly disambiguating the reference and separating himself and his peers from the students. Another frequent use of the exclusive mes 'we' in the medicine lectures is closely tied to the description of the practices applied in the medical institutions the lecturers represent (example 4). The speaker provides information regarding the availability of breastfeeding specialists at the hospital which she is affiliated with. Such use has the effect of creating distance between the experts (including the speaker) and the learners. This kind of reference has been recorded in the instructors' talk in the university lectures of hard (Yeo & Ting 2014: 30) and soft science domains (Fortanet-Gómez 2004: 57).
Regarding the inclusive uses, the most frequent reference scope across all lectures is the one that includes only the speaker and the audience ("speaker + audience only" in Table 3), with 45% of such occurrences detected in the corpus. All instances of this reference are indeed inclusive, as the lecturer refers to the actions performed by him/her and the student audience in the speech event, as demonstrated in example (5) Such use of the first person plural is known to be a very frequent and effective means of including and marking solidarity with the students (Yeo & Ting 2014: 32). By using the first person plural inclusive pronoun, lecturers position themselves and their audiences as equals in the speech event (ibid.). Where-as such use is predominantly concentrated in business administration (75% and 76% of all instances in BA1 and BA2, respectively), it is rather rare in the lectures on medicine, its frequency amounting to only 2% in M1 and 10% in M2. These stark differences might suggest that lecturers in business administration are more inclined to establish rapport with their student audiences.
The second most common referent of mes 'we' in my data is "a larger group of people (including speaker + audience)", with 21% of occurrences in the corpus. This is the broadest reference, as mes 'we' includes the lecturer, the student audience and the other people, not participating in the speech event. The scope of reference in this category varies from large groups of people, such as humans in general, as in example 6, to more restricted groups and communities, such as people involved in the study process at Vilnius University, as in example (7) This kind of reference is detected in all the lectures, with much higher estimates in medicine than in business administration. This result may be explained by the fact that the lecturers in medicine frequently referred to Lithuanian and European laws regulating the use of prescription medicine, vaccination procedures, and other related matters; consider example (8) The overall high frequency of this kind of inclusive reference is attested in English lecture data (Fortanet-Gómez 2004). Also being ranked the second most frequent in university lectures, such use is associated with instructors' intentions to promote learner involvement (ibid.: 57).
Mes 'we' inclusive only of the audience ("mes 'we' for jūs 'you' (audience)") amounts to only 4% in the corpus. The low frequency of such use of the first person plural is in line with the previous studies on university lectures (Fortanet-Gómez 2004;Yeo & Ting 2014). In my data, this reference scope is found only in the lectures on business administration (BA1 and BA2), where the lecturers used mes 'we' to express actions that can only be performed by the students, e.g.: (9) <...> keturioliktą uždavinį dabar spręsime su jumis <...> (BA2) 'We will now solve the fourteenth task with you' By analysing the context in which the words in (9) were uttered, it becomes clear that it is the students who will solve the task, as the instructor further only provides feedback to the students on whether their answers are correct or not. Both lecturers in business administration choose to make use of the inclusive mes 'we' in this and other similar instances possibly to signal that s/he will guide students in the process, thus creating a sense of cooperation, whereas in medicine no such instances were found.
The least frequent referent is "mes 'we' for I", with a frequency of 3% in the corpus. This scope of the reference is inclusive: instructors use we 'mes' in reference to their own actions. For instance, in the following situation presented in example 10, the lecturer uses the inflectional realization of the first person plural while uttering dirbsim tuo klausimu 'we will work on that issue'. Here she is obviously referring to herself since the discussed issue arises from a mistake the lecturer made while preparing the online exercise. The fact that the grammatical realization of the inclusive we is used to solely refer to the lecturer becomes even more apparent while analysing the linguistic context around the utterance: before that the lecturer says darau viską kad nematytumėt 'I'm doing everything so that you don't see', making it clear that she is in charge of creating the task (see example 10): It is interesting to observe such interplay between the use of singular and plural forms to express selfreference in the lecturer's spontaneous speech after realizing her mistake. It seems that the use of the plural form might signal the need to shift the focus away from oneself possibly due to feeling uncomfortable. This assumption is further reinforced by analysing the following context: the lecturer uses the first person plural with the general reference to all humans by uttering 'mokytis turime visi kur' 'we all have room to improve' to stress the fact that all people make mistakes. In her data, Fortanet-Gómez (2004: 58) has detected very similar instances of a lecturer switching the focus from I to we, interpreting this as an unconscious way of attempting to protect oneself. Such use of the first person plural with the speaker as the sole referent is sometimes referred to as "the royal we" in academic discourse studies (see Vassileva 1998;Fortanet-Gómez 2004;Yeo & Ting 2014).
Other referents provided in Fortanet-Gómez (2004) were not found in my data, such as the inclusive we for indefinite you or one (used in generalizations), as well as we for they (not inclusive of either the speaker or the audience). The absence of these reference scopes in my data was not surprising, as they were extremely infrequent and only detected in one lecture on anthropology in Fortanet-Gómez's data (2004). These instances may have occurred due to the personal style of a specific lecturer or the subject matter of anthropology.

Discourse functions
Following Fortanet-Gómez (2004), the discourse functions attributed to each use of mes 'we' can be divided into two major categories: functions related to the representation of groups of people and those used to construct metadiscourse. Both categories are further explained in detail in Sections 3.3.2.1. Representation of groups and 3.3.2.2. Metadiscourse.

Representation of groups
This category of discourse functions is the most dominant in my data and was detected in the uses of mes 'we' with the reference to larger groups of people (including speaker + audience) or to "speaker + other people" (excluding the audience). Discourse functions of representation are closely connected with the referent. As already explained, the reference scope of mes 'we' varies in the corpus, as the pronoun is used to represent, for instance, humans in general (as seen in example (6)), Lithuanian doctors (see example (8)), or more restricted groups of people, such as people involved in the study process organized at Vilnius University (see example (7)). The inclusive mes 'we' in all instances helps to reduce the distance between the lecturer and the student audience, creating a sense of cooperation when dealing with matters that concern both, the lecturer and the student audience. For example, such use of mes 'we' is observed whilst discussing which prescription medicine can be prescribed to cancer patients treated in Lithuania, something which is relevant for the lecturer (medical oncologist) and the student audience (individuals preparing to become general practitioners), the two professions that cooperate in the process of treating cancer patients (example (8)). Exclusive uses, on the other hand, highlight the expert knowledge of the lecturer and thus create a sense of distance between students and the lecturer, as illustrated in (3).
Interestingly, this category of discourse functions is mostly detected in lectures on medicine: the overall frequency is 98% of occurrences in M1, and 85% in M2, as opposed to only 21% and 7% detected in BA1 and BA2, respectively. Another tendency that emerges from my data, is that lecturers in medicine made use of both types of reference, i.e. exclusive (60% in M1, and 18% in M2) and inclusive (38% in M1 and 67% in M2), whereas lecturers in business management only employed the inclusive reference in the representation of groups. Certain factors may explain these tendencies. First, the discussion of medical procedures in both lectures on medicine often required explanations as to which medical practices (e.g. vaccinating and prescribing medicine) are in line with the regulations established in Lithuania. Such explanations were frequently followed with comparisons as to what is allowed in Lithuania vs. other countries. Therefore, instances of inclusive reference to larger groups of people were frequently incorporated in lectures on medicine. In addition, mes 'we' was frequently used exclusively by lecturers in medicine while explaining various processes involving the lecturer and his/her expert peers working in their field or their institution, yet excluding the student audiences. Thus, the exclusive reference has proved common in the data set. In contrast, lecturers in business administration did not focus on their professional backgrounds or the level of expertise they have in business administration in contrast to the student audience. Instead, they opted for using only inclusive reference in the representation of groups while, for example, providing information on the global and Lithuanian business climate or when introducing various laws or tax regulations implemented in Lithuania.

Metadiscourse
Metadiscourse functions are related to the organization of the lecture. All instances with metadiscourse functions are inclusive and are predominantly used to guide students through the speech event. This way lecturers may set forth what is about to come next and frequently relate their utterances to the already-mentioned information. Among the other metadiscourse functions, this use is very rare in the lectures on medicine and mostly concentrated in the lectures on business administration. The most common referent of mes 'we' with this function is that of "speaker + audience". It was predominantly used in the collocations mes fiksuojam/ užfiksuojame 'we record': (11) <....> čia mes užfiksavom tą operaciją keliaukime į sekančią <...> (BA2) 'Here we recorded that operation let us move on to the next one' In less common instances, the instructors used mes 'we' when the intended referent was the student audience, but not the lecturer, to guide the students through the speech event, for example, to signal what information should already be known by the students, as seen in (12) This function is effective in highlighting important aspects of the lecture as well as activating student knowledge and is found to be fairly common in university lectures, as observed by Yeo & Ting (2014: 31).
In some infrequent cases, mes 'we' was detected with the function "guide through the speech event" in situations where certain actions are in fact performed by lecturers, possibly to establish common ground and include the audience in their actions while signalling the upcoming direction of the lecture; for example, consider (13) There is a notable shift in the use of the singular first person (i.e. pagalvosiu 'I'll think about it') to the plural first person (i.e. pagalvosim 'we'll think about it') when expressing the lecturer's doubt regarding the correctness of the answer and the need for further mental processing of the matter. Such use reveals the lecturer's intention to promote student involvement while simultaneously guiding the audience by mentioning the upcoming course of the lecture.
Although not distinguished in Fortanet-Gómez (2004), the function of simulation was detected in my data and was primarily used in lectures on business administration (21% in BA1 and 23% in BA2).
The reference in all instances was inclusive of the speaker and the audience and was largely used in hypothetical situations, encouraging students to imagine specific scenarios in which certain actions would be performed. A rather common use of this function is detected in the collocational patterns such as mes užfiksuojame 'we record', mes parduodame 'we sell', mes investuojame 'we invest', mes nuomuojame 'we rent': Such inclusive use of mes 'we' is undeniably effective in lecturing: it promotes student understanding of the lecture material through the instructor's attempt to actively engage with students and encourage them to imagine being in real-life situations in which they would need to apply the knowledge they are gaining during the lecture.
Special attention should be paid to mes 'we' with the presentative function. In my data, mes 'we' with this function occurs only with the verb turėti 'to have' and is used when the lecturer provides new information. It is rather frequently detected in BA1 (22%), whereas only 4% and 5% of its use were found in BA2 and M2, respectively. This function was not detected in M1. In my data, the presentative function is frequently detected in the construction mes turim 'we have'. In all the instances with the presentative function mes turim 'we have' can be replaced by yra 'there is/there are' (see example 17). In addition, the results yield a frequent occurrence of the presentative function when there is a demonstrative determiner (e.g. tuos 'those') in the near context, which is in line with the observations in Fortanet-Gómez (2004: 62), as in example (17) Another function of the inclusive mes 'we' found in my data is "joint deduction", used to include the student audience in the interpretation of presented information. Like other metadiscourse functions, it is mostly used in lectures on business administration, with a frequency of 5% in both BA1 and BA2, and only 1% detected in M2 (medicine), e.g.: (18) <...> šita sąskaita būtų tinkamesnė <pause> nes čia manome kad skola trumpalaikė (M2) 'This account would be more appropriate because here we consider the debt to be short-term' A rather interesting finding is the use of the verb matyti 'to see' with this function, e.g.: (19) <...> matome kad tiksliau atvirkščias ko gero dalykas kad tikėtinų nuostolių dabartinė vertė yra didesnė už tikėtinus nuostolius <...> (BA1) 'We see that the exact opposite is probably the case that the present value of the expected loss is greater than the expected loss' Matome 'we see' in (19) is used in the metaphorical meaning of intellectual seeing/grasping. By using this verb, the speaker invites the addressee to see what the speaker thinks is important in his/her argument (Faccinetti & Adami 2008: 202). As demonstrated by Šinkūnienė (2012: 103), such use of matyti 'to see' is frequent in Lithuanian written academic discourse. Thus I expected to see this linguistic resource used in spoken academic Lithuanian, too. However, matyti 'to see' with this meaning has been detected only a few times in my data. A larger corpus could provide more possibilities in determining the use of matyti 'to see' in Lithuanian spoken academic discourse.
One more metadiscourse function, i.e. the use of mes 'we' to guide student audience through the course/discipline, is detected only in the lectures on business administration (its frequency is 6% and 5% in BA1 and BA2, respectively). In such uses of mes 'we', the lecturer performs the role of a guide and connects what is being said with the information already mentioned or still to be provided during the course, as demonstrated in example (20) In this example, the speaker relates the topic of the lecture with what has been explained previously. The inclusive use of mes 'we' helps to reduce the distance between the lecturer and the students and diminish the possibility of being perceived as an authoritarian lecturer. The choice of the verb keliaujam 'travel/move on' in this example supports the guiding function, revealing that the instructor conceptualizes herself as a guide and the lecture as a journey. The projection of a lecture as a journey is repeated in the introduction of lecture BA1, e.g.: (21) <...> pradėsim kelionę į sekiuritizaciją <...> (BA1) 'We will embark on a journey to securitization' Finally, the metadiscourse function "clarification" was detected only with mes 'we' used in reference to the speaker. It was employed to provide additional information to facilitate students' understanding, as can be seen in example (22) This function is rare in the corpus and distributed unevenly across lectures on business administration and medicine (5% in M2, 2% in BA2, 1% in BA1 and 0% in M1).
Contrary to lectures on business administration, certain metadiscourse functions of mes 'we' were either very infrequent or completely absent in the lectures on medicine. Instead of guiding students through the speech event/course/discipline, both lecturers in medicine appeared to be mainly focused on using mes 'we' when explaining the status quo of medical practices followed in Lithuania, the hospitals they work at or in their professional fields. Frequently, such uses were realized through the exclusive mes 'we' thus contributing to the lower lecturer-student engagement during those lectures.

Conclusions
The aim of this research was to gain a greater understanding of how speakers construct their disciplinary-situated identities and interact with their addressee(s) in Lithuanian spoken academic discourse. This corpus-based exploratory analysis has focused on the use of mes 'we' as markers of stance and engagement in lecturers' talk in online university lectures on business administration and medicine.
The corpus findings suggest the overall preference of the Lithuanian lecturers to use mes 'we' in their talk over other personal pronouns. However, quantitative and qualitative divergencies in the use of mes 'we' across the university lectures on business administration and medicine have revealed important differences in attempted lecturer-student rapport and the projection of stance in both disciplines. The results show that the lecturers in business administration used mes 'we' much more frequently, inclusively and in more diverse ways than the lecturers in medicine. In all instances, mes 'we' was used inclusively in the lectures on business administration, whereas the lecturers in medicine differentiated between the inclusive and the exclusive first person plural forms, with a highly noticeable inclination to employ the exclusive mes 'we'. The analysis of the referents and the discourse functions of the first person plural has revealed the lecturers' greater attempts to establish a close interpersonal relationship with the student audiences and promote interactivity in the classroom in business administration. These lecturers frequently positioned themselves as guides or equals with their students in order to improve and maintain their comprehension, help navigate the lecture/course and promote learner involvement.
In contrast, the instructors in medicine were found to be more concerned with projecting their roles as competent professionals in their specialized medical fields, frequently using the exclusive reference to indicate their belonging to their professional communities and highlight higher levels of expertise. Such use of personal pronouns has been attested by many studies, including the current one, to create distance in the lecturer-student relationship.
Due to the small sample used in the analysis, the cross-disciplinary differences outlined in this exploratory study could be subject to the idiosyncratic preferences of lecture delivery in both disciplines under investigation. Further study would benefit if more lectures representing each discipline were included in the linguistic analysis. In addition, a more diverse repertoire of disciplines would undeniably help generate greater insights into how Lithuanian speakers project their stance and seek rapport with their audiences in different academic communities. Regardless of these limitations, the description of the possible referents and discourse functions of mes 'we' outlined in this study might be beneficial to educators, grammarians, and lexicographers, as this is the first attempt to investigate the personal pronoun use in Lithuanian spoken academic discourse.

List of abbreviations
BA1 -business administration lecture on credit rating procedures BA2 -business administration lecture on internal audit tasks M1 -medicine lecture on new-born care M2 -medicine lecture on the medical treatment of cancer patients