Leadership, Management and the Welsh Language

Adopting a multilingual perspective, this empirical paper explores the varied use and meanings of referents of leadership and management among 24 Welsh senior managers. We argue the importance of recognizing the linguistic imperialism of the English language &#8211; and its dominant leadership discourse of the heroic individual leader &#8211; over locally signified referents of leadership and management in other languages and discourses. At present, the lexical item leadership lacks evaluation of its relevance and meaning within other languages and multilingual contexts. We add to this gap through a discussion of research that analyses 24 Welsh senior managers&#8217; reflections on referents of leadership and management and their varied meanings in their two official workplace languages, namely, Welsh and English. Based on the findings, we explore linguistic power dynamics represented in their reflections on this language diverse work setting and conclude with implications for the practice of leadership and management.

This paper adds to our understanding of referents of organizational leadership and management in Welsh and English through analyses of their diverse meanings for 24 bilingual senior managers in Wales and by reflecting on linguistic power dynamics within this language diverse work setting. Such linguistic analyses are timely and important in recognition of existing critiques (e.g. Collinson, 2011;Alvesson and Spicer, 2012) of organizational leadership as a western (Guthey and Jackson, 2011) conceptualization, dominated by a discourse that promotes masculine, middle class, white images of leaders, rooted in the cultural and linguistic boundaries of the assumed academic and business lingua franca: the English language (Jepson, 2010a). The English word leadership within this dominant discourse is said to derive from a mixture of Old German, Old English and Old Norse (Grint, 2010) with a shared focus on direction and travelling forth. This metaphorically powerful (O'Reilly and Reed, 2010) future, goal oriented meaning of leadership has within this discourse been closely tied to that of management, anchoring the meaning of the latter firmly in the day-to-day handling and controlling of objects and processes and rendering it inferior to the more powerful image of leadership (Ford et al., 2008).
The field of leadership studies has started to recognize and challenge the previously unquestioned dominance of this mainstream discourse of the transformational, charismatic individual leader (Collinson, 2011), highlighting the linguistic heterogeneity of leadership and management within the English language. Yet, one important linguistic issue remains largely ignored in the field of leadership studies (Schedlitzki et al., 2016) and therefore requires our attention: the discussion of linguistic imperialism of English (Phillipson, 1992) as an academic and business lingua franca (Tietze et al., 2003;Tietze, 2008), its influence on organizational discourses and its impact on publication and leadership and management development processes. Schedlitzki et al. (2016) have argued that the dominant meaning of the lexical items of leadership and management, rooted in Anglo-cultures and deeply embedded in dominant leadership discourse, has colonized organizational practice and academic research to the extent that it has limited our insights into and search for other lexical items, referents for and meanings of leadership. Recognizing Ozbilgin's (2011: 2) call for 'contextual studies of management', we propose that it is time for leadership studies to take a situated perspective that explores the linguistic roots and multiplicity of lexical items such as leadership and management within multilingual workplaces. This paper contributes to this endeavour through its exploration of lexical items and referents of leadership and management used by Welsh managers in their reflections on their multilingual work context. We discuss some of the shared meanings found within and across Welsh and English and how these relate to or create tensions with the dominant English-language based discourse of leadership and management. Our research objective therefore is to explore the varied use and meanings of referents of leadership and management and reflect on the linguistic power dynamics present in participant reflections rather than the development of a Welsh model of leadership. We conclude with reflections on the implications this has for the practice of leadership and management in multilingual work contexts.

Multilingualism and its place within leadership studies
The last ten years have seen a growing body of critical views that 'challenge the hegemonic perspectives' of mainstream, Anglo-centric leadership studies (Collinson, 2011: 182). These critical voices problematise the taken for granted power relations and dominant leader identity constructions produced through research conducted in largely white, middle class, western, English-speaking, male contexts (see Ford et al., 2008;Collinson, 2011;Fairhurst, 2011b;Evans and Sinclair, 2015). They are particularly concerned with the extent to which this dominant leader identity is culturally coded (Sutherland et al., 2014) as an individualised, white, masculine image with legitimised power over the subservient, feminised follower (Ford, 2010;Gordon, 2011). Popular theories embedded in this hegemonic perspective, such as Transformational Leadership Theory (Bass and Riggio, 2006), are argued to promote heroic leader fantasies that promise greater organizational performance (Alvesson, 2010;Alvesson and Spicer, 2012) but are unachievable and disadvantage those who do not fit into this ideal image. Evans and Sinclair (2015: 3) further remind us that 'the very idea of leadership itself is alien to some cultures'.
In response, Collinson (2011: 190) has encouraged leadership studies to value 'multiplicity, diversity, simultaneity and difference'. He argues that conceptualizations of leadership have tended to be overly positive and dualistic, creating a real need to study and understand multiplicity through an exploration of conceptual overlap, tensions and paradoxes (Collinson, 2014). Evans and Sinclair (2015) add that leadership research needs to break apart dominant and colonizing views of leadership and account for other culturally dynamic, complex patterns of leadership. Both calls have been partially answered by studies taking a power, gender and diversity perspective (e.g. Gordon, 2002;Ford, 2006Ford, , 2010Harding et al., 2011;Collinson, 2011;Muhr and Sullivan, 2013) and those exploring leadership in indigenous contexts (Warner and Grint, 2006;Bolden and Kirk, 2009;Julien et al., 2010;Turnbull et al., 2012;Evans and Sinclair, 2015) and providing linguistic analyses (Prince, 2006;Holmes, 2007;Koivunen, 2007;Jepson, 2010aJepson, , 2010bO'Reilly and Reed, 2010). Jepson's (2009aJepson's ( , 2010b empirical analyses of interviews with German employees in the chemical industry, for example, revealed several sample-wide themes that provide a socio-historical account of what leadership means to her participants in their reflections on their specific work context. Working with German language expressions (Mueller, 2007), Jepson (2010b) is able to explore the specific role of Kompetenz (competence) in the leader-follower relationship and trace its historical origin and roots in German society's beliefs in collective identity, inward-looking individualism and Ausbildung (education) as a means to freedom and selfenlightenment (Jepson, 2010b). This socio-historical reading of Kompetenz and its association with an organic paradigm (Barley and Kunda, 1992) of leadership stands in clear contrast to the dominant characterization of German leadership as 'autocratic'symbolic of a mechanistic paradigm of leadershipin cross-cultural management studies (Jepson, 2010b). Apart from highlighting this linguistic misrepresentation of a local signification of leadership, Jepson (2009a: 73) also demonstrated the dynamic and diverse nature of these language expressions, 'supporting Ailon-Souday and Kunda's (2003) argument that national identity is only one of many factors shaping an individual's' identity.
Similar research into linguistic differences, such as Prince's (2006) work on Taoism, has highlighted the importance of challenging universalist assumptions and the need to explore other meanings of leadership. Discursive approaches have also found a stronger foothold within the field and significantly developed our understanding of the discursively constructed and performative nature of leadership theory and its impact on leadership and management practice (Ford et al., 2008;Fairhurst, 2011aFairhurst, , 2011b. These contributions have particularly served to challenge the above mentioned dominance of an overly positive, masculine, white, middle class image of the ideal leader within leadership discourses and show its negative implications on a diverse workforce. It has helped to make visible the way in which this dominant discourse produces, reproduces and legitimizes power relations and inequalities amongst those who are deemed to be leaders, managers or followers.
Yet, these research studies have not fully addressed the need for considering multiplicity in conceptualizations of 20 D. Schedlitzki et al.
leadership within and across languages and its implications for our understanding of the varied meanings of leadership in multilingual work contexts. Schedlitzki et al. (2016) raise concerns with the unproblematized dominance of English-language based conceptualizations of leadership and management. They argue that the Anglocentricity of academic publication systems, in particular, has disadvantaged scholars working within other languages and as such limited our insights into the relevance and transferability of the lexical items leadership and management into other languages. Steyaert and Janssens (2013) warn that taking a multilingual perspective and exploring the heterogeneous nature of language means to be aware that 'there is no mother tongue, only a power takeover by a dominant language within a political multiplicity' (Deleuze and Guattari, 1987: 7). Hence, multiplicity of languages does not only refer to several languages being present but recognizes the variation within any given language and linguistic space. As a consequence 'there is no such thing as a universal English language (…) but that people construct English as it suits their purposes in a given context at a given time' (Steyaert and Janssens, 2013: 133). This further implies recognizing language use as 'the result of a political process' and researching 'this political process (…) in situ, focusing on how the multiplicity of languages is enacted as a negotiated multilingualism' (Steyaert and Janssens, 2013: 133). Such research into the 'interplay and hierarchy between languages' (Steyaert and Janssens, 2013: 133) helps to illuminate heterogeneity within specific languages and highlight the extent to which leadership in itself is a negotiated power relationship expressed through language and discourse. As such, a closer examination of the linguistic roots and meanings of leadership and management across and within different languages and discourses will provide us with a means to enriching our conceptual understanding of leadership as a situated practice. Evans and Sinclair (2015: 3) stress the importance of recognizing multiplicity as a means to working against the risk of essentializing insights into culturally situated practice or indeed 'enabling Western notions of leadership to be the uninspected norm against which all others are measured'. By giving voice to otherwise silenced notions of leadership meaning, we can create an alternative discourse (Sliwa and Johansson, 2014) that encourages contextually sensitive leadership practices. By making visible power relations and alternative leader images, such analyses may hence bring the sought for multiplicity and diversity into leadership studies and practice.
In recognition of the identified need to re-evaluate dominant, taken-for-granted, Anglo-centric conceptualizations of leadership and explore linguistic heterogeneity, this paper advances our understanding of the multiplicity of meanings that lexical items such as leadership and management can take in the multilingual context of Wales. Through the analysis of 24 qualitative interviews with senior managers working in organizations in Wales, the paper intends to make visible the 'multilingual side' (Steyaert and Janssens, 2013: 133) of leadership language within and across Welsh and English. Of importance here is the great diversity in usage and role of the Welsh and English languages within organizations in Wales, reflecting the complex socio-historical context of this language-diverse setting. Taking a power-based reading of the participants' reflections on language use allows us to highlight implications for practice. The paper begins with a brief contextual account of language multiplicity in Wales and then proceeds to a discussion of methods, analysis and findings.

The multilingual context of Wales
Modern Wales is a diverse society with two official languages -Welsh and Englishand several immigrant languages (e.g. Urdu, Somali, Cantonese, Polish, Irish). Its history and culture are deeply rooted in both ancient Celtic traditions and centuries of political rule by outsiders (Davies, 1994). This section and the empirical analyses in this paper will particularly focus on the relations between the two official languages Welsh and English, recognizing their predominant use within managerial contexts in Welsh organizations.
Jones- Evans et al. (2011: 220) stress that, particularly due to Wales' history as a dominated country, the 'Welsh language incorporates the unique experiences, thoughts, and expressions of a people over the last millennium and a half', marking the ability to speak Welsh as an important aspect of social identification. This is reflected in the prominent role that cultural activity in the form of music and poetry has taken in the history of Welsh society. Ancient Welsh traditions such as the annual 'Eisteddfod' a cultural competitionoften came into being as a reaction to external challenges and have since significantly contributed to the importance of the Welsh languageexpressed through song and poetryas a symbol of social belonging amongst Welsh speakers (Davies, 1994). This importance of cultural activity in Welsh society is reflected in historically recorded meanings of the Welsh equivalent to leaderarweinydd as a conductor (Geiriadur Prifysgol Cymru, University of Wales Dictionary). Indeed within the Welsh language, the functional noun arweinyddleaderdates back to the 16th century, whereas verb equivalents to the English leadingarwainand managingrheolhave been recorded as early as the 12th century.
The complex relationship with England is deeply interwoven in this history of the Welsh language and its role as a symbol of social belonging. It is important to Leadership, Management and the Welsh Language 21 recognize that, under dominant English rule, the dominated Welsh language was subject to public ridicule and banned from schools in the 19th century (Davies, 1994). As a consequence of this linguistic imperialism of the English language (Phillipson, 1992) within the public, educational and organizational spheres, the dominant image of business leaders bears the imprint of their ability to speak English, leaving Welsh language speakers in the minority in leadership roles and associated with the status of a 'subservient class' (Jones- Evans et al., 2011: 220). Jones- Evans et al. (2011) argue that those who consider Welsh as their first language tend to feel less comfortable speaking in English and are consequently less likely to achieve their full potential in English-speaking organizations.
Efforts to strengthen the status and use of Welsh and to elevate it to an equal position with English, have been present throughout the 20th century and continue today with the role of the Welsh Language Commissioner. More specifically, following the formation of the Plaid Cymru (Welsh National Party) in 1925 and its efforts towards seeking greater autonomy from the rest of the UK, Wales saw the foundation of the Welsh Language Society in 1962 as an attempt to save the Welsh Language from extinction (Davies, 1994). This was supported by a surge of Welsh-language speaking schools since the 1960s, the introduction of the Welsh Language Act in 1993 and the establishment of the Welsh Government in 1998all nurturing the Welsh national feeling and promoting active use of the Welsh language (Connolly et al., 2000;Jones-Evans et al., 2011). Yet, whilst Welsh is the first language in some areas in the North and West of Wales, English remains the majority language in Wales. Indeed, the 2011 census showed that only 19% of the overall population in Wales were able to speak Welsh (Jones- Evans et al., 2011). There is also diversity in dialects amongst those who speak Welsh across all Welsh countiesand particularly between the North, West and South of Walesleading to differences in vocabulary, pronunciation and grammar. Davies (1994) further notes the appearance of 'Wenglish', a Welsh English language dialect over the last century. For different generations in different regions of Wales, the Welsh language may thus represent more or less a symbol of social belonging (Jones- Evans et al., 2011), reflected in and reinforced by stark differences in Welsh language speaking abilities and active use across generations. This needs to be seen as further interacting with professional discourses within an organizational setting (Jepson, 2009b) whereparticularly in the private sectorthe English language is used as the sole de facto business language (Jones- Evans et al., 2011). This ongoing power play between different languages and discourses will arguably influence an individual's sense-making of leadership and management practice (Jepson, 2010a).
Beyond the Welsh context, Steyaert and Janssens (2013) have similarly stressed the cultural, historical, institutional and political dimensions of language use. The power play between languages and discourses in organizational settings has been generally found to have significant implications for an individual's career as language skills can impact on group and decision-making inclusion or exclusion (Piekkari, 2008;Sliwa and Johansson, 2010). Steyaert et al. (2011: 25) explore such power dynamics through the concept of 'linguascapes', a 'discursive space in which an organization (…) imagines how it can deal with its (…) multilingual composition by negotiating among various discursive options'. This concept highlights the potential for organizations to proactively deal with the political processes inherent in multilingual contexts.
Through the following empirical analyses, we explore meanings of leadership and management referents within the Welsh language and compare these to the English lexical items. We will pay particular attention in our reading of senior managers' reflections on language use in the workplace to possible linguistic power dynamics. We recognize the limitations of exploring only the interaction of the two official languages in Wales as this may not capture the full complexity and diversity of languages spoken in situ. We recognize multiplicity within these two languages and pay attention to discourses, but due to the small sample, we are unable to capture and analyse in detail diversity across Welsh and English dialects. Follow-up studies need to explore interactions of these official languages with other discourses, minority languages and compare leadership language use across different dialects.

Methodology
The empirical material we draw on for our empirical analyses here is taken from a qualitative research project into leadership and management referents in the Welsh Language, co-funded by a University located in South Wales and the Welsh Government. The research was conducted through qualitative interviews, exploring the use and meaning of Welsh and English-based lexical items linked to leadership and management within participants' responses. As such, the project sought to shed light on shared, local and contextual meanings of these concepts within and across languages. Qualitative analyses of the interviews enabled exploratory, thematic interpretations of the nuanced meanings of these lexical items and their uses within and across the English and Welsh language within the interview setting.
Each interview started with the use of word cards as a projective technique (Goetz and LeCompte, 1984). The interviewer handed a set of cards with a range of English
and Welsh lexical items related to leadership and management (see Table 1 for complete list) to the interviewee with a view to eliciting thoughts, feelings and examples about the use of these lexical items in connection with their work context and meaning attributed in the context of the interview setting. We recognize that the interview setting cannot capture examples of how participants use these lexical items in everyday discourse and as such their reflections on use may not coincide with how the lexical items are used in natural language settings. The word card method aimed to tap into thoughts on ideal forms of leadership and management as well as reflections on what participants do and/or currently see in practice, by probing conceptual thoughts, emotions and examples of actual experiences and practice. Spare empty cards were further used to encourage participants to write down any lexical items that they see as connected to the processes of leading or managing in Welsh or English. All interviews used the Welsh word cards first and then followed with the English word cards. Drawing on both Welsh and English lexical items intended to enhance insights into the multilingual work context of these participants. Once these card discussions were concluded, the interviewer engaged the interviewee in a semi-structured conversation and asked questions such as 'how would you describe a leader', 'how would you describe a manager', 'when do you use the Welsh words or the English words in practice'? This served as an opportunity for the interviewer to probe and ask followup questions to the card discussion.
This particular choice of method was informed by Jepson's (2010a) work with target language-specific expressions (see Table 1). An independent linguist and Welsh native speaker was employed to conduct the research and work with the academic team on the empirical analysis. The participants were 24 senior managers working in four public sector, one third-sector and four private sector organizations (see Table 2 for details).
Despite the overall high level of 'formal' Welsh language fluency, the degree to which interviewees felt comfortable talking about the meaning and use of the Welsh words compared to the English words varied widely across the sample. Those participants working at higher levels in organizations tended to be more engaged in discussing differences between the Welsh and English words. A reason for this could be their formal educational background, either the level attained or the level to which their formal education has been anglicized. The younger the participants, the more their education was through Welsh, though, paradoxically, they were often less articulate in their responses suggesting, perhaps, that this aspect may be more to do with register and professional lexis than degree of fluency. The organizational context also seemed to matter here with regards to organizational policies on the use of Welsh or English internally and with clients externally. Furthermore, there is no Welsh equivalent to some English words and therefore English words are evident in the original Welsh transcripts, highlighting the continuing dominance of the English language within organizational discourse and practice (Jones- Evans et al., 2011). This use of English terminology in the interviews, however, appears to be more prevalent amongst those managers either educated or with significant career experiences outside Wales.
It is worth noting that the word translations listed in Table 1 are a result of the native Welsh speaking researcher's identification of possible translations of the English lexical items into the Welsh language. Due to inherent differences in the semantics of both languages, the dictionary based meanings of English words used did not always correspond exactly to the Welsh referents typically translated and used on the cardsalthough as close an equivalence as possible was attempted in selecting the definitions. By back-translating (using the  Leadership, Management and the Welsh Language 23 Cysgeir Dictionarythe most current availablesee http://www.cysgliad.com/Cysgeir) one could obtain a better grasp of the semantic range available to the interviewees. Some are simple one-to-one equivalences; others are very broad in usage. Table 3 provides an overview of all the English equivalents found in Cysgeir related to all the Welsh prompt cards (the equivalent English prompt card is shown in bold). The word translations, the empirical material collected and analyses conducted are not treated as fixed in meaning and representative of an objective reality but rather as socially constructed in interactions (Sliwa and Johansson, 2014) between the interviewees, the researcher and the wider academic team involved in the empirical analysis. As such, the analysis process was iterative and ongoing (Lincoln and Guba, 1985), starting with the original meetings of the researcher and academic team where specific lexical items were identified and continued throughout the processes of word translation, empirical material collection and entailed three further stages of discussion and collective decision-making post interviewing. The native Welsh speaking researcher who had conducted the interviews met with the academic team to talk about his experiences in the field and the themes that he had encountered whilst doing, transcribing and translating the interviews. This meeting further helped the researcher to share knowledge on the nuances and specificities of the Welsh language with the academic team. The specific themes emerging from this first meeting were recorded and influenced the second stage of analysis where the academic team read carefully through the transcripts, coded these and discussed arising themes. These second stage interpretations of the empirical material were further compared back to the themes recorded in the first stage of analysis and grouped to produce a list of second-order themes (Gioia et al., 2013) focussed on meanings of specific English and Welsh words. The native Welsh speaking researcher was consulted on these second-order themes and a final stage of analysis then took this list and went back through the interview transcripts to ensure that nuances in meaning within and across both languages were captured. It is in this final stage of analysis that particular attention was paid to over-arching power-based themes.
We will focus in the following findings section on the discussion of three word pairs as they entail the strongest second-order themes present in the interviews.

Arweinydd and leader
Across all 24 interviews there was considerable diversity in the way participants signified the Welsh word arweinydd and the English word leader. Some respondents were immediately reflecting on the meaning of the words in context, that is, in relation to their own experiences and the organizational usage, whilst others were exploring these words in a much more abstract sense. Three types of responses emerged across the empirical material and seemed to be linked to the participants' connection with the Welsh language: 1. those that thought the words were the same; 2. those that saw a difference between the Welsh and English word; and 3. those that were just getting used to the Welsh word.
Those participants who saw a clear difference between the Welsh and English word also seemed to relate to Welsh as their mother tongue. Arweinydd was often described amongst those participants as somebody at the top of an organization, leading from the front, providing direction and as a role model. Whilst the ability to provide direction was often also associated with the English word leader, the participants suggested that a leader is not only to be found at the top but at any level throughout the Director, instructor Gweithredwr (first recorded in 16th century) Agent, executive, executor, operator The verbal form which this noun derivesgweithreduhas a very extensive semantic range, and can mean to act, implement, operate, execute. The word gweithredwr is also very commonly used for operator as well (e.g. in manufacturing).

Boss
This English word is commonly used as a loanword in the spoken Welsh language but has no place in the Cysgeir dictionary, which translates it as meistr (= master).
organization. This locating of arweinydd at the front or top was further more closely associated with cultural, nonbusiness specific roles. For example, one participant described arweinydd as a leader of a traditional ceilidh (noson lawen), a compere or a conductor: What comes to my head when I'm thinking of an arweinydd is Charles Williams leading a noson lawen [traditional ceilidh]. A leader he was nota compere or conductor is what arweinydd means in this sense. By the time I start thinking, arweinydd is a warmer word than leader. A leader to me is like the Boy Scouts, and one scout who is bigger than the others leads them on. An arweinydd to me is something different. Participant S (Private Sector, Male, 51 years) Similar to this respondent others also described arweinydd as softer, warmer and more caring than a leader, which they considered is a little more impersonal than arweinydd: arweinydd is more emotional and softer, more careful; leader is more hard, pretty straight. Certainly, arweinydd is more about care, its meaning is wider. Arweinyddiaeth is also warmer, more enjoyable. With leadership there's an emphasis on the leadera word that is harder, more direct, perhaps more inflexible. Participant M (Public Sector, Female, 53 years) I think of leader as being a little bit more impersonal than arweinydd. Also a little childish in tonemakes me think of 'follow my leader', 'Scouts' leader'. It seems a less childish term now that my own children are at nursery school. Participant F (Public Sector, Female, 40 Years) The distinction made above is reminiscent of the gender stereotypes dominant in leadership discourse (Painter-Morland and Deslandes, 2014) and could be read as attributing agentic (masculine, control) characteristics to the lexical item leader whereas arweinydd seems to be described in communal (feminine, caring) terms (Eagly and Carli, 2007). In addition, for participant E and N, leader takes on a public, political perspective whereas arweinydd bears a socio-historical meaning: Leaderthe word feels like something remote perhaps because the English language is like that to me. It's more grandiose, it does not belong to my world. I'm thinking about 'council Leader'snooty, posh, precious … arweinydd feels closer. Participant E (Third Sector, Female, 39 years) Leader feels colder, more distant. But they have the same kind of features. Arweinydd feels closer to me but perhaps it is due to it being in the language that I use often. The Labour Party leader is coming into my head, perhaps leader is more political. Arweinyddin a singing festival, eisteddfod, choirmaybe the word is more cultural. Participant N (Public Sector, Male, 42 years) What is starting to appear here is an interesting divide in the attribution of words to either work or social living contextswhere the English word is associated with one context and the Welsh with another. We need to recognize at this point that whilst most participants associated the English word more closely with business contexts, there was considerable diversity across all participants in how this was conceptualized, reflecting always their specific organizational and social, personal contexts.

Rheolwr / rheoli and Manager / Managing
With regards to the lexical items manager and rheolwr, there did not appear to be as much differentiation. However, when talking about the process of managing, three participants suggested that the Welsh word rheoli was clearer and stronger in meaning for them than the English word to manage: Rheoli is more firm, so you know more what it means. Gives more of a right to the person to manage if you say "Ti'n cael rheoli hwn" [You get to manage this], it gives power too. In contrast with "you can manage that … oh thanks" i.e. it's nothing special in English. Participant F (Public Sector, Female, 40 Years) [R]heoli is stronger. Rheoli can mean control in Welsh too, maybe that's why it sounds stronger. Participant A (Private Sector, Male, 40 years) Rheoli has at its core a directional role to it … to manage is at first glance 'to look after the whole, the bigger picture' whereas rheoli is to order to do things. To manage is softer than rheoli. Participant E (Third Sector, Female, 39 years) The interviews further seemed to reveal that the word rheoli is possibly better understood than arweinyddiaeth (leading) owing to managers attending management training programmes: But when there are external things -'training courses, etc., which bring in outside skills to the business, English is the medium. I've never been on a management course in Welshit is true for everyone else here surely. So even when someone thinks and acts and communicates in Welsh, they think about management skills in English. It is difficult to explain it. But I also think it might be easier to teach management courses in English because the vocabulary is more familiar to people. 'Sgiliau rheoli' [management skills] is a term that people understand but 'arweinyddiaeth [leadership]' is not something This last quote starts to address more directly the political dimension of multilingual work contexts. Several managers firmly stated the influence of management development programmes on their association of management and rheoli with business discourse and the significance of its delivery in the English language, situating the Welsh word arweinydd outside a business context.

Pennaeth and head
Several participants introduced the Welsh lexical item pennaethhead in Englishto describe an aspect of management and leadership that seems of particular importance to them within a work context. Participants referred to pennaeth as a title used frequently at various levels in an organizational context. Some felt it was closely related to leading whilst others linked it to managing. For participant M, pennaethas an organizational termis significantly more meaningful in Welsh: Headas in pennaeth, but doesn't work as well in English. More disjointed in English, pennaeth flows quite easily, no ambiguitypeople would know exactly what it means. With head it's not so rounded a term, possibly less authority, just doesn't convey the same clarity of meaning as pennaeth. Participant M (Public Sector, Female, 53 years) When do you use the Welsh words and when do you use the English words?
Responses to this formal interview question entailed detailed contextualization of the meanings mentioned previously in relation to the lexical items listed on the word cards, delving into a range of historical, cultural and political issues. Whilst responses were often linked to the definitions introduced above, they also reflected on some of the complexity of working in multilingual contexts and the extent to which they felt that the words leader and arweinydd are used distinctly differently. Participant F's answer illustrates this well: There's a difference in the way they are used. But that should not be. Because it gets mixed up with managing teams. We have team leaders who are basically managers of teams. In fact they manage and do not necessarily offer corporate vision and interpretation. A leader in the meaning of managing rightly. There's confusion that has developed around these terms, talking about the chief executive as arweinydd and leader [uses the English word] -'it should convey something completely different to the payroll team leader, but when leaders are at all levels as a term they do not necessarily see themselves as real leaders. For teams and project leaders, leader is perhaps a titlethey know that managing resources is their work in essence and they do not perceive the work as conveying leadership. Participant F (Public Sector,Female,40 Years) This quote seems to suggest tensions emerging where the word leader is associated with a structural role or label below senior management level in the organization as it creates conceptual overlap with the seemingly popular term manager. These potential tensions between meanings of the labels leader and manager in an organizational compared to a social context need to be further explored in relation to their socio-historical roots in Welsh history and professional, organizational discourses present. For participant B, this was linked to different meanings embedded in different discourses, where his references to leader seem to particularly reflect the dominant individual, heroic leader discourse present in mainstream leadership literature and dominating organizational discourse (O'Reilly and Reed, 2010): Leader is something like a great leader e.g. as prime minister … arweinydd is more something I'm doing. 'I'm a Leader' sounds like Moses in the Bible. I'm not a church person and maybe I've missed somethingmaybe arweinydd is a big something in the Bible. Arweinydd is more day-to-day but the leader is more sublime. Anyone can be an arweinydd but not in Englishleader is more inspirational. I do not see arweinydd as equivalent to leader. I'm more comfortable with arweinydd than leader, maybe the nature of the use of the word in Welsh. Leader in literatureis like what you get in English novels such as George Orwell's 1984arweinydd is not like that. Participant B (Third Sector, Male, 31 years) Participants also often referred to the historical dominance of the English language particularly as a business language and its implications for Welsh language speakers: It comes back to the social super-structure in Wales. Where traditionally people have not considered Welsh speakers as managers until recently. Certainly in the world of business and public business, the Welsh are usually teachers, ministers of religion etc. The heads of business are expected to come from England and be English speaking, that English is the language of business. Participants R (Public Sector, Male, 50 years) These reflections included the aforementioned dominance of English within leadership and management development programmes and some participants suggested that this weakened the status and position of Welsh language speakers in organizations and generally
discouraged the use of arweinydd compared to leader in organizational life: The English terms feel more solid and formal due to the training I have had in English. Welsh is also not an international language of business or day to day business language even in the majority of Welsh language institutions. The English words feel more formal and more accurate (…) My mind has been scheduled to consider the English terminology in business administration and that the terms then are more formal and robust than in Welsh. Participant R (Public Sector, Male, 50 years) Interestingly, this formal interview question seemed to evoke generalizations and some stereotyping across Welsh and English people, indirectly assuming cultural and linguistic homogeneity. This is reflected in participant X's cultural explanation of the relative disadvantage of the Welsh people and Welsh speakers as culturally rooted: There is a tradition in Wales of not showing yourself. I feel the difference, that people in England are more natural in confidence even though that capacity is not always with them. Coming from the system of public schools I suppose. The Welsh people have a confidence level much lower than their capacity/ability. (…) There is a pattern of a lack of venture in Wales -English people are at the top here as entrepreneurs. Welsh people are less risk-taking (…) It is also easier in England to have a variable diagonal range of work experience that will make you attractive to employers the indigenous career pathways in Wales are a lot more vertical with fewer steps on the ladder to get the appropriate leadership experience. Participant X (Third Sector, Male, 43 years) Finally, participant O expressed the extent to which there is a strong emotional engagement with Welsh words if this is perceived to be the first language of employees and the organization: They're completely different. In English jargon has devalued many common words in the business world and makes them superficial words which are often almost meaningless. The Welsh words have meanings which are far more intense and meaningful, and these words are more relevant to our workforce since Welsh is the first language of the majority of themover 80% here. Participant O (Private Sector, Male, 52 years) A sharp distinction was typically drawn by private sector participants, such as participant O above, between the voluntary use and promotion of Welsh as the main medium of internal communication operationally and the necessity and pragmatism of using English with the bulk of their customers, whom they perceived as being largely from or based elsewhere in the UK and abroad: The reps, customers will be discussing our external business in Englishour market like many other Welsh companies is a British one, so we use the English terms on that occasion, it is inevitable. Participant A (Private Sector, Male, 40 years) Recognizing the multiplicity in meanings of the Welsh and English words expressed in the interviews, we now turn to reflect further on the over-arching power-based themes found within the empirical material and highlight implications for leadership and management practice within this multilingual work context.

Discussion
The over-arching linguistic power based themes are interlinked with three core second-order themes. The first core theme emerging from our empirical analysis was the relative disassociation of the words leader or arweinydd with the participants' workplace. Indeed, those who saw a difference between arweinydd and leader associated the Welsh word more strongly with non-business related aspects of life such as choir practiceresembling historically recorded meanings of arweinydd as a conductorwhereas business life was more closely associated with the English word manager and its Welsh equivalent rheolwr. This separation of business and social life expressed in the interviews is not entirely surprising in light of the turbulent history of the Welsh language. Whilst Welsh was banned from education in the 19th century and organizational ownership increasingly in English hands, it was the cultural activities in the local, social context where the Welsh language continued to thrive. Another facet of this signification of arweinydd compared to the English word leader was the emotional connotation of both lexical items. Whilst both leader and arweinydd share a similar focus on direction, participants associated different feelings with these words. The softer, warmer feeling associated with arweinyddparticularly amongst those who looked upon Welsh as their mother tonguecould be read as symbolizing the comfort that these participants experience when using words that reflect most closely their social belonging (Jones-Evans et al., 2011: 220). We also noted that characteristics that are stereotypically coded as feminine in the leadership literature (Eagly and Carli, 2007)such as nurturingwere more closely aligned with arweinydd compared to the leader. It is in this gendered reading of the participants' reflections, that we start to see a connection to the dominant leader discourse reproducing masculine ideal leader images. We could therefore see the disassociation of arweinydd with business life as a Leadership, Management and the Welsh Language 27 result of the interaction of linguistic imperialism embedded historically in organizational life in Wales (Phillipson, 1992) and the colonization of Englishlanguage based leadership discourse with masculine images that render the cultural associations of arweinydd (e.g. as a nurturing compere) subordinated to the stronger, competitive signification of leader.
The second core theme looks at this complex role of the dominant English-language based leadership discourse in more detail. The rise of the hegemony of the heroic leader and its discursive colonization of organizations and executive education is well researched (e.g. Ford et al., 2008;O'Reilly and Reed, 2010). Ford and Harding (2007) and Ford et al. (2008) have particularly argued that the title manager has not only been replaced largely in organizations by the more popular term leader but that managers are also expected to become and take the identity of a leader. Whilst our research showed a familiar gendered reading of the lexical item leadership, it did not suggest the dominance of this discourse of leaderism (O'Reilly and Reed, 2010) but instead reflected the continued focus on managing and being a manager in the language used by our 24 participants. Participants seemed to associate this greater familiarity with the Welsh and English words for manager, compared to leader, with their predominant attendance of management rather than leadership development programmes. In addition, participants frequently drew on the Welsh word pennaeth head in Englishto describe a seemingly Welsh language specific concept that could encapsulate both processes of leading and managing. Incidentally, this Welsh word pennaeth historically predates all other Welsh words included in this research (Geiriadur Prifysgol Cymru, The University of Wales Dictionary).
The third core theme that emerged from the interviews relates further to the shared meaning between manager and rheolwr and the connection this draws to the continuing power of the English language over business life and education. One participant, in particular, suggested that the likeness of manager and rheolwr in meaning was a consequence of participation in English-language based management development programmes. Others agreed that the dominance of the English language in professional education seems to have allowed English concepts to become strongly embedded in organizational life, fundamentally shaping expectations held by organizations about effective management. This influence of English language based management constructs in organizational practice could be seen to further reinforce the linguistic distinction of the Welsh word arweinydd as rooted in a social rather than business context. The concept of the 'manager' or rheolwr could further be seen to represent a commoditized Anglo management development product, colonizing locally meaningful, historically rooted referents of managing and leading such as pennaeth.
Finally, the expressed disassociation of arweinyddleaderfrom the business context in these interviews may reflect possible tensions for individuals when working in organizational structures that use the label of 'leaders' in connection with roles of responsibility. Such labelling is popular within Anglo organizations highlighting the influence of the dominant discourse of leaderism (O'Reilly and Reed, 2010) and may lead to tensions between the exaggerated expectations discursively imposed on this label and the situated experience of those enacting such roles (Ford and Harding, 2007;Ford, 2010). Participant F made an explicit reference to the tensions emerging through pressures to assume and use words such as leader to describe one's own role when these are perceived to be at odds with local, socially rooted notions of arweinydd or pennaeth. Unlike Jones- Evans et al. (2011), our findings do not suggest a direct link between being or speaking Welsh and career progression into senior management or leadership positions. Yet, our participants' reflections on language use in the workplace do make visible some of the power dynamics at work where the dominant business and educational language (Steyaert et al., 2011;Steyaert and Janssens, 2013) has normalized the use of English words for management and leadership that are signified through dominant managerialist discourses, thereby affecting narratives of self at work (Sliwa and Johansson, 2014) and subordinating Welsh lexical referents and signification of leadership and management.
Our insights into the diverse use and signification of leadership and management amongst the 24 Welsh managers and the power dynamics of linguistic imperialism and discursive colonization bear implications for our knowledge of leadership and management within multilingual work contexts. It stresses the importance of challenging the false assumption of a universal meaning of leadership and management and highlights the plethora of locally rooted variations impacting perceptions of self at work and images of effective leadership and management. It further shows the importance of gaining further understanding of the power play of dominant languages and discourses at work and their potential to render locally meaningful referents of leadership and management subordinated. Seeing all organizations as multilingual work contexts, where either linguistic imperialism and/or discursive colonization is present, it invites research into dialogue within the workplace that may help redress this power imbalance and thus open up spaces for different ways of leading and managing effectively.

Concluding remarks
This paper sought to advance our understanding of linguistic multiplicity in meanings and uses of referents
of leadership and management in the context of Wales. It argued that there is a real need for contextual studies (Ozbilgin, 2011) as a means to strengthening our understanding of multiple and possibly competing referents and signification of the lexical items leadership and management. This is particularly important in light of the dominant, essentialist strand of leadership theory and discourse that takes the Anglo-centric meaning of the word and its relationship with management as universally applicable (Jepson, 2010a). Our analysis of empirical material from a qualitative research study with 24 senior managers working in nine Welsh organizations has enhanced our understanding of this multilingual work context in two ways. First of all, the analyses of the participants' reflections on referents of leadership and management used within and across the two official work languages Welsh and English, showed different dynamics between these lexical items compared to the dominant organizational discourse of leaderism (O'Reilly and Reed, 2010). Indeed, it suggested a separation of the Welsh referents, where arweinyddleaderwas closely associated with the social, cultural context whereas the business context tended to be dominated by references to the Welsh equivalent for managerrheolwrand the Welsh language expression of pennaethhead. The latter encapsulated not only the focus on a person in authority but was open to embrace both activities of leading and managing, again revealing a difference to English language based debates where managing and leading have usually been categorized as activities of either a manager or leader. Secondly, our analyses have helped to make visible power dynamics in the participants' reflections on meaning and use of referents for leadership and management within and across the two languages. Participant references to the dominance of management development programmes and their sole delivery in the English language could be seen to reflect the complex interplay of a historical linguistic imperialism of the English language within organizations and education in Wales, and the colonization of a continued focus on managerialism over organizational discourses. These development programmes can further be seen to impact narratives of self at work (Alvesson and Willmott, 2002;Sliwa and Johansson, 2014), reinforcing ideal images of managers that are potentially alien to or in contradiction with local, social values and practices and thus leading to the continued subordination of Welsh based lexical items such as arweinydd or pennaeth in connection with senior organizational positions.
Contribution to practice This paper has real implications for managers working in Wales as it 'makes multiplicity visible' (Steyaert and Janssens, 2013: 138) and enables historically rooted Welsh language expressions to be heard and understood. It further highlights the influence of historical linguistic imperialism over dominant organizational discourses of management and the confinement of the dominated Welsh referents of leadership to non-business contexts, supporting Evans and Sinclair's (2015) observation that leadership may not always be the most relevant concept.
More broadly, our research into language multiplicity ensures that leadership and management practice is understood to be contextually embedded in local, political, 'structural and cultural webs of relation' (Ozbilgin, 2011: 2). As such, organizations need to recognize that leaders and managers work within contexts of linguistic and cultural heterogeneity and need to be equipped with an understanding of 'the complexities of such plurality and differentiation' (Jack and Lorbiecki, 2007: S82). Recognizing diversity of languages and discourses alongside heterogeneity in gender, ethnicity, culture and identity is of great importance for organizations, enabling management to take a more holistic view of the nature and complexities of diversity. Our small study has highlighted the importance of making visible the power play of languages, where tensions arise between dominant labels of leadership and their dominated, local referents. The dominance of Englishlanguage based management development programmes in our study has reinforced ideals of a leader and manager that are embedded in dominant Anglo discourse and subordinated locally rooted notions of leading and managing. This may disadvantage potentially effective leaders and managers who do not fit or are unable to discursively narrate their self identities accordingly (Sliwa and Johansson, 2014). Recognizing these realities and implications of multilingualism may ultimately make existing power relationships visible and enhance active dialogue in organizations that enables language to be seen as polyphonic. This focus on language multiplicity and power dynamics should then form part of national and international management development programmes in organizations in order to raise awareness of the realities of multilingualism and to encourage dialogue as a core feature of successful leadership and management processes.
Future research Further research into the semantics of referent word use within multilingual contexts would enhance our understanding of locally situated perceptions of leadership and inherent power dynamics. These studies could replicate the word card technique we have introduced here and complement these through explorations of meaning associated with other media, such as, artefacts, imagery, paintings or poetry to elicit further thoughts and discussion in relation to leadership and management (Taylor and Ladkin, 2009;Schyns et al., 2013). In addition, it will be of great importance to bring Leadership, Management and the Welsh Language 29 a longitudinal, ethnographic element to this kind of research. An ethnographic study in a Welsh organization, for example, is likely to shed light on language practices and political processes of language performance, particularly linked to the words leader/arweinydd, manager/rheolwr and head/pennaeth, but potentially also highlighting their interaction with other minority languages, differences across dialects and professional and organizational discourses. Through observations of meetings and daily interactions within groups, the researcher is able explore the diversity of the meaning-in-use of these lexical items within specific professional and organizational contexts and highlight possible other social identifiers such as gender, race, sexuality and age. Building on this example of the Welsh context, we hence recommend contextual studies of leadership meaning-in-use in other international and language diverse work settings that focus particularly on how dominant and minority languages are being used and interact in situ and their influence on perceptions of leadership and management in real terms in organizations.