Abstract
This contribution explores transgenerational language change in a historical migrant community by qualitatively examining the correspondence of first- and second-generation Scottish immigrants coming to New Zealand in the nineteenth century. Taking a microsocial approach, the letters of a migrant family and one other migrant are explored for language maintenance and shift, to identify whether Scots language features were lost altogether or continued to be utilised for specific social, personal and stylistic goals, despite the English-dominant space that the migrants operated in. In tandem, the adoption of early New Zealand English (NZE) and te reo Māori lexis is analysed, to identify differences in usage patterns that might point to different degrees of integration and mobility. Finally, inter-writer and inter-generational differences are examined in relation to the mobility and social networks of the correspondents, to consider how this might contribute to any variation observed. For the investigation, manuscript letters were digitised, and relevant features identified, extracted and discursively analysed. Results show the continuation of heritage features through a combination of style-oriented goals and learned letter-writing practices, while the adoption of new lexis is shown to occur within specific semantic domains that reflect the social mobility of the migrants. However, language maintenance and shift are not uniform between the writers, elucidating the highly variable experiences of migrants, even within the same family. Rather, contact-induced language changes are sensitive to minute differences across individuals, underpinning the value of nuanced explorations of historical migration and language change.
1. Introduction
Migrant diasporic communities have seen increasing investigation within both contemporary and historical sociolinguistic analyses, with a growing recognition of the light they can shed on contact-induced variation and change, and the consequent development of new dialects, idiolects and sociolects. Migration settings often engender high degrees of contact between linguistically diverse speakers, creating scenarios unmatched by their everyday lives (Hendriks et al., 2018, p. 124). This provides us with a rich and nuanced landscape to explore, but much remains to be investigated when it comes to transgenerational language shifts in migrant vs. heritage languages across the world. This is particularly true of historical cases of migration, in which we are left with what has been fortunate enough to be preserved: in some cases, this is whole collections of correspondence, in other cases, a single letter or diary entry. The unevenness of data availability, combined with its written nature, has perhaps contributed to the lag in sociolinguistic investigation of migrant and heritage language variation within historical communities. This trend is fortunately changing, and this paper seeks to contribute to this scholarship, providing a diachronic perspective into the changing linguistic loyalties and practices of migrants whose heritage language is closely related to the language community. It investigates possible language maintenance or shift among the nineteenth-century Scottish migrants to New Zealand, by considering both the extent to which heritage features (Scots) are maintained by the first and second generations, and whether their letters contain host language lexis (New Zealand English (NZE) and Māori) alongside, or instead of, Scots.
Representing this community is the MacDonald family—a multi-generational, working-class Scottish migrant family who settled in the small, rural town of Halcombe on the North Island and took up farming. The correspondence of the first-generation father, George, and his four second-generation children, who include three sons (John, Alec and Albert) and one daughter (Flora), is examined for Scots, NZE and Māori features, which are then discursively analysed. To investigate the possible role of gender (and class) within this linguistic scenario, the letter of another, unrelated female migrant—Rachel Hepburn—is also included in the sample. Rachel falls within generational categories to some extent, as she was born in Scotland but immigrated at just five years old to the region of Otago, on the South Island of New Zealand, with her lower-middle-class family. Rachel, like the four MacDonald siblings, writes to a family member (her sister) in her letters. Together, the six writers represent varying stages of the migration process as well as varying degrees of connectedness to their ancestral homeland (Scotland) and their new place of residence. Moreover, they are based in two contrasting areas of Scottish settlement in New Zealand: the thinly populated North Island versus the overwhelmingly-Scottish province of Otago. Given the importance of social mobility and networks in migration settings, inter-writer differences within generations are compared in relation to the writers’ backgrounds and possible social or stylistic motivations.
This gives us an insight into language changes among the Scottish immigrants arriving on New Zealand’s shores during the mid-to-late nineteenth century. Most originated from the Scottish lowlands (around 70%) and came from agricultural or light industrial backgrounds, thus they were generally of the lower-middle or upper-working class. Consequently, Scots was likely to be the language of the home. Despite immigrating to New Zealand in numbers well above their population ratio, many Scottish migrants found themselves among a colonial society in which the majority spoke southern English varieties and did not have Scots as part of their linguistic repertoire. This undeniably had an impact upon their retention of the heritage language, which is borne out by the nature of the New Zealand English dialect today, which contains very few features of Scots. Alongside contact with migrants from other British locations, the choice to retain Scots was also coloured by linguistic developments that had been ongoing in Scotland. Scots had been undergoing anglicisation over the previous three hundred years, as the fallout of a series of historical events that led to a closer connection with England, coupled with the rising prestige of the incoming southern English Standard. This meant that by the nineteenth century, Scots had acquired the status of a ‘domestic’, ‘homely’ and ‘colloquial’ variety (Murison, 1964, p. 9), and had fallen out of use from most written genres. Once in New Zealand, this process only continued, as Scots became a private variety used entirely within the household, unsuitable even for informal communication with non-family members.
As a language that was already highly restricted in its use pre-migration, the further shift away from features associated with Scots would thus suggest that dialect levelling affected the entire sociolinguistic repertoire of Scottish migrants—not just formal registers but also the most personal and integral part of their linguistic identity. Its status and stigma no doubt played a role in conditioning the migrants’ shift away from their heritage variety, particularly for second-generation speakers who typically have greater impetus and ability to integrate and become socioeconomically mobile (Hannafin, 2025). This shift was plausibly aided by the highly similar nature of the two varieties, thanks to their shared linguistic roots (see Section 2.3) and the ongoing anglicisation of Scots. In many senses, the situation investigated here is akin to speakers adapting their dialect to the target variety, and this research adds a new diachronic insight into how this scenario may play out for different generations of migrants. As the results will make clear, language variation can occur across multiple linguistic levels simultaneously, reinforcing the typologically distinct nature of Scots and English and the non-teleological outcomes of dialect and language contact in migration settings.
This paper is structured as follows. Section 2 will consider previous research into migrant and transgenerational language change and contact, examining the role of identity shift and its ramifications for new dialect and language development in migration settings (Section 2.1). This will be accompanied by a brief overview of the development of New Zealand English and the new dialect formation theories that have been developed specifically from this variety. Previous research examining other groups of migrants to New Zealand (such as the Irish or English) is also referenced (Section 2.2). Following this, an overview of the historical background regarding the Scots language is provided (Section 2.3). Finally, a historical overview of the Scottish settlement of New Zealand is provided, with reference to the colonial scenario that emerged during the nineteenth century (Section 2.4). This is followed by Section 3, which details the materials and methods used for this study. Background information on the individuals who comprise the migrant family under investigation is given (with more detail for each migrant presented in Section 4), and specifics around the documents used for this analysis are outlined (Section 3.1). The process of digitisation and tagging of relevant linguistic features is then briefly summarised Section 3.2). Section 4 shows the results for each family member, and the comparative behaviour of Rachel Hepburn, in turn, before Section 5 discusses their behaviour and what this tells us more broadly about heritage language use and transmission across generations in migration settings. Finally, Section 6 concludes on the findings of this study, their relevance to this emerging field, and potential future directions.
2. The Social, Linguistic and Historical Context of Migration Linguistics and New Zealand
2.1. Migration, Language and Identity
The field of migration linguistics has seen increasing awareness, investigation and development over the previous two decades, though more studies specifically examining transgenerational changes are still emerging. In particular, there has been a large focus on the factors that influence language maintenance and shift among migrating groups (Wright, 2020, p. 481), and on the range of decisions faced by individuals in deciding to maintain the family language and transmit it to the next generation. There are some complexities to taking a blanket approach to migration and language change: migratory settings can be wildly different, and so too can their linguistic ecologies. The circumscribed nature of traditional notions of linguistic migration—in which citizens of a nation speak a singular, defined and uniform variety—fails to recognise that for an overwhelming number of speakers, their linguistic practice forms part of various continuums that change across space and time (Wright, 2020, p. 482). This notion applies particularly to the Scottish, whose language variety had been undergoing substantial change and shift already prior to immigration, and whose linguistic repertoires included features shared by English and Scots, as well as variable options from either language depending on the social and stylistic context.
There are multiple possible outcomes when groups of language users migrate—they may continue the language practices of their place of origin, they may shift towards the dominant or prestige variety within their new place of residence, or they may expand their linguistic repertoire to include both varieties or indeed a multitude of languages, dialects or sociolects depending on the environment they have migrated to. Perhaps unsurprisingly, language shift or maintenance in migrant communities shares many overlaps with wider linguistic developments observed in stable speech communities elsewhere, in which language users navigate the bi-directionality of prestige issuing from both heritage and standard languages (Kasstan et al., 2018, p. 388). As language users often specifically focus on economic benefits and improvement, it is not surprising that the socioeconomic value of a language has been identified as a key factor in determining both the linguistic vitality of migrant groups (Bourhis et al., 1981; Edwards, 1985), and the potential for stable bilingualism in a community (Landry & Allard, 1992; Romaine, 1989). Groups who migrate for economic reasons rather than to escape religious and political suppression frequently tend to shift to the language practices of the host group, and when this is coupled with emergent standard language ideologies that favour a singular language variety for the newly settled territory, integration pressures become exponentially greater and less flexible. Thus, non-standard or minority languages tend to be subsumed by national or prestige standards (Wright & Viggiano, 2020, p. 492). Considering the minority status of the Scottish migrants and their drive for social mobility, these factors were likely to be highly influential and may have encouraged language shift in order to expedite social mobility and enhance their acceptance into elite fields such as entrepreneurship.
Even when speakers have positive views towards their own language variety, if this is stigmatised by wider society as low-prestige, speakers may come to share these views in migration settings (Bettoni & Gibbons, 1988). Minoritisation, stigmatisation and prejudice can have converse effects, however. While they often encourage assimilation as the path to incorporation, this may generate a counter-reaction and feelings of solidarity with fellow migrants, which is then asserted through the use of linguistic features that index their shared backgrounds. Given that most immigrants are adults when they leave, their set of memories from the homeland is formed during youth, giving it a strong emotional and identity-defining status that they cling to and re-enact through symbolic practices, including their language use (Hickey, 2019, p. 6). This generates a continual feedback loop between practice and identity construction. Unsurprisingly therefore, the close connections between migration, language and identity have been identified and confirmed time and again in linguistic studies within the larger context of identity formation and integration (see, among others, Canagarajah, 2017; Koczan, 2016; Llamas & Watt, 2010; Preece, 2016; Regan et al., 2016; Schwartz et al., 2011). Historic cases of migration and identity are abundant, and research has focused in particular on the Early Modern period during which Western Europe embarked upon various colonial enterprises throughout the rest of the world (i.e., Dossena, 2019; Hendriks, 2018). Applying the sociolinguistic and variationist toolkit to contact and heritage linguistic studies has proven fruitful (see, for example, Bousquette & Brown, 2018; Cognola et al., 2019; Eide & Hjelde, 2015; Nagy et al., 2018; Sánchez-Muñoz, 2010), although Kasstan et al. (2018) note that migration linguistics has still seen comparatively less attention compared to other heritage–linguistics fields, and more focus on extra-linguistic factors is required. Continued use of the heritage language is particularly revealing for socially- and stylisticially-motivated variation, acting as an invaluable linguistic resource offering a specific means of collective identity construction, whilst enabling the transmission and retention of heritage features among second, third or even fourth generation immigrants (e.g., Bagwell et al., 2019; Chambers, 2008, p. 105; Litty et al., 2015). Indeed, the Scottish were often remarked upon for their ‘clannishness’ and exclusive communities in New Zealand (McCarthy, 2006, p. 9), which may similarly have fostered the continuation of Scots writing practices and cultural transmission.
Conversely, migrants’ linguistic repertoires may change or expand as they become members of different, loosely-connected communities, establishing new network ties (Croft, 2000, p. 87). The intensity of belonging to different social networks can also change over time and generations (Brubaker, 2006; Wright, 2020), as migrants negotiate their social and linguistic place within the wider host society (Litty et al., 2015, p. 186). Demarcation and integration do not have to be mutually exclusive; after all, there can be a range of behavioural acculturation strategies or orientations that migrants adopt (Berry, 1997; Karpava, 2025, p. 92; Ladilova, 2015, p. 177; Madsen & van Naerssen, 2003, p. 61). Moreover, not all individuals will experience migration the same way, and the interaction between personal or self-identity and collective or social identity needs to be kept in mind: migrants could at times feel part of their larger ethnic community and at times have personal motivations and ambitions that interacted with the self-construction of their identity through language (Auer, 2021, p. 288; Auer & Thorburn, 2021, p. 2; Chambers, 2008, p. 103). Thus, while some individuals might change their idiolects over the course of their lives, others can remain relatively stable or retain the heritage language or its features to a large degree (Hendriks, 2018). Their language thus becomes a window into the ”crosscutting dynamics between creative readjustment of traditions, resistance and assimilation” that are at work in the formation of collective identities (Stievermann, 2013, p. 14).
As the heterogeneity of immigrant groups has been increasingly recognised, patterns of both idiolectal continuity and change have been identified simultaneously within the same migrant community (Hava, 2018). Bonness (2019, p. 201), for instance, investigated an Irish immigrant family to New Zealand, and found differences in the brothers’ retention of heritage features that reflected their level of connection to the relatively close-knit Ulster community in the new settlement. Changes in social affiliations are hence mirrored in, and contribute to, changes in linguistic patterns, and these changes are highly variable and malleable rather than binary switches between different versions (Litty et al., 2015, p. 186). This can generate transgenerational changes in what defines the heritage language and ancestral ‘roots’ of multi-generational speakers (e.g., Bagwell et al., 2019; Karpava, 2025), in which certain features become more or less appropriate to mark identity or other stylistic practices (Lauersdorf, 2018, p. 23). This transcultural view highlights the creative possibilities and multiple cultural repertoires exhibited by migrants across generations, enabling them to perform hybrid linguistic and cultural identities that reflect being and belonging in varied ways throughout their life trajectory, and meet social or stylistic needs in specific contexts (Cubas et al., 2023; Liu & Lin, 2019).
Generational language shift among migrants has been confirmed in both contemporary and diachronic studies (i.e., Bagwell et al., 2019; Hava, 2018), which is unsurprising given the experiences of first-, second- and third-generation migrants are quite radically different at times. This will have consequences for their sense of self, identity and belonging to different communities, as well as their language or dialect acquisition (Hendriks, 2018, p. 151). Children born of immigrant parents have much more opportunity to acquire an insider’s knowledge of the habits, traditions, cultural practices and significantly, language(s), of the host country (Karpava, 2025, p. 93), though claims of belonging regarding their heritage are also more complicated and easily challenged in consequence (Hannafin, 2025). Parents’ positive attitudes towards the heritage variety often influences children’s acquisition and maintenance, or else shift and loss, of the language (Fishman, 2004), although a number of recent studies have suggested that this link is somewhat more ambiguous than would first appear (e.g., Gharibi & Boers, 2017; Kircher, 2022). The use of the heritage language, or whatever vestiges remain of it, is often a marked choice for post-immigration generations, and their ability to develop their transnational capital is constrained by time, place, social hierarchies and power relations, which might trigger pragmatic reasons for abandoning the heritage variety (Karpava, 2025, pp. 97–101). Nor will the next generation necessarily take up and continue any collective identity desires.
The extent to which these influences play out when we explore written data is currently unclear in the case of the Scottish, though previous sociolinguistic analyses into historical migration settings have suggested that both assimilation (c.f. Bonness, 2019 and see Section 2.2) and demarcation (c.f. Auer, 2021; Auer & Thorburn, 2021; Bagwell et al., 2019; Litty et al., 2015) are possible outcomes. As Section 2.2 and Section 2.3 will make clear, the former seems most likely, though Section 2.4, detailing the historical background of Scottish migration to New Zealand, questions quite how straightforwardly we can make this assumption.
2.2. Migration and Dialect Development in New Zealand and Beyond
Linguistic systems can be radically restructured under the impact of mobility resulting from historical migration, which gives us insight into both language and societal changes (Britain, 2012). Exploring transgenerational changes within migrant communities accordingly allows us to explore related linguistic processes, including dialect levelling, language maintenance and shift, the development of a heritage dialect, hybrid language use, and in particular, new dialect formation (NDF) (Auer, 2021, p. 283; Auer & Derungs, 2018, p. 1; Hava, 2018, p. 67). New Zealand English (NZE) has been a prime inspiration and test site for Trudgill’s traditional NDF model (see Trudgill, 2004; Trudgill et al., 2000 in particular), though this is not without dispute. His theory has tended to be quite deterministic and focused on the wider language pool, suggesting that the selection of features that made it into NZE was largely frequency-based. The features that were in the majority or spoken by more than one dialect group settling the new colony became the default standard. The minority variants and highly-marked, salient regional features on the other hand, were not preferred and were quickly abandoned by the second generation. While offering some compelling evidence and a systematic design for exploring NDF, the role of social factors and stylistic goals is clearly missing from such a model and does not neatly account for all variants present in NZE. Schneider’s (2007) Dynamic Model, also based on NZE, is more socially nuanced. It assumes that linguistic change in the post-colonial setting is driven by speakers’ wish to create their linguistic and social identities, accommodating to others by selecting from a shared pool of language features (Schneider, 2007, pp. 26–27)1. Both these theories, however, have been based primarily on early spoken data and do not consider the language use of first-generation immigrants (for which there were no speech recordings) or the possibility of inter-generational transmission. Nor have the influence of family networks and transnational identities seen much linguistic investigation, despite the obvious feedback loop between them. Indeed, migrant language use often says more about the social structure of the community than its linguistic composition (Chambers, 2008, p. 107), especially as geographical mobility can lead to social mobility and vice versa (Lauersdorf, 2018, p. 22).
Rather surprisingly, the Scottish have been curiously overlooked in the sociolinguistic investigation of New Zealand’s migrant groups, despite the wealth of historical research into their contributions to New Zealand society (i.e., Brooking, 1985, 2006; Bueltmann, 2011; Harper, 2003; Hearn, 2003; Lenihan, 2015; McCarthy, 2006, 2012). Some valuable linguistic analyses have been undertaken in the context of Irish (Avila-Ledesma, 2019; Bonness, 2017, 2019) and English settlers (Hundt, 2012, 2015; Hundt & Szmrecsanyi, 2012, 2015), though the latter focused on letters printed in a British newspaper, so it is unclear how much silent editing may have taken place in what became essentially public-facing documents. Bonness (2017, 2019) is the only scholar to have taken a transgenerational, sociohistorical approach to diachronic correspondence data, examining three members of the McElreath family who emigrated to New Zealand from the Ulster province of Northern Ireland. She explored how their language use changed during the course of several decades, as well as the linguistic practices of their children. She found evidence of an expanded repertoire that included options from across the linguistic spectrum, which included Irish English features, Māori loanwords, New Zealand English and Australian colloquialisms, and morphosyntactic variation as the migrants adapted their language use over the years. Their children, on the other hand, showed evidence of rapid levelling: they used the standard form almost categorically when it came to morphosyntactic structures (Brooking, 1985, pp. 143–145).
Looking beyond New Zealand to other diachronic studies, Auer (2021) found that the dominant English language had a clear impact on the Swiss migrant community of New Glarus, Wisconsin, first settled in 1845. The transfer from English was mostly on the level of lexis, and a large-scale shift to English took place after the community had become more vertical—pertinent concerns for later generations of Scottish immigrants to New Zealand who similarly experienced an all-encompassing shift to English norms and rapid, increased social mobility. Hendriks et al. (2018) investigated correspondence produced during the period of Early Modern Dutch urbanisation (1500–1800), which saw huge in-migration to certain urban locations in the Netherlands from other Dutch and German locations. They identified numerous self-corrections towards the heritage or the host language variant and considerable code-switching in the letters, as well as transfer of the same heritage language features into the host language between the first and second generations, suggesting the children reflected the practices of their parents (Hendriks, 2018; Hendriks et al., 2018). Bagwell et al. (2019, p. 39) similarly found evidence of historical transfer from German, right through to the fourth generation, when investigating the ethnically-German community of Dodge County, Wisconsin. There is thus repeated evidence of accommodation or maintenance leading to idiolectal change and eventually, dialectal development, in these historical studies. This underpins the importance of social factors in shaping the linguistic ecology of migration settings and, consequently, of new dialect formation scenarios. This highlights the need for further research into this area to give value to, guide and inform the existing frameworks of NDF.
2.3. The Scots Language in the Nineteenth Century
To understand potential patterns of Scots use in correspondence by nineteenth-century migrants, understanding the status of their language during this historical period is needed. Scots is a West Germanic language that shares the same parent language as English (Old English) and as a result, the two have retained a large common core of lexical items and orthographic practices (Aitken, 1984). Scots diverged from Old English around the twelfth century and underwent various linguistic changes to become a structurally and semantically distinct national language variety (Kopaczyk, 2012, p. 235), which was widely spoken and written in Lowland Scotland up until the beginning of the seventeenth century. However, by the time the Scottish were immigrating to New Zealand, Scots as a language variety was much more limited in range and application. This was the result of a prolonged period of convergence with English, spearheaded by the arising prestige status of the emerging southern English standard, and the closer links between Scotland and England as they became unified socially, religiously and politically. This saw a large-scale shift to a majority of English forms and features, which gradually displaced written Scots from various higher state functions. By 1700, Scots had seen a dramatic decrease in status and context-suitability across many textual genres (Aitken, 1979; Bugaj, 2004)—a process that Miller (2012) has described as the ‘dialectalisation of Scots’. During the eighteenth century, Scots became seen as uneducated, basic and domestic—a variety to be avoided by those with social aspirations. Written Scots was instead increasingly relegated to ‘the domestic, the familiar, the sentimental, the comic—in fact to the ambit of folk-life’ (Murison, 1964, p. 37)—in other words, Scots became primarily a spoken variety that was acceptable in the home, but not for most formal contexts. The associations and feeling of inferiority surrounding the language variety have remained, and they likely played a role in determining the language choices of the socially-mobile and aspiring Scottish migrants. Moreover, this history of anglicisation will have curtailed the density of Scots we can observe in the first place. The retention of Scots in correspondence into the eighteenth century has been identified in previous accounts, even among highly educated writers (c.f. Cruickshank, 2012; Elder, 2022; Gotthard, 2024; van Eyndhoven, 2023), though all of these have indicated that mere vestiges remain, and this is mostly on the level of lexis with some morphosyntax that was likely below the level of consciousness.
However, our writers were by and large from the lower social orders, and thus would have had less exposure to the southern standard than the upper classes. Scottish children from the lower social orders had a more limited access to education, particularly beyond the basic literacy taught in the parish schools, which tended to be mixed and incorporated Scots and English variants in their practices. First-generation migrants may, as a result, have retained certain heritage features. Indeed, Hickey (2019) found a high degree of vernacularity in correspondence when investigating historical Irish immigration to Australia, though this situation is not immediately comparable to that of the Scottish in New Zealand. However, given that proportions are likely to be slight, this will limit what could be transmitted in writing to the next generation. This does not mean, however, that the variety spoken in the home was not strongly Scottish-inflected or influenced, and possibly even Broad Scots itself. The spoken situation in Scotland by this time was likely one of diglossia, where speakers spoke either Scots or southern Standard English, depending on the situation, interlocutor, the level of formality required and their own role or positioning within those contexts (Johnston, 1997, pp. 438–440). Thus, within the context of the home, Scottish parents may have spoken their native Scots variety, and the extent to which the spoken variety was transmitted to their children’s writing will have depended on their level of education. Certainly in New Zealand, the written standard taught in schools was almost entirely southern Standard English or approaching this, which will have constrained the degree to which we can observe the use of heritage language features in correspondence.
An obvious issue is also the high degree of similarity between English and Scots. Scots and English, while historically different languages, were already functioning in a single sociolinguistic continuum2; thus, to speak of a bilingual dichotomy between ‘heritage’ and ‘migrant’ language features can be misleading. The mixing of varieties that arguably exist along a dialect continuum among diaspora is by no means unique (Hendriks, 2018), but this does mean the use of English over Scots may not have been experienced as a ‘shift’ for the Scottish migrants, given that many of the written forms they used were by 1850 as commonplace in Scots as they were in southern Standard English. Adopting more features from a variety that was already in a close relationship with the heritage language may not have been seen as problematic. Moreover, where a language has no standard written form it is more prone to instability, and in many cases migrant groups will not use languages recognised as lacking stable, codified forms (Wright, 2020, p. 482). Scots, having failed to reach the final stage of standardisation (Bugaj, 2004), falls into this category. Its lack of codification enabled highly variable spelling and written practices to persist throughout the Modern period, but also assisted the transition to majority English options in writing (Aitken, 1979, p. 89). These factors will all have had important ramifications for how the variety was viewed, experienced and used by our writers, as the results and metalinguistic commentary will make clear in Section 4 and Section 5. Yet, despite this historical narrative, we also have evidence of a New Zealand English regional dialect existing in the provinces of Otago and Southland—areas that were historically settled by the Scottish in much higher proportions than elsewhere. This variety contains features at the lexical, morphosyntactic and phonological levels that could plausibly be attributed to Scots (Bauer, 1996; Villarreal et al., 2021). This may be an example of what Chambers (2008, pp. 105–106) calls ‘Inverse Assimilation’, in which the use of heritage features can become stable across multiple generations and become markers of region rather than ethnicity3. While the MacDonald family did not settle in these regions, the other migrant, Rachel Hepburn, did. Comparing their use of Scots will be all the more salient, therefore, given their different levels of connection to the wider Scottish community. The settlement of New Zealand by the Scottish, more generally, and some of the concerns facing this group are discussed further in Section 2.4.
2.4. The Scottish in NZ
The Scottish who came to NZ were not unusual among their kinsfolk: between 1815–1930 some 3.25 million Scottish people emigrated overseas, and they made up around 20% of the immigrant population to New Zealand (Lenihan, 2015, p. 18). Already prior to this, the Scottish had experienced centuries of internal mobility within Britain (Harper, 2003, p. 71), and this trend continued once they reached New Zealand, with many settlers moving around consistently in search of better opportunities. The exception to this trend was the early settlement of the Otago province, where our writer Rachel Hepburn migrated to. In total, 80% of the population of Otago, and neighbouring Southland, was Scottish up to 1860 (Brooking, 1985, p. 159). This was a consequence of the Otago Settlement Scheme, in which the township of Dunedin, at the heart of Otago, was intended to be a specifically (Presbyterian) Scottish settlement (Lenihan, 2015, p. 24). The North Island, however, where the MacDonald family settled, saw a much lower concentration of Scottish migrants in general. Most of those immigrating were motivated by pull rather than push factors: New Zealand offered the chance of land ownership and economic independence, whereas the agricultural and factory workers involved in the light industrial trades of the Scottish Lowlands had little chance at upward mobility and advancement (Harper, 2003; Lenihan, 2015). It is perhaps unsurprising that the majority of emigrants were young, male and single, working and lower middle class, and from the industrialised Lowlands (Hearn, 2003, pp. 79–81). Despite their backgrounds in manual labour, many Scots did realise their social aspirations, becoming involved in both local and central government so successfully that they were over-represented in fields like politics, banking and manufacturing (McCarthy, 2012, p. 524). Their migration thus led to increased social as well as geographical mobility, which would have led to different social networks and degree of contact with fellow Scottish diaspora, and thus plausibly different linguistic outcomes in terms of language maintenance and transmission to post-migration generations.
Often, the desire to shift towards using the host language wholesale is motivated primarily by pragmatic reasons, while maintaining the heritage language is strongly tied to emotional factors (Mohamed, 2025, p. 164; Rosiak et al., 2025, p. 182; Wright, 2020). The Scottish in New Zealand are curious in this regard, in that their cultural and linguistic motivations are difficult to condense into one uniform approach. On the one hand, they were committed to modernity and capitalism, focused on furthering themselves and utilising the new social opportunities available to them (Brooking, 1985, p. 50). Such goals might seem incongruous with maintaining their own, localised identity, given that increased ‘verticalisation’, i.e., changes in social opportunities and community structure, have been strongly associated with language and identity shift (Auer, 2021, p. 288; Chambers, 2008, p. 102). Yet, the persistence of ‘Scottishness’ is visible in the associational culture that arose, with Scottish and Caledonian societies being some of the first ethnic organisations within New Zealand (Bueltmann, 2011, p. 16), remaining well-attended up until the present day. There is plenty of evidence that the Scottish deliberately sought each other out over and above other ethnic groups, and ethnicity appears to have acted as a substitute for the absence of kin and community within the pioneering environment (Brooking, 2006, p. 184). Previous historical accounts would suggest that the endurance of a Scottish identity was connected largely to visual symbols and organisational efforts, rather than language use. This might go some way to explaining why relatively few Scots features made it into the New Zealand English dialect (Trudgill et al., 2000). A similar case applies to Irish features entering Australian English: Hickey’s (2019) analysis suggests that Irish migrants were quick to drop heritage features given their stigmatised status, bar a few lingering remnants surviving into the twentieth century.
Yet, the strength of Scottish ‘transnationalism’ has also been noted in various accounts (i.e., McCarthy, 2006, 2012), in which the Scottish formed informal networks that enabled members to reference a common identity, history and connection, stabilising their collective identity over time. Similar developments of ethnic solidarity and new national identities over and above religious, regional and denominational traditions has been identified in other historic patterns of migration and transnationalism (see, for example, Auer & Derungs, 2018, p. 6; Karpava, 2025; Litty et al., 2015; Madsen & van Naerssen, 2003, p. 68; Stievermann, 2013, p. 10). Schneider (2007, p. 128) notes that the settlers in New Zealand probably still associated themselves with being British post-migration, but with an overseas experience, and their social identity was strongly connected to a ‘common territory of origin and a feeling of belonging there’ (Schneider, 2003, p. 245). Migration, after all, is not a fixed, one-way journey but involves ongoing connections between place of origin and destination (Hannafin, 2025, pp. 122–123). Similarly, ethnicization was never a linear process that played out solely in the colonies, but one that remained dependent on developments in Europe (Stievermann, 2013, p. 16). Thus, attitudes within Scotland around Scots and English, as well as Scottish national identity, will have had important consequences in the pioneering environment, as will become clear in the analysis. Connectiveness and belonging are mediated by the wider social and political context in which migrants operate (Hannafin, 2025, p. 123; Mohamed, 2025, p. 156), and so too, the Scots’ sense of belonging and home will similarly have operated through the dualism of social ambition versus upholding Scottish tradition.
The extent to which we might observe this sense of belonging in the language use of the Scottish remains unknown, and it is plausible that these connections remain vague as a consequence of the heavily-reduced nature of written Scots by this time. However, by stepping away from a purely frequency-based, quantitative perspective and instead considering linguistic behaviour through a qualitative, micro-social lens, the singular use of individual features, as well as the adoption of new features, can be stylistically loaded despite their scant appearance in the texts. It is this approach that is taken here, through considering the wider discursive context, the background of the writer (as far as we can reconstruct this) and the potential of the feature itself to index or reflect attitudes, performances or goals. Utilising this framework, this investigation explores three main ideas.
Firstly, the extent to which heritage features (Scots) are maintained by both first and second generation immigrants from the same family, and whether this indicates cultural transmission via spoken or written language norms within the home. Secondly, whether there is evidence of host language lexis (NZE and Māori) being used in these letters, and if so, whether their form and function differ across the generations. Finally, whether we see inter-writer differences within generations and the plausible role that social mobility and gender play in generating any variation observed. The process of investigating these writers and their private writings is detailed in the following methodological section, before the results of these investigations are presented in Section 4.
3. Materials and Methods
In order to investigate transgenerational change in nineteenth-century Scottish immigrants to New Zealand, a digitally-searchable collection of their correspondence is required. Correspondence is a valuable historical source to explore changes in migrant language use, as letters offer a wider social range than other text types and may display variability on multiple linguistic levels (Bagwell et al., 2019, p. 29), thanks to their varying addressees and reduced level of self-monitoring (Montgomery, 1995, pp. 27–33). They are also considered to be as close to speech as possible (Biber, 1995; van der Wal & Rutten, 2013), resulting from their frequently unplanned, informal and more ‘immediate’ nature (Hendriks, 2018, p. 127). This enables the highest chance of vernacular language practices, unlike print which tends to be formal, formulaic and edited. Of course, letters can be subject to the influence of epistolary practices, and the need to comply with politeness standards which dictated greater use of the standard language (Daybell, 2014; Dossena, 2019, p. 68; Elsweiler, 2021). The use of such conventions was usually linked to the writer’s level of schooling, and the amount of attention they paid to the composition of their letter, as well as the status of the addressee (Elsweiler, 2021, 2023; Hickey, 2019, pp. 1, 9). Sometimes however, writers (particularly migrants) could perpetuate the use of vernacular features via letter-writing formulae and phrases, long after these features had fallen out of everyday use, as these became fossilised in stock phrases and expressions (Pietsch, 2015, p. 224). Stylistic devices such as face-saving strategies, including closing psychological distance or evoking common ground through shared memories or collective experiences, can also be instigated through the use of the heritage language (Dossena, 2019, p. 75). Given the considerable distances that immigrant correspondence frequently had to travel, such strategies were paramount for maintaining firm and cordial bonds between family members. Accordingly, it offers a prime resource for exploring a historical case of transgenerational maintenance and change in a heritage community.
Currently, an online or publicly-available corpus of historical correspondence from any of the major immigrant groups to New Zealand does not exist, and linguistic research into Scottish family collections or individuals is also missing from the record. Accordingly, these materials had to be identified in New Zealand’s archival repositories, requested and photographed. They were then digitised with the help of specialised transcription software, before being uploaded to a corpus-compilation platform to store all the transcripts. This collection of transcripts is being compiled into the brand-new corpus SCOTIA—the Scottish Corpus of Original Texts by Immigrants to Aotearoa—which seeks to be the first fully-digitised collection of correspondence and diaries by Scottish settlers immigrating to New Zealand during the mid-late nineteenth century. This is an ongoing project and will not be explained in detail here given that this investigation solely considers the letters of a single family and individual. Instead, the following subsections will briefly elucidate the materials used and how these were examined to produce the results presented here4.
3.1. The Documents
The materials presented here form part of a larger investigation into changing language use among Scottish immigrants to New Zealand and were identified through close reading of previous historical and genealogical research into this immigrant group. From these sources, archival references were compiled and located in a number of repositories across New Zealand, including the Alexander Turnbull Library (ATL), housed within the National Library of New Zealand, Wellington, and the Hocken Library in Dunedin. For the purposes of this analysis, the MacDonald family was chosen, whose letters were sourced from the ATL, as well as the individual migrant Rachel Hepburn, sourced from the Hocken Library. Details about the individuals are provided in the Table 1 below:
Table 1.
Details on the collections and individuals included in the analysis.
George MacDonald was the father of John, Alexander (‘Alec’) and Albert MacDonald, and Flora Davy (née MacDonald) – Flora was married at the time of her letters in this collection. George writes to his wife Sarah MacDonald, while the three youngest children direct all their correspondence to the oldest brother John, who writes to his parents, siblings, and other extended family in New Zealand. A single letter from the Scottish migrant Rachel Hepburn was also included, to incorporate a further female writer into the sample of letters. As she writes to her sister Sarah, who was born and raised in New Zealand, Rachel offers a point of comparison to the letters written between the MacDonald siblings. She also provides a nice parallel perspective to Flora, as a fellow female migrant writing to a second-generation sibling. In addition, we can consider whether writing between two female correspondents might encourage different language practices than correspondence written between a sister and brother (i.e., Flora and John). All the materials were identified, requested and photographed (with permission) in the reading rooms of the ATL and Hocken Library. Once downloaded to a workspace, each document was tagged according to its archival shelfmark, ready to be digitised using the browser-based, online transcription application Transkribus (Kahle et al., 2017).
3.2. Transkribus
Transcribing even a relatively small volume of letters by hand involves many hours of work. However, the specialised transcription application Transkribus, developed at the University of Innsbruck (Kahle et al., 2017), was created specifically to enable the semi-automatic digitisation of historical handwritten (or printed) documents. Using PyLaia (Mocholí Calvo, 2018), a deep neural network learning toolkit for Handwritten Text Recognition (HTR), the software is able to ‘learn’ the script of an author by identifying individual characters in the manuscript, which it then provides as digital text. This approach can in theory be applied to any language, alphabet, time-frame or script. As a result, users are able to digitally transcribe photographs or scans of handwritten texts, achieved either by creating a custom-built HTR model or using a pre-existing model that most closely matches the language and time-frame in question. To build a model from scratch, around 15,000 words of handwritten text, which equates to roughly 100 pages of manually transcribed correspondence, is needed as input data. Transcribed pages are then split into a ‘training’ set and a ‘test’ set at a ratio of 4:1, which are fed into a model-building algorithm. The process essentially involves AI in its widest sense: the model matches its predictions against the true values in the test set, which are subsequently used to assess the performance and accuracy of the algorithm. This process undergoes several iterations by random sampling with replacement, and more training data can be added to improve the recognition rate. The model is usable once it achieves a comparatively small Character Error Rate (CER), i.e., the percentage of characters the model fails to recognise correctly. The HTR model can then be applied to previously unseen material. Naturally, the output generated requires manual correction, but the time investment is still considerably smaller than with manual transcription, saving many hours of work whilst producing comparatively accurate transcriptions.
For this analysis, training data were generated by manually transcribing 200 pages of correspondence sampled from various writers using the special transcription panel in Transkribus. Transcribing original manuscript material requires transparent and linguistically rigorous editing standards that ideally recognise the material nature of the documents themselves (Elspaß, 2012, p. 160; Fairman, 2015, p. 60). Accordingly, repetitions, abbreviations, superscript letters and crossed-out words were included when possible. Unknown words were marked as [ ] and comments in the margins or above the lines of running text were transcribed separately. Once this was complete, the data were fed into the algorithm. The HTR model trained on these data achieved a CER of 8.81%, which at first glance appears very poor, but by historical text standards, anything under 10% is usually considered very good. The output was then manually corrected and exported in TEI format. Transkribus also enables the user to tag words or lines in the transcription using TEI markup, which can be extracted during the export. In this case, heritage language features (Scots lexis and morphosyntactic structures), as well as host language lexis (NZE colloquialisms and Māori features), were tagged during the manual correction stage. Given the very low likelihood of Scots, NZE or Māori in these letters, as a result of both the three hundred of anglicisation that written Scots had experienced, and relatively infrequent use of vernacular features (such as NZE) in correspondence in general (c.f. Görlach, 1999, pp. 149–150), working from a pre-defined list of features was unlikely to be fruitful. This would only capture some of the potential variation that existed in these documents, particularly given that little is known about which features of written Scots remained in personal writing by the mid-nineteenth century, as this time period remains underexplored. Additionally, there has been no systematic investigation of the NZE and Māori words used in early migrant correspondence. As a consequence, relevant wordlists do not currently exist for any linguistic analysis.
Instead, features were tagged as “Scots”, “NZE” or “Māori” when they were encountered in the texts during the manual transcription process. These were identified by drawing on pre-existing knowledge and checking any uncertain examples in the Dictionary of the Scots Language (DSL, n.d.) the Oxford English Dictionary (OUP, 2025) and the Oxford New Zealand Dictionary (Orsman, 1997). Features were almost entirely lexical in nature, particularly for NZE and Māori. Some morphosyntactic and grammatical features of Scots were also identified alongside lexical items, such as the use of the Northern Subject Rule (e.g.,‘the birds sings’), ‘for to’ and ‘about + progressive’ constructions (e.g., ‘for to stay at home’, ‘I have been about working every evening’), the dropping of <-ly> from adverbs (e.g., regular for ‘regularly’) and Scots-specific uses of relative clause markers, such as which, that and how. In most cases, only a handful of instances were identified for each type. A well-known feature that was not attested in the correspondence data was the use of double modals (combinations of two modal verbs in a single tensed clause, such as might could or will can). These are a well-known feature of Scots and have been also been reported in use in NZE by recent corpus research (Morin & Coats, 2025). Despite these writers’ use of other Scots features, it appears that double modals were either avoided or not within their repertoire, however.
The finished transcriptions were exported in TEI format from Transkribus and uploaded to LaBB-CAT (Fromont & Hay, 2008); the browser-based corpus platform used to store and search all the transcripts5. LaBB-CAT enables the user to view individual transcripts page by page, and different search filters, such as ‘language choice’, can be applied. This filter, compiled from the TEI language tags created in Transkribus, highlights Scots, NZE or Māori features in each letter when applied. For the purposes of this analysis, the MacDonald family members and Rachel Hepburn were selected individually from the corpus, and instances of heritage or host language features were identified in their letters by applying the language search filter. These features were then analysed qualitatively in relation to the surrounding discourse. A quantitative analysis was not possible given the very low numbers of second-generation writers in the corpus and extremely low frequencies of relevant language features (both Scots and NZE/Māori). Yet, frequency is just one dimension of relevance to the study of social meaning (Moore & Podesva, 2009), and revealing insights can be gleaned from other aspects (Dossena, 2019, p. 69; Milroy & Gordon, 2003, pp. 28–29; Sankoff, 1980, pp. 51–52). A qualitative approach was considered to be more appropriate and provides a more nuanced insight into whether and how immigrants demonstrate sensitivity to the varying social influences they encountered upon relocation. Accordingly, the language use of each MacDonald family member and of Rachel Hepburn is presented consecutively in the following section, preceded by some basic background information on the migrant.
4. Results
4.1. The MacDonald Family
The main collection of letters examined was written by the MacDonald family—a Scottish family who settled on the North Island of New Zealand and took up mixed farming. Relatively little is known about the MacDonalds, whose letters were gifted by one Mr B. N. Lindsay, of Wellington, in 1984. The letters of the father, George MacDonald, demonstrate reasonably high letter-writing skills, suggesting he had acquired a basic level of schooling and literacy prior to his immigration. This would place the MacDonalds in the upper working class, though the family experienced some social mobility during their lifetimes, with all four children able to attend secondary school, and the oldest son accepting a carpentry apprenticeship following his education.
George MacDonald was born in Inverness and immigrated to New Zealand at an unknown date, though it must have been before 1869, as this is when he married Sarah MacKinnon at Wanganui. He farmed sheep and cattle in Halcombe, a small settlement in the Manawatū-Whanganui region of the North Island. This area saw some Scottish settlement, but far less than regions such as Otago and Southland on the South Island. George named his residence ‘Glencoe’ after the scenic and highly romanticised Scottish location, which might suggest an enduring connection to his homeland. His wife, Sarah MacDonald (née MacKinnon), was born at Fort William and came to New Zealand in 1864 on the steamer ‘West Australia’. George and Sarah had four children, including John Leonard, Alexander Hugh (‘Alec’), Albert Leonard and Flora May. In his early letters, George writes to his wife Sarah, who was residing in Wanganui with the children while George was developing Halcombe. These letters are mostly dedicated to encouraging her to relocate to Halcombe, presenting a positive view of the farm and living conditions, as there is some suggestion that she was reluctant to leave Wanganui. His later letters see the family living with him at Glencoe, after Sarah agreed to move south to the new settlement.
The oldest son, John Leonard MacDonald, was born in 1870 at Makirikiri North and became a farmer at Halcombe and Mokau. He married Charlotte Elizabeth MacDonald (née Johnson), who was born in New Zealand in 1876. He frequently signs off his letters as ‘Jack’ and is addressed as such by his siblings in their correspondence directed to him. They all seem to have held him in high esteem, yet their letters also suggest they were personally close to John, indicating that the siblings likely shared strong, cordial bonds. John’s correspondence dates mostly between 1887–1888, when he was seventeen–eighteen years old and had left the family farm. He writes to his parents and brothers during this time, and discusses his carpentry apprenticeship and local news. He also appears to have harboured scholarly interests, as evidenced by the other letters in the collection. Albert references books that John sends him, and remarks frequently upon John’s interest in learning French:
In addition, John learnt to play the fiddle and was a keen musician, further attributes suggesting a degree of social mobility and a desire to scale the social ladder. Unlike the other children, he writes to a slightly wider network of people that includes his parents, siblings, aunt and uncle, and his letters also reflect active involvement in family affairs.
| (1) | “So you understand French now do yow. What is this? I saw it in a book. “Le roi Edouard ordonne a’ Harold d’aller apprendre au duc Guillaume qu’il sera un jour roi d’Angleterre” I don’t understand it all”—Albert MacDonald to John MacDonald, 21 May 1886. |
Alexander Hugh MacDonald (though he is frequently referred to as ‘Alec’ or ‘Alick’ and often signs off as such) was the second-oldest son, born in 1872 at Halcombe. He was perhaps the most mobile of all the children, moving around frequently in search of new opportunities. His letters are addressed from various locations around the Manawatū-Whanganui region, including Stanway, Rata, Mangaonoho, Utiku and Taihape, while his final letter in the collection is written from Nowra—a small town in New South Wales, Australia. By this time, Alec was married, had two children, and owned his own block of land. This suggests a high degree of social and geographic mobility, and indeed the letters provide evidence of Alec taking up a number of different occupations and activities, including working at the local mill, clearing land for farming (known locally as ‘bushwacking’), cutting chaff, hunting and eventually farming. This letter collection is also more extensive than the other family members’, spanning the years 1887–1919. Alec writes solely to his older brother John, who had already left the family farm, giving news of the family and his own activities throughout. Unlike John, Alec appears much less interested in schooling and education, remarking upon one occasion in a letter to John: ”It was a jolly fine book you sent up to me, but I could’nt [sic] get over those three big names which were in it”. His letters are also more colloquial and chatty in style, including occasional town gossip, jokes and slang, as well as a noticeably more casual tone than the letters written by John. His last letter from Nowra sees a reduction in the use of such features, and a more formal style throughout, reflecting his age and changed life circumstances.
Albert Leonard MacDonald was the youngest son, born a year after Alec in 1873, at Halcombe. He remained unmarried and died in 1919. The correspondence examined here was written from Halcombe and Hawera, and spans the years 1887–1890, when Albert was between fourteen and seventeen years old. During these years, Albert was able to attend secondary school, and this forms a frequent topic of discussion in his correspondence with his oldest brother. He too seems to have had a keen interest in academic subjects and creative skills, as his letters discuss history, French and mathematics, and he consistently asks for updates on his brother’s ability with the fiddle.
Flora May MacDonald, as the youngest child and only daughter of the family, was born in 1881. She married Alfred (‘Alf’) Charles Davy in 1903 and moved to Yondover Farm, Taihape, which is situated about 80 kilometres north-east of Halcombe. She is represented by just two letters written to ‘Jack’ (John) from Taihape, in 1905, giving news of the farm and her family. Her letters suggest active engagement with all aspects of the farm, including milking, harvesting and hunting, indicating a level of interaction between the male and female spheres that was a regular feature of working-class farming communities in colonial New Zealand.
These writers will be examined in turn for the heritage and host-language features present in their letters. As the following sections will make clear, their use reveals the complex nature of how features could reflect stylistic practices, generational differences, mobility and the different social networks characterising these migrants.
4.1.1. George MacDonald (b. Unknown)
The father, George MacDonald, reflects very few host features in his correspondence with his wife, Sarah. This is interesting, given that the majority of his letters are devoted to painting a rosy picture of the new farm and the positive lifestyle that Sarah can expect to enjoy upon relocating to Halcombe. Arguably, this might encourage the use of new lexis reflecting the life and social mobility that it promised the family. However, the use of several heritage language features can be observed, though these are very limited. Given the historical developments affecting Scots, as touched upon in Section 2.3, this is somewhat expected, and reflects perhaps the status of written Scots as a variety heavily reduced in scope and use by this point in time. Scots is represented almost entirely by the lexical feature mind, in the sense of: ‘To remind (a person) or put (a person) in mind of, also with about, on, or with clause or infinitive’ (OUP, 2025) which had become chiefly Scottish by the Early Modern period, and it is still commonly used in Scotland today. The examples from George’s correspondence are provided below, with the lexical item italicised (my emphasis). The occurrences are in two different letters sent during the same year (1870) from Halcombe:
| (2) | “Mind dont be sending other things as I can man -age now till the end of the month and I know that you want it all at home.”—George MacDonald to Sarah MacDonald, Halcombe, no date, 1870 |
| (3) | “I hope to get a letter from you on Saturday. Mind post it before 5 O’Clock on Friday if yow can.”—George MacDonald to Sarah MacDonald, Halcombe, no date, 1870 |
| (4) | “Mind dont be sending me any more things as I am too much in your debt alrendy.”—George MacDonald to Sarah MacDonald, Halcombe, no date, 1870 |
In the above extracts, mind occurs in contexts where we see George ‘reminding’ Sarah to send certain items or to post her letters on time, which could be considered a face-threatening act by virtue of its demanding or imperative nature. In order to preserve positive face, soften the request, and prevent a possible breach between the correspondents, George might seek to use a common household lexical item, rather than the standard English equivalent. The latter could come across as stiff, stilted and unnaturally formal, particularly for letters exchanged between husband and wife. Given their shared Scottish heritage right down to the local level (both came from the same region of Scotland) it is likely that the spoken linguistic repertoire of George and Sarah included various Scots features on a number of different linguistic levels. Mind, as a very common, everyday item in Scots, was likely used in their daily conversation. While its use in writing was more marked, the ‘speech-like’ nature of correspondence may have encouraged its use, moderated by the writer’s level of schooling, as well as their ability and desire to perform politeness moves in the text. Unfortunately, we do not know much about George’s educational background, but his letters suggest a reasonable familiarity with southern Standard English spelling practices, bar the occasional error as in his spelling of already as ‘alrendy’.
The majority of the two letters from which these extracts are taken are dedicated to convincing Sarah to relocate to Halcombe, and it is clear from their contents that she is rather reluctant. The addition of several requests in a letter might thus serve only to increase the divide between George and Sarah and exacerbate unwanted tensions that could not be smoothed over face-to-face. The use of various pragmatic strategies to soften requests, in which the writer attempts to redress the threat to the recipient’s face by envisaging reciprocity, has been identified in other Scottish immigrant correspondence (Dossena, 2019, p. 74). We appear to observe a similar desire here, through the use of the shared vernacular in this instance. Accordingly, the use of a familiar item to index their shared past and reduce the level of imposition might seek to close both the metaphorical and physical distance between them. This created common ground they could relate to and also assuaged Sarah’s concerns about the unknown.
Another Scots feature crops up in a later letter, in which George is again trying to highlight the possibilities of the new property and location. In this instance, he uses the morphosyntactic feature always, which in Scots is frequently used in the sense of ‘still’. This feature stems from Older Scots (1375–1700) and was commented on as a feature specific to Scots and the far North of England in 1825 (OUP, 2025), while it had become obsolete in most English dialects by the late Middle English period. George uses always when describing his son Alexander (‘Alick’)’s work ethic:
| (5) | “Alick did a good bit of work on Monday before he went away and if he is always doing the same he would be as good as a man here.”—George Macdonald to Sarah MacDonald, 17 October 1880 |
Here, we can observe the use of a heritage feature in the context of persuasion and George’s attempt to paint a positive picture of Halcombe. Again, this might reflect his indexing of the Scottish roots he shares with his wife to encourage a beneficial association with his news. While this is just one feature, this may fulfil his stylistic goals—a single linguistic element can become enough to index a migrant with a particular group (Ladilova, 2015, p. 180). It is of course possible that this was done subconsciously, particularly given that morphosyntactic features are more likely to fall below the level of consciousness. Yet, while the act may have been unintentional, the motivation behind this usage is suggestive of similar behaviour to the other examples, given the context in which the vernacular occurs.
So far, George, as a first-generation immigrant, appears to have retained a select few heritage features and, for the rest, writes largely in fairly standard English. There are, however, two instances of New Zealand English in his correspondence, written in a letter a few years later when the family was living in Halcombe, while Sarah was visiting family in Wanganui. George is discussing life on the farm, and uses two NZE lexical items: ‘dagged’ and ‘Maori hens’, which are italicised in (6):
Dagged—‘to trim matted or dirty locks of wool from (a sheep) to reduce the risk of infestation or infection’, was originally an English regional term, but is now chiefly Australian and New Zealand English (OUP, 2025). This feature emerged with the groups of British migrants arriving on New Zealand’s shores and was selected as the variant used to describe a key fixture of the new colonial life that many settlers (including George) adopted. ‘Maori hens’ was the name the settlers initially gave to the Weka, a bird native to New Zealand that was similar in size to a chicken. It appears George has not adopted the Māori name for the bird, but rather the colonial NZE term, which gradually fell out of use as the Māori name became commonplace. In this extract, we see the use of the host language variety occurring specifically within the context of the new life George is building, and the new activities the family is engaged in, perhaps reflecting their integration into a different social strata and livelihood. This adoption of new lexical items could be almost performative, indexing the family’s increased social mobility—owning sheep and hunting game on their own land were activities that would have been out of reach for George and Sarah prior to immigration. Both came from the Scottish highlands, which were still largely under a semi-feudal system, in which land ownership was concentrated in the hands of a few powerful lairds, and pastimes such as hunting were inaccessible to the middling and lower classes (Divine, 1990).
| (6) | “I had the sheep in on Saturday + dagged them + killed one. Yesterday we dipped them at Sorrensen’s The boys were very tired after it but went out shooting after. Coming home, they got 2 Maori hens Alick shot the one and Bob caught the other.”—George Macdonald to Sarah MacDonald, Halcombe, 1885 |
Thus, this use of NZE lexis, which was certainly not part of their heritage language variety, suggests an expanding repertoire, and perhaps a desire on George’s behalf to assert his integration into the country and location that he speaks highly of. We can surmise that Sarah would understand these terms as no explanation is given, suggesting the correspondents have both become familiar with some of the new lexis around them. Unlike his previous letters, in which he had yet to persuade Sarah to make the move, now that the family is settled at Halcombe and developing their socioeconomic prospects, George has the ability and impetus to use the host language with his wife, rather than feeling the need to resort to their common ancestry as part of a politeness move or request strategy. The stylistic intentions of his letters have clearly changed, and so too has his linguistic repertoire in the process.
Nonetheless, his adoption of new lexis remains very limited, perhaps as a product of the sample of letters we are examining here, but also George’s status as a first-generation immigrant. He writes solely to his Scottish wife; thus, there was likely greater motivation to use Scots features that harked back to their shared roots or even reflected genuine linguistic practices that they shared between them. The use of Scots was likely to be more successful than NZE in face-saving strategies and politeness requests, given the natural association with their past and the home environment. Moreover, George’s exposure to the host language was limited to a degree, as he did not pass through the New Zealand schooling system and lived relatively isolated on the family farm, unlike some other trades or professions that might encourage more social interaction among migrants of various British origins. John and Albert’s letters also provide evidence of the close-knit network that their father maintained with Scottish family members, specifically their aunt and uncle, who had immigrated to the same region of New Zealand. Given this small but coherent community of writers that George was most frequently in contact with, it is perhaps logical that his adoption of NZE and Māori would be extremely limited. To observe, however, whether this influenced the next generation and their use of heritage features, the four children are now examined in turn.
4.1.2. John MacDonald (b. 1870)
John MacDonald, as the oldest of the siblings, demonstrates very standard language use in his correspondence and a polite, formal register throughout. He does not reflect any use of heritage features, despite having perhaps the closest contact with the extended Scottish family. He also uses very few host-language features, which are restricted to just two instances. Rather, his writing suggests a relatively high degree of schooling, and his letters follow largely southern Standard English models. This became the norm in New Zealand classrooms, certainly by the time he passed through the educational system. Even relatively minor ‘colloquial’ items are apparently marked for John: we can observe this in (7), where he describes his old boss to his aunt and uncle. His use of words deemed shorthand or informal by him is accompanied by scare-quotes, as if to imply that this was not his usual language choice, but rather a reflection of the speech (and by implication, character) of his boss and associates:
| (7) | “I saw nearly every one except my old boss. He was on the “booze” at the “pub”.”—John MacDonald to his aunt and uncle, Halcombe, 27 December 1888 |
There are just two instances of new linguistic features: one of which occurs later in life when he is discussing his children, and the second is found in a long anecdote about a lecture he attended. The examples are provided below (NZE features italicised).
| (8) | “Charlie has been offering him lollies to day & gets cross because he wont take them.”—John MacDonald to George MacDonald, Wellington, 14 September 1902 |
| (9) | “Two Freethought lecturers were sitting at the back and as they got as excitable as a couple of old mopokes, they caused considerable amusement.”—John MacDonald to parents and siblings, 7 June 1887 |
In (8), John is discussing his young son Charlie, who was attempting to give lollies—sweets or candy—to their newborn child. His use of the NZE lexical item lollies in the context of their new family life might reflect the transgenerational changes he has experienced as a second-generation migrant. While George would not have grown up with the term, John likely would have, given that it was in circulation in New Zealand as early as 1862 (Orsman, 1997, p. 453). As an item intrinsically linked to childhood and children, the context of the letter and his own exposure to the term growing up as a child in New Zealand might have encouraged adoption of the word and its use, despite the fact that he is writing to his Scottish father.
In (9), John uses the term mopokes to describe two excitable lecturers. Mopoke is a shorthand alternative spelling for ‘Morepork’, which is the NZE term for a native species of New Zealand owl (Ruru in Māori), known for its distinctive shrieks and cries. This feature references the host country two-fold—both as a lexical feature of the host language and as a local species of fauna that is unique to New Zealand. Although John overall uses very few NZE colloquialisms, in this instance, the local wildlife was perhaps the most accessible metaphor available to him. The passage also discusses an attempt to bring Spiritualism to New Zealand, which John concludes was none too successful, stating ‘So much for Spiritualism’. Using the colloquial term mopoke, which was becoming a colonial epithet at this time6, might thus reinforce his mild disdain for the people at the meeting.
Though we have just two examples, John does reflect similarities and differences with his father in his use of host-language features in writing. Both use new lexical items that were connected to flora and fauna, which is one of the earliest semantic fields to see borrowing among immigrant populations (though usually this relates to the adoption of loanwords from indigenous languages during what Schneider (2003, p. 255) labels ‘exonormative stabilisation’). However, John is able to use a new lexical feature for a particular stylistic and creative purpose, in a way that was probably less accessible to his father. George would more likely recourse to comparisons or metaphors reflecting the place he grew up in, which, in turn, were less salient to his offspring. Children will not necessarily directly reflect the practices of their parents after all, and this highlights the advantage of second-generation migrants in using the host language—they have a much greater ability to acquire its various nuances and culturally specific stylistic features. To identify whether this trend continues among the other children of the family, we will next analyse the writing of Alec, and compare his behaviour to that of his older brother.
4.1.3. Alexander MacDonald (b. 1872)
The second son, Alexander (referred to henceforth as ‘Alec’ in line with his, and his family’s usage), demonstrates the greatest amount of shifting along the plane of variation, incorporating both heritage and host language features to a considerable extent in his writings. In the following extracts for instance, he uses Scots features, as well as NZE and Māori lexis, even exhibiting variation between the latter two. Part of this might be explained by Alec’s lower level of schooling compared to his brothers. While he attended the same school as John and Albert, his letters indicate that he was working in the bush and mill already early on in his teenage years. Moreover, as noted before, Alec seems much less interested in scholarly pursuits, and the focus of his letters is largely upon his work activities, wages, hunting and romantic pursuits. Consequently, his style is more casual, and his letter-writing abilities also appear to be less developed than those of his other siblings throughout his lifetime. For instance, there is frequent evidence of crossed out words or misspellings, and uncertainty as to the spelling of a word, as emphasised in the following examples.
| (10) | “there were a great |
| (11) | “Albert and I were |
| (12) | “All the people about here have got the Dandy, Dainty or Danety fever I don’t know which.”—Alec MacDonald to John MacDonald, Rata, October, 1887 |
Alec’s casual style also comes through in his choice of words; for instance, his closing formula in (13) below is anything but formal, and unlike John, who deliberately marked his use of the word ‘pub’, Alec uses it without any overt recognition of its vernacular nature in (14).
| (13) | “I must now shut up as there is nothing more to say.”—Alec MacDonald to John MacDonald, Halcombe, 11 February 1889 |
| (14) | “You will be surprised to hear that old Kennedy is trying to get the Pub in Halcombe”—Alec MacDonald to John MacDonald, Halcombe, 31 July 1887 |
Alec may be writing much closer to how he spoke, rather than referring to standard written models taught in the classroom or the conventions followed by his parents and older brother. There is very little evidence of the standard health formula and politeness practices or requests that tend to characterise immigrant and historical correspondence (Dossena, 2019, pp. 73–74), which we see evidence of in George and John’s letters. Consequently, Alec has a much higher propensity to code-switch; Scots features can occur alongside NZE and Māori within the same letter and even within the same passage. Some examples are given below, in which I have underlined the Scots features and emphasised the NZE or Māori lexis.
Concentrating first on the Scots features, these appear on various linguistic levels, including the lexical level with sluck, likely a misspelling of ‘slock’—‘To put out or extinguish’ (DSL, n.d.), and stop—‘to remain, lodge, dwell, reside, abide’, which was recognised as a Scotticism by 1881 and is widespread in general Scottish English but restricted to certain dialects of English, having been replaced in approved usage by stay (OUP, 2025). On the level of morphology, we can observe the absence of <-ly> in regular ‘regularly’7, and also Scots discourse markers such as just now—found in various English regional dialects, but typically associated with Scots. Syntactically, Alec uses ‘got’ for BE in getting, and his WH-interrogatives (or lack thereof) reflect key distinctions between Broad Scots and standard English. We can observe this in his use of how for ‘why’ (a key Scots feature): ‘bush fires burnt the tram last week so that is how we are not working to day’, and his use of which for ‘that’: ‘there was a large comet as or shooting star I don’t I know which fell here 3 weeks ago’. In Scottish English, relative clauses introducing an event are preceded by which, never ‘that’. Finally, there is also potentially some Scots phonology represented in spellings such as tacken ‘taken’, which suggests the monophthongal rather than diphthongal pronunciation of the FACE vowel, matching Scots rather than NZE pronunciation choices.
| (15) | “I have wore out one axe allready and all for 15 bob a week and tucker… All the school children were tacken in for 9 pence, so we had a train over 1/4 of a mile long it sluck up 2 engines going up the Wangahu hill.”—Alec MacDonald to John MacDonald, Halcombe, 3 July 1887 |
| (16) | “I am |
| (17) | “I am still bushwhacking and am likely to be for another a month”—Alec MacDonald to John MacDonald, Halcombe, October, 1887 |
| (18) | “I had a mate come to stop in the |
| (19) | “I am now getting quite at home in my new billit sometimes I strikes a hear |
| (20) | “Of course if I died you would be all right, one danger is if business went cronk with me you would have to pay up whatever was left of the instalments due as they fell due.”—Alec MacDonald to John MacDonald, Taihape, 12 October 1900 |
By and large, these features are more likely to fall below the level of consciousness compared to lexical items, especially nouns, so Alec may not necessarily be aware of their status or be overtly performing ‘Scottishness’. Rather, he may be reflecting his spoken variety, a tangible possibility considering his lower interest in schooling and letter-writing norms. Regardless, he has adopted a number of heritage features, despite having perhaps the least connection to his wider Scottish family. While John writes to his aunt and uncle and resides with them for a time, and Albert and Flora ask after family members in their letters, Alec makes almost no reference to the family and his letters are largely focused on his friends, fellow workmates or other local figures. Moreover, Alec was highly mobile and moved around consistently in search of different employment opportunities that were agricultural or trades-based. He therefore had greater contact with a wider range of people than his immediate family, which, from the perspective of both migrant language and dialect formation studies, would predict greater incentive to assimilate to the language practices around him, as a consequence of his enhanced integration.
Undoubtedly, Alec’s background influenced his frequent use of NZE and Māori lexis, which appears largely as workplace-specific jargon, particularly terminology specific to farming, logging and milling. This includes tucker—‘board/provisions’ (OUP, 2025), tram—‘a rake of trucks or bogies running usually on wooden rails [..] and used to carry logs from the bush to the mill’ (Orsman, 1997, p. 855), bushwacking—‘felling or clearing bush with an axe, New Zealand English’ (OUP, 2025) and billit (billet)—a common term in use after the 1870s, transferred from military terminology and found particularly in Australian and New Zealand English (OUP, 2025), used to refer to a position or appointment; a permanent job (Orsman, 1997, p. 50). In addition, we see the interchangeable use of NZE and Māori to refer to the same item in use of the terms shanty and whare, both of which refer to a temporary hut or dwelling. This suggests Alec has both forms available in his repertoire and was able to switch between them quite comfortably, even within a single letter. While neither of these items are used in this way in contemporary New Zealand English, their appearance here suggests that there were a number of competing forms circulating during the nineteenth century, and Alec recognises their interchangeable nature at this moment in time.
His use of cronk, a colloquial term meaning ‘unsound, liable to collapse, also bad, inferior, ill-considered or worthless’ (Orsman, 1997, p. 184), suggests a propensity to use vernacular terms when he is being particularly casual and colloquial. Alec’s informal and chatty style has already been noted in the earlier examples, and we can see this in some of his other word choices, for instance, referring to friends or associates as a ‘mate’. This suggests that, while clearly looking up to his older brother, Alec also shared a close, familiar bond with John that is reflected in his language choices. Use of Māori occurs sporadically in flora and fauna, as in totara—‘Tōtara’ or Podocarpus totara, a species of tree belonging to the Tōtara group, which is native to New Zealand, as well as placenames such as Wangahu—‘Whangaehu’. Alec’s lexical usage of host language features thus patterns similarly to his father and brother—these relate to the workplace, housing, and flora and fauna, suggesting the adoption of new features occurred first and primarily in these semantic fields, which matches general observations of dialect formation in historical migration and colonisation scenarios (c.f. Schneider, 2003, p. 255). His misspellings or uncertainty about how to spell these words, as in the case of whare in example (18), adds further support that he was not following standard letter-writing models but rather representing his spoken variety, or what he heard around him.
There is one interesting case of self-correction in which Alec deliberately corrects his usage to the Scots equivalent rather than the standard English form. The example is provided below, with the Scots italicised (my emphasis):
| (21) | “We had Miss KateCambell from Wellington here at tea last Sunday evening. She is |
Alec self-corrects his letter to change ‘working’ to at service. At service refers to being engaged or employed as a servant (usually a domestic servant), and was chiefly Scottish in later use (OUP, 2025), certainly by this time period here. Although his reasons for deliberately changing the lexical item to the Scots equivalent remain opaque, it is possible that in recollecting the event, he has recalled the terminology used in the household between his parents and this fellow Scottish immigrant. Domestic service was a workplace he was unfamiliar with, and perhaps he lacked immediate access to the vocabulary used to describe this situation. His parents and their guest had clearly discussed her situation, and accordingly, Alec may have chosen to replicate that in his letter to John as the most appropriate way to describe Miss Cam[p]bell’s employment. This shares interesting parallels with Auer’s (2021) research into the Swiss migrant community of New Glarus, Wisconsin, which found evidence of linguistic insecurity among speakers, with some self-correction as to what was the ‘correct’ heritage form. Alec here, too, appears unsure which form to use, and opted for the Scots rather than the English verb, though he was clearly aware that they were equivalent terms.
Overall, Alec presents a much greater ability to shift along a continuum of Scots, Standard English, NZE and Māori features in his correspondence, despite presumably being exposed to the writing practices of his more linguistically conservative brother. Moreover, his mobility and contacts would predict the least use of Scots, yet Alec is actually the most prolific in using heritage-language features out of the MacDonald children. He is thus a prime example of the unpredictable and heterogeneous linguistic outcomes of second-generation migrants, and of the need to consider the influence of schooling as well as stylistic factors in shaping the presence or absence of heritage features.
4.1.4. Albert MacDonald (b. 1873)
Moving on to the youngest son, Albert’s letters are directed solely to John while he was still at school. His grammar and spelling suggest that his schooling was not yet complete, as there is some evidence of misspellings and self-corrections throughout. For instance, ‘buckets’ is written ‘bukets’, and he includes an additional, unnecessary <e> at the end of some words, only to cross them out (indicating awareness that this was incorrect), as in ‘a longe letter’ and ‘we got the bage’. Nonetheless, Albert also demonstrates a keen willingness to excel at school, noting the following:
| (22) | “We are going to get a silver medal if we can keep going to school from the begi |
He is also fascinated by his brother’s learning French, asking John to translate various passages for him, and trying to employ the little he has learned in his own letters, as in the following:
| (23) | “Dear Monsieur Jack […] Votre affectionate frere Albert MacDonald.”—Albert MacDonald to John MacDonald, Halcombe, June 1888 |
He also reflects a level of insecurity and concern about his writing to his oldest brother on several occasions, suggesting he perhaps recognised that there were certain expectations surrounding the construction of a letter, even if he was not always able to adhere to these. Some examples are provided below:
| (24) | “I think I have wrote a pretty long |
| (25) | “I hope you will excuse me if you cant read my writing as I am in a hurry to get done”—Albert MacDonald to John MacDonald, Halcombe, 26 September 1888 |
| (26) | “This is a fearful pen I’ve got, so I hope you excuse my scribbling”—Albert MacDonald to John MacDonald, Halcombe, 18 November 1887 |
These are just a few of several such observations by Albert, indicating that his letter-writing practice remained a concern for him, even when writing to his own brother. This may have had important ramifications for his use of heritage and host features, given that both Scots, as well as NZE, were more likely to be perceived as colloquial features that had no place in writing. Interestingly, there is an example of a Scots grammatical feature in (24) above, in which Albert uses the past preterite wrote rather than the expected ‘written’. This feature was not strictly Scots but rather a pan-dialectal feature shared with various English regional dialects, though it was not the prescribed form in southern Standard English (Gustafsson, 2002, p. 275), and is unlikely to have been taught at school. This suggests that Albert may have maintained some heritage features in his writing, despite writing to a sibling who likewise was born and raised in New Zealand, and who, furthermore, seems to have taken pains to adopt a formal and polite style in his writings that did not include dialectal features. It could be that Albert observed the epistolary practices of letters sent and received by the wider family, including that of his parents, and subsequently picked up upon Scots morphosyntactic features in the process, imitating these as part of ‘correct letter-writing’ (c.f., Daybell, 2014, pp. 61–63). There is some further support for this possibility in a few other extracts, shown below, in which Scots features are italicised:
| (27) | “I think it high time for you to be about thinking of writing to me again”—Albert MacDonald to John MacDonald, Halcombe, July/August 1890 |
| (28) | “Mind and write up quickly telling us if aunt will take the butter.”—Albert MacDonald to John MacDonald, Halcombe, 30 April 1889 |
In (27), Albert uses the about + progressive construction—a morphosyntactic feature of Scots used as the progressive form in making deliberate statements (Grant & Dixon, 1921, p. 114). The phenomenon seems to apply most commonly to the verbs think and doubt, and it appears to be an innovation of the modern period (Beal, 1997, p. 373), having been noted in late Modern Scots by Mitchell (1799, p. 79). It seems plausible that Albert could have encountered such a construction in the family letters, given that correspondence between immigrants and family members often included remarks upon the frequency of letters and the desire to receive further communications (Dossena, 2019). Very often in historical settings, letter-writers relied on the use of stylistic models, but also the letters circulating within the family or community in learning to construct a letter (Daybell, 2014, pp. 61–63; Elsweiler, 2023, p. 186). This enabled certain linguistic features to be perpetuated long after they had fallen out of use, and may have brought our immigrants into contact with the writing practices of the previous generation or family back in Scotland, encouraging transgenerational language practices and the maintenance of certain Scots features, regardless of their salience. The about + progressive construction did not make it into general New Zealand English, and is moreover a typically Scots feature; thus, it is unlikely that Albert picked this up through his schooling. Instead, by utilising a set epistolary phrase in his letter, he may have transplanted a Scots feature in the process.
In (28), he uses the lexical item mind, which we have seen earlier in the letters written by his father George. It is quite likely that Albert heard mind being used in the household growing up, and may have used the lexical item in his own speech. It is equally plausible, however, that he has adopted this written practice from his father when making requests. Requests form a frequent component of family letters, in asking for news or demanding certain actions (Dossena, 2019, p. 73; Elsweiler, 2021), and we have observed several such instances in George’s correspondence to his wife. Those letters, though directed to Sarah, may have also been read or viewed by the children, given that they were residing with Sarah in Wanganui at the time. Consequently, Albert may have generalised this feature as a necessary component of requests. Use of a politeness strategy or formula when addressing his own brother might seem somewhat out of place or unnecessarily formal, but throughout the correspondence, there is evidence that he wishes to impress his brother with his scholarly achievements; thus, following correct letter conventions may have been important to Albert. In addition, if this was a feature frequently used in their household, its use may not have been marked as dialectal, old-fashioned or archaic, but rather reflects the family-oriented community of practice that both brothers participated in, even after John’s departure from the family household.
Albert’s adoption of heritage features adds further evidence that some generational transmission of Scots did take place within the MacDonald family, possibly unconsciously through the medium of shared correspondence and letter-writing norms. His incomplete acquisition of the written standard, as suggested by the errors in his letters, may have enabled these features to occur in his correspondence, unlike John, who seems to have actively avoided the vernacular in his attempt to adhere to the formality he felt writing required. Similarly, Albert uses a much higher frequency and range of New Zealand English and Māori lexis than John. He is much closer to Alec in this regard, reflecting the hybrid space he inhabited, as a second-generation migrant familiar with both the heritage and host languages. Some examples of new lexis are provided below:
| (29) | “We got the bag |
| (30) | “Mr Callingham is the slaby down at Bell’s mill”—Albert MacDonald to John MacDonald, Halcombe, 6 September 1887 |
| (31) | “I was up there on Saturday and saw their whare and the place where they work I thought it was a pretty rum place”—Albert MacDonald to John MacDonald, Halcombe, 18 November 1887 |
| (32) | “I been up to the Wellington Observatory. It is like an old whare … Our ancient house is painted but the verandah pillars are bent with age… The Next day mother and I started for home again, having stayed a week at Wanganui, But I took good care to round the old place, butts, Paddy Dillen’s whare, Cemetry & . &. first.’—Albert MacDonald to John MacDonald, Halcombe, 6 April 1902 |
| (33) | “Have you any more of your own knocking about down there?”—Albert MacDonald to John MacDonald, Halcombe, 6 April 1902 |
The use of host language features occurs primarily when Albert is discussing local places or persons and their employment. Discussion of such aspects are accompanied by structural or workplace terminology, as in the case of dray—a wooden cart pulled by horses or bullocks (OUP, 2025), hoggets—‘a sheep in its second year; a yearling sheep; (also) a sheep from weaning until its first shearing, now chiefly New Zealand’ (OUP, 2025), slaby (slabby)—New Zealand colloquial and now historical for ‘a worker in a sawmill responsible for removing the waste pieces of timber left after sawing’ (OUP, 2025) and verandah—‘a roof-like structure built along the side of a building, especially one built over the pavement outside business premises, particular to Australia and New Zealand’ (OUP, 2025). There is also use of Māori with whare—a term that today refers to a (Māori) house or residence, but in colonial times acquired the meaning of a temporarily built hut in the bush or by the beach, among the Pākeha (European) settlers to New Zealand. Albert’s use of it here to describe the dwellings of fellow townsfolk or migrants is, like his NZE terms, linked to locally bound features and locations. This suggests that his access to host-language features was enabled and encouraged by the context of hyper-local topics linked to the home country, and we saw singular instances of this among his other family members.
It appears that this trend, initiated among first-generation immigrants (like George), persisted into the second generation and increased. As lexical items that the MacDonald brothers would have been highly familiar with—having grown up on a farm and being engaged in both farming and milling practices—such jargon was also perhaps the most logical lexis to use, rather than standard English or Scots terms, which may have seemed out of place and distanced the brothers from the communities that they were part of. Moreover, unlike the more formal contexts in which we see Scots being used in the previous examples, the tone here is quite casual and convivial, in which Albert ‘rounds the old place’ and labels their friends’ workplace as a ‘pretty rum place’. Rum is a colloquial and now somewhat dated term for something that is ‘odd, strange, bad, spurious and suspect’ (OUP, 2025), while (33) has the term ‘knocking about’—a construction found in other dialects but recorded earliest in New Zealand, meaning ‘to lie around somewhere, to be at hand or available’ (Orsman, 1997, p. 416), generally used in casual and informal speech. These examples both describe a familiar, congenial sense, and local colloquialisms may have seemed most appropriate to create this sense.
Albert and Alec thus appear to behave similarly, using new lexical items primarily in more colloquial and convivial contexts, in which they are describing local happenings, people or places, while Scots seems to be associated with the more formal aspects of a letter, such as opening and closing formulae or request strategies. Previous research on request strategies in Scottish letters has suggested that some authors were more likely to use Scots in these sections, perhaps as part of learned letter-writing practices (see, e.g., Dossena, 2019; Elsweiler, 2021, 2023). Given the close relationship Albert and John seem to have with one another, Albert’s linguistic behaviour might reflect the nature of their contact. He followed the letter-writing conventions plausibly learned in the home in a bid to demonstrate respect and adherence to politeness norms, whilst employing local features when discussing aspects intrinsically linked to their life and work in New Zealand. Together, this engendered a certain familiarity between them. To complete the MacDonald picture, however, we must consider the letters of the only daughter of the family, and whether gender may have played a role in language shift and maintenance within the Scottish immigrant community. Accordingly, Flora’s letters are now analysed and compared with those of another female Scottish migrant.
4.1.5. Flora MacDonald (b. 1881)
Flora MacDonald was the sole daughter in the family and is represented by just two letters, which were written when she was already married and living at Yondover farm with her husband Alf[red] Davy. Thus, some caution must be exercised when comparing her linguistic behaviour to that of her brothers, who write largely during their teenage and early adult years, when they were still mostly living at home. The influence of the home environment will therefore have been less pertinent for Flora, and her social networks will also have differed. These would have revolved around her husband and, potentially, the agricultural or trade-based community he was part of. Nonetheless, this enables us to consider the role of different networks and community connections in determining written language norms, and the influence of marriage with its expectations for women’s roles and behaviour—factors that can influence the adoption of new lexis or the retention of heritage features, especially if the marriage partner shares the same background or ethnicity (see, for instance, Mohamed, 2025).
Examining Flora’s two letters, they are notable for their lack of heritage and host language features, though we must bear in mind the very small sample size examined here. Unfortunately, we do not gain much insight into her schooling in these letters, though we know she completed secondary education. Nonetheless, there are some indications that Flora’s education was somewhat rudimentary. There are a number of misspellings and self-corrections, as well as non-standard spelling features that she frequently employs, such as the use of <w> for <u> in ‘you’, and a lack of punctuation and pronouns—though this may have been caused by space and time limitations. Despite this, Flora clearly values receiving correspondence from her eldest brother, noting in her opening formula to John:
| (34) | “Was delighted to receive such a long letter from you. Of cause I know you are very busy now So cant expect yow to write quite so often.” —Flora Davy to John MacDonald, Yondover Farm, 20 January 1905 |
Despite her basic schooling, the importance Flora attaches to letter-writing may be part of the reason why we see almost no instances of NZE or Māori lexis in her letters. As most early NZE was largely colloquial in nature, and often linked to men’s workplaces, it may have seemed impolite, improper or even inappropriate for a woman to use such terms, even with family members. While the pioneering nature of New Zealand’s early European settlement meant that men’s and women’s spheres of work overlapped, there was still some recognition of the divide, and certain cultural practices remained. It was, however, not unusual for women to actively help on the farm or business, depending on status and occupation, and there is evidence of this in Flora’s first letter. She notes the following:
| (35) | “Alf is driving the engine + has to be away at six in the morning to have steam up. So you can imagine it dont give him any time to milk nine cows before he goes away So I am going to milk his share for him”—Flora Davy to John MacDonald, Yondover Farm, 20 January 1905 |
Interestingly, this passage is followed by the one instance of NZE in her letters, in her use of the term batche— ‘to live as a bachelor; to live alone and do one’s own cooking and housekeeping’ (OUP, 2025), which was an Australasian colloquial term that usually applied to a man. The example is provided in (36):
| (36) | “He is getting 3£ per week at the mill Mr Beban will most likely be leaving us in a week or two. As the other men are going to batche, he will go with them.”—Flora Davy to John MacDonald, Yondover Farm, 20 January 1905 |
In the context of discussing men and the workplace, Flora thus deploys a NZE term. This might speak to the subconscious influence of this semantic field in the adoption of new lexis, as we have seen with her brothers. John would likely be familiar with the term, given that the practice was not uncommon in New Zealand during the nineteenth century, on account of a significant gender imbalance skewed towards men (Harper, 2003, p. 274). Flora thus uses local lexis when her letters describe contexts in which such terminology dominates, perhaps reflecting her shift into a specifically New Zealand frame of mind, but also her abilities in understanding the workings of the farming and working community surrounding her. Her competence in a male-dominated workplace are arguably underscored by the use of jargon associated with it, highlighting the new identity she has adopted since leaving the home and taking up a leading role within the farming community—a role she likely played less prominently in a household with three brothers.
There are also some instances of Scots features in her letters, indicating that she too has adopted certain heritage features, possibly through emulation of family correspondence. Some examples are given below, in which Scots features are emphasised:
| (37) | “They are going to call the baby Gordon. Gordon Waugh will sound very scotch will it not. She had a splendid time She got bad at 11 oclock & the baby was born at 1 oclock.”—Flora Davy to John MacDonald, Yondover Farm, 20 January 1905 |
| (38) | “Of caurse it will be all right if she can stay at your place. but she said she would never go back to stay with Mother again for love or money, as she was growled at to much. Gladys used to |
| (39) | “I sent you a parcel over a week ago but you never said any thing about it in the letter I wonder if you got it at all.”—Flora Davy to John MacDonald, Yondover Farm, 19 April 1905 |
In (37), Flora specifically refers to Scotland in noting the name that associates have given their baby, stating ‘Gordon Waugh will sound very scotch will it not’, and this is followed shortly after by a Scots syntactic construction, in the use of got rather than the expected ‘was’. This feature is not specifically Scots, as it is found in a number of other English regional dialects, but it was not the prescribed form in southern Standard English, certainly not for written English, which Flora is presumably aiming at. It is tempting to posit a link between the discussion of the baby’s Scottish name and the use of a Scots feature just one line later, though this must remain conjecture. Nonetheless, as a commonly-used feature of spoken Scots, it is highly likely that Flora picked this up in the home, and it has translated to her written correspondence in discussing family life and fellow Scottish migrants, mirroring the practices of some of her siblings.
In (38), we can observe variation between the southern Standard English ‘stay’, which is used twice in the passage, and the use of Scots stop, which we have also observed in Alec’s correspondence. This verb appears to be in common use in the family, and Flora is clearly able to switch between the two forms, suggesting the options were interchangeable for her and she varied freely in their application. Flora is also making a request of John in this extract, hoping that her associate Gladys can stay with John and his wife, and she accompanies this with various strategies to smooth over the imposition. This includes several positive face moves, stating that Gladys will be ‘delighted’ and that ‘she thinks a lot of both of you [John and his wife]’, as well as emphasising negative face to highlight the consequences of rejection: ‘Gladys used to cry cry in the night over it’. The use of a Scots feature here might function similarly to what we saw earlier in George’s letters, acting as a politeness and face-saving strategy that closed any possible breach, through indexing a shared family background. This would suggest that the transmission of heritage features to the MacDonald children—at least, those who sought to adhere to the conventions of correspondence—occurred largely through the vehicle of letter-writing strategies and phrases, rather than as a vocabulary of individual words.
We have some more evidence for this in the last example, which sees the use of never as a negation marker where standard English would more commonly have the simple negative. The preference of never in Broad Scots has been remarked upon in several previous accounts (i.e., Beal, 1997, p. 372; Miller, 1993, p. 115) as the normal negator of the past tense. It is widespread across dialects of English, but seems to remain largely below the level of consciousness for most language users (Cheshire et al., 1993, p. 67). Given its dialectal status, however, this feature was unlikely to have been taught in the classroom. Yet, it crops up in Flora’s closing formula, which, if anything, we might expect to be more formal than the rest of the letter. Immigrant letters frequently tend to include remarks about previous letters sent by the writer in the closing section (Dossena, 2019), as it was not uncommon for letters to be lost or misplaced in both the local and global postage system, thus comments such as that in (39) were not unusual. Again, Flora might have adopted this feature as part of a standard construction that was used by her father, perpetuating the use of heritage vernacularisms through these set phrases.
Despite these examples and her simple schooling, Flora is in many ways closer to John in terms of writing practices. Part of this can be explained by her attention to detail and the importance she attached to correspondence, but part of this might also stem from her status as a female writer with more limited exposure to the primary industries generating much of New Zealand’s early lexicon. To explore this further, the letter of one final (female) immigrant, who was unrelated to the MacDonalds, is examined below.
4.2. Rachel Hepburn (b. 1845)
Rachel Hepburn’s letter to her sister adds a point of comparison and a final perspective into transgenerational language practices among the Scottish migrant community. Relatively little is known about Rachel, but records state that she was born in Scotland and immigrated to Otago with her entire family when she was just five years old in 1850. She is thus in many ways a second-generation migrant, but would have acquired her earliest language use in Scotland. The Hepburn family were agricultural labourers, though they quickly established themselves as successful run-holders and acquired multiple large-scale properties around the Otago region. Rachel’s social mobility further increased when she married William Downie Stewart, a barrister. Her letters to Sarah were written when she was around eighteen years old and living together with her brother James at ‘Brooklands’—a homestead built for the sons of the family in the township of Goodwood, situated about 50 km north of the Scottish settlement of Dunedin. Unlike Halcombe, this area did have a significant concentration of Scottish immigrants settling in the area, and while it is not possible to tell from her single letter how much contact Rachel had with fellow Scottish migrants, the possibility was certainly higher than for the MacDonald family.
Rachel is in many ways more similar to John than to Flora and the other MacDonald brothers, in that her letters do not reflect the tell-tale signs of a lesser-schooled writer. There are almost no misspellings, self-corrections, ink blots or alterations, and there are just two plausible instances of vernacular language use. Rachel’s social position was higher than that of the MacDonald family, and she was not required to contribute to manual labour on the farm, which meant she had perhaps a greater opportunity to complete her secondary education. This is evident in the density of epistolary practices that she incorporates into her letter. Despite writing to her sister, she employs a lot of routine health formulae in the opening of her letter:
| (40) | “I was very glad to hear that you and yours were well when you wrote hope this will find you well in the same happy state still, am happy to say that we are all well”—Rachel Hepburn to Sarah Hepburn, Goodwood, 5 April 1863 |
Rachel’s care in constructing a letter and observance of letter-writing practices might explain why we see so few vernacular features. Southern Standard English was upheld as the model for polite, letter-writing norms in Britain (Elsweiler, 2022), and consequently, Rachel’s writing style might have sought to match that. Recall that Scots had acquired the status of a ‘homely’ and ‘domestic’ variety suitable for speech in the home, but was no longer appropriate for most formal or polite settings, including writing. While Rachel was only five years old when she immigrated to New Zealand, her parents were likely aware of these narratives and carried those ideologies with them to the other side of the globe. Despite the high numbers of Scottish migrants characterising the early settlement of Otago, immigration still meant separation from the Hepburns’ tight-knit, relatively homogeneous community back home, and greater potential for increased social mobility and thereby contact with a wider social spectrum of settlers. The Hepburn family certainly realised their social aspirations - quickly acquiring several large plots of land and becoming a successful farming dynasty with multiple properties. Consequently, this may have reinforced linguistic stereotypes, adding motivation for the family to adopt southern Standard English models in their writing. Combined with Rachel’s higher social status than the MacDonald family, this may have manifested in her presentation as a very standard letter writer. She shares some similarities with John MacDonald, who seems to have been the most dedicated to his education out of the four MacDonald children and continued his academic pursuits outside of his secondary years. In addition, his letters provide evidence of a formal, polite style. He and Rachel may both have subscribed to the ‘correct’ language models taught through the education system, and the desire to emulate a writing style that reflected their upwardly mobile status, neither of which included Scots options.
Rachel’s letters also provide evidence of a strict separation of male and female spheres, which appears much more absolute than Flora’s experience. While the latter actively helped her husband, Alf, on the farm, Rachel notes instead that ‘James has gone out to see after his sheep.’ The sheep are ‘his’ (i.e., James’ sheep) and at no point does Rachel indicate that she plays any role in helping with the livestock. Instead, she refers to the workload keeping her brother busy, and suggests that this is keeping her from being able to explore the area:
| (41) | “I also hope that he will not be so all the time that I am here for I want to go up the Bush that I have heard so much about but have never seen yet.”—Rachel Hepburn to Sarah Hepburn, Goodwood, 5 April 1863 |
This implies a certain lack of autonomy, and it is likely that Rachel’s social networks did not contain the multiple loose connections that encourages linguistic variation and change—though of course her access to the bush was not necessarily the same as her access to the township. Nonetheless, her socialisation was likely limited to the home and domestic spheres of interaction, as a consequence of the sharp segregation of the workplace and home between the sexes hinted at in her letters. This created less impetus for Rachel to adopt NZE and Māori lexis—a lot of which was linked to farming, milling, gold mining and other types of labouring—but also less ability for her to acquire it in the first place. However, despite this general trend, we can observe in (41) a lexical item that was connected to both Scots and NZE usage, in Rachel’s use of the term the Bush. The origins and geographical spread of this word are complicated—originally ‘bush’ could refer to an area of land with a dense growth of low vegetation and small trees, etc., which then came to be applied to a woodland or area of trees. The use of ‘bush’ with this meaning became chiefly Scottish and New Zealand English in later use, which might suggest that bush was a term brought over by the Scottish, which then entered general NZE. This possibility is strengthened by the wide geographical spread of this term in the singular form, chiefly with the. Use of ‘the bush’ is frequent in Australia, New Zealand, Canada and South, East and West Africa, as well as parts of the Caribbean and the United States—all areas that have seen patterns of Scottish migration and settlement. Conversely, use of the bush is less common in British English; thus, Rachel is unlikely to have been using or adopting southern Standard English norms in her use of the term. Originally, the bush referred to a region that is covered with its natural forest or vegetation that has not been used for agriculture (OUP, 2025), and this appears to be what Rachel is referring to here.
Given that the word is attested as chiefly Scottish in later use, Rachel could well have grown up using bush in the household, as the standard term used by her Scottish parents to refer to a woodland. This may then have been reinforced by the NZE language community, leading to her adoption of the singular term the bush. Thus, while Rachel differs from the MacDonald siblings in her lack of lexis relating to the workplace involving agricultural or manual labour, in the semantic field of flora and fauna we can observe use of a term that did not align with southern British norms. Given its origins, it seems unlikely that this was the wholesale adoption of a new word, but rather reflects a development in the language community as the pre-existing term bush underwent a semantic shift, and Rachel provides potential evidence for that change in progress. Although Rachel was not actively involved in the farming lifestyle of her brother, she was part of the wider colonial society of Otago and concurrently was able to use lexis related to broader concepts such as geography and landscape.
Rachel’s only other instance of dialectal or vernacular usage in her letter is similar in nature to the Bush—in that it has both Scottish and New Zealand provenance or use. Plausibly, this might suggest that Rachel only chose to maintain the Scots variants that were adopted by the new language community in New Zealand, as terms acceptable for everyday use and, therefore, not attracting social associations with being ‘uneducated’ or working class. Rachel uses the word tea to refer to a cooked evening meal. The use of ‘tea’ in this way is found locally in the United Kingdom but is especially northern and Scots, as well as being the usual term in Australia and New Zealand (OUP, 2025). The example is provided below:
| (42) | “Since writing the above James has come in and brought me a letter from Andrew after reading which we had tea”—Rachel Hepburn to Sarah Hepburn, Goodwood, 5 April 1863 |
As the model for written language in New Zealand became (as in Britain) southern Standard English, it is unlikely that this was something Rachel adopted through specific instruction, but rather that she is reflecting the emerging norms of the New Zealand language community, which chose to adopt tea as the standard term for an evening meal. Tellingly, this is a lexical item very strongly connected to the female domain, underscoring the separation of spheres and continuation of strict gender roles in the mid-nineteenth century within certain classes in colonial New Zealand, leading to different language patterns across men and women. In addition, the commonplace and domestic nature of tea meant Rachel’s sister would likewise have recognised it, just as John would likely have understood Flora’s use of batch. In summary, although Rachel shares many similarities with Flora—both women grew up in New Zealand as the children of Scottish parents, and both write to their siblings—we see a stark contrast in their use of vernacular features and the arenas they refer to. Flora’s singular use of NZE relates to the male sphere and masculine activities (quite explicitly in this case), reflecting perhaps her own position within the colonial society and her level of integration with the wider farming and working community. Rachel, on the other hand, uses vernacular lexis when referring to flora/fauna and to a concept inherently linked to a female-dominated arena in the form of domestic activities. Their different social status and level of schooling is also apparent through their written behaviour and different degrees of Scots. We see a higher propensity of Scots in Flora’s letters, alongside the tell-tale markers of a lesser-schooled writer, while Rachel’s letter contains no other or purely Scots features, and by and large is written in very standard English, with various politeness constructions and stock phrases included. Both women, however, are much more standard compared to Albert and Alec, suggesting plausible gender differences and experiences were operating as well.
Overall, the second-generation migrants suggest that a degree of transgenerational language transmission has taken place—whether this be through the written adaptation of spoken norms, imitation of epistolary practices and politeness conventions, or adopting the language attitudes and ideologies of their parents’ generation, as has plausibly happened for Rachel. Yet, this did not pattern uniformly across the writers, and the significance of the different factors identified as playing a role is discussed further in Section 5.
5. Discussion
Examining the language practices of a multigenerational family from a historical sociolinguistic perspective has given us insight into the complexity of outcomes that can characterise even a single migrant family. While all four siblings grew up in the same household and, moreover, took up largely similar professions, their retention of heritage features and their adoption of new lexis were not identical. Differences in education or scholarly interests, occupations, geographical and social mobility, as well as awareness of epistolary conventions and letter-writing practices, all seem to have played a role in conditioning the extent and types of features we see occurring in these letters. This highlights the multiplicity of the migrant experience—even within the same family, individuals might experience different degrees of language contact and mobility, contentment or lifestyle satisfaction. Not everyone has equal access to mobility as a resource (Skeggs, 2004, p. 49), and this is predicated, among other factors, on the strength of social networks and community involvement, as well as the strength of family structures (Chetty et al., 2014, pp. 1557–1558). This differed for the migrants investigated here. George and John, in particular, were tightly connected to the wider Scottish community and maintained strong bonds with the family, whereas Alec and Flora were largely independent and formed broader social networks.
Thus, we saw differences in their use of new lexical features, representing different degrees of integration into the local sphere. All writers demonstrated use of specialised lexis connected to location, wildlife and the workplace, but while flora and fauna was a domain most of the writers referred to (i.e., totara, mopokes, the bush), it was largely Alec and Albert who used terms related to the workplace, whether this was specialised equipment or structures (whare, shanty, dray) or positions of employment (slaby, billit). There was also a clear generational split from the first to the second generation in this regard, particularly when it came to Alec and Albert. Their linguistic repertoire was much more expansive than their father’s, and included various features that they could use flexibly to perform a range of stylistic and social functions in their letters. Such stylistic nuance is naturally more available to second-generation immigrants, who have greater access to the host language and its finer details. When it came to Rachel Hepburn, her largely modern English usage reflected her adoption of the variety emerging as the standard in her surroundings. Yet, their linguistic behaviour suggests that all six writers wanted to belong to the new colonial nation and its dominant English population, and sought to reflect this in their everyday linguistic practices. An important way in which feelings of belonging to a place are shaped and affirmed is through everyday routines, which includes, fundamentally, language use (Hannafin, 2025). Both the MacDonalds and Rachel demonstrated such motives in their adoption of the language norms that matched their social situation. As Scots and English were already part of the same sociolinguistic spectrum for our writers—changing their linguistic practices would not have involved a switch to an entirely different language mode, but rather, a mere shift along the cline towards greater use of English features, even in informal, family-oriented and domestic communication channels. This may have enabled a relatively quick and easy adjustment.
Perhaps unsurprisingly, while we do see transmission of heritage features, they are very low in frequency and range. Even in George’s letters, there is little that is explicitly dialectal in nature, despite his status as a first-generation migrant. If his correspondence was used as a model for letter-construction within the family (c.f., Daybell, 2014, pp. 61–63), then it is perhaps self-fulfilling that the children similarly use little Scots in their writing, as the rate of transmission would have been very low. As touched upon before, the ideologies surrounding Scots may have influenced George’s behaviour, which became more prominent upon emigration. Forms that are sociolinguistically unmarked can become marked as the nature of migration changes (Hendriks et al., 2018, p. 152), thus features that had been below the level of consciousness in Scotland may have become salient once our immigrants moved to New Zealand and were surrounded by dialect speakers from other regional areas. Such was the case for Irish English features in Australia, which quickly became stigmatised as indexical of low-status, and were avoided by the Irish upon migration (Hickey, 2019). Consequently, they were not adopted into the emerging Australian English variety. Considering that the language variety that emerged in New Zealand contains almost no trace of Scots features, despite the considerable proportion of Scottish immigrants, a similar story is plausible8. For second-generation speakers, factors that have proved to be important in attitudes to heritage-language maintenance include, among others, the linguistic behaviour of their parents (Karpava, 2025, p. 100). If negative attitudes towards Scots were thus already carried over to New Zealand with the first generation, this might go some way to explaining the overall trend identified here, as well as outcomes for the development of the New Zealand English dialect. A telling comment from another Scottish immigrant represented in SCOTIA adds further support to this:
| (43) | “I was surprised at the Scotch people here speaking so much English. I have since noticed that they try to imitate the English as near as they can in every thing as regards manners”—Thomas Adams to his brother, Dunedin, 18 June 1857 |
Interestingly, this comment was made by a writer in Dunedin, located in the Otago region of New Zealand, at a time when the Scottish dominated this settlement. It was not the case, therefore, that the speech habits and selection of dialect features were a pure numbers game alone (as previous theories of new dialect formation have argued, c.f. Trudgill, 2004; Trudgill et al., 2000), underscoring the pervasive influence of pre-migration historical developments in determining post-migration outcomes for a minority language. Considering the already-restricted use of Scots prior to immigration to New Zealand, the results presented here suggest that even an informal, family-internal variety can come under pressure from the speech community at large in a migration setting. Levelling in the sense of Trudgill (2004) appears to have affected the entire sociolinguistic repertoire of the Scottish migrants, including both the formal and informal ends of the spectrum. Consequently, language attitudes and the prestige status of varieties are as important in shaping the linguistic ecology of migration, perhaps as part of speakers’ desires to embody a new linguistic and social identity (c.f. Schneider, 2003). Working in tandem with this, the importance of the ‘home’ connection often depends on the level of contentment and integration into new social networks in the host country (e.g., Hickey, 2019, p. 7). Thus, it is perhaps not surprising that we see so little Scots use, as the contents of these letters suggest a relatively high degree of satisfaction among the immigrants, implying that we are investigating individuals with a positive outlook towards their host country. This may consequently have led to reduced Scots use, as they focused on upward mobility and integration with the local population.
Though we only have a very small sample size when it comes to the women, their behaviour suggests there were also gendered differences in second-generation language users. Flora and Rachel reflect a lower proportion of heritage and host-language features in their letters compared to the MacDonald brothers. Both seem to have placed considerable value on correspondence both to and from their siblings, which may have led overall to more attention paid to language choices and an attempt to adhere to the expectations of the genre. This manifested in routine opening and closing constructions and health formulae, as well as following southern Standard English spellings and lexical choices. Letter-writing was historically seen as a desirable trait for women of a certain status (Elsweiler, 2022), in which ‘polite’ language use was upheld as the model for correct written etiquette. Those ideologies may have persisted beyond Europe’s shores, on into nineteenth-century New Zealand, causing Flora and Rachel to produce correspondence that matched the expectations of the genre and their gender. Comparing the behaviour of Flora’s brother, Alec, strengthens this possibility. He seems to have placed less value on his written practice and thus was less sensitive to these sociolinguistic narratives, causing him to adopt a writing style that was closer to his spoken variety. Consequently, he peppers his writings with Scots features across a number of linguistic levels. This has interesting parallels with contemporary research, which has found that (unlike the first generation), second-generation women generally adopt the mainstream language to a greater degree than men (e.g., Clyne, 1991; Gal, 1979; Horvath, 1985). Concurrently, when surrounded by a high proportion of southern Standard English speakers, these women appear to have transmitted both the incipient ideologies of their parents’ generation and the language norms around them to their correspondence.
Their lack of NZE and Māori, on the other hand, can be partly explained by their reduced access to spheres where the greatest proportion of new lexis emerged during the early stages of New Zealand English dialect development. Early NZE lexis was highly concentrated in the agricultural, forestry, milling and gold-mining fields, as some of the main sources of employment during the colonial settlement of the country, while Māori lexis was largely limited to flora and fauna, placenames and the adoption of whare to refer to a roughly built hut. These were all largely arenas where women had less direct involvement. Flora had more interaction with the farming and milling community of practice than Rachel, and consequently, we do observe an instance of NZE in her letters that speaks to male realms of use. Rachel’s letters, on the other hand, suggest she was far removed from this, and unsurprisingly, the only use of dialectal lexis in her letters involves terms connected to the home or general geography. Moreover, these were Scots lexical items shared with northern English varieties, and they became standard in New Zealand English as well. Research on koineization suggests that regional forms found in just one dialect are disfavoured, whereas those found in two or more dialects are sociolinguistically unmarked, and are favoured by speakers for whom social integration is paramount (Goss & Howell, 2006; Howell, 2006). This might explain Rachel’s use of these features, and could clarify why we observe the persistence of certain pan-dialectal forms in our writers, such as the use of never for ‘not’, the preterite form of verbs and the use of stop for ‘stay’. Of course, variation might not necessarily point to dialectal language use, but rather incomplete mastery of the standard or ‘slips of the pen’, and some of the letters do show evidence of the lesser-schooled writing, with errors, crossed-out words, phonological spellings and erroneous grammatical structures. Nonetheless, as specific lexis made up the greatest proportion of Scots features, we cannot attribute this to rudimentary schooling alone.
Moreover, these authors were writing to their fellow family members, in which social aspirations, integration and polite decorum may have played second fiddle to the conviviality, familiarity and common background that they shared. Their Scottish ancestry was something communal, and, whether they were conscious of it or not, they may have replicated the phrases, lexicon, pronunciation and written practices they grew up with when corresponding with one another. Consequently, despite everything working against transmission of heritage features—the ideologies around Scots use, the highly similar status of Scots and English as sister languages, letter-writing norms and constraints, the emerging new dialect situation they grew up with, and the general drive for socioeconomic success—we do see the transmission of certain features into the second generation of the MacDonald family, with the exception of John. By way of contrast, he and Rachel Hepburn provide tentative evidence that this was not always the case, underscoring the complexity of migrant language scenarios, even in written form. Their letters indicate the importance of factors outside of frequency and identity in determining written language norms in early New Zealand society, and this is an area that merits further investigation, alongside more detailed consideration of how social networks may have played a role in language maintenance and shift among second-generation migrants.
6. Conclusions
Even though we have only observed the transgenerational language practices of a single migrant family from a historical and sociolinguistic perspective, this qualitative investigation has been able to uncover potential stylistic, performative and indexical uses of heritage and host-language features, as well as possible reflections of spoken language practice and epistolary norms. This fine-grained level of detail is lost in aggregate corpus data, in which a microsocial investigation becomes impossible. Yet the trends identified here are by no means predictable or readily accessible unless we drill down into individual practice, which can manifest contrary to expectations. Indeed, if anything, we might expect the oldest son to use the most Scots features. John seems to have felt the expectation of familial duties most strongly out of the four siblings, and had the greatest contact with the extended family—both factors that would theoretically encourage Scots use. Yet, it was the most geographically mobile child who reflected the greatest range and degree of Scots features. This analysis has emphasised the role of language ideologies, education, and adherence to polite language norms, or recognition of the standard, in characterising migrants’ linguistic choices—factors that have typically seen less attention than migrant mobility, identity, belonging and connection to the homeland. Such factors warrant our attention in both historical and contemporary investigations, particularly when examining a marginalised and stigmatised linguistic variety.
Building up a picture of transgenerational use through the lenses that language attitudes, ideologies and written conventions offer, also contributes to an understanding of the nature of New Zealand’s dialect development, and the surprising lack of Scots features. Transgenerational practices, including both the maintenance of heritage features and language shift through the adoption of new lexis or variables from other linguistic levels, can thus shed light on the complexities of language contact and its outcomes into new, hybrid or multilayered varieties. Examining this from a historical perspective, meanwhile, highlights the continuity in the migration experience from the past to the present day, in which similar concerns, practices and intentions shaped the language behaviour of a speech community across time and space. Regardless of the diachronic setting and the written nature of their communication, we see once more the ability of second-generation migrants to create new linguistic identities that embrace the present whilst maintaining a symbolic link with their shared past.
Funding
This research was funded by the New Zealand Royal Society Te Apārangi as part of a Rutherford Foundation Postdoctoral Fellowship grant, grant number RFT-UOC2301-PD.
Institutional Review Board Statement
Not applicable.
Informed Consent Statement
Not applicable.
Data Availability Statement
The data used in this research were derived from the following resources available in the public domain: the MacDonald family data presented are available in the Alexander Turnbull Library, located within the National Library of New Zealand, and items are available to request for public viewing at https://tiaki.natlib.govt.nz/ (accessed on 15 October 2025), reference number MS-Group-1462, Series-4096 (see Table 1 for specific shelfmark numbers relating to individual writers). The letters of Rachel Hepburn are available in the Hocken Library, Dunedin, and items available to request for public viewing at https://hakena.otago.ac.nz/ (accessed on 15 October 2025), reference number MS-0985-034/003.
Conflicts of Interest
The author declares no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.
Notes
| 1 | In addition, see Haugen (1966a, 1966b) for discussion around general models of standardisation that are applicable to this context. |
| 2 | What has been termed by Stuart-Smith (2004) as the ‘Scots-SSE bi-polar continuum’. |
| 3 | This can also be considered a form of re-allocation in the sense of Trudgill (1986, p. 110); one of the key new-dialect formation processes put forward in his early work. |
| 4 | Further details about the corpus and its composition will be published in due course—please contact the author for further information. |
| 5 | For details concerning the corpus-building process, please contact the author. |
| 6 | However, the meaning this term took on was different from how John uses it here. ‘Morepork’ was applied to a dawdler, a slowcoach, a stay-at-home or a stupid, tiresome person (Orsman, 1997; OUP, 2025), which suggests the term may have undergone a semantic shift during the colonial period. |
| 7 | Scots often does not add <-ly> to adverbs (Miller, 1993, p. 108). |
| 8 | Of interest is the absence of double modals, as mentioned earlier in Section 3.2. There is a lack of consensus about whether double modals in World English varieties, such as New Zealand, were brought over with the Scottish migrants or arose independently through innovation. While previous accounts have suggested they were not among the Scots features imported and transmitted to New Zealand (Bauer, 2007, p. 22), recent corpus research by Morin and Coats (2025) has documented 226 double modal tokens across 47 types in New Zealand English—providing large-scale evidence of their existence in the variety. Crucially, the corpus data did not suggest any particular geographic patterning—double modals are distributed across New Zealand with no regional clustering in regions with historically high Scottish settlement, like Otago and Southland. Given that none of the MacDonald family members used double modals, despite retaining other Scots features in their correspondence, these data cautiously provide additional evidence that would support the innovation hypothesis, i.e., that double modals emerged later in New Zealand English through independent processes, rather than being transmitted directly from Scottish immigrants. |
References
- Aitken, A. J. (1979). Scottish speech: A historical view with special reference to the Standard English of Scotland. In A. Aitken, & T. McArthur (Eds.), Languages of Scotland (Vol. 4, pp. 85–118). W&R Chambers. [Google Scholar]
- Aitken, A. J. (1984). Scots and English in Scotland. In P. Trudgill (Ed.), Language in the British Isles (pp. 517–532). Cambridge University Press. [Google Scholar]
- Auer, A. (2021). Of Zibele and Bélle: Patterns of language variation in the Swiss Language Island New Glarus (North America). In A. Werth, L. Bülow, S. Pfenninger, & M. Schiegg (Eds.), Intra-individual variation in language (pp. 283–304). De Gruyter. [Google Scholar]
- Auer, A., & Derungs, A. (2018). Preserving Swiss dialect features in the Diaspora: The case of New Glarus. In J. H. Petersen, & K. Ktihl (Eds.), Selected proceedings of the 8th workshop on immigrant languages in the Americas (WILA 8) (pp. 1–8). Cascadilla Press. [Google Scholar]
- Auer, A., & Thorburn, J. (Eds.). (2021). Approaches to migration, language and identity. Peter Lang. [Google Scholar]
- Avila-Ledesma, N. E. (2019). “Believe my word dear father that you can’t pick up money here as quick as the people at home thinks it”: Exploring migration experiences in Irish emigrants’ letters. Corpus Pragmatics, 3(2), 101–121. [Google Scholar] [CrossRef]
- Bagwell, A., Litty, S., & Olson, M. (2019). Wisconsin immigrant letters: German transfer to Wisconsin English. In R. Hickey (Ed.), Keeping in touch: Emigrant letters across the English-speaking world (pp. 27–41). John Benjamins. [Google Scholar]
- Bauer, L. (1996). Attempting to trace Scottish influence on New Zealand English. In E. Schneider (Ed.), Englishes around the world (Vol. 2, pp. 257–272). John Benjamins. [Google Scholar]
- Bauer, L. (2007). Some grammatical features of New Zealand English. New Zealand English Journal, 21, 1–25. [Google Scholar]
- Beal, J. (1997). Syntax and morphology. In C. Jones (Ed.), The Edinburgh history of the Scots language (pp. 335–377). Edinburgh University Press. [Google Scholar]
- Berry, J. W. (1997). Immigration, acculturation and adaptation. Applied Psychology, 46, 50–68. [Google Scholar] [CrossRef]
- Bettoni, C., & Gibbons, J. (1988). Linguistic purism and language shift: A guise-voiced study of the Italian community in Sydney. International Journal of the Sociology of Language, 72, 37–50. [Google Scholar] [CrossRef]
- Biber, D. (1995). Dimensions of register variation: A cross-linguistic comparison. Cambridge University Press. [Google Scholar]
- Bonness, D. (2017). The Northern subject rule in the Irish diaspora: Subject-verb agreement among first-and second-generation emigrants to New Zealand. English World-Wide, 38(2), 125–152. [Google Scholar] [CrossRef]
- Bonness, D. (2019). ‘[S]eas may divide and oceans roll between but Friends is Friends whatever intervene’: Emigrant letters in New Zealand. In R. Hickey (Ed.), Keeping in touch: Emigrant letters across the English-speaking world (pp. 185–209). John Benjamins. [Google Scholar]
- Bourhis, R. Y., Giles, H., & Rosentha, D. (1981). Notes on the construction of a ‘subjective vitality questionnaire’ for ethnolinguistic groups. Journal of Multilingual and Multicultural Development, 2(2), 145–155. [Google Scholar] [CrossRef]
- Bousquette, J., & Brown, J. R. (2018). Heritage languages in North America: Formal linguistic approaches. Journal of Language Contact, 11, 373–378. [Google Scholar] [CrossRef]
- Britain, D. (2012). Innovation diffusion in sociohistorical linguistics. In J. M. Hernandez-Campoy, & J. C. Conde-Silvestre (Eds.), The handbook of historical sociolinguistics (pp. 451–464). Wiley-Blackwell. [Google Scholar]
- Brooking, T. (1985). “Tam McCanny and Kitty Clydeside”—The scots in New Zealand. In R. A. Cage (Ed.), The scots abroad—Labour, capital, enterprise, 1750–1914 (pp. 156–190). Croom Helm. [Google Scholar]
- Brooking, T. (2006). Weaving the tartan into the flax: Networks, identities and scottish migration to nineteenth-century otago, New Zealand. In A. McCarthy (Ed.), A global clan: Scottish migrant networks and identities since the eighteenth century (pp. 183–202). Tauris Academic Studies. [Google Scholar]
- Brubaker, R. (2006). Ethnicity without groups. Harvard University Press. [Google Scholar]
- Bueltmann, T. (2011). Scottish ethnicity and the making of New Zealand society, 1850–1930. Edinburgh University Press. [Google Scholar]
- Bugaj, J. (2004). Middle Scots as an emerging standard and why it did not make it. Scottish Language, 23, 19–34. [Google Scholar]
- Canagarajah, S. (Ed.). (2017). The routledge handbook of migration and language. Routledge. [Google Scholar]
- Chambers, J. K. (2008). Sociolinguistics of immigration. In D. Britain, & J. Cheshire (Eds.), Social dialectology: In honour of Peter Trudgill (pp. 107–124). John Benjamins. [Google Scholar]
- Cheshire, J., Edwards, V. K., & Whittle, P. (1993). Non-standard English and dialect levelling. In J. Milroy, & L. Milroy (Eds.), Real English: The grammar of English dialects in the British Isles (pp. 53–96). Longman. [Google Scholar]
- Chetty, R., Hendren, N., Kline, P., & Saez, E. (2014). Where is the land of opportunity? The geography of intergenerational mobility in the United States. The Quarterly Journal of Economics, 129(4), 1553–1623. [Google Scholar] [CrossRef]
- Clyne, M. G. (1991). Community languages: The Australian experience. Cambridge University Press. [Google Scholar]
- Cognola, F., Baronchelli, I., & Molinari, E. (2019). Inter- vs. intra-speaker variation in mixed heritage syntax: A statistical analysis. Frontiers in Psychology, 10(1528), 1–20. [Google Scholar] [CrossRef]
- Croft, W. (2000). Explaining language change. An evolutionary approach. Longman. [Google Scholar]
- Cruickshank, J. (2012). The emergence of Scottish Standard English and the role of the second earl of fife. Scottish Language, 31, 111–127. [Google Scholar]
- Cubas, M. A., Al-Deen, T. J., & Mansouri, F. (2023). Transcultural capital and emergent identities among migrant youth. Journal of Sociology, 59(3), 754–771. [Google Scholar] [CrossRef]
- Daybell, J. (2014). Early modern letter-books, miscellanies, and the reading and reception of scribally copied letters. In J. Eckhard, & D. S. Smith (Eds.), Manuscript miscellanies in early modern England (pp. 57–72). Ashgate. [Google Scholar]
- Devine, T. M. (Ed.). (1990). Conflict and stability in Scottish society 1700–1850. John Donald. [Google Scholar]
- Dictionaries of the Scots Language Online SCIO. (n.d.). Dictionaries of the scots language/dictionars o the scots leid. Available online: https://dsl.ac.uk/ (accessed on 14 September 2025).
- Dossena, M. (2019). Singular, plural, or collective? Grammatical flexibility and the definition of identity in the correspondence of nineteenth-century Scottish emigrants. In R. Hickey (Ed.), Keeping in Touch: Emigrant letters across the English-speaking world (pp. 67–84). John Benjamins. [Google Scholar]
- Edwards, J. (1985). Language, society and identity. Blackwell. [Google Scholar]
- Eide, K., & Hjelde, A. (2015). Verb second and finiteness morphology in Norwegian heritage language of the American Midwest. In B. R. Page, & M. T. Putnam (Eds.), Moribund Germanic heritage languages in North America (pp. 64–101). Brill. [Google Scholar]
- Elder, C. M. (2022). Anglicisation in the letters of Marie Stewart, Countess of Mar and her family: A sociolinguistic perspective [Master’s thesis, University of Glasgow]. [Google Scholar]
- Elspaß, S. (2012). The use of private letters and diaries in sociolinguistic investigation. In J. M. Hernandez-Campoy, & J. C. Conde-Silvestre (Eds.), The handbook of historical sociolinguistics (pp. 156–169). Wiley-Blackwell. [Google Scholar]
- Elsweiler, C. (2021). Convergence and divergence in two historical varieties of English: Pragmalinguistic strategies in commissive and directive speech acts in Scottish and English letters (1500–1700). Anglistik, 32(1), 115–138. [Google Scholar] [CrossRef]
- Elsweiler, C. (2022). Gender variation in the requestive behaviour of Early Modern Scottish and English letter-writers? A study of private correspondence. Journal of Historical Sociolinguistics, 8(1), 55–88. [Google Scholar] [CrossRef]
- Elsweiler, C. (2023). Requests in two early modern Scottish letter-writers. In M. Schiegg, & J. Huber (Eds.), Intra-writer variation in historical sociolingusitics (pp. 181–200). Peter Lang. [Google Scholar]
- Fairman, T. (2015). Language in print and handwriting. In A. Anita, D. Schreier, & R. J. Watts (Eds.), Letter writing and language change (pp. 53–71). Cambridge University Press. [Google Scholar]
- Fishman, J. (2004). Language maintenance, language shift, and reversing language shift. In T. Bhatia, & W. Ritchie (Eds.), The handbook of bilingualism and multilingualism (2nd ed., pp. 466–494). Blackwell Publishing. [Google Scholar]
- Fromont, R., & Hay, J. (2008). LaBB-CAT (formerly ONZE miner). Available online: http://labbcat.sourceforge.net/ (accessed on 15 January 2025).
- Gal, S. (1979). Peasant men can’t get wives: Language change and sex roles in a bilingual community. Language in Society, 7, 1–17. [Google Scholar] [CrossRef]
- Gharibi, K., & Boers, F. (2017). Influential factors in incomplete acquisition and attrition of young heritage speakers’ vocabulary knowledge. Language Acquisition, 24(1), 52–69. [Google Scholar] [CrossRef]
- Goss, E. L., & Howell, R. B. (2006). Social and structural factors in the development of Dutch urban dialects in the early modern period. In T. Cravens (Ed.), Variation and reconstruction (pp. 59–83). John Benjamins. [Google Scholar]
- Gotthard, L. (2024). The rise of Scots do—Transfer or innovation? English Language and Linguistics, 1–31. Available online: https://www.cambridge.org/core/journals/english-language-and-linguistics/article/rise-of-scots-do-transfer-or-innovation/F2837F610B6F51D0EAF82DAAA7A88C1D (accessed on 15 January 2025).
- Görlach, M. (1999). English in nineteenth-century England: An introduction. Cambridge University Press. [Google Scholar]
- Grant, W., & Dixon, J. M. (1921). Manual of modern Scots. Cambridge University Press. [Google Scholar]
- Gustafsson, L. O. (2002). Preterite and past participle forms in English 1680–1790. Standardisation processes in public and private writing [Ph.D. Thesis, Uppsala University]. [Google Scholar]
- Hannafin, S. (2025). Language, accent, and the experience of belonging for the second-generation Irish from England. In S. Dunmore, K. Rosiak, & C. Taylor (Eds.), New approaches to language and identity in contexts of migration and diaspora (pp. 122–133). Routledge. [Google Scholar]
- Harper, M. (2003). Adventurers and exiles: The great Scottish exodus. Profile Books. [Google Scholar]
- Haugen, E. (1966a). Dialect, language, nation. American Anthropologist, 68, 922–935. [Google Scholar] [CrossRef]
- Haugen, E. (1966b). Linguistics and language planning. In W. Bright (Ed.), Sociolinguistics (pp. 50–71). Mouton. [Google Scholar]
- Hava, Z. Y. (2018). Assessing the link between language and identity: The construction of identities among Belgian-Turkish Migrants. The Journal of Humanity and Society, 8(1), 67–84. [Google Scholar]
- Hearn, T. (2003). Scots miners on the goldfields 1861–1870. In T. Brooking, & J. Coleman (Eds.), The heather and the fern: Scottish Migration and New Zealand settlement (pp. 67–86). University of Otago Press. [Google Scholar]
- Hendriks, J. (2018). The effects of complex migration trajectories on individual linguistic repertoires in the early modern dutch urban context. Neuphilologische Mitteilungen, 119(1), 121–144. [Google Scholar]
- Hendriks, J., Ehresmann, T., Howell, R. B. H., & Olson, M. M. (2018). Migration and linguistic change in early modern holland. Neuphilologische Mitteilungen, 119(1), 145–172. [Google Scholar]
- Hickey, R. (Ed.). (2019). Keeping in touch: Emigrant letters across the English-speaking world. John Benjamins. [Google Scholar]
- Horvath, B. M. (1985). Variation in Australian English. The sociolects of Sydney. Cambridge Studies in Linguistics London, 45, 1–200. [Google Scholar]
- Howell, R. B. (2006). Immigration and koineisation: The formation of early modern Dutch urban vernaculars. Transactions of the Philological Society, 104(2), 207–227. [Google Scholar] [CrossRef]
- Hundt, M. (2012). Towards a corpus of early written New Zealand English—News from Erewhon? Te Reo, 55, 51–74. [Google Scholar]
- Hundt, M. (2015). Heterogeneity vs. homogeneity. In A. Anita, D. Schreier, & R. J. Watts (Eds.), Letter writing and language change (pp. 72–100). Cambridge University Press. [Google Scholar]
- Hundt, M., & Szmrecsanyi, B. (2012). Animacy in early New Zealand English. English World-Wide, 33(3), 241–263. [Google Scholar] [CrossRef]
- Hundt, M., & Szmrecsanyi, B. (2015). Do-support in early New Zealand and Australian English. In P. Collins (Ed.), Grammatical change in English world-wide (pp. 65–86). John Benjamins. [Google Scholar]
- Johnston, P. (1997). Regional variation. In C. Jones (Ed.), The Edinburgh history of the Scots language (pp. 433–513). Edinburgh University Press. [Google Scholar]
- Kahle, P., Colutto, S., Hackl, G., & Mühlberger, G. (2017, November 9–12). Transkribus—A service platform for transcription, recognition and retrieval of historical documents. 14th IAPR International Conference on Document Analysis and Recognition (ICDAR) (Vol. 4, pp. 19–24), Kyoto, Japan. [Google Scholar]
- Karpava, S. (2025). The hybrid linguistic and cultural identity of second-generation immigrants in Cyprus. In S. Dunmore, K. Rosiak, & C. Taylor (Eds.), New approaches to language and identity in contexts of migration and diaspora (pp. 91–105). Routledge. [Google Scholar]
- Kasstan, J. R., Auer, A., & Salmons, J. (2018). Heritage-language speakers: Theoretical and empirical challenges on sociolinguistic attitudes and prestige. International Journal of Bilingualism, 22(4), 387–394. [Google Scholar] [CrossRef]
- Kircher, R. (2022). Intergenerational language transmission in Quebec: Patterns and predictors in the light of provincial language planning. International Journal of Bilingual Education and Bilingualism, 25(2), 418–435. [Google Scholar] [CrossRef]
- Koczan, Z. (2016). Does identity matter? Migration Studies, 4(1), 116–145. [Google Scholar] [CrossRef]
- Kopaczyk, J. (2012). Communication gaps in seventeenth-century Britain: Explaining legal Scots to English practitioners. In B. Kryk-Kastovsky (Ed.), Intercultural miscommunication past and present, Warsaw studies in English language and literature (pp. 217–243). Peter Lang. [Google Scholar]
- Ladilova, A. (2015). Language and identity of migrants: The role of the heritage language in the process of collective identity construction in a migration situation. Language and Dialogue, 5(1), 176–193. [Google Scholar] [CrossRef]
- Landry, R., & Allard, R. (1992). Ethnolinguistic vitality and the bilingual development of minority and majority group students. In W. Fase, K. Jaspaert, & S. Kroon (Eds.), Maintenance and loss of minority language (pp. 171–196). John Benjamins. [Google Scholar]
- Lauersdorf, M. R. (2018). Mobility and borders, proximity and distance. Neuphilologische Mitteilungen, 119(1), 21–38. [Google Scholar]
- Lenihan, R. A. (2015). From Alba to Aotearoa: Profiling New Zealand’s Scots migrants, 1840–1920. Otago University Press. [Google Scholar]
- Litty, S., Evans, C., & Salmons, J. (2015). Grayzones: The fluidity of Wisconsin German language and identification. In P. Rosenberg, K. Jungbluth, & R. O. Zinkhahn (Eds.), Linguistic construction of ethnic borders (pp. 183–205). Peter Lang. [Google Scholar]
- Liu, W., & Lin, X. (2019). Family language policy in English as a foreign language: A case study from China to Canada. Language Policy, 18, 191–207. [Google Scholar] [CrossRef]
- Llamas, C., & Watt, D. (Eds.). (2010). Language and identities. Edinburgh University Press. [Google Scholar]
- Madsen, K. D., & van Naerssen, T. (2003). Migration, identity, and belonging. Journal of Borderlands Studies, 18(1), 61–75. [Google Scholar] [CrossRef]
- McCarthy, A. (Ed.). (2006). A global clan: Scottish migrant networks and identities since the eighteenth century. Tauris Academic Studies. [Google Scholar]
- McCarthy, A. (2012). The Scottish Diaspora since 1815. In T. M. Divine, & J. Wormald (Eds.), The Oxford handbook of modern Scottish history (pp. 510–532). Oxford University Press. [Google Scholar]
- Millar, R. M. (2012). English historical sociolinguistics. Edinburgh University Press. [Google Scholar]
- Miller, J. (1993). The grammar of Scottish English. In J. Milroy, & L. Milroy (Eds.), Real English: The grammar of English dialects in the British Isles (pp. 99–138). Longman. [Google Scholar]
- Milroy, L., & Gordon, M. (2003). Sociolinguistics: Method and interpretation. Blackwell. [Google Scholar]
- Mitchell, H. (1799). Scotticisms, vulgar anglicisms, and grammatical improprieties corrected, with reasons for the corrections. Falconer & Willison. [Google Scholar]
- Mocholí Calvo, C. (2018). Development and experimentation of a deep learning system for convolutional and recurrent neural networks [Ph.D. thesis, Universitat Politècnica de València]. [Google Scholar]
- Mohamed, N. (2025). Narratives of (un)belonging: Language management and identity negotiations in two immigrant families in New Zealand. In S. Dunmore, K. Rosiak, & C. Taylor (Eds.), New approaches to language and identity in contexts of migration and diaspora (pp. 155–169). Routledge. [Google Scholar]
- Montgomery, M. (1995). The linguistic value of Ulster emigrant letters. Ulster Folklife, 41, 26–41. [Google Scholar]
- Moore, E., & Podesva, R. J. (2009). Style, indexicality and the social meaning of tag questions. Language in Society, 38, 447-–485. [Google Scholar] [CrossRef]
- Morin, C., & Coats, S. (2025). Double modals in Australian and New Zealand English. World Englishes, 44(3), 415–438. [Google Scholar] [CrossRef]
- Murison, D. (1964). The Scots Tongue—The folk-speech. Folklore, 75(1), 37–47. [Google Scholar] [CrossRef]
- Nagy, N., Iannozzi, M., & Heap, D. (2018). Faetar null subjects: A variationist study of a heritage language in contact. International Journal of the Sociology of Language, 249, 31–47. [Google Scholar] [CrossRef]
- Orsman, H. W. (Ed.). (1997). The dictionary of New Zealand English. A dictionary of New Zealandisms on historical principles. Oxford University Press. [Google Scholar]
- Oxford University Press. (2025). Oxford English dictionary. Available online: https://www.oed.com/ (accessed on 14 September 2025).
- Pietsch, L. (2015). Archaism and dialect in Irish emigrant letters. In A. Auer, D. Schreier, & R. J. Watts (Eds.), Letter writing and language change (pp. 223–239). Cambridge University Press. [Google Scholar]
- Preece, S. (Ed.). (2016). Routledge handbook of language and identity. Routledge. [Google Scholar]
- Regan, V., Diskin, C., & Martyn, J. (Eds.). (2016). Language, identity and migration: Voices from transnational speakers and communities. Peter Lang. [Google Scholar]
- Romaine, S. (1989). Bilingualism. Blackwell. [Google Scholar]
- Rosiak, K., Dunmore, S., & Taylor, C. (2025). Epilogue and reflections on migration, language, and identity research. In S. Dunmore, K. Rosiak, & C. Taylor (Eds.), New approaches to language and identity in contexts of migration and diaspora (pp. 182–184). Routledge. [Google Scholar]
- Sankoff, G. (1980). The social life of language. University of Pennsylvania Press. [Google Scholar]
- Sánchez-Muñoz, A. (2010). Different words for different contexts: Intra-speaker variation in Spanish as heritage language. In S. Rivera-Mills, & D. Villa (Eds.), Spanish of the US southwest: A language in transition (pp. 337–352). Vervuert. [Google Scholar]
- Schneider, E. W. (2003). The dynamics of New Englishes: From identity construction to dialect birth. Language, 79(2), 233–281. [Google Scholar] [CrossRef]
- Schneider, E. W. (2007). Postcolonial English: Varieties of English around the world. Cambridge University Press. [Google Scholar]
- Schwartz, S. J., Luyckx, K., & Vignoles, V. L. (Eds.). (2011). Handbook of identity theory and research. Springer. [Google Scholar]
- Skeggs, B. (2004). Class, self, culture. Routledge. [Google Scholar]
- Stievermann, J. (2013). Introduction. In J. Stievermann, & O. Scheiding (Eds.), A peculiar mixture: German-language cultures and identities in eighteenth-century north America (pp. 9–34). Pennsylvania State University Press. [Google Scholar]
- Stuart-Smith, J. (2004). Scottish English: Phonology. In B. Kortmann, & E. Schneider (Eds.), A handbook of varieties of English (Vol. 1, pp. 47–67). Mouton de Gruyter. [Google Scholar]
- Trudgill, P. (1986). Dialects in contact. Blackwell. [Google Scholar]
- Trudgill, P. (2004). New-dialect formation: The inevitability of Colonial Englishes. Oxford University Press. [Google Scholar]
- Trudgill, P., Gordon, E., Lewis, G., & Maclagan, M. (2000). Determinism in new-dialect formation and the genesis of New Zealand English. Journal of Linguistics, 36(2), 299–318. [Google Scholar] [CrossRef]
- van der Wal, M. J., & Rutten, G. (Eds.). (2013). Touching the past: Studies in the historical sociolinguistics of ego-documents. John Benjamins. [Google Scholar]
- van Eyndhoven, S. (2023). Eighteenth century Scots in correspondence during the Union Debates: An intra-writer perspective. In M. Schiegg, & J. Huber (Eds.), Intra-writer-variation in historical sociolinguistics (pp. 271–294). Peter Lang. [Google Scholar]
- Villarreal, D., Clark, L., Hay, J., & Watson, K. (2021). Gender separation and the speech community: Rhoticity in early 20th century Southland New Zealand English. Language Variation and Change, 33(2), 245–266. [Google Scholar] [CrossRef]
- Wright, S. (2020). Migration, linguistics and sociolinguistics. In C. Inglis, W. Li, & B. Khadria (Eds.), The Sage handbook of international migration (pp. 142–158). SAGE. [Google Scholar]
- Wright, S., & Viggiano, C. (2020). Language and incorporation. In C. Inglis, W. Li, & B. Khadria (Eds.), The Sage handbook of international migration (pp. 481–495). SAGE. [Google Scholar]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.