1 Introduction
Sociological fractionation, which Agha (Reference Agha2007) describes as part of processes of register formation, includes scenarios where ‘one group resists the scheme of values upheld by another (counter-valorization), or reanalyzes and thus transforms such values in fashioning norms for itself’ (Agha Reference Agha2007: 154). In this article I discuss the concept of sociological fractionation in relation to enregistered (Agha Reference Agha2003) Yorkshire dialect in nineteenth-century dialect writing (see also Cooper Reference Cooper2013, Reference Cooper, Honeybone and Maguire2020). In particular, I consider the ways in which we can observe the formation of social boundaries as expressed through commentary on Yorkshire dialect in nineteenth-century writing. For the purpose of defining dialect writing, I am following Shorrocks’ (Reference Shorrocks, Klemola, Kytö and Rissanen1996) distinction between literary representations of non-standard language into dialect literature and literary dialect, as the difference in intended audience for each type of text can allow us to draw conclusions about sociological fractionation. Dialect literature (DL) includes works that are written almost entirely in dialect. These texts are usually written by local writers and are usually aimed at local audiences. This is as opposed to literary dialect (LD), which includes the dialect that appears in sources such as novels and plays, where the vast majority of the text may be written in Standard English, but there are examples of characters speaking in dialect. In both of these text types, local speech is represented using non-standard respellings. DL tends to include texts which are almost entirely respelled, whereas LD will include examples of respellings for specific characters, but the majority of the text will be in Standard English.
Deschrijver (Reference Deschrijver2020) discusses the extent to which specific terms are shared and understood between different groups of speakers in relation to metalanguage. His study considers online discourse surrounding international economic issues and he observes that certain linguistic items are understood to different extents depending on who is engaging with the discourse. Deschrijver notes that the less a particular group of participants is familiar with a specific discourse, the more metalinguistic commentary is observable for individual linguistic items. He goes on to suggest that this can be taken as evidence of sociological fractionation, since the associated commentary on particular language forms highlights that certain linguistic ‘norms are unevenly shared between participants of a community’ (Deschrijver Reference Deschrijver2020: 133). This uneven sharing can lead to different ideologies forming in relation to language forms, and where we see these ideological differences, we can observe what Agha (Reference Agha2007: 157) describes as the ‘creation of social boundaries within society, partitioning off language users into groups distinguished by differential access to particular registers and the social practices they mediate’. This kind of social partitioning is observable in relation to Yorkshire dialect in nineteenth-century dialect writing as we see frequent and consistent use of particular Yorkshire features alongside discourses where we can see certain writers resisting the values upheld by others. It also highlights the kind of differential access to registers Agha describes, particularly in nineteenth-century texts which discuss Yorkshire dialect, where in many cases writers complain about the ways in which other writers represent the dialect and the kinds of stereotypical characters they portray. One reason for the differential access to Yorkshire dialect appears to revolve around whether or not writers were from Yorkshire. Some non-Yorkshire writers attempted to portray Yorkshire identities using enregistered Yorkshire dialect in ways that other, usually local writers felt were inaccurate and inauthentic. Such commentary on the portrayal of stereotypical Yorkshire dialect speakers therefore illustrates the unequal sharing of ideologies about language, and we can observe the formation of two main groups, an ingroup comprised of writers with more localised knowledge of Yorkshire dialect, and an outgroup mostly made up of writers who weren't from Yorkshire, which I discuss further in section 3.2.
My specific aims in this article are to consider what kind of sociological fractionation we can observe in historical dialect writing in order to understand how this contributed to the evolution of Yorkshire identities and ideologies about Yorkshire dialect in the nineteenth century. I also address the issue of how enregistered Yorkshire dialect is employed to create these identities, and how discussion of the use of the dialect in the nineteenth century can be taken as evidence for sociological fractionation. I begin this article by illustrating how Yorkshire dialect was enregistered in the nineteenth century and the processes through which a repertoire of language features became ideologically associated with Yorkshire. I then move on to highlight the evidence we can observe for sociological fractionation in historical commentary on Yorkshire dialect and how this can illustrate the formation of distinct groups of people who evaluate the dialect in different ways based on their having different ideologies about language. Finally, I discuss quantitative differences observable in representations of Yorkshire features in nineteenth-century dialect writing and show how these contributed to sociological fractionation.
2 The historical enregisterment of Yorkshire dialect
Enregisterment, as defined by Agha (Reference Agha2003: 231), is the ‘processes through which a linguistic repertoire becomes differentiable within a language as a socially recognized register of forms’. That is, a set of language features become ideologically distinct from other linguistic varieties. Agha goes on to state that these features have come to index ‘speaker status linked to a specific scheme of cultural values’ (Reference Agha2003: 231). Agha's definition of enregisterment draws on Silverstein's (Reference Silverstein2003) paradigm for indexical order. Silverstein (Reference Silverstein2003: 193) states that indexical order comes in degrees where language features align with context, and that the context itself ‘has a schematization of some particular sort, relative to which we can model the “appropriateness” of its usage in that context’. Johnstone, Andrus & Danielsen (Reference Johnstone, Andrus and Danielson2006: 82–3) develop this paradigm further, identifying first-, second- and third-order indexical links for features of ‘Pittsburghese’. First-order indexical links correlate with non-linguistic factors such as place, social class, gender, etc., but are not usually noticeable by members of the speech community. Second-order indexicals become associated with different speech styles, and variation is usually influenced by ideologies relating to social values such as class, correctness, aesthetics, etc. This allows speakers to style-shift and may also raise awareness of the ‘localness’ of certain language features. Finally, third-order indexical features demonstrate the highest and most explicit levels of speaker awareness. Third-order features become associated with conceptual identities which usually centre on place. This allows both local and non-local speakers to ‘use regional forms drawn from highly codified lists to perform local identity, often in ironic, semiserious ways’ (Johnstone et al. Reference Johnstone, Andrus and Danielson2006: 82–3). Due to the increased levels of speaker awareness and the explicit linking of linguistic forms with contextually driven social values and identities, we can consider both second- and third-order indexical features to be enregistered.
With regards to Yorkshire dialect, we can see that there are social and cultural values that are associated with a particular repertoire of features and that this repertoire was ideologically distinct from other forms of English in the nineteenth century. The features represented in historical dialect writing may be considered as having second- and third-order indexical links to particular social meanings since the texts themselves can be taken as evidence that people were aware of this kind of linguistic variation. We can observe enregisterment in historical contexts by considering evidence for metapragmatic discourse or what Johnstone et al. (Reference Johnstone, Andrus and Danielson2006: 80) call ‘talk about talk’ in sources such as dialect dictionaries and glossaries. In particular, the introductory material to these kinds of texts can be especially useful sources of data, as such material often includes commentary on both dialect speakers and the dialects themselves. This commentary serves to illustrate the social values that are associated with these varieties. The importance of dialect dictionaries in the enregisterment of regional dialects is highlighted by Ruano-García (Reference Ruano-García2020: 188), who explains that dialect dictionaries include social and cultural data about dialects and ‘testify to ideas about what the vocabulary of that variety was like by listing a repertoire of forms in an attempt to describe it’. The listing of forms in this way allows for ideas about dialects to spread and, as Beal (Reference Beal2009: 140) notes, provides wide audiences with a model for ‘the performance of local identity’. Similar metapragmatic commentary and dialect representation may also be observed, albeit to a lesser extent, in dialect almanacs and pamphlets. In general, these kinds of sources form part of the broader metadiscourses about language in the nineteenth century, because, as Agha (Reference Agha2003: 249) states, ideas about dialects are transmitted ‘through print artefacts – books, manuals, magazines, newspapers, etc. – that can be read at different times by different persons’. Moreover, sources like these are interpreted by different audiences in different ways, and this can lead to multiple and sometimes conflicting social values indexed by the same linguistic forms. This can be observed in the ways in which nineteenth-century writers interpreted Yorkshire characters in literary representations, as discussed by Beal & Cooper (Reference Beal, Cooper and Hickey2015: 37), who note that some nineteenth-century writers described Yorkshire dialect as deficient, disastrous and unsophisticated in works of literary dialect, which contributed to a negatively evaluated stereotypical persona. In section 3 I discuss other nineteenth-century Yorkshire writers’ responses to this kind of persona, and illustrate how they argued in favour of a more positive stereotype. Agha goes on to state that sources such as historical novels and literary works can also be taken as evidence of metadiscourse about accent. The kinds of characters represented in these texts, as well as in ‘derived genres such as music, drama’ (Agha Reference Agha2003: 257), serve to expand speaker awareness of these varieties and of these language features through frequent and consistent recycling and reiteration across different text types. Since such characters were often depicted as dialect speakers, they consequently help to spread the association between repertoires of language features and place.
When looking at this kind of historical dialect writing, we can see that the same language features are both represented and discussed time and time again. In considering features which occur frequently and consistently in historical texts, we can identify repertoires of features that are associated with specific social and cultural values, such as the enregistered repertoire of Yorkshire dialect set out in table 1.
Table 1. Enregistered repertoire of nineteenth-century Yorkshire dialect (adapted from Cooper Reference Cooper2013: 266)
This repertoire consists of multiple items which purport to represent lexical, morphophonemic and phonological features of Yorkshire dialect. These features were all represented frequently and consistently in a corpus of Yorkshire dialect consisting of DL and LD (see also Cooper Reference Cooper2013, Reference Cooper, Honeybone and Maguire2020). These same features were also very frequently commented on by historical writers, meaning that people were aware that these features existed, and demonstrate third-order indexical links to place and identity. The repertoire in table 1 identified by Cooper (Reference Cooper2013) emerged from a corpus of 46 texts and 44,605 words. This corpus illustrates that, although the same features were being used to represent Yorkshire dialect to both local and non-local audiences, these features occurred in different amounts. That is, certain features were more salient to local audiences than non-local ones. Cooper (Reference Cooper, Honeybone and Maguire2020) goes on to illustrate that not only were dialect features represented differently according to text type, but there were similar patterns in representations that aligned with areas within Yorkshire itself. This means that certain features were used to distinguish Yorkshire areas from one another, and the emerging regions were the North, East and West Ridings of Yorkshire (NR, ER, WR respectively), as well as the northern parts of the West Riding, which Cooper (Reference Cooper, Honeybone and Maguire2020) labels West Riding (North), or WRN. These areas were identified by considering the available metadiscourse on Yorkshire dialects in the nineteenth century and, as Agha (Reference Agha2003: 249) goes on to note, there ‘is ample evidence that metadiscourses of accent in the 18th and 19th centuries involved identifiable speech change trajectories through which accent values moved in space and time’. In this article I draw on an expansion of the corpus used for the previous studies mentioned above, which is now comprised of 106,463 words from 108 texts, where 63,500 come from DL and 42,963 are from LD. I am following the methodology set out in Cooper (Reference Cooper2013, Reference Cooper, Honeybone and Maguire2020), where 1,000-word samples were taken from each text or, where there weren't 1,000 words of dialect represented, frequencies were normalised per 1,000 words following Biber et al. (Reference Biber, Conrad and Reppen2006). I also discuss explicit commentary on Yorkshire dialect gathered from 62 texts comprising LD, DL, as well as more general commentary texts such as dialect dictionaries, glossaries and travel writing published throughout the nineteenth century, following Agha (Reference Agha2003) and Cooper (Reference Cooper2013).
An additional consequence of the recirculation of ideas about language is the formation of vernacular norms. These norms help us to identify sociological fractionation in historical contexts as particular variants tend to be used more or less depending on which group a writer belongs to, as discussed further below. Johnstone & Baumgardt (Reference Johnstone and Baumgardt2004) define vernacular norm formation as the processes through which non-standard vernacular linguistic items, such as regional dialect features, can become regularised. They illustrate that these processes are ideological and explain that speakers who engage in them ‘draw on and reshape local and supralocal ideas about language and dialect and their social meanings’, since their experiences with their local dialect give them the sense that they are experts, leading them to believe that ‘they have the right to evaluate local speech by displaying their knowledge of it’ (Reference Johnstone and Baumgardt2004: 115). Johnstone & Baumgardt go on to discuss the formation processes for vernacular norms in relation to ‘Pittsburghese’ as it is represented in writing by people posting in online discussion boards. They illustrate how participants engage in explicit metacommentary regarding which features count as legitimate ‘Pittsburghese’ and discuss aesthetic value judgements of the dialect. Johnstone & Baumgardt also talk about the use of yinz to distinguish speakers of ‘Pittsburghese’ from speakers of other varieties of American English, some of whom may use forms like y'all to represent ‘you (plural)’. Their results are therefore illustrative of fractionation since yinz is used to distinguish Pittsburghers from non-Pittsburghers, reinforcing ideological notions of an ingroup versus an outgroup. The nature of this distinction relies on perceptions of authority in relation to the dialect and ‘establishing who can say authoritative things about “Pittsburghese” and on what grounds’ (Johnstone & Baumgardt Reference Johnstone and Baumgardt2004: 141). Usually, the speakers with most authority are those who belong to the ingroup, which is comprised of local Pittsburgh speakers who use these online forums to demonstrate their perception that they have a detailed level of knowledge about the dialect. The same processes can be seen operating in relation to nineteenth-century Yorkshire dialect, and we see analogous commentary from writers seeking to be viewed as authoritative, as discussed in section 3.1.
A similar scenario is observed by Pearce (Reference Pearce2015) in his study of language ideologies surrounding the use of vernacular language features of North East English in online discussion forums. He describes a debate over the use of the local word mam as opposed to the standard mum. The former is very strongly associated with the North East by posters in the forums, whereas the latter is explicitly ideologically linked with the South of England. As in Johnstone & Baumgardt's (Reference Johnstone and Baumgardt2004) examples, Pearce's participants also used non-standard respellings of local language features in order to convey authority on the dialect. He goes on to note that this is one way in which participants distinguish themselves from speakers from outside the North East, since ‘participants use North East words and grammatical constructions together with respellings to demonstrate their local affiliations and identities’ (Pearce Reference Pearce2015: 118). In particular, people were using local dialect to ideologically position themselves in direct opposition to stereotypical perceptions of Southern British English speakers. As a result, we can interpret Pearce's findings as an example of sociological fractionation based on his suggestion that these participants are demonstrating what he terms ‘anti-austrocentrism’ (Pearce Reference Pearce2015: 127), where participants provide frequent negative commentary on notions of Southern English speakers and resist a perceived trend for the South to be favoured over the North. So, in the fashioning of a vernacular norm for themselves, North Eastern participants are seen to be resisting the social values associated with an ideologically othered group. Moreover, these ideological groupings create a sense of an ingroup, which has greater access to local linguistic knowledge displayed in the use of regionally restricted, non-standard written representations of local dialect. This kind of indexical linking of language forms and a conceptual ingroup is discussed by Ilbury (Reference Ilbury2019: 12), who explains that some gay British men employ certain features of African American Vernacular English (AAVE) on Twitter in order to align with an identity associated with a particular gay subculture. The AAVE features are represented using non-standard respellings such as dat ‘that’ and gurl ‘girl’ and function as an ‘ingroup code’ (Ilbury Reference Ilbury2019: 12) which is recognisable by others who associate with that group identity. In some cases, this kind of sociological fractionation can create and reinforce notions of an ideological outgroup as defined by members of the ingroup. The outgroup can be associated with standard linguistic forms, as Southern British English speakers were by Pearce's participants or, as discussed in section 3.2, they may be associated with perceptions of inaccurate dialect representation which is in turn associated with a stereotypical identity perceived to be inauthentic, as in the case of nineteenth-century Yorkshire dialect.
The stereotypical identities associated with the historically enregistered repertoires of Yorkshire dialect listed above can be observed in DL and LD where we can see similarities and differences in local perceptions of dialects and dialect speakers compared with those of wider audiences. Historical discussion of these perceptions ultimately contributed to the kind of fractionation Agha describes. The repertoire highlighted in table 1 combined with Cooper's (Reference Cooper, Honeybone and Maguire2020) identification of sub-Yorkshire repertoires illustrates that there was some consistency between perceptions of Yorkshire dialect in the nineteenth century, but that there was also variation embedded within the broader concept of ‘Yorkshire’. This further aligns with Agha's definition of sociological fractionation as he states that we can observe ‘competing schemes of valorization’ which ‘presuppose underlying commonalities’ (Reference Agha2007: 172). Competing schemes of valorisation for historical Yorkshire dialect can also be seen in writers representing localised areas within Yorkshire using differing repertoires. The underlying commonalities manifest in the inclusion of elements of the generic ‘Yorkshire’ repertoire identified by Cooper (Reference Cooper2013). However, these sub-Yorkshire repertoires also include more localised features which has the effect of simultaneously associating them with more localised areas and with Yorkshire as a whole, as all these repertoires are described as being ‘Yorkshire’ dialect, with some writers arguing that their variety is more legitimately ‘Yorkshire’ than others, as noted by Easther (Reference Easther1883: ix), who states ‘It is a somewhat amusing fact, that in a company of Yorkshiremen each thinks his own dialect the most genuine.’ Moreover, local and non-local writers conflicted with each other when it came to depicting Yorkshire speakers, and it is this conflict that can be taken as historical evidence for sociological fractionation. Therefore, we can see that sociological fractionation simultaneously applies to differences in valorisation and differences in localised repertoires. This is discussed further in section 3.2.
Nineteenth-century writing also provides us with explicit evidence of these ideologies as writers discuss Yorkshire dialect and Yorkshire speakers, the identities associated with them, and the contexts in which these identities appear. These discussions frequently centred on the appearance of Yorkshire dialect speakers in literature, particularly in terms of the specific dialect features used as well as the kinds of characters literary writers chose to associate with Yorkshire. The importance of literary representations of dialect speakers in relation to language ideology is highlighted by Hodson (Reference Hodson2016: 28) in her discussion of servants as depicted in nineteenth-century novels. She argues that the use of dialect-speaking servant characters in novels does ‘not just reflect existing cultural understandings of what it means “to speak like a servant” but also serve to reproduce and perpetuate those cultural understandings’ among audiences. Indeed, we can see evidence of the kinds of cultural understandings nineteenth-century audiences had in relation to Yorkshire dialect by considering historical commentary on dialect and dialect speakers in sources like those listed above. DL and LD texts are initially analysed quantitatively following the methodology of Cooper (Reference Cooper2013, Reference Cooper, Honeybone and Maguire2020) where relative frequencies of dialect features were considered diachronically across the nineteenth century. Additionally, frequencies of frequently and consistently occurring Yorkshire features in DL were compared with their frequencies in LD. Finally, Yorkshire features were analysed in terms of how frequently and consistently they were discussed in commentary material. Features were considered to be enregistered in historical contexts if they appeared in a ratio of between 40:60 and 90:10 DL:LD, where greater than or equal to 50 per cent of texts included tokens and more than 40 per cent of commentary texts featured discussion of tokens (Cooper Reference Cooper2013: 273). They are also analysed qualitatively in order to examine the metapragmatic discourse they present in relation to Yorkshire dialect and identities. Additionally, more general commentary texts such as dialect dictionaries, glossaries, travel writing, etc. will likewise be qualitatively analysed as these sources allow for the identification of competing valorisations evident in the commentary they provide. Competing repertoires are identified using a modification of Cooper's (Reference Cooper, Honeybone and Maguire2020) methodology for identifying sub-Yorkshire repertoires, where relative frequencies of non-standard respellings are considered diachronically and in terms of their occurrences in LD versus DL. This allows for the analysis of competing repertoires over time, as well as the identification of competing variants according to audience, where the ingroup predominantly produces DL for local audiences and the outgroup predominantly produces LD for wider audiences. I will now turn to a discussion of these sources to illustrate how we can use them to observe and understand sociological fractionation in historical contexts.
3 Evidence for sociological fractionation in historical dialect writing
3.1 Dialect commentary
Despite some broad agreement over the features used to represent a general ‘Yorkshire’ dialect in the nineteenth century, some writers contested this and debated the characterisations of Yorkshire speakers and the identities they were associated with. For instance, Robinson (Reference Robinson1862: ii–iii), a Yorkshire writer who establishes himself as part of the ingroup, states that some writers in ‘the book-world’ present Yorkshire dialect as though ‘dialectic distinctions throughout the county are nil, and to the speech from one end of the county to another, the general term “Yorkshire” is given’. Examples like this highlight Yorkshire writers like Robinson resisting the scheme of values upheld by others in the creation of norms for themselves. In this case, Robinson is attempting to fashion a norm wherein the perception of linguistic variation in the region is recognised. Robinson's use of the term ‘the book-world’ indicates that he is defining the outgroup relative to more local Yorkshire speakers. He later gives explicit commentary on people he perceives to be part of this outgroup and states his belief that many Yorkshire dialect dictionaries and glossaries are:
the work either of leisured clergymen, upon their annual visits to particular watering places, or of gentlemen from town, … visiting their friend the rector of some country parish. (Robinson Reference Robinson1862: vii)
These comments suggest a belief that such writers are not qualified to speak authoritatively about Yorkshire dialect due to their lack of local affiliation. Alongside this we see additional commentary resisting authoritative statements by outgroup writers where the dialect is negatively evaluated. For instance, Richard Stead (Reference Stead1878: 36) in his discussion of Holderness dialect in the ER, comments on the wider perceptions presented by the outgroup, and provides a defence of the dialect, which directly challenges the association of negative social values with Yorkshire, including being ‘broad’, ‘uncouth’, ‘vulgar’, ‘improper’ and ‘bad English’, describing it instead as ‘very pure, proper, and good English’ (Stead Reference Stead1878: 36). Stead states that he is ‘a native’ of the Holderness area in his discussion, which further reinforces the ingroup/outgroup dynamic illustrated by his commentary. This dynamic was further reinforced by ingroup and outgroup writers using different features to index Yorkshire identity in writing. Outgroup writers draw on a largely codified enregistered repertoire of general ‘Yorkshire’ features. However, as we can see in Robinson's commentary, this general ‘Yorkshire’ repertoire does not always index authenticity for some local writers due to the perception of linguistic variation within the county. As a result, we not only see writers in the ingroup deviating from Standard English spelling, but in many cases, they also deviate from the generally enregistered repertoire identified by Cooper (Reference Cooper2013). These deviations are observably quantifiable, as discussed in section 3.2.
The kind of identity that appears to be being resisted by the ingroup commentary is exemplified in a story recounted in the popular London-based publication the Cornhill Magazine. This story describes a scenario in which some Yorkshiremen visit London where they encounter an exhibition of models who are posing as mannequins. The writer states that the Yorkshiremen argue over whether the models are alive, until one of them shouts ‘That lass is wick (anglice alive). I seed her wink iv an eye’ (Anon. Reference Ruano-García1864: 95), prompting other audience members to call for them to leave. We see here that the Yorkshiremen are depicted as loud, obtrusive and disruptive. In addition, their speech is represented in non-standard respellings of local dialect. We also see a dialect word that is glossed, wick ‘alive’, which indexes that elements of Yorkshire dialect are unintelligible for certain audiences. Indeed, this is explicitly commented on by numerous writers. For instance, Parsons (Reference Parsons1834: 490) describes Yorkshire dialect as ‘curious and utterly unintelligible to a stranger’, and the description of people from outside Yorkshire as ‘strangers’ is also employed by Morris (Reference Morris1892: 20) when he discusses definite article reduction (DAR), noting a perception that the reduction of the article means that words that ‘otherwise sound familiar become almost unintelligible to strangers’. This serves to reinforce the ideological distinction drawn between Yorkshire speakers and perceptions of outsiders. Moreover, this is a clear example of the competing valorisations involved in sociological fractionation as Parsons and Morris, both Yorkshire speakers, describe Yorkshire dialect as unintelligible to outsiders. Morris goes on to state that Yorkshire speakers evaluate this positively, when he presents an anecdote of a Yorkshireman visiting London, whose ‘chief delight … was to walk into any shop that seemed specially to interest him, and air his broad Yorkshire speech’, to the ‘bewilderment’ of the London shopkeeper (Morris Reference Morris1892: 174). Conversely, an article in the Cornhill Magazine describes the unintelligibility of Yorkshire in a much more negative way. It states ‘To those unused to them, the dialects of all the Ridings would sound equally uncouth and unintelligible’ (Anon. Reference Ruano-García1864: 89). So we see the same indexical link between Yorkshire dialect and unintelligibility evaluated in both a positive and a negative way by different writers. Examples like this can therefore be taken as evidence of sociological fractionation in historical contexts.
Overall, we can see that these examples illustrate an awareness of differing perceptions of Yorkshire dialect speakers in the nineteenth century. This highlights the notion that there was an ideological ingroup as well as an outgroup, which both associated Yorkshire dialect with different social values, and shows that different third-order indexical links were activated for each group. The ingroup perceives Yorkshire dialect as pure, proper, good English, and Yorkshire speakers are proud that their dialect is unintelligible to anyone but them. The outgroup perceives Yorkshire dialect as being uncouth, improper, bad English, and its perceived unintelligibility is seen to be a direct result of this. Quantifiable evidence for these differences is also observable when we consider the ways in which Yorkshire dialect was represented in writing, especially when we compare texts that were written and intended for local versus international audiences, as in the case of DL and LD respectively.
3.2 Dialect literature and literary dialect
In his discussion of literary representations of Yorkshire speakers Hamilton (Reference Hamilton1841: 291) describes a tendency in drama and novels to include a ‘Yorkshire boor’ as a character. He also complains about the dialect representation, stating that their dialogue is written by people ‘without any knowledge of the vernacular speech’ (Hamilton Reference Hamilton1841: 291). Hamilton describes a Yorkshire identity that indexes being boorish, vulgar and unintelligent. These are notions which he resists in his subsequent discussion of this kind of character. He explains that these stereotypical ‘Tykes’ are a caricature of authentic Yorkshire speakers which would ‘befit a clown of any place, and would then require that clown to be a buffoon in order to utter’ the inaccurate dialect (ibid.). Hamilton suggests that some outgroup writers think that an authentic Yorkshire character can simply be named a ‘Tyke’ and that this will index a widely known stereotypical identity of a clownish, buffoonish character. Hamilton is therefore presenting a competing valorisation of the authentic Yorkshire speaker, where he suggests that authenticity is actually indexed by accurate Yorkshire dialect representation rather than clownish behaviour. A similar stereotypical identity is also described by Burnley (Reference Burnley1875: 10), who discusses the concept of the ‘stage Yorkshireman’, which he states is the kind of character who ‘grins and guffaws’ and demonstrates ‘outlandish dialect’. This character has, according to Burnley, ‘no counterpart in real life’, and is seen to index a lack of authenticity, particularly in relation to the use of dialect, which Burnley states is more akin to the stage dialect of Zummerzet, which is likely a reference to the stage accent ‘Mummerset’ (Altendorf & Watt Reference Altendorf, Watt, Kortmann and Upton2008: 198), where voicing of /s/ is common. He goes on to present the ideology that the ‘real, weighty’ dialect of authentic Yorkshiremen is absent in these portrayals, and in so doing illustrates that he is resisting the scheme of values represented by the ‘stage Yorkshireman’ and reiterates the norms associated with a different enregistered Yorkshire identity, as Burnley goes on to mention the pride Yorkshire speakers are perceived to have in their dialect, and then describes how Yorkshire folk are industrious, independent, hard-working and hospitable. Stead (Reference Stead1878) provides a detailed example of this kind of Yorkshire identity in his discussion of Holderness people. He states that they are ‘a remarkably industrious class’ (Stead Reference Stead1878: 88), and that the characteristic Holdernessman has a ‘sturdy independence – his unwillingness to bow down before any unauthorised superior’ (Reference Stead1878: 85). In addition, we also see discussion of the Holdernessman's sense of humour, which is combined with an additional discourse of authenticity, where Stead notes that the ‘Holderness man has a keen sense of humour, and especially loves a practical joke. The humour is dry and quiet, perhaps, but genuine for all that’ (Reference Stead1878: 91). He also describes farmers as being the elite class in Holderness, and describes their dialect as being more ‘proper’ than the ‘vulgar’ forms used by old people and labourers (Reference Stead1878: 47). These social values emerge as norms which align with those that were frequently and consistently associated with Yorkshire identity in nineteenth-century dialect commentary, such as those described by Beal & Cooper (Reference Beal, Cooper and Hickey2015: 42–3), which include independence, authenticity and having a sense of humour.
The ‘stage Yorkshireman’ stereotype appears in plays like Pirsson's The Discarded Daughter (Reference Pirsson1832), where we see the character of Lubin Gubbins, who is stated to be a Yorkshire character. However, in like manner to that described by Burnley, we can see that Gubbins uses features such as the voicing of /f/ and /s/ in volks and zee, which do not align with the features identified as ‘Yorkshire’ in Cooper (Reference Cooper2013, Reference Cooper, Honeybone and Maguire2020). Gubbins’ dialect is also negatively evaluated by the character of O'Leary, who states that he will not put up with Gubbins’ use of language. This follows a trend noted by Hodson (Reference Hodson2014: 84), who notes that ‘dialect speaking characters will often be treated with less respect than characters who speak Standard English’. Gubbins responds that he talks this way because he is a ‘simple Yorkshire lad’ (Pirsson Reference Pirsson1832: 20), and later states that he is a servant, which illustrates the typical kind of occupations Yorkshire characters tend to have in LD. Indeed, we see that there is a much broader range of Yorkshire occupations presented in DL, unlike LD, where Yorkshire characters tend to be one of the following: blacksmith, farmer, gamekeeper, gardener, parson or servant. In DL, the occupations aren't limited to industrial or working-class occupations, as they include the ones represented in LD but also: barber, brewer, chimneysweep, cutler, doctor, fishmonger, gaoler, grocer, joiner, lecturer, policeman, schoolmaster, shepherd, shoemaker, shopkeeper, sweeper, warehouseman, watchman and weaver. This means that there is a much broader range of identities being presented in DL than in LD, which represents the competing valorisations of Yorkshire dialect and Yorkshire identity in the nineteenth century and can be taken as further evidence for sociological fractionation. Furthermore, the ways in which occupations were discussed and presented differs between DL and LD. For instance, in LD we see the notion that characters are presented as ‘only’ a farmer, as in the novel Mary Anerley, ‘Nobbut a farmer am I, in little business’ (Blackmore Reference Blackmore1880: 114). Also, we see that certain farms were prevalent in Yorkshire villages, such as in the fictional Garriton in Harker's novel Phillip Neville of Garriton (Reference Harker1875), where a Yorkshire character refers to ‘t'Scar Top Farm’ as a prominent location in the scene (Harker Reference Harker1875: 4). However, in DL, the representations of farmers differ as we see farmers referenced as one among many Yorkshire occupations alongside butchers, etc., as illustrated in a dialect poem included in Howson's An Illustrated Guide to the Curiosities of Craven, ‘Theear's lambs at's killed wi't butcher's knife, An ducks bith’ hand o'th farmer's wife’Footnote 1 (Howson Reference Howson1850: 117). Additionally, in the Nidderdale Comic Annual, farmers are described as ‘well-to-do’, and being a part of the ‘aristocracy’ of ‘respectable’ villages such as Nidderdale in the Yorkshire Dales, ‘witch includes t’ parson, t’ docter, sundry oade maids and batchlers at livs o’ ther means, a few well-ta-dew shopkeepers, farmers, an’ sike like’Footnote 2 (Nydds Reference Nydds1877: 8).
Additional differences in the ways in which Yorkshire identities were represented in nineteenth-century dialect writing can also be seen in the ways in which the word ‘Yorkshire’ is used in LD compared with DL. Here we see every instance of the word ‘Yorkshire’ in LD in table 2 and in DL in table 3, illustrated in concordance lines generated by Simple Concordance Program (Reed Reference Reed2019), following the methodology of Grund & Smitterberg (Reference Grund and Smitterberg2014) who also used concordance software to analyse conjuncts in nineteenth-century English. Employing this methodology allows for the quantitative comparison of the DL and LD representations of Yorkshire dialect as these differing repertoires form part of the competing valorisations contributing to sociological fractionation. The focus on concordance lines helps to illustrate third-order indexical links being associated with the key words in each case, as well as the specific enregistered Yorkshire identities each variant is associated with, such as ‘Yorkshire Tyke’, ‘Yorkshire lass’ and ‘Yorkshire lad’, etc. The concordance lines for ‘Yorkshire’ illustrate that this word is not very frequent, as it occurs only 24 times in LD and 22 in DL. Of those, half of them come from one poem called ‘I'm Yorkshire Too’, which are the first 11 examples listed in table 3, reducing the range of texts where the name of the county is represented.
Table 2. Concordance lines for Yorkshire in literary dialect
Table 3. Concordance lines for Yorkshire and derived non-standard respellings in dialect literature
The LD representations all feature the Standard English spelling of ‘Yorkshire’, which aligns with the representations of nought as opposed to nowt discussed previously. In the LD, we can also see that ‘Yorkshire’ is predominantly used as an adjective, as in ‘Yorkshire lasses’, ‘Yorkshire folk’, ‘Yorkshire lad’, etc., where we see it used to index place in relation to certain characters. In addition, ‘Yorkshire’ is used to describe particular characteristics of the county and Yorkshire dialect speakers, as in ‘a strong Yorkshire arm’, where physical characteristics are depicted, and ‘a Yorkshire bite’. This latter term appears to allude to a somewhat pejorative term for a Yorkshire speaker as described by Marshall (Reference Marshall2011: 38). He notes that:
During the nineteenth century, ‘Tyke’ (sometimes ‘Tike’) and ‘Bite’ jostled with each other as the favoured slang term for a person from Yorkshire. ‘Tyke’ eventually became a neutral term that carried no other meaning than simply a Yorkshire person.
He goes on to explain that ‘the expression “Yorkshire Bite” ensured the retention of an association between Yorkshire people and low cunning’ (Marshall Reference Marshall2011: 41), meaning that the LD uses of ‘Yorkshire’ here also index negative evaluation of Yorkshire speakers. The term ‘Tyke’ does not appear in LD, but it does occur in DL, as shown in table 3, where we see it adopted in dialect poetry, as in the title ‘The Yorkshire Tike’ (Anon. 1886), and in the title of an annual published in dialect, T’ Yorkshur Tyke's Krusmiss Annual (Tiffany Reference Tiffany1872). This is a further example of competing valorisations as the outgroup use the negatively evaluated term ‘bite’, whereas the ingroup adopt the term ‘Tyke’ to self-identify.
However, in the DL texts we see examples of non-standard respellings such as Yorksher, Yorkshur and Yorrkshir, all of which suggest a different pronunciation in the second syllable, likely [ə], to give something like Yorksh[ə]. Robinson (Reference Robinson1876: 162) provides a similar rendering, Yorkshir, in his Mid-Yorkshire glossary, which he transcribes as [yaork⋅shu'r’], also suggesting a pronunciation similar to those indicated by the respellings in the DL corpus texts. This may have been to distinguish the pronunciation of ‘Yorkshire’ from pronunciations recommended in pronouncing dictionaries, such as Smith's abridged version of Walker's Critical Pronouncing Dictionary (Smith Reference Smith1864: 325), which lists ‘shire’ as ‘a county’ and describes a vowel pronunciation which suggests [iː]. Similarly, Stormonth (Reference Stormonth1874: 576) lists ‘shire’ in his pronouncing dictionary and the vowel pronunciation he describes suggests [ɪ] when appended to county names. Neither of these pronunciations align with the DL respellings, which indicates that Yorkshire writers were likely intending a different pronunciation, although there is no explicit commentary to support this. In addition, the representations in DL tend to include ‘Yorkshire’ used more frequently as a noun, as in ‘i’ all Yorkshire’, ‘West Ridin’ o’ Yorksher’, ‘e Saath Yorkshur’, ‘cum all ye young lads in Yorrkshir’, etc. We do also see some adjectival uses of ‘Yorkshire’ in DL, such as ‘Yorkshire Tike’, ‘Yorkshire folk’, ‘Yorkshire lass’ and ‘Yorkshire sympathee’, which are evidence of third-order indexical links to both place and identity. However, LD characters are simply referred to as ‘Yorkshire’, and the consistent use of the Standard English spelling of ‘Yorkshire’ in LD indexes a competing identity to those observed in the DL, where greater authenticity is indexed via the use of non-standard respellings and the references to more localised regions within Yorkshire. This represents a trend in the corpus data where competing valorisations of Yorkshire identity tend to be indexed by competing Yorkshire repertoires.
The use of Standard English spellings in LD also extends to some Yorkshire dialect features. For example, Forester's (Reference Forester1851) novel The Warwick Woodlands includes the Yorkshire character of Tim Matlock, who is a servant to another character. He is described as being bluff, as well as ‘direct in speech or behaviour but in a good-natured way’, and his language is explicitly evaluated and referred to as ‘the most extraordinary West-Riding Yorkshire’ (Forester Reference Forester1851: 7), which may imply a positive evaluation of this character's dialect. However, although Forester describes Matlock as hard-working, honest, and humorous, he also calls him a ‘piece of Yorkshire oddity’ who is a ‘character’ and a ‘queer’ choice of servant (p. 6). When these descriptions are considered alongside the mention of Matlock's ‘extraordinary’ dialect, we can see that they index associations with the dialect being unusual, strange or out of the ordinary, despite them being ‘good-natured’. Although this is not an overtly negative evaluation, we do not see the overt indexical links to regional pride observed in texts by ingroup writers. For instance, Robinson (Reference Robinson1862: xxix) states that Yorkshire folk have ‘a deep-rooted affection for their own speech’, which is an explicitly positive evaluation of the dialect. There is not a direct comparison in Forester's description of Tim Matlock; the links to positive social values are only made implicitly as a by-product of some of his character traits, rather than explicit metacommentary like that demonstrated by Robinson. Moreover, the specific reference to the West Riding may also index negative social values based on comments by ingroup writers such as Harland, who states that the dialect of Swaledale in the Yorkshire Dales is ‘altogether different from the barbarous jargon of the West Riding’ (Reference Harland1873: 2) in his Swaledale glossary. Beal & Cooper (Reference Beal, Cooper and Hickey2015) demonstrate that urban Yorkshire varieties tended to index more negative social values in the nineteenth century when compared to rural varieties. We may infer from this that Matlock's dialect is not intended to be interpreted as explicitly positive in Forester's novel. In addition, we can see that Matlock's dialect does include features of the enregistered Yorkshire repertoire, as illustrated in his dialogue where he states that he has nothing to say on a particular topic by saying, ‘Weel, Ay'se nought to say aboot it – not Ay!’Footnote 3 (Forester Reference Forester1851: 7). Despite the inclusion of weel, Ay'se, and aboot, the use of nought for ‘nothing’ demonstrates engagement with the historically enregistered repertoire of Yorkshire dialect identified by Cooper (Reference Cooper2013), although Forester does not employ the more commonly enregistered form nowt. This can be explained by the fact that there were multiple variants of this word in nineteenth-century writing, as illustrated by Robinson (Reference Robinson1862: 375), who demonstrates awareness of these variant spellings in nineteenth-century Yorkshire dialect writing, as he lists ‘NOWT. Nothing’ in his Leeds dialect dictionary. In his definition he states, ‘the spelling various, but in every case is a close approximation to this form’. This comment indicates that nowt is the variable form of this feature, with the alternate spellings being variants.
When we consider written representations of nowt ‘nothing’ variants over time, we can see that there is a lot of variation at the start of the nineteenth century in both DL and LD as illustrated in figure 1, which highlights eight different variants totalling 191 tokens in the corpus. This is similar to the variation recorded by Johnstone et al. (Reference Johnstone, Andrus and Danielson2006: 96) for the second-person plural of ‘you’ in Pittsburghese, where they note the existence of seven different variants including yinz, yunz and you-unz. They go on to state that there is a tendency for phonetic spellings such as yinz to become dominant over forms such as you-unz, which reflect the etymology of the regional variant. Ultimately, yinz essentially becomes the vernacular norm in Pittsburgh and is the non-standard variant which tends to be used when representing ‘Pittsburghese’ in writing.
Figure 1. Variants of nowt ‘nothing’ per 10,000 words in corpus data for each decade of the nineteenth century
The variation shown in figure 1 similarly reduces over time and by the end of the century, the only non-standard variant left in the corpus in the 1890s decade is nowt. The other variants are naught and nought, which are both Standard English spellings. However, although nowt appears to emerge as the vernacular norm over the course of the nineteenth century, we can see that there is a slightly different pattern if we consider the distribution of variants in DL produced for local audiences as opposed to LD produced for wide audiences, as shown in figure 2.
Figure 2. Percentage proportions of nowt ‘nothing’ in literary dialect and dialect literature
Figure 2 illustrates the relative use of nowt variants in both LD and DL. We can see that some variants were used almost exclusively in DL, such as naut, nawt and nowght. Others turn up in mixed proportions, such as nout, which predominantly appears in DL. Only noot appears in LD alone, and this represents one token in the whole corpus from a text printed in 1800. As there is no commentary on this specific form we cannot draw any firm conclusions from this without more data. The two standard variants naught and nought occur in much greater proportions in LD than in DL, and the vernacular norm nowt was represented in equal proportions. This pattern may be explained by considering additional commentary on authenticity in relation to representations of Yorkshire dialect. For instance, Stead (Reference Stead1878: 36) describes the ‘genuine old dialect’ in Holderness as being in opposition to ‘standard or ‘proper’ English’. Similarly, Burnley (Reference Burnley1875: 10) explains that the inauthenticity of the ‘stage Yorkshireman’ character is a result of ‘the distinctive features of the language of the Yorkshire rustic’ being missing. The use of the Standard English nought and naught in LD would appear to demonstrate the opposition Stead refers to, and the lack of non-standard variants aligns with Burnley's commentary about the lack of authentic features. Moreover, Robinson (Reference Robinson1876: 93) states that there are so many variant pronunciations of nought, some of which are ‘opposite’, that the ‘truly characteristic form is apt to be lost sight of’. This explains the variation in the DL, as writers use multiple competing variants over time, resulting in nowt, and aligns with Robinson's (Reference Robinson1862) comment that all spelling variants are some approximation of nowt. Figure 2 therefore illustrates competing valorisations presupposing underlying commonalities as the outgroup writers use Standard English spellings to index Yorkshire identity to wide audiences combined with ‘generic’ Yorkshire spellings, whereas ingroup writers attempt to index authenticity by using many different variants which tend towards one non-standard variant over time.
In analysing these data, we can take the discussion and representations of dialect presented above as evidence of sociological fractionation in the nineteenth century and the formation of at least two main ideological groups. One views itself as an ingroup and can be seen in the comments of writers like Hamilton, Burnley, Howson, Stead and Nydds, who were from Yorkshire themselves. Writers in this group complain about the representations of Yorkshire speakers when they are depicted as boorish with inaccurate dialect. Moreover, in their own representations of Yorkshire dialect, these writers employ elements of the sub-Yorkshire repertoires identified by Cooper (Reference Cooper, Honeybone and Maguire2020), likely in an attempt to index authenticity. On the other hand we see the ideological outgroup which is mostly comprised of speakers who are not from Yorkshire, and represent Yorkshire speakers using the ‘stage Yorkshireman’ stereotype. When these writers represent Yorkshire dialect, they tend to rely on features which are enregistered as generally ‘Yorkshire’ without the additional regional specificity identified by Cooper (Reference Cooper, Honeybone and Maguire2020), or they make use of Standard English terms as in the case of nought, which is used to index Yorkshire identity to a wide audience as opposed to nowt, which simultaneously indexes authenticity. The use of nought and not nowt in LD is a further example of competing valorisations underpinning sociological fractionation, as the Standard English spelling is used in texts aimed at wide audiences, whereas the most frequent and consistent non-standard respelling nowt is predominantly used in texts aimed at local audiences. It follows that dialect is central to debates surrounding Yorkshire identity in nineteenth-century writing, as evidenced by the commentary presented above. In addition, Johnstone & Baumgardt's (Reference Johnstone and Baumgardt2004) discussion of ‘Pittsburghese’ speakers debating vernacular forms online illustrates that there are differing social values indexed by these forms for groups of speakers who view Pittsburgh in either a positive or negative light. They conclude that differing ‘ideological contexts for norm-formative talk give rise to different ways of arguing for norms and hence different norms’ (Reference Johnstone and Baumgardt2004: 141), and where we see sociological fractionation occurring, the different fractions of society do indeed hold different norms. We see this reflected in both the different social values discussed by nineteenth-century writers, as well as the differing use of enregistered Yorkshire features in writing. The accompanying commentary from the ingroup resisting the schemes of values indexed for the outgroup may also be taken as evidence for sociological fractionation of the kind Agha describes.
4 Conclusions
The opposing views of Yorkshire dialect and identity in nineteenth-century dialect writing highlight the differing ideologies and schemes of social values held by each group. In fashioning norms for themselves, literary writers contributed to the emergence and reinforcement of the ‘stage Yorkshireman’, a ‘boorish’ character who predominantly uses generic, non-standard Yorkshire-esque dialect that doesn't readily align with the enregistered repertoires previously identified. As Marshall (Reference Marshall2011: 11) notes in his discussion of the historical creation of ‘Yorkshireness’ in relation to Yorkshire identity, the ‘stereotype of the hard-headed, often boorish “Yorkshire Tyke” or “Yorkshire Bite” was nationally established by the eighteenth century’, and these nineteenth-century texts participate in the maintenance of that stereotypical identity. However, in opposing this norm, the ingroup are seen to be resisting the scheme of values put forward and reiterated by outgroup writers and they are fashioning norms for themselves by presenting dialect representations that index a greater degree of authenticity to local readers, as well as depicting a broader range of occupations to illustrate a broader range of ‘Yorkshire’ identities. This ultimately shows that the phenomenon of sociological fractionation can be observed in historical contexts and develops our understanding of the social values associated with regional dialects for different historical audiences.
Open access
