301
Anne-Christine Gardner DOI: 10.4312/linguistica.63.1-2.301-336
Université de Lausanne
anne-christine.gardner@unil.ch
ENGLISH PAUPER LETTERS IN THE EIGHTEENTH CENTURY 
AND BEYOND: 
ON THE VARIABILITY AND EVOLUTION OF A NEW TEXT 
TYPE*
1 INTRODUCTION
Text typological studies have long attracted the interest of scholars exploring language 
variation and change in the history of English. Adopting a multidimensional approach, 
Biber/Finegan (1989: 512), for instance, show that letters rank among speech-like texts 
(cp. Culpeper/Kytö 2010: 18) and observe an evolution in letters from the seventeenth 
to the twentieth century in the shape of a drift from a more literate to a more oral style. 
While a significant amount of attention has already been paid to the familiar letter (e.g. 
Fitzmaurice 2002, Dossena/Tieken-Boon van Ostade 2008) and business correspond-
ence (e.g. Dossena/Fitzmaurice 2006), as well as formulaic sequences typical for epis-
tolary writing, also in languages other than English (e.g. Nevalainen/Raumolin-Brun-
berg 1995; Nevala 2004; Dossena 2006; Laitinen/Nordlund 2012; Rutten/van der Wal 
2012, 2014), the primary focus in this article lies on the value of formulaic sequences 
for the exploration of a new subtype of vernacular correspondence emerging in the 
eighteenth century, i.e. the pauper letter. 
Social change involving the creation and modification of a welfare system brought 
about the development of this new type of letter. Under the Old Poor Law, originally 
established in 1601, anyone in distress could appeal to their parish of legal settlement 
in order to obtain relief, which often took the form of financial aid, to be paid weekly or 
as a lump sum. Settlement could be gained through, e.g., birth, marriage or a completed 
apprenticeship (Whyte 2004: 280). Non-resident poor, i.e. those who had migrated to 
another parish, could also apply for support by writing to the overseers and church war-
dens, but risked being removed to their home parish where they would be relieved, but 
had fewer prospects (Sokoll 2000: 24; on migration patterns of paupers see Gardner et 
al. 2022).1 It was then often more economic for home parishes to grant relief to paupers 
at their current domicile, so-called out-relief (Sokoll 2000: 24). Letters abound where 
applicants use this circumstance as a negotiating factor when soliciting relief. 
* The author would like to thank Oliver Currie, Anita Auer, Mark Iten and two anonymous reviewers for their 
helpful comments on earlier versions of this article, which was written in the context of the SNSF-funded 
project The Language of the Labouring Poor in Late Modern England (2020–2024; 100015_188879).
1 In this article the term pauper is used both in the specific sense of ‘a recipient of relief under the provisions of 
the Poor Law or of public charity’, as well as with its broader meaning of ‘a very poor person’ (OED Online).
302
Only few pauper letters survive from the eighteenth century, starting in the 1730s 
(Sokoll 2001), but became much more numerous after a legislative change in 1795, as a 
result of which removal still remained a possibility, yet posed a lesser threat than before 
(for details see Sokoll 2001: 13, Sokoll 2008: 111–115). A noticeable increase in out-
relief applications can also be observed from the end of the first decade of the nineteenth 
century, when many labouring poor found it difficult to make ends meet, for instance 
after the “Year Without a Summer” (1816) (Brönnimann/Krämer 2016), which resulted 
in food shortages, and after the Napoleonic Wars (Beardmore 2020: 144).2 Collecting 
poor-relief correspondence for Essex, Sokoll finds that fewer than 2% of the letters date 
to the eighteenth century (13 out of 758 letters), and that there is a “massive concentra-
tion of the material in the 1820s and early 1830s” (2001: 19); this serves to highlight the 
scarcity of pauper letters from the eighteenth century. Nevertheless, pauper letters are a 
“rich demotic source” (Jones/King 2015: 56) and even as smaller datasets offer valuable 
findings, as shown by earlier research by e.g. Fairman (2008), Gardner et al. (2022), and 
Auer et al. (2023). Pauper letters unlock the voices and experiences of the labouring poor, 
providing us with a much-needed perspective ‘from below’ (Elspaß 2005, Auer et al. 
2015, Timmis 2020: 78–80), which remains under-researched in text typological studies.
The poor had already appealed for financial support in writing long before poor-
relief applications were penned. These texts were typically formal petitions by the resi-
dent poor addressed to magistrates and dealt with at quarter sessions (Hindle 2004: 408, 
Healey 2016: 84–6). Jones/King (2015: 76) describe a petition as a “closed question” 
or “two-way monologue” where the request is either granted or dismissed, whereas a 
pauper letter initiates a dialogue for negotiating relief through correspondence. Another 
major difference between petitions and pauper letters is that the former were normally 
encoded by “highly competent scribes” (Jones/King 2015: 75), while the latter were 
mostly written by the supplicants themselves or someone from their immediate social 
circle (Sokoll 2001: 64; Jones/King 2015: 72, 75; King 2019: 36). The historians just 
referenced refer to poor-relief applications as pauper letters rather than petitions, high-
lighting differences in the production circumstances and communicative setting, but 
acknowledge the potential stylistic influence of petitions. According to Sokoll (2001: 
57), “pauper letters, in their basic composition, rhetoric and gesture, may be said to be 
bounded by two contemporary types of epistolary expression: the ‘familiar’ letter on 
one hand, and the formal petition on the other”. In Sokoll’s view, relief applications 
resemble the familiar letter more than petitions (ibid.), while Jones/King (2015: 73–4) 
note that the question of “how and why the emphasis shifted so comprehensively from 
one form of appeal to the other over this period” remains unanswered. In earlier lin-
guistic research, in contrast, the labels ‘petition’ and ‘pauper letter’ are sometimes used 
synonymously (e.g. Laitinen/Auer 2014, Auer 2015). This article sets out to determine 
empirically, firstly, whether poor-relief applications should be classified as petitions or 
2 Only following the 1819 Vestry Act parish records were officially required to be kept in order, and increasingly 
assistant overseers and vestry clerks were paid to manage these records (Sokoll 2001: 23). Further reasons for 
the uneven diachronic distribution of pauper letters are discussed in King (2000: 414–5), Sokoll (2001: 19) and 
King (2019: 27).
303
as letters, and secondly, whether there was indeed a shift from petition style to letter 
style from the early eighteenth century towards the nineteenth century.
With respect to text classification, differing approaches (e.g. Biber/Finegan 1989, 
Kohnen 2001, Görlach 2004) have resulted in a “terminological maze” (Moessner 2005) 
– concerning terms such as ‘genre’, ‘text type’ and ‘register’ –, which has been compre-
hensively reviewed by Diller (2001). In this article, Kohnen’s definition of text types as 
“dynamic patterns of communication combining aspects of function, context and form” 
(2001: 198) is adopted, which is able to account for variability and evolution in a text 
type over time. For the textual classification of pauper letters, five essential features will 
be investigated (voice, greeting, petition-element, closing formula and self-reference) 
which serve to distinguish petition-style from letter-style writing. These features have 
been chosen because they have a text-constitutive function, meaning that “[w]ithout 
actually having to read the body of the text, the text type is revealed” (Rutten/van der 
Wal 2012: 178; see also Wray (2002: 101) and Rutten/van der Wal (2014: 82–6)). Such 
an approach makes it possible to perform a quantitative text type analysis on a larger 
dataset, meaning that close reading of the texts in their entirety is not required for an 
initial classification and content is eliminated as a factor determining text type. Coher-
ence across the dataset in terms of content is ensured by focussing on the first application 
letters surviving from all selected correspondence sets (see also Section 3). Since the 
five selected features are realised very differently in model letters and model petitions 
presented in letter-writing manuals, as illustrated in Section 4.1, they provide an imme-
diately discernible contrast between the two writing styles and are therefore sufficient 
for the purposes of text type classification.3 Closing formulae (e.g. in duty bound shall 
ever pray) and self-references (e.g. your humble servant) are typically composed of 
formulaic sequences, which can index social relationships and the writer’s identity (e.g. 
Del Lungo Camiciotti 2006; Laitinen/Nordlund 2012). As such they are highly revealing 
with regard to the social practice of letter-writing (Barton/Hall 2000: 1, Rutten/van der 
Wal 2012: 194–5), offering insights into how the poor viewed and portrayed themselves 
in the application process, what kind of education they received, how they acquired for-
mulaic sequences, and to what extent their use of these formulaic sequences differs from 
that of other social groups such as the better-educated parish officials.
In order to better understand the variability and evolution of pauper letters, inter- 
and intra-writer variation as well as self-corrections in closing formulae and self-ref-
erences (as they appear throughout the text as well as attached to closing formulae) 
will be examined in more detail. Particular attention will be paid to how the linguistic 
choices regarding the five essential features relate to the communicative setting and the 
writer’s social roles, as well as to the production circumstances of pauper letters, i.e. the 
question of who encoded the letters, and the educational background of the applicants. 
3 Follow-up studies may be interested in analysing rhetorical and pragmatic or politeness strategies 
used in pauper letters with a view to comparing petition-style and letter-style writing. However, earlier 
research has not identified a differing use of rhetorical strategies in the two writing styles (e.g. Sokoll 
2001; King 2019; Timmis 2020), which is a further reason for their exclusion in the present study, 
while the analysis of pauper letters from a pragmatic point of view lies outside the scope of this article.
304
While earlier research has touched on the language of pauper letters and the possible in-
fluence of petitions and models in letter-writing manuals on the writing of the labouring 
poor (e.g. Sokoll 2001, Auer 2015, Jones/King 2015, King 2019, Calvo Cortés 2020, 
Timmis 2020), these studies are based on smaller datasets or letters to different insti-
tutions, or they do not incorporate a more fine-grained diachronic analysis of pauper 
letters from a linguistic perspective; furthermore, a closer examination of intra-writer 
variation in applications for poor relief has not yet been undertaken. 
The following section discusses the (limited) educational opportunities the labouring 
poor had in order to acquire letter-writing skills, which has a bearing on the linguistic 
repertoire available to them. Section 3 describes the correspondence material on which 
this article is based in more detail, i.e. 203 letters written between 1730 and 1834, pri-
marily drawn from a corpus which is currently being compiled as part of the project 
The Language of the Labouring Poor in Late Modern England (LALP). The analysis of 
the poor-relief correspondence is divided into two parts. Section 4 is concerned with a 
diachronic investigation of the stylistic impact of petitions on pauper letters, providing 
a text-type classification for poor-relief applications. Section 4.1 identifies five relevant 
features (voice, greeting, petition-element, closing formula and self-reference) which can 
be found in pauper letters and where authors theoretically have the choice of writing in 
the style of a petition or of a familiar letter. Section 4.2 investigates actual usage pat-
terns in pauper letters regarding these features, with a view to shedding more light on 
the observation by King/Jones (2015: 73–4) that in terms of preferred writing style there 
was a shift from petition to letter towards the nineteenth century. Section 5 focuses on 
the two features where variation in pauper letters is most extensive, i.e. closing formulae 
and self-references. Inter-writer variation concerning these features will be examined in 
Sections 5.1 and 5.2, respectively, both within pauper letters and, by way of comparison 
in order to shed light on social practices of letter-writing, in a sample of letters by parish 
officials and advocates (e.g. servant, doctor, landlord) who wrote on behalf of poor-relief 
applicants. In further steps, intra-writer variation (Section 5.3) and self-corrections (Sec-
tion 5.4) in pauper letters will also be investigated, offering valuable insights into the ac-
quisition and usage patterns of stylistic conventions as exhibited in closing formulae and 
self-references. Finally, the article closes in Section 6 with a brief summary and outlook.
2 ACQUIRING LITERACY AND LETTER-WRITING SKILLS
Working-class families typically relied on additional income generated by their chil-
dren in order to make ends meet. With elementary education only becoming compul-
sory in England in 1880 (up to the age of 10, Stephens 1998: 79), children from poor 
backgrounds either received no schooling at all or had to abandon their education at 
an early stage, equipped with reading skills but unable to write, which was usually 
taught after reading (Stephens 1998: 2). An exception were National Schools run by 
the Church of England from 1811 onwards, where writing was taught first (Gardner 
2023a; Iremonger 1813: 272).4  Composition (e.g. of letters), a skill which would 
4 For more information on the limited educational opportunities of the labouring poor see e.g. Timmis (2020: 
51–71), Auer et al. (2023) and Gardner (2023a).
305
have been useful for poor-relief applicants, did not become part of the school cur-
riculum until 1871 (Vincent 1989: 89). In the eighteenth century, working-class chil-
dren often left school by the age of 8, and even by the mid-nineteenth century rarely 
received schooling after reaching the age of 10 or 11 (Stephens 1998: 2, Vincent 
2014: 274). As a result, many among the labouring poor were unlikely to be more 
than semi-literate initially.
However, many of the lower classes are known to have continued their education 
alongside work, either informally by themselves, with their peers or in mutual im-
provement groups, as well as more formally in Sunday schools (from the 1780s on-
wards) and evening schools amongst others (Lawson/Silver 1973: 189–95, 238–50, 
Stephens 1998: 3, 5). Data from Suffolk prison records, for instance, show that among 
the offenders the most literate are over 40 years old, and the most illiterate under 15. 
Repeat offenders typically acquired a new skill in prison school or in-between prison 
sentences, either becoming semi-literate or moving from partial to full literacy (Crone 
2018: 182–3). Lifelong learning thus played a significant role in the education of the 
labouring poor. Yet based on signature literacy, by 1840 still only approximately 27% 
of unskilled labourers and 21% of miners could read and write, while average signature 
literacy was at around 67% for men and 50% for women (Vincent 1989: 93, 97).
There is little evidence of poor-relief applicants directly relying on models presented 
in letter-writing manuals (Sokoll 2001: 59, Auer 2015: 142, King 2019: 36, see also Sec-
tion 4.1). At the time of publication the letter-writing manuals by Brown (1770?) and 
Cooke (1770?) cost one and two shillings, respectively, which would have been forbid-
dingly expensive for most of the labouring poor when the letters in the corpus document 
the hopes of many of receiving one or two shillings per week (see also (1)) in order to 
ensure survival. Letters and petitions were also printed in newspapers, which would be 
more readily accessible to the poor than manuals. However, it must be born in mind that 
the authors of manuals drew on actual practice (King 2019: 362–3, fn. 15), meaning 
that the art of writing letters and petitions must have been part of communal knowledge. 
As Timmis (2020: 80) convincingly argues, literacy should be seen “as a community 
resource”. Even if a person was not literate enough to (read and) write themselves (‘pos-
sessive literacy’), they would be able to approach someone in their community who was 
(‘accessive literacy’) (cp. Timmis 2020: 81). Research has shown that letter-writing could 
have been taught by family or community members, and received letters could have been 
used as models (e.g. Whyman 2009: 221, Auer 2015: 143, Timmis 2020: 90–1). Since 
letters were often read aloud even the illiterate could become familiar with certain for-
mulations typically found in epistolary communication (Timmis 2020: 91). Besides their 
family, friends and neighbours, other members of the local community could also provide 
assistance during the writing process, including landlords, employers, creditors, overse-
ers, tradesmen, teachers, clergymen and members of the military (Sokoll 2001: 65, Hou-
ston 2014: 81, Jones/King 2015: 67, King 2019: 35). In this vein, Jones/King propose the 
existence of a “shared linguistic register of appeal […] that cut across social, economic 
and structural boundaries” (2015: 69) and suggest that pauper letters are “the product of a 
community of voices that interacted to produce a vernacular of appeal” (2015: 76). 
306
3 DATA 
The letters examined in this article were collected as part of the LALP project, which 
aims to build a corpus of letters written by paupers who applied for poor relief under the 
Old Poor Law (c. 1795–1834). As a result of opportunistic sampling in archives across 
England, the project also unearthed letters composed earlier in the eighteenth century, 
as well as correspondence by parish officials and advocates writing about or in support 
of paupers. For this article, three different subcorpora were compiled, drawing on a col-
lection of letters which were transcribed by members of the LALP project as close to 
the original as possible.5 The first was designed for the diachronic study of inter-writer 
variation and contains pauper letters written between 1730 and 1834 (‘Diachronic Cor-
pus’), the second acting as a ‘Supplement’ with letters from advocates and officials from 
the earlier eighteenth century for comparison. The third corpus forms the basis for the 
analysis of intra-writer variation and is comprised of letter sets by six poor-relief appli-
cants (‘Individual Corpus’). The following two subsections elaborate on the sampling 
principles and make-up of these corpora, including the authorship of the sampled let-
ters. In total, 203 letters containing 39,674 words were prepared for analysis.6
3.1 Diachronic Corpus and Supplement
Linguistic choices in letters, especially concerning greetings, closing formulae and self-
references, varied depending on the degree of familiarity between writer and addressee 
(e.g. Nevala 2004). In order to ensure comparability between pauper letters, missives 
indexing familiarity on a personal level or intimacy (an infrequent occurrence) rather 
than the social relationship of applicant and parish official were not considered, so that 
the sample only contains writing produced in a formal setting. In terms of content, the 
sample was restricted to first applications for poor relief, or, where not available, the 
earliest re-application for relief obtained from the archives. 
Following these principles, the Diachronic Corpus of pauper letters comprises 
139 letters (26,509 words) based on the LALP collection, with three letters added 
from Sokoll (2001: Letters 280, 281 and 707) to supplement data from the earlier 
eighteenth century.7 For the diachronic analysis of pauper correspondence in Sec-
tions 4 and 5, the material was divided into three subperiods covering 25 to 30 years 
each (Table 1). The gap of 15 years between Subperiods I (1730–1759) and II (1774–
1799) allows for a comparison between the earlier and the later eighteenth century 
without running the risk of imposing an artificial boundary in a continuous corpus, 
while Subperiod III (1807–1834) contains only letters from the nineteenth century 
5 With the LALP corpus still under construction, suitable letters for the current investigation were sampled from 
35 of 39 historic counties of England, without any particular regional bias. Limitations in data availability, 
particularly in the eighteenth century, currently do not permit the investigation of regional variation in the 
occurrence of stylistic features.
6 Addresses (or superscriptions) providing directions, as well as annotations on the documents in other hands 
were excluded from the word count and the analysis.
7 The transcription of Letter 281 used for analysis is my own, based on the image provided by Sokoll (2001: 
Plates XIX–XX).
307
up to the implementation of the New Poor Law in 1834. To avoid overrepresentation 
of individuals and thus an imbalance within Subperiod III, only one letter from each 
pauper was sampled. As a result, the gap between Subperiods II and III had to be 
reduced to 8 years. However, 85% of the letters in Subperiod III are dated between 
1814 and 1834, which means that for most of the material a gap of 15 years to the 
previous subperiod is in fact observed. 
Table 1. Diachronic Corpus (pauper letters)
Subperiod Years Senders Number of letters Word count
I 1730–1759 Paupers 22 4,166
II 1774–1799 Paupers 17 3,207
III 1807–1834 Paupers 100 19,136
Total 139 26,509
The Supplement consists of 14 letters from the earlier eighteenth century by of-
ficials and advocates interceding on behalf of paupers (Table 2), and will be used 
to illustrate inter-writer variation between different social groups in Sections 5.1 
and 5.2.
Table 2. Supplement (letters by advocates and officials for Subperiod I)
Subperiod Years Senders Number of letters Word count
I 1730–1759 Advocates 5 990
Officials 9 1,537
Total 14 2,527
As regards authorship of the pauper letters, a basic distinction is made for the 
Diachronic Corpus between (a) ‘authentic’ letters which were probably written by 
the paupers themselves or by someone from their circle with limited schooling 
(cp. Sokoll 2001: 65), and (b) ‘non-authentic’ letters which were probably encoded 
by someone with significant training and education.8 These typically boast unusu-
ally neat and elaborate handwriting on even lines, sometimes with flourishes on 
individual characters, as well as careful layout, for instance observing deferential 
space (Sairio/Nevala 2013). The proportion of non-authentic letters in Subperiod 
II is much higher than in the two other subperiods (Table 3). However, this is the 
result of opportunistic sampling and probably also reflects the vagaries of archival 
survival. There is too little data to warrant the hypothesis that at the end of the 
eighteenth century paupers took recourse to professional support more often than 
in the other periods.
8 For a more detailed discussion of authorship and authenticity of pauper letters see Gardner (2023b).
308
Table 3. Authenticity of pauper letters in Diachronic Corpus
Authenticity
Subperiod I
(1730–1759)
Subperiod II
(1774–1799)
Subperiod III
(1807–1834)
Letters Words Letters Words Letters Words 
Authentic 18 3,459 10 1,792 87 16,683
Non-authentic 4 707 7 1,415 13 2,453
The letters by advocates and officials in the Supplement are not examined further 
with respect to authorship, but are presumed to be authentic, i.e. representative of writ-
ings from the respective societal group to which each author belongs. Part of a parish 
official’s duties was to engage in correspondence, sometimes with the support of a 
clerk, and advocates writing on behalf of the poor are believed to have acted on their 
own accord.
3.2 Individual Corpus
For the analysis of intra-writer variation in Section 5.3, six individuals were selected 
who sent a total of 54 letters to receive out-parish relief in the nineteenth century: 
Charls Ann Green, John Hammont, Sarah Hughes, Robert Kingston, Augustine Mor-
gan and Frances Soundy (Table 4). Owing to the scarcity of available archival records, 
a comparable study is currently not possible for eighteenth-century writers. However, 
the findings of this case study are also likely to apply to the earlier material because of 
the similarities in the limited schooling opportunities of the labouring poor and in the 
production circumstances of their letters. 
Table 4. Individual Corpus9
Applicant Current 
domicile
Parish of legal 
settlement
Period Number of letters 
(and hands)
Word 
count
Autographical
Charls 
Ann 
Green
London Wimborne  
(Dorset)
1818–1826 8 
(same hand)
894
Augustine 
Morgan
Beaminster Blandford Forum 
(Dorset)
1803–1806 6 
(same hand)
1,023
Frances 
Soundy
Battersea Pangbourne 
(Berkshire)
1818–1830 20 
(same hand)
7,792
Total autographical 34 9,709
9 Four letters (one each from Green, Hammont, Hughes and Soundy) are also part of the Diachronic Corpus 
(Subperiod III). ‘Current domicile’ refers to the parish in which an individual resides at the time of application, 
i.e. the parish to which they had migrated to, whereas ‘Parish of legal settlement’ indicates the parish in which 
they had originally gained settlement (see Section 1).
309
Applicant Current 
domicile
Parish of legal 
settlement
Period Number of letters 
(and hands)
Word 
count
Non-autographical
John 
Hammont
Manchester Brampton 
(Huntingdonshire)
1821–1824 4 
(4 different hands)
338
Sarah 
Hughes
London Pangbourne 
(Berkshire)
1829–1830 11 
(5 different hands)
968
Robert 
Kingston
Huntingdon St Andrew 
the Less 
(Cambridgeshire)
1820–1821 5 
(2 different hands)
432
Total non-autographical 20 1,738
TOTAL 54 11,447
The question of whether the letters by the six applicants are autographical or not, 
and if non-autographical whether they are authentic or not, is of central importance 
in the linguistic analysis. In the case of three of the applicants (John Hammont, Sa-
rah Hughes and Robert Kingston), several hands were involved in the composition 
of their letters over the course of time, and consequently their letters are deemed 
non-autographical. In other words, the person who penned the actual letter (writer) 
and the person whose name appears in the signature and who solicits support from 
their home parish (applicant) are not identical. It would appear that none of the three 
applicants were even able to write their own name, since the handwriting of the 
signatures (and even the spelling of the names) varies across the parishrent hands 
involved. Nevertheless, all non-autographical letters can be considered authentic, i.e. 
representative of the writing of the labouring poor, on account of the limited training 
evidenced in handwriting and layout (see previous subsection), with three excep-
tions: the third letter sent by Kingston and the last two sent by Hughes were clearly 
encoded by well-educated scribes. 
For two other individuals, Augustine Morgan and Frances Soundy, clues in the let-
ters allows us to conclude that missives are autographical, i.e. writer and applicant are 
identical. The production circumstances which Soundy describes in (1) leave no doubt:
(1) PS Sir I have wrote this un be none to any one But my salf (3_Soundy_1828_3)10
Likewise, Morgan must have been able to write as in one letter, not investigated 
here as it contains only brief instructions on how Morgan’s relief should be conveyed, a 
certain Joseph Barratt acts as a witness to his handwriting. Lastly, Charls Ann Green’s 
letters are also considered autographical. The eight letters sent in her name, and appeal-
ing for support for her family, were written by the same untrained hand over a period 
10 The filenames of pauper letters follow the model ‘Subperiod_Surname of applicant_Year’. In the Individual 
Corpus, several letters by the same applicant from the same year are numbered in temporal sequence with the 
addition of ‘_Number’ after ‘_Year’. Non-authentic letters are marked with a superscript N at the end of the 
filename.
310
of eight years. Following Sokoll (2001: 64) and King (2019: 37), it is unlikely that the 
same scribe would have assisted her in the writing of the letters during such a long 
period of time. An analysis of dialectal features in Green’s letters shows that the writer 
must have had their linguistic anchor in Dorset, where her parish of legal settlement is 
situated, which gives further support to the assumption that Green wrote these letters 
herself (Gardner et al. 2022). 
4 PAUPER LETTERS: CATEGORISATION AND EVOLUTION
This section explores the impact of petitions on linguistic choices in pauper letters, first 
identifying five features typical for petitions where the authors of pauper letters have 
the option of adopting these or of writing in the style of a familiar letter (Section 4.1). A 
diachronic analysis of these features in pauper letters then follows in Section 4.2, with a 
view to classifying pauper letters as a text type and exploring the evolution of this text 
type from the early eighteenth to the nineteenth century.
4.1 Petition-style vs. letter-style writing
Both petitions and letters share the same rhetorical roots, having developed from Clas-
sical models which were subsequently adapted in medieval times (Sokoll 2001: 57, 
Houston 2014: 73–5, Jones/King 2015: 59f.). Jones/King (2015: 60) illustrate that the 
rhetorical tradition of petitions is remarkably uniform across time, space and culture, 
which leads them to state that “the form of the written petition in history remained 
[…] consistent over more than 1,800 years”. In the eighteenth century, guidance works 
continued to perpetuate the model of the formal petition. The history and audience of 
English letter-writing manuals, the first being printed in the mid-sixteenth century, has 
been well-documented (e.g. Bannet 2005, Fens-de Zeeuw 2008, Auer 2015). The first 
such guides to specifically offer sections with model petitions appear to be The Uni-
versal Letter-Writer; or, New Art of Polite Correspondence by Thomas Cooke (1770?) 
and The New and Complete English Letter-Writer; or, Whole Art of General Corre-
spondence by George Brown (1770?). These sections are advertised within the lengthy 
titles of the works as Containing great Variety of Petitions on various Subjects, from 
Persons in low or middling States of Life, to those in higher Stations (Cooke 1770?) 
and The New Universal Petitioner, Comprehending The greatest Variety of Petitions, 
adapted to every Situation, with Directions for presenting them in a proper Manner 
(Brown 1770?). Example (2) reproduces a model petition from Brown (1770?: 195f., 
original italics), which presents the case of a poor widow applying to her parish for 
financial support.
(2)         To the Maſter, Church-Wardens and Overſeers of the Pariſh of --------.
The humble Petition of A. B.
      Sheweth, 
That your petitioner’s huſband was an honeſt induſtrious man, and lived many 
years in credit in the pariſh, where he ſerved every office, and paid ſcot and lot; 
but dying in diſtreſſed circumſtances, owing to his buſineſs having fallen off ſome 
311
years ago, ſhe is left utterly deſtitute. In this unhappy ſituation ſhe has preſumed to 
addreſs herſelf to you; and as ſhe has a little work to do, when able to go through 
with it, ſo ſhe ſubmits to you, whether the allowance of two ſhillings per week 
would not be better than going in to the workhouſe. Your petitioner humbly hopes 
that her caſe will be taken into conſideration.
And ſhe, as in duty bound, will ever pray.
This case is reminiscent of the difficult circumstances which cause the lower classes 
to turn to their parishes of legal settlement for out-relief. One such application, by a 
husband with a large family and a wife who is unwell (1_Breddy_1746), is reproduced 
in Figure 1 and transcribed in full in (3).
(3) Poole [^SEAL^]rch 25 1746
ſir
      My necessity att presencs doth oblige
me to trobel you in this afair my family
being Large and my wife Lying ill ſo
Long and my misforten in Lameing my ſelf
for time back have brouft me be hind hand                                
with my house rent and Except I have not
the money to pay I do Expect my Goods to be
ſold wich I Cannot do with out ſo I hope
that you and the Gentelmen will Consider
the Clamytys I have d had in my famyly and
if you pleas to ſend me 30 ſhillings it
will keep me from f[^{…} HOLE^]ar trobel and I hope
not to trobel you any moore from your
                ſaruant to Comand
                    John Breddy
as I Cant Come my ſelf I hope my
ſister will apear in my be half
On the basis of the model petition in (2) and a similar petition by a destitute widow 
in Cooke (1770?) (discussed in Auer 2015: 140–1), as well as the pauper letter in (3) 
and the other applications under investigation, it is possible to identify five essential 
features which are prescribed for petitions and to determine their realisation in ap-
plications written in the form and style of a familiar letter (Table 5). While petition-
style features are fairly uniform across different samples, letter-style writing presents 
a significant amount of variation. Petitions are typically written in the third person and 
open with a heading containing the title(s) of the addressee(s) followed by the title of 
the petition and the sequence Sheweth, That across two separate lines (petition-element 
for short). Letters, in contrast, are normally written in the first person and begin with 
salutation such as Gentlemen or Sir, lacking the petition-element. The closing formula 
312
expected in petitions is the “prototypical petitionary phrase” (Sokoll 2001: 59) involv-
ing a combination of (as) in duty bound and shall/will ever pray; your petitioner is 
used as a self-reference throughout the text and often also at the very end before the 
signature. In contrast, in letters there is considerable variation when it comes to closing 
formulae and self-referential terms (Sections 5.1 and 5.2). Although this list of fea-
tures (and examples) is not exhaustive, it serves to illustrate major differences between 
petition-style and letter-style writing, and as such provides as a useful starting point for 
the analysis of variation in pauper letters in the following section.
Figure 1. Letter by John Breddy, 25 March 1746, Poole11
11 This image is reproduced with kind permission of the Dorset History Centre (PE-BF/OV/13/1).
313
Table 5. Typical features of writing in petition style and letter style 
Feature Petition style Letter style
Voice 3rd person 1st person
Greeting Heading12 (To the Master/Minister, 
Church-wardens and Overseers of the 
Parish of …) 
Salutation (e.g. Sir, 
Gentlemen, Mr NAME)
Petition-element The humble Petition of NAME // Sheweth, 
// That13
X
Closing formula (as) in duty bound shall/will ever pray  
(or variation thereof)
e.g. I remain, so no more 
from
Self-reference your petitioner e.g. your humble servant
4.2 A continuum of styles
As King/Jones (2015: 73–4) suggest, there was a comprehensive shift in pauper let-
ters from petition style to letter style as the preferred form of epistolary expression 
in the later eighteenth and early nineteenth centuries. This section seeks to verify this 
observation and explore inter-writer variation in pauper letters from a diachronic per-
spective, considering to what extent these letters contain petition-style or letter-style 
features. The features under consideration are voice, greeting, petition-element, closing 
formula and self-reference (Table 5 in Section 4.1).
How many pauper letters in the Diachronic Corpus contain between 0 and 5 peti-
tion-style features is displayed in Figure 2 (overall proportions) and Table 6 (detailed 
diachronic perspective). Although the dataset for the eighteenth century is compara-
tively small and the results need to be treated with caution, three broader tendencies can 
be observed. Firstly, relief applications which do not rely on any petition-style features 
at all are the most dominant. Overall, merely 12 out of 139 pauper letters (8.63%) show 
more than one petition feature. Secondly, the proportion of letters containing petition-
style features is higher in the eighteenth century (c. 41% in both Subperiods I and II) 
than in the nineteenth century (21% in Subperiod III). Thirdly, there is a decrease in 
the number of petition features used in pauper letters over time. In the early eighteenth 
century 22.73% of all applications contain at least two petition features. This propor-
tion appears to drop towards the end of the century (11.76%) and diminishes even 
further to 5% in the early nineteenth century. The only application boasting all five 
petition features under investigation dates from the first subperiod (1_Nason_1758). 
These findings, although based on relatively few letters, seem to confirm the observa-
tion by King/Jones (2015) concerning a shift in writing style, but suggests that it could 
be a gradual development. This may have been encouraged by the change in legislation 
at the end of the eighteenth century and is more fully in evidence owing to the increase 
in letter frequency (and therefore source material) in the earlier nineteenth century, as 
described in Section 1. 
12 This term is taken from Włodarczyk (2013: 203f.).
13 A double forward slash // indicates a line break.
314
Figure 2. Proportion of pauper letters containing 0–5 petition features (Diachronic Corpus)
Table 6. Number of pauper letters containing 0–5 petition features by subperiod (Diachronic 
Corpus)
Number of petition features
Subperiod 5 4 3 2 1 0
I (1730–1759) 1 0 0 4 4 13
II (1774–1799) 0 0 1 1 5 10
III (1807–1834) 0 1 2 2 16 79
Total 1 1 3 7 25 102
Proportion 0.72% 0.72% 2.16% 5.04% 17.99% 73.38%
petition style               letter style
Figure 3. Continuum of styles in pauper letters
The inter-writer variation in pauper letters regarding the use of petition-style and/or 
letter-style features can perhaps best be described as a continuum of style, with petition 
style at one end of the pole and letter style at the other (Figure 3). A relief application 
containing all five petition features is situated on the petition-style end of the pole, 
and in opposition a missive without any petition features on the letter-style end of the 
pole. The more petition features a relief application contains, the further left on the 
continuum it is positioned.
 
0%
10%
20%
30%
40%
50%
60%
70%
80%
5 4 3 2 1 0
315
A model based on a cline is also able to represent relief applications where both 
petition style and letter style are employed within a single feature. This is most com-
mon in closing formulae (10x), followed by the features greeting and voice (both 6x) 
and self-reference (5x).14 Example (4) illustrates how one writer, Sary Young, uses 
both styles sequentially at the end of the letter. She effectively closes the application 
twice, first in petition style with the self-reference “your poor {patitioner} pertisoner” 
and the closing formula “will all ways be bound to pray For you all”, then in let-
ter style with the closing formula “[I] am” and the self-reference “your most Humble 
Sarvant to Command”.15 On three occasions the petition-style formula based on in duty 
bound appears in non-conventional places, twice in the running text (3_Elkes_1817, 3_
Wood_1827) and once after a signature (1_Jones_1730); in all cases letter-style closing 
formulae are employed before the final self-reference. Petition style and letter style can 
also occur together, as in the self-reference in (5) where “petitioner” and “parishioner” 
are both linked by a conjunction and jointly pre-modified by “your humble”. Example 
(6) exemplifies a combination of styles in the greeting, with the petition-style heading 
“To the Overseers of Shippon Mallard” followed by letter-style “Sir”.
(4) your poor {patitioner} pertisoner will all ways be bound to pray For you all 
& am your most Humble Sarvant to Command (1_Young_1755)
(5) your humble petitioner and parishsoner (1_Cross_1755)
(6) To the Overseers of Shippon Mallard Sir (3_Sheppard_1829)
A switch in voice from third to first person (petition style to letter style) is presented 
in (7) and (8), and a change from first to third person in (9). The names mentioned at the 
beginning of the applications in (7–9) appear again in the signatures, underlining that 
all applicants refer to themselves in both the first and the third person. Such switches 
in voice, while infrequent, have also been observed by Calvo Cortés (2020: 200) in 
petitions to the Foundling Hospital and the Bank of England written between 1785 and 
1815. Here the switches were from third to first person, which Calvo Cortés attributes 
to the introduction of personal circumstances. For 1820 settler petitions Włodarczyk 
(2013: 215) also records changes in voice, those from third to first person coinciding 
with a switch in social roles (e.g. from petitioner to mother or wife) and those from 
first to third person associated with closing formulae, triggered by formulaic phrases. 
Formulaic phrases and by extension particular words, as well as structural aspects, also 
seem to have triggered the switches in voice in the Diachronic Corpus, where they are 
present in both the beginning (7, 8) and at the end of letters (9). The switches do not 
appear to be content-related, but in three cases there is a switch from third to first person 
after a letter-element clause containing the verb or noun petition (e.g. as a verb in (7)), 
and one switch is located at the end of the letter with the introduction of a letter-style 
14 The petition-element is excluded from this list as this is a feature which is either present or absent and has no 
letter-style equivalent.
15 Curly brackets {} are used to indicate uncertainty in the transcription. The self-correction will be discussed in 
Section 5.4.
316
closing formula (8), adding a more personal dimension to the application through ego 
and topic involvement in “I hope”. Conversely, in (9) the letter is opened with an epis-
tolary greeting and formula “I have to inform you that”, while for the remainder the 
petition-style third-person voice is maintained. A final example of an unconventional 
and only brief switch comes from a non-authentic letter where a run-on formula de-
veloping from the previous clause (cp. Bannet 2005: 65) introduced by who triggers 
switch to the third person, which is then not followed through, leading to technically 
incorrect agreement in first-person am (10).
(7) Elizabeth Patchett in the City Of Worcester Petions you being in great 
distreſs and Oppreſd with a Large family, Craves your Aſstance for Relief as 
it’s not my Wishes to Come with all my family troublesome Wholly on ye 
[Parish] (2_Patchett_1777N)16
(8) this Sheweth that Jane Wildman is at this time in verry Great destreſs […] I 
hope to heare from you I reman […] (3_Wildman_1826)
(9) Dear Sir I have to inform you that Sarah White your parisher are So bad in 
health that […] (3_White_1831)
(10) I am Sorry that […] a Distrest widow, Who am with Duty and humble 
Submiſsion (2_Keely_1799N)
What reasons could lie behind the general dominance of letter-style writing, the de-
crease in petition-style features over time, as well as the mixing of both petition and 
letter styles within individual applications? Paupers seem to have been aware that by ap-
plying for relief they would enter into an epistolary relationship with officials from their 
home parish which would allow them to negotiate relief and re-apply through continued 
correspondence. At the same time some applicants at least must also have been aware 
that petitions as a text type existed as well, which were directed at other institutions 
for different reasons and characterised by a more formal writing style. The diachronic 
decrease in petition-style features could perhaps, in part, be the result of a dwindling 
number of pauper letters based on the petition style which could potentially serve as 
a model for new applicants, with lower-level community knowledge of this form of 
writing receding over time. Considering the significant numbers of applications in the 
nineteenth century, it is also possible that word spread that it was not necessary to adopt 
petition-style features in order to be granted poor relief. However, those writers who did 
emulate features of petition style were able to increase the formality of their writing, 
highlighting the social distance and unequal status between applicant (supplicant) and 
parish official, conveying their respect and deference for those higher up in the social 
hierarchy who had the power to decide over their fate. With this strategy these applicants 
probably hoped to better achieve their communicative goal of securing financial support. 
Another strategy for indexing formality is to use the label petition in an application, 
which occurs only in five letters (for the use of petition as a verb see (7)). While (11) 
could be interpreted as a shortened and modified version of the petition-element (but 
16 Examples from non-authentic letters are marked with a superscript N after the filename.
317
is not counted as such), the meaning of “Pettition” in (12) is more general. Rather than 
referring to a specific type of document used in the context of courts, however, these 
writers likely relied on the broader meaning of ‘a formal written request or supplica-
tion […] appealing to an individual or group in authority (as a sovereign, legislature, 
administrative body, etc.) for some favour, right, or mercy, or in respect of a particular 
cause’ (OED Online). 
(11) this with a humble petition to you (1_Robinson_1750)
(12) my Pettition is that you will only Relieve me a Little from the Severity of the 
Hard Winter (2_Williamson_1799 N)
Interestingly, despite the general decrease in the use of petition-style features to-
wards the nineteenth century, it is not the case that particular petition features fall out 
of use entirely, since all of them are attested in all subperiods. Rather, it seems to be 
a matter of individual choice which petition-style feature is employed, regardless of 
whether the letter represents the writing of the lower classes (authentic) or of the better-
educated members of society (non-authentic). The findings in this section contradict the 
seemingly too sweeping claim made by Jones/King (2015: 73–4) concerning a com-
prehensive shift in style since letter-style writing is shown to be dominant already in 
the earliest pauper letters. In conclusion, from a text-typological perspective poor-relief 
applications constitute not a type of petition, but a type of letter. The following section 
explores inter- and intra-writer variation in more detail, including differences between 
authentic and non-authentic letters, and focuses on the two features where variation is 
most extensive, i.e. in closing formulae and self-references. 
5 VARIATION IN CLOSING FORMULAE AND SELF-REFERENCES
This section is concerned with variation observable in closing formulae (Section 5.1) 
and self-references (Section 5.2) in letters by paupers from the Diachronic Corpus, 
and by way of comparison in letters by advocates and parish officials from the earlier 
eighteenth century (Supplement). This extended perspective on inter-writer variation 
within pauper letters and between the different groups of writers will be complemented 
in Section 5.3 by an analysis of intra-writer variation in pauper letters from the Indi-
vidual Corpus. Self-corrections occurring in closing formulae and self-references offer 
additional support for findings concerning the acquisition and processing of formulaic 
sequences (for a definition of the term see Timmis 2020: 100) and will be examined in 
Section 5.4. 
5.1 Inter-writer variation in closing formulae
In the Diachronic Corpus of pauper letters, 133 instances of closing formulae can be 
observed, occasionally occurring in combination, while 14 letters contain no formula 
at all (Figure 4). The two most frequent formulae overall, based on remain (13) and am 
(see (4)), do not become dominant until the nineteenth century. By this period remain 
has become associated with business correspondence (Austin 1973: 131, Dollinger 
318
2008: 282, Shvanyukova 2020: 96–7; see also Dossena 2006 for examples), and was 
also observed to be the most common closing formula in a larger corpus of pauper 
letters (Timmis 2020: 100). The formula based on am was already commonly used by 
educated writers in the early eighteenth century (Austin 1973: 131), and is also attested 
in later business correspondence where it “focuses the addressee’s attention on […] 
the explicit presence of the writer” (Shvanyukova 2020: 91). Considering the official 
nature of poor relief correspondence and, viz. am, the need for paupers to be ‘seen’ in 
their distress, the two most frequent closing formulae seem appropriate choices in this 
context. Less frequent, closing formulae based on oblige (14) are also attested in busi-
ness writing (Dollinger 2008: 272–273, Shvanyukova 2020: 91).
(13) iremain you afflickted Servn{t} (1_Rumbell_173X)
(14) and you will much oblige your humble petitioner and parishioner (1_
Cross_1755)
Figure 4. Closing formulae in authentic pauper letters (Diachronic Corpus)
The formulae from, no more and conclude, on the other hand, were firmly associ-
ated with the lower classes in the eighteenth century (Austin 1973: 119, 130–1).17 The 
fact that the paupers represented in the Diachronic Corpus continue to employ these 
in the nineteenth century speaks in favour of a generational transmission of episto-
lary conventions among the labouring poor. At the same time these closing formu-
lae, proportionally especially no more (15), remain less common than the two most 
17 When from follows no more, as in (15), this is counted as part of the realisation of the formula no more. The 
formula could also be extended, for instance by adding of so before and at present afterwards, as illustrated in 
(17).
 
0 5 10 15 20 25 30 35
remain
from
am
no more
bound to pray
oblige
conclude
Other
None
I (1730–1759) II (1774–1799) III (1807–1834)
319
frequent choices, which shows that the poor were also able to acquire knowledge of 
alternative and likely more appropriate forms. Alongside from, the petition-style clos-
ing formula based on bound to pray (see (4)) is proportionally the most frequent in the 
earlier eighteenth century where half of all occurrences of this formula can be found. 
This highlights the decline of petition-style features after this subperiod. Sokoll (2001: 
59) notes that there are only twelve occurrences of bound to pray in his collection of 
758 pauper letters from Essex. Timmis (2020: 110) similarly observes only six bound 
to pray formulae (as opposed to 134 examples of remain) in his pauper letter corpus, 
but intriguingly a much larger proportion in a collection of coeval prisoner letters; the 
formula is also a conspicuous feature in petitions to the Foundling Hospital (Calvo Cor-
tés 2020: 207). Such findings reveal the stylistic variation observable across different 
pleading genres addressed to different institutions and indicate that the closing formula 
bound to pray was upheld longer in petitions proper than in poor-relief applications. 
Formulae occurring only once in the Diachronic Corpus are subsumed under ‘Other’ in 
Figures 4 and 5 and include, amongst others, by (2_Patchett_1777N), run-on formulae 
and phrases with an imperative, as in (16), which is again a strategy also documented 
in business letters (Shvanyukova 2020: 91).
(15) no mor from A poour poper (1_Jones_1730)
(16) except this my humble patition and believe me to be your very hum{le} 
servent (3_Stagg_1807)
Figure 5. Closing formulae in non-authentic pauper letters (Diachronic Corpus)
 
0 2 4 6 8 10
am
oblige
bound to pray
remain
from
no more
Other
None
I (1730–1759) II (1774–1799) III (1807–1834)
320
Although only few non-authentic letters are included in the corpus, what is striking 
is the preference here for am already in the earlier eighteenth century (e.g. (4); Figure 5). 
Trained encoders of pauper letters seem to have been more aware than untrained writers 
of the general appropriateness of this closing formula in a formal context. Corroborating 
this finding, the same preference is shown during this period in letters by those writing 
on behalf of paupers, whether parish officials addressing their peers (5x) or other advo-
cates (3x), closely followed by not including any closing formulae at all (officials 4x, 
advocates 1x). The only instance of ‘lower-class’ no more in the Supplement appears in 
a letter by an advocate from 1758, fittingly by a servant (Merey Willioms), who like the 
pauper Catherine Jones she writes for (both as advocate and as encoder) also belongs to 
the group of lower-class writers. 
5.2 Inter-writer variation in self-references
The most common self-reference in the diachronic pauper letter corpus is a noun pre-
ceded by the pronoun your and an adjective. In keeping with epistolary traditions of the 
time, servant is by far the most frequent noun (Figures 6 and 7), expressing deference 
and social distance, particularly in combination with your humble (see (4)). However, 
petitioner and parishioner also make a noticeable appearance, especially in the earlier 
eighteenth century, sometimes in combination (see (14)). These nouns are particularly 
suitable choices in pauper letters in that they allow the writer to draw the attention of 
the addressee to their social roles of formal applicant and member of their home parish, 
which entitles them to apply for relief and also instils a sense of responsibility to act 
in the recipient. The connections between applicant and parish official is strengthened 
by the pronoun your, which according to King (2019: 33) conveys “embodied belong-
ing”. The noun pauper, employed in two letters (see e.g. (15)), places the focus on the 
financial distress of the applicant. Supplicant expresses a social role akin to petitioner, 
perhaps with a higher level of humility, but unlike petition and petitioner the Latin 
borrowing was probably not part of the register-specific vocabulary pertaining to poor-
relief applications which the poor would be familiar with – in the Diachronic Corpus 
at least supplicant only occurs in a letter encoded by a well-educated hand (Figure 7).
Again unsurprisingly, in all subperiods the most frequent adjective used in self-ref-
erences is humble (Figures 8 and 9), sometimes preceded by the superlative most (8x; 
see (4)) or the intensifier very (3x), and five times occurring in combination with obedi-
ent (cp. also Timmis 2020: 109). In the eighteenth century, obedient was only used in 
non-authentic letters by well-educated writers, and it was not until the nineteenth cen-
tury that less-educated paupers followed suit. During this period, humble and obedient 
are also regularly encountered in business correspondence, as are expressions of obli-
gation like obliged and obliging, and the semantically bleached intensifier truly, which 
aptly occurs in a non-authentic letter (Del Lungo Caminciotti 2006: 161, Shvanyukova 
2020: 93). Afflicted, poor, distressed and unfortunate allude to the struggles and re-
duced circumstances of the paupers. In (15) poor pre-modifies pauper, emphasising 
321
the financial difficulties.18 While afflicted can refer to both ‘distress’ and health issues 
(OED Online), no semantic preference can be detected in the pauper letters. Like af-
flicted, the more unusual devoted only appears (after a reflection on suffering and the 
transience of life) in a letter deemed authentic, but it was written by a hand on the upper 
hand of the spectrum with more advanced training, and epistolary phraseology can be 
found throughout the application.
Figure 6. Self-referential nouns in authentic pauper letters (Diachronic Corpus)
Figure 7. Self-referential nouns in non-authentic pauper letters (Diachronic Corpus)
18 This is one of only two instances in the Diachronic Corpus (both in Subperiod I) where an indefinite pronoun is 
used in place of your.
 
0 10 20 30 40 50 60 70 80
servant
petitioner
parishioner
pauper
Indeterminate
I (1730–1759) II (1774–1799) III (1807–1834)
 
0 2 4 6 8 10 12 14 16
servant
petitioner
parishioner
supplicant
I (1730–1759) II (1774–1799) III (1807–1834)
322
Figure 8. Self-referential adjectives in authentic pauper letters (Diachronic Corpus) 
Figure 9. Self-referential adjectives in non-authentic pauper letters (Diachronic Corpus)
Although a self-referential type based on your humble servant is most common, 
some writers do not fill the adjective slot (21x) or the noun slot (5x), and one writer 
simply concludes with your (2_Rosthron_1779N). Yours is used in only 14 letters and 
in all except two instances is followed by additional elements such as adverbs or 
phrases, with a potential diachronic trend towards yours respectfully, although fre-
quencies are low overall. Infrequent already in the early eighteenth century, till death 
 
0 10 20 30 40 50 60 70 80
humble
obedient
afflicted
poor
obliged
distressed
obliging
devoted
unfortunate
I (1730–1759) II (1774–1799) III (1807–1834)
 
0 2 4 6 8 10 12 14
humble
obedient
distressed
poor
obliged
faithful
I (1730–1759) II (1774–1799) III (1807–1834)
323
was associated with the lower classes (Austin 1973: 135–6) and correspondingly only 
occurs once in an authentic letter in the earliest subperiod of the Diachronic Corpus 
(17). Also employed as a post-modifier of servant on three occasions in the earlier 
eighteenth century (see e.g. (3) and (4)), the time-honoured formula to command (for 
examples from the fifteenth to seventeenth centuries see Nevala 2004: 95, 145, 153) 
seems to fall out of use towards the nineteenth century, with two attestations each in 
the earlier and later eighteenth century. In non-authentic letters, instead, in distress 
occurs once in the later eighteenth century, the former emphasising the pauper’s situa-
tion (cp. distressed above), and &c (‘etc.’) once each in Subperiods II and III. The ab-
breviation &c is used to represent the final self-reference, which according to Dossena 
(2010: 288–9) points “to the degree of codification that these formulae had achieved: 
so fixed and invariable were the formulae, that the encoder did not even need to note 
them down”. In authentic letters from the nineteenth century, &c is used once as well, 
alongside one instance of respectably and four of respectfully (cp. Timmis 2020: 110), 
which index social distance and respect. Overall, seven letters (all authentic) contain 
no self-reference formula at all, which means the applicants forgo a final chance in 
their letter to display respect, deference and submission (cp. King 2019: 235, 345) 
towards the recipient, which could have increased their chances of success. Four of 
these applicants also omit the closing formulae, and it seems probable that they were 
not aware that these were conventionally expected in letters, nor that by engaging with 
this type of humiliative discourse they would signal their adherence to the “shared set 
of norms and values” (Shvanyukova 2020: 99; see also Timmis 2020: 102–3) under-
pinning the poor-relief system.
(17) So no more at presan But I Remain Yover abl[^i OVERWRITES e^]ging 
Saruant til Deth (1_ChappellH_1741)
With respect to self-references, letters by advocates and officials from the earlier 
eighteenth century (Supplement) once again mirror majority usage patterns. They rely 
almost entirely on the formulaic sequence your (most) humble Servant or drop the 
self-reference altogether, in three of five cases the letter starting with a petition-style 
heading instead of a letter-style salutation. Once a doctor, John Lodwick, writing for 
“Perriſioner Roice” in 1741 adds “to Command” after the self-reference, in keeping 
with the temporal trajectory of this phrase indicated by the Diachronic Corpus.
To conclude, variability in closing formulae and self-references is much higher in 
pauper letters than in missives by advocates and parish officials. Much of the variation 
reflects the need for applicants to invest in their identity construction, outlining their 
reduced circumstances, as well as specifying their social roles and their relationship 
with the recipient. On the other hand, this can also be the result of the limited schooling 
received by applicants who, unlike parish officers and the more educated advocates, 
might not be sufficiently aware of existing epistolary norms. This will be explored fur-
ther in the next two subsections.
324
5.3 Intra-writer variation in closing formulae and self-references
The pauper letters in the Individual Corpus cast an interesting light on the inter-writer 
variation observed previously: paupers writing their own letters hardly wavered in their 
linguistic choices. In his six letters Morgan always concludes with the no more formula 
as in (18), three times without the self-reference a poor before his signature. Green is 
consistent in her use of I remain your humble servant across seven letters (e.g. (19)), 
even if the spelling is somewhat variable, only once omitting both formulaic sequences. 
Soundy unfailingly closes with the bound to pray formula in her twenty letters as in 
(20), variation occurring only in spelling and word order. Also writing on behalf of fam-
ily members, her preferred self-reference is your parishioner(s) (18x), your petitioner 
occurring only twice. In merely three letters she fills the adjective slot with unfortunate 
(1x) and distressed (2x), and once the self-reference is lacking; the second and third op-
tions do not deviate from typical usage patterns described in the previous section. 
(18) So no mour fro{m} a pour (3_Morgan_1805)
(19) I Rame your Houmble Servant (3_Green_1820_1)
(20) your Perrishoners In duty bound will ever Pray (3_Soundy_1823_2)
In non-autographical letters there is a much higher degree of variation. For instance, 
each of the four letters encoded for Hammont has a different closing formula and self-
reference, and each hand involved with the letters by Hughes and Kingston has their 
own stylistic preference. In line with autographical applications, however, the choices 
are fairly invariable within the sets encoded by particular hands when deemed authen-
tic, i.e. representative of someone with limited schooling. This can be seen, firstly, in 
the first six letters encoded for Hughes, all in the same hand, where the closing formula 
is always based on remain and the self-reference on your humble servant; and sec-
ondly, in the four letters encoded by the same ‘authentic’ hand for Kingston, where the 
two features are never realised. 
The three non-authentic letters in the non-autographical set yield additional insights. 
The final two letters encoded for Hughes were prepared by a trained hand, and although 
the self-reference remains the same (your obedient Servant), there is stylistic variation 
in the closing formula (am vs. run-on formula). As Auer (2015: 155) notes “[t]he better 
the schooling and the writing practice, the greater the stylistic variation will be”, which 
would apply here as well. The third of the five Kingston letters was equally penned by 
a well-educated hand; its rendering of closing formula and self-reference as “I Remain 
yours Respectfully” (3_Kingston_1821_2N) offers a stark contrast to the four authen-
tic letters (all by the same untrained hand) where the formulae are absent and the text 
abruptly closes with the signature. This particular case is interesting because the letter 
by the trained hand is the third of the five Kingston letters and positioned in the middle 
of a sequence of three letters sent between September and December 1821, with the 
earlier and later letters encoded by the uncoached hand. This indicates that Kingston, 
who could not even sign his name (Section 3.2), was probably not aware of the expres-
sions used in the letter by the trained hand and also did not pass on any formulations to 
325
the less experienced encoder. The stylistic variation in the set of letters sent by Hughes, 
who also could not sign her name, suggests that she was also not involved linguisti-
cally in the composition of the letters, at least not concerning the closing formulae. 
Otherwise, considering her level of schooling and the evidence from the autographical 
letters, the formulaic phrases would be more homogeneous. The cases of Kingston and 
Hughes present counterexamples, on a local level, to the “shared linguistic register” 
stipulated by Jones/King (2015: 69) and provide proof for the final scenario outlined by 
Sokoll (2001: 66) who proposes “a broad spectrum of possibilities, with some paupers 
dictating their letters, while others had them written without much say in their composi-
tion and still others where the sender dispatched them without ever knowing what had 
been set down in his or her name”. 
The evidence from autographical letters, and sets of letters by authentic hands among 
the non-autographical applications, strongly supports earlier findings on epistolary for-
mulae in writers with limited schooling, also including members of the lower classes 
from other European countries. The repertoire of formulae is relatively small and fixed 
(Elspaß 2005: 163, 172, 192), with formulaic phrases being stored as a single unit in the 
mental lexicon and retrieved as a whole during the writing process (Elspaß 2005: 170, see 
Timmis 2020: 97–100 for an extensive discussion). Variant spellings observed in clos-
ing formulae and self-references by paupers suggest that the writers probably acquired 
knowledge of formulaic sequences by hearing them rather than seeing them in writing 
(unless they happened to take a letter with non-standard spelling as a model), and they 
rendered the formulae in writing according to the spelling principles they were familiar 
with (cp. Fairman 2008: 206–207; Allen 2015: 211, quoted in Timmis 2020: 101). 
Limited schooling and incidental rather than comprehensive (oral) transmission of 
stylistic norms could also explain why pauper letters do on occasion contain petition-
style features, but why different applicants focus on (a varying number of) different 
features. Furthermore, as a result of an incomplete acquisition of epistolary norms pau-
pers may not have been aware that closing formulae like from, no more and conclude, 
or the addition of till death, had become associated with lower-class language and 
would not be used by the officials they corresponded with (Sections 5.1 and 5.2), or 
that a switch from first to third person after a letter-style element would be considered 
unconventional as well (Section 4). While the letters examined in this subsection date 
to the nineteenth century, the findings on the production circumstances of later pauper 
letters and the influence of limited education on the stylistic repertoire of the writers 
can likely also be applied to the eighteenth century, when there were even fewer insti-
tutional schooling opportunities for the labouring poor. 
5.4 Self-corrections in closing formulae and self-references
A number of self-corrections in pauper letters offer additional support for the hypoth-
esis that oral transmission was an important pathway for the acquisition of epistolary 
formulae. In closing formulae and self-references in the Diachronic and Individual Cor-
pora there are 24 corrections, 18 of which affect the spelling of individual words. 7 self-
corrections reveal struggles with the spelling of the verb remain in a closing formula, 
326
and the adjectives humble (2x) and obliging, as well the nouns servant, petitioner and 
parishioner (1x each) in self-references. Uncertainty concerning the correct vowel rep-
resentation, for instance, can be seen in exchanged vowels in “abl[^i OVERWRITES 
e^]ging” (1_ChappellH_1741), “H[^o OVERWRITES u^]mble” (2_Keeling_1788) and 
“pr[^ea OVERWRITES i^]shener” (3_Wall_1821). Such variant spellings can repre-
sent reflections of speech (Auer et al. 2033, Gardner et al. 2022, Gardner 2023b). Par-
ticularly revealing in this context is the self-correction in (4) where Young struggles 
with the standard spelling of petitioner, deleting her first attempt “patitioner” and opt-
ing for “pertisoner” instead with insertion of non-etymological /r/. This suggests that 
she had a rhotic accent and spelt phonemically (Fairman 2008: 206) owing to limited 
schooling and lack of written models. Rhoticity is likewise evidenced in other writers, 
for instance in the spelling “Pertishoner(s)” (3_Soundy_1823_1; 3_Soundy_1827_1; 
3_Soundy_1828_3) by Frances Soundy, who also wrote to a home parish in Berk-
shire, her likely place of origin, an area in which r-colouring was still attested in the 
mid-twentieth century (Upton/Widdowson 2006: 42–3). 11 further modifications be-
tray lapses in concentration on the word level when a writer thought ahead and had 
to supply a character that was omitted (9x), e.g. with an insertion in “Disttresed” (1_
Camp_1759), or when a repeated character is deleted by overwriting (2x), e.g. “Se[^r 
OVERWRITES e^]vant” (3_Spencer_1815). On the sentence level we find similar con-
centration lapses or mechanical mistakes (following Fairman 2008: 199) which are due 
to the (partial) repetition of a previously written word (1x) or thinking ahead (2x). In 
(21) the writer started with “y[our]”, part of the self-reference, but realised “from” was 
missing from the closing formula, and in (22) the verb “Will” originally appears before 
the self-reference is complete. With the petition-style closing formula duty bound vari-
ation in the word order is possible, but in twenty letters Soundy is relatively invariable 
(Section 5.3) and diverts from her usual order “self-reference + in duty bound will ever 
pray” (see (20)) merely three times, with “in duty bound” appearing either before the 
self-reference (1x) or between “will” and “ever pray” (2x). With the self-correction in 
(22) Soundy establishes her preferred word order.
(21) so no mor [^{y} RUBBED OUT^] from your […] (3_Bryan_1829)
(22) your unforchenate [^Will CROSSED OUT^] Perrishoner in duty bound will 
Ever Pray (3_Soundy_1818_1)
Only three changes register beyond the level of spelling and mechanical errors. In 
an undated letter (3_Morgan_18XX), Morgan deletes an indefinite pronoun from the 
self-reference slot, presumably the start of a poor, switching to his alternative strategy 
of not supplying any self-reference at the end of the letter (Section 5.3). However, as 
previously noted, there is no indication from the contents of his letters as to when he pre-
fers which option, and there is no change in preferences over time either. In a letter sent 
by Hughes, “Servt” is added above the line after “your Hunble” (3_Hughes_1829_2), 
which could mean a deliberate change from a self-reference without a noun, but most 
likely represents an omission since this particular hand writes your humble servant in 
327
the self-reference in six different letters for this applicant. Fixedness in the applica-
tion of epistolary formula probably also accounts for the change in a different letter 
by the same hand from “yours” to “your” after “i Remain” (3_Hughes_1829_4). It is 
noteworthy that no correction was made in closing formulae or self-references with the 
objective of switching from letter to petition style, or vice versa.
In total, 22 of 189 pauper letters investigated are affected by changes to closing 
formulae and self-references. While this seems to be a comparatively low proportion, 
especially in light of Fairman (2008: 198) finding 490 decipherable strike-throughs in 
c. 1,600 pauper letters, further research is required to determine to what extent epis-
tolary formulae are generally subject to modification during the writing process. The 
relatively low number of self-corrections in the present study is suggestive of the fixed-
ness of formulaic sequences and supports the theory that they are stored and retrieved 
as a whole from the mental lexicon (Section 5.3). That only three changes rectify me-
chanical errors on the sentence level – even though Fairman (2008: 208) notes that in 
his corpus such corrections owing to lapses in concentration (‘slips of the pen’, i.e. his 
categories ‘jump’, ‘echo’ and ‘repeat’) are roughly as frequent as spelling modifications 
– could signify that the effort in retrieval of formulaic sequences is fairly minimal, but 
also that when drawing on these pre-fabricated chunks the mental load is lighter and 
the writers are less easily distracted than when they are required to freely compose new 
sentences. Within formulaic sequences lapses in concentration manifest themselves 
most commonly at the word level, as shown by the insertion of omitted characters and 
the deletion of repeated ones. 
As a final point of note, self-corrections in closing formulae and self-references 
were only found in authentic letters, which again highlights the role of education in 
the acquisition of epistolary formulae. Better-trained writers will either have been able 
to write an error-free letter straightaway or, provided they had sufficient means, they 
prepared a clean copy from a draft, being aware of the notion that self-corrections were 
considered impolite and carried a social stigma. As early as 1756 (and possibly earlier, 
if earlier editions can be uncovered), self-corrections were described in a letter-writing 
manual as ‘not only a Reflection on the writer, but a Rudeneſs to the Perſon to whom 
they are written’ (Anonymous 1756; on draft writing see also Gardner 2018, 2023a).
6 CONCLUDING REMARKS
Through the study of five text-constitutive features this article has shown empirically 
that relief applications written under the Old Poor Law represent letters rather than peti-
tions in terms of form and style, and for the first time offers a text-typological account 
of pauper letters as a subtype of correspondence which shows stylistic similarities not 
only with the familiar letter and petitions, but also with business and official corre-
spondence. Within the text type of pauper letters there is significant variation, and a 
continuum of style can be observed ranging from letter style to petition style. However, 
in the analysed dataset letter-style writing was already the dominant form of expres-
sion in the earliest poor-relief applications. A low proportion of applicants does adopt 
features typical of petitions, yet mostly only one or two of the five identified here, and 
328
the usage of petition-style features decreases already in the eighteenth century. These 
observations contradict and offer a more nuanced view on the observation by Jones/
King (2015: 73–4) that a shift in stylistic preference took place towards the nineteenth 
century only. Although the dataset on which the present study is based is compara-
tively small, owing to the scarce survival of pauper letters from the eighteenth century 
and limited source availability, the article has uncovered interesting and suggestive 
diachronic trajectories which merit exploring further in a more extensive collection 
of letters, but find support in related research, as noted in the earlier sections, and also 
in Sokoll’s study on Essex pauper letters. In a much larger dataset of 758 letters he 
unearthed only one pauper letter clearly modelled on a petition, written on behalf of 
Ann Marsh in 1824 by a professional scribe (Sokoll 2001: 48, 193–4). The findings 
presented in this article corroborate Sokoll’s statement that “it cannot be emphasized 
too strongly that in stylistic terms and from their overall scriptual habitus, most pauper 
letters do not normally follow the contemporary model of the formal petition” (2001: 
59, original emphasis). 
While inter-writer variation across pauper letters is extensive regarding closing for-
mulae and self-references, letters by parish overseers and advocates exhibit significantly 
less variation and, similar to letters by the better educated (which includes non-authen-
tic pauper letters), reveal a noticeable affinity with formulaic language evidenced in 
business correspondence.19 The findings on intra-writer variation and self-corrections 
in pauper letters suggest that individuals typically acquired only one formulaic expres-
sion each for the slots of closing formula and self-reference, often only through oral 
transmission. Owing to limited schooling opportunities, stylistic conventions were not 
always learned entirely successfully as evidenced by self-corrections affecting spell-
ing, the partial adoption of petition-style features and their occasional application in 
unusual positions.
Although their literacy and epistolary skills may have been limited, the poor-relief 
applicants nevertheless were competent enough to be able to successfully engage with 
parish officials in correspondence in an attempt to alleviate their distress. The evidence 
from pauper letters, particularly concerning the lack of intra-writer variation and the 
types of self-corrections made, is significant for our understanding of the social practice 
of letter-writing, as well as how the lower classes acquired literacy and to what degree. 
The labouring poor emerge as resourceful individuals who penned letters themselves 
or identified individuals in their social circle who could act as scribes for them (see also 
Sokoll 2001: 65, King 2019: 36). This could support Vincent’s findings concerning 
household literacy in the nineteenth century, which signifies that at least one member 
of the household is literate; on the basis of signature literacy documented in marriage 
registers he determines that in the mid-nineteenth century, i.e. a generation after the lat-
est pauper letter investigated in the present paper and a generation before compulsory 
elementary education was introduced, “there was literacy in 75% of the new homes” 
(2014: 275). The concept of household literacy is equally relevant for the earlier periods 
19 Inter-writer variation in pauper letters with respect to orthographical and phonological features has been 
examined in e.g. Gardner et al. (2022) and Auer et al. (2023).
329
investigated in this article, and in fact preceding centuries. Following Hailwood (2023), 
already in the period 1550–1700 the literacy skills of the lower classes should not be 
underestimated. Examining whether witnesses left simple strokes, meaningful marks 
or symbols, initials or signatures in depositions, he argues that gradations between il-
literacy and literacy should be recognised, finding that 
[a]rtisans, husbandmen and labourers are all groups that are considered to be 
overwhelmingly illiterate using traditional signature studies techniques, but here 
they emerge as subsections of society in which between 50 and 60 per cent of in-
dividuals demonstrated more than the most basic writing skills. They could form 
meaningful and recognizable letters or symbols with a pen in hand. (Hailwood 
2003: 59) 
Furthermore, Hailwood (2003): 61) maintains that “even a minimal amount of 
formal schooling could have been sufficient for an individual to acquire ‘letteracy’: 
the ability to identify individual letters, if not much more”. The writing competencies 
observed in pauper letters from the eighteenth and nineteenth centuries are testament 
to the continued resourcefulness of the lower classes and their access to foundational 
education at the very least.
The findings of this article and limitations in the amount and type of data analysed 
have also created new research questions and opened up new avenues for future studies. 
For instance, why the already infrequent recourse to petition-style features decreased 
even further over time merits further consideration. After all, these features could use-
fully be employed to increase the formality of their writing and to show respect towards 
the socially distant addressees, i.e. the overseers and other parish officials. Perhaps 
the petition as a model became less popular or available, or, more likely, letters and 
epistolary style became even more firmly established by convention as the appropriate 
mode for poor-relief applications. Such applications initiated correspondence with par-
ish officials, in some cases spanning many years, which the poor used to negotiate their 
case and plead for (continued) support (Sokoll 2001, 2008). It would also be interest-
ing to adopt the framework and methodology by Biber/Finegan (1989) and determine 
whether the decrease of petition-style features in pauper letters can be linked to the 
drift towards a more oral style which the two authors observed in letters more generally 
(1989: 515). 
For a fuller perspective on these changes the period of investigation needs to be 
extended back into the seventeenth and sixteenth centuries as well, requiring extensive 
archival work (and luck) in order to uncover relevant material from the labouring poor. 
Further research is also necessary to establish the potential stylistic impact of business 
and official correspondence on pauper letters, and to determine to what extent overarch-
ing norms for official writing can be identified across different areas of letter-writing. 
A comprehensive diachronic investigation of closing formulae and self-references in 
different types of correspondence from the sixteenth to the nineteenth century would 
help elucidate these questions further and also show when certain specific formulaic 
330
expressions may have become restricted to e.g. lower-class writing over time. Once a 
version with normalised spelling of the LALP corpus of pauper letters is completed, it 
will be possible to conduct a more extensive analysis of stylistic variation, increasing 
the number of features and adopting a stylometric approach, to obtain a more compre-
hensive view of inter- and intra-writer variation in pauper letters, and achieve a more 
detailed text-typological differentiation between pauper letters, petitions, business and 
official correspondence, and the familiar letter. 
All in all, this article aimed to contribute to current debates in historical sociolin-
guistics and has showcased pauper letters as a text type in its own right which help-
fully contributes to the study of language history ‘from below’ by documenting the 
language of members of the lower spectrum of society from whom relatively few lin-
guistic sources have survived and made available so far. The rich findings gained from 
investigating intra-writer variation and self-corrections in pauper letters invite further 
research considering a wider range of linguistic features in order to gain a deeper un-
derstanding of the social context in which pauper letters were written. Considering the 
fact that similarly valuable insights have emerged in previous studies on the language 
of the better-educated (e.g. Gardner 2018, 2023a), analysing intra-writer variation and 
self-corrections in handwritten documents from the past should become standard in 
historical linguistic research.
References
Primary sources
ANONYMOUS (31756) The compleat letter writer; or, new and polite English secre-
tary. London.
BROWN, George (1770?) The New and Complete English Letter-Writer; or, Whole Art 
of General Correspondence. London.
COOKE, Thomas (1770?) The Universal Letter-Writer; or, New Art of Polite Corre-
spondence. London.
Secondary sources
AUER, Anita (2015) “Stylistic variation.” In: Anita Auer/Daniel Schreier/Richard 
Watts (eds), Letter Writing and Language Change. Cambridge: Cambridge Univer-
sity Press, 133–155. 
AUER, Anita/Catharina PEERSMAN/Simon PICKL/Gijsbert RUTTEN/Rik 
VOSTERS (2015) “Historical sociolinguistics: the field and its future.” Journal of 
Historical Sociolinguistics 1/1, 1–12.
AUER, Anita/Anne-Christine GARDNER/Mark ITEN (2023) “Patterns of linguistic 
variation in Late Modern English pauper petitions from Berkshire and Dorset.” In: 
M. Schiegg/J. Huber (eds), 133–156.
AUSTIN, Frances (1973) “Epistolary conventions in the Clift family correspondence.” 
English Studies 54: 129–140.
331
BANNET, Eve Tavor (2005) Empire of Letters: Letter Manuals and Transatlantic 
Correspondence, 1688–1820. Cambridge: Cambridge University Press.
BARTON, David/Nigel HALL (2000) “Introduction.” In: David Barton/Nigel Hall 
(eds), Letter Writing as a Social Practice. Amsterdam: John Benjamins, 1–14. 
BEARDMORE, Carol (2020) “The Marquis of Anglesey: working and social relation-
ships on a Dorset estate (c. 1812–1844).” Rural History 31/2, 135–150. 
BIBER, Douglas/Edward FINEGAN (1989) “Drift and the evolution of English style: 
a history of three genres.” Language 65: 487–517.
BRÖNNIMANN, Stefan/Daniel KRÄMER (2016) “Tambora and the ‘Year Without a 
Summer’ of 1816: a perspective on earth and human systems science.” Geograph-
ica Bernensia G90.
CALVO CORTÉS, Nuria (2020) “Variations from letter-writing manuals: humble 
petitions signed by women in Late Modern London.” In: Andreas H. Jucker/Irma 
Taavitsainen (eds), Manners, Norms and Transgressions in the History of English: 
Literary and Linguistic Approaches. Amsterdam: Benjamins, 183–212.
CRONE, Rosalind (2018) “Educating the labouring poor in nineteenth-century Suf-
folk.” Social History 43/2, 161–185.
CULPEPER, Jonathan/Merja KYTÖ (2010) Early Modern English Dialogues: Spoken 
Interaction as Writing. Cambridge: Cambridge University Press. 
DEL LUNGO CAMICIOTTI, Gabriella (2006) “From Your obedient humble servants 
to Yours faithfully: the negotiation of professional roles in the commercial corre-
spondence of the second half of the nineteenth century.” In: M. Dossena/I. Taavit-
sainen (eds), 153–172.
DILLER, Hans-Jürgen (2001) “Genre in linguistic and related discourses.” In: Hans-
Jürgen Diller/Manfred Görlach (eds), Towards a History of English as a History of 
Genres. Heidelberg: Universitätsverlag Carl Winter, 3–43. 
DOLLINGER, Stefan (2008) “Colonial variation in the Late Modern English business 
letter: ‘periphery and core’ or ‘random variation’?” In: M. Dossena/I. Taavitsainen 
(eds), 257–287.
DOSSENA, Marina (2006) “Forms of self-representation in nineteenth-century busi-
ness letters.” In: M. Dossena/I. Taavitsainen (eds), 173–190.
DOSSENA, Marina (2010) “‘Be pleased to report expressly’: the development of a 
public style in Late Modern English business and official correspondence.” In: Ray-
mond Hickey (ed.), Eighteenth-Century English: Ideology and Change. Cambridge: 
Cambridge University Press, 293–308.
DOSSENA, Marina/Susan FITZMAURICE (eds) (2006) Business and Official Cor-
respondence: Historical Investigations. Bern: Peter Lang.
DOSSENA, Marina/Irma TAAVITSAINEN (eds) (2008) Diachronic Perspectives on 
Domain-Specific English. Bern: Peter Lang.
DOSSENA, Marina/Ingrid TIEKEN-BOON VAN OSTADE (eds) (2008) Studies in 
Late Modern English Correspondence: Methodology and Data. Bern: Lang.
ELSPAß, Stephan (2005) Sprachgeschichte von unten: Untersuchungen zum geschrie-
benen Alltagsdeutsch im 19. Jahrhundert. Tübingen: Niemeyer.
332
FAIRMAN, Tony (2007) “Writing and the ‘Standard’ 1795–1834.” Multilingua 26, 
167–201.
FAIRMAN, Tony (2008) “Strike-throughs: what textual alterations can tell us about 
writers and their scripts, 1795–1835.” In: M. Dossena/I. Tieken-Boon van Ostade 
(eds), 193–212. 
FENS-DE ZEEUW, Lyda (2008) “The letter-writing manual in the eighteenth and 
nineteenth centuries: from polite to practical.” In: M. Dossena/I. Tieken-Boon van 
Ostade (eds), 163–192.
FITZMAURICE, Susan M. (2002) The Familiar Letter in Early Modern English: A 
Pragmatic Approach. Amsterdam: John Benjamins.
GARDNER, Anne-Christine (2018) “Downward social mobility in eighteenth-century 
English: a micro-level analysis of the correspondence of Queen Charlotte, Mary 
Hamilton and Frances Burney.” Neuphilologische Mitteilungen 119/1, 71–100.
GARDNER, Anne-Christine (2023a) “Petitioning for the education of the poor: 
self-corrections as stylistic choices in a Late Modern English draft letter.” In: M. 
Schiegg/J. Huber (eds), 157–180.
GARDNER, Anne-Christine (2023b) “Speech reflections in Late Modern English pau-
per letters from Dorset.” English Language & Linguistics 27/3: 491–516.
GARDNER, Anne-Christine/Anita AUER/Mark ITEN (2022) “Language and mobil-
ity of Late Modern English paupers.” In: Michael C. Frank/Daniel Schreier (eds), 
Contacts and Migrations. Heidelberg: Winter, 45–70.
GÖRLACH, Manfred (2004) Text Types and the History of English. Berlin: Mouton 
de Gruyter.
HAILWOOD, Mark (2023) “Rethinking literacy in rural England, 1550–1700.” Past 
& Present 260: 38–70.
HEALEY, Jonathan (2016) “‘By the charitie of good people’: poverty and neighbourly 
support in seventeenth century Lancashire.” Family & Community History 19/2, 
83–94.
HINDLE, Steve (2004) On the Parish? The Micro-Politics of Poor Relief in England, 
1550–1750. Oxford: Oxford University Press.
HOUSTON, Robert A. (2014) Peasant Petitions: Social Relations and Economic Life 
on Landed Estates, 1600–1850. London: Palgrave Macmillan.
IREMONGER, Frederic (1813) Suggestions to the Promoters of Dr. Bell’s System of 
Tuition with an Account of the Hampshire Society for the Education of the Poor. 
Winchester. 
JONES, Peter/Steven KING (2015) “From petition to pauper letter: the development of 
an epistolary form.” In: Peter Jones/Steven King (eds), Obligation, Entitlement and 
Dispute Under the English Poor Laws. Newcastle upon Tyne: Cambridge Scholars 
Publishing, 53–77.
KING, Steven (2019) Writing the Lives of the English Poor, 1750s–1830s. Montreal: 
McGill-Queens University Press. 
KOHNEN, Thomas (2001) “On defining text types within historical linguistics: the 
case of petitions/statues.” European Journal of English Studies 5/2, 197–203.
333
LAITINEN, Lea/Taru NORDLUND (2012) “Performing identities and interaction 
through epistolary formulae.” In: Marina Dossena/Gabriella Del Lungo Camici-
otti (eds), Letter Writing in Late Modern Europe. Amsterdam: John Benjamins, 
65–88.
LAITINEN, Mikko/Anita AUER (2014) “Letters of artisans and the labouring poor 
(England, c. 1750–1835): approaching linguistic diversity in Late Modern English.” 
In: Simone E. Pfenninger/Olga Timofeeva/Anne-Christine Gardner/Alpo Honka-
pohja/Marianne Hundt/Daniel Schreier (eds), Contact, Variation and Change in the 
History of English. Amsterdam: John Benjamins, 187–212.
LAWSON, John/Harold SILVER (1973) A Social History of Education in England. 
London: Methuen & Co. 
MOESSNER, Lilo (2001) “Genre, text type, style, register: a terminological maze?” 
European Journal of English Studies 5/2, 131–138.
NEVALA, Minna (2004) Address in Early English Correspondence: Its Forms and 
Socio-Pragmatic Functions. Helsinki: Société Néophilologique.
NEVALAINEN, Terttu/Helena RAUMOLIN-BRUNBERG (1995) “Constraints on po-
liteness: the pragmatics of address formulae in early English correspondence.” In: 
Andreas H. Jucker (ed.), Historical Pragmatics: Pragmatic Developments in the 
History of English. Amsterdam: John Benjamins, 541–601.
OED Online. http://www.oed.com [accessed on 17 June 2022].
RUTTEN, Gijsbert/Marijke VAN DER WAL (2012) “Functions of epistolary formulae 
in Dutch letters from the seventeenth and eighteenth centuries.” Journal of Histori-
cal Pragmatics 13/2, 173–201.
RUTTEN, Gijsbert/Marijke VAN DER WAL (2014) Letters as Loot: A Sociolinguis-
tic Approach to Seventeenth- and Eighteenth-century Dutch. Amsterdam: John 
Benjamins.
SAIRIO, Anni/Minna NEVALA (2013) “Social dimensions of layout in eighteenth-
century letters and letter-writing manuals.” In: Anneli Meurman-Solin/Jukka Tyrk-
kö (eds), Principles and Practices for the Digital Editing and Annotation of Dia-
chronic Data. Helsinki: Varieng.
SCHIEGG, Markus/Judith HUBER (eds) (2023) Intra-Writer Variation in Historical 
Sociolinguistics. Bern: Lang
SHVANYUKOVA, Polina (2020) “‘With the greatest sincerity’: expressing genuine-
ness of feeling in nineteenth-century business correspondence in English.” Multi-
lingua 39/1, 81–104.
SOKOLL, Thomas (2000) “Negotiating a living: Essex pauper letters from London, 
1800–1834.” International Review of Social History 45, 19–46.
SOKOLL, Thomas (ed.) (2001) Essex Pauper Letters, 1731–1837. Oxford: Oxford 
University Press.
SOKOLL, Thomas (2008) “Verhandelte Armut: Mobilität, Kontrolle und Selbstbe-
hauptung im englischen Armenrecht, 1780–1840.” In: Sebastian Schmidt (ed.), 
Arme und ihre Lebensperspektiven in der Frühen Neuzeit. Frankfurt am Main: Peter 
Lang, 85–118.
334
STEPHENS, William Brewer (1998) Education in Britain, 1750–1914. Basingstoke: 
Macmillan.
TIMMIS, Ivor (2020) The Discourse of Desperation: Late 18th and Early 19th Century 
Letters by Paupers, Prisoners, and Rogues. New York: Routledge.
UPTON, Clive/John D. A. WIDDOWSON (1996) An Atlas of English Dialects. Ox-
ford: Oxford University Press.
VINCENT, David (1989) Literacy and Popular Culture: England 1750–1914. Cam-
bridge: Cambridge University Press.
VINCENT, David (2014) “The invention of counting: the statistical measurement of 
literacy in nineteenth-century England.” Comparative Education 50/3, 266–281.
WHYMAN, Susan (2009) The Pen and the People: English Letter Writers 1660–1800. 
Oxford: Oxford University Press. 
WHYTE, Ian (2004) “Migration and settlement.” In: Chris Williams (ed.), A Compan-
ion to Nineteenth-Century Britain. Malden: Blackwell, 273–286.
WŁODARCZYK, Matylda (2013) “Self-reference and ego involvement in the 1820 
Settler petition as a leaking genre.” In: Marijke J. van der Wal/Gijsbert Rutten (eds), 
Touching the Past: Studies in the Historical Sociolinguistics of Ego-Documents. 
Amsterdam: Benjamins, 201–242. 
WRAY, Alison (2002). Formulaic Language and the Lexicon. Cambridge: Cambridge 
University Press.
Abstract
ENGLISH PAUPER LETTERS IN THE EIGHTEENTH CENTURY AND BE-
YOND: ON THE VARIABILITY AND EVOLUTION OF A NEW TEXT TYPE
This article explores a new subtype of vernacular correspondence emerging in the 
early eighteenth century, the pauper letter, providing a textual classification of this 
new text type and an account of its variability and evolution into the nineteenth 
century. The study is based on 189 letters sent between 1730 and 1834 by the poor 
applying for support from their parish of legal settlement, with a focus on the po-
tential influence of the form and language of petitions in the realisation of five fea-
tures which can be found in pauper letters (voice, greeting, petition-element, closing 
formula and self-reference). From a diachronic perspective, letter-style writing was 
preferred already in the earliest pauper letters, and the presence of petition-style fea-
tures further decreases over time. In consequence, pauper letters should be classified 
as letters rather than petitions. The analysis of inter-writer variation across pauper 
letters and different social groups shows that in pauper letters there is significant 
stylistic variation and that the writers employ strategies, in particular self-reference, 
to index their social roles of applicant and parishioner or to highlight the difficult 
circumstances in which they find themselves, as they appeal to the responsibility 
of the parish officials to offer assistance. In contrast, in 14 letters by parish officials 
and advocates writing on behalf of the poor formulaic sequences are generally more 
335
uniform than in pauper letters. Stylistic choices by writers with more experience 
(including paupers) tend to be oriented towards majority usage patterns and show an 
affinity with stylistic expressions also attested in business correspondence. A study 
on intra-writer variation reveals that individual encoders have a fixed repertoire of 
formulaic expressions expected in closing formulae and self-references. Owing to 
their limited schooling opportunities, paupers typically only acquired incomplete 
knowledge of conventional expressions and their appropriateness. Self-corrections 
by paupers in closing formulae and self-references support earlier findings that for-
mulaic sequences were often learned through incidental oral transmission and stored 
and processed as a whole.
Keywords: historical sociolinguistics, pauper letters, text types, formulaic language, 
self-corrections
Povzetek
ANGLEŠKA PISMA REVNIH V 18. STOLETJU IN POZNEJE: O RAZNOLIKO-
STI IN RAZVOJU NOVE BESEDILNE ZVRSTI
V članku se ukvarjamo s pismom revnih kot novim podtipom vernakularne korespon-
dence, ki je nastal v zgodnjem 18. stoletju. Predlagana je besedilna klasifikacija te 
zvrsti in predstavljena njena raznolikost in razvoj do 19. stoletja. Raziskava temelji 
na 189 pismih, ki so jih revni poslali med 1730 in 1834 in v katerih so župnijo, ki so 
ji uradno pripadali, prosili za podporo. Posebej nas zanima morebitni vpliv oblike in 
jezika teh prošenj na izraženost petih značilnosti, ki jih najdemo v tovrstnih pismih 
(slovnična oseba, pozdrav, prošnja v ožjem smislu, zaključna formula in avtorefe-
rencialnost). Z diahronega stališča se avtorji že v najzgodnejših pismih odločajo za 
pisemski slog, medtem ko prisotnost peticijskega sloga skozi čas še nadalje upada. 
Zato bi bilo smiselno imeti pisma revnih za pisma, ne za peticije. Analiza variantnosti 
med posameznimi pisci in različnimi družbenimi skupinami kaže, da so pisma revnih 
slogovno zelo raznolika in da pisci z uporabo določenih strategij, predvsem avtore-
ferencialnosti, kažejo na družbeno vlogo, ki jo imajo kot prosilci in župljani, ali pa 
skušajo poudariti, v kako težkih razmerah so se znašli, ko se sklicujejo na odgovor-
nost župnijskih uradnikov, da nudijo pomoč. Nasprotno pa je za 14 pisem, katerih 
avtorji so bili župnijski uradniki in zastopniki, ki so pisali v imenu revnih, značilna 
večja enovitost uporabljenih stalnih zvez kot v pismih revnih. Slogovne izbire bolj 
izkušenih avtorjev (vključno z revnimi) stremijo k vzorcem prevladujoče rabe in vse-
bujejo podobnosti s slogovnimi izrazi, ki jih najdemo tudi v poslovni korespondenci. 
Raziskava o znotrajavtorski variantnosti razkriva, da imajo posamezniki ustaljen na-
bor stalnih zvez, pričakovanih v zaključnih formulah in na avtoreferencialnih mes-
tih. Spričo omejenih izobrazbenih možnosti revni običajno pridobijo zgolj okrnjeno 
znanje o ustaljenih izrazih in o ustreznosti njihove rabe. Samopopravljanja revnih, ki 
jih najdemo v zaključnih formulah in pri avtoreferencialnih izrazih, govorijo v prid 
336
predhodnim ugotovitvam, da so bile stalne zveze pogosto naučene prek naključnega 
ustnega prenašanja in da so se shranjevale in procesirale kot samostojne enote.
Ključne besede: zgodovinska sociolingvistika, pisma revnih, besedilne zvrsti, formu-
laični jezik, samopopravljanje