The Pragmatics of Poverty in the Essex Pauper Letters, 1731–1837

This paper discusses the way early nineteenth century English paupers used language for the pragmatic purpose of securing charitable relief. The paper is based on two historical sources: (1) The Essex Pauper Letters (Sokoll in Essex pauper letters, 1731–1837, Oxford University Press, Oxford, 2001), which consists of letters written by paupers applying for charitable relief, and (2) the Mayhew Corpus, a corpus of interviews with the destitute of London carried out by Sir Henry Mayhew in the 1850s. The paper focuses on certain grammatical differences between the language of the pauper letters and the language in the Mayhew Corpus. From this analysis, it emerges that the pauper writers made markedly less use of certain vernacular features than speakers in the Mayhew Corpus. The features not used to any great extent in the pauper letters but present in the Mayhew Corpus are: vernacular relative pronouns (as and what); vernacular preterites and past participles; a-prefixing; and non-standard verbal ‘s’ ending. It is argued that the infrequency of these features in the pauper letters indicates that the pauper writers were orienting towards the emergent notion of Standard English. However, in contrast to this argument, we find that multiple negation, a low prestige vernacular feature, occurs with similar frequency in both The Essex Pauper Letters and the Mayhew Corpus. The main argument of the paper, in the light of this apparent contradiction, is that, in some cases, the pauper writers’ attempts to orient towards prestige forms faltered as they were dealing with the emotive issues of health, welfare and money.


Introduction
This article aims to show, through a comparison of aspects of the written and spoken language of the poor in nineteenth century Essex and London, how the Essex paupers sought to use language strategically to secure relief from poverty. At first sight, the aim to compare the written and spoken language of the poor in the nineteenth century may seem quixotic, for two reasons. Firstly, while 'ordinary writings' have been uncovered in recent years (Lyons 2007), not a great deal of material written by the poor and destitute has survived from the era in question; this is partly because even basic literacy teaching had only just been introduced and partly because, in general, the poor had little need or motivation to write texts of any kind at this time. Secondly, there is no direct evidence of the speech of paupers at the time as audio recording was unknown.
What I argue in this paper, however, is that there are two historical sources of exceptional quality which do permit comparison of the language written by the poor and the language spoken by the poor in the nineteenth century: the Essex Pauper Letters (1731-1834) and the Mayhew Corpus of 1850s spoken data. These sources, I aim to show, allow us to address the question of how the paupers marshalled their linguistic resources for the pragmatic purpose of writing letters seeking vital charitable relief. Specifically, we will discuss how the pauper writers conspicuously eschewed certain vernacular grammatical features in their letters, but still made some use of the low prestige vernacular feature of multiple negation. In this way, the paper reflects the aim of historical socio-pragmatics which, Del Lungo Camiciotti (2014: 20) argues, 'sheds light on the ways in which texts are used to achieve particular goals in social interaction'. As the term 'vernacular' features large in the ensuing discussion, it is important to define at the outset the way I am using it. I am following the definition provided by Biber et al. (1999Biber et al. ( : 1121: The term vernacular, referring to the popular, untaught variety of a language found in colloquial speech, will serve to cover a range of phenomena in popular speech which may to a greater or lesser extent be felt to lack prestige and to be inappropriate for serious public communication, especially written communication (my italics).

The Pauper Letters Ordinary Writings
While the focus of this paper is on pauper letters, the study needs to be seen in the wider context of increasing research in Europe in recent years into 'ordinary writings', particularly those of the nineteenth century (Lyons 2007). The study of ordinary writings is described by Lyons (2007: 13) as 'an expanding field of inquiry, which involves cultural historians, ethnographers, sociologists, educators 1 3 The Pragmatics of Poverty in the Essex Pauper Letters, 1731-… and specialists in literary studies and the study of literacy'. It is striking that linguists are not included in this list, though there has been work from a linguistic perspective by, for example, Fairman (1999Fairman ( , 2007 and Laitinen (2015) on the pauper letters. Other forms of ordinary writing include working class autobiographies, petitions, memoirs, family chronicles and diaries (Sokoll 2006;Lyons 2007). The general tenor of such research has been to show that 'the poor and labouring classes were functionally literate and used language to assert and reflect their agency' as far back as the late eighteenth and early nineteenth centuries (Gestrich and King 2013: 12-13). The specific significance of the pauper letters among these genres is underlined by Gestrich and King (2013: 13): Within the wider historiography of the rise of reading, writing, and publishing, pauper letters push back the date of an effective literate culture for ordinary people, and illustrate how, for many decades in the eighteenth and nineteenth centuries, oral and literate cultures survived side-by-side, each reinforcing the other.
It is particularly relevant for the purposes of this paper that Laitinen (2015: 201) points to the value of the pauper letters for the study of vernacular language at a time when the construct of Standard English was emerging.
The material [pauper letters] is a living record and evidence of vernacular speech at the time when the general ideology of standard English was being formed by the privileged and the most powerful.
We have, then, in the Essex Pauper Letters, and in other collections of such letters e.g. the Kirkby Lonsdale Township Letters (Taylor 1997); the Letters of Artisans and the Labouring Poor ; and the Sussex Weald Letters (Allen 2015), some evidence of the way the pauper class used language in writing a genre which could literally be a matter of life and death to the writer and/or his/her family. This paper, as noted above, focuses on a particular collection of letters, The Essex Pauper Letters, 1731-1834 (Sokoll 2001) (henceforward EPL) as around 750 letters are included in the collection, which have been published as a book (Sokoll 2001). Running to around 140,000 words, they can be said to constitute a small de facto corpus.

The Legal and Economic Background of the Pauper Letters
As an introduction to the discussion of these letters which follows, a sample letter written by one of the most persistent pauper writers, David Rivenall, is quoted below. While I describe him as a persistent writer, it should be noted that it is obvious on reading all his correspondence that certain of the letters bearing his name (in various spellings e.g. Rivenell; Rivenall) have been written on his behalf by others. This was not an uncommon practice in letters of this kind (Sokoll 2006;Fairman 1 3 2007). This particular letter illustrates starkly the desperate plight some of the pauper writers found themselves in.
Sir I am under the Necesity of writing to You again As not having an Answer from the last I sent You respecting the interment of my Child Which is out of my power to do I hope You will Answer this by return of post and send mee some money that I may bee able to get it put in the ground or Otherways I must apply to the Parish Where I am for relief I expect to loose two more every hour I am Sir Your most Abdt hbl St Davy Rivenell [sic] Please to direct to mee Kings head Comersiael Road To appreciate such letters fully, we need to set them in socio-historical context. We need, then, to consider legal and social aspects of the historical background to the letters. In terms of the legal context, the historical roots of pauper letters lie in the Poor Law of 1601. Under this law, the obligation to relieve the poor passed from the churches and monasteries to the state (Fairman 1999). This obligation, Fairman (1999) points out, was then devolved by the state to 15,000 parishes and administered by annually elected 'overseers' operating under the auspices of local magistrates. The main provisions of the Poor Law are set out by Sokoll (2006): in principle, with rare exceptions, paupers only had the right to receive poor relief from their home parish while living in their home parish. If paupers fell on hard times, then, their first port of call had to be their home parish rather than the parish where they happened to be living. The presumption for most of the period in which the Poor Law operated (1601-1834) was that, to receive poor relief, the paupers would return to their home parish (which was in the county of Essex in the case of the EPL). In 1834, the Poor Law was amended such that paupers could no longer receive relief at a distance and their only recourse was to the Workhouse.
Against this legal background, the conjunction of two historical developments created the pauper letter genre. Firstly, the incipient industrial revolution at the turn of the eighteenth and nineteenth centuries led to a migration of large numbers of the rural labouring classes from the countryside to urban centres in pursuit of employment. Secondly, the increasing provision of rudimentary education for the lower social classes led to wider basic literacy. Thanks to their mobility and their newly acquired basic literacy, when the paupers fell on hard times in their new parish of residence, they had both a need to apply for aid to their home parish and the means to do it: 'the pauper letter'. The opportunity to apply for relief by letter had the advantage for the paupers of obviating the need for a troublesome journey or permanent relocation back to the parish to make their claim. While the provision of poor relief at a distance was not part of the 1601 Poor Law, from the point of view of the overseers in the home parish, the payment of relief to paupers in their parish of residence could also be advantageous, as the parish could get away with paying less to the paupers if they remained in their parish of residence rather than returning to the home parish (King 2007).

3
The Pragmatics of Poverty in the Essex Pauper Letters, 1731-…

The Education of the Paupers
It is important now to consider how well the rudimentary education the paupers received at that time prepared them to compose a genre upon which their lives literally depended. The education provided in this era was undoubtedly of varying quality- Gillard (2011 n.p.) reports that in 1816 the figures for 12,000 parishes were as follows: '3500 had no school, 3000 had endowed schools of varying quality, and 5500 had unendowed schools of even more variable quality'. The emphasis in schools for the poor was very much on passive literacy and would typically involve spelling books as well as reading and copying from the bible (Fairman 2007). Pupils in these schools would most often spell phonemically (Fairman 2007;Allen 2015) and they would have no experience of composing texts such as familiar letters or petitions as these formed no part of the curriculum. The education provided for the poor at this time was functional in the sense that it only prepared the poor to work for those in power. This severely limited aim of education for the poor was often openly stated. Fairman (2007: 183), for example, draws attention to the rules of the Charity School of the Parish of St Mary-le-Bone in 1794, which stated that the education provided in the school should be 'such only as fits them for the condition they are to hold in life'.
A picture emerges then of an institutional goal for the poor of minimal literacy and a highly variable quality of preparation for even this minimal target. Even in cases where literacy had been reasonably well taught, there may well have been a considerable gap between the time the paupers left school and the time when they had to compose letters asking for relief, time enough for their literacy to atrophy (Allen 2015).

Standards, Norms and the Pauper Letters
An analysis of the language of the pauper letters needs to be set in the context of the language norms and genre models of the time. In terms of norms, it is generally recognised that the late eighteenth and early nineteenth centuries were an important period for the development of the construct of Standard English. A plethora of prescriptive grammars published during this period bear witness to this development (Linn 2006). In terms of nineteenth century notions of Standard English, it is important to note that Allen (2015: 202) stresses 'exclusiveness and superiority, especially class-based superiority' as salient features of the construct. Similarly, Watts (2015: 4) argues that 'The nineteenth century use of the term "standard language" is interchangeable with terms such as "the national language", "the language of the educated", "the language of refinement" …'. The question arises, particularly in the light of the discussion of education for the poor above, as to whether this emergent Standard English had any effect on the way the paupers wrote their letters. The paupers certainly used some constructions which would not have been part of their everyday repertoire (Watts 2015). Even though these constructions were not always appropriately or accurately used, Watts (2015) argues, the attempts to use them mark an orientation on the part of the paupers towards the emergent Standard English.
Frequently, such attempts to perform the standard miss the mark, but are, for this very reason, important indications for ways in which non-standard speakers perceived the social significance of a variety of English that was regularly projected by those in power in the first half of the nineteenth century as the only 'legitimate' form of language.
However, Pietsch (2015) argues that many of the awkward formal phrases the paupers attempted were neither part of their everyday language nor part of the emergent Standard. Such phrases, Pietsch (2015) argues, may well be 'hand-me-downs' from previous generations which were traditionally considered to be prestige forms. For this reason, Pietsch (2015: 223) points out that 'such texts [pauper letters] require a wholly new approach to the classification and interpretation of their status within the spectrum of varieties beyond the traditional terms of "standard" and "dialect"'. Other descriptive terms have indeed been proposed. Fairman (1999: 65), for example, prefers the terms 'schooled' and 'open English', which he defines as follows: For centuries English has been passed on in socially or economically recognised situations, for example from master to apprentice in a scriptorium or printing workshop, or from teacher to pupil in a school or drawing room, using books as authority for what might be termed 'best practice', whether in orthography, punctuation, pronunciation, lexis, meaning, grammar or discourse. By 'Schooled', therefore, I mean any English passed on at any time by such Schooling. Schooled English is typically written, but speech can be Schooled too and has been influenced by Schooled writing. 'Open' refers to English which is free of schooling but Open to other influences.
A classification of pauper letters proposed by Allen (2015: 207-208) adopts a continuum (often used in Creole studies) ranging from basilectal to mesolectal, which she explains as follows: • At the basilect end of the continuum, not only are misspellings frequent, but writers also demonstrate a high level of misunderstanding of word structure and morphology • In the mesolect range, the letters reveal a general awareness of standard forms, of a model, of conventions, but it is clear that insufficient familiarity with the norms of the standard make its use impossible.
In the light of the discussion above, we could place Rivenall's letter (2.2.1) towards the mesolectal end of the range, towards the oral end of an oral-written continuum or towards the vernacular end of a vernacular-(emergent) standard continuum.
There are, of course, norms and models in terms of genre as well as grammar for letters, and there was awareness of such norms among some of the pauper writers. As Watts (2015: 10) puts it:

3
The Pragmatics of Poverty in the Essex Pauper Letters, 1731-… … it is also striking how some of the pauper letters display evidence of an awareness of letter-writing conventions, e.g. how to address the addressee appropriately, how to finish off the letter, how to format the address at the top of the letter, and so on.
Indeed, there is intriguing evidence in a letter from a certain Mrs Reilly in 1828 (no. 638) of meta-awareness of the demands of the genre: Dr Sr Like all troublesome indigent persons I begin my Letter, by beging you to excuse the trouble I am about to give you … In terms of the genres of letter which paupers could potentially draw upon to develop this kind of awareness and apply it to their own letters seeking poor relief, Sokoll (2006) points to two relevant genres which were well established at the time: the familiar letter and the ancient genre of the petition. The structure of the pauper letter does seem to bear a close similarity, Sokoll (2006) argues, to the classical rhetorical model of the petition in which the following stages were conventional: • greeting the recipient (salutatio) • appealing to his or her goodwill (captatio benevolentiae) • giving details of the particular case (narratio) • making a specific request (petitio) • bidding the recipient farewell (conclusio).
We can return to the letter from David Rivenall to illustrate this. The stages are given in bold after the section in italics to which they relate.  Sokoll (2006) that paupers were consciously following such a model, or even that every letter contained those elements in that sequence. Nevertheless, it is possible to discern at least some elements of this sequence in most of these letters, as indeed we can in the letter above

Compilation and Analysis of the Corpora The EPL Corpus
Compiling the EPL corpus simply involved digitising the 758 letters in Sokoll (2001). As was the case with the Mayhew Corpus, demographic information about the writers is not recorded, though the signatories of the letters, and sometimes their addresses, are noted. It is important to recall here that the signatory of the letter is not always the author as some of the pauper writers asked other people to write for them (Sokoll 2001). There are around 230 signatories of the 758 letters in the EPL as multiple letters from a particular signatory are often included in the collection.

The Mayhew Corpus
The value of the Mayhew Corpus for this paper lies in the fact that the data is from, I argue, a period, location, and a social class comparable with the EPL context. It is not, of course, a perfect point of comparison. London is not in Essex, though it is near, and many of the Essex paupers took up residence there in pursuit of work. It is not an identical period, though only 16 years separate the end of the pauper letter era (1834) and the beginning of the project which generated the Mayhew data (1850). The interviewees who provided Mayhew's data were not necessarily paupers, but his focus was explicitly on 'the poor'.
The Mayhew Corpus is drawn from the four-volume London Labour and the London Poor (LLLP) written in the 1850s by Sir Henry Mayhew working, at least in the early stages of the project, for the Morning Chronicle as 'Special Correspondent to the Metropolis'. Mayhew was commissioned to write a series of articles aiming 'to give a full and detailed description of the moral, intellectual and physical condition of the poor of England' (Morning Chronicle, October 18th, 1849). It is these articles which form the basis of the two million word London Labour and the London Poor. Significantly for our purposes, LLLP contains, among the reports, the statistics, and narrative, a substantial amount of data from the interviews Mayhew conducted with the destitute of London. From such evidence as there is, it seems that Mayhew carried out most of the interviews for LLLP himself, accompanied by an assistant taking notes. Typical topics included wages, working conditions, expenses, working hours and housing issues (Nunes 2012).
Mayhew's subjects largely fell into six categories: street sellers; street buyers; street finders; street performers, artists, and showmen; street artisans, or working pedlars; and street labourers (Nunes 2012). Unfortunately, Mayhew's logs of the interviews he conducted have not survived (Douglas-Fairhurst, personal communication) which means that we do not have much demographic information about the respondents. In terms of the ethnic origins of the respondents, Mayhew does record very small numbers of the following: Irish, Jewish, Italian and Hungarian. Given that race or ethnicity is not otherwise specified, and that Jewish people could easily be from London, it seems reasonable to apply 1 3 The Pragmatics of Poverty in the Essex Pauper Letters, 1731-… Occam's Razor and assume that the interviewees whose origins were not specified were local London people. I compiled the Mayhew Corpus by digitising about 110 direct speech interviews from London Labour and the London Poor, which yielded just under 200,000 words. Crucially for our purposes, Mayhew (1861: 22) made a clear effort to capture the vernacular, recording the words of his interviewees, as he put it, 'near verbatim, omitting oaths and slangs (sic)'. We should note that the 'slangs' Mayhew refers to are the in-group words of costermongers and his claim that the interviews were 'near verbatim' needs to be treated with caution; he does not record, for example, hesitations, repetitions and slips of the tongue. All excerpts from the MC are reproduced as they were in LLLP so that readers can judge for themselves the verisimilitude of this data.

Analysis of the Corpora
The process of analysis of both the MC and the EPL was manual and painstaking. I would draw attention to three aspects of the methodology. First of all, I was heavily involved in the transcription of the material which constitutes the corpora-this means that I inevitably began to notice certain recurrent features just through the act of typing. This manual approach allows you to recognise a recurrent feature even when there is variation in spelling e.g. wrote; wrought and rote in the EPL. (VARD2 software can cope with spelling variation but needs to be 'trained' first-http://ucrel .lancs .ac.uk/vard/about /). When I needed to do a frequency count, therefore, I knew I had to search for different spellings of the same word or phrase. A second feature of the methodology was that I read all the texts in both corpora several times. A benefit of this approach is that in so doing you familiarise yourself to some extent with the social and historical background to the corpus data, which is so important in the contextualisation of historical language sources (Rissanen 2008). A third point worth noting is that, as I had analysed the MC before the EPL, when I came to the analysis of the EPL, I expected, because of the comparable social backgrounds, to find similar vernacular features to those in the MC. It was during this process that I began to notice that certain vernacular features from the MC, which I expected to find in the EPL were not present to anything like the same extent. The focus on absence as well as presence of vernacular features is perhaps unusual, but as we know from the Sherlock Holmes story about the 'dog that didn't bark' in the story, The Curious Incident of the Dog in the Night-time, absence can sometimes be as significant as presence in forming a theory. In this instance, I will argue that, from studying both absence and presence of vernacular features, we gain some insight into what the paupers regarded as high and low prestige language forms and how they deployed them, often writing under intense emotional pressure to achieve a particular pragmatic goal. While there is no clear evidence that the more literate letters were successful in gaining charitable relief, what is important for our purposes is that this seems to have been the perception of the pauper writers. Why else would they produce often convoluted efforts to adopt a non-vernacular style? 1 3

Vernacular Forms in the MC and EPL: Presence and Absence
In the next section we return to the Sherlock Holmes analogy, the 'dogthat didn't bark', or in ths case, 'dogs' i.e. the vernacular features frequent in the MC (and described in historical grammars at the time) but rare in the EPL. We consider one syntactic feature-relativizers-and three features of verb morphology: preterite/ past participle forms, verbal 's' ending and a-prefixing, as illustrated below: • I learned them style and grace what I knowed myself • I recollects nothing more than this • It's the march of hintellect wot's a-doing all this-it is, sir

Standard and Vernacular Relativizers
I have chosen to use the term 'relativizer' rather than 'relative pronoun' in this section as it allows us to include in our discussion the forms 'as' and 'what', which are not generally recognised as relative pronouns, but which play an important role for the purposes of comparison between the MC and the EPL. The relativizer system in Present-day English seems straightforward at first sight since there is general consensus as to the members of the group and their functions. They have also been in the language for a long time and were certainly all in use in the nineteenth century. The picture turns out, however, to be a little more complex than it seems for three reasons, as Timmis (2017: 128) outlines: Firstly, existing forms have previously been used in different ways: which, for example, has been the most frequently selected relativizer to refer to people at certain periods, including the nineteenth century (Johansson 2006). Secondly, the frequency of certain relativizers has varied a great deal historically. Thirdly, standard relativizers have co-existed with non-standard relativizers such as what, as, and zero-relativizer with a subject. The examples of relative clauses from the MC below show forthright views on gender relations and an eccentric reading of evolutionary theory; more importantly for us, they all use as-relativizer: • A man as is married is obligated to keep his wife if they quarrels or not • they hears things there as oughtnt to be said to young gals • Ive heerd him talk about the first man and woman as was made and lived-it must be more than a hundred years ago In the two examples below, it is curious that the vernacular what-relativizer is juxtaposed with hypercorrection in hintellect and hopposition.
• It's the march of hintellect wot's a-doing all this-it is, sir • I did once, though, strike up hopposition to a street preacher wot was a holding forth in the New-road, and did uncommon well When we turn to the EPL, we find two interesting differences from the MC in terms of the distribution of relativizers, illustrated in Table 2 below: Which is clearly the most common relativizer in the EPL, while the vernacular relativizers, what and as are, save one example, completely absent. The most convincing explanation for this disparity in relativizer use between the MC speakers and the EPL writers, I am going to argue, is that in this aspect of language use the pauper writers were orienting towards what they regarded as prestige forms and eschewing vernacular forms. This argument is supported by Johansson's (2006) observation, based on her nineteenth century corpus of written English, that Wh-relativizers are a particular feature of nineteenth century formal written English (Johansson 2006). Significantly for the purposes of this research, in Johansson's (2006) sub-corpus of letters, the preponderance of wh-forms was particularly marked (86% wh-forms; 14% that-forms). Wh-relativizers, Johansson (2006: 139) remarks, were 'the literary norms' of the era. By contrast with the wh-forms and that, as-relativizer and whatrelativizer were, and are, rare in writing. It is the vernacular character of these two features, Poussa (1991: 306) argues, which explains their absence from most grammatical descriptions of the relativizer system. This apparent orientation of the pauper writers towards the standard relativizers may, however, have come at a cost as 1 3 around 20% of the which-relativizers in the EPL are wrongly used. Most problems seem to lie with one or more of the following points: identifying a clear referent; using a redundant pronoun in addition to the relativizer; using the right preposition where one is required, or omitting a preposition where it is required. These difficulties are illustrated below in extracts from the EPL: • I hope never to be aney moore truble to you Wich if my rent is paide now at Standon I Shall belong to that parish • I am very sorry thay you Gentlemen of th commity should Cause a man like me from my family which I love my children a dear as you do • she has got over her laying In which you no it is of no use for me to have her with me in a strange Place with out anything to do with • my work has been So Short I Was Bolige to go the Parish Which sir I hope you Will not take it Amiss but thank God I am better

Vernacular Strong and Weak Forms
As was the case with the relativizer system, the apparent stability of the current standard system of strong and weak forms conceals a story of marked variability, both historically and in present-day dialects. What the story reveals, argues Hickey (2012: 3), is the 'arbitrariness of what was to become standard usage'. Similarly, Anderwald (2011: 85, cited in Timmis 2017 points out that the historical development of the strong and weak verb system raises a number of important questions relevant to this study: • Who determines what is 'correct' in English? • Have these norms of correctness changed over time? • Do 'normal' speakers adhere to these norms? • If 'normal' speakers do not adhere to these norms, what possible reasons are there for this variability? Anderwald (2011: 85).
It is challenging to track the vernacular aspects of variability in strong and weak verb forms historically as we are largely dependent on written sources for data (Anderwald 2011) where one might expect greater consistency. It is particularly fortunate, then, that we are able to call on the MC for evidence of vernacular strong and weak forms in circulation during this era.
There are three kinds of variation between standard and vernacular strong/weak forms, which are described by Timmis (2017)  3. More rarely, the verb is weak in SE but strong in VE e.g. SE treat-treatedtreated; VE treat tret tret.
The occurrence of strong and weak vernacular forms in the EPL and the MC is summarised in the table below (Table 3): It is apparent from the table that there is quite a disparity in terms of both tokens and types of vernacular strong/weak forms between the MC and the EPL. The disparity seems even more marked if we consider that 27 of the 38 tokens in the EPL are forms of the verb write, which appears recurrently as I have wrote (in many spellings). In the MC the most frequent variation from SE is knowed as preterite (24), followed by done as preterite (20). It seems safe to conclude that the figures provided above indicate that the pauper writers oriented towards the standard in this domain.

I can't tell how many brooms I use; for as fast as I gets one, it is took from me. [MC]
A further aspect of verb morphology which provides a striking contrast between the MC and the EPL is the verbal 's' ending i.e. the addition of an 's' inflection on the stem of the verb where SE has no such inflection, as illustrated in the example above and the section below from the MC: I sometimes spends a few browns a-going to the play; mostly about Christmas. It's werry fine and grand at the Wic, that's the place I goes to most; both the pantomimers and t'other things is werry stunning. I can't say how much I spends a year in plays; I keeps no account; perhaps x or so in a year, including expenses, sich as beer, when one goes out after a stopper on the stage. I don't keep no accounts of what I gets, or what I spends The use of numerous instances of vernacular verbal 's' ending in a long turn by a single speaker is typical of the MC as respondents describe their routine to Mayhew. This needs to be kept in mind when considering the overall frequency figures below.
Although the examples in the extract above are all first person singular, which in fact accounts for the majority of cases in the MC, there are also many instances in first person plural and a few in second person singular. A particularly common instance is the form 'we calls' when the respondents are referring to the terminology they use for their equipment, activities and tricks. In the case of third person plural, it is not really possible to say whether this is a case of 'default singular' or vernacular 's' inflection, but these instances are included in the table below. The 'Northern Subject Rule', which suggests that vernacular 's' ending is not found directly after a pronoun, does not seem to operate in the MC data. The contrast between the MC and the EPL could not be starker with regard to verbal 's': there are 243 instances in the MC produced by 68 different speakers and only one in the EPL (Table 4). As with the other vernacular features discussed above, the verbal 's' ending on all persons is a feature which long pre-dates the EPL and is still to be found in regional dialects (Godfrey and Tagliamonte 1999;Shorrocks 1999). It was, however, clearly avoided by the pauper writers in their letters, presumably on the grounds that it was a marker of the vernacular.

A-prefixing
Sir I should be much ablige to you if you would Send me Sarah players money to we am a-going to move Some Distence off A-prefixing, as illustrated in the example above from the EPL, has a long history in the language and is still to be found in certain areas of the USA such as the Ozark mountains (Montgomery 2009). The most common form of a-prefixing is the addition of 'a' to the present participle, though it also occurs occasionally attached to adjectives and the bare infinitive. There are 30 examples of a-prefixing per 100,000 words in the MC (a total of 55 produced by 40 different speakers), which means that it is not a particularly frequent feature but nor can it be dismissed as an anomaly or an idiosyncrasy: it is found among different speakers and is particularly common in the narratives of Punch and Judy (and similar) artists as they relate to Mayhew the stories they tell in their shows. As the a-prefix has no obvious ideational value, we need to account for its presence in the MC (and elsewhere).
Explanations of a-prefixing centre on affective concerns-it is argued, for example, by Frazer (1990) and Montgomery (2009) that a-prefixing is a style marker which signals a shift into the vernacular and may index solidarity with the interlocutor. It has also been argued (Feagin 1979: 110) that a-prefixing relates to the nature of the narrative: … the meaning of the prefixed present participle is that of the addition of emphasis, immediacy, and vividness to the verb phrase itself and, more generally, to the discourse. 1 3 The Pragmatics of Poverty in the Essex Pauper Letters, 1731-… A more specific discourse explanation for a-prefixing is offered by Frazer (1990: 90) who relates it more particularly to 'narratives of unusual drama or excitement' including, potentially significantly for both the EPL and MC data, 'crisis situations'.
The nature of the MC data, which includes both fictional dramatic narratives (Punch and Judy shows) and real life crises, seems to provide quite fertile ground for a-prefixing as the examples below illustrate: • Punch being condemned to suffer by the laws of his country, makes a mistake for once in his life, and always did, and always will keep a-doing it In a genuinely tragic tale, for example, a former farmer from Limerick describes the advent of famine: • I filt the famine a-comin. I saw people a-feedin on the wild green things, and as I had not such a bad take.
Given the nature of the EPL, and the discussion above of drama and crisis in relation to a-prefixing, we might at first sight expect a-prefixing to be common in the EPL. It is, however, not common: there are only nine examples i.e. 6 per 100,000 words as opposed to 30 per 100,000 words in the MC. As was the case with all the features discussed in this section, we find that a feature is generally eschewed by the pauper writers though it has a long history, was evidently in use in the vernacular of the time, and is still in use now. As was the case with the other features, we can only account for this disparity, I would suggest, by the argument that the feature is such a marker of vernacularity that is was avoided by the pauper writers in their attempts to orient towards the standard.

Multiple Negation: Affect and Prescription
if there was never no receivers there wouldn't never be no thieves [MC] The grammatical phenomenon of multiple negation, illustrated above, is described by Biber et al. (1999: 178) thus: Two or more negative forms may co-occur within the same clause to express a single negative meaning.
This very straightforward description of multiple negation, however, is immediately followed by the observation that the feature is 'socially stigmatized' (Biber et al. 1999: 178). Similarly, Carter and McCarthy (2006: 736) begin with the straightforward descriptive statement that '… in many non-standard dialects of English, double and multiple negatives are frequent with words such as never, nobody, nothing and nowhere' before quickly adding the caveat that multiple negation is 'a very sensitive issue'. For our purposes, we need to note that multiple negation is 'a very old pattern' (Biber et al. 1999: 178) and that it has long attracted social opprobrium on the grounds of its putative illogicality. Lowth (1762: 126) leaves us in no doubt about the supposed illogicality of multiple negation: 'Two negatives in English destroy one another, or are equivalent to an affirmative'. Multiple negation was certainly stigmatised in the early nineteenth century as Pegge (1814: 54) observes with characteristic levity: The most notorious imputed crime is, the use of redundant Negatives … This is a luxuriance of no modern date among the Cockneys; but it is not of their own manufacture; for there is evidence enough in the history of our Language, drawn from the old school, to shew that this mode of speech, this accumulation of Negatives, is no new-fangled tautology. One Negative is now accepted by us, and reputed as good as a thousand. The present Cockneys think otherwise; and so did the Ancestors of us all.
Multiple negation is manifestly in use in the MC where there are 85 occurrences produced by 37 different speakers (42.5 per 100,000 words) involving a variety of negators, as illustrated below: • I haven't got neither no father nor no mother,-never had, sir; for father's been dead these two year, and mother getting on for eight • there isn't no use a talkin' to them ere fellows-they're all tosticated now, and they doesn't care nothink for nobody … They're a awful lot, and nobody ill niver do anythink with them • why you see I aint no great schollard for Ive never been in no schools, only always hard at work, and knows nothing about it.
The three most common multiple negative combinations in the MC are listed in the table below (Table 5): Another noteworthy feature of multiple negation in the MC is its co-occurrence with ain't. There are 12 such instances, as exemplified in this self-deprecating comment: 'I ain't no great schollard'.
Multiple negation, as Table 6 shows, is also a recurrent feature in the EPL where there are 73 instances produced by 42 of the 230 different writers (49 per 100,000 words).
While the frequency of multiple negation in the MC and the EPL is broadly similar, there are differences in the frequency of particular multiple negation combinations which, I argue, turn out to be of interest. A salient multiple negation in the EPL is not + no more: there are 19 such instances with no more, and four with the semantically equivalent no longer. Eighteen of these 23 instances appear in remarks to the effect that the writer will give the overseers no further trouble with letters or demands e.g.
• I hope that the Gentlemen will do A little for me and I wont trouble them all the winter no more On further analysis, it was apparent that there were other semantic links with multiple negation in the EPL, as illustrated in the table below. The categorisation has to be tentative since, in a sense, all comments in the EPL are directly or indirectly related to the topic of money. Nevertheless, the table does show the different aspects of their plight which the pauper writers chose to emphasise in their letters (Table 7).
From the evidence above it seems that, in the case of multiple negation, the paupers did not orient towards standard forms and eschew the vernacular to the same degree as they did, for example, in the case of the other vernacular forms we considered above. This is particularly interesting because, as noted above, there is evidence to suggest that multiple negation was a stigmatised form during the pauper letter era. Given that, in other aspects, the paupers seem to have striven to impress with their orientation towards the standard, it is at first sight curious that multiple negation, a low prestige form, slipped through the net to the extent it did. The most plausible explanation for the frequency of occurrence of multiple negation in the EPL, I argue, lies in the semantic areas in which it is applied: these areas, as the table above shows, were, for the most part, critical areas of health and welfare. The crises which the paupers faced in these areas may well have led the paupers to abandon their attempts at the standard in their urgent need to express the severity of their case. In some cases, under pressure, the pauper writers' grip on Standard English seems to have faltered and a vernacular form appears to have (re)surfaced.

Conclusion
We have seen in this paper how two concurrent developments-urbanisation and the development of basic literacy among the poor-conspired to produce the pauper letter era in the late eighteenth century and early nineteenth century. A further concurrent development, the emergence of highly prescriptive notions of Standard English, and the social attitudes associated with it, contributed to shaping the pauper letter genre. Most significantly for this paper, the pragmatic purpose of the letters seems to have had a great influence on the style adopted. In terms of grammatical style, we noted that the pauper letter was something of a hybrid form. Among the letters, and often within the same letter, we can find attempts at formality, awkward though such attempts may be, mixed with barely literate spelling and occasional 'outbreaks' of unrestrained vernacular. Comparison with the MC, vernacular data from 1850s London, allowed us to suggest that certain vernacular features which we might expect to find in the pauper letters were actually rare. 'The dogs that didn't bark' which emerged from this comparison were vernacular relativizers; vernacular strong and weak forms; verbal 's' ending and a-prefixing. A plausible explanation for the infrequency of these features in the EPL, we argued, was that they were highly marked as vernacular and thus eschewed by the pauper writers in their belief that orientation towards the standard would carry more weight with the arbiters of their fate. This is consistent with the observation by Kautzsch (2000: 48) that 'semiliterates can be aware of the stigmatization of certain non-standard features in writing'. Similarly, Laitinen (2015: 195) notes that pauper writers were aware of 'how certain discourse features are indexed as socially more acceptable than others'. We need to credit the pauper writers, then, with a good degree of language awareness, just as Millar (2012: 176) does: 'most speakers of dialect had a developed sense of what was readily comprehensible over a wide area which could easily be transferred to writing'. The question arises as to why the pauper writers should avoid these vernacular forms when they are present in other forms of vernacular writing. In relation to the mid-nineteenth century American Civil War Letters, for example, Ellis (2017) refers to the frequency of non-standard features and a lack of self-consciousness on the part of the writers in using these features. A two-fold explanation can be offered: the pauper writers were corresponding with overseers who were higher in the social hierarchy and they often had a wish to portray themselves as respectable and, therefore, worthy recipients of aid. As Laitinen and Auer (2014: 203) put it: 'These letters from lower social orders to parish overseers of middling sort not only travelled physical distances, but they also crossed social spaces'. However, perhaps unexpectedly in the light of their tendency not to use these frequent vernacular features, we found that multiple negation-then, as now, a highly marked vernacular feature-is as frequent in the EPL as it is in the MC. A preliminary explanation for this was that multiple negation carries high affective value which over-rode the orientation towards the standard. This is analogous to Labov's (1972: 354) 'danger of death' argument that emotive topics were more likely to elicit vernacular speech, though in our case we are talking about writing. A similar argument is made by Taavitsainen and Melchers (1999: 17): 'Excessive emphatic use or emotional pressure may trigger codeswitching between standard and dialect'.
It could be objected that we also argued that the vernacular feature known as a-prefixing carries affective value. On this basis, we might expect a-prefixing, like multiple negation, to be a fairly frequent feature in the EPL, but it is not. We can, however, explain this apparent contradiction, I would argue, by pointing out that multiple negation, by its very nature, deals with absence, want and lack, the very issues which lay at the heart of the pauper letters. A-prefixing may also have affective value, but it does not necessarily express absence, lack or want, and so was unable to break through the pauper writers' orientation to the standard to the same extent as multiple negation does. In the case of multiple negation, by contrast, for some of the pauper writers some of the time, this vernacular feature was extremely important for the pragmatic purpose of expressing their plight. Their efforts to orient towards the standard seem to have weakened under the strain of expressing this plight. In short, when it came to the crunch, affect trumped the strictures of the prescriptive grammarians.