"It is me" vs. "It is I". Case Study on the effects of prescriptivism

Table of contents

List of graphs and tables

Abstract

Introduction

2. Literature Review
2.1. English grammars, grammarians and grammar writing: the creation of the standard
2.2. The field of Normative Linguistics
2.3. Pilot study

3. Initial hypotheses, aims, and research questions

4. Methodology
4.1. Precept Corpus
4.2. Usage Corpus 26.

5. Results and discussion
5.1. Diachronic change in attitudes
5.2. Labels, reasons, and topics
5.3. Target audience
5.4. Author’s place of birth
5.5. The dichotomy of gender: grammarians and target audience
5.6. The effects of prescriptivism on language use

6. The ‘controverted phrase’ nowadays

7. Conclusion

Reference list

Appendix I

Appendix II

Appendix III

List of graphs and tables.

Figure 1. Distribution of attitudes towards ‘it is me’ in the precept corpus

Figure 2. Overall choice of the correct form in the precept corpus

Figure 3. Diachronic distribution of grammars in the precept corpus

Figure 4. Attitudes over time from the 17th to the 20th century, raw numbers

Figure 5. Attitudes in time from the 17th to the 20th century, percentages

Figure 6. Topics in relation to which ‘it is me’/‘it is I’ is addressed Table 1. Labels on the construction ‘it is me’, sorted diachronically

Figure 7. Target audience of the grammars in the precept corpus

Figure 8. Target audience in relation to attitudes towards ‘it is me’

Figure 9. Distribution of grammarians in the precept corpus according to place of birth

Figure 10. Distribution of attitudes according to author’s place of birth, in raw numbers

Figure 11. Distribution of attitudes according to author’s place of birth, in percentages

Figure 12. Distribution of male and female authors in the precept corpus

Figure 13. Distribution of attitudes in female-written grammars

Figure 14. Diachronic distribution of ‘it is me’ in the Old Bailey Corpus

Figure 15. Distribution of ‘it is me’ in the Old Bailey Corpus according to speaker sex

Abstract

English grammar writing started in the late 16th century and flourished during the 18th century. It was common for grammarians during this era to adopt a prescriptivist attitude towards language; that is, they would dictate norms of correct usage of the language rather than describing language use. This trend lost weight from the 19th century onwards.

One of the constructions prescriptive grammarians usually criticised was it is me, with an object case pronoun, as opposed to the traditionally correct it is I, with a subject case pronoun. The aim of this paper is to look at the reasoning behind grammarians’ choice of correct form, and assess the effects of prescriptivism on actual language use. This was done by looking at a self-compiled precept corpus made of 66 historical English grammars, ranging from the 16th to the 20th century, as well as a usage corpus (ARCHER and Old Bailey) with data from the same eras.

The results show prescriptivist negative comments towards it is me decreasing diachronically, and while grammars definitely did not advocate for the use of the object case pronoun in this construction, people still used it in their speech, but not, however, on their writings, which clearly indicates a division between the uses of these two constructions. It is me has, nonetheless, succeeded in becoming part of the present-day Standard.

This dissertation hopes to make a valuable contribution on the field of Normative Linguistics, and to be used as ground for future research.

This dissertation is dedicated to Dr. Nuria Yáñez-Bouza, who introduced me to the topic during my undergraduate degree and has been a helpful guide ever since. Especial thanks to Dr. Tine Breban, who supervised this work and was with me every step of the

way. To Harry Styles, for being who he is and inspiring me to keep going. To family and friends who have put up with my grumpy, secluded summer. All the love.

- X.

1. Introduction.

The production of English grammars started in the late 16th century with the publication of William Bullokar’s Pamphlet for Grammar (1586), as reported by Alston (1965) in what still is the standard source of reference for the history of grammar writing in England to this day. After the publication of this grammar, production took off in an intermittent manner; and it was not until the 18th century that grammar writing really flourished in England, when there was a “growth of interest in living languages and science” (Sundby et al. 1991: 4).

There was a lot of variation among early grammarians and the grammars they wrote, but a common point for most of them was their prescriptivist tone. Prescriptivism is defined by the Oxford English Dictionary as “The practice or advocacy of prescriptive grammar; the belief that the grammar of a language should lay down rules to which usage must conform.” (s. n. prescriptivism). Prescriptivism reached its peak in eighteenth century grammars, with Robert Lowth (1710-1787) - author of one of the most prominent grammars of the eighteenth century, A Short Introduction to English Grammar (1763) - being its highest exponent (Tieken-Boon van Ostade 2010); while the other side of the coin, descriptivism (describing how language is used without imposing rules) became the preferred trend from the nineteenth century onwards. Vorlat (1979) differentiates three types of grammars.

(1) descriptive registration of the language, without value judgements and including ideally - as a very strong claim - all language varieties; (2) normative grammar, still based on language use, but favoring the language of one or more social or regional groups and more than one written with a pedagogical purpose; (3) prescriptive grammar, not based on usage but on a set of logical (and other) criteria. (Vorlat 1979: 129, cited in Yáñez-Bouza 2014: 15).

Some experts, however, see the differences among grammars not as a binary opposition, but rather a continuum, as the differentiation between normative and prescriptive grammars is usually blurred; nonetheless, they are usually put together under the prescriptive category. Historically, and in general terms, the seventeenth and nineteenth centuries correlate with descriptivism, while the eighteenth century is the era of prescriptivism (Yáñez-Bouza 2014).

A lot of constructions that have become a part of the present-day English standard were criticised by prescriptivist grammarians; in other cases, prescriptivism successfully stopped people from using certain forms. The aim of the present paper is to contribute to the field of Normative Linguistics (see section 2.2) with a new case study on 17th-, 18th-, 19th-, and 20th-century grammarians’ attitudes towards the construction it is me, with an object case pronoun, and nowadays part of the standard - as opposed to it is I, with a subject case pronoun and traditionally regarded as the correct form. This study also focuses on the effects of prescriptivism on actual language use via the analysis of corpora of historical written and spoken English, to determine whether prescriptive comment had any effect on usage, and how and who it affected.

2. Literature review

2.1 English grammars, grammarians and grammar writing: the creation of the standard.

“Standardization is a process that is constantly in progress in any language that undergoes it” (Milroy 1994: 19). In order to discuss prescriptivism and grammar writing, it is important to start by describing the standard, as this is the variant prescriptive grammarians will refer back to in their comments - how it is created, what it constitutes, and how it may or may not change over the years.

There are at least two separate definitions of the standard according to Stein (1994): firstly, the standard is the result of a process of standardization, and it is the variety of the language that carries the most prestige, thus being used as written, literary, and religious language, among other uses listed by Görlach in 1988 (as cited in Stein, 1994: 2), and that is, essentially, a national language. This seems to be the more widespread notion of the standard. A second definition of the standard was created regarding countries that have no nationally accepted varieties. In these cases, the attitudes of the speech community towards a specific variety are what conforms the standardization process for the most part (Garvin 1964: 522) - the forms that are widely regarded as correct or educated choices (usually the ones used by the upper classes of society) are what constitutes the standard in this second notion. The differences between these two definitions is that the first one is far more technical, as it implies undergoing a process of codification. The second one, on the other hand, is sometimes regarded as a definition of different standards, but not standard language (Stein 1994: 3). This is the case of, or example, the fourteenth-century Chancery standard, a written form of English used by government bureaucracy and other official purposes, according to the 1066 and all that website (created by Alex Chubarov, senior lecturer at the University of Coventry). This standard constituted a norm but was not a standard language in the technical sense of the definition, as this also implies a certain “sense of nationhood or common identity” (Milroy 1994: 19).

Roger Lass argues that a “successful standard must be widely comprehensive, socially highly valued, and ‘codified’ to some extent” (Lass 1987: 65). The creation of a standard must follow two subsequent steps in its codification - that is, the laying of language rules: usage, definitions and pronunciation. The two stages of codification are selection and regulation (Lass 1994: 82). The selections stage implies choosing a dialectal type - be it a regional or social construct - to work as the basis for the new standard (usually the speech of the upper class living in a capital city). Regulation, on the other hand, describes the process of development of a normative form and acknowledgment of minimal variations, which then becomes de supraregional form. This is, for example, the case of Received Pronunciation in present-day British English.

Following Milroy’s (1994) guidelines, a standard, in a strict sense, must fulfil the following conditions:

a) It is imposed from above;
b) It involves legislation;
c) It contains and ideology of standardization.

The third condition is probably the most salient one in the case of English, and it is also the one that relates directly with sociological issues (Stein 1994: 4). The eighteenth century saw a rise on the interest for correctness and the setting of a standard, which is why “the pace of comment on the English language quickened” (Klein 1994: 31) during that era. Grammar writing flourished and prescriptive commentary abounded. There seems to be a consensus of opinion that the process of standardization for the English language started in the late seventeenth and eighteenth centuries, and the process was mostly completed by the 1800s (Lass 1994: 81). Standardization in English was very much a matter of choices: variants that were rejected and the ones that became a part of the standard. The next section will focus on some of these dichotomies.

One of the main criteria for deeming a certain word or structure correct rather than ungrammatical was its alleged politeness, related to the status of those who used the word, and the settings in which it was used. The English Standard was, then, shaped out by the speech of the upper classes. There was a volition “to polish and refine the English tongue” (Defoe 1697: 223, cited in Klein 1994: 31). Politeness had to do with all aspects of social interactions, and it assumed a certain word, mannerism, or behaviour could have serious political, moral, and social implications (Klein 1994: 43). This is also why a lot of the features of the language considered ungrammatical by grammarians were labelled as vulgarisms, and why a lot of criticism had to do with their perceived informality.

Grammar-writing is key in the process of codification and standardization of a language, as these books are the ones that lay down the rules of usage of grammar, and also lexicon, and elocution during historical eras of English, especially the 18th century. Dictionaries and other language manuals are also important steps in the road to standardization. However, this paper is only concerned with grammars; a good account for the development of dictionary-writing during the late modern English period can be found in Beal (2004: 35-65).

It has been a question for the past years whether the first grammarians, and thus the ones who would more heavily inforce prescriptivism, had any authority whatsoever in dictating rules of language usage. Nowadays, grammarians are language experts in the sense that they usually hold a linguistics degree and publish their language research in professional journals, while they also belong to professional linguists associations. In the eighteenth century, when prescriptivism was advocated for, however, grammarians were not language experts, “simply because experts, as they have come to be considered nowadays, did not exist for practically any field” (Chapman 2008: 21). Taking these limitations into account, it could be argued that the grammarians were, indeed, language experts who paved the way for modern linguists; and while present-day linguists also normally reject prescriptivism, the ideas imposed by 18th-century prescriptivist grammarians are still engrained in the collective notion of the standard, as is the case in point in the present paper.

“For present-day linguists, most eighteenth-century views on language change and correctness look antiquated and naïve, yet those ideas remain preserved in today’s prescriptivist tradition” (Chapman 2008: 35).

Even if grammarians were not expertly trained in the working of the language, their opinions and comments - whether communal or individual - still remain relevant to this day.

On a final note, it is also important to note that a lot of the time grammarians shared linguistic views because they must have belonged to the same discourse community, as Watts (2008) argues and Yáñez-Bouza (2008) corroborates in her study. A discourse community is composed of a set of individuals who reveal common interests, goals and beliefs through their discourse practices, and their works presents a certain degree of institutionalisation (Watts 2008: 51). So even if there was no standard at the time the grammarians were creating their works, they were creating their own inside their own community, which was then imposed into anyone who would read their grammars via prescriptivist comment.

2.2 The field of Normative Linguistics.

Normative Linguistics is a recent field of study that developed in the last few decades - there were rarely any studies dealing with prescriptivism and its effects before the 1970s - and has quickly gained weight among the linguistic community, developing a series of regular conferences - Perspectives on Prescriptivism (2003, 2006, 2009, 2013), as reflected in Yáñez-Bouza (2014: 12). The aims of this field are to gain insight into grammar writing in the history of English: what were the factors that led to grammar writing without the creation of an academy of the language, what was considered to be Standard English during the Modern English era (1500-1900), what were the rules that shaped the present-day standard, and how were they justified. It is also interesting to see how prescriptivist rules compare to present-day Standard English.

Another major aim in Normative Linguistics is the volition to assess (or rather, reassess) the alleged influence of prescriptivism on actual language use: did speakers and writers adhere to these rules? Which social groups paid more attention to prescriptivism? And why?

The field presents a cross-disciplinary approach to the study of the language. The framework is comprised by both Historical Linguistics, because the periods that interest this field of study are the Early Modern and Late Modern English periods (1500-1700 and 1700-1900 respectively); and also Sociolinguistics, because it studies language variation and change. The context where the field operates is that of Grammaticology, as Normative Linguistics focuses on grammar writing and also extracts most of its data from grammars. Finally, the methodology used in the research relates this field to that of Corpus Linguistics, because corpora is used as evidence.

There are two different types of corpora that are used in Normative Linguistics: usage corpora and precept corpora; and I will be using both of them in the present paper. The usage corpus represents actual language use, normally from written texts, as spoken records of historical English are almost impossible to obtain. The data used for the usage corpus is usually readily available, already compiled corpora, e.g. ARCHER (A Representative Corpus of Historical English Registers) or the Helsinki Corpus. The researcher can choose to focus on an individual author or group of authors, or a particular text type; and they could also make use of self-compiled corpora, like a collection of private correspondence. For the usage corpus it is also important to take into account the time gap, that is, the amount of time that it would have taken for a prescriptivist comment to have its effect - if a grammar was written in 1756, the effects would be seen a couple of years later rather than immediately. Not only that, but it is also crucial to observe the trends before and after the period of the precept corpus (see next paragraph), to observe the change in trends - if any.

As for the precept corpus, this is the corpus that studies the norms and rules of the language, usually in the form of grammar books, but also writing manuals, dictionaries or treatises of rhetoric. These type of corpora are usually self-compiled making use of online collections such as ECCO (Eighteenth Century Collections Online), since the focus of these analyses is usually on the 18th century due to the boom in grammar writing and the start of the codification and prescription stages in the process of standardisation.

There have been several large scale studies aiming to assess the effects of linguistic thought on actual language usage, and re-assess language myths with empirical evidence. Some linguists have placed emphasis on the importance of this type of study, as proven on the following quote.

Data on the effectiveness of prescriptivism and the reasons for its success or failure are clearly needed since some investigators view the work of prescriptive linguistics as misguided and futile, citing cases where the linguistic community has ignored prescriptions, whereas other view prescriptivism as awesomely successful […]. MacKay

(1980: 364, cited in Auer & González-Díaz 2005: 317)

Three different effects of prescriptivist comment on language usage have been identified in the literature.

Precept triggering language change.

The effects could have been permanent or temporary. A permanent example would be the loss of morphological levelling in the past and past participle paradigm of strong verbs, e.g. sing, sang, sang becoming sing, sang, sung; although Milroy and Milroy (2012) claim that this variation is still attested in some British English dialects, and it is said to have gained social ground among London middle-class teenagers in the 1990s, which could move this phenomenon into the temporary category.

A phenomena that falls, without a doubt, into the temporary category of precept as a trigger for language change is that of the inflectional subjunctive, e.g. if she were really interested, she would have called. The state of the subjunctive throughout the history of English has been extensively studied, but there was a gap in the literature corresponding to the Late Modern English period, which was filled by Auer and González-Díaz (2005). In the 18th century, Auer & González-Díaz say, the subjunctive was not recognised as a separate mood by all grammarians, and only some of them (out of their 27-grammar precept corpus) commented on the development of the inflectional subjunctive, noticing its decline while advocating for its use, and associating it with politeness in Britain, thus giving the use of the subjunctive a sense of social position and level of education. In their usage corpus analysis, Auer & González-Díaz find that educated letter writers often used the inflective subjunctive, which is explained by the fact that letters were considered a literary form in the eighteenth century. In other genres, however, there was a progressive upward trend during the second half of the 18th century and/or the first half of the 19th century; and “these progressions strongly suggest that eighteenth-century grammarians’ comments were effective in influencing the development of the inflectional subjunctive” (Auer & González-Díaz 2005: 325). From this study we can see that prescriptivist comments take time to reach language use, but they eventually do it; even if on a non- permanent basis, since the use of the subjunctive has, once again, declined from the 19th century onwards.

Precept reinforcing an existing trend in the language - the change was already taking place in the community.

In these cases, the stigmatised form is usually still used in colloquial or non-standard present-day English. The effects of prescriptivism could, once again, be permanent or temporary.

In the eighteenth-century, you was and you were (as singular forms) functioned as paradigmatic alternatives, regardless of degree of formality or politeness. However, a change “originated below the level of social consciousness” (Laitinen 2009: 215) started taking place in the community, and the development on the use of you was “was thwarted by the developing normative grammar tradition which played a vital role in the stagmatisation of you was, […] limited to non-standard speech from the mid to late eighteenth century onwards” (Laitinen 2009: 215), which makes this phenomenon fall into the permanent category of change by precept.

On the other side of the coin we can find preposition stranding, i.e. when the preposition happens in a place other than adjacent to its object, usually at the end of the sentence.

This syntactic arrangement was heavily criticised by eighteenth-century grammarians, even if it successfully survived until the present day. Yáñez-Bouza studied “the various ways in which preposition stranding was understood and commented upon within the eighteenth-century normative grammatical tradition” (2008: 252) and concluded that this was, indeed, a matter of debate in the eighteenth century - “preposition stranding is something to talk about, then as well as now” (Yáñez-Bouza 2008: 277).

Precept having a marginal effect, or none at all.

In these cases, prescriptive comment had no effect - or a moderate one - on the constructions being advocated or criticised. One example is that of phrasal verbs, analysed by Rodríguez-Puente (2012) in a usage corpus-based study that concluded that, despite grammarians’ remarks opposing the use of phrasal verbs, their occurrence or non- occurrence is usually related to the type of text one encounters; much as it is today, eighteenth- and nineteenth-century writers also preferred the use of phrasal verbs in informal settings.

The present study is also expected to fall under that category, since it is known, from a present-day point of view that the construction it is me has successfully entered the standard.

2.3 Pilot study.

Last year during my undergraduate degree I carried out a study that has served as the basis for the present paper. In that study, supervised by Dr Nuria Yáñez-Bouza, an expert on the field of Normative Linguistics, I examined 30 grammars from the late Modern English period; 14 from the eighteenth century and 16 from the nineteenth century. The present study is not only an expansion on my undergraduate dissertation, i.e. multiplying the number of grammars that were considered in the analysis, but also an enhancement, as usage corpora was not taken into account in the previous study. The data used in the pilot study will also be analysed again and recoded.

The methodology used in the pilot study is similar to the one I have used for the Precept Corpus part in the present paper, which will be explained in section 4.1. The corpus was self-compiled and all the grammars were accessed online or on pdf scanned versions that my supervisor provided me with. The texts were read through, paying attention to specific sections where the issue at hand was most likely to be mentioned. Then, a database was constructed with all the resulting variables.

The variables included for the analysis of the grammars in the pilot study are the same ones that will be looked at in the present paper: diachronic change in attitudes, labelling of the construction, reasons given to either reject or advocate for it is me, the topic or section of the grammar where the issue was mentioned, the target audience of the grammar, and the grammarian’s place of birth and gender. These parameters will be explained in further detail in the methodology section (4) of the present paper.

The main conclusions drawn from this analysis were as follows. Diachronically, negative attitudes towards the use of it is me decrease the closest we get to the present day, even if advocating for the use of it is me is still rare - only 3% of the grammars on the corpus display this attitude, all of them dating from the second half of the nineteenth century. The overall results, nonetheless, show that a lot of grammars do not mention the issue at all, which could mean that it wasn’t as popular as one might think, or maybe due to an understanding that the construction was so vulgar that there was no interest in acknowledging the variation in a linguistic text.

[...]