British Pronoun Use, Prescription, and Processing

Linguistic and Social Influences Affecting 'They' and 'He'

L. Paterson

• Publisher: Palgrave Macmillan
• Year: 2014
• Print ISBN: 9781137332721
• eBook ISBN: 9781137332738
• Topic: Languages & Linguistics
• Subtopic: Sociolinguistics
• Index: Languages & Linguistics

This study considers the use of they and he for generic reference in post-2000 written British English. The analysis is framed by a consideration of language-internal factors, such as syntactic agreement, and language-external factors, which include traditional grammatical prescriptivism and the language reforms resulting from second-wave feminism.

1

Exploring Epicene Pronouns in History

The standard British English personal pronouns have been relatively fixed for the last 200 years. The pronouns are marked for person as there is a distinction between the first-person forms used for the speaker/writer (referred to as I, me, my, mine, and myself) and the forms used for their addressee(s) (you, your, yours, yourself, and yourselves) which constitute the second person. Finally, the third-person pronouns (he, him, his, himself, she, her, hers, herself, they, them, their, theirs, themselves, themself, it, its, and itself) refer to participants external to interaction. Unusually for the English language, the pronouns are also marked for case – for example nominative he is different from accusative him – which indicates a pronoun’s relationship to the verb in a clause. The personal pronouns are also marked for number and have singular and plural forms. For example, when a speaker/writer refers to themselves plus another participant, the plural form we is used instead of the singular form I. There are some exceptions in the second person. For example, you does not change its form between the singular and the plural, and as a result you can refer to an individual or a group, depending on context. Also, you does not change case form depending on whether it is the subject or object of a verb. So we can have the sentences ‘You hit me’ and ‘I hit you’ where the first-person pronoun changes depending on who is doing the hitting, but the second-person form stays constant. However, despite these discrepancies, person, number, and case marking are all quite robust features of the pronoun paradigm.

Whilst number and person marking are predominantly consistent across the paradigm, the third-person singular (animate) forms are the only forms obligatorily marked for gender. As a result, and as illustrated in the introduction, the use of any singular animate antecedent coindexed with a third-person pronoun forces a choice between he and she, whether or not the biological sex of the intended referent is known. However, gender marking on the third person is rare cross-linguistically (see Bhat 2004), and thus it is not a language universal (Greenberg 1963) needed for effective communication. This forced choice of grammatical gender and its relation to biological sex – which is anomalous to the rest of the English language – alongside the corresponding absence of a gender-neutral third-person singular pronoun are the primary focal points of this book.

The expectation that British English would have a standard gender- neutral third-person singular form comes from the uniformity of the rest of the paradigm. This choice of gender/sex is not an issue for first- or second-person forms. Why should it be the case that the third-person singular forms are different? Why must a person referred to in the third-person singular be marked for what amounts to biological sex (as grammatical gender is not part of present-day British English)? One potential way to get around this issue would be if third-person pronouns were selected based on gender performance (Butler 1990). Then the choice between he and she would be relatively straightforward. However, given the apparent difficulties that some writers have when selecting third-person pronouns for people who are transgender/transsexual (see Love 2004: 91), it is clear that the selection of he or she is based on a person’s perceived sex at birth:

(1) I still occasionally slip and call my daughter he (Griggs 1998: 65).

(2) I use both him/her and he/she when talking about Betty, and which pronoun I use is sometimes determined by context but sometimes just by how she looked in my head when I was thinking of him (Boyd 2007: 18).

(3) We like your asking much better than if you guess and get it wrong, and we get especially unhappy if you use the pronoun ‘it’ (Herman 2009: 51).

Nevertheless, even if pronoun choice were based on performative gender, there would still be no clear-cut way to select a pronoun for a generic antecedent, such as parent, judge or celebrity, none of which are marked for grammatical gender in their lexical specification. Appropriate pronoun selection would only be possible for definite singular referents whose gender performance was already known to a speaker or writer. There would still be an issue when one had to refer to a referent of unknown sex or gender.

The absence of a third-person singular gender-neutral animate pronoun can be theorised using Hartmann and James’ (1998: 84) definition of a lexical gap – an ‘absence of a word to express a particular meaning’. According to Cruse (2006: 93) such gaps are hypothesised when ‘a language might be expected to have a word and express a particular idea, but no such word exists’. Furthermore, Cruse (2006: 93) argues that a ‘lexical gap has to be internally motivated: typically, it results from a nearly-consistent structural pattern in the language which in exceptional cases is not followed’. In terms of pronouns the exceptional case in question is the overt marking of gender in the third-person singular, which is anomalous to the rest of the paradigm. This means that there is a potential gap in the pronoun paradigm for a gender-neutral animate third-person singular form which corresponds to the lack of gender marking across the rest of the paradigm. However, subscription to the notion of a lexical gap does not underpin all research on gender-neutral pronouns: Newman (1992: 228) cites Corbett’s (1991) study which suggested that epicene (gender-neutral) pronouns ‘are very rare – if they exist at all’ and argues that there may not be a gap in the pronoun paradigm, dismissing the concept that ‘a gap must be filled or avoided’ as a ‘misguided notion’ (Newman 1992: 469). Corbett’s argument is returned to later in this chapter.

In order to refer to gender-neutral animate third-person singular pronouns succinctly, the term epicene is used here, following Newman (1998: 357), for pronouns that are ‘coreferent with a singular antecedent’, which in turn refers to ‘a referent of unknown or interdeterminate sex’. Thus, an epicene pronoun is classified as a pronoun which does not convey gender or sex information and is coindexed with an animate singular antecedent. Working with this definition, the focus here is on two candidates for epicene status that have been widely debated in discussions of epicene pronouns (henceforth the epicene debate). Generic he (4) refers to the use of the masculine singular pronoun taken to include the feminine if an antecedent is of unknown sex. Singular they (5) refers to the use of the formally plural pronoun with singular antecedents.

(4) The child should do his homework as soon as he arrives back from school.

The baker must always wash his hands before touching the dough.

(5) The child should do their homework as soon as they arrive back from school.

The baker must always wash their hands before touching the dough.

The first section of this chapter contextualises epicene pronouns within the whole personal pronoun paradigm. Significant to the epicene debate is the classification of the personal pronouns as a closed-class of words, an argument which goes almost unanimously unchallenged in previous scholarship. Closed-classes are resistant to change and unlikely to admit new members. If the paradigm is fixed then surely it will be impossible to fill the potential gap for a third-person singular animate gender-neutral form and epicene research would be null and void. However, a review of the history of the personal pronouns illustrates that, despite their closed-class status, paradigm changes have occurred throughout the development of the English language. Furthermore, when attested pronoun changes are analysed together it becomes clear that a combination of language-internal and language-external forces can work together to influence the content of the closed-class personal pronoun paradigm.

Considering these patterns of change in relation to singular they and generic he, and drawing on theories of language acquisition, supports the proposal that the acquisition of singular they is entirely possible (whereas there are problems with generic he). The paradigm as a whole is rote learned during L1 acquisition based on the input that children receive, meaning it is not possible to apply other acquisition techniques, such as overgeneralisation, to the personal pronouns due to their irregular forms and case marking. Furthermore, theories of closed-class acquisition may explain why some pronoun changes have been integrated into the personal pronouns and been adopted into the standard paradigm whilst others have not.

The second section of the chapter focuses on how, despite their closed-class status, third-person singular pronouns are atypical in their characteristics. Although they display some characteristics that are common to closed-classes, such as their use as function words, the pronouns do not possess all characteristics associated with closed-classes, especially in relation to the supposed low semantic value of closed-class forms. For example, the empirical data discussed in the second section of this chapter indicates that there is a high level of consensus across scholars in support of the position that generic he has a default masculine value. This ‘inherently masculine meaning’ (Newman 1992: 453) suggests that the pronoun carries its own semantic value, which is highly unusual for a closed-class word.

In the final section of the chapter singular they and generic he are directly compared. By focusing on how the two candidates (dis)agree with different types of antecedent the syntactic differences between the two primary contenders for epicene status are highlighted. The semantic value of he is difficult to avoid and the form is only rarely used in coreference with an indefinite pronoun. On the other hand, they is used across a range of antecedent types, from indefinite pronouns to definite NPs, and, when used as a singular form, only selects one potential referent. Potential antecedents for singular they and generic he are split into four categories (indefinite pronouns, NPs with quantifiers, indefinite NPs and definite NPs) in order to analyse whether syntactic definiteness impacts upon epicene choice. However, syntactic definiteness is just one of the issues considered here. Firstly, the closed-class nature of the personal pronouns is examined.

Characteristics of the personal pronouns

It is widely established within linguistics that languages are made up of open- and closed-classes (Schultz 1975; Shapiro and Jensen 1986; Cutler 1993; Chafetz 1994; Segalowitz and Lane 2000; Rizzi 2004). Furthermore, it is generally accepted that the English personal pronoun paradigm is an example of a closed-class because it consists of a set of fixed function words, which are used to express grammatical relations. Thus, on the surface, the personal pronouns fit neatly into Herron and Bates’ (1997: 217) definition that ‘closed-class words are used primarily to express grammatical and semantic relations, are generally very high in frequency and low in semantic content’. However, the line between open- and closed-class words is not a language universal, and Geurts (2000: 728) notes that ‘there is no sharp divide between content words and function words’. Similarly, Rizzi (2004: 438–9) suggests that there are ‘some intermediate cases’ where forms can display both open- and closed-class characteristics.

Despite the potentially shifting boundaries between open- and closed-classes, there is evidence that the distinction between the two is more than just linguistic theory, as results from brain-imaging studies suggest that open- and closed-classes are processed differently (Neville, Mills and Lawson 1992; ter Keurs, Brown, Hagoort, and Stegeman 1999). In order to test how the brain responds to open- and closed-class words Shapiro and Jensen (1986: 321) investigated whether the left hemisphere ‘selectively supports a distinct closed class recognition device’ (1986: 321), making it faster at processing closed-class forms than open-class forms. They presented 16 participants with 48 non-words – these are words created from English roots but which do not occur in the language – to test their processing time. In Shapiro and Jensen’s study the non-words had either an open- or a closed-class word as their stem, as in lostner and mostner respectively (1986: 321). They tested one visual field at a time based on the assumption that data presented to the right eye is initially processed by the left hemisphere and vice versa. Shapiro and Jensen’s results showed that differences in reaction times for identifying open- and closed-class words were ‘more robust for presentations to the RVF [right visual field]’ (1986: 323). In other words, the left hemisphere of the brain was more sensitive to the open/closed-class distinction, a finding which Shapiro and Jensen linked to its ‘role in processing grammatical structure’ (1986: 324). Their results support the general hypothesis that the brain processes open- and closed-class words differently.

Pulvermuller (1999) suggested that ‘lexical access for function words involves the perisylvian region’ (Segalowitz and Lane 2000: 378), which supports the argument ‘that distinct neural structures’ are used when processing open- and closed-class forms (Rizzi 2004: 440). Segalowitz and Lane state that ‘the mental lexicon respects the differences between the two word classes in some way’ (2000: 377), and Rizzi (2004: 440) argues that open- and closed-classes are ‘dissociable in acquisition and pathology, accessed differently in language use’. Extrapolating this theory to epicenes, it can be argued that pronouns, as a closed-class, are processed differently from open-class words, which in turn suggests that the closed-class nature of pronouns is a key issue within the epicene debate.

One of the primary differences between open- and closed-classes is that the former, such as nouns and verbs, easily admit new members, whilst the closed-classes, or function words, such as prepositions and determiners, help to create grammatical structure and are resistant to change. In the case of pronouns this grammatical structure comes in the form of case marking, which signifies the relationship between the pronoun and the verb (nominative I and accusative me, for example). Characteristically, closed-classes are ‘highly resistant to change’ – a view based, at least partially, on the premise that ‘pronouns and other function words, are parts of the basic machinery of the language’ (Green 1977: 152). Considering epicene forms specifically, Segalowitz and Lane (2000: 376) argue that although a gender-neutral third-person pronoun would be useful, adding to a closed-class ‘through cultural change is very slow...

Table of contents

• Cover
• Title
• Copyright
• Contents
• List of Tables
• List of Figures
• Acknowledgements
• Introduction
• 1 Exploring Epicene Pronouns in History
• 2 Epicenes in the Twenty-First Century
• 3 Epicenes and Social Movements
• 4 Prescriptions, Standards, and Epicenes
• 5 Accounting for Epicene Choice
• Conclusion
• Notes
• Glossary
• References
• Index