Data and method
This study aims at showing how third person singular pronominal reference (he, he or she, they and their inflected forms him, her, his, hers, herself, etc. which have the same semantic ranges) is reflected in newspapers and in academic writing. More specifically, it focuses on the relation between a dependent variable (i.e. a form of the epicene pronoun) and two independent variables: a) the linguistic context (i.e. type of antecedent) and b) the type of genre (journalistic and academic). The pronominal forms taken into account in the analysis are he, he or she and they, and their inflected forms. Combinations of he and she (for example, she or he, s/he, etc.) were all counted as instances of he or she. The pronouns she, it and impersonal they have not been taken into account as their frequency of occurrence in the corpus is very low.
The analysis is based on two different corpora of contemporary English language writing: a corpus of academic writing and a corpus of journalistic writing.
The collection of academic texts comes from textbooks from the social sciences covering the period 1995-2005. The texts analysed make up a corpus of 2,500,000 words and comprise the following areas of study: Business Administration, Economics, History, Law, Linguistics, Media Studies, Psychology, Anthropology, Political Sciences and Sociology.
The newspaper corpus has 2,500,000 words and comprises online editions of UK and US newspapers from the first half of 2007. It consists of a balanced sample of tabloid and broadsheet dailies including The Sun, The Daily Mirror, LA Daily News, The Washington Post, The Independent, The New York Times, The Daily Star and The US Financial Times. This wide selection of newspapers has been made with a view to minimising the possible effects of editorial policies in matters of language usage.
In all newspapers, advertisements were disregarded and quotations of direct speech were not considered in the search for epicene pronouns.
Content analysis (Krippendorff 1980) has been employed to explore and observe how epicene pronominal forms are used in these publications and how these forms can be correlated with the type of linguistic context and with the type of genre. The categorisation of the items collected for the analysis has been expressed in terms of manifest content, that is, those items-the different epicene variants-which are physically present in the texts and which can be regarded as low-inference items.
The analysis has been carried out taking into account the occurrence of the three pronominal variants with three different types of antecedents:
a) indefinite, when the noun phrase is preceded by the indefinite article a/an:
e.g. “An individual is entitled to have access to his records.” (Social
Science textbook, March 2004)
b) definite, when the noun phrase is preceded by the definite article the or a possessive adjective :
e.g. “...it is as if the reader himself is being warned against
making judgments.” (The Independent, May 2007)
c) quantifiers, forms such as nobody/no one, each, any, every, etc.:
e.g. “No one wants to take big risks with their money.” (The
Washington Post, July 2007)