1.1 What is this book about?
Evaluation refers to the expression of attitudes, opinions and feelings (Thompson & Hunston, 2000). Because it is “both centrally significant and pervasive in communicative discourse” (Partington, Duguid & Taylor, 2013, p. 47), delving deeply into how it works in discourse can provide a unifying perspective from which to understand fully how different text types are linguistically and rhetorically constructed.
This book is about the pivotal and yet dynamic role of evaluation in the construction of three major newspaper genres: hard news stories, editorials and feature articles, as represented in the South China Morning Post of Hong Kong (henceforth the SCMP). In other words, it explores how the three newspaper genres are differently shaped and defined by evaluation. To this end, it looks closely at how journalists employ explicitly evaluative language in multiple ways to perform a wide range of discourse functions. In so doing, this book sheds new light on key aspects of newspaper discourse from the perspective of evaluation, notably,
- the multi-functionality of evaluation within newspaper discourse,
- genre specificity within newspaper discourse and
- topic specificity within newspaper discourse.
Alongside this threefold goal is the overarching question:
In what ways does evaluation contribute to the construction of hard news stories, editorials and feature articles in newspaper discourse?
Evaluation in each of the three newspaper genres will be studied from the perspective of the factor considered most instrumental in its construction (in bold below):
- How does evaluation contribute to story design in hard news stories (Chapter 4)?
- How does evaluation contribute to argumentation in editorials (Chapter 5)?
- How is evaluation used to construct human interest in feature articles (Chapter 6)?
The three newspaper genres will also be compared and contrasted in terms of epistemic evaluation (i.e. boosting and hedging):
- How does epistemic evaluation contribute to constructing the three newspaper genres (Chapter 7)?
To investigate these questions, this book draws on a 900,000-word, comparable corpus of newspaper texts, arranged by genre and topic, and a new framework of analysis with a strong orientation to values in newspaper texts, especially news values. It also places major emphasis on structural analysis of texts in the corpus (henceforth the SCMP Corpus).
1.2 Why focus on a single newspaper?
It may be wondered why the present book focuses on a single newspaper and not on two or more newspapers. No doubt, similarities and differences between newspapers in terms of evaluation use have been the subject of innumerable studies, including Alba-Juez (2017), Bednarek (2006a), Blanco (2016, 2020), Huan (2017) and Pounds (2010). Yet it does not follow that studies devoted to one single newspaper cannot offer a new perspective on evaluation use, as illustrated by the work of Belmonte (2019), Liu (2010) and Mahlberg and O’Donnell (2008). In fact, several innovative corpus-based studies on textual positions in hard news stories (Hoey & O’Donnell, 2008a, 2008b, 2015; O’Donnell, Scott, Mahlberg, & Hoey, 2012) also draw on data taken from a single newspaper.
This book adopts the second approach and explores the role of evaluation in constructing a total of three newspaper genres, as represented in a single newspaper. Founded in 1903, the SCMP is Hong Kong’s largest English-language broadsheet newspaper and widely regarded as the city’s newspaper of record. It is also “arguably the city’s most important title internationally” (McLaughlin, 2020), having “reported on China and Asia for more than a century with global impact” (SCMP, 2021). Yet it does not seem to have received much linguistic attention, especially when it comes to evaluation.
Admittedly, no single newspaper – however influential it is locally or internationally – can claim to be linguistically representative of all newspapers because each newspaper has its own linguistic identity, which is at least partially reflected in its writing style. As Crystal (1997, p. 67) puts it,
Institutions, as well as people, need to be considered in relation to the definition of style as “individual identity”. There are certain distinctive linguistic characteristics of newspaper language, for example, which will be found in all instances of the genre […]; but each paper [newspaper] has its own linguistic identity too, which makes it different from the others.
Strongly implicit in this statement is that in the study of newspaper language, the issue of representativeness needs to be considered in relation to the notion of linguistic variation. (See Section 1.3.1 for more discussion about variation.) In theory, then, the more newspapers we compare in a single study, the more complete picture we can gain of their differences and similarities in terms of such aspects as language use. Yet, despite their strengths, studies of such nature tend to suffer from certain limitations (e.g. underuse of structural analysis) and thus may not be able to reveal everything significant about newspaper discourse.
Although it may also have some potential limitations, focusing on a single newspaper in a book-length study has several distinct advantages. First, this approach allows a more in-depth analysis of how evaluation (both epistemic and affective) contributes to creating hard news stories, editorials and feature articles, which have not previously received the same amount of linguistic scrutiny in the literature. Second, it frees up space to examine under-appreciated aspects of evaluation use, notably textual positions and topic specificity, in a large corpus of newspaper texts, arranged by genre and topic domain. Third, it makes it possible to compare in-depth how evaluation (e.g. epistemic modality) varies across the three different genres in the same newspaper, without the analysis being made unnecessarily complicated by the simultaneous need for detailed comparison with another newspaper.
This is by no means to say that the discussion about evaluation use throughout this book is confined only to the SCMP. While the bulk of the data discussed in this book is derived from that single newspaper, the discussion draws on also the existing large literature on evaluation and newspaper discourse, and reference is made, where appropriate and possible, to findings from other studies in order to pinpoint similarities and differences between the SCMP and other newspapers in terms of evaluation use. Paradoxically, this approach of focusing on a single newspaper, then, promises to add a new dimension to the study of evaluation in newspaper discourse.
1.3 Key terms
Before moving on to the data and methodology used for this book, it is necessary to look at some key terms especially relevant to newspaper discourse:
- variation,
- news,
- newspaper genres and
- evaluation.
They are useful in helping us better appreciate the discussion in the rest of the book.
1.3.1 Variation
This book deals with how evaluation varies across hard news stories, editorials and feature articles, as represented in the SCMP. Hence, much of the discussion that follows (especially in Chapters 4–8) can be seen in terms of variation (or specificity).
Variation (or linguistic variation) refers to linguistic differences observed in different situations (Trask & Stockwell, 2007). It is an integral part of linguistic behaviour and ascribable to different factors such as purpose, participants and topic. Much linguistic variation is “highly systematic” (Reppen, Fitzmaurice, & Biber, 2002, p.vi), and, consequently, texts we use in different settings and for different purposes show “different distributions of linguistic features” (Conrad, 2015, p. 309). Academic texts, for example, are shown to vary across disciplines (e.g. Biber, 2006; Biber & Conrad, 2019; Hyland, 2004a). Newspaper texts are no exception; they show systematic linguistic differences, especially across genres.
That linguistic variation is an integral part of newspaper discourse means that “[t]he language of newspapers is a heterogeneous and complex object of analysis” (Semino, 2009, p. 453). In addition to genre (e.g. editorial writing), other factors also determine newspaper language, including newspaper type (e.g. broadsheet), subject domain/topic (e.g. politics), culture (e.g. Chinese) and individual style. Thus, evaluation is not merely genre-specific. It is also topic-specific, culture-specific, style-specific and so on. In short, variation is an integral part of newspaper discourse, a fact stressed by Crystal (1997, p. 392):
With such a range of content, there is no likelihood of finding a single style of writing used throughout a paper [newspaper], nor of finding linguistic characteristics that are shared by all papers.
All this suggests that variation is a complex and yet worthwhile object of study. Hence this book on variation in evaluative language use within newspaper discourse.
1.3.2 News
Most of us today would probably think of news simply as anything new or information reported or shared through such channels as newspapers...
