Ideology in Disguise: Place Name Metonyms and the Discourse of Newspaper Headlines

by Jenny Lewin-Jones and Mike Webb
University of Worcester; University of Worcester

Sociological Research Online, 18 (4) 18
<http://www.socresonline.org.uk/18/4/18.html>
10.5153/sro.3185

Received: 10 Jan 2013     Accepted: 2 Aug 2013    Published: 30 Nov 2013

Abstract

'Place metonyms' are figures of speech which use place names as shortcuts, for example Whitehall to mean the British civil service, or Europe for the institutions of the European Union. The paper examines place metonyms in the headlines of two British newspapers, the Sun and the Guardian. Using evidence from a 12-month period in 2011–12, a headline-by-headline linguistic analysis is used to work out the denotations and wider connotations of each metonym. This critical discourse approach suggests that such place metonyms in headlines have three problematic effects: firstly they may conceal agency and responsibility within some public bodies, secondly for some social institutions, they give an exaggerated impression of unity and homogeneity, and finally for a further list of institutions, they offer relentless pejorative evaluative colouring. These effects are found not only in the right-of-centre Sun but also to some extent in the more progressive newspaper, the Guardian. The authors speculate that it may be difficult for readers of newspapers to think critically about place metonyms in headlines. In particular, place metonyms may subtly reinforce any impression that public institutions are fixed entities, not susceptible to challenge, and may facilitate the polarised value-judgments that are characteristic of 'headlinese'. Such social constructions support some of the central tenets of neo-liberal, capitalist ideology, and so subtly add to the news media's distorting representations of public matters.

Keywords: Metonyms; Metaphor; Place Names; Media; Headlines; Language; Discourse; Ideology; Nationalism

Introduction

1.1 This paper explores the use of place name metonyms in British newspaper headlines. Place name metonyms (or place metonyms) are tropes or figures of speech in which a place name, such as a country, a city, or a building, represents something associated with that place. For example, in most contexts Hollywood means the US film industry rather than the Los Angeles suburb, and Iraq commonly indicates the Iraq war.

1.2 On the face of it, place metonyms in headlines are simply helpful shortcuts referring to (for example) an industry, an organisation, or an office holder. However, this paper asks whether, with significance beyond this, they present worrying dangers. In particular we are interested in the link between place metonyms and ideology, especially the articulation of and distortion of power structures.

1.3 The inter-relationship between the linguistic and socio-political dimensions of headlines is of particular interest to us, the authors of this paper, because it reflects our academic backgrounds, one of us a linguist and the other a social scientist. Previous collaborations between linguists and social scientists have proved fruitful, for example Sealey and Carter (2004). We therefore use a linguistics-based Critical Discourse Analysis, beginning with the headlines themselves, then considering the ideological and social context within which they operate and what this tells us about the struggle for supremacy of ideas.

1.4 The paper begins by examining the literature on place metonyms and on headlines. We then analyse place metonyms in headlines in two British newspapers, the Sun and the Guardian, considering each metonym's meaning not only in individual headlines, but also, speculatively, the cumulative effect of its use in series of headlines over time. This enables us to identify three problematic effects of place metonyms, and to reflect on their likely impact on readers. Finally we suggest what this might tell us about the contest for power in society and the subtle ways in which a dominant ideology can establish and replicate itself.

Place metonym headlines: literature review and initial questions

2.1 In democratic theory, newspapers are important as a means of informing citizenry and challenging the powerful. Although the circulations of print newspapers are declining, while their on-line versions rise (Key Note 2012; Ofcom 2012), nonetheless print newspapers still have considerable reach and there is a danger that research will start to neglect them. Moreover print and on-line versions use similar news stories and news judgements (Redden & Witschge 2010: 185–6), and although their exact headlines differ, the headlining processes are likely to be similar.

2.2 Newspaper headlines have value for producer and audience, for example in attracting readers' attention, indicating the topic of the story, and suggesting its approach (Semino 2009: 444), or setting its tone (Saxena 2006: 31). However, headlines are not merely at-a-glance summaries to 'tell' a story: they also 'sell' the story (Wheeler 2005: 112). In national newspapers they are therefore usually written by specialist headline-writers conforming to the newspaper's preferred style, for example 'clever, original and witty headlines' in the Guardian, according to its style guide (Marsh & Hodsdon 2010: 152), with fewer features of oral discourse than in the Sun's headlines (Fritz 2007: 38).

2.3 In general, headlines have a truncated style which has been called 'headlinese' (Mårdh 1980). Omitting grammatical words such as determiners and auxiliary verbs, compressed noun phrases or pre-modified nouns are used not just because they are short, but because they are dramatic (Swan 2005: 211–2). Ellis (2001: 55–6) suggests that features and opinion pieces have a higher percentage of headlines in the passive voice, but although there may be some differences between 'hard news' headlines and those for soft news, features, sports, and opinion pieces, it seems useful to consider them together.

2.4 Headlines are based on edited selections and so reflect particular points of view, providing a frame or angle on the rest of the story (Bednarek & Caple 2012: 100). Bagnall (1993: 122) said the tabloid press engages in 'comment-by-headline', while Rafferty (2008: 224) similarly views the Sun's headlines as 'an "end" in themselves, a "take" on the story, rather than a signpost or précis of the report.' Headlines can be quite provocative, featuring, as Conboy (2006: 161;2010: 128) points out, hyperbole and sensationalism. Stylistic devices by which opinions are expressed include pseudo-quotes, absence of attribution, and strong, intense, emotional/evaluative lexis (Bednarek & Caple 2012: 100–101; Semino 2009: 445). Whose views are given prominence does then matter: for example, in one Dutch case study (van Dijk 1988: 227) the authorities were found to dominate headlines, taking first position in them.

2.5 However, it is not always easy to immediately identify the meanings built into headlines, not least because their compressed styles of expression depend on readers having the pragmatic knowledge to be able to decode them (Biber 2003: 180). Indeed some headlines are deliberately opaque or ambiguous, to puzzle the reader or create humour (White 2011; Bucaria 2004), or lack information in order to arouse curiosity (Ifantidou 2009: 699). The looser, 'teasing' style of headline with unclear meanings has been associated with downmarket newspapers (Mårdh 1980: 184–6) and with feature items (Hicks & Holmes 2002: 75), but can be found with news stories and in up-market newspapers too, possibly encouraged by the 'teasing' style of promotion in other media (Lasky 2000: 113). The ambiguity is achieved using a variety of stylistic and rhetorical devices, such as puns and word play, alliteration, rhyme, proverbs and intertextual references (Bednarek & Caple 2012: 100–101; Semino 2009: 445), as well as allusions (Shie 2011) and humour (Bucaria 2004).

2.6 The trope or rhetorical device of specific interest here is metonymy. Metonyms are appealing to the editorial staff as they not only make for varied, interesting language, but are 'condensed expressions' (Jakobson 2002: 42) conveying complex associations with brevity. The relationship in metonymy is not one of resemblance, so they do not require an imaginative leap as metaphor does. Rather the relationship is direct association, or contiguity, within one semantic field (Richardson 2007: 66–68), in other words a metonym 'maps' or 'projects' one concept onto another in the same 'domain' or area of experience (Barcelona 2005), thus acting as an 'index' of the designated idea, the one referred to.

2.7 This paper concerns metonyms which are based on place names. Some place metonyms may be merely innocuous shortcuts with straightforward denotations, along the lines suggested by authors (such as Panther et al. 2009) who identify various categories of place metonyms. Typical categories include 'place for event' (e.g. Iraq for the Iraq war), 'place for industry' (Harley Street for prestige British private medical practice), 'place for institution' (City Hall, or even just the City, for the local council), 'place for person or office-holder and staff' (L'Élysée for the President of France), and 'destination for goal or purpose' (building Jerusalem, for example)

2.8 The trouble with these categorisations is that they do not examine the complexities of what is being represented, for place metonyms also make assumptions and 'generate meanings and understandings' (Cornelissen 2008: 81). Higgins (2004), for example, suggests that in Scottish newspapers Holyrood is a 'location token' where 'concealed in a simple description' are a complex of ideas about Scotland and its Government. More generally, understanding place metonyms relies on knowledge of the cultural context (Musson & Tietze 2004: 1307). This cultural knowledge may reveal that many place metonyms are imbued with wider connotations or associations, some of them problematic. This danger is particularly important as metonymies and the assumptions they carry could become 'automatic, unconscious mappings, pervasive in everyday language' (Barcelona 2000: 5).

2.9 Thus, place name metonymy in headlines is worthy of critical investigation as it may be one of a number of tropes that carry ideological content.

Methodology and data

3.1 To investigate these concerns we examined headlines in two British newspapers: the Sun, a mass market 'tabloid' with the largest circulation and which has a right-of-centre ideological stance, contrasting with the Guardian, an up-market centre-left 'broadsheet'. Using a database from the print versions of the papers, we found all instances of selected place names in headlines of all kinds, whether news items, editorials, sport, or features, for a 12-month period mid 2011 to mid 2012.

3.2 Although the study is a synchronous one, not examining changes over time, we initially chose a 12-month period to ensure that a range of stories would be included in the dataset. However, where there were many headlines featuring a particular place name, we reduced the sample to 40. This could have been done by looking at every nth headline, but in practice we found we still obtained the required variety simply by working with the first 40 of each set.

3.3 We then manually selected those instances where we thought the place name was used as a metonym. Although this approach was time-consuming, like Fairclough (2003: 6) we saw benefits from intensive and detailed qualitative textual analysis, rather than assuming computer corpus-search algorithms are always superior. We could, for example, reflect carefully on the subtleties of each place name and how it was used, and were also able, incidentally, to eliminate false mentions such the instances of Washington which referred to the actor Denzel Washington.

3.4 To decide whether a place name was used as a metonym, we often used a substitution or commutation test (see Halverson & Engene 2010), asking how else the headline could have been worded. For example if Hollywood in a headline could meaningfully be replaced by the phrase the mainstream US film industry, then Hollywood is being used as a metonym.

3.5 One of the most challenging decisions, as previous studies found (for example, Markert & Nissim 2007), is whether or not a country name is a metonym. Country names can indeed often be metonyms rather than geographical location references, though this claim may seem surprising at first sight because the use of country names is so naturalised that it is difficult to recognise the layers of social construction involved. An example of a country name as a metonym is in Nelson's famous message before the battle of Trafalgar: 'England expects every man to do his duty'. Another is in our headlines database:

Jersey is already physically separated from the mainland UK, so the headline only makes sense if Jersey and UK are metonyms for the governments or (sub)nations located there.

3.6 The metonyms we analysed are summarised in the accompanying table. The incidence of any metonym will presumably vary over time: for example, one category from the literature not prominent in our selection was 'place for event'. The period examined, in 2011–12, included many instances of London 2012, or occasionally just London, referring to the 2012 London Olympic Games, which accounts for the high number of mentions of London in the table. In this period there were few other events represented in headlines by place metonyms, although the severe economic depression, austerity and Euro crisis may have increased the references to Brussels.

Table 1: Place metonyms in headlines in the Guardian and the Sun over a 12 month period in 2011–12 (frequency and percentage). Our selection of metonyms, ranked in descending order by number of occurrences in the Guardian
Place name metonym:A:
Guardian:
number of occurrences of the metonym in 12 months**		B:
Sun:
number of occurrences of the metonym in 12 months** 		C:
Guardian:
% metonyms in all mentions of that place
D:
Sun:
% metonyms in all mentions of that place
Britain* 50320671 76
London*473566532
U.S.*336303826
Europe* 2037041 64
Hollywood*123879895
Whitehall 37897 100
Brussels 361692 70
Beijing 31774 100
White House2698769
Downing Street25193 6
Westminster 23372 100
Washington 21175 33
Anfield 13833 53
Essex 126633 65
Berlin1042933
Mecca325252
Buckingham Palace00--
* For Britain, London, U.S., Europe, and Hollywood, the numbers are estimates, because where there were over 40 headlines, a sample of 40 was examined, and the % of metonyms in that sample was used to work out the likely number of metonyms in the total.
** The 'occurrences' columns for the Guardian and the Sun are broadly comparable, except that for the final part of the period the Sun also included a Sunday edition; however, the effect of this on the figures was small, so they have not been adjusted.

3.7 The percentages of place name mentions which are metonyms are shown in columns C and D. Some place names appear almost always as metonyms: Hollywood, Brussels, Whitehall and Westminster, for example. These locales would rarely merit being featured in British newspapers were it not for the institution with which they are associated. However, even with the country name Britain, in around three quarters of the instances it is a metonym for something more specific.

3.8 Thus we began with a paradigmatic approach, examining the selection of metonyms used, and the significance of those choices in preference to non-metonymic expressions. However, the most productive part of the research was when we added in the syntagmatic dimension: here, we considered headline-by-headline the ways in which each metonym was used linguistically in combination with the other elements of the headline, and the semantic impact of that (the impact on meanings). This critical analysis suggested that place metonyms may have three problematic effects: firstly a general concealment of agency within institutions, secondly for some social institutions an exaggerated impression of unity, and finally for a further list of institutions a relentless pejorative evaluative colouring. Our paper now looks at each of these effects in turn.

Concealment of agency within institutions

4.1 The first way in which place metonyms may mislead is by fading out evidence about how organisations work and about who actually takes the decisions in them.

4.2 Even before encountering this problem, interpreting the meaning of a metonym is not straightforward, as in some cases the metonym may have more than one denotation (or metonymic shift). This variation in meaning can be between newspapers. For example, in the Guardian we found that 74% of instances of Beijing meant the Chinese government, whereas in the Sun 70% of the instances referred to the Beijing motor show:

4.3 In those instances, regular readers might anticipate the likely meaning of Beijing on the basis of previous occurrences of the metonym, however this may not always work as metonyms often have different denotations within the same paper, making the deciphering task more complex. In the Guardian, Westminster usually meant the British parliament or parliamentary institutions, but on two occasions it was an abbreviation for Westminster local council. Similarly, in the Sun Essex was used metonymically 22 times, but 2 of these meant Essex cricket club, 9 indicated characteristics associated with Essex, and in 11 cases, matters associated with TV programmes which had Essex in the title.

4.4 Alternative denotations are arguably not a problem if the audience is sufficiently sophisticated to be able to decode them. For instance, Palace has a variety of meanings including Lambeth Palace and Kensington Palace, but in the following headline (not one from our 12-month period) the Sun's sports pages show confidence in readers' ability to interpret its play on words between Buckingham Palace and Crystal Palace football club:

4.5 However, the more serious problem discussed here is that even within a particular metonymic shift, there may be vagueness or fluidity over what the metonym actually denotes. For example, the place metonym Downing Street, referring to the British Prime Minister's residence (no.10), might mean the Prime Minister in person, but it could mean the PM's office, or indeed any or all of the government units which reside in no.10 Downing Street:

4.6 A similar example of a place metonym whose denotation is ambiguous is Whitehall, as in these headlines:

Here it is unclear what ministry or ministries Whitehall refers to, and whether it refers solely to the civil service, or to the ministers with their civil servants. Indeed the final headline almost suggests that Whitehall and Government are sometimes used as synonyms.

4.7 Yet another example of the lack of clarity when a metonym has various, inconsistent meanings is the White House. In the first of the pair of headlines below, White House appears not to include the President (Obama) himself, whereas in the second, White House is used in a way which does focus on the person of the President:

4.8 We therefore question some of the categorisations found in the literature, where metonyms are (for example) differentiated according to whether they stand for 'place for institution' or 'place for people': the reality may be more complex, and it is necessary to think more carefully about which human 'agents' are being referred to. In all the examples so far, the building name in effect deletes the reference to occupants, thus transforming the reference to agency (Musson & Tietze 2004: 1312).

4.9 The lack of clarity about agency is particularly apparent with cities or countries, as with such metonyms there is often no indication whether the metonym refers to the people in the location or a local council or another supposedly representative body, or a conflation of all of them. An example of this is the following, where it is not evident what London refers to:

Similarly with countries, in our headlines we found that Britain as a metonym can mean any of: all the people in Britain, British social life, British public life, British sporting teams and figures, the British Government, the British economy, or sometimes some indeterminate combinations of those – the precise meaning is often blurred.

4.10 An especially opaque metonym is Europe, as in this headline:

Europe here could mean the people of Europe, or the European economy, or it could mean the national governments of European countries involved in the Euro, or the European Union and its institutions, or some or all of those. Even where Europe evidently refers to the European Union, it is rarely clear whether that means the EU member states, or the EU Commission, or some other EU institutions. Europe is also, as in the example above, often juxtaposed with another element of the headline, implying a dichotomous rather than an overlapping relationship between the two.

4.11 The syntax of the headline may add a further layer of concealment: with a passive construction, for example, the metonym is not the subject of the sentence. Of course, passive structures have advantages, saving space, and establishing the topic with immediacy by putting the metonym to the front of the headline. However, the passive construction may conceal who the agent is or hide the fact that the agent is unknown, as in these examples:

Here, as well as failing to specify who or what bodies in the EU are being warned, the passive construction of the headlines changes the emphasis – it is unclear who is doing the warning (it might just be the newspaper itself!). It is also unclear whether the full sentence would be 'is warned', or 'was warned', and such ellipsis, the omission of key words, is a further source of vagueness in headlines.

4.12 Place metonyms' concealment of the responsible agents depersonalises or 'impersonalises' (Machin & Mayr 2012: 79) decision-making. Attention may be further diverted from this abstraction when the metonym is combined with anthropomorphisation, treating it as human. This re-introduces a personal element, but this re-personalisation is arguably fake and over-simplified. Some examples are:

4.13 The Brussels examples referred to earlier (Brussels warned … Brussels urged …) also involve personalisation. The verbs 'warn' and 'urge' would normally be used with a human being on the receiving end, so the use of these verbs adds a metaphorical layer. The 'organisations as humans' metaphor is a widely used one (Cornelissen 2008: 93; Brdar and Brdar-Szabó 2009: 248), and gives Brussels human characteristics. The city name is already being used as a metonym for unspecified institutions of the European Union, so this melding of metaphor and metonymy leads to some complexity.

4.14 The pseudo-personification of an institution is complete when the metaphor is that of a body, with human emotions or physical reactions, for example:

Personalisation may also be used with opposition, and this commonly gives rise to a discourse dominated by metaphors from the source domain of fighting or war, as in these instances:

4.15 Thus, in many cases the use of place metonyms leads to an obfuscation about how institutions work, diverting attention away from the people responsible. They misleadingly suggest that decisions and actions are products of impersonal institutions rather than of individuals within them who could be held to account, and this is sometimes hidden by fake personalisation.

Exaggerated impressions of unity and homogeneity

5.1 The concealment of agency just described opens the way to a second way in which place metonyms may distort meaning: they can hide difference and disagreement, and so falsely imply unity and collectivity where it does not exist.

5.2 We have already mentioned that where the country name is used as a metonym, the implication is that it is more than just a random collection of people and activities in the geographical location: they are bound together as a nation, a common entity. However, an exaggeration of unity within decision processes may also be at work when a place metonym focuses attention on an institution, as for example with the metonym Downing Street in the following headline:

Here, the phrase Downing Street disputes may give the impression that all the Prime Minister's officials are as one in disputing the evidence, whereas it is conceivable that 'behind the scenes' there are groups of people, some of whom take a contrary view.

5.3 The idea that organisations are unified entities may be reinforced by the juxtaposition of one with another, for example:

Somehow in these cases, the binary opposition in the headline leads little room to question whether each of the two elements in the headline does actually operate as a unit.

5.4 The emphasis on unity is often bolstered when a singular verb is used, indicating that the place metonym is being treated as a singular rather than plural noun. Place metonyms naturally push the writer towards this pattern of use as places are usually singular, even when with the actual names (governments or football clubs, for example) it would be more natural to use a plural. For example, in both the Sun and the Guardian, ALL the uses of Whitehall we found in headlines were singular, thus giving the impression of unity, as in this instance:

5.5 Place metonyms' suggestion of unity is also sometimes reinforced by the use of the pronoun we or us (or the adjective our): a kind of positioning process. Sometimes, of course, the pronoun we refers to the newspaper itself:

At other times it can refer to the subject of the article:

5.6 However, commonly the pronoun we seems to refer to the people of Britain (or perhaps the readers of the newspaper – the distinction is unclear), trading on, and reinforcing, a picture of national unity. An example is:

In some headlines, the pronoun we conflates the subject of the article with elements of Britain as a whole, thus further reinforcing the identification of the reader with the (usually national) figures or organisations in the headline:

5.7 A further linguistic feature of headlines that sometimes suggests collectivity is the use of a possessive. The Sun, for instance, often uses Britain's with a superlative in a way which highlights an underlying belief that people in Britain have a common culture, for example

Applying the commutation test, the phrasing of these headlines could have been Friendliest British B&B, Fattest woman in Britain, but the use of Britain's rather than British or in Britain seems to reinforce collectivity, ownership, and national pride.

5.8 By contrast, in cases where the reader will probably not identify with the entity denoted by the place metonym, the possessive form can be used to indicate distance from the reader. In the following headline, the phrasing is Beijing's candidate, not the Beijing candidate, which has the effect of emphasising that the candidate has chosen by, and belongs to, the regime:

5.9 Thus, an additional problematic effect of place metonyms is that they involve generalisation, and especially when used as singular nouns, with particular pronouns, and sometimes as possessives, they present an anodyne view of the nation states and organisational entities which dominate public life, suppressing the likely reality of internal divisions within them. The wider significance of this will be seen later in the paper, but for the moment its impact is to allow further generalisations about the entity described, which leads to our final problem.

Repeated evaluative colouring

6.1 The final way in which place metonyms may give a distorted impression of the world is that, being based on generalisations, they can invite or embody 'evaluative colouring'. The colouring especially occurs when place metonyms develop connotations: associations evoked beyond the metonym's immediate denotation, thus giving it 'emotional charge' (Bloor & Bloor 2007: 129).

6.2 A metonym's emotionally-rich undertones are often built up over a series of usages or headlines, initially via lexical choices associated with it. As Stubbs (2002: 202) points out, '… a connotation is rarely carried by a single word, but is distributed prosodically across a textual sequence'. In examining headlines we found this did not tend to happen through collocation, the repetitive accompaniment with the metonym of one particular word. Rather, a metonym's undertones come from its use over time with a range of either exclusively disparaging or exclusively favourable words. However, the words associated with any particular metonym were often from one lexical field or over-arching conceptual metaphor. A common metaphor for politics was that of a fight, although in the Guardian, elections to the White House were also headlined using the metaphor of a race, with favourites and winners and losers.

6.3 Because metonyms are nouns, so the evaluative colouring typically comes from the verb. For example, Whitehall, a place metonym for the British civil service, is used in the Sun with verbs which suggest how wasteful the civil service is:

Here, the specific instances of extravagance form part of a larger scale framing (a bigger picture) of the civil service as wasteful (and possibly, in the case of the first of those headlines, as a male preserve). The repeated syntactical pattern of Whitehall + verb + amount of money in different headlines on different days reinforces this stance.

6.4 Even more striking is the Sun's use of Brussels, a place metonym for the European Union. In the Sun, Brussels is seen as a powerful external force, interfering with domestic, national concerns. The language used, mostly from the lexical field of fighting or verbal warfare, is blatant and often quite vivid, and the message over a series of headlines particularly relentless:

6.5 The Guardian's lexical accompaniments to the metonym Brussels are more subtle, but nonetheless have an evaluative effect similar to those in the Sun:

Those examples eschew neutral verbs such as tell, say, or requests in favour of verbs which carry illocutionary force or additional meaning (Bednarek 2006: 57): warned and urged imply that Brussels has behaved badly and is being admonished for its actions, demands implies that it is using its might to get its own way.

6.6 In the next example, the verb take on is significant, as the Spanish Prime Minister is seen as challenging scary bully Brussels, almost like David taking on Goliath:

Meanwhile in the following, gave in to Brussels again emphasises a powerful, dominant institution, with the adjective flagrant further suggesting that it is so powerful it can afford to be brazen even when acting outside the law: Brussels is used in such a critical context that Whitehall, itself the subject of pejorative colouring in other headlines, can sometimes be used as a contrast to show up Brussels as unreasonable: In both the Sun and the Guardian, all the headlines about Brussels were about difficulties, failures, warnings, Brussels demanding power, and so on: there were no headlines suggesting that Brussels was (for example) a source of useful solutions or helpful planning.

6.7 Another instance where place metonyms are presented in a critical light are the place metonyms for those national governments which are regarded by the newspaper as particularly pernicious regimes. In the Guardian, for instance, when Beijing is used to mean the Government of China, it is represented as being a belligerent, ambitious entity:

6.8 Perhaps surprisingly, headlines about Washington often have a tone which is as derogatory as the ones featuring Brussels or Beijing. Washington too is seen as powerful and controlling:

It undertakes missions, rather like the terminology in spy movies: Moreover verbs such as admits and claims suggest a regime which is secretive and not to be trusted:

6.9 There is a striking contrast between the critical presentation of Washington as underhand and controlling, and the more favourable treatment of the place metonym the US. In these headlines, the US is often represented as an innocent victim, affected in problems by the actions of others, rather than being the prime mover:

Where the US needs to take action, this is portrayed as a defensive move, signalled here by the choice of the word bolster and the personalisation imagery used:

6.10 Eventually, after an extensive series of headlines over time establishing a particular connotation, a place metonym may come to have those connotations without any help from associated words in the headline. This might explain what had initially been a puzzle to us: the absence in the headlines of any of the stock clichés we had expected, such as Brussels bureaucrats (we found not even one instance of this). However, it may be that the word Brussels automatically brings to mind images of 'bureaucrats', not just because of previous headlines, but possibly also because of another aspect of semantic prosody, inter-textual references: rhetoric such as Brussels bureaucrats may, we speculate, already be familiar to the reader from sources such as politicians' speeches.

6.11 Essex is another example of a place metonym which has come to have connotations without the writer any longer needing to employ evaluative colouring devices, as in these headlines:

Here, Essex (a county near London) has the disparaging connotation of something cheap, vulgar and down-market, but aspirational. The second of those headlines relies on the reader recognising the contrast between Essex and Ecclestone, the latter referring to Bernie Ecclestone, who effectively controlled formula 1 motor-racing and whose name came to represent a wealthy, up-market lifestyle.

6.12 Metonyms such as this draw on generalisations about particular places or whole groups of people, based on the exaggerated impressions of homogeneity previously discussed. Where the metonym is used as an adjective, as in the Essex examples, the generalisation is foregrounded, making it the main focus. Our table in the methodology section above showed that the metonym Essex was used in Sun headlines more than in the Guardian, hinting at the possibility that the Sun may have a particular liking for using place metonyms for the explicit purpose of description and evaluation.

6.13 Thus, evaluative colouring present with many place metonyms means that headlines are even less neutral than they might initially appear.

Possible audience responses

7.1 So the use of place metonyms is problematic: some may over-simplify, mystify or obscure, or appear to be neutral abbreviations while actually giving headlines a particular slant. The crucial question, though, is how readers interpret such representations.

7.2 'Reception' of headlines does matter. Although there are relatively few studies of headline-reading, there are suggestions that many readers focus on headlines more than on the articles themselves (Dor 2003: 696–7). Indeed, Couldry, Livingstone and Markham (2010: 89, 101) found, from diaries kept by media users, that some people get their news by glancing at or 'flicking' through headlines.

7.3 The classic debate is whether audiences decode texts 'actively', in other words thoughtfully, critically and sometimes oppositionally and subversively, or whether they do so 'passively', accepting of the viewpoints presented to them – or, of course, a 'negotiated' position intermediate between these extremes.

7.4 Most theorists in the critical or related traditions suggest the likelihood of at least some, albeit limited, resistance by consumers and by implication, resistant readings of texts. Gramsci, for example, suggests consumers 'redirect' the output of culture industries in ways producers cannot foresee (Storey 2012: 84), Habermas is optimistic about resistance to colonisation of the lifeworld (Habermas 1991; Joseph 2006: 94), and even Foucault, although focussing on subordination, believes that wherever there is power, there may be resistance (Lindgren 2000: 303).

7.5 Thus, some readers may engage with headlines critically: Pursehouse's (1991) interviews showed Sun readers 'negotiate' the interpretation of what they read, while more recently, experiments by Brône and Coulson (2010) suggest that newspaper readers are aware of ambiguity in headlines. Lack of trust in journalists may also engender a scepticism about what they write: as Fenton and Witschge (2011: 155) say, the reduction in trust in journalists 'follows closely on the heels of a history of the marketisation of news that has contributed to the culture of spin, as well as the close relationship of mainstream news to public relations machinery and the power of commercial pressures to guide news content.'

7.6 But it is difficult to be definitive about reader response, as it will vary. Madianou, a noted audience researcher, says that as well as the text itself, interpretation is shaped by factors such as the ideological climate, the individual's social capital, and their exposure to alternative sources (Madianou 2009: 328). A particular reason for suspecting that audience interpretation of headlines may vary is that they are written in language designed to induce emotions (Taiwo 2007: 218), including anger, disgust, fear, joy, sadness, and surprise (Burget et al. 2011), and the personal nature of emotions means that varied responses are likely. Connotation-rich language in particular, common in headlines, may be open to different and shifting associations for different audience members (Goddard 2011: 27).

7.7 Clearly there is a need for more research into the reading of headlines and of place metonyms in them, and to do that Critical Discourse Analysis could draw upon a range of audience-centred instruments such as focus groups and topic-centred interviews (Wodak et al. 2009).

7.8 However, there is good reason to be cautious about claims that audiences are critically active. Anderson and Weymouth (1999: 169) for example, in their study of newspaper discourse about the E.U., see 'consumers of information as being less liberated and autonomous than has been claimed …, and…discourse as being less open to interpretation than some observers…believe.' Readers also have a number of strategies at their disposal to avoid being challenged, for example seeking out sources that reinforce their pre-existing views (Watson 2003: 72), or reading only the sections of the newspaper which interest them.

7.9 Our own experience, as authors of this paper, is that while reading and interpreting the sample headlines, we sometimes found it difficult to penetrate and interpret some headlines. Some, such as the ones involving cities or countries, proved particularly bewildering. To judge by our own response, busy newspaper readers may not find it easy to read place metonyms headlines critically, needing a sophisticated understanding of the political process to interpret what is behind some of them.

7.10 Metonyms based on place names are arguably particularly pernicious because they are so seductive. Working via emotion, they may help to suspend readers' critical faculties, as places have an affective appeal: people tend to have attachments to particular places, and feel alienated from others. Place metonyms explore the complex and unfamiliar by means of the reassuring, familiar and appealing images of places and buildings, perhaps making it more difficult to think critically about what they actually refer to, and some conjure up images of grand buildings or districts suggesting enduring institutions whose basis is difficult to question. The seductive effect of country metonyms may be particularly pronounced: Billig (1995: 108), in his study of the language of nationalism, shows how routinely flagging the homeland in newspaper headlines and first sentences helps to make the homeland homely. All in all, the use of place metonyms may provide a further stumbling block in the way of the reader's ability to critique and react to what they are reading.

7.11 Thus we tentatively think that for many readers it may be a daunting task to see through place metonyms, to penetrate their meaning and resist their assumptions. Although place metonyms are only constructs, their appeal could sometimes lead readers to suspend critical judgment more than would be the case with headlines which did not use them.

Ideological interpretations

8.1 Place metonyms may therefore play a part in distorting thinking about the world, first of all by introducing judgements about it. Essex, for example, has become a label which, often cloaked with humour, is used for class-based and gendered mockery (Biressi & Nunn 2013: 42), and similarly metonyms such as Europe or Beijing may encapsulate and activate stereotypes and prejudices. In general, discourses often delineate who is an insider and who is not (Gee 2012: 158–9; van Dijk 2006: 139), and metonyms with their multiple layers of meaning may help construct those dichotomies.

8.2 Metonyms which refer to favoured establishment institutions, such as Whitehall or Downing Street, present a further problem. Those metonyms offer over-simplifications, short on detail. They 'conjure away' the actors involved (Reisigl & Wodak 2001: 58) and camouflage the likely reality of internal disagreement and division, reifying the organisations referred to and perhaps making them appear impervious to challenge. Thus, they subtly suggest unified forces beyond the control of ordinary citizens, against which resistance would be futile.

8.3 Place metonyms may even mislead by sometimes focussing on the 'wrong' institutions. Newspapers and their headlines are perhaps attracted by the concrete particulars of life, and buildings and places inevitably feature prominently, for places and spaces are fundamentally important to people and their identities. However, in contemporary 'network society' (Castells 2009b; van Dijk 2012), especially in the global world of finance capitalism, power arguably no longer resides in the buildings of government or parliamentary institutions that metonyms tend to focus on. If so, place metonyms may 'ossify' notions of the power of particular institutions, some focussing on locations from which power has ebbed, thus making it even more difficult for citizens to understand and challenge the forces that control their lives.

8.4 We are not suggesting that there is a deliberate conspiratorial strategy to use place metonyms to obscure where power lies, rather that their habitual use leads to that effect. The impetus for including place metonyms in headlines may be no more than writers' search for simplification and interest. However, some of the over-simplifications conveniently support assumptions and distortions at the heart of right-of-centre ideologies, which is at least very convenient for newspapers of that persuasion.

8.5 The nation-state metonyms (such as Britain) seem particularly problematic as they embody all three of the problems we identified, concealment of agency, exaggerated impressions of unity, and evaluative colouring. They facilitate a discourse which reduces the complexity of international events to 'us'; and 'them' (Wodak 2009: 583, 585). Thus many place metonyms, focussing as they do on familiar places, may encourage a parochial, ethno-centric focus on news events The metonym 'Britain', for instance, arguably leads to the conflation of 'the British government' and 'Britain' the country, making it difficult to critique commonplace but ill-founded observations such as 'Britain can't afford it'. The result, all too often, is an in-built tendency to rationalise a regime of relatively low government spending.

8.6 Taken together the impacts we have analysed are congruent with the dominant, right-of-centre or populist ideology. For example, the contrast between the favourable treatment of the place metonym the US and the critical presentation of Washington tends to reflect the perceptions of right-wing populist or Republican viewpoints in the USA, where Federal government and Federal agencies are viewed as an unnatural interfering regime whose 'big government' role needs to be kept to a minimum. Similarly, the connotation-laden use of the metonym Brussels reinforces the more blatant and explicit messages from the right, that those who work in Brussels are meddling bureaucrats. In general, the institutions which most often have pejorative representations are those that people on the right typically fear could be used to challenge the status quo, including government's administrative arm, the civil service.

8.7 Our expectation had been that different newspapers would vary substantially in the way they use place metonyms. The ones we studied, the Guardian and the Sun, did indeed use place metonyms slightly differently. However, our analysis is that the effect was similar in both papers. It was no surprise to us that in a right-of-centre newspaper like the Sun metonyms would distort the reality of power.  However, unexpectedly, we found that the Guardian's place metonyms also tended to mislead. Arguably, that is more pernicious because the effect of the metonyms in the Guardian is more subtle and it is easy to overlook their ideological edge.

8.8 That the effects appear not only with headlines in the right-of-centre Sun, but also with ones in the supposedly liberal and investigative centre or centre-left Guardian suggests the message may be hegemonic, and therefore difficult to challenge. It is particularly insidious that the processes are at work even in the Guardian, whose readers, expecting a liberal newspaper, may not be so alert to the possibility of distortion in the presentation of its news stories. Potentially misleading information about the location of power, and the other problematic uses of place metonyms, does not reflect the ideal of the press informing citizens about the world they live in.

8.9 So what, in general, is the significance of place metonyms? Place metonyms are most likely chosen for headlines because they in some way reflect the newspaper's news values, or selection criteria. To use Conboy's phrase, they are a type of 'mnemonic hook' (Conboy 2007:108) that enable readers to locate events within the newspaper's values. Indeed they are likely to fit several of the standard news values such as cultural proximity, personalisation, prominence, and predictability (the latter, stories which fit an expected pattern). Thus, place metonyms are like most news, not new, but confirming what is already known (Matheson 2005: 18).

8.10 Place metonyms often have connotations, and any language which has connotations can frame a text ideologically (Richardson 2007: 48): figurative language in particular can disguise hidden beliefs (Wodak & Meyer 2009: 8). Place metonyms, with their condensation through nominalisation (Young & Fitzgerald 2006: 264), are perhaps a particular instance of what Marcuse called 'abridged use of language' (Marcuse 1964/1991: ch.4), sliding contradictions past the reader in the guise of harmonious ideas. Marcuse gives an example using a place metonym, showing how a phrase like Georgia's governor can fuse contradictions together 'in one indivisible and mutable structure which, in its innocence and immediacy, overwhelms the reader's mind' (Marcuse 1964/1991: ch.4; also cited in How 2003: 92).

8.11 Herman and Chomsky (1994[1988]) argued that an underlying elite consensus structures all facets of the news, including the framing of issues. More recently, David Harvey (2005: 40–1) suggested media helped construct 'consent' to the newly resurgent neo-liberalism, and its penetration into 'common-sense' understandings. We think that the headlines containing place metonyms play a part, albeit small, in those processes, helping to foster and naturalise a distorted, elitist or class-based view of the world, and reflecting some of the central tenets of neoliberal capitalist ideology.

Conclusion

9.1 In summary, we suggest that small, almost unnoticeable and apparently harmless elements of language can have significant effects. One of the ways in which the dominant neo-liberal ideology remains so utterly resilient is by reaching out into even these minor constituents of stories which frame the world. Metonymy, one such component, was formerly rather over-shadowed by the academic study of metaphor, but is now receiving greater attention (for example Cramer 2008; Bierwiaczonek 2013; Catalano & Waugh 2013), and we are happy to contribute to this re-emphasis.

9.2 We also argue that it is important to keep studying newspapers. Their readerships, even if declining, are still substantial, and the influence of tabloid style has spread, even beyond newspapers (Biressi & Nunn 2007). In the 'remix culture' (Castells 2009a: 132) they are generators of news that filters into other news media, so even as the focus in Sociological research shifts to online social media (see McKie & Ryan 2012), there is still a need to explore newspapers as an input of ideas into those forums.

9.3 Sociology has a key role here, especially with the 'the sociologising of discourse research' (Keller 2013: 1), for Sociology's essential purpose is surely to challenge everyday, 'common-sense', 'taken-for-granted' and ultimately potentially misleading views of the world. Our research has made us optimistic about the value of combining linguistic and sociological analysis to make constructions such as news headlines more transparent. Tropes do not operate in isolation: they work systematically in headlines in conjunction with other linguistic devices, and so it is productive to examine how they are configured, but to consider this alongside the social, political and economic context from which they spring.

9.4 We think the study's methodology had some merits. Despite the advantages of computer analysis of corpora, there are still virtues in close manual analysis of multiple text elements such as headlines, especially via the commutation test: reflecting on how else the text could have been phrased, and the effects of its actual wording. Yet in the context of debates about the possibility of bridging the divide between micro and macro analytic approaches to discourse (e.g. Heller 2001 cited in Baxter 2010: 113; Koteyko 2006), we have also been able to suggest how our linguistic micro analysis might relate to the broader discourse, and to issues of ideology and power.

9.5 We have left it for others to trace the discursive practices themselves: how news headlines get constructed, how they are actually read, what their emotional effects are, and what use people then make of the metonyms from them. There is also much more to be said about the whole of the discourse to which the place metonyms relate, and how it develops diachronically over time.

9.6 Nonetheless, our work has shown that seemingly innocuous but nuanced language can selectively and simplistically reinforce the dominant perspective on matters which should be more open to public debate. By critiquing and exposing such language features we hope to subvert them. Thus we might in a small way be helping people to critically engage with the concepts and constructions they meet, surely the ultimate purpose of both Sociology and Critical Discourse Analysis itself.

References

ANDERSON, P. & Weymouth, A. (1999) Insulting the Public? The British Press and the European Union. Harlow: Addison Wesley Longman.

BAGNALL, N. (1993) Newspaper Language. Oxford: Butterworth-Heinemann.

BARCELONA, A. (2000) 'Introduction: the cognitive theory of metaphor and metonymy', in Barcelona, A. (Ed.), Metaphor and Metonymy at the Crossroads: A Cognitive Perspective. Berlin: Mouton de Gruyter.

BARCELONA, A. (2005) 'Metonymy', in Strasny, P. (Ed.), Routledge Encyclopedia of Linguistics. London: Routledge.

BAXTER, J. (2010) 'Discourse-analytic approaches to text and talk', in Litosseliti, L. (Ed.), Research Methods in Linguistics. London: Continuum.

BEDNAREK, M. (2006) Evaluation in Media Discourse. Analysis of a Newspaper Corpus. London: Continuum.

BEDNAREK, M. & Caple, H. (2012) News Discourse. London: Continuum.

BIBER, D. (2003) 'Compressed noun-phrase structures in newspaper discourse: the competing demands of popularisation vs. economy', in Aitchison, J. & Lewis, D. (Eds.), New Media Language. London: Routledge.

BIERWIACZONEK, B. (2013) Metonymy in Language, Thought and Brain. Sheffield: Equinox.

BILLIG, M. (1995) Banal Nationalism. London, Sage.

BIRESSI, A & Nunn, H. (Eds.) (2007) The Tabloid Culture Reader. Maidenhead: Open University Press.

BIRESSI, A. & Nunn, H. (2013) Class and Contemporary British Culture. Basingstoke: Palgrave Macmillan.

BLOOR, M. & Bloor, T. (2007) The Practice of Critical Discourse Analysis. London: Hodder.

BRDAR, M. & Brdar-Szabó, R. (2009) 'The (non-)metonymic use of place names in English, German, Hungarian, and Croatian', in Panther, K.-U., Thornburg, L.L., & Barcelona, A. (Eds.), Metonymy and Metaphor in Grammar. Amsterdam, John Benjamins.

BRÔNE, G. & Coulson, S. (2010) 'Processing deliberate ambiguity in newspaper headlines: double grounding', Discourse Processes, 47(3) p. 212–236.

BUCARIA, C. (2004) 'Lexical and syntactic ambiguity as a source of humor: the case of newspaper headlines', International Journal of Humor Research, 17(3) p. 279–309.

BURGET, R., Karasek, J. & Smekal, Z. (2011) 'Recognition of emotions in Czech newspaper headlines', Radioengineering, 20(1) p. 39–47.

CASTELLS, M. (2009a) Communication Power. New York: Oxford University Press.

CASTELLS, M. (2009b) The Rise of the Network Society (2nd edition). Chichester: Wiley-Blackwell.

CATALANO, T. & Waugh, L. (2013) 'The language of money: how verbal and visual metonymy shapes public opinion about financial events', International Journal of Language Studies, 7(2) p. 31–60.

CONBOY, M. (2006) Tabloid Britain: Constructing a Community through Language. London: Routledge.

CONBOY, M. (2007) The Language of the News. London: Routledge.

CONBOY, M. (2010) The Language of Newspapers: Socio-Historical Perspectives. London: Continuum.

CORNELISSEN, J. (2008) 'Metonymy in language about organizations: a corpus-based study of company names', Journal of Management Studies, 45(1) p. 79–99.

COULDRY, N., Livingstone, S. & Markham, T. (2010) Media Consumption and Public Engagement: Beyond the Presumption of Attention (revised edition). Basingstoke: Palgrave Macmillan.

CRAMER, P. (2008) 'The frequency and discourse features of the public metonym', Critical Discourse Studies, 5(3) p. 265–280.

DOR, D. (2003) 'On newspaper headlines as relevance optimizers', Journal of Pragmatics, 35(5) p. 695–721.

ELLIS, B. (2001) The Copy-Editing and Headline Handbook. New York: Perseus Books.

FAIRCLOUGH, N. (2003) Analysing Discourse: Textual Analysis for Social Research. London: Routledge.

FENTON, N. and Witschge, T. (2011) '"Comment is free, facts are sacred": Journalistic ethics in a changing mediascape', in Meikle, G. & Redden, G. (Eds.), News Online: Transformations and Continuities. Basingstoke: Palgrave Macmillan.

FRITZ, E. (2007) Audience Orientation in News Stories: a Comparison Between the Guardian and the Sun. Norderstedt: GRIN.

GEE, J. (2012) Social Linguistics and Literacies: Ideology in Discourses (4th edition). Abingdon: Routledge.

GODDARD, C. (2011) Semantic Analysis: a Practical Introduction (2nd edition). Oxford: Oxford University Press.

HABERMAS, J. (1991) The Theory of Communicative Action. Cambridge: Polity.

HALVERSON, S. & Engene, J. (2010) 'Domains and dimensions in metonymy: a corpus-based study of Schengen and Maastricht', Metaphor and Symbol, 25 p. 1–18.

HARVEY, D. (2005) A Brief History of Neoliberalism. Oxford: Oxford University Press.

HELLER, M. (2001) 'Undoing the macro/micro dichotomy: ideology and categorisation in a linguistic minority school', in Coupland, N., Sarangi, S. & Candlin, C. (Eds.), Sociolinguistics and Social Theory. London: Longman.

HERMAN, E. & Chomsky, N. (1994 [1988]) Manufacturing Consent: the Political Economy of the Mass Media. London: Vintage.

HICKS, W. & Holmes, T. (2002) Sub-editing for Journalists. London: Routledge.

HIGGINS, M. (2004) 'Putting the nation in the news: the role of location formulation in a selection of Scottish newspapers', Discourse and Society, 15(5) p. 633–648.

HOW, A. (2003) Critical Theory. Basingstoke: Palgrave Macmillan.

IFANTIDOU, E. (2009) 'Newspaper headlines and relevance: ad hoc concepts in ad hoc contexts', Journal of Pragmatics, 41(4) p. 699–720.

JAKOBSON, R. (2002) 'The metaphoric and metonymic poles', in Dirven, R. & Pörings, R. (Eds.), Metaphor and Metonymy in Comparison and Contrast. Berlin and New York: Mouton de Gruyter.

JOSEPH, J. (2006) Marxism and Social Theory. Basingstoke: Palgrave Macmillan.

KELLER, R. (2013) Doing Discourse Research. London: Sage.

KEY NOTE (2012) Newspapers Market Report 2012. Available from: < https://www.keynote.co.uk/market-intelligence/view/product/10515/newspapers>

KOTEYKO, N. (2006). 'Corpus linguistics and the study of meaning in discourse', The Linguistics Journal, 1(2) p. 132–157.

LASKY, M. (2000) The Language of Journalism. Volume One: Newspaper Culture. New Brunswick, NJ: Transaction.

LINDGREN, S. (2000) 'Michel Foucault', in Andersen, H. & Kaspersen, L. (Eds.), Classical and Modern Social Theory. Oxford: Blackwell.

MACHIN, D. & Mayr, A. (2012) How to Do Critical Discourse Analysis: a Multimodal Introduction. London: Sage.

MADIANOU, M. (2009) 'Audience reception and news in everyday life', in Wahl-Jorgensen, K. & Hanitzsch, T. (Eds.), The Handbook of Journalism Studies. London: Routledge.

MADIANOU, M. (2012) 'News as a looking-glass: Shame and the symbolic power of mediation', International Journal of Cultural Studies, 15(1) p. 3–16.

MARCUSE, H. (1964) One-Dimensional Man: Studies in Ideology of Advanced Industrial Society. Boston: Beacon. Available from: < http://www.marcuse.org/herbert/pubs/64onedim/odmcontents.html>

MARCUSE, H. (1991 [1964]) One-Dimensional Man: Studies in Ideology of Advanced Industrial Society (2nd edition). London: Routledge.

MÅRDH, I. (1980) Headlinese: On the Grammar of English Front Page Headlines. Malmö: CWK Gleerup.

MARKERT K. & Nissim M. (2007) 'Metonymic proper names: a corpus-based account', in Gries, S.T. & Stefanowitsch, A. (Eds.), Corpus-based Approaches to Metaphor and Metonymy. Berlin and New York: Mouton de Gruyter.

MARSH, D. & Hodsdon, A. (2010) Guardian Style. London: Guardian Books.

MATHESON, D. (2005) Media Discourses. Maidenhead: Open University Press.

MCKIE, L. & Ryan, L. (2012) 'Exploring trends and challenges in sociological research (the coming crisis of empirical sociology)', Sociology, 46(6) p. i–vii.

MUSSON, G. & Tietze, S. (2004) 'Places and spaces: the role of metonymy in organisational talk', Journal of Management Studies, 41(8) p. 1301–1323.

OFCOM (2012) Declining Circulation of Print Newspapers Occurs Alongside Shift in Emphasis to Online Versions. Available from: < http://stakeholders.ofcom.org.uk/market-data-research/market-data/communications-market-reports/cmr12/internet-web/uk-4.58>

PANTHER, K., Thornburg, L. & Barcelona, A. (Eds.) (2009) Metonymy and Metaphor in Grammar. Amsterdam: John Benjamins.

PURSEHOUSE, M. (1991) 'Looking at the Sun: into the nineties with a tabloid and its readers'. Cultural Studies at Birmingham 1: pp. 88–133 (Cited in: Allan, S. (2010) News Culture (3rd edition, p.129). Maidenhead: Open University Press).

RAFFERTY, E. (2008) 'Headlines', in Franklin, B. (Ed.), Pulling Newspapers Apart: Analysing Print Journalism. Abingdon: Routledge.

REDDEN, J. & Witschge, T. (2010) A new news order? Online news content examined', in Fenton, N. (Ed.), New Media, Old News: Journalism and Democracy in the Digital Age. London: Sage.

REISIGL, M. & Wodak, R. (2001) Discourse and Discrimination: Rhetorics of Racism and Antisemitism. London: Routledge.

RICHARDSON, J.E. (2007) Analysing Newspapers. An Approach from Critical Discourse Analysis. London: Palgrave.

SAXENA, S. (2006) Headline Writing. London: Sage.

SEALEY, A. & Carter, B. (2004) Applied Linguistics as Social Science. London: Continuum.

SEMINO, E. (2009) 'Language in newspapers', in Culpeper, J., Katamba, F., Kerswill, P., Wodak, R. & McEbery, T. (Eds.), English Language: Description, Variety and Context. Basingstoke: Palgrave Macmillan.

SHIE, J. (2011) 'Allusions in New York Times and Times Supplement news headlines', Discourse and Communication, 5(1) p. 41–63.

STOREY, J. (2012) Cultural Theory and Popular Culture (6th edition). Harlow: Pearson.

STUBBS, M. (2002) Words and Phrases: Corpus Studies of Lexical Semantics. Oxford: Blackwell.

SWAN, M. (2005) Practical English Usage. Oxford: Oxford University Press.

TAIWO, R. (2007) 'Language, ideology and power relations in Nigerian newspaper headlines', Nebula, 4(1) p. 218–245.

VAN DIJK, J. (2012) The Network Society (3rd edition). London: Sage.

VAN DIJK, T. (1988) News Analysis: Case Studies of International and National News in the Press. Hillsdale, NJ: Lawrence Erlbaum.

VAN DIJK, T. (2006) 'Ideology and discourse analysis', Journal of Political Ideologies, 11(2) p. 115–140. Available from: <http://www.discourses.org/OldArticles/Ideology%20and%20discourse%20analysis.pdf>

WATSON, J. (2003) Media Communication (2nd edition). Basingstoke: Palgrave Macmillan.

WHEELER, S. (2005) 'News subbing simplified', in Keeble, R. (Ed.), Print Journalism: a Critical Introduction. London: Routledge.

WHITE, M. (2011) 'Cracking the code of press headlines: from difficulty to opportunity for the foreign language learner', International Journal of English Studies, 11(1) p. 95–116.

WODAK, R. (2009) 'Language and politics', in Culpeper J., Katamba, F., Kerswill, P., Wodak, R. & McEbery, T. (Eds.), English Language: Description, Variety and Context. Basingstoke, Palgrave Macmillan.

WODAK, R., De Cillia, R., Reisigl, M. & Liebhart, K. (2009) The Discursive Construction of National Identity (2nd edition). Edinburgh: Edinburgh University Press.

WODAK, R. & Meyer, M. (2009) 'Critical discourse analysis: history, agenda, theory and methodology', in Wodak, R. & Meyer, M. (Eds.), Methods of Critical Discourse Analysis (2nd edition). London: Sage.

YOUNG, L. & Fitzgerald, S. (2006) The Power of Language: How Discourse Influences Society. London: Equinox.