Killer, Thief or Companion? A Corpus-Based Study of Dementia Metaphors in UK Tabloids

ABSTRACT This article examines the metaphors that are used to represent dementia in British tabloid newspapers over a ten-year period (2010–2019). The analysis takes a corpus-based approach to metaphor identification and analysis, utilizing in particular the corpus linguistic technique of collocation analysis. Metaphors are considered in terms of the ‘targets’ they frame, which include the following aspects of dementia: (i.) prevalence; (ii.) causes; (iii.) symptoms and prognosis; (iv.) lived experience; and (v.) responses. A range of metaphors are identified, with the tabloids exhibiting a particular preference for metaphors which construct dementia as an agentive and violent entity and people with dementia as passive victims, and which foreground preventative responses to dementia such as pharmacological intervention and individual behavior change. It is argued that such metaphors have the potential to contribute to dementia stigma and place focus on preventing or eliminating dementia while backgrounding responses which may help people to “live well” with the syndrome in the here-and-now. Metaphors which frame dementia as a companion or which the experience of dementia as a journey are put forward as potentially less stigmatizing alternatives which might better reflect the particularities of this complex public health issue.


Introduction
Dementia is a syndrome characterized by a series of diseases that cause impairment in memory, reasoning, perception and communication (World Health Organisation [WHO 2022]). This study examines the metaphors used by British tabloid newspapers to frame dementia. The analysis takes a corpus-based approach to metaphor analysis, based on a purpose-built corpus containing national UK tabloid articles about dementia published between 2010 and 2019 (inclusive). The analysis considers which aspects of dementia are foregrounded by the tabloids' metaphorical choices (as well as those aspects which are backgrounded or elided), and the particular views or understandings of the syndrome that are prioritized as a result. These framings, in turn, are interpreted in terms of their potential implications for the public's understandings, and indeed experiences, of dementia. Following this introduction, the next section begins by providing a more detailed overview of dementia, outlining its prevalence and health consequences and describing existing research into the ways dementia is represented in the media. The second half of the section then introduces metaphor and framing, paying particular attention to the potential for metaphorical framings to shape individuals' healthrelated knowledge, beliefs and behaviors.

Dementia
The nomenclature "dementia" is an umbrella term that covers a range of discrete diseases, the most common of which is Alzheimer's disease. People experiencing each of these variants are likely to present with different symptoms and the experiences of any type of dementia will be unique to the individual affected by it. Thus, for example, the experiences of one person living with Alzheimer's disease are typically distinct from those of another person living with Alzheimer's disease (Harman & Clare, 2006). Dementia is a progressive syndrome, which means that its symptoms develop and intensify over time.
Dementia is widely understood to be one of the greatest public health challenges facing society today. There are around 55 million people living with dementia worldwide (WHO, 2022), and this figure is projected to rise to 152 million people by 2050 (WHO, 2022). In the UK -the context of the present study -there are currently understood to be around 900,000 people living with dementia; or, 1.35% of the country's population (Wittenberg, Hu, Barraza-Araiza, & Rehill, 2019). This figure is set to rise to 1.59 million people by 2040 (Wittenberg et al., 2019). This purported increase in the prevalence of dementia has been interpreted and described by many -including media commentators, politicians and scientists -in apocalyptic terms, as a "rising tide," a "silent tsunami" and a 'time bomb" (Zeilig, 2014). Yet, such statistical niceties do not provide a straightforward indication that the numbers of people affected by dementia are necessarily growing in real terms, or at least at the same rate as available prevalence statistics might imply. In fact, a recent systematic review of studies regarding dementia prevalence found moderately consistent evidence to suggest that the incidence of dementia may actually be declining in high-income countries (Prince et al., 2016). At the same time, any evidence for the rising prevalence of dementia might be attributable to a number of different factors, including aging populations and the increased capacity of medical professionals (and the technologies they use) to detect and therefore diagnose dementia. The picture regarding the prevalence of dementia is, it seems fair to say, complex and not particularly well understood.
There is currently no known pharmacological cure for dementia. This means that people living with the syndrome will also die with it. Between 2016 and 2020, dementia was reported by the UK's Office for National Statistics (ONS) to be the leading cause of death in England and Wales (Office for National Statistics [ONS], 2020). Like its purportedly rising prevalence, though, the increase in dementia-related mortality can be linked to a number of factors. Significantly, the aforementioned ONS report also attributed the evident rise in recorded dementia mortality to changes in UK death certificate-writing practices, whereby deaths formerly attributed to other conditions have recently been coded as exclusively dementia. Furthermore, during this period NHS General Practitioners were controversially given financial incentives to diagnose dementia (BBC, 2014). Yet, by far the most problematic aspect of these mortality rates is that dementia itself does not actually cause death; that is to say, although people die with dementia, they do not die of it. In fact, when people with dementia die, their death is typically caused by co-occurring conditions -most often pneumonia -rather than being caused by dementia itself (Brunnström and Englund, 2009).

Cultural and media representation of dementia
Going beyond biomedical understandings of dementia, it is also important in a study such as this to consider dementia as a cultural phenomenon. This includes consideration of how the syndrome is represented in socio-cultural context. In Western societies, dementia is commonly conceived of as a cultural metaphor for the aging process itself, as a stage in life that has to be resisted at all costs. The personhood movement in dementia studies (see Kitwood, 1997) has drawn particular attention to the propensity for prominent, historical and enduring cultural conceptions of dementia to contribute to stigma. Sociologist Goffman (1963) referred to stigma as a "spoiled identity." Durkheim (1895Durkheim ( /1982 argued stigma to be unavoidable and relative -that even in societies with no crime, "venial faults" would create scandals and that if societies could judge and punish, such acts would be defined as criminal. Representations of dementia which have been linked to the creation and sustenance of dementia stigma include, perhaps most prominently, the reduction of the person with dementia to their brain and (deteriorating) cognitive faculties (Harvey & Brookes, 2019) and the framing of the syndrome as representing a loss of self and "social death" (Sweeting & Gilhooly, 1997), as something to be feared (Basting, 2009) and, relatedly, as a metaphorical battle which needs to be fought by any means (Zimmermann, 2020). Such discourses are by now long-standing and well established, and can be traced back over decades (Zimmermann, 2020). Furthermore, such is their dominance that these conceptions of dementia also transcend particular cultures and modes of communication (as demonstrated, for example, by the essays compiled by Swinnen and Schweda (2015)).
As well as understanding its biological influences, understanding how dementia is culturally constructed "contributes to the fight against stigma and can improve the wellbeing of people with dementia and their caregivers" (Swinnen & Schweda, 2015, p. 11). Indeed, given that there is presently no cure for dementia, there is a pressing need for research, including linguistic research, which aims to shed light on the public's experiences and understandings of the syndrome, including how such experiences and understandings might be shaped by media framings (Lock, 2013). This is important, since popular media -including news media -remains one of the primary means through which people access health information and then form their impressions and understandings of health issues. Dementia is certainly no exception here, given that it appears to represent something of an "unfailingly newsworthy topic" for the print media (Brookes, Harvey, Chadborn, & Dening, 2018, p. 371). This is reflected in the amount of interest that media coverage of dementia has received from social scientists and public health specialists. For example, Kirkman (2006) analyzed the contexts in which Alzheimer's disease was portrayed across a 5-year sample of articles taken from 15 New Zealand newspapers, with the analysis showing that Alzheimer's disease was covered in various sub-registers including notices, scientific reports, obituaries, personal stories, "missing people" reports and reports dealing with residential care for people with Alzheimer's disease. Kirkman concluded that print media served to perpetuate negative stereotypes associated with Alzheimer's disease, dementia and aging. van Gorp and Vercruysse (2012) analyzed the frames emergent from images related to dementia in articles from six Belgian newspapers as well as a sample of movies, literature and healthcare texts. They reported the most common frame to be one that represents humans as being composed of two distinct parts: a material body and an immaterial mind. More recently, Harvey and Brookes (2019) examined the visual discourse surrounding dementia in commercial stock images. This study found that the values, assumptions and beliefs articulated by these images represented people experiencing the syndrome in objectifying and de-humanizing terms, emphasizing disease and deficit at the expense of whole personhood, for example by frequently symbolizing dementia through images of disembodied, wrinkled pairs of hands.
The studies discussed above paint a decidedly negative picture of media representation of dementia. Such representations have been broadly characterized as relying on age-related stereotypes, and consequently as perpetuating of dementia and age-related stigma (see also : Swinnen & Schweda, 2015). Of direct relevance to the present study, other discourse-based studies of dementia have elucidated some of the metaphors that the mass media and other types of text producers have employed when representing dementia. Before reviewing some of these studies, I will first provide a more detailed introduction to metaphor itself, considering its relationship to communication about health and illness.

Metaphor and the framing of dementia
For the purposes of this study, I adopt Semino's (2008, p. 1) definition of metaphor as "the phenomenon whereby we talk and, potentially, think about something in terms of something else," where the two 'things' involved are different but are perceived by the language user to share some similarity. Cognitive metaphor theorists, such as Lakoff and Johnson (1980), have demonstrated that metaphors can play a central role in not only communication but also cognition, as they have the power to "express, reflect, and reinforce different ways of making sense of particular aspects of our lives" (Semino, Demjén, & Demmen, 2018, p. 625). Thus, metaphors have the capacity to "frame" our understandings and experiences of reality, including health and illness (Tay, 2017).
The present study examines the role of metaphor in framing dementia in the press. The concept of framing was originally developed within the field of sociology (Goffman, 1974). Bringing together a series of definitions, Atanasova and Koteyko (2017, p. 652) usefully define a "frame" as "an organising principle . . ., a central organising idea . . . or an interpretative package" that enables individuals to make sense of issues by turning "meaningless" aspects "into something meaningful" (Goffman, 1974, pp. 21-22). Studies applying the concept of framing in analyses of media data are often influenced, more and less directly, by the work of Entman (1993), whose perspective on framing places particular emphasis on the role of selection and salience. He argues that "[t]o frame is to select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation for the item described" (Entman, 1993, p. 52). For Entman, then, a frame has four functions or elements which may or may not all be present within a particular text: (i) to define an issue, (ii) to suggest its causes, (iii) to propose solutions and (iv) to assign moral evaluations (see also : Atanasova & Koteyko, 2017;Baker, Brookes, Atanasova, & Flint, 2020).
Given that metaphors are often used to express abstract, subjective, sensitive and taboo subjects, it stands to reason that talk about health and illness is often replete with metaphorical language. This point is addressed by Semino (2008), who also observes how metaphors surrounding health and illness can be found at multiple levels; [m]etaphor is relevant to the linguistic and conceptual construction of illness in more than one way. As a physical or mental condition to be understood and treated, illness can be modelled metaphorically by experts like any other phenomenon that is the subject of scientific investigation. [. . .] However, illness is also an individual, personal state of being, which is normally associated with physical discomfort or pain, and with feelings of anxiety, fear, isolation and, potentially, shame. As such, it belongs to the kind of complex, subjective and poorly delineated experiences that tend to be conventionally verbalised and conceptualised through metaphor. Semino (2008, pp. 175-176) Considering their potential to shape health-related knowledge, beliefs and actions, there is clear value in studying the metaphors that are used to communicate about topical health issues such as dementia. Indeed, recent studies have shed light on some of the metaphors that are used to represent dementia not just in the media but in other types of texts, too. Adopting a perspective from critical gerontology, Zeilig (2014) explores the metaphors in stories about dementia across different text types, interpreting these in terms of their cultural and historical influences. Studying a collection of films, TV documentaries, news reports, theater, memoirs, novels and poems, Zeilig observes the use of metaphorical language to frame dementia as, for example, a "rising tide," an opponent in a battle, and a monster. She also notes the characterization of dementia as a crisis and a burden. A similar observation is made by Peel (2014) in her analysis of 350 newspaper articles. Taking a "discursive thematic approach," like Zeilig, Peel noted the use of a "panic-blame framework", wherein dementia was characterized using "catastrophic" terminology, for example being framed as a "tsunami" and something "worse than death." Peel also observed a focus on individualistic behavior and lifestyle changes as in discussions of how to prevent -or, in the words of the articles in her data, "stave off" -the syndrome. Comparing these depictions to interviews conducted with informal carers, Peel observed that the carers in her data were much less likely to employ such hyperbolic discourse and rarely framed dementia in terms of individual responsibility.
More recently, Brookes et al. (2018) studied the multimodal discourse representations of dementia in a series of UK national newspaper articles printed in response to the publication of statistics revealing dementia to be "the leading cause of death" in England and Wales. In addition to scrutinizing the images used to represent dementia in the articles, this analysis also highlighted the use of war and violence metaphors in headlines which branded dementia as the nation's "biggest killer," as well as metaphors of sport and competition in the characterization of dementia as winning the race to claim "top spot" as the UK's leading cause of death. Bailey, Dening and Harvey (2021) carried out a corpusassisted analysis of UK media discourses around dementia in newspaper articles published between 2012 and 2017. Their analysis identified metaphors which, the authors argued, supported a biomedical view of dementia and foregrounded pharmacological treatments and research.
The aim of the present study is to build on this insightful, though slim, body of research which has illuminated media metaphors of dementia. It aims to do this in a number of ways. First, this study provides the first quantitative (and qualitative) study focusing exclusively on dementia metaphors. While the aforementioned studies have identified instances of metaphor use when exploring dementia representation, their analyses have not set out to systematically explore this feature in the way that the present study does. Furthermore, these previous studies have tended to be based on relatively small datasets. By analyzing a corpus representing dementia coverage over a ten-year period, the present study examines a dataset that is not only more widely representative of press coverage of dementia than previous datasets, but is also more up-to-date, running to the end of 2019. This is an important consideration when researching a topical and ever-developing health issue like dementia. An exception to the use of small datasets is Bailey et al.'s (2021) study which, like the present study, used a corpus linguistic methodology. The present study builds on this research by systematically analyzing metaphors specifically (this was not the main analytical focus of Bailey et al.'s study) and by basing insights on a larger and more recent dataset (representing ten years, up until the end of 2019).

Methodology
This analysis takes a corpus-based approach to identifying and analyzing the use of metaphors in tabloid framings of dementia. Corpus linguistics is largely a group of methods which use computational techniques to analyze linguistic patterns in large collections of naturally occurring language use (Brookes & McEnery, 2020). Such datasets are known as "corpora" (singular corpus) and are assembled with the aim of representing a language or linguistic variety on a broad scale. Corpora tend to be much larger and more representative than the types of datasets that are typically analyzed in studies employing manual, purely qualitative analytical methods. Corpus linguistic techniques can also provide unique insights into the language in a corpus, relating for example to the frequencies of words and the statistical salience of patterns of word co-occurrence (i.e. collocations).

Data
The data analyzed in this study is a purpose-built corpus representing UK national tabloid coverage of dementia over a ten-year period (2010-2019). Using the online news repository, LexisNexis, I obtained all articles which mentioned "dementia*" in their headline and/or three or more times in their body (note: the asterisk acts as a wildcard to capture longer strings, e.g. "dementias"). Search results were grouped by moderate similarity and, once downloaded, required some cleaning to remove article metadata. The resulting corpus contains 6,751 articles (3,961,272 words), distributed across five UK national tabloids (Express, Mail, Mirror, Star, Sun -including print, online, Sunday and "sister" publications). See Table 1 for a breakdown of the newspapers in the corpus.

Analytical approach
The approach to metaphor identification adopted in this study is based on collocation analysis. As conceptualized in corpus linguistics research, "collocation" refers to the association between words based on patterns of co-occurrence. Collocation is usually judged to exist using a word association measure that tells us how often two or more words occur within close proximity of each other within the corpus, and whether this association is notable as a sizable effect (i.e. that the words in question have a measurably strong preference to occur together as opposed to being randomly associated). Using WordSmith Tools (version 8; Scott, 2020), I began by generating a list of collocates for the word dementia (frequency: 50,868) -i.e., the "target" or "recipient" of the metaphors. For words to be judged as collocates, I stipulated that they had to occur within a five-word window either side of dementia at least 10 times and be assigned a Mutual Information (MI) score of 3 or above. This threshold represents something of a standard in corpus-based discourse studies (Baker, Gabrielatos, & McEnery, 2013;Gabrielatos & Baker, 2008;Hunston, 2002) and, for the purposes of the present study, gave a number of collocates that was small enough to be manageable for qualitative analysis while being sufficiently large to show a range of different metaphorical representations. I then generated concordances for all resulting collocates and manually analyzed their co-occurrences with dementia to ascertain whether or not they tended to be used metaphorically. My identification of metaphors was guided by the Metaphor Identification Procedure (MIP; Pragglejazz Group, 2007). For collocational pairings with a frequency exceeding 100, I analyzed a sample of 100 uses. In such cases, I balanced the samples so that they reflected the distribution of the given collocational pairings across the newspapers in my corpus. For example, if 70% of the total occurrences of a given collocational pairing occurred in the Mail, 25% in the Guardian, and 5% in the Mirror, then this distribution would be mirrored in the make-up of the sample used for the analysis of that collocational pairing. Within newspapers, uses of collocational pairings were sampled at random.
Each use of all candidate metaphorical collocates within the samples was analyzed following the MIP, as summarized by Semino et al. (2018, p. 632): "an expression is regarded as metaphorically used when its 'contextual meaning' contrasts with a more physical and concrete 'basic meaning,' and where the former meaning can be understood via a comparison with the latter." Following Semino et al. (2018), I also included simile and other figurative comparisons, in line with Steen et al.'s (2010) definition of "direct metaphor." Collocates judged to have been used metaphorically in at least 50% of cases were considered metaphorical, while those which evidenced no metaphorical use, or which were used metaphorically but in fewer than 50% of cases, were discarded. This step ensured that the words analyzed were used metaphorically at least as much as they were used with non-metaphorical senses, while having the practical benefit of keeping the analysis manageable by removing the need for me to wade through large numbers of non-metaphorical cases. My codings were checked by another researcher with training in linguistics to ensure accuracy and plausibility.
The metaphorical collocates were then allocated to one or more of the following categories, which reflect the specific "targets" of the metaphors: (i.) prevalence; (ii.) causes; (iii.) symptoms and prognosis; (iv.) lived experience; and (v.) responses. I then qualitatively analyzed all metaphorical uses of the collocates (or, if their frequency exceeded 100, I took 100 metaphorical uses). This qualitative analysis set out to identify the frames (Goffman, 1974) that were activated through the use of the metaphorical collocates. In particular, my analysis considered which aspects of dementia are foregrounded by the metaphorical framings (and which are backgrounded or elided), and what views or understandings of dementia are prioritized as a result. In the discussion following the analysis, these framings are interpreted in terms of their potential implications for the public's understandings, and indeed experiences, of dementia.
A limitation of the data analyzed in this study is that it focuses just on tabloid coverage and not coverage from broadsheets. This was a practical decision driven by limitations on space, and to allow the analysis to be more focussed. I decided to focus on tabloids as these tend to be read more widely than broadsheets in the UK (PressGazette, 2019), and so arguably play a more prominent role in shaping societal understandings of dementia. A further limitation of the data is that it is unbalanced in terms of the extent to which each tabloid is represented (for example, the Mail contributes many more articles and words to the corpus than the Star does). This imbalance does at least reflect the landscape of UK press coverage of dementia which, on the basis of Table 1, is clearly unbalanced in terms of how much each newspaper contributes to the discourse. A limitation of the approach taken in this study is that it will lead to the identification only of those metaphors which are lexicalized frequently within close textual proximity to mentions of dementia. A more comprehensive approach would involve considering metaphors which also occur beyond this tight collocational span and which may, in fact, not be represented at all among the strongest collocates of dementia. However, for the purposes of the present study, the collocation-based approach provided a productive but practical route into the data.

Findings
A full list of metaphorical collocates can be found in Appendix 1. The analysis identified evidence of a range of metaphorical domains which contributed to the framing of the following aspects of dementia: (i.) prevalence; (ii.) causes; (iii.) symptoms and prognosis; (iv.) lived experience; and (v.) responses. For the facility of analysis, the findings reported in this section are structured according to these metaphorical targets. However, as will become clear, many of the metaphorical patterns identified relate to one another in how they frame dementia and, at some points, overlap.

Prevalence
Dementia's prevalence was framed using metaphors belonging to a diverse range of domains, all of which framed the UK's prevalence of dementia as being high and rising. Most frequently, the articles could be interpreted as drawing on a domain of sports and competition to frame dementia as the leading cause of death in the country. This was indicated in the majority of uses of the following collocates (frequency of collocation with dementia shown in brackets): leading (242) and leader (49). The notion that dementia is the "leading" cause of death could be considered a more conventionalized trope (for example, many English dictionaries define leading as "first or most important"). However, its uses could also be more self-evidently metaphorical, for example in cases where dementia was framed as overtaking (overtaken (19), overtook (10)) other conditions to become the leader or the leading cause of death in the country. As the following example shows, this metaphor could be mixed with a metaphor of violence, with dementia being framed as the biggest "killer" in the UK (a metaphor considered in more detail later). Note: examples chosen were deemed to be representative of the metaphorical patterns being discussed. The node is highlighted in bold and lexicalizations of the metaphors are underlined.
dementia has overtaken heart disease as the biggest killer (Express 2017) Violence metaphors could also be used to frame dementia's prevalence more directly, where the incidence of the syndrome was quantified in terms of the number of people that had been hit (162) by the condition, or what its death toll (53) was. Related to the violence metaphor, dementia was also construed as a weapon -a time bomb (bomb (22), ticking (10), timebomb (47)) -which foregrounds more specifically dementia's rising prevalence and implies that this will continue to grow, ultimately to disastrous effect. BRITAIN faces a dementia timebomb with one in three people born this year expected to develop the condition, research shows.

(Mail 2015)
Dementia's growing prevalence was also metaphorically rendered as an epidemic (66). In around a fifth of cases, this epidemic was also described as "silent," which arguably attributes to it an insidious quality.
DEMENTIA is a silent epidemic, with 11 people a day now diagnosed with the disease here and around 55,000 current cases. (Sun 2018) Since the word epidemic belongs to the domain of illness, this characterization of dementia may not at first appear metaphorical. However, strictly speaking, the term epidemic denotes the spread of an infectious disease, which dementia is not. Consulting the general language British National Corpus, the "epidemic" metaphor tends to be used to characterize an increase in the frequency of a phenomenon or activity that is evaluated negatively, such as crime and the illegal use of drugs. Thus, like the "ticking time bomb" metaphor, the epidemic metaphor is used to provide a negative evaluation of the growth in dementia's prevalence.

Causes
Dementia's causes could be framed using the domain of movement, with certain behaviors and risk factors being construed as speeding up the development of the syndrome (accelerate (18), accelerated (11), speeds (27)).
BEING overweight dramatically speeds up the onset of dementia, a study has found.

(Sun 2015)
As well as focusing on the environmental risks associated with dementia, the newspapers also reported on the role of certain biological factors, particularly protein variants and parts of the brain to which dementia onset is attributed. These risks were framed using violence metaphors, whereby the 'risky' parts of the brain are anthropomorphized as a treacherous rogue (18) or a weapon (dagger (10)) which attacks other parts of the brain. This latter metaphor is presented as a borrowing from scientific discourse.

Symptoms and prognosis
Metaphorical representations of dementia symptoms and prognoses characteristically rely on the domain of violence, wherein dementia is anthropomorphized as an actor which inflicts forms of violence on those diagnosed with it. Memory, cognitive function and even personhood are all framed as things that are stolen (robbing (13), robs (18), steals (14)) by dementia.
Victims of dementia, which robs sufferers of their memory, rely on social care as drugs are unable to slow the progression of the incurable disease. (Mail 2017) Other symptoms of dementia are framed as forms of violence in more generic terms. For example, dementia is described as something which ravages (12) and strikes (41). In terms of its prognosis, the newspapers frequently foreground a fatalistic view of dementia, framing it as a murderer that kills those diagnosed with it (kill (54), killed (83), killer (108), killers (12), killing (45), kills (22)).
DEMENTIA kills more Brits than any other illness, new figures show.

(Sun 2017)
How we fail victims of our worst killer illness, dementia; There are now 850,000 dementia sufferers and that number can only increase as our population ages. (Mirror 2017) This fatalistic framing of dementia is supported by the anthropomorphic characterization of it in terms of qualities which imply dementia to be a violent actor who inflicts such violence, and theft, willfully and deliberately (brutally (11), cruel (83), cruelest (16), deadly (44), heartless (17), savage (10)).
Dementia is the cruellest of diseases. It robs you of your faculties, your dignity, your identity.

(Mail 2019)
The other way in which dementia's symptoms are metaphorically rendered is as the 'loss' of memory and cognitive function. The notion of memory loss might be considered a conventionalized metaphor in discourse surrounding dementia. However, it is notable that, in the articles, this metaphor is extended, through the collocates lose (72), loses (18), losing (111) and lost (125), to foreground the agency of the person with dementia, who is framed as having metaphorically lost the particular functions under focus.
Many people with dementia gradually lose their ability to walk and perform simple tasks as their condition progresses.

Lived experience
Consistent with the metaphorical framing of dementia as a cruel and brutal actor who inflicts forms of violence, including murder, on people diagnosed with it, the lived experience of dementia could also be characterized in terms of conflict. People with dementia were framed as variously 'battling', 'facing' and 'struggling with' dementia (battle (425) Other constructions were less agentive, though, on the part of people with dementia and instead foregrounded their victimhood. In such cases, people with dementia were referred to as a victim (115) or victims (143), were described as being in dementia's grip (18) or hold (79), as being stricken (176) and struck (95) by dementia, and as being stripped (26) of their cognitive function and sense of personhood.
Doctor Who actor Christopher Eccleston tells how dementia stripped his father of his personality (Mail 2017) Meanwhile, the framing of people with dementia as succumbing (10) or having succumbed (14) to dementia arguably implies an initial struggle or resistance but an eventual failure or even surrender. Another set of metaphors which similarly frames people with dementia as the objects or victims of processes are found in stories which focus in particular on cases where people with dementia have been purportedly abandoned by their relatives. This process is metaphorically likened to that of waste disposal (dump (10), dumped (31)). The negative appraisal of life with dementia is also conveyed through metaphors which frame the experience of the syndrome in horrific or even apocalyptic terms, likening it to living in hell (18), a horror (13) movie or a nightmare (10). A negative evaluation of life with dementia is also conveyed more subtly through the collocate descent (12), whose uses indicate a spatial metaphor which frames life with dementia as downward movement into the condition, which is metaphorically rendered as a bounded region.
Yesterday in her heart-breaking series about her husband's descent into dementia, Nula Suchet described her anguish as she watched his health deteriorate.

(Mail 2019)
Not all metaphors necessarily imply negative evaluations of life with dementia or foreground the victimhood of people diagnosed with it, though. A particularly frequent example is the metaphorical characterization of the relationship between dementia and the person with it as one of companionship. This was indicated in uses of the collocates live (377) and living (1,198). The companionship metaphor is more flexible than those seen so far in this section, as it could be used to frame the relationship between dementia and the person diagnosed with it in more or less harmonious terms, as these examples demonstrate (note that the second example derives its negative evaluation from the blend of companionship with the biblical scenario of being 'hell').
Doctors advised him that to keep living well with dementia he should stay active, eat healthily and socialise.

(Mirror 2019)
Living with dementia is hell, and I wouldn't wish it on anyone else. (Mail 2012) Likewise, the final metaphor considered in this section, which framed the experience of dementia as a journey (35), could be used to appraise life with the syndrome in more or less positive terms, as well as to foreground mixed experiences and dementia's progressive nature.
When choosing a game to play with dementia patients, it's important to note what stage of their dementia journey they are at, as not all games will be suitable.

Responses
The final area of representation explored in this analysis concerns responses to dementia. This is a broad category of representation, which mostly comprises responses from society as a whole, from biomedicine, and from individuals (in terms of reducing their risk of developing the syndrome). The most pervasive metaphorical domain in this category, in terms of both its frequency and the number of individual collocates that constitute it, is violence. Most of the metaphors in this domain are used to frame the response to dementia from society as a whole, as well as from biomedicine in terms of the development of a pharmacological cure. This response is framed in terms of conflict in general (against (529), beat (261), beaten (19), beating (52), beats (16), combat (52), defeat (30), tackle (84), tackling (48)), as well as more specifically in terms of a battle (battle (425), battled (22), battles (31), battling (167)), fight (fight (370), fighting (49), fights (24)), as well as being a mission (15) and a war (44). These conflicts can, as noted, be pitched as taking place between dementia and scientists but also between dementia and society as a whole. The conflict is framed variously as being fought successfully and unsuccessfully at different points in time. In all cases analyzed, the key to victory in the conflict is the development of pharmacological interventions for dementia, as well as securing funding for this.
New hope in the war on dementia as scientists discover diabetes drug may also reverse memory loss (Mail 2018) A small set of violence metaphors were used to frame individuals' responses to dementia. Specifically, these metaphors were used to advise readers on behaviors that could WARD off dementia (ward (234), warding (11), wards (57)).

(Sun 2019)
These movement metaphors were also used in representations of the promise of studies and pharmacological trials, which were framed as having the potential to slow, stop or reverse dementia's movement. The construction of dementia as a moving entity can be linked, in turn, to the broader cultural framing of disease as an invader (see Gwyn, 2002), whereby the disease (in this case, dementia) is positioned as external to the individual and the body is presented as a battleground upon which doctors fight disease or, in this case, individuals fend it off.
Relatedly, the development of a dementia cure was characterized as a race (58) between dementia and scientists.
The race against dementia is a much bigger challenge. This is the biggest personal challenge of my life and I will make sure that whilst I am still on this earth, I put all my efforts into finding a cure for this horrendous illness. (Express 2016) Other metaphors imply that a cure for dementia exists but has yet to be found by scientists who are framed as undertaking a quest (11) or hunt (55) for clues (16) or the key (258) to curing the syndrome and devising methods for earlier diagnosis.
BRITAIN will spearhead the hunt for a dementia cure by doubling the funding for cutting edge research.

(Express 2019)
Blood clues on dementia MOLECULES in our blood could help identify those most at risk of dementia. (Mirror 2018) Although the vast majority of metaphors focused on biomedical responses to dementia, followed by individual responses in terms of behavior modifications, this was not the case for all metaphorical framings of responses. In a small minority of cases, the articles drew on a metaphor of gambling to critique inconsistency in the provision of social care, with the public positioned as being at the mercy of a postcode (22) lottery (30) or Russian (18) roulette (16) regarding the social care services available to them as well as how prepared their local hospitals were to treat patients with dementia.

Discussion and conclusions
This study has identified the most prominent metaphors around dementia in the British tabloids across a period of ten years. The analysis focused on five related areas of dementia representation: prevalence; causes; symptoms and prognosis; lived experience; and responses. These aspects of dementia were found to be framed using a variety of metaphors which characterized dementia, including its prevalence, causes and symptoms, variously and amongst other things, as a violent actor, a murderer, a thief, an invader, an athlete, and an infectious disease. Meanwhile, experiences of dementia were characterized in terms of victimhood, combat, companionship, biological treachery, losing possession of memory and function, descent into a bounded region, and being disposed of like waste. Responses to dementia were characterized as battles, races, attempts to resist invasion and restrict movement, hunting and gambling. In this section, I will argue that many of these prominent metaphorical representations have the potential to frame dementia in ways that stigmatize people with the syndrome. I will also argue that people with dementia and the public in general might benefit from metaphorical framings which, rather than foreground the fatality dementia and the need for a pharmacological cure, pay greater focus to selfhood and the person living with dementia, and as a response foreground the ways that people can be supported in living socially fulfilled lives with the syndrome in the here-and-now.
As noted in Section 2, when viewed through the lens of framing, metaphor can be considered to have the power to foreground certain elements of the phenomena they represent while backgrounding or eliding others. A hallmark of almost all of the metaphors identified in the foregoing analysis is that they foreground the agency of dementia -in many cases anthropomorphizing it as a social actorwhile background or precluding altogether the agency of those diagnosed with the syndrome. As an agentive force, dementia moves and invades, races against and hides from those seeking to remedy it, and competes with other conditions to become the leading cause of death. People with dementia, meanwhile, are largely denied agency, being instead rendered as the target of the actions both of dementia and those seeking to cure it. In certain contexts, people with dementia were granted agency, though the actions attributed to them were negatively freighted, such as being depicted as "losing" their memory and personhood and as "succumbing" to dementia. This pattern in the assignment of agency is arguably most visible in metaphors belonging to the domain of violence, which were used to frame dementia as a violent actor which attacks people with dementia, kills them, and steals from them their cognitive function and very personhood. Such representations contribute to a wider discourse of dementia as social death. Moreover, the manner in which dementia undertakes such acts of violence is presented as deliberate and malicious -carried out in a manner described as "brutal," "cruel" and "heartless." Violence metaphors were pervasive across the corpus, contributing to the framing of dementia with respect to all five areas of representation considered.
Violence metaphors (also termed "war," "martial" and "militaristic", etc.) are pervasive not only in coverage of dementia but in the language associated with health and illness in general, with criticisms of such tropes by now long-standing and wide-ranging. Sontag (1978) cautioned against the use of such metaphors, arguing that they can have stigmatizing effects for those affected by the illness in question. In particular, Sontag argued that violence metaphors have the potential to frame people affected by illness as the "enemy" of the "war," and their bodies as the metaphorical "battlefields." This is an argument that has since been extended and discussed in relation to a wide range of health issues, including for example cancer (Semino, Demjén, Hardie, Payne, & Rayson, 2017), SARS (Wallis & Nerlich, 2005) and obesity (Brookes, 2021a(Brookes, , 2021b. In the case of dementia, such concerns may be compounded when we consider that we have seen how the causes of dementia are frequently located within the individual, in some instances with parts of their brain being characterized as treacherous weapons which inflict violence on the rest of the brain. Sontag and others have argued that a further issue with violence metaphors is that, if disease prevails and the person fails to recover from it, that person could then be viewed as having "lost" the battle. Given, as noted, that dementia is a chronic syndrome with no known cure, it seems likely that this is a battle that people diagnosed with the syndrome will inevitably lose. Scholars concerned with metaphor in relation to a wide range of illnesses have variously supported and challenged the association between violence tropes and stigmatization. A prominent counterargument is that people can find violence metaphors empowering, as they have the potential to grant individuals a greater sense of agency in their illness experience, helping them to feel like they have greater influence over their prognosis (see: Demmen et al., 2015). However, this argument is problematic in the case of dementia, not only because a cure for the syndrome has yet to be developed, but also because, as discussed, the prospects for agency for people with dementia are limited if not elided altogether in the ways that violence and other metaphors are used by the tabloids. People with dementia tend to be framed as victims of violence, with combative resistance more likely to come from people without a diagnosis who are advised on how to "fight" dementia by reducing their risk through behavioral change. A potential consequence of this neoliberal discourse is that if people are diagnosed with dementia, they could be considered to have "lost" the battle by failing to adequately take responsibility for their risk.
I would thus argue that the use of violence metaphors to frame dementia does indeed have the potential to contribute to dementia stigma. Dementia stigma is, I would argue, propagated not only through the use of violence metaphors but also by other metaphorical framings identified here, including those of horrors and nightmares and the dehumanizing equation of people with dementia to waste being disposed of. These tropes foreground dementia's most disastrous and fear-inducing aspects while backgrounding, if not shutting down altogether, the possibility of "living well" with dementia following diagnosis, as many people do with the right support (Sabat, 2018). The effects of stigma on identity and within society more broadly are well-documented. For example, Becker (1963, pp. 33-34) argued that the stigmatized aspects of a person's identity tend to be noticed first, while Epstein (1998) suggests that any stigmatized aspect of a person's identity is likely to subsume other aspects of their identity. A consequence of this, Epstein (1998) contends, is that the behavior of a stigmatized person is likely to be attributed to those aspects of their identity that are stigmatized rather than any other aspect of their identity. Falk (2001) argued that stigmatization occurs because it helps in the creation of in-groups and out-groups within societies, with stigmatised identities serving as the basis for the establishment of 'othered' out-groups.
Stigma remains a salient feature of the experience of living with dementia (Swaffer, 2014) and its effects on people's lives and society as a whole can be severe. If we take the consequences of stigma noted above, we can see how dementia stigma may contribute to a societal view of people with dementia as "other". Dementia stigma can also cpmtribute to a sense of 'spoiled identity' in the person with dementia, whereby the syndrome is seen first and all of a person's traits and behaviors are attributed to it. Taken together, these kinds of societal representations can lead to people with dementia being negatively labeled and stereotyped, which can in turn lead to discrimination and a loss of power (Link & Phelan, 2001). As well as drawing on our understanding of the impacts of stigma in general terms, recent research has highlighted the impacts of dementia stigma specifically. For example, dementia stigma has been found to engender feelings of shame in people with the syndrome, making them reluctant to seek a diagnosis and medical support, as well as making them less willing to get involved in research (Burgener & Berger, 2008). Dementia stigma has also been found to distort service standards at all levels of UK healthcare, including frontline provision and resource allocation (Benbow & Jolley, 2012). Dementia stigma can also create fear, misunderstanding and negative attitudes among the general public. For example, one in five UK adults mistakenly believe that dementia is an inevitable part of aging; two-thirds mistakenly believe dementia risk cannot be reduced; and one in five believe they would find it hard to communicate with a person with dementia (Alzheimer's Research UK, 2018). More recently, a report by Alzheimer's Disease International (2019) suggested that 80% of people are fearful of dementia.
While people with dementia were metaphorically rendered as providing little if any opposition to the syndrome, agency was ascribed to those responsible for developing curative, pharmacological solutions. This focus not only results in emphasis being placed on prospective treatments that are not currently available, but it also serves to simultaneously shift focus away from other, non-biomedical, non-pharmacological responses, such as arts-based therapies, as well as other initiatives designed to help those with a diagnosis and their relatives to manage the syndrome. Finding a cure for dementia is, of course, a worthwhile endeavor. However, the development of a curative treatment is likely to do little to improve the quality of life of people living with dementia in the here-and-now. Relative to the financial support that is given to pharmacological research, social care for people with dementia is already chronically under-funded (Alzheimer's Society, 2021). Yet, given the power of media framing to garner public and policymaker support for particular health-related initiatives (McCombs & Shaw, 1972), metaphors which foreground prevention in the form of individual behavior change and yet-tobe-developed pharmacological remedies, all the while backgrounding the role of social care in helping people to live with dementia in the here-and-now, are only likely to widen this imbalance. Indeed, metaphors which I have observed to frame the development of a pharmacological cure as a hunt or a puzzle with clues may imply the promise that there is a cure for dementia "out there," if only scientists were able to find. The gambling metaphors used to expose inconsistencies in care quality offer a refreshing focus on social care, though these are notably far less frequent than those which foreground pharmacological responses.
One defense of the metaphors identified in this study is that such tropes offer useful linguistic and conceptual apparatus with which to communicate about what is -in dementia -a complex and difficult-to-understand syndrome, in the process conferring a sense of "symbolic order" over it. Indeed, as noted above, many people have a poor understanding of dementia and fear it more than many other serious health issues (Zeilig, 2014). Yet, we may question the extent to which the metaphors which characterize the tabloids' coverage are indeed likely to improve public understanding of dementia. For example, the metaphors used to represent dementia's prevalence as wide and growing to disastrous effect -including the framing of dementia as a ticking timebomb and the leader in a race with other conditions -were seldom accompanied by any discussion of the epidemiological niceties, discussed earlier, which are likely to give us an inflated impression of the syndrome's prevalence. Indeed, as noted, recent research indicates that the prevalence of dementia in high income countries such as the UK may have actually fallen (Prince et al., 2016), quite contrary to the apocalyptic and doom-laden metaphors drawn on by the tabloids. However, such optimistic counterarguments run contrary to the news value of negativity and would arguably make for less sensational headlines, and the news agencies studied here may contend that their priority is to sell newspapers and generate traffic to their websites. Indeed, it is important to bear in mind that the news creators featured in the data work within commercial organizations. However, it is also important to reflect critically on the likely trade-off between raising profits and awareness, and the ramifications this may have for the lives of people living with dementia as well as the public more generally.
I would argue that people with dementia, their relatives and the general public would all be better served by metaphors which help to challenge dementia stigma and raise genuine awareness of the syndrome. Reducing dementia stigma could, as we have seen, result in more timelier diagnoses, the delivery of more appropriate care, improvements in the quality of life for people with dementia and their relatives, and less fear and misunderstanding of the syndrome amongst the public. Swaffer (2014, p. 710) points out that language use, along with inclusivity and the creation of dementia-friendly communities, is "important in the reduction of stigma, and without positive change, stigma will continue to be a significant burden on people with dementia." With this in mind, some of the metaphors identified in the foregoing analysis may be productive to these ends -in particular, those which frame the experience of dementia as a journey or companionship can be viewed as respectively articulating the progressive and chronic nature of the syndrome. These tropes also leave open the possibility that individuals' relationships and journeys with dementia can be positive, negative or a mixture of the two. In this sense, these metaphors might be able to account more flexibly for differences in subjective experiences of dementia, as well as variations in an individual's experiences; indeed, just as dementia is experienced differently between individuals, a single person can experience their dementia differently from day-to-day.
It is difficult to issue recommendations around metaphor use, though, as the interpretation of such figurative language use is subjective, and their effects on our experiences and understandings of health topics can vary from individual to individual. Indeed, in a recent study of metaphors used in dementia blogs, Castaño (2020, p. 126) found that bloggers could use militaristic metaphors to "show determination or to praise themselves or others for their tenacity and effort". Likewise, the blogs analyzed in this study also evidenced that 'journey' metaphors are not, by default, more positive than 'war' metaphors if, as in the case of dementia, individuals' prognoses lead them to conceive of the journey as a "one-way journey" marked by a progressive and unavoidable decline in their physical and cognitive abilities. More recently, Putland (2022) used semi-structured interviews and focus groups to consider how people with dementia and their carers and/or relatives evaluated a prominent visual metaphor for dementia (one which maps autumnal trees losing leaves onto the brain/head). The analysis identified three recurring but conflicting responses to the metaphor; that it, (i.) accurately depicts dementia, (ii.) explains a biomedical account of progressive brain deterioration, and (iii.) reinforces inaccurate and/or "hopeless" frames of what having dementia involves. Further research like this, which explores directly and systematically how people with dementia and the public more broadly respond to metaphorical representations is clearly essential if we are to understand in empirical terms how such metaphors are received and the impacts these might have on how people perceive and indeed experience the syndrome.
Notwithstanding this gap in the present study, as a general guiding principle I would suggest that British tabloids should endeavor to better incorporate into their coverage the perspectives of people with first-hand experience of dementia. A step toward this goal could involve media text producersand producers of other types of public discourse around dementia, for that matter -working more closely with dementia charities and advocacy groups. Such groups frequently work with people with lived experience of dementia and, as such, have in place the infrastructure to involve people with dementia and their relatives in such activities in ways that are ethically responsible and empowering for the individuals concerned. Presently, such voices are largely absent from new coverage. Better representation on this front may allow media organizations and other communicators to be guided by the metaphors that people directly affected by dementia themselves find useful for capturing and communicating their lived experiences of this complex health issue.