Metaphors of Depression. Studying First Person Accounts of Life with Depression Published in Blogs

ABSTRACT

This work analyzes the conceptual metaphors of depression in a corpus of 23 blogs written in Catalan by people suffering major depressive disorder. Its main aim was comparative, in order to check whether metaphors detected in previous studies were also used in a new genre and a new language. Their use was confirmed, thus reinforcing the metaphors’ relevance and their conceptual (i.e. non language-dependent) nature. Furthermore, the study broadens the scope of the conceptualization of life with depression with a set of metaphors not attested before, mostly related to social, communicative and medical factors. The results suggest that the containment and constraint that characterize a crucial part of the metaphorical discourse of depression are not only imposed by the disorder itself, but also by contextual factors (such as stigma, lack of communication, or the medical practice perceived as a repressive power) that can have a significant impact on the lives people with depression lead. They also suggest that the very nature of blogging as a genre allows these people to provide more accurate depictions of their condition, thus providing a more comprehensive account of metaphors of life with depression and potentially empowering them.

Introduction

This paper presents a study on the conceptual metaphors found in the discourse of people suffering from major depression in their writings on blogs. Our research is conducted in the framework of a project titled MOMENT: Metaphors of Severe Mental Disorders (Coll-Florit et al., 2018), whose main goal is the application of the conceptual metaphor theory (CMT) to the mental health field.

In recent years, there has been an increasing amount of literature analyzing the use of conceptual metaphors in a wide range of medical conditions, such as cancer (Gibbs & Franks, 2002; Semino et al., 2017; Semino, Demjén, Hardie, Payne, & Rayson, 2018), strokes (Boylstein, Rittman, & Hinojosa, 2007) or diabetes (Goering, 2015). Regarding linguistic analysis on mental health, there are studies in the field of discourse analysis that investigate both patient narratives and mental health practitioners’ discourses (Harper, 1995; Leishman, 2004; Mancini & Rogers, 2007; Zeeman & Simons, 2011). Moreover, a large and growing body of literature in the fields of psycholinguistics and neurolinguistics has investigated the language of people diagnosed with schizophrenia, focusing on their irony identification ability and metaphor comprehension capacity (Bruce & Bodenstein, 2005; Deamer et al., 2019; de Bonis et al., 1997; Iakimova, Passerieux, Laurent, & Hardy‐Bayle, 2005; Langdon & Coltheart, 2004; Mossaheb et al., 2014). However, as noted by Gibbs and Franks (2002), in this last type of study researchers tend to only pay attention to creative and idiosyncratic metaphors, omitting the study of conventional metaphors.

Besides, a considerable amount of literature has been published on metaphor and psychotherapy (cf. systematic reviews by McMullen, 2008; Mould, Oades, & Crowe, 2010; Tay, 2016), where the usefulness of metaphor as a tool to facilitate communication between client and therapist is highlighted. In particular, as pointed out by Tay (2016, p. 274), two main approaches can be distinguished: one in works carried out by mental health researchers, which adopt a functional approach and focus their attention on the effectiveness of metaphor in promoting positive changes in therapeutic processes; and another approach in works by linguistic researchers, which are more concerned with developments in metaphor theory and whose main goal is to describe the contextual richness of metaphors in psychotherapy, rather than their therapeutic function. Nevertheless, some studies by linguists also include recommendations and implications for professional practice. For example, Ferrara (1994) proposes that therapists use metaphors previously elicited by clients to foster their agency. This suggestion goes in line with Semino et al. (2017), who propose the use of empowering metaphors in interactions with people suffering from cancer.

With regard to conceptual metaphor detection in discourses about severe mental illnesses, apart from the works about depression that we are going to present in more detail below, in the field of cognitive linguistics studies have been carried out on obsessive-compulsive disorder (Knapton, 2016a, 2016b; Knapton & Rundblad, 2018) and on schizophrenia, in particular the analysis of metaphors of schizophrenia produced by patients and mental health professionals in documentary films about this mental disorder (Climent & Coll-Florit, 2017; Coll-Florit, Miranda, & Climent, 2019) and the study of metaphors in accounts of voice-hearing (Demjén, Marszalek, Semino, & Varese, 2019).

Several previous works have analyzed metaphors of depression; these have mostly focused on English, and to a lesser extent on Spanish and Chinese. Those focusing on empirical data have analyzed different kinds of corpora such as therapy session recordings (McMullen & Conway, 2002), interviews with people who had experienced depression (Charteris-Black, 2012), radio broadcasts on depression (Semino, 2008), animation films (Fahlenbrach, 2017; Forceville & Paling, 2018), the press (Reali, Soriano, & Rodríguez, 2016) and books – fiction and nonfiction (Demjén, 2011; Reali et al., 2016; Schoeneman, Schoeneman, & Stallings, 2004). However, no first-person accounts published in blogs have been analyzed yet, and this is the main goal of the present study. Specifically, we analyze 23 blogs written by people diagnosed with severe depression. These texts have not been filtered by any therapist or interviewer. In other words, they are texts in which people with depression explain their experience directly.

Kotliar (2016, p. 1203) expounds that in the last decade researchers have highlighted the ways in which the Internet acts as an empowering agent for people affected with different illnesses, though research about online expressions of mental illness is scarce, and about depression it is practically non-existent (Cavazos-Rehg et al., 2016; Lachmar, Wittenborn, Bogen, & McCauley, 2017). In fact, Kotliar (2016) is the only work that deals with depression narratives in blogs, using a thematic analysis methodology. However, there is no previous work analyzing conceptual metaphors in depression narratives in blogs. Therefore, our aim is to fill this gap.

More specifically, our study has a comparative goal: it aims to check whether the metaphors of depression detected in previous studies and in other discursive genres are also used in a corpus of first-person testimony blogs, and in what proportions. In parallel, our study seeks to check whether these metaphors are also used in a language not analyzed so far, since our study focuses on blogs written in Catalan.

According to the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) (American Psychiatric Association [APA], 2013), which is one of the leading publications used by mental health professionals for diagnoses, major depressive disorder (or depression, as it is commonly known) is a serious mood disorder (sadness, loss of interest) which can affect cognition, language, motricity (range and speed) or behavior, and physical symptoms include fatigue and changes in appetite and/or sleeping patterns. Depression is associated with social decline and social malfunction (as summarized by Kupferberg, Bicks, & Hasler, 2016). This means that people with depression have difficulties making social commitments, maintaining quality relationships with others, and feeling a sense of belonging or affection. This social isolation, either to protect themselves from anticipated rejection, or as a result of lack of interest in others or not having the skills needed for an adaptive response, results in action that helps them regain control over themselves.

To add etiological meaning to the DSM-5 diagnostic criteria for depression, Patten (2015) presents eight possible models of depression based on analogies to other well-known diseases (for example, the author presents depression as a degenerative condition, an outcome of toxicity or an injury, among others). Each of these explanations can lead professionals to a specific therapeutic approach, but they also lead the persons diagnosed to different experiences of living with it, and hence different metaphors to talk about it. In fact, metaphors have been used by professionals as progress markers in therapy, signaling changes (Levitt, Korman, & Angus, 2000). And metaphors have also been used by diagnosed persons as a way of gaining agency, to deal with circumstances beyond their control (Gesicki & Nelson-Becker, 2018; Westerbeek & Mutsaers, 2008) and to shape and reflect their relationship with the depression they live with. The point here is to identify whether the person perceives depression as a part of him/herself or as something independent. These are two different approaches to this relationship which have been identified by the literature, although there is not a clear distinction after the analysis of diagnosed people’s narratives (Westerbeek & Mutsaers, 2008).

In this work we assume that patients perceive depression fundamentally as a part of themselves, thus they experience not only biological factors but also psychosocial factors. For instance, the experience of stigma and discrimination can have a big impact. We expect that these contextual factors, which go beyond the purely biological, will emerge in our corpus; this is because the corpus constitutes blog discourses and in these kind of texts, since they are unmediated and written in the first person, the authors tend to explain what is most intimate, most critical or what deeply affects them. As pointed out by Seale, Charteris-Black, MacFarlane, and McPherson (2010, p. 596), “individuals might use the relative anonymity of the Internet to reveal things that they would not discuss in a face-to-face research setting”. As a consequence, our hypothesis is that new metaphors about life with depression, not detected previously in other textual genres, will be revealed.

The rest of this paper is divided into five sections. Section 2 reviews the state of the art regarding CMT and depression; Section 3 describes the corpus under study and the metaphor identification and annotation methodology we used; Section 4 presents the main outcomes of our analysis; Section 5 is devoted to discussing the implications of our findings; and, finally, Section 6 presents our conclusions.

State of the art on metaphors of depression

In 1986, a few years after the emergence of the CMT (Lakoff & Johnson, 1980), Antonio Barcelona published the first study on the metaphors of depression in English (Barcelona, 1986). This qualitative study was based mainly on his own examples, without a systematic collection of informant data or documentary sources. However, it already showed the main metaphors of depression that would later be documented in empirical studies. According to Barcelona, the main metaphor that structures the concept of depression in English is the directional metaphor “ HAPPINESS IS UP/UNHAPPINESS IS DOWN ” (e.g. “ I am in low spirits/Mary is down in the dumps ”), followed by the perceptual metaphor “ HAPPINESS IS LIGHT/UNHAPPINESS IS DARKNESS ” (e.g. “ I feel under a cloud ”).

Regarding metaphors of depression per se, Barcelona highlights the conceptualization of depression as a “ FORCE ” (natural or violent), a “ BURDEN ”, a “ NUISANCE ”, a “ LIVING ORGANISM ” (among which a beast), an “ ENEMY ” in a fight, or a “ BOUNDED SPACE ” (generally a container). In relation to this last source domain, the author also highlights the conceptualization of the person suffering from depression as a container of emotions, formulated as Depression is a substance within the person container. Finally, the author presents the metaphor of the life with depression as a “ JOURNEY ” (e.g. “ He is at the worst stage in his depression ”).

That same year, Stanley Jackson published a study on depression and melancholia (the ancient term for this mental condition). Although it does not fall directly within the field of cognitive linguistics, the author identifies two major recurrent metaphors in the history of writings on melancholy and depression, used since the Classical Greece: being in a state of darkness and being weighed down or weighted down (Jackson, 1986, pp. 396–397).

The first empirical work on metaphors of depression is McMullen and Conway’s (2002), an analysis of the discourse of 21 people diagnosed with major depression. The corpus they studied consisted of therapy session recordings, in English, in which the patients described their experiences of depression to the therapist. The main objective of the study was to determine whether the metaphors described by Jackson (1986) were applicable to current descriptions of depression. Their results show that the prevailing metaphor is “ DEPRESSION IS DESCENT ”, which subsumes 90% of the depression metaphor instances in the corpus. In particular, the authors note that such metaphors are not expressed in terms of a weight or a strain but in terms of an unrestrained movement in a physical space, from a high location to a lower location or place. In general, patients describe the place they descend to as dark, underground and deep, with recurring examples such as pit, hole or gutter. The remaining 10% of the instances correspond to examples of the three following metaphors: “ DEPRESSION IS DARKNESS ” (usually expressed in terms of cloudy or rainy weather), “ DEPRESSION IS WEIGHT ” (generally unrelated to descent) and “ DEPRESSION IS CAPTOR ” (a metaphor not mentioned in Jackson’s account, expressing a sense of restriction or coercion).

Another relevant work regarding metaphor and depression is Semino (2008, pp. 178–190), which detected and formulated metaphors used in a radio phone-in program on clinical depression. The corpus analyzed amounted to 8,000 words and included contributions from 14 callers: nine individuals with depression, two parents who cared for children diagnosed with the condition, one doctor and two more people not directly related to the illness. The study shows that the domains most used by people speaking about their depression are “ UP/DOWN ”, “ ENCLOSED SPACE ”, “ JOURNEY ” or “ PHYSICAL ENTITY ” (among them one “ BURDEN ”). More scarcely, depression is equated to an “ ENEMY ” in a fight. Additionally, in some cases the individual with depression is conceptualized as a “ MALFUNCTIONING MACHINE ”.

The work of Charteris-Black (2012) is particularly relevant among studies in English. One of its main goals was to determine whether there are differences between men and women in the types of metaphors used to talk about depression. To do so, the author used a corpus of 22 interviews with people who had suffered from depression (11 men and 11 women) and had recovered from it at the time of the interview. The corpus for this quantitative study consisted of a 300,000 word balanced sample (approximately 150,000 for each subgroup, men and women). The analytical procedure was based on automatic detection of candidate metaphorical expressions (ME) by means of keywords that previous research had shown to be commonly used metaphorically by people experiencing depression, e.g. darkness, descent and weight. At the next stage, these keywords were analyzed in the context of the concordance by establishing whether they had a more basic meaning, i.e. following Pragglejaz Group (2007) method.

Although some differences are found between men and women in how they use metaphors, the results do not show significant differences in the frequency and types of metaphors they use to talk about depression: “Gender is not therefore an important influence on metaphor type and men and women have the same expressive resources for talking about depression” (Charteris-Black, 2012, p. 207). As to types of metaphors found, three of the four predominant metaphors in Charteris-Black’s (2012) corpus correspond to those already identified by McMullen and Conway (2002). As in this previous study, the most common is “ DEPRESSION IS DESCENT ”; likewise, “ DEPRESSION IS WEIGHT ” and “ DEPRESSION IS DARKNESS ” feature in third and fourth position, respectively.

The most remarkable feature of Charteris-Black’s (2012) study is what the author refers to as metaphors of containment and constraint. These metaphors are second in frequency in his corpus and account for 24% of the ME identified. They are expressed in terms of a container in two ways: first, the individual with depression is conceptualized as a container of sad emotions; and second, depression is conceptualized as a container surrounding and constraining the individual, as expressed by words such as come out, pour out, escape and release. According to Charteris-Black, these metaphors are completely different from McMullen and Conway’s (2002) “ DEPRESSION IS CAPTOR ”, which necessarily entails conceptualizing depression as an animate entity; a container is clearly inanimate.

Regarding this point, it should be noted that the depression as a container metaphor had already been formulated by Barcelona (1986), although in a more intuitive way, and had also been mentioned by Semino (2008). The main contribution of Charteris-Black’s work is the fact that it allows empirical validation using a corpus of interviews, while also providing a detailed and unified explanation of the types of container, the disorder and the self, with implications for therapists’ professional practice.

It is also important to note that Charteris-Black’s (2012) category of containment and constraint metaphors has a certain overlap with the way McMullen and Conway’s (2002) “ DEPRESSION IS DESCENT ” metaphor is conceptualized. This is a category in which McMullen and Conway (2002) include not only the expression of the physical movement of descent, but also the destination, often conceived as a container (e.g. well). However, the overlap is only partial, as there are numerous cases of the conceptualization of depression as a container that do not involve descent, for example bubble. Table 1 shows a comparative summary of all the source domains from the studies on metaphors of depression in English discussed above.

Table 1. Comparative summary of the source domains in the main studies on metaphors of depression in English

Barcelona (1986)	DARKNESS	BURDEN/NUISANCE	UP/DOWN	BOUNDED SPACE	LIVING ORGANISM	JOURNEY	ENEMY (WAR)	FORCE
Jackson (1986)	DARKNESS	WEIGHTED DOWN
McMullen and Conway (2002)	DARKNESS	WEIGHT	DESCENT		CAPTOR
Semino (2008)		PHYSICAL ENTITY (THING)/BURDEN	UP/DOWN	ENCLOSED SPACE		JOURNEY	ENEMY (WAR)
Charteris-Black (2012)	DARKNESS	WEIGHT	DESCENT	CONTAINMENT and CONSTRAINT

Another type of approach is that of the works based on the in-depth metaphor analysis of a single subject. In this regard, some of the most relevant studies are Schoeneman et al. (2004) and Demjén (2011), which analyze works written by authors suffering from depression. More specifically, Schoeneman et al. (2004) analyze the memoirs of the writer William Styron, in which they identify the main metaphors of depression we have described so far. And Demjén (2011) focuses on metaphors of mental states in Sylvia Plath’s work, a study in which the identification and analysis of the “ SPLIT SELF ” metaphor stands out inasmuch as it is a metaphor not mentioned in the previous studies.

It should also be noted that studies have been conducted on metaphors of depression in films. The first to study this topic was Fahlenbrach (2017), who identifies two main source domains to conceptualize depression in short animated films: “ DARK PLACE ” and “ BEING ENCLOSED BY RAIN ”. Likewise, Forceville and Paling (2018) analyze visual metaphors for depression in wordless animated films. In this case, the two distinctly dominant domains are “ DARK CONFINING SPACE ” and “ DARK MONSTER ”. Therefore, we observe that some of the main metaphors found in textual corpora are also relevant in films.

Metaphors of depression have also been studied in other languages. In Spanish, Reali et al. (2016) carried out a corpus study based on CREA, a corpus of written Spanish which includes press and books (fiction and nonfiction). In particular, they focused on texts from four Latin American countries: Argentina, Colombia, Chile and Mexico. Their corpus analysis was based only on sentences where the target word depresión (depression) appeared. Thus, the analysis started with an automatic pre-selection of potential ME, as in the quantitative study by Charteris-Black (2012), but in this case only using the keyword depresión. Their results show that the two most frequent metaphors are “ DEPRESSION IS A PLACE IN SPACE ” (generally bounded, deep and dark, as reported in previous studies) and “ DEPRESSION IS AN OPPONENT ” (including cases of monsters and beasts).

With regard to Chinese, it is worth mentioning the study by Yu (1995) about the related emotion of sadness, since the author identifies three of the main source domains found in the English studies on depression: “ SADNESS IS DOWN ”, “ SADNESS IS DARK ”, “ SADNESS IS HEAVY ”. Another related work in Chinese is Pritzker (2007), where metaphors of brain and heart are analyzed in three interviews with people suffering from depression. The fact that very similar metaphors (in some cases actually the same ones with different formulations) are found in the three languages points to a non language-dependent nature of several metaphors of depression, something which could be confirmed by extending this research to new languages.

Finally, it should be noted that related experimental studies have been carried out in the field of psycholinguistics. These last studies did not deal precisely with metaphor types for depression, but with the metaphorical capacity of people diagnosed with depression. In this sense, the recent study by Kauschke, Mueller, Kircher, and Nagels (2018) reveals that the understanding and production of metaphorical expressions is completely preserved in patients with depression, with no significant differences from the healthy control group.

Corpus and methodology

In this section we present the corpus under study and ethical issues, followed by the methodology used for detecting and formulating conceptual metaphors.

Corpus

The corpus under study consists of blogs fostered by a mental health patient organization headquartered in Catalonia. These blogs are written by the organization’s members, people who have experienced or are going through a mental disorder. In all instances, they are first-person accounts of the experience of living with a mental health condition and the overall aim is to give voice to the collective and raise mental health awareness.

More specifically, this study includes a total of 23 blogs authored by individuals who described themselves as having been diagnosed with major depression. The language the blogs were written in is Catalan. We gathered posts published between 2016 and 2018 (the year when the blogs started to be published and the year when this research started, respectively). Posts where the author did not describe personal experiences were discarded, e.g. those where they comment on films or literary works. The resulting corpus has a total of 13,641 words.

With regard to ethical issues, we have opted for preserving the anonymity of the subjects, following the criteria of Semino et al. (2018, p. 49): “there is some consensus that anything a person posts to a forum that is open to public view on the web can be used as research material without seeking informed consent from the individual contributor, as long as anonymity is fully preserved”. Nevertheless, we informed the organization hosting the blogs of our plans. Ethical approval was obtained from Open University of Catalonia (UOC).

Methodology

In keeping with our hypothesis, which was based on a person-oriented approach to mental health, we analyzed not only the metaphors of depression itself, as did the works reviewed in the state of the art section, but also those aspects related to the experience of living with a severe mental disorder, i.e. all topics relevant to severe mental disorders, such as symptoms, medication, communicative aspects or related emotions.

This study was based on a fully manual corpus annotation, i.e. without previous automatic pre-detection of keywords. In particular, the method for metaphor detection and formulation in corpora by Coll-Florit and Climent (2019) was employed. This method involves (i) manual pre-selection of candidate MEs; (ii) the use of a standard method for metaphorical focus identification; (iii) the use of compendia of metaphors for labeling metaphor domains; and (iv) if no suitable models are found in these compendia, the use of labeling strategies based on dictionary definitions.

Annotators are issued a detailed annotation guide and a set of auxiliary documents, among them a list describing thematic fields specific to severe mental disorders, and two metaphor compendia, one specific to mental health and one general-purpose. The thematic fields document describes the topics considered relevant for the analysis of discourse in mental disorders as compiled from Climent and Coll-Florit (2017) and Coll-Florit et al. (2019). The compendium of mental health metaphors consists of metaphor formulations with representative examples compiled from extant literature on metaphor and mental health, including all the works on metaphor and depression reviewed in the state of the art section. The general-purpose metaphor compendium is the Master Metaphor List (Lakoff, Espenson, & Schwartz, 1991).

The method is divided into two main phases: selection of candidate ME (working hypothesis formulation) and analysis of ME (hypothesis verification/rejection and domain annotation).

Phase 1: selection of candidate ME

The analyst reads the sentence and its context to capture the general meaning and decides whether the sentence contains one or more candidate ME for mental health on the basis of two conditions: (i) intended metaphoricity and (ii) thematic relevance. As a result of this stage, clauses with candidate MEs are extracted from the general text.

Phase 2: analysis of ME

The resulting set of candidate ME is analyzed in a maximum of two steps. The first involves checking the candidate ME against the compendia of metaphors detected in previous works, with priority given in this case to metaphors of depression. Candidate MEs are compared to examples and formulations included in the compendia and, if a possible match is found, then the Metaphor Identification Procedure (MIP) (Pragglejaz Group, 2007) is applied to the clause in order to validate (or reject) metaphoricity (see Coll-Florit & Climent, 2019, pp. 54, 61–63). In positive cases, this procedure results in both the selection of metaphorical focus and the labeling of metaphor domains as per the compendium.

If the potential ME does not fit any of the metaphor labels in the compendia, a reverse second step is employed: the analyst first applies the MIP to all the words in the clause and if a word is deemed metaphoric, it is annotated as the metaphorical focus. Next, the analyst labels the domains of the underlying metaphor using several inference strategies, which are mainly based on substitution by dictionary keywords (see Coll-Florit & Climent, 2019, pp. 54–58, for a detailed explanation and examples of such strategies). For the application of the MIP and the domain formulation strategies, since the corpus is in Catalan, the dictionary used as the reference source is DIEC2 (Institut d’Estudis Catalans, 2019), published by the official authority on this language.

The reliability of this method was empirically assessed in an inter-annotator agreement test, which is described in detail in Coll-Florit and Climent (2019, pp. 65–69), where two linguists annotated a test corpus of 4,143 words produced by two subjects with severe mental disorders, published in a book describing personal stories and experiences of mental disorders. Results showed a high degree of agreement (97.6%; kappa value 0.79) regarding the annotation of metaphorically used words. In the labeling of the conceptual domains, the analysts proposed the same formulation for both metaphor domains in 71% of the cases and agreed in at least one domain in 87% of the cases.

In this work, in line with previous works in manual metaphor corpus annotation (Semino et al., 2017, 2018), the analysis was carried out by three linguists in a process where one lead analyst’s annotations were subsequently verified independently by the other two analysts to ensure coding accuracy and consistency. 302 of 314 ME annotations made by the first analyst were accepted by the revisers (96.18%). This high level of consensus can be attributed to the fact that all three linguists had been working together in the field of mental health metaphors. Lastly, a psychologist participated in the data interpretation stage to ensure the robustness of the analysis from a mental health expert’s point of view.

Results

A total of 302 metaphorical expressions were detected corresponding to three broad domains related to mental health: (i) metaphors of depression; (ii) metaphors of interpersonal communication and social context; and (iii) metaphors of medicine and professional treatment. As can be seen from the data in Table 2, 69.53% of the ME are metaphors of depression, which have three target domains: “ LIFE WITH DEPRESSION ”, “ DISORDER ” and “ PEOPLE WITH DEPRESSION ”; namely, the process of living with a mental disorder, the disorder itself and the person with depression as an individual.

Table 2. All of the metaphorical expressions identified in the corpus

	ME	%
Metaphors of depression	210	69.53%
Metaphors of interpersonal communication and social context	70	23.17%
Metaphors of medicine	15	4.96%
Other	7	2.31%
Total	302

Metaphors of interpersonal communication and social context represent 23.17% of the cases and encompass the following three target domains: “ PREJUDICE ”, “ COMMUNICATION/LACK OF COMMUNICATION ” and “ PEOPLE OF THE SOCIAL CONTEXT ” (mainly family and friends). Finally, metaphors of medicine account for 4.96% of the cases and they also have three target domains: “ DOCTOR ”, “ MEDICAL CONSULTATION ” and “ TREATMENT ”.

As had been expected, the metaphors of depression are the most prevalent in the corpus. However, what is most noteworthy about these results is the significant presence of metaphors of communication and society and, although less frequent, metaphors of medicine. Altogether this reveals the importance of social and contextual aspects in the experience of major depression.

In the following subsections we will analyze in more detail the results for each of the broad domains under scrutiny: metaphors of depression, metaphors of interpersonal communication and social context and metaphors of medicine.

Results for metaphors of depression

We understand metaphors of depression in a broad sense, including as target domains not only the conceptualization of the disorder itself but also the person suffering from depression and his/her life. Table 3 shows the ME distribution for each of these target domains in our corpus.

Table 3. Target domains of metaphors of depression

Target domains	ME	%
LIFE WITH DEPRESSION	97	46.19%
DISORDER	70	33.33%
PEOPLE WITH DEPRESSION	43	20.47%
Total	210

As can be observed, 46.19% of the ME relate to the life with depression, in other words, the experience of the illness as a process, which is mainly conceptualized in terms of either a journey or a war. Secondly, 33.33% of the ME reify the disorder, which is mostly conceptualized as a living organism. Lastly, 20.47% of the ME refer to the individual with depression himself/herself, who is mostly conceptualized as a container of emotions or a bounded space. We present and exemplify these results in more detail below.

Metaphors of life with depression

As shown in Table 4, life with depression is mostly conceptualized through two main source domains, “ WAR ” and “ JOURNEY ”, with a predominance of war metaphors.

Table 4. Source domains of metaphors of life with depression

Source domains	ME	%
WAR	54	55.67%
JOURNEY	41	42.26%
Other	2	2.06%
Total	97

The conceptualization of conflict situations through war metaphors (1) and the conceptualization of processes that have an eventual goal or purpose through journey metaphors (2) are widely documented in CMT. The work of Semino et al. (2017) is particularly relevant in this regard, as it is specifically dedicated to analyzing these two types of metaphors in the discourse of health professionals and cancer patients. Our data confirm that war and journey metaphors are used to structure not only the subjective experiences of physical illnesses such as cancer, but also the process of living with a mental health condition.

• (1) a. La meva lluita continua avui en dia.

  My fight goes on today.

• b. El que compta és no rendir-se.

  What counts is not surrendering.

• c. Quan culpen el teu trastorn et llencen un míssil a la teva línia de flotació.

  When your disorder is blamed, you get a missile launched at your waterline.

• (2) a. Aquest camí no és curt.

  This is not a short road.

• b. I aquí seguim, sense tenir molt clar el rumb.

  And here we are, not quite sure of the direction.

• c. Les persones coneixedores del meu trastorn mental van remar a favor del fet de voler viure sola la maternitat.

  People who knew about my disorder rowed in support of my wanting to live motherhood alone.

In fact, as we have pointed out, the metaphors for war and journey were already previously detected in the works by Barcelona (1986) and Semino (2008) on metaphors of depression. Specifically, regarding the source domain “ WAR ”, these works document conceptualizations of depression as an enemy or the opponent in a struggle. But what is more interesting in our results is precisely the types of ontological correspondences emerging from the data (i.e. the specific mapping of entities between the source and target domains): not only is the illness an enemy (3.a), stigma and discrimination are, too (3.b).

• (3) a. [La meva vida] és una lluita per vèncer el monstre que em domina.

  [My life] is a fight to defeat the monster that commands me.

• b. Seguirem lluitant contra l’estigma.

  We will continue to fight the stigma.

Metaphors of war can therefore have different functions: when the illness is the enemy, the individual with depression is often conceptualized in an inferior position with respect to it, and the illness is seen as an entity exerting control over him/her; therefore, the sufferer’s lack of control or capacity for agency is emphasized. However, when prejudice is the enemy, people with depression do not usually appear as victims or defeated, but quite the opposite: they are seen as tenacious fighters who struggle to change a reality that they do not like. Just as Semino et al. (2017) point out, war metaphors are not negative by default, since they can have empowering uses.

Metaphors of the disorder

In this subsection we present the results for metaphors that have the disorder as their target domain. As shown in Table 5, the most frequent metaphor, found in 30% of the expressions in this category, is “ DEPRESSION IS A LIVING ORGANISM ”. This is usually instantiated in terms of a monster, an evil being or ghost (4.a), or a person with whom you live (4.b) and to whom you attribute characteristic human actions (4.c).

• (4) a. Els nostres fantasmes interiors.

  Our inner ghosts.

• b. Durant un parell d’anys he conviscut amb una depressió severa.

  For a couple of years I’ve lived together with a major depression.

• c. La meva depressió no fa vacances …

  My depression doesn’t take holidays.

Table 5. Source domains of metaphors of the disorder

Source domains	ME	%
LIVING ORGANISM	21	30%
DESCENT	12	17.14%
THING	8	11.42%
WEIGHT	7	10%
UNBALANCE	6	8.57%
CONTAINER	5	7.14%
FORCE	5	7.14%
DARKNESS	4	5.71%
PLACE	2	2.85%
Total	70

Secondly, metaphors related to loss of control over emotions stand out, where the disorder is conceptualized as an uncontrolled downward movement (5), a type of metaphor that has been coded in previous studies as “ DEPRESSION IS DESCENT ”. On the same subject, McMullen and Conway (2002, p. 181) state that: “representing our experience of what we call depression in images of descent, we might be compounding the sad affect that appears to be the core of depression with associations of failure and loss of control”. Likewise, Charteris-Black (2012, p. 214) argues “it seems that losing control is an essential stage in the communication of depression”.

We also found the metaphor “ DEPRESSION IS UNBALANCE ”, where the disorder is conceptualized as a loss of stability (6). As pointed out by Johnson (1987, pp. 88–89) when describing the metaphorical extension of the BALANCE image schema: “we tend to seek temporary homeostasis where we are emotionally balanced, stable […]. The ideal is a balanced personality. This requires that sufficient weight be given to each of the parts or dimensions of our character. The emotions must be sufficiently under control […] Problems can weigh on our minds throwing us out of balance”. We suggest that this conceptualization may be directly related to “ DEPRESSION IS DESCENT ”, since the loss of balance can cause a fall into a negative situation where emotions are out of control.

• (5) a. Em vaig acabar enfonsant en una depressió.

  I ended up sinking into a depression.

• b. Caiguda, recuperació, caiguda …

  Fall, recovery, fall …

• c. No estem sols ni ens els moments de les pitjors davallades.

  We are not alone even in the moments of the worst descents.

• (6) Tenia inestabilitat, negativitat, era poc constant.

  I felt unbalanced, negative, I was inconstant.

Thus, although the metaphor “ DEPRESSION IS DESCENT ” is not as pervasive as in the studies of McMullen and Conway (2002) or Charteris-Black (2012, p. 214), the presence of metaphors related to the loss of control over emotions is still significant.

Other metaphors for the disorder identified in our corpus are “ DEPRESSION IS A THING ”, usually a nuisance (7); “ DEPRESSION IS A WEIGHT ” (8) and “ DEPRESSION IS A FORCE ” that paralyzes, oppresses or shakes you (9). Additionally, we found examples of “ DEPRESSION IS A CONTAINER ”, primarily deep and dark containers, with the most recurrent and paradigmatic example being a well (10), and “ DEPRESSION IS DARKNESS ” (11).

• (7) Ningú hauria d’espolsar-se el teu patiment.

  No one should brush aside your suffering.

• (8) Les meves sensacions, lluny de millorar, es van tornar molt pesades.

  My feelings, far from improving, became very heavy.

• (9) Vaig tenir una sacsejada força intensa.

  I had a quite intense shaking.

• (10) Em sentia al fons d’un pou profund.

  I felt like I was at the bottom of a deep well.

• (11) [La gent de l’associació] van ser un far en un mar d’ombres per a mi.

  [The people at the association] were a beacon in a sea of shadows for me.

This shows that the main metaphors identified in previous studies of English corpora are also found in our Catalan corpus.

Metaphors of people with depression

Finally, in this section we present the metaphors of the individual with depression himself/herself. As can be seen in Table 6, the predominant metaphor is “ PEOPLE WITH DEPRESSION ARE CONTAINERS ”, which is the category of over half the ME for this target domain.

Table 6. Source domains of metaphors of people with depression

Source domains	ME	%
CONTAINER	23	53.48%
SPLIT SELF	11	25.58%
THING	5	11.62%
MACHINE	2	4.65%
Other	2	4.65%
Total	43

In particular, the individual with depression is conceptualized as a container of negative emotions (12.a), generally, as theorized by Charteris-Black (2012), in terms of containment and constraint, i.e. a closed tridimensional space (12.b) that oppresses him/her and from which he/she cannot escape (12.c). Therefore, this metaphor is directly related to “ SADNESS IS CAPTIVITY ” (McMullen & Conway, 2002).

• (12) a. El dolor va per dins.

  The pain is on the inside.

• b. El fet de tancar-me en mi mateix només va agreujar més la situació.

  The fact that I locked myself inside myself only made the situation worse.

• c. No podem fugir de nosaltres mateixos.

  We can’t run away from ourselves.

The second most frequent metaphor here, represented in 25.58% of the ME identified, is the “ SPLIT SELF ” metaphor (Lakoff & Johnson, 1999; Demjén, 2011, pp. 17–20), in which the individual with depression is conceptualized as a divided entity, or with two people coexisting inside him/her (13). In these conceptualizations, the Subject, i.e. “the locus of a person’s Essence, that enduring thing that makes us who we are” (Lakoff & Johnson, 1999, p. 269), is the healthy one and the other part of the self is the ill one.

• (13) a. Feia tot el que podia per combatre’m, però la part malalta em superava.

  I was doing my best to fight myself but the ill part of me was overcoming me.

• b. Quan estic malament sóc una altra persona.

  When I’m ill I’m someone else.c. Tornar a ser la personaque jo sé que era i que soc. To go back to being the person I know I was and I am.

As noted above (cf. state of the art section), this last metaphor had only been mentioned in one of the previous studies: the work by Demjén (2011) on metaphors of mental states in Sylvia Plath’s work. Thus, the results of our study confirm that this is a significant metaphor in the discourse of people suffering from depression. The “ SPLIT SELF ” metaphor has already been documented in the study on metaphors of schizophrenia (Coll-Florit et al., 2019) – in keeping with schizophrenia’s literal meaning, split mind. Its appearance in this study reveals it as a self-conceptualization metaphor used not only by people diagnosed with a psychotic disorder (such as schizophrenia), but also by people with an affective disorder (such as depression).

In some metaphors the sufferer conceptualizes himself/herself as an easily breakable or non-valuable thing (14).

• (14) Em feien sentir culpable i fràgil.

  They made me feel guilty and fragile.

These metaphors are comparable to those found by Semino (2008, p. 183) “in which the sufferer is presented as a machine that is not working properly”, although those exemplified in (14) are more general since machines are a specific kind of thing – but in both cases, those things show weak properties.

Results of metaphors of interpersonal communication and social context

Although people living with depression have social functioning difficulties (see the Introduction section for an overview), previous studies on metaphors of depression have omitted certain factors that can have a decisive effect on their experiences of living with a mental disorder, such as prejudice or relationships with people in their social surroundings. Thus, one of the main contributions of this study is the incorporation of metaphors related to these factors: interpersonal communication and social context. Specifically, we have identified 70 metaphorical expressions, for which the target domains are “ PREJUDICE ”, “ COMMUNICATION/LACK OF COMMUNICATION ” and “ PEOPLE OF THE SOCIAL CONTEXT ” (primarily family and friends).

Regarding the frequencies observed, as shown in Table 7, metaphors of prejudice are the most prevalent, represented in 42.85% of the ME in this category; this shows the high incidence of discrimination in people’s experiences of living with depression. More specifically, the most frequent metaphors are: i) “ PREJUDICE IS A WEIGHT ”, where the prejudice is conceptualized as a heavy object that people with depression cannot hold (15.a); ii) “ PREJUDICE IS A FORCE ”, mainly a force that exerts pressure on them (15.b); and iii) “ PREJUDICE IS A MARK ”, where the prejudice is conceptualized as a label or visible mark worn by them (15.c).

• (15) a. Els prejudicis i, en definitiva, la desconeixença sobre el tema han estat una llosa.

  Prejudices and, in general, lack of knowledge on the subject, have been a stone slab.

• b. Em sento pressionada constantment a “aguantar” les emocions, a reprimir-les, per no ser titllada de boja, histèrica, depressiva.

  I feel constantly pressured to put up with emotions, to repress them, so as not to be labeled crazy, hysterical, depressed.

• c. Ser un mateix, aprendre a dir no, no doblegar-se i no deixar-se etiquetar per ningú.

  To be yourself, to learn to say no, not to bend to anyone, not to be labeled by anyone.

Table 7. Metaphors of interpersonal communication and social context

Target domains	ME	%
PREJUDICE	30	42.85%
COMMUNICATION/ LACK OF COMMUNICATION	21	30%
PEOPLE OF THE SOCIAL CONTEXT	10	14.28%
Other	9	12.85%
Total	70

The second most frequent type of ME, directly linked to prejudice, is a group of metaphors whose target domain is communicative aspects. Communication (and its opposite, lack of communication) with society is primarily conceptualized in terms of closeness and distance (16.a). In particular, lack of communication from society to the individual with depression marks such a distance that it ends up causing the individual’s isolation (16.b). Therefore, the individual’s relationship with society is also metaphorically conceptualized in terms of limitation and restraint.

• (16) a. Segurament, el fet que ens marquin les distàncies, sense voler comprendre més sobre el que realment ens passa, és moltíssim més complex.

  Surely, the fact that they are establishing distances with us, without wanting to understand more about what really happens to us, is much more complex.

• b. Les relacions socials no ens fan res més que mal i acabem, en moltes ocasions, en una situació d’aïllament social.

  Social relationships only hurt us and we often end up in a situation of social isolation.

• c. El patiment perquè ets exclòs, infantilitzat o menystingut, no té res a veure amb el fet de tenir un diagnòstic de trastorn mental.

  Suffering because you are excluded, infantilized, or undervalued has nothing to do with your having a mental disorder diagnosis.

Furthermore, society is presented as a third container – in addition to the containers of the self and the disorder (Charteris-Black, 2012) – from which the individual with depression feels excluded (16.c). Hence, these metaphors reveal that the individual’s desire is to dissolve not only the limitations imposed by the disorder itself, but also those conditioned by the lack of communication with society. Thus, these metaphors show that, despite the need of people living with depression to be alone, feeling excluded from society constrains them.

Finally, we found a smaller set of metaphors for the most relevant people in sufferers’ social contexts, where family and friends are conceptualized primarily as pillars that support them (17).

• (17) Jo tinc un suport millor que el que reben altres persones amb trastorn mental.

  I have a better support than other people with mental disorders.

Results of metaphors of medical practice and diagnosis

Lastly, in this section we present metaphors related to medical practice and diagnosis. As pointed out above, this is the area of mental health for which we found the fewest ME in our corpus, representing only 4.96% of all the ME identified (cf. Table 2). We found conceptualizations of medicine as a repressive power that denies people with depression’s agency (18.a), the doctor as a prosecutor (18.b) or the doctor as a captor that locks the individual with depression into a diagnosis, therefore conceptualizing the diagnosis as a bounded space (18.c). Thus, once again, we have metaphors related to containment and constraint.

• (18) a. Només cal […] parar atenció al que diem i no anul·lar-nos.

  You just have to pay attention to what we say and not cancel us out.

• b. Després del seu interrogatori de gairebé 30 minuts, i d’un parell de testos estandaritzats [sic] que em va fer omplir, l’única cosa que em va dir era que parés de plorar.

  After his almost 30-minute interrogation, and a couple of standardized tests he had me fill out, all he told me was to stop crying.

• c. Et tanquen dins un diagnòstic.

  They lock you into a diagnosis.

In short, most of the metaphors found in this category are strongly critical of psychiatric practice and the medical system since they are seen as limiting the patient’s capacity for agency. A plausible explanation for this result is that the writers of these blogs have gone through a medical system that is more problem-focused than person-focused, and as a result they are looking for more empathy and understanding of their suffering from medical personnel, as well as a society that welcomes them as equals, overcoming the stigma often associated with mental health diagnoses. Moreover, this result is directly related to the findings in the work by Westerbeek and Mutsaers (2008) on depression narratives in autobiographical texts by patients. These authors found that, due to the fact that antidepressants have many side effects and patients usually have cyclic relapses, patients tend to lose faith in the medical system: “the conclusion that antidepressants and other forms of therapy cannot help them makes most of the authors question the biochemical explanation as the only possible explanation” (Westerbeek & Mutsaers, 2008, p. 40).

Discussion

Our study presents a comprehensive and quantified account of the main types of metaphors used by people affected by a depressive disorder. Specifically, the corpus under study provides further evidence of all the metaphors of depression that had been identified in previous studies and, going beyond that, unveils previously undetected types of metaphors.

Firstly, 30% of the metaphors we detected have target domains that do not belong to depression per se, but to related aspects such as communication, social context or medicine. The blog writers use these metaphors to express the existence of important factors in their experience of living with depression, mainly factors that make it worse: prejudice, lack of communication, and the disabling powers exercised by medical practitioners. These metaphors – in conjunction with the conceptualization of stigma as an enemy in metaphors of war – suggest the importance of contextual aspects in the experience of living with a mental illness; just as McMullen and Conway (2002, p. 179) point out, citing Kleinman and Kleinman (1986): “depressive illness discloses not only how an individual relates to society, but also how society affects individuals”. As a corollary to this, support from the most relevant parts of their social context is seen as a factor that can alleviate suffering.

And secondly, we find that our corpus gives importance to the human being who is experiencing depression (not just the abstract concept of “depression”), as there is a significant presence of metaphors that specifically conceptualize the individual with depression; and, on the contrary, there is a relative scarcity of some of the metaphors for the disorder itself that have been highlighted in previous studies, such as depression as darkness or depression as a container.

Regarding this latter type of metaphor, in our corpus it is much more usual to conceptualize the individual with depression as a container than the disorder as a container – specifically, the individual as a container of negative emotions. Nevertheless, our data also shows the conceptualization of two new containers that can contain the individual with depression: 1) society conceptualized as a container from which he/she feels excluded, mainly due (once again) to stigma and discrimination, and 2) the medical diagnosis as a container in which the individual feels locked up. Therefore, these metaphors reveal that people with depression are concerned not only with the limitations of the disorder itself, but also with the restrictions imposed on them by society and the medical system. These results therefore confirm Charteris-Black’s (2012) thesis that patients are conceptualized predominantly in terms of containment and constraint, but they also expand upon this thesis by detecting new kinds of containers restricting them.

This study also extends the scope of metaphors of depression in other directions: it detects a new captor, the psychiatrist; a new weight and force, prejudice; and it ratifies the significance of the “ SPLIT SELF ” metaphor in the discourse of mental illnesses other than schizophrenia. It is also interesting to see that our study broadens conceptualization of loss of control by people with depression in two new aspects: first, with metaphors motivated by the “ BALANCE ” image schema, where “ UNBALANCE ” stands for loss of control; and second, observing metaphors related to medicine and health professionals that conceptualize a restriction on people with depression’s capacity for agency, thus associating disempowerment with a loss of sufficient control over their lives. Table 8 summarizes the main conceptual metaphors of life with depression found in our corpus.

Table 8. Main conceptual metaphors of the experience of living with depression found in our corpus. Cases not accounted for in previous studies on metaphors of depression are highlighted with an asterisk

THE DISORDER IS	A LIVING ORGANISM (PERSON, BEAST, MONSTER) DESCENT *UNBALANCE A CONTAINER DARKNESS A THING A WEIGHT A FORCE A PLACE
PEOPLE WITH DEPRESSION ARE	CONTAINERS SPLIT SELVES * (FRAGILE, NON-VALUABLE) THINGS MACHINES
LIFE WITH DEPRESSION IS	A JOURNEY A WAR • THE DISORDER IS AN ENEMY • *THE STIGMA IS AN ENEMY
*PREJUDICE IS	A WEIGHT A FORCE A MARK
*LACK OF COMMUNICATION IS	DISTANCE
*SOCIETY IS	A CONTAINER
*FAMILY AND FRIENDS ARE	SUPPORTS
*THE MEDICAL PRACTICE IS	A REPRESSIVE POWER • THE DOCTOR IS A CAPTOR • THE DOCTOR IS A PROSECUTOR
*THE MEDICAL DIAGNOSIS IS	A CONTAINER

In this study we hypothesized that the nature of the textual genre studied, first-person accounts published in blogs, would reveal new metaphors. Regarding the effect of blogging as a textual genre, Kotliar (2016) argues that its accessibility, the promise of anonymity and the lack of an interviewer’s gaze or any editing of online narratives (as “online extensions to personal diaries”) allow people with depression to form more accurate and satisfactory depictions of their mental condition. In the same way, Seale et al. (2010, p. 600) claim that online writing provides livelier insights into experiences than those gained in researchers’ constructed environments, and people with depression are more likely to discuss sensitive or taboo subjects when writing online.

Hence, Kotliar (2016) posits that this medium of communication acts as an empowering agent for people affected with different illnesses. According to this author, the genre helps the individual with depression to face etiological inexplicability and break out of depression’s social isolation by forming communal bonds with their readers and other bloggers. He argues that the process may even help toward self-redefinition and recovery, and upon this basis he recommends using this medium in therapeutic processes.

Therefore, the finding of new metaphors for life with depression, including contextual factors such as social prejudice, confirms our hypothesis, thus supporting the idea that the blogging genre allows people with depression to express what really concerns them freely, and therefore offers a better basis for detecting their conceptualizations. Besides, our results are in line with the proposal by Kotliar (2016) that blogging can act as an empowering agent, since its ability to offer agency allows people with depression to regain a degree of control over their life and emotions, and this increased sense of control can alleviate their suffering and may therefore be beneficial.

As mentioned above, McMullen and Conway (2002) and Charteris-Black (2012) posit that patients’ negative feelings – especially relating to loss of control over their emotions – are key to their communications of depression. These authors’ reference to control over emotions is not surprising, given that depression is largely defined by the presence of negative feelings and emotions such as sadness, hopelessness, guilt and worthlessness, and is associated with others such as anxiety (APA, 2013). With this in mind, it is very striking that our corpus does not provide metaphors that conceptualize these emotions – i.e. it has no metaphors with emotions such as “ ANXIETY ” or “ FEAR ” as target domains. This makes us think that emotions must be implicit in sufferers’ metaphorical conceptualizations of depression.

In this respect, Goatly (2011, pp. 165–168) points out that one of metaphors’ main functions is to express emotion, although the study of this aspect has been largely neglected – perhaps because theorists have instead focused on conceptual meaning. For instance, he points out that in the extreme case of swear words such as hell “the mapping or transfer of features is not a matter of conceptual or ideational meaning, but the transfer of negative feelings about […] eternal punishment to the meaning of the swear word”.

Thus, possibly, emotions are implicit in those metaphors related to the encompassing concepts of control and loss of control. Many metaphors in our corpus express different variants of loss of control: unbalance, uncontrolled descent, inescapable seclusion in a space, threat by a frightening evil being or captor, the exercise of a repressive power by the medical status quo, social exclusion, etc. These metaphors are formulated to a great extent by using source domains such as “ WAR ”, “ ENCLOSED SPACE ”, “ EVIL BEING ”, “ DARKNESS ”, “ DESCENT ”, “ BODY MARKING ” or “ EXCLUSION ”. We suggest that such domains may be associated with negative emotions such as anxiety, fear, confusion or sadness, which, as in the case described by Goatly (2011), are transferred to the target domains (“ LIFE WITH DEPRESSION ”, “ THE DISORDER ”, etc.).

Our corpus also provides a number of metaphors evoking control, such as those evoking an empowered and mindful fight against stigma, or the conscious knowledge of the path in a journey. Semino et al. (2017) have shown how metaphors used in the context of cancer can have two sides: violence metaphors can express and reinforce patients’ negative feelings, but they can also be used in empowering ways when expressing a perception of themselves as engaged and effective agents within the illness experience. This suggests that the default negative emotions often associated with “ WAR ” metaphors (for example by Hendricks, Demjén, Semino, & Boroditsky, 2018) may be overridden or at least palliated by an empowering emotion of pride. As for “ JOURNEY ” metaphors, they are the paradigm for reflecting and reinforcing positive feelings when conveying a sense of purpose, control and companionship, but can also be used in disempowering ways, for instance when the epistemic correspondences transfer notions such as obstacles or aimless direction, as we found in our corpus. Although studies based on Internet narratives do not provide accurate data on the stages of people’s depressive disorders, future research based on other corpora could attempt to evaluate whether these two metaphors, as well as their different uses and functions, correlate with the different stages in people with depression’s life trajectories.

Conclusions

This study has broadened our understanding of metaphorical conceptualizations of depression by its sufferers. Firstly, it has confirmed in a new language and a new textual genre the existence of the full set of metaphors that had been accounted for in previous studies. This reinforces these metaphors’ relevance and their conceptual (i.e. non language-dependent) nature. Secondly, new types of metaphors that people with depression use to conceptualize their experience have been found, most of which correspond to social, communicative and medical factors. These dimensions had not previously been taken into account in metaphor studies on depression. Their detection confirms our hypothesis that analyzing a textual genre that facilitates freer expression by people with depression would reveal new metaphors related to contextual factors of life with depression.

This new typology of metaphors suggests that the containment and constraint that characterizes an important part of the metaphorical discourse of depression is imposed not only by the disorder itself, but also by other contextual factors (stigma, lack of communication, the repressive power of medicine) that can have a significant impact on those diagnosed with depression. However, we have also detected metaphors that evoke control and reinforce blog writers’ capacity for agency, thus suggesting that the blogging genre can have empowering and thus positive effects on the lives they lead.

Acknowledgments

This research was conducted in the framework of MOMENT: Metaphors of severe mental disorders. Discourse analysis of affected people and mental health professionals, a project funded by the Spanish National Research Agency (Agencia Estatal de Investigación, AEI) and the European Regional Development Fund (ERDF), within the Spanish Government’s National Programme for Research Aimed at the Challenges of Society. Ref. FFI2017-86969-R (AEI/ERDF, EU).

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Spanish National Research Agency (Agencia Estatal de Investigación, AEI) and the European Regional Development Fund [FFI2017-86969-R].