Can love conquer all in business media discourse? An assessment of gender-specific metaphorical use

Laurent Nicaise

 Université Libre de Bruxelles (


There has been increasing interest in the use of metaphor in business media discourse, both in the nature of the metaphorical expressions and their connection with ideology. However, there has been growing recognition that the impact of gender on the production of metaphorical expressions of WAR and LOVE, two key markers of group identity in business and economics, has not been fully investigated. To explore that issue in more depth, this paper extends existing perspectives on gender-related use of metaphors by integrating a multivariate analysis that allows for a better assessment of the individual impact of the author ’s sex on metaphor selection and its relation to other factors.

Die Verwendung von Metaphern im Wirtschaftsmediendiskurs hat in den letzten Jahren wachsendes Interesse auf sich gezogen, das sowohl in der Natur der metaphorischen Ausdrücke als auch auf deren Verkettung mit ideologischem Gehalt/Verbindung mit Ideologie liegt. Indes wurde bis dato die Auswirkung des Geschlechts auf die Produktion von Ausdrücken im Metaphernbereich KRIEG und LIEBE/ auf die Produktion von Ausdrücken im Bereich der Kriegs- und Liebesmetaphorik, zwei wesentliche Gruppenidentitätsmerkmale in Handel und Wirtschaft, noch nicht umfassend untersucht. Zur Vertiefung dieses Aspekts soll in dem vorliegenden Beitrag der geschlechterspezifische/geschlechterbezogene Gebrauch von Metaphern eingehend erforscht werden. Anhand einer multivariater Analyse wird versucht, eine bessere Beurteilung der individuellen Auswirkungen des Geschlechts des Autors auf die Wahl von Metaphern und deren Beziehung zu anderen Faktoren vorzulegen.

Download kompletter Artikel (PDF): 



Seite 65


Can love conquer all in business media discourse? An assessment of gender-specific metaphorical use

Laurent Nicaise, Université Libre de Bruxelles (


There has been increasing interest in the use of metaphor in business media discourse, both in the nature of the metaphorical expressions and their connection with ideology. However, there has been growing recognition that the impact of gender on the production of metaphorical expressions of WAR and LOVE, two key markers of group identity in business and economics, has not been fully investigated. To explore that issue in more depth, this paper extends existing perspectives on gender-related use of metaphors by integrating a multivariate analysis that allows for a better assessment of the individual impact of the author’s sex on metaphor selection and its relation to other factors.

Die Verwendung von Metaphern im Wirtschaftsmediendiskurs hat in den letzten Jahren wachsendes Interesse auf sich gezogen, das sowohl in der Natur der metaphorischen Ausdrücke als auch auf deren Verkettung mit ideologischem Gehalt/Verbindung mit Ideologie liegt. Indes wurde bis dato die Auswirkung des Geschlechts auf die Produktion von Ausdrücken im Metaphernbereich KRIEG und LIEBE/ auf die Produktion von Ausdrücken im Bereich der Kriegs- und Liebesmetaphorik, zwei wesentliche Gruppenidentitätsmerkmale in Handel und Wirtschaft, noch nicht umfassend untersucht. Zur Vertiefung dieses Aspekts soll in dem vorliegenden Beitrag der geschlechterspezifische/geschlechterbezogene Gebrauch von Metaphern eingehend erforscht werden. Anhand einer multivariater Analyse wird versucht, eine bessere Beurteilung der individuellen Auswirkungen des Geschlechts des Autors auf die Wahl von Metaphern und deren Beziehung zu anderen Faktoren vorzulegen.


Gender – social identity – corpus linguistics –business media discourse – metaphor

1. Introduction

Counting metaphor frequencies in various corpora has been a rewarding research object in recent years. Yet, for want of reliable data, one of the variables impacting metaphor selection, the author’s sex, has been left almost untouched. There is little doubt that the use of metaphorical expressions varies according to the author’s sex and concomitant gender. One particular metaphorical domain has been identified as a marker of gender identification, namely the WAR metaphor. Previous studies on this particular metaphorical domain show divergent results, ranging from Eubanks' findings that “no salient gender pattern emerged with respect to Trade Is War” (2000: 162) to Koller's observation (2006: 252) that the WAR metaphor is highly gender-specific and ideological since it constructs an aspect of the world as “an act of large-scale aggression enacted mostly by men”.

Such mixed results obviously require further research, all the more since, up to the present day, no attempt has been made to integrate the variable ‘author’s sex’1 into a multivariate quantitative analysis, encompassing as many independent variables as possible, so as to measure the impact of the author’s sex on the selection of metaphorical expressions of WAR and LOVE and the hierarchy between these independent variables. Recently, efforts have been made to avoid monocausal explanations of differences in metaphor use. In one of the most rigorous attempts to relate metaphor use to a range of factors, Semino and Koller (2009) have shown that the differences noted in the metaphorical expressions used by former Italian Prime Minister Silvio Berlusconi and one of his political opponents, Emma Bonino, can be imputed not only to gender but also to political orientation, goals, topics, institutional roles and national audiences. I strongly agree with the authors’ claim that gender-related variation can only be thoroughly interpreted by taking into account the interaction of a range of factors.

Since the vast majority of these factors can be coded as categorical variables and the outcome is often binary (the author either uses or does not use a particular lexical item), I would suggest conducting binary logistic regression to quantify the individual impact of each factor and its relation to others. When the corpus consists of a collection of press articles, I would further propose that the relation between the use of these metaphorical expressions and readership statistics, such as education, profession and income in the target audience, should be included not only in the qualitative but also in the quantitative analysis, as journalists attempt to some degree to adjust their discourse to the socio-economic features of their readers. To cover business media discourse from various angles without privileging a particular approach to the discipline or ideological inclination, all articles published in a full financial year and dealing with market reports, analyses or investment advice were gathered by consulting the digital archives of four newspapers and two weekly magazines in Belgium. The corpus compiled for the initial analysis includes ± 450000 words, distributed among nearly 1000 articles selected randomly from these publications.

The present article is part of a broader research project, now at the postdoctoral stage, investigating the use and functions of metaphor in the financial press. The initial goal of the doctoral project was to develop a multifunctional model to predict and explain the use of such metaphors. Particular attention was paid to identifying underlying ideological premises and their relation to an array of factors, such as language community, gender, seasonal influence and degree of specialisation. The present additional research focuses on the implications of gender-specific metaphor selection as a signal of group identification in a broader socio-economic context.

2. Theoretical and methodological considerations

Metaphor selection may be regarded as a signal of group identification: the choice of metaphorical clusters will be influenced by the social group with which the language user identifies. Through a process of conceptual blending (see Fauconnier and Turner 2003), metaphor not only shapes a view of the world but also transports that view as a social reality. That is why metaphors will often be employed by a social group to reify their own ideological and discursive models.

Observational data in financial reports seem consistent with psychological theories of social identity models (see Tajfel and Turner 1986). Social categorization is a cognitive process in which individuals are classified as group members: individuals who feel a bond with a particular group resemble in many respects other members of the same group, while they share fewer similarities with members of other groups. This classification is probably the result of a universal evolutionary process designed to make our complex world more surveyable (see Cliquet and Thienpont 1999).

In addition to a personal identity, economic decision-makers and journalists have thus a social identity, which is defined by who does and who does not belong to their own group, even if each individual may be a member of more than one social group (e.g. family, religion, language community, association, workplace). Social identity appears to be important in the financial world since the members of the in-group enjoy various benefits related to recruitment, remuneration and promotion. In order to activate the feeling of solidarity with the in-group, the individual may try to adopt the characteristics of the group and adapt their behaviour and language use accordingly. Given that female decision-makers and commentators are greatly outnumbered by men in the financial world, one may assume that men make up the ‘in-group’ and women the ‘out-group’.

In the case of media business discourse, Koller (2006) assumes that in-group membership is shaped to some extent by the WAR metaphor, drawing on masculinised cognitive and discursive resources which serve to establish women as an out-group. Likewise, Hearn (1992) suggests that women feel less comfortable with the conflictual discursive model that the conceptual metaphors of WAR and SPORTS help to normalise.

Hence, it may be argued that ideology makes social groups cohere by exercising control on the actions and discursive practices of their members. However, groups are far less homogeneous today than they once were. In our everyday society, hegemony is an unstable and temporary equilibrium. The relevant question is: to what degree will women adhere to traditionally male values, in so far as these can be read off a text?2

Research data suggest that there may be conflicting strategies involved. One possible approach consists of complying with the existing male discursive practices that have been institutionalised in a particular domain, e.g., in this case, the financial world and business media discourse. An illustration of this strategy can be found in the language of female law-enforcement officers, who try as much as possible to hide their female linguistic and behavioural characteristics (see McElhinny 1995). Whether this communicative approach serves women’s interests is a matter of debate: Crawford (1995) maintains that this strategy often leads to the problematisation of female language and legitimisation of gender inequality.

The opposite strategy is to openly challenge traditional male discursive norms by promoting alternative female discursive patterns, which, as a rule, are more based on cooperation and solidarity than on conflict and hierarchy. This approach appears to be less common. Walsh (2001: 6) believes that this strategy may contribute to an undesirable gender split in the public domain by reinforcing the tenacious stereotype that women are better suited for functions with a lower social status.

Walsh’s data suggest that most women shift consciously between male and female discursive patterns, in particular in domains that are still dominated by male values. Walsh argues that the motives are not political but pragmatic, as women endeavour to accommodate the ambivalent expectations to which they are subjected in the public domain.

This hybrid tactic can be seen in a corpus of political speeches in the European Parliament, where Footit (1999) does not record significant differences between male and female politicians with regard to the frequencies of metaphorical expressions of WAR. However, Footit states that women tend to employ metaphorical expressions related to small-scale conflicts while men allude more often to all-out military campaigns. This could signal that women feel compelled to espouse typically male discursive norms in male-dominated domains, but that this adaptation process is not absolute.

The subtle differences in the use of the WAR metaphor between male and female journalists cannot be described by means of the statistical analysis presented here. A more in-depth qualitative analysis would be required to elicit the finer subtleties. Moreover, it is crucial to realise that female journalists, just like any other social group, do not respond equally to discursive practices in male-dominated domains. Since the present analysis does not screen journalists on a personal level, these distinctions are lacking in my model.

In order to further assess the dynamics of this hybrid tactic, an allegedly feminine metaphorical domain, the LOVE metaphor, was added to the model. The basic assumption is indeed that by assimilating the (sub)language of a social group, outsiders will acquire, to some extent, the cognitive schemata of the new group (Cohn 1987). Koller (2004) assumes that the members of the out-group will seek to secure easily identifiable features of the in-group, such as the conceptual metaphor BUSINESS IS WAR, in the hope of joining the authoritative discourse of the in-group and, by extension, the economic and political privileged class. However, some features of the original social group will remain, such as the LOVE metaphor. Chouliaraki and Fairclough (1999) believe that such hybridity is becoming an irreducible characteristic of modern complex discourse, caused by the fact that negotiation processes of norms and values are more arduous than ever in our postmodern society.

Koller (2005) perceives some heterogeneity in the metaphorical clusters encountered in the financial press. An example would be the expression “she did some serious housekeeping and bolstered the morale of the troops” (p. 205), which combines the diametrically opposed domains of WAR and HOUSEKEEPING within a single sentence. Such metaphorically conflicting clusters are expected to be observed in increasingly dissenting discourse, where hegemony is challenged, in the present case by the access of a growing number of women to the financial world and business media.

Hybridity is intimately connected with the current political and media landscape in Belgium and probably elsewhere in Europe, which is characterized by an increasing tendency towards ideological broadening – newspapers and magazines have more autonomy than before and political parties seek to address a broader group of potential voters.

Koller (2006: 240) notes that business media discourse is made even more complex by a high degree of readership orientation: financial journalists try to reproduce the conceptual models that they perceive in their audience. This feature holds true for media discourse in general but is particularly relevant for business reports, as financial journalists act as intermediaries between companies and readers, who may or may not invest in these companies. The journalistic objectives of financial newspapers, which, in the corpus studied here, provide investment advice to their readers, are driven, at least to some degree, by the goal of assisting their readers in increasing the value of their portfolios or taking other financial decisions. Consequently, the metaphorical expressions found in the corpus are not the sole property of the journalists but are driven by external agents as well (business people and readers).

From a corpus-linguistics point of view, the material selected for the qualitative and quantitative analysis includes 967 articles selected randomly from six Belgian publications focusing partly or exclusively on business and finance: De Standaard (Dutch-speaking and conservative), De Morgen (Dutch-speaking and progressive), Trends-Cash (Dutch-speaking and liberal), La Libre Belgique (French-speaking and conservative), Le Soir (French-speaking and progressive) and L’Investisseur (French-speaking and liberal). In the Belgian political environment, the ‘liberal’ parties tend to favour a free-market policy, while the ‘progressive’ political formations are inclined to adopt a more community-based view. The financial year 2005 was chosen due to its representative nature in historical perspective, i.e. a normal share price evolution without considerable upward or downward motion.

The corpus comprises a total of 448764 words, distributed evenly among the six publications concerned. The material includes the headline and the body of the article. The collected articles were then ordered according to the date of publication, assigned a number and finally submitted to a selection made by a random number generator, so as to guarantee the representativity of the sample.3 Editorials were excluded due to the risk of a bias associated with this highly rhetorical type of discourse. Although it would be quite interesting to look at generic differences between editorials and other types of financial articles, editorials in the financial press usually display more conscious and ostensible methods that appeal to the emotions, including a greater deal of metaphors. This may introduce a bias since this paper is primarily concerned with more unconscious tactics employed by women journalists to conform to dominant discursive expectations.

The procedure incorporates three stages: in the first stage, an onomasiological perspective that searches manually all the possible realizations, whether metaphorical or not, of seven pre-established target concepts (DAILY SHARE PRICE EVOLUTION, SHORT-TERM TREND, ECONOMIC SITUATION, DEVELOPMENT OF COMPANY RESULTS, COMPETITION, RESTRUCTURING and TAKE-OVER BID) was opted for. In the second stage, the metaphorical expressions were identified by means of the Metaphor Identification Procedure developed by Pragglejaz (2007), which is designed to ascertain whether a lexical unit has a more concrete basic meaning that contrasts with the meaning in the given context and whether this contrast can be understood by a comparison between the two meanings. Finally, these metaphorical expressions were classified in a wide range of metaphorical domains (WAR, SPORTS, WEATHER, HEALTH, MOBILITY, MACHINERY, LOVE, ANIMALS, FOOD and ORIENTATION) but the discussion in the present article primarily deals with WAR and LOVE.

The approach is multivariate. As the variable ‘author’s sex’ should not be isolated from other key factors impacting the selection of metaphors, seven other variables were included in the model: the above-mentioned pre-established target concepts, political orientation (liberal, conservative or progressive), type of article (market report, analysis or investment advice), season (winter, spring, summer or autumn), language community (French or Dutch), specialisation (specialised or nonspecialised press) and the six aforementioned publications. These variables were selected on the basis of their intrinsic interest or because they had been identified and discussed in previous research (see e.g. Boers 1999 for the variable ‘season’ and Caers 2008 for ‘political orientation’). For the sake of completeness, some of these factors will be mentioned briefly at the end of the results section, yet the discussion in the present article concentrates on the impact of the author’s sex.

The data were analysed by means of logistical regression analysis, which makes it possible to predict the presence or absence of metaphorical expressions from this set of eight independent variables and to assess the hierarchy between these variables.

I shall now discuss how the WAR and LOVE metaphorical domains are articulated in my corpus.

3. The WAR metaphor

The metaphorical domain of WAR is not uniform: it contains a blending of physical aggression and military strategy. As Koller (2004) emphasizes, metaphorical expressions often contain aspects of both features. Financial journalists often depict the stock market and the broad economy as a war, with belligerent parties, friendly or hostile take-over (bids) and offensive or defensive strategies:

(1) Boeing/Airbus: fin de l'armistice? (La Libre Belgique, 12/04/2005)

(Boeing/Airbus: End of the cease fire?)4

These and numerous other metaphorical expressions of WAR found in my corpus are consistent with the claim that metaphors often have a metonymical basis (see e.g. Barcelona 2000 and Goossens 1990). One of the main functions of the metonymy is to reduce the complex aggregate of financial players (shareholders, analysts, brokers, CEOs, …) to an easily identifiable unit, such as ‘Wall Street’, the renowned location of the US financial markets. The place stands for the diversity of people operating in that place or connected with it through various channels. This unified metonymic referent then serves as a target domain that will be understood with the help of distinctive source domains: Wall Street is fighting back, moving forward, is ailing, is on the ropes, … Furthermore, White (1997) argues that the market is frequently conceptualised through the medium of personification, enabling the journalist to view companies, organisations, national economies, financial institutions and stock markets as human entities involved in mutual fighting.

In many instances, financial journalists comment on the TAKE-OVER PROCESS in terms of WAR, which can only be won by the strongest side:

(2) ABN Amro komt dus als overwinnaar uit een strijd die vijf maanden duurde, en die heel financieel Europa in zijn greep hield. (De Standaard, 16/09/2005)

(ABN Amro has emerged as the winner of a fight that lasted for five months and had financial Europe in its grasp.)

Another target concept continually associated with WAR is COMPETITION:

(3) Mobistar entend conserver ce trésor de guerre pour affronter la concurrence impitoyable tout en procédant à des rachats. (L’Investisseur, 22/08/2005)

(Mobistar intends to keep this war chest to fight the ruthless competition while taking over other companies.)

Our current economic model is highly hierarchical. The power of the market leader is defined by the amount of defeated competitors. There is almost a natural(ised) order until the time has come to challenge this order, for instance after a competitor has made a technological breakthrough or developed a more efficient commercial strategy. The end of this new battle ushers in a new era. Metaphorical expressions of WAR may be used to justify this perpetual fight or, by opposition, to try to maintain the supremacy of the current leader. When a leading company or national economy has to ‘fight a war’, an atmosphere of solidarity may be fostered by the frequent use of metaphorical expressions of WAR. Boers and Demecheleer (1995) note that when people read that their national economy and companies are attacked by external enemies, they are more inclined to tolerate extraordinary measures, such as changes in the tax system, working more for less money and collective redundancies.

Wars exist in all shapes and sizes. Hunt and Menon (1995) distinguish between major wars, significant local wars and small wars. Major wars are widespread conflicts between world powers and their key allies, e.g. the First and Second World War. Vital economic interests are at stake, especially the interests of the major economic powers (the US against China for example) or multinational giants (Boeing against Airbus):

(4) Conflict Airbus-Boeing wordt zwaarste handelsoorlog ooit. (De Morgen, 01/06/2005)

(Conflict Airbus-Boeing is turning into the heaviest trade war ever.)

In military conflicts, decisions are made by world leaders that determine the strategy and have the authority and charisma to impose their views. In the economy, the strategy is shaped by presidents, finance ministers and influential CEOs. From the perspective of this metaphorical model, major wars can only be won if the leaders display exceptional moral strength and their subjects are prepared to make the necessary sacrifices:

(5) Nous devons être continuellement compétitifs et agressifs. Nous devons être comme les Russes défendant Moscou face à Napoléon. Nous devons nous battre comme les guerriers de Sony que nous sommes, a proclamé le nouveau PDG de Sony. (Le Soir, 23/09/2005)

(We must remain competitive and aggressive at all times. We must be like the Russians defending Moscow against Napoleon. We must fight like the Sony warriors that we are, Sony’s new CEO declared.)

Serious regional wars break out when countries or large regions want to protect vital interests, such as the natural resources in Africa. The motives are similar to those of major wars but the impact on the rest of the world is more limited. Large segments of the economy and financial world operate on that level: companies compete with one another in different geographical regions and market segments (for example energy and automobile). Here, key-decisions are taken by top managers at the business unit level:

(6) Les investissements publicitaires piqueraient du nez au premier trimestre. En cause, la hausse du prix de l'énergie, l'euro et la guerre dans la grande distribution. (La Libre Belgique, 08/04/2005)

(Advertising investments have been falling sharply in the first quarter, due to rising energy prices, the euro and the war in the big retail.)

Small-scale wars are local armed conflicts, skirmishes, insurrections, ... In the economy these conflicts correspond to the organic competition between individual companies, departments and even internal competition within large companies. These daily operations are left to managers and do not receive particular attention unless competition exceeds normal limits and alters the profitability or even viability of the company:

(7) La prise en charge de ce format permettra à Global Graphics de mieux affronter la concurrence dans le segment à faible coût du marché d'impression. (L’Investisseur, 18/05/2005)

(Supporting this format will allow Global Graphics to better fight against competition in the low-cost printing sector.)

In addition to traditional metaphorical expressions of war, other types of physical aggression are possible in financial-economic reports:

(8) Ce fut le concordat... et puis la faillite. Michel Capron, chercheur à la Fopes (UCL), parla de « duel à mort au-dessus d'un volcan ». (La Libre Belgique, 15/03/2005)

(First Chapter 11, than bankruptcy. Michel Capron, researcher at Fopes (UCL), speaks of a “duel to the death on top of a volcano”.)

Most metaphorical expressions of WAR are conventional, as are the other metaphorical domains analysed in this study. This could, at least in part, be attributed to the circumstances under which journalists often have to operate: information needs to be processed faster than ever, so journalists have to rely on conventional metaphors (Koller 2006: 250). Yet, some critically-minded linguists, such as Holmgreen (2003), suggest that military metaphors support the rhetorical strategy of free market ideology by encouraging the readers to identify the agents responsible for a crisis outside the system itself, instead of questioning the free market mechanisms.

One may wonder, indeed, what the financial world would be like without the WAR metaphor. This question cannot be answered reliably, even if one may assume that the WAR metaphor barely leaves any space for sustainable economic strategies and cooperation to solve economic problems.

Koller (2004), on the other hand, argues that drawing heavily on the WAR metaphor helps to maintain the social domain of business discourse as a male arena. According to Koller, it is the nature of war, i.e. a highly masculinised form of engagement, which makes it almost paradigmatic in media reports on business and marketing. The intensive use of the WAR metaphor thus supports a discourse community characterised by masculinity, entrenching the war patterns and securing the gendered power relations in business discourse.

4. The LOVE metaphor

While the WAR metaphor emphasises the conflictual features of business relationships, the LOVE metaphor tends to focus on cooperation, mutual trust, exchange of information and commitment. According to Hunt and Menon (1995: 87), the metaphoric transfer involved in the MARRIAGE metaphor suggests that emphasizing these aspects of mutual trust and commitment could increase effectiveness in doing business.

O’Malley and Tynan (1999) agree with Hunt and Menon and have found that by entering a romance-like relationship based on commitment and mutual trust, uncertainty can be reduced and resources can be bundled to create more value and wealth.

The LOVE metaphor is particularly abundant in take-over discourse. In the minds of financial journalists, a take-over process often resembles a mating game:

(9) De paringsdans rond de London Stock Exchange (LSE) sleept nu al aan sinds midden december, toen de Deutsche Börse haar eerste bod uitbracht. (De Standaard, 02/02/2005)

(The London Stock Exchange (LSE) has been doing a courtship dance since mid-December, when Deutsche Börse launched its first bid.)

The seduction attempt can result in a relationship:

(10) Pacman épouse un ‘tamagochi’. Bandai, l’inventeur des fameux ‘tamagochi’ va se marier avec le créateur du jeu ‘Pacman’ Namco. (Le Soir, 03/05/2005)

(Pacman marries a ‘tamagochi’. Bandai, the inventor of the famous ‘tamagochi’ is to get married to Namco, the creator of the game ‘Pacman’.)

Or in a rejection:

(11) De beurs van Frankfurt bood vorige maand ruim 2 miljard euro op LSE, maar Londen wees die avances af. (De Morgen, 10/02/2005)

(The Stockmarket of Frankfurt offered as much as 2 billion euros to buy LSE last month, but London turned down their advances.)

5. Results

Relationship between the author’s sex and the use of WAR and LOVE metaphors:

Table 1



Relationship between the reader’s sex and the use of WAR and LOVE metaphors:


Table 2

(% male readers within source)



Trends-Cash (57.8%)



De Standaard (55.3%)



De Morgen (50.2%)




Hierarchy between the author’s sex and the other variables:

Table 3


The metaphorical domain of WAR makes up 4.7% (p < 0.000) of the total number of metaphorical expressions used to describe the pre-selected target concepts in this research. This makes WAR imagery a well-established metaphorical domain, yet less frequent than, say, HEALTH (19.2%; p < 0.000) and MOBILITY (14.9%; p < 0.000). but more than, say, SPORTS (1.6%; p < 0.000) or WEATHER (1.9%; p < 0.000). Metaphorical expressions of LOVE are certainly less common (1.4%; p < 0.000).

Beyond that, metaphorical expressions of LOVE are three times more frequent in articles written by female journalists (5.4% versus 1.8%; p < 0.000). More surprisingly however, is that female journalists use roughly as many metaphorical expressions of WAR as their male colleagues (6.9% compared to 6.2%; p < 0.000). The different metaphorical expressions of WAR and LOVE can be observed across a limited yet sufficient number of female authors (20 distinct authors) and are not a feature of one journalist’s personal style; nor do the ‘house rules for journalists’ since the variable ‘publication’ itself is not statistically significant.

These latter outcomes are surprisingly beyond my own expectations, based on some works in the literature discussed above and on the observation that, all in all, male journalists use slightly more metaphorical expressions (69.4% versus 65.3%; p < 0.000) to describe the seven pre-established target concepts selected here, which rules out the possibility that higher frequencies of metaphorical expressions of WAR and LOVE may be simply attributed to higher proportions of metaphorically used words in general.

The present data seem to confirm the claim that women make linguistic efforts to become member of the dominant in-group. However, the markers of the original group identity do not entirely disappear, which may explain why the LOVE metaphor remains more common in financial articles written by women.

In this respect, there appears to be a relationship between the frequency of metaphorical expressions of WAR and the proportion of male readers. Concentrations of metaphorical expressions of WAR are higher in publications with a higher percentage of male readers (p < 0.000): De Morgen (5% of WAR imagery for 50.2% of male readers5), De Standaard (5.6% for 55.3%) and TrendsCash (8.8% for 57.8%).

These findings concur with Koller’s observation (2006: 251), based on a sample of articles collected in Business Week and The Economist, that “the group written about is largely convergent with the group written for – readers are obviously meant to recognize themselves in the journals and papers”. Group identification is fostered by echoing the conceptual models that financial journalists perceive in their target audience, and the WAR metaphor plays a key-role in the process, since metaphor formation contributes to the interpretation of experience and guides actions in a variety of domains, here, the financial world.

As expected, the use of LOVE imagery correlates inversely (p < 0.000): De Morgen (1.2% of metaphorical expressions of LOVE), De Standaard (0.8%) and TrendsCash (0.5%).

Unfortunately, correlation between the use of metaphorical expressions of WAR / LOVE and education, profession and income in the target audience (as defined by the CIM) is poor. This may be explained by the observation that the three publications involved in this study are read primarily by the ‘well-educated upper classes’, which can be inferred from the distribution among ‘social groups’ in the CIM statistics. The CIM defines a social group by computing a ratio between the professional occupation of the main contributor to the family income and his or her education level. The population of readers is divided into eight groups, the value of 1 corresponding to the highest professional and education level. Social groups 1 and 2 are overrepresented in the publications concerned, ranging from 57% in De Standaard to 61% in De Morgen. Such minor differences in the publications’ readership demographics make an analysis of the relation between the frequencies of metaphorical expressions of WAR / LOVE and the social group membership of the readers involved irrelevant.

It is intricate to make a clear differentiation of metaphor use as influenced by the author’s sex and gender on the one hand and culturally gendered metaphors on the other. If the variable ‘author’s sex’ is crossed with the variable ‘language community’ – which in the Belgium context defines cultural identity up to a certain degree -, it transpires that Dutch-speaking male journalists use slightly more metaphorical expressions of WAR (6.8%) and fewer of LOVE (0.9%) than their female colleagues (WAR = 6.2%; LOVE = 1.3%). Unfortunately, these results are not significant (p = 0.529 for WAR and p = 0.658 for LOVE). Neither does the qualitative analysis performed on the type of metaphorical expressions being used, provide any indication of culturally gendered use of specific metaphorical expressions of WAR and LOVE.

However, the French-speaking press does display significant frequency differences in this case (p < 0.000): French-speaking female journalists employ more metaphorical expressions of LOVE (15.9%) and WAR (8.6%) than their male counterparts (WAR = 4.6%; LOVE = 4.1%). Further analysis reveals that the higher concentrations of metaphorical expressions of WAR and LOVE in the financial articles written by French-speaking female journalists can be imputed almost exclusively to the subject discussed in the article: with regard to the use of metaphorical expressions of WAR, it is remarkable that female journalists formulate primarily the concept of COMPETITION (100% of all the metaphorical items used to describe this concept) and to a lesser degree TAKE-OVER BID (21.9%) in terms of WAR; as for the metaphorical domain of LOVE, female journalists describe the concept of TAKE-OVER BID chiefly in terms of LOVE (75%), while they do not use a single metaphorical expression to convey the other target concepts (p < 0.000). By contrast, their male colleagues use fewer metaphorical expressions of LOVE (57.6%) and more WAR imagery (25.8%) to describe the concept of TAKE-OVER BID (p < 0.000). Male journalists also employ metaphorical expressions of WAR (7.1%) to convey the concept of RESTRUCTURING, while female journalists do not (p < 0.000). Although culture should not be equated with the language community in the Belgian situation, it seems that cultural considerations may be triggering contrasting metaphoric conceptions according to the author’s sex and corresponding gender.

I shall now discuss the hierarchy between the author’s sex and the other independent variables. When the author’s sex is included in a multivariate model designed to predict the use of metaphorical expressions of WAR, it transpires that the most influential factor is the target concept itself (Wald value of 704.503; p < 0.000)6, which far exceeds the weight of the next highest Wald value of 40.441 (p < 0.000) for the effect of the variable ‘language community’.These two variables unmistakably outweigh the variables ‘political orientation (Wald value of 7.8; p = 0.02) and ‘author’s sex’ (Wald value of 0.628; p = 0.731). In the prediction model for metaphorical expressions of LOVE, the variable ‘target concept’ appears to have the highest impact (Wald value of 182.293; p < 0.000), followed by ‘language community’ (Wald value of 35.651; p < 0.000) and the author’s sex (Wald value of 8.881; p = 0.012). These statistics demonstrate that the variable ‘author’s sex’ and concomitant gender cannot be isolated from other variables in an analysis of metaphor use. Furthermore, the regression model developed here indicates that the effect of the author’s sex, if significant at all, is overridden by the target concept and language community in the case of WAR and LOVE imagery and also by the political orientation in the case of metaphorical expressions of WAR.

For the sake of completeness, some results regarding the direction of the observed effects will be mentioned. It is noteworthy that journalists writing for the liberal-minded press (o.r. 1.69)7 use more WAR imagery than their colleagues in the conservative (o.r. 1.569) and progressive press (reference category). In addition, odds ratios Dutch-speaking/French-speaking press indicate that Dutch-speaking journalists (o.r. 2.655) are more inclined to employ metaphorical expressions of WAR but are particularly reluctant to use LOVE imagery (o.r. 0.196; p < 0.000). Finally, high proportions of WAR imagery are recorded when the concepts TAKE-OVER BID (reference category) and COMPETITION (o.r. 13.054, p < 0.000) are involved, while metaphorical expressions of LOVE are almost exclusively used with the concept TAKE-OVER BID (o.r. 1421.655; p < 0.000).

Although some differences in the sex-related use of metaphorical domains exist (male journalists tend to use more metaphorical expressions of HEALTH and MOBILITY for example), these probabilities exceed the 0.05 significance level in the logistical regression analysis, which indicates that the use of metaphorical expressions of LOVE is significantly more related to the author’s sex than is the case for other metaphorical domains.

6. Discussion

One may wonder, to what extent the individual female journalist is capable of resisting discursive control. It would be presumptuous to claim that this question may be answered only by means of the present corpus-based linguistic study. A key factor here could be the existence of a detailed style guide for journalists. Yet, although all publications provide their journalists with a range of instructions regarding content, terminology and style, some of the financial journalists with whom I spoke did not report receiving such explicit guidelines on the use of specific metaphors.

It is even doubtful whether female journalists writing for ideologically heterogeneous publications are able to reflect on metaphorical choices, considering the time pressure under which articles have to be written. The result is often a “quickly produced newspaper prose” (Eubanks 2000), including a great many conventionalized metaphors, which tend to support, deliberately or not, the existing power structures, and in the case of WAR and LOVE metaphor, to sustain a fairly strict gender dichotomy.

The comparative rarity of metaphorical expressions of LOVE may be partly attributed to the narrow scope of the MARRIAGE metaphor in business reports. The main application of the LOVE metaphor is related to the traditional conception of a monogamous marriage between two partners who remain faithful and committed to each other. Not abiding by the rules of monogamy will threaten marital relationships. It is not pertinent to debate here whether this conventional view fits with human nature. In any case, Hunt and Menon (1995: 87) argue that this concept of a monogamous marriage metaphor is inadequate in economic discourse, especially marketing. Most companies are indeed undoubtedly polygamous: they ‘flirt’ with different partners, build partnerships with multiple distributors, suppliers and even competitors.

The findings presented in this article go some way to indicating that when one is exposed to discourse in a particular field, i.e. women journalists in the financial press, one will assimilate and reproduce the corresponding mental models to a certain degree. Nevertheless, the degree to which any individual is free to process discourse is difficult to assess: personal cognitive freedom can vary substantially depending on a range of factors, such as cultural and political background, income, education, job, social milieu, lifestyle, etc. Joseph (2006: 126) maintains that the message designed to persuade one’s audience is not unconditionally accepted by discourse recipients but is subject to a complex negotiation that enables recipients to put the message into perspective: “ordinary people do not simply accept what those in power tell them, but question it, are sceptical about it, resist it, appropriate it and tweak it in order to suit their own ends”.

7. Conclusion

It has been argued in the paper that the use of metaphorical expressions of WAR and LOVE may be considered as a marker of in-group membership and social identification. Since metaphors guide people’s abstract reasoning and transport a world view as a social reality, metaphorical expressions of WAR and LOVE are likely to be employed by a social group to reify their own ideological and discursive models. In the case of financial-economic media discourse, in-group membership builds predominantly on the WAR metaphor, drawing on the masculinised domain of FIGHTING and grounding this conflictual model even further.

The corpus-based quantitative analysis of about 1000 financial-economic articles in the Belgian press reveals that an author’s use of metaphorical expressions of WAR and LOVE is determined to a significant degree by the target concept and to a lesser extent by gender, language community and economic-political preferences – factors that could be considered to fall under the common denominator of ideology.

Obviously, the method presented in this paper is only an explorative attempt to assess the direct and indirect influence of a range of variables on the selection of metaphorical expressions of WAR and LOVE with the help of a corpus combined with a multivariate statistical tool. Additional parameters, such as genre, syntactical features and word class (see Berber Sardinha 2008 for the calculation of metaphor probabilities based, among other variables, on word class) could be included to refine the analysis. It would be useful to evaluate the connection between the use of metaphorical expressions of WAR / LOVE and some of the characteristics of the target audience (e.g. income class and education), that turned out not to be significant in this research project but, in my view, still needs further research, since the poor correlation may be due to the nature of the specific publications on which the present study is based. To investigate this relation, I would recommend considering publications with more divergent socio-economic profiles.

Finally, although there appear to be some interactions between the metaphorical expressions of WAR and LOVE, no clear patterns emerged concerning the coherence or incongruity between these two metaphorical domains. This may be due to the focus of my analysis, which rests more on the hierarchy between the variables effecting the use of particular metaphorical items than on the conceptual links between these metaphorical domains within a text. It may also be attributed to the nature of my corpus, which differs from marketing discourse, which, as Koller (2008) shows convincingly, has undergone a fundamental shift in terms of social practice and corresponding discursive features. Although contemporary marketing discourse is still dominated by the WAR metaphor, the RELATIONSHIP MARKETING metaphor, particularly the ROMANCE and FAMILY metaphors, constitutes an emerging alternative conceptualisation, apparently triggered by a move from marketers to consumers, who, in today’s highly competitive, internet-driven global market, need to be courted to increase one’s market share. No such transition appears to have taken place in the types of discourse studied in my corpus: although financial journalists occasionally venture into alternative LOVE conceptualisations, in particular in TAKE-OVER discourse, this does not seem to have affected the cognitive underpinnings of financial reports and advice in a meaningful manner, as this domain is still subjet to a prevailing win-lose approach to investing.

If – as proposed by a growing number of cognitive linguists – metaphor selection guides people’s thoughts and defines to a certain degree in-group membership, then the observations made here may be taken as indirect evidence of the connection between metaphorical expressions of WAR / LOVE and the issue of gender-specific power relations in business media discourse.


Barcelona, Antonio (2000): “Types of arguments for metonymic motivation of conceptual metaphor”, in: Barcelona, Antonio (ed.): Metaphor and Metonymy at the Crossroads, Berlin/New York, 31-58.

Berber Sardinha, Tony (2008): “Metaphor probabilities in corpora”, in: Zanotto, Mara Sophia/Cameron, Lynne/Cavalcanti, Marilda C. (eds.): Confronting Metaphor in Use: An applied linguistic approach, Amsterdam/Philadelphia, 127-147.

Boers, Frank (1999): “When a bodily source domain becomes prominent: the joy of counting metaphors in the socio-economic domain”, in: Gibbs, Raymond W./Steen, Gerard (eds.): Metaphor in cognitive linguistics, Amsterdam/Philadelphia, 47-56.

Boers, Frank/Demecheleer, Murielle (1995): “Travellers, patients and warriors in English, Dutch and French economic discourse”, in: Belgisch Tijdschrift voor Filologie en Geschiedenis 73, 673-691.

Caers, Eric (2008): A contrastive study of reporting in The Guardian and The Economist: Metaphors in the transition from Keynesian economics to monetarism, PhD thesis, Antwerp.

Chouliaraki, Lilie/Fairclough, Norman (1999): Discourse in Late Modernity, Edinburgh.

Cliquet, Robert/Thienpont, Kristiaan (eds.) (1999): Ingroup-outgroup Behaviour in Modern Societies. An Evolutionary Perspective, The Hague/Brussels.

Cohn, Carol (1987): “Sex and death in the rational world of defense intellectuals”, in: Signs 12, 687-718.

Crawford, Mary (1995): Talking Difference: on Gender and Language, London.

Eubanks, Philip (2000): A War of Words in the Discourse of Trade: The Rhetorical Constitution of Metaphor, Carbondale, IL.

Fauconnier, Gilles/Turner, Mark (2003): The Way we Think: Conceptual Blending and the Mind’s Hidden Complexities, New York.

Footit, Hilary (1999): Women and the language of politics. Paper presented at the Conference on Women and Politics: debating ways forward, Middlesex University, 18 June.

Goossens, Louis (1990): “Metaphtonymy: The interaction of metaphor and metonymy in expressions for linguistic action”, in: Cognitive Linguistics 1, 323-340.

Hearn, Jeff (1992): Men in the Public Eye, London.

Holmgreen, Lise-Lotte (2003): “Setting the neo-liberal agenda: How metaphors help shape socio-economic "realities"”, in: 5, 90-115.

Hunt, Shelby D./Menon, Anil (1995): “Metaphors and competitive advantage: Evaluating the use of metaphors in theories of competitive strategy”, in: Journal of Business Research 33, 81-90.

Joseph, John E. (2006): Language and Politics, Edinburgh.

Koller, Veronika (2004): Metaphor and Gender in Business Media Discourse: a Critical Cognitive Study, Basingstoke.

Koller, Veronika (2005): “Critical discourse analysis and social cognition: evidence from business media discourse”, in: Discourse and Society 16, 199-224.

Koller, Veronika (2006): “Of critical importance: using corpora to study metaphor in business media discourse”, in: Stefanowitsch, Anatol/Gries, Stephan Th. (eds.): Corpus-Based Approaches to Metaphor and Metonymy, Berlin, 229-257.

Koller, Veronika (2008): “Brothers in arms: Contradictory metaphors in contemporary marketing discourse”, in: Zanotto Mara Sophia/Cameron, Lynne/ Cavalcanti Marilda C. (eds.): Confronting Metaphor in Use: An applied linguistic approach, Amsterdam / Philadelphia, 103-125.

McElhinny, Bonnie (1995): “Challenging hegemonic masculinities: female and male police officers handling domestic violence”, in: Hall, Kira/Bucholtz, Mary (eds.); Gender articulated, London, 217-244.

O’Malley, Lisa/Tynan, Caroline (1999): “The utility of the relationship metaphor in consumer markets: a critical evaluation”, in: Journal of Marketing Management 15, 587-602.

Pragglejaz Group (2007): “A practical and flexible method for identifying metaphorically used words in discourse”, in: Metaphor and Symbol 22, 1-39.

Robinson, Paul (2006): Military Honour and the Conduct of War: From Ancient Greece to Iraq, London.

Semino, Elena/Koller, Veronika (2009): “Metaphor, Politics and Gender: a Case Study from Italy”, in: Ahrens, Kathleen (ed.): Politics, Gender and Conceptual Metaphors, Basingstoke, 36-61.

Tajfel, Henri/Turner, John C. (1986): “The social identity theory of intergroup behaviour”, in: Worchel Stephan/Austin, William G. (eds.): The psychology of intergroup relations, Chicago, 7-24.

Walsh, Clare (2001): Gender and Discourse in Politics, The Church and Organizations. Harlow, England/New York.

White, Michael (1997): “The Use of Metaphor in Reporting Financial Market Transactions”, in: Cuadernos de Filología Inglesa 6, 233-245.

1 A distinction is made between the ‘author’s sex’, used as an independent variable in the multivariate model to refer to the biological features of the journalist, and ‘gender’, by which is meant the socially constructed language use (i.e. masculine or feminine) of the journalist, in conformity with the literature on gender.

2 Clearly, a degree of subjectivity is involved in the notion of ‘traditionally male values’. The notion is used here in the context of the WAR metaphor. Although an increasing number of armies now admit women to serve active duty, the vast majority of the fighting has most often been done by men and although women have been known to enforce standards of conduct during war, military codes of conduct, especially with regard to honour, are characterised by male values. In that respect, Robinson (2006) observes that, in the context of war, such codes are very often defined by men’s desire to prove their ‘manliness’.

3 Available on []. This tool generates random numbers by extracting values from the computer system’s time, a relatively reliable procedure for randomly assigning objects to a sample in a statistical experiment.

4 The English ‘translations’ of the French and Dutch illustrations are intended to help the reader not familiar with French or Dutch. These translations are not part of the original and were prepared by the author. These are not meant to trigger a discussion of equivalence in terms of the equivalence debate in translation studies.

5 The figures for the amount of male readers are provided by the CIM, the Belgian Centre for Information on the Media. See Media Surveys Press and Cinema 06-07.

6 The Wald test is used to determine whether an effect of a specific factor exists or not and to measure the hierarchy between the selected factors. Larger values signal a more pronounced impact. The associated p-value indicates how (un)likely a given observation is. In the present analysis, the data that do not reach the 0.05 significance level should be ignored.

Although most of the press articles in the corpus deal with stocks listed on the Belgian stock market (Euronext Belgium), increasingly globalised financial markets imply that some of the content may have been drafted on the basis of news agency releases in another language, especially English. There is some evidence of this impact on metaphor use, particularly when the issue of global market sentiment is discussed, as exemplified by the paradigmatic imagery of the struggle between bulls and bears. Some of the journalists that I interviewed during my research project acknowledge the influence from external English-speaking sources and occasionally from the other national language. Although it would reveal interesting data on the matter of cross-cultural interaction and authorship, it is not feasible at this stage of this multivariate research analysis to generate reliable statistics on the extent to which some of the content studied here was partially edited, rewritten or reworded in the target language.

7 o.r. (odds ratio) is a measure of association between two binary data values. The odds ratio represents the odds that an outcome (for example developing cancer) will occur given a particular exposure (for example radiation, smoking or consuming large quantities of red meat), compared to the odds of the outcome occurring in the absence of that exposure. In the present study, odds ratios estimate the effect of a variety of factors, such as the author’s sex, on the use of LOVE and WAR metaphors.