Abstract
Historically, impartiality has been imposed as the norm of professional journalism. Yet, be it conceived in terms of non-partisanship or balance, it offers a limited approach to the evaluation of the quality of news. This article revises the traditional approach to bias: As neutrality is impossible and truth does not lie in the middle, accuracy is better served by fairness than by a delusive position of impartiality. An alternative model promoting fairness is thus proposed, which is based on the criteria of consistency and justification of position-taking. Based on the work of socio-linguist Labov, this model is not without methodological challenges. We apply the model to a newspaper article for illustrative purposes and as a starting point for discussion, showing that, unlike impartiality, fairness is altogether an attainable and desirable standard.
‘I shall never admit for any affection towards countree or Kyn, to be so partial, as wil wittingly either bolster the falsehood or bery the truthe …’ (quoted in Stephens, 1988: 256–257). This commitment to refrain from partiality can be found in a 1548 English news book. Although historians of journalism have shown that impartiality has been imposed as the norm of the journalistic profession in a limited space and time frame, roughly corresponding to the Anglo-American cultural realm of the turn of the 20th century (Kaplan, 2002; Schudson, 1978), this quote suggests that, before and beyond its imposition as a professional standard, impartiality has been advocated as a value because partiality is equated with falsehood. Impartiality is here conceived in terms of the journalist’s avoidance of leaning toward political forces. Historically, avoidance of leaning has been translated into non-partisanship, meant as the scientific depiction of the world as it is and, when this appeared to be an impossible ambition, as neutrality, that is, the adoption of neutral or euphemistic language intended to prevent stigmatization of the news stories’ actors (Schudson, 1990). Impartiality later evolved into balance meant as symmetry, that is, an equilibrium in the characterization of the actors and the citation of their supporters’ voices, achieved through a process of compensation, notably in the report of misdeeds committed by one side by misdeeds committed by the other side.
Be it defined as non-partisanship or as balance, the notion of impartiality seems to be enacted by journalists and measured by scholars in a binary mode, wherein partiality is equated with being (too) ‘pro’ one side and ‘against’ the other. This article challenges such an approach not only for being reductive but also for introducing what it was supposed to combat, namely, bias. I develop a critique of impartiality in three steps. First, I reject the binary conception of impartiality and define instead five situations of partiality corresponding to five distinct sorts of bias. Each situation is illustrated by concrete cases drawn on scholars’ research, especially on the coverage of violent conflicts, for which the polarization of the debate and the temptation of taking side are particularly significant.
In the second phase of my critique, I extend the critical analysis to a more fundamental rejection of impartiality on its philosophical foundations. My central argument is that impartiality is not necessarily the path to truth, while truth has historically been defined as the ultimate goal of journalism (Klaidman and Beauchamp, 1987).
Finally, refutation of the standard of impartiality invites reflection on alternatives. Impartiality implies that journalists must refrain from making judgments, which is altogether impossible and undesirable as this would result in unfair reporting. As values are necessarily expressed in news stories, fairness, rather than impartiality, is what journalists should consider the best attitude in their quest for truth. This article thus develops a framework for a proposed solution to evaluate fairness in news reports, notably based on the works by socio-linguist William Labov. It provides one application for illustrative purposes, drawn on prior research on the French press coverage of the second Intifada.
Two facets of impartiality and five situations of bias
Typologies of bias and their limits
Bias: an undertheorized object
Entman (2007) characterizes bias as ‘that curiously undertheorized staple of public discourse about the media’ (p. 163): Although bias raises the attention of citizens and activists, media scholars have failed to seriously conceptualize and measure it. Certainly, scholars have attempted to assess definitional criteria of media bias. Williams (1975) considers that bias must be volitional, influential, threatening to conventions, and sustained (cited in D’Alessio and Allen, 2000: 133). Yet, these criteria are not easily open to an evaluation: for instance, how to measure the ‘volitional’ character of bias?
Rather than a constructed definition, other scholars have proposed operational typologies. Entman (2007) thus considers three types of bias: ‘distortion bias’, referring to ‘news that purportedly distorts or falsifies reality’; ‘content bias’, constituted by ‘news that favor one side rather than providing equivalent treatment to both sides in a political conflict’, and ‘decision-making bias’ which considers ‘the motivations and mindsets of journalists who allegedly produce the biased content’ (p. 163). Entman proposes to abandon the first option because of the ‘irresolvable questions about truth and reality’ (p. 166) and to focus on ‘content bias’ and ‘decision-making bias’, supposed to offer more operational dimensions of bias.
Content and selectivity bias
In the past decades, decision-making bias has raised particular attention among scholars, who contend that external processes of selectivity slant the news even more than content bias, which would be a more obvious internal form of bias. In their meta-analysis of 59 quantitative studies on partisan media bias, D’Alessio and Allen (2000) identify three main types of bias, one of which is a selectivity bias, close to Entman’s ‘decision-making bias’: ‘gatekeeping bias’, which refers to the external world of the journalist’s decision-making rather than to the outcome and the internal measure of bias in extant texts (p. 133). Therefore, although theorized later, in the last decades of the 20th century, selectivity biases, such as framing, agenda-setting, and priming, have led to much effort at building models of assessment (Entman, 1993, 2010; Entman et al., 2009; McCombs and Ghanem, 2001). Yet, as Matthes and Kohring (2008) acknowledge, the assessment of frames still raises theoretical and methodological problems. Fundamentally, these problems originate in the fact that ‘decision-making bias’ refers to the journalists’ intentions – what they have in mind when selecting news stories and framing them in one way rather than the other – which are nearly impossible to ascertain. 1 Paradoxically, while content bias was the first discussed in the literature, it now inspires fewer endeavors, and its conceptualization has hardly been renewed since Lippmann.
For these reasons, we propose to approach bias in its admittedly limited internal dimension, what D’Alessio and Allen call ‘statement bias’ and ‘coverage bias’. ‘Statement bias’, as opposed to non-partisanship, occurs when the journalist expresses himself or herself in a way that would favor one side. ‘Coverage bias’ – the opposite of balance – is when the journalist does not give the same weight to both sides’ points of view. As we further develop, non-partisanship and balance are two aspects of the standard of impartiality, which had been imposed as the norm of professional journalism, at least in 20th century America. Actually, non-partisanship and balance do not only constitute two dimensions of impartiality but also represent two phases in its historical development.
History of objectivity: From non-partisanship to balance
Impartiality imposed itself as the norm of the journalistic profession at the turn of the 20th century. It was then conceived in terms of the journalist’s avoidance of leaning toward political forces. More concretely, it was defined as non-partisanship in the characterization of the actors and their actions in news reports (Schudson, 1978, 1990). Journalists are expected to focus on facts and refrain from expressing judgment for or against one side in a political dispute. But as this conception grew on the journalistic profession, it also evolved, especially after World War I propaganda demonstrated that facts were not given in advance but rather constructed and influenced by values and stereotypes. Faced with the limits of the ‘naïve empiricism’ (Schudson, 1978) characterizing non-partisanship, Lippmann proposed in the late 1920s to extend impartiality from the external actors to the journalists themselves, who were expected to help the audience interpret the events without being themselves the producers of ideas (Schudson, 1990). Such scientific detachment (Lippmann, 1931) was to be realized through the technique consisting of getting both (all) sides of a story. In this view, if reports cannot be free of interpretations, journalists must at least give an equal voice to those competing for the interpretation of the events and must refrain from expressing their own position. This view is epitomized by Cronkite’s (Interview with Jimmy Hoffa on CBS News, April 1964) famous saying that ‘in seeking truth, you have to get both sides of a story’.
Thus, while on a theoretical level, non-partisanship and balance are two possible ways of defining impartiality, on a practical level, the norm of impartiality evolved from non-partisanship (no statement in favor or against one side) to balance (equal weight to competing actors and their supporters’ voices). We speak of an evolution in the sense that the conception of non-partisanship as focus on facts is now largely rejected as naïve and impracticable. However, remnants of non-partisanship can be found in contemporary journalism advocating neutral or euphemistic wording. Interviews with American and French journalists about the norms guiding their practices (Boudana, 2010a) indeed show that a number of practitioners still believe that, in a world full of uncertainties, journalists should refrain from judging actors and their actions and must therefore prefer neutral words, such as ‘activists’ rather than ‘terrorists’.
In any case, be it defined as non-partisanship (neutral characterization) or as balance (equal evaluation, expressed directly or through citations), the concept of impartiality is flawed. Paradoxically, whereas impartiality was meant to combat bias, its two concrete expressions may introduce biases in news stories. Our ambition here is to challenge the norm of impartiality and to propose alternative measures of bias. Our critique starts with showing that favoritism and unbalance offer limited views on bias.
Denunciation of partisanship as favoritism is too reductive
When defined in terms of non-partisanship, impartiality means the refusal to favor one side. In this conception, the journalist’s expression of sympathy for one party and of antipathy for the other would be like two sides of the same coin. Being too favorable or too unfavorable to both sides are not perceived as possible forms of bias. As an illustration, we can quote the following extract from a newspaper article (Le Priol, 2000) reflecting on the controversy arisen by the French media coverage of the second Intifada (2000–2005):
Letters received by France 2 accuse its news reports of ‘manipulating the opinion’, in a way that is too favorable to the Palestinians or (more rarely) to the Israelis. That is why the TV channel mediator, Jean-Claude Allanic, is not far from thinking that ‘no matter what we do or say’, in such conflict, TV will always be suspected of partiality.
Here, the journalist opposes those arguing that the French press is ‘too favorable to the Palestinians’ to those considering that it is ‘too favorable’ to the Israelis. He never mentions other forms of bias. However, one of the main findings of my research on this case (Boudana, 2010b) is that the French press has an anti-Israeli position but is not, for all that, pro-Palestinian. Instead, it gives a rather bad image of the Palestinians: The representation of Israelis as victimizers and colonizers goes with processes of infantilization and bestialization of the Palestinians who are deprived of voice and rationality (Boudana, 2010b). The French press coverage of the second Intifada thus presents a case of negative partisanship (bias against the Israeli side but not favorable to the Palestinian side) that complexifies the picture of partiality.
Balance can introduce bias when artificially imposed in the narrative
The second important argument against the traditional view on impartiality concerns cases of balance. Here, impartiality does not mean cleaning the text from any judgment, but rather equally judging both sides, either by compensating misdeeds committed by one side by misdeeds committed by the other, or by giving equal weight to supporters of antagonistic sides. Impartiality is then taken as synonymous of balance between both sides. However, as reality is not necessarily balanced, this process of compensation can be artificial and unfair.
A pertinent critique of this compensation process is found in Jean-Paul Gouteux’s study of Le Monde’s coverage of the 1994 genocide in Rwanda. According to Gouteux (1999), scholars have long established the French government’s intention of supporting Habyarimana’s dictatorial regime and its negation of the reality of the Tutsi genocide. But the support of the French newspaper of record, Le Monde, to this negation is less known. Gouteux argues that Le Monde systematically padded out the respective roles of each protagonist, presenting the genocide as interethnic massacres, implicating the regime and also the rebels. He points out the domination of an ethnic reading of the conflict, which eludes the political responsibilities and reduces the conflict to a centuries-long tribal war. Specialists of the issue have nonetheless repeatedly refuted such vision of Rwanda, recalling that Hutus and Tutsis share the same language, culture, and territory. Gouteux proposes two explanations for this media bias: One is the media support to the politics of President Mitterrand and the other is what he calls ‘intellectual firmly fixed bad reflexes’ when dealing with African issues. In this case, balance, in terms of a depiction of equal violence emanating from both sides, represents a distortion of the reality of the conflict, in which one group was victim of a genocide perpetrated by the other.
Five situations of bias
In the light of our prior discussion, it appears that the binary categorization of partiality as ‘favorable to A at the expense of B’ is too reductive in its approach to bias, while balance (‘equally favorable or equally unfavorable to A and B’) cannot be considered a good indicator of unbiased news reporting.
The question of impartiality is more complex. Based on the definitions of partisanship and balance that we discussed, five situations of bias can be considered: Partisanship can be positive or negative while balance can be positive, negative, or partial (in this latter case, balance is not between two antagonist sides but between the positive and negative aspects of only one of the protagonists).
1. Positive partisanship. The journalist is pro-A and ignores B (or is anti-B only because – and to the extent that – it valorizes A).
This is the classic approach to bias, as formulated in the quote from Le Priol’s article.
2. Negative partisanship. The journalist is anti-A and ignores B (or is pro-B only because – and to the extent that – it de-valorizes A).
This is the case of the French press coverage of the second Intifada (Boudana, 2010b).
3. Positive balance. The journalist artificially balances the positive actions and arguments characterizing A and B.
This is the case when journalists uncritically give voice to both sides although one side has more solid arguments.
The American press coverage of the global warming issue is a good illustration of such artificial balance. Boykoff and Boykoff’s (2004) study demonstrates that, while there is a large consensus among scientists that human activity has a significant impact on global warming, the majority of the articles published in American mainstream newspapers give equal weight to the arguments of scientists claiming that global warming is caused by nature. Here, we see that mechanisms of artificial compensation are based on the bogus argument that truth lies in the middle.
4. Negative balance. The journalist artificially balances his or her denunciation of misdeeds and deceptive arguments characterizing A and B, and their supporters.
This is the criticism expressed by Gouteux in his analysis of the French press coverage of the genocide in Rwanda and one of the major problems identified by scholars advocating Peace Journalism (Galtung, 1998; Lynch, 2006; Maslog and Lee, 2005). By depicting only the black on both sides, journalists, it is argued, are not only unfair to both sides, but they favor hatred and violence rather than conflict resolution.
5. Partial balance. The journalist presents the positive and negative elements characterizing A but ignores B (or let B’s actions and arguments be framed by A).
This criticism has notably been addressed by scholars analyzing the place of official versus non-official voices in news stories and particularly those considering that the State represents the primary definer of news (Bennett, 1996; Hall et al., 1978; Sigal, 1973). This dimension of bias has become a central issue with the development of framing analysis in media studies (Entman, 1993; Entman et al., 2009; Gitlin, 1980). Here, the internal and external dimensions of bias are meeting (see our discussion of the typologies of bias).
While the discussion on these five situations of bias shows that the traditional approach to impartiality is actually partial, the concepts of non-partisanship and balance can be challenged more fundamentally on philosophical grounds.
Theoretical foundations of the rejection of impartiality as a journalistic norm
Truth does not lie in the middle
Media scholars commonly equate balance with objectivity and fairness. This is the case, for example, in Drew’s (1975) study:
Objectivity for this study is defined as fairness or balance in decision making, information seeking and presentation of information. The objective reporter should present a quantitatively balanced story containing both negative and positive information about the source. (p. 219)
As our examples suggested, equating balance with objectivity and fairness is challengeable. In a prior work (Boudana, 2011), I argued that objectivity has a claim to truth that is not always compatible with a position of neutrality and detachment, on which non-partisanship and balance find their basis. I am here extending the argument to the notion of fairness and propose that, although the criterion of balance is promoted in the name of fairness, its application can actually lead to unfair reporting.
Drew imposes as self-evident that a ‘quantitatively balanced story’, that is, the presentation of an equal number of positive and negative pieces of information, is necessarily fair. But what if reality is made of more negative elements? The selection of an equal number of positive elements would then be artificial. Drew’s assumption of an equation between fairness and balance rests on the postulate that truth always lies in the middle, that is, between negative and positive dimensions of an action and between conflicting voices. Media ethics analysts have criticized the philosophical grounds of such an assumption. In particular, Cunningham (1999) purports that it is based on the ‘deterioration’ of Aristotle’s ‘Golden Mean’ theory, according to which ‘goodness’ would be a ‘sort of middle ground or juste milieu between two extremes of excess or defect’ (p. 5).
Hackett (1984) goes a step further in this criticism of balance: ‘far from being in some absolute sense neutral, news balance generally leads the media to reproduce the definition of social reality which has achieved dominance in the electoral political arena’ (p. 234).
Impartiality is here denounced as a delusive ideology which, under the appearance of centrism, actually serves the elite’s interests and socio-political status quo.
Impartiality/objectivity 2 denounced as a strategy
By leading audiences to the belief that the media are politically untainted and by giving in this way credibility to journalists, what Friedman (1998) calls the ‘fantasy’ of objective journalism serves to ‘naturalize’ the ideology of the dominant group. ‘Balance’ between competing politicians and the presentation of ‘both sides of the story’ may actually serve to divert the audience’s attention from the questions of why these individuals (usually officials) are granted the right to define the issue, why the issue is being cast in these terms, or why it is an ‘issue’ at all. Here again, we see how content bias meets the selectivity bias, while the latter has become a privileged research object for media scholars who often neglected content bias as too obvious.
The major criticism of (fake) balance has been expressed under diverse forms since the 1950s (Boudana, 2011) and extended from the accusation of political motives to the denunciation of journalistic occupational routines. Several scholars have argued that, far from representing an ideal, impartiality is a mere strategy – what Tuchman (1972) calls a ‘strategic ritual’ – aimed at protecting the journalists from the risks to which they are exposed in their daily practice (Altschull, 1995; Glasser, 1984: 15; McNair, 1996: 33). By showing impartiality and by referring to credible sources, journalists can produce unobjectionable news and can meet deadlines, thus coping with what are essentially economic constraints: expenses entailed by delays and costs of lawsuits in cases of offensive journalistic stories.
In other words, far from being an ideal, impartiality would be a manipulative ideology and an occupational strategy. Facing this radical criticism, can we still make sense of the notion of bias and think of ways of measuring it?
Toward a practical approach to fairness: The criteria of consistency and justification
I will not enter into an endless discussion of the philosophical stakes of the concept of fairness and the related notion of ‘judgment’. I will be content with proposing to define fairness as the ability to make judgments free from discrimination. What is of importance and interest is the idea that fairness does not mean ‘neutrality’, but the universal application of moral values, as opposed to the application of two sets of rules. To support this conception, I recall Raymond Aron’s (1938) phrase: ‘Objectivity does not mean impartiality but universality’. Fairness thus implies critical distance, which is not synonymous of detachment in the sense of withdrawal or avoidance of any emotional involvement. My approach is situated in Kantian ethics and, more specifically, refers to Kant’s (1998 [1785]) concept of universalizability, as defined in his Groundwork for the Metaphysics of Morals (p. 53). Based on this concept, an action is morally acceptable if it can be generalized (that is, applicable to each and all) and is free of contradiction (that would result from a ‘Do as I say, not as I do’ position).
Therefore, as the following presentation of the methodology emphasizes, the evaluation of fairness is not based on the journalists’ position of neutrality but on the consistency and justification of their position-taking, in their characterization of the actors and the actions constitutive of the reported events, and in their treatment of the voices that are given a platform for expression. Consistency does not mean to even-handedly praise or blame all actors – that would be (artificial) balance – but, rather, to apply the same criteria of moral or ethical evaluation to all actors. For instance, if the killing of civilians in an armed conflict is reprehensible, journalists should clearly identify those responsible for such killings. They should not try to find misdeeds on one side to compensate for the mass-massacre of civilians committed by the other side, as some French journalists did in the case of the Rwanda genocide. If judgments cannot be avoided, at least they are now consistent – that is, subject to universalizable principles – and justified – by admissible legal, ethical, or moral criteria.
Methodology
Fair characterization of the actors and their actions
Definition in terms of assignment of responsibility
My approach draws on the socio-linguist William Labov’s (1997) analysis of narratives (pp. 395–405). Labov purports that ‘[I]n accounts of conflict between human actors, or the struggle of human actors against natural forces, the narrator and the audience inevitably assign praise and blame to the actors for the actions involved’ (p. 401). Narrative may thus be ‘polarizing’, when some protagonists are blamed and others are supported, or ‘integrating’, when blame is passed over.
Labov’s assertion that ‘the narrator and the audience inevitably assign praise and blame to the actors for the actions involved’ finds supports in many literary and media scholar works. Analyzing the plot components of 100 Russian folk tales, Propp (1984) identified their irreducible narrative elements, arriving at a typology of 31 functions and only 7 broad character types, including the hero, the villain, the helper, and the princess. These types are easily transposable in journalistic narratives. Propp’s analysis of folk tales is in line with the view developed by constructionist frame researchers, who regard journalists as storytellers whose treatment of the news has many of the qualities of myth. Hartley (1982) argues that journalists fulfill a ‘bardic function’ and that ‘media in modern society is not unlike that of the maker and teller of tales in oral cultures’ (p. 104).
These researchers focus their work on the identification of formulae recurrent in news stories. For instance, Barkin (1984) shows that news stories steadily contain elements of tension and conflict between villains and heroes and that these elements solicit responses from the audience: The audience supports, sympathizes with, or turns against the characters of the story. For the particular coverage of armed conflicts, one can refer to the warrior archetype as described by Campbell (1972): This archetype defines a group that has to be protected from the enemy-as-monster so that order of human life is preserved. The identification of an enemy-as-monster enables one to define the problem as threat to a community and the warrior archetype is embodied by the politicians and/or the journalists themselves who suggest a solution to the problem. This solution consists of the eradication of the enemy for no compromise can be achieved with the monster.
If assignment of responsibility is inherent in (news) story, then the key question for the evaluation of fairness becomes, Is the assignment of praise and blame justified?
Method of evaluation
I use Labov’s (2001) ‘participant analysis tables’ that connect each participant in the story to the actions that the narrator attributes to him or her (via the verbs in the text). The participants in the story are listed and associated with the nouns and adjectives characterizing them and with the verbs characterizing their actions. The journalist’s assignment of praise and blame lies in the nature of the deeds attributed to the actors and the connotations of the words used to refer to these deeds.
Here, the question of labeling poses one of the thorniest problems in the evaluation of fairness. There is no consensus on the question of whether journalists should use neutral terms to characterize the actors and their actions or can be judgmental because of the accuracy and fairness lying in the fact of calling a spade a spade. Interviews with journalists demonstrate that not only do they substantially disagree with one another on this issue, but they even show indecisiveness in many cases (Boudana, 2010a). I would argue that, practically, there is no such thing as a neutral label. If by neutral one means euphemistic, then the label still involves judgment in the sense that it tends to minimize the moral condemnation of the actor so labeled.
I thus propose a discussion on each case according to the criteria of justification and consistency: Does the journalist justify her characterization of the actor by the actions that she attributes to this actor? Does the journalist apply the same policy to all characters or does she have double standards depending on the identity of the (groups of) actors?
Fair treatment of the voices
Definition of the voices
I call ‘voices’ those – be they direct participants in a story or external observers – expressing themselves in, and participating in the construction of, the journalistic account. The voices animate the story by expressing their feelings or ideological positions, in reaction to a given event reported by the journalist (Boudana, 2009).
The challenge, for journalists, consists of selecting voices that have a claim to impose their definition of the event – in particular, the actors themselves – and of having these voices compete without distorting their arguments.
Method of evaluation
The method consists of listing the voices mentioned in the article and of proceeding as in the evaluation of the actors. As a narrator, the author of the article should be regarded as a voice that might give her own opinion but also judge the voices she selected. The journalist has to fairly report each voice’s words and fairly puts it in competition with other voices, before she can possibly take a justified position toward this voice. Here, I evaluate the justification of the journalist’s evaluation of the voices and of their arguments: What is the apparent criterion for the selection of the voices? Given this element, did the journalist select only weak voices? Did the journalist distort the arguments (argumentum ad absurdum for instance) of a voice? Then, one can judge whether the journalist’s support to, or attack on, the voice is justified.
To make sense, this criterion of justification is to be applied universally, that is, independently from the journalistic genre (e.g. editorial page vs factual report) or the type of assertion (i.e. statement of facts vs statement of opinion, for the journalist’s opinions too must be submitted to the demand for justification).
Illustration
Insofar as the methodological tools proposed are supposed to work as a practical alternative to non-partisanship and balance, I should conclude on an application to a concrete case. This application is meant to show how the evaluation of fairness is effective through the criteria of consistency and justification applied to the ways in which journalists characterize participants in their news stories and the voices that they mobilize to give more substance to their narratives. Furthermore, the major goal is to demonstrate that, despite widespread criticism, objectivity – if defined in terms of accuracy and fairness – is still altogether desirable and possible for professional journalism: Its evaluation can be based on criteria that are applicable and open to external judgment, as we now let the readers judge of the following evaluation of a news item.
This item is a newspaper article selected from a sample of 167 articles used in my research on the French press coverage of the second Intifada. As I only have space here for the development of one case, I offer it for illustrative, rather than demonstrative, purposes. I will, however, refer to my evaluation of other articles for the sake of comparison, as fairness is evaluated in terms of degrees or levels of justification and consistency. I selected one of the fairest articles in the sample although it is unrepresentative of the corpus studied, because my point is to show how the fairness previously discussed on theoretical grounds can find expression, materialization, in an actual published paper, which was submitted to the regular constraints that scholars invoke as obstacles to fair reporting.
Article covering an episode of the second Intifada
The article relates a crucial episode of the second Intifada, namely, the assassination of a leader of the Hamas by the Israeli army. In the raid, 15 Palestinian civilians were killed (Mann, 2002):
[My translation from French] Sud Ouest Wednesday 24 July 2002, p. 4 ‘Day of rage’ in Gaza Ouri Mann in Jerusalem Middle-East. Fifteen Palestinians, including nine children, were killed on Monday in an Israeli raid that targeted the chief of the Hamas. Palestine yells out in anger.
For many years, Salah Shehade, 50 years old, had been on top of the list of the Palestinians wanted by Israel. Since 1989, he was the chief of Ezzedin al-Qassam, the Hamas armed wing responsible for hundreds of lethal terrorist attacks [attentats]. He died on Monday evening, when a 1-ton bomb, dropped from an F-16, pulverized a building of several floors in a populous neighborhood of Gaza.
Tsahal was not particular about details [n’a pas fait dans la dentelle]. The spouse, daughter, and body guard of the fundamentalist chief perished, as well as 11 Palestinians, among them 9 children, while 150 persons were injured, including 15 seriously. For Prime Minister Ariel Sharon, who deplored the death of innocent people, ‘this is one of the most successful operations […] There is no compromise to be made with terrorism: it has to be fought’. He and the Minister of Defense, Binyamin Ben Eliezer, authorize the operation without prior meeting of the Security Cabinet. The Israeli military staff claim that they were unaware of the presence of civilians in the bombed building but, despite the bloody toll, it welcomed ‘the liquidation of a very dangerous enemy’.
‘A war crime’. Shehade is an emblematic figure. Since 1984, he is one of the close relations of Sheikh Ahmed Yassin, the spiritual chief and founder of Hamas. Gifted with a phenomenal memory, he knows the Koran by heart. As such, he is authorized to deliver fatwas, that is, religious rulings that have authority over the faithful, who generally received their education from Islamic schools. He also has great organizational skills. The two recent military offensives led by Israel in the West Bank did not prevent him from re-implanting cells in the Northern area, or from supplying them with weapons and powerful explosive devises, made of chemical products used in fertilizers.
Israel thus pleads the right to self-defense. But its raid was severely condemned on the international scene. The European Union judges it ‘totally inacceptable’ while United Nations (UN) Secretary-General Kofi Annan calls for ‘respect of the humanitarian law standards’. The Palestinian Authority, for its part, speaks of ‘State terrorism’ and of a ‘war crime’ that must be reviewed by the International Criminal Court and UN Security Council. The Palestinians also bring to the forefront that Tsahal used an F-16 of American manufacturing and that this raid compromises the efforts made for a return to calm after 22 months of Intifada.
‘An agreement was in sight between the Palestinian Authority and Sheik Yasin so that the Hamas stops its attacks on the Israeli territory’, affirms Zyad Abu Zyad, an elected member of the Palestinian Legislative Council. According to other information, from his prison in Israel, the chief of the Fatah in the West Bank, Marwan Barghouti, was about to order the ‘Hudna’ (truce) to the al-Aqsa Brigades.
Riposte. Now the anger is brewing up. The Palestinians broke dialogue with the chief of diplomacy Shimon Peres concerning ‘gestures’ on the part of Israel for the alleviation of the difficult economic situation in the occupied territories. They also proclaimed a ‘day of rage’ and ordered general strike. Angry, 50,000 Palestinians flooded into Gaza streets yesterday for the funerals of the 15 victims. All voices as one, the secular extremists from the Democratic Front for the Liberation of Palestine (DFLP) and the fundamentalists from the Hamas and the Islamic Jihad cry for vengeance.
Two armed Palestinians were shot yesterday morning after they infiltrated the kibbutz (collectivist organization) of Kisufim. ‘This is our first riposte’, the Jihad indicated. Three combatants from Ezzedin al-Qassam were also killed near Nablus. The Israeli security forces are in state of alert. Before the raid in Gaza on Monday, they had thwarted the plan of the Hamas armed wing for the perpetration of several simultaneous suicide attacks in Israel, coupled with rocket firing of Qassam 2. But, yesterday evening, an artisanal rocket shot from Northern Gaza Strip exploded, without casualty, in a Southern Israeli town:
Illustration: Gaza During the funerals of the victims, among whom a two month-old baby Photo AFP
Fair characterization of the actors and their actions
Table 1 synthesizes the journalist’s characterization of the actors and their actions.
Participants’ table.
The treatment is fair. Unlike many articles composing the sample, this article does not lapse into the confinement of the protagonists in pre-determined psychological characterization: Sharon as the insane butcher worth of Nero or the Palestinians as infantilized victims. The portrayal of Shehade is rich and complex: The journalist at once expresses his admiration for Shehade’s skills and leadership and supports the idea that he indeed represented a threat for Israel.
Rejection of Manichaeism does not prevent the expression of moral judgment: indignation at the Israeli officials (Sharon, Ben Eliezer, and the army), responsible for the killing of Palestinian civilians, and compassion for these victims. This compassion is expressed through the importance granted to the toll of victims, which appears as the very first information in the introductory paragraph and is recalled in the body of the text. It is also conveyed by the insistence on the presence of children among the victims – a fact that is emphasized in the caption of the photograph, which refers to the funerals of a 2-month-old baby.
If the expression of Palestinian ‘anger’ as the dominant feeling might recall other articles that present Palestinians as savages who only communicate through verbal and physical violence, here this feeling appears to be justified as ‘rational’ given the toll of victims that the journalist recalls.
One can only regret the focus on the religious dimension in the characterization of Palestinian groups involved in terrorist attacks: ‘secular’ versus ‘fundamentalist’. If this division has political and sociological relevance, it overlooks the nationalist dimension.
All in all, the journalist’s implicit judgment of the actors is fair as it is supported by the reported context of action.
Fair treatment of the voices
Table 2 synthesizes the way the journalist characterizes the voices and introduces them.
Voices’ table.
DFLP: Democratic Front for the Liberation of Palestine; ICC: International Criminal Court; UN: United Nations.
The treatment is fair, although information on two voices is lacking. First, when the journalist reports that Marwan Barghouti ‘was about to order the “Hudna”’, he does not identify his source of information. He contents himself with a vague reference: ‘According to other information’. Second, when the journalist writes that ‘before the raid’ the Israeli security services have ‘thwarted’ a plan of large-scale terrorist attack, he does not explicitly refer to the voice behind this alleged information.
This article reports Sharon’s reaction, without distorting it: ‘For Prime Minister Ariel Sharon, who deplored the death of innocent people, “this is one of the most successful operations”’. Rather than directly quoting Sharon, the other articles composing the sample tended to present the toll of Palestinian casualties and then to refer to Sharon as openly rejoicing. Here, we know that Sharon expressed regret for the death of civilians and considered the operation a success only because the targeted man was indeed killed. And this is significant although some might cast doubt on Sharon’s sincerity: The journalist does not try to guess what the actor’s real intentions are, which is fair insofar as intentions can never be proven (see Note 1). And again, this does not absolve Sharon from blame since the journalist insists on the consequences of Sharon’s decision and on the reprobation it triggered.
While most of the other articles studied focused on the international voices of condemnation (to the point of framing the whole event in terms of the international reprobation of Israel), this article grants a larger platform to the actors of the conflict. The Israelis are given the opportunity to defend themselves before they are condemned so that they are indeed fairly judged. Above all, the journalist renders well the fact that the Palestinians expressed their reactions at different levels (officials and people) and by different means (public declarations, strike, demonstrations, and physical violence). The Palestinian official voice, namely that of the Palestinian Authority, is rightfully emphasized, given the reported event (i.e. an Israeli raid that killed Palestinian civilians): The characterization of the act as ‘State terrorism’ and ‘war crime’ is taken up and this last qualification is even used as a subheading. We see here how a news story can altogether be critical and fair.
Discussion
At the turn of the 20th century, objectivity has emerged as the journalistic standard that would best serve journalists in their quest for truth, defined as the ultimate goal of the profession. While challenged, rejected, and redefined in the past decades, objectivity is often equated with impartiality, which is itself an evolving notion. First, synonymous with neutrality and non-partisanship, impartiality was meant to help journalists grasp the social world in its entirety and depict it with scientific, unconnotated language. Yet, as historical events demonstrated the weakness of such ‘naïve empiricism’ (Schudson, 1978), the concept of impartiality became synonymous of balance: If interpretations are unavoidable, then journalists should even-handedly present both sides of the story. While remnants of non-partisanship can be found in journalists still advocating neutral or euphemistic language, the process of compensation at work in balance appeared to be the best practical way of achieving impartial reporting.
However, non-partisanship and balance are not free of bias. As truth does not lie in the middle, reality is neither neutral nor balanced and so these criteria can rarely lead to accurate descriptions. Without accuracy, journalists are inevitably unfair to the participants in their news stories. If we concur with Labov and others that narratives necessarily imply assignment of responsibility to some of their participants, then it appears that fairness, rather than impartiality, is the best approach to accurate reporting.
Fair treatment considers each event as requesting from the journalist a renewed effort at characterizing the actors and their actions and at selecting voices to put in competition. This is not necessarily achieved through the identification of victims versus perpetrators, but, if so, it has to be done via a process of justification of the role casting in the narrative. As a result, each narrative might not be balanced, but the position-taking is fair in the sense that it is justified by the accurate report of actions attributed to each side and of the arguments formulated by competing voices. Journalists must justify their position-taking by the consistent, that is, universal, application of legal, moral, or ethical principles in a Kantian approach.
The case analyzed in the last part of the study was meant to illustrate how fairness could be evaluated through the criteria of consistency and justification. Undoubtedly, it only serves illustrative purposes, and generalizations from this single article to other articles or different issues would be hazardous. Nonetheless, this limited experience makes two main contributions to our research project.
The first is that Labov’s tables of analysis are a powerful tool for the systematic study of actors and their actions, voices and their utterances, as presented – and sometimes distorted – in journalistic narratives. To go from the assessment of the journalistic characterization of actors and voices to the evaluation of the journalist’s fair treatment of these actors and voices, one has to apply the criteria of consistency and justification to Labov’s tables.
The second contribution is to show that, despite the obstacles that scholars are prompt to emphasize, fair reporting is possible. In the case selected, the journalist did achieve a high level of fairness in his reporting on a sensitive case involving antagonists at war. The editorial line and political leaning of the newspaper, which are often designated as insurmountable constraints, did not prevent the journalist from performing quite well.
Future research should apply this model to a corpus of newspaper articles to evaluate the fairness in the coverage of a given issue, but also to assess the validity and reliability of the model based on Labov’s tables. It should try to include this model in a more general framework that would encompass the selectivity bias, which Entman and other scholars have endeavored to theorize and measure. Paradoxically, while content biases have been identified and discussed for much longer, their assessment has been neglected in the past decades, with research projects focusing on the external dimension of bias. Selectivity bias, referring to processes of framing or agenda-setting, has indeed been the center of attention of many scholars and has inspired many attempts at building methodological tools of measurement. However, our analysis of content bias suggests that it is intimately related to selectivity bias, and our work thus calls for further research that would integrate our methodological tools in a broader model, taking all forms of bias into account.
Footnotes
Acknowledgements
I am very grateful to Paul Frosh for his comments on earlier drafts of the manuscript and to the two anonymous reviewers of Journalism, whose comments and recommendations have helped me to considerably improve this work.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.