What concerns us here are television images. These belong, to be sure, to a semiology of the image, but they also belong to a pragmatics of showing. They call for an analysis that is less the one we would devote to a painting than the one we would devote to that painting in a museum. Like any exhibited painting, the television image is selected, framed, positioned, lit, designated, commented upon, given a title. Like any exhibited painting, the television image is the object of an “institutional retention”; it serves as an emblem for the institution that promotes it. But institutions differ in their ends. Some images will become museal, others find themselves journalized.”
Why speak of “journalized” images and not journalistic images? “Journalistic” is a pejorative term. “Journalized,” by contrast, proposes a normative reference. The added value of journalized information, as compared with the kind of information one can now find on all sorts of internet sites, is that of the filtering carried out by journalism as a professional body, as a normative institution, as an institution bound to the norms of democracy. To the question “What would happen if journalism did not exist?” the sociologist Michael Schudson replies: “We would have to invent it.” But what, then, would have to be invented? What must we retain when we want to reinvent journalism? The journalistic? Certainly not. What remains, then, is “journalized” information, which is characterized as information subjected to criteria. The filtering process that is called “gate keeping” represents a validation of what gains access to the public sphere (cross-checking of information, representativeness of opinions, respect for the deontological rules separating public and private). Information that does not meet these criteria is theoretically turned back. It does not deserve to pass into the public sphere. If it is problematic with regard to one or another of the validation criteria, yet nevertheless seems important, the information is submitted to an “investigation,” at the end of which the criteria will either be satisfied or the information rejected.
And yet we know full well that a great deal of information that does not deserve to pass into the public sphere passes into it all the same, and that journalistic information (the kind that gains access de facto to the public sphere) is not always journalized information (the kind that enjoys, de jure, the right to appear there). The gap between the journalistic and the journalized refers to pathologies in the functioning of the public sphere, pathologies that it is important to identify and to combat. When we are asked to mobilize in defense of journalists, we are asked to mobilize in defense of journalized information, and not to call for “tabloids” or to promote sheets of denunciation. What, then, in matters of the image, are the criteria of “journalized” information? How can they be described, and what ties do they maintain with the pragmatics of showing?
The Stakes of Showing
Showing as Gesture
Faced with a tragic event, the act of showing takes part in a facilitation—of mourning, or of denunciation, or else of that unwholesome jubilation the Germans call “schadenfreude.”
The televisual image is in fact accompanied by an essential dimension: it proposes acts of looking. The gaze that we are invited to share certainly conveys visual contents, but it constitutes first of all an acting. This acting belongs to the sphere of gesture. A gaze is never neutral. There is a way of seeing that expresses proximity or sympathy. There is another that expresses distance or remoteness. Ways of seeing may be adopted or, on the contrary, rejected with horror. It seems to me, then, that these fundamental attitudes that J. L. Austin calls “behabitives” are at the heart of the activity that consists in showing. Like ideological narratives, images translate gestures or postures: Being with or facing. Being near or far. Including, Embracing, or Excluding, Repelling.
This is why, however indispensable it may be, content analysis renders so poor an account—on its own—of the coverage of events such as, for example, September 11, 2001. Contents are inert realities, decontextualized realities. In situ, one image responds to another image. One discourse to another discourse. To extract them from the continuity in which they are inscribed amputates them of their entire dimension as acts.
The worldwide circulation of images is not simply a circulation of visual contents. Different public spaces engage in a labor that consists in peeling images away from the acts of looking in which they figured, and in pairing them with a new act of looking. The same images are mobilized by different narratives, and these narratives are themselves carried by different globalizations. To be sure, there exist images shared by all, crossroads-images. But crossroads are conducive to collisions. The same images lend themselves to divergent evaluations, to opposing gestures. However astonishing they may seem within a given public sphere, certain “aberrant readings” can come to dominate another. (One can pass from compassion for the September 11 dead to jubilation at the existence of those same dead.) It is through such “aberrations” that history is made. It is important to know, not simply what the images say, but what the different public spheres do with what the images say.
Does this then amount to saying that images are infinitely malleable? No.
Often images resist. Their content, however polysemic it may be, does not lend itself to the act of looking that one wants to inject into it. Such is the chief interest of the detailed analysis of images, and of a type of reading exemplarily carried out by Marc Ferro, and advocated in a different context by another historian, Carlo Ginzburg. It is a question of an indexical reading of the image; of a reading of the image against the discourse that puts it into act. Often, indeed, the image informs only on the second reading. We have an example of this each time we replay the images of a television newscast. We then see these images part company with the discourses that framed them. What was happening, then, on the first reading? Let us take an example.
“Behabitives,” “Verdictives,” Rituals. Suffering at a Distance
What television can offer that is most precious is time. This means that a live narrative corresponds to a sumptuary expenditure. Such an expenditure marks the importance of the event. In certain cases, and if it is sufficiently autonomized with respect to the habitual discourse of the news, the narrative of an event leaves the domain of information to gain access to that of ritual. It is interesting to know which sorts of images thus leave their role as news to be assigned to this ritual role, and what they are going to speak of.
They will often speak of sufferings, according to a mechanism very well described by Boltanski in La Souffrance à Distance (Distant Suffering).[^1] At the outset there is the compassion of the spectator before a suffering that he can see in detail and often in real time, on his screen, but to which he can bring no response, no comfort. This impotent empathy leads to the indictment of the one guilty of this suffering, to his denunciation, to his condemnation, and to a mobilization against this guilty party.
One can then instrumentalize the suffering that is exhibited. It suffices, indeed, to show the sufferings of some and to keep silent about those of others. One must then translate sufferings too great, and therefore invisible, into individual pains, breaking them up into sufferings that are visible, into sufferings small enough to be personalized. One must, in other words, organize a “drip-feed” of sufferings.
One must then, playing on the range of verdictive utterances, distinguish between victims and torturers, between the innocent and the guilty, between relative culprits (excusable by circumstances) and absolute culprits. One must finally, playing this time on the range of “behabitives,” sort out the near and the far; distinguish between those we include in the WE, and, facing this we, those who are to be designated as the OTHER. The narratives about September 11, for example, succeeded in sculpting a figure of the United States that, for the French, became—in less than six months, and well before the Iraq war—that of the other. Such a sculpture corresponds fairly closely to the range of “behabitives.”
On the one hand, it appeals to those gestures that rest on the Latin root “cum” (comprehension, commiseration, compassion, condolence). On the other, it appeals to those that are built on the Latin root “de” (unveiling [dévoilement], denial [démenti], denunciation [dénonciation], derision [dérision]).[^2]
Constructing Public Opinion
All of this directly poses the question of the effects of the media on the construction of public opinion. An uninteresting question, many researchers assert, rolling their eyes whenever one still dares to speak of “effects.” Negligible effects, many journalists keep saying, having ended up memorizing in defensive fashion the lesson of the “Two-step Flow.”
They have misunderstood the lesson. The thesis according to which the media have limited effects in no way excludes that some of their effects, in certain contexts, may be immensely powerful. Those very factors that in general limit the power of the media can, in certain cases, multiply it. The role of opinion leaders consists, in fact, in serving either as filters or as amplifiers. There either does or does not exist a relay public, minorities actively engaged in a role of constructing reception. The instrumentalization of sufferings aims precisely at mobilizing existing publics, at creating new publics, and at launching them into action. Thus can sufferings to come be anticipated and condemned in advance; thus can invisible sufferings be represented or mimed by publics that want to make others aware of them. Such is the role of a new type of demonstration.[^3]
What is most striking in the post–September 11 situation is that we are in a situation of exacerbation, in a situation where there flourishes what Bateson qualified as schismogenesis. Certain narratives of events engage the will to—and the capacity to—understand the motivations of various actors. Others are conceived in such a way as to introduce the maximal distance between the “we” one claims to belong to and the other one constructs. Everything in them is done to push toward the escalation of conflicts, toward the adoption of extreme and irreconcilable positions. Must this escalation, then, affect the very practice of information by creating parallel factualities, strictly subordinated to the friend/enemy “behabitive,” to the imperative of “cui bono”? Must one, on the contrary, return to the notion of a shareable information, and thus to that of a possible objectivity?
A Performance That Would Be Called “Objectivity”
Can one still speak of objectivity without exposing oneself to ridicule? Someone will have to take the risk of doing so. I shall take this risk with reference to Clifford Geertz’s quip: “It is with objectivity as with absolute sterility in surgery. Such sterility does not exist. Must one for all that perform surgery in the sewers?”
Of course, objectivity is constructed. Of course, objectivity possesses a genealogy. It even has several. It already engages a dramaturgy of “showing” in the stagings to which the empirical science of the eighteenth century devotes itself. It also possesses a more commercial genealogy. It is, in fact, what allows agencies wishing to sell the same news to different newspapers to broaden their clientele by avoiding offense to the convictions of those newspapers’ publics. Objectivity is therefore a construction, and it suffers from an original sin, from the flaw constituted by a commercial heredity. Is this enough to dismiss it?
Let us first underscore what such a construction makes possible. Objectivity—even as a commercial notion—refers to the existence of differentiated publics. It allows these publics to form an idea of certain events on the basis of narratives that must agree on a certain number of relevant elements. These events are thus rendered available for a common debate. Their availability serves as a precondition for discussion. Different publics have at their disposal the same series of facts to interpret, without the choice of facts prejudging their interpretation; without various publics having the possibility of seceding, each of them furnished with “facts” tailored to the measure of its tastes. It is therefore the possibility of a debate on shareable foundations, referring to the notion of a common world, that the notion of “objectivity” makes possible. However, if it makes possible a debate that is not a fight to the death, if it makes it possible to attenuate the risks of “schismogenesis,” must this putting-in-common of narratives for the benefit of differentiated publics still be called “objectivity”?
The notion of objectivity has been subjected to a barrage of criticism. It must be reformulated. I propose to do so in terms that escape most of the criticisms formulated up to now, since I place at the heart of my argument the constructed character of the practices it engages. The “objectivity” that I shall attempt to reformulate here does not engage an adequation between a reality and a discourse that proves impossible. It allows us to designate the norms that lead to a certain type of performance. This normative ideal, and this type of performance, come to complete what is taught in schools of journalism (multiplicity of sources, cross-checking, independence… etc.). I propose, then, to describe here, very schematically, this sort of oxymoron that objectivity-as-performance constitutes.
The performance objectivity is defined fundamentally as a commitment, a contract, a promise. Serving as a support for trust, this contractual dimension overdetermines all the others.[^4] It is then declined into more specific contracts.
## A Commitment of Loyalty
The first of these contracts is a contract of loyalty. This loyalty is manifested toward rules. It is fidelity to the rules that, in a given society, or for a given public, make a piece of information relevant or important. Relevance is then that of events (which one can report or pass over in silence) or that of social problems (which one can place on the agenda, or abandon to their fate).
These rules of relevance are often implicit. Nevertheless, they are fairly easily inducible. Journalistic practices never cease to illustrate them. Let us say that if a given type of event is highly relevant when it concerns Rwanda, it cannot cease to be so when it concerns Sudan; or that if racism causes scandal in the case of group A, it cannot lose all interest as soon as it concerns group B.
Respect for the social relevance of events and problems poses the question of their visibility. One can thus formulate as a rule that: Everything that is relevant for a given society must be visible. Conversely, only what is relevant must be visible. Finally, the degree of visibility must be proportional to the degree of relevance, or of importance. It is with reference to criteria of relevance that one can call into question incomprehensible silences or, on the contrary, wonder at monstrous valorizations. Let us imagine—as the Norwegian researchers Johann Galtung and Mari Ruge did—that one were to draw maps of the world, designating countries or continents on the scale of their representation by the media. The result would very likely be astonishing. To be sure, the country to which the media in question belong will always figure as a center mapped in the extreme of detail, but one would also see countries change into pinheads, entire continents vanish, and minuscule spaces expand to the dimensions of a universe in which each neighborhood or each sidewalk would be depicted in detail, as if it were Joyce’s Dublin or Cavafy’s Alexandria. To this hallucinatory geography, the point is obviously not to respond here with a parody of exact science, but to underscore the existence of a normative articulation between three notions: social relevance; the contract of loyalty by which journalism commits itself to respecting it; visibility.
Visibility is a “doing,” a putting-into-visibility, the result of an act of showing. “Journalized” information commits itself to ensuring a public visibility to situations that satisfy the criteria of relevance. Such a commitment constitutes a first form of fidelity.
## An “Expositive” of Legibility
The second contract envisaged here also engages a dimension of relevance, but it bears no longer on the what of showing, but on its how. It is a matter of reporting the elements necessary to the understanding of the situation, of reporting all these elements and these elements only. One need not be a positivist to remark that the absence of certain crucial information, and the overabundance of useless information, harm the understanding of a situation. For example, the census of obese Americans is not directly relevant if the point is to understand the American intervention in Kabul. By contrast, the omission of a big bad wolf would no longer make it possible to identify the story of Little Red Riding Hood.
The contract of legibility consists in reporting the interaction shown in terms by which it can be identified. (For example, the narrative of a race implies at least two runners; that of a fight, two combatants.) The putting-into-visibility engages a performance that consists in ensuring the intelligibility of what is shown. In Peircean terms, “showing” would consist in producing an interpretant thanks to which the situation evoked would become legible. In Austinian terms, the image of an event is never a simple constative, but an expositive, and it consists in saying: “Here is what I mean by the situation X.”
The situations concerned by the news do not generally concern objects of the physical world, but those objects of the social world that are actions or interactions. In the case of television news, it is always a matter of someone doing something to someone else, or against someone else, or with the help of someone else, or in spite of someone else. Each interaction refers to a set of constitutive rules that make it possible to identify it. These constitutive rules can be respected by the expositive of showing. The interaction is then identifiable. But they can also fail to be respected.
In this case, the expositive performance can give rise to two sorts of slippage, can provoke two sorts of pathologies of legibility. The first consists in creating illegibility. One can thus show a reaction without showing that to which the reaction responds, a riposte without an attack, a reply without an interlocution. The second consists in creating a fallacious legibility, in identifying an interaction other than the one that has been filmed. Thus can the interaction “beating” become the interaction “brawl” as soon as the one beaten is unsympathetic and the beaters are, moreover, identified as likeable. Or again, changing scale, the interaction “genocide” can be replaced by the interaction “ethnic conflict” (which is what Le Monde did, Edwy Plenel recounts, in its initial approach to the events in Rwanda).[^5]
Just like a musical interpretation (more or less loyal toward the score), the journalistic performance can present itself as a faithful performance, or as a betrayal. “Fidelity” here consists in framing the interaction in such a way as to allow the receiver to identify its “constitutive rules” (Searle). “Betrayal” consists in making another set of rules glimmer, often with reference to a pre-judgment bearing on what the presented event ought to be. This pre-judgment presents many dangers, the chief of which is to blind one to new information; to facilitate its capture by the stereotype. Let us take an example particularly interesting for its visual dimension.
On March 26, 2005, around seven o’clock in the morning, I tune in to the news on the Euronews channel, and come upon one of the sequences that make for the channel’s originality. This daily sequence is made of shots broadcast without the slightest commentary, and in which the image (supplied with its soundtrack) “speaks for itself.” The exercise consists in offering a pinch of reality, a moment of everyday life, a sequence whose banality refers to the temporality of the imperfect tense: a reality that, in the place where it unfolds, is quite ordinary may become exotic or enthralling for distant spectators…
But this morning the image is different. No imperfect tense. We are at the heart of a present. We are at the heart of an event. We are in that space everyone knows as the “occupied territories.” A sequence shot makes it possible to see, in the foreground, soldiers in fatigues, weapons in hand. About ten meters from the camera stands a little boy of about eleven, terrified. The soldiers are shouting orders that we do not understand. The little boy obeys these orders docilely. He is told to get rid of a kind of vest he wears around his body. He throws it down. He is then told to take off his shirt. With trembling hands, he takes it off. He is told to lower his trousers. The little boy obeys. The child is publicly humiliated by the soldiers, no doubt while worse is awaited. There is no commentary at all. We move on to something else.
This image is accompanied by no commentary. But it is legible without commentary. It answers to the scenario of a collective fantasy. Freud would call this fantasy “a child is being beaten.” Here one would say rather: a little Palestinian is being brutalized by Israeli soldiers. The three roles foreseen by Freud are well established. The child is humiliated,