Article
A visual framing analysis of British press photography during the 2006 Israel-Lebanon conflict
Media,War & Conflict 3(1) 67-85 ©TheAuthor(s) 2010 Reprints and permission: sagepub. co.uk/journalsPermissions.nav
DOI: 10.1 177/1750635210353679
http://mwc.sagepub.com
®SAGE
Katy Parry
University of Liverpool, UK
Abstract
This article develops a model of visual framing analysis through an examination of the photographic representation of the 2006 Israel-Lebanon conflict. The intentions of this investigation are twofold: first, to compare and contrast the use of press photographs in two 'quality' British newspapers - TheTimes and The Guardian; second, to develop and test a model of visual framing analysis which takes seriously both the visual elements and verbal context of the photographs. The study employs a detailed content and framing analysis of all press photographs relevant to the conflict, with the following questions in mind: What is the character of photographic representation of both sides in the conflict and how does it fit with various moral evaluations and political interpretations of the war? Which groups are shown in an empathetic light? Is there any coherence to the framing of the war, or are many alternative explanations presented?
Keywords
conflict, Israel, Lebanon, photography, photojournalism, visual framing
On 15 July 2008, two years on from their reported capture, the bodies of reservist Israeli soldiers, Ehud Goldwasser and Eldad Regev, were returned to Israel as part of a prisoner exchange with Hizbullah. According to Israeli news sources, post-mortems confirmed that the soldiers had died in the original attack. Goldwasser and Regev were apparently kidnapped by Hizbullah fighters following an ambush on 12 July 2006, sparking Israel to declare their abduction an 'act of war', as The Guardian newspaper reported the following day:
Israel threatened to bomb Lebanon 'back 20 years' yesterday as it launched a ground and air assault after Hizbullah captured two of its soldiers and killed others. The attack escalated a
Corresponding author:
Katy Parry, Department of Communication and Media, University of Liverpool, Roxby Building, Chatham Street, Liverpool L69 7ZT, UK. Email: katy.parry@liv.ac.uk
68
Media,War& Conflict 3(1)
crisis that began last month with the abduction of another soldier now held in the Gaza strip where Israeli forces also launched a fresh assault yesterday and killed a family, including seven children. (McGreal, 2006)
The abduction was reported as the immediate impetus for military action by Israeli forces and yet the reasons for the conflict can also be viewed within a wider context of ongoing conflict in the Middle East. The ensuing conflict lasted for over a month, with a UN ceasefire taking effect on 14 August 2006. It is not the intention here to examine the legitimacy of the various protagonists' actions and objectives in the conflict, but rather to analyse the photographic depiction of the related events and people in the British quality press.
The aims of the investigation are twofold: first, to compare and contrast the use of press photographs in two 'quality' newspapers - The Times (traditionally the newspaper of record in the UK and owned by Rupert Murdoch's international News Corporation) and The Guardian (a liberal leftwing newspaper owned by the Scott Trust);1 second, to develop and test a model of visual framing analysis which 'takes seriously' (Rose, 2001: 15) both the visual elements and verbal context of the photographs. I posit that news photographs are traditionally an under-investigated area for news media scholars, in part because of the difficulties associated with their 'polysemic' attributes (Hansen et al., 1998: 191; Entman, 2004: 56). However, viewed in their original news context, I argue that photographs can give salience to particular framing(s) of news events offered in newspapers through their selection and omission, depiction, symbolism and lexical context (caption and headline). This article considers the detailed content and framing analysis of all photographs relevant to the conflict in The Times and The Guardian newspapers from 13 July to 23 August 2006 and has the following questions in mind: What is the character of photographic representation of both sides in the conflict and how does it fit with various moral evaluations and political interpretations of the war? Which groups are shown in an empathetic light? Is there any coherence to the framing of the war, or are many alternative explanations presented?
Photographs in the news
When considering visual communication and photojournalism, one should distinguish the practice of photojournalism from other areas of photography, recognizing its particular context and accepted uses. To understand news photographs we need to be aware of both the constraints and routines of those involved in the various stages of production, and the wants and expectations of the readers. 'What is photojournalism commonly supposed to be? Unbiased. Factual. Complete. Attention-getting, storytelling, courageous' (Becker, 1995: 5). Just as the rather opaque general 'news values' and the more distinct ideological or party political values of a publication constrain all reporters to a certain degree, so pho-tojournalists have to abide by rules of providing 'unambiguous' and 'meaningful' news images to secure inclusion (Galtung and Ruge, 1965). An understanding of expectations is crucial in a competitive news market and the visual reporting will both reflect the newspaper's own interpretation of events and be congruent with its audience's values: 'Although the media are required to reflect all sides of a story, retaining values and beliefs of the target audience is not only expected, it is also indispensable' (Fahmy, 2005a: 150).
Parry
69
Knightley (2003) writes that 'although in most cases the camera does not lie directly, it can lie brilliantly by omission' (p. 14). Recent studies on war imagery have pointed to the selectivity and framing of photographs in the mainstream press. In his analysis, Griffin (2004) finds that:
the present analysis of photographic war coverage in American news-magazines lends further support to the idea that news photographs prime and reinforce prevailing news narratives rather than contribute independent or unique visual information, (p. 399)
Chiming with Entman's (2004) findings on photographic depictions of US military action in Central America, Griffin (2004) finds little photographic evidence to add coherence or salience to counter-frames, contending that the range of photographs are 'arguably more severely restricted than the language of reporters and columnists' (p. 399). While Griffin does not refer explicitly to frame/framing analysis studies or methodology, he is concerned with news frames and in the thematic depictions of war, noting that photographs provide an 'emblematic' short-hand account of the written reports, selected on the basis of their perceived impact and symbolism (pp. 383^1).
The highly selective use of press photographs, along with their brief captions, may present a strong, forceful idea about a distant conflict. By omitting other possibilities, there is a danger of one-sided representation. This is perhaps most likely when the nation in question is directly involved in the conflict, so we may ask: What about the visual reporting of a supposedly neutrally-observed conflict? Which groups are shown in an empathetic light? Who is shown to suffer? When killing or destruction is depicted, how is it explained?
Framing news photographs
There are two key notions of 'framing' that are pertinent to this study: the compositional framing of the individual photograph and the application of framing analysis to visual images.
'It is the position of point-of-view, occupied in fact by the camera, which is bestowed upon the spectator. To the point-of-view, the system of representation adds the frame'' (Bürgin, 1982: 146, original emphasis). In this quotation, Bürgin goes on to propose that the 'frame' of the photograph organizes the objects depicted within 'into a coherence', a 'decisive moment'. Framing in this sense applies to the compositional qualities of the image. In taking a photograph, the photographer's message about what is captured within the frame is 'this is important' (a notion that is further emphasized via cropping and printing in a newspaper). According to Bürgin, the photograph is decoded 'instantaneously' and 'naturally' (p. 147), without an awareness by the reader that the photograph is not a compact reproduction of 'reality' but an example of a signifying system, whose conventions are so familiar that we do not realize we are adhering to them in looking. Photographic choices such as camera angle, focus and distance can be overlooked by the reader or viewer when making judgements relating to the figures depicted, but this analogical quality has consequences for the reader's response. The framing of a photograph, the time-bound capture of a particular composition, is one instant captured in a 'decisive moment', a window through which the implied spectator sees the world as shaped by the
70
Media,War& Conflict 3(1)
photographer's point-of-view. Politicians and political writers make claims about photographs which 'say it all' or 'sum up the atrocity' yet, as Perlmutter (1998) writes, such metonymy is illogical and yet enticing for journalists: 'The seeming naturalness of the imposed metonym masks the variability of its context ... narratives seem compelling, familiar, and are assiduously replicated' (p. 17).
This definition and directive purpose of compositional framing is replicated in the English language to express choices of opinion or 'outlooks'. Burgin (1982) goes on to suggest that 'the structure of representation - point-of-view and frame - is intimately implicated in the reproduction of ideology (the "frame of mind" of our "points-of-view")' (p. 146). It is this slippage between the two meanings of 'framing' and 'point-of-view' that is of interest here. The photojournalist Susan Meiselas commented that while she can choose how to 'frame' the image in terms of selection and shooting, she has no power over how an image is 'framed' in the larger context of appearance, for example, how it will later appear in a newspaper (Meiselas, 1987: 33).
Such framing in the 'larger context' brings us to the second definition of framing. This definition is derived from its employment in the concept of 'framing analysis' (or 'frame analysis') and has been hinted at in Perlmutter's description of iconic images as metonyms. Indeed, one of the earliest scholars to apply framing analysis to media communications, Gaye Tuchman, compared news frames to window frames. Tuchman (1978) wrote that the: 'characteristics of the window, its size and composition, limit what may be seen. So does its placement, that is, what aspect of the unfolding scene it makes accessible' (p. 209).
Two often-cited definitions of news frames are presented by Gitlin and Entman. Gitlin (1980) stresses the routines and practices of journalism in his definition: 'Media frames are persistent patterns of cognition, interpretation, and presentation, of selection, emphasis, and exclusion, by which symbol-handlers routinely organise discourse, whether verbal or visual' (p. 7, original emphasis). Entman (1993) defines frames as the selection of certain aspects of reality 'to make them more salient in a communicating text in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation' (p. 52). Such descriptions echo the definition of framing within photographic discourse, yet here the definitions go further in suggesting not just the promotion of a primary subject, but also the promotion of a 'problem definition, causal interpretation, moral evaluation, and/or treatment recommendation'. Whereas Gitlin stresses media frames as 'patterns of cognition ... interpretation', etc., Entman extends their remit in suggesting that frames serve to explain causal factors of events and provide moral evaluations or recommendations. Gitlin sees the employment of frames as a practical necessity, part of the routine of news reporting which has to convert complex issues and events in the real world into understandable, digestible, concise news events. While these frames or patterns are 'persistent' over time, this does not mean that their meanings are totally fixed, rather that they are routinized in nature to the extent that they provide short-hand frameworks for understanding various events.
It should be noted that framing is not always conceived as a hegemonic force with no competition. Dominant values are not always consistent, and indeed resistance could come from an oppositional sphere.
Parry
71
For many events, there may be more than one frame suggested, and one needs to ask questions about the prominence of competing frames in the same news report. All senders - whether journalists or sources - should be regarded as sponsors of frames. (Gamson, 1989: 158)
Wolfsfeld (1997) takes this idea further with his political contest model. In this model, all political players fight for control of the political environment, in which the news media are key. In his model, the news media 'serve as public interpreters of events and as symbolic arenas for ideological struggle between antagonists' (p. 54, original emphasis). While Wolfsfeld looks at the structural dimensions to explain the power relationships between political actors and media actors, the cultural dimension of his analysis 'focuses on how norms, beliefs and routines all have influence on construction of media frames' - political contests are also struggles over meaning as well as access (p. 5).
The concept of visual framing analysis in relation to media and conflict provides a relatively new method in approaching the study of visual texts. Messaris and Abraham (2001) point to a central role for visual images in framing research, with images reinforcing cultural stereotypes which may not even be referred to in the lexical-verbal text.
The iconic ability to seemingly reproduce nature means that visual images are capable of producing documentary evidence to support the commonsensical claims of ideology, and in turn to use the very appearance of nature (seemingly factual representations) to subtly camouflage the constructed, historical, and social roots of ideology, (p. 220)
While the visual depiction may not be so easily 'read' in a syntactical and monosemous manner, experimental analyses, which test how the presence or manipulation of images can shape attitudes, have shown how either the inclusion of a forceful image or the variance in head angle and gesture of the human subject can impact on critical or positive evaluations of the depicted individual or group (Domke et al., 2002; Petersen, 2005; Arpan et al., 2006). The authors of such studies note the relative lack of scholarship dealing with the visual element of news coverage and call for a methodology to include visuals at the heart of media content analysis; 'while scholars in recent years have begun to devote increasing attention to people's use of core values and mental categories to sift through news messages ... the role of visual images is virtually unexamined' (Domke etal., 2002: 133).
More recently, scholars have turned their attention to developing robust methodologies for visual framing analysis (for example, see Fahmy, 2004; Perlmutter and Wagner, 2004; Fahmy, 2005b), with coverage of the Iraq conflict attracting particular consideration (King and Lester, 2005; Schwalbe, 2006; Fahmy 2007; Fahmy and Kim, 2008), while others are more cautious in their visual analysis, stressing its 'suggestive' nature (Entman, 2004: 56). I do not detail approaches or findings from these studies in this article but would like to highlight that they offer exemplary and influential cases for the development of visual framing analysis.
72
Media,War& Conflict 3(1)
Methodology
Content analysis
Rather than a content analysis based predominantly on the lexical-verbal text, this study uses the photographic image as the primary unit for analysis. I now provide a summarized version of the coding schema, detailing the most significant variables applied in this study.2
Coding schema
1. Newspaper (The Guardian, The Times)
2. Date
3. Page
4. Relevant headline? (headline text entered in database)
5. Caption? (caption text entered in database)
6. Graphic nature: in most cases this will have a 'not applicable' ('n/a') value coding. However, where casualties are depicted, this variable will provide further information as to the visual impact of the photograph.
Potter and Smith explained two ways to identify graphic imagery. The first is close-up versus long shot. A close-up image of a violent act is more graphic than a long shot of that same act ... Second is the degree of physical alteration to the victim. For example, in portraying a body, in a nongraphic image the victim may lie with his or her eyes shut; in a graphic image, the victim may be lying in a pool of blood. (Fahmy, 2005a: 148)
The implied social distance and degree of physical alteration are the key determinants identified by Potter and Smith, and adopted as a measure by Shahira Fahmy (2005a) in her survey of photo editors' and photojournalists' attitudes to visual coverage of 9/11 and the Afghan War. I would argue that further factors are crucial in determining the graphic quality of press photos; namely, the degree of framing or cropping, so that we only see body parts rather than the whole body (here labelled as 'disembodied'), and whether or not we see the face of the victim (see Hallin, 1986; Taylor, 1998).
7. Distance? Distance for central subject focus of the image? Here, the typology as explained by Kress and Van Leeuwen (1996: 130) - extreme close-up, close-up, medium close, medium, medium long, long - is applied to the main human subject(s) of the photograph.
8. Photo subject/theme: main heading (e.g. political people, military people, destruction): sub category (e.g. Ehud Olmert, Israeli soldier, post-missile rubble). Enter value from a drop-down menu corresponding most closely to the sub category (see Table 1):
Parry
73
Table I. Example subject codes
Photo subject no.	Main heading	Sub category
01	Political people	Ehud Olmert
22	Military people	Israeli soldier
71	Destruction	Post-missile rubble
Code for more than one subject if necessary (87 variables in total)
Framing analysis
Similar to Schwalbe (2006), this study employs an inductive approach in its choice of frames; they are selected based on knowledge of the context and rationales particular to this specific war, rather than more generic frames (see Table 2). Each photograph is allotted a related frame code if deemed relevant for that particular photograph - more than one frame code can be applied to any single image. The selected frame also receives a 'tone' coding (reinforced/mixed/undermined) to indicate a clear reference to the frame in a reinforcing or deflating manner; or alternatively an ambiguous reference will be 'mixed'.
Framing: two framing questions are asked of the coder:
9. Photographic framing: Which frames can be applied to the image and can the imagery alone3 be described as 'reinforcing' or 'undermining' the frame? Are there visual elements that evoke cultural ideas or values related to the frame? If no frame is evoked, code as 'not applicable' and leave the 'tone' variable blank. Coding will involve a drop-down menu for any relevant frames (e.g. Israeli responsibility for casualties or destruction) with appropriate tone coding (reinforced/undermined/mixed).
10. Linguistic framing: does the caption or headline reinforce or undermine the (above) framing? That is, does the lexical choice in the background information provide extra detail to guide interpretation? Include instances here of lexical framing even if the image by itself does not warrant an evaluative framing code. Use the tone code as before, congruent with the extra information supplied in the headline, sub-heading or caption.
More than one frame can be applied in both controls. The same frame can also be used in both photographic and linguistic framing categories.
The coder uses the same variables for both photographic framing and linguistic framing: see Table 2 for further explanation of the frames and how they fit with Entman's typology. Photographs of casualties are perhaps the most likely to convey strong messages, especially where blame is attributed for the injuries or death. Clear attribution of responsibility for destruction of homes and/or civilian casualties indicated in the caption or headline is an important element in framing the conflict.
Integrating the various evaluative frames in Table 2, two distinct meta-frames can be deduced.
Table 2. Framing codes
Frame type	Specific Lebanon-Israel frame	Expected imagery or caption?
Problem definition	Iranian/Syrian support for Hizbullah	Usually in written context but could be indicated by associated flags or political poster
	Israeli disproportionate response	Code for particular reference, could be linked to destruction/casualties (could also only be apparent through collation of many instances)
	Hizbullah aggression	Negative military imagery, focus on destruction, reason given for the start of war
	Israeli aggression	As above
Causal interpretation	Israel's right to defend itself against terrorism	Hizbullah illegitimacy, terrorism links, threat to Israeli security
	Zionist imperialism	Expansionism of Israel (not used in practice)
Moral evaluation	Empathy/humanization of Lebanese civilians	Civilians in portrait, medium close-up, especially if named or injured
	Empathy/humanization of Israeli civilians	As above
	Israeli responsibility for casualties	Injury, death, destruction of homes or infrastructure with attribution
	or destruction	explicit or implicit
	Hizbullah responsibility for casualties	As above
	or destruction	
	Negative humanitarian consequences for	Refugees, humanitarian deliveries, etc. with no apparent attribution
	whole region	of blame
Treatment	Military might wins out for Israelis	Weaponry, victory symbolized (seizure of flags, enemy weaponry
recommendation		or land)
	Military might wins out for Hizbullah	As above
	Peaceful reconciliation possible	National leaders shown achieving peace talks, or with civilians
	Peaceful reconciliation: international solution	International leaders shown in good light, making progress
	Israeli government in crisis	Olmert under fire from those in own country, unflattering depiction
	Hizbullah/Lebanese government in crisis	As above for Lebanon/Hizbullah
Other	Heroism/empathy of Israeli soldiers	Soldiers shown relaxing, under pressure, named
	Heroism/empathy of Hizbullah	As above
	Heroism/empathy of other soldiers	As above
	Historically determined outcome, brutal cycle	No ceasefire or peace with both parties to blame
	Historical: appeal to past conflicts/glories	Link to past wars through old footage (not used in practice)
	Other	For any frames not considered above
Parry
75
Frame one: Lebanese perspective. The problem lies with Israeli aggression and imperialism in the region and the disproportionate use of military action against civilians as well as militiamen. Such acts of aggression are seen in the indiscriminate bombing of civilian areas causing wanton destruction of infrastructure, homes and lives. With substantial funding from the United States, Israel exerts asymmetrical power in the region, avoiding international legal constraints which are stringently applied to other states.
Frame two: Israeli perspective. The problem lies with ongoing attacks on Israel by Hizbullah fighters, who hide among civilians and therefore bring about further hardship for those they claim to be fighting for. Supported by Iran and Syria, Hizbullah threaten the security of Israeli citizens and employ terrorist tactics to disrupt Israeli life and kill civilians. The only solution for Israel is not to back down in the face of terrorism and the threat to its very existence.
Alternative frames. Refugees need to escape as the Middle East countries yet again become embroiled in irreconcilable conflict: what British leaders are doing to resolve the situation (where the story is given a national interest slant). The international community is needed to force reconciliation and rescue civilians. Further diplomatic talks are required to break the destructive cycle of violence.
Findings
Both newspapers published a similar amount of visual coverage, with 99 fully coded photographs in The Times and 112 in The Guardian (no Sunday papers were included).4 If results are restricted to the front page of the newspaper, The Times printed only 8 photographs, while The Guardian had 19 in the same period. This does not indicate how often the conflict appeared as a news story on the front page but it does indicate that The Guardian was more likely to add prominence to the reporting of the conflict with front-page visualization. The detailed coding schema allowed for a great number of variables to be recorded and combined in 'queries' using the MS Access database. However, I will focus here on the 'photo subject', 'graphic nature', 'distance' and 'framing' variables.
Content analysis: photo subjects
Rather than restricting the coding of the photo subject to one type of actor or landscape feature, each photograph could have multiple subjects assigned to it, providing they formed a substantial part of the captured image. One of the most striking aspects of the coverage is the focus on civilian lives during the conflict. For both newspapers, Lebanese civilians were the most frequent subject(s) of photographs (featured with no other actor types in 23 per cent of photos in The Times and 24 per cent in The Guardian). Including the additional photographs in which humanitarian workers are featured with Lebanese civilians, this rises to 25 per cent in The Times and a substantial 30 per cent in The Guardian. The two newspapers had similar subjects as their predominant photographic
76
Media,War& Conflict 3(1)
Table 3. TheTimes:the seven most prominent photographic su		bjects
Photographic subject: sub category	No. of photos	% of photos
Lebanese civilian	23	23.2
Israeli soldier	19	19.2
Rubble/post-missile	15	15.2
Armoured vehicle	9	9.1
Israeli civilian	7	7.1
Refugees from Lebanon	7	7.1
Guns	5	5.1
Table A. The Guardian: the seven most	prominent photographic	: subjects
Photographic subject: sub category	No. of photos	% of photos
Lebanese civilian	27	24.1
Rubble/post-missile	21	18.8
Israeli soldier	17	15.2
Coffins	7	6.3
Humanitarian workers with Lebanese	7	6.3
Armoured vehicle	7	6.3
Airstrike	7	6.3
themes (see Tables 3 and 4 for the top seven subjects), but with Israeli soldiers more prominent in The Times (19.2%) than The Guardian (15.2%), along with Israeli civilians (7.1% compared to 5.4%).
Prominence of individual subjects can reveal only so much about the content of press photographs. When the photo subjects are collated and analysed within their main heading categories, the differences in representation between the two newspapers start to emerge (see Tables 5 and 6).
Table 5. TheTimes: photo subjects by main	headings	
Main heading	No. of photos	% of photos
Civilians/medical	47	47.5
Military people	28	28.3
Destruction (inc. airstrike)	22	22.2
Military hardware (ex. airstrike)	20	20.2
Political people	17	17.2
Flags/political posters	10	10.1
Mixed (political military with others)	2	2.0
Parry
77
Table 6. The Guardian: photo subjects by main headings		
Main heading	No. of photos	% of photos
Civilians/medical	55	49.1
Destruction (inc. airstrike)	41	36.6
Military people	22	19.6
Military hardware (ex. airstrike)	17	15.2
Political people	9	8.0
Flags/political posters	6	5.4
Mixed (political/military with others)	3	2.7
Both newspapers' photographic selections are dominated by depictions of civilian life or injury, including medical or rescue workers; what we might call the human cost of war (49.1% in The Guardian and 47.5% in The Times). Almost half of all photos relating to the war in both newspapers showed civilians as a main focus, whether injured, going about their everyday lives or as a personal portrait. The prominence of the other main categories starts to reveal signs of divergent coverage between the two newspapers. Whilst The Times is more likely to show military personnel (28.3% compared to 19.6%) and military hardware (20.2% compared to 15.2%), The Guardian reiterates the destructive cost of war with damage to homes, bridges and other buildings in 36.6% of all photos (compared to 22.2% in The Times). In both cases the military personnel depicted are largely Israeli, with only two photos depicting Hizbullah fighters in The Times and one in The Guardian (the other featured troops being UN or Lebanese). The near-invisibility of one side of combatants in this conflict is glaring. Hizbullah's photographic omission in the British press fuels their status as illegitimate, shadowy actors. Whether as underground, heroic guerrilla fighters or mercenary terrorists, the mythology of masked gunmen is a tradition continued in this (lack of) coverage. In The Times, the caption under the photo of a lone fighter, with gun raised in the air, reads 'Hezbollah's support in Syria and Iran is the main problem' (18 July 2006: 8) - the framing in that case is quite clear. The Times' second photo appeared on 21 July 2006, in all likelihood an archived image, showing a group of soldiers with their arms raised in a salute, not unlike a Nazi salute, with the main headline 'Hezbollah is fighting to the death, but who is it?' emphasizing their extremism and mysterious nature (p. 9). The sole photograph in The Guardian (29 July 2006) is taken from an extreme low angle, from the side, with the man's face covered by a scarf and his gun positioned between the camera and his body; the caption reads 'A Hizbullah fighter watches smoke rising after a rocket attack in southern Lebanon' (p. 1). In addition to editorial constraints and choices regarding their depiction, we can note two reasons for the near invisibility of Hizbullah fighters: the need to protect their anonymity and restricted (Western) journalistic access. This is in great contrast to the depiction of Israeli soldiers who are shown relaxing with comrades, involved in conventional military operations (searching, patrolling), returning after 'daring' night-time raids, and in victorious poses with captured Hizbullah flags, for example.
78
Media,War& Conflict 3(1)
Table 7. Frequency of photographs featuring graphic imagery
Value No.	Graphic nature?	TheTimes (%)	The Guardian (%)
1	Not applicable	76(76.8)	78(69.6)
2	injury not obvious (referred to in caption)	6(6.1)	7(6.3)
3	bodily harm evident	2(2.0)	1(0.9)
4	face shown, injured	7(7.1)	8(7.1)
	death: see face, no blood	1(1.0)	2(1.8)
6	death: see face, with blood	0(0.0)	1(0.9)
I	death disembodied	1(1.0)	4(3.6)
8	disfigurement	1(1.0)	0(0.0)
9	metaphorical (e.g. an empty shoe)	1(1.0)	2(1.8)
10	mourning	2(2.0)	1(0.9)
1 1	coffins/bodybags/covered bodies	2(2.0)	8(7.1)
	Total	99(100.0)	1 12(100.0)
Notes: Tints denote the scale of graphic imagery.
Graphic nature
The issues surrounding the graphic depiction of casualties are numerous and their impact is not straightforward. Though it could be argued that showing victims of war helps to humanize them and invoke empathy in the reader or viewer, extreme violence can also have an anaesthetizing effect (Sontag, 1979: 20,109). In 2003, Susan Sontag revised her earlier opinion on this effect, expressed originally in the collection On Photography: 'As much as they create sympathy, I wrote, photographs shrivel sympathy. Is this true? I thought it was when I wrote it. I'm not so sure now' (p. 94). Comparing still photography to the image-flow of television, Sontag acknowledges the limited function of photographs to explain or add understanding, but recognizes the 'vital function' that they perform: 'The image says: This is what human beings are capable of doing - may volunteer to do, enthusiastically, self-righteously. Don't forget' (p. 102).
In this study, 23 out of 99 photographs in The Times featured imagery of a graphic nature (23%), including depictions of mourning and coffins/covered bodies, compared to 34 out of 112 in The Guardian (30%). Only 4% of photographs in The Times showed death in some way, including covered bodies (2%), compared to 13.4% in the Guardian, most of which (7.1%) were also covered (body bags or coffins).
Combining these results with the 'distance' variable, The Guardian was also more likely to show 'physical alteration' or death in close-up shots. Isolating the results for values 3-8 in Table 7, in which death or injury is apparent, a majority of Guardian photos were between 'medium close-up' to 'extreme close-up' (56.3%), while The Times had 41.6%. Considering the small number of photographs in this selection of graphic images, it would be hasty to make any generalizing statements about major differences in the coverage between the newspapers, although The Guardian does appear to be more likely to print images of a graphic nature. Only The Guardian showed death on its front page.
Parry
79
Without conducting a comparative study with other conflicts, it is difficult to ascertain whether the violence shown during this conflict is more common or graphic than for other distant wars. However, other studies on US news media have indicated that photographs showing death or injury are rare. In his analysis of Time magazine, Entman (2004) found no pictures of civilian deaths from US military action in Libya, Panama and Grenada, although these are examples with direct US national involvement. He does however recognize the potency that such images might have: 'To show as many pictures of dead civilians' bodies as of heroic U.S. armed forces would seem inappropriate to the elite consensus and reported mood of public celebration' (p. 60). Griffin (2004) finds very few images of Afghan or Iraqi civilians at all in his analysis of US news magazines in the 1991 and 2001 conflicts, other than as 'stereotypes of the Islamic world', although there is some effort to include Iraqi civilians in 2003, mostly as recipients of aid, and not as casualties: 'the human and economic costs of war were largely absent from news portrayals' (pp. 393, 399). While Fahmy and Kim (2008) found that both the New York Times and The Guardian pictured 'civilian life' in around 21 per cent of all Iraq-related photographs during the invasion period from March to May 2003, only 2.6 per cent of photographs depicted a 'civilian casualty' (p. 452).
The human cost, and particularly the depiction of dead or injured children, would appear in this case to form a substantial part of this coverage when compared to other studies. It is suggested that the lack of direct involvement, or rather culpability, for civilian deaths increases the likelihood of their depiction reaching the national press.51 would specifically call attention to issues of media access, perceived audience empathy (based on geographical and cultural distance or 'otherness') and national involvement in the conflict as key to determining the salience of photographs and their graphic nature. Research based on decision making in the news process (Bissell, 2000), as well as on survey data of photo editors' and photojournalists' attitudes and perceptions (Fahmy, 2005a), has found that decisions about the inclusion of graphic material are largely based on personal political orientations and connected to the particular circumstances in which the human suffering occurs. However, as indicated by Fahmy's findings, visuals that are considered 'graphic' and their supposed impact are also somewhat changeable, dependent on the perceived value of the human lives depicted (pp. 159-60).
Framing analysis
The differences between newspapers, indicated in the results for 'photo subjects' and their 'graphic nature', are reinforced in the findings for photographic and linguistic framing. Rather than reporting results for the individual frames, I will refer to the meta-frames indicated in the methodology section. By collating all the instances of frames with an empathetic stance towards Hizbullah or the Lebanese people, and which receive a 'reinforced' tone code (along with the 'undermined' pro-Israeli frames), the number of photographs that framed the conflict through a supportive position for the Lebanese/ Hizbullah can be gauged (see the 'Frame one: Lebanese perspective' in methodology section). Likewise, the collation of the opposing frames in support of the Israeli government, military or people is obtained by reversing the 'reinforced' and 'undermined' tone codes for each frame (see 'Frame two: Israeli perspective').
80
Media,War& Conflict 3(1)
Table 8. Frequency of photographic framing codes
Photographic framing	The Times	The Guardian
Lebanese perspective Israeli perspective	25 frames for 25 photos (25%) 30 frames for 30 photos (30%)	65 frames for 52 photos (46%) 28 frames for 25 photos (22%)
Table 9. Frequency of linguistic framing codes		
Linguistic framing	The Times	The Guardian
Lebanese perspective Israeli perspective	48 frames for 37 photos (37%) 46 frames for 34 photos (34%)	88 frames for 66 photos (79%) 34 frames for 28 photos (25%)
Both levels of framing analysis demonstrate that The Times balanced photographs supportive towards the Israeli position with those supportive of the Lebanese very closely in terms of numbers of photographs and applied frames. In contrast, The Guardian showed a greater degree of empathy towards the Lebanese, through its depiction of the human cost of the war for Lebanese people, with over twice as many photographs classed as supportive in both photographic and linguistic framing results. Although the most commonly applied single frame for both newspapers was 'Empathy/humanization of Lebanese civilians', The Guardian was also more likely to be coded for 'Israeli responsibility for casualties or destruction', when culpability was clearly attributed, whereas The Times also had a greater share of 'Heroism/empathy of Israeli soldiers' and 'Hizbullah aggression' codes. A particularly moving image, of the same scene for which Guardian photographer Sean Smith later won a 2007 Press Photographer's Photograph of the Year award,6 shows a distraught boy next to his injured mother, under the headline, 'Blasted by a missile on the road to safety: Family ordered to flee were targeted because they were driving minivan' (24 July 2006, front page) (see Figure 1). The caption reads: 'Tears of a son: Ali Sha'ita, 12, is distraught as he tries to comfort his mother, who was wounded when an Israeli missile hit their vehicle, killing three and injuring 16.' His face and clothes covered in blood, next to his barely conscious mother, the complete anguish on the boy's face is further expounded by the caption and headline. Ali is also a named victim of an Israeli missile, his family's fate yet more tragic as they were 'on the road to safety'. By way of contrast, children pictured in The Times tended to be female, attractive and with clean, well-bandaged injuries.
Discussion and conclusion
Viewing the Israeli state and its supporters as 'sponsors of frames' (Gamson, 1989: 158) that compete with Hizbullah or the Lebanese people in the 'symbolic arena' (Wolfsfeld, 1997: 54) of the British press, the selection of photographic subjects, their depiction and graphic nature, along with the accompanying captions, reveal differing levels of success.
Parry
81
Figure I. 'Blasted by a missile on the road to safety: Family ordered to flee were targeted because they were driving minivan' (The Guardian, 24 July 2006, front page). Photographer Sean Smith. Copyright Guardian News & Media Ltd 2006.
The Times and The Guardian visually constructed this far-off conflict for their readers by focusing on some similar themes (the human cost), albeit with a greater emphasis on the lives lost in The Guardian. At a glance, The Times appeared to balance both competing framings and present a sense of neutrality, in which a roughly equal number of pro-Israeli and pro-Lebanese depictions were shown - this is succinctly illustrated in a double-page feature in which five people from each country are interviewed and their stories related (28 July 2006: 6-7). However, The Times' own figures for actual deaths, injuries and displaced people showed that up to 600 Lebanese (compared tol9 Israelis) had died by this point in the conflict (pp. 6-7). I would argue that the implied notion of 'balance' in these circumstances does not in fact reflect the scale of destruction in Lebanon but reveals editorializing decisions that serve to downplay the Israeli military force used against the Lebanese population. The Times' less graphic portrayal of victims could also reflect the impositions of its owner's preferences. Don McCullin quotes Rupert Murdoch (in relation to The Sunday Times) demanding 'no more starving Third World babies; more successful businessmen around their weekend barbecues' (Taylor, 2000: 134).
What looks like a bias or 'skewed' coverage in The Guardian arguably indicates an attempt to inform readers of large-scale devastation in Lebanon. Shocking images are perhaps voyeuristic and sensationalist, but they also add emotional impact to the written narratives of civilian casualties and experience. That voyeuristic pull may even mean that readers hesitate before turning the page and so become engaged and informed about a distant conflict. Photographs of suffering can provide a rare connection with victims who otherwise would have no voice or access, and at the same time make us aware of the knowledge that our political leaders are privy to, and may/can do something about. Ideally, and perhaps rather idealistically, the pictures are an 'invitation to pay attention'
4895
82
Media,War& Conflict 3(1)
(Sontag, 2003: 104). As Susan Meiselas explains of her own photojournalistic approach, the photographer is 'opening a door slightly' making a 'sketch' which 'enfranchises the reader in the journalistic process ... It's as if the reader should not only witness but be implicated by the reality of change, not only the idea of it' (Meiselas, 1987: 41).
The second intention of this study is to test a methodology for visual framing analysis. The findings detailed here only just begin to mine the rich levels of data available in the full coding schema. The content analysis allows for further research on both compositional elements (gaze, angle) and editorial decisions in the selection, placement and verbal framing of photographs (wire source, sorting by photo size, discourse analysis of headline/caption). The high frequency of photographs depicting civilian life meant that the majority of news photographs in both publications received at least one framing code, particularly 'empathy/humanization'. The allocation of frames admittedly adds a more subjective level of analysis than the content analysis. However, the criteria for the connections made between photographic elements and frames are overt and clearly defined. The criteria are also applied consistently between the two newspapers, and so comparative results can arguably be of value even if the extent to which a frame can be explained photographically is queried. This approach, in which the content analysis provides clear measures and values in support for the interpretive moves made in the admittedly more subjective framing stage of analysis, fulfils the criteria of providing an 'explicit, codified and public' research process (King et al., 1994: 8; see also Robinson, 2002: 138-9).
I would also point to some concerns and limitations of the study. In addition to the framing analysis, a comprehensive approach could include more detailed scrutiny of certain representative photographs. There is a danger of becoming too removed from the original images via over-fragmentation of the photographic details through visual semi-otic analysis. Their emotional pull should not be discounted. Aggregated numbers alone cannot tell us about the significance of a photograph, but it is also dangerous to talk about the impact of any single image in a wide-ranging manner. Findings from detailed content and framing analysis could therefore be supplemented with more qualitative 'stylistic' methods of analysis (Hall, 1975: 15). This small study analysed only two newspapers in a single country - it would be valuable to compare it with press samples from countries which differ in geographical distance and cultural proximity. Useful studies that take a comparative approach to press visuals include Fahmy and Kim's content analysis of the Iraq invasion in the US and UK press, comparing The New York Times and The Guardian (2008), and Fahmy's (2007) analysis of the 'Saddam statue toppling' event that examined 43 newspapers from 30 countries. The structured nature of the content and framing analysis developed here allows for cross-country comparisons and replicability.
I also reiterate that this is not a study of the entire coverage, and its scope is not intended to cover the entirety of each newspaper's textual coverage, rather to focus on the photographs used and their immediate context. Ideally, framing studies such as this would benefit from interviews with photographers and other people responsible for the selection of such images and considerations of decency and suitability. However, the present study offers refinement of a technique which incorporates serious analysis of the news photograph and its context, including the formation of meaning through the interaction of images with words. In this sense, the study builds on scholarship of visual framing analysis as a methodology for examining press photography.
Parry
83
Notes
1 Although also traditionally known as 'broadsheet' newspapers, both The Times and The Guardian had changed their format at the time of this study period; The Times as a tabloid-size newspaper and The Guardian with its prize-winning Berliner format.
2 The codebook with full listings of the coding schema, variables and values is available from the author. A Microsoft Access database was used to record data and conduct queries and searches.
3 Whilst this level of coding is based primarily on the photographic elements, the caption can be used to identify protagonists or location. I recognize that without such factors, any coding would be very difficult to establish. However, any loaded terms or explanatory information provided in the caption should not be considered until the next stage in coding.
4 Unfortunately three newspapers were missing from The Guardian sample (30 July, 1 and 5 August) and one newspaper from The Times (27 July). The Guardian included slightly more photos than The Times, and this could have increased even further if the sample was complete. As the total sample included 36 days of coverage (13 July to 23 August 2006, excluding Sundays), the missing data should not significantly affect findings. The average number of photos relevant to the conflict per day was 3.4 for The Guardian, and 2.8 for The Times.
5 A current research project that I am conducting will establish if this is the case with British coverage of the 2003 Iraq invasion, across seven national newspapers.
6 Peter Nicholls of The Times also won the 'photo essay' award for his pictures of Lebanon, indicating the high quality of the photographs in the two selected newspapers (see http://www. theppy.com/content/results2007).
Acknowledgements
This research was funded by the AHRC as part of a doctoral grant. I am very grateful to Farida Vis, Julia Hallam, John Corner and reviewers for their comments on earlier drafts.
References
Arpan, L. M. et al. (2006) 'News Coverage of Social Protests and the Effects of Photographs and
Prior Attitudes', Mass Communication & Society 9(1): 1-20. Becker, H. S. (1995) 'Visual Sociology, Documentary Photography, and Photojournalism: It's
(Almost) All a Matter of Context', Visual Sociology 10(1-2): 5-14. Bissell,K. L. (2000) 'AReturnto "Mr. Gates": Photography and Objectivity\ Newspaper Research
Journal 21(3): 81-93. Burgin, V. (1982) Thinking Photography. London: Macmillan.
Domke, D., D. D. Perlmutter and M. Spratt (2002) 'The Primes of Our Times? An Examination of
the "Power" of Visual Images', Journalism 3(2): 131-59. Entman, R. M. (1993) 'Toward Clarification of a Fractured Paradigm', Journal of Communication
43(4): 51-8.
Entman, R. M. (2004) Projections of Power: Framing News, Public Opinion, and US Foreign
Policy. Chicago: University of Chicago Press. Fahmy, S. (2004) 'Picturing Afghan Women: A Content Analysis of AP Wire Photographs during the
Taliban Regime and after the Fall of the Taliban Regime', Gazette: The International Journal
for Communication Studies 66(2): 91-112. Fahmy, S. (2005a) 'US Photojournalists' and Photo Editors' Attitudes and Perceptions: Visual
Coverage of 9/11 and the Afghan War', Visual Communication Quarterly 12(3-4): 146-63.
84
Media,War& Conflict 3(1)
Fahmy, S. (2005b) 'Emerging Alternatives or Traditional News Gates? Which News Sources Were
Used to Picture the 9/11 Attack and the Afghan War?', Gazette: The International Journal for
Communication Studies 67(5): 381-98. Fahmy, S. (2007) '"They Took It Down": Exploring Determinants of Visual Reporting in
the Toppling of the Saddam Statue in National and International Newspapers', Mass
Communication & Society 10(2): 143-70. Fahmy, S. and D. Kim (2008) 'Picturing the Iraq War: Constructing the Image of War in the British
and U.S. Press', Gazette: International Communication Gazette 70(6): 443-62. Galtung, J. and M. H. Ruge (1965) 'The Structure of Foreign News', Journal for Peace Research
2(1): 64-91.
Gamson, W. A. (1989) 'News as Framing: Comments on Graber', American Behavioral Scientist 33(2): 157-61.
Gitlin, T. (1980) The Whole World Is Watching: Mass Media in the Making and Unmaking of the
New Left. Berkeley: University of California. Griffm, M. (2004) 'Picturing America's "War on Terrorism" in Afghanistan and Iraq', Journalism
5(4): 381-402.
Hall, S. (1975) 'Introduction', inA. C. H. Smith (ed.) Paper Voices: The Popular Press and Social
Change. London: Chatto & Windus. Hallin, D. C. (1986) The Uncensored War: The Media and Vietnam. Oxford: Oxford University
Press.
Hansen, A. et al. (1998) Mass Communication Research Methods. New York: New York University Press.
King, C. and P. M. Lester (2005) 'Photographic Coverage during the Persian Gulf and Iraqi Wars in Three U.S. Newspapers', Journalism & Mass Communication Quarterly 82(3): 623-37.
King, G., R. O. Keohane and S. Verba (1994) Designing Social Inquiry. Princeton, NJ: Princeton University Press.
Knightley, P. (2003) The First Casualty: The War Correspondent as Hero, Propagandist and Myth-Maker from the Crimea to Iraq. London: André Deutsch.
Kress, G. and T. van Leeuwen (1996) Reading Images: The Grammar of Visual Design. London: Routledge.
McGreal, C. (2006) 'Capture of Soldiers Was "Act of War" says Israel', The Guardian, 13 July: front page.
Meiselas, S. (1987) 'The Frailty of the Frame, Work in Progress: A Conversation with Fred
Ritchin', Aperture 108(Fall): 31-41. Messaris, P. and L. Abraham (2001) 'The Role of Images in Framing News Stories', in S. Reese et
al. (eds) Framing Public Life, pp. 215-26. Mahwah, NJ: Lawrence Erlbaum. Perlmutter, D. D. (1998) Photojournalism and Foreign Policy: Icons of Outrage in International
Crises. Westport, CT: Praeger Series in Political Communication. Perlmutter, D. D. and G. L. Wagner (2004) 'The Anatomy of a Photojournalistic Icon:
Marginalization of Dissent in the Selection and Framing of "A Death in Genoa'", Visual
Communication 3(1): 91-108. Petersen, T. (2005) 'Testing Visual Signals in Representative Surveys', International Journal of
Public Opinion Research 17(4): 456-72. Robinson, P. (2002) The CNN Effect: The Myth of News, Foreign Policy and Intervention. London:
Routledge.
Parry
85
Rose, G. (2001) Visual Methodologies. London: Sage.
Schwalbe, C. B. (2006) 'Remembering Our Shared Past: Visually Framing the Iraq War on US News Websites', Journal of Computer-Mediated Communication 12(1). URL (consulted Nov. 2006): http://jcmc.indiana.edu/voll2/issuel/schwalbe.html
Sontag, S. (1979) On Photography. Harmondsworth: Penguin.
Sontag, S. (2003) Regarding the Pain of Others. London: Penguin.
Taylor, J. (1998) Body Horror: Photojournalism, Catastrophe and War. Manchester: Manchester University Press.
Taylor, J. (2000) 'Problems in Photojournalism: Realism, the Nature of News and the Humanitarian
Narrative'', Journalism Studies 1(1): 129-43. Tuchman, G. (1978) Making News: A Study in the Construction of Reality. New York: Free Press. Wolfsfeld, G. (1997) Media and Political Conflict: News from the Middle East. Cambridge:
Cambridge University Press.
Biographical note
Katy Parry has recently completed her PhD thesis at the University of Liverpool and is now working on a new AHRC-funded project, studying 'Political Culture and Media Genre' with Kay Richardson and John Corner at the same institution. Her research is primarily concerned with the use of photography in the UK press depiction of war, with a particular interest in repositioning the concerns of political communication to include news photography at the forefront of investigations. Katy is co-author (with Piers Robinson, Peter Goddard, Craig Murray and Philip M. Taylor) of Pockets of Resistance, a book on British media coverage of the Iraq war to be published in 2010 by Manchester University Press. Address: Department of Communication and Media, University of Liverpool, Roxby Building, Chatham Street, Liverpool L69 7ZT, UK. [email: katy.parry@liv.ac.uk]
Copyright of Media, War & Conflict is the property of Annenberg School for Communication and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.