The language of objectivity: Reuters’ internal editorial discussions on terminology in the Arab–Israeli conflict,1967–1982

Objectivity in British journalism: The BBC and Reuters

In his comprehensive survey of the literature on journalistic objectivity, Maras (2013) argues that news objectivity cannot be simplified into a transnational standard synonymous to impartiality or neutrality. Boudana (2011, 2016) has argued that objectivity, essentially the professional commitment to provide accurate reporting, is certainly not the same as neutrality, when defined narrowly as not taking sides, or impartiality, when defined as an effort to balance both sides of a story.

Others have also contested the view that objectivity is a commonly shared professional standard, showing instead that its acceptability and translation into practice depend on national journalistic cultures. Schudson (2001), for instance, considers that scholars may have too quickly concluded that British 20th century journalism endorsed the American ideal of objectivity, whereas it may actually ‘be a kind of half-way house’ between American and more partisan European journalistic traditions. Hampton (2008) and Sambrook (2012) state that although often used interchangeably, the term ‘impartiality’ has been used more often in Britain than ‘objectivity’.

The public commitment in Britain to ‘impartiality’ has been most notably identified with the BBC. Established in the 1920s under Royal Charter, the BBC treated ‘impartiality’ as ‘an obligation’ from its inception. Adopting a neutral language was a key element of this obligation, and a BBC 1949 staff handbook included a list of potentially offensive labels and expressions to be avoided (Hampton, 2008: 481–482). In its 2015 Editorial Guidelines, the BBC reaffirms the principles of ‘accuracy and impartiality’ and appropriate ‘use of language’, especially in the section on ‘War, Terror and Emergencies’.

Although a global news agency, Reuters, established in Britain in 1851, has been influenced by British journalistic traditions, and it also eventually promoted impartiality. However, unlike the BBC, its motivations were less public service–oriented than commercial: ‘in the case of Reuters this ideal resulted from both a semiofficial identification with the nation and the desire to provide raw news to partisan newspapers of all leanings’ (Hampton, 2008: 478). In a 1921 Bulletin, Reuters considers that, unlike newspapers, it should refrain from expressing opinions: ‘It is a tradition with us that we have no sentiments of our own, that we have no policy and no likes and dislikes. We are the impartial recorders of facts and events’ (Bulletin, 1921: 34, quoted in Hampton, 2008: 478–479).

Another motivation for impartiality was the difficulty of news agencies to answer public criticism: ‘A newspaper has far greater licence and it can always defend itself from attacks by rejoinders in its own columns. REUTER, on the other hand, is, to a great extent, defenceless’ (Bulletin, 1921: 4, quoted in Hampton, 2008: 479). This defensive position was evident in reporting the Middle East in the period under discussion. In a 1970 note, chief news editor Ranald Maclurkin insisted that ‘the World Desk here in London has its orders to scrutinize Middle East copy more carefully than material from any other part of the world to avoid unobjectivity creeping in’. He added that mutual Arab and Israeli complaints served ‘to show the sensitivities we meet from both sides doing our best to produce an honest and completely impartial news service’ (quoted in Goodman, 2016). As will be seen, the editorial correspondence of Reuters researched in this study contains numerous similar discussions of ‘objectivity’ and ‘impartiality’, often used interchangeably, mostly to defend Reuters from accusations of partiality by governments or subscribers.

Language of objectivity: Archival exploration of a news organization’s performance

To avoid appearance of side-taking, news service necessitates neutral language. But, as critical theorists argue, ‘the power to signify is not a neutral force in society’ (Hall, 1982: 137). Language is the site of social–political struggles over meaning, which the media contribute to enforce. Notably examining BBC and Reuters’ coverage of conflicts, Barkho’s (2013) edited book, dedicated to the theoretical and practical intricacies of ‘impartiality’, calls for more research on the question of language:

there is a need for impartiality and objectivity that seeks to reveal how language can be used as a tool to hide bias, and at the same time how critical language investigations can help bring to surface the hidden and sometimes dubious values of our accounts of others. (p. 6)

While media scholars have engaged in such ‘critical language investigations’, they have proposed tools to measure objectivity (or lack thereof) by default, that is via bias detection, and by focusing on the outcome, thus ignoring the process of objectivity ‘manufacture’. Certainly, many studies have shifted interest from news content to journalistic work and working environment and have produced research based on interviews and ethnographic observations (notably, Boyer and Hannerz, 2006; Reich, 2008; Velthuis, 2015). But these studies may result in bias inherent in direct contact of researchers with human subjects of research (Chenail, 2011) and do not allow evaluation of objectivity at organizational level and with historical perspective. Indeed, they are circumscribed to individual journalists (those interviewed) and to recent news items (those the interviewees remember so they can account for decisions made about them).

More fundamentally, these studies do not confront the methodological challenge described by Latour (1999) as ‘blackboxing’, that is ‘the way scientific and technical work is made invisible by its own success’ (p. 304). For instance, Horvit (2006) determined that Reuters had been more neutral than other Western news agencies in covering the pre-war debate over Iraq in early 2003. However, her content analysis does not consider the editorial process that might have lead to this outcome – a product of competing forces, tensions and contradictions.

To resolve this methodological challenge and scrutinize the ‘blackboxed’ editorial process itself, we studied internal reports, notes and letters retained in the Reuters Group Archive (RGA). Their outstanding value is that they record in real-time the editorial difficulties faced by Reuters and efforts to overcome them. Historians and communication scholars have already made wide use of this extensive archive to explore the commercial development of the Reuters news agency in the 19th and 20th century, its evolving relationship with the British government and empire, and its present-day services as a global news provider (Ilan, 2013; Putnis and McCallum, 2013; Read, 1999; Silberstein-Loeb, 2014). Our research field focused on a dozen voluminous files of editorial correspondence in the 1960s and 1970s between the London headquarters and its offices and correspondents in Egypt, Lebanon, Jordan, Syria and Israel. Using the qualitative tools of historical enquiry, we searched the documents for all recorded discussions of lexical issues relating to objectivity and impartiality and brought them into our own analysis. The limits of the material are of course that only discussions preserved in writing are available for analysis. However, these were numerous enough to shed invaluable light on the news agency’s lexical policy and tensions in that period. We also examined for comparative purposes another dozen editorial files relating to other prominent conflicts Reuters was covering, as well as British foreign office files held at the National Archives of the United Kingdom (NAUK), relating to Middle East coverage of Reuters or the BBC in that period.

The effect of Arab–Israeli conflict on Reuters news organization, 1967–1982

Immediately after the Israeli victory in the 1967 war brought extensive Arab territories under Israeli military occupation, the conflict acquired the global media presence it holds to this day. Israel became a much more important international news centre than previously, evident in the late 1960s in the striking expansion of the regular ‘foreign’ press corps, including agencies such as Reuters and its rivals (Goodman, 2016). Geopolitical relations between Israel and neighbouring Arab states, Egypt and Syria, often attracted great interest. This peaked during the 1973 war and during the protracted peace negotiations between Egypt and Israel in the late 1970s. However, it was the post-1967 rise of the Palestinian national movement and its independent armed struggle against Israel that generated most headlines, reflected in increasing importance of Reuters’ office in Beirut as news provider. Much Palestinian activity was in open defiance of Arab governments, plunging Jordan into a bloody civil war in 1970 and Lebanon from the mid-1970s. Palestinian attacks on European soil, especially hijacking of aeroplanes, significantly added to the intensity of global media attention (Morris, 1999; Sayigh, 1997).

Heightened interest and controversy over the conflict challenged the Reuters news agency organizationally and editorially in maintaining its reputation for objectivity. Endless complaints from Arab subscribers about real or perceived ‘bias’ of Arye Wallenstein, the veteran head of Reuters in Israel, brought the agency’s management to establish in 1969 a permanent post for a British ‘expat’ correspondent in charge of reporting from Israel. After Wallenstein’s sudden death in 1975, no longer would an Israeli staffer be chief of Reuters in Israel (Goodman, 2016). Similar changes occurred in Beirut. The veteran head of Reuter’s regional centre, Palestinian Elias Na’was, was replaced in 1969 by Ian Macdowall, a rising star in Reuters. After 2 years in Beirut, during which Israeli subscribers repeatedly complained about his alleged pro-Palestinian ‘bias’, Macdowall returned to London and was appointed chief news editor, responsible for another decade for reporting from the Middle East (Goodman, 2016). Shortly before he died in 1991, Macdowall revised for publication the handbook of journalism he had compiled internally to increase ‘the clarity, precision and consistency’ of reporting by Reuters (Macdowall, 1992). Much of this is about choice of terms and its impact still apparent in the present-day Reuters manual (Reuters, 2008).

Editorializing and emotive language

Editorializing and choice of wording had to be homogenous and decided at corporate level, not left to journalists’ discretion. This follows the Anglo-American conception of journalism as team work (Reich and Boudana, 2013).

A lengthy editorial memo in the mid-1950s stated the ‘primary task’ of Reuters was ‘to provide its subscribers with facts – swiftly, accurately, clearly’, and ‘all sides of a controversy must be fairly covered’. Facts had to be ‘thoroughly checked’ but ‘sourcing’ was the crucial reporting duty, ensuring Reuters was not blamed for false or misleading news and maintained its reputation for accuracy and impartiality. Anything not an ‘incontrovertible fact’ or a ‘matter of personal observation’ by a Reuters’ correspondent had to be strictly ‘attributed to the source from which it comes’ (Reuters, 1955).

Related to repeated emphasis on ‘facts’ and ‘sourcing’ was another professional obligation stressed in internal editorial directives: awareness of language. Resorts to verbs such as ‘admit’, ‘point out’, ‘emphasize’ or ‘stress’ were all emphatically banned from the news file – unless quoted – so Reuters would not be associated ‘with a point of view of the speaker being reported’. Desk editors were warned that ‘interpretive’ verbs such as ‘accused’, ‘denounced’ or ‘condemned’ required ‘careful handling’ and the simple word ‘said’ was usually ‘adequate’. Also banned were phrases like ‘slashing attack’, ‘strong protest’, ‘striking speech’ and ‘powerful plea’, whose ‘editorializing’ nature also threatened ‘objective reporting’ (Campbell, 1958).

Another general term in Reuters for overly sensational or seemingly partial words was ‘emotive language’. This was raised in September 1970, after a Palestinian group hijacked four American and European passenger airliners and flew them to Jordan and Cairo. An attempt on an El Al plane failed. One hijacker arrested, Leila Khaled, had already shot to international media fame in the previous year after hijacking a plane to Damascus. The Reuters office in Tel Aviv received an angry phone call from Ted Lurie, editor of Jerusalem Post, complaining that in decades of relying on Reuters, he had not seen news that ‘angered him so’, expressing concern about Reuters’ ability to supply international news ‘free of propaganda’. His indignation was about a profile of Khaled sent by Macdowall from Beirut (Lurie, 1970; Wallenstein, 1970c). Macdowall had described Khaled as ‘a beautiful girl guerilla’, ‘a fiery revolutionary’ and ‘a shapely brunette’, whose ‘spectacular hijacking’ of a plane in 1969 made her ‘an instant heroine among the Palestinian people’ (Macdowall, 1970). Reviewing the Israeli complaint, the chief news editor Michael Charvet conceded that ‘the general tone of writing’ did imply that Reuters ‘sympathises with her activities’, though the profile contained ‘no fact that cannot be substantiated’. Charvet agreed, ‘Khalid is a good looker; is a revolutionary, is a heroine to the Palestinians who are refugees and is on record as saying she enjoys hijacking airliners’. Nevertheless, in an editorial message sent to all Reuters desks, Charvet stated that ‘the profile as a whole’ with ‘its use of emotive words’ and calling Khaled ‘by her first name’ gave readers the impression that ‘the writer rather admires her’. Charvet asked correspondents to ‘redouble’ their efforts ‘to steer a calm, detached course’ for Reuters ‘under the assault of bitter and often hysterical partisanship from all sides in the Middle East crisis’ (Charvet, 1970a, 1970b).

The crisis in Jordan soon developed into the long-expected showdown between King Hussein’s regime and Palestinian organizations controlling large slices of his kingdom, supported by Syria. Wallenstein reported to Reuters headquarters in London of further criticism from Israeli editors that Reuters reports were being written ‘in a vein of sympathy’ towards the Palestinian armed organizations in the Jordanian crisis (Wallenstein, 1970d, 1970e). In a confidential letter to the editor-in-chief of Reuters, Brian Horton, Wallenstein pointed to the ‘numerous sins’ and ‘straying’ from ‘the path of objectivity and impartiality’ of the Beirut office during the Jordanian crisis, highlighting in his opinion what could only ‘appear to be a consistent attempt to back a cause, that of the Arab guerilla groups’. One of Wallenstein’s complaints was a ‘choice of words’ betraying sympathies of the writer, such as ‘unleashed’, or ‘ruthlessness’, to describe the Jordanian army’s attacks. Wallenstein contrasted this with the sympathetic depiction of the Palestinian side: the unexpected ‘courage’ with which ‘young Fedayeen’ fought, the ‘skilful’ use of ‘guerilla radio broadcasts’ and the assertion that ‘Palestine Guerillas tonight spurned a Jordanian offer’ (Wallenstein, 1970f).

In fact, reporting of the Reuters teams in Amman and Beirut during the Jordanian crisis won tributes from the Reuters desk and media subscribers, constantly beating other news agencies. Macdowall, in particular, was personally commended for orchestrating the operation, dispatching no less than 200,000 words over a fortnight (Reuters News Letter, September 1970, 1–13; October 1972, 28). Horton believed the overall standard of reporting was very high and that the Beirut office endeavoured to keep a proper balance while ‘deluged by claims and counter claims in a fast developing major story’. He admonished Wallenstein for ‘petty’ objections to use from Beirut of words like ‘unleashed, spurned and others’. Horton agreed that these were the kind of ‘over-dramatised journalese’ that Reuters has ‘always discouraged’ but though deplorable ‘from the comfort of our armchairs’, they were excusable occasionally ‘in all fairness’ under the tremendous, tiring and unrelenting pressure of work in Beirut and Amman during the crisis (Horton, 1970a).

Nevertheless, Horton raised with Macdowall, during a meeting one month later, ‘the problems of emotive language’ in the Beirut news report. Horton noted that Macdowall showed ‘some sensitivity in discussing the issue’ but seemed to accept Horton’s suggestion ‘that Beirut probably became too involved in the story and that the main trouble was caused by the occasional injudicious choice of words, particularly adjectives’ (Horton, 1970b).

Defining and naming borders and capitals

While ‘editorializing’ or ‘emotive language’ were general linguistic offences, some words created specific problems for being politically charged, or having become so, especially in the international environment in which Reuters reported and from which it derived its profits. An editorial note on ‘objectivity’ in May 1956, reminded that ‘Reuters readers world-wide represent many points of view’ and ‘words’ that were ‘harmless in one context’ could ‘give offence in others’ (Imeson, 1956).

A linguistic pitfall for Reuters particularly in the Middle East was geographical. The general international Reuters policy regarding place names was to follow officialdom which, in the international context, meant the United Nations (UN). The UN recognized that the term ‘Persian Gulf’ was known in the Arab world and only referred to on ‘patriotic’ grounds as the ‘Arabian Gulf’. As Elias Na’was, chief of Reuters office in Beirut, told his staff in March 1967, the solution found with London was using ‘Arabian Gulf’ in regional news distribution in Arabic and ‘Persian Gulf’ in copy filed in English to Reuters in London, unless ‘Arabian Gulf’ was quoted directly, in which case, it had to be followed by ‘Persian Gulf’ in brackets (Na’was, 1967).

After the 1967 war, Reuters was confronted with even greater geographical challenges in defining the boundaries of ‘Israel’ and the territories it now occupied. In April 1968, the Reuters office in Cairo alerted senior editors to reports from Tel Aviv and other international Reuters offices which referring to Israeli attacks across the Jordan ‘into Jordanian territory’, seemed to treat the occupied West Bank as ‘Israeli territory’, and ‘the Jordan river as a frontier of Israel’ (Little, 1968). Besides wrongly inferring that the occupied territories were an internationally recognized part of Israel, the Arabs gravely disliked it. Macdowall too discovered this shortly after first arriving in Beirut in summer 1969 and he asked for ‘alarm bells’ to ring on the World Desk whenever ‘Israel’s borders’ appeared describing what were ‘ceasefire lines’ (Macdowall, 1969). The chief news editor assured him the World Desk had been warned several times and that another editorial warning note had been issued (Maclurkin, 1969).

This remained a potentially sore issue, watched closely by governments. In August 1984, a feature by Reuters correspondent in Israel Ethan Bronner was issued by Reuters with the dateline ‘Allenby Bridge, Israel-Jordan border’. The Reuters correspondent in Amman, Ibrahim Noori, was summoned to the Jordanian Ministry of Information and informed the dateline was ‘a serious error of fact, history and geography and injured the dignity of Jordan’. Correction was demanded and Reuters issued a note to subscribers (Duriaud, 1984).

Israelis too had geographical sensitivities and one related to Jerusalem as the claimed capital of Israel. According to UN decisions, Jerusalem was to be ruled under international jurisdiction. Before June 1967, few foreign governments recognized Israel’s claim and, after the war, hardly any accepted the Israeli annexation of East Jerusalem. However, the phrase ‘the Tel Aviv Government’ was mostly associated with the Arabs and their supporters. Wallenstein believed its insertion by a sub-editor on the London desk in March 1970, to paraphrase an Israeli ministerial statement, was ‘inexcusable’ (Wallenstein, 1970b). An editorial instruction to all desks reminded that ‘surely everyone knows that the seat of the Israeli Government is Jerusalem, not Tel Aviv’ (Maclurkin, 1970a). A year later, Wallenstein warned again that phrases like ‘Tel Aviv’s response’ were ‘offensive to Israeli editors’. One severely criticized its use in a Reuters report (Wallenstein, 1971).

Occasional references to Jerusalem as Israel’s capital were equally likely to raise Arab displeasure and the chief news editor instructed ‘to avoid dispute’ by not labelling ‘either city’ as Israel’s capital and use the name of the country, ‘Israel’, as Wallenstein had suggested (Charvet, 1971a, 1971b, 1971c, 1971d). In January 1974, another note to all desks and correspondents concluded:

Tel Aviv is not the capital of Israel, and Jerusalem’s status is in contention. We should not use the name of either city as a synonym for Israel (as in ‘the Jerusalem Government’). When describing East Jerusalem we must use impartial phrases like ‘captured from Jordan in 1967’. (Carter, 1974a)

Geographical difficulties were not unique to Middle East coverage. Reuters encountered similar problems on the other side of the world, when naming a government according to its geographical location. In July 1973, all Reuters offices in Asia were instructed to avoid ‘unnecessarily upsetting’ the South Vietnamese government by referring to it not in direct quotes as ‘the Saigon government’ or ‘the Saigon administration’, Viet Cong terminology which the South Vietnamese were ‘sensitive’ about. By the same token, it added, ‘the North Vietnamese government’ should also be referred to as such (Hart, 1973). Soon the war in Vietnam would end with the complete demise of South Vietnam and the turning of Saigon into Ho Chi Minh City, in Reuters reports too. In the Middle East, however, the ongoing conflict continued to keep editors on their toes even as to the difference between ‘captured’ and ‘re-captured’. In June 1977, Macdowall exploded at Israeli staffer Hugh Orgel’s reference to ‘recapture’ of East Jerusalem in 1967. The chief news editor instructed he be told ‘this kind of Israeli propaganda has no place on our file’ (Macdowall, 1977b).

Reference to the term ‘Palestinian’

Editorial use of ‘Palestinian’ posed another challenge caused, after the Arab state’s debacle in the 1967 war, by the re-emergence of international discussion concerning Palestinian national identity and territorial rights. In June 1969, Prime Minister Golda Meir publicly argued in interviews with the Anglo-American press that the Palestinian people never existed in the past as such (Giles, 1969). Understanding the damage from this position, the Israeli foreign office instructed its representatives on the existence of a ‘Palestinian people’ if they wished to define themselves so, but demanded a solution to their national aspirations be found elsewhere, mainly in Jordan (Ben-Horin, 1969). For their part, both the Palestinian armed organizations and their reluctant Jordanian hosts pressured the British embassy in Amman in June 1970 to impress on the BBC to refer to the ‘Arab guerrillas’ as ‘Palestinians’, or at least ‘Palestinian Arabs’. By describing them as ‘Arab’, the BBC was accused of toeing ‘the Israeli line that “there are no Palestinians”’ and it was biased for the BBC to call someone ‘a “Palestinian refugee” only until he takes up arms to recover his homeland, when he ceases to be deemed “Palestinian”’ (Strachan, 1970).

By 1972, with Macdowall as chief news editor, the Reuters editorial management had adopted this position. In May that year, a Sabena airliner on its way to Israel was hijacked by members of the Black September group, who were shot dead at Lod airport by Israeli commandos. The Reuters report from Israel describing the hijackers as ‘Arab guerillas’ rather than ‘Palestinian’ was subsequently queried by head-office. The chief reporter in Tel Aviv, Michael Arkus, explained it was impossible to know during the incident whether they were Palestinian or, as in previous incidents, members of other Arab nationalities. Therefore it was better ‘to use the general word Arab guerilla rather than risk making a mistake’. While conceding this was ‘a difficult one’, the senior editor Nick Carter argued that it implicated ‘all Arabs’, and that no non-Arabs objected being called ‘Palestinian guerillas’ as ‘all of them work for the Palestinian cause’ (Arkus, 1972; Carter, 1972a, 1972b). The Reuters editorial management seemed particularly sensitive to this issue, compared to its leading Anglo-American news agency rival (Associated Press, 1972). However, by asking Arkus for ‘the real objection to the word Palestinian’, Carter probably hinted that the Israeli official line or the influence of Wallenstein were behind avoidance of the term (Carter, 1972b).

Macdowall continued to impress the matter on subsequent chief correspondents in Israel. In May 1975, he inquired with Wallenstein’s British replacement, John Chadwick, whether there was ‘any good reason why Tel Aviv should follow Israeli practice in referring to Arab guerillas instead of the more precise Palestinian guerillas?’ (Macdowall, 1975). In March 1977, Macdowall likened a report from Tel Aviv referring to the Popular Front for the Liberation of Palestine (PFLP) as an ‘Arab guerrilla organisation’ to describing Euskadi Ta Askatasuna (ETA) as a ‘Spanish’ rather than ‘Basque’ organization. He asked bureau manager Bernard Edinger to ensure use of the term ‘Palestinian’ in the future (Macdowall, 1977a). A year later pressing the Tel Aviv office to use ‘Palestinian’ rather than ‘Arab guerrillas’, Macdowall argued that the fact that other Arab nationals served in them did not make the groups ‘any less Palestinian, just as the presence of Englishmen and Welshmen does not make Scottish regiments any less Scottish’ (Macdowall, 1978).

In January 1978, in the wake of Egyptian President Sadat’s visit to Jerusalem, Macdowall informed bureau manager Pat Massey that in considering some of the ‘semantic’ problems in the Middle East, the London editorial management decided that ‘the Arab inhabitants of the West Bank’ should be ‘simply’ referred to in Reuters stories as ‘Palestinians’. However, noting earlier correspondence from Massey, the chief news editor agreed that ‘this usage from Tel Aviv could be viewed askance by the Israelis, who dispute that there is a Palestinian identity’. He therefore exempted the Tel Aviv office from this ruling, but also asked Massey to ‘avoid where possible’ the phrase ‘Palestinian Arabs’, preferring ‘a neutral phrase’ like ‘Arabs on the West Bank’ (Macdowall, 1978). That many readers would notice such fine terminological points is open to doubt. But these internal discussions show that, wishing to avoid offending any side’s sensitivities, Reuters editors generally pursued objectivity but sometimes simply ended up adapting terminology to sensibilities of local clients.

‘Terrorist’ and ‘guerrilla’ labels

Discussion of these labels epitomizes, in particular, the struggle for signification between those involved. In British-ruled Palestine during the Arab rebellion in the 1930s and the Jewish revolt in the 1940s, Reuters and other British media generally followed the authorities in describing insurgents as ‘terrorists’, or other officially encouraged criminalizing terms such as ‘bandits’, ‘gangs’, ‘thugs’ and ‘murderers’ (Carruthers, 1996). However, by the mid-1950s as decolonization advanced, Reuters management began to question the western cultural foundations of its terminology. In May 1956, an editorial directive on ‘objectivity’ from senior editor George Imeson reminded all desks that the agency’s readership was global and ‘material relating to problems of colonialism, race and so on’ needed handling with ‘special care’. Even with the best intentions it was ‘easy to slip unconsciously into usages which seem to indicate a degree of bias to people of radically different background and outlook’. Referring to the bloody colonial conflict in Algeria, Imeson wrote:

Recently, for example, we spoke of ‘rebel savagery’ in an apparently straightforward narrative story from French North Africa. Would we talk about ‘French savagery’ in similar circumstances? Apart from the label of ‘savagery’, the word ‘rebel’, which used to be quite acceptable, appears to be no longer. So we should avoid talking about ‘rebels’ in French North Africa. A more ‘respectable’ word, perhaps, is ‘insurgents’ and we can, therefore, talk about ‘Nationalist insurgents’. This will differentiate those actively in arms against the French from other Algerians who are still using constitutional means to espouse the nationalist cause.

‘Terrorist’, Imeson (1956) added, ‘is another word to avoid in most contexts’. Nevertheless, Reuters reports in the late 1950s and 1960s repeatedly referred to ‘terrorists’ attacking British security forces and civilians in colonies such as Cyprus and Aden, as well as using the label ‘terrorist’ to describe intentional attacks on unarmed civilians in conflicts in which the British were not involved, such as Algeria and Vietnam.

After the 1967 war, the term was very occasionally used without quotes in Reuters reports from Israel too, but Wallenstein was well aware of its sensitive nature and on the whole took care to portray perpetrators of armed attacks as ‘saboteurs’ (Arad, 1967), a term which his editors in London accepted, though some Palestinians resented this term too (Dajani et al., 1968). However, the terms ‘guerillas’, ‘commandos’ and ‘fedayeen’ (those who sacrifice themselves) were soon often used by Reuters and other western news organizations to describe the Palestinian organizations engaged in warfare with Israel and in increasingly bitter internal rivalries in Jordan and Lebanon. These were terms of choice for the BBC too, despite pressure from the Israeli Embassy in London to depict ‘fedayeen’ organizations operating in and from Jordan as ‘terrorists’ rather than the more positive ‘guerrillas’ (Long, 1970).

Visiting Israel in November 1969, the chief news editor Ross Maclurkin told the Tel Aviv office to stop referring to ‘Arab terrorists’ as this was ‘an unobjective word which should have no place in the Reuter file’ (Maclurkin, 1970b). However, the term continued to be argued heatedly within the agency. A few months later, after dozens of Egyptian civilians were killed by an Israeli air-strike near Cairo in February 1970, Ian Macdowall reported from Beirut a Lebanese newspaper banner-headline decrying ‘Another Deir Yassin’, explaining this referred to ‘the slaughter’ of men, women and children carried out by ‘Jewish Irgun terrorists’ in a Palestinian village in 1948. Wallenstein complained that this was contrary to the recent rule ‘so forcefully laid down’ banning the word ‘terrorist’, and beside potential harm to the reputation of Reuters, it was ‘unfair to have different rules for the same game’ (Wallenstein, 1970a). Maclurkin agreed that ‘copy from Arab centres’ should also not refer to ‘Jewish terrorists’ and promised to impress on Macdowall ‘the need to strike out emotive words or phrases’ (Maclurkin, 1970b).

To be sure, in the early 1970s, the terms ‘terror’ and ‘terrorist’ were still regularly used by the mainstream Anglo-American press to describe Palestinian attacks on Israeli civilians and especially international airline traffic. Exceptionally, it was also employed to condemn Israeli raids on civilian targets such as Beirut airport, the downing of a Libyan airliner over Sinai and Israeli bombing and shelling in Lebanon causing extensive loss of life (The Economist, 1972; The Guardian, 1968; The New York Times, 1968; The Sunday Times, 1973; The Times, 1974). Reuters would have none of this, in the Middle East or elsewhere. A note for journalistic applicants published also in the internal newsletter, pressed again that correspondents should exclude their own ‘expressions of opinion’ and avoid words implying ‘value judgments’ with the example that ‘freedom-fighters’ and ‘terrorists’ are normally both ‘guerillas’ (Carter, 1974b).

Meeting senior editors at the London head-office in September 1973, Wallenstein held it was possible to define whether ‘certain actions’ were ‘terrorist’ but agreed Reuters’ policy of avoiding words like ‘terrorist’ and ‘other emotive phrases’ was correct ‘in a situation such as that in the Middle East’ (Macdowall, 1973). Similarly, the Lebanon office was told again in 1974 that writing ‘slaughtered’ rather than ‘killed’ in another background story on Deir Yassin exposed Reuters to ‘charges of bias’ (Fenby, 1974). John Chadwick at Tel Aviv was told that ‘outrage’ was banned in reporting arrest of seven Arabs thought part of a Palestinian group that planted bombs in town-places in Israel and the west bank (Macdowall, 1975).

No less important for Reuters was to check rephrasing of its copy turned into press by its subscribers. In July 1977, the Jerusalem Post ran a Reuters story on a visit by Bernard Edinger to the Gaza prison, where he was allowed to interview inmates about recent press claims they had been tortured during interrogation. The Jerusalem Post replaced the Reuters phrase ‘Palestinian guerilla prisoners’ to ‘Palestinian terrorist prisoners’ (Edinger, 1977). After consulting a senior London editor in London, Edinger wrote a letter to the Jerusalem Post, duly published, noting that Reuters had used the term ‘guerilla’ and not ‘terrorist’ in order to maintain ‘impartiality’ (Fenby, 1977). In March 1979, the writer Dilip Hiro visiting Israel noted that the Jerusalem Post changed ‘guerilla’ into ‘terrorist strikes’ in a Reuters report from Lebanon and was told by a member of the newspaper’s staff that this was its policy (Hiro, 1979). Reading his article, the London head-office was concerned that Reuter-credited copy was ‘being significantly changed for political or policy purposes’ and asked Tel Aviv Bureau chief Pat Massey to chastise the offenders (Blackman, 1979). In more recent times, Reuters management (and also their rival Associated Press) similarly stopped Canadian subscribers from implanting the term ‘terrorist’ under the agency’s byline in reports of Palestinian armed activity following the second Intifada (Austen, 2004).

By the late 1970s, it must be stressed that Reuters’ firm rejection of the label ‘terrorist’ was not only apparent in the Arab–Israeli context but, unlike earlier decades, also in reporting Britain’s own conflicts, such as the violent backyard in Northern Ireland, where the minimizing term ‘The Troubles’ masked a conflict costing some 2000 lives that decade. In November 1980, the British Foreign Office queried Reuters use of ‘guerrilla’ to describe the Irish Republican Army (IRA). The British press presented them as ‘terrorists’ for the British audience but Reuters catered for an international market so the Foreign Office understood Reuters’ wish ‘not to take sides’. However, it argued that ‘guerrilla’ itself was ‘not a neutral term’, or accurate ‘academic’ description of the IRA, and could be ‘particularly misleading’ in the United States, where IRA propaganda presented their members as ‘romantic heroes’ of a ‘national liberation movement’ (Gordon Lenox, 1980).

Gerald Long, the General Manager of Reuters, explained in a lengthy reply the agency policy of dispassionate ‘vocabulary’. He admitted that the term ‘guerrilla’, chosen after internal deliberation concerning the IRA, was not satisfactory but ‘all other available words’ were also ‘unsatisfactory’. The word ‘rebel’, for example, was ‘a badge of honour in countries that owe their foundation to rebellion’, including the United States itself. In contrast, the term ‘insurgent’ was so little in popular use that it had ‘remained untinged with passion’. The problem, Long explained, was that ‘perhaps people are not sure what it means, and lack of precision is the price one pays for the lack of emotion’ (Long, 1980). This was a most important observation as indeed accuracy, no less than fairness, has been regarded by academics and journalists alike as one of the twin values that should guide journalistic practices (Boudana, 2016).

In any event, Long later informed the Foreign Office that, after review, use of ‘guerrilla’ would neither be forbidden nor encouraged. Reuters would try to use descriptive terms such as ‘gunmen’ or ‘attackers’ that ‘no one can quarrel with’, though even ‘gunmen’, he earlier suggested, ‘might be held to have an unhealthy tinge of Sean O’Casey romanticism to it’. Nevertheless, Long flatly rejected yet again the term ‘terrorist’ suggested by the Foreign Office as ‘too full of emotion for reports that attempt to be detached’ (Long, 1980, 1981).

It is striking that neither in this letter nor during the Lebanon war which broke out in 1982 did Reuters suggest or use in this sense the supposedly-neutral term ‘militant’, which has become so current in the 21st century.