NEWS & ANALYSIS

Where the Sunday Times went wrong

Text of the report by the Review Panel into problems at the newspaper, December 2008

Sunday Times Review Panel Report, December 2008

Paula Fray
Anton Harber
Franz Kruger
Dario Milo

Executive Summary

In September 2008, a four-person panel was appointed to review the Sunday Times' editorial processes following a number of high-profile story retractions. The panel was asked to make recommendations "to enable the Sunday Times to produce bold, incisive journalism that maintains the utmost credibility with its audience".

The panel interviewed over 80 staff members, reviewed documentation, observed editorial processes, studied international best practice and undertook four case studies of problematic stories.

The panel found that the paper's stature and a string of successes in previous years had led to complacency, and that over time some policies, practices and structures had become inadequate.

The organisational structure had become top-heavy and thin on the frontlines of news generation, with a lack of clarity on roles and responsibilities, particularly among the proliferation of senior editorial managers. There was no chief sub-editor -regarded as a vanguard for quality control - and there was a need to strengthen the office of the Managing Editor.

There was a need for resources to be redirected to the start of the process - story generation. The weekly Johannesburg newsroom diary meeting was seldom happening, and too many decisions were being deferred to a large conference.

Rewriting by seniors was raised as a problem, with many expressing unhappiness with the process of rewriting a lead story into a "splash". Sub-editors and junior reporters felt disempowered and marginalised in this process.

Internal communication was seriously inadequate. Greater communication is required between managers and general staffers, as well as Johannesburg and the bureaus. Staff development was also identified as an essential issue - not just in terms of induction processes but also ongoing training that would help minimise potential errors and - most importantly - the rewrite process.

There was general dissatisfaction and lack of clarity regarding the expectations and rewards of the company's performance management system. The key performance assessments (KPAs) were identified as a potential deterrent to getting exclusives as reporters stuck to predictable stories.

The panel considered various editorial policies - such as the Policy on Gifts and Freebies - and found a need for them to be reviewed, updated and for gaps to be filled. Additional policies are needed on plagiarism, corrections and apologies; privacy; the use of unconventional methods of reporting; and the handling of taste, decency and violence.

The Sunday Times uses an Accuracy Check form which is designed to ensure reporters meet accuracy and fairness requirements. There is, however, no consistency in the use of the form and few consequences for not using it, or for getting facts wrong. The disciplinary process was described as uneven and this appears to have impacted on morale.

There is agreement that the form is a useful tool - but also criticism that it had been reduced to a bureaucratic formality. It needs to be supplemented with a more comprehensive system of quality-monitoring, with firm checkpoints and clear responsibilities for signing off on stories.

The panel believes the paper's approach to sourcing is a crucial flaw in current Sunday Times policy and practice. There is an urgent need to return to the forms of attribution typical of journalism.

The Sunday Times has a set process for stories that have potential legal issues, yet neither the Transnet nor the first Land Bank stories were sent for external legal advice. There was general concern about the lack of ongoing training in respect of legal (and other) issues.

The response to stories which elicited complaints also points to a lack of policy on complaints and corrections, with different views in the various departments.

The panel believes that secrecy around investigations should be kept to an absolute minimum. When secrecy is required, a small team of senior people should be assigned early in the process to guide the project.

The scope of this report was limited to the news operation of the Sunday Times. The perception exists that The Times and Times Online have drained the weekly's resources, but this view was challenged. The panel found real frustrations related to the demands for online content.

The panel noted that while there were many criticisms of the systems and processes, the editor-in-chief emerged as well-respected and well-placed to lead the newspaper out of the current situation.

The panel has made several recommendations, key of which are:

• The appointment of a Public Editor responsible for investigating complaints as well as engaging Sunday Times readers and the general public;

• A review of the Sunday Times staffing structure with greater emphasis on news generation, the clarification of roles and responsibilities, the strengthening of the Managing Editor's office and the reinstatement of position of chief sub-editor;

• A more rigorous internal communication process;

• The implementation of a structured induction process, as well as proactive newsroom-based training programmes;

• A review of existing editorial policies and an update to cover new policies;

• The Accuracy Check form should be reviewed and incorporated into a more comprehensive system to ensure editorial integrity.

The panel believes there should be open debate on these recommendations, as well as a structured review of their implementation.

Chapter 1: Introduction

Within a few weeks in 2008, the Sunday Times retracted and apologised for a number of high-profile stories, leading to concern that the newspaper's credibility and standing was being damaged. Most notable of these were:

• "How fat cats looted bank billions" (11 November 2007) and "Land Bank boss ‘must be charged'" (20 January 2008), which was the subject of two adverse Press Ombudsman's rulings, run prominently on 24 August 2008; and

• "Transnet sold our sea to foreigners" (24 August 2008), followed by a front page retraction (7 September 2008).

In the wake of these stories, a number of adverse public comments were made about the Sunday Times, denigrating its credibility. These stories came in the wake of others which had also been controversial, and raised questions about the newspaper's professionalism and accuracy. Editor-in-Chief Mondli Makhanya wrote in a column on 7 September 2008 that the paper had been "under fire from various quarters in recent weeks" because of these stories and needed to "maintain the intimate trust relationship we have with our readers".

To do this, the paper would embark on "a process of reviewing the way we do our journalism, and strengthening our verification and authentication mechanisms". He promised an "honest, critical look at ourselves" and said a panel of experts would be commissioned to help with this process.

The four individuals on this panel were approached to fulfil this role on what became known as the Sunday Times Review Panel:

• Paula Fray, Regional Director of Inter Press Service (IPS) and former editor of the Saturday Star;

• Anton Harber, Caxton Professor of Journalism, Wits University, and former editor of the Mail and Guardian;

• Franz Kruger, Senior Lecturer in Journalism, Wits University, former national editor of SABC radio news, author of Black, White and Grey: Ethics in South African Journalism; and

• Dario Milo, media lawyer at leading firm Webber Wentzel and author of Defamation and Freedom of Speech.

At an initial meeting with the editor, a brief was agreed, and that brief has defined the scope and parameters of this report (see Appendix A). The brief set out the purpose of the panel as follows: "To review the systems and processes at the Sunday Times to gain an understanding of how recent stories which fell short of standards of journalistic excellence were printed." The panel was asked to make recommendations for future action "to enable the Sunday Times to produce bold, incisive journalism that maintains the utmost credibility with its audience". For us this means breaking strong news stories of public interest which command the attention of readers, are well told and accurate.

The panel was specifically asked to:

• Identify shortfalls in the editorial process from conceptualisation to publication;

• Review the use of the Accuracy Check and other quality control systems in place;

• Review specific case studies;

• Review post-publication processes for complaints;

• Review relevant policies such as sourcing; and

• Consider any other issues relevant to the credibility of the newspaper.

The panel's report would be presented to the editor-in-chief and the panel undertook to make itself available to present its findings to staff.

This report is the outcome of that brief. It takes the following form:

• The Executive Summary provides an overview of the report;

• Chapter 1, this introduction, sets out our task and methodology, the scope and limitations of our work and the spirit in which we have done it;

• Chapter 2 summarises the issues which emerged in interviews and documentation, under the headings Structure of the Editorial Staff, Morale and Communication, Policies, Editorial Processes, Human Resources, and Post-Publication Issues;

• Chapter 3 sets out four case studies we chose to focus on;

• Chapter 4 surveys relevant international best practice on the issues outlined above; and

• Chapter 5 contains our Conclusions and Recommendations.

The Spirit of this Report

In his column in which he announced this panel, Makhanya said that the paper was "not about to give up its tradition of bold reporting and laying out the truth in its most naked form". The paper's reports had to be "irrefutable", but at the same time "we do not want to be held back from unearthing nefarious activities".

This is the spirit in which this report was prepared. Our brief charged us with enabling the Sunday Times to continue to produce bold, incisive journalism, but in a way that maintains the newspaper's credibility with its audience.

We have striven for this balance. It would be all too easy to propose measures that would promote a culture of excessive caution, which may ensure there are fewer errors, but which stifles the newspaper's reporting and softens its editorial voice. A much harder purpose is to encourage care and attention to detail and professionalism, while still promoting the kind of journalism which has made the Sunday Times one of the country's most important newspapers. The measures we have considered and proposed are designed to ensure the paper can rebuild its credibility without closing the important space it provides for strong and impactful journalism.

The Sunday Times occupies an important place in South African journalism and politics. As the largest Sunday newspaper, it has an unmatched reach and influence. When the Sunday Times hurts, South African journalism hurts, and so may the country's democracy. The paper has built credibility and standing over many years, and a reputation for high-quality and incisive journalism. In this report we have set out to shore up these foundations.

We are not looking to point fingers or attack individuals. As our brief states, our focus has been on policies, processes and systems, and how these can be improved to provide a strong base for journalism of the highest professional standard.

We are conscious of the fact that there are those who would use the paper's recent problems to damage it and encourage it to temper its voice. We have no truck with such a view. Indeed, our aim is to strengthen the newspaper and its bold voice by ensuring it is built on a base of impeccable credibility and professionalism.

The Sunday Times staff are justifiably proud of the many excellent stories they have produced, which far outnumber those which have raised problems. Our focus, however, must be on the problem stories, as these undermine the force and impact of the others.

Methodology

The panel's approach to its task had five elements:

a. Interviews

In a meeting with staff on 9 September 2008, the panel outlined its brief and invited everyone to present us their views individually or collectively over the coming weeks. All were asked to attend a series of group discussions, in which we engaged with staff based in Gauteng at all levels over the following two weeks. We met with groups loosely made up of junior and mid-level reporters, senior writers and columnists, editors and section heads and sub-editors and the news desk. Panel members travelled to Durban and Cape Town to hold similar meetings with bureaux bureau staff there.

Staff were provided with an e-mail address for the panel and invited to send any information or views which might be useful. Two staff members took advantage of this invitation.

This was followed by a number of individual interviews, including with the publisher, editor-in-chief, deputy editor, political editor, managing editor, some of the deputy managing editors, senior reporters involved in the stories we were examining and the paper's external legal advisor. We also engaged with some individuals who asked to see us alone.

Over 80 staff members were interviewed. Two of those we requested to interview were unavailable for personal reasons: investigations editor Jocelyn Maker, and senior sub-editor Claire Robertson. We also failed to secure an interview with Megan Power, the other author of our case study "Mbeki took R30-m", despite repeated attempts. We offered assurances that we did not wish to know confidential sources, nor would we push her to answer questions she found ethically difficult to answer, to no avail. We were told she was reluctant to see us without her colleague Jocelyn Maker. This was most unfortunate, as it supported a perception among staff that some seniors at the paper were "untouchable" and this was why they were able to get copy into the paper without the normal scrutiny.

b. Documentation

The panel requested and received a range of documentation, including:

• Policy documents - a Staff Policy Guide which includes policies on gifts and freebies, and which deals with issues of race, religion and cultural difference; copies of the Press Ombudsman's Code of Conduct; and the newspaper's Accuracy Check;

• A sample of completed Accuracy Check forms (from Gauteng and Durban);

• The recent Climate Survey conducted at the newspaper;

• A staffing organogram;

• Copies of all versions of stories to be used in our case studies (reporter's version, sub-editor's version, final version, etc), where available, as well as copies of all versions of other stories randomly selected;

• Copies of all corrections carried in the newspaper for the previous three months;

• Copies of relevant newspapers; and

• The company's Performance Assessment template.

c. Observation

Members of the panel observed editorial conferences and some spent time watching the news desk operate on deadline.

d. International best practice

Many of the issues the panel has considered are not unique to the Sunday Times or South Africa and we found strong parallels in the experiences of international newspapers from which we could draw pointers to international best practice. We drew from, inter alia: the New York Times' 2003 investigation into the Jayson Blair scandal; reports of the Poynter Institute in Florida, USA, notably from their Journalism Without Scandal conference of 2003; the Associated Press Managing Editors (APME) book Building Trust in the News; and the Organisation of Newspaper Ombudsmen. For editorial policies on issues such as corrections, the panel drew on the policies of a number of respected journalism institutions, including the New York Times and the Canadian Broadcasting Corporation (CBC).

e. Case studies

The panel set out to identify stories which could serve as case studies for the examination of the causes of problems. The choice was widely canvassed in interviews and discussion before the following were chosen:

• The Land Bank story: "How fat cats looted Land Bank billions" (11 November 2007, front page), "Land Bank boss ‘must be charged'" (20 January 2008, lead, page 15), and the Press Ombudsman's ruling, "Land Bank reports ruled a breach of the Press Code" (24 August 2008, front page).

• "Transnet sold our sea to foreigners" (24 August 2008, front page), followed by a retraction (7 September 2008, front page);

• "Mbeki took R30-million and gave some to Zuma" (3 August 2008, front page), and the editorial "The arms-deal truth must out, once and for all" (page 20); and

• "Tito's niece, 13, throws a stylish bash" (7 September 2008, lead, page 3).

The first two choices presented themselves, as these reports were the immediate cause of the panel's creation. The Mbeki story was chosen because a number of Sunday Times staff expressed concern with the treatment of the story. It became clear in discussions that this report highlighted a number of important issues which the panel needed to deal with. The Mboweni story was picked because it was not a front-page splash (lead story) and, therefore, did not go through the standard treatment of such a story. It also raised a different set of questions, namely, how the paper handles ethical issues, notably those involving privacy.

Limitations of this report

For practical reasons, it was necessary to limit the scope of this report. Although the Sunday Times operation is intertwined with the new daily paper, The Times, and the paper's online presence at www.thetimes.co.za, this report has focused on the Sunday paper. The panel could not ignore the overlap between these products and their newsrooms, but considered the daily and online products only insofar as they impacted on the Sunday Times.

No survey of readers or other stakeholders was envisaged. Research was limited to the views of the staff and management of the Sunday Times. Although we had at least one approach from an outside interest group, the panel decided it could not canvass outside opinions without massively widening the scope of our work and extending the time-frame considerably.

Disclaimer: Franz Kruger sat on the Press Council's Appeal Panel for one of the Land Bank complaints, and therefore recused himself from discussions of this case study.  

Chapter 2: The issues which emerged

In our interviews, there were certain issues which were raised repeatedly by a range of staffers. In this chapter, we summarise and analyse those we considered most pertinent to our brief.

Culture

The Sunday Times holds a special place in South Africa: its front pages have enormous impact on the national discussion, setting the agenda for the week and often longer. Over the years, it has produced some ground-breaking journalism. Several stories have initially been denied strenuously, only to be justified in the long run. Examples cited by Makhanya were the report that a rape charge had been laid against the ANC's then deputy president, Jacob Zuma, and the report that then minister in the presidency, Essop Pahad, had put enormous pressure on the parliamentary finance committee to drop the arms deal inquiry.

These successes, Makhanya told us, had led to a certain arrogance. "We got complacent about how good we were. We can't get it wrong," he said. We heard similar sentiments from several other members of staff. Arrogance of this kind can be dangerous, and perhaps this is the single greatest factor causing the recent missteps.

There is also a measure of arrogance in the paper's attitude to the rest of the SA media. It's only a story if it has been in the Sunday Times, the approach seems to be. This has made it possible for reporters to pitch stories that have been run in other media. This is sometimes completely legitimate: a skilled journalist may see the potential in a small mention in another newspaper. But it can lead to the newspaper running old stories as if they are new. One example is provided by the posters on the SABC story in the Sunday Times of 23 November 2008. While the story's angle - the possible choice of Zwelakhe Sisulu to chair a new board - was news, the posters declared that the ANC was to fire the existing board. That intention had been reported and extensively discussed for some months.  14

The newspaper has developed a particular editorial style in which potential front page "splashes" are rewritten and re-angled to fit that style. We discuss the editorial approach in more detail below, but believe that its style has sometimes come to be used in a way that is out of touch with the paper's audiences and with the mainstream of journalism. This sometimes also leads to inaccuracies being unwittingly introduced.

We will recommend that the paper launches a concerted campaign to open its doors to its audience, using a range of communication techniques to develop a discussion with its readers about what it is doing. This is in line with practice on major newspapers worldwide.

The paper's unique position in South Africa has another consequence, which in turn contributes to some unfortunate trends.

For many journalists, a job at the Sunday Times is the high point of their career. The newspaper rarely advertises vacancies: it head-hunts people it regards as outstanding. As a result, there is comparatively little movement in the staff complement. Once you've got there, you stay. The organisation is dominated by people who have been there for a very long time - decades in some cases. This in turn breeds a certain cliquishness, which many staffers described to us as problematic.

It has also created problems for career-pathing, in the sense that staff run out of places to go. Some interviewees told us they felt that some promotions had been made simply because people were bored in their old positions, without the newspaper having a clear role for them to move into. We will discuss the structure in more detail below, but it does seem as if some positions have been created around people. The better practice is for the newspaper to identify the need for a particular role, and then find the right person to fill it.

There is an unusually high level of family and personal relationships between staff. Some are related, others are married, or involved with, or divorced from each other,  15

or a host of other variants. Human nature is what it is, and most workplaces struggle with some degree of this phenomenon. Some organisations find it necessary to develop policy that prevents people who are involved with each other from working in the same section, or at least reporting to each other. We simply wish to point out that this is another factor affecting the workplace atmosphere.

We encountered a high degree of unhappiness in the interviews we conducted. We will discuss morale in more detail below, but we believe strongly that it has had a major impact on the problems the paper has experienced. If people are unhappy, they don't work effectively.

Of course, journalists are well known as grumblers, and we were conscious of the need to try to work out how much of what we heard was normal workplace complaining, and how much should be discounted as driven by office-political agendas. We did our best to sift carefully.

One strong positive should be mentioned: There were all kinds of accusations hurled at a range of people, but we heard hardly any criticisms of the editor. Makhanya is clearly a well-respected figure, setting a tone of collegiality and professionalism. We believe that he is insufficiently supported, both on the editorial and on the management side. But we believe that he is well-placed to lead the newspaper out of the current situation.

Structure

The Sunday Times news production is described as a "very big operation" which has evolved to accommodate changing newsroom needs and dynamics. In particular, appointments have been made in line with the vision and needs of editors over the past few years.

While reflecting a dynamic culture, the evolving structure of the news production unit means reporting lines have been blurred and positions skewed in regard to actual needs.  16

This has impacted on the unit's ability to function optimally. During the interviews, the panel heard:

• "One person can't keep a very strict eye on everything that happens - there are occasions when people don't know who they are reporting to exactly";

• "The newsroom is top-heavy - there are a lot more editors than people actually doing reporting";

• "There are four times as many deputy and associate editors as we had five or six years ago. This has impacted on the number of people who are journalists/writers";

• "The news desk is small - the desk is big - but a lot of other people get involved in the story";

• "The news desk is a lonely place. There are too many backseat drivers and this puts us under immense pressure";

• "The subs (sub-editors) room is separate, but reporters and subs need to interact more";

• "We have a lot of experience which we are not using."

An organogram was prepared following a request from the panel. In some cases, it was not possible to determine roles and responsibilities of positions, with many responsibilities being shared across management. Many of the staff members shown the organogram had not yet seen it.

The organogram also reflects a centralisation of reporting - despite the many management positions, a number of positions report directly to the editor-in-chief. These include the deputy managing editors (DMEs) for politics, big projects and investigations, as well as the sports and news review editors.

The newsroom is headed up by a deputy managing editor: news, with bureau chiefs in Durban and Cape Town and an acting news editor in Johannesburg. The deputy editor oversees the DMEs for news, features, production, legal/campaigns, as well as the foreign editor.  17

The subbing process is overseen by the DME: production, who reports to the deputy editor. There is no chief sub-editor for the main body, but the managing editor is supported by a design editor, a chief sub-editor: Business Times; and two deputy production editors.

Linked to the number of editors is a lack of clarity on roles and positions. In a number of instances, individuals' personal understanding of their roles differed from that of their line manager or peers.

The absence of a chief sub-editor was targeted as a major reason for the marginalisation of the sub-editors' pool. While the managing editor: production was often cited as being the chief sub-editor, his role was also described as that of an overseer across the various products rather than that of traditional chief sub-editor.

• "The managing editor job should be much more clearly defined than it is - critical things that normally fall into that office fall by the wayside."

The traditional role of chief sub-editor as primary quality controller of the final published product has been dispersed among various sub-editing role players - leading to greater responsibility on the news desk.

Various stakeholders identified the late arrival of potential as a primary reason for the inability to thoroughly check content ahead of publication. As reflected elsewhere, the news-gathering process is light at the start and heavy at the end.

The national news editor receives assistance from editors on the desk who have other primary responsibilities, and who then channel stories when they have not been involved in briefing or debriefing the reporters. While this assistance for the news desk is much-needed, the introduction of new editors into checking processes can increase the potential for errors - particularly in rewriting.

Greater resources at the start of the process - story ideas brainstorming, briefing and debriefing - would enable a more substantial start to the editing week. This requires additional support for the news editor at the outset.  18

There was a clear call to redirect resources:

• "We need to reduce the number (of editors and move them) to where resources are needed in the newsroom."

• "I think the stories that appeared are the result of not having sufficient people on the hard end of journalism and too many people on the other end. There are not enough breaking stories; stories don't arrive on time and then you're making decisions on the page one splash late in the day ... (We) need to look at how resources are allocated so that unproductive areas have resources reallocated."

It became clear in our interviews that Business Times was seen as a particularly problematic section of the paper. The issues raised included:

• Staff shrinkage. We were told that there were now four reporters, whereas there had been eight only a few years ago.

• Additional load. This is the one section of the paper which takes responsibility for the daily edition's business coverage as well as the Sunday's, and this has put excessive strain on the department. Reporters said they were now unable to get to their Sunday stories before Friday and this was having an impact on the quality of their output.

• Some Business Times reporters refused to administer the Accuracy Check on principle, creating an inconsistency in policy across the newspaper.

• There was tension between the Business Times editor and his staff, with many of the latter feeling that their editor, as a newcomer to business reporting, had less expertise than those who reported to him (This editor has since been replaced).

Business Times is an important part of the newspaper and needs to be edited by someone with knowledge and experience in this area. It is clear that resource issues need to be addressed to ensure that the pressure of the daily paper does not continue to compromise the quality of the weekly.  19

Impact of The Times

There were differing views on the impact of the launch of the sister daily paper, The Times. While the editor-in-chief liaises with the daily on diary issues, there is no sharing of diaries between the two publications.

However, the perception exists that The Times has drained the weekly's resources. Some individuals, previously key to the Sunday paper, were moved across. "The pressure on resources is immense. Since the emergence of the daily the quality of the weekly has dropped."

For the most part, the daily paper has recruited its own team. As we have indicated, Business Times is one department which has to service both the daily and the weekly needs, and staffers were adamant that it has negatively affected the Sunday paper. Since this is seen as a model for how the two papers relate to each other, it needs careful consideration.

Reporters are particularly frustrated by the demands for online content: "We are expected to produce for online when we don't have the equipment." Some complained of inadequate training to produce multimedia for online.

Morale/Communication

Morale at the Sunday Times was directly impacted by the events under review. It was compounded by the limited communication back to all staff. While managers directly involved in the process were clear on what had happened, there was a consistent complaint in the newsrooms and bureaus that they were generally kept in the dark.

Although newsroom and management meetings were scheduled for Johannesburg, staff reported that they were infrequent. Some noted that communication with readers was also limited.

There was a general call for greater communication:

• "There is no debrief. The only feedback is on the website with the comments from readers."

• "No one knows what anyone else is doing or who they report to ... who is accountable to whom or for what."

• "We heard that Paul (Stober) is going and that he is being replaced, but we still have not been told."

• "We have a source drought. We have to call in favours because people don't think we're credible."

• "Morale is at an all-time low."

It was noted that communication was not as good as it should be between the bureaus and Johannesburg. And there needs to be stronger feedback. At the moment there is no post-mortem of the paper.

There is a fundamental lack of communication in terms of management conveying messages to more junior staff members. In bureaus, there was an expressed concern that their bureau chiefs were not being included in conference meetings and were unable to motivate stories or get newsroom information first-hand. One manager told us that information was being conveyed to the seniors, but did not appear to be reaching other staff.

According to the managing editor, issues are communicated through newsroom meetings and, when necessary, management might be called in directly to speak to staff. However, these meetings address only staff available in the newsroom and additional methods of communication - email, newsletters, etc. - must be entrenched to ensure that messages are communicated to all members.

It should be noted that the bureaus have regular meetings to discuss the paper and diary. However, communication with Johannesburg was not regular and increased the perception that these branches were marginalised.

Communication channels were unclear and use of the intranet limited.

News diary sharing was pinpointed as the source of the "silo" effect that limited sharing of knowledge and ideas. It was noted that the Transnet story was not shown to Business Times ahead of publication, when such input could have helped tell the story in a more coherent and accurate way.

The panel believes communication processes need to be enhanced using all the available tools at newsroom management's disposal. These include - but need not be limited to - regular formal meetings, e-mailed briefs to all newsroom staffers, intranet postings and face-to-face communication.

Management

The Panel review follows a company-wide climate survey in which staff expressed concern about several management issues. These included:

• Need for management training;

• Career planning for staffers;

• Independent grievance process;

• Review of the performance management system and allocation of merit increases;

• Improved communication, including reinstatement of key meetings;

• Improved relationship between company management and the newsroom floor; and

• The addressing of complaints related to equipment;

Many of these issues were still being raised by staff in our discussions.

Performance Management

Performance indicators were raised repeatedly by staff, with general dissatisfaction on their implementation in regard to additional indicators for work on the online site.

There was a general perception that the new key performance assessments (KPAs) were introduced "in a hurry" and without proper consultation. Reporters believe the financial incentive for contributions to the daily and online editions was misleading.  22

Certainly, we found these incentives hard to understand and it was clear that staff did not understand it.

There remains a general lack of clarity regarding the Performance Assessments and related financial incentives. In fact, many reporters indicated they were willing to forego financial incentives for certain performance indicators:

• "The Accuracy Check doesn't create the right culture. It has become an administrative nightmare. It's a bureaucratic process for performance assessment - its rubbish - but it's in our KPAs. I'm prepared to lose those points."

• "We don't have equipment to do the multimedia that is part of our job description."

The KPAs were also identified as a deterrent to getting "exclusives":

• "It is encouraging mass production rather than quality ... we've lost focus on getting those splashes because we don't invest enough time in people working on the splash."

In short, performance assessments were described as "unevenly applied", "problematic" and dependent on the manager.

Reporters in the regions also expressed concern that it was easier to meet the KPA requirements for page leads and splashes in Johannesburg when stories were assigned.

Some reporters were cynical that the KPA requirements outweighed the required annual number of stories for the Sunday Times and that this reflected a focus on quantity rather than quality.

While the relations between effort, performance achieved and rewards play an essential role in staff motivation, it should also be noted that if the required performance seems unrealistic, performance-related rewards will not have impact.  23

The lack of clarity about the KPAs, as well as the perception that some KPAs cannot be achieved, has therefore limited the value of the performance management system. In reviewing the KPAs, effort should be made to emphasise breaking news and exclusivity.

Disciplinary Action

The disciplinary process was described as uneven and this appears to have impacted on morale. There was a general perception that disciplinary action was seldom taken against reporters who failed to complete an accuracy check or who had errors in a story.

• "It's a general issue there are not enough consequences when things go wrong";

• "Juniors have to take responsibility but the more senior you are, the less the consequences."

The panel found that the Accuracy Check, for example, had different levels of compliance from full compliance to ad hoc and, in some departments, minimal.

The lack of enforcement reinforces the perception that the Accuracy Check is a bureaucratic requirement.

"Getting it wrong needs to matter much more," said one reporter.

Staffing

While there is general agreement that more staff is needed at the frontlines of news production, there is a debate as to whether the Sunday Times needs additional people or a reallocation of resources.

Staff were also concerned - and attributed this to the reallocation of resources to The Times - that there was an ongoing freeze on posts.

At various times, there were indications that the newsroom needed at least three or four additional reporters split between mid-level and senior skills.

Staff Development: Induction and Training

An induction process performs far more than simply an introductory function. It sets the tone for subsequent attitudes, commitment and motivation and, critically, is the start of the training and development of new recruits. In doing so, it reduces uncertainties, addresses anxieties which can impact on productivity and offers support to new entrants.

An induction process needs to provide a sequence of planned activities that allows newsroom staffers to acquire the necessary skills and knowledge that will enhance their capacity as Sunday Times journalists.

It should range from knowledge of Sunday Times processes and policies to elements unique to the paper such as:

• How to write a Sunday Times story;

• Identifying splash ideas; and

• Understanding the Sunday Times reader.

The Sunday Times appears to have a well-established administrative induction process as evidenced by the Sunday Times Policy guidelines, but we heard of many cases where there had been no induction or it was haphazardly implemented. The induction process needs to go much further in order to recognise the needs of the individual.

For this reason, it is recommended that the induction process be outlined by the Managing Editor and monitored and implemented by the appropriate line manager.

And, this must recognise the needs of the individual. Effective "socialisation" of well-chosen staff has long-term financial benefits for the organisation. The training of Sunday Times newsroom staff must be viewed as an investment.  

The Sunday Times has a skills plan prepared annually by the Managing Editor. The panel was not able to review the annual training audit or plan, but feedback indicated a largely reactive approach to what training was available rather than what was needed. Training that is linked to KPAs and is measured and evaluated by line managers would have a direct impact on the newsroom. For example, a reporter sent on a crime reporting workshop should be required to write crime stories that reflect new learning in the short term.

Various skills gaps were identified during our review process. The newsroom was described as having a "skills challenge" while reporters, in turn, called for greater management training (including communication skills): "Journalists are appointed to management without any proper training."

During the period under review, the news desk attended a Poynter Institute news management course.

It should be noted that a number of stories from senior writers are processed without major changes to angle or flow. However, the panel observed rewriting "into a Sunday Times" story of several articles. Such rewrites - and re-angling of stories on deadline after conference discussions - could be minimised through in-house writing training that focuses on the elements that make a Sunday Times story unique. Greater clarity is needed on what these elements entail.

Limitations on the news desk hamper the ability to mentor and coach reporters although the panel did observe news desk incumbents processing reports in consultation with reporters.

Training must therefore be ongoing - with annual reviews of changes, new challenges and editorial focuses.

Editorial Policies

We were provided with the Sunday Times Staff Policy Guide, which consists of:

• A section describing the newspaper's history and ownership;

• A detailed "Policy on Gifts and Freebies";

• A "Code of Conduct for Sunday Times Staff" in dealing with issues of race, religion and cultural difference", with an addendum to the reporting checklist;

• "Extracts from the Press ombudsman's Code of Conduct"; and

• The "Sunday Times Accuracy Check".

We are not aware of any other written policy documents.

It is very striking that the policies seemed generally unknown to staff. We showed the document to participants in many of the group discussions we held, and rarely found people who had seen the document. The exceptions were the accuracy check and the policy on freebies, which will be discussed in more detail below.

The policy on gifts and freebies

This is a very comprehensive policy, setting out in some detail how the newspaper's journalists are expected to avoid conflicts of interest. The policy sets out that gifts are to be handed in, then auctioned off and the proceeds given to charity. In fact, it is an excellent approach, and is rightly held up as an example to other media houses.

Nevertheless, the policy should be reviewed. In some respects, it is unclear. For instance, the provision on "CDs, tapes and books" says these may not be retained without payment, and then immediately says a book reviewer is allowed to keep the review copy. This should be clarified. It is also doubtful that the newspaper pays publishers for review copies, nor is this strictly necessary.

In some other respects, the policy seems unrealistic: For instance, the requirement to declare "any shares bought or investments made" seems too wide: it might include even a car, painting or unit trust. What is missing, on the other hand, is a bar on writing about shares one owns. The policy's effectiveness is undermined by provisions that are too wide to enforce.

The policy might make some use of the practice of disclosing benefits to readers.

Aspects of the policy seem not to be observed: it makes provision for an "Editor's Register" where certain kinds of benefits are to be declared. However, we found no sign of that register.

Some core features of the policy - such as the duty to hand in freebies, which are then auctioned off to benefit charity - were well known by staff. But many other details were not known.

Code of conduct on issues of race, religion and cultural difference

This seems to be a useful and clear statement of policy, although it may need to be looked at afresh in the context of an overall review of policy.

Press Ombudsman's Code

The version provided in this document needs to be updated: there have been several changes to the code since this policy document was compiled. It is also not clear why staff should have only extracts of the code, rather than the whole document.

Accuracy check

For some years, the newspaper has used an Accuracy Check form, which reporters are obliged to fill in for every story they file. It is designed to ensure they have met the requirements to ensure the accuracy and fairness of their story and they are meant to go through it with a peer. This is a system which has drawn much praise and is at the backbone of the newspaper's drive for accuracy.

The panel was provided with two different versions of this Accuracy Check form. The first version is a one-page form headed "Sunday Times Accuracy Check on Report". The second is a two-page form provided to us with the Sunday Times Staff Guide, which does not include the "Background" section and contains more detail than the first. For instance, it contains more detail under the "Content" and "Legals" headings than version 1. Version 2 appears to be the one that is employed, as confirmed by our analysis of sample accuracy checks for 2007.

We discussed this process with reporters and seniors, and reviewed completed forms from the last three years. It was clear to us there is no consistency in the use of the form. Some reporters stated that it was "filed and then forgotten about", there were time constraints in completing it, and it was not useful because "it is blind to tone". There were complaints that some reporters appeared to be exempt from the check.

Reporters on Business Times indicated that they no longer used the Accuracy Check form. It also appears that it is not completed by Lifestyle journalists, nor is it always employed by reporters in the politics department. Some senior reporters stated that they would not do the accuracy check, and did not find it helpful at all.

On the other hand, many members of staff - especially heads of department - were of the view that the Accuracy Check served an important and useful function, though it was conceded that "some departments are more conscientious than others". Some expressed the view that the accuracy check comes too late in the process; it should not wait for the article to be on-page.

Our examination of the completed forms confirmed the view that implementation was haphazard. Only a small proportion of the stories in the paper appear to have gone through this Accuracy Check system. It appeared to be implemented efficiently by the bureaus, but only by sections of the Gauteng newsroom, most notably those working for the Metro before that section was closed down earlier this year. Of those forms filled in, many showed meticulous care for detail, while some were done hurriedly and incompletely. We found that an occasional one was used to record the concerns of reporters that changes were being made in the editing, as if to distance themselves from the final product. It is notable that these were filed away without consideration of the contents.

It was also clear to us that in some cases the Accuracy Check form had been reduced to a bureaucratic formality: boxes were ticked off hurriedly to meet the obligation for the check, without the reporter paying much attention. We were also told that this sometimes replaced the standard interrogation of copy by an editor. Editors would ask if the Accuracy Check had been completed rather than ask the questions themselves, and this allowed material to slip through with less scrutiny than required. We should emphasise that we also learnt of many cases where the check had been done rigorously and usefully, and saw forms where the checker had clearly picked up errors in names and titles, but we were struck by the danger that the form can provide a bureaucratic substitute for the editorial interrogation of a story.

We were told that a reporter could face disciplinary action if there were queries on a story and it was found that she or he had not filled in the form, but nobody knew of any case where this had happened.

This system was the subject of vigorous debate in our panel. It was clear that it was not being implemented rigorously, except in pockets of the newspaper, and was at best only partly effective, having failed to pick up problems in the stories which led to this inquiry. But we were of the view - supported by international best practice - that it can be a useful and important tool for improving accuracy.

Our conclusion is that the Accuracy Check Form is useful if it is recognised to be just one such tool and does not substitute for the direct questioning and scrutiny of stories by seniors, which is an essential and irreplaceable part of the editing process. It is a valuable mnemonic for reporters for what they need to ask themselves before they submit their reports, but should not be seen as more than this. Much more important is what happens after the reporter has ticked off these questions and submits the story. The story needs to go through a series of checkpoints in which individuals with clear responsibilities take responsibility for ensuring not just that the story is "Sunday Times" style, but is accurate, fair and balanced.

In the section below on editorial process, we go through some of the problems with the current checking systems and make recommendations on how this can be improved, including a more comprehensive system of checkpoints.

Sourcing

No written policy seems to exist on sourcing and people gave different accounts of the newspaper's approach. On the one hand, we were told that the newspaper made sure that readers knew where information came from. Instances where this did not happen were put down to mistakes or bad practice prevalent in certain departments.

Other interviewees said reporters were encouraged to state things as fact if they were sure of themselves. This was seen as being stronger, more authoritative.

Our examination of the newspaper showed clearly that this is the approach very often adopted. Where there is sourcing, it is often tucked away deep into the story, a very brief reference that could easily be missed.

We accept that overly laborious sourcing can get in the way of "telling the story". But this practice leaves the newspaper open to various problems. For one thing, when something is stated without attribution, it is made on the newspaper's authority. If it turns out to be wrong, it is the newspaper's own credibility that is damaged.

It is particularly risky to make predictions without attribution, as when the Sunday Times reported as fact that Baleka Mbete would be President after the removal of Thabo Mbeki (21 September 2008). Of course, the prediction turned out to be wrong. It would have been much safer to attribute the statement to wherever it came from, even if the source could not be named directly.

Also, it is simply more honest to show the reader where particular aspects of a story come from. It enables readers to assess for themselves how much weight to attach to the information.

There are also significant legal risks that arise where information is not attributed.

The current practice makes it easy for reporters - under deadline and competitive pressure - to persuade themselves that they are sure enough of something to state it with certainty, when more careful scrutiny might show the evidence to be insufficient.  31

We feel that this lackadaisical approach to sourcing is a crucial flaw in current Sunday Times policy and practice. There is an urgent need to return to the forms of attribution typical of journalism.

There is also a lack of clear policy on the use of anonymous sources. Most people we asked about this articulated the principles becoming generally accepted: that such sources should be treated with caution and only used when absolutely necessary; that attempts should be made to persuade people to go on the record; that there was a need for particularly careful corroboration, etc. However, we feel it would be useful to flesh out these points and set them out in a written policy.

It has become more clearly understood how much power there is in anonymous leaks, and how careful newspapers have to be to avoid being duped into furthering some agenda or other. Many newspapers now require reporters to get authorisation from a senior editor to use an unnamed source, as recommended by the South African National Editors' Forum guidelines on confidential briefings. The Sunday Times needs to establish a practice that ensures similar levels of caution with sources.

A related area that could do with clarification is how far reporters need to go to seek comment from somebody who is subject to critical reporting. Are there circumstances where a right of reply need not be offered? These kinds of questions also need to take into account the relevant legal implications of not offering a right of reply.

Finally, the paper should stop the practice of giving a byline to reporters who are simply pulling together material from other sources. This is not a common practice, but it happens often enough to warrant attention. Typically, the stories revolve around big international stars in the entertainment world, and it is perfectly reasonable for the newspaper to rely on published material. But a byline should indicate that the writer has researched the material, not just cut and pasted it from other sources.  32

Missing policies

Besides the question of sourcing, there are several other areas where the absence of policies is very noticeable. The following is not an exhaustive list, but areas we have identified include:

Plagiarism: although the handling of a couple of incidents show the newspaper to be clear and tough on the issue, it would be useful to spell out the paper's approach. This would be particularly useful for columnists and other outside contributors.

Accuracy: this should spell out the consequences to staff of mistakes.

Corrections and apologies: this will be discussed in more detail below.

Privacy: the Sunday Times has always presented its readers with generous quantities of entertaining reporting on the lives of the rich and famous - and sometimes the less famous. It would be useful to formulate a statement of its approach to privacy. Does absolutely anything go? If not, where are the limits beyond which privacy will be respected?

The use of unconventional methods of reporting

The handling of taste, decency and violence

We are not suggesting the writing of huge tomes that seek to pin down absolutely every eventuality. Short, clear and realistic statements are most useful. But it is not enough to have policies that live in people's heads: as we have found, this creates the strong possibility of different versions in different heads. A serious media organisation needs to have a set of clear policies that people across the newspaper can use and refer to.

Once these policies have been written, it is essential that they be communicated clearly to staff.

The Editorial Process

This panel looked at the editorial process that produces Sunday morning's paper, to look for any sticking points. We did this by observing various editorial meetings through the course of several weeks, watching production late on Saturday afternoon, and asking people's impressions in many discussions and interviews. What follows is our attempt to make sense of what we saw and heard.

Story management

Any newspaper will experience increased pressure as deadlines approach. For a Sunday newspaper, it is inevitable that Friday and Saturday will be busy days. However, the Sunday Times has so much last-minute pressure on a Saturday afternoon that it undermines effective editing. Time and time again, we were told that this was a key problem: stories were being finalised too late, and mistakes were slipping through as a result. Sub-editors were particularly vociferous in this regard, but almost everybody made the point. Late-breaking stories will always bring a last-minute rush, but we were told that much of the late copy was material which could have been submitted earlier in the week. Acknowledging the issue, Makhanya said many efforts had been made to push copy through earlier in the week, with insufficient success.

The solution has to lie in strengthening the editorial processes earlier in the week. Much of this work has to be done by the news desk, bureau chiefs and other section heads. We were struck by the fact that the Tuesday morning editorial meeting with reporters rarely happens. Reporters are often simply told to e-mail their story ideas. Makhanya said this was a temporary gap, and ascribed it to people newly appointed to the desk finding their feet.

It is clear, though, that a key function of the news desk is to manage reporters' ideas, from the very beginning onwards. This means finding mechanisms for brainstorming, and cultivating an atmosphere that encourages original thinking. It means giving constructive and useful feedback that helps focus ideas. Of course, it also sometimes means making it clear that an idea won't work.

Planning is crucial: news editors need to keep track of ideas, to make sure that the week does not start with a blank slate. We were told about "splash lekgotlas" held to identify possible lead stories. And it is clear that in areas like investigations, the time horizon is sometimes measured in months. However, medium-term planning is also critical. It is important to keep a close eye on ideas aimed at publication two or three weeks hence.

Our general impression was that news managers were relatively disempowered. Far more authority rests with conference, which will be discussed in more detail below. We were told more than once that reporters would have to wait for a go-ahead from conference before working on a story. And this decision might only be relayed late on Tuesday, after the 3:30pm meeting, or sometimes on Wednesday. Obviously, if news editors, bureau chiefs and others were able to take decisions of this kind, it would help get reporting moving earlier in the week, lessening the pressure on Saturday afternoons.

The national and Johannesburg news desk seemed to be under enormous pressure, without enough tools to do the job. We suggest that this desk, in particular, needs strengthening.

Conference

The Sunday Times has developed a particular style of operating that places enormous authority in conference. Conference finalises the diary, decides angles and story placements, chooses the splash (front page lead story) and writes major headlines. The net effect is that many editors find their decision-making constrained. We have described how news managers have to wait for conference before giving a reporter the go-ahead to work on a story. Conference also reaches deep into roles usually left to the subs desk.

One of the strongest themes to emerge from our interviews was how difficult people found conference. It is striking that these were the sentiments of senior editors. Among comments we heard were:

• "Conference on main body doesn't like to consume more than two facts at a time."

• "Conference is quite hostile - there are 15 people who all think they know better than you. Some individuals you don't want to take on."

• "It is not always a constructive discussion."

• "People at conference sometimes ask stupid questions."

• "Debates at conference can be very destructive."

• "Conference doesn't understand the story and is asking the wrong questions."

• "It hurts to have people who produce nothing who then run you down."

About the mildest adjectives used to describe conference were "tough" and "robust". Somebody described how congratulations were once exchanged in conference over a decision not to lead the paper with a particular story when the relevant reporter left the room: he had been "defeated". Another person told how they preferred to hold back their stories until the conferences later in the week, so that rivals would not have the extended time to knock down their story. In one of our meetings, we saw how quickly somebody was slapped down for saying it was sometimes difficult to contribute openly to discussion. The speaker's point was thereby neatly illustrated. This is not a constructive, collegial atmosphere.

Our own observation was that editors come to conference with insufficient clarity on angles, probably in expectation that conference will decide. Even late on Saturday, angles were changed radically without the relevant story manager engaging much in the discussion. One editor told us he was "uncomfortable" in offering a story as a splash contender: "I want conference to decide this is a strong contender." This illustrates our point that conference is usurping the authority of editors.

A large meeting of this sort is not always the best forum for effective decision-making on details, especially under time pressure. Conference should be reviewing, or overseeing, these decisions, rather than making them.

One of the discussions we witnessed at conference was strongly geared to making the strongest point possible with stories under discussion, sometimes to a point far beyond what seemed justifiable. More cautious voices were muted or absent.

The Sunday Times is known as a competitive environment: people are expected to compete for prime space, page leads, section leads and page one. A certain measure of competition is not a bad thing, but it seems that it sometimes reaches levels at the paper that are unhealthy. It is clear from some of the problematic decisions made by conference, and which led to this report, that voices of caution are being drowned out.

Asked their response to these observations, senior editors said the following: "It is true that conference plays an important role in finalizing diary. However, the news editors do have enormous influence in the way the story is pursued and presented.

"We do regard conference as a place to interrogate stories and make sure that they pass scrutiny. We try to encourage a robust discussion of all aspects of stories in order to avoid errors or holes that will rebound on the newspaper once they are published. While it may sometimes be difficult to have the stories you manage come under scrutiny, the discussion is intended to bolster the reporting, not to undermine the news editor. We would prefer to find ways to bolster the confidence of news editors rather than to do aware with the key role that conference, in our view, plays.

"We agree that there have been a number of occasions when individuals have been unacceptably abrasive or have made undermining comments. However, this is not the way conference generally operates and such behaviour does not translate into influence.

"We also feel that there is an element of self-disempowerment by some managers ... who may sometimes prefer to blame conference when the weakness in fact lies in the story pitched. The weakness may also lie in the editors' ability to make uncomfortable and unpopular decisions about the merits of a story prior to conference.

"We disagree that conference is responsible for introducing facts into stories in order to ‘splashify' them. Re-angling of stories is sometimes discussed and, when desirable, implemented by the manager in consultation with the reporter.

"News editors are never told to tell reporters to wait for conference ...

"We don't think that it is a problem for people to ask stupid questions in the sense that stupid questions are inevitable in open discussion. There may be many dead- end questions and thoughts, but through discussion clarity is achieved. There are also often many useful comments and suggestions made, and information shared.

"It is only the headline of the front-page lead that is written in conference. It is worth noting that reporters can and do challenge the headlines on their stories, although this rarely happens for the lead headline on page 1."

Having considered this important input, the panel remains of the view that consideration needs to be given to ways of re-engineering conference to ensure that voices of dissent and caution are heard, and that crucial decisions be made by those individuals empowered to do so and who have the facts at hand, rather than by committee where not everyone may be fully informed on the details of the story and responsibilities are dispersed. Conference should keep an oversight on decisions, but not supersede news and section editors.

"The helicopters"

Elsewhere, we discuss the lack of clarity about roles, particularly among some senior staffers. In the context of this discussion, it is worth pointing out the impact of this problem on the editorial process. On deadline, these staffers hover around the news desk (hence the term "helicopters" used by staff). We observed no fewer than seven editors one Saturday afternoon - and that was before the evening shift came in. Some had a clear function and were working very hard, but too many of these editors seemed only to be checking pages. As a result, more "post-facto" editing on page was done than is useful and efficient.

The "helicopters" are effectively backseat drivers, offering little support but much criticism. One frontline editor said to us: "It's a very lonely job managing a page one story."

Editing, rewriting and "sexing up" stories

The panel looked at different versions of (randomly selected) stories as they made their way through the editing process, and were struck by the extent of rewriting we saw. In several cases, two sharply different versions were generated on the news  38

desk. We do not know the particular circumstances that led to this, but do wonder whether it is efficient to have news editors doing the same job twice.

Of course, stories are legitimately subjected to editing on any newspaper. Stories, and in particular the story angles, must be sharpened. The condition is always that any rewrite must remain true to the facts. Makhanya and his senior editors were adamant that this basic rule was strictly observed. We were told that the newspaper had a particular style, and would edit copy to suit it. In many cases, we saw the news desk consulting very closely with the relevant reporter as their stories were worked on.

Nevertheless, it was striking that this did not accord with the perceptions of people further down the chain of command. Many people said there was a tendency to decide on an approach without being fully aware of the facts. The comment of one junior reporter was typical: "We have to write stories to match headlines." Another said: "Editors often come out of conference with preconceived angles."

This attitude was also reflected in what people said about the rewriting process. We were told many stories of editors writing material into stories without justification. Some examples were:

• "Six horses" became "six white horses"

• The adverb "unceremoniously" was added to a statement about Mbeki's removal from office. It was only taken out after strenuous objections by the (senior) reporter.

• "unmarried" became "divorced".

• In looking at versions of stories, we found that the second newsdesk version of the "Kerry Winter missing in Dubai" story added in the statement that the boyfriend "reasoned that if he couldn't have her no one else would". In the reporter's version, family members described him as obsessive - but there is no evidence in any version that he held this belief. Of course, the line could have been added in consultation with the reporter, but it looks very much like random pop psychology.

• Similarly, we found a schoolgirl who suddenly became a "shy" schoolgirl.

• The newspaper reported the resignation of a senior official from a major public entity as being politically motivated. This was done in the teeth of strenuous objections from the reporter, who said he had 12 sources saying the development was not political. In the end, he had his byline taken off the story. Even if he was wrong, and other people had better information, it is not a happy situation for a reporter to feel so strongly that a story of his is being distorted.

• We were told that quotes were occasionally rewritten.

• One senior writer said he preferred not to have his stories chosen as the splash since "splashifying" them often meant distortion.

• We found the Accuracy Check forms where reporters indicated their concerns about a story were overridden by "editorial decision". We do not have further details, however.

• Our case studies also illustrate the point. With regard to both the Mbeki story and the Transnet story, we took the view that a legitimate story was obscured by overwriting. These are discussed in more detail below.

Some of these examples may be dismissed as trivial matters of detail, others may rest on a misunderstanding. But it is hard to dismiss the accumulation of examples, and to disregard their impact on credibility. They reflect rewriting that crosses the line, and it is very striking that there is a high degree of unease among staff themselves about it.

The rejoinder from some editors was that it was the reporter's responsibility to sign off on the final version of the story, and that they should object to any inaccuracy. It seems to us, though, that the atmosphere is not conducive to this, and many reporters live with a degree of unease in the belief they are dealing with normal Sunday Times style.

One editor argued strongly that it should be a strict rule that, whenever possible, the rewriting should be done with the reporter present. This would have the multiple advantages of reducing errors, ensuring the reporter keeps ownership of the story, and making space for mentoring and coaching. We support this suggestion.

We believe the newspaper has persuaded itself that the set of norms it calls "Sunday Times style" remains within reasonable bounds. To most outsiders, however, this style sometimes goes further than the usual newspaper approach of highlighting the dramatic.

Subbing

The extraordinary power of conference has an impact on the subs desk too. Our meeting with the subs was dominated by complaints that they were disempowered and that that they had little influence on the newspaper. "There is a widespread sense of grievance on our floor. We are not involved, we are not asked," somebody said. A senior sub said he was so used to not being taken seriously that he only raised issues he spotted in order to cover his back. He said he was accustomed to being told not to worry, since conference had approved whatever it was he was uneasy about.

This is unusual: subs are usually very influential and provide an important check and balance on stories. The chief sub is generally, in the words of one interviewee, "the troll under the bridge".

In the case of the Sunday Times, stories are allocated to pages by the news desk, subject to conference's approval. Once they are put through by the desk, they can - and often do - go straight out for layout and copy editing. This means that the story may not be looked at by another senior editorial eye until it pops up on the page: an obvious problem with the process. Traditionally, the chief sub would play that role.

There is now no chief sub at the paper (although the organogram we were given shows a chief sub, the person shown in that post denied it was his job). The roles usually played by the chief sub have been divided between a number of people: the deputy managing editor (production), the deputy production editor and the design editor. It is likely that the withering away of this role is also a result of the power and functions undertaken by conference.

We endorse the point made by several people that the paper badly needs a chief sub. One interviewee described this as "a point of reference and authority with some distance from the news room". This cannot be a hands-off manager of production - it needs to be somebody with the skills and energy to get deeply and practically involved with copy.

It would also be helpful to clarify other roles on the subs desk. It is not clear, for instance, why the revise sub should be titled "deputy production editor".

Subs complained about a lack of communication. They wanted clearer communication about the decisions of conference, for instance, which obviously affect their work - a task that would normally be undertaken by a chief sub.

At the same time, subs said there were too many people checking pages. One said: "I have never seen so many people go through copy." In some cases, the claim was made that section editors were editing insufficiently before sending copy through, but then doing a great deal of editing on proof. Stories would be removed or radically changed at this stage, which obviously wastes the subs' time. This was due to indecisiveness at the top of some sections, it was said. In fairness, this was not something we observed first hand, however.

Accuracy and quality assurance

It was clear to the panel that what was missing was a clear path for a story and page to follow through the system, with set checkpoints where individuals have clear responsibility for ensuring the quality of the material. We propose that the system needs to be shaken up to ensure the story passes along a well-set path: from reporter to department head, chief sub-editor, a sub-editor, a revise sub-editor (or equivalents), and then be signed off by the most senior person on duty that day, or someone designated to this task by the editor. Each person needs to take responsibility for ensuring the story meets Sunday Times' requirements (in style and professionalism) and held accountable for doing so. No story should avoid any of these checkpoints. Other seniors who spot potential problems in stories or pages should always direct their comments or suggestions through the responsible person on duty, to avoid the "helicopter" phenomenon. The signing off process should be formal so that a story can be traced back along this path and individuals held accountable for approving stories which should not have passed unchanged.

Legal checking

The panel interviewed Susan Smuts, who has the internal responsibility of legal checking, and Eric van den Berg of Bell Dewar & Hall attorneys, who does the majority of the newspaper's legal work. The panel was informed that when Smuts is not able to attend to legal matters, managing editor Herbert Mabuza attends to these issues.

The panel was told by Smuts that she should see stories that raise legal issues, and that the news desk was good at identifying these stories. If Smuts cannot resolve the issue, then she contacts external advisors.

Van den Berg is briefed on an ad hoc basis by the newspaper, usually on controversial splashes, and invariably without being provided with sufficient time to study the underlying evidence that may justify the proposed article. He was not consulted at all before publication of the Transnet story, or the first Land Bank story. On the other hand, he was intimately involved in the Mbeki story, which was, however, a story that was being investigated for a number of months. He was not consulted on the page 20 leader accompanying the Mbeki story.

It is striking to us that external legal advice was not sought for the Transnet story , the first Land Bank story and the editorial that accompanied the Mbeki story; all three clearly raised legal issues and that ideally required external advisors to study the evidence backing up the claims made. (Legal advice was obtained for the main Mbeki front-page story). A group of senior editors told us that with both the Land Bank and Transnet stories, the newspaper did not believe that they were especially "novel in a legal sense" and were confident that the paper would have the defence of truth in the public interest. They also commented that the paper should have been able to work through the weaknesses in the stories without the help of lawyers. "When we are wrong on facts, there is little a lawyer can do to help."  43

We have some difficulties with the merits of this approach. Firstly, the lawyer is a fresh eye on a problematic story and would generally take a cautious approach to the verification of facts. While the paper is always free to disregard a cautious view, it remains valuable for them to hear and consider it. Secondly, a lawyer reviewing a controversial story will be able to assess the defence of truth in the public interest with reference to admissible evidence available to the paper at the time of publication. For instance, in the first Land Bank story, the forensic report that formed the basis of the claims (see below) was not available to the newspaper at the time of publication. A similar point applies to the Transnet story. In these circumstances, a defence of truth would probably not be successful as the paper might have difficulty proving it. Thirdly, even if this defence holds good, a lawyer may suggest changes which can lessen the chances of being sued without changing the substance of the story.

Also, having a system in place in which lawyers are approached for their views on potentially problematic stories which contain significant criticism of individuals or corporations points to the reasonableness of the newspaper, and would be of assistance in any legal action which may follow.

At the very least, a legal opinion can inform the editor of the risks involved in the story, and ensure the editor is fully cognisant of this when making his or her decision.

Although it is understandably seldom possible, due to time constraints, to involve external advisors to the degree where they can fully verify evidence, we recommend that special effort be made to seek pre-publication advice on controversial splashes after the conclusion of the Saturday morning conference, if not earlier.

A general concern that was expressed by journalists, some editors, and Van den Berg, was the lack of ongoing training in respect of legal (and other) issues. Such legal training as exists is ad hoc, usually reactive, and uncoordinated.

As we have already stated, the lack of training on legal (and other) issues is a material failing on the part of the newspaper. On the legal side, it is contemplated in the Accuracy Check that journalists are able to identify obvious legal issues that may arise at the outset. Myriad and complex legal issues frequently are triggered in Sunday Times stories, particularly in the contexts of court reporting and investigative stories. Journalists and editors must be able to flag these issues timeously, to allow Smuts and, if necessary, external advisors, to have a meaningful opportunity to reach a view on whether the story is publishable. For instance, in one of our case studies considered below, "Tito's niece, 13, throws a lavish bash", it appears not to have been identified by either journalists or editors that a potentially complex privacy issue was triggered; far more searching interrogation ought to have been applied to whether there was sufficient public interest in the story to justify its publication. Smuts was only asked to advise on the story after the newspaper was threatened with legal action in a telephone call from someone purporting to be an attorney representing the family.

Procedures should, therefore, be put in place for annual intensive and practical legal training for journalists and editors, on topics and with the use of case studies to be identified by the newspaper. Further, a real effort should be made to seek legal advice, with sufficient lead time to allow meaningful debate and consideration.

Post-publication: complaints and corrections

Our interviews indicated that journalists at the Sunday Times were vague on the newspaper's policy on corrections. While it was clear that, generally, Susan Smuts took the decision as to whether a correction should be published, some departments appeared to handle the issue of corrections without reference to Smuts (e.g. Travel & Food).

Many journalists and some editors expressed the view that the corrections and retractions issued by the newspaper were done on an ad hoc basis, and often poorly explained where the newspaper went wrong, leaving the public none the wiser. An instance cited in this respect was the Transnet retraction, which some said fell short because it did not explain exactly where the newspaper had gone wrong with the story. This is probably a result of the internal division as to whether an apology ought to have been published at all, but it does illustrate to the panel the need for a clear policy on corrections.

Some journalists felt there was a need for a post-mortem process on stories that were not accurate, so that they could learn from the mistakes. We discerned a definite sense that many felt that there was little if no feedback on stories that went wrong. There was also a belief among some staff that apologies were sometimes made only to avoid litigation.

Senior staff told us that this belief was unfounded and that apologies were only made when the paper got something wrong. This disconnect between the views of some staff and seniors seems to us to reinforce a need to have a written policy on corrections.

We have been provided with a sample of recent corrections published by the Sunday Times. We have noted that these corrections are generally well-drafted and explain the error made and what the correct position is.

From the sample corrections that were provided to us, it was evident that many corrections had to be made in respect of misquotes. While we have not studied the instances giving rise to these, and can, therefore, express no view as to why the mistakes were made, the need to make corrections in respect of such basic journalistic skills as correctly reflecting quotes, raises two issues.

First, the Accuracy Check is failing in circumstances where quotations are simply recorded inaccurately or mistakes are introduced in the editing process. There is no justification for editors or sub-editors materially altering quotations, but if this does occur, journalists doing the final accuracy check on-page must ensure that the correct quotation is reflected.

Second, and in common with the accuracy practices adopted by the Chicago Tribune and the New York Times, a database needs to be kept of mistakes and why they occurred. This allows for patterns of errors to be discerned, and training and other mechanisms can then be adopted to arrest these developments. The individuals concerned can be counselled and disciplined for substantial and persistent inaccuracies.

The Chicago Tribune model for generating this database is worth considering. The journalist responsible for the error has to fill in an electronic correction form describing the error, explaining how it occurred and how it came to the newspaper's attention, and suggesting how it could have been avoided. The system that is in place then characterises the error into one of a number of categories - "news-gathering error", "editing mistake", "display error" (e.g. headline or photograph error), "erroneous material from syndicates or outside news services", "simple error" (e.g. typos), and "unavoidable error" (e.g. where the source was wrong). This allows the newspaper to act on trends and patterns and has resulted in improvements in accuracy.

In summary, the Sunday Times needs a formal corrections policy indicating to whom and how readers must complain. The policy would stipulate that all corrections need to appear prominently, promptly and in a consistent place in the paper; errors should not only be corrected when complaints are made, but whenever the true position comes to the attention of the newspaper; once complaints are received, errors should be corrected promptly: if possible, in the next edition of the newspaper, and immediately the decision to correct is made, on the web; all errors - even those that appear trivial to the newspaper - should be corrected.

Even though the Sunday Times in practice achieves many of these goals, it is important that a formal policy be crafted and used as a reference point.

Secret projects

The secrecy surrounding investigative projects has caused considerable unhappiness. Some of these stories only surface on the Saturday before publication, making it difficult for them to be subjected to normal editorial scrutiny.

We accept that some projects need especially careful treatment, and we also accept that Makhanya is usually involved from an early stage, and applies very rigorous standards. However, it was clear to us that it was a cause of ill-will because of a feeling among some, including senior staff, that they were being excluded from crucial decisions. More seriously, it means that major stories may not have the benefit of adequate scrutiny by someone who has not been close to it until it is too late to effect significant changes. The "Mbeki took R30-m" story was most often cited as an example of this.

Secrecy around investigations should be kept to an absolute minimum. It should only be applied in exceptional cases. And where it is decided on, a small group of senior people could be assembled earlier in the process to guide the project, and think about how it will appear in print.  48

Chapter 3: Case Studies

Case Study I: The Land Bank Story

These reports arose out of a Cabinet announcement in November 2007 that, following a Forensic Audit Report into the Land Bank, a series of measures would be taken against Bank officials, including referring the matter to the police and prosecuting authority for further investigation. This statement was released on Friday 9 November 2007, and the story was assigned to investigative reporter Wisani wa ka Ngobeni . Ngobeni did not have, and could not get, a copy of the audit report, and had just over 24 hours to secure his story.

"How fat cats looted the Land Bank" appeared as that Sunday's front-page splash with the subhead "Cabinet calls for criminal charges after R2-bn is siphoned off to fund associates' business schemes" (11 November 2007)

It gave details of the audit report, though it gave no indication that Ngobeni had not seen it. It was a long and detailed account, but the relevant parts for our purposes are:

• Former Land Bank Chief Executive Alan Mukoki was placed at the centre of the story, which stated that he and his executives had deviated from the Land Bank's mandate and funded companies which had nothing to do with agriculture and which benefited a number of people associated with the Land Bank or its officials;

• It said that Mukoki, his executive and one board member should face criminal charges;

• Sam Mkhabela was described as "a Land Bank board member who has since resigned... (and) benefited" from a loan; and

• It included an interview with Mukoki, who pointed out that the audit report had not been signed; that the Land Bank board had deviated from their mandate only after not getting a response from the Ministry to their proposals for a turnaround strategy and waiting longer than the required 30 days, and that there was nothing wrong with his business associates or Land Bank executives benefiting from loans.

In a follow-up story the following week, "Heads roll for loans to fat-cat buddies" (18 November 2007), most of the allegations were repeated. It quoted the Department of Agriculture and Land Affairs putting the extent of the fraud at R900 million (as opposed to the figure used the previous week of R2-billion), but added that "Land Bank officials" had told the Sunday Times that this was only the amount disclosed in the financial statements, and did not reflect the total.

Two months later, the paper returned to the story with a Page 15 lead, "Land Bank boss ‘must be charged'" (20 January 2008). This time, reporter Ngobeni quoted directly from the report. He repeated many of the allegations in the original story and added substantial detail, but corrected some of the figures: the amount involved was now given as R1,1-billion. Mukoki declined to comment and the story recorded that Mukoki and Mkhabela had laid complaints with the Press Ombudsman about the earlier stories.

In his complaint to the Press Ombudsman, Mukoki challenged a number of details of the story, but the elements of his submissions relevant to our purposes were that the Sunday Times had failed to report that the Forensic Audit was issued with qualifications. He also claimed his views were incompletely reported or he was misquoted.

The Sunday Times conceded that there were errors in the report, which were the result of the fact that Ngobeni did not have a copy of the Forensic Audit and was relying on sources telling him what was in it. The newspaper also accepted that it had misquoted Mukoki in parts. However, the newspaper dismissed the factual errors as immaterial and stood by the overall import of the reports. The newspaper claimed that the qualifications in the audit report were standard auditors' disclaimers and could not be reported in full.

The Ombudsman found for the newspaper in many respects, but against it on some important issues. Those relevant here are the Ombudman's finding that the newspaper had been right to report the matter in the public interest and had been enterprising in finding out the content. However, the newspaper had not exercised sufficient care and should have reported the Forensic Audit's content as claims and allegations rather than fact. The Ombudsman found that the newspaper should have acknowledged its mistake over the amount of money involved; and that Mukoki's side of the story should have been carried more fully.

The newspaper was ordered to carry an edited version of the ruling.

Mkhabela's complaint to the Ombudsman was that he was linked to the alleged fraud and said to have benefited improperly, but he denied these allegations and said they had not been put to him before publication.

The Ombudsman found for Mkhabela in all major respects: that the Sunday Times "went too far in treating the allegations in the forensic report as fact and in sensationalising the report into confirmed allegations of fraud". The Ombudsman also found that the newspaper had failed to correct a number of errors in the article and had reported corrected figures (in its last story) without acknowledging that the previous figures were exaggerated. It also criticised the Sunday Times for neither putting the story for comment to Mkhabela, nor trying to correct this after the initial publication. The finding carries the sharp remark: "This makes for sensationalist journalism."

Again, the paper was ordered to publish a summary of the findings. That both the findings were published prominently on the same edition as the controversial "Transnet sold our sea to foreigners" story (on 24 August 2008) added considerably to the newspaper's embarrassment.

This case study illustrates a number of important issues:

1. Accuracy and correction

Ngobeni had to write the story without access to the Forensic Audit report. He told the panel that he had two sources who independently described the contents to him, but said he would still not have been satisfied if he had not interviewed Mukoki before publication and put these allegations to him. The story, done under extreme pressure, was 99% correct, he pointed out.

However, it was the 1% which was used by the complainant to attack the newspaper and discredit what was in many respects a good piece of reporting. The original errors were understandable, given the difficulties of deadlines and the unavoidable reliance on second-hand accounts of the contents of the Forensic Audit. It is apparent that the reporter did his utmost to secure accuracy in good faith and showed skill in getting many details of the report. More difficult to understand is why, when the Sunday Times received the audit report and it became apparent that some of the reported figures were inaccurate, these were not corrected in the paper. The paper did report some of the correct figures in its report of 20 January 2008, but without ever acknowledging that it had previously erred. The result was that there were contradictory numbers in the public arena, with little indication of which were reliable.

Makhanya told us it was the policy of the paper to publish corrections when errors were discovered, even if there had been no complaint. In this case, there had been more than one complaint and the matter was already before the Ombudsman. One journalist told us, however, that the newspaper generally made many errors under pressure which were never corrected. Sometimes the newspaper dealt with this by doing a follow-up report with the correct figures, but most often it ignored the mistakes.

Senior editors told us that they were cautious about publishing corrections when a formal complaint had already been instituted as the corrections would usually be dealt with in the resolution of that complaint by the Ombudsman. They pointed out that it had been the practice of the previous Ombudsman to negotiate a suitable and agreed retraction and apology for the aspects of the story which were wrong. The new Ombudsman took a different approach, preferring hearings where each side argued its case before the panel. This left the issue of corrections hanging in the air during a much longer process. The editors acknowledged that their approach might need rethinking in the light of this change.

The panel's view was that it was still preferable for the newspaper to correct errors promptly once it had established to its own satisfaction that errors had been made. This would not apply where there is still dispute over the facts, which the seniors say was the case with aspects of this story. But where there are clear errors, no matter how trivial, to correct them would show goodwill, remove them from the Ombudsman's consideration and allow him to focus on the more substantive issues.

The uncoordinated attitude to the correction of errors was reflected in the representation to the Ombudsman. Even when the newspaper acknowledged mistakes, these were dismissed as immaterial in the context of the complaint as a whole; it was not found necessary to explain why the mistakes had not been corrected. No correction was offered or they said they would await the Ombudsman's guidance on the need to correct it. A newspaper committed to accuracy should be cautious about dismissing any error as immaterial and should be confident in providing the date and place where the correction was carried.

What this revealed was a lack of clear policy and procedure on corrections, leading to what might appear to readers to be a cavalier attitude to the accuracy of details in the story.

2. Sourcing and attribution

The first two reports (those of 11 and 18 November 2008) purported to relate what the Forensic Audit contained, without giving any indication that the reporter had not seen it and was relying on sources who were describing the contents. This made the report vulnerable to claims that it was inaccurate or at least misrepresented the true facts, and hence subject to criticism. This was done to protect sources, we were told, as there were only a few people in the Land Bank who had seen the report and the paper did not wish them to be identified.

International best practice is to protect the identity of a source when necessary, but to offer as much description of them as possible in order to allow the reader to judge the credibility of the source. Even if it was impossible in this story to identify the sources as coming from the Land Bank, it is hard to justify never mentioning the sources at all and giving the impression that the facts were taken directly from the document. This meant that a layer of protection for the reporter and the paper - that some errors were the result of the second-hand nature of their reporting, a matter not of their choice - had been removed, and gave ammunition to those wishing to attack the newspaper's credibility.

This points to a need to clarify and strengthen policy on the attribution of facts and claims, particularly to rule out the use of unattributed claims.

3. Facts and allegations

The Ombudsman ruled that the Forensic Audit contents should be treated as allegations rather than facts unless and until they were tested in court. This sets a high standard, as an auditor's official report of this nature would normally be assumed to be thorough, accurate and reliable.

The newspaper will, therefore, need to set a firm standard about differentiating between allegations and established fact in future.

4. Fairness and right of reply

The Ombudsman was clear that the reporter had erred in not putting the allegations to Mkhabela before printing the story. The reporter said he had tried and failed to find a number for Mkhabela and he was mentioned only in passing in the story.

Nevertheless, the allegation against Mkhabela was serious and should have been put to him. Not doing so made the newspaper vulnerable to criticism, exposed it to legal action and undermined the story. It is notable that even after the initial report, no attempt was made to contact Mkhabela. A follow-up interview would have ameliorated the original oversight, and may have stopped his complaint to the Ombudsman in its tracks.  54

Editor Mondli Makhanya told us that he thought that in retrospect the story should probably have been held a week to allow more time for it to be completed. Others felt that a number of journalists were chasing the story and it would not have held for a week. The editing process should have picked up that there were individuals against whom serious allegations were being made who had not been contacted. There should have been more careful consideration given to either holding the story or publishing it with these names omitted. This points to a need to tighten up on editing procedures to identify and deal with instances where serious allegations are being made without the subject being given a chance to respond.

In conclusion, we emphasise that the reporter did a reasonable job under pressure, but the outcome points to a failure of procedures and policies to ensure accuracy, fairness, proper attribution of claims and the quick correction of errors.  55

Case Study II: Mbeki's R30m

On 3 August 2008, the Sunday Times filled its front page with material related to the arms deal. The main headline read: "Mbeki took R30-million and gave some to Zuma", with a strap above it saying "Secret report fingers both men in arms-deal bribe". A subsidiary headline referred to a "six-month Sunday Times investigation" which had revealed a series of related issues, identified in four bullet points. The rest of the page was given over to a summary of the investigation, angled on the R30-million claim, and an image of a submarine said to have major flaws. In the bottom right hand corner of the page, readers were promised: "For more sensational allegations, read the Sunday Times again next week."

Inside, some of the detail was provided. The report began: "A German shipbuilding giant paid President Thabo Mbeki R30-milllion to guarantee it won the submarine contract in South Africa's billion-rand arms deal. Mbeki gave R2-million of this to Jacob Zuma and the rest to the ANC. This staggering new allegation has surfaced for the first time in a secret report compiled in 2007 by a UK specialist risk consultancy." Further on, the story dealt with other claims: that Tipp-Ex had been used on arms deal evaluation documents, and that this had been questioned by the attorney-general; that Mbeki had brushed aside concerns about the deal's affordability. A separate story dealt with a submarine that turned out to be "a dud".

An editorial called for Mbeki's dismissal and full disclosure, either through a judicial inquiry or preferably through criminal prosecution. The editorial said Mbeki's "hands were dirty", which was why he was so evasive over the arms deal previously, and called him a "partner in crime" with Zuma.

The story was an immediate sensation. It was strongly denied by the Presidency, and there were hints Mbeki might take legal action against the newspaper. The German company involved took out a full-page advertisement denying the story. Public discussion of the claim was widespread and intense, and much of it quickly focused on whether the newspaper had produced enough evidence for its sensational headline and editorial.

This panel decided to include this story as one of its case studies, even though it was made clear right at the start that the newspaper believed strongly in the report and the way it was handled. Among other arguments in defence of the story, we were told it had never been properly challenged. We were told by various people that the fact that the Presidency has not pursued any action proves the story's strength. While other stories were acknowledged to have been problematic, the newspaper was happy to defend this one to the hilt. Nevertheless, the controversy around it seemed to us to justify looking at it.

We were unable to speak to the reporters involved in the six-month special investigation, in one case for health reasons. However, we were able to discuss the story in some detail with Makhanya, the newspaper's lawyer, Eric van den Berg, and the deputy managing editor (legals/campaigns), Susan Smuts. All had close involvement in the story. It also came up in several general discussions with staff.

We were told that one of the difficulties with investigating this story was the sensitive nature of the investigation, and the fact that it was ongoing. We made it clear, however, that we did not need to know any detail that might compromise sources or further investigation. In any event, the most important conclusions to be drawn from this case are simply based on what appeared in print.

It is clear that the newspaper believed at the time - and still does - that the accusation against Mbeki is rock solid. In print, though, the only real evidence presented was a paragraph in a report compiled by an unnamed "UK specialist risk consultancy", which quoted an also unnamed "former South African official who had access to such information" as saying the bribe had been paid. The paragraph further said that Mbeki had told "investigators" about the distribution of the money. To our mind, this is at best third-hand information.

We were told that there was corroborative evidence. Van den Berg, who had often been consulted during the investigation, said the reporting team had taken great care, and had found information backing up other claims in the report, so buttressing its general reliability. Makhanya told us that the investigative team had obtained strong direct evidence corroborating the claim against Mbeki from other sources. We were unable to verify the strength of this other evidence, and accept the newspaper's right to protect its sources.

The problem is, however, that readers were not given any inkling that there was other evidence beyond the extract from the risk consultancy's report. All they were shown was a third-hand allegation - an anonymous source quoting another anonymous source. The newspaper talked of a six-month investigation into the arms deal. While this produced evidence of other issues, notably problems around a submarine, readers would have focused overwhelmingly on the sensational claim against the head of state. And on that point, the investigation showed little evidence.

As Makhanya acknowledged in his discussion with us, the newspaper failed to persuade its readers. Because the political import of the accusation was immense, much more effort was needed to buttress the report. It became too easy for critics to dismiss it, and the newspaper's credibility suffered as a result.

The newspaper's attitude to the claim against Mbeki is clear from its handling of the story. The main front-page headline treated the claim as fact. It is true that the strap indicated the source, but its wording ("secret report fingers ...") put little distance between itself and the allegation. The editorial went further, reading, in part: "Today we reveal why Mbeki was so angry, afraid, evasive and frustrated by the constant pursuit of the truth by better men: his hands were dirty." The tone of the podcast put on the paper's website in which Makhanya discussed the story followed similar lines.

We were told that the strap made it clear this was just a claim, but we believe that most readers looking at the front page would be left with the clear impression that the allegation was proven fact, and that the newspaper had adopted the allegation as its own. In a case of this kind, the newspaper should use quote marks or another method used in headline shorthand that indicates something is simply a claim.

Elsewhere, we have commented on the newspaper's tendency to downplay the sourcing of stories. The attitude seems to be that where the reporter is sure of something, the newspaper should swing its full authority behind the story and state  58

it with certainty, without bothering too much with indicating where the information comes from. Sourcing often comes late in a story, and has to be inferred.

It seems that this attitude lies behind the handling of the Mbeki story. The paper believed the story, and presented it as fact. The problem was readers weren't persuaded.

As in other cases, we feel this story would have been legitimate if it had been handled differently. Clearly, its treatment was significantly affected by the cloak of secrecy thrown over it. It popped up, insufficiently formed, very late in the day, leaving no time to refine the approach. One of the benefits of a large newsroom is the range of views and voices that can be brought to bear on a complicated editorial task. The loss of wider input is the price of secrecy.

The follow-up

The front page promised further "sensational allegations" in the next edition. In the event, the paper ran reports about the lack of success of offset deals, little of which appeared to be new. While this was a follow-up to the broader topic of the arms deal, readers would have expected more on the most sensational allegation - that against Mbeki. Makhanya told us that the reaction to the Mbeki story had scared off some sources, which had made it impossible to run the planned follow-ups. Be that as it may, the newspaper suffered further damage by being unable to deliver satisfactorily on its promise.

Case study III: Transnet

On 24 August 2008, the headline of the splash in the Sunday Times read "Transnet sold our sea to foreigners". The subhead read "New law could result in R20-m claim from buyers". The strap stated: "Exclusive: Cabinet not told deal included ocean area up to Robben Island". The story was accompanied by a diagram containing an insert of the Waterfront, with the caption "What they said they sold", within a larger diagram, with the caption "What they sold". The larger diagram indicated that the area sold by Transnet was the entire Table Bay area, extending from Blouberg Strand and Green Point to Robben Island, and including the coast line between Bloubergstrand to the Waterfront. The key claim made in the article was that Transnet secretly sold prime Cape Town coastal land and a vast sea area when it sold the V & A Waterfront to investors from Dubai and London, and that the sale included the transfer of 22km of coastline and 90km² of sea from Table Bay to Robben Island. The article also alleged that if the Integrated Coastal Management Bill was passed, Transnet would face a potential claim of R20-billion from the purchaser for breach of contract.

Two weeks later, on 7 September 2008, also on the front page, the Sunday Times published a retraction headed "Transnet and the Sunday Times", which stated that the headline, the diagram and the statement about the extent of the area of sea sold "went too far".

The processing of the story

We met with the reporter, Mpumelelo Mkhabela, in Cape Town. He was the only staff member of the Sunday Times to have studied the underlying documentation. He told us that the angle he had envisaged for the story was that Transnet was arguing strenuously in Parliament that the Integrated Coastal Management Bill would affect Transnet's contractual arrangements in regard to its disposal of the V & A Waterfront to foreign investors. This was because the Bill sought to ensure that all coastal area was public property, and this would have made it impossible for Transnet to fulfil its obligations under the agreement of sale of the V & A Waterfront; this agreement clearly contemplated that the sale included land that would be reclaimed from the sea.

Mkhabela obtained the substantial and complex underlying contractual agreements and Parliamentary submissions on Thursday 21 August 2008. By that stage, the feedback from the Johannesburg conference was that the story, as pitched by Wally Mbhele, the politics editor, was interesting. Not much attention had at that point been paid to the size of the area that Transnet could reclaim from the sea (and that it had transferred to the buyer). This is reflected in Mkhabela's first draft of the story; its first paragraph reads:

"The ANC in Parliament is forging ahead with legislation which state-owned company Transnet claims could sink its R7-billion deal to sell the V & A Waterfront - and a large sea area - to a consortium of London and Dubai-led investors."

The size of the coastline and water area - stated to be 22km and 90km² respectively - are mentioned in this first draft some distance from the beginning of the story, and in these terms: "Transnet claims that it owns 22km of coastline and water area of 90km² stretching from Table Bay to Robben Island". There are no claims in the original draft that the sale was "secret", or that the lobbying by Transnet of the National Council of Provinces was "unprecedented" or "frantic", or that it amounted to Transnet seeking to "torpedo" the Bill.

At various stages of the editing process, the story was significantly re-structured - including the claims that the deal was "secret", that the lobbying was "unprecedented" and "frantic", and that it amounted to Transnet seeking to "torpedo" the Bill. We have been provided with what appears to be eight versions of the story, apparently by different editors and sub-editors.

It appears that a diagram that described the area under contention was requested by the Saturday late afternoon conference, and Mkhabela then produced the diagram that was used on the front page, and which he thought was correct. Mkhabela also told us that he thought that, while the original angle of his story had been changed in the editing process, the new angle - as well as the headline - was still justified by the underlying documents. This view was supported by the political editor and others.

Many editors who spoke to us expressed the view that there was not enough time at the conference to interrogate the Transnet story. Makhanya confirmed that not enough questioning took place at this conference, also in relation to the diagram setting out the size of the sea that was claimed to be sold. Some said they would have welcomed an opportunity to study the underlying documents, and in particular the claim that Transnet sold the sea.

The underlying documentation

We have examined in detail the underlying documentation. For our purposes, we point out here that the documentation indicates that:

(1) the diagram representing the land that Transnet could reclaim from the sea was not in the possession of the Sunday Times prior to publication;

(2) the briefing note of the Department of Environmental Affairs and Tourism ("DEAT") stated that Transnet had control and management (as opposed to ownership) over large areas beyond the port boundaries, including in Cape Town 22km of coast line and water area of 90km²; it was from this document, apparently, that the claim made in the newspaper as to the size of the area sold by Transnet was made; and

(3) John Dludlu of Transnet, in e-mail correspondence with Mkhabela, stated that the diagram indicating the extent of the sea area that Transnet was entitled to reclaim had previously been made available to the media. Indeed, a previous story on the issue had been published in the Cape Times.

The Sunday Times' response post-publication

Transnet's lawyers sent a letter of demand dated August 27, 2008, to the Sunday Times, demanding a retraction of all allegations and an unconditional apology, in the August 31, 2008, issue of the Sunday Times.

Susan Smuts, in a letter to Transnet's lawyers the next day, acknowledged receipt of the letter of demand and stated that the newspaper was in the process of investigating the matter.

We were told that Heather Robertson and Susan Smuts were tasked by the newspaper with determining, based on the documentary evidence, whether a retraction should be published by the newspaper. We understand that a detailed conference call was convened which included Smuts, Robertson, Mkhabela and Mbhele. While Mbhele was aware that a retraction would be published, Mkhabela told us that the first time he read the retraction was when he saw it in the newspaper on 7 September 2008. He also stated that, at the time we interviewed him, he was unaware of whether the Ombudsman's complaint by Transnet had been proceeded with, and he had not been requested to assist with formulating a response to the complaint.

We were also provided by Eric van den Berg of Bell, Dewar & Hall with a copy of his advice that an apology should be published because the evidence did not justify the allegation that Transnet sold the area depicted in the diagram.

Ultimately, as stated, there was significant internal division as to whether a retraction should be published by the newspaper, resulting in the unspecific retraction that was published on 7 September 2008.

A number of observations can be made concerning this case study.

The story was in the public interest and it was right to have published it. The difficulty lies in the form in which it was published. We believe that elements of the story were sensationalised without reason by the newspaper. The headline to the story is not necessarily inaccurate, because the documentary evidence backs the claim that, according to Transnet, it was entitled to sell those areas of the sea that it could legitimately reclaim, to Lexshell. We see no merit in Transnet's contentions that the sale was one of shares and not assets, nor in its argument that Lexshell - whose majority is controlled by foreigners - is a South African entity. But the dominant impression given by the story - the accompanying diagram, the strap suggesting that the deal was secret (for which we have seen no justification), and the subhead (with its claim - which appears to be too speculative to be given such prominence - that the new Bill could result in a R20-billion claim from the buyers), together with the headline - is not accurate. The facts did not, in our view, justify the sensational angle that the newspaper adopted.

With such a complex and technical story as the Transnet story, far more explanation should have been given to readers as to issues such as the process of land reclamation from the sea. We caution that informed editing of such a piece of journalism is difficult without access to at least the key documents that are being relied on.

Perhaps most strikingly, the re-writing of the story, from the reporter's version to the final version, illustrates significant interventions by editors. There is obviously a need for editors to intervene to sharpen language and make the copy flow. But in our view the re-writing of the text in this case appeared to go beyond sharpening the language. The editing gave the story a sensational angle that the reporter appears not to have intended, and that the documentary evidence - seen only by the reporter - did not justify. Without having seen the evidence, the editing should not have taken the course it did. Likewise, the reporter should have objected to the edits to the story.

A number of claims made in the report are not justifiable by reference to the evidence in our possession. While the reporter did a very good job of gaining access to and distilling complex documents, there was no justification for claiming the extent of the area to be reclaimed by Transnet (22km of coastline and 90km² of sea) and the diagram was clearly incorrect. The newspaper was right to retract the diagram and the sizes of the areas mentioned. The newspaper should also have explained to its readers - if necessary in a follow-up article - where it stuck to its guns.

If the newspaper wished to refer to the size of the coastline of 22km and the size of the sea area (90km²), it should have attributed that reference to the DEAT's briefing document, and placed it in proper context. In other respects, the article attributes quotes and statements to sources (such as Transnet's, the DEAT's and Lexshell's submissions to Parliament) very well.

An Accuracy Check was not done for the story, nor was the story sufficiently interrogated by senior editors. A well-maintained system of accuracy checks should have alerted the journalist and editors to the fact that allegations were being made in the article that could not be backed up. Some rigorous questioning and examination of documents by at least one senior not close to the story would have picked up the problem.

Finally, the newspaper's ultimate retraction in our view illustrates the need for a consistent and more rigorous approach to corrections and for better communication between editors and journalists. While Mkhabela was made aware that the paper intended to publish a correction and was party to numerous discussions about the problems with the story, he should have been given a copy of the retraction before it was published.

Case Study IV: Tito Mboweni's Niece

On Sunday 7 September 2008, The Sunday Times page three lead was "Tito's niece, 13, throws a stylish bash" with a sub-heading "Fancy goodie bags and a former Miss SA for the girl whose uncle signs the banknotes".

The story was passed on to the news desk by the social editors who had heard it discussed on the local social scene. The story was assigned to a reporter who is known for writing "good human interest" stories. Attempts to confirm the story by deadline were unsuccessful and it was taken over to the next week.

At the time, Reserve Bank Governor Tito Mboweni was making calls for consumers to tighten their belts. Furthermore, it was noted that there was a growing trend of young non-Jewish teenagers having "coming out" parties: "It was a social trend: your 13th is the new 21st - that was the spark to the story. It just so happened that Mboweni was the name and that was how the story was sold," said one editor.

Initially there was no confirmation that the teenager was related to the governor, but details of the event were confirmed by a number of sources to various persons at the paper.

It was, says the reporter, a story about a 13-year-old getting a dream party. The reporter confirmed details of the story that were cross-checked with different sources.

The reporter contacted the father on Friday morning and it was agreed that she would call back after he had discussed the story with family members. When the reporter called later in the afternoon, the father declined to be interviewed, saying: "If she was a celebrity or I was a celebrity of any of us was a celebrity, I would say there's a platform. But we are not even politicians."

The reporter says she tried to speak directly to one of the guests or other family members. News editor Jessica Bezuidenhout also spoke to the father to convince him to speak - to no avail. As the story had already been held one week, it was decided to submit it for publication. "We were getting further and further away from the actual event."

After the reporter had left the office, a woman claiming to be the family lawyer called the news desk. She phoned several times - noting at one stage that the event would not be a story if the teenager was not Mboweni's niece.

At this stage, the story was referred to the legal editor, Susan Smuts.

There was also, says the news editor, discussion about whether this was a good story. The issue of whether there was public interest or invasion of privacy "may have come up more strongly if there was an Mboweni angle".

The story was reworked with the reporter - first by the news editor and then with the foreign editor.

The reporter was comfortable with the story and the rewrite but uncomfortable with the possibility that they might be making "something out of nothing". The reporter conducted an initial accuracy check, as well as an accuracy check with the final version of the story.

There was general consensus in the newsroom that the story was newsworthy even though the initial spark of looking at a trend of 13th birthday parties was downplayed in the published article. "The story is newsworthy because of Mboweni, but it is also still about a kid having a dream party."

Bezuidenhout says they were mindful of the fact that the story was about a young girl, but she believed that it was treated gently.

"We do these stories all the time," says the editor. "It was not a privacy issue. There are areas we don't go with children, but this was a public function."

There was no comeback on the story and it elicited various comments online.

Senior editors told us that while they did not have a written policy on privacy, they were guided by jurisprudence and journalistic ethics on these questions. "We would not have named the child if the story had been in any way embarrassing or humiliating or personal or judgemental. We do not believe that we disregarded concerns about privacy. Rather, we were not of the view that the story would breach the child's right to privacy."

Was the story in the public interest? There is no policy guideline to govern stories of this nature: every story is discussed and judged on its own merits. There has also been no ongoing legal training.

However, the story did trigger possible privacy issues. While the news editor did discuss this with the reporter, there could have been greater consideration of the privacy issue. There needs to be greater clarity about what can and cannot be published in regard to children in particular. If the child had not been named, then no privacy issue would have existed. We are surprised that there appears to have been no substantive discussion on whether the child should be named.

In addition, the story highlights some of the sourcing issues we have raised. The food and clothing are described as if the reporter had seen them first hand. It would have been more appropriate to attribute this information fully.

Chapter 4: International Practice

In this section, we look at some useful experiences and practices from international news media. We begin by looking briefly at the Jayson Blair affair, a major scandal that embarrassed the New York Times (NYT). It's not that we believe there are significant parallels between that affair and the situation the Sunday Times finds itself in. However, the NYT responded to the damage to their credibility with a very thorough - and public - self-examination. The scandal also sparked a much wider discussion in the American media. The internal and wider discussions, the insights articulated and the concrete measures taken to rebuild the trust of audiences, we believe, are relevant to the Sunday Times. It is striking that some of the areas of concern that emerged in the US do have resonance for the Sunday Times.

After a short discussion of the scandal and its aftermath, we turn to some examples of international thinking and practice around themes of relevance to the Sunday Times.

Jayson Blair

In May 2003, a 27-year-old reporter by the name of Jayson Blair resigned from the New York Times after he was found to have indulged in large-scale "journalistic fraud", in the phrase used by the NYT. The newspaper investigated his reporting, and published a four-page spread on the results (full text at http://www.nytimes.com/2003/05/11/national/11PAPE.html?ex=1226120400&en=14ebc97b7bb0d983&ei=5070. The paper said: "He fabricated comments. He concocted scenes. He lifted material from other newspapers and wire services. He selected details from photographs to create the impression he had been somewhere or seen someone, when he had not." In an editorial comment, the paper apologised to readers, the subjects of distorted reports and journalists whose work had been stolen.

Five weeks later, the two most senior editors resigned because they took responsibility for what had happened, and because as a result of the scandal, long- simmering unhappiness about their management style had become prominent. A committee, chaired by Assistant Managing Editor Allan Siegal, investigated the specific circumstances of the Blair case, and looked at newsroom culture, communication, staff development and other issues (full report at http://www.nytco.com/pdf/committeereport.pdf).

The committee's report was accepted by new executive editor Bill Keller, who accepted its proposals to:

• Appoint an ombudsman or public editor;

• Appoint a "standards editor" at senior level, responsible for developing standards and ensuring that staff are educated about them;

• Appoint a "staffing and career development officer" at senior level;

• Ensure that editors are assessed for their ability to lead people and manage resources, as well as their journalistic skills; and ensure the implementation of an effective performance appraisal system;

• Rationalise dateline and byline policies;

• Ensure that each desk had a system to track errors;

• Review guidelines for the use of anonymous sources;

• Review management structures and clarify responsibilities at top editorial level, and review the centralised copy desk; and

• Improve communication

About a year later, Keller asked the committee to review progress, and a second report was compiled, Preserving our readers' trust. (available at http://www.nytco.com/pdf/siegal-report050205.pdf). Keller responded to the report with a list of additional measures (available at http://www.nytco.com/pdf/assuring-our-credibility.pdf). They were:

• To improve dialogue with readers:

o Senior editors would make themselves available for a regular online Q&A with readers;

o The paper would respond more systematically to public attacks;

o Public appearances by staff, on radio, TV and elsewhere, would be more carefully managed; and

 o A system would be established to allow readers to contact reporters directly from the website, while protecting reporters from spam and other undesirable material.

• To improve accuracy:

o Journalists were encouraged to check back facts and conclusions with sources, without giving them licence to edit copy;

o While accepting the need for anonymous sources in some circumstances, the paper would make sure their use was an exception rather than routine;

o A system to track corrections would be established; and

o It would be made easier for readers to draw errors to the paper's attention.

• To clarify the distinctions between news and opinion writing:

o Standardised formats for different kinds of writing would be developed; and

o Contact between magazine and news sections would be improved.

• To widen the definition of diversity to include other dimensions of life, beyond the standard ones of gender, race and ethnicity.

Poynter conference: journalism without scandal

In July 2003, the Poynter Institute convened a conference in response to the Jayson Blair affair, under the title Journalism without scandal. Reports on the conference are available at http://www.poynter.org/content/content_view.asp?id=41405

Five themes were identified:

• Leadership and Culture: The conference said that editors must develop and apply journalistic standards, communicate them to staff and readers, ensure they are passed on through training and articulate their application to non-print forms of journalism.

• Accountability and accessibility: The conference said that newspapers need to become more open and accessible to readers.

• Attribution and sourcing: The conference developed some guidelines to narrow the use of anonymous sources, and discussed the handling of sources in narrative forms.

• Corrections and clarifications: The conference said: "Newspapers must develop a comprehensive and aggressive policy of inviting and publishing corrections of fact and clarifying errors of context."

• Bylines and datelines: The conference developed guidelines to make sure the use of these was consistent and clear to readers.

Some themes in international best practice

1) Accuracy and fact-checking

Newspapers are developing new methods to minimise errors of fact, many of them designed to track mistakes and corrections in order to find patterns and deal with them. Examples are at Zero Hora, in Brazil, and the Chicago Tribune (details of the Brazilian example at http://www.poynter.org/column.asp?id=101&aid=137822).

Users of such systems say it is important to get everybody to buy into these systems, to make accuracy a joint project. If it becomes a policing matter, that aim is undermined.

Other measures include

• Doing post-publication "accuracy audits" on some stories on a regular basis, by checking every detail in the report;

• Including accuracy as part of performance appraisals; and

• Developing electronic accuracy check tools.

Further ideas and suggestions are in Chapter 2 of the APME book, Building Trust in the News. http://www.apme-credibility.org/Building_Trust_2006.pdf

2) Corrections:

Journalists need to be accountable to their listeners, readers and viewers. One of the best and most effective ways to do that is to admit mistakes and promptly correct them. - Alicia Shepard, National Public Radio Ombudsman

The issues of accuracy and accountability come together in the area of corrections, which newspapers need to take seriously. Journalists sometimes point out that errors are inevitable. Given the volume of information that has to be processed daily or weekly, that is probably true. But the implication can't be acceptance and complaisance, but a serious attempt to prevent mistakes, and to rectify them when they do creep in.

Readers understand that mistakes are possible. But credibility suffers real damage when a newspaper is seen to be slow and grudging in correcting mistakes. There is no room for an approach that hopes people won't find out. Credibility is enhanced when a newspaper is seen to correct errors swiftly and unreservedly.

Aspects of strong corrections policies include:

• Corrections are dealt with systematically. There is one person who deals with all mistakes the newspaper becomes aware of. Any complaint is dealt with as quickly as possible.

• The newspaper proactively corrects mistakes, even if nobody has complained. Staffers are encouraged to point out errors.

• It is made easy for readers to draw mistakes to the newspaper's attention, and they are positively encouraged to do so. Contact details for complaints are clearly and prominently publicised.

• Corrections are handled systematically online. This does not mean invisibly fixing the error on the site. Many newspapers ensure that the original article has a clear notice attached that points to, and if necessary links to, the correction. In addition, a separate and easily accessible ‘corrections' section on the site can be used to collect all corrections, with links back to the original article.

• There are different approaches to the question of pointing out when the error stems from an outside source. Some newspapers take responsibility for all mistakes, regardless of where they came from. Others will point out the source if it is relevant. A correction is not the place for pointing fingers internally, however - it makes no difference whether the reporter or the sub-editor made the mistake.

Further ideas and suggestions are in the APME book, Building Trust in the News. http://www.apme-credibility.org/Building_Trust_2006.pdf

3) Transparency / Accessibility:

Freedom of the press is critical to our democracy, but journalistic independence can beget arrogance, or the perception of it, which alienates readers. News leaders, therefore, must create a culture in which connections with readers - and non-readers, as well - are valued, sought, created and maintained. The news process must be as transparent as possible, informed by and accountable to public concerns, while retaining journalistic integrity and independence. - Poynter report on Journalism without Scandal conference.

A significant amount of the new thinking on credibility issues facing the media internationally has focused on a new challenge: newspapers have to put time and energy into building their relationship with their audiences. Journalists can no longer see their responsibility ending with the published story. Media houses that want to retain the trust of their audiences need to work at it. Three chapters in the APME book Building Trust in the News (cited above) are devoted to these issues. We borrow their headings:

Building reader connections

Traditionally, a newspaper speaks through the pages of its daily or weekly edition, and listens by looking at readers' letters. This is no longer enough: modern technology has opened many new doors which newspapers should use. More importantly, readers expect a better quality of interaction.  74

Many readers phone the newspaper. A study done at the Lincoln Journal Star found that around half of the callers were looking for information, a quarter were offering information on possible news items, and just over 4% had a complaint. It is important to make sure that callers are courteously and efficiently dealt with. A caller must feel that the call has been useful, even if they do not find exactly the solution they seek. People answering the phones must be equipped to find the information being sought quickly. Dealing with readers on the phone is a core function, not an irritating distraction.

E-mail obviously offers huge possibilities. Some newspapers publish reporters' e-mail addresses together with stories they have written, although mechanisms are necessary to keep out spam and nuisance mail. Alternatively (or in addition), an online form can be created through which readers can ask questions or offer information. It is important to encourage people to use such mechanisms. Some newspapers have built e-mail lists that are sometimes used by reporters to seek unusual information, or to test ideas.

Face-to-face contact can only reach small numbers of people, but is fantastic for building relationships. Some newspapers have built readers' panels, others offer newsroom tours, or invite members of the public to spend a day in the newsroom. Round table discussions are organised. Opportunities for editors and other staff to attend events and speak publicly are welcomed as further opportunities to listen to audiences and build the newspaper's profile.

Formal audience surveys have been conducted, to establish not only which fridges readers have, but what they think of the newspaper and what they expect of it.

Explaining journalism

For many readers, the practices and habits of journalists are thoroughly opaque, and decisions are hard to understand. The ANC line "they will do anything to sell the paper" has much wider resonance than media people would like to believe. A recent Wits University research project found that a very high proportion of people believe that journalists routinely buy information. The myths out there are very damaging to credibility. The answer is simply for journalists to do more to explain themselves.  75

Steve Smith, editor of the Spokane-Review, talks of leaving "fortress newsroom". There are many ways of doing this:

• Some newspapers arrange for senior editors to make themselves available online for a Q&A on a weekly basis.

• The Spokane-Review has established a feature on its website called News is a Conversation, where eight reader bloggers comment on the newspaper every day.

• Notes on news conferences have been posted, the conferences themselves have even been webcast.

• Staff profiles have been posted on websites.

• The Register in Orange County, California, set up a newsroom at a fair where people could come and watch the paper being produced.

Becoming accountable

Just as journalists should explain themselves more, they should also be more willing to answer to criticisms. The two points are closely interlinked. (Issues around the handling of corrections have been dealt with in more detail above.)

• Contact details for staff members should be easy to find, preferably through searchable lists on the paper's website.

• Editors should keep an eye on themes that may emerge in readers' letters, calls or complaints, and seek out the people making these points to understand the issue better.

• The paper can explain decisions that are likely to be controversial before the outrage even begins.

• The Rockford Register Star had reporters, sources and readers fill in a report card on the paper's credibility.

Further ideas and suggestions are in Chapters 3, 6 & 7 of the APME book, Building Trust in the News. http://www.apme-credibility.org/Building_Trust_2006.pdf

4) Public Editors/Ombuds:

An ombudsman works independently within news organisations at the interface between readers, listeners and viewers on one side, and journalists and editors on the other. It is the only kind of self-regulation that can have the effect of building trust between a specific news organisation and its readership or audience. It does that through the systematic and impartial handling of complaints, and the open discussion of issues raised by readers. It offers a real chance to build a new, stronger relationship between journalist and reader. - Ian Mayes, former ombudsman at The Guardian

Ombudsmen, sometimes known as ombuds or public editors, are senior people whose job it is to respond to complaints and provide an independent view of standards of journalism. They must be structurally independent of the editing hierarchy, and are generally full-timers, although some smaller newspapers have saved money by bringing in academics or the like on a part-time basis. They usually write a regular column, and sometimes write regular commentary for internal use (Maariv, Israel), or are brought in to discuss particular issues (Mail & Guardian).

The system is particularly popular in the US, where (significantly) there are few media councils. But with the current downturn in newspaper circulations, many of these roles are being scrapped as part of cost-cutting measures. In South Africa, only Die Burger and the Mail & Guardian have an ombudsman. Some other newspapers advertise such an office, which is one of a range of responsibilities a designated editor has.

Critics say the role is redundant since it is the job of editors to make sure the paper is responsive to readers. The fact that the person is paid by the newspaper is said to undermine the claim to independence. However, supporters argue that it is useful to have somebody whose mandate is to represent an independent view. And it is not clear who else would pay for a post of this kind if not the newspaper itself.

There is more information on the site of the international Organisation of Newspaper Ombudsmen, including links to columns by all members: www.newsombudsmen.org.  77

The site includes two very useful articles: "Increasingly, newspapers call on ombudsmen to cure what ails them", by Lucia Moses in Editor & Publisher (2000) (http://www.newsombudsmen.org/moses.html) and

"Newspapers need ombudsmen (an editor's view)", by Charles W Bailey, Washington Journalism Review (1990) (http://www.newsombudsmen.org/bailey.html).

5) Attribution and sourcing:

Our responsibility to the reader is to make clear where we got our information. - Poynter report on Journalism without Scandal conference

As a reader, I want to know where stuff comes from, even if it's in a source I really trust. I just want to know, 'How'd you get that?" You get such a great feeling of credibility and authority when you're reading a book or a newspaper story where you see where everything came from. - Mike Miller, P1 editor of the Wall St Journal

Attribution is a core characteristic of journalism, and is standard practice on the great newspapers of the world. It allows readers to judge the strength of the information being presented, and also helps to create some distance between the information and the newspaper. Most newspapers make an exception where the reporter personally witnessed the event. "The usual rule of thumb here is that if you weren't there you've got to attribute it," said Miller.

The use of anonymous sources has been much discussed internationally. A 2005 survey by the Associated Press Managing Editors found that 44% of readers were less likely to believe a story if it was based on anonymous sources. Some 42% said it would make no difference while 11% said it enhanced credibility (full details at http://www.apme-credibility.org/readersspeakanon.html).

In the UK, the "lobby system" of reporting Parliament, with its complete reliance on unattributable information, came under heavy criticism for being vulnerable to manipulation and spin.  78

It is generally agreed that some sources legitimately insist on anonymity, and that some areas - like the political dynamics inside political parties - cannot be reported without some anonymous sources. Nevertheless, many media organisations are trying to limit the practice. CBC-Radio Canada says: "Our credibility hinges on our transparency. Information from anonymous sources should therefore be used only in truly exceptional circumstances, when we have no other way to report a piece of information deemed reliable and essential."

• The use of anonymous sources is seen as a last resort. Some newspapers use the phrase "X was granted anonymity because ..." to indicate it is not automatically available. Sometimes it is possible to push a source onto the record, or to get at least partial information on the record.

• Even where names can't be used, sourcing tries to give some information on how the person relates to the organisation in question, or to the debate. The more specific the sourcing can be, the better.

• Information from anonymous sources has to be subjected to particularly careful scrutiny. Corroboration is essential.

• Confidential sources must have first-hand information.

• Sources promised confidentiality may need to be identified to an editor, who then is of course also bound to protect the person. But since the newspaper as an institution is bound by a promise of this kind, it has the right to satisfy itself that the undertaking was worth making.

• If a source has been promised protection, it goes without saying that this must be given.

CBC-Radio Canada policy on this issue is at http://cbc.radio-canada.ca/accountability/journalistic/appendixA.shtml.

The New York Times policy is at

http://www.asne.org/images/nytpolicyonconfidentialnewssources.pdf.  79

Chapter 6: Conclusion and Recommendations

Conclusion

This panel has spent some three months looking at the Sunday Times, and it has been a fascinating exercise. The newspaper has a rich and proud history, and can be justifiably proud of the contribution it has made to South Africa's media and public life, and indeed its democracy. In the course of this exercise, we met many outstanding individuals, and saw some great work.

As one interviewee pointed out in the context of a particular story, reporting can be 99% right and everybody focuses on the 1% that is wrong. We were asked to identify the reasons for that small proportion of missteps. Unfortunately, credibility is hurt by even a few stories that are successfully challenged. And there is no doubt that the newspaper's standing has been affected by a few such reports.

The panel used a range of tools to identify the factors that shape the newspaper's character and functioning, in order to understand what has allowed these stories to appear. Our position as outsiders was our disadvantage and our advantage. On the one hand, we would have missed some things. On the other, we were able to take a fresh look at the situation. In any event, our insights relied very heavily on what staff members told us.

We conclude the review with a strong belief that a core problem is the organisational structure that has developed on the editorial side of the Sunday Times. Simply put, the structure has been allowed to grow wild over many years, and has now reached a stage where it obstructs effective editorial decision making. From long before the current leadership arrived, promotions were granted without enough attention being paid to what the organisation really needed, and what exactly a new title would mean. Now, the structure is like the inverted pyramid journalists use as a writing device, with too many people with big titles at the top and too few people doing the actual reporting at the bottom. In many cases, there is confusion about roles and an overlap of function, and more people report directly to the editor than he can realistically manage.

We saw evidence of this in many places. People talked about too many seniors with unclear roles, and we often had contradictory information on functions from different people. There was no organogram before we asked for one - and the one that was produced was unable to turn the functional confusion that exists into a coherent picture. Some staff members disputed what the organogram claimed their role to be. Late on a Saturday afternoon, we saw the multitude of senior editors hovering around, checking pages. Our group interviews were divided into groups according to seniority: the group of juniors was tiny, that of seniors was huge. An outward sign of this confusion is the extensive use of titles (most noticeably the deputy managing editor) that are not used elsewhere.

All this has led to an over-reliance on conference, which takes far more decisions than it should. The diary is decided there, as are angles, placement of stories and headlines on major stories. Almost everyone who deals with conference found it difficult, hostile, sometimes ignorant, and at the same time unable to ask the right questions. The centralisation of power in conference firstly makes it difficult for others to question decisions, since, after all, they represent the consensus of the paper's leadership. Secondly, it disempowers editors.

As our detailed recommendations (below) indicate, we are proposing a complete rethink of the editorial structure, starting from a zero base. We feel that the newspaper needs to think through its requirements completely afresh, define the different roles and design logical, effective reporting lines. Then it should match people to those roles, and in the long run hold staff to account for doing the jobs they are appointed to do.

In this review, we highlight one particular gap in the present situation: the lack of a chief sub, and also recommend that the office of the managing editor be strengthened and defined more precisely.  81

We acknowledge an organisational review of this kind will be difficult, but believe it is the best way forward.

It will also have an impact on the process, which we found to be light at the start and heavy at the end. We acknowledge that all news organisations struggle to get copy in early enough. However, many staff members told us that some major stories - and particularly those that caused problems - arrived too late in the week to be worked on as thoroughly as they should be. If the newspaper puts more effort into brainstorming and developing ideas at the beginning of the week (or even before), it will have more and better stories at the end.

We found real problems of low morale and poor communication, and recommend various steps to address these issues.

The panel also looked at the paper's policies and editorial approach. We found large areas where there was simply no policy at all, and recommend that it sets out a comprehensive statement of editorial policy.

We were particularly concerned at the approach taken to sourcing. It is clear that the newspaper has developed a style of downplaying or withholding details of where information comes from, in the belief that it sounds more authoritative. The problem with this approach is that it leaves the newspaper vulnerable if the information turns out to be false. It is also out of step with best practice internationally. We are recommending that the editorial policy pays particular attention to this issue.

We also thought hard about the editing process, and accusations that stories were "sexed" up. We found that the newspaper has often stretched the limits of acceptable editing too far, and has made itself vulnerable to accusations of sensationalism. We were often told about Sunday Times style, and editors uniformly felt this always managed to stay true to the facts. However, it was striking that many writers felt a sense of unease about what was done to their copy. Our case studies also illustrated the point. We recommend a more precise approach to editing: it's only powerful facts that really make powerful stories.

Due to the nature of our assignment, we looked particularly closely at the Accuracy Check. We found that when used conscientiously, it was an effective tool. However, it was clear that its use was only patchy, sometimes becoming a mere bureaucratic checklist. Most disturbingly, it became a crutch that editors relied on. If the form had been filled in, it meant that a particular kind of checking did not need to be done by the editor.

We are recommending a new and more comprehensive accuracy system, of which the existing check (which in any event should be reviewed) would be a part. The details are spelled out below.

Many staffers, including the editor, told us the newspaper had become arrogant, and we found it was in some ways cocooned from the rest of the media. We are recommending a number of measures to throw open the doors to greater interaction with audiences and other stakeholders.

And so, lastly, we are recommending the appointment of a public editor, who would manage this interaction, as well as take responsibility for complaints and corrections.  83

Recommendations

Handling of the report

We recommend:

• That the panel present this report to staff, and the editor his response, and that staff be encouraged to debate it.

• That at least the executive summary of this report, and a response from the editor, be published in the Sunday Times, and the full report be made available to the public on the Internet.

• That a mechanism be put in place to investigate progress on the issues raised in this report and the implementation of recommendations within 6-12 months.

Management and structure

• The staffing structure should be reviewed from a clean slate: first identify the newspaper's needs, then develop a new structure to meet those needs, and thereafter fit individuals into the structure.

• The new structure should have more reporters and fewer editors/managers. It should define clear roles and responsibilities (job definitions), including more delegation of authority.

• Within this new structure, provision should be made for a chief sub-editor.

• The managing editor's office should be strengthened. The essential task of this office should be to ensure the smooth and efficient running of operations. The office should play a strategic role in this structure, pro-actively preparing the Sunday Times to meet its challenges.

• The role of conference should be reviewed so that it does not take all decisions and instead becomes a forum for effective discussion.

• Staff should be given a written outline of performance assessment systems and related financial incentives to ensure there is consistency and no confusion. Management training should be given for the performance assessments.

Communication and meetings

• A plan needs to be drawn up and implemented for more effective communication throughout the newsrooms (including Durban and Cape Town).

• This should include the reinstatements of the weekly Johannesburg newsroom meeting as a forum for post-mortems, idea generation and general communication; and regular management meetings - with bureau editors via Skype video - to pass on information, provide a forum for discussion and communicate successes and problems. All meetings should be followed up by email communications that are also posted on the intranet.

Induction and training

• A structured induction process for new staff will ensure that policies, as well as the consequences for breaking these policies, are known. The managing editor should provide guidelines for induction, and it should be implemented by department heads.

• A pro-active newsroom-based training programme should be drawn up to address needs. This should include regular updates on legal issues, and guidelines on writing for the Sunday Times.

Policies

• A comprehensive statement of editorial policy should be drawn up, and then made known to staff. Policies should be clear, simple and enforced, and responsibility for their implementation should be clear. The policy should cover:

o Plagiarism: Under this heading, the newspaper should rule out the practice of reporters getting bylines for stories they have simply pulled together from other secondary sources. A byline should indicate that the writer did the interviews and other research him or herself.

o Privacy, and in particular the protection of children

o Handling of issues of taste, decency and violence

o Special attention should be given to sourcing. It should state the need to source information as clearly as possible at all times. It should set out parameters to govern the use of anonymous sources: measures to minimise their use, while recognising that they are sometimes unavoidable; the special need for additional corroboration, to situate sources even where they can't be specifically named. It should set out who is entitled to give permission for the use of anonymous sources.

o Corrections (see below)

o The freebies policy, duly reviewed and clarified

Accuracy Check

The Accuracy Check form should be reviewed and incorporated into a more comprehensive system to ensure editorial integrity, consisting of a number of elements:

• The Accuracy Check should be done earlier and more tightly focused on monitoring details in the story;

• A clear path for stories and pages should be set out, with firm checkpoints where individuals have responsibility for ensuring the quality of material. At each stage, the responsible person should sign off. At the very least, this should include a head of department, a chief sub-editor, and the most senior editorial person available or designated by the editor for this task.

• These individuals should be held accountable for approving a story/page.

• The paper should consider a system of post-publication audit, where stories are randomly selected and retrospectively checked for accuracy and balance.

Corrections

• The paper should write a detailed corrections policy, which allocates responsibility for decisions on when corrections should be published, their placement, speed of response, handling on the website and other matters. Corrections should be published in the first available edition.

• The paper should establish a database to log all corrections and errors, from which the Public Editor should extract patterns and trends for purposes of disciplinary action, counselling and training.

Legal issues

• Annual intensive training on legal issues should be provided to journalists and editors and clear guidelines provided to ensure that journalists can identify legal issues early on in the process of generating a story, and alert the responsible person (the senior designated to deal with legal issues) to this possibility immediately.

• The responsible editor and/or external legal advisors should be provided with the Accuracy Check and relevant documentation justifying the story.

Public Editor

The Sunday Times should appoint a Public Editor, who should have two main areas of responsibility:

• S/he should be responsible for investigating complaints about accuracy, fairness and other aspects of the paper's reporting, and deciding how and when a correction should be published.

• S/he should develop a programme to make the paper and its staff more accessible to readers and the general public.

• The Sunday Times should establish an accessible manner in which the public can contact the Public Editor - by e-mail or telephone, and this information should be published prominently every week.

• Measures to improve interaction with the public could include:

• Publishing reporters' email addresses;

• Regular online Q&As with editors;

• Visits to the newsroom by interested parties;

• Staff profiles published online; and

• Other ideas have been listed in the section on international practice.

• The Public Editor should communicate regularly with staff at the newspaper and with the public in regard to issues of accuracy and fairness at the newspaper and generally.

Other matters

• Stories should be edited together with the reporter, unless pressing reasons prevent this.

• The newspaper should make it very clear that overwriting will not be tolerated, and action should be taken against editors who introduce errors of fact, emphasis or tone into a story.

• Secrecy around investigations should be minimised. Where it is unavoidable, an informal group of people should be set up well before publication date to guide the project, subjecting it to particularly careful scrutiny and discussion.

• An audit of equipment needs should be undertaken to ensure reporters have the basics needed for their jobs (such as audio recorders and sufficient computer memory to store recordings).

Appendix

SUNDAY TIMES REVIEW PANEL

Brief

The purpose of this panel is to review the systems and processes at the Sunday Times in order to gain an understanding of how recent stories which fell short of standards of journalistic excellence were printed. The panel will make recommendations for future action to enable the Sunday Times to produce bold, incisive journalism that maintains the utmost credibility with its audience.

The panel will therefore:

• Identify shortfalls in the editorial process from conceptualisation to publication;

• Review the use of the Accuracy Check and other quality control systems currently in place;

• Review specific case studies;

• Review post-publication processes for complaints;

• Review relevant policies such as sourcing;

• Consider any other issues relevant to the credibility of the newspaper.

It will do so by:

• Reviewing all codes and quality control systems currently in place;

• Reviewing current processes and implementation of codes and policies;

• Developing and examining at least three case studies of articles from origination to post-publication;

• Holding large group discussions with staff and managers;

• Holding small peer group discussions with staff and managers;

• Examining international best practice;

• Examining the processes and policies related to legal checks.

The final report will be presented to editors and the panel will be available to present the findings to staff.

Source: www.businessday.co.za Transcribed from PDF. Please check against the original.

Click here to sign up to receive our free daily headline email newsletter