Appeal Hearing Decision: Business Day vs Trillian Capital Partners

Decision: Appeal Hearing

Applicant: Business Day

Respondent: Trillian Capital Partners

Matter No: 33335/05/2017

1.         Business Day (“appellant”) lodged an appeal in this matter, and Trillian Capital Partners (“respondent”) also lodged a counter-appeal.  That notwithstanding, the parties will be referred to as “appellant” and “respondent” respectively.

2.         The respondent had filed a complaint against the appellant in respect of a story which appeared in the appellant’s edition of 26 April 2017, with the headline “GE accuses Eskom of rigging tender – Claims contract favoured Chinese company; Politically connected Trillian supported bid”.  The story also appeared on appellant’s website with more or less similar headlines: “GE claims Eskom favoured Trillian bid – General Electric claims contract favoured Chinese company and politically connected Trillian supported bid”. Actually, there was an earlier online headline which was removed by the appellant within a few hours of its appearance, and replaced by the one stated above. That headline is not in issuePart of the story was about the role the respondent played in the  tender that was awarded by Eskom to a Chinese company known as Dongfang, the validity of which was being challenged by one of the failed bidders, namely, General Electrict (‘GE”). Murray & Roberts (“M&R”) was the other failed competitor. In his Ruling dated 24 June 2017, which is being appealed against, the Press Ombud summarized the respondent’s complaint as follows:


TCP (respondent) complains that, construed together (headlines and articles), the reportage was devised to impel readers to wrongly conclude that Trillian was involved in improper conduct or corruption by assisting Eskom to rig the outcome of a tender awarded to the Chinese company Dongfang.

The company also complains that the reporter did not adequately reflect the contents of its responses to his questions”.

3.         The respondent wanted the appellant to publish a corrective story, and an apology.  The Ombud dismissed all the complaints except one, namely, that the applicant omitted to publish respondent’s response in which it was denying certain allegations respondent felt had been levelled against it.  In this respect, the Ombud held that the appellant breached articles 1.1, 1.2 and 3.3 of the Press Code.  The sanction imposed was an apology. Appellant felt the sanction was too severe, and sought leave to appeal against it; however, leave was denied.

Respondent’s Counter-appeal

4.      It would be more convenient to deal with the respondent’s counter-appeal first. Central to the dispute between the parties is a report known as a High Level Review which was prepared by the respondent at the request of Eskom. Eskom had to consider tenders from the above mentioned 3 companies, and the report was to assist it in deciding on these bids, which were for the replacement of a boiler at one of Eskom’s power stations. The report dealt with a very crucial element of the tender, namely, the cost element.  The bid by Dongfang differed with those of its competitors in that, although higher, it was at a fixed amount and therefore did not have an escalation element. Regarding the other two competing bids, the report showed that there was a risk that costs could subsequently turn out to be much more because of the escalation factor built into them. It also indicated that with the Dongfang bid, Eskom would find it easier to raise the funding because of its fixed cost element.

5.      The respondent complained that the story falsely implied that respondent improperly influenced Eskom’s decision to award the tender to Dongfang; and also that appellant falsely said that respondent favoured Dongfang.  These complaints can conveniently be dealt with together. It was argued for the respondent that the headline and the first three consecutive paragraphs cast the respondent as corrupt.  That cannot be so, for the reason given below.

6.  The headlines and sub-heads: The print headline and its sub-heads read: “GE accuses Eskom of rigging tender

·                Claims contract favoured Chinese company

·                Politically connected Trillian supported bid”.

The online headline and sub-headline read:

GE claims Eskom favoured Chinese firm’s bid, with backing of Trillian

.         General Electric has gone to court to stop the utility implementing a contract with Dongfang”.

The appellant argues that the main headline is borne out by the story, the first sentence thereof, and by other paragraphs further into the story; and that the factual accuracy of the headline was never challenged by the respondent.  The first sub-headline is borne out by what GE is quoted as saying: “At the very least, it gives rise to a reasonable apprehension of bias in favour of Dongfang Electric on the part of Eskom”. The appellant points out that the “politically connected” allegation was never complained of, and says its assertion that it is common cause was not challenged by the respondent. That the respondent’s report “supported” Dongfang’s bid, in the manner described above, cannot be denied.

All the arguments in relation to the print headline and its sub-headlines hold good in relation to the online headline. Besides, the headline is an allegation by GE, contained in the court papers.

If any corruption is suggested, it would be on the part of Eskom, because it is Eskom which is being accused of “rigging” the tender.  The rigging could have been in tandem with Dongfang, and not necessarily with the respondent.

Respondent also questions any reference to it in the headlines and the story,  on the basis that respondent was not part of the court case between General Electric and Eskom.  But the truth of the matter is that the story could not have made sense without reference to the respondent’s report; there had to be some background or context to the court case reported on. In any case, the story was not exclusively about the court case; reference had to be made to the role played by the report, given its significance.

7.  That the content of the story suggested corruption on the part of the respondent; that is, that respondent improperly influenced Eskom and favoured Dongfang:

7.1. It must be accepted, for a start, that the report was meant to influence Eskom’s decision; that must have been why Eskom commissioned it in the first place, and for which it surely must have paid.  That does not, however, amount to improper influence.  It is a well written document.  Even though it describes itself as provisional, and has some caveats, it is very emphatic about the fact that the bids of other companies (GE and M&R) were more risky because of the cost escalation factor and that, with Dongfang’s bid, it would be easier for Eskom to raise funds. On the face of it, the report was objective and not meant to favour Dongfang in a corrupt manner.  In fact, in the story, Eskom explains why it relied on the report. The report cannot therefore be regarded as an instrument of improper influence.

7.2 The respondent argues that the Ombud erred in his Ruling that breach of the Code is demonstrated only if the reportage was “devised to impel readers to conclude” that respondent was involved in “improper conduct or corruption by assisting Eskom to rig the outcome of the tender”.  Respondent argues that the Ombud set the bar too high.  We do not agree.  While the Ombud might indeed have used the clause “devised to impel readers”, it is clear that he was not setting the bar higher than usual. He was applying the usual standard by asking the usual question: how would an average reader understand the story?  This is the question we need to ask, and answer, in the present case.  In doing so, we need not again define what is understood to be a “reasonable reader”; the test is well known.  It is our view that no inference contended for by the respondent can be drawn by an average reader from the story, let alone that such an inference would be the only reasonable one to be made. The contents state the truth, and state in so many words that the respondent did not take part in the adjudication. For the same reason there was no need for a disclaimer against any adverse inference, as contended for by the respondent. Respondent made heavy weather of the statement in the story that respondent gave “the thumbs up”  to Dongfang’s bid. We have already discussed the force of the report; it must have been seen, objectively considered, as being strongly in favour of Dongfang; importantly, with strong motivation.  The phrase means no more than that the report, objectively looked at, speaks in favour of Dongfang. The article went into details explaining, if not justifying, why the report favoured that company.

8.    We are of the view that the Ombud dealt well with the above complaints, and find no justification to overturn his Ruling. For the reasons given by the Ombud, and those stated above, the counter-appeal fails.

Appellant’s Appeal

9.      Prior to the publication of the story, the journalist sent a number of questions to the respondent. The latter gave certain responses. It later complained that the appellant failed to include some of those responses, which respondent says constituted denials of certain allegations against it. The Press Ombud found, under the heading “Not adequately reflecting TCP’s responses”, that:

The journalist “neglected report to Trillian’s denial [in response to what the Mail and Guardian had reported] that it had:

·                been appointed by Eskom to find a supplier to replace a broken boiler;

·                any prior knowledge of or relationship with Dongfang and therefore did not identify that company for the job; and

·                been involved in a conflict of interest (for evaluating a sort-listed [sic] bidder after sourcing that bidder).”

He found that these omissions were in breach of articles 1.1, 1.2 and 3.3 of the Code of Ethics and Conduct and, as a sanction, directed the appellant to apologise to the respondent. It is against that ruling the appellant is appealing.

10. The respondent submitted that the Ombud’s factual findings on the materiality of the omitted response were correct and should not be lightly disturbed on appeal.

11.   The appellant points out that the part of the story on which the respondent relies to support the complaint about the issue of conflict of interest, is this:

In 2016, the Mail & Guardian reported Trillian had helped Eskom in talks to settle its insurance claim for the boiler after it exploded in 2014.  The article said Trillian was appointed to find a supplier to replace it”.

Appellant says it is true the story said as stated above; but it says it was entitled to so report and that it sought comment from the respondent, and the response, which was indeed published, was as follows:

Trillian said this week it had planned to submit a proposal with another Chinese firm, HYPEC, during a previous bidding round that was cancelled.  It declined to comment on its role in insurance negotiations with Eskom.” We agree with appellant’s submission that respondent’s published response could only be understood by the reader as a denial of any possible conflict of interest. The appellant made the following submission in its heads of argument: “In context, the statement and response do not convey that Trillian has a previous relationship with Dongfang; sourced it for the job; and then assessed its bid in what would be a conflict of interest.”  This submission is correct. To this must be added the fact that the story specifically mentioned that respondent did not take part in the adjudication of the bids: “This week, Trillian said it had merely produced a ‘cost-benefit analysis over a two day period’ for the Duvha bids and was not part of the evaluation team.”

12.      From the portions of the story we reproduce above, as well as the appellant’s submissions, we are satisfied that respondent’s complaint that material denials were omitted, was not justified; we therefore disagree with the Ombud that the appellant breached articles 1.1, 1.2  and 3.3 of the Code. Appellant’s appeal therefore succeeds.

13.      The following  Orders are made:

13.1 Respondent’s cross-appeal is dismissed.

13.2  The finding made by the Press Ombud in his Ruling of 24 June 2017 that the appellant breached articles 1.1, 1.2 and 3.3 of the Code of Ethics and Conduct, is set aside.

Dated this 2nd day of October 2017.

Judge B M Ngoepe, Chair, Appeals Panel

Mr P van der Merwe, Member, Public Representative




1.         I concur with the Orders made in the Majority Decision. However, regarding the respondent’s cross-appeal, I would have approached the matter differently, though with the same results. In dismissing respondent’s complaint that the report transgressed article 3.3 of the Press Code, I have a different motivation from the one in the Majority Decision.

2.         The appellant denied that “an inference can reasonably be drawn based on these [the first three]paragraphs or the article as a whole, that Trillian was involved in tender rigging and corruption”. The Majority Decision found:  “It is our view that no inference contended for by the respondent can be drawn by an average reader from the story, let alone that such an inference would be the only reasonable one to be made. The contents state the truth, and state in so many words that the respondent did not take part in the adjudication.”

3.         This is in line with the Press Ombud’s finding:

“Having taken all of the above facts and argument into account, I do not believe that the article(s) impelled readers to wrongly conclude that Trillian was involved in improper conduct or corruption by assisting Eskom to rig the outcome of the tender. It is an established principle that a publication cannot be held liable for conclusions drawn by some readers if the statements complained of are factually correct.”

4.         The appellant also relied on the “precedent” set by the Appeals Panel in ANC v Daily Dispatch, dated 3 November 2015, par. 25 and applied in Mail & Guardian v Jesse Duarte, Press Appeals Panel, 21 Jun 2016, par. 7-8:

“[N]either the Press Code nor South African law hold the author of a factual article liable for any adverse inferences that are drawn by the reader. Critical to this defence is that the facts (not the inferences) must be verified.”

5.         I respectfully submit that this “precedent” does not fully reflect the correct legal or ethical position. In my view, in South African law a publisher of an article can be held liable for defamation if a reasonable reader would make an inference that is defamatory of the plaintiff, except if the publication has a defence to rebut wrongfulness. According to the Press Code the dignity or reputation of an individual should be overridden (which, I submit, would include adverse inferences) only if it is in the public interest and in circumstances set out in articles 3.3.1 to 3.3.5 which mirror the common law defences in defamation cases.

6.         Contrary to the Majority Decision (and also the Press Ombud) I, however, accept the respondent’s submission that a reasonable reader would, after reading the article and especially taking into consideration the juxtapositioning of Trillian in the headline/subheadline and in the first three paragraphs, infer that the respondent was in some way implicated with the alleged “tender rigging”. This would then certainly have a negative impact on the respondent’s reputation.

7.         But this does not necessarily mean that the appellant was in breach of article 3.3 of the Press Code – it could have a defence to override the reputation of the respondent.

8.         As it would be difficult or impractical to prove or disprove that the respondent had adversely been involved in the alleged tender rigging, the appropriate defence for the appellant would be as stated in article 3.3.4 of the Press Code: “It was reasonable for the information to be communicated because it was prepared in accordance with acceptable principles of journalistic conduct and in the public interest.”

9.         Contrary to what the respondent argued, I find that the appellant’s report met the requirements set in articles 3.3 and 3.3.4. It was reasonable to publish the particular facts in the particular way at the particular time. The appellant based its story on credible sources and documents, which they reported on reasonably; the respondent was given an opportunity to comment; and these comments were published fairly.

Dated this 2nd day of October 2017

Mr F Groenewald, Member, Press Representative