On appeal from:Free State High Court, Bloemfontein (Hancke J sitting as court of first instance):
The following order is made:
a. The two condonation applications by the appellant are granted, but the appellant is ordered to pay the costs incurred by the third to sixth respondents in opposing these applications.
b. The appeal is partly upheld and the third to sixth respondents are ordered to pay the appellant’s costs of appeal.
c. The order of the court a quo is set aside and replaced by the following:
‘1. The plaintiff’s claims against all six defendants are dismissed.
The plaintiff is ordered to pay the costs incurred by the third to sixth defendants.
The third to sixth defendants are ordered to pay the costs incurred by first and second defendants.’
BRAND JA(Maya JA, Mhlantla JA, Majiedt JA and Meer AJA concurring):
 This appeal has its origin in a defamation action originally instituted by the appellant in the Free State High Court. The appellant, Mr Mogale Modiri, is a businessman of Bloemfontein in the province of the Free State. On 3 March 2004, an article appeared in the Daily Sun newspaper which is distributed, inter alia in the Free State province, under the title ‘Mangaung Crime Crackdown’. The article commenced by informing the reader that the ‘Mangaung police are getting on top of the crime situation in the Bloemfontein area’. It then proceeded to convey some general information about police business which it ascribed to a senior police officer, Superintendent Adam Wiese. Thereafter it continued with the following statements which eventually gave rise to the defamation action:
‘Daily Sun readers in the area are asked to help the police in catching Stanford Modiri, who is allegedly involved in drug dealing, cash-in-transit heists and car theft.
Wiese said: 'We will catch him, but it would be great to have some help. The problem is that he uses other people to do his dirty work for him.'’
 Departing from the premise that Superintendent Wiese had said what the article attributed to him, the appellant at first brought his action against Wiese’s employer, the Minister of Safety and Security, and against Wiese personally as the first and second defendants. On appeal they are the first and second respondents, to whom I shall refer as ‘the police respondents’. In their plea the police respondents emphatically denied, however, that Wiese ever made the statements ascribed to him by the writer of the article. Following upon this denial, the appellant sought and obtained the leave of the court a quo to join the present third to sixth respondents as defendants in the action. I shall refer to these respondents collectively as ‘the media respondents’. The third to fifth respondents are, in the order of their citation, the editor, the owner and the publisher of the Daily Sun. The sixth respondent is a newspaper journalist and the writer of the challenged article, Ms Yoliswa Sobuwa. In their plea the media respondents contradicted the statement by Wiese, that he did not tell Ms Sobuwa what she attributed to him in the article. They insisted that he did. In any event, they denied that the article was either defamatory, wrongful or published with the intent to defame.
 In the court a quo the trial came before Hancke J. The first witness for the police respondents was Superintendent Wiese. In essence, he stood by the denial, foreshadowed in his plea, that he ever made the statements about the appellant that Ms Sobuwa ascribed to him in the article. What had happened, Wiese explained, was that Ms Sobuwa came to see him in his office on 26 February 2004. She was in the company of Mr S Z Bahumi who was known to Wiese as a member of the National Intelligence Agency (NIA). Wiese was not told that Ms Sobuwa was a journalist and he assumed that she was a colleague of Bahumi. According to Wiese, his acquaintance with Bahumi stemmed from a previous incident during 2001 when firearms and other items were stolen at the police station where Wiese was then the commanding officer.
 At the time of these incidents of theft, there were suspicions that some of the policemen at the station were involved and that these policemen could have some connection with the appellant. This was recorded in a letter written by Wiese to police headquarters in Pretoria on 18 February 2002. According to this letter, two of the policemen suspected of involvement in the incidents of theft were associated with an individual who was employed by the appellant, while the appellant, in turn, was known in police circles for his alleged involvement with armed robbery, vehicle theft and drugs. In short, Wiese confirmed in cross-examination that there is a close correlation between the allegations in the letter, on the one hand, and the contents of the impugned article, on the other, with regard to the appellant’s involvement in crime.
 Probably because of this letter, the appellant’s name came up in the conversation between Bahumi and Wiese where Ms Sobuwa was present. During the course of this conversation, so Wiese testified, Bahumi said that the NIA had a file on the appellant and that according to information available to him, the appellant was involved in drug dealing, cash-in-transit heists and car theft but that it was difficult to apprehend him because he made use of others to do his dirty work. In short, according to Wiese’s testimony, the statement that Ms Sobuwa attributed to him in the impugned article, mostly derived from Bahumi. Wiese denied, however, that either he or Bahumi ever invited readers of the Daily Sun to assist the police in the apprehension of the appellant. That, Wiese said, could only come from the writer of the article. What also transpired from Wiese’s evidence was that Bahumi had passed away some time prior to the commencement of the trial.
 A further witness called to testify on behalf of the police respondents was Senior Superintendent Gerber, who was a member of the Organised Crime Investigation Unit of the police, known as the Scorpions. His evidence turned on a comprehensive written application which he prepared in July 2005 for permission to initiate an investigation under the name project Vulindlela, into a crime syndicate. The application document was based on information available to the police at the time. According to this information the syndicate, of which the appellant was the confirmed leader, involved itself in motor vehicle thefts and related crimes. The crimes referred to in the application were committed in various places in the country, including Bloemfontein, on a regular basis since 1984. In motivation of the special project, which would require considerable expenditure and manpower, the application stated that over a number of years the police had been unable to apprehend the leaders of the syndicate through conventional investigation methods. The primary difficulty, so the application explained, was that despite the fact that upon their arrest the actual perpetrators of the crimes identified their leaders, including the appellant, they were unwilling to testify against these leaders in court. In consequence, the leaders were able to continue their community-threatening illegal activities with impunity.
 Gerber further testified that the application to embark upon project Vulindlela succeeded and that, as a result of the ensuing investigations, leaders of the syndicate, including the appellant, were arrested in September 2005 and appeared in court on charges of motor vehicle theft. At the time, these arrests and appearances were widely reported in the press. However, because the potential state witnesses refused to testify, the charges had to be withdrawn. After the arrests, Gerber said, the incidence of motor vehicle theft in the Bloemfontein area declined from about 130 to about 60 per month. In cross-examination Gerber could not say when the enquiry which preceded the application started, but that information about the appellant’s alleged involvement had come to his personal knowledge when he joined the vehicle theft unit in about 1991. Moreover, Gerber said, the information involving the appellant as a ringleader in criminal activities had been available in the police circles for a number of years.
 In the court a quo, the media respondents closed their case without presenting any evidence. Moreover, Hancke J found that no criticism could legitimately be levelled against the witnesses who testified on behalf of the police respondents. Consequently, he accepted that Wiese never made the statements defamatory of the appellants that were ascribed to him in the impugned article. In the result the appellant’s action against the police respondents was dismissed with costs. On appeal the appellant did not contend that the court a quo had erred in dismissing his claim against these respondents but submitted that it should have ordered the media respondents to pay their costs. In the result, the police respondents took no part in the appeal proceedings. In this court the only remaining issue with regard to the police respondents therefore related to whether it is the appellant or the media respondents who should be held liable for their costs in the court a quo.
 With reference to the media respondents, Hancke J found that, although the article complained of contained a number of inaccuracies, the sting of the defamatory part was substantially true and its publication for the public benefit. In the result, he upheld the ground of justification raised by the media respondents, generally known as the defence of truth and public benefit. He therefore also dismissed the appellant’s claim against the media respondents with costs. The present appeal against that judgment is with the leave of the court a quo.
 On appeal the media respondents no longer disputed – in my view rightly so – that the article included statements that were per se defamatory of the appellant. That raised the presumption that these statements were both wrongful and published with the intent to injure. The media respondents therefore attracted the onus to establish a defence which excluded either wrongfulness or intent. Though at some stage there was doubt as to the exact nature of that onus, it has by now become settled law that the onus on the defendant to rebut one or the other presumption, is not only a duty to adduce evidence, but a full onus that must be discharged on a preponderance of probabilities (see Hardaker v Phillips 2005 (4) SA 515 (SCA) para 14; Le Roux v Dey (Freedom of Expression Institute and Restorative Justice Centre as amici curiae) 2011 (3) SA 274 (CC) para 85). In their pleadings the media defendants denied both intent to defame and wrongfulness. But in the absence of any evidence on their behalf to rebut the presumption of the former, it seems to me that intent to injure must be regarded as being established. It therefore matters not that, because we are dealing with media defendants, fault in the form of intent is not required and that negligence would suffice (see eg National Media Ltd v Bogoshi  ZASCA 94; 1998 (4) SA 1196 (SCA) at 1214C-E; Mthembi-Mahanyele v Mail & Guardian Ltd 2004 (6) SA 329 (SCA) paras 44-46). By virtue of the media respondents’ failure to prove absence of intent, the appellant has cleared a higher hurdle of fault than the required one. The outcome of the appeal thus turns exclusively on the element of wrongfulness. Hence the only question is whether the media respondents have succeeded in establishing any one of the various grounds of justification that they raised.
 In their plea the media respondents relied on a number of recognised grounds of justification, including truth and public benefit, fair comment, reasonable publication, and qualified privilege on the basis of a right or duty on their part to publish the defamatory statements and a corresponding right on the part of the readers of the Daily Sun to receive the same. Any one of these would, if established, serve to exclude wrongfulness. The one that found favour with the court a quo was that of truth and public benefit. If that finding were to be upheld, it would be the end of the matter. I therefore turn to that enquiry. In this regard the appellant’s contentions as to why the court a quo erred in upholding the defence of truth and public benefit were essentially threefold. First, that the media respondents did not lead any evidence in rebuttal of the presumption of wrongfulness. Second, that the inaccuracies in the article precluded any reliance on the defence under consideration. Third, that the media respondents could not rely on the information of the appellant’s alleged criminal activities testified to by the police witnesses, because it had not been demonstrated that the article was based on that information.
 The appellant’s first contention seems to depart from a confusion of the element of wrongfulness with that of intent. Though both the presumption of intent and that of wrongfulness arise from a single event, that is, the publication of a defamatory statement, the two presumptions are essentially different in character. The presumption of intent to injure relates to the defendant’s subjective state of mind. By contrast, the presumption of wrongfulness relates to a combination of objective fact, on the one hand, and considerations of public and legal policy, on the other (see eg Neethling v Du Preez; Neethling v The Weekly Mail  ZASCA 203; 1994 (1) SA 708 (A) at 768I-769A; Le Roux v Dey 2011 (3) SA 274 (CC) paras 121-125). Generally speaking, a rebuttal of the presumption relating to the subjective state of mind of those who acted on behalf of the defendant will therefore require some evidence to be led on the defendant’s behalf. By contrast, the objective nature of the enquiry into wrongfulness signifies that the subjective beliefs of the defendant are of no consequence. Thus understood, it becomes apparent, with reference to the defence of truth and public benefit, for example, that both elements of this defence can in principle be established on the basis of facts not deriving from the defendant’s own witnesses. Hence the failure by the media respondents in this case to call any witnesses did not automatically preclude them from relying on this defence.
 As to the appellant’s second contention based on the admitted inaccuracies in the impugned article, it is a matter of settled law that the defendant is not required to prove that the defamatory statement was true in every detail. What the defence requires is proof that the gravamen or the sting of the statement was true. Inaccuracies in peripheral detail do not rule out the defence (see eg Johnson v Rand Daily Mails 1928 AD 190 at 205-206; Independent Newspapers Holdings Ltd v Suliman  3 All SA 137 (SCA) paras 34-38). The underlying logic appears from the judgment of Wessels JA in Johnson. The reason, he explained, why truth and public benefit is recognised as a defence, is because a plaintiff is not entitled to recover damages in respect of an injury to a reputation which he does not deserve. Consequently, the defendant ‘need not justify immaterial details or mere expressions of abuse which do not add to its sting and would produce no different effect on the mind of the reader than that produced by the substantial part justified’. The gist or sting of a statement is determined with reference to the legal construct of a reasonable reader. It is the meaning that the reasonable reader of ordinary intelligence would attribute to the statement (see eg Basner v Trigger 1945 AD 22 at 32; Sindani v Van der Merwe 2002 (2) SA 32 (SCA) para 11). The test is thus an objective one. Evidence of how the plaintiff, or for that matter, any actual reader of the article understood the statement is of no consequence.
 The appellant contended that on the application of the test thus formulated, the reasonable reader would understand the article to convey that he was guilty of serious criminal activities and that he used other people to do his dirty work for him. I do not agree with this analysis. In my view a reasonable reader would understand the article to mean that a police officer had told the journalist that:
(a) on the basis of information available to the police, they suspected the appellant of being involved in serious criminal activities, including cash-in-transit heists and car thefts;
(b) the same police officer told the journalist that, although the police would eventually apprehend the appellant, they were at that stage unable to do so through lack of evidence, because he made use of others to do his dirty work; and
(c) the police officer therefore called upon the assistance of the Daily Sun readers to provide that evidence.
 In Independent Newspaper Holdings Limited v Suliman  3 All SA 137 (SCA) para 24, this court declined to accept the proposition that the reasonable reader is bound to equate a statement that a person is suspected by the police of committing a crime with a statement that the person has actually committed that crime. This is so because the legal construct of the reasonable reader knows that, while many persons arrested and charged with criminal offences are eventually convicted, guilt or innocence is determined by a court on the basis of admissible evidence and that, not infrequently, the person charged is acquitted in the end (see eg Mirror Newspapers v Harrison  HCA 50; (1982) 42 ALR 487 (HC of A) at 492). Though, generally speaking, it is therefore per se defamatory to say of a person that he or she is suspected of criminal conduct (see eg Hassen v Post Newspapers (Pty) Ltd 1965 (3) SA 562 (W) at 565B-C; Suliman para 31), it is not the same as to say that he or she is guilty of that crime. That must be even more so in a case like the present where the published statement made it clear that, despite their suspicions, the police were not even in a position as yet to arrest the suspect because they lacked the necessary evidence to do so.
 Understood in this way, I believe the court a quo rightly found that the gist of the article was objectively true. The application by Gerber to embark on project Vulindlela confirms all the essential elements of the impugned article; ie that the appellant was suspected by the police of participating in serious crime and that he made use of others to do his dirty work. The appellant’s counter-argument rested on the fact that the application was prepared more than a year after the publication of the article. But I believe there are two answers to this counter-argument. In the first place it is clear from the contents of the application itself and from Gerber’s testimony, that the application reflected information and suspicions held by the police for a number of years. Secondly, the application neatly dovetailed with the letter written by Wiese on 18 February 2002, which corroborates that these suspicions about the appellant’s participation in serious crimes, including vehicle theft and dealing in drugs, predated the publication of the article.
 This brings me to those allegations in the article that proved to be untrue. Of these, as I see it, there are only two. First, there is the allegation that Wiese asked the readers of the Daily Sun in the area to assist in the apprehension of the appellant. According to Wiese he never said that. But the question is whether this untrue statement can ever be regarded as part of the sting of the article. I do not think so. This, I believe, can be demonstrated by asking whether the defamatory meaning of the article would have been less serious if these words were omitted. As I see it the answer to this question is clearly ‘no’.
 The second untrue statement in the article was the one identifying Wiese as the journalist’s main source of information regarding the appellant. Again, however, I do not regard this untrue statement as part of the sting. I find support for this view in the decision of this court in Times Media Ltd v Niselow  1 All SA 567 (SCA). The defamation complained of in Niselow was contained in an article published by Times Media. It was to the effect that Prof Boffard of the Johannesburg General Hospital had said that the food prepared by Niselow for a large group of children ‘smelled awful and looked appalling’. On appeal this court accepted that Prof Boffard never uttered these words. Yet it held that this did not matter, because the untrue statement was not part of the sting. The sting of the article, so this court held (para 25) was that the food prepared by the respondent smelled awful and looked appalling. What Times Media had to prove was the truth of that statement, not that it was made by any particular person. The same sentiment, I believe, finds application in this case.
 Closely linked to this point regarding the journalist’s source of information is the appellant’s third contention that the media respondents could not rely on information which emerged from the testimony of the police witnesses because it had not been demonstrated that the article was based on this information. But, as I see it, the contention raises the rhetorical question ‘why not?’. Once the media respondents had established that the sting of the article was true, as in my view they did, it matters not where the information relied upon by the journalist came from.
 Turning to the further element of the defence, namely, that of public interest, I must admit that I find this a difficult issue to decide. The difficulty arises from the subtleness of the element itself. No exact definition of the concept is readily available in textbooks or decided cases. On the facts of this case, the issue appears to be further complicated by the following statements of Marais JA on behalf of the majority in Independent Newspapers Holdings Limited v Suliman  3 SA 137 (SCA) para 47:
‘That said, I think that the consequences of a premature disclosure of the identity of a suspect can be so traumatic for and detrimental to the person concerned when he or she may never be charged or appear in court and is, in fact, innocent, that greater weight should be assigned to the protection of the constitutional right to dignity and privacy and the common-law right of reputation, than to the right of the press to freely impart information to the public. It is not as if the press will be permanently deprived of the right to identify the suspect. Once he or she appears in court his or her identity may be disclosed with impunity. . . . But, generally speaking, and subject to the considerations I have mentioned in paragraphs 45 and 46, I do not believe it is in the public interest or for the public benefit that the identity of a suspect be made known prematurely.’
 I appreciate that these statements may be understood to provide authority for the proposition that, as a general rule and save for exceptional circumstances, it will not be for the public benefit or in the public interest to publish the identity of a person suspected of criminal conduct, unless and until that person has actually been charged in open court. On the facts of the Suliman case and read in the wider context of Marais JA’s judgment as a whole, I do not believe, however, that he intended to lay down such an immutable rule. To do so would, in my view, negate the role of public benefit as a constituent element of the justification ground of truth and public benefit (see also eg Manyatshe v M&G Media Ltd  ZASCA 96 para 18).
 As explained by the Constitutional Court in Le Roux v Dey 2011 (3) SA 274 (CC) para 122, common law grounds of justification play a pivotal role within the framework of our Constitution. The reason is that it is primarily in the province of justification that the common law allows the courts to strike a proper balance between the often conflicting fundamental rights of freedom of expression, including freedom of the press, on the one hand, and the rights to freedom of privacy and dignity, including reputation, on the other. Under the rubric of truth and public benefit, the balancing act turns mainly on the element of public interest or benefit. If a defamatory statement is found to be substantially untrue, the law does not regard its publication as justified. Publication of defamatory matter which is untrue or only partly true can never be in the public interest, end of story. But, the converse does not necessarily hold true. Our law does not regard publication of a defamatory statement as justified merely because it is true, precisely because the court may, in its performance of the balancing act, find that in the particular circumstances of the case, the freedom of expression is outweighed by the victim’s right to privacy or dignity.
 In the case of mere suspicion held by the police the court may well come to the conclusion, in a particular case that the right to dignity of the suspected person outweighed the publisher’s right to freedom of expression. This may happen in a case where, for example, it is found to be true that the police held the suspicion, but that the suspicion was based on no more than unfounded allegations by a meddlesome neighbour or antagonistic police informant. If in that case publication of the unfounded suspicion then wrecked the reputation of the suspected person or destroyed his or her career, the defence of truth and public benefit is most likely to fail. On the other hand a blanket ban against publication of police suspicion may very well impede the press in the performance of its vital function ‘to ferret out corruption, dishonesty and graft whenever it may occur and to expose the perpetrators’. (See Khumalo v Holomisa  ZACC 12; 2002 (5) SA 401 (CC) para 23.)
 In performing its balancing act the court must therefore decide the public benefit issue with specific reference to the facts of the case before it. Needless to say that these factual situations may vary infinitely. From this, I think, two consequences must follow. First, that the courts should refrain from restricting themselves in the performance of their balancing act by departing from ‘starting points’ or ‘general rules’ or ‘guidelines’. Though these may be of assistance to other courts and practitioners, some degree of uncertainty is unavoidable, precisely because we are concerned with a balancing act which has to be performed on a case by case basis in circumstances that may vary widely. The second consequence is that a decision as to whether or not publication of a defamatory statement was for the public benefit in a particular case – whether in Suliman or in this case – cannot constitute any binding authority in other cases which are distinguishable on their facts.
 Reverting to the facts of this case, I believe that publication of the suspicion held by the police was for the public benefit. This was not an instance of suspicion derived from information by a nosy neighbour or based on flimsy grounds. Nor was the police suspicion confined to a single, isolated incident. According to the police evidence which stands uncontroverted, they had information from those actually involved in the commission of crimes that the appellant was their ringleader. Moreover, as a result of this information the appellant had been in the sights of the police force and apparently also of the NIA for more than ten years. Despite this convincing case against him, the police were unable to apprehend the appellant, not because the evidence against him appeared to be unreliable, but because the witnesses were unwilling to testify. As a result the police were compelled to launch a special project, at substantial expense, by the Organised Crime Unit to put an end to the community-threatening criminal activities of the syndicate of which the appellant was the suspected leader.
 If these strong suspicions proved to be true, the further possibility was that the appellant would never appear in court on any charges for want of witnesses willing to testify against him. Hence the opportunity to publish his identity as that of a suspected criminal, as envisaged by Marais JA in Suliman, would never arise. As I see it, the appellant could not insist on enjoying the reputation of an honest businessman who is beyond any suspicion, which he did not or ought not to possess. In addition, the publication of these suspicions could serve the purpose of persuading members of the appellant’s community to come forward with potential evidence against him which the police so eagerly sought. The fact that the police did not actually ask the journalist to invite public assistance, plainly did not detract from this possibility. In consequence I find that the defence of truth and public benefit had been rightly upheld and that the defendant’s claim based on defamation was therefore rightly dismissed by the court a quo.
 This brings me to the costs of the police respondents in the court a quo. In this regard it will be remembered that the appellant was ordered to pay the costs of all the defendants, including the police. As to why the court a quo had erred in doing so, the appellant argued that his claim against the po
Please Login To View The Full Judgment!
lice was based entirely on the statement in the impugned article that attributed the defamatory matter to Superintendent Wiese. In addition, so the appellant argued, the media respondents maintained, both in their plea and throughout the trial proceedings, that the defamatory statements were in fact made by Wiese. But for this persistence by the media respondents in a version which the court a quo eventually found to be untrue, so the appellant’s argument concluded, he would not have pursued his action against the police respondents.  In my view the media respondents gave no persuasive answer to these arguments. The fact that publication of the defamatory statements in the end proved to be justified, provides no excuse for the media respondents’ reliance on a factual version that their information derived from the police, which proved to be untrue. Since it is that untrue version which led to the involvement of the police respondents in the action and the costs resulting from that involvement, I can see no reason why the media respondents should not be liable for these costs, which were solely attributable to them. To this limited extent the appeal must therefore succeed. But the fact that the appellant’s success on appeal is limited does not mean that he was not substantially successful. It follows that in my view, the media respondents should be held liable for the costs of appeal.  What remains is the costs of the two condonation applications by the appellant in this court. The first resulted from the late filing of his notice of appeal and the second from his late filing of his heads of argument. Both applications were opposed by the media respondents on the basis that the reasons advanced by the appellant’s attorney for his failure to comply with the rules of this court, were largely unsatisfactory. I agree with this argument. I also find some of the explanations disturbingly inadequate. Yet, I do not believe that they were so unacceptable that it would justify the refusal of condonation without regard to the merits of the appeal. In the end, the outcome of the condonation applications therefore turned on the appellant’s prospects of success. Since the appeal should, in my view, be upheld in part, it follows that that condonation applications should also succeed. However, because the media respondents were not unreasonable to oppose these applications, in the light of the unacceptable explanations, I believe they are entitled to their costs of opposition.  In the result the following order is made: a. The two condonation applications by the appellant are granted, but the appellant is ordered to pay the costs incurred by the third to sixth respondents in opposing these applications. b. The appeal is partly upheld and the third to sixth respondents are ordered to pay the appellant’s costs of appeal. c. The order of the court a quo is set aside and replaced by the following: ‘1. The plaintiff’s claims against all six defendants are dismissed. The plaintiff is ordered to pay the costs incurred by the third to sixth defendants. The third to sixth defendants are ordered to pay the costs incurred by first and second defendants.’