Whether or not Bulelani Ngcuka spied for the security police during the 1980s is a question that may be of interest to political historians but is it an issue of public concern that should be investigated officially?

Only if there are reasonable grounds to suppose that Ngcuka's decisions as a public prosecutor may have been influenced by fear of exposure of his past, surely.

In other words, in pursuing corruption allegations against Jacob Zuma and against Mac Maharaj, might Ngcuka have been mainly motivated by a desire to damage the public stature of people who possessed evidence of his complicity in security operations against the ANC?

Or, alternatively, was his behaviour in these corruption investigations influenced by private anxieties about his past? For example, was the official decision not to press charges against Zuma - despite the existence of ostensibly quite persuasive evidence against him - caused by Ngcuka's fear of exposure as a traitor?

Even if Ngcuka was a spy, these possibilities seem very unlikely. Even though it happened 20 years ago, Ngcuka must be aware that he was the subject of an ANC investigation. There were plenty of rumours circulating at the time about the people whom the ANC thought were spies. The most obvious way to protect himself from exposure would be to shield those with evidence against him from any official investigation into their conduct rather than singling them out for attention.

If the government has any reason to believe that Ngcuka's leadership of the Directorate of Public Prosecutions has been influenced by personal anxieties then the focus of any official investigation should be the operations of the directorate, not Ngcuka's past. After all, it is quite possible for Ngcuka to have both once been a spy and to be a good public prosecutor today influenced only by concerns of public interest.

In this latter capacity, the decision that has aroused the most controversy is the directorate's refusal to charge Zuma.

An official inquiry might compel the directorate to provide a clearer explanation of this refusal in public. It did not seem to this writer to be an unreasonable decision. The evidence that has been published which suggests that Zuma may have been trying to solicit a bribe from Thales, one of the arms contractors bidding to supply the government, looks fairly damning to a layman, certainly. But a court would probably need stronger grounds than third party reports about Zuma's behaviour in correspondence. Zuma could claim that such documentation misrepresented him. Unless the DPP can predict a reasonable likelihood of finding witnesses to testify against Zuma, charging him would be pointless, even if the documents were deemed to be authentic.

The terms of reference of Judge Joos Hefer's investigation are unclear. If the judge is simply attempting to establish whether or not Ngcuka spied for the ANC, then his mission is fairly pointless. For the inquiry to be in the public interest the judge would need to establish whether Ngcuka's past has influenced his management of the prosecutorial authority. No reports to date have suggested that this is what the judge is trying to establish.

It appears rather unlikely that Ngcuka was a spy. He, as Mac Maharaj has suggested, was investigated by the ANC at the beginning of the 1980s for being a spy, and was allegedly recruited to inform the police about the affairs of the National Association of Democratic Lawyers. He was subsequently detained and then imprisoned for refusing to serve as a state witness in a security trial. As he himself has argued, would the police have allowed the confinement of one of their own agents?

Ngcuka could hardly report on Nadel's inner life from prison.

It is most unlikely that Hefer will be able to form any definite conclusions. Judges tend to be very cautious in their assessment of evidence. The policeman who Maharaj believes was Ngcuka's handler denies having had any dealings with Ngcuka. He will probably maintain such denials.

The ANC's 1981 investigation into the South African security police spying operations certainly uncovered genuine police agents. It also implicated innocent people. Several of its findings were derived from evidence extracted from people who were tortured by the ANC's security officials. Such evidence cannot provide the basis for any official investigation.

From the perspective of the ANC's leadership, the revival of these accusations against Ngcuka have already served one useful function. They have distracted media and more generally the public's attention from the much more serious issue of whether members of Nelson Mandela's and Thabo Mbeki's governments influenced the award of contracts as a consequence of receiving gifts from Schabir Shaik, a major beneficiary of government business.

The allegations against Ngcuka have also helped to line up the ANC's popular constituency behind the top party leadership - the behaviour of the delegates at last month's congress was an especially telling reminder of just how morally damaging such accusations can be.

To be fair, these effects may not reflect the cabinet's original intentions in setting up the Hefer investigation, the move may have simply been prompted by panic. From a public interest perspective, though, Judge Hefer's mission is ill-considered. It represents yet another instance of the ANC using official resources to address what are essentially private, party concerns.

Tom Lodge is professor of political studies at Wits University.

With acknowledgements to Tom Lodge and The Star.