The evidence that Stephen Ellis refers to about Mandela’s membership of the Communist Party is pretty definitive. It is based upon recollections of party members some of whose testimony is derived from first hand experience. Joe Matthews was a member of the clandestine SACP’s leadership in the late 1950s and early 1960s. He told the Cape Town based Russian researcher Irina Filatova in 2004 shortly before he died about attending central committee meetings at which Mandela was present. This corroborated statements taken by police from party members they had arrested in 1961 and 1962 which hitherto historians have dismissed as unreliable. The Jack and Ray Simons’ papers held at the University of Cape Town include minutes of an SACP meeting in Lusaka in 1982 at which a member of the Central Committee, John Motshabi, reminisced about Mandela’s recruitment, referring to events that would date it to 1961. So, there are independently elicited statements from a range of authoritative sources that indicate that Mandela belong to the underground party in 1961.

Does it matter? It certainly did 50 years ago. In his court testimony Mandela said that he did not become a Communist, though he qualified his statement. If by a Communist was meant a person who was a member of the party and who believed in the theory of Marx, Engels, Lenin and Stalin and who adhered strictly to party discipline, then, Mandela claimed, he had not become a Communist.

This may not have been a total misrepresentation. Party member or not, Mandela remained an independent personality , temperamentally resistant to the rigidities of organisational “discipline”. In his pan - African travels between January and July 1962 amongst the various politician she met in different parts of the continent Mandela encountered a widespread anxiety about the ANC’s links with its Congress Alliance partners and its connections with the Communist Party. He would return to South Africa with the message that the ANC should play a more assertive role in the Alliance and that it should project a more obviously “African” profile, an argument that he had already tried out on Yusuf Dadoo and Villa Pillay, SACP leaders in London, who had reacted very unfavourably. From the outset, then, Mandela was prepared to question prevalent orthodoxies.

Much later, in the late 1970s, in prison, Mandela would fall out with Govan Mbeki over the issue of whether the ANC and the SACP should remain separate organisations. By this stage Mandela was willing to maintain arguments that had he been outside the country would have put him completely at odds with party leadership, on whether to cultivate relations with homeland politicians such as Chief Buthelezi, for example, an option that Mandela favoured. In his first autobiography, written secretly in prison in 1975, Mandela included a critical review of Umkhonto’s initial campaigning, suggesting that the new body had “drained the political organizations of their enthusiastic and experienced men”. This was not a view welcomed in exile circles and in London SACP leaders shelved Mandela’s manuscript, vetoing its publication. This was an extraordinary decision really, given that this was a time when the ANC badly needed the kind of public attention that the appearance in book - shops of Mandela’s text would have generated. So Mandela joined the party in 1961 but had probably in his own mind decided to distance himself from it by 1962: certainly his African travels would have alerted him to just how damaging to the ANC any public knowledge about his party affiliations would have been. Later, half - way through his prison term, he was taking up positions that in the world of exile politics would in the case of lesser personalities would have caused their expulsion from the party.

So how does our new knowledge of Mandela’s Communist party membership alter our understanding of the more general history of which it is party? For Stephen Ellis it suggests that we should be very critical of the ways in which the ANC’s decision to embark upon armed struggle have been represented conventionally. Mandela’s recruitment into the party in 1961 might mean that at the time that he was still campaigning for a national convention and calling for a massive strike to back this demand he was in fact acting in bad faith. He and the other key figures within the ANC’s top leadership had already committed themselves to guerrilla warfare, through their membership of the SA CP ,itself resolved in favour of “armed struggle” at a party conference in December 1960.

For this to be the case we would have to be certain about when Mandela joined the party. Motshabi’s recollections recorded in the minutes of the 1982 meeting suggested that Mandela was recruited at about the time that J B Marks was being proposed as party Chairman: given that Marks was elected Chairman at the Party’s 1962 conference, that would suggest later rather than earlier in 1961. In which case Mandela might have joined the party after he had already made up his mind about armed struggle, a decision that he has generally maintained he made at the beginning of June. We do know that between June and December Mandela remained in hiding in a series of safe houses arranged by SACP members, passing much of the time reading a series of classic authorities on insurgent warfare. This seems the most likely time for him to have joined the party. He already had a rough grasp of the essentials of Marxist ideas, acquired again through reading and discussions with Moses Kotane at the beginning of the 1950s: as he noted much later in his 1994 autobiography at the certainties offered by “the scientific underpinnings of dialectical materialism” were for him at this time powerfully compulsive.

If Mandela was a committed and disciplined Communist at the time of his trial that might also mean that we have to read his famous statement from the dock as an expression of political expediency rather than a statement of principle. Again, though, we cannot be certain. In fact there is a wealth of evidence to show how through the 1950s Mandela’s political thinking was subject to competing ethical and emotional and strategic imperatives, as was the case with the movement of which he was part. In particular, in Mandela’s politics, the tensions between liberal and Marxist modernisms and his affection for and loyalty to patrimonial social order would never be completely resolved, not in 1961, not in 1994, and, who knows, not today neither. Nor had the Communist Party wholly made up its mind about the use of armed force by the end of 1960.

Now that we have so much more knowledge than we used about who was and who wasn’t a communist the grounds might seem that much stronger for viewing the ANC as simply a transmission agency for directives emanating from a vanguardist elite. There remain, though, good reasons to reject such a view, not least because the ANC itself was far from being an effective hierarchical body: in the 1950s it was much more a composite of different regional political cultures. Also we know too little about the party and its own life during this period, though from what we do know it was much less internally resolute than is sometimes suggested in the published recollections of “stalwart” veterans.

It is certainly true that the party’s influence over ANC leadership decision - making expanded in exile, with or without the authority that might have derived from Mandela’s political affiliations. Whether that left it so strategically bereft in 1990 is not so obvious. Communists themselves had been pulled between a “hard” insurrectionary predisposition and a more conciliatory acceptance of the compromises that might have to arise from negotiation: Thabo Mbeki was in the latter camp and under his patronage planning for a post - apartheid constitutional order had already made some headway by the time of Mandela’s release. And, as we have seen, some of the most pragmatic navigators of South Africa’s reintegration into the global economy have been yesterday’s comrades.