Home page > 3. Debate > Building new parties of the left > The experience of building DS and the PT, from 1979 to the first Lula (...)
Save this article in PDF Print article Printable version


The experience of building DS and the PT, from 1979 to the first Lula government

Monday 28 May 2012, by João Machado

João Machado was a leader of DS throughout the period in question. This text draws on parts of a written contribution to the Seminar dedicated to Daniel Bensaid, held at the IIRE in January 2012. However, Bensaid’s role in DS’s discussions was not singled out just for that reason – he really played a crucial part, especially in the later period.

Creation of the PT

The idea of creating the PT (Workers’ Party) in Brazil was put forward at the end of 1978 and it began to be organized in 1979. With the agreement of its Manifesto in February 1980, the move to legally constitute the party was formally launched. At that time (the last phase of the military dictatorship installed in 1964) there were only two legal parties in Brazil: ARENA which supported the government and the MDB) Brazilian Democratic Movement) which was in opposition.

The party’s initial documents already spoke of socialism and attacked capitalism, but its central idea was to build a party independent of the bosses, of the workers, which would express their interests and not manipulate them. Its Founding Principles used – without mentioning the source – Marx’s famous formulation for the I International (“the emancipation of the working class must be the act of the workers themselves”).

In large part the creation of the PT reflected the strike movement in Brazil at the end of the dictatorship, and was pushed forward by a trade union current called “authentic trade unionists” (which basically meant class-struggle trade unionists). From the beginning, however, several revolutionary political organisations also took part alongside these trade unionists: some had an influence on its initial launch – especially two organizations from the Trotskyist tradition, the Socialist Convergence (CS – Morenoist) and the Trotskyist Workers’ Fraction (FOT – a small organization coming out of a split from the Lambertists). The FOT had influence mainly because its leader, Paulo Skromov, President of the Leather Workers’ Union, was one of the main leaders of the movement to launch the PT in 1979 and the first part of 1980.

For its part, DS (Socialist Democracy), which would become the Brazilian section of the FI, was only founded officially with this name, at the end of 1979, when the movement for the PT was already under way, but its members took part in the movement even before DS came into existence. It played a decisive part in the initial organisation of the PT in two states (Minas Gerais and Rio Grande do Sul), and later broadened its national presence. To understand the importance of those two states, we could say that São Paulo, Brazil’s most populous and industrialised state, was always by far the most important base for the PT, followed by the three states of Minas Gerais, Rio de Janeiro and Rio Grande do Sul.

On 31 May and 1 June 1980, the PT’s National Launch Meeting was held, and the process of signing up members began.

At the time, the procedures for legally registering a political party in Brazil were extremely difficult. A party could only be registered if it existed in several states with a minimum number of members that totalled several hundred thousand. To achieve this, the participation of the revolutionary organisations was decisive, as was that of a growing number of activists from Catholic Church Grass-roots Communities and other left-wing Catholics. So although the influence of the “authentic trade unionists” (especially Lula) was dominant in the leadership, sectors further to the left had considerable weight from the start.

It is important to point out that all activists in Brazil who were to the left of the parties of the bureaucratised international communist tradition (i.e., to the left of the PCB – the official communist party – and the PC do B – originally Maoist, then “Albanian” up until 1989) joined the PT in those early years, although not necessarily right at the launch.

In August 1981, the PT completed the membership and organization procedures required to obtain legal registration, and held its “1st National Meeting” (i.e. its first congress).

The year 1983 saw the creation of the “Articulation” (also known as the “Articulation of 113”, because its manifesto had 113 signatures). This was a block formed by Lula and other leaders to ensure themselves a majority of the PT (at the time it had about 60% of the delegates to the National Meetings). Apart from the trade unionists close to Lula, this block included personalities, members of parliament originally elected by the MDB, left Christian activists and others from organizations of the revolutionary left that had dissolved, and also some that had not dissolved. From then on this block was to be a constant feature, under various names (Articulation, Unity in Struggle, Majority Camp) and with varying compositions. It is important to be clear that its political positions were always quite varied; up until 2003, it always included sectors that were further to the left alongside others with social-democratic positions and, from the 90s, some with clear social-liberal politics. The influence of the Cuban CP within Articulation was strong throughout the 1980s. Until 1989, Lula was in the “centre” of this block. After that, he came to lead its most right-wing sector, which favoured more integration into the bourgeois electoral system and the existing “political order”. This position of Lula’s, however, was not publicly apparent – he avoided getting involved in the internal party debates.

The creation of the PT was, in a sense, completed, with the founding of the CUT (Single Workers’ Confederation) in 1983. In spite of the name, it was never the only one, but it did become the main Brazilian trade union confederation. It was always closely tied to the PT. It was always led by a block called “Trade Union Articulation”, linked to the Lula wing of “Articulation” in the PT, which always bit more than half of its leadership.

The creation of the PT gave Brazilian workers a political expression of their own, just as their social weight was on the increase; it was a move towards political independence for the class. The PT also defined itself as pluralist, with a reasonable degree of internal democracy, and as a socialist party.

Growth and changes in the PT

It is true that, from the beginning, the PT had some important weaknesses, in particular the lack of any depth in its discussion of what it meant by socialism and its very loose organisational structures. Nonetheless, it was able to grow and consolidate its position as the main reference point of the Brazilian left up until 1989, when Lula almost won the presidency as a clearly left-wing candidate. The PT also became an international reference point. As a result, much of the Brazilian left downplayed the party’s weaknesses and became unable to imagine any future outside of the PT.

That same year of 1989 saw the beginning of a drastic change internationally, with the collapse of the so-called “socialist camp” and the big crisis of the left that followed. From then on, the neoliberal wave intensified around the world, including in Brazil. This had economic and political-ideological consequences, as well as a profound social impact: wage earners and the social movements were weakened. At the same time, in the case of Brazil, from 1988 the PT’s insertion in existing institutions became increasingly significant (not just in parliament, but in local municipal and, later, regional state governments). Naturally, this increased the pressure for the party to adapt to bourgeois institutions – which was all the more difficult to resist given the PT’s relative weakness politically and ideologically. We should also remember that the PT’s weight in the trade union apparatuses brought with it bureaucratic pressures from the party’s birth, and that from the end of the 80s the “Trade Union Articulation” was generally to the right of the “Articulation” in the PT; the pressures for the PT to adapt to the bourgeois order were stronger in the trade union area than in political and parliamentary arena.

Lula’s electoral defeat in 1989, coinciding with the beginning of the big crisis of the international left, marked a turning point for the PT. Lula and his circle came increasingly to believe that it was possible for him and the PT to win by moderating their programme and broadening their alliances. In fact gradually, and especially after the defeat of Lula’s second presidential bid in 1994, the core electoral strategy came to be based on a complete denial of the class antagonisms that are inevitable in capitalist society. This ruled out the alternative route to electoral victory – based on greater popular mobilization, extending what had already been achieved in 1989, and understanding elections as a moment of confrontation between class alternatives.

From the beginning of the 1990s, Lula and his immediate circle were among those pushing for “moderation” and a diluting of the PT’s socialist references, and seeking ever stronger links with bourgeois sectors. The PT was losing its rebellious edge and becoming increasingly integrated into the existing order. Many neoliberal ideas began to take hold among its leaders. At each national election, the party appeared less to the left. Each presidential defeat (Lula lost again in 1998, after the defeats of 1989 and 1994) was taken to show that “moderation” and the “broadening” of alliances had still not gone far enough.

This political evolution had an organizational counterpart. While in the early days of building the PT there was a concern with organizing branches and having a party with a militant intervention, this was gradually overtaken by the idea of a party organized around elections. In the 90s it became common to have campaigns that were organized “professionally”, that is fundamentally by people who were paid to do it, rather than by political activists volunteering their time.

On the other hand, the evolution of the PT as a whole became very uneven and internal political differences increased hugely. In the 1990s, there was in general a clear polarization between a left and a right in the PT, especially after 1993 when the majority block – the Articulation – split into a majority (which took the name of “Articulation-Unity in Struggle”, the sector of Lula and José Dirceu) and “Left Articulation”. This split, in fact, made it possible for a short period (between 1993 and 1995) for the Left Articulation, along with other left currents, especially DS, to form the majority of the PT’s national leadership. This left majority was possible in spite of the left losing one of its important currents, the Socialist Convergence, in 1992 (formally the CS was expelled from the PT for failing to accept the rules on internal tendencies in the party), and even though part of the left had moved rightwards after 1989.

The most right-wing sector of the party became more clearly defined with the creation of the PT’s “Majority Camp” in 1995. Bit by bit, most of this “camp”, especially its leadership, were altering their social allegiances and their political references: as they developed closer links with business, they distanced themselves from socialist positions. During the preparations for the PT’s 2nd Congress (1999), there was even an attempt formally to abandon the reference to socialism (José Dirceu declared that “socialism was a ’living corpse’ hanging round the PT’s neck”). However, this attempt did not succeed, because it did not have majority support even within the “Majority Camp”. Nonetheless, the fact is that a growing section of the PT leadership came to identify with the new tendencies in international social democracy, and therefore became integrated into the sphere of neoliberal ideology.

On the other hand, in the case of sectors more to the left of the PT, things evolved in a very different direction. These sectors were also affected by the crisis of the international left and by the greater institutional pressures, but not in the same way. A part of the left, as we have mentioned, moved rightwards. But among those who stayed to the left, what predominated was resistance to the pressure of neoliberalism and the maintenance of socialist ideological references. The creation of Left Articulation in 1993, was an important expression of this resistance.

It is worth noting that, even while bits of the left regularly shifted to the right from the beginning of the 1990s, the left of the PT continued to exert considerable influence on the party’s positions. At the National Meetings in 1995 and 1997, the division between right and left was almost half and half. What is more, even within the “Majority Camp”, the evolution was quite varied. Many of its supporters continued to oppose social liberalism and identify with socialism. Furthermore, during the government of F.H. Cardoso, its role in opposition forced the PT as a whole to distance itself from neoliberalism – which to a considerable degree concealed the changes that were under way.

The evolution of DS through the 1980s

DS, as we have said, was created in 1979, basically out of the unification of two groups of militants (the biggest in Minas Gerais, the second in Rio Grande do Sul). Some members coming from the POC-Combate (which had been linked to the International Majority Tendency of the IV International years before) took part. Altogether, in the whole country, it had 60 members.

DS already had links with the IV International – two representatives of the IV International took part in the founding conference, Francisco Louçã and Socorro Ramirez, who a few years later would leave the FI. The new organization, even before being officially founded, had had a representative as observer at the XI World Congress of the FI, in 1979. However, it only formalized its request for membership of the IV International in 1984, and was recognized as the Brazilian section of the FI at the World Congress in 1985.

In 1981, a small organisation coming from the CS fused with DS (in reality, it joined DS), and the organization took a new name, ORM-DS (Revolutionary Marxist Organization-Socialist Democracy). In 1982, the FOT (which had taken the name CLTB, Brazilian Trotskyists’ Liaison Committee) joined DS (by that time, Paulo Skromov had lost the central role he had played in the PT leadership at the time of its no foundation in 1979-80; he was to leave DS a few years later).

From its founding congress DS adopted a position of combining building itself with building the PT. It was therefore not a question of some sort of “entryism”. It was a question of building on two levels: building the PT as an independent workers’ party (which meant seeing it as a party and not as a “legal front” or any other such formulation) and building DS as the section of the IV International, understood as part of the PT and not in competition with the PT. We characterized the PT as a party whose future was open, whose orientation had been in dispute since its foundation, but the outcome of which was not predetermined. It could evolve and turn into a revolutionary party, but such an evolution would depend on the most left-wing sectors of the party winning the battles over its political direction. This general line was first synthesised clearly in 1980, in a pamphlet with the title, The PT and the Revolutionary Party in Brazil.

From the beginning, relations between DS and the leadership of the IV International were strong. Apart from Francisco Louçã, who returned to Brazil several times after 1979, Daniel Bensaid and Michael Löwy took part in many of DS’ discussions and activities, as well as other leaders of the FI or its sections. Michael played an important part in discussing the the basic, initial orientation for building DS; Daniel, on the other hand, was the non-Brazilian FI militant who was most present in Brazil between 1980 and 1990, and again between 2002 and 2004.

Between 1980 and 1990, the FI sought to strengthen its organizations in Latin America, organizing annual “Latin American Political Bureaux of the Latin American sections” and promoting trips by various leaders to countries of the region. The two members of the FI Bureau that took part regularly in these efforts were Daniel Bensaid and Charles-André Udry. In this framework, DS developed very strong relations with the Mexican PRT, which in the 1980s was the strongest section of the FI in Latin America.

A particularly important Conference for DS was held in 1988. In 1987, the PT’s V National Meeting had adopted a decidedly left-wing line, and also “rules for internal tendencies” in the party (which aimed to prevent the existence of “parties within the party” but which also extended the rights of minorities). To comply with these rules, DS replaced its previous statutes (a word that suggested a party) with “basic norms”, which outlined essentially the same organizational principles. At the same time, the organisation’s name went back to just “Socialist Democracy”, as it had been at the founding conference. Apart from being a more appropriate name for a party tendency, this was in practice the name the organisation had continued to use all along.

Another important decision resulted from the positive balance sheet that DS made of the PT’s evolution. We agreed a characterization of the PT as a “revolutionary party in construction”. Comrades of the Mexican PRT, in particular Sergio Rodriguez, who was the PRT leader most active internationally, and who had most contact with the Brazilians, suggested that we call the PT directly a “revolutionary party”, in order to make clear our commitment to the party and our open approach. Up until then, we had spoken of a “mass workers’ party”, an “independent working-class party” or a “class party” – but not a “revolutionary party”. What we said was that the PT could become a revolutionary party, depending on the general course of the class struggle in Brazil and on the disputes within the party. Faced with the suggestion from the Mexican comrade, the DS leadership judged that characterizing the PT directly as a “revolutionary party” would be an exaggeration (among other reasons, because forces that were clearly not revolutionary continued to have a lot of influence in the party, including in its majority articulation), but opted instead for the formula, a “revolutionary party in construction”, as a way of strengthening our identification with building the PT as such.

It is interesting to record that Daniel Bensaid (who was not at that DS Conference) later questioned this formulation. Basically, he said that a “revolutionary party in construction” didn’t mean very much and that, in terms of reinforcing our commitment to the PT, there was no advantage in this characterization. On the other hand, the formula risked disarming DS members in relation to the problems the PT would still face: if we judged that the PT could become a revolutionary party (as we always had done), that meant we also judged that it could go in a different direction.

The argument can be broadened out. To speak of a “revolutionary party in construction” tends to mean we pay less attention to the opposite positions that exist within the party and, what is more, obscures the qualitative change that would be necessary for a broad party of the class to become a revolutionary party as such. In other words, it minimizes the qualitative change involved in moving from a workers’ party that defends their interests to a party that is organized not only around the need to fight for a different (socialist) society, but also around the need to overthrow the capitalist state through a revolution. This approach implies a clear understanding of the limits of bourgeois institutions and the struggle within them, something that was never shared by the PT as a whole.

The resolutions of DS would not cease to address these questions, in the years that followed, just because we were now talking about a “revolutionary party in construction”. But that synthetic formula had a strength of its own, ad exerted more influence than the analyses that accompanied it. The expression, a “revolutionary party in construction” was later abandoned, but some confusion remained over whether we saw the PT as a “revolutionary party”. Many members of DS got used to thinking that there would never be a conflict between their DS (and IV International)-identity and their PT-identity.

This confusion was reinforced by a common interpretation of the significance of changing the organisation’s name back to DS (in place of ORM-DS) and the adoption of those “basic norms” to fit in with the PT’s “rules on internal tendencies”. Many of our members thought the change was more than a formal one – that the nature of DS had changed. The intention of those who drafted the resolutions and the “basic norms”, to make a purely formal change, without altering the nature of DS’s relations with the PT, was not clearly understood by many.

At the end of the 80s, immediately before the shift in the PT’s evolution began after 1989, DS had about one thousand members, with the support of a larger number of PT members. In general, DS got about 10% of the delegates to the PT’s National Meetings (congresses). It had considerable importance in the left of the PT, and in the party. It developed significant social roots (especially in Rio Grande do Sul; in Minas Gerais, its initial stronghold, it was weakened by a voluntarist policy of moving members to other states), it had an important presence in the CUT (the trade union central), and it maintained its strength in the student movement, where many of its members had come from in the beginning. It began to have a group in parliament (especially with the election of Raul Pont as a member of the state assembly in 1986 and as a federal deputy in 1990). On the one hand, it was an organisation very closely identified with the IV International and its positions. In fact this identification had grown stronger. On the other, the PT in that period had been evolving to the left, and DS certainly played a role in that. It is surely reasonable to conclude that the policy of building DS as a section of the FI, combined with building the PT as a party, had been rather successful up until then.

DS from 1990 to 2002

It is probably in the period after 1990 (or maybe after 1995), that is after the PT began to move towards less left-wing positions and to consolidate its adaptation to bourgeois institutions, even before the election of Lula and organization entering the government, that the line for building DS began to have serious problems.

The PT began to acquire a growing presence in the institutions of the bourgeois state (especially after 1988, when it elected the mayors of three state capitals – São Paulo, Porto Alegre and Vitória) and began to look like a real contender for the presidency of the Republic. This combined with the crisis of the international left and its references, and the mounting neoliberal offensive. Although it was far from a linear process (between 1993 and 1995, as we said, the left won a majority in the PT leadership), the PT moved rightwards, especially after 1994.

The left of the PT also increased its participation in the structures of the bourgeois state. Rio Grande do Sul, apart from being the state where DS was strongest, was also the state where the PT’s institutional presence was greatest (first in the city hall of Porto Alegre, then from 1988 in the state government as well). This did not make the PT in Rio Grande do Sul less left-wing than in the rest of the country – on the contrary, at that time the PT in Rio Grande so Sul was the more to the left than anywhere else (which can be explained by the participation of both DS and the Left Articulation – there the majority of the old Articulation had stayed with Left Articulation). Curiously, the PT in Rio Grande do Sul was both the most “institutional” and the most left-wing in the country, throughout the 90s and at the beginning of the 2000s.

There is one thing from this period that should be recalled. For some months in 1994, Lula was ahead in the opinion polls for the presidential election. There was never a debate in the FI about whether DS should take part in a government. But it was something that worried Ernest Mandel. In an international meeting he called my attention to the risk involved in such a hypothetical participation, since he thought it unlikely that Lula, in government, would adopt a left line of confronting the bourgeoisie and imperialism. At the time, I was not convinced. It was a time when the PT had a majority of its leadership from the left of the party (and DS was part of that party leadership). The dominant view among the leadership of DS was that a Lula victory would trigger a big class confrontation, both because of the general situation in the country and because of the general views that prevailed in the PT, even if this might go against the wishes of Lula himself.

The discussion on taking part in a possible Lula government did not continue; from the middle of 1994 it became clear that Lula would not win. The topic would only return at the end of 2002, when the balance of forces was already much less favourable to the left.

There was another discussion in the FI around the 1994 Brazilian elections. In his report to the International Executive Committee that was preparing the 1995 World Congress, “A New Historical Epoch”, Daniel Bensaid pointed out that the PT had fought the 1994 elections with a more moderate programme than that of Popular Unity in Chile – and with the agreement of the members of the Brazilian section.

In the PT we’d had a lot of arguments about the programme, but the left had not proposed an overall alternative line; the programme adopted was a compromise between the left that was then in a majority in the leadership, and the sector around Lula himself. With hindsight, this may have been a mistake. The left’s majority on the leadership was, in part, illusory: Lula continued to be the dominant political influence in the party. It is impossible to tell what would have happened if Lula had won, but the agreement we made then on the programme certainly contributed to obscuring for us the extent of the differences that already existed within the PT.

All of this, of course, needs to be seen in the context of the time. In fact, the subject of Daniel Bensaid’s report was precisely the beginning of a more difficult period for the left, on the defensive before the powerful neoliberal offensive.

On the other hand, that report, “A New Historical Epoch”, became from then on an important reference. Various texts by Daniel Bensaid and Michael Lowy were translated and distributed; several of them were published in a collection, Marxism, Modernity and Utopia, edited and introduced by José Corrêa Leite (Sao Paulo, Editora Xama, 2000), which became central educational tool for DS.

Since at least the second half of the 1990s and the beginning of the 2000s, many of us in DS were aware of the importance of this “institutionalization of the PT” (that is, the process of its adaptation to the bourgeois institutions of the Brazilian state in the course of the 1990s), and also of the fact that this process had affected the left of the party, DS included (indeed, DS had acquired very significant institutional weight). However, at the beginning of the 2000s, we believed that the impulse from the new internationalism, of which the World Social Forums were one of the most visible expressions, could be strong enough to restore the revolutionary perspectives of the left – in particular of DS – and help to reverse the dangerous process of adaptation. Daniel Bensaid, along with other members of the FourthInternational, shared this view, as can be seen in the optimistic vision of the Forums’ mood that he laid out in Une Lente Impatience.

The fact is that in that period, although our analyses (including some reflections by leaders of the IV International) pointed out the worrying evolution of the PT, we did not make any basic changes in our line for building the organisation.

Lula’s election and the DS majority’s break with the IV International

The election of Lula to the presidency, at the end of 2002, brought things to a head. It was achieved with a series of fresh adaptations to the logic of bourgeois state institutions, by Lula and the majority of the PT leadership: the choice of a leading businessman as the vice-presidential candidate, offering the “markets” guarantees that “contracts” would be honoured, etc. The left of the PT, especially DS, opposed this – the PT member who symbolised this resistance, throughout 2002, was Heloisa Helena of DS. Her candidacy for governor of her state, Alagoas, was blocked by the PT’s national leadership in order to facilitate its alliance with the Liberal Party.

The relation between DS and the Lula government was then posed, at the end of 2002. This would be one of the main concerns of FI leaders, and of Daniel Bensaid in particular. Shortly after Lula’s victory was confirmed, he phoned me to express his concerns (I was his main point of contact in Brazil). He pointed out that it was unlikely the Lula government would confront the ruling classes or carry out any deep reforms, in a way that might justify participation by the left, especially by DS – an argument similar to Ernest Mandel’s years earlier. But if I was doubtful in 1994, that wasn’t the case in 2002. I told him I was convinced that the Lula government would be even worse than he thought, that I was totally opposed to DS taking part in it, but that there would be a hard discussion in DS. Given the history of relations between DS and the PT, and the general mood produced by Lula’s election, it would not be easy to refuse to take part in the government.

There began a process that would test the revolutionary coherence of DS, as well as its relations with the FI and indeed the very role of the FI as an international revolutionary organisation. The FI did not practice international centralism, and did not take positions on questions of political orientation in individual countries. It did not pretend to be a “World Party of Revolution”, as the III International had been and as it had itself aimed to be in its early years. But the simple fact it saw itself as an international, revolutionary socialist organisation, albeit structured as a centre for reflection and exchange, and as a network of sections, meant it had to play a real part in the discussion of an issue with international implications. One way of doing this is for different members to express their positions and concerns. Another, which we accept for questions of a programmatic nature, is for the international bodies to adopt positions that differ from those decided by the section involved; the sections should publicise these positions, but they are not obliged to implement them.

In spite of the great public impact of Heloisa Helena’s stance, and the internal opposition of a part of its leadership, DS decided in favour of taking part in the government. There was, however, a kind of compromise, which took the objections into account: participation was tied to a “dispute over the orientation of the government”, and it was emphatically stated that there might have to be a break with the government in the future. What is more, in the months that followed, we made various compromises in the leadership to try to keep the discussion, and DS, going. For those more to the left in DS, those most opposed to taking part in the government, these compromises were justified because we were convinced that as the process unfolded the nature of the Lula government would become clearer and clearer.

In January 2003, many members of the IV International were in Porto Alegre during the WSF, including Daniel. He made an interesting report on the discussions we had there:

In January 2003, the atmosphere at the third Forum was very different from that the year before. The PT had just lost the state government in Rio Grande do Sul, while Lula had won the presidential election with more than 60% of the vote (in the second round – JM). After twenty years, not such a long time after all, the São Bernardo machine tool operator had become Latin America’s first working-class president. His victory was that of the PT, a party that had come from nowhere at the end of the 70s. It was also, in part, our victory. The new government was a coalition. The PT had the lion’s share, but it was flanked by uncomfortable and compromising allies. (…) Our comrade Miguel Rossetto carried the heavy responsibility of the Ministry of Rural Development and Agrarian Reform, alongside a Ministry of Agriculture given to a direct representative of big landed interests.

Much of the time was devoted to meetings with our Brazilian comrades who were coming together for the first time since the government was formed. For some of them, not surprisingly, it was still a moment of electoral illusions. But this ambiguous victory was full of contradictions. While urban social struggles had been on the wane for a decade and the PT had just suffered some worrying setbacks (including the loss in Rio Grande do Sul), Lula had won hands down on the back of a very personal campaign, mainly because people were fed up with the bourgeois parties. To reassure his allies and the markets, he had conducted a moderate campaign, full of guarantees in advance to the IMF, and surrounded himself with people that business could trust. Nonetheless, some comrades seemed to see in his government a sort of institutional dual power, pitting the economic and financial ministries (under neoliberal influence) against the social ministries like Agrarian Reform, Cities and the Environment. A sort of government with two heads and two hearts!

In less than a year, it became clear that the balance between the two was far from symetrical. (…) (Une Lente Impatience, p. 317-8).

Daniel commentary continued to emphasise the important part played by Heloisa Helena in the debate.

Soon after the WSF, during the World Congress of the FI in February 2003, the discussion on the government continued. The Brazilian delegates had, for the most part, a more critical view than DS as a whole. Among the 5 delegates, apart from myself and Heloisa, there was the then state deputy, Luizianne Lins, who along with the rest of the comrades in her state – Ceara – was on the left of DS. Luizianne would later abandon her more left-wing positions, after she was elected mayor of Fortaleza in October 2004, even though her candidacy was won by the left of DS against not only the PT majority but also against the majority of the leadership of DS itself. At that time we reached an agreement among the delegates: we would not conduct within the world congress a debate for and against DS participating in the Lula government; rather we would try to clarify the complexity of the issue and emphasize the conditional character of taking part and the possibility of breaking with it at any moment.

Daniel Bensaid [1] was the FI leader who participated most in the Brazilian debate. He did not speak in the debates on Brazil in the congress. In a personal conversation, however, he told me he thought everything pointed towards an approaching split in DS. I agreed with him. I was convinced of this. But I thought it was quite possible that in such a split those of us opposed to taking part in the Lula government might be in the majority. After all, the Lula government represented a turn against the PT’s traditional positions. It was plainly a bourgeois government, and this should become clear in the months to come. The incompatibility of a revolutionary marxist current taking part in a bourgeois government was something DS had asserted from the beginning, and it would be strengthened by the debates in the FI.

In the course of 2003, Daniel travelled twice more to Brazil, the second time to take part in the DS conference at the end of the year. He also wrote an important article for Rouge on the Brazilian situation, “Fear triumphs over hope” (02.10.2003), which was promptly translated into Portuguese and circulated in Brazil. The title, which indicates its content, was a reversal of one of Lula’s main campaign slogans, “hope triumphs over fear”. As well as making a very severe judgement of the government, the article also takes a stand against the offensive launched by the PT leadership to demand “discipline” from the party’s members of parliament in the vote on pension reform:

The meaning of this disciplinary offensive, at the expense of the pluralism that has been the PT’s strength, is clear: the party must choose between being a political voice for the social movements and a transmission belt to promote government measures in society. What is at stake is the future of a class-struggle party, born out of the huge radicalisation of social struggles from the end of the 1970s.

Its transformation into “new PT”, a sort of bossa nova version of Blair’s “third way”, cannot be achieved in the coming months without strong resistance from “old PT”, for it is government policy that represents the real breaking of discipline, in relation to the resolutions of the last PT congress held in December 2001.

The article was, of course, in support of the positions on the left of DS, although Daniel made a point of keeping dialogue open with the organisation as a whole.

At the same time, various sections of the FI launched an international manifesto against the expulsion of Heloisa Helena and others members of parliament from the PT.

In November 2003, the DS Conference was held. This conference adopted quite a left-wing resolution. Daniel Bensaid represented the FI, and only spoke in the closing plenary. He made clear his opposition to taking part in the Lula government, which of course was seen with some reserve by a section of the delegates.

In December, the PT leadership expelled Heloisa Helena and three federal deputies from the party, in spite of a very broad campaign against these expulsions. In response, these members of parliament, along with some groups of militants who left the PT and other from different backgrounds, launched the movement to form another party (which would later be called PSOL – the Party of Socialism and Liberty).

In January 2004 (27/01/2004), Daniel Bensaïd wrote (in consultation with Francisco Louçã, who was also very involved in the Brazilian discussions) a letter to two leaders of DS, leaving it up to them whether they would circulate it to other members of the leadership. It strongly insisted that it was unacceptable that Heloisa, after being thrown out of the PT, should also be thrown out of DS, and then pointed to the fundamental strategic questions:

- a balance sheet of the Lula government and its future;

- the main pillars of an alternative to the government’s unchanged, social-liberal economic and social policy, and to its international policies, in the current balance of forces at home and abroad.

- the assertion of DS, programmatically and organisationally (as decided by its conference) as the backbone of a left alternative to the government’s postions.

He went on:

If DS is not clear on these questions, it risks drifting from day to day, carried on the latest tide, reduced to impressionist commentary on Lula and the government’s latest initiatives or declarations, instead of developping a clear orientation of internal opposition in the PT. Without this, the aim to build DS as a “bigger”, more autonomous organisation (also passed at the national conference) will remain a dead letter.


I hope that a catastrophe can still be averted.

In February 2004, on another visit to Brazil, Daniel attended a meeting of the DS leadership. However, by this stage the split in DS was irreversible, and the process moved further in the months that followed. There was, however, one last attempt by Daniel and other FI comrades to exert a positive influence on the debate and preserve a common FI framework. In January 2005, Daniel wrote another letter to Brazilian comrades, together with Michael Löwy and Francisco Louçã. Most of the points made were not new – the letter centred on a balance sheet of the Lula government and the need to break with it – but the arguments were more detailed. On the other hand, given the split in DS, which was already underway, the letter suggested maintaining relations between the IV International and the parts into which the Brazilian section was dividing. It recognised the legitimacy of more than one orientation for FI supporters in Brazil:

- To contribute, in the case of those comrades who wish to do so, to building the PSoL, while avoiding the pitfalls of infantile ultraleftism (…);

- To foster dialogue between left-wing currents inside the PT and small independent forces like the PSoL. A certain complementarity could then be established among the critical left inside and outside the PT, avoiding attacks on each other and respecting each others’ different tactical choices. This concerns particularly the comrades of our own current: even if they are implicated today in different choices and dynamics, they should make an effort not to burn their bridges and to keep their future options open.

This alternative presupposed that the majority of DS could accept a position that would cause them many problems in their relation with the Lula government and the PT. In fact it would only be acceptable to the majority leadership of DS if they were seriously considering the possibility of breaking with the Lula government and helping to build a new party (the PSOL).

The letter from Daniel, Francisco and Michael was distributed to DS members during the 2005 WSF, only by cadre who were critical of taking part in the government. The DS majority decided to avoid the issue. They also avoided debating with the two representatives of the international leadership sent to the WSF for just that, François Sabado and Olivier Besancenot.

The position of the letter’s three signatories was backed by a resolution of the International Committee of the IV International (of 27.02.2005) – the first resolution to explicitly criticise the position of the DS majority, which said, in particular:

1) The experience of two years of the Lula government clearly demonstrates this government’s orientation and the policies it is carrying out. This is a coalition government with representatives of capital, dependent on the parliamentary right. It is a government implementing neoliberal economic and financial policies and thus incapable of responding to the essential problems of poverty and social exclusion in Brazil and confrontation with imperialism. These two years also show that the internal dynamic of its policies cannot be changed. (…)

3) In these conditions, policies meeting the demands and requirements of the popular classes - wage increases, creation of millions of jobs, defence of public services, sweeping agrarian reform, a budgetary and fiscal policy in the service of social priorities rather than the financial markets - are being put forward in opposition to the Lula government’s policies.

4) The government’s general orientation turns left-wing ministers into mere insurance policies or hostages for overall policies that are not their own. These two years of experience show clearly that building an anti-neoliberal, anti-capitalist socio-political workers’ bloc is in contradiction to support for and participation in the current government.

5) Since the formation of the Lula government, there have been reservations, doubts or disagreements in the International on the subject of DS’s participation in the government and the modalities of this participation (role in the social movements). Nevertheless, once DS had taken its decision, taking account the arguments put forward by the majority of Brazilian comrades, the International decided at the beginning of this process not to vote on any resolution, and to accompany this experience. (...) The International has thus avoided posing the issue of participation in the Lula government in dogmatic terms, without taking account of the country’s characteristics, the PT’s history and its links to social and trade union movements. After the experience of these two past years, and taking note of what has been laid out in points 1 through 4, there can no longer be any doubt that occupying positions in the Lula government, whether at the ministerial level or in other posts involving political responsibilities, is in contradiction with the construction of an alternative in Brazil consistent with our programmatic positions.

Further to this, on 01/03/2005, the International Committee passed a motion supporting the general line of Daniel, Francisco and Michael’s letter.

The majority of DS refused to organise a discussion of these positions among its members. In April, they held a new Conference, this time without the presence of militants who had been involved in building the PSOL. This Conference adopted an indirect response to the FI: a very ambiguous resolution on “Internationalism in the 21st century”, which represented in fact a distancing of DS from the IV International.

A minority of the members who took part in this Conference did not accept the line of staying in the Lula government and in the PT, and continued to identify with the IV International. A few months later, these members broke with the PT, and the majority joined in building the PSOL (along with other sectors who broke with the PT in September 2005).

This brought to a close one stage of the FI in Brazil (the stage of DS and the building of the PT), and began a new one (reorganisation, recomposition and indeed rebuilding of the Brazilian socialist left, after the blow suffered with the Lula government). Daniel and the FI played a decisive part in ensuring that his new stage could begin in the best possible conditions – even though these conditions have proven even more difficult than they seemed in 2004-2005.

Towards a balance sheet

After a long period of building DS as revolutionary organisation and the PT as a mass workers’ party, we have to recognise that we suffered a grave blow with the beginning of the Lula government. The PT ceased to be an independent party of the Brazilian working class (and became a transmission belt for the government and the state), and the majority of the Brazilian section split from the IV International. Why?

Looking at it from another angle, the majority of an organisation built on the basis of a programmatic identification and many direct links with the FI, which used many texts by FI members and FI resolutions as the basis of its political education, accommodated to a government that had nothing to do with any kind of socialist project.

By the end of 2003, DS had some 2000 members – the tally made at the 2003 Conference. Of these, a little over 500 were linked to the sectors that kept their relation with the FI and broke with the PT between 2004 and 2005. About three quarters of the members stayed in the PT. It is worth recalling, however, that this choice was very uneven across the regions. In Rio Grande do Sul, the state that had almost half the DS membership, and where the members were most involved in local government posts, in the structures of the PT, in parliamentary offices and trade union bodies, almost 90% of the members stayed in the PT.

However, of the almost 500 DS members who broke with the PT, many (almost half) did not remain politically organized after leaving the PT and either never joined the PSOL or left it soon afterwards. What is more, after the poor electoral results in 2006 (poor for FI supporters, because we failed to retain our two members of the federal parliament), we suffered further losses. Some members went back to the PT (this happened especially in Ceara, where in addition to the Lula government, there was the strong pull of Luisianne Lins in the mayor’s office). Others did not return to the PT, but left the PSOL and Enlace (the current in the PSOL that organises FI members) and continued to be active in the social movements, or dropped out of political activity.

The reorganisation of the FI section within the PSOL was therefore weaker than we might have hoped – and than Daniel, for example, did hope, as indicated by his very favourable reference to the revolutionary coherence of many DS members in Une Lente Impatience, published in 2004, when, in the same book he was already drawing a definitive and negative balance sheet of the Lula government. With hindsight, we can see that this reference to the revolutionary coherence of DS members in 2004 was more positive than what later occurred. Like many of us, Daniel overestimated the extent of the break with the Lula government that some DS members were still to make.

Why didn’t more members of DS break with the PT? It is useful to discuss this question, leaving aside the hypothesis that those who stayed in the government were right and that the very idea that a socialist revolution is necessary (and possible) is unrealistic.

Certainly part of the explanation has to do with the general evolution of the class struggle, the impact of the neoliberal offensive on the left, objective questions, and so on. To begin with, building a new party meant losing the political advantages and influence that the PT provided; it meant restarting in difficult circumstances. What is more, after being weakened in 2003-4, and even more in 2005 with the “mensalão” (scandal over political graft), in 2006 the Lula government recovered and increased its support among the people and the social movements, thanks to the impact of its social assistance programme and the improved economic situation.

In this situation, one key question was the fact that DS had, at the time, hundreds of members working full-time for the party, the CUT or for local and regional governments linked to the PT, and from 2003, in federal government too. This was especially true in Rio Grande do Sul state, which had almost half the membership of DS. It was always going to be difficult to resist the attractions of government and the pressure of power; in the concrete situation of DS between 2003-2006, it was even more difficult. Perhaps the greatest confirmation of this pressure and power of attraction was the evolution of Luisianne Lins and some of the members in Ceará, who after winning the mayor’s office in Fortaleza shifted from an apparently firm position of rupture with the PT to the opposite position and an abject realpolitik.

It is certainly true too, that those of us who broke with the PT after Lula’s election made mistakes in the internal struggle, both in the PT and in DS. But for those who want to evaluate the IV International’s project to build a strong revolutionary organization in Brazil, and to contribute to the building of a mass revolutionary party, the most interesting thing is to look at the problems we had before, and which meant that, at the end of 2002, DS was not reasonably prepared (it might be too much to have expected perfect preparation) to face a situation as difficult as that created for the left in the PT by the election of Lula.

There had already been discussion on several occasions in the DS leadership of a possible break with the PT. Formulations like “sectors of the PT cannot be assimilated into a revolutionary project” regularly appeared in DS conference resolutions. But the fact is that for many members a break with the PT was very difficult to imagine, and even part of the leadership had difficulty grasping this discussion. Others simply weren’t prepared to face the difficulties of a hard reconstruction process and of doing politics in much reduced material circumstances, not to mention the material circumstances of their own lives.

So how did we get to that point?

Undoubtedly, as always, we made mistakes. One of these was pointed out by Daniel Bensaid when he criticised the formulation, a “revolutionary party in construction”: an excessive identification with the PT and an underestimation of the conflicts to come. This underestimation continued throughout the second half of the 90s, when the PT slowly lost the more radical characteristics of its early years. Another critical question was the lack of a clearer analysis of the implications of taking part in governments within the framework of a bourgeois state.

For example: we never made a balance sheet, in DS as a wbole, of the body of experiences of DS itself taking part in municipal governments (especially in Porto Alegre but in other cities too), or in state governments (in Rio Grande do Sul, but also, for certain peridods, in other states). We had some discussions on aspects of these experiences – for example on popular participation, and in particular on the “participatory budget” - but we never got as far as an overall balance sheet. In part that was because there wasn’t time: the experience of state government in Rio Grande do Sul ended in 2002, just as Lula was elected president.

Another question we never seriously confronted in DS was the discussion of election campaign finances. However, this is obviously a key question. From quite early on, the PT’s campaigns drew on contributions from businesses. What is more, from the second half of the 90s, and even more clearly after 2001 and 2002 (when two PT mayors were assassinated in circumstances that remain obscure), we had some information that the means of fund raising in some town halls run by the PT were less than “orthodox”. [2]

A useful way of thinking about this is to note that up until 2003-2004, DS members had two basic identities: the PT and the IV International. The latter summed up a more general, revolutionary socialist identity; it was the form that revolutionary and socialist convictions took for DS members. It was what made our political struggle part of an ethical-political commitment that went beyond the day-to-day issues. That formulation we used for a time, of a “revolutionary party in construction”, and the insufficient attention paid to the problems in the PT’s evolution, tended to make us forget that the two identities could come into conflict. In fact, contrary to many members’ assumptions, there was no straightforward and permanent compatiblity between the two identities.

When the two identities came into open opposition, with the Lula government, the PT identity had in its favour a social and material force – in every sense of the expression – that could only be counteracted by a much stronger revolutionary identity, which could have existed only if it had been worked on much more beforehand, and accompanied by a greater emphasis on the PT’s (growing) limitations and by a more thorough evaluation of its experiences in government.

These criticisms do not mean we were wrong to put our efforts into building the PT, nor that our general line, at least up until the 1989 campaign, was mistaken. Until then it was true that the PT was a party with a very clear, left orientation that was evolving in a generally positive direction – and that DS was growing within the PT, as it built it.

What was much more questionable was that we maintained this line of the 1980s almost unchanged throughout the 1990s, even after the defeat in the elections of 1994, which gave a new impulse to the search for “broader” alliances and greater moderation in the party, in order to “reduce the resistance” of the ruling class to the PT.

Even the best of lines cannot be maintained for ever! In fact, as Daniel Bensaid pointed out in Une Lente Impatience, already in 1989 there had been a fundamental change:

The shock wave of the 1980s was far from imaginary there (in Latin America – JM). The extension of the Nicaraguan revolution to Guatemala and El Salvador seemed imminent on several occasions. There were popular uprisings in Bolivia and Santo Domingo. (…) This dynamic was broken. After ten years of war in Central America, the double electoral defeat in 1989, of the Sandinistas in Nicaragua and of Lula in the presidential election, put an end to this promising sequence of events (Une Lente Impatience, p. 296). In Brazil, the shift that began in 1989 was completed in 1994 with Lula’s second defeat in presidential elections. The neoliberal offensive gained maximum force and the obsession with moderation to reduce the resistance of the ruling classes took complete hold of Lula and his sector in the PT. That was the time to rethink our political line and, above all, correct the optimism of before.

This analysis of the Brazilian experience in building a section of the FI inside the PT does not put into question the orientation of building “broad parties” (which, of course, can be quite diverse) in certain circumstances. But it does show how important it is to take account of the concrete situation, especially in terms of deciding how to combine the these two levels of party building, which may come into conflict with each other – as they did in Brazil, when the “broad party” became very broad indeed, to the point where it came to government at a time when social mobilization was in retreat.


[1] An informal leader, since he was not at the time a member of the FI leadership bodies.

[2] The most plausible explanation for the two murders (the mayors of Campinas and Santo André, both important cities) is that the mayors were trying to block or limit schemes for raising funds from companies working for the municipality. These are the explanations suggested by the mayors’ relatives, although they are denied by the PT leadership. The investigations of both cases are still ongoing.