SciELO - Scientific Electronic Library Online

vol.1 special edition author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand



  • Have no cited articlesCited by SciELO

Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google


Cuadernos del CLAEH

Print version ISSN 0797-6062

Cuad.CLAEH vol.1 Montevideo  2006


'Brazil needs to change'. Change as iteration and the iteration of change in Brazil’s 2002 presidential election1


"Brasil necesita cambiar". El cambio como iteración y la iteración del cambio en la elección presidencial de Brasil de 2002



Francisco Panizza

Department of Government, The London School of Economics and Political Science

Translated by María Cristina Cafferatta
from Cuadernos Del CLAEH, Montevideo, n.90, p. 25-44, 2ª série, año 28. 2005.




Luiz Inácio Lula da Silva’s triumph in Brazil’s 2002 presidential election was construed in terms of a promise of radical change, and so against the holders of the status quo. This article argues that in fact a more subtle political game was a stake in the election, a contest over the meaning and limits of change itself. The article examines how the various players Lula da Silva, outgoing president Fernando Henrique Cardoso and the PSDB’s presidential candidate José Serra invoked and iterated a discourse of change to define and redefine the political dividing lines that marked out the electoral dispute and attempted to set or fix the distinctions between their political positions.

Keywords: Brazil, campaign, change, continuity, discourse, election, iteration, Lula, rupture.


El triunfo de Luis Inácio Lula Da Silva en las elecciones presidenciales de Brasil del año 2002 fue construido según los términos de la promesa de un cambio radical y en oposición a los defensores del statu quo. Este artículo argumenta que, de hecho, la apuesta en las elecciones fue un sutil juego político, una contienda sobre los propósitos y los límites del cambio en sí mismo. Examina cómo diversos actores —Lula Da Silva, el expresidente Fernando Henrique Cardoso y el candidato presidencial por el PSDB, José Serra— invocaron e iteraron un discurso de cambio para definir y redefinir las líneas políticas divisorias que caracterizaron la disputa electoral y pretendieron precisar y acotar las distinciones entre sus posiciones políticas.

Palabras clave: Brasil, campaña electoral, discurso político, elecciones 2002, historia política, Luiz Inácio Lula Da Silva, José Serra, Fernando Henrique Cardoso.




Luiz Inácio Lula da Silva’s triumph in Brazil’s presidential election in October 2002 invoked the image of a radical turn in the country’s politics, perhaps only comparable with the triumph of Chile’s Unidad Popular in 1970. Many analysts saw it as a defining moment in Brazilian political history. One of the country’s most renowned intellectuals,the sociologist Francisco De Oliveira, described the election as a landmark comparable only with defining events such as the abolition of slavery, the proclamation of the Republic, and the Revolution of 1930.2 Another scholar, the historian JoséMurilo de Carvalho, claimed that Lula’s victory marked ‘a turning point in the country’s republican history’.3 Political scientist Wanderley Guillerme Dos Santos argued that 2002 saw the country’s ‘first critical election’ in that it was ‘a realignment of the electorate from a pro-establishment to anti-establishment stance’.4

But on closer examination, the nature and scope of Lula’s victory, as well as popular expectations about the new president, are more difficult to pin down than the claims, images, electoral figures and even the winning candidate’s remarkable personal history suggest. Observers of the electoral campaign noted that Lula and the government’s candidate José Serra had similar electoral platforms, both centred on promises of economic growth, export promotion and the creation of jobs. Taking issue with Dos Santos’s analysis, Luiz Werneck Vianna denied that the election was fought over alternative views on the country’s future: he pointed out that all the candidates converged on the social-democratic centre and that there were few programmatic differences between Lula and Serra.5

Trying to pin down the meaning of Lula’s electoral victory, a number of scholars have emphasised Lula’s personal attributes as signifiers of the change brought by his election as president. Perry Anderson points out that Lula embodies a popular life experience and ‘a ‘‘bottom up’’ trajectory of social and political struggle without equal for any other contemporary head of government’.6 Similarly, Murilo de Carvalho bases his claim about the significance of Lula’s victory on the Partido dos Trabalhadores’ (PT) candidate’s popular origins and biography that makes him an outsider (um estranho no ninho) to the elite that had ruled Brazil since independence.7 But, the powerful symbolic charge of Lula’s persona is no less ambiguous than his electoral discourse. It is true that Lula’s life story combines a narrative of political struggle and personal success.8 For many of his followers he became an iconic image of the struggle of the Brazilian people for their rights as workers and citizens in one of the world’s more unequal countries. But the retirante (internal migrant from the northeast) turned lathe operator and union leader also came to symbolise the social mobility of the country’s modern working class, a condition taken by former president Fernando Henrique Cardoso as a token of the democratic nature of Brazilian society, conveniently forgetting the country’s vast socio-economic inequalities.9 And noting that both Lula and Serra had their personal roots in the industrial suburbs of São Paulo (São Bernardo and Moóca, respectively), José de Souza Martins remarks that for almost a century São Paulo’s industrial heartland was the focus of one of the country’s most successful processes of social mobility and that to that extent, Lula and Serra alike embody a bourgeois ideology of social improvement through hard work, family life and education.10

What then is the meaning of the political change symbolised by Lula’s electoral victory? How can it be conceptualised, particularly when the PT’s economic programme had so much in common with that of the government’s candidate, José Serra? Is it possible to claim that Lula’s triumph represents a foundational moment in Brazil’s history and simultaneously concede that not much is likely to change or that any change will be at best limited and piecemeal? Is Lula an anti-establishment figure or is he the best embodiment of a tradition of social mobility, ultimately legitimising the status quo? And if Lula’s government represents the dominated taking history in their own hands, how could prominent members of the country’s traditional elite andbusiness class have endorsed his candidacy, and his triumph be digested so easily by the markets?

A possible answer to the above question could follow the familiar themes of co-optation and betrayal, or marketing and presentation, supplanting the party’s historic values. Perry Anderson acerbically notes that Lula’s new ‘peace and love’ electoral image was, ‘in anticipation, the language of defeat’.11 An alternative explanation would focus on the constraints imposed by the markets, which made it necessary for the PT’s presidential candidate to moderate his tone and now severely limit the new government’s scope for implementing an alternative economic policy. A third explanation would see Lula’s avoidance of hard policy choices during the campaign, and even his change from an angry union leader to an Armani-clad suave politician, as little else than a marketing ploy to reach beyond the PT’s core supporters.

The above explanations are not mutually exclusive and contain elements of truth. However, they are also less than convincing and do not give the full picture. Did Lula betray his party radical roots? While many PT supporters are uneasy with the new government’s early adherence to fundamental tenets of the previous government’s economic programme, claims of betrayal are not only premature but also unfounded. There are cases in other Latin American countries of presidential candidates campaigning on an anti-neoliberal discourse only to enforce harsh neoliberal economic policies once in government – Alberto Fujimori and Carlos Menem are cases in point (Stokes, 1999) but this was not the case with Lula’s electoral campaign. As will be examined in detail below, Lula made ‘change’ the nodal point of his campaign; but he also made no secret of his commitment to the International Monetary Fund’s (IMF) agreed targets for the budget primary surplus, thus committing himself to tight control of public expenditure. He made it clear that his government would honour the country’s contracts (i.e. would not default on the external debt) and would preserve economic stability, all of which were important elements of the Cardoso government’s economic model. No promises were made on traditional left-wing policies, such as increasing public spending, the renationalisation of privatised utilities, or the return to the policies of state economic intervention. Even the financial sector, the bogeyman of the left during the Cardoso years, was spared in the PT’s electoral campaign, as Lula met with bankers to reassure them that they had nothing to fear from a future PT-led government. In short, if the PT’s moderate programme arguably represented a betrayal of the party’s socialist roots, at least Lula did the ‘betrayal’ during the electoral campaign, instead of waiting to get into government a' la Menem or Fujimori.

Was Lula a prisoner of the markets? Markets constraints are all too real, and clearly the PT’s leadership was acutely aware of the power of the markets during the electoral campaign. But if he had agreed that, as some left-wing economists have argued, the commitment to the market-friendly, fiscally tight and high interest rate policies of the Cardoso government would inevitably lead to economic failure and political disaster, would not Lula at least have considered challenging market orthodoxy rather than seeking to placate the markets during the campaign and beyond? Not even politicians go with their eyes open to the slaughter. The question is not so far-fetched considering that the shadow of Argentina’s De La Rua, a president for ‘change’ who nonetheless continued the orthodoxy of the Menem era with disastrous consequences, loomed large over the PT’s strategists.

Was Lula’s new moderate discourse just a marketing ploy? Marketing is part of modern politics and by all accounts Lula had a brilliant strategist in Duda Mendonça. But Lula’s politico-electoral strategy cannot be reduced to a matter of political marketing. A marketing strategy based on selling a product on completely spurious grounds is very difficult to sustain, particularly when subject to the relentless attacks that are part of any electoral campaign. And as we will see below, the Serra camp tried very hard to present Lula’s moderation as just TV-deep and failed.

While taking on board market constraints and marketing strategies, a more rounded analysis of the nature of Lula’s electoral appeal needs to examine how change became the key signifier of the electoral campaign. This article argues that while Lula’s working class roots, his denunciation of the Cardoso government’s economic model, and the PT’s radical tradition would suggest that the election was fought on the PT candidate’s promise of a radical rupture with the status quo against the enemies of change, in fact a more subtle political game was a stake: a contest over the meaning and limits of change itself. To this purpose the following section discusses two discursive constructions of change. The article then examines the political context in which the battle for change was fought in the 2002 election. Subsequent sections analyse the various politicodiscursive strategies of PT’s presidential candidate Lula da Silva, outgoing president Fernando Henrique Cardoso, and Partido da Social Democracia Brasileira’s (PSDB) presidential candidate José Serra.12 The analysis shows how each of these actors iterated change to define and redefine the political dividing lines that marked out the electoral dispute and set or closed the distinctions between their political positions. The article concludes that Lula’s electoral discourse successfully articulated the themes of ‘change as rupture’ and ‘change as continuity’ to maintain the hopes of his followers, address the reservations of those who had been reluctant to vote for him in the past, and allay the fears of the markets.


‘Change as Rupture’ and ‘Change as Continuity’

Any promise of change involves a political struggle over the meaning of change itself. One traditional way of understanding political change is to associate it with a new beginning, with an event that represents a complete rupture with the existing order. This imagining of change is usually related to epoch-making historical episodes such as revolutions, the overthrow of dictatorial regimes, or wars of independence. Imagining change as a new beginning may also be part of less traumatic political scenarios. In a non-revolutionary, post-colonial era, change as a new beginning can be imagined to signify a break with a present condition associated with severe economic or political crises, or with governments that are deeply unpopular or have lost their legitimacy. In other words, ‘change as rupture’ is the promise either of a radical new order or of the reinstitution of order in societies facing civil wars, political upheavals, economic and social breakdown, or illegitimate government. It is usually part of the discourses of revolution, populism and class struggle, and operates by dividing the political space into two antagonistic camps: past and present, friends and enemies, us and them, dominators and dominated, the nation and its enemies, order and chaos, the government and ‘the people’. In this discursive operation, the negative pole of the dichotomy (the enemy of change) becomes a radical otherness. As a result, all differences within the existing order are erased, because in oppressing all actors the shared enemy simultaneously renders all of ‘us’ the same (Bowman, forthcoming): we ‘are all equal’ in our antagonism to capitalist oppression, the communist dictatorship, the enemies of the people, corrupt politicians, the ancien regime, economic chaos, the colonial master, the evil empire, or godless America. Thus change as rupture requires that its adherents work to overcome the Other that blocks the promise of the fullness of time and identity that will be achieved in the new order.

But political change can also be invoked not as rupture but as continuity dislocated by a process of iteration. Iteration is the logic that ties repetition to alterity. Jacques Derrida argues that the essential iterability of any sign (linguistic or non linguistic) means that it can always be detached from the chain of signification into which it is inserted and grafted onto alternative chains: ‘Every sign […] an be cited, put between quotation marks; in so doing it can break with every given context, engendering an infinity of new contexts in a manner which is absolutely illimitable. This does not imply that the mark is valid outside context, but on the contrary that there are only contexts without any centre or absolute anchoring’ (Derrida, 1988: 12). Iteration thus involves a recovery that does not leave the original unchanged, as the very act of repeating something incorporates something new, constituting a difference that turns the repetition into some form of re-institution (Arditi, 2003:90).‘Change as iteration’ is not so much the elimination of hard dividing lines between two antagonistic political orders, but the disruption of an existing order by grafting into it new elements that alter hierarchies of power and shift the edges of political frontiers. Thus conceived, change as iteration operates as the anticipation of something to come in the spaces between the political logic of the permanent revolution and the technocratic logic of the end of history. It is a dimension of the practices that make institutions operative and that contribute both to their subsistence and to their transformation through time. In political discourse it is the subversion of literality by ambiguity, of the text by context, of what is said by what is left out, of what is beyond doubt by what is qualified, of what is asserted by what is implied, of antagonism by difference.

Two consequences derive from these imaginings of change: first, continuity can never be erased from the promise of a new order. However, much it presents itself as rupture, all change involves iteration and thus any new beginning always retrieves traces of the past. Secondly, and conversely, by being played out in different contexts, repetition necessarily involves change and opens up possibilities for hegemonic struggles and the dislocation of relations of power. As I will show below, the interplay between change as new beginning (or change as rupture) and change as iteration (or change as continuity) was crucial in the political struggle to define the Brazilian electoral agenda in 2002, and ultimately in the success of Lula’s electoral appeal.


The Changing Political Frontiers of Brazilian Elections

Elections are usually fought on the themes of continuity and change (they are, after all, institutionalised identification contests). However, these themes are played out in very different ways according to the institutional frameworks within which the electoral contest takes place and the political conjunctures in which it is fought. In the case of Brazil, presidential elections are the only true national moment in the country’s political life and thus the only institutional instance in which it is possible to appeal directly to the citizens as ‘one people’ and not just as a conglomerate of fragmented local clienteles. The countrywide and personalised nature of presidential contests allows citizens to identify more easily with leaders who may be committed to a programme of change, as opposed to the conservative political elite that have traditionally dominated local politics, particularly in rural areas, through patronage and clientelism (Roett, 1984). At times of political instability presidential elections have an in-built plebiscitarian element that favours political polarisation and the promise of a break with the status quo (Mangabeira Unger, 1990). However, other institutional features of Brazilian elections undermine the presidential election’s potential to simplify the political space into two opposite camps. The structure of the party system in Brazil means that it is difficult for a presidential candidate to run as the candidate of a single party, leading instead usually to the establishment of multi-party electoral alliances around the candidates. These coalitions are often composed of parties with considerable politico-ideological differences among themselves. And while successful presidential candidates should poll an absolute majority of votes either in the first or second rounds, their parties would normally receive a far smaller share of the votes for Congress: in the past three elections no ruling party has won more that 22% of seats in the Chamber of Deputies.

If presidential candidates need multi-party support to win the election, presidents also need to form large multi-party coalitions (coalition presidentialism) to control Congress, usually including parties of different ideological leanings in the ruling coalition (Abranches, 1993). Moreover, since 1994 presidential elections have taken place simultaneously with congressional and state elections, in which political alliances have very different political configurations from the presidential election. It is not unusual for a party that supports a presidential candidate at national level to be formally or informally allied to a party supporting a rival candidate at state level. So, while presidential elections may favour the division of the political space into two opposite camps, the wider electoral rules of the game and the states’ politico-electoral alignments blur the potential polarisation derived from the nature of presidential elections and transform elections in a complex game of intersected alliances that potentially undermines or even erases political polarisation.

In Brazil’s recent electoral history, strong political polarisation gave the 1984, 1989 and 1994 presidential elections a quasi-plebiscitarian nature, in which campaigns were fought on the promise of change as a break with the existing order or as the restoration of order in the face of political or economic dissolution. The 1984 election was a regime-change, indirect election in which the dividing line between democracy and the military government saturated the political space. What I have labelled ‘change as rupture’ was also evident in the 1988 election, fought against a backdrop of hyperinflation, accusations of corruption against outgoing president José Sarney, and a general crisis of governability. Against this political background, the winning candidate Fernando Collor de Mello fought a populist campaign in which he successfully polarised the electoral space between himself, as an outsider fighting on behalf of the socially excluded, and a corrupt and self-serving political establishment (anti-tudo que está aí, against everything that is out there) (Panizza, 2000a). The 1994 election also acquired a quasi plebiscitarian nature but in a different manner. In this case, the Other, against which a new beginning was posited, was not the political establishment but the disorder represented by hyperinflation and economic mismanagement. The election was centred on the successful restoration of economic order achieved by the highly popular Plano Real (the Real anti-inflation plan) and was won by the plan’s political father, Fernando Henrique Cardoso.

The PT, and Lula as its co-founder and leader, developed into a national political force in the polarised and unstable context of the 1980s and early 1990s. A grassroots party with a socialist-oriented programme, the PT campaigned on behalf of the politically and economically excluded, denouncing the feeble nature of Brazilian democracy, the corruption of its political establishment, the unjust economic order, and the country’s dependence on the diktats of the IMF. The PT was never a revolutionary party, and its radical rhetoric was always tempered with respect for the democratic rules of the game and a considerable dose of pragmatism (Nylen, 2000). But the popular roots of this party without precedent in Brazilian history, and its links with radical social movements, gave its message of change a hard edge that set it apart from the catch-all, centrist, political forces that had traditionally dominated Brazilian politics, and also from the majority of the electorate, as shown by Lula’s three successive electoral defeats in 1989, 1994 and 1998.

Each defeat occurred in a different political context but what they have in common is that they each showed that Lula had failed to persuade a majority of the citizens that he could be trusted to govern a country besieged by political and economic problems. In 1989, Collor de Mello defeated Lula by successfully portraying him as a dangerous radical who would sink the country into political violence and economic chaos. In 1994, Lula was the frontrunner to win the election but failed to appreciate the popular appeal of the Real stabilisation plan, launched just before the formal start of the electoral campaign. His attacks on the Real made it easy for Cardoso to present Lula as a threat to the new economic order, and he was heavily defeated. In the 1998 election, polarisation was less in evidence in the electoral campaign. However, the election was still largely centred on the 4-year success of the Real stabilisation plan and on the danger that a PT government would return the country to the economic disorder of the recent past. Moreover, the election was fought against the background of the mounting economic threat brought by the fallouts of the Asian and Russian financial crises. President Cardoso capitalised on the threat posed by these crises by presenting himself as a reliable statesman who could steer the country through stormy economic waters, in contrast to Lula’s lack of experience and unreliability. So, while less allembracing, the antagonism of the 1998 election was similar to that of 1994: between order, as represented by Cardoso, and the threat of a return to economic chaos, as represented by Lula.

After 8 years of government led by president Cardoso, the 2002 election had all the characteristics of a plebiscite on his government. This was clearly the political dividing line around which all the opposition candidates, including Lula, chose to campaign. In the televised debates the three main opposition candidates, Ciro Gomes (Partido Popular Socialista, PPS), Anthony Garotinho (Partido Socialista Brasieiro, PSB) and Lula da Silva (PT), concentrated their fire on José Serra (PSDB) as the candidate of a government who had manifestly failed to fulfil its promises of high economic growth and social justice. But other political and economic factors blurred the election’s potential for political polarisation. Though Brazil remained a country deeply unequal and dangerously vulnerable to external financial shocks, in the 8 years of Cardoso’s two presidential terms (1995–2001) it had changed from a country characterised by weak democratic institutions, political volatility, and economic disorganisation to a society where politics was more institutionalised and the economy relatively more stable (Panizza, 2000b).

The 2002 election was therefore not fought in terms of regime change as in 1984, of a moral crusade against a corrupt political establishment as in 1988, or of the preservation of order against the threat of hyperinflation as in 1994 and, partially, 1998. Rather it was an election fought within an increasingly institutionalised political system, against a government whose legitimacy was not in question, which by Brazilian standards had low levels of corruption, and which had managed to maintain at least a semblance of economic stability. While there was widespread disillusion with the government, the popularity of the outgoing president remained relatively high at around 26% during the electoral campaign, and president Cardoso was voted in a poll the most successful president in modern Brazilian history.13 While in the past the political system had been shown to be vulnerable to populist outsiders or pseudo-outsiders such as Collor de Melo, 2002 was an election in which all the main presidential candidates were political insiders and were backed by established political parties.


Lula’s Politico-electoral Strategy

Lula had a difficult politico-electoral strategy to follow. On the one hand, it was crucial for him to polarise the election between continuity (represented by Serra) and change (represented by himself in dispute with other opposition candidates). But at the same time he needed to reassure the markets, and also a substantial section of the citizenry that had benefited from the relative economic stability brought by the Cardoso administration, that some key elements of the outgoing government’s economic model would be preserved. As the candidate of a party that had forged its identity on the promise of radical change, his dilemma was how to campaign on a discourse of change in a country that was becoming steadily more stable. Should he call for radical change to attract those disillusioned with the failings of the Cardoso administration, or should he abandon the message of radical change to attract the voters who supported Cardoso in 1994 and 1998 and were afraid that radical changes in economic policy would bring back economic disorder? Moreover, as in 1998, the 2002 electoral campaign was fought under the shadow of an economic crisis. Facing a similar conjuncture the PT’s candidate needed to avoid falling into a similar trap. How could he avoid being accused by Serra, the government’s candidate, of talking up the crisis and so maximising the destabilising effects that a PT victory would have on the financial markets?

In dealing with these strategic dilemmas Lula’s electoral discourse had also to reach three different addressees. First, he had to reassure PT supporters that he remained committed to the party’s historical project of radical change, as expressed in numerous party documents, including the resolutions of the PT’s most recent congresses. Secondly, he had to convince the majority of the electorate, who had rejected him in the past three elections, that he could be trusted with the presidency and that a PT government would introduce meaningful changes without throwing away the political and economic gains made under president Cardoso. Thirdly, he had to reassure the economic agents (the markets) that he would not antagonise their fundamental interests, so that they would not precipitate a financial crisis by withdrawing their funds from the country.

Lula’s strategy was to make change (mudança) the nodal point of his electoral campaign. Time and again he iterated the call for change like a mantra. A crucial document of his campaign, the so-called ‘Carta ao povo brasileiro’ (Letter to the Brazilian people) is paradigmatic of this discursive strategy. This letter was published while the outgoing government was in delicate negotiations with the IMF for assistance to prevent the precipitation of a financial crisis. A public commitment from all main presidential candidates that a future government would abide by the terms of the agreement was seen as crucial for the success of these negotiations. In the letter, Lula started by claiming that ‘Brazil wants to change’ and he continued by re-affirming that the Brazilian people wanted ‘change for good’ (mudar para valer), a term that evoked the notion of change as rupture. He presented change as a new beginning by rejecting any explicit continuity (continuísmo) with the status quo. He reinforced this image by claiming that ‘the current model was exhausted’ and that there was ‘a powerful popular will to put an end to the current economic and political cycle’. He articulated change to a number of themes that had figured prominently in the PT’s traditional discourse, such as economic growth and social justice. He strongly criticised several aspects of the Cardoso government economic policies, particularly the economy’s structural vulnerabilities and the fragility of public finances. But traces of a continuing present were already to be found in this radical imagining of change, as he also reasserted his commitment to fight inflation and to pay the external debt at the cost of a tight fiscal policy. He stated that his government would honour the country’s financial contracts and obligations, and stressed his disposition to ‘preserve whatever primary budget surplus would be necessary in order to prevent an increase in the internal debt that would destroy confidence in the capacity of the government to honour its commitments’.14

The paradox of the letter lies in that while it was couched in the language of change as a new beginning, the markets interpreted Lula’s promise of change not as the threat of a break with the economic model but as a commitment to continuity by enacting market-friendly policies, such as honouring the country’s external debt and upholding fiscal discipline. The letter was widely perceived as a turning point in the relationships between the PT’s candidate and the markets, which slowly but surely began to change their perceptions of the threat to their interests represented by a possible Lula victory.15 So, arguably, a letter that had as its explicit addressee the Brazilian people, and claimed that the people rejected any form of continuity with the existing economic model, had the markets as its ultimate addressee; and a document that used change as its key signifier was construed by these hidden addressees as a commitment to continuity in some key macroeconomic policies.

Does this made Lula a manipulative politician saying one thing while meaning the opposite? Not particularly. The significance of the letter is precisely that Lula meant what he said. But words only mean what they mean for those who interpret them. Lula did not lie in the letter, nor did he promise one thing while intending another. What Lula did in the letter was to use change as a floating signifier (Laclau, 1990: 28; Torfing, 1999: 301) by articulating it to different political elements that over-endowed it with meaning, as shown in the following excerpt: ‘Another path is possible. It is the path of economic growth, with stability and social responsibility’.16 Each of these terms appealed to different audiences, all of whom were of crucial importance for Lula’s politico-electoral strategy: the promise of restoring economic growth was particularly important for the business sector as well as for the middle classes, which had traditionally supported Cardoso but were suffering under his policies of high interest rates; honouring the external debt was fundamental for the financial operators, which were afraid that the new government could default on the debt; while social responsibility (i.e. social justice) appealed to the PT’s traditional followers. Addressees would make sense of the letter (i.e. imbue it with meaning) by identifying themselves withLula’s change according to their own cognitive maps.

What was Lula’s Other in the letter? Clearly it was ‘the current model’ or, as he put it in slightly different terms, ‘the current economic and political cycle’, suggesting a relation of antagonism (change as rupture) grounded on both political and economic terms. Lula’s discourse equated ‘the model’ to a chain of negative features, such as economic stagnation, financial vulnerability, a compromised sovereignty, the continuation of high levels of corruption, and ‘frightening levels of social crisis and lack of security’. The antagonism between Lula’s promise of change and ‘the model’ (as its radical Other) set up the political dividing line that was central to the electoral campaign: between him, as the candidate of the opposition, and Serra, as the government candidate identified with the model’s negativity, an antagonistic relation on the basis of which Lula’s sought to fight and ultimately won the election.

However, Lula’s discourse disrupted the antagonism set up in his institution of ‘the current model’ as the Other of change by committing his government to honour the country’s external debt, pledging to control inflation and, particularly, to implementing the harsh fiscal adjustment agreed by Cardoso with the IMF. In the past, these policies had been subject to the PT’s critique of the so-called neoliberal model of the outgoing government. The party had traditionally associated these policies with subservience tothe IMF, the privileging of the financial sector over the needs of producers, lack of social investment, and low pay for public servants. Thus, by iterating his call for change in different contexts while committing himself to the preservation of important elements of the so-called ‘current model’ (o atual modelo), Lula resignified the meaning of change itself. Some policies were detached from their traditional association with neoliberalism and re-attached to the signifier ‘change’, a change that, by way of this iteration, came to articulate to the actuality of change represented by Lula’s victory not only the promises of a more fair and equal society but also elements of continuity with the outgoing government’s macroeconomic policies.

Other campaign documents and speeches follow a similar pattern. Writing in the Folha de São Paulo of 5 October 2002, Lula starts by claiming that ‘Brazil needs to change’ and associates change ‘with a national project that benefits the totality of the people and the rescuing of the centuries-old social debt’. But he also points out that ‘the moment of change has arrived through a process of pacts and negotiations’ and renews his commitment to ‘responsible macroeconomic policies, low inflation and economic stability’ and with ‘targets that would create the conditions for renewed economic growth’, an implicit reference to the IMF-agreed primary surplus target.17

Lula’s first public speech after winning the run-off election and his inaugural speech upon assuming the presidency are further examples of the same discursive strategy. In his post-electoral victory speech Lula interpreted his election victory as a vote for change: ‘Yesterday, Brazil voted for change’, he began. He followed this by stating that his victory meant the endorsement ‘of an alternative project and the start of a new historic cycle for Brazil’ (emphasis added), in other words, change as a new beginning. His speech evoked the traditional PT themes of the struggle in favour of the excluded and of the victims of discrimination, thus connecting his triumph with the party’s traditional banners. He said that he saw his life’s mission as ensuring that every Brazilian could have three meals a day, the promise that set up his ‘Fome Zero’ (Zero Hunger) programme; another powerful image of regeneration. But, as he had already stated in the letter, the newly elected president tempered the promises of a new order by introducing elements of continuity through his commitments to honour government contracts, controlling inflation and fiscal responsibility, thus taking on board the electors’ desire for economic stability and the need to reassure the markets.18

The pattern is repeated in Lula’s inaugural speech as president. The speech starts by instituting ‘change’ as the key signifier of his government: ‘Change (Mudança). This is the key word, this was the great message from Brazilian society in the October elections’.19 As in the letter, he made of the so-called ‘model’ the Other of change by associating it with a chain of negative elements that needed to be eliminated in order to allow the creation of a fairer society.20 In other passages, however, change is presented as a gradual process in which some demands will have to be delayed and elements of the present model preserved: ‘change is a gradual and continuous process, not just an act of the will’; ‘We have to keep our many legitimate social concerns under control so that they can be addressed at the right pace and in the right moment’. And in the speech, as in the letter, he made a explicit commitment to economic stability, to the ‘responsible management of public finances’ and to ‘an implacable fight against inflation’.21

Lula’s iteration of change as a floating signifier conveying both rupture and continuity allowed him to broaden his electoral appeal. But other key actors also joined the political battle to define the meaning of change. To further understand the ways in which the different iterations of change operated during the electoral campaign we need to examine the politico-electoral strategies of president Cardoso and of the PSDB candidate, José Serra.


Cardoso’s Politico-electoral Strategy

Cardoso's political strategy during the electoral campaign sought to blur the antagonism between his government and Lula. He carefully cultivated the figure of the statesman, presenting himself more as a head of state than as a head of government. While he supported his PSDB party's candidate, José Serra, his endorsement of Serra was less than fulsome. When in the early stages of the campaign the candidacy of Serra was faring badly and it appeared that he would be beaten into third place by the PPS’s candidate, Ciro Gomes, Cardoso studiously began to build bridges with Lula. He hinted that in a run off between Lula and Gomes he would support the PT’s candidate. He stated that Lula represented the forces of progress against the country's backward looking oligarchies. He rebuked Serra’s suggestion that Lula’s lack of an university education made him unfit to govern the country.22 Later, in a meeting with all the presidential candidates to inform them about the economic situation of the country and the ongoing negotiations with the IMF, he had a well-publicised one-to-one meeting with Lula. He conducted discreet meetings with some of the PT's leading politicians including former party president José Dirceu.23 The special relationship that developed between Cardoso and Lula during the electoral campaign and the political distance between the president and Serra even led some analysts to question who was Cardoso’s ‘true candidate’.24

The President’s attitude towards the main opposition candidate had a clear political purpose: by building bridges with the main opposition candidate, Cardoso was seeking to ensure that an Serra’s eventual defeat would not be construed as a total defeat for Cardoso’s own government and, particularly, for him as a person. By keeping himself above electoral politics and particularly by smoothing the process of transition as much as possible he consolidated his image as a builder of democratic institutions.

Cardoso had also an interest in preserving economic stability during the transition period and in the economic success of a moderate PT government. As Clovis Rossi wrote in Folha de São Paulo, he was aware that the future would be the judge of the past25 any economic collapse brought by the markets’ mistrust of the incoming government would unavoidably taint his legacy of economic stability, as in Argentina the failure of De La Rua coloured judgments of Menem’s second presidency. Conversely, a moderate PT government could be perceived in terms of change as continuity: the continuation of a gradual process of democratisation, political institutionalisation, and economic stabilisation initiated by Cardoso’s own government. According to this narrative, Cardoso’s reforms had successfully undermined the power of the traditional regional oligarchies and isolated the radical left, effectively bringing the PT into the centrist consensus that characterised his administrations. By asserting, as he did in a speech to the National Confederation of Industries days before the second-round election, that the country’s path would be maintained whoever was to be his successor (que o caminho de pais sera mantido seja quem for seu sucessor), Cardoso was not only reassuring business about Lula’s fundamental moderation, he was also implying that the path chosen by his government was the only possible and the best for the country.26

Cardoso invoked the notion that, far from representing a rupture with his government, Lula’s election was part of a broader process of gradual change through democratisation and stabilisation in which he and the new president were both participants. Thus, he remarked that ‘he had known Lula for 30 years’ as he underlined their common past as left-of-centre adversaries of the military government and vouched for the new president’s sound political character. He pointed out that the election proved that there was social mobility in Brazil and commented that this showed the quality of Brazilian democracy.27

Significantly, Cardoso attacked Lula’s suggestion that if elected he would seek to establish a social pact between the country’s social and economic actors. ‘Speaking as a sociologist’ Cardoso argued that a pact would be neither necessary nor pertinent, as pacts were only relevant at times of economic or political crisis and Brazil was not a country in crisis. As Cardoso noted in the aftermath of Lula’s second round victory: ‘I don’t feel I have been defeated. Serra was not the candidate of the government but of the PSDB. In a mature democracy, there is no space for very big disagreements between parties. Rupture is not possible. If the PT has changed direction I salute it’ (emphasis added).28


Serra’s Politico-electoral Strategy

The less-than-complete identification between the President and his party’s candidate was reciprocated by Serra, who campaigned on the rather ambiguous slogan of ‘continuidade sem continuismo’, a play of words that has no literal translation into English but can be loosely translated as ‘continuity with change’. Serra made no secret of his serious differences with the government’s economic policy. Throughout the campaign he kept his distance from the government: Cardoso hardly appeared in Serra’s television campaign, while paradoxically figuring twice in Lula’s TV adverts.

Despite Serra’s strategy of avoiding a close identification between his candidacy and the outgoing administration, the election and particularly the run off between him and Lula was a political scenario in which the electoral field would naturally become polarised between government and opposition. And while the government’s approval rating of around 26% was not negligible after 8 years in power, it was clear that the Government’s candidate would be the likely loser amid such polarisation. Thus, Serra sought to constitute a different polarisation, which would identify Lula’s change with a radical Other, as in the 1994 and 1998 elections. This antagonism aimed at presenting Serra as the candidate of stability and economic development against Lula as the embodiment of ‘change as rupture’ and therefore of uncertainty and fear of the unknown: as one PSDB politician put it: ‘fear of chaos, fear of capital flight, fear of inflation, fear of administrative incompetence’.29 Serra used three arguments to set up astrong dividing line between himself and Lula: the first warned that under a PT government, Brazil would be in danger of becoming a Venezuela, a Colombia, or an Argentina.30 The dangers represented by these countries were (i) a populist government that polarised society and threatened democratic institutionalisation (Venezuela), (ii) an upsurge of crime and lawlessness brought by the PT’s inability to tackle organised crime and its association with violent movements such as the Movimento dos Sem Terra (MST, The Landless Peasants’ Movement) (Colombia), and (iii) the PT’s inability to maintain economic stability leading to debt default and economic collapse (Argentina), respectively.

Secondly, to make the threat of political and economic instability credible and to counteract the PT’s moderate image during the campaign, Serra used a device typical of political rhetoric by establishing a relation of appearance and essence between what he called ‘the PT on the TV’ (Lula peace and love, or change as continuity) and the ‘PT on the streets’ (the angry PT of the Movimento dos Sem Terra and other radical movements,or change as rupture), and he claimed that the latter represented the true face of the PT.31 Perhaps the key moment of this strategy of setting up an alternative between order and the ‘unknown’ was a TV advert featuring the actor Regina Duarte. Here, Duarte said: ‘I am afraid. It’s been a long time since I had this feeling. We have two candidates. One I know, he is Serra; the other I don’t recognise, because everything he said has changed. That gives you a fright’ (emphasis added).32

Serra’s third discursive device to constitute Lula as the Other of order and good government was to compare his own governmental competence and willingness to engage in rational argumentation against Lula’s lack of experience, emotional appeal and refusal to debate. Serra suggested that Lula was unwilling or unable to engage in rational debate and thus was ill-prepared to govern. During his campaign Serra stressed his achievements as health minister and in other public posts as proof of his capability, experience and political will. Governing Brazil, Serra said, required ‘love for reason and the truth’. Voting for a candidate who had refused to discuss his ideas ‘would be a leap into the dark’ (emphasis added); again, change as rupture.33

As shown by the election results, Serra’s strategy was ultimately unsuccessful. Some analysts have blamed his failure on his alleged lack of charisma. Others have pointed to Serra’s aggressive campaign against two other rivals, Roseana Sarney and Ciro Gomes, which while highly effective in seeing off both rival candidates contrasted negatively with Lula’s ‘peace and love’ image. This was particularly relevant, as the Brazilian electorate does not like US-style negative campaigning. However, there were more fundamental reasons for Serra’s electoral defeat. His strategy of taking distance from the government while at the same time being the candidate of the president’s party left him in an identification vacuum. Marking his distance from the Cardoso administration while at the same time being unable to campaign as an opposition candidate made it hard for him to tell the Brazilian people a credible story that made sense of their recent past and gave them hope for the future. Opposition candidate Anthony Garotinho captured this failure when he pointedly asked Serra during a TV debate: ‘As the candidate of this government that is coming to a sad end, what hope can you offer to the Brazilian people?’

Why was Serra’s strategy, of identifying Lula with the threat of a radical change that would bring disorder and uncertainty, not more successful? His failure had to do with Lula’s skilful articulation of change as rupture and change as continuity that neutralised Serra’s attacks. In this strategy, Lula was also helped by the way the economic crisis that constituted the economic background to the electoral campaign, as shown by the fall in the value of the real and the rise in the country risk, played for both candidates. It was very difficult for Serra who was, after all, the candidate of a party that had been in power for 8 years to claim simultaneously that the country was going through a serious economic crisis (and thus that Lula’s radicalism and lack of experience would make things worse) and that the crisis had nothing to do with the Cardoso government. To convince the electorate that voting for a PT government would bring economic instability, it would have been necessary to convince the voters that the country was economically stable under the Cardoso administration and that it was only the threat of a change of government that was endangering stability. Given that this was clearly not the case, and that in the 1998 election president Cardoso played the stability card only to devalue the real 2 weeks after starting his second mandate, the argument that the opposition was responsible for the country’s economic problems was not particularly credible. Moreover, the strategy of identifying Lula with economic chaos was made even more difficult by Cardoso’s repeated statements that the country’s stability would not be threatened by the electoral results and that whoever were to win the election there would be no radical change in economic policy. Conversely, Serra became exposed to the same accusation made against Lula in 1998, namely that his campaign of ‘economic alarmism’ was worsening the crisis. Lula was quick to pick up on this point when he accused the Government of exploiting the fragility of the Brazilian economy for electoral purposes: ‘What the government cannot do is to keep playing terrorist games with the Brazilian economy that is in a fragile state’.34

By contrast, Lula used the crisis to justify the maintenance of some key aspects of the economic model. It has been argued that crises open the possibilities for radical change and policy reform. But in Lula’s discourse the crisis acted as a constraint to the radical and rapid change of the model. He argued that a period of transition would be required: ‘a careful and well-thought (criteriosa) transition between what we have today and that what society wants’. Again, change as continuity. So, while for Serra the crisis could not be fully acknowledged as part of the present but only as a threat for the future (if the opposition was to win the election) for Lula the ongoing crisis was a reason for maintaining some aspects of the status quo that he had learned from past experience were worth preserving or were to risky to change.35



This article has explored how three key actors in the electoral process the PT candidate Lula da Silva, former president Cardoso and PSDB candidate Serra iterated change in their political discourses. Candidates sought to fix the meaning of change in a way that would favour their electoral strategies. In this quest, each of them traced and retraced the political dividing lines that defined the electoral contest and the play of differences that gave change its different meanings.

The ambiguous relationship between continuity and change was perhaps best expressed in Serra’s slogan of ‘continuidade sem continuismo’. As we have seen, Serra fell through the identification gap between the two terms. This made him neither the heir of Cardoso’s vision nor a credible bearer of change. But ambiguity can also deliver rich political rewards. Lula campaigned on a call for change, but his appeal was tempered with reassurances about the preservation of policies that in the past the PT had associated to the economic model that was the other of change. He succeeded where Serra had failed, namely in his ability to articulate two apparently contradictory meanings of change in a way that allowed him to appeal to constituencies that had different expectations about his government.

Perhaps one of the reasons for Lula’s success was that the iteration of change in his discourse allowed the Brazilian people to imagine not only a better future, already symbolically present in the significance of his victory, but also a reinstitution of the past. His triumph enabled the emergence of a new narrative in which the history of Brazil, instead of being the preserve of elite hegemony as had been traditionally presented in both academic and political discourse, could be reinterpreted as a progressive process of democratisation that incorporated popular struggles for the enlargement of citizenship as symbolised by Lula’s personal and political journey.

Some of Lula’s new-found friends on the right and some of his critics on the left alike would probably agree that if Serra failed in representing ‘continuidade sem continuismo’, Lula succeeded in representing a certain dose of change with continuity. In politics, the two terms coexist in a mixture of contradictions, tensions, hopes, fears, illusions and disillusions. There is an unstable relationship, and it is difficult to predict which one will come to define the legacy of Lula’s government. Surely, the very different actors to whom Lula so skilfully appealed during his electoral campaign will wish the scales to tip to different sides.

As developments during the first 15 months of Lula’s government show, his strategy is not without danger. But, change and continuity cannot be measured only in terms of macroeconomic policy. It may be that there will be no major changes in the economic model under president Lula. But if the political context in which the new president has to operate is taken into account, Lula’s accession to the presidency represents in itself a significant change. Unlike Serra, Lula did have a compelling story to tell the Brazilian people: a story of hope symbolised by his own life as an internal migrant, a worker, a trade union leader and as a presidential candidate who defeated the political establishment against all odds. The story is still unfolding and its end still uncertain.



Abranches, S. H. (1993) Strangers in a common land: Executive/legislative relations in Brazil. In S. Marks (ed.), Political Constraints on Brazil’s Economic Growth, pp. 105–129. Transaction Publishers: New Brunswick and London.

Arditi, B. (2003) Talkin’ ‘about a revolution: the end of mourning. Parallax 9 (2): 91–95.

Bowman, G. (forthcoming) Constitutive violence and the nationalist imaginary: The making of ‘The People’ in ‘Palestine’ and ‘Former Yugoslavia’. In F. Panizza (ed.), Populism and the Mirror of Democracy. Verso: London.

Derrida, J. (1988) Signature event context. In Limited Inc, pp. 1–23. Northwestern University Press: Evanston, IL.

Laclau, E. (1990) New Reflections on the Revolution of Our Time. Verso: London.

Mangabeira Unger, R. (1990) A Alternativa Transformadora. Como Democratizar o Brasil. Editora Guanabara Koogan SA: Rio de Janeiro.

Nylen, W. R. (2000) The making of a loyal opposition: The workers’ party (PT) and the consolidation of democracy in Brazil. In P. R. Kingstone and T. J. Power (eds), Democratic Brazil. Actors, Institutions and Processes, pp. 126–143. University of Pittsburgh Press: Pittsburgh, PA.

Panizza, F. (2000a) Neopopulism and its limits in Collor’s Brazil. Bulletin of Latin American Research 19: 177–192.

Panizza, F. (2000b) Is Brazil becoming a ‘boring’ country? Bulletin of Latin American Research 19: 501–525.

Roett, R. (1984) Brazil: Politics in a Patrimonial Society. Preager: New York.

Stokes, S. (1999) What do policy switches tell us about democracy? In A. Przeworski, S. Stokes and B. Manin (eds), Democracy, Accountability and Representation, pp. 98–130. Cambridge University Press: Cambridge.

Torfing, J. (1999) New Theories of Discourse. Laclau, Mouffe and Zizek. Blackwell Publishers: Oxford.



1 I thank Benjamín Arditi, Vicente Palermo and two anonymous reviewers for their helpful comments.
2 'Lula é refundação do Brasil, diz sociologo'. Folha de São Paulo (FSP), 29 October 2002.
3 'Entrevista: Dificuldades de Lula serão propocionais á esperança que criou'. FSP, 20 October 2002.
4 'Ciências Sociais: Intelectuais divergem sobre diferenças entre Lula e Serra'. FSP, 24 October 2002.
5 Ibid.
6 'Balanço: FHC deixou saldo negativo, diz historiador'. FSP, 10 November 2002
7 'Entrevista: Dificuldades de Lula …'. FSP, 20 October 2002
8 As Lula himself put it in his inaugural speech: 'Each of us Brazilians knows that what we have done until today is not little, but also knows that we can do much more. When I look at my own life as a retirante nordestino [internal migrant from the North East], as a kid that sold peanuts and oranges in the docks of Santos, who became a lathe operator and trade union leader, who one day founded the Workers' Party and believed in what he was doing and now assumes the position of the nation's head of state, I see and I know, with absolute clarity and conviction, that we can do much more.' – 'Discurso do presidente Luiz Inácio Lula da Silva durante o ato de posse no Congresso Nacional'. FSP, 2 January 2003
9 'Eleição de Lula mostra que há mobilidade social no Brasil'. FSP, 29 October 2002
10 José de Souza Martins, 'O triunfo do suburbia'. FSP, 17 November 2002
11 'Balanço …'. FSP, 10 November 2002
12 Effectively both Lula and Serra were candidates of multi-party coalitions. However, the PT and the PSDB were the leading parties of their respective coalitions and were the parties to which the candidates were affiliated.
13 FSP, 13 October, 19 October and 20 October 2002.
14 'Carta ao povo brasileiro'. Partido dos Trabalhadores. Dirétorio Nacional, 22 June 2002. ‹›.
15 See, for instance, 'Corretora recua e agora recomenda Brasil'. FSP, 5 October 2002
16 'Carta so povo brasileiro'
17 Luiz Inácio Lula da Silva, 'Um projeto de nação'. FSP, 5 October 2002
18 Transição: 'Brasil votou para mudar', discursa Lula. FSP, 29 October 2002.
19 'Discurso No Congresso'. FSP, 2 January 2003.
20 'Facing the exhaustion of a model which, instead of generating growth produced stagnation, unemployment, and hunger; in view of the failure of a culture of individualism, of selfishness, of indifference to the other, of the disintegration of the families and the communities; in view of the threats to national sovereignty, of the precariousness of public security, of the lack of respect for the elderly, and the despondency of the youth; in view of the economic, social and moral stagnation of the country, Brazilian society chooses to change'. FSP, 2 January 2003.
21 Ibid.
22 'FHC dá apoio a Serra, mas faz acenos a Lula'. FSP, 6 October 2002.
23 'Troca: FHC e Lula criam ''ponte para o futuro'''. FSP, 27 October 2002.
24 'Intelectuais divergem sobre diferenças entre Lula e Serra'. FSP, 24 October 2002.
25 'O futuro julga o passado'. FSP, 10 October 2002.
26 'Eleições 2002. Serristas unificam ''teoria do caos'''. FSP, 17 October 2002.
27 'Eleição de Lula mostra que há mobilidade social no Brasil'. FSP, 29 October 2002.
28 Ibid.
29 'Eleições 2002. Serristas …'. FSP, 17 October 2002
30 'Dias de Decisão'. FSP, 20 October 2002
31 Segundo turno: Em três meses PT põe paý´s em crise, diz Serra'. FSP, 10 October 2002.
32 'Eleições: Lula vai a 61% e abre 29 pontos sobre Serra apo´s cinco dias da TV'. FSP, 20 October 2002.
33 'José Serra: Peço seu voto', FSP, 5 October 2002 and 'Serristas unificam ''teoria do caos''', FSP, 17 October 2002.
34 'Governo explora economia frágil, diz Lula'. FSP, 12 October 2002. FSP, 5 October 2002.
35 Lula, 'Um projeto de nação'. FSP, 5 October 2002.