‘Any sociological study of voting behaviour’, Matei Dogan has written,
is inevitably caught in a vicious circle. A thorough analysis should not be limited to such gross categories as the working class and middle class.… Nevertheless, the more one takes into account the actual social diversity, the more one is limited to rough guesses and surmises.… Thus, paradoxically, the more one looks for precision in the social infrastructure, the more one risks being imprecise in analysing the social bases of each party.1
The inherent limitations of the present study oblige us to proceed by ‘gross categories’, and the numerical data for different countries must not be taken to the decimal point because of the different criteria employed. Analysis will nevertheless make it possible to identify the main trends, especially given that some similarities or contrasts are too marked not to be significant.
Working-Class Support and the Majority Vocation
One of the postulates of the ‘catch-all’ thesis, which highlights the mutation in partisanship in developed societies, is that during the 1960s parties of social integration gradually abandoned any ambition to mould populations intellectually and morally, and simply sought to expand and diversify their electoral base. In the event, the tendency to form voter, ‘catch-all’ parties, largely confirmed in terms of the actors’ programmes and strategic objectives, is only very partially confirmed at the level of the socio-electoral support for European social-democratic parties.
Whether we are talking about the Scandinavian countries, Austria, Great Britain, or even the FRG or Belgium, the sociological content of social democracy’s electoral make-up is unambiguous. In the 1960s, parties of the social-democratic type remained parties of the working class on two counts. First, the largest part of their mass electoral base derived from the working class (Table 3.1). Second – and above all – the majority of this class, and sometimes a very large majority, supported the parties in question (Table 3.2). Despite a tendency to become increasingly interclassist in their physiognomy, parties of the social-democratic type were constructed during the period under examination as specific coalitions whose main ingredient and central bloc remained the working class.
Table 3.1 Social composition of the social-democratic electorate
| SAP 1968 | SD 1966 | DNA 1969 | SPD 1972 | Labour 1966 | SPÖ 1955 | |
| Workers | 67 | 72 | 65 | 50 | 81 | 61 |
| White-collar workers/managers | 29 | 23 | 26 | 35 | – | 25 |
| Self-employed | 4 | 5 | 9 | 8 | 19 | 14 |
Sources: For the Scandinavian social democracies, see Gøsta Esping-Andersen, Politics against Markets: The Social Democratic Road to Power, Princeton University Press, Princeton, NJ 1985, pp. 128–9. For the SPD, Gerard Braunthal, The West German Social Democrats, 1969–1982, Westview Press, Boulder, CO 1983, p. 165. For the British Labour Party, Bo Salvik and Ivor Crewe, Decade of Dealignment, Cambridge University Press, Cambridge 1983, p. 89. For the SPÖ, Wolfgang Muller and Peter Ulram, ‘The Social and Demographic Structure of Austrian Parties, 1945–93’, Party Politics, vol. 1, no. 1, 1995, p. 148 (for the SPÖ, the category of ‘self-employed’ contains self-employed, professionals and farmers. The percentages have been established by us after excluding housewives, pensioners and the unemployed.)2
Table 3.2 Working-class electoral penetration of five social-democratic parties (1960s)
| Country | Date | Party | Working-class vote % |
| Sweden | 1968 | Social democrats | 77 |
| Denmark | 1968 | Social democrats | 65 |
| Norway | 1965 | Labour | 69 |
| Great Britain | 1964 | Labour | 64 |
| FRG | 1969 | Social democrats | 58 |
Sources: See Table 3.1.
I think the essential point to emphasize is that the most successful social-democratic parties electorally are precisely those that are most distinctively working-class. The long social-democratic ascendancy in Scandinavia was inextricably bound up with the decisive influence of the socioeconomic cleavage and, as a consequence, the development and persistence of class voting. ‘One of the paradoxes of Scandinavian political development’, Francis Castles has written, ‘is … that the reformist Social-Democratic parties of Scandinavia offer the only empirical confirmation of Marx’s view that the politics of social class leads to the inevitable political victory of the working class.’3 The majority status of these parties was often linked less to their capacity to attract the middle classes than to their impressive entrenchment among the working class. The case of Swedish social democracy is exemplary here.
The hegemonic role – in the Gramscian sense – of the SAP, established before the Second World War, was based, among other things, on the ‘unique advantage’ (in Eric Hobsbawm’s phrase) of the party’s entrenchment in the working class. Despite its quasi-hegemonic influence, its spirit of openness, its remarkable trade-union implantation among white-collar workers, and despite the fact that the SAP ‘has never been a socially isolated labor party’ – despite all that, and contrary to conventional opinion, the SAP’s electoral influence remained comparatively limited outside the working class.4 Fundamentally, the enduring political hegemony of the SAP – possibly the most powerful party western Europe has ever known – had as its spinal column the stability, at a very high level, of social-democratic influence among the working class.
Throughout the 1960s, the relative progression of the SAP among non-manual strata – significant in itself, given their growing numerical importance – accompanied and supported the extraordinary stability of social-democratic penetration of the working-class milieu (Table 3.3).
Table 3.3 Social-democratic penetration among manual and non-manual workers in Sweden (1956–76)
| Date | SAP share of the manual vote | SAP share of the non-manual vote |
| 1956 | 74 | 22 |
| 1968 | 74 | 32 |
| 1976 | 62 | 28 |
Source: Christine Buci-Glucksmann and Göran Therborn, Le Défi social-démocrate, François Maspero, Paris 1981, p. 281.
The 1976 elections, and the bourgeois parties’ accession to power after forty-four years of social-democratic governments, marked the end of – or possibly an interlude in? – the SAP’s impressive domination of Swedish political life. As Table 3.3 demonstrates, the SAP’s electoral crisis was essentially due to the defection of a significant number of workers, and hence to losses deriving from the party’s ‘natural’ social base. This electoral erosion, which crystallized very clearly during the difficult years of the economic crisis, reflected the erosion of social-democratic hegemony in Swedish society.
The British Case
From the viewpoint that interests us here, the case of the British Labour Party is even more interesting.
In the 1960s, Labour’s arrival in power after three consecutive defeats depended on two things. The first was the stronger propensity of manual workers to vote Labour – a propensity that approached the symbolic 70 per cent mark at the 1966 election. The second was Labour’s increased penetration among the new salaried strata. But just as the 1964 and 1966 victories were primarily the indisputable result of the party’s advance among the manual workers, so the defeat of 1970 derived from extensive disaffection among working-class voters, Labour’s share of the non-manual vote having scarcely changed relative to 1966, and having increased compared with 1964.
If, on the basis of the data in Table 3.4, we establish the average Labour share for the 1960s according to elections won/lost and the categories manual/non-manual, we can better illustrate the relationship between the manual vote and the Labour Party’s majority status. Table 3.5 clearly indicates that there is a close connection between the level of the working-class vote and the majority status of the party. The picture that emerges from these data allows for little interpretative hesitation: Labour’s victories were strongly correlated with increased working-class support, and its defeats with a reduction in this support.
Table 3.4 The British Labour Party’s penetration among manual and non-manual workers (1959–70)
| Date | Labour share of manual vote | Labour share of non-manual vote |
| 1959 | 62 | 22 |
| 1964 | 64 | 22 |
| 1966 | 69 | 26 |
| 1970 | 58 | 25 |
Source: Bo Salvik and Ivor Crewe, Decade of Dealignment, Cambridge University Press, Cambridge 1983, p. 87; my adaptation.
Table 3.5 Elections won and lost by Labour and the manual/non-manual vote (1960s)
| 1960s | Share of manual vote (average) | Share of non-manual vote (average) |
| Won: 1964/1966 | 66.5 | 24 |
| Lost: 1959/1970 | 60 | 23 |
Source: See Table 3.4; my calculation.
The German Case
In the social-democratic universe, the SPD constitutes a paradigm of its own. Despite their great working-class tradition, German social democrats suffered the repercussions of the religious divide and the division of the country, and were restricted to a level of penetration of the working-class population markedly inferior to that of their Swedish, or even British, counterparts. While it had a majority within the working class, the SPD stood below the 60 per cent threshold of the working-class vote.
The SPD’s progress to power was a long process, marked by a gradual and almost continuous improvement in its influence among all segments of the population, and in the first instance the salaried middle classes.
The German social democrats achieved a competitive status that allowed them to rule the country5 in two phases. In 1969, an initial quantitative and qualitative leap forward derived from a differential advance, whose main distinguishing feature was the SPD’s spectacular breakthrough among salaried middle strata. In 1972, by contrast, professional categories shifted in their entirety towards social democracy, demonstrating that the party of Willy Brandt had enhanced its powers of attraction in an extremely diversified electoral field (Table 3.6).
Table 3.6 Electoral penetration of the SPD (1953–72)
| 1953 | 1961 | 1965 | 1969 | 1972 | |
| Workers | 48 | 56 | 54 | 58 | 66 |
| Employees/civil servants | 27 | 30 | 34 | 46 | 50 |
| Self-employed | 11 | 14 | 18 | 17 | 23 |
| Farmers | 4 | 8 | – | 16 | 10 |
Source: Ilias Katsoulis, in Helga Grebing, L’histoire du mouvement ouvrier allemand, Papazissi, Athens 1982, p. 413.
Overall, we could say that the SPD – which, since the Second World War, had been of a less distinctively working-class character than the most important social-democratic parties – owed both its strong electoral growth after 1969, and its constitution as a ‘natural’ party of government, largely to the new salaried strata.
Hence this party appears to be very sensitive to fluctuations in the vote of the middle strata, and this was to be the source of many anxieties in the future. But it continued to be a political formation whose electoral and social base remained, in its majority, the working class.
Social Democracy and the Middle Classes
Referring by definition to a scalar – and ternary – structure of social stratification, the term ‘middle class’ derives its force from an intuitive perception of the reality of social classes which is very widespread and long-standing. From the standpoint of electoral sociology, however, it conceals extraordinary disparities of behaviour and ideology. In accordance with the variety and fragmentation of the social groups composing them, it is preferable to speak of the middle classes in the plural – something I shall do throughout the present work.
The Uncertain and Uneven Adhesion of the Salaried Middle Strata
The first question we should address concerns the evolution in the postwar period of the propensity to vote social democrat among the middle classes, provisionally construed as an internally undifferentiated statistical unit. Now, the social-democratic or labour constituency grew significantly with time, the arithmetical progression of these classes in the population being accompanied by an increase in social-democratic influence among them (Table 3.7).6
Table 3.7 The middle-class vote in Norway, Denmark and Sweden
| Farmers | Small businessmen | Senior managers | White-collar/Middle managers | Total electorate | |
| Denmark 1971 | |||||
| Radical Left | 0 | 2 | 12 | 18 | 10 |
| Social Democrats | 2 | 19 | 9 | 31 | 34 |
| Centre Parties | 79 | 27 | 43 | 21 | 35 |
| Conservatives | 19 | 52 | 36 | 31 | 20 |
| TOTAL | 100 | 100 | 100 | 101 | 99 |
| Norway 1969 | |||||
| Radical Left | 1 | 0 | 9 | 5 | 6 |
| Labour | 9 | 42 | 21 | 40 | 44 |
| Centre Parties | 84 | 27 | 28 | 31 | 31 |
| Conservatives | 6 | 31 | 42 | 25 | 19 |
| TOTAL | 100 | 100 | 100 | 101 | 100 |
| Sweden 1970 | |||||
| Radical Left | 2 | 5 | 1 | 2 | 3 |
| Social Democrats | 9 | 18 | 24 | 47 | 47 |
| Centre Parties | 85 | 57 | 53 | 40 | 40 |
| Conservatives | 5 | 21 | 22 | 10 | 10 |
| TOTAL | 101 | 101 | 100 | 99 | 99 |
Source: See Francis Castles, The Social Democratic Image of Society, Routledge & Kegan Paul, London and Boston, MA 1978, p. 109.
However, the electoral fate of parties of a social-democratic type, at least in Scandinavia, was very different depending on whether the relevant subset was the ‘self-employed’ or salaried strata (Table 3.7). In their great majority, the ‘self-employed’, whether farmers, small businessmen, or members of the liberal professions, cast their votes for the parties of the centre and the right; the same applies to senior managers.7 In contrast, the electoral behaviour of the salaried middle strata, especially the lower fraction, was more nuanced: a significant proportion of them – something in the region of 40 per cent – turned to social democracy. Thus, in their electoral behaviour the salaried middle strata presented themselves as the principal partner – albeit an uncertain and hesitant one – of the working class within the social-democratic coalition. It should, however, be stressed that the propensity to vote social democratic among the salaried middle strata was almost identical to the national average in the three countries concerned (Table 3.7). This demonstrates the ambiguous and, in a sense, ‘undecided’ character of their electoral behaviour.
This general impression is largely confirmed in West Germany, where, as we have indicated, the SPD, a less working-class and more interclassist party than its cousins in other countries, comfortably passed the 40 per cent threshold in the category employees/middle management from 1969 onwards (1969: 46 per cent; 1972: 50 per cent).
During the 1960s and up until recently (with the exception of the elections of 1997 and 2001), Labour followed a rather different pattern of electoral development, making Great Britain a partially ‘atypical’, borderline case. Compared with other countries, British society is characterized by a more assertive and rigid differentiation between workers and the middle classes as regards lifestyles and, more generally, what are called ‘life chances’. This means that there has traditionally been a very strong middle-class vote for the Conservative Party. ‘In largely rallying to the Conservative banner,’ Monica Charlot has written, ‘the middle classes defend a status and a lifestyle that they consider to be under threat from a majority working class.’8
Thus, Labour’s marked inferiority among not only the upper middle classes and self-employed but also among salaried middle strata, which contrasts with the politico-electoral ‘uncertainty’ of the latter in other countries with a strong social democracy, makes Britain a rather special case. Yet in Great Britain, too, a more subtle analysis, based on socioprofessional categories other than the conventional distinction between manual and non-manual workers, yields a picture with greater contrast. In the 1964–70 period, Labour’s penetration in the category of employees (which, in its majority, forms part of the popular classes) is situated at a level very nearly equivalent to that of the social-democratic-type formations examined in this work (31 per cent in 1964, 42 per cent in 1970).9
Analysis based on examination of ‘gross categories’ indicates that from the perspective of electoral sociology, the space occupied by the middle classes, rather than being a zone of convergence and absorption of ‘extremist’ attitudes in some kind of intermediate electoral behaviour, is traversed by pronounced disparities and oppositions. The middle classes are politically extremely divided – something that is underestimated by approaches that employ a ‘scalar’ schema of social stratification. These approaches attribute a resolutely intermediary role to the middle classes, which are regarded as forces of moderation and consensus – a kind of third party challenging a supposedly bipolar social antagonism. Now, this schema does not correspond to the complexity of societies in advanced capitalism. From the viewpoint of electoral sociology, the image of an ‘intermediate’ middle class, privileged depository of social peace, is a myth. It is a myth based on – and, in turn, justifying – the notion of solidarity between medium-sized and small capitalists on the one hand, and the lower ‘middle’ classes on the other – a ‘solidarity’ that does not withstand the test of actual political behaviour. In addition, Nicos Poulantzas’s hypothesis of a ‘quasiclass’ – the petty bourgeoisie – with two fractions, the traditional petty bourgeoisie and the new petty bourgeoisie, is justified by their common polarization vis-à-vis the bourgeoisie and the working class. However, Poulantzas’s approach underestimates the incontrovertible empirical fact that these two components of the petty bourgeoisie are highly polarized, electorally and culturally, among themselves, before being polarized vis-à-vis the proletariat and the classical bourgeoisie.10 In this respect, the key distinction seems to us to be that between those who own their means of labour (medium and small businessmen, shopkeepers, artisans, liberal professions, etc.), as well as senior managers, and the middle and lower bands of the salaried middle strata (junior and middle managers, civil servants, employees, salaried intellectual professions). Thus, the urban middle classes are politically structured around two major poles: on one side, the traditional middle and petty bourgeoisie, which is a pole of minimal support for social democracy and the left in general; on the other, the salaried middle strata (broadly, the famous ‘new petty bourgeoisie’), which act as an intermediate pole of support for social democrats, and as the second major pillar, alongside workers, of their electoral strength. Within the category of salaried middle strata, which is an internally fragmented group, two components – or, better, two types of voters – seem to constitute the epicentre of social-democratic influence.
The first is essentially found among employees, all those who do not occupy a position in the social division of labour which confers on them a certain status and authority. The orientation of this category to the left is ‘instrumental’, and reflects its relative ‘marginality’ within the middle classes – a ‘marginality’ that brings it closer to the working class. The firm entrenchment of the social-democratic parties among employees and minor public officials underlines the popular character of the social-democratic enterprise.11
The second type of voter comes from the intermediate or higher scales of the salaried middle strata. This type is characterized by the practice of a professional skill, often involving duties of supervision and authority, and is principally – but not exclusively – located in the public and para-public sector. It broadly corresponds to what sociologists call ‘new middle strata’. This group comprises the professions of teaching, social work, health, and public officials, as well as artistic professions. Its value-orientation, regarded by Frank Parkin as ‘altruistic’,12 is often associated with a highly structured left-wing ideology. A ‘radical-altruistic’ electorate, it corresponds in part to what the ‘post-materialism’ school, following the pioneering work of Ronald Inglehart, was subsequently to characterize as an electorate with post-materialist values.13 The most distinctive characteristic of this well-educated group (which refers not to a unified social category, but to a socially disparate jumble) is its ‘cultural liberalism’, according to 1980s French electoral sociology, or ‘radical liberalism’, according to David Jary.14 This liberalism or radicalism is linked to a relatively well-articulated left-wing political ideology. In fact, during the 1960s this group became the privileged vector of the two ideological dominants that would seem to be specific to the left-wing electorate: ‘traditional left-wing values’ and ‘cultural liberalism’. Both are strongly correlated with voting for the left – the former more so than the latter, obviously.
The shortage and heterogeneity of available data certainly complicate any attempt to establish the characteristics and real arithmetical weight of this group in the social-democratic electorate with any rigour. It nevertheless seems to be established that the most consistent share of support for social democracy from the higher and middle ranks of the middle classes derived from the new middle strata. This support was an expression – and, at the same time, a demonstration – of the ability of social-democratic parties in the postwar period to expand well beyond the poor and poorly educated strata of the population. Social democrats’ tendency to pose as a vehicle of ‘cultural modernization’ in the European societies of the 1960s encouraged – and was encouraged by – the left-wing leanings of this group. It was also furthered by a certain cultural ‘archaism’ on the part of the centre-right parties (particularly Christian Democrats and conservatives). Hence the support in question was used by the social-democratic elites as the marker of a certain identity, as an index of modernity. In this instance, where ‘modernity’ was the value connoted, the support of new middle strata became a source of strength – a source that was a good deal more important than their electoral weight in the population (ultimately limited) betrays. In the 1950s, and especially in the 1960s, social democracy – mature social democracy – truly and fully acquired its specificity solely through this ‘modernizing’ aspect. In the 1960s, social democracy established itself, and blossomed, as a force for social equality and redistribution, but also as a ‘modern’ and culturally liberal force – more modern and liberal, at any rate, than its conservative opponents. Thus, the firm entrenchment of parties of a social-democratic type among minor employees and new salaried strata, particularly their most educated fraction, highlights and confirms the simultaneously popular and culturally modernizing character of the social-democratic approach in the initial postwar period. The social democrats’ ability to ally egalitarian demands with liberal demands for cultural modernization underlay their strength and consolidation as parties of government.15
We cannot offer a sharper image of social-democratic penetration of the salaried middle strata here. The rapid analysis above does not, therefore, enable us to avoid ‘contemplation of an ocean of uncertainties’, as the French political scientist Philippe Braud might say. Even so, everything indicates that two types of non-manual voter, socially and culturally distinct, gave their support to social-democratic parties. Everything also indicates that this advance among the middle classes allowed social-democratic and labourist formations to constitute themselves as natural governmental forces, and to play an important, or even central, role in the political system. However, they did not establish themselves as dominant forces among these classes, not even within the segment most inclined towards them, the salaried middle strata (treated here as a statistical whole without internal differentiation). In their electoral behaviour, these strata were intermediate between a working class strongly inclined to the left, and ‘self-employed’ and upper middle classes strongly inclined to the right.
Thus, even during the 1960s the adhesion of the middle classes remained a minority and uncertain affair. The structure of the social-democratic electorate confirms the uncontested predominance of the working class.
Electoral Social Democracy: Working-Class or Interclassist?
Class Parties and Parties of the Class
Given the arithmetically minority status of the working class stricto sensu,16 working-class parties in the 1920s and 1930s were soon faced with a crucial, ideologically formidable electoral dilemma: either to remain class parties and run the risk of being doomed to permanent minority status; or to transform themselves into interclassist parties and claim a majority position in the political system. Condemned to ambiguity, to be strategically Janus-faced, the social-democratic parties ‘cannot remain a party of workers alone and yet they can never cease to be a workers’ party’.17 What ensued, particularly after the Second World War, was the gradual dissolution of their class appeal and their constitution in accordance with a format that tended towards the interclassist. Social-democratic parties are coalition parties, but not all ‘coalitions’ are identical.
What constituted the specificity of each coalition was the fascinating and peculiar interaction of the combinations that were constructed, the particular mix in the components of a partisan electorate. Otherwise, the electorates of left-wing parties in the countries of advanced capitalism – constituted, as a general rule, by the crucial contribution of two social groups, workers and salaried middle strata – would all have to be treated as equivalent.
The social character of a party is largely conditioned, first, by its social penetration (the percentage of the class or social category that supports the party in question); second by the ‘concentration’ or ‘social cohesion’ of its electoral base (the percentage of its electorate that belongs to a single social class); and – even more, perhaps, since what is at stake is then the party’s development – by the evolution of these over time.
The great majority of parties of the social-democratic type are, in Richard Rose and Derek Urwin’s classification (developed using the criterion of cohesion or concentration, not penetration), either homogeneous class parties, or parties with ‘mutually reinforcing loyalties’, constructed around a working-class and a secular component.18 This criterion, however, does not make it possible to account for the fundamental difference – very rich in its implications – between class parties and parties of the class. This distinction is highly significant for understanding the representative capacity of social-democratic political forces, and the place they occupy within the party system. A class party is one whose electoral base is socially homogeneous (in the sense, following the criterion of cohesion, that the great majority of its voters belong to a single social class). But a class party is not necessarily a party of the class (because it may be the case, following the criterion of penetration, that only a small minority of the relevant social class supports this party). Consequently, a left-wing party can be constituted as a working class party without thereby being the party of the working class. It can also be a party of the working class without thereby being a working-class party.
The criterion of social penetration, prioritized here, enables us to measure and grasp the entrenchment – the real influence – of the social democrats in society and, in so doing, their role in the political system.
Social Democracy as an Enlarged Coalition of the Working Class
If we follow the indications given throughout this chapter, a number of conclusions follow.
1.In all the cases examined, the central electoral bloc in the social-democratic electorate is focused on the working class, social-democratic penetration of which is very deep. Alongside the central bloc, a second bloc is formed by the salaried middle strata, which, in terms of electoral support, represent an intermediate pole. Thus, the electoral strength of social-democratic-style parties during the 1950s and 1960s was articulated around a central working-class pillar and a secondary pillar, based on the most ‘popular’ and ‘radical-liberal’ segment of the middle classes.
2.Given that a clear majority – and in some cases an imposing majority – of the working class supported social-democratic parties, with the assistance of a substantial minority of the middle classes, we shall call these political formations enlarged coalitions of the working class from the standpoint of electoral sociology. We shall distinguish two categories:
(a)The first (Sweden, Norway) is characterized by a social-democratic penetration that approximates or exceeds two-thirds of the working-class vote. These are the parties of maximum working-class penetration, and their prototype is obviously the SAP.
(b)The second is characterized by a penetration situated around the three-fifths mark (Great Britain, Denmark) or, in the borderline case of the FRG, clearly exceeding the threshold of 50 per cent of the working-class vote. These are parties with high working-class penetration,19 and it is possible to identify two variants: strong working-class concentration – not penetration – whose most characteristic example is the British Labour Party; and a more interclassist format, of which the SPD is the most developed specimen.20
3.Parties of a social-democratic type are clearly distinguished in the 1950s and 1960s from working-class parties which are class parties, but not parties of the class (e.g. the PCF and other communist parties). They are also distinguished from parties which, though they belong to the great European social-democratic/socialist family, are neither class parties nor parties of the class, but interclassist formations (e.g. the SFIO, the French PS, the PSI, PASOK in the 1970s, the Portuguese PS).
4.The social fabric of electoral behaviour, which we have explained very briefly, demonstrates that social-democratic-type parties did not conform to the model of the catch-all party in their electoral sociology.21 In the initial postwar period, these parties certainly strove to present themselves as great ‘popular’ formations, as ‘coalition forces’, in electoral competition. But if the programmatic profile, the social-democratic scaenae frons, evolved firmly in a catch-all direction, the hard core – the sociological pillar – was based on the working-class matrix. During the 1960s, the massive sociological reality of social democracy remained essentially centred on the working class.
5.Unquestionably, the so-called ‘catch-all’ policies of the social-democratic parties proved effective not ‘although’, but precisely ‘because’, they were interclassist. But at the same time, these interclassist policies proved effective not although, but because, they were centred on the working class. The sub-standard performance in the 1950s and 1960s of parties without deep working-class roots, like the SFIO and PSI, constitutes a contrario evidence for this claim. In reality, what a number of political scientists – including myself, for the sake of terminological convenience – have referred to as the social-democratic ‘catch-all’ strategy of the 1950s and 1960s was not a catch-all strategy, or a working-class strategy, or a ‘dual’ strategy (which, by definition, is impossible). It was a ‘semi-working-class’ and ‘semi-catch-all’ strategy: a strategy of ‘sociological broadening’.22
In short, the gradual crystallization of a significant audience among some segments of the middle classes (employees, new salaried strata), and the preservation of the working-class mould, form the two sides of social-democratic vitality, of the politically stable and electorally triumphant social democracy of the 1950s and 1960s. Compared with the beginning of the twentieth century and the interwar years, social democracy changed substantially in this period. But the architecture of its electoral space retained the structure – and depth – of yesteryear. The change was gradual and partial; the mutation was not a metamorphosis: it represented an evolution, not a revolution.
From this perspective, the very significant contribution of the salaried middle strata to the social-democratic electoral coalition was that of a pivotal social force. Without it, social democracy would never have had a majority. But without the working class, the very idea of social democracy would have been inconceivable. Its electoral core consisted in the working class, the primary and primordial source of power and effectiveness for a social democracy that was increasingly interclassist.23
Notes
1.Matei Dogan, ‘Political Cleavages and Social Stratification in France and Italy’, in Seymour Martin Lipset and Stein Rokkan, eds, Party Systems and Voter Alignments: Cross-National Perspectives, Free Press, New York and Collier-Macmillan, London 1967, p. 129.
2.The exclusion of housewives, pensioners and the unemployed leads to a certain underestimation of blue-collar workers in the social structure of the SPÖ. According to the data cited by Muller and Ulram, 75% of SPÖ voters in 1961 and 68% in 1969 came from a blue-collar occupational milieu (occupational milieu is the occupation of the ‘head of the household’) (Wolfgang Muller and Peter Ulram, ‘The Social and Demographic Structure of Austrian Parties, 1945–93’, Party Politics, vol. 1, no. 1, 1995, p. 148).
3.Francis G. Castles, The Social Democratic Image of Society, Routledge and Kegan Paul, London and Boston, MA 1978, p. 118.
4.Göran Therborn, ‘A Unique Chapter in the History of Democracy: The Social Democrats in Sweden’, in Klaus Misgeld et al., Creating Social Democracy: A Century of the Social Democratic Labor Party in Sweden, Pennsylvania University Press, Pittsburgh 1992, p. 11.
5.In reality, the SPD has been supported by even a relative majority of the electorate only once – in 1972, when it secured 45.8% of the vote against 44.9% for the CDU-CSU. Even during its best period (1969–80), the SPD always came second and its electoral averages range from 43.3% in federal elections to 40.8% in European elections (as against 46% and 49.2%, respectively, for the CDU-CSU). See Henri Menudier, Les Élections allemandes, 1969–1982, Centre d’Information et de Recherche sur l’Allemagne Contemporaine, Paris 1982, pp. 37, 81.
6.See also Gerassimos Moschonas, La gauche francaise (1972–1988) à la lumière du paradigme social-démocrate. Partis de coalition et coalitions des partis dans la compétition électorale, doctoral thesis, University of Paris II, 1990, p. 61.
7.In Norway we observe a stronger propensity, by European standards, for the old middle classes to vote for the left. Given, as in most European countries, the weak arithmetical weight of these classes (11% of the population in 1957; 8% in 1989), Norway may be regarded as a particular case, without thereby being distinguished from the general trend. Italy and Greece may also be regarded as separate instances. A feature of Italian and Greek society is the high proportion of self-employed work. The influence of the left (PCI and PDS in Italy, PASOK and KKE in Greece) among the old middle classes is far from negligible.
8.Monica Charlot, ed., Élections de crise en Grande-Bretagne, Presses de la Fondation Nationale des Sciences Politiques, Paris 1978, p. 19.
9.Anthony Heath, Roger Jowell and John Curtice, How Britain Votes, Pergamon Press, London 1985, pp. 32–3.
10.For a deeper, more detailed analysis of the electoral bond between the left and the middle classes, see Moschonas, La gauche française, pp. 59–70, 249–82, 300–12.
11.Working with a very large sample of non-manual Labour voters, Colin Railings demonstrated that 70.7% of its non-manual electorate belonged to the category of ‘employees’ and 56.3% regarded themselves as part of the working class. He also showed that the more a middle-class voter resembled a working-class voter in sociological terms, the more likely she or he was to vote Labour (‘Two Types of Middle Class Labour Voter’, British Journal of Political Science, vol. 5, 1975, pp. 108–9). Butler and Stokes, for their part, reported that in 1963 68% of those in the category of ‘lower non-manuals’ regarded themselves as belonging to the working class (Political Change in Britain, Macmillan, London 1970, p. 70). The ideal-typical model of this type of non-manual Labour voter, a modified and expanded version of Richard Rose’s model, comprises the following five characteristics: (1) possession of a minimal level of education; (2) non-performance of supervisory tasks; (3) membership of a trade union; (4) council house tenancy; (5) working-class origin in respect of intergenerational mobility (Colin Railings, ‘Political Behaviour and Attitudes among the Contemporary Lower Middle Class’, in J. Garrard et al., eds, The Middle Class in Politics, Saxon House, London 1978, p. 205; Richard Rose, Class and Party Divisions: Britain as a Test Case, University of Strathclyde, Glasgow 1969).
12.Frank Parkin, Middle Class Radicalism, Manchester University Press, Manchester 1968.
13.Ronald Inglehart, The Silent Revolution, Princeton University Press, Princeton, NJ 1977.
14.David Jary, ‘A New Significance for the Middle Class Left? Some Hypotheses and an Appraisal of the Evidence of Electoral Sociology’, in J. Garrard et al., eds, The Middle Classes in Politics, Saxon House, London 1978, p. 140.
15.Alain Bergounioux and Marc Lazar, La Social-démocratie dans l’Union européenne, debate in Les Notes de la Fondation Jean Jaurès, no. 6, Paris 1997, pp. 10–11.
16.Adam Przeworski and John Sprague, Paper Stones: A History of Electoral Socialism, University of Chicago Press, Chicago 1986, p. 35.
17.Adam Przeworski, ‘Social Democracy as a Historical Phenomenon’, New Left Review, no. 122, 1980, p. 44.
18.Richard Rose and Derek Urwin, ‘Social Cohesion, Political Parties and Strains in Regimes’, in Dogan and Rose, European Politics, Macmillan, London 1971, p. 221. The authors regard the Swedish and Danish social democrats, and the British and Norwegian Labour parties, among others, as homogeneous class parties; and formations like the SPD, SPÖ, PCI and PCF as possessing mutually reinforcing loyalties.
19.It might be argued that the distinction between social-democratic-type parties with maximum working-class penetration and those with high penetration corresponds, in large part, to the distinction between successful and less successful parties of the working class.
20.Regarding these two variants, constituted by the criterion of social cohesion, my distinction coincides in the main with the conclusions of Rose and Urwin in European Politics, pp. 221–5.
21.My analysis broadly coincides with the conclusions of Bartolini: ‘if a “golden age” of class alignment and of social homogeneity in socialist support ever existed, this was certainly not the early phase, but rather the intermediate period of the 1930s-1950s’ (Electoral, Partisan and Corporate Socialism. Organisatonal Consolidation and Membership Mobilisation in Early Socialist Movements, Estudio/Working Paper 83, Juan March Institute, Madrid 1996, p. 44).
22.See Moschonas, La gauche française, pp. 493–4. Kirchheimer does not identify the catch-all party with the interclassist structure of its electorate (see Otto Kirchheimer, ‘The Transformation of Western European Party Systems’, in Joseph La Palombara and Myron Weiner, eds, Political Parties and Political Development, Princeton University Press, Princeton, NJ 1966, p. 185). For a presentation and critique of the thesis of the catch-all party in Kirchheimer, see Steven B. Wolinetz, ‘The Transformation of Western European Party Systems Revisited’, West European Politics, vol. 2, no. 1, 1979, pp. 4–28. See also Karl Dittrich, ‘Testing the Catch-all Thesis: Some Difficulties and Possibilities’, in Hans Daalder and Peter Mair, Western European Party Systems, Sage, London 1985, pp. 257–66, especially pp. 258–9; also Daniel-Louis Seiler, De la comparaison des partis politiques, Economica, Paris 1986, especially pp. 94–9.
23.The precondition for the realization by social democrats of their ‘dual’ function as parties of the working class and national parties of integration (Leo Panitch, ‘Profits and Politics: Labour and the Crisis of British Capitalism’, Politics and Society, vol. 7, no. 4, 1977, p. 478), resides in their working-class entrenchment. It is thus not surprising if the most spectacular successes or the most crushing defeats of social-democratic parties have often been associated with fluctuations in the vote of ‘low-paid social groups’. For – to invoke Geoffrey Hodgson – I shall say that this is ‘surely the lesson’ of the 1960s: ‘Those who must live by the labour movement can also die by that movement’ (Labour at the Crossroads, Martin Robertson, Oxford 1981, p. 122). Some decades later, this was perhaps no longer true.