A model used in Australian labour historiography, which suggests a division of labour between political and industrial labour movements, respectively institutionalised in the ALP and the unions, presumes the ALP is the political leader of the movement. But, even before the 'third party' experiences of the last two decades, this leadership was challenged from within the working class, most sharply towards the end of and just after World War II. This paper discusses the basis and extent of that challenge, in particular in the electoral arena. It also considers the policy restraints on this challenge and the implications of this for politics within the working class.

In the years towards the end of World War II and immediately after, many workers in Australia became more class conscious. One aspect of labour movement activity related to this was a strike wave, but the close attention previously given to industrial disputes and union leadership battles in the period is alone insufficient to study the politics of these years in the working class: what, for example, workers read, what parties they joined and participated in, and how they voted ­ which receives particular attention here ­ also all changed during this period.

     Tom Sheridan's Division of Labour is the major study available of the labour movement in the years immediately after the war and an invaluable contribution to any discussion of politics in the working class at the time. It is the basis for the description of the strike wave in this paper. Sheridan's analysis is limited, however. He attributed the strike wave to a conflict between a workforce that was spontaneously aggressive in its drive for improvements in wages, hours and other working conditions, and the Australian Labor Party (ALP) government headed by Chifley, which resisted that drive. He also associated the successes of the Communist Party of Australia (CPA) in the period solely with the industrial labour movement: its (very significant) 'influence among unionists held up and rose slightly during the first two post-war years', because, he argued, the party's talents and energy supported the period's militancy.^[1]

     The limits of Sheridan's analysis are imposed by his model of a labour movement divided into industrial and political wings. Union activity is political to the extent it is exercised within or otherwise oriented to influencing the ALP. The political activity of the working class is placed in an overall framework in which the ALP is the point of reference, including even that of workers in other political organisations. The ALP is thereby made the necessary political leader of the class, in line with the common understanding that between the end of the first decade after federation until the 1980s at least, a fundamentally two-party system - of the ALP and non-ALP parties ­ existed in Australian politics. Once a broader working-class radicalism is involved, however, the possibility of a challenge to the two-party from within the working class is posed and the model is undermined. Of particular interest, then, is the activity of and support for the Communist Party of Australia, since that party appealed in various ways to radicalising workers, despite its Stalinism.^[2]

     According to Sheridan, however, the view of the CPA leadership about the party's political influence in the working class in the post-war strike wave moved from a realistic understanding that it had little to the 'almost total delusion' of leadership claims. Its left turn from 1947, in which it sharply attacked the ALP and asserted the CPA as the leadership of the working class through, for example, what Tom O'Lincoln describes as 'aggressively independent election campaigns',^[3] accelerated this movement, Sheridan said. He cited as an example of this movement ­ that is, as unrealistic - a remark of the CPA assistant general secretary, Richard Dixon, at the party's March 1946 Political Committee meeting. Dixon said there was a 'separation of very considerable sections of the workers from the Labor Party, and very considerable doubts in the minds of workers about the Labor Party'.^[4]

     A number of developments suggest Sheridan's assessment of working class politics must, however, be qualified. Between 1945 and 1947 ALP Industrial Groups were initiated in New South Wales, Queensland, South Australia and Victoria to intervene in the union movement (in Western Australia, the union peak council was inside the ALP's structures: an attempt by 15 unions to form an independent labour council were successfully resisted). This attempt by the ALP to establish avowed control of the unions was unique.

     According to Sean Scalmer, the Industrial Groups were the response of the party's parliamentarians to what they perceived to be a novel threat to them from the union movement. Their leadership in the formation of the Groups followed their tendency to act against the labour theory of democracy, which involved the labour movement entering politics through providing information to the ALP and controlling its parliamentarians. Scalmer argued the ALP parliamentarians' response had become necessary when in government they opposed workers' desires expressed through ALP structures such as its conferences. The party could not, as it had before, unify labour's political interests. Driven by the war effort, it overlooked the views of unionists and party members, and ignored newly emerging political forces. So, according to Scalmer, women and intellectuals became important political constituencies for the CPA. Thus the CPA became a political alternative to the ALP. ^[5] This was despite the generally conciliatory approach of the CPA towards the ALP before the left turn. It had supported the Chifley government's re-election. It was against the ALP's disaffiliation of and refusal to affiliate some CPA-led union branches ­ at least eight in Queensland and the ARU in Victoria ­ which had, for example, supported the election campaigns of non-ALP pro-worker candidates.^[6]

     For politics in the working class, however, the character and course of the strike wave and the relationship of this to CPA policy was still more important. The strike wave extended throughout the country, including into the normally industrially quiescent South Australia and Tasmania. It involved many workers who were either unorganised, only recently organised, or not traditionally militant or effectively organised, including those suffering from racial and gender oppression. These included: Aboriginal workers in Darwin and the Pilbara, women (clothing workers, tram conductors and waitresses, for example), bakers, BHP iron and steelworkers and some operative staff, postal workers, pastoral workers, and tobacco workers. In the strike wave, industrial and political activity also connected, through the demands of some workers' actions and other agitation, but especially in the question of party organisation within the working class.

     Thus, the strike wave of the 1940s approached a mass strike more than any other in Australian history. This is not contradicted by: its relatively prolonged character; nor the defeats or compromises workers took or made in some disputes, since these occur in any strike wave. As Ken Buckley and Ted Wheelwright put it, the period 'offered a matchless opportunity for working class advance'. However, that they pose this only in terms of reducing exploitation of workers, and argue the ALP federal and state governments held workers back from this, should be noted.^[7] The strike wave had its 'prehistory'. The level of industrial action rose from 1943 onwards. Women workers, coalminers and waterside workers were among those most active.

     The CPA, in its effort to be the 'leading war party', related all industrial questions to efforts to increase production. It therefore supported strikes only in defence of fundamental issues or to forestall unorganised outbreaks. Otherwise, it opposed them even to the extent of organising scabs and expelling workers from their unions. This cost it support, most spectacularly among the Balmain ironworkers, who struck to defend the position of a Trotskyist job delegate against their union's federal leadership. ^[8]

     At the end of 1945 and the beginning of 1946, strikes countered employer aggression against wartime gains such as union recognition, job control and the employment of militants. The two major strikes of this period centred on BHP's steelworks and on meatworks in Queensland. Other strikes, such as among powerhouse workers and of footwear workers, who opposed the sacking of communists at the Enoch Taylors factories in Sydney, were also significant instances of militancy. The CPA supported these reactive struggles, although with reservations about unbridled militancy. Its policy until the 1946 federal election was to restrain industrial action.

     After the September 1946 federal election the strike wave surged again. The metalworkers campaign, in which the Amalgamated Engineering Union (AEU) played the leading role, was fully resolved only in June 1947. At this time the CPA, along with other militants, supported industrial struggles that expressed workers' expectations of the ALP government.

     Some important post-war demands, such as the 40-hour week and an end to the pegging of marginal payments for skills, were largely won in this upsurge of struggle. Of course, the militants still suffered frustrations: for example, Ted Rowe, a communist AEU leader, admitted drawing the union's members who weren't already activists into strike committees was difficult; among the rest, he said, the craft tradition that understood militant action as simply depriving the employers of their skilled labour was strong. However, the AEU had refused to accept a proposed resolution to the metalwork dispute that favoured its members over the non-craft members of the Federated Ironworkers' Association.

     The strike wave therefore began to decline from the middle of 1947, coinciding with the CPA's left turn. The major industrial disputes of 1948 and 1949, in Victoria, the Queensland railways and the coalmines, resulted from further delays or difficulties in the implementation of or exclusions from the gains now won elsewhere, or from opposition to legislation or courts which sought to restrict industrial action.^[9]

     The relationship of the strike wave to CPA policy during these years contradicts Sheridan's argument about clashes between the party leadership and its leading trade unionists. He argued this was because the latter's views were moderated by the influence of the overall union membership and the primacy of union militancy as a political orientation among the ranks of CPA members.^[10] However, for much of the period under discussion ­ from 1943 to 1946 ­ working class action was often more militant than that supported by CPA officials in the party and the unions alike. Even after the left turn there were still instances of industrial militancy which aligned with the CPA leadership's outlook, such as the reaction in Victoria to essential services legislation and the pressure of the coalminers for action in support of their claims.^[11]

     Thus, the support for the CPA among workers in the mid-1940s cannot be explained simply as a close alignment of it with industrial militancy, although the occurrence during the strike wave upsurge in 1946-47 might help explain a brief consolidation of party membership that followed. The CPA, instead, combined several features that appealed to radicalising parts of the working class. It had a tradition of militancy. It continued to encourage workplace organisation ­ and, for example, as late as June 1949, 5000 NSW railway workers rallied under shop committee leadership to the CPA's platform at the Domain in Sydney. It appears, too, to have maintained a radical image, even when it encouraged class collaboration and proposed unity with the ALP during the war or after the war acted as a ginger group to the ALP focusing its criticism on the latter's right wing. Its assertion of the hegemonic outlook that, in Craig Johnston's words, 'the working class had the opportunity to seize the leadership of the anti-fascist war',^[12] and its dominance of working class party organisation outside the ALP may have played a part in this.^[13]

     On the basis of this appeal the CPA became a significant threat to the ALP political hegemony in the working class.^[14] The CPA challenge to the ALP for the leadership of working class industrial organisations and movement was an important part of this, because of the significance of this sphere in the day-to-day political activity of the class. However, the influence the CPA gained there did not necessarily express support for the broader outlook of the party, especially since only a few unions, chiefly the Waterside Workers Federation and the Seamen's Union, frequently engaged in industrial action for political demands.^[15] With regard to its overall policy, party recruitment and membership retention, its presentation of electoral candidates and its newspaper distribution all more directly gauged support, although in different ways: the first of these substantially measured those willing to actively propagate the views of the CPA, the second marked the limits of the success of its appeal and the third indicated to what extent the party was gaining a hearing.

     CPA membership peaked at an estimated 23,000 in 1944. However, even when it fell to 16,280 in 1945 and 13,450 in 1946, it still approached one-quarter of the ALP's individual membership. The ALP's membership also included union affiliations, of course. However, the political activity of CPA members would generally have been more than that of ALP members from branches or affiliates.^[16]

     A number of factors complicate an analysis of the CPA electoral results.^[17] One such factor is that competing 'Labor' trends with electoral clout existed. Their significance as expressions of a breaking up of ALP influence and of a radicalisation in the politics in the working class is varied and not immediately clear. Lang Labor in NSW was anti-communist and won several state seats. The State Labor Party, also in NSW, was associated with the Communist Party: the proportions of its support from Communist and Labor voters are unclear. There were also independent Labor candidates, sometimes backed by dissident ALP branches, who often received large votes and were elected to the Queensland parliament and at the state and federal levels from the Coburg area in Melbourne. However, while these unofficial Labor parliamentarians and branches collaborated with Communist front organisations, none was consistently radical, although Tom Aikens, from the expelled Hermit Park ALP branch in Townsville, worked with the sole Communist MP, Fred Paterson, and opposed the state ALP government's emergency legislation in 1948.^[18]

     Another unusual feature of the times was the breakdown of the urban-oriented conservative party organisation, previously the United Australia Party, and the rebuilding of this as the Liberal Party. As a result, many electorates were contested solely by ALP and CPA candidates, or otherwise only with other Labor candidates. In this circumstance anti-Communist and other conservative voters may have supported an ALP or other Labor candidate to oppose the CPA candidate, but some ALP supporters, not fearing an ALP loss, might equally have voted for the CPA candidate to press the ALP to carry out a more radical policy.

     Finally, the CPA ran candidates for relatively few lower house seats and not at all in the Senate, even though its membership was most likely capable of carrying out a much broader campaign. Thus, its electoral support across the country was not directly tested. The seats it did stand for in an election were often quite varied, especially in federal elections, and seem to have failed to include many in which it was likely to have gained greater votes - from some coal districts and parts of northern Queensland, and some factory and port suburbs- than others in which it did stand. Moreover, only in a few cases did it put up a candidate for the same electorate in one election after another.^[19]

     In the 1943 federal elections, CPA candidates ran in 17 of the 74 electorates and averaged about eight per cent of the vote in these. In the 1946 elections, they ran in 14 seats and averaged a little more than seven per cent of the vote. When compared with the results of the 1934, 1949 and 1951, when more lower house candidates and Senate tickets were run, with the average voting results for the former lower by at least one-third, and the latter gaining more than two per cent of the vote each time, the 1943 and 1946 results suggest a national CPA vote of up to five per cent was possible at these times. This is not an insignificant 'third party' vote ­ it compares well with many since the 1980s - but it is still only a small proportion of the votes of all workers.

     In federal and state polls, however, signs of a situation similar to that in the coalmining community of Cessnock, as described by Robin Gollan ­ 'a political community removed one step to the left of the rest of Australia [where] the politics are almost exclusively working class: the Labor Party is the conservative and the Communist the radical party'^[20] ­ can be observed. If the seats where 70 per cent or more of the vote went to CPA and any Labor candidates are considered, then in elections held between 1943 and 1945, at least one-quarter of the combined vote for those candidates was for the CPA in half these electorates. Overall, for these seats the CPA vote averaged more than 20 per cent. But in similar electorates in 1946 and 1947, except for those in coalmining districts, some North Queensland seats and Adelaide's port, the CPA vote had diminished towards relative insignificance, and its average vote had dropped to 15% or less (see Table 1). However, the fall in CPA electoral support seems to have accelerated through 1946 into 1947. In the sole election held in 1945, in November in Victoria, the CPA vote across a number of inner-city seats was still more than 20 per cent (see Table 2).

     The range of falls in CPA votes between the federal elections of 1943 and 1946 where it had candidates in the same seat each time was also less that for the various state electorates in 1947 where the CPA had a candidate in the previous election (see Table 3), although this may reflect a peak of support in 1944.

Table 1: CPA vote of all CPA and Labor votes, when latter 70% or more in electorate
Electorates (election) 1943-1945		Electorates (election) 1946-1947
Bowen (Qld 1944)	53%	Bowen (Qld 1947)	56%
Herbert (Qld 1944)	43%	Herbert (Qld 1947)	31%
Port Melbourne (Vic 1943)	39%	Bulli (NSW 1947)	20%
Williamstown (Vic 1945)	3 5%	Port Adelaide (SA 1947)	17%
Kennedy (Qld 1944)	34%	Hartley (NSW 1947)	15%
The Tableland (Qld 1944)	33%	Cessnock (NSW 1947)	15%
Balmain (NSW 1944)	33%	Kurri Kurri (NSW 1947)	14%
Richmond (Vic 1943)	32%	Hindmarsh (fed. 1946)	13%
Bulli (NSW 1944)	31%	Stuart (SA 1947)	12%
Collingwood (Vic 1943)	31%	Hindmarsh (SA 1947)	12%
Clifton Hill (Vic 1943)	26%	West Sydney (fed. 1946)	11%
Carlton (Vic 1943)	26%	King (NSW 1947)	10%
Lakemba (NSW 1944)	26%	Newcastle (fed. 1946)	10%
Richmond (Vic 1945)	24%	Paddington (NSW 1947)	10%
Port Melbourne (Vic 1945)	24%	Phillip (NSW 1947)	9%
Waratah (NSW 1944)	22%	Yarra (fed. 1946)	9%
Carlton (Vic 1945)	22%	Cook (fed. 1946)	8%
Clifton Hill (Vic 1945)	19%	Newtown (NSW 1947)	7%
Ballaarat (Vic 1943)	17%	Redfern (NSW 1947)	6%
Yarra (fed. 1943)	15%
Newcastle (fed. 1943)	14%
Melbourne (fed. 1943)	13%
Kalgoorlie (fed. 1943)	11%
Dalley (fed. 1943)	8%
Mean per electorate	26%	Mean per electorate	15%
CPA vote across all above	22%	CPA vote across all above	12%

Table 2: CPA votes in the 1943 and 1945 Victorian elections
1943 election			Of all CPA and Labor votes	1945 election		Of all CPA and Labor votes
(Electorates fought in both elections)
Carlton	22%		26%		22%	22%
Clifton Hill	19%		26%		14%	19%
Port Melbourne	39%		39%		24%	24%
Richmond	32%		32%		24%	24%
Ballaarat	12%		17%		6%	12%
(Other electorates)
Collingwood	25%		31%	Williamstown	27%	35%
Footscray	19%		33%
Prahran	13%		25%
Upper Yarra	11%		30%
Mean per electorate	21%		29%		19%	23%
Across all	21%		29%		20%	22%

Table 3: Change in CPA vote between elections in 1943-1945 and 1946-1947 where candidates in consecutive elections
1946 election compared with 1943 election
Adelaide (fed)	3%
Kooyong (fed)	-23%
Newcastle (fed)	-37%
Yarra (fed)	-43%
Perth (fed)	-48%
Herbert (fed)	-55%
1947 election compared with 1944 election
Bowen (Qld)	-11%
Bulli (NSW)	-35%
Adelaide (SA)	-43%
Herbert (Qld)	-50%
The Tableland (Qld)	-50%
Kennedy (Qld)	-50%
Balmain (NSW)	-68%
Lakemba (NSW)	-69%
1947 election compared with 1945 election
Port Melbourne (Vic)	-74%

A similar timing in the fall of support for the CPA is suggested by the fate of the campaign to lift sales of its Sydney-based newspaper Tribune. After the end of the war an increase in sales of 10,000 was projected. At first partial successes were reported, but by October 1946 the newspaper reported 'a steady decline over the last few months' in its circulation.^[21]

     Thus, a defence against the charge that CPA left turn was mistaken - for example, the one mounted by Tom O'Lincoln, which argued that instead, 'a pattern of manipulative and bureaucratic behaviour . was an important reason' for the defeats the CPA suffered during and after the 1949 coal strike^[22] ­ seems rather by-the-by. Regardless of the party's views or long-prevalent Stalinism, such defeats were likely because the political radicalisation in the working class that developed towards the end of World War II began to decline from 1946. This decline reached significant proportions in 1947.

     The rise and fall of the working class radicalisation at the end of World War II also puts the course and results of the strike wave in a new light. Strike waves are often understood to create the conditions for rising working class political consciousness.^[23] Yet in this case, the wave of radicalisation preceded that of industrial action. By the time of the 1946-47 industrial upsurge, however, capital could more readily make concessions to workplace-related demands to the extent that these were no longer seen as encouraging a working-class challenge to the rule of capital. This development was reinforced by some others. First, the mass of the working class experienced an ever-extending period, from the end of the war, of relative economic stability and low unemployment. In Queensland and in the coal industry ­ sites of major 1948-49 strikes - this was less true than in most other parts of the country and in other industries. For most workers, however, such an experience would have increasingly contradicted concerns about the return of economic depression. Second, some groups of workers who might have played a positive role in the political radicalisation of the working class did not do so. The outstanding example in this regard is that of the Broken Hill miners, who in the past had been a leading militant and radical section of the class. They were in the midst of a decade and a half of growing employment within a labour market protected by local employment requirement and spiralling contract and lead bonus earnings, did not take part in the strike wave and their shop committees tended to decline from 1946 onwards, partly because of a pay deal that withdrew union sponsorship of them.^[24] Third, after 1947, the ALP's industrial intervention was significant. In the coal strike, according to Sheridan, the key to the majority of rank-and-file coalminers losing confidence that they could win was the decision of NSW land transport unions, led by the ARU branch, to move stockpiled coal, and the AWU's threat to work open cut mines. Both of these unions were under ALP leadership. More broadly, the mobilisation of the ALP Industrial Groups was now routing the previously relatively successful Communist interventions in the ACTU and, also, in some unions and regional labour movement organisations. Participation in union bodies could fall rapidly with Grouper success ­ for example, in the Boot Trade Employees' Federation. From 1949 many shop committees, where the CPA had played a leading role in the development of union workplace organisation for up to two decades, were weakened when they lost official support.^[25]

     For some time before the middle of 1947, however, the character of the strike wave, the relative weight of political leaderships in the union movement and initial concerns about post-war economic instability had combined to create more favourable conditions for the working class radicalism that found an organisational expression through the CPA. The party's left turn, so much discussed, does not appear to have been nearly as important as the party's policy before it and how this might have slowed the growth of the party's influence. Sheridan's conclusion noted:

The government's labour policy was explicitly based on the assumption that there should be no major redistribution of income from profit to wages. In the years when price control was effective, such a redistribution was exactly what most unionists implicitly expected as their due in the postwar world. From almost any point of view, of course, this expectation would appear to demonstrate the political naivety and lack of understanding of the contemporary ALP among Australian unionists. The Chifley government's plans simply did not encompass so daring a notion. The unions' failure to identify and clearly publicise this fact was compounded by the moderates' increasing distaste for direct action .

The task Sheridan suggested for the unions was, however, one not for workers' industrial organisation but their political organisation, which would have been the CPA. Moreover, such an approach was thought of: for example, Jack Blake, the party's Victorian state secretary, proposed the CPA pursue united action campaigns more vigorously, while withdrawing support for the government and engaging in a 'consistent campaign of enlightenment' about reformism and the ALP. This was rejected by the CPA leadership. Therefore, the CPA continued to lack the orientation to rank and file agitation of seeking to conduct the class struggle on the political plane, which had helped bring about its earlier achievements in the coalfields, maritime communities and parts of northern Queensland, and so failed to consolidate more broadly the creation of communities where the politics was almost entirely divided between conservative and radical working class outlooks.^[26]

     A model of the Australian labour movement which divides it into industrial and political wings in the institutionalised form of unions and the ALP distorts our understanding of it. Recognition of the movement's historical political domination by the ALP is needed. In the middle of the 1940s, however, working-class political radicalism, principally expressed through the CPA, was at the point of beginning to challenge for hegemony in the class. This suggests the fundamental framework for understanding politics in the working class should be a consideration of its division between opportunist and revolutionary trends and of the social bases of these trends.