Since the 1983 general election there has been a marked change of political mood in Britain: a recovery of morale by Labour; a series of misfortunes for the Tories. At the same time there has been comparatively little change in the underlying political realities. This contrast between mood and reality can be misleading and even in some respects dangerous. The very hard questions posed by the defeat of last June can be softened or allowed to slip away in the name of necessary resilience or a merely foolish optimism, each strengthened by the passage of time. It is for this reason that we have still to take seriously the questions defined by Eric Hobsbawm, with his usual plainness and lucidity. I disagree strongly with the answers he suggests or implies, but I find myself at an even greater distance from those who think they can dispose of the questions by simple and self-righteous cries against ‘coalition’, which in this context is often little more than a swearword.

It is true that it has long been the purpose of right-wing and centrist commentators to get rid of the Labour Party as a significant independent political force. This is also the explicit intention of the Alliance parties. Indeed for more than thirty years there has been a sustained attempt to form a left-of-Tory party or grouping in which the socialist component would be minimal or altogether excluded. Moreover it should not be forgotten that elements of this attempt have come from within the Labour Party including, at times, its leadership. It would be a simple matter for the whole of the current discussion to be absorbed by this long and dangerous campaign.

But this would be too easy. The current problem is not so much in the campaign itself as in the deteriorating political situation to which it is a response. It is a fact that we have a very dangerous right-wing government elected by only 43 per cent of the electorate. Yet appeals against its legitimacy, on this ground, cut little ice against its actual monopoly of state power. It is also a fact that on all recent and current evidence there are now three parties or party groupings each capable of gaining at least 20 per cent of the popular vote. Both Labour and the Alliance believe that they can make the other go away and there are some signs, among the ideologists of each party, that they give higher priority to driving out the left-of-Tory rival than to opposing the main enemy. They rationalize this by the belief that the only way to defeat the Tories is to dispose first of the alternative anti-Tory vote. But it is then reasonable to ask how this displaced emphasis will affect real political opinion, when it is Tory support that has really to be reduced. Reasonable also to ask what will happen if, regardless of the energy devoted to such efforts, it turns out that a three- or four-party system has come to stay. This connects with the alarming central fact that the ‘Labour vote’ had declined to 28 per cent: a low point only reached in any comparable situation during a period when there were, as now, three contending parties. It is tempting to link this obvious correlation between so low a vote and the existence of three parties to the tactic of giving priority to eliminating or reducing the third party. But this is especially dangerous for the Labour Party because it obscures or postpones the question that has really to be answered: what kind of party is it? Is it a broadly socialist party, or is it (as it became during the generations of Liberal decline) the only realistic left-of-Tory coalition?

For we have in fact been living with one kind of coalition politics for many years, inside and – on both wings – outside the Labour Party. The real question raised by Hobsbawm is whether that kind of coalition – the Labour ‘broad church’ and its friends – can be successfully continued or revived, or whether some new kind of coalition is now necessary, given the rise of a third party grouping. It seems clear that he would be glad to see either, though with a preference for the first. It is here, especially, that the grounds of the whole argument need to be widened. On the hitherto limited electoral terms no useful resolution of the argument seems to me possible. The true context of any practical politics is always the general social and economic situation, and only secondarily the party dispositions and shares of the popular vote which follow from this. One obvious weakness of recent electoral analysis, as with the analysis which followed Labour’s defeat in 1959, is that it treats current distributions of votes as if they were primary data from which the social and economic situation, or at least the main responses to it, can be inferred. There is a related habit of inventing social entities in the form of ‘the Labour vote’ or ‘Labour voters’ and so on. It is true that there are some significant numbers of people who vote consistently for this or that party over a long period, and that relative increases or decreases in these numbers are significant. But these groupings taken together are never anywhere near the sum of the electorate. A pattern of multiple shifts this way and that, even at times appearing to cancel each other out in a relatively unchanged general distribution, has been characteristic of British politics since the 1950s and has increased markedly in recent years. These multiple shifts have been in evidence even since the June 1983 election.

Moreover, although speculative reasons for shifts are often advanced, very little is known about their causes. Some appear to be settled shifts in social affiliation; others are almost certainly short-run impulses. In practice, it is impossible to infer from these, or from relative changes in more sustained groupings, either the true social situation or the requirements of political practice. This is especially the case when there are more than two electorally significant parties. It is instructive that the erstwhile ‘Liberal vote’ and the more recent ‘Alliance vote’ have been exceptionally ‘unstable’ in these terms. This is only one of several reasons why we should not begin a political analysis from ‘shares’ of votes, but from the more important and more objectively discoverable general situation.

Whether Britain is viewed in isolation or, as should be the case, the uncertainties of ‘Britain’ are seen in the context of a critical and uneven world politics and world economy, the general situation is even more unstable and volatile than any pattern of voting.

Britain, in so far as it is still an autonomous economy, is now a weak and exposed sector in a severe and prolonged world crisis. As far as it is still an independent political nation, it is a junior partner in a system of military alliances, and an uncertain partner in the attempts at a Western European community which is itself now in crisis.

Within these very broad determinations, all of which indicate continued and dangerous uncertainty and instability, there can be varying assessments of what will happen within the sphere of British electoral politics. The orthodox view from the left assumes a steady, chronic degeneration of the existing social order: a still failing economy, continued mass unemployment, a greatly reduced welfare state, an extending domination of capitalist values.

The orthodox view from the right, though it speaks of recovery and regeneration, transposes these damaging changes into the vital reconstitution of a profitable capitalist economy. This is not to be confined to Britain alone, but to be extended by the worldwide deployment of British capital. Continuing mass unemployment, reduced welfare costs, weakened trade unions and the defeat of socialist ideas and organizations are the deliberate conditions of this reconstitution. So also is the pursuit of an aggressive and heavily armed foreign policy in the struggle for control of key areas of the newly industrializing and dependent poor world.

Another orthodox view from the left goes beyond the usual account of ‘Tory mismanagement’ and understands current Tory policy as a hard and, in its own terms, rational programme. It then assumes that there must eventually be a gathering of political opposition to its inevitably heavy costs, of a kind that will return the Left to power. This is very much better than the older assumption that there will be sufficient opposition to mere ‘mismanagement’: ground on which Labour, from its recent record in government, is in no strong position to fight. But it may still not be strong enough, for there is a third possible assumption, different both from the hard Right and from the orthodox Left in either of its versions. What is interpreted from the left as a process of chronic degeneration, and from the right as a process of profitable capitalist reconstitution, is seen from this third perspective as imposing strains on the whole social order which will radically change the terms of British politics. A minority of Marxists believe this will lead eventually to some kind of pre-revolutionary situation in classical terms. But alongside them other Marxists see the danger of a further break to the right, with much harder authoritarian controls to contain the pressures which the capitalist reconstitution must inexorably increase.

This perspective does not often appear directly in current arguments for left coalitions or similar arrangements, but I think it is no coincidence that its kind of response – in effect an updated version of the popular front – is now coming from some eminent Marxists. Nor, in any real historical perspective, should this kind of thinking be dismissed by mere labelling. There are undoubtedly possible circumstances in Britain and in the rest of Western Europe in which the organization of a popular front would indeed be a priority. The historical lessons of the defeat of the Left in Italy in the early 1920s and in Germany in the early 1930s cannot be forgotten, and must indeed still be taken seriously. But any such proposal must be assessed in contemporary terms. The historical memories and the futurist projections have to be related to where we are. It still matters very much which of these assumptions of the nature of the next ten years we adopt, but in any case we have to move on from retrospective accounts of the movements of votes and into this real area of political analysis.

One useful way of testing the arguments about electoral policies and arrangements is to set them in their real or probable political context. Let us suppose, for example, that the 57 per cent of votes against the present Conservative government had not been distorted by an absurd electoral system but had produced a majority of non-Conservative representatives. On what kind of political programme would they have been able to agree? The answer is, I believe, disconcerting to both main sides in the current argument, but at first sight it is most disconcerting of all to those who dismiss the whole discussion with cries against ‘coalition’. For it would almost certainly have been possible to form a government with the following main heads of policy: first, deliberate reflation of the economy by extending the public sector borrowing requirement to some figure between the Labour and Alliance proposals; second, cancellation of cuts in the welfare services, education, transport and general infrastructure; and third, refusal – or at least delay – of Cruise missile installation, and introduction of Polaris into general disarmament negotiations. Obviously there would be other areas in which simple agreement would be out of the question – wages policy, trade union legislation, the electoral system itself. But the three areas cited above are of such commanding importance that in practice few people would turn down the chance of realizing them, whatever other disagreements remained. Moreover, such policies would, without question, produce some marked improvements in our present circumstances. They have only to be compared with current Tory policies for that to be evident.

It is from this conclusion, and from seeing the alternative as prolonging current Tory policies while the non-Tory vote remains so divided, that the coalitionists – whether explicit or not – draw their strongest arguments. But of course any such coalition is hypothetical. There are, as Hobsbawm fairly said, substantial objective and subjective obstacles to its practical realization. One of these is the electoral system itself, currently a major difference between the prospective partners. Even where this is modified, as in the 1983 election, by a good deal of tactical voting (which, it should be noted, itself distorts the crude figures of shares of the vote), there is no way of effecting a practical coalition of policies except by negotiations before the election. Yet such bargaining instantly challenges the full national ambitions of the separate parties, and is indignantly rejected.

What then happens is an appeal to balance the political advantages of some new arrangement – positive advantages, as in the three heads of policy; negative advantages, in that at least the Tories would be out – against what can be seen as merely residual and hidebound party positions. Alternatively, these probable gains are seen to outweigh the likelihood that one of the potential partners, Labour or Alliance, can in the next four years so thoroughly defeat or reduce its anti-Tory rival that it will itself gain majority power to execute both the main heads of policy and its other, more particular, commitments. Put this way, the appeal seems very strong.

So what’s wrong with it? The answer lies in the politics rather than in the electoral calculations. The main objection follows from the fact that, in the present electoral system, such arrangements have to be made in advance. For even if we could agree that current objections to such arrangements are decisively outweighed by their political advantages, the actual effect would be a relative freezing of current policy positions: not only agreement on the main heads of policy, which in real terms would not be difficult in current circumstances, but agreement also that certain other kinds of policy, while they could of course be independently retained, would be relatively played down so as not to threaten any proposed, plausible agreement. Thus there would at least have to be some loose coalition of policies before any practical electoral arrangements were possible.

It can, of course, be argued that this is desirable, as a way of achieving at least the major policies. Or it can be said that this kind of advance agreement is very much more desirable than a post-election coalition resulting from a hung parliament, in which broadly similar policies would probably be agreed – but behind closed doors between the leaderships rather than in the open between full parties. In fact we cannot now say which of the two election results relevant to this argument is more likely next time: a Tory government even on a reduced minority vote, or a hung parliament in which some kind of coalition would inevitably occur. Perhaps one of the best reasons for having this argument in the open now is to force us to think not only about the relatively improbable pre-election agreements but also about the much more probable post-election problems in which Labour would be embroiled if it failed to get its own majority.

Everything, in the end, must come back to the real politics, and here there is a more challenging argument. For we have to ask what real differences there are between these proposals for what can be called the big coalition – Labour and the Alliance in any of its possible forms – and what has to be called the smaller coalition, which is that version of the Labour Party which draws on the same types of argument as the explicit coalitionists – pressing for electoral unity around certain main heads of policy while other differences lie on the table – as a way of maintaining the practical coalition of diverse tendencies which the Labour Party has long been. There is one clear difference between the two kinds of coalition. To the extent that the Labour Party maintains or extends its recently improved democratic structures so that policy decisions are of an openly discussed and contested kind, the implicit coalition of socialists and social democrats within the party will be continuously active, as distinct from manoeuvred coalition agreements between separate leaderships. At the same time we can all imagine circumstances in which the appeal for electoral unity could be made to override this active process, and some of the same arguments as those of the big coalitionists would then be deployed: the need to maximize the vote against the real enemy; the subordination of contentious policy decisions to the unifying imperatives of electoral organization.

One version of these arguments is acceptable – at least initially. All of us who have experienced the defeats of Labour and the Left not just as analysts or observers but where they really hurt, in the lives of our own people, are understandably determined not to be defeated again. In fact, from this determination a new kind of politics can really begin. But there is another version of what are apparently the same arguments which leads us directly back into the old politics of defeat. The distinction between the versions is not primarily in relation to elections: it is in relation to actual policies. For what has to be said very clearly is that if Labour’s major policies are broadly the same as those of the Alliance, it would be very foolish indeed not to seek some mutually beneficial electoral arrangements. (There is an obvious practical limit to how many centre or left-centre parties there can be: even two is already too many in the present electoral system.) Of course it is indignantly denied in both camps that there is any such identity. Minor, marginal or dispensable differences are maximized as a condition of becoming the only true contender for the sensible anti-Tory vote. And if this sensible ground is where majority public opinion now is, but two or three parties are contending to represent it, there are only two reasonable tactics: driving out the rivals or coming to an arrangement with them. Hobsbawm assumes such a ground, and while he would doubtless prefer Labour to drive out the Alliance he is realistic enough to face its great difficulties and to contemplate the alternative.

Yet to be reduced to a choice between these two limited tactics would be the most important defeat which the Left could now suffer. The reasons for this are political. They become clear if we look again at the main heads of policy which might now be agreed in some anti-Tory coalition, or which might alternatively be most strongly emphasized in Labour’s drive for its own government. None of the policies is in any distinctive sense socialist. Reflation of the economy, in the terms usually proposed, is a continuation of Keynesianism. Restoration of the welfare state is in the broad tradition of the liberal and social-democratic consensus out of which the Labour governments emerged. Limited measures of disarmament are within the broad internationlist and peacemaking consensus of the same tradition at its best. None of these identities proves any of the policies wrong or insufficient. Yet the underlying identity, while it is held at this practical level, indeed makes dividing and splitting votes between parties which broadly adhere to it foolish. What is wrong, from any socialist position, is the definition of such policies as adequate for any sustained recovery or advance. Nor is it enough to add in some of Labour’s more distinctive policies, such as opposition to the Common Market, opposition to incomes policies or repeal of anti-union legislation. None of these makes the basic body of policy more realistic or coherent. It is in the main heads of policy themselves that there must be significant and convincing socialist development, if we are to substantiate Labour’s claim to be a unique alternative not only to Tory policies but also to the liberal/social-democratic consensus now electorally available in the Alliance. Some people think that this development can be brought about by a bold announcement of a commitment to socialism. But for many years this has been not the solution but the problem. Nominal socialist commitment has broadly coexisted with actual liberal/social-democratic policies, and this has led to confusion inside the party as well as to justified criticism of it by others as incoherent and unconvincing. What really needs to happen in the next four years is a radical reconstruction of all the main directions of policy in the light of the most open and informed contemporary socialist analysis.

So in the case of the general direction of the economy, it is necessary to move beyond ‘one-nation Keynesianism’ in each of its component terms. It is clear from the experience of the French socialist government that economies of the size of Britain or France cannot pull away on their own from the main forces of the international capitalist economy. Thus coordinated or integrated policies with other left governments are a condition of any sustained success and need to be carefully discussed and agreed in advance, at least in general outline. One obvious route for this is within the EEC; it could mean changing the residual option for negotiated withdrawal to a policy of forcing the pace for a coordinated socialist reconstruction of a group of European economies and thereby reforming the EEC itself. But the simple Keynesianism also needs to be changed. Centralized management of credit and the money supply, as in simple reflation policies, has to be only a part of a much more developed process of democratic economic planning and control, including hard selective policies on investment, prices, taxes and incomes.

This connects with the need to go beyond a simple welfare state. In the current employment and demographic crisis it is impossible to isolate an area of services and benefits recommended only in their own terms (pensions and compensation for early retirement are only the most obvious examples), without reference to the direction and priorities of the general economy. Opportunism in this area has already been heavily paid for politically, and it is essential to rework the principles and costs of the system as part of a general policy on investment, taxation, employment and benefits.

On the third head of policy, in relation to peace and disarmament, it is necessary to transcend the limited holding initiatives which are now being emphasized and rework from the beginning a coherent and sustainable international policy. This must include realistic relations between the requirements of British security and the problems of membership of either a nuclear or a non-nuclear military alliance. It must also include policies for the reconstruction of economic and political relationships with the Third World – a greatly neglected area. It is here especially that general monetary and economic policies beyond Keynesianism interlock as matters of political struggle with policies to change the role of British finance capital in the international economic order.

These are brief examples in relation only to the three main issues previously defined. A whole range of other social policies has to be developed in coherent relationship to them. It is then easy to say that getting even provisional agreement on them by majorities in the Labour Party would be a formidable political task. For it would be particularly important that the policies be developed well beyond the status of conference resolutions into genuinely practical programmes. There would need to be professional detailed and continuously updated work drawing on the research resources and practical experiences of the whole labour movement.

Yet even this would not be enough. The whole point of this new political direction would be the attempt, by informing and educating each other in the hard realities of the contemporary world, to launch the widest possible public process of reconsidering and (where necessary) changing every popular assumption, habit and attitude.

As it happens this would also be the best possible kind of electoral campaign, with the organization of an electoral machine genuinely powered by an expanding socialist awareness and conviction. Otherwise, as so often before, there will be arguments about who is to drive, which maps will be used, even how best to tune the machine. And still there will not be enough bloody petrol.

This kind of politics is, I am sure, the only immediate and practical alternative to the politics of overt or covert coalition. I am not saying that the answers to come out would be the ones known and already advanced by the left of the party. On the contrary, that assumption – still common on the Left – is merely the way to more divisive factionalism. Until the work has really been done and had come to include detailed, convincing answers to the objections already made and acted upon by the right of the party and beyond it, there is no ready-made socialist programme to translate into simple resolutions and majorities.

The long neglect of fundamental research and political education has produced an uneven but unmistakable mixture of half-formed policies and half-convincing protests. Much of the most essential detailed work is being done outside or at the edges of the party – in the peace movement, in the women’s movement, in the ecology organizations – and all these bear especially on that politics of the future to which Labour must now redirect itself from the depths of defeat. But there is also promising work inside the party itself as well as great potential within the trade unions and their research departments for work on the central issues of employment and investment – an area in which all convincing policies need to be very specific. Through these various channels and through the Socialist Society, the Fabian Society and constituency political education officers, the necessary work can be done. The true intellectual resources of the labour movement have never been so rich and diverse, and the political problem is to bring them to bear in general public debate rather than internal dispute.

The campaign must be much more than ‘bringing the message’ or even ‘winning the intellectual argument’ (though that, too, is necessary). In one linked area after another what we really have to discover – and as far as possible agree on – is what that intellectual argument actually is: the fully contemporary intellectual argument for socialism. This is why coalitions must be opposed. Whether it’s the big or the smaller version, the advocates of either have in effect abandoned the struggle to transform belief and opinion. In a cold climate, they say, the many but now disparate remnants of decent and sensible opinion must huddle together, pooling their surviving resources against the Tory storm. I can see how easy it is to feel like that or to respond hopefully to a few brave words flung back against the wind. I also know that the kind of campaign for renewal which I have been describing has been proposed before, and has never fully happened. Are these words, too, no more than a cry against the wind?

It is for many of us to answer. If this new kind of politics is too hard for us, if there is too little time or if we already believe that these more radical tactics must fail, there are still answers – indeed now common answers. Ruling the new politics out, or merely paying it lip service without the practical changes which need to go with it, leaves plenty of room for other kinds of political activity: we can sustain the smaller coalition without any real work on policies, or reach out for the larger coalition, adapting ahead of its formal arrangements by trimming or underplaying those innovative socialist policies which are known to be incompatible with it. But we can then draw a clear line, to our mutual advantage, between socialists and coalitionists. We can begin to see where we really are, and what we have to change.

First published in the original New Socialist, this essay is republished with permission from Verso Books. “Socialists and Coalitionists” is available in Resources of Hope: Culture Democracy and Socialism, edited by Robin Gable and published by Verso.