During the referendum campaign – just as in 1975 – Cabinet ministers have been allowed to dissent from the government’s pro-Remain position. Chris Malone examines why the suspension of Cabinet collective responsibility is such a significant moment in British political history, and the ways in which ministers habitually reconcile their individual positions with the collective line. Despite the ‘blue-on-blue’ attacks of recent weeks, they will likely revert to the fiction of internal consensus when the campaign is over.
In his speech to the Commons on January 5, the Prime Minister announced that the longstanding convention of Cabinet collective responsibility would be suspended in the run-up to the EU referendum. Following the conclusion of his renegotiation, ministers would be granted the freedom to openly express conflicting views on the issue – in other words, to dissent from the Government’s official position if they so decided.
To the surprise of few, that declared government position turned out to be one of continued EU membership, and the feared Eurosceptic dissent from within Conservative ranks was swift, with a number of prominent party MPs publicly backing the Vote Leave campaign. Alongside backbenchers such as Boris Johnson, this also included a number of serving Cabinet ministers and members – Michael Gove, Chris Grayling, Theresa Villiers, John Whittingdale, Priti Patel, and for the short time preceding his ostensibly-unrelated resignation, Iain Duncan Smith. Whilst the charismatic Johnson’s allegiance may prove more politically significant, the open divergence of the latter group, breaking the otherwise-united front of cabinet government, remains of greater constitutional import.
Cabinet collective responsibility, a convention with roots as far back as the 18th century, is a key mechanism by which the executive branch of government is ensured to speak with one voice. Under the agreement, itself a prerequisite of Cabinet membership, ministers subject themselves to the combined restraints of secrecy and unanimity in decision-making. How the Cabinet comes to form its final policies remains shielded from the public gaze, and alongside this confidentiality, what they ultimately decide is taken to be equally binding on all members, regardless of their individual positions and preferences during that preceding deliberation. Cabinet decisions are shared decisions, formulated and upheld as a group. The only way to express dissent is therefore to leave the group altogether, by resigning from one’s ministerial position. This uniform stance emanates from the Cabinet but extends to similarly constrain the government as a whole, including its more junior ministers and post-holders.
The convention’s efficacy is clear. Aside from protecting sensitive information, inside the Cabinet confidentiality creates a safe environment for the frank expression of divergent views and untested ideas, benefitting the scope and quality of policy discussion. Alongside this, outward solidarity – both in what Cabinet members say and in how they vote – maintains the external perception of strong, united governance. As the government’s supreme decision-making body the Cabinet needs to give clear direction and exert leadership over the parliamentary rank and file, not to mention the country at large. A cabinet prone to regular public disagreement over major policy issues would be both ridiculed and practically ineffective. We need only look at the uncomfortable contortions of the current shadow cabinet on issues such as Trident and UK intervention in Syria to see the wider political damage open disunity and mixed messages can create.
David Cameron’s suspension of the convention is thus more than a trivial point reserved for aficionados of parliamentary procedure. There are real questions as to what its relaxation means for governance. Are we witnessing a constitutional shift as significant as those debated in the EU referendum itself?
It’s essential, firstly, to outline some limits on what has actually been relinquished and what has not. The secrecy of Cabinet discussions was never intended to be relaxed under the PM’s waiver. We have not and will not see the contents of formal Cabinet meetings laid bare, in contravention of the usual thirty-year embargo on their minutes. Nor is his declaration broad in scope. This is not the dissolution of cabinet government in favour of issue-by-issue advocacy by individual ministers. Rather, as his letter of clarification to colleagues states, ‘this wholly exceptional arrangement will apply only to the question of whether we should remain in the EU or leave’. Outside of this very specific policy debate in the context of the referendum, and even on other closely-related matters concerning Europe and the EU, Cabinet members ‘continue to be subject to the normal rules of collective responsibility and party discipline’.
No matter how limited its remit, however, the move to interrupt collective responsibility and allow ministerial dissent against the government is far from routine. Genuine equivalents in the modern era are exceedingly rare, and always contentious. The most frequent suspensions have come during periods of coalition government, when policy divisions have traced clearly along party lines. Alongside the 1932 Cabinet split over tariff policy, such exemptions to unanimity – so-called ‘agreements to differ’ – were similarly outlined during the 2010-15 coalition in relation to electoral reform, university funding, Trident renewal, nuclear power and married tax allowances, with opposing ministers permitted to outline alternative views or abstain from related parliamentary votes.
Yet the atypical conditions of coalition government are markedly different to those of a majority, single-party administration. With longstanding policy differences already on record, there is simply not the same public expectation of internal concord, and thus fewer consequences of dissonance being revealed.
Instead, the most salient precedent to today’s state of affairs occurred, with striking parallels, when the UK last held a referendum on EU (then EEC) membership in 1975. With Cabinet colleagues ‘free to advocate a different view’ to the official recommendation of Harold Wilson’s Government, the sight of serving ministers Roy Jenkins and Tony Benn facing off against one another on BBC Panorama became one of the enduring images of the campaign, with divisions laid uncomfortably bare on national television. Present ministers, unlike their 1975 counterparts, may so far have avoided direct confrontation, but successfully navigating this abnormal schism over an extended campaigning period remains a challenge fraught with difficulty.
In light of the PM’s announcement, much public commentary initially focused on the practical necessity of the convention, and how a break from this would be managed. Would the civil service offer advice to both sides of the campaign? Would dissenting ministers be able to answer questions on the EU in the Commons, or be required to continue attending Cabinet meetings in which ‘in’ campaign planning occurred? Below the institutional level, less formal tensions have similarly arisen. Can individual relationships within the Cabinet endure the prolonged discord and reputational mud-slinging split campaigning inevitably involves? Will political survival, aspirations for promotion or even future leadership be predicated on which side of the debate ministers line up on, and which wins? While the first set of questions have been necessarily settled the latter continue to rage.
Nevertheless, the full significance of withdrawing Cabinet collective responsibility is not simply procedural, practical or even constitutional, but philosophical. The British Cabinet may be composed of separate individuals, but key to its political role is that it functions as a group agent, a body whose members together form genuinely collective attitudes. Philosophers, following Margaret Gilbert, sometimes refer to such groupings as ‘plural subjects’, in the sense that the subject of a particular attitude (its ‘owner’) is no longer an individual – ‘I believe’ – but instead a number of individuals, the group as a whole – ‘We believe’.
We might wonder what all the fuss is about. Surely all that it means for a group to judge, believe, desire or intend something is that its individual members judge, believe, desire or intend it – a mere aggregation of what the human beings populating it think? Saying, for instance, that ‘the electorate desires change’ is surely to suggest that all, or at least some significant proportion, of the voters constituting that multitude so desire. Talk of groups themselves having attitudes is on this view no more than useful but fictitious shorthand, a simplifying metaphor. Is this not a case, therefore, of academic philosophers over-theorising what is a pretty simple and trivial observation, of positing new conceptual entities when they aren’t required?
It’s by no means unreasonable to assume this. In many cases the attitudes we would assign to a group ‘as a whole’ do happen to reflect the mutual or majority personal attitudes held by that group’s individual members. But deeper analysis can quickly reveal that this isn’t always the case, and that the two attitude levels are in fact logically distinct.
Take the example of a central bank committee, whose members are equally divided over whether to increase the national interest rate by 0.5 percent or 1 percent, and so decide to split the difference at 0.75 percent. No individual committee member originally believed a 0.75 percent increase was optimal, and on a personal level – that is, thinking simply as John, Sarah, Robert, and so on – they may each continue to believe in the merits of their own position. The agreed policy on the interest rate is merely a compromise, yet it is one which each member, in unison with the others, has nonetheless agreed to accept – indeed, to commit to – as being their collective stance, the belief of the committee taken together.
This collective attitude is no empty metaphor. It is the attitude which now guides the behaviour of those committee members every time they are acting in their official capacity. When a committee member briefs the media, or begins to plan the future agenda, or works to implement the bank’s policies, it is this collective position that now acts as their lodestar; it is a belief that 0.75 percent is the appropriate figure that prescribes their group-representing conduct. Their personal beliefs concerning the interest rate – what they may think in private, argue for at the next internal review meeting, or discuss over their family dinner table – are no longer the relevant ones to reference when shaping their contribution to the collective project. Harmony at the group level does not necessitate unanimity at the personal level. In fact, as demonstrated, it doesn’t require even a single person to be personally aligned.
It doesn’t take much reflection to see that this ‘group-mode’ thinking and behaviour happens all of the time, not just with bank committees, but company boards, parent-teacher associations, scientific bodies, commissions of inquiry, interview panels, unions – whenever a policy, view or appointment is decided on behalf of a group in its own right. It’s also what we see happen with the Cabinet. When ministers speak on record they speak as members of that government body, they express the collective view of that ‘we’, not the personal view of an ‘I’ (though occasionally still using the latter pronoun).
The importance of this may be easily overlooked, but the Cabinet operating as a plural subject – a ‘we’ – is the very thing that allows its identity as a unified actor to endure across changes in personnel. Ministerial reshuffles frequently modify the membership of the Cabinet, with some MPs exiting the body and others – possibly with very different views – filling the vacancies. Despite this fluctuation, we still perceive agential continuity in the group; we think it right to hold this body accountable for its past decisions even when a number of different faces now sit around the table. It is the existence and persistence of group-level commitments, belonging to the collective itself, that enables this.
If the attitudes routinely ascribed to ‘the Cabinet’ (and more broadly ‘the government’) really were just shorthand for the amassed personal attitudes of current ministers, then group attributions would be little more than snapshots, valid only as long as people with those attitudes populated the body. Moreover, while those specific individuals would continue to be personally responsible for their past views and choices, neither new members nor the collective as a whole would be. Political accountability would be solely personal and potentially short-lived.
Instead, collective agency provides for both the sharing and carry-over of responsibility. On entering the Cabinet new members commit themselves to the maintenance of its existing group-level attitudes, the standing commitments of previous incarnations. In other words, they are expected to support and defend even those policies formed prior to their appointment. These policies can of course change (and frequently do), but such occurrences are perceived as just that – a change in the collective attitudes of the same group agent, not the extinguishing of one and formation of another. From ministers’ own perspectives, a policy shift signals a change in what we think, not a change from what they, some separate group, thought – merely changing attitude content rather than ownership.
This enduring responsibility of course has limits, notably failing to traverse electoral changes in governing party. When membership is modified wholesale, commitments to collective positions – the very things that bound past members together as a ‘we’ – are indeed allowed to expire, and need to reconstituted afresh. We do not think it legitimate to hold the current Conservative Cabinet to account for decisions made by past Labour ones. Though each iteration is still ‘the Cabinet’ in a formal, institutional sense, there is a clear break in their agential identity – each is a different plural subject, a different ‘we’.
As such, the Cabinet’s collective qualities are far from self-sufficient. Group agents are human constructions that require concerted will and effort to maintain; the Cabinet as an ongoing bearer of attitudes and responsibility does not exist independently of human behaviour. To maintain genuinely collective attitudes, members need to publicly endorse, support and justify the stated stance of the collective, refrain from disputing or contradicting it, and generally do whatever is most apposite in demonstrating the existence and operation of a group-level view. If they consistently fail to do this when acting in their representative capacity, then effectively those group attitudes will not exist. There is no other foundation.
The convention of Cabinet collective responsibility is the institutional mechanism for ensuring this necessary behaviour occurs, and when properly enforced it makes this political body one of the most secure examples of collective agency we can find anywhere. Ministers are locked into their roles as spokespeople for the group’s attitudes as a formal condition of their continued membership. Yet with the convention’s suspension for the EU referendum we instead see a very different state of affairs. In place of this desired cohesion we witness a Cabinet in the throngs of open disagreement, with supposed colleagues lobbying for rival policy agendas while at the same time publically questioning each other’s judgement, record and even trustworthiness. This must surely threaten the Cabinet’s very coherence as a collective actor.
What, then, is going on conceptually when the Cabinet’s united front fractures, when it speaks with multiple voices? Least damaging, and possibly reconcilable with the body’s continuance as a group agent, is if defiant ministers were interpreted as expressing mere personal contrary opinions – their views simply as individual citizens, or politically as just ordinary, constituency MPs. Nothing prevents personal and group-representing statements being explicitly differentiated: ‘Personally, I favoured awarding the literature prize to Smith, but as a whole we, the committee, decided Jones was the worthy winner.’ Though undermining any convenient impression of internal consensus, so long as this collective stance continued to be reaffirmed alongside personal views, the attitudes of the group itself would remain operational. It would still be accurate to say that the committee backed Jones.
The PM’s guidelines for relaxing collective responsibility follow precisely this track, outlining special permission for ministers ‘to take a different personal position from the official position of the Government.’ Emphasis here is on the word personal; permission does not extend to withdrawing support for the Government position entirely, including as group representatives. Dissenting ministers, quite sensibly, have thus been keen to present their critical remarks as being of this limited nature, with Michael Gove signing off a recent letter of denunciation as the MP for Surrey Heath versus his Cabinet title of Justice Secretary.
Nevertheless, despite both sides’ best efforts it is hard to paint the current discord as being so constrained. As noted, expressions of individual disagreement are benign to shared agency only so long as they are coupled with a continued stance of group-member support for the corresponding collective view: ‘I prefer that book but we together prefer this one’. Yet examining the increasingly hostile language of the referendum debate, this essential second element – revalidation of what the body as a ‘we’ ought still to think – has been practically non-existent from dissident Cabinet members. The PM’s own guidance, stipulating that such ministers are not expected to contradict their ‘long-standing and sincerely held views’ when speaking from the frontbench, has no doubt contributed to this particular absence.
Instead, the rebellious comments of Brexit ministers have at times actively chipped away at this group-level unity. The above-mentioned letter of June 5 not only criticises the government for previous U-turns on EU budget payments, but argues that the public therefore ‘cannot trust’ its promises at present. The latter clearly challenges rather than affirms what the collective attitudes of the executive are. But moreover, throughout this discourse the government is exclusively referred to as ‘it’, not as ‘we’ – as if it were a separate grouping those ministers did not belong to. Offering neither support for current group-level positions on the EU, nor acceptance of joint accountability for past ones, this must be interpreted as a refusal to still consider the Cabinet or wider government a plural subject of such attitudes.
Ministers backing Remain have, by contrast, tended to ally their own personal remarks with a continued espousal of the executive’s supposed group-level position. With the two presumably in alignment this is of course far easier to do. Nevertheless, with other colleagues breaking rank it is no longer philosophically accurate to say that either the Cabinet or government as institutions really do have a stance on the EU referendum, whether for or against. If genuinely collective attitudes do still exist within these bodies, then they are in force only amongst those conformist members who sustain them with representative behaviour.
In other words, if there is a ‘we’ that collectively believes the UK should remain an EU member, then the catchment of that plural subject no longer matches the recognised boundaries of corresponding institutions. The formal and agential identities of both the Cabinet and government, indelibly aligned under the usual convention of collective responsibility, have instead been allowed to come apart.
What does this revelation mean for British politics? Will the Cabinet stop functioning, governance disintegrate, policy stances break down into incoherence? Practically speaking, the answer is clearly no. Five years of functioning coalition demonstrated the incredible durability and flexibility of cabinet government on a day-to-day level. Politics will find a way to go on, irrespective of philosophical creases. Indeed, even from that more abstract perspective, the failure of ministers as a whole to sustain a collective stance on the question of the EU need not signal the end of their broader collective unity.
The Cabinet’s enduring group agency is bound together by many strands, not a single thread. So long as open disagreement is limited in time and scope to the confines of the current debate, and collective responsibility continues to be observed in relation to the vast majority of other policy areas, there will be sufficient issues on which the group continues to think as a ‘we’ to allow it to continue functioning as such. If the increasing tendency of Brexit ministers to slide, without sanction, into questioning wider government policies continues, such unity might begin to unravel more seriously, but much will depend on how this dissent is managed once the referendum is over.
Politics is as much practical and symbolic as it is formal and principled. The suspension of collective responsibility pulls back the veil on a helpful illusion of internal consensus, but little contemplation is needed to recognise it as a fiction to begin with. Ministers commit to collective positions because it is an effective way to transcend continued differences. Yet while tried and tested conventions are essential to the long-term resilience and effectiveness of government, they can – to a limited degree – be stretched, reshaped and put aside when the issues of the day demand it. So long as these issues are ultimately resolved, few are likely to complain. There really is no collective government position on the EU referendum, if the government consists of all its ministers. But for the sake of political practicality, we should be comfortable with pretending there is.
This post represents the views of the author and not those of the BrexitVote blog, nor the LSE.