Scenes of emergency: Dis/re-assembling the promise of the UK emergency state

The paper traces the development of UK ‘state of emergency’ legislation through three ‘scenes of emergency’: the introduction of the Emergency Powers Act in 1920, a revision to the Act in 1964, and discussion within government departments about possible changes to emergency powers in 1973. Through these scenes, and contra to existing work on the state of emergency as an occasion for the intensification of sovereignty, I show how the introduction of and revision to ‘state of emergency’ legislation were occasions for a double concern – with the excessiveness of the state, as per Foucault’s analysis of liberalism, but also for the excessiveness of events. In ‘scenes of emergency’ a specific ‘state effect’ was dis/re-assembled: the promise of the providential state that protected life through control of events. As emergency legislation was subject to deliberation and contestation, other versions of the state surfaced: beginning with the interested, classed, state and the tyrannical state as emergency powers were introduced and ending with the anxious state that loses faith in the efficacy of emergency powers in a world of changing events. As well as arguing that work on governing emergencies should be orientated to ‘scenes of emergency’ in which that which governs relates to excess, the paper suggests that assemblage approaches to the state should be concerned with dis/re-assembly.

time a declaration had been made in three years, the then Secretary of State for the Home Department, Robert Carr, concluded with a wish to end the powers the Government held: "Finally, I need hardly say that it is the profound wish of the Government, and of the public at large, that these disputes should be speedily and fairly settled, so that the Government can ask for the emergency powers to be revoked. I assure the House that the Government will make as much, but no more, use of these powers as is absolutely necessary to fulfil our primary duty to the people of the country as a whole, and that we shall bring the powers to an end as soon as we believe it is safe to do so." 1 The use of emergency powers was justified by Carr on the grounds of necessity. The "present situation" authorised the declaration, alongside the assumption that the state had a "duty" to protect the "essentials of life". Declaring an emergency was presented by Carr as a secondary act that happens after the onset of and in the midst of a 'present situation'. Connected to the characterisation of action as "prudent" and "wise" 2 rather than excessive were gestures of reluctance and self-limitation by the Secretary of State. As well as the promise of revoking them "as soon as we believe it is safe to do so", Carr goes on to stress the possibility of the non-use of the various regulations. The message on behalf of the Government read by Lord Windlesham in the House of Lords expressed a similar reluctance: "The use of the powers will be limited, as always, to what the essential public interest requires." 3 But in a 'present situation' in which the action of a trade union was identified by the Government as the cause of threatened disruption and the use of emergency powers had become normal, the charge from the opposition was that government action was political. Opposition MPs accuse the government of "almost" having, "declared war on the miners" 4 and of being "reprehensibly dishonest" by using the miners' action as a "political alibi in order to prejudice public opinion against them." 5 In response, the Secretary of State returned to his characterisation of the situation: "There is no doubt in my mind that, because of the situation arising out of industrial action being taken in the coal and electricity industries on top of the uncertainty about oil, we are right to have taken these powers, and right to have taken them now. If the Opposition persist in their charge that there was some sinister political motive for this action, or for its timing, that, I believe is more a commentary on their approach to these matters than on ours, and I am prepared to let the country judge which is the right course to take." 6 Introduction: The emergency state Emergency declarations such as those issued in 1973 are typically associated with the foreclosure of deliberation, contestation, and other habits and actions taken to be necessary for democratic action (see Scarry, 2011).The claim of necessity by the state -and associated affect of urgency and the presence of an interval for action (Anderson, 2015) -authorises exceptional action. However, the use of 'state of emergency' legislation in 1973 precipitated an intensified scene of contestation. It was precisely the claim of necessity (and the linked claim of disinterest on the part of the state) that the opposition subjected to heated and repeated contestation. The anti-emergency claims by the opposition were various -that a state of emergency was not necessary at that time, that it was not necessary as other action was possible and desirable, and that the government made the declaration for cynical or interested reasons -but what reoccurred was an attempt to disrupt the status of necessity as the authorising ground for the declaration of an emergency.
In this paper, I describe three scenes across the twentieth century in which UK emergency legislation was subject to deliberation and contestation: the introduction in 1920 of the legislation the 1973 declaration was based upon, an amendment to the original legislation in 1964, and discussion within government throughout 1973 on the need to revise the legislation. Drawing on archival research, the paper tells the hitherto untold story of twentieth century emergency legislation in the UK (although see Agamben (2003: 19) for brief comments). To date, critical work in geography, security studies, and international relations has focused on the contemporary arrangement of emergency governance in the wake of the war on terror, placing the UK in the context of changes in how events and conditions are governed in a turbulent world (see Amin, 2013;O'Grady, 2018;Zebrowski, 2015). Complementing this work, I focus on the emergency legislation that pre-dated the current 2004 Civil Contingencies Act (on which see Anderson and Adey, 2012). By staying with the contestation that surrounded emergency legislation during the twentieth century, I show that actual or imagined emergencies were problematic for the UK state: occasions where the state reasserted a key promise -to protect life and act neutrally -and occasions where that promise became fragile. As the state reasserted the promise that it would respond effectively and neutrally, other versions of the state proliferated. In the above scene, imaginative 'stateeffects' (after Mitchell 1991) surface through competing emergency and anti-emergency claims: the prudent state, the provocative state, the cynical state, etc.
In addition to telling the story of emergency legislation in the UK and showing how scenes of emergency serve as occasions for particular forms of contestation rather than only the intensification of sovereign power, I explore what the state is and becomes in relation to emergency. I do so in order to supplement emerging assemblage related approaches to the state by advocating a focus on scenes that confound lines between disassembly and reassembly. Current work on the state in the wake of Feminist and poststructural approaches attempts to show how the effect of the state -the affective-imaginative and material appearance of the state as a unitary actor separate from society -is achieved through a host of mundane practices (e.g. Coleman and Stuesse, 2016;McConnell, 2009;Painter, 2006). As Desbians et al. (2004: 242) put it: "the state is not a unitary object but is, rather, a set of practices enacted through relationships between people, places, and institutions". The mistake is presuming what Mitchell (1991: 78) calls the "coherence, unity and absolute autonomy" of the state (see also Painter, 2006). As with many other research objects in contemporary human geography, the task of analysis becomes to describe and thus reveal the multiplicity and contingency of a seemingly unitary object. The hope of this approach is that revealing contingency weakens the unitary object, and opens up the possibility of change. This leads to a problem, though: how to account for the affective, material and imaginative presence of the state as a unitary actor? The problem haunts Mitchell's (1991) vocabulary in his influential paper on state effects. After Foucault, he describes 'the state' as an "almost transcendental entity" (Mitchell, 1991: 94, emphasis added), a "practical yet ghost-like effect" (91), and focuses on how a line is drawn "internally" within a "network of institutional mechanisms" (78) to produce the "apparent autonomy of the state as a freestanding entity" (91). The same problem surfaces across recent work that develops from Mitchell. The state, as Woodward (2014: 23) describes it, "arises like a tune in one's head that can be neither shaken nor identified". It is in relation to this problem of unity in the midst of contingency that assemblage-related approaches have offered a solution. They provide a conceptual vocabulary and a disposition or ethos which emphasises the gathering together of disparate parts in "open wholes" (DeLanda, 2016). By describing processes of assembly, rather than formed assemblages, they also avoid any presumptions about form (so avoiding treating the state as a unitary whole). The paper begins from this approach, but develops it by arguing that assemblage related approaches need to be supplemented by a focus on specific occasions in which processes of assembly and dis-and re-assembly blur and become indeterminate. 'Scenes of emergency' are one such occasion in which 'the state' is simultaneously being disassembled and reassembled as the excess of events is made present.
These arguments develop across four sections. The first section supplements recent work on 'governing through emergency' with assemblage-based approaches to the state, by arguing that understanding the liberal emergency state requires staying with how it encounters and relates to the excess of events. This leads to an emphasis on processes of dis/re-assembly in 'scenes of emergency'. The rest of the paper traces how the promise of the state was (dis/ re)assembled in 1920, 1964, and 1973. By staying with the emergency and anti-emergency claims that animate the three scenes, I argue that emergency legislation becomes an occasion for the surfacing of anxieties about the authoritarian state and then the unprepared state as the liberal concern with the excessiveness of the state mixes with concern for the excessiveness of events. The paper concludes by setting out the implications of the orientation to 'scenes of emergency' for understanding the emergency state, and articulating the contribution of a focus on dis/re-assembly to existing research on the (de)formation of the state.

Section 1: Forms of emergency
Recent research has documented the role of emergencies, disasters and other disruptive events to the emergence, endurance and intensification of the modern 'state effect' (Mitchell, 1991): the state as a "totalising totality" (see Ophir, 2007). The affective/ imaginative-material effect of the state as a powerful, unified, totality -or what Ophir (2007: no pagination) calls "an imaginary singular person and an imaginary integrated totality" -is in part enacted through the state's promised control over the occurrence of ordinary and catastrophic events within a territory.
For Ophir, the emergence of this modern 'state effect' is shared by both of the chronologically and formally distinct state apparatuses through which states govern emergencies: the 'catastrophic state' that threatens harm and the 'providential state' that promises protection. Both state apparatuses -and Ophir uses the term state apparatuses and state formations interchangeably -'tend towards' emergency, in the sense of being orientated towards an external event or situation. However, for Ophir, the providential emergency state and the catastrophic emergency state differ formally in their relation with life and death: "The catastrophic state governs an already abandoned population; the body and living conditions of any individual or of many at once may be hurt, damaged, or destroyed altogether without changing anything in the state's modus operandi. The providential state governs -or may sometimes simply reach without governing-people whose right to live, sometimes even to live well, it has recognized . . ." (Ophir, 2007: no pagination) Ophir argues that although there is a chronological difference between the two apparatuses, they have now fused in the midst of the extension of the 'war on terror'. Nevertheless, in distinction from Agamben (2003), Ophir stresses the formal distinction between care and destruction as relations with life. The coexistence and entanglement of providential and catastrophic relations in particular apparatuses is, for Ophir contra Agamben, a contingent matter. What is shared across those relations with life is, though, that the 'state' emerges through its relation with emergencies and the harm, damage, loss and so on that emergencies threaten. By offering a formal distinction, Ophir pushes us to understand the different ways in which a 'state effect', in Mitchell's (1991) terms, emerges as states governs (through anticipation, response, recovery, and other practices and logics) events or situations that threaten it. For the providential state organised around the ideal and promise of absolute control to achieve absolute prevention, for example, emergencies become an occasion where, as Ophir (2007: no pagination) puts it, "the gap between this perfect potentiality [of the providential state] and the impaired and partial reality is exposed". Typically, this 'exposure' of a gap through formalised practices such as public inquiries or via demands by affected publics leads to practices that aim to improve the state's capacities to predict and respond.
The assumption being that the state is the kind of actor that could and should offer protection and control events (notwithstanding that this expectation and promise is unevenly attached to by different publics, including racialised groups who live in disastrous proximity to the workings of the catastrophic state that brings violence; Grove, 2014;Wakefield and Braun 2014).
Ophir's work is important because it reminds us how central events and responses to them have been to the reproduction of the modern 'state effect', although we might also remember how emergencies can be occasions for the refusal, contestation or breakdown of that effect (after Schmidt, 2018, on the 'refusal' of state-effects). However, an important consequence of Ophir's formalism is that he does not focus on how exactly emergencies pose problems for particular states. We can extend Ophir's work by focusing on how emergencies can be troubling occasions for any liberal state which, even after the contracting of the material actuality and promise of the state through neoliberalism, promises protection to a population (on the relations of 'failed states' to emergency see Duffield, 2007).States govern events or situations as emergencies because, in some way, they threaten harm, loss, damage or suffering to 'something' the state values and recognises (whether that 'something' be an infrastructure, people, living beings, or an atmosphere of control and continuity). The claim by the state is that harm and damage is happening and/or is imminent, but an 'interval' (Anderson, 2015) of time remains for action. Nevertheless, in an emergency something that appears to be outside of the state, some kind of excess, disrupts and threatens to break and end its capacities to act and secure life.
The trouble that emergencies pose for and to the state has not been central to existing critical work on emergencies which, after Agamben (2003), has in the main focused on emergencies as occasions for the intensification of sovereign power. Staying with the trouble that emergencies pose requires that we shift from a formal analysis of specific relations with life to how 'the state' takes form in relation to emergencies -where 'the state' is at once a material, imaginative, and affective formation. Recent work on the assembling of specific ways of governing in, through and by emergency demonstrates the specific forms through which 'state effects' are achieved in relation to events (e.g. Adey, 2016;Fredriksen, 2014;Grove, 2014;O'Grady, 2018). As such, assemblage influenced approaches to the state in and through emergency shift analysis from an Agamben informed concern with the relation with life to the ensemble of elements (including relations with life) that are held together for emergencies to be governed.
The emergency state might, then, exist as a similar kind of object of analysis to any other seemingly coherent 'thing' that assemblage as concept or ethos has been applied to. The orientation is to assembly; the situated, provisional, processes through which an 'open whole' comes to form (e.g. Dittmer, 2017;Mu¨ller, 2015;Squire, 2015). However, returning to and staying with the problem that emergencies pose to the state might interrupt this now almost habitual gesture of understanding anything and everything that appears to be coherent in terms of ongoing processes of assembly. Whether actual or anticipated, emergencies are occasions where things fall apart, or potentially fall apart. What falls apart varies, it may be people's lives or a fantasy of the state's omnipotence, a plan that is rendered unworkable or an existing piece of legislation that becomes out of date, but in all cases something may end or change as the state relates to excess. By which I mean that in an emergency or in anticipation of emergency some kind of excess that does not fit with a current arrangement becomes a problem for the state and exerts a material and/or imaginative and/or affective force (see O'Grady (2019) on elemental forces as one form of excess). Excess is present affectively at the event of an accident, for example, in the confusion and disorientation of responders as they attempt to make sense. It may be present in the halting words, pauses and recollections that make up testimony at an inquiry as what happened is retrospectively pieced together. Excess also happens as politicians justify a new piece of emergency legislation in parliament on the basis that existing legislation is no longer appropriate for today's complex world, or in a control room as staff work collaboratively to work out what, if anything, is happening.
The emphasis on the problem that the excess of events poses to the state as it attempts to govern emergencies shifts attention in assemblage related work to the complex entanglements between assembly and processes of dis/re -assembly. Emergency states are constantly being disassembled but also reassembled in relation to excess. Despite the emphasis on the provisionality and precariousness of assemblages which follows from the orientation to agencement and the emphasis on assemblages as 'open wholes', processes of de/re assembly have received less attention in assemblage related work on emergency and elsewhere. Notwithstanding the importance of de/re-coding and de/re-territorialization to Deleuzian versions of assemblage theory (see DeLanda, 2016;Dewsbury, 2011), the emphasis has been on the non-linear emergence and composition of assemblages. 7 By contrast, what I am calling 'scenes of emergency' involve actual and/or possible disassembly as the excess of events overwhelms, ends, disrupts, intrudes, questions, or otherwise acts. As used here, disassembly is an umbrella term for processes of removing, ending, subtracting, fraying etc in which previously assembled and stable parts, and/or relations between parts, are changing in ways that may shift assemblages between states. Consequently, processes of disassembly may result in something seemingly formed falling or coming apart. By 'scenes of emergency' I mean any occasion where states have proximity to the problem of excess and therefore disassembly may happen: testimony in an inquiry, debate during parliament, discussion between government departments, exercises that enact response, as well as the space-times where incidents and accidents happen. More precisely, a scene of emergency is future orientated -including when past events are made present (e.g. in an inquiry) or a current event is happening (e.g. as an event is unfolding). They involve anticipatory practices and statements which make futures present in a manner that enables the excess of events to be made actionable. In public inquiries, car crashes, planning meetings and other 'scenes of emergency', the state encounters events and those events shift between different 'modes of eventfulness' (Berlant, 2011) -from threatening catastrophes to recognised occurrences, from vague worry to looming threat, from stable situation to outbreaks, to name but some modes.
However, the state is not only being disassembled in 'scenes of emergency'. It is also being reassembled -in the sense that parts and/or relations are reconnected in a bid to enable the continuity and continuation of phenomena. By which I mean that some aspect of the statethe credibility of a promise, for example -is on a threshold. It is in-between falling apart and ending and being preserved or remade and continuing. By invoking the spatio-temporal image of a threshold, I am trying to attune to an in-between state where change is happening but the outcome is uncertain and an assemblage has not-yet qualitatively shifted. In this threshold, processes of dis and reassembly combine to produce emergency states. Importantly, disassembly and reassembly vary across the state's material, imaginative and affective dimensions, not necessarily happening across all three simultaneously. It might be, for example, that the emergency state persists as a more or less coherent material formation with an imaginative unity, but the cluster of promises that envelop it may 'fall or come apart' and have to be remade. Consider, for example, a public inquiry into the response to an event as a scene in which the state's promise to respond frays and falls apart, as the event is made present again and structural and situational failures are revealed amid anger, recrimination and other affects of loss and injustice. The state as material actor persists, but its promise is lost, even if momentarily, and even allowing for the already unevenly distributed attachment to the state's promises. Of course, an inquiry is also a scene of optimism in which that promise can be repaired -that failures can be identified and the state can improve its response. 'Scenes of emergency' are also, therefore, occasions where aspects of the emergency state may be reassembled. Of course, in some 'scenes of emergency' the state may 'fall or come apart' as a material actor, as well as affective-imaginative formation, and not be reassembled. Tracking the becoming-event of the 2018 hurricanes in Puerto Rico, Bonilla (2020), for example, follows how people's attachment to the state gradually waned as the emergency state failed to materialise in an extended interval of emergency. As people waited, harm intensified, and the state failed to respond, what fell apart and was lost was both the state as a material formation and as a cluster of promises that people attached to.
My focus in this paper is on 'scenes of emergency' in which the promise of the providential state that protects from emergencies is disassembled and reassembled. 8 By which I mean that what is at stake is the existence and legitimacy of the state's claim that it both can and will protect the population. The promise involves both the state's will to act -that the state is neutral and wishes to act to protect -and its capacity to act -that the state has the potential to respond. 'Scenes of emergency' vary. The scenes in this paper are all of a specific type: discussion in parliament and between government departments surrounding the need for emergency legislation. Excess is not materially present in these scenes. There are no dead bodies or raging fires. Rather, the excess of events is imaginatively and affectively present through verbal and written statements about future events and future states. The excess of events is present through practices which create future presents of damage and destruction that challenge the credibility of the promise of the neutral and/or capable state. These are not, let me stress, scenes in which 'the state' as a material actor falls apart, particularly given the constitutional role of parliamentary critique and scrutiny and ubiquity of intra and inter department discussion and deliberation (see Neal, 2019). 9 On the contrary, parliamentary discussion and intra-departmental deliberation are key state practices through which the UK state in a representational democracy persists. Indeed, the UK state is assembled through ongoing acts of proposition, justification, contestation, and deliberation in relation to future presents. What makes them 'scenes of emergency' is that even as the state is being assembled the cluster of promises that surround the state are at risk of 'falling or coming apart'.
I draw on two main archival sources to describe the 1920, 1964 and 1973 scenes, based on a year of archival work on UK emergency legislation and planning. 10 First, Hansard records were used to analyse parliamentary debates in 1920, 1964, and 2004 as emergency planning legislation was introduced or amended, as well as each of the twelve times an emergency was declared in the twentieth century under the (amended in 1964) 1920 Emergency Powers Act.
All parliamentary speech surrounding the introduction of or changes to legislation over the 20th century was gathered. Analysis focused on how legislation and the use of legislation was justified by governments and the range of responses to legislation by other parties and parliamentarians. Second, material was gathered from the UK National Archives on how central government organised in relation to and planned for emergencies over the same period of time, with particular emphasis on material related to the government departments which have held central responsibility for emergencies, principally the Home Office (including the Emergency Planning Unit and Home Office Emergencies Organisation) and the Cabinet Office (including the Civil Contingencies Unit from 1972 onwards). 11 The National Archive material was used to track changing institutional arrangements for governing emergencies, justifications for changes, and deliberations around possible changes. The first scene returns us to the introduction of emergency powers in the UK.

Section 2: Scenes of emergency
The 1920 Emergency Powers Act is typical of the advent of Emergency Powers across liberal democratic regimes in Europe and the United States in the interwar years. Emerging from the delegations of emergency economic authority to the executive during World War 1 and building on the 19th century use of emergency authority by elites to end labour and socialist unrest that supposedly threatened civil insurrection (Mitchell, 2011;Scheuerman, 2000), the Powers are explicitly orientated to trade unions and the activity and situation of strikes (and thus separate from military conflict or armed rebellion). Unlike the 1914 Defence of the Realm Act, the 1920 Act granted powers over economy and life during emergencies not limited to war.
The state of emergency is a technique that, at least as presented, aims to resolve the problem of what happens when the state assumes a duty to maintain life but is faced by situations that threaten serious disruption but are also unique. Put differently, the state of emergency as legal-political technique is one way of recognising and incorporating the excess of events into the purview of the state. Declarations of emergency by the sovereign were authorised by a particular kind of event: "Her Majesty may by proclamation declare that a state of emergency exists if at any time it appears to Her that any action has been taken or is immediately threatened by any persons or body of persons of such a nature and on so extensive a scale as to be calculated, by interfering with the supply and distribution of food, water, fuel, or light, or with the means of locomotion, to deprive the community, or any substantial portion of the community, of the essentials of life." (Emergency Powers Act, 1920) As with all state of exception legislation, the Act is simultaneously a recognition of law's (and the state's) limit and an attempt to encompass the exception that threatens harm or damage within law (and the state) (Agamben, 2003). The effect of the legislation is to reproduce the state-effect Ophir (2007) associates with emergency: the state as a total entity with the capacity to act. However, parliament is also a scene for the accusation that the powers are excessive and the disassembly of the promise of the state as neutral, disinterested, party. Introduced in the midst of industrial tensions and implicitly referring to trade unions ('any persons or body of persons") (on which see Keith and Hennessy, 1983), the credibility of the promise of the neutral state is challenged through a set of antiemergency statements and claims by opposition politicians.

Scene 1: 22 October 1920
On first reading of the Bill, H.H. Asquith -then leader of the opposition Liberal Partyargued that it was unnecessary in the absence of an emergency: "The Government have in their possession all the powers that are needed for any risks or dangers which at present confront us. That being the case, surely the elementary dictates of political prudence, not to say statesmanship, indicate that it would be better, for the time being, to rely upon those powers, and not at a most inopportune time, to pass permanent legislation of the kind contemplated by this Bill. It is not a matter of urgency at all. There is no emergency which calls at this moment for legislative action." 12 The anti-emergency claim by Asquith was threefold: that there was no emergency, that the state already possessed all the powers it may need, and that it was an "inopportune time" given the turbulent post war public mood. Consequently, for the opposition, the Bill was a "provocation" that was "calculated to arouse the suspicions of the working class" in midst of the "crisis in the mining industry." 13 One MP accuses the Government of not taking into account the "psychology of the situation" and warns that "there will be created an atmosphere outside this House which will be distinctly harmful to the progress of friendly negotiations." 14 Another MP cautions in similar terms of the consequences for the present dispute of a "mood of suspicion" amongst "men who do believe this is an attack on trade unionism." 15 Emergency powers were not only dangerous for the extra-legal effects that the passage of a Bill might have had in the midst of a tense atmosphere, though. In later discussions in the House of Commons, other MPs claimed they were also dangerous for how they fundamentally reshape the balance of forces within society. Some MPs accused the state of other, hidden motives: notably a class politics that aimed to crush the power of the labour movement. It is worth quoting at length one objection, as it folds into the scene of emergency past events by charging the Government with a "cunningly veiled attack" on the trade unions. "In 1911 there was not only a railway strike, but practically half the dockers were on strike. The whole port of Liverpool was closed. There were ironclads in the river. There were tremendous riots and disorder. There was no suggestion that this kind of legislation was necessary. In 1912 the whole of the mines of the country were stopped for between six and seven weeks, and it was never suggested by even the most reactionary that legislation of this kind was necessary. Let anybody go into these particular conditions, and he is bound to see that this is not an open but a cunningly veiled attack upon the powers now-possessed by trade unions, and that the Title of the Bill should rather be "a Bill to revoke the whole of the Statutes since 1824 conferring legal powers upon trade unions." 16 Something of the then present atmosphere of heightened tensions, as well as affectively imbued memories of past conflicts, became part of the scene in parliament. By invoking a series of past situations that did not necessitate the introduction or use of special powers, the Emergency Powers Bill is enacted as an instrument of class politics and the state becomes a cunning, duplicitous actor.
The introduction of legislation becomes a scene in which the promise of the disinterested state is disassembled as other claims are made: the state as a classed, interested, actor with a threatening capacity to reorder society. There is, though, a second version of the state that emerges in the wake of the disassembly of a promise: an anxiety about the becoming authoritarian of the democratic state that is enacted through concerns about what the state is enabled to do by the Bill. Here we see an intensification of the concern with what Foucault (2008: 322) calls "the irrationality peculiar to excessive government" that he argues animates liberalism. Emergency Powers becomes the point, to paraphrase Foucault (2008: 320), where the state governs "too much" and the liberal democratic state becomes the monster that must be kept in check.We gain a sense that emergency legislation is itself a threat when the Bill moves to committee stage. Unusually, committee stage happened four days after the first reading of the Bill. Committee stage starts with an objection to the first words of Clause 1 of the Bill -"at any time" -, in part as a protest at what opposition MPs consider to be the compression of the time for deliberation. Mr Walter Smith argues that there is a mismatch between the seriousness of powers and the time given for "close examination and scrutiny": "There has been no opportunity whatever for giving that close examination and scrutiny to this Measure which its importance warrants, and yet the Government propose to make it a permanent part of their legislation to be used whenever the Government of the day deems a state of emergency has arisen which justifies its provisions becoming operative. I venture to suggest that that is a very dangerous procedure. We may have a sudden election during a period of passing excitement, and a Government may be returned to power which might use the provisions of a Bill such as this very largely for political purposes against their opponents." 17 For Smith, emergency powers require the time for practices of deliberation: "discussion of every detail and for the examination of every line and word." 18 What animates dissent is a concern by opposition MPs with the potential misuse of powers. Expressed by MPs as the Bill becomes an object of explicit concern is the threat of a different type of exceptional situation than that invoked by proponents of the Bill. The possible exception is the misuse of the Powers by a tyrannical, illiberal state.

Scene 2: 20 February 1964
The introduction of emergency powers in 1920 was an occasion for anti-emergency statements that disassembled a promise: the neutral state acting through necessity. Through those statements, the state was enacted as an actual threat -to the existing powers of the trade unions -and a potential threat -to the 'liberty of the subject' and democracy itself. Anxiety, mistrust and anger enveloped the introduction of emergency powers, perhaps resonating with then moods of a 'present situation' of industrial tensions and class based conflict (Keith and Hennessy, 1983).These concerns enacted the liberal suspicion about the state governing too much that Foucault (2008: 319) highlights, but assemble them with two distinct state imaginaries: the all-powerful despotic state and the interested, classed, state. For proponents in government, by contrast, emergency powers are not merely necessary but also prudent given the possibility of exceptional situations, specifically strikes, leading to revolution (see Jeffery and Hennessy (1983: 9) on the "underlying fear of revolution" within the state during that period). Emergency becomes an occasion where the question of the excessiveness of the state, its "necessity and usefulness", is played out and subject to deliberation (Foucault, 2008: 319).
The legislation was the basis for the twelve times a state of emergency was declared in the UK, all in relation to industrial disputes. In 1964, a small but significant change was made to the 1920 Emergency Powers Act. As well as making permanent the Defence (Armed Forces) Regulations of 1939 that allowed for the use the armed forces in temporary employment in agriculture or other work considered to be of "national importance", the then conservative government proposed a change should be made to the definition of emergency in the 1920 Act. As we have seen, the concern in the 1920 Act was with a specific causal agent -'persons or body of persons' -and, although not stated as such in the Act, with one type of situationstrikes and industrial disputes. The 1964 Act amended this definition by replacing the phrase 'persons or body of persons' with the word 'events', so it reads: "Her Majesty may by proclamation declare that a state of emergency exists if at any time it appears to Her that any action there have occurred, or are about to occur, events of such a nature and on so extensive a scale as to be calculated, by interfering with the supply and distribution of food, water, fuel or light, or with the means of locomotion, to deprive the community, or any substantial portion of the community, of the essentials of life" (Emergency Powers Act (amended) 1964, emphasis added) The amendment widened the scope of emergency powers to the open-ended 'events', irrespective of the type of event or the cause for the event.
In the second reading of the Bill on 20 February 1964 the Secretary of State for the Home Department, Henry Brooke, justified the expansion in the following terms, invoking a different event before which the providential state may be powerless: weather: "It was during the prolonged bad weather of last winter that I first started to think about the limitations on our powers to take emergency action in a national emergency. Clause 1 amends the Emergency Powers Act, 1920. When that Act was passed the country was in the throes of readjusting itself to peace-time after the First World War. As many of us remember, it was a time of industrial unrest. That was the problem in view, and so the Act was directed at emergencies resulting from labour troubles. It was confined to them.
Its purpose was, and is, to enable the Government of the day to take measures to maintain supplies which are essential to the life of the community. But there is no obvious reason why the Government should be able to take these measures only when supply difficulties are threatened as a result of labour troubles. The threat may come from other causes; from what the insurance policies refer to as an act of God, or indeed from developments in other countries beyond our control.
If since 1920, all Governments have been granted special powers to secure essential supplies for the public in times of industrial emergency, it seems common sense to make similar provision for dealing with any type of emergency in the future which similarly threatens the essentials of life." 19 Describing the Bill as a "wise exercise in foresight" and an "insurance policy," 20 Brooke justified the extension to 'events' as necessary because of the spectre of possible future 'exceptional' events that threatened the always fragile providential state's capacity to respond. In this scene, the state confronted a different form of excess -events as such, rather than the actions of a trade union. Although a range of 'new' emergencies were invoked by proponents of the amendment, including mechanical breakdown of nuclear power stations and stoppage of essential supplies from abroad, it was 'weather' that stood in for events that threaten the state: "I would like the House to know that after last winter the Government undertook a comprehensive review of all the arrangements everywhere for coping with bad weather. As a result of that review, many steps have been taken-steps to improve equipment, to acquire additional emergency equipment, to coordinate plans at all levels, and to encourage preparations for bad winter conditions. But there is a limit to the money which it is sensible to authorise on emergency provisions which may never be used, and the best plans may be defeated in an emergency on a scale which out-soars anything ever previously contemplated.
The problem is to strike the right balance. With the action already taken and planned, we think that we have gone as far as we practically can by administrative means and forethought to take precautions against severe and prolonged snow and frost. But one must recognise that circumstances may arise so utterly exceptional that all precautions might be inadequate. That is why we think that emergency powers ought to be available as a last resort, not confined to industrial disputes only." 21 The proposed change in legislation was legitimised around the invocation of a series of 'extreme' or 'abnormal' possibilities that exceeded government's capacity to plan, or what was referred to by Brooke as 'administrative means and forethought'. He invoked a twin problem for the providential emergency state: the possible future event that is 'exceptional' and 'unforeseen'. Whilst not quite equivalent, the two ways of registering occurrence and impact mean that rationalisations for the Bill are organised around a slight tweak to the version of emergency that that was embedded in the 1920 Emergency Powers Act -the event as a 'real exception'. As well as rarity and severity of impact, the shift to the generic 'events' intensifies the problem of the unforeseen. Unlike strikes, with their defined actors and precipitating conditions ('tensions'), 'events' may happen without warning. 'Acts of God' are so named precisely because of the problem of ascribing causality to them. They appear to come from nowhere. For this reason, Brooke described the Bill as "an insurance policy against contingencies-remote contingencies, perhaps, but real ones nevertheless." 22 Parliament became, then, a scene of emergency: the excess of events (in the form of the 'exceptional' and 'unforeseen') was present as a problem for the state. What was enacted was the promise of the providential, prudent, state which may need extra powers. As in 1920, the introduction of the Bill was also a scene of contestation where the promise of the providential state was disassembled. The tyrannical state was not expressively present, but was there in the background to claims that the state already possessed the resources needed to govern emergencies. One MP challenges the home secretary to either "think of one example when it would have been desirable to make these regulations" at some point since 1920 or give a future example, even a "barely imaginable" example, of a "possible situation" that would require regulations to be made under the Act. 23 As he did so, he offered a different account of past events that render them undramatic -matters of "inconvenience" or "discomfort": "I would test the position by asking the right hon. Gentleman to think of any situation relating to the happening of a natural cause, like the cold weather of last year, in the 44 years that have elapsed since the 1920 Act was passed when it would have been desirable or necessary to make regulations under this Bill had it been law. Consider the cold weather last year, which occasioned a great deal of discomfort and inconvenience. Is the right hon. Gentleman saying that there would have been regulations made to deal with that situation had the Bill been law? If so, what sort of regulations would have been made?" 24 In comparison with the contestation which greeted the 1920 Act, the criticism of the amendment is focused on the past or future presence of events. As Soskice remarks, he does not have "any violent objection in principle to the Bill." 25 Indeed, he accepts the need for the state to be "be ready for emergencies", but considers the case for the amendment to rest on "wholly hypothetical contingencies which are unlikely to be translated into reality according to all the experience we have had." 26 For other MPs, the legislation was necessary because unforeseen events might happen: excess reoccurs through emergency statements and the state is assembled as potentially unprepared as well as possessing a responsibility to be prepared. An MP supportive of the measure recalled an event as he supported the change in wording: "I can look back upon disasters which were quite unexpected. I remember with vivid clarity the 1953 floods along the East Coast, which grievously affected my constituency, among others. Nobody imagined that there was any danger of flooding upon that scale at the time, and I am sure that if my right hon. Friend had introduced this legislation prior to those floods people would have said, "What a ridiculous thing to talk about. Such a thing could not happen." But it did.
Furthermore, in the complex industrial life in which we live today products are being manufactured which have never been manufactured here before, and new scientific techniques and devices are being adopted. It is, therefore, right to visualise the possibility of a quite unforeseen disaster occurring." 27 The Act became, then, a prudent means of responding to the problem of the 'unexpected' or 'unforeseen' and the limits of the state's imagination. The Bill was one part of a wider shift involving the expansion of the problem of the 'unforeseen' to any event alongside an expanded sense of not only what threatens to disrupt, but also the contingencies that a 'prudent' or 'responsible' government should take 'reasonable steps' to prepare for. By 1964, the existence of emergency legislation has become taken for granted within parliament and contestation is a matter of disagreement about whether or not the state is sufficiently prepared: the "too little" of the state, which Foucault shows is the flipside of the concern with "too much" (Foucault, 2008: 322).

Scene 3: 1973
What is played out in the mid-1960s is not simply, then, the liberal concern with the excessiveness of the state. As in 1920, that concern meets an anxiety about the excessiveness of events and the credibility of the promise of response. The assembly and disassembly of the promise of the prepared state when faced with events reappears within the UK state at various times after 1964, although the Emergency Powers Act is not replaced until 2003-2004 (on which see Anderson and Adey, 2012). What results are various scenes of emergency in which proposals to expand the scope of emergency powers are deliberated against a background of the 'changing circumstances' of events. One significant occasion is in 1973, the year of the last declaration of a state of emergency that I began the paper with. In the midst of ongoing industrial disrupts, global economic crises, and a changing economy, the Industrial Relations Policy Committee ask the newly formed Civil Contingencies Unit (CCU) to review the requirements for emergency powers. What surfaces in this scene -a scene constituted through letters, notes and a report -is a concern with the volatility of emergency powers. This is the state that risks governing "too much" (Foucault, 2008) and therefore cannot control the effects of its own actions. Here we find another iteration of the concern with the excessiveness of the state -but it is channelled into speculation on the unpredictable effects of declarations and regulations.
Underpinning many of the submissions to the review was a need to avoid unduly "restrictive" definitions of the circumstances on which an emergency might be proclaimed. However, there was no unanimity within the state about the changing threat. The Department of Employment, for example, argued that "An impressionistic view of the risks of industrial action ahead is not, I suggest, a good starting point for considering whether amendments of the 1920 Act is necessary." 28 Nor was there agreement about what should be protected by the providential state. One typical concern was whether the list of circumstances upon which a proclamation can be issued should be widened to include interference with communications or health services. In a 'note' from the Cabinet Office described by its author as "a first shot at defining this rather nebulous problem [of the requirements of emergency powers] and how it might be tackled" the complexities of expanding to 'telecommunications' were discussed: "How widely or narrowly should they be defined? Are "communications" just "telecommunications" or, bearing in mind the importance which the Government attaches to public opinion, and the changing circumstances of the communications media, is there any case for taking emergency powers in this field, too? Are there other categories which should be included? Is it now desirable to take account of the effect of action against key computer installations? Are these sufficiently covered under the existing categories of services or should some additional provision be made?" 29 The deliberation was around, then, the sufficiency of existing powers faced not only with changing events but with attempting to govern a changing society in which it was no longer clear what exactly counted as the 'essentials of life'. For example, expanding powers to telecommunications is rejected by the CCU in part based on a letter from The Ministry of Posts and Telecommunications, which reported no advantage in amendment because of the range of alternative measures enabled by other pieces of existing sector specific legislation. 30 The promise is very similar to in 1964: the prepared, reasonable, state in a changing world of events and conditions that threaten the capacity to prepare. As well as this doubt about whether the state will protect life, what is disassembled in this scene is faith in the utility of emergency powers for the state. Partly, this is because departments place faith in existing, ordinary powers. But, it is also because changes in the organisation of economy reduce the effectiveness of emergency powers that depend, in different ways, on some degree of compulsion from a centre. For example, the Department of the Environment wrote "It is difficult to envisage the way the application of emergency powers could assist the health service, or communications or computer installations, because generally all these depend on expert staff who cannot be easily withdrawn from other sources." 31 Likewise, the Department of the Environment stressed the interrelated and complex nature of emergencies. Their letter is worth quoting at length: "The Act is essentially defensive in character, presupposing that the Government can provide some alternative service. In the event it is clear that this is no longer the case in many instances, partly because services which can be hit by strikers are now more complex and partly because they are more interrelated so that a comparatively limited action can repercuss throughout a complete activity . . . In the broader sense it seems unlikely that the current Act could be modified in a way to provide an effective instrument against a deliberate and sustained attempt to cause widespread disruption to the life of the community or to the economy. Hence the Departmental view that it is best to leave well alone and not to seek extension of powers under the Act." 32 Other departments similarly emphasised the limited contribution of emergency powers to governing the type of disruption that are strikes, and highlighted other ways of governing through the achievement of consent. The Welsh office wrote: "Post-war history does suggest that emergency powers are only on particular and a small number of occasions necessary to contribute to the achievement of the Government's objectives, and the contribution is in a substantial part psychological. In all strikes, there has been a high level of 'makeshift' performances by other members of the industry or service concerned, by local authorities and by others. Indeed, the effects of most strikes seem to have been eased largely through a miscellany of makeshift arrangements and not by the specific use of emergency powers." 33 There was also a trace of something else: that the occasion of the use of emergency powers was a scene of intensified vulnerability for the state. It is when the capacity to govern is directly placed in question and when other sources of authority emerge. Following a warning by the Department of Employment that prohibiting picketing at essential installations would lead to "deliberately contrived confrontations with the Law" in which the "resulting situations would be ugly," 34 the MoD wrote: "To the extent that these are useful weapons of crisis management we must be careful not to blunt their efficiency by attempting to resort to them in circumstances in which, as Douglas Smith [Department for Employment] has pointed out, the probable consequences would be a confrontation in which the will of the Government could not easily be enforced" 35 The risk was not only that emergency powers are ineffective as pragmatic means of governing an increasingly interconnected economy, but that they may both intensify the event they are designed to end and, in doing so, reveal the fragility of the state.

Concluding comments: Disassembling the emergency state
We reach an impasse: emergency powers have been extended beyond strikes and are presented as necessary because of the potential weakness of the state faced with events but in 1973 doubts surround their efficacy and effects. What is questioned, in the year of the declaration that I began the paper with, is their utility as a means of governing events in an increasingly complex economy. Alongside this practical objection to their use, there is also a background sense within the state that the use of emergency powers has itself become a dangerous occasion of vulnerability. Partly, this is because they may prove to be inadequate when faced with events. More than this, though, there is a sense that emergency powers are unpredictable. Their use might lead to a loss of morale in the army, unwanted confrontations, and so on. What could in 1964 be likened by a government minister to 'insurance' with its connotations of prudence has become, by 1973 and in the midst of crisis, an unstable but still used power, unpredictable in its effects, always threatening to escalate events.
The paper has focused on how one part of the emergency state -the promise of neutral and effective response -is disassembled and reassembled. What we see across the scenes is concern with two types of excess -of the state and of events. In part, this enacts the liberal concern with whether the state governs "too much or too little", or what Foucault (2008: 322) calls the "irrationality peculiar to excessive government". Concern with the excessive emergency state is enacted through anticipatory claims about the authoritarian state, in particular, but also the ineffective state. However, the relation with emergencies is not reducible to the liberal suspicion of government -what is not critiqued is the necessity or possibility of responding to emergencies per se (to paraphrase Foucault, 2008: 319).To go back to Ophir (2007), the expectation that the providential state should have control of events is re-enacted in scenes in which the excess of events is made present and actionable. The state suspects it might not be governing enough, that it might not be able to meet events. Here, it is not only the state that is a "supplement" (Foucault, 2008: 319) to society, but emergencies that function as a dangerous "supplement" around which the promise of neutral and effective response exists on a threshold, being dis/re-assembled.
The paper has two wider implications in addition to the need for more detailed work on forms and practices of contestation around state of emergency legislation across different state formations, and through different state practices. First, the paper has stayed with 'scenes of emergency' in which the state relates to the excess that the term 'emergency' invokes and gestures towards. It is in and through 'scenes of emergency' that the state renders events governable whilst, at the same time, the state's capacity to act and control events is dis/reassembled. The scenes described in this paper -composed of future orientated practices of discussion and contestation surrounding emergency powers legislation -are but one type of scene. This opens up a research agenda on the different 'scenes of emergency' through which emergency states are (dis/re)assembled and the problem that the events or conditions that are governed as emergencies pose for the state. A scene of emergency is not every time an emergency is materially, imaginatively or performatively present. What defines a 'scene' is that some kind of excess is present that reveals or exposes the limits of the state's capacity to protect life through anticipation and response. How, then, are events made present through state practices of legislating, planning, exercising, responding, and so on? In what ways is that which is outside of the state's capacities reincorporated as emergency assemblages form, deform and reform? Second, in relation to emergent work on assembling the state, the paper has stayed with occasions in which some kind of state effect -here the promise of response -is being disassembled and reassembled. Recognising the intimacy of processes of dis and reassembly, pushes assemblage related work on the state to focus on occasions when 'state effects' are at a threshold. Where the term threshold names a spacetime of (more or less abrupt, more or less intense) transition in which new state effects may be beginning, and existing state effects may be ending. Thresholds are, then, occasions of indeterminacy in which what the state is and does is placed in question, whether momentarily or for the duration of an ongoing crisis. They will vary across different states and state practices and functions. In the scenes in this paper, what is dis and reassembled is the promise of a specific type of prepared state that can control events and acts with neutral benevolence to protect life. In different scenes of emergency, other elements that compose emergency assemblages may be dis and re-assembled: the plan or protocol that requires revision, equipment that does not work, as well as the immaterial promises and threats, ideals and fantasies, which animate and enable the state.