In law the army was the king’s, or queen’s. Politically it had for long been controlled by ministers whose responsibilities had come to be owed to Parliament, and Parliament had come to control its finance, but the extent of those controls, the remaining role of the monarch, and the autonomy of the army high command, together formed a persistent if minor theme in Victorian and Edwardian constitutionalism.1The army’s prerogative status was used rhetorically from time to time by army high command as a justification for autonomy from politicians, and right to the end of our period it was possible for army officers, in times of stress, to seek solace that the unpalatable orders they anticipated receiving had the king’s personal approval and were not merely the implementation of government policy.2 Like the Church, the army had its own internal system of law. The common law judges accorded it a high degree of autonomy, in part as a reflection and reiteration of its prerogative status, but predominantly because they thought external adjudication would prejudice good discipline. Though this attitude was tested in the common law courts from time to time, it remained essentially unchanged throughout our period.
1. The Contours of Civilian Control
Through into the mid-century, command, control, and administration of the army rested on an eighteenth-century Whiggish construct of checks and balances that from the late 1820s politicians of both parties would like to have replaced with a more centralized, rational, and cheaper system.3Broadly, command rested (p.402) with the military through the general commanding-in-chief, usually known as the Commander-in-Chief. For most of the nineteenth century his was in practice a joint appointment by the sovereign and the Prime Minister, for a period or for life, hence apolitical in the sense of not changing with each government. It was a prerogative post, a delegation of the sovereign’s own powers as commander of his forces, a designation which, in the eyes of the Commander-in-Chief at least, could be taken to suggest that he owed his duties directly to the sovereign, not to be mediated through politicians.4 With command went appointment, dismissal, and disciplining of officers, and it was fear of what might happen if those powers fell to politicians that explains why through until at least the 1860s proposals for ‘direct subordination of the army to a civilian minister invited violent reaction from the Crown, military figures and political opinion’, as John Sweetman put it.5
Command, however, had its limits. Disposition of the troops within the realm was controlled by the Secretary at War, later the Secretary of State for War, not the Commander-in-Chief.6 Civilians also initiated the use of troops to quell civil unrest, in that their deployment depended first upon the magistrates to make a request, and then upon the Home Office (later the War Office, later still the Home Office again) to agree either in the specific instance or by general instructions.7 How far civilian command extended to the disposition of troops once called out was a delicate question more suited to negotiation than to regulation. Statements of the official position ranged from Home Secretary Gathorne Hardy’s in 1869, that disposition was for the officer in command, to Adjutant-General Sir Redvers Buller’s in 1893 that if the magistrate was firm the commanding officer must obey, through to an inter-departmental committee’s the following year that it was for the magistrate to formally commit the military to action but that thereafter the commanding officer’s discretion was absolute.8 Queen’s Regulations (and the inter-departmental committee) stressed that detachments should ensure that a magistrate accompany them, but as was only too evident both to the military and to the committee of inquiry that Buller was addressing, the commanding officer (p.403) could not insist. It was in any event a regulation only, and could be waived at will. In the midst of the very serious labour disputes of 1911 Home Secretary Churchill went one stage further by suspending the requirement for an initiating requisition from civil authorities: if the need existed, troops could call themselves out.9
Command was limited also by law, in the dual sense that an order had to be lawful before disobedience to it constituted an offence against army discipline—not a matter that was tested often in the nineteenth century—and, more significantly, that it was an almost axiomatic feature of military subordination to civil power that the civil authorities might insist on trying soldiers for offences against the general law irrespective of whether they were also offences against military law.10 In the nineteenth century this principle was no longer disputed, but it did cause the military some concern when their duty to assist the suppression of civil disturbance exposed them to possible prosecution for the use of excessive force.
One response was to try to shift the moral responsibility to the magistracy: in 1835 the eighteenth-century guideline, that soldiers should fire only under orders from magistrates or in conditions of necessity, was tightened by amending King’s Regulations to make a magistrate’s order essential.11 In the much calmer times of 1868 it was changed back again, but in 1893, in the aftermath of fatal shootings during the Featherstone colliery strike, Sir Redvers Buller testified that he could not recall an occasion of ‘necessity’.12 Nonetheless, necessity it presumably was that led the troops freed by Churchill to the shooting of two strikers at Llanelli in 1911. Another response in the 1890s through into the first decade of the new century was from the military for a reduction in personal responsibility when under civil command.13 It was firmly rebuffed by Secretary of State Haldane, who undertook only to align the King’s Regulations more precisely with the austere principles of the common law.14The symbol of submission to law was important; (p.404) in practice it could always be mitigated by prosecutor’s discretion and the ability of the Attorney-General to enter a nolle prosequi, though this has not been systematically researched.
In the Whig model, command was balanced by administration. In 1820 administration was the function of civilians responsible to Parliament, principally through the Secretary of State for War and Colonies, the Secretary at War, and the Master-General of Ordnance, who was a soldier but a political appointment and member of the cabinet. The administrative side included supply, first of money, which was the business of the Commissariat (accountable to the Treasury, and responsible also for provisions and transport in the field), secondly of equipment, which was the responsibility of the Master-General of Ordnance and the Board of Ordnance, which also controlled the artillery, the engineers, the survey, and, from 1822, the barracks. Political control itself was thus fragmented, indeed all three Secretaries of State were involved in one way or another, plus the Treasury. Demarcation lines were vague, the precise duties of figures such as the Secretary at War and the Judge-Advocate General (another political appointment) difficult to state, and lines of responsibility were sometimes notional. But, as Hew Strachan writes, ‘to unify the control of the army would increase its potential power in the state, whether as an independent pressure group applied to Parliament or as a further subject of Parliamentary sovereignty’, so the system of dual control survived for fear of the alternatives and proposals for administrative rationalization were always resisted.15
Parliament legislated for the army annually, a regular opportunity for intervention in such internal matters as the nature of regimental courts martial and the use of flogging as a punishment. There was temptation also to use the occasion to pursue wider political goals, most spectacularly when the Conservative opposition in 1913 was known to be contemplating, but did not pursue, an amendment to prevent troops being used to force home rule on Ulster.16 For the most part, however, the main thrust of the annual Mutiny Act was only to make lawful what statute and common law otherwise prohibited. It permitted the sovereign to maintain an army of specified size at home in peacetime, it gave power to enlist soldiers, to billet them, to discipline them outside the procedures of the common law, and to empress transport for their movement within the realm.17 It did not say how the army was to be composed, commanded, controlled, directed, (p.405) or equipped.18 During Victoria’s reign that territory was contested between Her Majesty’s military servants, sometimes buttressing their claims to expertise with assertions of direct allegiance to the Crown—their interpretation of what it meant to fill a prerogative post—and Her Majesty’s ministers of state, with their responsibility to Parliament. The Queen herself was a partisan for army autonomy and an opponent of Commons’ intrusion. The Commons, of course, controlled the money. Its powers had been greatly strengthened by Burke’s Act of 1783, which had brought regimental allowances into the annual estimates and increased the Commons’ power of audit, giving ample opportunity for debate.19
Largely on the back of concerns about financial control, senior officials collected a medley of statutory powers concerning money and army property, the Secretary at War and the Ordnance Board especially.20 Administrative reform remained politically difficult even when the shocks of the Crimean war broke much of the resistance to it, and when it came it appeared in the statute book as a minimalist transfer of powers, a matter of mere housekeeping. Neither the Ordnance Board Transfer Act 1855 nor Cardwell’s War Office Act 1870 so much as mentioned the Commander-in-Chief, since his directive and command functions flowed from the prerogative rather than statute, though demarcation of his role and his relation to political authority were matters of supreme political concern. They were dealt with by prerogative instruments of one sort or another, studiously vague and not meant for public discussion.21 In 1854/5 this resulted from the Queen’s intervention. In 1854 a separate Secretary of State for War was created, who absorbed the duties of the Secretary at War and assumed responsibility for the Commissariat and the Ordnance department. He also took over control of the militia and other auxiliaries from the Home Secretary, a shift that had previously been resisted on constitutional grounds—that the function of the militia ought to be kept separate from that of the army, and its basis remain local rather than national.
This was a considerable centralization. But through pressure from the Queen the Commander-in-Chief’s position remained separate, and he gained command of the military side of the Ordnance Board’s activities—the Royal Artillery, the Royal Engineers, plus responsibility for various formerly semi-autonomous (p.406) agencies such as the medical department previously attached to the Board. There was thus a rationalization of both parts of the dual command, but thanks to the Queen there was no unifying board, nor did the Secretary of State have access to military advice other than through the filter of the Commander-in-Chief. The vagueness of the demarcation line with the Secretary of State was to work to the advantage of the Duke of Cambridge, the Queen’s cousin, who was appointed Commander-in-Chief in 1856. The force of his personality and his ability to build on his relation to the Queen to reinforce the prerogative basis of the post enabled him to keep matters of command at arm’s length from the Secretary of State for the next 40 years.22 Formally his relation to the Secretary of State for War was settled in 1870, when as part of Cardwell’s centralization of all administration under the War Office he accepted that, constitutionally, he advised the Secretary of State and not the Queen, and was responsible to him, though, again, there was no sign of this in the statute. Politics continued as normal, however; reconciliation of military expertise with party government could not be achieved merely by Acts of Parliament or orders-in-council, and in practice the Duke continued to claim a special constitutional relationship with the Queen that freed him from the normal constraints of a subordinate official.
When the Duke was finally persuaded to retire in 1895, his successor was moved into an amorphous supervisory role, the heads of his military sub-departments made directly responsible to the Secretary of State, and formal roles delineated for a War Office Consultative Council and an Army Board to propose promotions, senior appointments, and expenditure, all under the Secretary of State’s suzerainty.23 All this was done, as Gwyn Harries-Jenkins writes, with one stroke of the pen—by order-in-council, with an explanatory memorandum to follow.24 It was, as he writes, just the first, if much belated organizational step in transforming the traditional military into the new military of an industrialized society. Legally all were simply rearrangements of Her Majesty’s servants like many another.
(p.407) 2. Judicial Process
Whoever wielded His Majesty’s powers over his military servants was safe from judicial scrutiny. That is not to say that courts were not asked to adjudicate. Dismissal from the service might itself bring litigation, or it might be that the manner or cause left such a stain on an ex-officer’s character that he sought redress as a matter of honour, or his concern might be for the consequences, such as loss of a pension, or its amount, or the recovery of accrued pay. Throughout our period, however, the courts were consistent: in Lord Esher’s succinct summary, the Crown’s military engagements were voluntary on its side (though binding, of course, on the other).25
It made no difference that the power concerned might plausibly be seen as resting on statute rather than the prerogative, or that deprivation of a pension or a retirement allowance could be seen as loss of property.26 Provisions in the Mutiny Act that might have been interpreted as giving an officer a right to pay and allowances once accrued were interpreted instead merely as imposing sanctions upon paymasters.27 Because the Crown could part with its military servants merely because it wished to, courts held themselves unable to inquire into the grounds of dismissal or the procedures used.28 The courts treated the Crown’s civil engagements as voluntary too, but in military cases they tended to use military justifications and, as will be seen below, their reticence led also to only the most rudimentary scrutiny of courts martial proceedings. In essence, the courts held that if the behaviour complained of affected only a soldier’s military status, or the incidents that flowed from that status, redress was to be found solely through military channels, not at common law.29 ‘It would be a very lame application of the doctrine that military men are to dispose of military questions, to apply it only if it is proved by the result of a trial that a military question arises’, said Willes J. in 1866.30
(p.408) Litigants in dismissal cases tried two overlapping ways around that doctrine. First they alleged malice, which in common law theory was a claim that the defendant was acting from personal motives and not within the role allocated to him by law; hence that the harm he did to the plaintiff was not done as a soldier, hence he could not claim the benefit of the exclusive military code. This was the claim in the important eighteenth-century case, Sutton v. Johnstone, where the defendant was alleged to have maliciously exposed the plaintiff to a court martial.31 The rule of law implications invoked by the plaintiff had attracted the court of Exchequer, only to be rejected by Lord Mansfield and Lord Loughborough, who feared the potential for making common law courts into general courts of appeal from courts martial. Officers, who alone stood between the subject and a licentious soldiery, were to be protected from harassment in the courts. Nonetheless, there is a case reported at nisi prius in 1843 where Cresswell J. nonsuited a plaintiff for failure to plead malice, and another going a stage further in 1865, when Erle CJCP let the question of malice go to a jury, so the principle of Sutton v. Johnstonemay not have been deeply embedded.32
The second route to attracting common law attention and the hoped-for sympathies of a jury was to seek vindication for loss of reputation through an action for libel or slander. There was a range of targets, from the bringing of charges, through to things said at courts martial, or at courts of inquiry, or in reports, correspondence or public announcements. Common law privilege attached to courts martial and, so it was held, to the other situations as well, but it would be lost if the defendant had been malicious—and again that issue exposed the plaintiff’s senior officers to scrutiny by a jury.33 In 1806 Mansfield CJCP had echoed Sutton v. Johnstone in rejecting the applicability of common law altogether, hence ruling that malice could not even be put to the jury, but in 1820 Dallas CJCP was more equivocal.34 Though in the circumstances of the case he held the communication under challenge to have been absolutely confidential, he left it unclear whether he too was rejecting the applicability of common law or finding a defence within it. In 1859, however, Lord Campbell CJ allowed military proceedings short of a court martial to be examined by a jury for malice, and in 1863 Cockburn CJ did the same, later justifying his view as necessary for the prevention of abuse, declaring his confidence in juries to do justice.35
(p.409) It could be argued, then, that there was a shift under way from regarding the affairs of military men as beyond the remit of the common law courts. But Cockburn CJ failed to convince his own court, which endorsed instead a strong version of Lords Mansfield and Loughborough’s analysis: relations between military men were wholly governed by an exclusive code of military law founded upon their presumed consent and the need to preserve discipline, a code providing its own means of redress through complaint to commanding officers.36 In 1875 the Lords declined to decide between these analyses, but from then through to the end of our period the ‘code’ analysis of military law was orthodoxy up to and including Court of Appeal level.37 It is possible that judges had a more interventionist attitude in the mid-century—the materials are too scanty to be sure—but if so it was only an interlude.38
Maintenance of Military Discipline
This was the legal background against which the maintenance of military discipline should be seen. Responsibility lay with the Commander-in-Chief, aided and regulated by the annual Mutiny Act and the Articles of War the Act authorized. Political responsibility for those codes, however, was the Secretary at War’s, later the Secretary of State’s, though in the 1860s (if not earlier) much of the work was done by the Judge-Advocate General.39 Because the Act was annual there was scope for parliamentary pressure, as with attempts to reduce flogging, for example, and that could also impact on the articles, though formally they did not come Parliament’s way.40 At all times control of the formal disciplinary(p.410) codes thus rested with civilians—though presumably the War Office worked in co-operation with the military through the Adjutant-General, the head of the Commander-in-Chief’s staff.41
Eighty years on, and a statutory consolidation later, Sir Henry Thring, parliamentary counsel and architect of the consolidation, saw the 1803 Mutiny Act as the culmination of a long process of replacing prerogative articles of war by statutory authorization, because it was the first to apply to the army wherever it might be.42 This was a partial and formal truth only. It was true in that the sovereign needed legal authority to create offences punishable by death, flogging, or imprisonment, and to authorize their trial otherwise than by common law procedure, and true in that the most serious offences were delineated in the Mutiny Act, but it was not true that everything in the articles could be sourced back to the Act, save in the most general terms. The Act authorized articles to be made for the discipline of the army, and instructed judges to take notice of them, but that did not entail that articles were valid only to the extent that they elaborated some particular provision of the Act. Instead, as Vernon Lushington, Deputy Judge-Advocate General, explained to a royal commission in 1869, articles were valid unless positively contradicted by the Act, that is, they continued to be treated as though they were made under the prerogative.43 Though this flexibility rather disconcerted some members of the royal commission, it was good law, or, at least, it was good military law; whether it was good common law was not tested.44 It was certainly usual practice, as when the Common Pleas, in an unusually interventionist mood, had held that a soldier could not be imprisoned for disobeying an order to attend school, because the order was too remote from military concerns to be lawful, the response had been to amend the articles, not the Act.45
The articles duplicated much that was in the Act, and were thus much more compendious. Occasionally the codes were remodelled, in 1829 and again in 1860, and annual amendment seems to have been easy—the Act could be amended to introduce and then to fine-tune a statutory offence of stealing from a comrade, for example, or to validate some enlistments that had broken the rules.46 The result was an increasing refinement of both codes, the articles in particular (p.411) drawing praise from Cockburn CJ in 1867 for their elaboration and precision.47 But while courts of common law treated only the Act and the articles as law, officers responsible for prosecuting breaches of military law needed familiarity with the Queen’s Regulations, which detailed how the Act and articles were to be applied, and sometimes extended them and plugged their gaps, and with the numerous circulars and general orders issued in the Commander-in-Chief’s name; for as the standard text on courts martial explained, in fixing the meaning of the articles the expressed pleasure of His Majesty was decisive.48 Where all else failed there was still a role for military custom too; it was that, said Vernon Lushington, that must explain commanding officers’ summary power of detention, since neither the Act nor the articles did.49
The royal commission accordingly recommended a statutory simplification and consolidation of the two codes into one, which materialized a decade later in the Army Discipline and Regulation Act 1879, shortly afterwards amended and further consolidated as the Army Act 1881.50 Those Acts preserved the power to make articles—the Duke had insisted to the royal commission that they were necessary to preserve elasticity—and with touching constitutional precision that power was vested in the Queen, whereas the power to regulate procedure under the Act was vested in the Secretary of State.51 It is unlikely that this difference would have affected how new rules of either description were generated within the War Office, but what happened next would differ, for procedural rules had to be laid before Parliament, whereas articles did not. In the event a detailed code of procedure soon covered every aspect of courts martial from arrest to confirmation of sentence, taking in along the way such minutiae as the seating of members and the swearing in of shorthand writers.52 By contrast no articles were made, and the writers of the official manual that accompanied the reform henceforward (p.412)treated the basis of military law as wholly statutory.53 In that sense military law had lost its distinctive character as a blend of statute, prerogative, and custom.
Military law was administered by courts martial consisting of army officers. From 1829 there were three main types: general courts martial for the trial of commissioned officers and the most serious offences by soldiers, district or garrison courts martial for intermediate offences, and regimental courts martial. The long-term tendency was to diminish that last category: by the introduction of district courts martial in 1829 and by the later expansion of commanding officers’ powers of summary punishment.54 General courts martial were valid only if attended by a judge advocate, who was not a member of the court, but whose role, first by custom later by regulation, included advising the court on law and, until 1860, the settling of the charges to be brought. There was nothing quite like him in other courts, but clerks to justices of the peace were not entirely dissimilar. For important trials the Judge-Advocate General might send his permanent deputy if he had one, more usually he would appoint a judge advocate ad hoc; semi-permanent deputies were often appointed for garrisons or campaigns overseas. At first there was the same requirement for district courts martial, but it was quickly removed in 1830, the judge advocate’s duties devolving instead to the court’s president, its presiding officer.55
According to Simmons on Courts Martial, the leading text from the 1830s through to the 1870s, it was not unusual for civilian lawyers to have been appointed judge advocates before 1802, when a War Office circular insisted on military men only.56 Thereafter judge advocates were officers who had trained themselves in military law, or, from the 1860s, had been trained at staff college.57 By long tradition civilians were not allowed to speak for soldiers at a court martial.58 There was no bar to a lawyer being the friend the accused was allowed to have with him to suggest questions he might ask or statements he might make—like a twentieth-century Mackenzie friend—indeed in the most difficult cases the judge advocate (p.413) sometimes had counsel sitting behind him to whisper advice; but the court itself should not have to suffer them.59
In the 1860s procedural changes were made to encourage greater legal regularity in courts martial.60Judge advocates had gradually shifted into a wholly neutral role at trials, with particular responsibility for summing up the evidence. In 1860 they shed the final vestigal traces of their former function as prosecutors, though the charges might still be drawn in the Judge-Advocate General’s office. 61 In 1864 a deputy judge advocate was appointed for a five-year term for each of the three military regions of the British isles, to attend all general courts martial in their region and such district courts martial as they chose or were ordered to. There were possibilities for tension here; even before 1864 Simmonsworried that courts martial tended to devolve too much to the judge advocate, whereas Thomas Headlam, who was Judge-Advocate General from 1859 to 1866, wanted the judge advocate’s relation to the panel at a court martial to be ‘almost’ that of judge to jury.62 Army ethos, however, was that trial should be by officers, so there were difficulties in formally elevating a judge advocate over a court’s presiding officer, who would usually outrank him.63 The sensitive balance reached in the 1860s survived through to the end of our period, by which time it was written into the statutory Rules of Procedure: on points of law and procedure ‘the court should be guided by his [the judge advocate’s] opinion, and not overrule it, except for very weighty reasons’—more an admonition, perhaps, than a rule.64
No court martial could pronounce sentence itself. In all cases it had first to be confirmed by the commanding officer of the troops concerned—which once (p.414) caused Sir Henry Thring to muse that courts martial might not be courts in the true sense at all—indeed until the consolidation in 1879 no court martial could even pronounce its verdict.65 For general courts martial confirmation was by the sovereign personally, after receiving advice from the Judge-Advocate General as to whether the court had properly observed the law.66 Commanding officers also scrutinized the sentence to ensure uniformity; where the sovereign was the confirming officer it was the Commander-in-Chief who advised on sentence. Some of this process was statutory, or partly statutory, but until the consolidation most rested simply on military custom. In addition to the confirmation process, the papers of district courts martial were sent to the Judge-Advocate General’s office, where they were scrutinized for error.
These processes were important, but open to different interpretations. In parliamentary debates near the end of our period there was a strong current of opinion that the Judge-Advocate General was meant to protect soldiers from abuse, that this was a necessity given the harsh content of military law, and that it was also why he had (or, by then, ought once again to have) a seat in the Commons so that he could be held accountable. More cautiously, the Secretary of State, Arnold-Forster, added that the Judge-Advocate General also existed to protect courts martial from the illegality of exceeding their statutory powers—which historically has a truer ring to it, even if it had been giving way since the changes in 1860.67