I

The formulation and execution of national strategy demands the exercise of judgement. Aligning strategic aspirations with national resources obliges decision-makers to estimate the opportunity cost of each possible option – economic, fiscal, social, and political – both in the present and future. Even as the context and consequences of their choices are broad, some of the technical issues which they must consider can be narrow. In matters of national defence, decisions about the allocation of resources can call forth intense debate over esoteric technologies and fundamental principles at one and the same time. The present article offers a case in point.

In late Victorian England, the ‘two-Power standard’ constituted the lynchpin connecting naval policy with national politics. At that time, naval power reigned as the supremely British way of warfare, with the battleship as the primary means and ‘command of the sea’ as the ultimate end. The strength of the fleet was determined by parliament under the 1889 Naval Defence Act to constitute a force of battleships at least equal to the combined strength of the next two largest navies. When the standard was adopted, the cost to the Exchequer was approximately £13 million out of effective government spending (excluding debt service) of £67 million. By 1905, driven by major expansions in the French and Russian navies, naval expenditure had risen to consume £37 million of £121 million. Such figures attest to the centrality of the Royal Navy in British political life during this period, and hint at how intertwined were questions of naval policy with state finances and domestic politics.Footnote ²

In late 1904, a major change in warship construction policy initiated by the Board of Admiralty led to a new metric for calculating the two-Power standard. On 22 December, First Lord of the Admiralty the earl of Selborne, the political head of the Royal Navy for the past four years, approved the appointment of a special committee to review the designs of the new-model warships proposed to be laid down under the 1905–6 construction programme.Footnote ³ In March 1905, the Admiralty approved the report of this so-called ‘Committee on Designs’ and announced their intention to order one battleship, later named HMS Dreadnought, plus four (later reduced to three) Invincible-class armoured-cruisers, later known as battle-cruisers.Footnote ⁴ Whereas construction of the battleship was accelerated so that she was completed in late 1906, the battle-cruisers did not enter service until mid-1908. As a result, public attention was fixed from the beginning on the Dreadnought.Footnote ⁵

Conflating public perception with that of the actual decision-makers in government, historians have generally viewed the story of the ‘battle-cruiser’ as a footnote to the supposedly bigger stories of the ‘dreadnought revolution’ and ensuing ‘naval arms race’. Specialist naval historians have not disputed this portrayal because they have applied purely naval metrics to the subject, assessing the essential design concept (big guns, high speed, and thin armour) as flawed and the construction of the type as technologically errant, a verdict shaped and seemingly supported by the catastrophic losses in 1916 at the Battle of Jutland. Viewed this way, the historical debate over battle-cruisers, such as there has been, turns upon why they were built and what their intended function was. Non-specialists would be forgiven for thinking that if ever there were a subject that fell exclusively within the limited domain of ‘naval historians’ – technocratic in nature and narrow in historical significance – then this would appear to be it.

In fact, understanding the genesis of the battle-cruiser is far more important – and interesting – than mainstream historians might realize, because it had broad political, financial, and social implications. To appreciate those, we must comprehend the issues as they appeared to decision-makers at the time, in all their complexity. By failing to grasp simultaneously the technical details and the broader fiscal-political context, historians have erroneously drawn a clear distinction between ‘dreadnoughts’ and ‘battle-cruisers’. Before the First World War, however, contemporary observers were more struck by their similarities than by their differences: both mounted an all-big-gun (12-inch) main armament and were propelled by turbine machinery which allowed them to steam faster and for longer than any contemporaneous large warship. The size and unit cost of each was approximately the same (£1¾ million apiece). In parliament and press, both types were spoken of as ‘dreadnoughts’. As we shall see, this conflation allowed politicians to count battle-cruisers as well as battleships towards the two-Power standard, precluding the need to build as many of either and thereby releasing millions of pounds for welfare spending.

This article begins with a document: one well known to historians of pre-war British defence policy, but too often misunderstood in ways reflecting problems within the field of naval history generally and the study of the battle-cruiser specifically. It then turns to reviewing the evolution of historiography on the battle-cruiser. This section includes an extended critique of a recent book published by Oxford University Press which, though favourably reviewed, in fact impedes our understanding. The article concludes by analysing a previously unnoticed parliamentary exchange which sheds new light upon the precise configuration of the aforementioned naval-political lynchpin – the two-Power standard.

II

Why the battle-cruiser? Traditionally, historians approaching this question have taken as their starting point Admiral Sir John ‘Jacky’ Fisher, who bestrides the naval history of the period like a colossus. The battle-cruiser unquestionably was his brainchild. As is well known, Fisher took office as First Sea Lord on 20 October 1904 intent on building ‘new types of fighting ships’ incorporating greatly increased firepower and speed, and he had already secured the First Lord of the Admiralty's consent to appoint a ‘committee on new designs of War Vessels’.Footnote ⁶ To begin the story with Fisher, or attribute the battle-cruiser to him alone, is a mistake, however. The £7 million earmarked for the construction of four such vessels (though ultimately only £5¼ million was spent building just three) represented a considerable fiscal and political opportunity cost, and thus any attempt to explain the decision to build these expensive new-model warships, at a time of acute financial limitation, must go beyond a single officer's desires to explain their endorsement by the Board of Admiralty as well as the cabinet's approval.

The turn of the twentieth century was a period of uncertainty in imperial defence. Shifts in diplomatic alignments, the emergence of a globalized world economic system, and the development of revolutionary new technologies like intercontinental communications, to name a few factors, contributed to the creation of a fluid geo-strategic environment. At the same time, British leaders discerned a relative decline in national power and began seriously to debate countervailing action. The formation in 1902 of the Committee of Imperial Defence, a non-executive body tasked with prioritizing defence problems and sifting the options, exemplifies the recognition by the national leadership that managing Britain's path forward through such a complex and threatening environment required inter-departmental co-ordination and high-level political consideration: these were no longer matters for the war-fighting departments alone.

The critical conundrum was how to pay for Britain's imperial strategic aspirations. Since 1889, naval spending had surged. Initially, state revenues had kept pace, obviating the need to raise taxes. From 1895, however, with naval expenditure continuing to grow, the Treasury began expressing concern at the unfavourable fiscal trends. In the aftermath of the Boer War (1899–1902), matters rapidly came to a head. Falling tax revenues, the result of a post-war recession, combined with the government's inability quickly to deflate military (army) expenditure, plus difficulties in servicing the national debt, prompted Treasury calls for reduced naval spending. In the spring of 1904, the credit of the state sank to the point where the cabinet of A. J. Balfour felt compelled to accept the chancellor's demand for Admiralty economies.Footnote ⁷ This switch, from curb to cut, had profound implications.Footnote ⁸

In early May 1904, approximately a fortnight after the chancellor's plea for naval cuts, Lord Selborne invited Sir John Fisher to become First Sea Lord.Footnote ⁹ Though a controversial selection, ‘Radical Jack’ was known throughout the government as an outstanding administrator, and besides he was practically the only senior flag officer prepared to countenance major cuts in naval expenditure. For Selborne, this was the burning issue. ‘The first thing we must do’, the First Lord re-emphasized to Fisher just before he took his seat on the board, ‘is to establish the frame-work of the estimates for [19]05–6. When once we have done that then the Departments will be able to go on filling up the details and we shall be free for the consideration of the multifarious subjects we desire to deal with.’Footnote ¹⁰

On 6 December 1904, Lord Selborne distributed in cabinet a memorandum announcing the biggest naval strategic shake-up in nearly a century.Footnote ¹¹ Circulated just weeks before the appointment of the aforementioned ‘Committee on Designs’, this document sheds important light upon the prevailing direction of the Admiralty's strategic thinking and related ideas on operational requirements. At the core of the ‘scheme’ was a worldwide redistribution of British naval forces, made possible by the simultaneous implementation of a package of money-saving administrative reforms prepared by Admiral Fisher. Historians broadly agree that this memorandum signalled a seismic shift in British strategic posture, but disagree as to the intent and outcome. The traditional interpretation argues that it shows the Admiralty responding to the emerging German menace by taking the precaution of concentrating the fleet in home waters to face the most dangerous potential foe.Footnote ¹² A more sophisticated variant adheres to the notion of concentration against Germany, but superimposes a second imperative. It argues, essentially, that Britain's leaders, confronted with imperial overstretch and struggling to find the resources for defence commitments, consciously prioritized homeland security over protection of imperial interests in distant waters.Footnote ¹³ Metaphorically, the Admiralty recalled the legions from the outer marches of the empire. Both standard interpretations of this memorandum thus characterize its message as pessimistic and focused upon Germany.

In fact, its tone was infused with purposeful optimism, and it contained no suggestion that Germany was regarded as the predominant threat, or that strategic commitments at the periphery would have to be abandoned. It opens by acknowledging fundamental changes in the global strategic environment over the previous thirty years, the consequence of essentially two factors. The first was the emergence of various new naval powers (both at the periphery and at the centre), which was clearly an unfavourable trend. The second change stemmed from advances in technology, which Selborne characterized not as another negative development but as a positive counter-balance to the first. As surely as the Navy's strategic environment contained new threats, in other words, technology appeared to offer new solutions.

In discussing the policy implications of recent advances in technology, Selborne attached markedly greater importance to cruisers than battleships. ‘In the British Navy all the older battleships have been replaced by modern ones’, he reported, ‘but still more significant and far reaching in its consequences is the fact that this country is now rapidly becoming possessed of a number of modern armoured cruisers.’ These new warships, he explained, which displaced practically the same as battleships, ‘bring in their wake a revolution in respect to the composition of our cruiser squadrons’. Furthermore, this ‘revolution has come suddenly; four years ago there was not one such ship in commission’.Footnote ¹⁴ To be clear, the First Lord was referring here to armoured-cruisers with their sides protected by an armoured belt running the complete length of the hull.

The importance which Selborne attached to armoured-cruisers reflected Admiralty concern about the threat they posed to British oceanic commerce. Since the mid-1890s, the French (followed by the Russians) had been building large numbers of these new-model warships with the avowed intent of deploying them in wartime to wage a ‘Guerre Industrielle’ against Britain – a modernized guerre de course strategy – calculated to undermine British economic stability by disrupting her oceanic trade. Not only were these warships faster and better armed than existing Royal Navy cruisers, but their side armour (thanks to recent advances in metallurgy) was impervious to shells fired from the 6-inch guns that composed the latter's main batteries. In effect, they rendered all existing British trade protection cruisers obsolescent. From 1897, there ensued an arms race in armoured-cruisers among the British, French, and Russians, which was arguably more significant than the later and better-known arms race in battleships between the British and Germans.

The true intent and focus of the 1904 redistribution, as outlined in Selborne's memorandum, was to achieve better defence of the imperial periphery, not the homeland. The scheme called for the replacement of the existing scattering of autonomous ‘station fleets’ with centrally located ‘flying squadrons’ comprised of armoured-cruisers that could be directed to trouble spots by wireless telegraphy. What Selborne described, in other words, was a switch from forward deployment to rapid reinforcement, in which fast armoured-cruisers, rather than heavily armoured battleships, played the pivotal role. Three months later, Selborne reaffirmed to parliament that ‘I am more anxious on the subject of cruisers than I am on the subject of battleships.’Footnote ¹⁵ Whereas conventional interpretations of the fleet redistribution memorandum suggest that the Royal Navy changed strategic ends (from maintaining its global supremacy through forward deployment to protecting supremacy in home waters) while keeping its technological means (battleships), in fact the document indicates that the Royal Navy was preserving its traditional strategic end (maintaining global supremacy) while changing its means (from forward deployment to rapid reaction, and from battleships to cruisers plus better communications).Footnote ¹⁶ In effect, Selborne was boasting that his department had found a way to square the financial circle.

III

Let us turn now to the work of four historians who have wrestled with the issues discussed above. Arthur Marder established what became the standard explanation for the ‘dreadnought era’, and the battle-cruiser therein. Between 1940 and 1970, he produced nine volumes that endeavoured to tell the complete story of British naval policy during the period 1880–1919.Footnote ¹⁷ His conclusions gained widespread acceptance and as a consequence subsequent historians relied on his work when framing their own.Footnote ¹⁸ In 1973, Ruddock Mackay published a biography of Admiral Fisher in which he pointed to new documents that contradicted Marder's narrative – but he regarded the anomalies he had discovered as inconsequential. In 1989, Jon Sumida published a monograph that openly challenged Marder's interpretations, not only his explanation for the origins of the battle-cruiser but also his account of the ‘dreadnought revolution’. In 2012, the topic was revisited by another historian, Matthew Seligmann, who offered a new interpretation that disputed the conclusions of all three previous scholars.

In his foundational monograph, The anatomy of British sea power, published in late 1940, Arthur Marder offered the first considered scholarly explanation for the development of the battle-cruiser – albeit only in passing. His main concern was the genesis of the Dreadnought, which he characterized as a ‘revolution in battleship design’ – defined as an advance in naval architecture so great that rivals were compelled to follow suit.Footnote ¹⁹ Just as Dreadnought represented a leap forward in battleship design, so Invincible represented a parallel development of the cruiser. ‘The principle of the battle-cruiser was not new. The fast armoured cruisers of the pre-dreadnought era differed from the pre-dreadnought battleships of that date in much the same way as the battle-cruisers differed from the dreadnoughts, that is, they were faster but less powerful.’Footnote ²⁰ Put simply, battle-cruisers were bigger and better armoured-cruisers.

In explicating their intended role, Marder relied upon the testimony and advice of Admiral Sir Reginald Bacon, an officer who back in 1905 had served as Fisher's naval assistant and afterwards been appointed the first captain of HMS Dreadnought. In 1929, Bacon produced a two-volume biography of Fisher that dwelt at length upon the deliberations of the Committee on Designs (of which he had been a member). In 1940, Bacon twice reaffirmed his testimony, in a published letter and in his autobiography.Footnote ²¹ In the acknowledgements of Anatomy, Marder thanked Bacon for reading (and thus presumably commenting upon) his draft manuscript.Footnote ²² In his text, he echoed Bacon by presenting the functions of the battle-cruiser as straightforward and orthodox:

Almost as an afterthought, Marder added that the battle-cruisers were never intended to join the line-of-battle – a remark pregnant with implications – but in so doing failed to elaborate or indicate who had suggested this possibility.Footnote ²⁴

Perhaps the more striking aspect of Marder's account of the battle-cruiser was his sheer lack of interest in it as compared to the construction of HMS Dreadnought, which he regarded as far more important. In his chapter detailing the deliberations of the Committee on Designs, Marder devoted some twenty pages to considering the design of Dreadnought and just two paragraphs to that of Invincible.Footnote ²⁵ When Marder revisited the subject in 1961, producing the first of what would become a five-volume history of the Royal Navy spanning the period 1904–19 (From the Dreadnought to Scapa Flow: the prelude to war [FDSF]), he repeated his explanation for the battle-cruiser practically verbatim from Anatomy and cited no new evidence.Footnote ²⁶ The ‘primary duty’ of the battle-cruiser, he summarized elsewhere, was ‘to find the enemy's battle fleet and report its position, course, speed, and movements’.Footnote ²⁷ In thus presenting the battle-cruiser as a subordinate component of a battle-fleet, Marder implicitly ruled it out as a capital-ship.Footnote ²⁸

Marder's lack of interest in battle-cruisers was consistent with, if not necessitated by, his narrow conceptualization of strategy, the naval policy process, and the period in general. Central to his narrative (in both Anatomy and FDSF) was the belief that, from late 1902 onward, British naval policy was driven by an anti-German imperative. Marder regarded it as self-evident that by 1905 the Anglo-German arms race was well underway and this competition defined the overarching strategic and diplomatic environments. ‘By the autumn of 1902’, Marder wrote in his first book, British ‘public opinion, the government, and the Admiralty were as one in viewing the German fleet as a potential menace far greater than the fleets of the Dual Alliance.’Footnote ²⁹ Before year's end, he amplified, ‘the German navy menace hardened into an ideé fixe; thereafter no one in the Admiralty took the threat posed by the numerically larger French fleet very seriously’.Footnote ³⁰ In the introduction to the second book, he reiterated that ‘Although I have in the present volume been concerned with many facets of the history of the Royal Navy during the pre-war decade, I have kept my eyes fixed on Ariadne's thread – the British aspects of the Anglo-German naval rivalry.’Footnote ³¹ He never questioned his core assumption of an anti-German imperative. This conviction, assumed not demonstrated, served as the prism through which he viewed all evidence.

Another important facet of Marder's analysis was his decision to structure FDSF around the career of Admiral Sir John Fisher. In the preface to the first volume, he dubbed the period 1904–19 as ‘the Fisher Era in the Royal Navy’.Footnote ³² He elsewhere asserted that ‘In effect, if not in fact, he [Fisher] was the Board of Admiralty’, the implication being that only Fisher's opinion mattered.Footnote ³³ Marder's narratives portrayed Fisher as a heroic reformer who provided the vital leadership that brought an antediluvian Royal Navy to combat readiness before the First World War.Footnote ³⁴ Upon taking office as First Sea Lord, in Marder's telling, Fisher initiated a string of interrelated (and long-overdue) measures to improve the Royal Navy's battle-readiness. ‘The efficiency and strength of the Navy, one ready for war at a moment's notice’, he reaffirmed in 1961, was Fisher's ‘megalomania’.Footnote ³⁵ All these reforms, Marder explained, were geared to one aim, namely to maximize the operational effectiveness of the battle-fleet for the inevitable showdown with Germany. For Marder, therefore, with the Admiralty's strategic eye purportedly fixed upon Germany, HMS Dreadnought represented the centrepiece of Fisher's grand scheme and the crux of pre-war naval policy.Footnote ³⁶ Indeed, he periodized FDSF according to the development of HMS Dreadnought: her design forms one of the two terminal points of his study, with the end of the First World War supplying the other.Footnote ³⁷

In building his case, however, Marder encountered a stumbling block: he found no evidence directly linking the design of the Dreadnought to Admiralty perceptions of a German menace. In his first book, he admitted, ‘Not once in any of the Fisher papers, or the Admiralty archives or the report of the committee on designs is Germany mentioned even indirectly.’Footnote ³⁸ Undaunted, he improvised an argument whereby the design of the Dreadnought had been driven by exogenous technical considerations stemming from recent improvements in gunnery technique that permitted accurate shooting of big guns at hitherto unimaginable ranges.Footnote ³⁹ The superior fighting characteristics of ‘dreadnoughts’ over ‘pre-dreadnought’ battleships provided necessary and sufficient explanation for the adoption of the all-big-gun battleship.Footnote ⁴⁰ Thanks largely to Fisher's ‘genius’, the Royal Navy stole a march on their rivals in building next-generation battleships. Marder's account thus combined technological determinism with teleology: the Dreadnought was an inevitable upward step in the progressive development of battleships.Footnote ⁴¹

In his later work, however, Marder minimized and even disregarded his inconvenient evidentiary problem by directly linking the construction of Dreadnought to official perceptions of a German menace.Footnote ⁴² For instance, in FDSF, he declared that the design of Dreadnought ‘ushered in the most intensive phase of the [Anglo-German] rivalry’.Footnote ⁴³ More explicitly, he wrote:

These slips reflected the fundamental discrepancy within Marder's approach to his subject. On the one hand, he treated foreign-policy considerations (namely the German threat) as the primary determinant of naval policy. On the other, he regarded capital-ship construction (namely HMS Dreadnought) as the heart of naval policy. Yet, by his own admission, he had found no causal relationship between the German threat and the decision to build the Dreadnought.

Nevertheless, Marder's interpretation seemingly possessed internal consistency.Footnote ⁴⁵ If the German naval threat in the North Sea was the primary determinant of British naval policy, then it made sense that the British would have built battleships to fight German battleships; that the most famous battleship (HMS Dreadnought) and the man most closely associated with it (Fisher) should define the story chronologically; and that the Admiralty would have concentrated the fleet in home waters at the price of sacrificing imperial commitments. It also conformed to other interpretations of the period written by diplomatic historians which emphasized Britain's ‘continentalist’ turn away from empire and depicted the abandonment of ‘splendid isolation’ as recognition of British relative decline. Perhaps not coincidentally, Marder's doctoral supervisor was the eminent diplomatic historian, William L. Langer.

The continuing appeal of Marder's legacy rests largely on the perception that he had mastered the relevant archives.Footnote ⁴⁶ John Keegan no more than expressed common wisdom when he opined that Marder achieved ‘standards of archival research and organization of material which defy betterment’.Footnote ⁴⁷ We now know that Marder's archival research in pre-1914 documents was limited. Anatomy was built chiefly upon secondary material, seasoned with a few primary documents misfiled in the Admiralty Library plus interviews with retired officers and officials. In September 1938, just as he was set to leave Britain after completing his research, Marder was granted special access to the Admiralty archives.Footnote ⁴⁸ Given at most four weeks, and shown only certain documents selected by the staff (he was not allowed to see a catalogue or amend his agenda in light of what he found), Marder plundered what he could.Footnote ⁴⁹ Subsequently, and to the distress of Admiralty officials, he allowed people to believe he saw much more than he really did.Footnote ⁵⁰ But the constricted character of his archival research into the pre-war Royal Navy was not generally known and thus his interpretation of naval policy remained unchallenged for many years.

IV

In 1973, Ruddock Mackay published a new biography of Admiral Fisher in which he queried several assumptions underpinning Marder's analysis.Footnote ⁵¹ After cataloguing Fisher's private correspondence followed by unfettered research in the recently opened Admiralty records, Mackay uncovered documents that contradicted Marder's version of events. If there was a ‘Fisher Era’, Mackay concluded, then surely it began not with the design of HMS Dreadnought in 1905, but several years earlier while the admiral commanded the Mediterranean Fleet (1899–1902).Footnote ⁵² He presented evidence showing that it was during these two-and-a-half years that Fisher's ideas on capital-ship design solidified. Additionally, Mackay found documents dated February 1902 in which Fisher mapped out his ideal fleet distribution, proposing arrangements that were practically identical to those outlined by Selborne to cabinet in December 1904. The date of these papers is highly significant, because in 1902 – notwithstanding Marder's insistence to the contrary – Mackay found that Fisher unequivocally regarded France (in alliance with Russia), not Germany, as the most likely and dangerous naval foe, and that he saw the most pressing naval defence problem as the danger to British oceanic commerce posed by marauding armoured-cruisers.Footnote ⁵³ When Fisher became First Sea Lord, his correspondence did not exhibit an especially anti-German attitude.Footnote ⁵⁴ In short, the evidence did not support Marder's presumption of an anti-German imperative before 1905 driving either Fisher's vision or Admiralty policy.

Mackay identified additional problems with Marder's representation of Fisher's ideas on capital-ship design – something that touched upon the Admiralty's construction programme and thus imperilled the central pillar supporting the Marderian edifice. In Fisher's writings, Mackay discerned a consistent preference (at least from 1892) for building faster, less well-armoured capital-ships mounting guns capable of hitting at long ranges.Footnote ⁵⁵ Further, with respect to the design of the Dreadnought, Mackay found the story to be much more complicated than previously represented, leading him to query the veracity of Bacon's account (upon which Marder had so heavily relied). He further noted Fisher's propensity to blur the distinction between battleships and armoured-cruisers and employment of the word ‘dreadnought’ as a blanket term for both.Footnote ⁵⁶ Mackay discovered something else that Marder had missed, that even while the Dreadnought was under construction, Fisher tried to supersede her with a ‘fusion’ warship incorporating the best characteristics of the all-big-gun battleship and armoured cruiser.Footnote ⁵⁷ But what did all this signify?

Mackay elected not to press these lines of inquiry: his intent had been to produce a biography, not a monograph on British naval policy. His sense of both period and subject shaped by familiar landmarks, he was perplexed by what he had found and attributed the anomalies he had discovered to ‘confusion’ in Fisher's thinking. With respect to the battle-cruiser, this related ‘not to the details of the design taken in isolation, but to the role for which the ship was constructed’.Footnote ⁵⁸ In so doing, Mackay fell into the same teleological trap as had Marder by reasoning that Dreadnought was the logical evolution of the battleship, and by assuming that naval policy revolved around preparation for decisive battle between fleets consisting of battleships. Mackay deemed Fisher's promotion of Dreadnought as a wise decision because subsequent events demonstrated her to be a ‘revolutionary’ warship (and thus a ‘good thing’). By contrast, the loss of the battle-cruisers at Jutland demonstrated that the type was conceptually flawed (and thus a ‘bad thing’).Footnote ⁵⁹ Put another way, Mackay defined the fleet as constituted around 1916 as the culminating point of an evolutionary period extending back into the nineteenth century and used it as a kind of mental yardstick. He then evaluated developments in the period 1889–1919 according to whether they conformed to or deviated from the 1916 norm. In so doing, Mackay conflated two separate questions: First, what did Fisher believe (and why)? Second, was what he believed a good idea? These questions remained unanswered for sixteen years.

V

In 1989, Jon Sumida published In defence of naval supremacy. Footnote ⁶⁰ He was the first scholar systematically to examine the Admiralty archives for the period 1889–1914, as opposed to consulting only archival files explicitly labelled as pertaining to capital-ship design. He argued that the formulation of naval policy during this period was far more complex than historians had realized, requiring them to engage the policy process at much greater breadth and depth than hitherto attempted or deemed necessary. His signal contribution was not to provide new answers (though he did) but to redefine the questions, problems, and sources that historians of naval policy needed to consider.

Sumida's most dramatic finding was that Fisher opposed the construction of the Dreadnought, wanting instead only the Invincible armoured-cruisers. This proposition represented more than just a challenge to Marder's interpretation of British capital-ship construction policy and thereby to his entire thesis: it constituted a direct attack upon Marder's underlying methodology. Most naval historians have never been comfortable with Sumida's conclusions (even as they struggled to contradict his formidable evidence), not least because he argued against what seemed to be the single most obvious ‘fact’ about pre-1914 naval policy – namely, that there was a ‘dreadnought revolution’ at its heart.

So, how did Sumida reach this contrary verdict? Fundamentally, he did not regard the Anglo-German naval competition as a self-evident starting point for examining early twentieth-century British naval policy. His research suggested that the single most important determinant at that time was not the German threat (foreign-diplomatic) but money (domestic-financial). He did not deny that other factors mattered, or that fiscal pressures were more severe in some years than in others, but merely that in the years that formed the fulcrum of Marder's analysis – roughly 1897 to 1905 – financial issues more than any others governed what the Royal Navy did, and that these financial issues had very little to do with Germany. Of the causes of the crisis in state finances precipitated by the rise in naval expenditure, the most important was unquestionably the Franco-Russian cruiser threat. In short, Sumida found that key policy decisions were taken in an environment of finite monetary resources and that this factor more than any other defined the essential context.

From this new financial perspective, Sumida re-examined Fisher's ideas on capital-ship design. Incorporating more new evidence, he confirmed Mackay's argument that Fisher strongly emphasized speed and long-range hitting capability in capital-ships, going back well before his time in the Mediterranean and thus predating the supposed priority of the German threat. In capital-ship terms, the significance of Fisher's attraction to speed was obvious: one of the key differences between battleships and battle-cruisers was that the latter were faster. But the significance of Fisher's preference for long-range guns on capital-ships was less obvious, mainly because there seemed to be no choice or controversy involved: what naval officer would not want long-range hitting capabilities?

In fact, Sumida found, there actually was considerable choice and controversy involved. In so doing, he validated Marder's assessment that the gunnery issue was important, while showing that Marder's understanding of the issue was over-simplified. Sumida's examination of trends in the naval gunnery during the late 1880s and 1890s yielded findings that did not conform to established understandings. He found, for instance, that during this period developments in naval gunnery tended to boost the relative advantage of medium-calibre (6-inch to 10-inch) over big (12-inch) guns. More surprisingly still, the calibre of naval ordnance did not correlate to its maximum effective range. During these years, naval opinion regarded medium-calibre quick-firing guns as long-range weapons; the main armament (i.e. the big guns) were intended to smash armour at shorter ranges. The technical but straightforward explanation for this seemingly counter-intuitive state of affairs was that before 1905, technological constraints on mounting and loading arrangements made the rate of fire from big guns slower than that of medium-calibre guns. As a result, the aim of medium-calibre guns could be corrected more quickly, thus permitting greater accuracy at longer ranges and greater probability of hits.Footnote ⁶¹ Given its obvious significance for the supposed inevitability of the all-big-gun Dreadnought, this seemingly narrow point is essential – and it was not understood by historians before Sumida.

Like many other officers, Fisher subscribed to the view that battleships should carry medium 9·2-inch or 10-inch guns instead of 12-inch big guns. In a letter submitted officially to the Admiralty in 1901, Fisher extolled the tactical superiority of capital-ships which could steam faster, shoot quicker and hit more frequently, while noting ‘the comparatively few rounds that can be fired with accuracy from the 12 inch guns when closing the target at great speed’.Footnote ⁶² ‘The armament we require’, he proclaimed in 1902, ‘is the greatest number of the largest quick-firing guns in protected positions. They call it the secondary armament; it is really the primary armament!’Footnote ⁶³ Fisher's long-standing views on this technical-tactical issue largely explain why, at the beginning of July 1904, he was still calling for new battleships to be armed with a battery of sixteen 10-inch guns and armoured-cruisers with sixteen 9·2-inch guns. Not until November 1904 did he shift his preference – though not irrevocably – to uniform 12-inch armament for both. And he did so not in response to the German navy, but in response to a report he saw in late July 1904 on recent gunnery experiments.Footnote ⁶⁴ In other words, there was nothing obvious or self-evident about the state of gunnery or Fisher's views thereon: both needed to be studied in far greater technical and chronological detail than previous historians had realized.

Connected to his work on gunnery, Sumida also revised our understanding of the use and development of cruisers. Between 1897 and 1904, the distinction between battleships and first-class armoured-cruisers became increasingly blurred, as the latter grew in size and cost, and improvement in medium-calibre artillery (described above) conferred upon them greater combat capability. As Sumida emphasized, a first-class armoured-cruiser displaced practically the same as a contemporary battleship and cost almost the same; its hull was actually larger (thus requiring bigger docks), and it required more crew so was more expensive to operate. Many Admiralty officials – not just Fisher – viewed armoured-cruisers as akin to fast battleships. Though more conservative officers were uncomfortable with this idea, between 1897 and 1904 these vessels were designed with an eye to their incorporation into the line of battle.Footnote ⁶⁵

This idea gained additional credibility in 1903 with the (very secret) development of armour-piercing capped (APC) projectiles. As Selborne explained in 1903 to the chancellor of the Exchequer, when requesting the money to retrofit all capital-ships, the new shells would go through ‘almost any armour’.Footnote ⁶⁶ APC shells were authorized only for the warships ordered under the 1903–4 programme.Footnote ⁶⁷ The knowledge within the Admiralty that APC shells could perforate the armour of any existing battleship at current anticipated battle ranges (3,000 yards) and even at future possible battle ranges (8,000 yards), buttressed those like Fisher who claimed that fast, lightly armoured capital-ships could plausibly engage slower, more heavily armoured vessels without serious disadvantage.

To Fisher's mind, Sumida argued, these recent technological developments offered a way out of the looming financial-strategic crisis confronting the Navy. By the time he became First Sea Lord, Fisher had been persuaded that perpetuating a conventionally balanced fleet to contain the threat posed by the navies of France and Russia – and possibly Germany as well – would cost more than Britain could afford.Footnote ⁶⁸ His proposed solution was to combine the functions of battleships and cruisers into a single hull. Though individually more expensive, their greater versatility would permit the building of fewer total units and thus result in overall cost savings. Fisher's new-model, multi-role, qualitatively superior warship would form the centrepiece of a new, more affordable naval force structure.

Within the Admiralty, Sumida showed, the underlying logic of ‘fusing’ the two types was well enough understood if not entirely accepted. For Selborne, the First Lord, the battleship remained the final arbiter of naval power. ‘I quite agree that the battleship type and armoured cruiser type are merging into each other’, he noted in the margin of one of the papers submitted to him by Fisher on 20 October 1904, ‘but I am quite sure that…the Power which had the battleships must win’.Footnote ⁶⁹ Upon departing the Admiralty in early 1905 to take up the post of governor general in South Africa, Selborne warned the prime minister of what he had come to regard as Fisher's extreme viewpoint: ‘I always said that the battleship held the field, that the battleship counted for more than anything else, and that no number of cruisers could be substitutes for them. Fisher did not believe this when I left the Admiralty’.Footnote ⁷⁰ Even after the Franco-Russian armoured cruiser threat receded after 1905 – though it had by no means disappeared – Fisher continued to press for fast capital-ships. For the rest of his term of First Sea Lord, until January 1910, he persevered in seeking complete achievement of his vision.Footnote ⁷¹

By revealing that Fisher's supposedly outlandish ideas possessed solid foundations and others in authority were thinking along similar lines, Sumida showed that the mainstream of naval thinking was far broader than previously thought. Simultaneously, he defended Fisher from charges of inconsistency by demonstrating the rapidity with which the many inter-dependent variables affecting naval policy changed: the abrupt deterioration of British state finances, the simultaneous development of multiple important naval technologies, and the kaleidoscopic nature of international relations. That Fisher made frequent and rapid modifications in his vision was therefore a sign not of mental confusion but of intelligent adaptability. It was not Fisher who was confused in his thinking, but – sic semper – naval historians who were confused in their understanding of it. Above all, Sumida exposed the shallowness of the evidentiary foundations on which the traditional narrative of British strategy was built: the Marder interpretation could not possibly be right.

VI

In 2012, the battle-cruiser issue was again the basis for a general reconsideration of early twentieth-century British naval policy.Footnote ⁷² Matthew Seligmann implicitly endorsed Sumida's conclusion that previous historians had misunderstood and underestimated the significance of the battle-cruiser – but he offered a different raison d’être for their construction. He argued that Fisher conceived them expressly for catching German passenger liners, ocean-going vessels that were capable of steaming further and faster than any British cruiser then afloat and specially designed to serve in wartime as auxiliary cruisers (69, 77–8).

Seligmann's battle-cruiser thesis is predicated upon two suppositions. First, in his own words, that ‘the German threat became an important consideration in Admiralty thinking from the very outset of the twentieth century’ (171) – a plea for a return to Marderian orthodoxy. Second, contra Marder, that ‘British anxieties about Germany’, and thus Admiralty policy, were driven not by concern at the prospect of battle with Tirpitz's steadily expanding High Seas Fleet, but rather commerce raiding by Germany's ‘extensive fleet of large Atlantic liners’ (6, 171).

Showing that some officials at the Admiralty were disconcerted about Germany is neither original nor significant. The so-called ‘revisionists’, Mackay and Sumida, never denied that from the turn of the twentieth century the Admiralty became increasingly concerned by the rise of German naval power. What they did not accept was that the Admiralty's concern about Germany was so paramount that it could be analysed in isolation from any other concerns: what Marder (and now also Seligmann) treats as an absolute threat, they reframed as an emerging threat that Britain's naval leaders deliberated upon relative to others.Footnote ⁷³ Sumida's work on financial limitation is important here, because it demonstrated that the Admiralty had to define and quantify threats precisely relative to other threats in order to obtain the financial resources necessary to maintain the capabilities deemed essential to the preservation of naval supremacy. Yet Seligmann proceeds as though Admiralty evaluations took place against a background of limitless financial resources and non-existent political opportunity costs (74, 87).

Seligmann's claim that Admiralty policy during this period was ‘moulded’ (45) more by fear of German auxiliary cruisers than by concern over the growing number of battleships in the High Sea Fleet is implausible even on its face. In advancing his thesis, Seligmann fails to make clear that in 1902, the date by which the Admiralty supposedly had become fixated upon the threat from across the North Sea, Germany's ‘extensive fleet of large Atlantic liners’ (6) consisted of two vessels – albeit with two more nearing completion. By 1905, therefore, while the Committee on Designs was deliberating, Germany's fearsome armada of these leviathans thus totalled four. It should further be noted that, in dramatizing the novelty of the threat these ‘uncatchable’ liners supposedly posed (71–4), Seligmann seems unaware that the Admiralty had for decades been pondering how best to meet fast (French and Russian) auxiliary cruisers that (also) could outpace most Royal Navy warships.Footnote ⁷⁴ In other words this ‘problem’ was not new.

Setting aside these threshold flaws in Seligmann's case, let us evaluate his arguments and supporting evidence on their own terms. He agrees that Fisher was ‘the instigator and driving force behind the battle-cruiser. He devised the concept, appointed the committee that worked out the details of their design, and consistently pressed for these vessels to be ordered in ever greater numbers’ (68). The problem, Seligmann postulates, is that unfortunately ‘Fisher took no steps to provide a clear statement concerning the purpose of these vessels’ (68) and ‘uttered little more than platitudes about the purpose of his new armoured cruiser at the time he proposed it’ (76).

That ‘Fisher neglected to spell out the intended function of these new vessels’ notwithstanding, Seligmann claims, ‘his intent can be gauged in other ways’ (75). Arguing that historians may infer Fisher's intentions from other sources, Seligmann utilizes two sets of evidence. The first consists of statements made by other members of the Committee on Designs who are known to have been close to Fisher, and whose opinions may be regarded as proxies for his own. It is notorious, Seligmann declares, that Fisher handpicked the committee and ‘packed it with his supporters – people who he believed could be relied upon to back his proposals’ (75). He claims that ‘three of the most prominent naval officers in the committee’ – Captain Reginald Bacon (naval assistant), Captain John Jellicoe (incoming director of Naval Ordnance (DNO)), and Rear Admiral Prince Louis of Battenberg (outgoing director of Naval Intelligence (DNI)) – left unambiguous evidence showing they were ‘supporters’ of building battle-cruisers specifically to chase German liners (75, 68). The second block of evidence derives from post facto events: ‘once the new ships incorporating these recommendations had been ordered – then and only then did the First Sea Lord begin to articulate the role he intended such vessels to play’ (76). Seligmann thus argues that Fisher's subsequent words and actions are indicative of his intentions beforehand, and that he conceived the battle-cruiser primarily in response to the German liner threat.

Let us look at these two sets of evidence. In quoting Seligmann, we shall employ block quotations and introduce no new ellipses. Extended quotation permits readers to see the documentary evidence in its original context, and the context in which Seligmann uses each piece to advance his arguments.

VII

‘If anyone should have known why these vessels were built’, Seligmann writes, it was Reginald Bacon (70). As assistant to the First Sea Lord and a member of the 1905 Committee on Designs, Seligmann reasons, this officer was in the best possible position to divine Fisher's true motives and intentions (68–70). In advancing Bacon's explanation for the battle-cruiser, Seligmann relies not upon any contemporary documents but entirely upon Bacon's 1929 biography of Fisher and 1940 autobiography. Seligmann deploys the gist of his argument and evidence on a single page, providing two block quotations, one from each volume (69). He writes that in the biography: Bacon provided a clear, detailed and precise account of the reasoning behind the construction of the first battle cruisers. As he explained, the need for such vessels stemmed from the German strategy of arming its fast liners:

Bacon reiterated this explanation ten years later in his autobiography. Here he recalled:

‘Bacon's testimony was unambiguous’, Seligmann states: ‘Battle-cruisers were conceived as a specific and direct response to intelligence assessments that the Germans planned to attack British commerce with fast armed liners’ (69). But how dependable is this (uncorroborated) testimony?

In summarizing Bacon, Seligmann conveys to his readers the impression that his primary witness was reporting Fisher's opinions. In fact, Bacon was not. The passages taken from his Fisher biography are written in the passive voice, and are thus not directly attributable. In his autobiography, by contrast, Bacon was explicit. Yet, the quotation Seligmann reproduces actually begins thus: ‘My arguments were quite simple and absolutely incontrovertible.’ Note the key qualifying opening words, omitted by Seligmann, ‘My arguments’: Bacon says that the rationale he describes was his own, not Fisher's.

Seeking to cement his witness's supposed authority, Seligmann states that Bacon ‘provided a clear, detailed and precise account of the reasoning behind the construction of the first battle-cruisers’ (69). Yet, even the passages he cites (reproduced above) are riddled with errors. For instance, Bacon's claim that the Committee ‘knew’ the German liners were carrying guns in peacetime is wrong; Seligmann himself acknowledges that the Admiralty did not consider this possibility until 1906 (30). Bacon misdates the Cunard subsidy agreement as 1900; it was negotiated in 1902 and signed in 1903. Bacon says the Cunarders ‘had never been built to fight’ – but the whole point of their subsidy was to ensure that they were built to fight. Indeed, citing a post-war report, Seligmann acknowledges that ‘they were a model of how a merchant ship might be prepared ahead of time for use as an auxiliary cruiser’ (60, 165).

More fundamentally, Seligmann's justification for employing Bacon as a proxy for Fisher – that Fisher never beforehand explained the intended function of the battle-cruisers – is mistaken. He overlooks the Admiralty statement appended to the terms of reference and circulated beforehand to the various members of the Committee on Designs, pre-defining the essential characteristics of each warship type.Footnote ⁷⁷ Under the heading ‘Armoured Cruisers’, at the top of the third paragraph, we see that:

This is an official contemporary document and it plainly says that the Invincible-class was designed to overmatch not German merchant liners but foreign 24-knot armoured-cruisers. In 1904/5, only the French Navy was building (or even contemplating) such vessels.Footnote ⁷⁹

Although this document is unsigned, a trail of evidence links Fisher directly to its drafting, revision, and pre-circulation. Three weeks before the formation of the Committee on Designs, in a letter on this subject to Lord Selborne dated 3 December 1904, the First Sea Lord nominated thirteen men and outlined the instructions he wanted issued them.Footnote ⁸⁰ Fisher deemed it essential that beforehand:

In an accompanying memorandum, Fisher amplified the necessity of providing committee members with clear prior guidance and presented himself as uniquely qualified:

In effect, to get his way, Fisher was defining capital-ship construction as a purely professional matter.

Selborne, however, would have none of it – precisely because he recognized that capital-ship construction existed in a broader context beyond the competence, constitutionally and otherwise, of the Royal Navy's professional head. The First Lord declined to give Fisher the free hand he sought in the composition of the Committee on Designs. At Selborne's direction, membership was expanded and reinforced with more gold braid (detailed below). Selborne further insisted that responsibility for ‘laying down the basis from which the committee is to work’ and defining ‘the fighting requisites’ of each warship type did not belong exclusively to the First Sea Lord but to the entire Board of Admiralty.Footnote ⁸² Two days later, indeed, in his cabinet memorandum on fleet redistribution reviewed above, Selborne alluded to the impending formation of the Committee on Designs and stressed that ‘the Board of Admiralty [will be] first laying down as a basis what they consider to be the fighting requisites of the desired types of war vessels and the governing features of each type to which the other features shall be subservient’.Footnote ⁸³

In replying to Fisher, Selborne further decreed that when formulating these ‘fighting requisites’, the board must consult certain senior officers: ‘I should like to thrash out these very general principles, first with you & naval lords & [Vice-Admiral Sir Arthur] Wilson & DNI [Rear-Admiral Battenberg] & [Vice-Admiral Lord Charles] Beresford.’Footnote ⁸⁴ The ‘formula’ arrived at – as Selborne termed it – thus ‘will be the basis from which the Committee will work & the meaning of which you can expound to [sic]’ as its chairman.Footnote ⁸⁵ And indeed the minutes of the first meeting (held on 3 January 1905) confirm that after Fisher opened proceedings by reading the terms of reference, he proceeded to expand upon the preliminary statement – which, as demonstrated, he had drafted and the board had subsequently modified, and which other evidence shows had been already circulated.Footnote ⁸⁶

This set of documents is highly significant, historically and historiographically. First, it shows that Fisher was not allowed to ‘pack’ the Committee on Designs. It is therefore problematic to assume that anyone on that committee necessarily represented Fisher's views. Certainly Fisher tried, but Selborne exerted his authority over the composition of the membership and the instructions issued them beforehand. Second, and more importantly, it shows that the essential features of Dreadnought and Invincible were defined not by the Committee on Designs but beforehand, by the Board of Admiralty in consultation with fleet officers, and thus responsibility for them rests with nearly a dozen men, not with Fisher alone. Selborne certainly saw it this way.Footnote ⁸⁷ Third, the fact that the board issued the committee with such precise instructions, hammered out through negotiations with senior outside officers, renders it implausible verging on impossible that Fisher was not beforehand obliged to justify his creations.

Seligmann also does not deal with a contradictory viewpoint expressed by Bacon in a March 1910 paper titled ‘The battleship of the future’ – a document discussed at length by Sumida.Footnote ⁸⁸ Therein, Bacon stated that new long-range torpedoes would reduce the appeal of the traditional line of battle, while new capped projectiles capable of penetrating any practicable thickness of armour would encourage warships to carry only light armour. Bacon predicted that ‘the battleship as now known will probably develop from a single ship into a battle unit consisting of a large armoured cruiser with attendant torpedo craft’.Footnote ⁸⁹ Here, Bacon unquestionably was describing the battle-cruiser as a capital-ship that would fight other capital-ships. Most historians would agree that it makes more sense to prefer a document written sooner after an event over one written much later. Instead, Seligmann privileges Bacon's recollections composed a quarter-century after the design of the battle-cruiser over the views he is recorded – in a printed document extensively cited by another scholar – as having expressed just five years afterwards.

In light of this document and the numerous contradictions within Bacon's uncorroborated recollections, his credibility as a witness must be discounted.

Seligmann contends that then-Captain John Jellicoe also supported the idea of constructing battle-cruisers specifically to chase German merchant liners. At the time of his appointment to the Committee on Designs, Jellicoe was already in line to become the Admiralty's next DNO. According to Seligmann, ‘for some time’ (75) Jellicoe

A couple of pages later Seligmann writes that:

Jellicoe's ‘enthusiasm’, Seligmann declares, ‘speaks volumes about their [battle-cruisers’] intended purpose’: they were intended as an alternative to the ‘unarmoured fighting ships’ that were the Cunarders (78). Yet, when scrutinized, Seligmann's sweeping characterizations of Jellicoe's thoughts ‘for some time’ on complex technical issues depends upon just one bowdlerized paragraph from one letter.

Let us look more closely at this letter. Dated 24 January 1906, it is written from the Admiralty (where Jellicoe was then DNO) and addressed to Captain Reginald Tupper, a friend at Malta commanding a battleship in the Mediterranean Fleet. In order to allow readers to judge for themselves, Seligmann's interpretation of this letter, the single paragraph from which he gleans so much insight into Jellicoe's thoughts, is reproduced here – unedited – with the underlining as in the original:

In the first three sentences, Jellicoe was not expressing his own opinion, as Seligmann has it, but reporting what he had heard regarding the committee's deliberations about the proposed change in policy. (Tupper, incidentally, had helped to frame the original policy – an important context overlooked).Footnote ⁹¹ The observation that the Navy would be liable for a million was not Jellicoe's but the committee's – nor indeed was it even original to the committee.Footnote ⁹²

Seligmann's interpretation runs into additional problems as we move into the second half of the quoted paragraph. Here, Jellicoe does venture his opinion – though the abbreviated nature of his prose makes it difficult to decipher his meaning. Notwithstanding this difficulty, on balance the internal evidence suggests that Jellicoe was alluding to replacing the Cunarders not with battle-cruisers but with smaller unarmoured cruisers – the development of which was under consideration at this time.Footnote ⁹³ Several qualifiers in the text point towards this interpretation. First, Jellicoe wrote that he was not averse to employing unarmoured warships – provided they possessed sufficient speed, which he put at 26–7 knots rather than the committee's 25 knots. Second, he referred to the building of ‘a 25 knot cruiser’ – not an ‘armoured’ (i.e. battle-) cruiser. Third, where he mentioned unarmoured fighting ships, he underlined the word fighting: ‘fighting ships’ are warships like cruisers, not auxiliaries like the Cunarders. Fourth, he mentioned the possibility of these ‘unarmoured fighting ships’ operating in the vicinity of TBDs: in 1905 torpedo-boat-destroyers were not ocean-going vessels and thus would never be encountered in the waters patrolled by the Cunarders. As this ambiguous paragraph is Seligmann's only piece of evidence for Jellicoe's supposed views, his second witness should also be dismissed.

The third officer whom Seligmann cites as a proxy for Fisher's position is Rear-Admiral Prince Louis of Battenberg. As we have seen already, Selborne dashed Fisher's hopes of ‘packing’ the committee, and thus it is problematic to equate membership with proxy status. It is especially problematic in Battenberg's case, since his name was not among those that Fisher nominated for membership: Battenberg is absent from the seven ‘brains’ that Fisher proposed for the committee on 19 October 1904, and it is missing also from the list handed Selborne on 3 December.Footnote ⁹⁴ His omission from the latter is unsurprising in light of evidence that Fisher and Battenberg had fallen out over the latter's insubordinate behaviour during the Dogger Bank incident (21/2 October 1904).Footnote ⁹⁵ Almost certainly it was Selborne who proposed Battenberg (whose opinion he valued and whom he regarded as a ‘friend for life’) to the committee,Footnote ⁹⁶ as he unquestionably did Rear-Admiral Alfred Winsloe.Footnote ⁹⁷

We may nevertheless allow that until recently Battenberg had been within Fisher's orbit and shared his enthusiasm for armoured-cruisers.Footnote ⁹⁸ As DNI, from November 1902 until January 1905, Battenberg composed scores of minutes and memoranda that touched upon cruiser policy. These demonstrate that he consistently promoted the building of large armoured ahead of small unarmoured cruisers for the protection of trade.Footnote ⁹⁹ So keen was he to see more in commission, indeed, that to find more crews he favoured reducing older battleships to the reserve. In November 1904, Battenberg chaired an Admiralty committee that recommended dropping the solitary battleship from the 1905 programme (i.e. HMS Dreadnought) to permit the construction of a fifth Invincible-class battle-cruiser.Footnote ¹⁰⁰

As DNI, Battenberg certainly studied the potential threat to British commerce posed by German liners. Seligmann attaches particular importance to a memorandum he wrote in March 1903. The first part of this document suggests, as Seligmann has it, that Battenberg preferred fast warships (61, 72–4). But he equivocated, going on to recommend the continuation of subventions for British companies to build fast merchantmen, albeit under much stricter terms.Footnote ¹⁰¹ In any case, showing that Battenberg wanted armoured-cruisers and that he was apprehensive of the German liner threat is not the same as showing that in 1905 he wanted battle-cruisers primarily because he feared the threat from German liners. The evidence that Seligmann cites for the latter claim consists, again, of a single paragraph from a single letter (81–2). Addressed to Arthur Lee, who had served alongside Battenberg at the Admiralty, Prince Louis herein claimed (retrospectively) that he had opposed providing the Invincible class with 12-inch guns. According to Seligmann, Battenberg's comments demonstrate that ‘[Battenberg's] enthusiasm for this type of vessel [battle-cruiser] had related specifically to his wish to use warships rather than merchant ships to hunt German raiders’ (82).

It does nothing of the kind. The full text of the paragraph in Battenberg's pencilled letter to Lee reads thus:

From the last five lines, it seems reasonable to infer that Battenberg had not been in favour of all-big-gun armoured-cruisers (though voted for them). But to claim that this letter, in and of itself, shows that he had wanted them specifically to chase German liners, is untenable – especially since Battenberg's complaint about the excessive size of the armament relates to fighting enemy cruisers, not ocean liners. The last of Seligmann's witnesses is removed.

VIII

Let us turn now to Seligmann's second block of evidence: the steps that Fisher took after the Invincible-class battle-cruisers were ordered, which, Seligmann says, showed that his motive in building them had been to chase German liners.

Towards the end of 1905, Fisher established two ad hoc committees: one for a ‘review of Admiralty policy on the use of auxiliary vessels’, the other to consider ‘the future employment of mercantile cruisers’ (76–8). As Seligmann says, both recommended ‘phasing out’ the employment of armed merchantmen to catch their German counterparts, in favour of battle-cruisers. Drawing a straight line between two points – the Committee on Designs in January 1905 and the formation of these ad hoc committees in November 1905 – Seligmann argues that circumstances surrounding the setting up of the latter may be taken as corroboratory proof that the original intended purpose of the battle-cruiser was to hunt German liners (76–9). It would be wearisome to explain all that is wrong with this argument, a perfect example of reading history abstractly and backwards: certainly battle-cruisers could hunt down liners, but neither this potentiality nor inferences drawn from the findings of any subsequent committee mean that this was their primary or original purpose.

In asserting that these two committees were established merely to rubber-stamp ex post facto Fisher's construction policy (76), Seligmann overlooks crucial context. Much happened between January and November 1905. For instance, a few weeks before the committees were appointed the Admiralty let lapse several long-standing contracts with some half-dozen British shipping lines under which the latter had been providing subsidies to build merchantmen for wartime conversion into auxiliary cruisers, thereby creating a policy vacuum.Footnote ¹⁰³ Additionally, during these eleven months, the strategic-operational context originally underpinning battle-cruiser policy changed dramatically. The annihilation of the Russian Fleet at Tsushima (May); the unexpectedly rapid end to the Russo-Japanese War (September); and news that Russia, consumed with internal problems and verging on bankruptcy, had cancelled a large number of new armoured-cruisers (October). These vital considerations, especially the submergence of the Russian threat combined with warmer relations with France, led majority opinion at the Admiralty to conclude during the winter of 1905/6 that the danger to British trade had receded and whatever foreign armoured cruiser peril remained could be dealt with by existing British vessels, at least for a few years.Footnote ¹⁰⁴ In November, the Admiralty consented to deleting the fourth Invincible from the 1905 programme.

Fisher, however, demurred. As Sumida showed, the First Sea Lord responded by lobbying for the construction of a single ‘fusion’ type of armoured warship. To this end, at a special meeting of Admiralty staff held on 2 December 1905, he decreed that henceforth ‘the terms “battleship” and “cruiser” had been discarded, and the term “Armoured Vessels” was substituted in place of both’.Footnote ¹⁰⁵ This change in nomenclature was embodied shortly thereafter in a Parliamentary Paper known as the Cawdor memorandum.Footnote ¹⁰⁶ This evidence and more suggests that, whatever his reasons for wanting the two committees, Fisher was not thinking of battle-cruisers primarily as a response to German ocean liners, but – as Sumida has it – as multi-role ‘capital-ships’.

In developing his retroactive evidentiary argument, Seligmann offers readers a tangible and seemingly conclusive piece of evidence: a supposed minute by Fisher on a report written by Captain Edmond Slade detailing the results of a war game run at the Portsmouth War College during the winter of 1906–7. He highlights that in this report, Slade wrote that ‘the German liners ‘easily avoided’ the cruisers sent in search of them on account of their ‘superior speed’ ’. Seligmann trumpets that ‘Next to this, Fisher penned the terse but unambiguous remark that that was the ‘Raison d’être…of the high speed of the Invincible type’’ (79). In the footnote Seligmann doubles-down, writing that this minute was penned on a document titled ‘“Précis of War Game played at Portsmouth Naval War College, lasting from November 1906 to January 1907”, enclosed in Slade to Fisher, 11 February 1907’. He adds that ‘Fisher's undated minute is, therefore, probably from February 1907’ (79 n. 39).

To paraphrase Voltaire, the rubric in question was not penned, not a minute, and not by Fisher. What Seligmann actually cites is nothing but a printed sub-heading (placed in the outside margin alongside the text) that is integral to Slade's report.Footnote ¹⁰⁷ There are many such sub-headings throughout the document – no fewer than seven on the page in question. In Admiralty documents of this period, they were a commonplace reading aid. For example, they may be found punctuating the Report of the Committee on Designs. There is nothing anywhere on this page or elsewhere in this document that could be remotely construed as attributing these words to Fisher.

Problematic also with this alleged minute is Seligmann's exclusion of nine words by ellipsis. The rubric in fact reads: ‘N. B. - Raison d'etre of the new type of fast unarmoured cruiser and of the high speed of the Invincible type’ (emphasis added).Footnote ¹⁰⁸ Note the mention of a ‘new type of fast unarmoured cruiser’ (referenced above in the discussion of Jellicoe's letter to Tupper). This detail is significant because it is well known that Fisher despised unarmoured cruisers for imperial defence missions – one of his favourite similes was to compare them to ‘ants’ which in time of war would be eaten up by armoured cruiser ‘armadillos’ – and opposed their construction. It is thus inconceivable that he could have ‘penned’ an approving marginal reference to them. Indisputably, the words were Slade's, the author of the report.

In this connection, Seligmann's use of a January 1907 memorandum written by Captain Charles Ottley, the then-DNI, is illustrative.Footnote ¹⁰⁹ By the end of 1908, Seligmann maintains, there existed within the Admiralty a ‘broad consensus that Germany had a plan to attack British commerce with auxiliary cruisers’ (33, see also 80), which is critical to his thesis that the battle-cruiser was designed specifically in response to this threat. Ottley's 1907 memorandum is one of only three documents cited in support of this claim. In it, according to Seligmann, Ottley ‘spelt out the alarming lessons’ drawn by the Admiralty from a war game demonstrating the serious danger posed by German liners. He quotes Ottley as follows:

In this passage, Ottley's position seems unambiguously to support Seligmann's interpretation – but the surrounding passages tell another story.Footnote ¹¹⁰ The paragraph directly above reads:

The next paragraph (quoted above by Seligmann) in fact opens thus: ‘A “war game” now being played at the War College at Portsmouth, affords an apt illustration of this [emphasis added]. Under the hypothetical… [see above].’ And directly below the quoted paragraph, Ottley wrote:

The first point to note is that the subject of Ottley's memorandum was neither a discussion of the war game nor a review of the German liner threat. Rather, this (twenty-four-page) document is a defence of Admiralty cruiser policy. The reference to the liners, which occupies less than a page of text, transparently was intended to serve as ‘an apt illustration’ of a very different point. Ottley here was not drawing attention to the ‘success’ of the German liners, but the uselessness of small cruisers championed by Admiralty critics – the real point being that they were so inadequate that they could not cope even with an ‘auxiliary’ cruiser!

The second point to note, and fatal to Seligmann's work at multiple levels, is Ottley's declaration that the supposedly ‘uncatchable’ German liners could ‘ultimately be caught’ by a British ‘County’ class cruiser, of which sixteen were in service. Seligmann's thesis, it will be recalled, is that the battle-cruisers were conceived specifically to chase German liners, because no British cruisers then afloat were fast enough to catch them. Here, he not merely suppresses inconvenient evidence (on the same page as the paragraph he quotes) but presents the text out of context and in such a way as to reverse its meaning. There is a case to be made that Ottley's assessment was incorrect – but Seligmann does not even begin to make it.

Without a sound evidentiary foundation, Seligmann's edifice collapses. To be clear, the documents scrutinized above include not a selection of the contemporary evidence he deploys in support of his thesis about the genesis of the battle-cruiser, but its entirety.

IX

At the turn of the twentieth century, Britain's strategic environment manifested a high degree of fluidity and diversity, in recognition of which Admiralty and cabinet officials had to keep in mind a number of possible threats, singly and in combination. Rapid and costly technological change contributed to the contingency and dynamism confronting decision-makers. Under the circumstances, any policy decision was bound to be complex. Historians should be wary of simple explanations for complex policy issues, and alert to the probability that technology may be introduced on one premise and then be adapted to another over time. It will not do to reduce British naval policy at this time to ‘dreadnoughts and the German threat’ as a component of the debate over ‘the origins of the First World War’ or to study it in ignorance of broader fiscal and political contexts.

By the same token, given the Royal Navy's centrality in British fiscal-political life, it is a mistake for mainstream historians to ignore seemingly esoteric naval issues. Scholarship on pre-war British policy has paid too little attention to the significant tension between warfare and welfare inherent in debates over the Naval Estimates. The genesis of the battle-cruiser was a matter of considerable significance that embodied a compendium of political, fiscal and social, as well as naval operational and technological imperatives. Their construction sat at the nexus of defence debates over shifting strategic priorities, the necessity for social reform and state funding thereof, and, most directly, parliamentary controversies surrounding the continued applicability, relevance and calculation of the two-Power standard.

During the pre-war decade, most commentators regarded this standard as a necessary political safeguard of adequate naval funding for a sufficiency of battleships and thus the maintenance of British maritime supremacy. For others, however, it verged on an anachronism: devised in an earlier era under very different circumstances, now ruinously expensive and a barrier to the achievement of enhanced imperial security through spending less money more intelligently. Yet, all sides agreed that suddenly renouncing the standard courted disaster, both in naval and political terms. By 1905, in other words, the two-Power standard had become, not a means to an end, but the end in itself: a strategic shibboleth too expensive – whether in fiscal, political, or social terms – to maintain, and yet too venerable to abandon.

For nearly three years, between 1905 and 1908, the Liberal administration that succeeded the Conservative government of Arthur Balfour successfully avoided committing itself to upholding – or even discussing – the two-Power standard.Footnote ¹¹¹ The Liberal leadership knew all too well that this subject had the potential to split the party. This prolonged evasion was made possible by the magnitude of Russian naval losses at the May 1905 Battle of Tsushima. Nine British battleships were already under construction to augment the fleet; it seemed that, in upholding the two-Power standard, the Empire apparently could afford to live off its naval capital for several years to come.

In early 1908, however, Germany confirmed its intention to amend its Naval Law, the net effect of which was to accelerate its naval construction. Shortly thereafter, and unnoticed by historians, the leaders of both British political parties tacitly agreed to reconceptualize the two-Power standard. This was done in two ways. First, in calculating adherence to the standard, they agreed to count only warships commissioned after HMS Dreadnought – instead of, as hitherto, all ‘modern’ battleships.Footnote ¹¹² In effect, the politicians endorsed the notion of a ‘dreadnought revolution’, concocted by Fisher in mid-1906 as a political device to preserve the Navy's construction budget from draconian cuts at a time when the strength of the fleet was above two-Power levels.Footnote ¹¹³ Second, and even more significantly, the unit of measurement was changed from ‘battleships’ to ‘capital-ships’, the latter being explicitly defined as including both Dreadnought and Invincible. Readers will recall that, as originally authorized in 1905, and laid down in 1906, the Invincible class had been categorized as cruisers and, as such, did not count towards the then two-Power standard.

In March 1908, Lord Cawdor (First Lord, 1905–6) told parliament that ‘Invincibles [sic] are to stand in the battle line against the battleships of other countries; they are practically battleships.’Footnote ¹¹⁴ The following week, A. J. Balfour, leader of the opposition, recognized these changes in the definition of the ‘standard’, insisting they had been made upon professional advice (i.e. the Admiralty):

In July, Arthur Lee, the senior Conservative party spokesman on naval affairs, observed ‘that next year's programme of capital ships should be at least five vessels of the Dreadnought or Invincible class’.Footnote ¹¹⁷ The discourse culminated in November with the (Liberal) prime minister, H. H. Asquith, assuring parliament that ‘the Government accepts the two-Power standard of Naval strength as meaning a preponderance of 10 per cent over the combined strengths, in capital ships, of the two next strongest Powers’.Footnote ¹¹⁸ In sum, the Conservative former prime minister, the former First Lord, and his former parliamentary secretary, as well as the current Liberal prime minister, his First Lord and parliamentary secretary went on record saying that battle-cruisers as well as battleships would henceforth be classed as capital-ships and counted towards the standard.

This redefinition was pervasively political, and it underscores the necessity of grasping the interactions between seemingly narrow naval decisions and broader context. When Asquith made his statement, Germany had as many dreadnought battleships built and building as Britain. Only by counting battle-cruisers as equivalent capital-ships could the Liberal government protect itself from accusations of endangering British naval supremacy through underbuilding. Including the battle-cruisers within the two-Power standard therefore allowed the government to build fewer battleships – releasing state revenues needed to fund the (1908) Old Age Pensions Act. Precisely how many fewer ships, and thus the magnitude of the fiscal relief, is hard to calculate – a matter of warship completion dates, depreciated service lives and secret accelerations in foreign building programmes. Such details were the substance of the confusing political debate attending the 1909 naval scare. But the key point is this: what may seem to have been a technical taxonomical change about battle-cruisers, with some minor implications for British naval and maybe diplomatic historians, must have major significance for historians interested in British state finances, party politics, and social reform. Perhaps naval history is too important to be left to the naval historians.