Divisions of massed facts pitted against bold raiders

What possible excuse can there be for another book on the first world war? This is not a question that historians would ask, but our colleagues in other disciplines might legitimately wonder. As with any other historical subject, there can be three justifications. A major new work might attempt to provide a new and more complete synthesis than any existing work. Alternatively, it might radically take issue with the existing consensus on the basis of fresh research. Finally, it might adopt a radically different perspective that raises new questions.

These three books provide good examples of each approach. Hew Strachan has taken the first route. A general work that takes more than 1,100 pages to get the reader to the end of 1914 is obviously making a strong claim to be definitive. This is academic history of the highest order. The bibliography runs to 50 pages and includes everything of major significance written in English, French and German over 80 years. No historian in any language has matched this for sheer thoroughness. This makes Strachan's work indispensible for the specialist but daunting for the non-specialist.

Strachan's style is readable and elegant. His judgements are judicious and intelligent. But unavoidably he ends up telling most readers more than they really want to know. I honestly wonder how many are going to persevere with a chapter of 180 pages on financing the war. This is a shame because it is extremely important scholarship.

The extent to which this volume presents the first world war as a "world war" is a mixed blessing. The European campaigns are taken to the end of 1914, but the campaigns outside Europe are followed through for longer. Fifty pages on the war in the Pacific and 60 pages on the war in Africa are excellent and can be read by specialists with substantial profit. Strachan's knowledge of the French literature in particular makes his chapter on Africa perhaps the best thing ever written on the subject.

Yet at the same time there is a real danger of distortion because the eastern and western fronts combined receive only the same 110 pages. As a result, peripheral skirmishes end up carrying the same weight as major battles. This is to some extent redeemed at the end where a remarkably concise essay on "the ideas of 1914" should stand as one of those pieces of writing on the first world war that is genuinely required reading. This might be said of the volume as a whole for anyone with a serious academic need to know about the war. It will undoubtedly become the most complete and balanced history of the war ever published.

A very different vision emerges from John Mosier's book. Gleeful revisionism can sell very well, as has been demonstrated by Niall Ferguson. In fact, revisionism might be the wrong word for both Mosier and Ferguson, who seem bent on reinvigorating the older orthodoxy of a botched and disastrous British and French war effort and celebrating Teutonic efficiency. Mosier's subtitle in particular seems guaranteed to cause maximum offence in Britain: "How the Germans Won the Battles and America Saved the Allies". A generation of British military historians, plodding forward to rehabilitate the British Expeditionary Force (BEF), is now being mown down.

The argument that America made the decisive contribution to winning the war is not in itself unreasonable. So is Mosier convincing? Like Ferguson, Mosier is fixated on kill ratios on the Western Front. There is certainly a case to answer here, although both Ferguson and Mosier may be placing too much faith in German official casualty returns.

During the first three years of the war, the German army killed its opponents with disproportionate efficiency. Mosier is particularly good in his detailed examination of the Franco-German battles of 1915, which have generally been neglected by English-speaking historians sidetracked by Gallipoli. Mosier makes a good case for a German operational ascendancy in 1915 that bled the French army almost to death for no territory gained. A perfectly reasonable point, but there is a danger in simply extrapolating from this to a description in which the entire war becomes "more of the same". A second and greater danger is of uncritically translating the tactical into the strategic and reading this as a German victory.

A plausible master narrative of the war has the German high command incrementally losing touch with strategic reality; and Mosier, drawing heavily on the German official histories, follows this path. But as the book proceeds, the judgements become less researched, more partisan and ultimately badly distorted. While an overall case can be made that the British effort on the Somme was a bloody debacle, there is no excuse for ignoring a mass of detailed scholarship on the battle. This is compounded by increasingly fanciful and unsupported statements such as "a generation of young German officers had grown up reading tales of the Wild West... the idea of having to fight a country where the average farmhouse had as many guns as the average German infantry platoon was far from reassuring". Given that Mosier has spent the first half of the book extolling how well armed the average German infantry platoon was, this implies that American farmhouses were habitually equipped with light machine guns and mortars. The unspectacular performance of US regulars in their various recent "small wars" may in fact have weighed more heavily with German officers than memories of Karl May.

By 1918 Mosier's theses become fantastical. It is a sure sign that polemic has taken over from analysis when the same information starts being used to "prove" opposing points to fit the argument. Thus heavy American casualties become evidence of the decisive role being played by American troops whereas heavy British casualties demonstrate that they were being defeated by the Germans. A passing German reference to the quality of one American division becomes a generalisation about the German high command's fear of the entire US army, while frequent German references to the quality of other allied divisions and their criticisms of American troops are ignored.

Alliance warfare is a complex business, and mono-causal explanations of victory are inherently suspect. The "real miracle of the Marne" was indeed in 1918, but the miracle was in the hearts and minds of the battered French poilu and not in the useful but ultimately secondary contribution of two American divisions at Chateau Thierry and Belleau Wood. Fresh American troops did discourage the Germans, but more vitally they encouraged the French. Furthermore, the shift of focus eastwards bought valuable time for the battered BEF to rest and refit. By the end of July, it was abundantly clear that General von Ludendorff's gamble to bring the war to a victorious end had failed.

Mosier's interpretation of the situation is surreal: that the Germans had gained an advantage in buying space for an orderly retreat. In fact, the Germans had created a massive and vulnerable salient with a devastated battlefield behind and railheads dangerously distant. Germany lacked the oil and vehicles to maintain the front. The combined effects of huge casualties and the onset of the Spanish influenza meant it lacked the manpower reserves to counter-attack effectively, while the dissipation of the mirage of victory had dented the morale of its troops.

Inadvertently, Mosier repeats Ludendorff's mistake: that the casualties of the spring offensive had to mean the British army was finished as an effective force. Beginning on August 8 at Amiens, he was proved terribly wrong. Attempting to downplay the subsequent "hundred days" by saying that the BEF was simply following a retreating army will not work. In the course of August and September, the British not only recaptured all the territory they had lost earlier, they also smashed through the main German defensive line of 1917 and in the process fought the bulk of the German army and filled fields with captured troops and guns.

This record ought not be minimised, but the problem with the more Anglocentric celebrations of this as decisive is that, by and large, the advance occurred in the wrong place. The British had a long way to go before they reached anywhere genuinely important. Indeed, they ended the war much where they had started it - outside Mons. The best hope of a quick military end was a Franco-American breakthrough further east to cut the main German lateral railway line. For whatever reason, it did not happen. The American effort was split between two operations, the one the Americans wanted at St Mihiel and the one Marshal Foch wanted at the Argonne. Mosier argues that if the Americans had been left alone to break out of St Mihiel, they could have made the decisive effort in October. This is a tenuous argument. In terms of strategic logic, Foch was right.

My attitude is not one of simple denial of Mosier's original assertion, more of scepticism about his specific arguments. Without American aid of all kinds in 1917-18, the British and French might not and perhaps probably would not have avoided defeat. It is incredibly doubtful that they could have won the war without the influx of American troops, who were able to multiply the offensive pressures faced by Germany and take over large stretches of the line.

In the end, Germany was not brought down by a rapier thrust on the battlefield but by a massive bludgeoning; the impact of blockade, the collapse of its allies, domestic paralysis and poor morale and the pressure on the Western Front. Germany faced a complete systems collapse. Germany could have lost the war through any of these causes; it ended by losing it to all of them simultaneously. There was an inevitability here: Germany had juggled its resources with some skill, but they were ultimately inadequate.

If there is a weakness in the volume edited by John Bourne, Peter Liddle and Ian Whitehead, it is the absence of an essay on France in their comparative collection on the experience of the two world wars. Nowhere was the experience of the two wars more starkly different. While the collapse of French self-belief did not precede military defeat in 1940, it followed very rapidly and still represents a sharp contrast with the earlier endurance of the Third Republic. The British also have rather different memories of the two wars, the first is presented as the definitive bad war, the second in a much more favourable light. One of the pleasures of the half-dozen essays on Britain in this volume is the degree to which this false contrast is punctured. A particularly good essay on religion in Britain works very well as a comparison of the wars and brings out the extent to which religious belief was rather important to the British in both wars.

The essays are rather mixed in their intentions. There are some good comparative syntheses and some that are more narrow and monographical. They range from the broad canvas of a magisterial overview of total war in the 20th century by Immanuel Geiss to a more miniaturist approach, with Naoko Shimazu concentrating on middle-class Japanese women in the second world war. As the second volume of a two-part work, the concern is with social and cultural history rather than military history.

One startling reflection is the realisation that the century's bloodiest war might in fact be neither of the world wars as conventionally defined but the Sino-Japanese conflict. Diana Lary gives a range of between 20 million and 30 million Chinese civilian deaths. When the Chinese Nationalists deliberately breached the Yellow River dykes in 1938, nearly 900,000 civilians were drowned, a substantially higher death toll than the total for the UK during the first world war. Millions more were killed by the Japanese in deliberate atrocities. The Chinese death toll probably equals the more famous tragedy of the USSR. Although not according to Sergie Kudryasov in this volume, who has somehow come up with a death toll for "Russia" in two world wars of 45 million. This is frankly unbelievable. A deep strand of unreconstructed Russian chauvinism runs through his essay, including a polemic about how the newly independent Baltic states have built contemporary identities around their pro-Nazi past.

As this illustrates, the memory of the 20th-century wars remains a live issue. It also demonstrates why books of all three sorts must continue to be written. Exhaustive research acts as a counter-balance to the wilder enthusiasms of polemics, particularly polemics operating within celebratory national frameworks. At the same time polemics are required to shift research agendas and challenge complacencies. Finally, broader chronological and geographical comparison provides better perspectives. European conflicts need to be properly related to global upheavals, and the emergence through two wars of American global hegemony needs to be properly understood. So the bad news for the sceptics is that the war history to end all war histories is not likely to happen any time soon.

Adrian Gregory is lecturer in modern history, University of Oxford, and fellow of Pembroke College.