THE GHOSTS OF CANNAE

Robert L. O'Connell

Random House, NY. 310 pgs., 2010 review by John Sloan

I have great admiration and respect for an author who undertakes a scholarly study of a difficult subject: All the more so when the topic is one that has occasioned numerous accounts over the centuries since its occurrence. Such is the case with a new book on the Battle of Cannae. When I was approached to consider preparing a review of this book, my first thought was 'why? why do we need another book on such an oft-treated classic. And in this case we have the excellent book by J. D. F. Lazenby Hannibal's War,.and several books on the Punic Wars. Can something new be revealed? What approach can the author take that may prompt readers to buy yet another story of one of the most famous battles in history?
Perhaps the chosen title reveals a bit of the author's purpose and what approach he has taken. This is a good book which I highly recommend but with reservations. The author has built a strong edifice on the basis of the ancient literary sources with superstructure provided by several of the best of recent authorities on Roman warfare, including Lazenby, which I discuss here. The content is narrative, analysis and conjecture. As the author notes, this last is considered required due to the sketchy nature of the original sources. In the main both analysis and conjecture are worthy of consideration, but, as I hope to show, the language style detracts from the value. Moreover, the two themes of the book are that the Roman losses at Cannae make this battle the most bloody in 'Western military history' and that the poor mistreated souls "Ghosts" who survived Cannae (apart from those sold into slavery) vindicated themselves in Scipio's victory at Zama. Both of these ideas are wrong as I will note below.
But in its style the book reflects the current approach to historical writing that I call the "Lady Gaga" method rather than a serious analytic study of the details of the battle. Apparently the publisher who commissioned Professor O'Connell to write this book wanted to include it in a series on battles, hence the title and the structural focus on Cannae as a battle. But the content can more properly be considered a book on the Second Punic War; or at least a partial study. This broader content is essential for setting the context and stage for consideration of the battle and especially for providing the author with support for his evaluation of the significance of the outcome. There are 6 maps, a useful 'cast of characters', extensive notes, glossary of Latin (but not Greek) military and technical terms, and index. The writing style probably appeals to today's audience, but I find it a bit florid and given to hyperbole. This is especially so in the author's required evaluation of both the immediate significance of the battle, itself, and especially its lasting influence including today. No question it was a huge loss for the Romans. But I have to agree with E. S. Creasy that the Battle of the Metaurus, in which Hasdrubal Barca was defeated, was not only the decisive battle of the war but among the decisive battles of world history. Nevertheless it is indeed Cannae that has captured much interest today due to Hannibal's tactical brilliance and the incredible slaughter of Romans that occurred. And it is the discussion of this influence on later generations, rather than the immediate significance of the outcome that is the chief theme of the text.
It is well known that historians choose to write on topics that interest them and their contemporaries. And they frequently do so with the point of view of their times and also may use their comments on a topic from a prior era as a foil for expressing their opinions on contemporary public policy. One may compare historical writing to the manner in which Rennaissance artists depicted Biblical and Classical scenes with their subjects dressed in current fashions. In the book under review we find much commentary and conjecture based on psychological analysis of the Romans and Carthaginians. It appears that the author (and perhaps publisher) were highly influenced by the trend in current writing on military history started by John Keegan, and followed especially by David Hansen - that is to conjecture on the psychological make up of the subject's participants.
In summary the book is worth reading as an account of Cannae and the Second Punic War. But it would be a much better book if the author had left out all the pop-psychology, hyperbolic language, and efforts to find some continuing influence and relevance for the study of the battle.

The author's bibliography - notes.
The historical foundation rests on Polybius and Livy with additions from Plutarch, Appian and scattered references from others.
The modern sources most quoted are Adrian Goldsworthy - The Punic Wars; J. F. Lasenby - Hannibal's War: A Military History of the Second Punic War, and The First Punic War: A Military History; H. H. Scullard Scipio Africanus, and A History of the Roman World; Gregory Daly - Cannae: The Experience of Battle in the Second Punic War; Theodore Dodge - Hannibal: A History of the Art of War Among the Carthaginians and Romans down to the battle of Pynda; and occasionally Hans Delbruck, Warfare in Antiquity . Many other authorities are cited for specific issues. Notably missing (unfortunately) are Theodor Mommsen and (fortunately) Liddel Hart.

Content: The book is divided into 10 chapters.
I - Traces of War
II - Rome
III - Carthage
IV - Hannibal's Way
V - The Fox and the Hedgehog
VI - Cannae
VII - Aftershocks
VIII - The Avengers
IX - Resurrecting the Ghosts
Epilogue - The Shadow of Cannae

Chapter I - Traces of War
The author immediately sets his stage by describing the outcome of the battle for the Romans in graphic (gory) terms. We see clearly what type of book this will be. It follows the current trope established by John Keegan (in The Face of Battle, Viking Press, 1979) to depict warfare in psychological terms that the author may prescribe for the combatants. O'Connell notes that the Romans on this terrible day lost more casualties than any other Western army in history. Well, I have to disagree. The Romans lost by general count over 86,000 Roman troops plus 40,000 allied troops plus almost as many camp followers and service personnel at the battle of Arausio in 105 B.C. And then the Russian losses at Borodino were about equal to the Romans at Cannae plus the French lost very many as well. But the uniquely horrible nature of Cannae will be one of the central themes of this account. The other theme is the deplorable condition enforced on the pitiful survivors enforced by the implacable Roman senate, that is exile to Sicily. This fact is obviously true - but the POW from several other, later, Roman losses were also exiled to Sicily. And at Zama the survivors of Cannae formed less than half of Scipio's army whose victory was achieved by his gaining the support of Numidian cavalry.
O'Connell also sets his scholarly stage right off also, by leading off with the personal view as might have been seen by Polybius, the great historian and principal source for our knowledge of the conflict. He follows this with an evaluation of Polybius himself and his possible motivations.
The next section is an excellent discussion of the litterary sources. But he appears to give inadequate value to the non-written sources. He comments, "Ancient historians were united in their belief that force was the ultimate arbiter of human affairs...." I might add, not only ancients - Mommsen aptly characterized the era by noting that a society, city or group then was either the "hammer or the anvil" when it came to relationships with others. By the way, O'Connell ignores Mommsen as a reference. He does not, in my opinion, discuss the sources that Polybius in turn relied on. Despite his note about what Delbruck thought of Appian, he cannot resist interjecting Appian's idea that there might have been 'a carefully plotted ambush' on an open plain. And he ignores Delbruck's careful reconstruction of the size of the armies involved.
Next, the author discusses the relevance of the battle, and our interest in it, in terms of modern history and the professional study Cannae has had over the years. This effort to create relevance I have to question, as Cannae never assumed a significance in the study of the history of warfare at the U. S. Military Academy as far as my memory or inquiries can determine and I have the actual text used in the course. (More below).
This section is followed by an extensive discussion of the history and nature of warfare itself from prehistoric times. But the theme is almost entirely psychological - about the nature of killing - and the 'unfortunate' human propensity to be a killer - with little attention to politics. Setting the psychological stage for Cannae is all well in good, but it is the political context that is critical. So I don't see much value in this exercise.

Chapter II - Rome
Again, the chapter begins with a psychological picture of the Roman army on its final march to doom. An excellent picture of Roman society appears. Indeed the Romans were a tribe devoted to war of the most brutal kind. As he notes, 'warfare was in essence the local industry'. The author gives a good summary of the structure of the Roman relationship to the other ethnic groups in Italy and of the real structure of Rome's government. But in my view Roman religious practices are not discussed sufficiently.
The most important part of this chapter is the description of the Roman army - its individual soldier, his weapons, and its legion. O'Connell, along with the rest of Roman equipment, describes the gladius hispaniensis and references Goldsworthy - The Punic Wars (page 47) - that it was probably introduced into Rome by Spanish mercenaries during the First Punic War. This is controversial.

But in Goldsworthy's other books he hedged this idea. In keeping with his methods following Keegan O'Connell launches into a detailed description of the psychological aspects of close-quarter fighting with edged weapons. The author then poses some interesting ideas about the forces of the Roman allies - the alae - who traditionally formed in legion-like units on the flanks of the two legions in a consular army. (more about this below). Next is a discussion of the Roman velites - light infantry troops. Finally the cavalry, which was comprised of the wealthiest class. True enough that this class also was the political class from which the Roman senior officer corps was formed. But I believe it is a stretch for the author to opine "the fact that the route to command among Romans was through service on the back of a horse raises questions about the leadership of what was essentially an army of foot soldiers." I don't believe there is anything in the sources that would indicate that the aspiring Roman politician who rose through several military-political offices served in the cavalry once he was elected to the lowest political office. Moreover, being on horseback and a cavalry man are two different things - Grant rode a horse on occasion but he did not have a 'background' in cavalry. Besides, for instance, at Ticinus, where Scipio the younger rescued his wounded father it does not appear that they were acting in a cavalry role, although as commanders they were on horseback. The problem for Roman cavalry was that the Roman individuals were not true horsemen. There is more, but we move on.

Chapter III - Carthage
Again, the chapter begins with a snippets showing Hannibal awaiting the Romans on the eve of the battle. But it rapidly digresses into a look into the collective Carthaginian head. "they have left almost nothing in the way of literature," Well the Romans took care to insure that.
The author has a very sarcastic and negative view of the value of elephants in battle in this chapter and throughout the book. After dismissing them he comments. "... marking Hannibal as the last of and greatest of the Punic pachyderm true believers. " Well, as the last significant Punic general I suspose this carefully worded comment stands. But elephants were used in battle by 'true believers' for centuries after Zama as Nossov shows. They were used quite effectively by the Romans themselves in the Macedonian wars.
A major part of this chapter is devoted to a valuable resume of the First Punic War and Hasdrubal's initial efforts to regenerate military power for Carthage.

Chapter IV - Hannibal's Way
A brief excursis on Hannibal's character is followed by a summary description of his march from Spain across southern Gaul and over the Alps. But first we read this, "There is little doubt that this was a sane, even psychologically healthy, individual." Hannibal that is. Pure gratuitous comment based on practically nothing but the author's own conception of what a 'sane' individual might appear to be. It is useful that the author notes Hannibal's Greek, Hellenistic culture. O'Connell also describes the Celts and the Roman opinion of them. And he stresses that Hannibal well understood both peoples and counted on generating significant support among the Celts. The description of the march through southern Gaul and across the Alps is excellent. He declines to opine on which pass Hannibal used.

Chapter V - The Fox and the Hedgehog
Hannibal of course is the fox and the Romans (especially Fabius) are the hedgehog. A description of the campaigns prior to Cannae follows, and especially a discussion contrasting Hannibal and Fabius. At the outset we read this: "In the end he was smacked down by the central non sequiur of the Western war of war: victory in battle does not necessarily mean victory in war." Here we have a theme 'western way of war' popularized by John Keegan and David Hansen. But surely that victory in battle does not necessarily mean victory in war is not 'western' in any sense nor is that idea the central focus of Keegan's popular notion. O'Connell accepts without comment the view that Hannibal "lost much of his army in the Alps" an idea dispelled by Delbruck. The descriptions of the battles at Trebia and Lake Trasimene are clear. He notes without comment that Hannibal deployed "eight thousand light infantry forward..." at the Trebia. (More on this incident later - see Polybius.) O'Connell then gives a good summary of strategies, policies and events on both sides during the dictatorship of Quintus Fabius Maximus. Significant is that he points out the 'rearming' of the African troops with captured Roman weapons and especially that he notes (quoting Head) the translator's error of the Greek word for pike or spear so commonly found in Polybius' history text.

Chapter VI - Cannae
The chapter is only 37 pages long and in it is very little analysis of the tactical operations themselves. But the diagram of the battle on page 151 is excellent. The chapter is mostly an effort to describe the 'terrible' psychological impact on the Romans. For instance, "The same crowding into helplessness must have been taking place on the flanks..." Lots of 'must have been' comments in this book. The author remarks, 'Not only does the process beggar description, but exploring the details of the massacre might seem to serve little purpose beyond pandering to some blood lust with a kind of pornography of violence." But this approach is exactly what this book is all about. He continues in this section, in fact, to "pry into the details of this exercise in mass homicide." Well all combat in warfare may be labeled 'homicide'. He continues "We live in an age when killing is cheap, virtually automated that was far from he case at Cannae." Well, sure, the killing then was not automated, but killing (and death itself) in Roman times was cheaper than today. Or putting it another way, death was a commonplace and killing even a spectator sport. The Romans enjoyed watching people torn to pieces in the Arena and burried alive two sacrifical victims at the height of the Punic War. He estimates that 100 men "had to be dispatched every minute. Yet even this astonishing figure under estimates the swiftness and profusion of the slaughter." Another author estimates a much higher number. I find this figure not at all astonishing when considering how many thousand pairs of opponents were engaged in hand-to-hand combat simultaneously. Plus the Roman mass was being subjected to overhead sling shot fire. Consider how many casualties occurred each minute at Borodino. What IS astonishing is the relatively small casualty rate among the Carthaginians. Even counting only the number of Roman pila that might have been thrown it would appear that this weapon was very ineffective.

Chapter VII - Aftershocks
The narrative and assessment continues with the events and policies of the following several years. The Romans reverted to a strategy of attrition, while Hannibal, unable to attempt another tactical annihilation, continued his own strategy of political attrition based on political inducements to find allies in Italy in Sicily and in Macedonia. The narrative shifts to politics and Carthage and Rome and to early operations in Spain.

Chapter VIII - The Avengers
This chapter begins with Scipio Africanus' rise to power starting with his assignment to Spain. What I have never been able to understand is how Scipio, one of the survivors who escaped death at Cannae, also escaped relegation to the two legion's worth of exiled survivors who were sent to Sicily. Apparently family connections were important, as usual, as the exiled survivors themselves noted in their petition to be recalled. It is possible and indeed likely that most of the survivors were from the velites - the poorest property class.
Our author continues with a good summary of events in Italy. He gives good credit to the generalship of Marcellus. He gives a clear description of the Battle of the Metaurus and its immediate military results . He notes the condition of the Roman populace before and after the battle, but not the significance of the Roman victory ( that it was the decisive battle of the war). One needs to read Mommsen and Cresey for this. Also, Mommsen faults Scipio for failure in his primary mission, which was to keep Hasdrubal in Spain. Our author than returns with Scipio to Italy and the political conflict over a campaign to Africa.

Chapter IX - Resurrecting the Ghosts
The chapter is about Scipio's campaign in Africa and the Roman victory over Carthage. The 'Ghosts' are the Roman soldiers exiled after Cannae to Sicily. The author's view that the Roman senate's treatment of these folks was despicable forms one of the principal handles he creates for discussing the battle. Zama took place 14 years after Cannae, 14 years of significant warfare, during which the Roman survivors of several other major defeats were also sent to Sicily, so we may wonder how many of the veterans of Cannae were serving in Scipio's army by then. But certainly the veterans were by then highly trained. In fact Scipio spent over 2 years training them prior to Zama. O'Connell relies on Livy for most of his discussion of Scipio's preparations for movement from Sicily to Africa. (Polybius' continuous surviving text ends after Cannae.) The entire campaign is clearly described. So too is the immediate aftermath- the Roman expansion.

Epilogue - The Shadow of Cannae
The chapter begins with a survey of the possible influence of knowledge about the battle on subsequent military theory and practice. The author assumes that the battle was 'decisive', which I claim it was not. After stating so unequivocally that the battle was influential for millennia, he writes that the lack of emphasis (even interest) by subsequent classical historians "probably indicates a proclivity to overlook what was by all accounting a miserable episode for Rome". Maybe they assigned the battle its rightful place in consideration of its eventual outcome in the context of the Roman conquest of the entire Mediterranean basin. For instance Livy notes that Cannae was not as significant a defeat as Ailia. Polybius in particular was focused on describing and explaining how and why Rome in the course of about 53 years expanded from partial control of central Italy to the entire "civilized" world, in the course of which Cannae was but a minor setback. And the Roman loss at Arausio resulted in the rise to power of Marius. O'Connell notes that during the Middle Ages even knowledge of the battle was confined to Byzantium. But the ideas of flank attack and encirclement are so basic to military experience that one could hardly attribute them to emulation of Cannae. One doubts if the Kipchak, Scythians, Mongols and steppe nomads in general had studied Cannae, not to mention the American Indians. The author ascribes 'an apparent conversion' to the study of Roman organization and tactics undertaken by the famous House of Orange. But the organization and battlefield tactics of the Roman legion are not the same thing as trying to emulate Hannibal.
He shifts to the modern era with focus on Alfred von Schlieffen and the German general staff. However, he notes that von Moltke achieved a double envelopment of the French without coaching from Hannibal. Here O'Connell mixes up two distinct aspects of Cannae that were subjects of military thinking. One is the double envelopment at tactical and even operational level. But the other, more of interest to von Schlieffen, is the strategy of annihilation versus the strategy of attrition. Cannae was not an example of either, since it was purely an example of tactical envelopment and subsequent annihilation within a general strategy of attrition. Plus, Hannibal was also noted for his use of deception. O'Connell notes correctly that von Schlieffen's text titled "Cannae" was reprinted at the U. S. Army Command and General Staff College in 1931. He may have missed that the same text was reprinted again at the U.S. Army War College in the 1980's as one of an extensive library of military classics in hopes of stimulating more consideration of the operational level of war. I have this text and will comment on its importance and possible relevance to Cannae in the list of references.
The author skips over the successful German encirclement of Russian forces at the Battles of Tannenberg and Masurian Lakes. Then, in discussing World War II he comments, "Even in the face of disaster, the Germans stuck to the theme...." But the operational double flank attack resulting in repeated encirclements was the crowning success of the German Operation Barbarossa. Yes, as he notes, the German effort at Kursk was a failure. But then the repeated Russian double envelopments that formed the central operational concept of their victorious march westward are ignored. (Not that the Russians needed instruction from Hannibal.) But the comparison of strategies of annihilation and attrition was the subject of considerable Russian military thought between the wars. O'Connell continues with his negative view, "Yet the Germans were not alone in responding to Cannae's siren song." He claims, derisively, that the American 'dream' of Cannae lived on after WWII. Well, McArthur's encirclement of the North Koreans as a result of the landing at Inchon was no dream. Nor was the Chinese encirclement of Americans during their counter- offensive.
O'Connell makes a point of claiming that Cannae was a significant topic of study at the U.S. Military Academy and that General Schwarzkopf, who graduated in the Class of 1956 apparently remembered this. Provided here is the actual text of the course book chapter on Hannibal in which Cannae is described. As for Norman Schwarzkopf's comment about Cannae in relation to Desert Victory, he obviously was not thinking about a double-envelopment nor a war of annihilation. Rather he was thinking of the manner in which Hannibal deployed his heterogeneous forces to maximize their various attributes and skills and the remarkable analogous way that Schwarzkopf did the same thing in his deployment of the varied national units for Desert Storm. And his direct comments were about Hannibal's generalship as an example of correct employment of the enduring 'principles of war'.
Finally, O'Connell comments that "the possibility of maneuver warfare seems, for the moment, distant.": and that modern intelligence might preclude Hannibalic results. But the various terrorist groups in Afghanistan are indeed fighting a war of maneuver and have achieved encirclements at the tactical level more than once. But no one today outside the fanatic head of government in Iran is thinking about annihilation.

Reviewer's comment:
SEE; For general discussion of the Punic Wars which contains an extensive bibliography- for biography of Scipio and his family and of Hannibal. And here is a list of general and specific references for the Battle of Cannae and for the course on History of Miltary Art at U.S. M.A. For easy access I have included the relevant chapters from Mommsen's master work on the History of Rome