'It is astonishing to see so great a number of men.' Hannibal senses his anxiety and decided to turn it in his own way: 'Yes, Gisgo, you are right, but there is one thing you have not noticed.' 'What is that, sir?' asked the puzzled officer. 'In all that great number of men opposite us there is not a single one named Gisgo!' My interest in attempting to re-fight the Battle of Cannae developed quite by accident. Having spent the high summer months wrestling with an adaptation of the popular and playable FIRE AND FURY ACW Rules for a quasi-historical campaign set in the mid-18th century, I decided that a break from the concerns of cannon ranges, effects of musketry, and brigade composition for the warring if fictitious countries, would be a good idea. One day, while paging through WARFARE in the CLASSICAL WORLD, my attention was piqued by the narrative and maps describing the Battle of Cannae. 2 The more I read and re-read these pages, the more I wondered if this engagement might be reconstructed on the wargame table. The purpose or idea was reconstruction not as an exercise in history, but as a miniature wargame; a wargame with rules dedicated to providing a reasonably realistic, historical representation of the battle on that fateful August day in 216 B.C. In other words, it would not be much of a wargame if the Romans did not have at least a fighting chance. 3 The selection of a rules set for this wargame project was not a problem. ARMATI (both basic and advanced sets) by Mr. Arty Conliffe was my first and only choice. 4 I had purchased the basic set after reading a review and other articles about the rules in the pages of MWAN (Midwest Wargamers Association Newsletter). So impressed was I with the game mechanics and tables of organization for centuries of combatants, that I proceeded to submit several wargame reports to that very same newsletter, in which I described the fun I had wargaming with the rules. The purchase of the advanced module came some time after these reports, so it seems fitting that I be inspired, once again, to take word-processor in hand, and submit this report for consideration and review to fellow wargamers; Ancients enthusiasts especially. The selection of a primary source for information about the battle, proved to be more difficult than the selection of rules, however. Digging through my meager collection of military history texts, I was both fascinated and frustrated with the material uncovered. On the general points of the narrative, the sources were in agreement: Hannibal's infantry "invited" the compact legions into a three-sided box, and his heavy cavalry, after routing the Roman knights, closed the top of that same box, with obvious and disastrous results for the legions and their command. 5 It was with the particulars of the battle however, that the versions of events seemed to match the number of works consulted. 6 I possessed a copy of Livy's account, but lacked anything written by his primary source, Polybius. Indeed, due to the date of the battle, I could safely assume that I would not have access to any records from eye-witnesses. A trip to the local Border's Books and Music retail store solved that problem, well, partially anyhow. It was there that I discovered HANNIBAL'S WAR, by Professor J. F. Lazenby. 7 The professor's considered narrative compares and contrast the ancient authorities, and blends in critical scholarship. This text then, was my primary historical source for the re-creation of the Battle of Cannae on the wargame table. For the mechanics of that recreation or transfer from text to table, I relied heavily upon the guidelines provided by Mr. Donald Featherstone in his wonderful book, BATTLE NOTES for WARGAMERS. TERRAIN Mr. Featherstone states: "To refight any historical battle realistically, the terrain must closely resemble both in scale and appearance the area over which the original conflict raged . ." 8 Just as the selection of rules for this planned re-fight was a simple matter, so was the reconstruction of the battlefield. Professor Lazenby provides both an examination of Livy and Polybius, with respect to the choice of deployment on the right bank of the Aufidus, as well as a topographical map, showing the elevation of ridge lines that surrounded the probable site of the battle. 9 For the purposes of this wargame reconstruction, the table was unremarkable: there were smallish areas of lichen and scrub scattered about, the left table edge (from the Carthaginian side) may be designated to "be" the Aufidus, or one can place blue-colored card or paper, about two inches in width, down the near-left side of the table. As the course of the battle did not involve the river in anyway, there is no reason save an aesthetic one, to represent the river on the wargame table. The same may be argued for the slight ridge or rise that makes up the right table edge (again, from the Carthaginian side). 10 And so, the terrain needed for this specific table is very simple. A green cloth, with the occasional magazine underneath - to break up the flatness of the ground - would suffice. For this re-fight, these minimal terrain features were added to the 10' x 7' table. (Please refer to MAP I for terrain placement) NUMBERS AND DEPLOYMENT The second main guideline from Mr. Featherstone concerns the representation of troops. He states: "To refight any historical battle . . . the troops (must) accurately represent the original forces." 11 He further advises that it is essential, "that the tactics and formations of the original battle are reproduced and that all troops are obliged to conform to their known standard methods of fighting". 12 Unlike the recreation of the terrain at Cannae, this guideline proved a little more difficult to follow. The reason: the aforementioned disagreement between sources with respect to the actual numbers involved on both sides. After much debate, a few rough sketches, and some basic math, I settled on a compromise between the reasoning of Lazenby, the table contained in WARFARE, and Lawford's narrative in THE CAVALRY. 13 Without going into the mundane details of my numerous drafts of force composition, the reader will see in the following orders of battle, that I may have erred on the smaller side, with respect to total numbers. For example, I have figured for only 14 legions (field strength - I believe) in the field (seven Roman and seven Latin / Allied), leaving two legions, plus sick and injured, in the two camps. For the army of Hannibal, I elected to make the Celtiberians the most numerous contingent. 14 For those gamers not familiar with the ARMATI rules, I opted to mount the Roman and Carthaginians (15 mm scale), according to "epic" unit sizes. It is one of three choices Mr. Conliffe allows when putting together armies for battles with this rules system. In the "epic" size then, a unit of Hastati - one of four types of infantry making up a legion, is comprised of 24 figures on a stand measuring 8.0 cm x 4.5 cm. As one might well imagine, it is a fairly solid-looking block of troops. Consider then, the veritable wall built when 14 legions (42 stands, not including Velites) are placed in one body on the table. Those gamers that are familiar with the ARMATI rules, it will be apparent that I have taken a liberty or three with the rules detailing command and control. The rationale behind such amendments shall be explained in a later section. DEPLOYMENT Another key aspect of this re-fight project was the proper disposition of units on the field of Cannae (read: my wargame table). Though not remarkable at first examination, this battle stands out as uncharacteristic for both Rome and Carthage as their armies were deployed in an unusual manner. (For the Romans, the deployment was actually more unfamiliar than unusual.) 15 Hannibal, after re-crossing the Aufidus, placed his heavy cavalry on his left flank, and the Numidian contingent on his right. The infantry line was also flanked, but by deep phalanxes of veteran African spear men. The center, comprised of Spanish, Celtiberians, and some Allied units, was spread thin and pushed forward so that it formed a kind of crescent-shaped line between the phalanxes of Africans. The light troops were dispersed all along the front of the main line. 16 Hannibal and his sub-generals were with their respective commands; the cavalry commanders were the only attached generals however. (Please refer to MAP I for dispositions of the Carthaginians) The Romans too, deployed with cavalry on either wing. But the mounted arm was of little concern to Varro or the other Roman commanders. Their center held the promise of victory, as Lazenby notes: "the Roman tactics were, evidently, to try and smash through Hannibal's line by sheer weight of numbers, and what had happened at the Trebbia must have given them some hope that their infantry could win the battle for them". 17 With a center comprised of approximately 14 legions, deployed in depth and very close order, such tactics seem reasonable. To portray this formation on the table top, the legions were formed into three masses or blocks. The first, on the right flank, was comprised of just two legions: one Roman and one Latin / Allied. In the center, a massive block of eight legions: four Roman flanked by four Latin / Allied. Finally, on the left flank, under personal command of the commanding Consul for the day, were four more legions: two Roman, flanked by two Latin / Allied. The cavalry wings and Velites completed the deployment. Unlike the Carthaginians, every Roman Consul (general / sub-general) was attached to his specific command. (MAP I) AMENDMENTS In the second paragraph of this report, I stated the overall purpose of this undertaking. With apologies as well as thanks to Mr. Conliffe, and just simple gratitude to Mr. Featherstone, I would like to take a couple of pages and review the amendments to the ARMATI rules that were felt necessary for this reconstruction. Again, the idea is to re-fight the battle not as an exercise in history, but as a miniature wargame; a wargame with rules dedicated to providing a reasonably realistic, historical representation of the battle on that fateful August day in 216 B.C. In other words, it would not be much of a wargame if the Romans did not have at least a fighting chance. The reader will please forgive the following departure from the report narrative. Sometimes rule amendments and / or additions can be dealt with more clearly if approached in an outline format. I would also like to apologize for the length of the notes to this particular section. The material explaining the rationale behind the amendments to Army Command and Control seemed more suited for notes as opposed to inclusion in the body. Anyhow. The narrative of this report will resume with the account of the re-fight of the Battle of Cannae. All Advanced Rules / Optional Rules applied for this reconstruction, with the exception of the "Evasion by Light Troops", listed in Section 16.0, page 39, of the Basic ARMATI Rules. A. Generals Name Ability Rating Key Unit Value Hannibal +2 3 Mago +1 1 Hasdrubal +1 2 Hanno +1 1 Varro n/a 2 Servilius +1 1 Paullus +1 1 All other rules for General figures / stands apply. 18 B. Sequence of Play Dicing for Move Option: Players will roll for movement initiative on the Left, Center, and Right areas of the table. (The Carthaginians will consider Mago's force to be part of the Center.) This rule amended to reflect the historical narrative of the battle. Therefore, the Romans will be at (-1) on either flank, but "enjoy" a (+1) advantage in the center of the field. C. Army Command and Control Rating: All rules apply concerning Army Command and Control Rating. However, special exception had to be made for this particular scenario. The reader should remember the purpose of the effort is not to discredit the ARMATI rules system(s), but to build upon an already strong foundation, adding the "decoration" required in order to more closely follow the history of the battle under consideration. 19 D. Veteran Units: The orders of battle have certain units designated as Veterans. These units are marked with a (VET) in the order of battle, and by a chit with the letter "V" on the table top. All rules concerning Veteran Units still apply. 20 E. Routed-Into Morale Checks: Non-Veteran Roman Legionnaires are broken by rout, if they roll a 1-2 on a 1d6.Non-Veteran Latin Legionnaires are broken by rout, if they roll a 1-3 on a 1d6. F. Combat versus Celtiberians: In melee against Celtiberian (WB) units, the Roman player must roll two (2) more than the Carthaginian player in order to reduce that unit by one (1) break point / strength point. If the Roman player rolls only one (1) higher than the Celtiberian (WB) unit, then the result of the melee is a "push back" of two inches (2"). G. Melee Break-Off: Cavalry may attempt melee break-off when engaged with enemycavalry or infantry. Cavalry attempting such against enemy cavalry will subtract one (1) from their die roll. A commander may elect to break-off a unit, part of a division, or an entire division. The Break-Off Response die roll result, howeverwill be applied to that unit, part of a division, or entire division. Roman / Latin legionnaires are not allowed to break-off from melee(s). 21 Celtiberians will add one (1) to their die roll when attempting to break-off melee with Roman / Latin legionnaires. H. Legion Formation Melee Penalty: In the first turn of any melee, the Roman / Latin legionnaires will fight without formation penalty. In second and later turns, the die roll will be modified by (-1), to reflect the inability of the front rank troops to properly employ their weapons. 22 I. African Phalanx Wheel: If the opportunity does in fact arise during the course of the re-fight, then the stands of African infantry will "wheel" from their in-depthdeployment, to the left or right, and advance upon the enemy with that new frontage. A special marker or arrow should be placed by the phalanx to indicate its current facing. There is no reduction to FV for the phalanx after making this "wheel". 23 J. Breakthrough Movement: 1. Due to the unfamiliarity of the formation, legion units securing a melee victory will be followed on the breakthrough by all units following that particular unit. 2. Breakthrough Movement MAY result in another melee if a massed enemy unit resists the routed-into check. In this instance, the unit conducting the breakthrough movement will have impetus, and roll with a (+1) modifier in the ensuing melee round. Note: Melee to be resolved immediately, before continuing with melee direction. Roman Formation: Normally, the Roman legion of this period would deploy in four lines: the first, comprised of Velites; the second line, of Hastati; the third line, Principes; and the fourth line or reserve, comprised of Triarii. To reflect the unusual and unfamiliar formation Varro created, the Velites will still be ranged in front of the legionnaires, but the remaining three units comprising the legion will be deployed in depth. This, accepting the unevenness of the block of troops, dueto the number of Triarii. 24 At first look, these rule amendments may appear to tilt the balance of the wargame too far in the favor of the Carthaginian commander(s). The fact does remain, however, that the Carthaginians are outnumbered substantially in heavy infantry, and that their center is composed of a large contingent of "flighty" troops (Celtiberians). With the parameters put forward in this scenario, the burden of the combat falls upon them, and upon Hannibal. Legionnaires, whether veteran or not, are tough to handle in close combat. As researcher of and participant in this re-fight of Cannae, I feel comfortable with the number and content of the rule amendments. The only sure way to determine if the balance is lacking, is to "fight" the battle. 25 Cannae: Wargame Refight NOTES 1 Ernle Bradford, Hannibal (New York: McGraw-Hill Book Company, 1981), 113. In his excellent work, Hannibal's War: A Military History of the Second Punic War, (Norman: University of Oklahoma Press, 1998), 81, Professor J. F. Lazenby provides this version, also taken from Plutarch (Fabius 15.2-3): "another thing which has escaped you notice Gisgo, is even more amazing - that although there are so many of them, there is not one among them called Gisgo." 2 Philip de Ste. Croix, ed., Warfare in the Classical World: An illustrated encyclopedia of weapons, warriors and warfare in the ancient civilisations of Greece and Rome (New York: St. Martin's Press, Inc., 1980), 120 - 123. 3 Here, I foreshadow the reference to Mr. Donald Featherstone's work, Battle Notes for Wargamers (New York: Drake Publishers, Inc., 1974), 9-23. 4 For a brief moment, I considered utilizing the De Bellis Multitudinis rules by Mr. Barker and Mr. Scott. My mind was changed when the question of scale was reviewed. In the DBM rules, a stand or element represents roughly 256 men in a four-man deep block. At this "figure" scale, the representation of a single Roman Legion would require approximately 19 stands / elements. Representing the infantry field force at Cannae then - some 14 legions - would require 266 stands / elements of miniatures. In contrast, the ARMATI rules allow for a larger number of troops to be represented by a single stand or element. The range of scale for unit strengths in ARMATI, runs from 400 to 1500 men / men and horses. The Roman Legion then, becomes much more manageable at just four stands / elements. This is very convenient, as it provides one stand / element for each troop type within the legion: Velites, Hastati, Principes, and Triarii. 5 Livy, "The History of Rome, Book XXII", in Men at War: The Best War Stories of All Time, ed. Ernest Hemingway (New York: Bramhall House, 1979), 712-717, provides a simple and direct narration of the day's events. Scholars and historians of this century use Livy's account, and the more extant material of Polybius, for their narratives and interpretation of the battle. The editors of Warfare, Ernle Bradford in Hannibal, and James Lawford, "Origins", in The Cavalry, ed. James Lawford (New York: Bobbs-Merrill Company, 1976), 38-41, are just a few examples. 6 Referencing these same texts listed above, one finds conflicting views on the actual placement of the battlefield, the deployment of the opposing forces, the numbers of the opposing forces, and the composition of same, as well as argument over the military genius of Hannibal with respect to the evolution of the course of the battle. 7 Lazenby, Hannibal's War, 75-86. 8 Featherstone, Battle Notes, 10. 9 Lazenby, 79, and Map Number 9 in MAPS appendix. 10 Another example of the "fun frustration" with respect to sources providing conflicting information / evidence, is the location of the town / grain depot of Cannae. Lazenby places Cannae very near the river bank, well behind the Carthaginian deployment. In Warfare, however, Cannae is found on the ridge lines, in the left-rear of the Roman deployment. Lawford, in "Origins", places Cannae on the right side of the river, but on a plain apparently, and puts the Carthaginian main camp right next to the town. 11 Featherstone, 11. 12 Ibid., 11. 13 Lazenby reviews the numbers and sources for same on pages 79, 80, and 81. The table from the encyclopedia is attractive to the wargamer, as it gives a breakdown of the armies by element and type (120) - though this is directly opposed by Lazenby. Lawford's narrative is most interesting, as he constructs a very convincing mathematical argument, based on his estimations of army strengths, that the Romans would have engaged the African phalanxes at the onset, and possible even the cavalry wings of the Carthaginian host. (39-41) 14 Lazenby, 81. 15 Lazenby, 80. Compare the close-order of this formation with the usual disposition of Roman and/or Allied Legions at this time: Warfare, The Roman Army (3rd century B.C.), 110- 112. 16 Ibid., 81-82. It is interesting to speculate on the intelligence of the Roman Consuls in command, and on the "unwritten" etiquette of ancient warfare as exemplified by this battle. First, the Roman command: Why Varro did not recognize the complete imbalance between his Roman Nobles and the Spanish Heavy Cavalry is. For the arm-chair general, rather mind-boggling. Second: one is made to pause and wonder why Hannibal was not harassed or flat-out attacked as he brought his army back across the Aufidus to deploy against the waiting Romans. Surely, his superior numbers of horse would have been negated by the water obstacle. The infantry, I am making an educated guess, would have suffered even more at the hands of the ranging and self-supporting maniples of legionnaires. 17 Ibid., 78. This provides a possible answer to the "stubbornness" of the Roman command. Still, it cannot be denied that the battle at Trebbia was a resounding defeat for the Romans. 18 A review of the orders of battle shows that each main segment of the opposing armies is under the command of a General or sub-General. In order to add a little flavor to this re-fight, I decided that each commander should have an assigned ability rating, as well as a value in terms of Key Units. For the Romans then, Varro was given a "zero" ability rating, so he would not modify any melee rolls for units he attached himself to. As he was the Consul in command on this day of battle, I decided that he would be worth two more points toward the Roman Army Break Point total. Servilius was apparently more of a soldier and had more experience, so his sub-general stand had a value of "one" for both ability and key unit worth. The same value was assigned to Paullus, commander of the Roman right wing. Hannibal, being the pivotal figure in all of this, was given an ability rating of "two", and a key unit value of "three". This may seem a bit much, but from what reading I have been able to complete about the battle and about the 2nd Punic War, Hannibal was indeed the driving force behind the Carthaginian effort. For the Carthaginian sub-generals, I simply used the standard rules for generals, listed in the Basic ARMATI rules booklet. 19 The orders of battle for this wargame depict forces well in excess of those allowed for pick-up games in the Advanced ARMATI rules. A "typical" later Carthaginian army under the Advanced rules would total some 57 points. The force total for this re-creation, amounts to a staggering 164 points. This is nearly three times the total for a core force plus allowed bonus points / units. The total for the Roman host on this particular field is even more astounding: 246 points. This, compared to a point total of 58 for core force and bonus points / units. For players who wish to increase the size of their tabletop forces, Mr. Conliffe advises a similar increase in Control factors and Break Point total. Following this guideline, would produce a Carthaginian army of (possibly) 15 Heavy Divisions, nine (9) Light Divisions, with a Break Point total of six (6). The initiative rating for this army would remain unchanged. The Roman army, then, would consist of (possibly) 20 Heavy Divisions, 12 Light Divisions, and have a Break Point value of eight (8). As with the Carthaginians, the initiative rating for this army would remain unchanged; a value of four (4). Part of the research for this battle involved a consultation - via correspondence - with the author of the ARMATI booklets. Mr. Conliffe was gracious enough to reply to the inquiry, and suggested that I "go by feel" on the issue of command and control. In essence, each army is (was) sub-divided into smaller armies, each under a separate command. This was done not to undermine the authority of the main commanders - Varro and Hannibal - but to reflect, historically, the division of command within the actual opposing forces. Both the Carthaginians and Romans then, come in well below the number of Divisions allowed for such large forces. The decision with respect to Break Point totals for the armies involved as much, if not more thought. As the collapse and rout of the Roman Knights and Allied cavalry was integral to the eventual massacre of the legions, it seemed reasonable at first, to suggest that the elimination of these key units, or a majority of them, result in victory for the Carthaginians. Given the imbalance in cavalry between the Roman right flank and Carthaginian left however, this line of thinking seemed rather unfair. The infantry contest was also central to the eventual Roman catastrophe, not in terms of elimination of Roman units (stands) per se, but in terms of how deeply they are drawn into the center of Hannibal's' position. And yet, one cannot deny the adverse effect the unfamiliar formation must have had upon the legion infantry, so . . . well, one could well get a headache from considering all these options. In the end, a compromise was reached. It was determined that the Roman Break Point total would be 10: five (5) points coming from the key unit of cavalry and five (5) from key units of infantry. This total was inclusive of any command figures captured or killed during the course of the re-fight. The Break Point total for the Cathaginian army was set at seven (7). This is one more Break Point than allowed for a force just over three times the size of a core force with additional units. The main rationale behind this total was the number of Celtiberian units in the Carthaginian line, and their generally poor historical performance (in gaming too) against the legions of Rome. 20 To be sure, this results in a preponderance of veteran units on the board. The Romans are "given" four veteran legions (three Roman / Citizen and one Latin / Allied), and these are placed in the center of the main line. As Lazenby notes, ". . . the centre of the Roman line would have been filled by the veterans who had fought against Hannibal in Apulia and Campania . . ." (80). The number of veteran units in Hannibal's army is the product of a long march to and through the Alps, of almost two years in Italy before Cannae, and, of victory in two major battles before the morning of Cannae. 21 This rule imposed due to the deep formation of the Roman line. It seems reasonable to conclude that the first few ranks were heavily engaged in the fighting, and that the subsequent ranks may have interfered with the ability of the first few in their eagerness to get into combat with the enemy. See Keegan's assessment of massed combat in "Agincourt", The Face of Battle (New York: Random House, Inc., 1977), 94-107, and 112-116. Though Sir Keegan is dealing with medieval battle, centuries after Cannae, some of the same parameters hold true. 22 As above. 23 Lazenby, 83. 24 For both Roman and Allied legions, the Triarii complement was calculated to 600 men. This was half the number of Hastati, Principes, and for the Roman legions, roughly half the number of Velites. While Mr. Conliffe does suggest that Triarii units be represented by two-sections, thereby equaling the frontage of a regular FT or HI unit, I decided to maintain the "scale", and keep the Triarii elements at just one-section. This did make for a rather odd-looking legion formation - something like a square with a smaller square at its back or top. 25 The reader may or may not be interested to know that in the first attempt at this reconstruction, before the addition of amendments, the Roman Cavalry wings held up very well against attacks by Spanish HC, and the fight in the center was over very quickly . . . in clear favor of the Romans. Back to Table of Contents -- Courier #79 To Courier List of Issues To MagWeb Master Magazine List © Copyright 2000 by The Courier Publishing Company. This article appears in MagWeb (Magazine Web) on the Internet World Wide Web. Other military history articles and gaming articles are available at http://www.magweb.com