Book Review: Lost Battles by Philip Sabin

I came late to Lost Battles but have been so impressed by what I read that I wanted to post this review—even though it is completely ‘out of sequence’ with my planned activities (Fortunately my self-determined plan is entirely flexible!).

I became aware of the rules and book thanks to the excellent game reports and videos of Aaron Bell (Here’s No Great Matter blog)—his are the rare case of videos by a wargamer that are worth watching! :). In addition I had some correspondence with Tony Emery (Parade Ground 20 mm blog), who is part of an email group of 20 mm/1/72 scale devotees that I have come into by association. Tony provided that little extra bit of encouragement which convinced me to purchase the book.

Thank you Aaron and Tony.

The book was first published in 2007 and had an earlier incantation as Strategos (2003), which is still available as Strategos II from the Society of Ancients. Sabin has published widely regarding the use of wargames in tertiary education and, as far as I am aware, discussed and developed his ideas over several articles in ‘Slingshot’. Lost Battles represents the fourth version of his game-model of ancient warfare, with Phalanx and Legion preceding Strategos (StrategosII being simply a reprint and ‘no different from the previous edition’).

When the book arrived in mid-June, I had a couple of others on the go, related to my ‘scheduled’ themes for the year. I naturally flicked through it, reading sections of pages here and there. I knew immediately that I was going to find it most interesting and that it was going to ‘push in’ to the queue! I like the idea of reading a thoughtful, well-researched analysis of the battles of the ancient ‘era’ that most interest me (the ever-popular three to five hundred years from around 300 BC to about 100 AD—possibly stretched to 200 AD). It seemed to be right up my street, particularly as my interest in wargaming stems from historiography and developing a greater understanding/sense of a period. More so since the sections that I had read during my flick through suggested that Sabin writes in a most readable manner too.

My first impressions were not wrong, in fact they were amplified.

Sabin says his aims are “…to complement rather than replace the existing scholarly techniques [with a game system], each of which has contributed very significantly to our understanding of the engagements concerned. …to build on rather than to challenge…and if it can raise a few neglected questions and help a little in resolving some of the many outstanding controversies, it will have more than served its purpose” (p. 11). His starting point and focus on history is a critical part of the appeal for me. He notes later in the book that “…it should even be possible for a single user to reconstruct the battles, by the simple expedient of playing each side in turn. Such ‘solitaire playability’ is a key issue for conflict simulations, because their users tend to be interested less in competitive play than in exploring for themselves the dynamics of the historical conflict concerned” (p. 63). That is me to a tee. This does not prevent two-person or even multi-player games, in fact my impression from reading the book is that the system would be well suited to both of those, better even than it is for solo play.

The book opens with a discussion of sources and approaches to trying to discover more about the ‘lost battles’ of antiquity. Sabin relates N. Whatley’s five approaches or ‘aids’ as he terms them to assisting with determining the veracity of ancient sources; modern geography and topography, applying modern military wisdom, physical reconstruction [re-enactment], using selected fragments to make a coherent story and the generic study of armies and navies to determine methods and styles of fighting. He discusses each in turn, telling us that his approach uses much of the fifth, blended with the second and third. Sabin discusses in detail the problem that anyone who has read any of the ancient accounts is aware of: the wild variation in numbers between combatants (often with far larger forces on the losing side) and in losses. So often we are given the impression of an Aragorn-style affair where 10 000 defeated 100 000, killing over half of the enemy while only suffering ‘but a few scratches’! I exaggerate for literary effect, but you know what I mean.

“So what are we to make of these and other similar figures? … The real question is where and on what basis we should strike the balance between scepticism and credulity” (pp. 12–13). He tells us how Hans Delbrück, writing at the turn of the 20th century merely dismissed them as poppycock and proceeded to equate forces. Sabin suggests that this is more related to the nature of war at the time of Delbrück’s writing since equality of arms and training gave greater weight to the size of the forces involved (and generally produced comparable casualty lists on the two sides). He notes that more recent conflicts, particularly in our time, have demonstrated the ability of training, tactics and/or equipment to enable a far inferior force to prevail over a larger one “… if we reject the judgements of writers in the ancient world itself, we are hardly in a position to replace them with sounder judgements of our own after a passage of over 2 millennia” (p. 14). What to do? Philip Sabin’s approach is two-fold. He accepts as “…broadly accurate the numbers given for better-documented battles such as Cannae, Magnesia and Pharsalus…” (p. 14) then using the model thus developed to test against some of the battles with more widely disparate figures. Secondly he weights quality over quantity “troop quality was clearly much more significant than either numbers or casualties…” (p. 15) as well as the situation of the troops rather than the casualties they have suffered—the all important effects of ‘morale and élan’. He later adds leadership to these two. These, of course, are the crucial aspects that wargames rules have done a better or worse job of modelling since kriegspiel and HG Wells.

While he does not state it overtly, it is clear that Philip Sabin’s rules have been developed with two tenets in mind: historical accuracy and simplicity. An Ockham’s razor approach, if you will. So, in the next four chapters (Armies, Movement, Fighting and Command) he introduces the reader to the rules by discussing the background and rationale and the simplest, easiest solution that he has derived to represent it in his model of ancient warfare.

Sabin begins the chapter ‘Armies’ with a discussion of how to handle the concept of ‘units’, a particularly difficult idea to apply to anything but the Roman and possibly Greek armies. He comes to a simple, workable, pragmatic and entirely arbitrary decision to divide an army into a maximum of 20 units (based a lot on wargaming experience and how many separate divisions can be handled reasonably by a player). While this is like ‘plucking from air’ I suspect that he came to it after a lot of backwards and forwards thinking and calculating as to how it would generate armies and ‘units’ and, the mix as a whole to relate to historic representations. I’ll discuss this further later.

Sticking with Sabin’s system for now. Having generated a starting point he quantified it around the idea of fighting power and troop quality to create an equivalence of three broad troop qualities (levy, average and veteran) by number of men to generate a fighting value. By trial and error this established a base whereby an average infantry ‘unit’ represents 500 men, doubled that number for levy and halve for veterans. These values are scaled, using an army multiplier to produce ‘units’ of cavalry and infantry that represent from 250 to 8 000 men enabling different sized battles to be modelled while maintaining the size of the army at around 20 units.

In keeping with his Ockham’s razor-style approach, troop types are limited to infantry or cavalry, each with light and heavy classifications. Chariots and elephants round out the base troop types. A few special sub-categories are added to allow for particular types of troops that were prevalent and important in the warfare of the period, namely legionaries, hoplites, phalangites, archers (all being sub categories of heavy infantry), cataphracts (as a sub-category of heavy cavalry), scythed chariots (as a sub-category of chariots).
Light troops are restricted to skirmishers, so legions, hoplites, phalangites, warbands, immortals, formed archers and so forth are all heavy infantry. A table at the end of the chapter provides a clear summary of his preceding discussion.

A picture (in this case table) tells a thousand words.

A specific case that was of particular interest to me was to consider legionaries, for most aspects of the model, as either veteran or levy. This immediately brought to my mind trained legions and levy/penal legions. While Sabin introduced it as a means of equating strengths it seems to fit perfectly logically with training and reported performance of legionaries.

The chapter ‘Movement’ begins with determining the placement of armies on the battlefield and the terrain. Once he has established the size and representation of a battlefield, he then specified how troops may move around this area (in his case a 5 x 4 grid of 20 ‘zones’). An absolute piece of brilliance in this chapter is to included the importance of deployment for battle as part of the system. This is modelled very much in the manner in which it was done, a few units at a time and limited by space (and time). The inclusion of this aspect is a real attraction of Sabin’s rules to me as it relates to important reported examples that I have read such as the daily deployments before Pharsalus until Caesar was able to get Pompey to accept battle (or to be talked into it by his generals) and its supreme importance to the outcome at a battle such as Ilipa (to list but two).

Prior to reading the book, I got the impression that the combat in Lost Battles was highly simplified and stylised. It is both of these for sure, but I now have a far greater appreciation of the system and think that I’ll be a lot more comfortable with it than I had expected to be. Units that are in the same zone may attack enemy units up to the attack limit of three to five. The attack limit is a factor of the scenario that is determined by the frontage that each zone represents and the army multiplier (i.e the troop density in the zone) and so whether the troops can ‘get in’. The chance of scoring a hit is set for troop type versus troop type, with modifiers for first attack, attack bonus (a command option), a few specific troop characteristics and wet weather. While Philip Sabin notes the concept of the passage of the lines in having several units in an area (zone) of the battlefield making attacks, it also makes a lot of historic sense in terms of the mixing of troops, particularly in ‘barbarian’ armies, and of ‘piling in’.

Command is perhaps the greatest strength of the rules. This is fitting given their grand tactical level. Sabin uses the novel approach of taking a composite view of commanders in an army so that “the player’s input should represent this collective command capacity” ( p. 64). This is a most ingenious way of addressing the omnipotence of 100-foot general. Furthermore he states that “the main role of the command system should be to constrain the player…” (p. 64). This had more worried at first as I do not generally like systems of pips, prompts or command points. However, it seems to have been better thought through in Lost Battles than in many other sets of rules. The command capacity is determined by the size of the army and quality of the troops (calculated as fighting value of the army divided by ten), plus the value of any generals (with restrictions) and a die roll. It generally costs one command point to do most things, two for more advanced actions. Hence, it is entirely possible for an army to be ordered to advance in strength and to conduct attacks, but it requires better troops and/or commanders to move larger groups, have a double move or an attack bonus (attack with greater élan/purpose/direction or such). Generals provide command exemptions that may be used to activate units and may also initiate attack bonuses. Generals may also attempt to rally units (at personal risk). Sabin divides generals into commanders (not attached) and leaders (attached to a guard unit). Leaders may be lost if the guard unit withdraws, routs or is shattered. Brilliant generals (e.g. Hannibal) have the option to flip the order of turns and so ‘seize the initiative’ and move first rather than second (or vice versa).

Having read the first five chapters I appreciated the strategic level of Sabin’s system but could also see how tactical aspects are dealt with. I did not try to remember all the details of the ‘rules', instead reading these chapters in the same manner as I would read any set of wargame rules for the first time; to get sense of them. It seems to me that, while deliberately ‘high level’, abstract and grand tactical, Sabin has produced some quite brilliant mechanics which make his rules a far better historic representation than most I have seen and used. While Sabin’s breaking down of armies into ‘units’ may at first sight seem little different from any other set, the integration of the concept of ‘units’ with the battlefield grid and the number of units allowed to attack from a zone creates fronts or areas of an action. Hence it is not at all akin to having the ‘1/21st warband’ carrying out actions, but instead some skirmishing warriors attack, then the more ‘formed’ warband comes in, followed by others, perhaps switching ‘ranks’, piling in or coming around one or other side. While really abstracted, it seems to be a most reasonable representation. It is clearly one that works at the overall level of the narrative of a battle, but also in exploring, disposition, tactics and timing of events (as presented in subsequent chapters, plus use of the system by others).

Another aspect that impresses me is that the model is intended to be flexible. “The beauty of the model is that it is flexible enough to accommodate such changes [scholarly theories on army size, deployment and battlefield terrain], as the fighting value system tracks the worth of any combination of forces, the attack limit system allows for tweaks to the army frontage, and the handicap system [a way of equalising victory points] should offset all but the grossest asymmetries between the two sides” (p. 77).

Of course, these impressions come (largely) from just reading the rules, although watching Aaron Bell’s excellent videos of his refight of Trebia helped a lot to get a sense of the rules ahead of reading them.

The next step is to give them a go myself and I am really keen to do just that.

While reading the chapters about the model, I did not attempt to recount nor commit to memory all of the specific details of the ‘rules’. In chapter six, Sabin provides a playtest of Cannae—the first four turns at least—that takes the reader through applying the rules to create the scenario and playing it out. He then provides further examples of how to apply the model to historic actions with his discussion of 35 major battles (490 BC–48 BC). This is rounded out by the summary of the rules in an appendix.

Map from turn 2 of Sabin's example of a playtest of Cunnae

His historical accounts of the 35 major battles and discussion of how to apply the model to them is divided into eight chapters: Athens and Sparta (i.e. Greco-Persian and Peloponnesian wars), The Age of Xenophon, Alexander the Great, The Successors, Carthage and Rome, Hannibal and Scipio, Rome Moves East and Julius Caesar. In each chapter Sabin briefly describes the campaigns/wars and then, for the selected key battles discusses the available information regarding terrain, size and make-up of the armies and how his model might be applied to them. He ends with an order of battle for each and a map is included indicating possible starting positions of units if the deployment option is not used. He also discusses options that can be explored with the rules to examine and to test various theories, interpretations or assessments of each battle.

Reading this book out my sequence, i.e. before I am ready to begin a structured study of particular ancient campaigns, meant that I came to it having only read bits and pieces of ancient works, modern books and almost none of the academic literature. Fortunately this is not a huge problem as, while Sabin refers to numerous battles, authors and works ancient and modern, he is clear about the aspects that he is referring to and so one can follow his discussion without needing to have read the literature.

The maps in the book are clear enough, but are not reproduced to a high quality that I would expect (have liked).

Typical of academic publications, the book is not particularly cheap. The ‘budget’ paperback version that I purchased, printed on pretty basic paper and with its soft binding, still cost me around $55. The figures in the book are ‘okay’, but are a little blurry and all in black and white. For the same sort of outlay I have many, hardback books printed with top quality stock and with colour pictures. Fortunately, the below average ‘packaging’ is well and truly outweighed by the top quality content.

‘But wait, there is more’

The book is ‘worth it as a book’, as Tony had said to me! Lost Battles is like three books in one: an analysis of ancient warfare, a potted history of 35 major ancient battles from Marathon to Pharsalus and a set of wargame rules. Sabin’s discussion of sources, historical background and related military matters is most interesting and edifying and worthy of a book in its own right. Definitely good value.

Caesarian Romans ready to go

Further Caesarian Romans a bit of a way off.

Gauls barely out of the box!

Having read it and been enthused and stimulated by the content I am now keen as mustard to have a go at playing/simulating a battle or three! I plan to bring ancient games into my schedule in 2023, with the 2080th anniversary of the beginning of Caesar’s Gallic campaign, beginning with Bibracte (around June, if possible). I have a reasonable number of Romans painted, but no Gauls/Celts/Britons (or Germans). Perhaps I’ll be able to have some ready a bit sooner; such is the excited state that this book has left me in.

Rating

References

Sabin, P (2013) Lost Battles Reconstructing the Great Clashes of the Ancient World. First Published 2007. Bloomsbury Publishing, London. 298 pp.

Whatley, N (1964) On the Possibility of Reconstructing Marathon and Other Ancient Battles. The Journal of Hellenic Studies 84, 119-139.

Battle Reports Using Lost Battles

The links below are to reports of games using Lost Battles from three blogs.

Game reports on Here's no great matter

Aaron's video of Trebia

Game reports on Parade Ground 20 mm

Game reports on Ancient & Mediaeval Wargaming