Tag Archives: Ligny

Understanding Ligny, 1815

Ferraris map of the Ligny field in the 1770s

One of my current projects is the battle of Ligny on 16 June 1815, one of the precursors to Waterloo, which was two days later. The usual script is that in this battle Napoleon comprehensively beat the Prussians under Blücher, and with a bit of better luck would have destroyed them, so winning the whole campaign. But, as I said my earlier article, a number of mysteries surround the battle, of which the most important was how the Prussians lost when they had such a strong numerical advantage.

But researching the battle faces some tough challenges. The biggest is the lack of French sources. It is far from clear when particular formations were committed and exactly where. Prussian sources are plentiful – I think they had a system of after-action reports – but they are subject to the usual systemic biases that will under-report poor performance. One particular frustration is that I can’t find any detailed casualty figures below corps level (and not even that for the Prussians), even though some authors make reference to these. Casualty figures are an important source of evidence, as noted by the French 19th Century writer Andant du Picq, as they are less subject to fakery (if you ignore after battle bulletins). If any reader knows where this data can be found, please get in touch.

My initial focus is on what I am calling “small Ligny” – the battle for the village of Ligny itself, and the chain of villages and hamlets from St Armand to Wagnelée to the west, involving the Prussian I and II Korps, and French III and IV Corps (minus Hulot’s division) plus Girard’s division from II Corps, from about 3pm to 7pm, when French Guard formations started to get involved. This will lead on to some wargames with my army-level rules. I am also intrigued by the possibility of some divisional level games.

For my first stage of research, I have used work by Dutch historian Pierre de Wit, which unfortunately does not seem to be available as pdf downloads online any more. This is dense stuff but closer to the primary sources than anything else I have seen in English (or French come to that – I can just about access text in that language). The main information I have sought from this is which units were committed and when, in terms of each hour of the battle, corresponding to game moves. My objective is both to understand the battle and to calibrate my rules.

Ligny village

“Small Ligny” organises itself into two main zones – Ligny itself to the south and St Armand to Wagnelée to the west. Let’s take Ligny first. This is quite a compact village, either side of the Ligne brook. On the south bank to the west is Ligny chateau, which was capable of being defended all-round by a garrison of a battalion in strength. There are a number of other substantial farms and a church with yard which became focal points of the battle. The brook was a significant obstacle, with one stone bridge at the eastern end of the village, and a couple of other less substantial crossing points. To the west of the village was relatively clear terrain, on a hillside, which is where the Prussians located a grand battery. To the east was a sunken road, orchards and so on, on the way to Sombreffe, which was clearly difficult terrain, and which does not seem to have been seriously contested.

The village was initially garrisoned by one small Prussian brigade (but remember that Prussian brigades equal other countries’ divisions) of six (large) battalions, of which two were initially held in reserve. One of these occupied the chateau, and held it until after 6pm, when exhaustion and ammunition loss forced a retreat. Over the course of the battle the Prussians fed in 14 more battalions, meaning that some 16,000 men were committed. The French committed just two divisions, in 18 smaller battalions, amounting to about 9,000 men. The outcome can be called a draw: the Prussians still held the village, or most of it, but were exhausted. When Napoleon committed the Guard and heavy cavalry they did not resist – the occupants pulled back to the next line of defence.

This is a very striking achievement, and goes some way to explaining the Prussian defeat. Just what happened here? This is important not just to understanding the battle, but also how to simulate battles on the tabletop. The direct sources tell us little. We are left to speculate, or hypothesise, using circumstantial evidence.

The first point of interest is that while the Prussians committed many more men, the number of battalions on each side was roughly similar. I have seen arguments among wargamers who suggest that when doing tabletop simulations the number of battalions is more important than the number of men – and that using standard units to represent each battalion, regardless of size, gives a fair representation. This view, which I have aways found suspect, is given support by this episode. But interestingly, this battle was not fought by coherent battalion formations, as battalions broke up into smaller tactical units. Still the battalion statistic does point to another factor – the French ratio of cadres (officers and NCOs) was probably higher, though I don’t have statistics on officer numbers for the Prussians. Battalions vary more in size of rank and file than they do in cadres – which tend to be dictated by the internal structure (number of companies, etc). And its clear that the Prussian officer corps was stretched by the fact that they tended to use more junior ranks to handle similar sized formations to the French in 1815 (while the British tended to field more senior ones). The ratio of cadres in regular formations (it’s a different matter in elite units) might make a significant difference to battlefield performance, and in particular to stamina – how long they could keep going when sustaining casualties.

And stamina is the critical issue here. The Prussians kept on having to feed in fresh troops to keep the battle going, while the French could recycle theirs. My guess is that the French the French used one division at the start (3pm), and withdrew it after about an hour, replacing it with the second. Which was in turn replaced by the rallied first division an hour or so after that. So the French are getting two bites of the cherry for each of their units, to the Prussians’ one. So far as I can tell casualties, in dead and injured, were roughly similar on both sides (about 3,000 for this part of the battle). The tactical situation may also have made it easier for the French to pull back and refresh units (replenishing ammunition in particular). To pull back the Prussians either had to go up the hill behind the village, exposed to French artillery fire, or along the road to Sombreffe, and into III Korps’s zone. They do not appear to have done it until the original brigade (Henckel’s) was withdrawn after 6pm.

There is another factor when considering quality of troops and stamina. Nine of the 20 Prussian battalions were landwehr (7,000-8,000 men), and all them from the Elbe and Westfalia provinces. These had only been incorporated into Prussia in 1814, and these units hadn’t been forged in battle – and nor were they so inured into Prussian military tradition. By contrast the landwehr units in the Prussian III and IV Korps were from the established territories of Kürmark, Silesia and Pomerania, which had been part of the great battles of 1813. It is estimated that the Prussians lost 8-10,000 men as deserters at and after the battle. It is thought that the bulk of these were from these landwehr units – though some did come from the more recently raised regular units, like the 25th Infantry regiment. There were 24 landwehr battalions in the two corps; 9 were engaged in Ligny, and 11 (3 after 7pm) in the west. So a very large chunk of the deserters must have come from the landwehr units in the Ligny battle. And that means that half or more of these troops must have fled. I have not found direct corroboration of this. There is a mention in de Wit of the landwehr troops wobbling a bit early in the battle and having to be rallied. But there were no mass routs. But Henckel’s brigade, with three landwehr battalions out of six, lost half its men in the battle according to one historian; it started with 5,000 men; if we say they took 1,000 dead and injured (being the battle the longest, casualties would have been higher than average), it means that 1,500 of the 2,500 landwehr deserted.

What happened? My guess is that at first the landwehr units would have engaged with reasonable effectiveness, but as the battle wore on the feeling among the men that they had done their part, and the imperative to survive and return to their homes, started to dominate, and they found ways of lying low. Substantial numbers may have been able to drift the rear areas. Once the Prussian army started to pull out in the fading light, these men flooded out along the road from Sombreffe to Namur. What this boils down to a very low stamina level in these units. If we try some sort of quantitative evaluation in wargames terms, we might class the French troops as “veterans”, the Prussian regulars, with their weaker cadres, as “trained” and the newly-raised landwehr as “raw”. If we weight veterans as one third more than trained, and raw one third less, we get a weighting for the French of 12,000, and the Prussians of 13,600. This is clearly much closer, and allows other factors, like stronger French artillery, to be brought into account.

Interestingly, not this analysis shows that the Prussians did not derive a great deal of benefit from being on the defensive, in a garrisoned village. According to French accounts, the first French assaults were beaten off with heavy losses, but they then managed to gain and exploit a foothold. This goes against most wargames rules. I have set up two or three games of the Blücher system based on Ligny, and it was hopeless for the French. Attacking the village was battering their heads against against a brick wall, and they soon ran out of infantry. Another interesting wargames point from this part of the battle is that the French artillery were able to rake the slopes behind the village. This almost certainly this involved a degree of overhead firing – and this was tactically important. Most rules systems allow this, but some (like Blücher) don’t.

St Armand to Wagnelée

This was a bigger and more complicated battle. The contested area was three villages along the line of the Ligne brook. To the south was St. Armand, apparently quite an open village, though with a substantial church and yard. Next to the north, with very little gap, came Longpré, where most of the fighting took place. This included two substantial chateaux – La Haye, heavily contested, and l’Escaille to the east, which the French never reached. Most historians of the battle call this village “St Armand la Haye”, but both current maps and the Ferraris map from before the battle call it Longpré. I think using this name is better for clarity. Next north after a small gap is Wagnelée. There was no serious attack on this village that I can see, but it was an important access point to the battle for the Prussian troops. Between Longpré and Wagnelée, at a crossroads, there was a hamlet of just few houses and an inn, which historians usually call “St Armand le Hameau”, but which is more correctly called Beurre-sans-Croûte. Historians generally refer to the whole area as St Armand, but this can lead to confusion.

St Armand, which wasn’t substantially garrisoned, was the subject of the first French attack, which was initially beaten back by Prussian forces waiting outside, but after the first hour it was occupied by the French and not seriously contested – but possibly after a second French division ws committed. Both sides concentrated their efforts after this in and around Longpré. The Prussians mounted attacks from Wagnelée into the open ground behind Longpré, leading to some open battles including cavalry support. The initial Prussian garrison (in Longpré) was just three battalions, plus some jager companies. But as the fight developed they committed some 29 more in the period I am looking at (and 6 more after that), giving 32 battalions or about 27,000 men. The French committed four divisions, and about 20,000 men in 39 battalions. At the end of this the Prussians had clearly won, and the French forces were close to collapse, forcing Napoleon to commit the Young Guard and most of the Chasseurs of the Old/Middle Guard to this sector. However the Prussians did not achieve what they had clearly hoped for: a breakthrough that would threaten Napoleon’s left flank.

We don’t have quite the same puzzle here Ligny village. The Prussians used fewer landwehr units (just 6 battalions); they also cycled their troops to refresh ammunition. They were in fact on the offensive for most of the time, unlike Ligny. Using the same weighting formula as for the Ligny analysis gives the French 26,700 men to the Prussians 25,300. Given that the Prussians ended up on top, it shows a better relative performance by them. That doesn’t seem to be because of better leadership than at Ligny, though. There are two cases of substantial Prussian attacks mis-firing and being defeated through poor coordination. There seems to have been no leadership at corps level, with brigade leadership undermined by the ad-hoc partial commitment of formations. Instead, the French leadership seems to have not to have been of the same standard as that for Gérard’s IV Corps. The III Corps commander, Vandamme, was very experienced but never made it to Marshal; there are numerous cases of questionable judgement across his long career. And one division, Girard’s, which led the attack on Longpré, was not under his direct command. This formation was over-committed and effectively destroyed, with one regiment fleeing in rout. Meanwhile Habert’s division was (arguably) under-committed, though their participation is not clear (it is known that the Swiss battalion that was in this division was not used) – but Vandamme’s orders may well have been unclear. Also the second division from III Corps (Berthézène’s) sees to have been very early to the fight, in contrast to Ligny. All this suggests a less measured management of resources by Vandamme.


It will be interesting to see how my rules work when I try this scenario out. This exercise will doubtless pose further questions. I have not paid so much attention to the artillery for example – but this is best done once the lie of the land is clearer, and that means modelling this on the tabletop. It is amazing how often historians fail to understand how terrain limited the use of artillery in particular battles (for example how hard it would have been to use artillery to reduce the British strongholds of la Haye-Sainte and Hougoumont at Waterloo). This phase of analysis does point to some places where the rules need a review. For example rallying can’t be done close to the enemy – which create problems for units defending terrain, like the Prussians at Ligny and (perhaps) Girard’s division at Longpré. Also how to feed in fresh units into an undity battle for a built-up area, and the role of strong-points – when to represent and when to abstract away.

A further thought concerns lower-level rules, which use battalions as their principal unit. This battle should be a fertile source of scenarios at this level. But it isn’t because rules tend to deal with built-up areas in far too abstract a fashion, usually giving too much benefit to the occupier. Lasalle 2, my go-to rules, would be hopeless. To get the proper feel of the battle you need to represent the structure of the villages – the streets, farms and churchyards and so on – rather than using undifferentiated terrain areas. It also probably means giving a role for company-sized formations. This is a problem that I might try giving some thought to. One episode, though, the attack by Tippelskirch’s brigade on the French flank, which included cavalry support, has the makings of a good game at this level though. What adds to the attraction is that I have actually made a representation of this formation, with four-base battalions, the core of my Prussian army collection.

Grouchy’s Waterloo by Andrew Field: Ligny

I’ve had this book for some years, and I’ve grazed from it, especially its account of Ligny. But recently I reread it in its entirety, and that serves to help me refocus on Ligny. I last studied this battle in detail in 2018, producing this article, in which I voiced my frustration with English language historians. That included this book.

This book, of course, is only tangentially about Waterloo – but you need to get the W-word into the title for it to sell, especially if the N-word or the other W-word (i.e. Wellington) doesn’t work. There is a discussion of why Grouchy never made it to Waterloo, which I suppose is link enough.

What this is actually about is the right flank of Napoleon’s assault on Belgium in 1815, and Grouchy’s role in it in particular. As with his other works on this campaign, Mr Field’s brief is to present French sources, which are typically under-represented in English language writing. That gives it a lot of value – but one of my main frustrations with most military history is that it is written predominantly from one side’s sources, and from their point of view. For all that, Mr Field does mention Prussian sources where relevant. My main frustration with the work is the one I raised in my earlier post: he “lets the sources do the talking”. He often uses the quotations as a substitute for his own narrative, and he rarely tries to pull the accounts apart to throw light on what is likely to be inaccurate or belong to another episode in the battle.

The main attraction of this book is its accounts of the battles of Ligny and Wavre, alongside the combat at Gilly. For all the one-sidedness of the sources, the accounts are as clear as any I have read, and his overall judgements seem sound enough. It is notable that he emphasises that the Prussian army wasn’t crushed, as breathless French-sourced accounts tend to suggest. He does indulge in critiques of the various commanders’ decisions though. Most historians do this, though I prefer a little more of AJP Taylor’s “What happened and why” – still it does help to understand that there may have been better choices.

Ligny is my main focus. One of my planned focus points for 2023 (I hesitate to call it a New Year resolution) is this battle, moving towards reconstructing it on the tabletop using my newly developed rules. Alas probably solo. I had allowed myself to be a little diverted by Waterloo, and especially the Prussian role there. I am left with a number of puzzles about this battle:

  • Why did Wellington, on his visit to the site before the battle, suggest that Blücher had deployed his men on forward slopes exposed to French artillery, drawing the riposte that Blücher “liked his men to see their enemy”. In fact the Prussians were so well concealed that Napoleon was confused as how many corps he was facing. Prussian reserves did have to cross a forward slope to reach Ligny, and did suffer – but their initial deployment was out of sight. Mr Field does not mention this. So much of the British reverse slope mythology is built on this episode – and yet I can’t believe that Wellington completely fabricated the story (though there are no corroborating witnesses).
  • Where did Vandamme’s corps start the battle? Almost all maps of the battle show it to the west of the St Armand complex, to the right of Girard’s division of Reille’s corps. That leaves a large gap between it and Gérard’s corps. Mr Field remarks on this gap but moves on. I think it is likely that the corps was in fact to the right of Gérard and astride the road out of Fleurus towards St Armand. This is what the language of the French sources suggests, and it also explains how these units came under artillery fire as they approached, as the first-hand accounts suggest, though these aren’t entirely reliable. An early Prussian map shows the corps in the westerly position and I think that most people have simply followed this.
  • How did the Prussians lose? The big question. They had more men and a decent defensive position. They just seemed to burn through their men more quickly in the two built-up areas. But why? Reinforcements had to expose themselves to French artillery, but that hardly seems enough to account for this. Mr Field does not address this question directly, but his assessment of the casualties on both sides does throw some light on it. They were similar, once you take out the desertions from some Prussian units. That suggests that the casualties in the street fighting were roughly similar – but that the stamina of the French was much better (a bit like the success of the British infantry at Albuera). I think the presence of so many recently established landwehr units (the Westfalian and Elbe units were formed only in 1814) may account for this. It also suggests that the advantage to defenders of built-up areas, a prominent feature of Blucher rules among many others, needs to be rethought.
  • What happened to the Prussian units that fought in Ligny? Reading the account of the battle, you would think that the Prussian units fed into the village disappeared, but this is clearly nonsense, as they turn up later, but depleted, at Waterloo. Mr Field adds something rather interesting. At the point of the famous Guard attack on the village, there seemed to be few Prussians actually in occupation – accounting for its rapid success in taking the village. This suggests that the Prussian command had been pulling out exhausted units, and had effectively abandoned the village by the time the Guard struck, leaving the defence principally to the cavalry as the French came out of the village.
  • Could d’Erlon’s intervention have been decisive? The standard French account of any lost battle is that but for one missed opportunity the battle would have been a triumph. In this case d’Erlon’s failure to arrive with his whole corps, and the hesitancy of the troops that did make it, was the missed opportunity. They could have cut off the Prussian retreat and helped nearly annihilate the Prussian army. The first problem with this is that they did not turn up where Napoleon had intended them to – on the road from Quatre Bras. This would have taken them right into the Prussian rear. But this was never a possibility (and Blucher would not have stood at Ligny, or not in the deployment he did, had this been a realistic possibility). Wellington’s army was in the way – a fact that Napoleon had no understanding of. The second problem is that he turned up pretty late, at about 6pm. As Mr Field points out, the Prussians hadn’t collapsed at this point and (though he doesn’t say this), may already have been contemplating withdrawal. D’Erlon may simply have hastened the Prussian withdrawal, rather than annihilating the army.

These are the questions that I hope my efforts will throw some light on!

Marshal Ney at Quatre Bras – a huge disappointment

In a recent post I complained at the poor quality of analysis in military history writing for the Napoleonic wars. That was when I was doing some earlier research into the Battle of Ligny on 16 June 1815. I have just bought a new book on the simultaneous battle of Quatre Bras: Marshal Ney at Quatre Bras by Paul Dawson, published in 2017 by Frontline Books. It is enough to drive me to despair.

There is real tragedy here. Mr Dawson is one of the few authors that does original research, and tries to bring fresh evidence into the picture. In this case it is muster returns for the French Army, which are in the French archives and have been untouched by historians to date. And if you look hard you will find the occasional fresh insight based on his voracious appetite for documentary evidence. And this easily might have been a much better book. What it needed was a robust dialogue between the author and an experienced editor to produce a book that would have been much clearer and easier to read. That is not how modern publishing works, however.

The basic style of the book the book is to present piles of evidence – eye witness accounts and (separately) data from the muster rolls – though not other evidence from maps and the ground itself. He gives primacy to the French sources, but Netherlands sources get their due, as do British sources too – though since my interest in the well-worn British sources is not great, I cannot give a view as to its completeness on that account. Like many modern authors, he seems to take the view that these accounts should speak for themselves, and he adds very little interpretation and challenge. This is a bit odd since in the introduction he explains how careful one has to be in interpreting these reports. The exception to this are the various communications between Napoleon, his chief of staff Soult, and the various field commanders, and especially Ney. Like pretty much every author covering Quatre Bras, he takes a great deal of interest in these. Which is fair enough, since these are central to the two main controversies which have raged from the day of the battle to the present. First is why Ney to did not press his attack much sooner in the day. And the second  concerns the movements of d’Erlon’s I Corps, which was diverted from Quatre Bras to Ligny and back again, managing not to play more than a minor role in either battle.

Mr Dawson ends up by saying very little new on either controversy. His writes history on the principle that it should allocate blame, and he heaps it on both Ney and d’Erlon. He gets quite outraged at times in condemning the laxity or perfidy of the various actors. I find this jarring in a modern author, though he is not unique (Digby Smith does this too). I prefer the great historian AJP Taylor on this: history should be about explaining what happened and why. Alas Mr Dawson is weak on both the what and the why. Which is a bit of a pity, because his book provides an all-too-brief peak at some rather interesting lines of enquiry on both controversies. The first is whether it actually made much military sense for Ney to press ahead with an early attack, given his relatively weak forces and lack of knowledge of the enemy dispositions. It could have led to the annihilation of II Corps for no real gain. And without II Corps, could Napoleon have taken on Wellington, even if he had been able to take the Prussians completely out of the picture? A second point presented by Mr Dawson but not followed through is that I Corps started the day very scattered (over 20 miles, he claims), and that the staff of its leading division, Durutte’s, had defected the previous night. However, when describing the corps’ movements later in the day (from about 5pm), he presents it as largely concentrated and ready to take part as a whole in either of the day’s battles. 20 miles is a long way to walk in a single day in full kit – so just how quickly the four divisions were able to concentrate near the field of battle looks like an interesting line of enquiry. If only one or two divisions were in practice available before dusk, then that would be an interesting new perspective. And if it the conventional account is in fact correct, why does he lay on the scattered nature of the d’Erlon’s corps with a trowel earlier in the book? The book is full of such inconsistencies.

While on the subject of d’Erlon and Ligny I found a couple of other points irritating. When the leading elements of the corps approached the battlefield, it threw Napoleon and his staff completely because they were approaching they were approaching from the south rather than the west. To such an extent that he halted the last phase of his attack to try an find out who the troops were because he did not think they could be d’Erlon. Mr Dawson completely fails to mention this, even though he has done a good job of laying the groundwork in explaining why Napoleon would have been so surprised. He was expecting them to come from Quatre Bras, because he did not realise that Ney had held back his attack, and that Wellington’s forces had arrived there in strength. This meant that d’Erlon’s corps was not in the right place to attack the Prussian rear. However Mr Dawson then goes on to accept at face value the claim that if Durutte’s division had pressed its advance more vigorously, it would have been catastrophic for the Prussians and stopped them taking part in Waterloo. This is exactly the sort of thing that defeated French commanders always claim in order to say that they could have saved the day if only they had been allowed to. Durutte only had one division with a bit of cavalry support; the Prussians might have had enough troops to present a rearguard long enough to allow the failing light to complete their retreat; or Drutte mat simply have been in the wrong place too late. Ligny is beyond the scope of Mr Dawson’s book perhaps. In which case the right thing to say is that he cannot offer an opinion on the claim – something he is happy enough to do elsewhere in the book. He can’t quite get the balance between being an impartial presenter of evidence and the wish to get his opinions off his chest.

What of his account of the battle itself? This is the book’s biggest failure. On the big controversies he at least presents arguments, even if he is often repetitive and laboured. For the battle itself you get a very muddled account. On the famous charge of Kellerman’s cuirassiers, for example, at one point he suggests that it might have happened much earlier in the battle than it is often supposed. This is very strange, because he has spent much of the earlier narrative telling us that it was a miracle that the troops reached the battlefield at all, let alone two hours early. In the course of his account of Kellerman’s corps he does have interesting things to say, to be fair. He highlights worries over royalist loyalties amongst the officers of the Carabiniers. He also makes it clear that three of the four brigades reached the battlefield with the battle in progress, when normally historians say just one did.

But the biggest problem concerns his new evidence, the muster rolls. Mr Dawson extracts previously unpublished casualty figures for French units from these rolls. Alas there is clearly a problem with these figures. So, for example, his reported casualties of the 2nd Leger , which led the French attack, were just 31 with 3 killed. And yet Martinien lists one officer killed and 13 wounded, which suggests a much heavier toll. For the 108 Ligne, also Bachelu’s Brigade, and heavily engaged, even he can see the records are incomplete: It records 23 killed, 5 prisoners, but no wounded. He lamely says “we cannot give any further comment”; which does not stop him adding these incomplete figures in to his overall casualties for the corps. What about extrapolating the numbers of wounded from those killed? Or comparing with Martinien’s lists of officer casualties? Another example, which he makes much more of, is that the 8th and 11th Cuirassiers suffered just 49 casualties between them. But pretty much all eye-witnesses from both sides suggest a much heavier toll (Kellerman himself estimated 200); Martinien lists 17 officers killed and wounded. In many (but not all) cases the casualties reported by the muster rolls look far to low. This would have been quite an interesting point of discussion. But instead Mr Dawsontrea treats his new evidence at face value, as a gold standard. He suggests that only one infantry brigade of the six involved was seriously engaged, that two barely took part in the battle, and that the French suffered half the casualties of the Allies. Though it is not uncommon for eye witnesses to exaggerate casualties, this all looks a little steep.

Paul Dawson is a diligent researcher, who takes more trouble with compiling evidence than most current authors. Alas he seems bereft of the analytical skills needed to interpret it. This book may have some value as a secondary source for lazy historians like me. Other than that it is a waste of time, I am afraid. It is hard to read: long tracts of direct quotation, argument that is laboured and repetitive and yet often seems to miss obvious points. And a lot of his evidences, like his casualty figures, poses questions which he makes no attempt to answer. And you have to endure him sounding off his armchair criticism of people long dead as if they were contemporary politicians. This does not bode well for Mr Dawson’s much bigger work on Waterloo, just published, which I have also acquired. But I will give it a fair crack.

Ligny 1815. The failure of English language historians

I like to focus the development of my Napoleonic rules around specific battles. Waterloo and Quatre Bras are regulars. My most recent finished rules were developed to refight Vitoria. As my latest rules stagger towards the playtest phase, I am focusing on the battle of Ligny on 16 June 1815.

This battle was fought on the same day as Quatre Bras as Napoleon attacked three out of the four corps of Blücher’s Prussian army. In spite of being outnumbered for most of the day, he scored a remarkable but costly victory – which might have been enough to win him the campaign, had not the Prussians withdrawn towards their British-led allies, instead of on their line of communication. In spite of its importance, and a scale that matches Waterloo, this last victory of Napoleon is rather neglected by historians. But it is an interesting battle nevertheless. At the heart of it is a mystery: how did the French perform so well against a numerically superior enemy?

I have four English language books focusing substantially on the battle. There is Andrew Uffindel’s The Eagle’s Last Triumph, my edition published in 1994, though I think there is a later version out there. Then there is Peter Hofschröer’s 1815, the Waterloo Campaign, the first in a two part account of the Waterloo campaign published 1998. Next is John Franklin’s Osprey: Waterloo 1815 volume 2, published 2015. Lastly, relatively fresh off the press, is Andrew Field’s Grouchy’s Waterloo, published 2017. Alas all these accounts are deeply flawed. There is a further important resource: Pierre de Wit’s website, The Waterloo Campaign.

I am at the beginning of my study of this battle, and I have focused mainly on is opening stages. But it is enough to  confirm the usual flaws in Napoleonic military history in the English language. Actually these flaws are almost certainly not confined to the Napoleonic era, and not tot he English language either; it just what I know. There are three problems: “lamppost syndrome”, a lack of forensic analysis, and poor maps.

By lamppost syndrome, I am referring to the story of the drunk found at night scrabbling around under a street lamp. “What are you doing?” he is asked; “Looking for my keys,” he replies. “Did you drop them there?”; “No, but I can see here”. Historical writing concentrates too much on where the evidence is, and especially if there are first hand accounts. And yet important things happen in places where these accounts do not exist, and you can’t understand what is going on until you try to work out what happened where the evidence is thin. The second problem I have called lack of forensic analysis. The word “forensic” really means associated with criminal justice, but it has now assumed a wider meaning that I am using here. And that is a careful piecing together of the witness evidence with other evidence, and an understanding of what is physically feasible. In the history of this era this  evidence includes the lie of the land, casualty figures, and an understanding of the technology and human capabilities. On top of this must come the persistent examination of motive: why did somebody do that? All this is familiar enough to viewers of television crime dramas or readers of detective fiction. Given the popularity of these genres, it is very surprising that so few modern military historians want to turn detective. Not all writers are as bad as each other. For an example of how it should be, there is  Rory Muir’s masterpiece on the battle of Salamanca, though even this lacks a decent map, a problem I will come to.

But what we usually get is mistakes by one author being repeated by the next uncritically, and very little in way of genuinely new perspectives. There is one interesting example in the early stages of Ligny: the defence of the village of St Armand, where the battle started. The normal story is that the village was defended by three battalions of Jagow’s brigade, and the French attack encountered bitter resistance. In fact a careful study of the evidence (as Mr de Wit makes clear) shows conclusively that these three battalions were in the nearby village of St Armand la Haye (or Longpré), which was more defensible and closer to the Prussian positions. And all the other Prussian infantry is accounted for. In fact the French faced artillery fire from the hill behind the village, and few skirmishers, and that was all. They quickly took the village, but found it impossible to move beyond it. The Prussians (Steinmetz’s brigade) then counterattacked.

So how do our four accounts handle the episode? Mr Uffindel gives us a lot of drama (“corpses littered the streets”) and sticks to the story that they had to fight hard to push out the unidentified Prussians. Overall his account is extremely thin on this stage of the battle. Mr Hofshröer, a German speaker, makes much more use of Prussian sources. He initially says that Jagow’s three battalions were in St Armand, he then goes on the say that the French found the village largely unoccupied, and then gives a very muddled account, including quotations from officers of those three battalions. At various points he suggests the fighting was in St Armand, St Armand la Haye, and then the neighbouring St Armand le Hameau (or Beurrre). Mr Hofshröer is a controversial author and in my view completely unable to tackle his subject forensically. His value is in his extensive quotation of German language sources. Mr Franklin is a more careful author, but tends to focus on French sources. He suggests that the French had to fight hard to capture the village, with the first brigade of Lefol’s division having to call in the second. Again there is extraneous detail (“the front files were decimated”) . Finally Mr Field: he is explicitly majoring on the French sources, and he likes to quote at length; there are two accounts of this episode: one from General Lefol, and one from Captain Gerbet of the 37 Ligne. This reveals the source of  a lot of the colourful detail of Messrs Uffindel and Franklin. After these extensive quotes Mr Field says that the village (though a strong position, he says) was lightly held and the French did not face determined resistance. But he makes no attempt to reconcile this with Gerbet’s account of a rather fierce struggle. I suspect it conflates episodes from later in the battle. Incidentally the statement that St Armand was a strong position is not my view, and that is one reason why the Prussians decided not to hold it strongly. As with other works, Mr Field is unwilling to pull apart his witness statements, but at least he is more transparent than other authors, and he is careful with his facts.

And the maps? Wargamers love a good map, from which they can create a decent table. 19th Century historians did too, but they generally compiled them without properly surveying the ground, and with only a schematic representation of relief. A simple matter, surely, to take modern contour maps, and use these as a basis for updated maps? Alas, far too often not. For Vitoria I had to do this for myself, and the two maps I did are now nearly top of the Google ratings for maps of the battle. All the offerings on Ligny are flawed.

Mr Uffindel doesn’t try. He illustrates his work with schematic diagrams that do not attempt to give a feel for the terrain. Mr Hofshroer gives us two detailed maps. One gives us a representation of relief, but taken from Ferraris map of the 1760s, and with no detail of the extent of the villages, and with a later highway missing; this is not actually all that helpful: what you need is a contour map. There is then a reproduction of a Prussian military map, with lots of detail, including the initial troop dispositions. But no contours. The Prussian dispositions look accurate to the battalion – but it shows Jagow’s three battalions in St Armand and not la Haye. This may be the source of the error in other accounts; perhaps these battalions went there first but were moved to somewhere less exposed. Hofshröer does have a proper modern map showing both Quatre Bras and Ligny, which is decent enough but lacks detail for the individual battles, though it illustrates what Napoleon intended with d’Erlon’s corps very well. Franklin has only Osprey’s 3-D maps, which promise more than they deliver. You can’t see the folds of the ground. He has Habert’s division of Vandamme’s corps in a different place to everybody else, interestingly enough. This is clearly wrong in my view. It is shown on the right of the corps, in a position that looks exposed to artillery fire; it came into action on the left. Of course it may have started out on the right.

And what of Mr de Wit? This is a very valuable resource, as he squeezes as much as he can from from the evidence. It is pretty heavy going, though. He can be quite forensic, but he suffers severely from lamppost syndrome. This is less a defect for Ligny, so far, than it is for the Prussian advance at Waterloo, which has big gaps. There are no maps. He does include some very nice surveys of the terrain, mentioning anything from the era that has survived, and including some old photographs of various features.

And so, like Vitoria, I am going to have to piece together my own account, and map. This will take a while, but it is a part of the hobby I love.