Attrition

[Robert Dunlop]

Jon, obviously I have tried to quote from primary sources as much as possible. I do not regard my analysis as complete. Thank you for respecting such information as has been put together. Like you, I am interested in continuing to evaluate the evidence. It is very helpful that we come at this from different perspectives.

Your point about not getting Haig off the hook is very important. The fundamental issue remains. Irrespective of who planned what, the Somme was a very bloody series of battles. Was it possible to loose fewer men and still achieve the same outcome - ultimately the end of war with victory over the Central Powers? I would like to pick this point up again once I return home.

Similarly, I am happy to offer comments on Gommecourt, FWIIW. This will have to wait until I get back, which is later this week. I would prefer to stick to the one thread. These issues are all related in my opinion. Eventually, we may get onto the Third Ypres campaign as well. I resisted commenting on the recent Plumer vs Gough debate because it has been more helpful to concentrate on the Somme first. However, it is all tied together.

Robert

[Jonathan Saunders]

Robert - look forward to your further words in due course.

Jon

[Old Tom]

Hello,

Once again may I add my thanks to Robert, and others, for the quality of explanation in this thread. It seems to me that a few issues need to be addressed further as they have a great bearing on the outcome of the results achieved during the Somme campaign. I suggest these are:

Command in relation to the previous experience and training of the CinC, the Army and Corps commanders.

The use of artillery in relation to the weapon available and their control. Including the position of senior artillery officers as advisors rather then as commanders.

The state of training of the New Army formations in relation to the scale of the attack launcehd on day 1. Or to pose the question was it necesary to commit the army to an attack on, was it, 25,000 yds, would one, or more, attacks on narrower fronts allowed greater concentration, perhaps penetration and the infliction of as many casualties.

These would involve a measure of hindsight, but should be interesting. I intend to have a go having thought a bit more. To quote a half remembered instructor 'Time spent on reconnaissance is seldom wasted'

Old Tom

[Robert Dunlop]

...can I ask what you think of the thinking behind, the planning, the preperation, and the theoretical value of the feint at Gommecourt?

Jon, I know much less about this action. Your question prompted me to review it in some more detail. So far as I can tell, the first mention of Gommecourt came in Haig's reply to Rawlinson's first plan:

"Simultaneous activity against the Gommecourt salient, designed to hold the attention of the enemy's artillery and reserves on that side [of Rawlinson's proposed attack], is advisable so far as it can be arranged for. The Third Army will probably be able to give some assistance of the same nature further to the north." (GHQ OAD 710/1)

Rawlinson responded with:

"The possibility of including Gommecourt amongst the first objectives to be attacked is, to my mind, purely one of men and guns. I did not include it in my original plan of attack because I considered that a 20,000 yards front was as much as the troops at my disposal could undertake with reasonable hope of success.

The necessity of keeping sufficient troops in reserve to maintain a sustained offensive for a considerable period renders this very important. [At this point, Rawlinson pointed to the lesson from Verdun of enabling sufficient fresh reserves to keep the momentum of an attack going.]

There is no doubt that if the attack on the Gommecourt salient could be undertaken as a small independent operation, it would be of considerable assistance to my main attack; but I do not consider that the means at my disposal render this possible for me to undertake, and I would suggest as far as this Army is concerned, limiting the activity against Gommecourt to a demonstration in which would be included the discharge of smoke and gas." (Amended Plan Submitted to GHQ - 19th April, 1916)

Haig ensured that Rawlinson did not have to plan or execute the attack:

"A simultaneous attack on the Gommecourt salient will be carried out by the Third Army troops under the orders of the GOC that Army." (GHQ OAD 16th May, 1916)

As we know, the attack on the salient took place on 1st July. Two British divisions were involved. Both obtained footholds in the German trenches either side of Gommecourt but neither the 46th or 56th Divisions could link up. More importantly, the combination of heavy German artillery counter-preparatory fire caused the attacking troops to be isolated, while the German counterattacks succeeded in clearing the German lines again. The attack failed in its immediate objectives, with heavy losses inflicted on both divisions (46th Division - 2,455 killed, wounded and missing; 56th Division - 4,314).

I guess the key question is why, oh why was an isolated attack carried out whose sole purpose was to distract the Germans. The failure was predictable - why sacrifice lives in this way? It raises the issue of incompetence. If not, then callous disregard for human life.

It is absolutely clear that British High Command only wanted the Gommecourt attack as a diversion. Snow (GOC VII Corps) had suggested an alternative "...that a threat from Arras would more effective and less costly" (British Official History). The purpose of the attack, however, was laid out in the OAD to Third Army: "...to assist in the operations of the Fourth Army by diverting against itself the fire of artillery and infantry that might otherwise be directed against the left flank of the main attack near Serre." The British Official History notes that "when, four days before the assault, Sir Douglas Haig asked General Snow how he was getting on, the commander of the VII Corps was able to reply, 'They know we are coming all right.'"

Why persist in attacking Gommecourt? Why have a diversionary attack at all, even if Snow's prediction about Arras was correct and casualties would have been less? Surely any casualties would have been a waste outside the main battle?

I can offer some thoughts. Long before the Battle of the Somme, the British had digested the lessons of Neuve Chapelle, Aubers Ridge and Festubert. It should be noted that many of the eventual shortcomings of these battles were predicted in advance. In any case, when the Battle of Loos was being planned, Gough noted:

"At this period [July, 1915] we had in France only 71 heavy guns of a calibre of six inches and over, and at a conference presided over by Mr Lloyd George in July on the question of munitions the conclusion had been arrived at that the only hopeful form of attack was one on a front of 25 miles carried out by 36 divisions and covered by 1150 heavy guns and howitzers [ie approx 50 heavy artillery pieces per mile]."

Why the importance of attacking a very wide front? Broadly, there were two key issues that had to be dealt with. One was the enormous increase in artillery and the firepower associated therewith. Gommecourt precisely indicated what happens when an attack occurs on a too narrow front. The defender's artillery pieces can converge on the area of the attack, particularly those batteries that are several miles away to the north and south of the actually focal point. The flanks would always be the most vulnerable part of the attack. Given the range of the artillery, the defender's batteries outside the flanks of the attack could fire well down the line of the attack. Only within the centre of a broad frontage attack would the defender's batteries be limited to those immediately ahead of the attack, in general terms. If these few batteries (relatively speaking) could be neutralised, then the attacking forces might create a wide enough lodgement to enable a breakthrough.

The second major issue was the reinforcements. These could be divided, broadly, into three types: immediate (present in the support lines and available immediately in that local area), nearby (not far from the battlefield and capable of being deployed across a relatively wide area), and distant (ready to move but need transporting by rail so need at least a day to get to the battlefield). For the attacker, reinforcements presented a major threat. The number one problem was their ability to seal off and destroy any gains in enemy line. This threat was chiefly posed by the immediate reinforcements, who could get to the attacker most quickly. Only if the captured territory could be retained did the second threat come into play - the threat of stopping any breakthrough or sealing off any breakout. The nearby reinforcements could achieve this, along with the distant reinforcements when they arrived. These reinforcements posed a threat to any captured territory but their ability to counterattack would always be delayed by the time take to get into position. In general, delayed counterattacks were much much less successful if the attacker had managed to consolidate.

The deployment of reinforcements required early recognition of site and extent of the threat from the attacker. Unless there was complete surprise, the defender could move more reinforcements into the various categories before the battle even commenced. In this situation, the attacker had to rely on other measures to negate or minimise the threat from the counter-attack troops. One measure was to try and destroy those reinforcements in the immediate area. This was the job of artillery. The second measure relied on trying to stop the reinforcements from moving. Two options were available for this. Artillery and aircraft could interdict reinforcements in their forming up areas and along known routes of rapid movement, particularly roads, road junctions, railways, rail stations, and major communication trenches. The second option was to paralyse the decision-making power of the defender. Cutting lines of communication and destroying command centres was the job of artillery. This would blind the commanders who had to decide where and when the reinforcements had to be sent. Operation Michael demonstrated how effectively this could be done. Less obvious was the fact that a wide attack frontage would also create indecision in the minds of the defending commanders. Given that communications would be impaired, both by the reduction in routes of communication as well as the chaotic nature of incoming information, higher level decision-making about the deployment of nearby reinforcements would be affected if multiple sectors were under attack.

Back to Gommecourt, but in the next post... time to give the RSI a rest

Robert

[Robert Dunlop]

I have created a series of diagrams to illustrate the rationale behind a wide attack frontage and an additional diversion. The first diagram represents the status quo:

The next diagram illustrates the effects of a narrow frontage attack:

Note how the defender can concentrate both artillery and immediate reinforcements against the flanks of the attack. Given the narrow width, the whole salient comes under severe pressure.

Robert

[Robert Dunlop]

If the attack frontage is widened sufficiently, the defensive forces are more dissipated. Note the reduced concentration of artillery fire in the centre of the attack:

Now the 'Gommecourt' effect. The last diagram illustrates how a separate attack off to one flank can potentially reduce the pressure on the main attack frontage, particularly on the flank:

Robert

[PhilB]

Looks very similar to simply widening the front of attack? Phil B

[Robert Dunlop]

The diagrams are stylised. In practice, artillery would not be distributed as evenly. As the war progressed and counter-battery fire improved, the effects of the defender's artillery would be further reduced. Also the immediate counter-attacks were often subjected to intense fire, as happened on several occasions in the Third Battle of Ypres for example. If a narrow frontage attack was needed, it had to be protected by vigorous counter-battery fire across a much wider frontage than the attack itself. This is somewhat easier at the beginning of a campaign. As the attacker progresses further into a salient, it becomes harder to use artillery in this way. The fire tends to be divergent, not convergent as it is for the defender. The problems taking Delville Wood illustrated this principle.

With respect to Gommecourt, the British Official History notes that the British preparations caused the Germans to reinforce this sector. The 2nd Guard Reserve Division was pulled into the line, which meant it wasn't available to repulse the attack further south. Also, six heavy artillery batteries were moved to cover Gommecourt. Presumably some of the artillery batteries already in the line were directed towards Gommecourt and not further south.

Robert

[Robert Dunlop]

The potential problems with the Gommecourt attack were recognized. I don't know for sure but Charteris' account about potentially stopping the attack of one corps may have related to VII Corps. See

If this is true, then the final decision to go ahead lay with GHQ, as represented by Charteris, but it was based on the feedback from the divisional and corps commanders, not the unshakeable whim of the C-in-C.

Robert

[Robert Dunlop]

The other point I meant to make is that Gommecourt was a pronounced salient. As soon as I saw the map, it was clearly a recognized target for an attack - textbook stuff. If you wanted to create a serious diversion, then Gommecourt was the place to do it. The fact of it being a salient would have meant the Germans took the threat of any attack very seriously.

Robert

[Jonathan Saunders]

Robert - I have printed off your reply and diagrams on Gommecourt and will study them before responding.

Many thanks,

Jon

[Jonathan Saunders]

If you wanted to create a serious diversion, then Gommecourt was the place to do it.

Robert,

Many thanks for your comments. Having raised the subject with you, and you having given the time to post a comprehensive reply, I am afraid time has been an issue for me to look at Gommecourt in depth and for this I apologise.

I understand your focus is that Gommecourt was used as a feint to draw reserves away from the main attack on 1 July, yet my point of view is that it actually distracted British resources away from the preparation and execution of the main attack. Bearing in mind we had about 50% of the ratio of artillery to yardage of trench that we had enjoyed at Neuve Chappelle, then I don’t think we should have been extending the frontage at all. Third army artillery would have been better served if it had been brought to bear on Serre. Also for Gommecourt to be a worthwhile gamble/diversion, then Serre had to be a success, otherwise there was no point to Gommecourt. As we know, Serre was a catastrophic disaster.

Also I have to strongly disagree that Gommecourt was the place to create a serious diversion – I would say it was the last place. The OH describes Gommecourt to the effect “it wasn’t a Salient but a mini fortress”. The extent of the fortifications at Gommecourt (and further south at other strong points on the Somme) were well known to the Allies from 1915 actions. I believe Rawlinson had described Gommecourt as virtually impenetrable and as far as I can make out he wanted nothing to do with Gommecourt if it involved Fourth Army. Allenby was equally dismayed and favoured a diversionary attack in the Arras region, although when Allenby recognised that Haig was immovable in his choice of Gommecourt as a diversion, he acquiescenced to Haig’s view as he was like to do when confronted by a superior.

Gommecourt should have been isolated, not attacked as a diversion. If there was to be a diversion then this had to be in a relatively weak place in the line, not one of the strongest parts of it – and with the defenders given ample warning of what was going to happen. Also I believe Gommecourt did not so much draw reserves or artillery away from the Somme but attracted them from the north – artillery that otherwise would have been out of reach of the Somme battlefield.

I am afraid it isnt a very comprehensive reply but I think it outlines some of my thoughts.

Regards,

Jon

[PhilB]

It`s always struck me that the Gommecourt Salient was asking to be nipped out. It was the obvious move and the Germans had prepared most effectively for it. And we didn`t disappoint them! Phil B

[Robert Dunlop]

Jon and Phil, I am travelling again but will try and post some responses over the next week.

Robert

[Robert Dunlop]

...I understand your focus is that Gommecourt was used as a feint to draw reserves away from the main attack on 1 July, yet my point of view is that it actually distracted British resources away from the preparation and execution of the main attack.

Jon, the purpose was to draw German resources away. To do this, British resources had to be used. Your point, a very valid one, is that the British forces could have been better used maximising the effectiveness of the main attack.

Bearing in mind we had about 50% of the ratio of artillery to yardage of trench that we had enjoyed at Neuve Chappelle, then I don’t think we should have been extending the frontage at all.

I know that this is an argument put forward by Prior and Wilson too. FWIIW, I am less convinced by this. It was not the tonnage of shells fired that was important. It was how the shells were used. Wire cutting was crucial to any success. In the absence of tanks and given the amount of wire that needed to be cut, then the hurricane bombardment of Neuve Chapelle was never going to work for wire cutting. Adding more field guns to the main line of artillery was not going to make a significant difference to the speed with which the wire was cut - it was still going to take days.

Would the heavy guns have made an extra difference? Again, not in my opinion. Dumping heavy shells on the German defences for days was not the key to success. The key was, at the moment of the infantry assault, to suppress the German defenders for as long as possible and to neutralise as much of the German artillery as possible - having cut the wire. In many places, British artillery tactics at the moment of attack were wrong. Simply adding more guns to the existing attack would not have made the tactics right. A higher tonnage of shells would still have been used in the wrong way. At the very least, the attack around Gommecourt indirectly 'neutralised' some of the German guns that would have otherwise been free to attack the left wing of the British attack.

Robert

[Robert Dunlop]

Also for Gommecourt to be a worthwhile gamble/diversion, then Serre had to be a success, otherwise there was no point to Gommecourt. As we know, Serre was a catastrophic disaster.

Jon, another really important point. How should the 'success' of a diversion like Gommecourt be judged? You have raised the very local perspective. Gommecourt should have enabled Serre to be captured. This is a valid comment. Serre was part of the key to any success against the high ground immediately south and east of the Ancre, which was Rawlinson's stated initial target.

But what if Gommecourt, and Serre, were crucial for the success of the British attack around Montauban, and the French attack south of the river? Prior to July 1st 1916, the BEF had succeeded in gaining lodgements in German-held territory but these were never big enough to leverage. The first day of the Somme gave the BEF and the French enough room to not only resist local counterattacks but, more importantly, begin prising the Germans off the Pozières ridge. Eventually, the Germans had to relinquish the whole area, including Bapaume. Gommecourt and Serre played a part in this.

Might the BEF have done better without Gommecourt and Serre? I think it would have been extremely difficult to push more men into the attacks on Thiepval, Beaumont Hamel, etc. More infantry was not the answer in these locations. As stated before, better handling of the artillery would have been much more helpful. As Ralph Whitehead and Jack Sheldon have demonstrated so well, the German defences around Beaumont Hamel and Thiepval were fabulous. Really determined and intelligent counterattacks finally nullified any gains around the latter. But the Germans were unable to defend everywhere at once. And that was the eventual undoing of the whole Somme front. There is no doubt in my mind that the width of the Entente attack was critical in enabling the success.

Robert

[Robert Dunlop]

Also I have to strongly disagree that Gommecourt was the place to create a serious diversion – I would say it was the last place. [shifted] Gommecourt should have been isolated, not attacked as a diversion. If there was to be a diversion then this had to be in a relatively weak place in the line, not one of the strongest parts of it – and with the defenders given ample warning of what was going to happen.

Jon, if the purpose of a diversion is exactly that, then it is important to announce that it is going to happen. However, it is equally important that the 'announcement' is seen as credible by the enemy. Otherwise, if there is no announcement then we are talking about a surprise attack, which is quite a different thing. If the threat is not credible, then no diversion will occur. Why would a point in the line be 'weak'? All things being equal, it would be because that point in the line was deemed less important by the enemy. Therefore the enemy would be less likely to divert resources to defend it. This is what happened with Fifth Army prior to March 21, 1918. The sector covered by Fifth Army was deemed less important than the salient and other areas further north. It was weaker, offering the Germans precisely the target they were looking for. Operation Michael, however, gained the Germans the wasteland of the Somme.

Gommecourt is precisely the sort of target to attack, as a diversion. The nature of the attack also lent itself to being credible. The BEF sought to pinch out the salient, attacking both flanks at the base. Precisely what the Germans would have planned in the same situation.

Allenby was equally dismayed and favoured a diversionary attack in the Arras region, although when Allenby recognised that Haig was immovable in his choice of Gommecourt as a diversion, he acquiescenced to Haig’s view as he was like to do when confronted by a superior.

I just want to pick up on this point, if I may. It has been suggested that Haig was incapable of accepting advice or counter-argument. As I have shown earlier, Haig accepted many of Rawlinson's recommendations, even forgoing Haig's own views on things like the duration of the bombardment, etc. This aspect of Haig is not usually portrayed.

Robert

[Jonathan Saunders]

Robert - many thanks for your further comments, which i will consider more carefully over the next couple of days.

Just a couple of quick points:

*with regard to better use of the artillery then I totally agree with you. You are right to say throwing more shells on the German defences would cause no further damage to the deep dugouts however the lack of counter battery work leading up to 1 July leaves me dumbfounded. Lack of a sufficient number of long range guns was a problem but some solution should have been found.

* Of course a diversionary attack has to be announced in some way whilst maintaining the integrity of the attack as if it was part of the "big push". Announcing an attack - which in essence was going to be light weight and purely a diversion - against one of the strongest parts of the German line was not IMHO a reasonable thing to do. We are told Haig was the Educated Soldier, well what did Haig think was going to happen at Gommecourt?

[Robert Dunlop]

Also I believe Gommecourt did not so much draw reserves or artillery away from the Somme but attracted them from the north – artillery that otherwise would have been out of reach of the Somme battlefield.

Jon, the British Official History notes that:

'Both the 46th and 56th Divisions, however, suffered specially from enfilade fire from artillery not on the corps front: the latter from batteries to the south-east near Puisieux (3 miles south-east of Gommecourt) on the VIII Corps front; and the former even more from the very numerous guns on the 37th Division front...'

If this is true, then the Gommecourt diversion succeeded in one of its objectives, where the counter-battery work did not.

Robert

[Robert Dunlop]

...the lack of counter battery work leading up to 1 July leaves me dumbfounded. Lack of a sufficient number of long range guns was a problem but some solution should have been found.

Jon, as you rightly point out, the counter-battery work (or lack thereof) was really crucial. Clearly, there were successes. The BOH notes:

'The artillery of both the 12th and 28th [German] Reserve Divisions suffered greatly in the preliminary bombardment, and more seriously still on the 1st July. According to one reliable history, the German batteries in the valleys north of Mametz and Montauban were destroyed with the greater part of their ammunition, and very few guns could be withdrawn to the 2nd Position. The official monograph goes into some details:

"One battery in the 12th Division had lost a gun on the 30th June. Gradually the other three were put out of action". Another "lost two guns before midday. The other two followed in the afternoon." In a third battery, two were lost, and in a fourth, "all the howitzers became gradually unserviceable. The greater part of the other batteries of the Group had considerable losses. Their fighting power fell off more and more." In the 28th Reserve Division "the batteries had suffered severely in the artillery battle, and could give little assistance. A great number of the guns were smashed up... Particularly the batteries in the Caterpillar Valley had lost numerous guns. When the British attacked there were only then field and thirteen heavy batteries in readiness and these had numerous unserviceable guns." All the field guns of the 28th Reserve Division were "rendered unserviceable by British fire or other causes". The British artillery fire on the 1st July in this part of the field is described as "devastating" (verheerend).

By midday on the 1st July four of the heavy batteries on the XIII Corps front were completely unserviceable, and later another is mentioned as sharing the same fate.'

Effective CB work was possible in this area. Lack of long range super-heavy guns and howitzers is cited as a problem in other areas of the front. The other thing to note is that the success to the south relied heavily on HE shells smashing the guns. Gas shells proved to be more effective for CB fire, in that they provided an area effect. This approach was used more extensively later in the war.

I have yet to fully understand why CB was not as effective further north. Simpson's book 'Directing Operations: British Corps Command on the Western Front 1914-18' certainly makes it clear that the corps' GOCRAs had the primary responsibility for the overal artillery plans for each corps, though divisional commanders had significant input. Farndale makes it clear that other corps devoted much less resource to CB work. I don't know and it is an issue that I want to explore more fully. In doing so, it is not just a question of whether the resource and attitudes were in the right place (or not) on the British side on July 1 and the days preceding. It is very important to understand how the Germans manoeuvred and reinforced their artillery in the lead up to July 1. To what extent were the German commanders reacting to the threat along the whole line of attack versus the focus on the defensive Schwerpunkt of Thiepval and Rawlinson's desired Pozières ridge? Were there enough artillery reinforcements available to bolster the areas around Montauban or south of the Somme? Or did the German commanders have to overlook what was happening further south, either deliberately because they misread the situation, or because the problems seemed too hard to solve and were therefore ignored, or some other reason/s?

South of the Somme River, the discrepancy in Entente versus German artillery was even more noticeable. Despite the fact that Verdun was still in progress, the French had amassed 80 heavy batteries against the German's 8.

Robert

[Robert Dunlop]

Announcing an attack - which in essence was going to be light weight and purely a diversion - against one of the strongest parts of the German line was not IMHO a reasonable thing to do. We are told Haig was the Educated Soldier, well what did Haig think was going to happen at Gommecourt?

Jon, I have double-checked Haig's diary, the Operational Orders I have access to, and some other sources. In general terms, Haig knew that some attacks (possibly the whole attempt on July 1) might fail and there would be significant casualties. Gommecourt does not appear to have featured as a special instance. The potential (lack of) performance of VIII Corps was a worry before July 1. The expected outcome of the attack on Gommecourt, as you know, was:

'...to assist in the operations of the Fourth Army by diverting against itself the fire of artillery and infantry which might otherwise be directed against the left flank of the main attack near Serre.'

I have not come across any acknowledgement from Haig about the efforts of VII Corps on July 1. It is highly likely that he did send a formal acknowledgement - he did so on many other occasions.

The real issue, however, that I think you are raising is how could Haig even consider the thought of sending the men of 46th and 56th Divisions to their certain deaths just for a diversion. Stepping up from the narrow focus on VII Corps, and accepting that some sort of battle had to be fought on the Somme, surely there were other examples, like Gommecourt, where an 'Educated Soldier' would have said something like: 'Let's keep this down to a minimum. Only attack the weakest point. Don't do more that we absolutely have to.' This is the heart of the debate about the Somme, Arras, Third Ypres, etc. Simply put, though this is not a simple issue, it comes down to whether you want to have a large number of smaller battles, with fewer casualties (potentially) at each battle, or a small number of large battles (or campaigns). I would like to suggest that, short of England not doing its duty, then the total number of casualties would have been the same, irrespective. Perhaps. There are arguments in favour of the small number of large battles approach versus the same number of casualties in more battles over a longer time period. But these will have to wait - work beckons.

Robert

[bmac]

I have been away on holiday so have only just come to this topic. I am also at a slight (though pleasant) disadvantage as I do not have to hand a copy of my little book on Gommecourt (I have sold them all and am waiting for new supplies ).

A few observations. In early April, Third Army were asked to prepare a variety of plans to support/help the Fourth Army's attack. Proposals were: an attack around Vimy or one at the Monchy au Bois salient to the north of Gommecourt. Monchy was the one approved and Kiggell asked them for plans. It was not until towards the end of April that Allenby was asked to prepare a plan for Gommecourt. Until that point the division occupying this sector, the 48th (South Midland) Division had been preparing a two brigade attack from in front of Hebuterne. The Official History, perhaps using the Vimy information, suggests that Allenby and Snow, were dismayed by the idea of an attack at Gommecourt and proposed, instead, an attack on a 'weak or vulnerable part of the enemy's front' (page 454). There is, actually, no record of them objecting after the instruction to attack Gommecourt was passed on. They had been asked for and provided their suggestions, i.e. Arras and Monchy before the attack on Gommecourt was proposed and not after the order had been given. The words in the Official History are, possibly, Edmonds being wise after the event and, as on numerous occasions, the comment that they should instead have attacked a 'weak or vulnerable part of the enemy's front' comes from a letter written by a respndent to a request for comments on the OH's original draft. In this case, the words arer drawn from a letter written by Allenby's Chief of Staff, Bols, to Edmonds during the preparation of the OH in the 30s.

Ironically, given OHL's persistent view that the main thrust of the British attack would fall well to the north of Gommecourt (a view which persisted for several days after the delivery of the 1st July attack with reinforcements being sent to Rupprecht's 6th Army rather than von Below's 2nd Army) a well publicised diversion at, say, Vimy, where German defences were then not of the same quality as those on the Somme and where there was a clear tactical advantage to be gained by taking the ridge, would have been sure to draw significant numbers of men and guns to this area. Instead, the 2nd Guard Division which had been there (and, in another irony, had defended the Hohenzollern Redoubt against the attack of the 46th Division on 13th October 1915) was taken out of reserve (but not 2nd Army reserve) and sent to Gommecourt in mid-May, bringing with it additional artillery which was soon reinforced by further heavy and field guns some from a Bavarian Reserve Division.

The effect of this reinforcement was that the 52nd Division which, until then had been defending a line from near Monchy to south the Serre, was able to withdraw the 169th and 170th Regiments from the defence of the salient. Indeed, the whole division side-stepped to the south but with no shift in its southern boundary. In other words, the reinforcement of the Gommecourt salient led directly to the reinforcement of the Serre and Redan Ridge sectors with an increased concentration of both infantry and guns. These additional guns were, therefore, available to help crush the 31st and 4th Division's attacks with the OH describing the barrage on the 31st Division's trenchs as like a 'thick belt of poplar trees' (page 443).

These guns did not, therefore, participate in defeating the 56th Division until the 31st Division's attack had been destroyed. At this point many of these guns switched to delivering devastating enfilade fire on the 56th Division from south of Puisieux. Meanwhile, the 46th Division were suffering from similar fire from the 111th Division's artillery in Adinfer Wood to the north as well as from the heavy and field guns of the 2nd Guards.

In essence, therefore, by selecting Gommecourt, regarded by many as the strongest part of the entire Somme defence line, GHQ actually reinforced the northern end of the main attack and must have contributed, therefore, to the shambles that took place north of the Ancre on 1st July.

As to the attack itself, the two divisions were instructed to make the infantry preparations as obvious as possible, which the 56th Division certainly did by advancing their front line in late May by some 400 yards. Oddly enough, though, the heavy artillery kept a remarkably low profile until 24th June, in spite of the fact that the majority of the batteries were totally new to the area and had little or no chance to register their guns with any thoroughness a problem compounded by the inexperience of the majority of the gunners and the poor weather obscuring the RFC observers of No 8 Squadron RFC.

In spite of the bravery of the men of the two divisions the 'diversion' was a bloody disaster and its planning materially contributed to the failure to the south. The OH comments that "it seems improbable that GHQ realised the strength... of the Gommecourt salient". If true this is something of an indictment of the Staff. What is also true, however, is that one of the chosen divisions (the 46th) was exhausted and enervated by a long stretch in the front lines at Neuville St Vaast (where mining was a popular field sport) in March and April. It was also losing men sick to diphtheria, typhoid and paratyphoid (originally under the generic name of 'trench fever') in signficant numbers. When they arrived at Fonquevillers they found a trench system that had virtually collapsed. The previous occupants, the 48th, had been unable to man the whole stretch of front and had fallen back on a series of outpost lines between which the trenches were filled with barbed wire and had then collapsed. This entire system had to be reconstructed by the same men who were to attack. In addition, they had to prepare the whole front and rear areas for a major attack. Owing to the topography, the new trench system was a disaster, liable to flooding and constantly ankle and sometimes knee deep in mud.

It is arguable that the 46th Division was not fit (nor did they have numbers) to launch a sacrifical attack of this nature. It was known that there was not sufficiant artillery available to the neighbouring divisions to achieve any significant results against the artillery at Adinfer. It was known very early on that reinforcements of men and guns had been drawn into the salient. In many ways, therefore, the objective of the attack had already been achieved. The heavy and field guns which could, then, have been used to prevent these men and guns playing any role in the defence of Serre were in place. Smoke, demonstrations, feint attacks, etc., would have pinned the defenders to their posts. The actual attack on which "nothing depended" (OH p 454) was unnecessary and futile.

Allenby and Snow were following orders. Haig decided upon the Gommecourt attack and demanded it be made obvious and then followed through. His is the responsibility for the outcome and he must also, therefore, take responsibility for the knock-on effects at Serre and beyond. In my humble opinion

Right, time for lunch!

[Jonathan Saunders]

Bill - thanks for adding your thoughts.

Robert - apologies for still not having responded to your comments but I just havent had the time to consider my reply in depth. Hopefully will get back to it soon.

Jon

[Robert Dunlop]

Just a quick note to bring this thread back to forefront. It is relevant to the current discussion on the Somme.

Robert

[PhilB]

Once a war of attrition has been decided on, is it necessary to fight large, set piece battles? Did the German Army fatally weaken in 1918 because of the battles or because 4 years of war had taken its necessary toll? Could the same result have been achieved by "aggressive line holding" rather than attempts at breakthrough? Or does the role of C in C demand big ambitions? Phil B