Skip to main content

Full text of "Development of Tactics--World War"

See other formats


Google 



This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project 

to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 

to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 

are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other maiginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 

publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing tliis resource, we liave taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 
We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain fivm automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attributionTht GoogXt "watermark" you see on each file is essential for in forming people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liabili^ can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 

at |http: //books .google .com/I 



Development of Tactics 

World War 

BY 

V! Mu.. BALCK 

Lieutenant General, German Army 



TIANILATD Br 

HAiOtY Bell 



With Eighteen Sketches 



* ^ « 



• 


. * - • - • 

■" ■ -■••'. . . • 

• • • - 

- e - • 

• - 

J ... 
' ' ' ' 




THE GENERAL SERVICE SCHOOLS PRESS 




FOBT LlAVBNWOBTH, KANIAS 


/ 


1922 






I 


• 

• 



TABLE OF CONTENTS 

Chapteb XII— The Year 1918 260 

Transition from Position to Mobile Warfare .260 

Methods of Hostile Defense and German Offensive 262 

The Spring Offensive 270 

Resmning the Offensive 281 

The Last Defensive Battle 286 

Conclusion 294 



I. Training in Peace and Reality in War 

''Our long garrison life has spoiled us, and effeminacy 
and desire for and love of pleasure, have weakened our mili- 
tary virtues. The entire nation must pass through the 
School of Misfortune, and we shall either die in the crisis, 
or a better condition will be created, after we have suffered 
bitter misery, and after our bones have decayed/' Thus 
wrote the late Field Marshal von Gneisenau under the im- 
pressions of the experiences of war in the Year 1806. Only 
the bloody seriousness of war furnishes a final receipt for 
long peace labors. If the training of the troops is accord- 
ing to correct principles, so that they can perform anything 
and everything war demands of them, they will not have 
to forget on the battlefield anything tiiey have learned in 
time of peace (I.D.R. 477) ; also they will not have then to 
learn anything new. Anyone who possesses actual ex- 
perience in war may more easily attain such education than 
another who knows war only from books, or whose remem- 
brances of war have been dimmed during long peace time. 
The Austrians in 1866 and the British in the Boer War, 
had materially larger experiences in war than had their 
opponents, and still neither the one nor the other was inured 
against disillusions; it may be because they arrived at erro- 
neous decisions based on prior experiences, or it may be 
because the conception of actual war had been lost by col- 
onial warfare. "La petite guerre gate le mUitaireJ* "The 
Austrians," wrote Marshal Foch when he was Director of 
the War Academy — ^*'had so far conducted war without un- 
derstanding war, the Prussians had studied war and there- 
fore understood it, even without having conducted war." 

Just as dangerous as effeminacy is the danger of let- 
ting habitual custom gain the preponderance, as is shown 
by a senseless repetition in training of a well known exer- 
cise or a measure that has proven successful in the face of 
the enemy. The army of 1806 — ^as von der Goltz calls it 
"a very industrious, orderly, willing army" — ^had gone to 
sleep on the laurels of Frederick the Great. Afraid to break 



2 DEVELOPMENT OF TACTICS— WORLD WAR 

away from the creations of Frederick, we had allowed them 
to become obsolete by adhering to the strong impetus given 
by unreliable foreigners, and consequently also adhered to 
the old time line tactics and the sterotyped formations. The 
first made the army tactically, the latter operatively clumsy, 
an enormous disadvantage opposed to the French Army that 
was not bound down to those tactics. Not in line tactics (for 
in adopting line tactics the British army was victorious in 
the Peninsula War over the French column tactics) but in 
the inaptitude of commanders and troops was to be found 
the cause of defeat on the battlefield. In the ^'echelon at- 
tack'' practiced in time of peace at every opportunity, later 
commanders perceived the surest guarantee of victory. In 
Germany we also trained, up to the publication of the Regu- 
lations of 1888, by battalion, regiment and brigade. In addi- 
tion to the free fighting, frequently badly transformed into 
what we called the ''Turkish'' method there was a hard and 
fast formation, which would hardly have been proper in 
the sixties. The finger marks of this formation was an at- 
tack, badly prepared by fire, carried out with insufficient 
means, and the final result of which was retreat, followed by 
flight in the face of cavalry and forming squares by bat- 
talions. Even the so-called "Schlichting's Seven Wonders" 
caused fatal, normal formation tactics, until the new regu- 
lations swept these misconceptions aside. But no matter 
how advanced these new regulations were, the habitual cus- 
tomary training of earlier times (for instance the prepon- 
derous importance of adhering to regular distance) kept up 
its domination. 

Prior to the World War the German army worked faith- 
fully and it was very far from falling into the errors Gneis- 
enau had condemned in 1806. But it was not free from a 
certain formality, which, as early as 1892 Major von Mala- 
chowski pointed out in an efficient study as "Review" tac- 
tics. By pointing out the difference between peace engage- 
ments and actual engagements in war he showed the "'Re- 
view" tactics as fatal for the field training of the troops. In 
the words of Gneisenau "The endeavor to lead the troops 
well on large parade and muster days is very damaging," 
and then continuing, "the tactics which expects everything 
from regulation uniformity, artificially produced on the 



PEACE TRAINING AND REALITY OF WAR 8 

drill ground, or at the green table, that cannot be used in 
war is '^Review" tactics. In most cases it extends only to 
a play, or criticism-proof battle exercise. During longer 
periods of peace it is the most dangerous enemy to field train- 
ing, by continually attempting to push the field training 
into the background at all points.'' And finally he adds 
wamingly 'The mask of 'Review' tactics continually 
changes, the inherent quality of the matter is always the 
same." Malachowski then turned against endeavors in 
the army which had for their object to assure a smooth 
course of exercises. ''No normal tactics can replace the 
military qualities and tactical perception of the leader," he 
says, "it is absolutely fatal to field training, by continually 
leading away from the simplicity and the actuality of 
things." There is but one auxiliary means to meet such ab- 
normal growths, and that is reverting to military historical 
experiences. The "Review" tactics are satisfied with super- 
ficial (exterior) forms, but forgets entirely that war is con- 
ducted by men against men and that in war the moral in- 
fluences are principally of the most decisive importance. 

It is very probable that the cavalry especially suffered 
from "Review" tactics, for it rode charges in autumn man- 
euvers till shortly prior to the World War. These charges 
gladdened every military eye, but they could never have 
been executed in that manner under actual fire. On the 
other hand, we do not desire to bring forth, as an example 
to be governed by, the British cavalry in the Boer War, 
which was nothing but mounted infantry. 

Still less did the artillery count in our exercises, unless 
it laid weight on appearances, as its firing capacities and 
results could not be portrayed. 

Actual experiences in war, which each one can gain 
only in a limited way, are of inestimable value; but they 
can produce fruitful effect only if they are thoroughly proven 
and utilized by the study of military history. And still, how 
quickly do actual war experiences fade away! We have 
experienced that fact in our own army. At the opening of 
the war in 1864 we had lost what had been gained in the 
Wars of Liberation at so much effort and so much loss of 
blood. The experiences of Worth and St. Privat should un- 
doubtedly have been sufficient to teach our army what we 



4 DEVELOPMENT OP TACTICS— WORLD WAR 

considered new experiences in the Russo-Japan War. In 
front of Verdun, in 1916, we had for instance, experiences 
in plenty concerning how horse batteries could follow the 
infantry across positions — and shortly prior to the Spring 
offensive in 1918, trials on a large scale were repeated to 
gain the same old results. 

It is not at all difficult to cite a military historical ex- 
ample for every tactical operation. '^Especially difficult 
will be a clear objective judgment, when the question is one 
of an unfortunate experience in war by our own army. The 
general adaption of military history to such examples often 
leads to the most serious errors which, having become ac- 
cepted rules, spread like an epidemic, which will even take 
hold of thinking heads, and still will not allow their meaning 
to be perceived." Thus we can account for the numerous 
erroneous decisions drawn from events. The Austrians 
drew from their experiences during the War in Upper Italy 
in 1859 the necessity of brutal shock tactics; the British, 
after the Boer War, were not very far from denying the 
possibility of attack; the desire, to avoid losses, took pre- 
cedence over the requirement to annihilate the enemy. And 
finally, the Russians in Eastern Asia had again to gather 
the same bitter experiences they had gained before at 
Plevna. Nothing but thorough study of military historical 
events, omitting the special experiences we obtain ourselves 
on the respective theaters of war, can prevent commanders 
from entering a new war with erroneous views of the in- 
herent qualities of present day fire effect gained during long 
periods of peace. Up to the present day every war has 
brought surprises, which the troops could master, not 
through hard and fast rules and formations, but only through 
the training they underwent in time of peace ; and this very 
fact must be reckoned with in peace time training. Only the 
inflexible will to be victorious, without regard to sacrifices 
demanded by battle, will overcome all difficulties. Only 
thus may we prevent the danger of immediately accepting 
every new thing happening in a strange theater of war as 
our guide, and adopting it as a cure-all for success. But, 
on the other hand, there is a caution, that ought to be con- 
sidered, contained in the words of the British lieutenant 
general, Sir Ian Hamilton, based on his impressions of the 



PEACE TRAINING AND REALITY OF WAR 5 

battles in Eastern Asia : — ''What a blessing ; the larger and 
the prouder an army is, the more immobile is it in its firmly- 
rooted power of sticking to fundamentals so that finaUy, as 
a unit, it becomes inapt to absorb the experiences of other 
armies. Military attaches can discover the most important 
points for training and employment in a foreign army and 
urgently recommend their adoption. The majority of their 
comrades pay just as little attention to them as did Napo- 
leon in pay attention to the reports of Stoffel concerning 
the Prussian army prior to the Franco-Prussian War.'' And 
quite similarly wrote the General of Infantry, von der Goltz, 
in the second edition of ''Rossback and Jena;'' — ''Even the 
South- African War has created doubt if we are stiU on the 
right road in the execution of our infantry fighting ; will our 
long and dense skirmish lines, with closely following-up sup- 
ports, carefully nested in the terrain, in the face of rapid 
fire, not go to pieces, as did in other times, the massed 
Prussian lines, under the fire of the French skirmishers?" 
Herewith were pointed out currents of "review" tactics, 
which threatened to limit the freedom left by training regu- 
lations. However, the German infantry could with satis- 
faction look back on the events of the war in Eastern Asia, 
for did not the Japanese infantry, trained according to Ger- 
man methods, victoriously fight under very difficult condi- 
tions? The Russo-Japanese War had set aside the uncer- 
tainty caused by the Boer War in the tactical views, and 
primarily the doubts concerning the possibility of executing 
the infantry attack. It made short and thorough work in 
removing the overestimation of adherence to forms, and of 
overestimation of the value of the fire power of the defense. 
However, the superior German leadership in 1870-71 and 
the passiveness of the Turks in 1877 did not show these two 
maxims as clearly as was the case in the engagements of 
the British and Japanese. Thus, these wars did not teach 
anything else than what could be gleaned from the experr 
iences of the Franco-Prussian and Russo-Turkish wars, 
namely : "Conduct of War means attack, attack means car- 
rying forward the fire." 

Attack and defense are on the same footing; any one 
who wants to be victorious and who desires to gain an ad- 
vantage by physically overcoming the enemy, must attack. 



6 DEVELOPMENT OF TACTICS—WORLD WAR 

But if we desire to resort to the attack^ we must train our 
troops accordingly. The bayonet training — almost preached 
by Dragomiroff — ^would have been the very thing in Eas- 
tern Asia, if it had been accompanied by thorough training 
in fire fighting. The form is of importance only, if it mater- 
ially increases the enemy's losses. The will to be victorious 
may also offset a discrepancy in numbers ; not the stronger, 
but the one that is more energetic, has the best chances of 
success. During a long period of peace, which easily causes 
overestimation of material factors, we cannot too fre- 
quently emphasize the fact that the decision to attack is 
based on the task, not on the relation of strength. This 
notwithstanding the fact we usually learn the actual 
strength of. the enemy only after a battle, in many cases 
only after the end of the war, and that all troops are in- 
clined to overestimate the strength of the enemy with whom 
engaged, and also that intrepid attack weakens the enemy 
and makes him dependent on our decision. Finally, no one 
knows if the enemy is actually able to make use of his 
forces. 

These maxims the German army appropriated to its 
own use. Its location between the two most important mili- 
tary Powers of Europe, in connection with an army which 
had delayed the opportunity to demand the utmost power 
of its people in the expectation of a decisive battle, forced 
the German army leadership to pay special attention to the 
attack against hostile superiority in numbers. In the ab- 
sence of experiences in war by ourselves, our army had to 
draw on the sources of military history for guides in its 
training, and this was done with success. Military history 
offers the possibility to properly learn the decisions that 
were of decisive importance, which, in exercises on the map, 
or on the drill ground seldom come into account. For con- 
duct of war we have to learn from foreign experiences, our 
own experiences come too dear and almost always too late. 
**Military history is no manual containing well-formed 
theories, is no volume to pass the time in reading, but is a 
careful teacher, who enables us, if we are attentive, to view 
things and to conceive their value, as we would never have 
seen in life. At any moment we may face in the same, or 
an entirely changed form, questions demanding responsi- 



PEACE TRAINING AND BEALIXY OP WAR 7 

ble, decisive and nevertheless immediate action. Of course, 
military history offers us in outline only the events. But 
it also offers what the very^ best theory can never offer, the 
picture of frictions in war, the picture of the influences of 
the doubts, of the urgency, of the incalculable chances, of 
the surprises, of the obstacles; it recounts the road which 
the commander and practical military knowledge have to 
take to overcome these difficulties; it prepares the normal 
counter-poise for the moment of action; it should prepare 
us also for the unexpected. Military history should take 
the place of actual experiences in war. Our life is not long 
enough to gather these experiences up to the moment of 
action.'' It does not suffice to merely follow up the regu- 
lations laid down ; these are only the basis of experiences 
during a definite, and passed, interval of time, which por- 
tray themselves in a far different manner in the brain of 
the victor than in the conception of the vanquished. Regu- 
lations must never descend to the plane of a code laid down 
for punishments; their details must not be allowed to in- 
terfere with freedom of action. Regulations are for the 
purpose of creating independent thought, but they have 
to be studied in conjunction with military history, and 
only what the commander inserts into them in the matter 
of his personal will and skill, makes them the guide for 
the conduct of battle. The new arms with which the armies 
were equipped in the beginning of the sixties, favored, in 
France, the creation of position tactics, caused von Moltke 
to consider the combination of the operative offensive with 
the tactical defensive, while the Austrians, having a one- 
sided, and in addition an entirely false estimation of the 
French offensive conduct, in vain endeavored to have vic- 
tory perch on the Imperial standards by means of brutal 
shock tactics. Of course, those shock tactics were suc- 
cessful at Custozza over a badly armed and morally in- 
ferior army. Tactical theories turn into fatal brain illu- 
sions if they influence the commander in his decision beyond 
the situation. 

Mahan, in his "Influence of Naval Power on History," 
writes as follows: "We will make the observation that 
changes in tactics will occur not after the introduction of 
new arms as is necessary, but also that the period of time 



8 DEVELOPMENT OP TACTIC&— WORLD WAR 

between the two changes is relatively very long; this, un- 
doubtedly, is caused by the fact that the improvement ot 
arms has its origin in the skill of one or two individuals, 
while for the change in tactics the inclination to stick to 
customs on the part of an entire class has to be overcome, 
which class endeavors to adhere to what exists now. This 
is a grave misfortune. It can be overcome only by ac* 
knowledging each change willingly and voluntarily.'' The 
history of tactics in the 19th Century shows this ''incli- 
nation to stick to established customs,'' from the disputes 
between field and review tactics, and we could cite more 
than one case thereof. 

The reason that, no matter how farseeing any regu- 
lations are constructed in their inception, they j become 
obsolete after a time, may be found in the very spirit of 
the training regulations. Emperor Napoleon measured that 
time to be ten years. In any case, frequent changes are a 
mistake, if we^do not want to interfere with the tranquility 
of tactical development and if we want to avoid friction in 
the composition of our mobile army consisting of the reg- 
ular levies, reservists and landwehr. On the other hand, 
regulations must follow suit, if the conditions on which 
they are based have changed. In his military phantasies, 
the Prince de Linge wrote in 1788: — ''A paragraph that 
ought to be incorporated in all regulations, and that we 
omit, I do not know for what reason, is that we should occa- 
sionally act contrary to regulations. We must teach action 
contrary to regulations just as well as we must teach dis- 
order among troops, for it will happen in battle." 

To be out of touch with present day requirements is 
always dangerous. Troops will have to pay later on with 
streams of blood for knowledge gained under the fire of 
the enemy. Of what use was it to the Austrians in 1866 
that they charged in utter disregard of death, imbued by 
the firm will to be victorious, but did so in tactical forma- 
tions which were then obsolete and in the face of the newest 
improved arms? The willingness to sacrifice themselves 
on the part of the troops and strictest discipline, encoun- 
tered an impassable obstacle in the rapid fire of an unshaken 
infantry. The experiences in war on the part of our regi- 



PEACE TRAINING AND BEALITY OF WAR 9 

ments show that bullets quickly write a new tactics, that 
bullets make short work of obsolete formations and create 
new ones. But at what cost ! In the Franco-Prussian War 
superior leadership and a better artillery permitted us to 
pay the price. 

At the outbreak of war the usual custom is to prepare 
troops, that have been trained under obsolete regulati<nis, 
for fighting on a strange theater of war by certain 'Tield 
Service Regulations.'' After the battle of MontebeUo in 
1859 Napoleon in made his troops quickly acquainted with 
the peculiarity of the terrain and the method of fighting 
on the part of his opponents, and the Austrians neglected 
to do this. In 1866 Benedek was forced to change the tac- 
tics that he had recommended in his field service regular 
tions, before his excellent views could be of any general 
benefit to the army. The tactical instructions of Euro- 
patkin were without any effect whatever in the Russo- 
Japanese War and individual experiences were disregarded. 
It is very desirable that the first experiences in battie be- 
come the general property of all concerned as rapidly as 
possible. 

. In 1870 we did not do this, and all units had to gather 
their own experiences. As late as August 18, 1870, the 
86th Fusilier Battalion advanced in columns toward the 
center, though the campaign in Bohemia had shown that 
that formation was completely obsolete. In the World War, 
the experiences gained at some one point were printed and 
thus quickly became the general property of all. The im- 
pulse to gain and spread experiences worked especially well 
on tranquil fronts. Extraordinarily much was accomplished 
by establishing schools in order to make officers of aU ranks, 
under officers of the special arms, familiar with the latest 
experiences. But this method sufficed only when the troops, 
by the method of their training in peace, had gained the 
necessary ability to adapt themselves thereto. Training 
regulations should not emphasize the matter of formation, 
they should induce every one to practical co-operation. 

Troops that are thus trained and trained theoretically, 
and who are not afraid of losses, will soon evolve new tac- 
tics. What a difference there was between the method of 



10 DEVELOPMENT OF TACTICS— WORLD WAR 

attack of the Guard at St. Privat and at Le Bourget^ and the 
charge of the Royal Grenadiers on Chateau Geisberg and 
of the Baden Body Grenadiers on the railroad embankment 
at Nuits ! The Russian experiences at Plevna were similar 
to ours in August, 1870, except that with the Russians the 
fear of losses gained the upper hand; they considered the 
fire effect of the Turks as a certain unchangeable factor, de- 
clined the freedom of operations, and sought only means to 
lessen their losses. In no instance must the troops at the 
first sudden impression of the hostile fire effect accept that 
fire effect as a stated factor and passively submit to it with 
the thought that it cannot be helped, and that the main 
consideration is to lessen the losses. In that case we forget 
that the impressions are the same with the enemy. The 
effect of these impressions is naturally larger if we have, be- 
fore the war, underestimated our opponent. The Russians 
sought to draw lessons for mechanical conduct in battle, as 
they also did subsequently in Eastern Asia ; but entirely dif- 
ferent factors co-operated decisively. The application of the 
best lessons as to utilization of our own fire power, as to 
formations and conduct against the influence of moral fac- 
tors, is impossible. We must never lay down the law, or 
accept as a maxim for our action, that procedure which 
proved itself in war, or in battles, as immediately the best, 
or which brought about the victory. Conditions, under 
which that procedure was correct, play too large a role, and 
among these numerous and different conditions the moral 
status of the troops and of the commander play the leading 
role. What one may permit himself to do and what, be- 
cause he permits himself to do it, leads to success, is for 
another who adopts it, the very cause of defeat. Studying 
wars and battles does not furnish recipes for victory. It 
increases the knowledge of the commander of troops only 
when it causes him to perceive the connection between 
cause and effect. It is not difficult to perceive the effect, 
that lies open to the view, but to correctly cull the causes, 
requires not only a clear view and study, but thought and 
knowledge of the characteristics of war. It is important 
also to know that chance plays a great role in all military 
actions, and that chance favors permanently him only, who 



PEACE TRAINING AND REALITY OP WAR 11 

deserves it. It is therefore wrong to condemn any action 
because without "luck'' success could not have been possible. 
Such an event teacheis merely that we must have the mental 
conception that luck easily passes one by, and that we must 
have manhood enough to grasp it in passing. 

Superior in rifles and guns, trained in fighting the 
mountain inhabitants of the Indian f rontier, and in defense 
against numerous swarms of the false Mahdi, the British 
entered the campaign against the Boers with full confidence. 
No one doubted a glorious victory ; but, in the dark December 
days of 1899 ill-success succeeded ill-success, not only in the 
matter of defeated attacks, but — ^what the heart of a soldier 
had to feel most poignantly — eyen capitulations in the open 
field. Up to then it had been unheard of in British military 
history, that in the first six months 182 officers and 4984 
men capitulated, while only 168 officers and 2124 men were 
killed or wounded. It is said that the tactics pursued by the 
British had their origin on German ground. But, German 
troops defeated, with German tactics, in South-Africa, an 
enemy equally as good as the Boers ; our troops there in any 
case understood how to quiclcly adapt themselves to new con- 
ditions. Colonial wars and inapt training in the home coun- 
try had led the education of British commanders into wrong 
channels. Absence of the firm will to insert even the very 
last man for victory, dearth of willingness to assume re- 
sponsibility on the part of the higher commanders, absence 
of independence on the part of the subordinate commanders 
and the inability to execute attacks by combined units sup- 
ported by artillery, are the real causes of the British dis- 
asters. As the British had, at the start, underestimated 
the Boers, the weight of the impressions gained was the 
heavier. Underestimation is followed as a matter of course 
by overestimation. In England, they went even so far as to 
question the possibility of any attack, and only gradually 
the spirits rose again. We must bear witness to the fact 
that the British in the World War attacked especially well. 
Through the misconception of inherent things in the Boer 
War, we attempted in Germany to follow the forms of the 
Boer attacks for a time (as a matter of fact, the Boers never 
attacked), until we again thought of our own experiences 



12 DEVELOPMENT OF TACTICS— WORLD WAR 

and employed only the most valuable of the experiences 
gained in the South African War. 

A similar thing happened in the Russo-Japanese War. 
The tactical causes of defeat were the same as in the Boer 
War, and thereto were added faults to be found in the char- 
acter of the Russian peoples. Bitterly -did they pay for 
their underestimation of the enemy. And this very fault is 
hard to avoid, as leaders and troops take their own achieve- 
ments as a model. Only sharp criticism of our own action 
can save us from underestimating the enemy. 

On the Japanese side also, the first successes had been 
bought with relatively small losses and just as in the Franco- 
Prussian War, a change of the procedure of attack took 
place, as we can see from following up the battles of the 
First Army. As with us in the year 1870 the endeavor was 
to arrange a looser formation. ''But the course of war, with- 
out definite rules, caused all tactical formations prescribed 
by regulations to be forgotten. As was the case with the 
Germans in 1870, the Japanese battle action gradually lost 
its uniform character and adapted itself again and again to 
the changing situation. In larger as well as in smaller units, 
down to the battle unit, the same applied. Thus, the Japan- 
ese tactics became more and more clearly acting according to 
circumstances, without regulation formations, and without 
ofiicial basis, all depending on the peculiarity of the leaders.'' 

Knowledge of the inherent peculiarities of a people, of 
their military establishment and tactical views is liieref ore 
absolutely required of every leader. Thus, the tactics of 
the French 1st Corps at Worth ought not to have contained 
anything surprising to us, as those tactics had been min- 
utely discussed in German military literature prior to the 
war, and the fire eflTect of the chassepot rifle, which we be- 
littled (probably for reasons of training) , ought not to have 
been unknown to the troops. The annihilating effect of hos- 
tile arms must not be allowed to offer something new to our 
troops. Proper target practice must make known the effect 
of arms to the troops and get them accustomed to have the 
artillery and other arms fire over their heads. As far as 
possible, the primary battle impressions must not be for- 
eign to the troops. Dragomiroff's method of training, to 



PEACE TRAINING AND REALITY OP WAR 13 

strengthen the will and the heart, was correct, though it 
had been completely misunderstood in Russia. The leader 
must be able to see his troops bleeding, but he must also 
know how to save them from unexpected high losses. The 
more individual the training has been in time of peace, the 
more the will to gain victory has been fostered, the more 
the readiness to assume responsibility has been developed, 
the better will the troops behave under the first impressions 
in battle. The individual will perform more, the more he 
is trained in the use of his arm, and the more he is imbued 
with the feeling that he is superior to the enemy in its use. 
In prior wars the British, the French, and the Russian 
boasted of their superiority to all other armies, in the use 
of the bayonet. This conviction in a way hurt them, as it 
merely fostered the desire to seek the decision in hand to 
hand fighting. Similarly, in the World War the conviction 
of the superiority of the German infantry in close range 
fighting which was proven in the very first battles, and there- 
after spread throughout the army, played a great role in the 
stubbornness of defense and in the relentless attack. Under 
the surprising impression of the effect of hostile fire we 
quickly discount the effect of our own fire, when it becomes 
the "safety valve of our own nervousness," as when the skir- 
misher fires only to deaden his hearing, or to be doing some- 
thing. Here the influence of the subordinate commanders 
and of quieter and better educated men must make itself 
felt, to attain a slow, well-aimed fire. The moral value of 
the individual is the decisive factor. 

We must not forget the importance of drill. It may be 
possible that other peoples do not feel its beneficial influ- 
ence. But we require drill by all means. We should never 
forget that it was nothing but the drill which helped the 
Guard at St. Privat gain the victory with the employment of 
obsolete formations and in spite of enormous losses. When 
imminent danger of death overpoweringly awakens the 
striving for salvation, when the frightful impressions of the 
battle cloud the clear understanding, then the often-heard 
word of command must electrify the soldier and make him 
capable to perform his duty. Therefore we cannot go too 
far in demanding the utmost endeavors in time of peace ; but 



14 DEVELOPMENT OP TACTICS— WORLD WAR 

not in the forms of a senseless marching around the parade 
or drill ground, but in the forms used in battle. We foster 
the drill, not for inspections, but for war. 

However, the army has no community life for itself 
alone. Prior to the World War, the army was the school 
through which a very materiial portion of the people went. 
Army and people are, and must be, one. Education of the 
entire people for efficiency in war and willingness to sacri- 
fice is a political and national necessity, and at the same time 
the main basis for the successful training of an army equal 
to all conditions. 

We were not saved from surprises in the World War, 
as will be shown later on. But we did not treat them merely 
passively, we became master of them and utilized them. 
That was possible only through the manner of our peace 
training, through rapid dissemination of all experiences, 
and through never slacking up in our work during the rest 
pauses between the battles. ''It has always been moral 
factors which have decided a war, even before it began, and 
it will always be moral factors that will do so. In compari- 
son with them everything else, organization, numbers, arma- 
ment, and even leadership, are but secondary considera- 
tions. And that is even true of leadership, as has just 
been said, because in an army in which every individual is 
trained to perform his highest duty, the leadership cannot 
be bad. And even if leadership sometimes commits an error, 
which is but human, what are those errors as compared with 
the efficiency that overtops everything, and the moral 
strength ? 

"But let us not believe that at any time any technical 
inventions, even if they are ever so enormous, will be able 
to change even the very least bit of the nature of war. It 
is true that they may change the forms, but they will never 
touch war's inner core. The capability to be victorious is 
not naturally inherent to any people. It cannot be bor- 
rowed from nature for eternity. No, the capability for 
victory must be educated into the people and into the army 
and must be maintained by strict performance of duty. 
Maintenance is frequently more difficult than acquisition. 
Two races, two nations, may be physically equal and have 



\i 



PEACE TRAINING AND REALITY OF WAR 15 

the same intelligence, but if in case of one of them the capa- 
bility of greater sacrifices by the individual, and thus by the 
entire people, is less than is the case with the other, it is 
as certain as the sun rises in the morning and sets at night, 
that war will pass its hard judgment on the lesser sacri- 
fice." 



II. Mobile Warfare 

TRAINING AND ORGANIZATION 

The Grerman leadership sought the annihilating battle 
as quickly as possible ; and in order to be able to conduct it 
at the decisive point, satisfied itself with defensive proce- 
dure on minor fronts, or with giving ground, prior to the 
decisive battle. The training of the infantry in time of 
peace was in full conformity with those intentions. Train- 
ing in peace was based on the excellent drill regulations of 
May 5, 1906, which of course had their origin in 1888, but 
which had been revised several times. Special preference 
to training in mobile warfare sought to foster the spirit of 
attack on the part of the infantry in every possible manner. 
The leaders saw the basis of all military success in indepen- 
dence, properly checked by tactical training and education, 
in increased firing ability and marching capacities, as well 
as in drill intelligently carried out. We cannot blame the 
troops for their inclination to a so-called ''attack-agitation'' 
in carrsring out the attack. This however prevented the 
co-operation between infantry and artillery which would 
have produced the fullest effect. Insufficient time was given 
to battle reconnaissance, to the establishment of means of 
communication, and it cannot be denied that there was a 
great preference for dense skirmish lines. But this peace 
training had such a permanent effect that the regular soldier 
and ttie reservist could not be distinguished in their capa- 
bilities within a few days. The officers of the furlough class 
were prominent by their knowledge and faithfulness to 
duty. The effect of the training course, held before the 
war on the drill grounds, could be plainly felt. Only the 
aims should have been set higher. It must be attributed to 
the youth and incomplete training of many officers who had 
been promoted on the battlefield, that they not always found 
the proper ''tone" in their relation with the older soldiers 
whom they had been with but shortly before in ranks, and 
to whom they were now superiors. The difficulty in pre- 
serving the status as superiors, placed the new officers in 

16 



I 



MOBILE WARFARE 17 

a difficult position often erroneously considered as super- 
cilious ; it probably might have been better to transfer these 
officers, on commission, to other units. It was far from those 
officers to be cold to their subordinates, though they were 
blamed for that. Many of the newly appointed officers 
failed. Instead of setting an example to their men, they 
thought more of their ease and pleasure, which could be 
procured behind the front by any one with sufficient means. 
The subordinate had sharper eyes than the superior. The 
danger to consider single instances as the general rule, lay 
near. It cannot be denied that in selecting the provisional 
applicants for commissions we did not always exercise suffi- 
cient care, and on the other hand, we were frequently too 
slow in promoting acting officers. How frequently, however, 
have we seen that in battle the men preferred to follow the 
youngest lieutenant than the older well-tried noncommis- 
sioned officer. 

The necessity to carefully husband our material for 
commissioned officers which compelled us to cut out a 
"leader reserve" before battle, encountered small compre- 
hension ; the great loss in officers show that very plainly. 

In the course of the World War the conditions of re- 
plenishment became worse. Younger levies at home had 
grown up uninfluenced by the older men in the field. The 
irresponsibility of youth in Germany increased and the short 
time for training at home by officers that were not of the 
best grade offered no counterweight. To this were added 
difficulties in the matter of subsistence. Increases in wages 
made the relations more acute between employers and em- 
ployees and between superiors and subordinates. Through 
this new levy, and through the influence of men of the field 
army, home on furlough, mistrust was sowed between the 
recruits arid the officers, v hich was not fuUy perceived in 
the first two years of the war. It was strange, but in the 
officers, people saw only "Prolongers of the War," who were 
believed to expect higher rank from the longer duration of 
the war, while as a matter of fact, a major general, for in- 
stance, remained a full year longer in that grade than he 
would have in time of peace, before being promoted to lieu- 
tenant general. The talk that officers procured at the cost 



18 DEVELOPMENT OF TACTICS— WORLD WAR 

of the men^ advantages which were denied the men, was 
easily believed. We do not want to deny, by any means, 
that irregularities happened, that through the method of 
appointing officers much was made easier for them. But in 
general the officers can be proud of the fact that they took 
good care of their men. Mistrust existed, factory workers 
believed that preference in the matter of granting furloughs 
was given to agricultural workers, while the latter objected 
to the numerous claims of well paid factory workers for ex- 
emption from service. All this ill feeling was increased by 
the advanced cost of all necessities at home ; and many men 
returned to the field army disgusted with conditions at 
home. 

As was the case during the Franco-Prussian War, the 
losses in officers were extremely high in the first battles, so 
that young officers quickly became company and even battal- 
ion commanders. It was soon perceived, tiiat it would have 
been wise to train every officer in time of peace for the du- 
ties of the next higher grade. Intense desire to attack on 
the part of the troops and bravery of the officers, never fear- 
ing bodily harm, helped the troops to overcome many diffi- 
cult situations. ''All hail to that army where untimely 
bravery is frequently shown ; it may be a rank growth, but 
it is a sure indication of fruitful ground." 

The German defense held to a single line, fortified in 
depth as much as possible — and generally strongly occu- 
pied—sought the decision by the attack of its main reserve. 
Advanced positions, which had been occupied under the ex- 
periences of the fortress warfare (Regulations for fortress 
warfare, August 13, 1910, p. 123) assumed a doubtful im- 
portance and the question arose, due to increasing develop- 
ment of aerial reconnaissance, of advanced positions being 
replaced by false works. Under the conviction that infan- 
try that can attack well, can also defend itself well, we had 
not paid as much attention to the technique of the defensive 
battle as should have been done. Leaders and troops paid 
little attention to the defense. The soldier does not love the 
spade. Troops that had participated in fortress maneuvers 
however were better trained for the close range battle in the 
use of hand grenades and minenwerf ers. 



I 



MOBILE WARFARE 19 

The infantry was armed with the rifle model '98 (mag- 
azine under the barrel and pointed bullet) . The complement 
of machine guns was not sufficient. There were only 6 
guns to the regiment. The fire effect was excellent, but 
the sled mount was too heavy; protective shields intended 
to offer protection to the men in the fire fight and during 
fire pauses, betrayed their location by their size. 

The intrenching equipment had been augmented shortly 
before the war by the addition of an intrenching tool wagon 
for each regiment (field train) containing 230 long handled 
spades. With the small intrenching equipment, suitable 
only for digging trenches of little depth, 1 company could 
construct a trench of 150 meters. The supply of wire cut- 
ters and wood working tools was inadequate. 

Hand grenades, introduced by the Russian infantry, 
were used only by pioneers. 

The "protective coloring" of the field equipment was 
not correctly done. The various arms betrayed themselves 
by different colored insignia. Officers fell because of their 
equipment, order clasps, belts, sashes, and especially the 
officers' long side-arms. 

To each infantry regiment were attached 6 telephone 
squads with 18 kilometers, of light wire (not field wire as 
in case of telephone battalions), and there were an addi- 
tional 12 kilometers of light wire on the infantry ammu- 
nition wagon. Of special importance for mobile warfare 
was the equipment of mobile field kitchens. The Russians 
had these rolling kitchens but not the French nor British. 

The Jager battalions, composed of picked men, and 
intended for the support of the army cavalry, had been 
reinforced by 1 cyclist and 1 machine gun company each. 
They performed excellent services due to their increased 
fire, their training and the excellent material of which they 
were composed. 

The German cavalry had excellent well trained animals, 
though a little sensitive against extremes of temperature. 
The troopers, trained by 3 years* service, were armed with 
the carbine and tubular steel lance as principal weapons; 
the "thrusting" saber, fastened to the saddle was little 
suited to the Grerman cavalry for their "strike and cutting 



20 DEVELOPMENT OP TACTICS— WORLD WAR 

method." Our cavalry could meet the French and British 
cavalry with full confidence. France believed that the Ger- 
man cavalry had a great love for fighting dismounted. The 
regulations of March 4, 1909, rightly warned against fight- 
ing on foot in only a half-hearted manner. These drill 
regulations also mentioned the possibility of carrying on at- 
tacks on a larger scale dismounted (G.D.R. 456) but laid par- 
ticular emphasis on the general rule, that the cavalry fight 
should in general be mounted (C.D.R. 389) ; also that cavalry 
must always try to solve its task mounted. "Only where 
the lance is out of place, the cavalry resorts to the carbine. 

In an engagement of all arms, small cavalry units 

may achieve local success by vigorous action at the right 
moment. Decisive participation in the course of a battle, 
either in the beating off of a hostile attack or in the support 
of our attack, is possible only by insertion of large masses 
of cavalry. The enemy is very vulnerable to a cavalry at- 
tack on his fiank or rear. Merely threatening an attack in 
such a direction has a serious effect on the hostile troops. 
But the cavalry must not be satisfied with mere threaten- 
ing. Operations against the lines of communications of the 
enemy to the rear can have valuable results ; but these opera- 
tions should never divert the cavalry from its battle task. 
If a battle ensues, then the watchword for each large and 
small cavalry unit must be "participation in the victory." 
(C.D.R. 893-395.) Success against unshaken infantry is con- 
sidered possible "if the cavalry has come up close and can 
attack by surprise" (C.D.R. 441). An attack against shaken 
infantry will always be successful. Attack direction and 
attack formation are of less importance in this case. (C.D.R. 
443.) Where possible, charges against infantry from sev- 
eral sides may be executed with depth formation, in which 
the leading waves are deployed. In attacks against artil- 
lery the advice was given to attack the flanks rather than 
the rear. Artillery under cover can be charged in front also, 
under certain conditions, without material loss (C.D.R. 444) , 
and heavy artillery was pointed out as an especially vulner- 
able target (C.D.R. 448). In the service of reconnaissance 
the attack was especially emphasized, the impression created 
on an enemy by the offensive action of patrols and recon- 



MOBILE WARFARE 21 

noitering detachments was valued especially high. It was 
believed that no valuable information could be gained by 
making detours. The enemy undoubtedly would endeavor 
to hide important measures from our view, which could only 
be overcome by force of arms. To foster the offensive spirit, 
stress was laid on the fact that our mounted messengers 
could find their way back only after the hostile reconnoi- 
tering detachments had been driven from the field. 

The composition of the larger units (army cavalry) 
was consistent with the demands of reconnaissance and 
battle ; higher cavalry commanders were instructed to cause 
the consolidation of several cavalry divisions on the battle- 
field into 1 cavalry corps (C.D.R. 523) , in so far as this had 
not been done during the course of the operations (C.D.R. 
229). The cavalry division consisted of 3 cavalry brigades, 
each of 2 regiments, 1 machine gun battalion, 1 pioneer de- 
tachment, and 1 information detachment. Some cyclist 
troops and a Jager battalion were attached. 

During the year prior to the war, the German artil- 
lery had been increased so that each infantry division had 
the disposition of 2 regiments, of 6 batteries and 2 light 
ammunition columns each ; 1 battalion thereof being equip- 
ped with light field howitzers. Corps had the disposition of 
a heavy howitzer battalion of 4 batteries. We could count 
on the attachment of mortars (21-cm.) and heavy guns in 
battle. At the beginning of the World War the field bat- 
teries had 6 guns and caissons ; horse batteries, 4 guns and 
caissons, heavy field howitzer batteries, 4 howitzers with 
8 caissons, and mortar batteries, 4 mortars and caissons. 
During the course of the war all field batteries were limited 
to 4 guns, without any disadvantages accruing thereby, as 
far as military literature shows. In the 6-gun battery 
there were 136 rounds per piece available, in the French 
4-gun battery 312 rounds for each piece. The field gun, 
model '96, with its very effective shrapnel^ was considered 
the main fighting weapon, and less importance was attached 
to the shell, as shown by the supply thereof carried by the 
battery. Tlie light field howitzer with high explosive shell 
was to be employed principally against artillery targets be- 
hind cover, villages, and troops in dense forests. 'It is able 



22 DEVELOPMENT OP TACTICS— WORLD WAR 

to pierce most of the cover found in the field, and to fire on 
captive balloons and airplanes at high altitude." (F.A.D.R., 
356.) The heavy field howitzers, drawn by heavy draft 
horses, broken to fire, was considered an especially effective 
gun to fire on artillery under cover and on strohgly fortified 
positions. Only when the crisis of battle demanded it were 
the heavy field howitzers advanced at the trot. The princi- 
pal projectile was the percussion shell (instantaneous and 
delayed fuse) . The employment of the mortar battery was 
made more difiicult by the parts of the gun being carried 
separately on the march — ^barrel, cradle and trail. It re- 
quired about five minutes to assemble the piece. Equipped 
with a mobile and efiicient heavy artillery the German AJrtny 
was presumed to have a material superiority over its oppo- 
nents. It could, ''delay by its long range pieces the advance 
of thi^ enemy; could force him to make detours; could block 
or keep open defiles." 'It also can take up the battle against 
heavy hostile artillery and thereby facilitate going into posi- 
tion of light batteries at effective range." (F.A.D.R. 388.) 
As the battery that is in the open can be rapidly destroyed, 
the development of the telephone and the indirect laying 
method, caused the field artillery to generally be found fir- 
ing from covered positions. Causing the officers of the fur- 
lough class to attend courses of instruction in field artillery 
schools of fire, had disseminated knowledge of indirect lay- 
ing and the new firing procedure to a very great extent. 
Regulations demanded that the field artillery should at 
times disregard the advantages of covered positions, and 
should fire from the open to bring about the decision in the 
infantry battle. (F.A.D.R., 367.) In such emergency even 
the hottest infantry fire was not to be feared. Emphasis 
was laid on the fact that the battle activity of infantry and 
artillery must be of the closest and that the main task of 
the latter is to support the infantry (I.D.R. 44, 446, F.A.D.R. 
366). Co-operation of these two arms was demanded, but 
it was left to the training in peace time to find the suitable 
means therefor. (I.D.R. 447, F.A.D.R. 368.) That require- 
ment left much to be desired. We intended to force the de- 
fender to leave his position and show his troops by having 
our infantry advance and attack (F.A.D.R. 494, H.A.D.R. 



MOBILE WARFARE 28 

464). Then the task was left to the artillery to annihilate the 
enemy with its sheaf of projectiles. Contrary to this method, 
the French Regulations contemplated but a portion of the 
artillery firing on the attacking infantry, while the mass of 
the guns were to engage the hostile artillery. Conduct of 
the artillery battle was left to the commander of the artil- 
lery. Employment by regiment and by battalion was the 
rule, without however prohibiting the employment of sin- 
gle batteries and platoons. It was recommended that in- 
stead of dispersion, employment by groups be resorted to, 
for the better utilization of the terrain for facilitating fire 
control (F.A.D.R. 366) . Hostile artillery was to be engaged 
by field fire and heavy artillery combined, though we did not 
deceive ourselves in thinking that locating batteries under 
cover would be easy. If we did not succeed in locating the 
hostile batteries, firing on observation posts that were visi- 
ble might promise success. (F.A.D.R. 438.) In any case the 
infantry had to become used to the belief that the attack 
must be made even when the artillery did not succeed in 
gaining fire superiority. 

The infantry division formed the battle unit; 2 to 6 
infantry divisions were formed into an army corps or group, 
which could be reinforced in artillery according to the sit- 
uation. All fighting arms were combined in the infantry 
division, while corps headquarters reserved to itself the dis- 
position of the heavy artillery, information detachments, 
columns and trains. Dividing the infantry division into 
2 infantry brigades of 2 regiments and 1 artillery brigade, 
also of 2 regiments (a corresponding number of infantry 
and artillery regiments) undoubtedly assisted in the proper 
co-ordination of these arms. Tliis co-ordination was inter- 
fered with by the assignment of a light field howitzer bat- 
talion. Tlie organization of the infantry into 2 brigades 
was disadvantageous in so far as the brigade unit became 
disrupted in deploying for battle from the march column. 
One brigade commander became superfluous. A better so- 
lution, and one corresponding to the battle object, was at- 
tained by decreasing the units into 3 infantry regiments 
and 3 artillery battalions (1 of the latter a light field how- 
itzer battalion) . Experience soon showed this organization 



24 DEVELOPMENT OP TACTICS— WORLD WAR 

corresponded best with the demands of battle, even if the 
overhead, in comparison with numbers of troops, became 
larger. Many instances happened where field guns and 
light field howitzers were found in the same battalions. 

It was a grave error to send the 22d and 27th Reserve 
Corps in October, 1914, to Flanders and Poland, after a short 
period of training, without attaching strong artillery to 
them. Only a few commanders, with war experience, could 
be assigned to them by transfer, at the last hour and the 
number of seasoned soldiers in those corps was very small. 
Intense enthusiasm could not offset training and equipment. 
These corps went to pieces in Flanders under heavy losses. 
This was the more serious, as in these corps the men were 
from the well educated classes and we lost considerable ma- 
terial that could have been used later on for commissioned 
officers. We thus gained the same experiences as we did in 
the Wars of Liberation, in forming volunteer corps. Events 
of war proved that new organizations of young troops, Land- 
wdir and Landstrum, require a more liberal supply of ar- 
tillery, to give them the same fighting power, as active and 
reserve organizations. No one can foretell in arranging war 
organizations when units will be compelled to fight at a de- 
cisive point. 

The divisional cavalry originally organized into 3 squad- 
rons, still possessed a certain fighting strength. This 
strength proved itself unnecessary the more the cavalry 
confined itself to message service and near reconnaissance. 

In the matter of technical troops, only 1 pioneer com- 
pany with division bridge train was attached; this sufficed 
for mobile warfare, but was not sufficient as the change 
was made into position warfare. There were no division in- 
formation detachments, and only 1 sanitary company. 

Ersatz units, at mobilization received an insufficient 
instructor personnel. When the loss in officers became so 
great these units very soon lost their active officers by 
transfer. This loss was very bad for the training of the 
Ersatz troops. The retired officers detailed as instructors, 
did not comprehend the requirements of modern war. This 
increased the difficulties in training at home, so that units 
in the field were satisfied to have only primary drill and 



MOBILE WARFARE 25 

training take place at home, while the actual training for 
work in the field was undertaken by divisions independently, 
in field recruit depots gradually established. In any case 
these replacements could be furnished quicker from divi- 
sion . depots than to be brought from home. For this 
the army authorities should have made arrangements in 
time of peace. All experiences gained in war proved that 
field recruit depots were necessary in the theater of war. 
The training in the field recruit depots was enhanced by 
the military atmosphere; by the immediate use of cam- 
paign experiences and lessons ; and by life within the midst 
of hostile peoples. According to the progress of their train- 
ing the men were formed into companies. Strict orders had 
to be issued that the men subsequently were sent to join 
their former organizations or in case or recruits, to whose 
home district they belonged. This was especially important 
in consideration of furthering esprit de ccnrps. In the battles 
on, the Hochberg near Rheims in 1917 one battalion of my 
division was almost annihilated. Within three days it had 
been filled up again from our field recruit depot, and by 
transfer of officers, and was again ready for employment. 
We generally allowed eight weeks for training in the Er- 
satz battalion, four of which was for company training. In 
that short a time nothing thorough could, of course, be ac- 
complished. The greatest difficulties encountered were to 
make good the losses in officers. The Ersatz battalions com- 
menced very early to train younger acting officers and of- 
ficer aspirants. As compared with the old noncommis- 
sioned officers, who in most cases were far more advanced 
in years, they lacked experience in military service and in 
the field. It might probably have been better to promote 
efficient first sergeants and sergeants major to commis- 
sioned rank, as rapid promotion of young men causes bad 
blood very quickly. In the field mental education alone is 
without value. If the young officers were assigned to a 
unit of well trained troops, their value in the field grew 
gradually. It was no illusion whatever that such a short 
time of training could not suffice to create a unit of war- 
trained troops. A superficial training, deceiving the lay- 
man, can be attained, but we can never create the cohesion 



26 DEVELOPMENT OP TACTICS— WORLD WAR 

and the morale, which assures the primary battle value of 
troops. Campaigns clearly showed the importance of a well 
trained peace army. Even the training of Kitchener's 
armies does not justify a militia system. Training was had 
under the pressure of military events. There was a nucleus 
of units that had been trained in peace, furthered by the 
sporting sense of the entire nation. The primary training 
was not for the difficult mobile war, but for mere position 
warfare, in which the troops were supported by superior 
artillery. The war situation permitted the high command 
to designate the time for recruits to be sent to the front 
according to the status of their training. Complaints that 
the army, with its young officers, was not able to cope with 
the difficult situations, were frequently heard. Sporting 
rifle clubs and military training in the schools and colleges 
of the nation are absolutely necessary for an army that has 
but short period of service in peace time. Athletic and 
similar training are not only for bodily training, they are, 
for the mind also, of great value. They not only further 
the will power but demand within a stated time the great- 
est exertion of all mental and physical powers. Thus they 
directly increase the value of our youth for service in the 
field. 

Labor Battalions modeled after those of Japan, were 
organized by us early in the war. These units relieved the 
fighting troops from detaching large numbers of men and 
preserved their fighting strength. The Russians learned 
this lesson to their sorrow through the difficulties they en- 
countered in bringing up recruits in Eastern Asia. Men, 
who were unsuited for active service in the field under arms, 
were enrolled in the labor battalions, and under efficient 
officers, performed excellent services. 

The infantry, as a matter of course, had to bear the 
heaviest burden. In active officers of infantry, the 7th 
Army Corps lost 70%, 60% of officers of the furlough class, 
and 40% of the men. The corresponding numbers in field 
artillery were 45%, 35% and 7%. 

The tactical views of the Austrian Infantry, after they 
had been purged of false ideas of the lessons of the Boer 
War, were in general in accordance with the tactical views 






MOBILE WARFARE 27 

of the German leadership. The Austrians did not fully un- 
derstand the method of German training, in which prime im- 
portance was laid on good training in firing and marching. 
However a portion of the Austrian Infantry was well trained 
for mountain warfare. In time of peace each regiment had 
2, in war 3 machine guns with tripod and collapsible shields 
carried on pack animals. It was believed machine guns 
could replace artilleiy at short range and their employment 
in defense as a fire reserve in widely separated positions 
was recommended. Regulations of 1911 laid emphasis on 
the value of training in close formation in order to increase 
discipline. This however without limiting the front to be 
covered in the various exercises. The Regulations desig- 
nated ''the inflexible will" as the power that would mainly 
decide the battle: — ''In case the attacker does not succeed 
in gaining fire superiority, even after inserting all his 
rifles, the commander must always remember that even in 
apparently hopeless situations success will come to the side 
which is the more stubborn. Tenaciously persisting is far 
better, and causes less losses, than retreat When dur- 
ing a costly, stubborn infantry engagement the enervating 
influences of the battlefield have caused the utmost exhaus- 
tion on both sides, that side will gain the victory whose iron 
discipline and stronger will power enables him to better 
withstand that exhaustion, and who continues the fight 
with unshakeable persistence, until he has forced the enemy 
to desist Upon the sure, firm will of the comman- 
der reflects all actions of the troops. If he vacillates in his 
decision, his uncertainty is transmitted to his subordinates. 
He must let his will permeate to his subordinates, and must 
always take care that his will turns into action." The im- 
portance of independence was especially emphasized. In 
the excellent words of our field regulations, par. 38, "Inac- 
tivity is criminal." It was pointed out that the decision 
being made and knowing its objective, the energetic exe- 
cution of the decision makes high demands on character. 
That often after success these demands must be increased. 

"After every battle the vital point is for the comman- 
der, and for the troops as well, to force themselves to over- 
come the mental, physical and moral tension, which, after 



28 DEVELOPMENT OP TACTICS— WORLD WAR 

f atigrue and danger, may easily lead to being satisfied with 
a half-success. Only a f orcef ul» inflexible will can overcome 
this weakness. In most cases the commander must demand, 
after the success, a further extreme effort by all forces to 
complete the victory and annihilate the enemy. Only thus 
can he avoid renewed heavy sacrifices in subsequent bat- 
tles." 

The regulations of 1903, written under the fresh im- 
pressions of the Boer War, placed too high a value on fire 
effect, and credited fire as the only decisive element. The 
regulations of 1911 took into account the fact that the oppo- 
nent would display tenacity and persistence, which all lead- 
ers demand. ''Infantry, imbued with the will to attack, 
physically and morally well trained and well led, can fight 
successfully under the most difficult conditions. The infan- 
try batters down the enemy with its fire, then with the bay- 
onet breaks down his last resistance.'' Good infantry must 
always look to the bayonet fight as the last resort, it is fre- 
quently indispensable to gain the decision. The final appeal 
to the bayonet, omitted in prior regulations, is found again 
in the new regulations. In its preparation for war the army 
suffered much from the small appropriations. A portion of 
the infantry (Landwehr) consisted mainly only of skeleton 
organizations so that the excellent plans for training of the 
drill regulations were of little use, the more so as the re- 
plenishment of officers had not been sufficiently prepared. 
The number of guns and machine guns of units did not cor- 
respond to the requirements of modem times. Still less 
than in Germany was the population, fit to carry arms, made 
available for service in the army. 

The Austrian Cavalry was excellently trained as such. 
Its regulations corresponded to those of the German cav- 
alry. The artillery had been for years the ^te arm. Theo- 
retically well trained, it suffered from having a small num- 
ber of guns. In an army corps there were only 8 gun bat- 
teries, 2 light and 2 heavy field howitzer batteries, a total 
of 84 guns against 160 in the German army. Austria's ar- 
tillery had not been sufficiently trained for the tasks of fire 
control in large scale battles. 



MOBILE WARFARE 29 

At the opening of the war the Imperial Austrian army 
was not a valuable mstrument for war. The value of the 
army was decreased by the composition of the States from 
which men, with different racial qualities, and different poli- 
tical aims were drawn. German and Hungarian troops were 
decidedly superior to Slavic troops. High losses in the first 
battles, for the replenishment of which little preparations 
had been made, the absence of an old, well-tried corps of 
noncommissioned officers, the mixture of non-German with 
German troops, with no common language, the unreliability 
of the Slavic troops, an indisputable effeminacy on the part 
of the officers, decreased the value of the army still fur- 
ther. ''In the instruction of officers training in will power 
was neglected, knowledge and education were rated higher. 
More dutiful subordinates, than independent, forceful sup- 
eriors were trained ; we systematically accustomed the offi- 
cer to be dependent and to await leadership. In the general 
staff, knowledge was the prime requisite ; the general staff 
dominated the forms of troop leading and the issue of or- 
ders, but was insufficiently acquainted with the instrument 
with whose help it intended to utilize its knowledge.'^ In 
1916 an exchange was tried with German officers; it was 
hoped to give the Austrian army a greater degree of power 
of resistance through the increased influence of Grerman offi- 
cers in its training, and through the insertion of Grerman 
troop units into the Imperial Austrian units of the army. 
But it was too late for that. The army lacked strict disci- 
pline, lacked the enormous driving power, and lacked the 
sense of duty, of the German troop units which never failed 
in the first four years. 

TACTICS OF MOBILE WARFARE 

The rapid advance of the German Armies in the West 
made a material change in the method of fighting impossi- 
ble ; the troops attacked as they had been taught in peace ; 
after the first experiences the skirmish lines were kept thin- 
ner. Troop leadership sought to give plenty of time for the 
absolutely necessary battle reconnaissance, and for a better 
co-operaljon with the artillery. Proper utilization of the 
terrain progressed under the stimulus of the hostile fire. 



30 DEVELOPMENT OF TACTICS— WORLD WAR 

On the eastern front the training of the German infantry, 
was so superior to its opponent that here also a change in 
the' method of fighting was not considered necessary. It 
must be said that the Russians had many excellent quali- 
ties. Its infantry especially showed the benefit of the ex- 
periences gained in the Busso-Japanese War in its use of 
the hand grenade and field fortifications. While demanding 
the strictest discipline, our regulations had educated the 
troops in initative and the will to attack. It had given them 
the means to adapt themselves to the changing forms of 
battle. Where high losses occurred, the reason could in- 
variably be traced to non-observance of the maxims laid 
down in regulations. 

A material difference in the conception of the infantry 
attack in Germany and in France existed prior to the World 
War in the value laid on utilizing infantry fire. Regula- 
tions for the training of the French infantry published im- 
mediately prior to the World War did not mention the neces- 
sity of gaining the infantry fire superiority, the artillery 
rather was to hold down the hostile guns, and then to facili- 
tate the advance of the infantry by heavy fire on the de- 
fender's infantry. "Infantry is the main arm; it fights 
through fire and on the move. Only the forward movement, 
leading to hand-to-hand fighting, is decisively irresistible, 
and usually an effective and strong fire has to open the road 
to pierce the enemy. .... Artillery fire, which has only a 
very slight effect on an enemy under cover, can never by 
itself drive the defender out of his position. The advance 
of the infantry must compel the opponent to show himself 
and to offer targets. The artillery supports the advance of 
the infantry, by annihilating everything that might hold up 
that advance." Infantry must resort to its protecting arm 
if the hostile fire prevents a continuation of the forward 
movement, which, as soon as possible, should be resumed. 

Our German regulations considered "the most excellent 
means for working up, (prerequisite to going forward) to 
be gaining fire superiority" (170), "which will be perceived 
by the decreasing hostile fire, or by the enemy firing too 
high" (336). In this, errors were possible, as fire pauses, 
ordered by hostile commanders, due to firing with incorrect 



MOBILE WARFARE 81 

sight elevations, might only too easily be mistaken as due 
to the results of our fire. The demands of the Infantry Drill 
Regulations (874), to not make the execution of the infan- 
try attack mainly dependent on first gaining artillery fire 
superiority, seldom furnished a basis in field exercises to 
execute an attack with our own infantry weapons. We 
waited until our attacking artillery had gained fire super- 
iority. During the Russo-Japanese War it was reported of 
Japanese attacks that they had been executed without the 
Japanese artillery having gained fire sui)eriority. Reports 
from the Balkan war made us study the statements accord- 
ing to which Bulgarian and Serbian infantry had worked 
their way up to within 200 to 300 meters of the enemy with- 
out firing a shot. By deplo3ring into small detachments and 
under efficient support by the attacking artillery, a well 
trained unit might get close to the enemy and take up the 
fire fight only at close range. This was successfully achieved 
by the 43d Infantry Brigade (General v. Hiilsen) on Sep- 
tember 9, 1914, at Gerdauen. 

French reports available at this time praise the rapid 
advance of our infantry and their excellent utilization of 
the terrain, without waiting for the artillery preparation. 
They state that the fire effect of the infantry was annihil- 
ating and that the machine guns were used effectively at 
every opportunity and with surprising skill. The effects 
of the field artillery were considered less favorable. The 
reports further say that the superiority of the Grerman 
tactics was especially noticeable in the open terrain during 
the battle of Saarburg. The hostile divisions were anni- 
hilated in a frightful frontal attack, though the French in- 
fantry utilized the terrain most excellently and could barely 
be distinguished. The Alpine Chasseurs and the colonial 
troops fought very well. TOie skill of the French in defend- 
ing a village was especially noticeable, While at the start 
the fire of the French infantry and machine guns did not 
come up to our expectations. That may have been because 
we valued the artillery effect too highly and believed that 
it could overcome every halt that was forced on us. Though 
the fire was good by itself, the French sought every oppor- 
tunity to attempt a flanking movement, and was seen to be 



82 DEVELOPMENT OP TACTICS-EWORLD WAR 

especially adept in finding good firing positions, unperceived 
by the enemy, in houses, bams, sheds, etc. 

The French positions also had a great depth, the lead- 
ing line consisting of single rifle trenches, connected by 
false works, and the foreground excellently arranged. Farm- 
ing implements that had been left in the fields, scarecrows, 
and trees were utilized to designate the range. 

In the British Infantry were found many excellent 
marksmen, who acted independently. Much attention was 
given to flanking flre; loopholes and embrasures in the 
trenches, arranged so as to allow firing obliquely to the line 
of retreat had the preference ; trenches were generally well 
hidden, were constructed in sawtooth shape or wave-like, 
so that the attacker could be taken under oblique fire, while 
the enemy's fire, straight from the front against the loop- 
holes had no effect. The trenches usually lay 300 to 600 
meters below the crest or behind hedges. The troops made 
much use of night firing. 

Subsequent to the first battles much was heard in Ger- 
many of the insidious conduct of the French and Belgians ; 
it was said that each soldier (?) had a suit of civilian cloth- 
ing in his knapsack, to evade difficult situations or to carry 
on the war as a bushwhacker. Long range ricochets caused 
the suspicion that inhabitants had fired from houses, and 
it was also assumed that the entire population took part in 
the information service. Cases of that kind certainly did 
happen, but should not be taken as the general rule. There 
is no doubt that the Germans were much inferior to their 
opponents in the emplo3anent of permitted and forbidden 
war ruses and tricks, and our men did not consider it fair 
that the enemy should resort to placing sharpshooters in 
trees, and neither did they consider the conduct of the 
British sharpshooters posted in cabbage fields as permissible. 
At the start our men were not suspicious enough and they 
trusted the inhabitants too much. Subsequently they saw 
in each inhabitant a traitor. 

The German rifle fire had an enormous effect, probably 
being the result of carefully firing each round in the manner 
taught in peace in connection with the training in the con- 
tinuous rifle fire, increasing and decreasing according to 



MOBILE WARFARE 88 

the situation. The main fire fight was carried on at about 
800 meters range, then the lines advanced in long, broad 
rushes, to close with the enemy quickly without awaiting 
for our artillery to gain fire superiority in each instance. 

Under the impression created by this advance, hostile 
fire effect soon diminished, and in most cases the French 
did not await the contact but retreated when the attacker 
came to within 500 meters, thus offering excellent targets 
to the artillery. Our troops felt superior to the enemy in 
short range fighting. The example set by the officers and 
by a few courageous men, who in time of peace had not 
been numbered amongst the best subordinates, was of de- 
cisive influence. In battle everyone went straight ahead and 
everyone fired straight to the front. In the very first bat- 
tles the importance of independence could be seen. 

Fire control, developed in peace time to perfection with 
its precise words of command, was practiced only in the 
most favorable instances at the opening of fire. The noise 
of battle is generally such that we cannot hear our own com- 
mands. Targets — almost exclusively a strip of the terrain 
from where the noise of firing seems to come — and eleva- 
tion, will be designated by the platoon commander as long 
as practicable, but very soon the fire control will slip out of 
his hands, and his place is taken by the squad leader and 
finally by the individual skirmisher. This is of no very great 
importance, provided the individual correctly observes the 
effect, husbands his ammunition, increases the fire rapidity 
when the target becomes more favorable, and decreases it 
when the target fades, and ceases fire at once when the 
target disappears. The command '^e slower, fire quicker'' 
merely indicates inattention or insufficient training. 

In carrying the fire forward to the enemy, in gaining 
the fire superiority by infantry, supported by the fire of 
the artillery, the Infantry Drill Regulations saw the surest 
means of success, and demanded that infantry open fire 
only at mid range even in a terrain devoid of cover. In bat- 
tle the conviction gained ground more and more, that to 
bring the defender into a condition where he could be 
charged, was the princiiml task of the artillery and that, 
considering that the infantry could rapidly entrench, fire 



34 DEVELOPMENT OF TACTICS— WORLD WAR 

preiMuratioii by infantry was a minor matter. Besides the 
terrain, the possibility of our own and of the enemy's fire 
effect controlled the manner in which infantry worked it- 
self up to charging distance. But even under the most 
favorable conditions artillery will not be able to open a 
road for the infantry to victory, even if reliable connection 
is maintained during the entire attack between the two 
arms. 

Insufficient support of artillery forced the infantry to 
resort to the spade to hold the ground it had gained. The 
infantry soon perceived that incautious conduct on the part 
of the staffs and troops soon drew the hostile artillery fire, 
and that fire was then directed also on points where in the 
opinion of the enemy commanders, German troops were 
moving and where they might be halting. The fire was 
then in the shape of a sweeping fire up to the longest ranges. 
This occasionally caused material losses to an incautious 
unit, but a cautious unit could avoid it, as the method of 
fire was distinguished by regular uniformity. The French 
artillery designated buildings, woods, and rows of trees as 
"artillery traps." Frequently the fire commenced with a 
well prepared fire-surprise and the effect then might be 
very great. The artillery took pleasure — ^being covered in 
front against the German artillery — in firing into neighbor- 
ing sectors. 

Reconnaissance and scouting that attracted no atten- 
tion was now of importance, in order to cross the long and 
mid-ranges by smaller units with irregularly formed skir- 
mish lines, to gain a firing position at about 400 to 500 paces 
from the enemy. The skirmish lines were made denser by 
supports rushing forward from cover to cover. It was also 
found to be well to bring the reserves up in skirmish waves. 
Massing the reserves at points that could be seen from afar, 
proved to be an error. The success of the attack was based 
on the efficiency of the lowest commanders; battalion and 
regimental commanders had their hands full in inserting 
the troops and keeping up connection with the artillery, as 
well as with efficiently bringing up the reserves, and could 
not therefore supervise all the minor details of troop lead- 
ing. 



MOBILE WARFARE 86 

Night engagements very soon assumed increased im- 
portance; their success frequently became questionable by 
reason of insufficient preparation and by the fact that no 
attention was being paid to the lessons of war that had been 
compiled during peace. Dense skirmish lines, with scouts 
far to the front, were found to be of advantage, or any for- 
mation in close order with narrow front extension. It was 
found best to execute the advance with pieces unloaded. 
All distances were shortened. The decision was sought in 
the charge with cold steel and without shouting. If the 
enemy was encountered, an immediate charge was always 
of advantage. As the enemy undoubtedly was prepared to 
fire, it was well to draw his fire by false movements and to 
conduct the charge in another direction, but in any case to 
clear the roads. Charges that were to pierce deep into the 
enemy, required formation in depth. Conduct in case of 
artificial illumination (agreed-on signals or signs for our 
own illumination or our own information) and co-operation 
with searchlights (light signals, flank protection) required 
special training. In the defense, fire must be opened only 
if the enemy is recognized beyond doubt. The practice, re- 
commended by Regulations, of laying guns and rifles during 
daylight for firing at night, was hardly ever resorted to; 
night fire was as a rule executed by machine guns, not by 
riflemen. Frequently it was found better not to occupy the 
i^irmish trenches, because the defender, standing lower was 
of a disadvantage against the attacker standing higher; it 
was found to be best during the night to have the unoccu- 
pied trenches as an obstacle in front of the line. 

The newly organized reserve corps employed in Flan- 
ders in October, 1914, did of course utilize the experiences 
of the Western front, but suffered from the difficulties in- 
herent to all new organizations, when they have been insuffi- 
ciently trained and Incompletely equipped. The first battles 
caused heavy losses among the young, inexperienced troops, 
but by early summer of 1915 they had been trained in the 
following attack method. Approach to within about 800 
meters of the enemy with patrols, which reconnoiter and 
find out everything necessary for the battle activity. Ad- 
vance by the company on a narrow front in skirmish waves, 



36 DEVELOPMENT OP TACTICS— WORLD WAR 

companies 200 meters in rear of each other. Intrenching in 
the first firinfiT position, reinforcing with machine guns, and 
men making the firing line denser to an interval of two 
paces between skirmishers. Working forward by squads 
up to the next firing position, which again will be entrenched. 
From the firing position close to the enemy the charge starts 
under all available fire protection, the captured position be- 
ing immediately arranged for defense against a counter- 
attack that is sure to come. 

The French First Army (Dubail, November 27, 1914) 
demanded that troops, as the result of the first lessons 
gained in the war, abandon, the march column 10 to 12 kilo- 
meters from the enemy and continue the further advance 
deployed; that the infantry, in open terrain, should not 
show any unit in close order at 10 kilometers range from the 
hostile infantry, but work up to 500 meters in the smallest 
units and these intrench when they could no longer advance. 
In covered, close terrain and in hazy weather, each battalion 
in the first line was to send ahead 1 company, followed by 2 
companies echeloned to the flanks, and followed in turn in 
their center by the reserve company. For fighting in woods 
it was laid down that the companies in the second line should 
at once turn against the enemy. It was recommended that 
in night fighting an advance be made by half platoons in 
column of fours. 

ft 

A German report concerning the French method of at- 
tack supplements these Regulations. "Frequently individ- 
uals rushed forward, assembling again in squads at the 
nearest cover. Stretches of open terrain were crossed in 
this manner in very thin lines, echeloned, and offering a 
very poor target. The supports and even the reserves fol- 
lowing the skirmishers separated into small groups, never 
more than a platoon, with large intervals. The endeavor 
seemed to be to reach the mid ranges without material losses 
and there to form skirmish lines that could take up the bat- 
tle in force. The fire of these skirmish lines usually was 
very strong, but was of little effect as soon as it met bur in- 
fantry fire." 

The French infantry was excelleijtly supported by its 
artillery, which utilized its long range guns to the utmost. 



MOBILE WARFARE 87 

Its skill was very great in finding and taking up covered 
positions, in frequently changing positions and in observa- 
tion. Batteries, platoons and single guns took position with 
very irregular, different intervals and in echelon. It appears 
that the French entirely abandoned the normal position 
with regular intervals. Detaching of platoons or pieces for 
the direct support of the infantry or for the purpose of 
flanking the leading lines was frequently observed. The 
French laid the greatest value on flanking fire. The en- 
deavor was very plain on the part of the French artillery, 
to cover itself f rontally against German artillery firing 
straight to the front by this method of taking position, either 
using the terrain (ridges, villages, buildings) or by the use 
of masks (strips of wood, rows of trees, hedges). Each 
French gun carried a ^aw in its limber, by the use of which 
it procured for itself a line of sight through these masks 
and, covered f rontally, fired obliquely, and mostly into the 
firing sector of its neighbor, and thus effectivdy flanked 
the neighbor's attack target. Therefore the French bat- 
teries were hard to locate by the Glerman batteries, and could 
but seldom be perceived by our infantry in the front line, 
causing it frequently to assume that it was being fired on by 
its own artillery. 

CHANGES IN ORGANIZATION, EQUIPMENT 

AND ARMAMENT 

Based on the experiences of the war, changes in or- 
ganization and armament had been made in the summer of 
1916, which were to make their influence felt on the method 
of fighting. The divisions were organized into S regiments 
of infantry, 1 regiment of artillery of 9 batteries (3 of 
them light field howitzers) and 2 pioneer companies; the 
strength of the divisional cavalry had been reduced to 1 
squadron, but the information troops had been permanently 
increased. Heavy artillery was assigned to the division 
from time to time according to need, so that in tranquil posi- 
tion warfare each division generally had 3 heavy field how- 
itzer batteries, 1 mortar battery and 2 heavy gun batteries. 
The number of machine guns had been materiidly increased, 
each battalion receiving 1 machine gun company of 6 guns. 



88 DEVELOPMENT OP TACTICS— WORLD WAR 

By changing the construction of the gun carriage (wagon) 
the guns per company could subsequently be doubled. The 
introduction of automatic rifles whose importance the Ger- 
man infantry had found during the fighting on the Somme, 
took place in the Summer of 1917 ; the introduction of ''fir- 
ing cups" for throwing grenades from the rifle, took place 
in the Spring of 1918, after specially constructed grenades 
had proved to be too heavy for mobile warfare. 

Each man was protected by a steel helmet, in addition 
received a gas mask and in many cases a long-handled spade, 
which was valued also as a means of hand-to-hand fighting. 
Every man was equipped with hand-grenades ; the originally 
adopted ball of disc grenades were not good, and they were 
replaced by grenades more easily handled, and the lighter 
^gfS grenades, which were carried in the attack in sandbags 
slung across the shoulder. There is no doubt that both sides 
overestimated the effect of the hand grenade and attached 
more value to it than to the rifle. An automatic pistol — 
"long pistor — (16 per company) was supplied for trench 
warfare. For connection with the artillery colored light 
rockets served, and for connection with airplanes colored 
lights and large cloths (panels) were used. Shocktroops, 
minenwerfers and flame throwers taken from the position 
war, could also be employed in the mobile war. The artil- 
lery had made great progress in flash and sound ranging 
methods. Much more use was made of the shell, than had 
been expected before the war. In gas shells artillery pos- 
sessed a fighting means, independent of the direction of 
the wind, to neutralize hostile batteries and to gas stretches 
of the terrain. Airplanes increased in importance through 
armament, equipment with cameras and wireless apparatus. 
The means of communication were materially improved. In 
addition to the telephone there came into use intermittent 
lights, ground and wireless telegraph, light and sound sig- 
nals, information projectiles, carrier dogs and pigeons. In 
February, 1918, the divisions received wireless battalions 
(also for ground telegraphy), and the personnel with the 
troops was organized into troop message detachments as 
follows : each infantry regiment, had 1 regimental informa- 
tion platoon, consisting of 1 officer, and 13 men; each in- 



MOBILE WARFARE 39 

f antry battalion had 1 battalion information platoon, consist- 
ing of 1 officer and 21 men ; each independent battalion (cav- 
alry rifle regiment) 1 battalion information platoon of 1 
officer and 30 men. Employment, traffic and co-operation 
was in charge of the information officer under the direction 
of the regimental (battalion) commander. 

As early as 1916 the French commenced to organize 
their battalions into 8 infantry and 1 machine gun com- 
pany of 8 guns, and to attach automatic rifles — ^up to 16 — 
to the companies. By the law of September 27, 1916, the 
company organization was fixed. Each company, not count- 
ing officers, had a subsistence strength of 194, the fighting 
company proper having a strength of 168 men. This num- 
ber did not include 4 older noncommissioned officers and 22 
men, used for runners, signal men and infantry pioneers. 
The men throwing hand and rifle grenades were designated 
^'grenadiers," the gunners of the automatic machine guns 
were designated 'fusiliers," and the rest as 'Voltigeurs." 

The fighting company was divided into 4 platoons of 2 
half-platoons (sections) each. Of the half-platoons, the 
first always contained rifle grenadiers, hand grenade throw- 
ers and fusiliers, the second contained the voltigeurs, 4 
rifle grenadiers and 2 cartridge carriers (paurvoyera) . All 
men were trained as hand grenade throwers. The fighting 
company numbered 32 hand grenade throwers, 16 rifle gren- 
adiers and 8 (12-16) gunners for automatic machine guns. 

The rifle grenade (Vivien B easier es) flred from a flring 
cup, was very effective. With 16 rifles arranged with the 
appliance a barrage could be thrown with 160 rounds per 
minute at from 80 to 160 meters. In village flghting they 
replaced absent artillery, cut off the retreat of the enemy, 
prevented supports coming up, and defeated counter-at- 
tacks. Fire unity was sought. The possibility of fire by the 
automatic rifle on the move forced the enemy under cover 
in the flnal phases of the charge. Of course, the machine 
gun could not be entirely replaced, but the automatic rifle 
was especially useful in accompanying the infantry, to se- 
cure terrain that had been taken, to stop counter-shocks, 
and enable the bringing up of machine guns without undue 
haste. 



40 DEVELOPMENT OP TACTICS— WORLD WAR 

The 37-mm. gun attached to the battalion was so mobile 
that it could follow the infantry in all battle situations ; its 
fire was exact, easy to regulate, had a range up to 1500 me- 
ters, and could be fired from a covered position. The pro- 
jectile had the effect of a shell, as solid shot pierced three 
sandbags, or steel plates. It was especially suited for anni- 
hilating invisible machine guns. In the attack waves the 
gun was not to be used, as there it would be easily seen and 
destroyed. 

In order to gain the highest efficiency of the company 
after its first organization, all war implements must act in 
co-operation, the supply of ammunition must be assured 
and the complement well trained. In the training the ma- 
chine guns and the 87-cm. gun directed their fire on targets 
above ground, the hand and rifle grenades against skirmish- 
ers or targets under cover. 

In the deployed line the interval between files was 4 
to 5 paces. The platoon had an extension of 60 to 75 me- 
ters. In depth it was formed in two waves following each 
other at 10 to 15 meters distance, the grenadiers and fusil- 
iers in the first wave, the grenade carriers and voltigeurs 
in the second. They were followed by the nettoyeur» 
(trench moppers-up) at 10 to 20 meters distance, and the 
latter, at a distance of 40 to 50 meters, by the company 
reserves. The fighting front of one compaqiy was 800 
meters. 

By the law of September 10, 1917, the peculiar organi- 
zation of the platoon was even extended to the half -platoon 
''as after the filling up of the first line the battle is con- 
ducted by half-platoons ; therefore these should not only 
in case of need, but at the very start, combine all infantry 
auxiliary and fighting means and have the auxiliary weap- 
ons within themselves be able to take formation in depth.'' 
Each half -platoon had 1 sergeant and 14 to 18 men, with 
one automatic rifle attached. After the automatic rifles 
had been distributed to the half -platoons, the 8 guns re- 
maining comprised the materiel reserve — 4 to the company, 
4 at the disposition of the division. There was no differ- 
ence made in half-platoons between grenadier (hand gren- 
ade throwers) and voltigeurs. The half-platoon consisted 



MOBILE WARFARE 



il 



French Battalion in Attack Formation 
Extension — 400 to 500 meters 



1st Company 



2d 



2d Platoon 



l8t W« 



I 

I 

S^on I 



let Platoon 
Section 



jf G ^ M a 



2d 

8d WaTB *• i 



jrfS9 



I « 

4tfa Platoon 
S C^M C S 



4th WaTa 

5111 wa^;J; u^;^ 

I 



2d Platoon 
Section 



lat Platoon 

I 



I SMtlMl 






>i II 

* I « ^\ I * 

8d Platoon I 4th tlatoon , 



Sd Plato 



Lbgbnd: 



i»« 



^ 8-4 Hand Grenade 
i^ Throweis 

S 1 Cpl ft 8 pvtB 
gto (automatic * 
rifles) 

^ 8-4 Rifle Grenade 
■■ throwen 

J'UfF'lame thrower 
X^ Company Comdr 
V with telephone 

operators ft liffht 
siflrnal equip- 
ment 

4 Platoon Comdr with 
teL operator 

4 Sersreant ft Section 
comdr 

•^f^titter bearers 
^ _. Minenwerfera 
r|\S/ Cleaners-up 



jffsi jtMf msPI^ jffS) \jirsiim\ Mfsi -m^^ jost 

Bn. Comdr with Tel ft It. Signal Eauipment 



hnaj^ I M. G. Co. 

nil 



2d.Platooi 



iv^ i-vv- ir^ i-H' 

\/r^ |r|y |r|^ |r|v 



4t]^Platooy» 
• ^ect^ 



8d Slatoon^ 



>« 



Sectaon Section 

H^ )pV \pY H^ 

I III i^gfmutSl^Bxm. gmt 



Compaajr 



42 DEVELOPMENT OP TACTICS— WORLD WAR 

of a complement (1 corporal, 3 men, fusiliers), of auto- 
matic rifles, of grenadiers V. B. (grenade gun) with 2 to 

3 firing cups divided between the two squads. It was 
directed that an 61ite company be formed in each regiment. 
In the attack the half -platoon was formed into 2 waves with 
10 to 15 meters distance, in the first wave the rifiemen 
and the automatic machine gun squad, in the second wave 
the rifle grenade throwers and skirmishers. In the attack 
the platoon inserted its half-platoon either alongside or 
behind each other ; in the latter case the .first half -platoon 
was the vague d'oLssandt (attacking wave), the second the 
vagtie de renfort (support) , distance 60 to 100 meters, ex- 
tension (half-platoons in rear of each other) 40 to 45 
meters. The nettoyeurs (trench moppers-up) were fur- 
nished by other than the attacking unit, which after the 
completion of their task formed the security detachment 
of the trenches. 

In England the same mixing of men took place. Each 
tactically independent platoon (28 to 44 men) consisted of 

4 squads; riflemen, hand grenade throwers, light machine 
gun (Lewis gun) and rifle grenade '^ throwers. Half of 
the men, were trained in the use of the machine gun, the 
other half in the use of the rifle grenade. 

TRAINING REGULATIONS FOR FOOT TROOPS 

At the close of the battles of the Somme in the Au- 
tumn of 1916 the lessons of the mobile and position war- 
fare that had been gathered were thoroughly examined. 
These were published to the troops in January, 1917, in the 
shape of an outline of ^Training Regulations for Troops 
in the Field." The mobile war has proven the correctness 
of the general rules of our previous training and field 
exercises, and thus it was only a question of minor changes 
of our Field Service Regulations. The few directions how- 
ever contained in our F. S. R. concerning position warfare, 
required material revision and extension. The impor- 
tance of drill, as means to an end, to teach the individual 
the absolute necessity of the strictest obedience to his 
superior, had come to the front everywhere, especially in 
difficult battle situations. Troops, trained and .educated 



MOBILE WARFARE 48 

in a strict school, fought well. This fact had to be ad- 
hered to in the new Regulations, but on the other hand, 
the fact was not to be lost sight of that thoughtless drill, 
carried to excess on the drill ground, had a damaging effect. 
The increase in auxiliary arms and the necessity of making 
the "shock-troop procedure" the very life of the troops, led 
to the publication of a second outline edition in January, 
1918. 

The Regulations treated of training, not leadership. 
Rigid adherence was held to careful, strict individual train- 
ing as a basis for the schooling of the unit, and all was left 
out which was not absolutely required in war. The train- 
ing of the individual was to be furthered at every oppor- 
tunity, either while at rest, or in the trenches. Special 
attention was paid to rigidity, exactness and order, in all 
close order drill. Double time in cadence, present arms, 
the manual of arms for the charge, and the use of signal 
flags was omitted. 

In extended order the formation in platoon columns 
was omitted, the line was the principal formation for the 
company. Drill ground formations were no more to be 
expected of the company, than skirmish fire in close order. 
Volley fire and the charge, gaining direction on guides sent 
to the front, the transition from line into column of squads 
while on the move, as well as different kinds of deployment 
for firing were done away with. A new designation was 
given to the single column (column of files) and also col- 
umn of twos, as column and double column. 

The most important formation in close order was the 
company column, in which the platoons are in column of 
squads, in double column, or single file with 1| pace inter- 
val, which can be increased according to need (see plate). 

The company must be able to take up, in addition to 
the regulation formations, a formation adapted to the ter- 
rain and the available space. The formation of the pla- 
toons need not be the same in that case. The main point 
is that the company must be ready for employment in any 
direction according to existing conditions. By attaching 
the automatic rifles, the company received a materially 
changed appearance* Training Regulations (first edition) 



44 



DEVELOPMENT OF TACHCS— WOBLD WAS 

OOMPAMT OOLUIIN 



I 






20' 



^» 



■^ r 



lAk 



0009 t^/^-f 



6BO(t)e6 



»:<^f:i»i 



rrm 



^A^A^A^^ 



-Hrrm 



o^^i^^ 



-HTTTI 



^I'i'i^^ 



I'm M i 



»i»5i^fi 



rrm 



^^^^ 



■ MM 

4eoc(>d 



'I'^i^'i^ 



Willi 

6 



Vli2i^^'S« 



CEED 



^'i^'i^ 



Strength of the Company 

6 officers 
22 NCO 

150 men 




•I-'i'i^^ 



fTTTI 



':»5»i*5»i 



HTTn 



1 machine sun wasron 



I 



Company Commander 



■ Lieutenant 

Qlst Sergeant 

BPile^oeer 

B Squad leader 

^Machine gun commander 

^ (also right guide) 

BLeft guide 
S Sanitary NCO 

r 

.V Armorer 



C3 Front rank file 

ORear rank file 

O Runner, tel^hone operator 

'f Litter beaiers 

BMusician 

O Machine gunner 

4>M. G. Captain 

^M. G. complement, 5 to 8 

# Assistant Armorer 

# Driver 



MOBILE WARFABE 46 

provided for the formation of a 4th 'platoon, in which the 
charging troops, the hand grenade throwers, the sharp- 
shooters, and the automatic riflemen were consolidated, ''so 
that the fighting power of the company would be kept at 
the highest degree/' This plan permitted at the opening 
of decisive battle, selected leaders and men to be a special 
platoon at the disposition of the company conmiander. But 
more correctly, Training Regulation Number 2 objected to 
composing the platoons of rifle carriers and specialists as in 
England and France. Every man was trained in the use 
of hand grenades and rifle ; ttke squad was trained for skir- 
mish fighting as well as for shock-troop fighting. The au- 
tomatic rifles were inserted on the right wing or at the 
head of the platoons and half -platoons. 

The machine gun complement had to be selected on the 
basis of the reliability and independence of the individuals. 
For each machine gun at least 1 complement and 2 er- 
satz cotaiplements should be trained. The men were armed 
with pistols, Uie ersatz men with rifle or carbine. The 
complement (Numbers 1 to 4) and the best ersatz comple- 
ment formed a machine gun squad under a machine gun 
squad commander. On the march the machine guns were 
loaded on the machine gun wagon of the company. All 
officers and noncommissioned officers of each company were 
trained in the service of machine guns. Special importance 
was laid on timely opening of fire, so the machine guns 
were inserted on the right wing or at the head of their 
platoons or half-platoons. 

The battalion consisted of four companies, one auto- 
matic rifle platoon, one information platoon, and machine 
gun company. The battalion assembled in companies 
alongside or in rear of each other, the companies in march 
column, in line or in company column. The machine gun 
company consisted of three platoons of two half-platoons 
each, the two guns of the half -platoon carried on one gun 
wagon. The following formations were designated for the 
machine gun company: the line, single file, and march col- 
umn. Drill was never to be in the line formation. It was 
proper to have the platoons march in line in rear of each 
other or in single file column abreast of each other. When 



46 DEVELOPMENT OF TACTICS— WORLD WAR 

the guns were taken off the wagon in each instance it was 
ordered whether the gunners were to carry the protective 
shields, the water, or cartridge chests. To decrease the 
weight of the heavy machine guns, the sled was replaced 
by an auxiliary mount or even by a sandbag. Loading and 
unloading at night with gas-mask was practiced. Move- 
ments of heavy machine guns, after having been unloaded 
from their wagons, was generally executed with the gun 
assembled, but in all cases without cartridges in chambers 
differing from the automatic rifle, in which movements 
with loaded guns was permissible. 

All movements were at the walk. Trotting was con- 
fined to exceptional cases. By order of the company com- 
mander, when the battle situation demanded haste in en- 
tering the action, portions of the complement rode on the 
machine gun wagons (for instance nonconomissioned offi- 
cers and gunners Nos. 1 to 3 of one platoon distributed on 
the 6 wagons). Employment of machine guns in battle, 
will be discussed later on. 



HI. Position Warfare in the West, 1914-1 917 

ORIGIN AND NATURE OF POSITION WARFARE 

The prime importance of the initial success upon the 
entire course of a campaign, as illustrated by the wars of 
the 19th Century, which generally ran a rapid course, led 
Germany to rapidly assume positions in readiness at the 
outbreak of war and to simultaneously insert all available 
forces. Strategic reserves were discarded. They con- 
sisted of troops (new organizations) condemned to a slow 
mobilization^ or of allies who entered the war later on. If 
neither of the opponents has sufficient forces to break the 
other's resistance, position warfare results. It lies in the 
very nature of the offensive war that, after relentlessly 
gatdiering the fruits of the initial success, the heavy ex- 
penditure of strength forces the victor to a halt, also, in 
order to protect what he has gained. 

Such a culmination of victory is shown in the World 
War by the first battle on the Mame with the resulting re- 
treat and defense, which was fought against a strong su- 
periority of the Entente. In the subsequent course of the 
World War, which developed for the Central Powers into 
the most magnificent operation of all times on interior lines, 
the situation forced them upon the defensive on one front 
in order to beat the enemy by an offensive on another. 
The length of the lines was so extensive that the disad- 
vantages of a central location were of no importance. A 
favorable railroad net made it possible to shift the forces 
from one to the other theater of war. Thus position war- 
fare had its origin in the East and in the West. 

Military history shows that only by inserting fresh 
forces, or by voluntarily abandoning a large part of the 
position followed by an offensive against the pursuing ene- 
my, is the inertia of position warfare overcome and mobile 
warfare resumed. The effect on the morale of the enemy 
and our own troops and the difficulties of overcoming rapid- 
ly constructed field fortifications, requires serious consider- 

47 



48 DEVELOPMENT OF TACTICS— WORLD WAR 

ation prior to voluntarily abandoning a portion of our posi- 
tion and starting a counter-attack against the enemy who 
may have taken our position. 

The extension of battle fronts of today seldom permits 
any envelopment, and leads naturally to a penetration of 
the position, the execution of which makes great demands 
upon leadership and troops. The commander requires 
strong forces to gather the fruits of a penetration after 
the first battle success. 

No commander will voluntarily choose position war- 
fare, and will hardly consider blocking off in time of peace 
a certain portion of the terrain after the method of the 
Roman Lines or by means of a Chinese Wall. The decision 
lies in the attack. The fortified lines of the 18th Century 
did not meet expectations'. They held confined extraor- 
dinarily large numbers of troops and these never did make 
a good stand. Today aerial reconnaissance can furnish 
the basis for the attack of fortifications. Another point 
must not be left out of consideration and that is that the 
sphere of action of the enemy into the home terrain by 
means of airships and long range guns has increased im- 
measurably, whereas in the 18th Century .the effective 
range of hostile guns reached only some few hundred me- 
ters within our position. By such procedure (choosing 
position warfare) we abandon every possible war objective, 
delay the decision into the distant future and increase the 
economic pressure on our own people. Sufiicient protection 
of valuable economic terrain can be secured only by carry- 
ing the war in the enemy's country. 

Brief examination of the military events in the past 
forty years may cause us to assume that position warfare 
became the rule; mobile warfare, the exception. But we 
must in each case carefully ascertain whether the position 
warfare was selected on account of inferiority of numbers, 
in order to delay defeat until other States could participate 
(expected British participation in the Russo-Turkish War, 
1877-78), or until the arrival of reinforcements (Russo- 
Japanese War, 1904-05), or whether the position warfare 
was forced by special conditions. Position warfare length- 
ens the war and thereby increases the suffering and, in 



POSITION WARFARE IN THE WEST, 1914-1917 49 

spite of the smaller losses in separate engagements, the 
aggregate losses total far more than results from a battle 
in the open. Only overestimation of our auxiliary means 
and underestimation of the moral influences, could lead 
to the conclusion that in the future position warfare will be 
the rule. Only an attack (not considering the influences 
of an economic war) is able to break the will of the enemy. 
For any commander seeking a decision, position warfare is 
but an auxiliary means used to let the enemy wear himself 
out and to gain time. The earlier the attack is undertaken 
and the greater the forces employed, the less time will the 
defender gain. Most time will be gained by the defender if 
the attack is made with insufl9cient means (Plevna, 1877). 
The commander must be thoroughly prepared for position 
warfare but must strive with all means at his command for 
mobile war and carry that through. 

Field Marshal von Moltke, as is known, had advised as 
early as 1866 a strategic offensive in connection with a tac- 
tical defensive, and in 1874 had again insisted on that pro- 
cedure; but in war, not counting the battles on the Loire 
(Coulmieres, Beaune la Rolande, Loigny) the Field Mar- 
shal had been unable to transmute his ideas and reconmien- 
dations into action, under the pressure of events. But 
Moltke surely had in mind a procedure that was to be vol- 
untarily adopted, even before the climax of the victory was 
attained. After the September battles on the Mame in 
1914, and in the summer of 1918 we were forced (in France) 
to choose the defensive after a successful offensive. We 
by no means appreciated the advantages of that situation. 
In 1914 the frontal attacks of the enemy were defeated on 
the Aisne, and then commenced a race ''to the sea" each op- 
ponent with the intention of gaining the flank of the other. 
The Entente had gained freedom of action by stopping our 
forward movement, and they were able to recuperate from 
the heavy fighting of the Summer of 1914. In England an 
army had been created from nothing and their divisions at 
the front had been equipped with heavy guns and modern 
battle means. 

In position warfare attacker and defender are opposed 
to each other at close range and both of them are forced 



50 DEVELOPMENT OP TACTICS— WORLD WAR 

to the same construction of positions. While the attacker 
has to take the terrain as he finds it, the defender seeks to 
increase his fighting strength by choice of a favorable ter- 
rain, fortified by artificial means and he certainly will not 
give up that position without good reasons. Therefore pos- 
session of the terrain gains increased importance in position 
warfare. As the attacker desires to avoid the frontal at- 
tack, he seeks to lead the attack against a flank or, as was 
the case in the 18th Century with smaller armies, to maneu- 
ver the enemy out of his position and to attack him in the 
open. The defender is thus offered the opportunity of 
counter-attacking with strong forces, holding the position 
with weak forces merely as a shield. However, the impor- 
tance of possession of the terrain held good only as long as 
the defensive means of the defense proved themselves sup- 
erior. All battles show that the attacker is successful, and 
even with far less losses, in gaining the first penetration. 
The defender has to decide whether to hold to the terrain 
under great loss or to avoid a decision, abandoning his labors 
and the terrain — ^but which have forced the enemy to time- 
consuming expenditure of material and personnel — and to 
resume the battle at some other place. 

It cannot be stated whether modern weapons favor the 
attacker or the defender, but the attacker has the advan- 
tage of being able to choose the place and time for the 
attack and to bring the assaulting troops into the effective 
fire zone of the defender only a short time before the hour 
of the assault. On the other hand the defender must keep 
his troops in readiness at all times awaiting the attack. 
For that reason^ surprise on a large or small scale increases 
in importance. Attack preparations that have not been ob- 
served avoid counter-measures. A surprise in the conmience- 
ment of the assault delays the hostile barrage fire and pre- 
vents timely occupation of the hostile fire trenches. But we 
cannot, reckon on surprise as a certainty. It is difficult to hide 
from aerial observation the placing in readiness of person- 
nel and materiel employed in an attack, on a large scale, 
even if we can succeed with our surprise at other points. 
The statements of a single deserter may easily nullify all 
preparations. Thus we cannot plan an attack on surprise 



POSITION WARFARE IN THE WEST, 1914-1917 61 

alone. Surprise is merely a material auxiliary means for 
success. In attacks on a very large scale the statements of 
a deserter immediately prior to the attack still may have 
no materially bad consequences. Transmission of messages 
requires much time, as well as the transition from decision 
to execution. Our military channels seldom have anything 
important to report. Even ''reliable reports/' meaning such 
as confirm a pre-arrived at opinion, cause little attention to 
be paid to other reports that are not in conformity there- 
with. 

A certain monotony is combined with position warfare, 
which can easily lead to indifference and apathy and finally 
to acceptance of a 'Teace within the precincts of a castle'' 
or 'leave m^ alone and I surely wiU do nothing to you" 
policy. Such a conception which is as far as heaven is 
from the earth from the conception of actual war, must be 
combatted with all possible means. The most extensive/ 
damage must be done to the enemy. In the foreground, 
up to the hostile obstacle, no patrols must be permitted. 
Fortification labor in the open and movements in the open 
by detachments or vehicles must be made impossible by 
fire. If that is not done we will finally see an agreement 
between the pickets and thus "trench*^ friendships" will 
be created which in aU cases are of the gravest danger to 
loyal troops. In the French divisions the maxim was gen- 
erally adhered to that troops, once relieved, should never 
again be placed in the line at the same point, thereby in- 
creasing the difficulties of fraternizing. Monotony will 
be prevented by firing as a general rule on all visible targets, 
laying a sweeping fire by day and night upon the hostile 
rear areas and frequently executing operations of some 
kind. In position warfare, as we have learned, discipline, 
capacity for marching and desire for operations suffer 
under the hard work of throwing up fortifications and 
officers of all and every grade have to do their very best to 
prevent a "tiredness for war." The value of cover is easily 
overestimated. As numerous complete reports indicate 
the troops accustomed to the cover of a position system in 
many instances can no longer accommodate themselves to 
the uncertain conditions of mobile warfare. 



52 DEVELOPMENT OF TACTIGS— WORLD WAR 

THE POSITION BATTLES UP TO THE FIRST 

ATTACK ON VERDUN 

When, after the battle of the Mame in the Autumn of 
1914, the German troops resorted to the spade, they had 
sufScient high-trajectory artillery. It therefore was suffi- 
cient, as the enemy lacked minenwerf ers, to arrange for 
protection in the trenches against smaU arms and shrapnel 
fire. The importance of aerial reconnaissance was insuffi- 
ciently appreciated at that time. 

The first positions in which infantry and artillery 
worked in many cases without proper connection, showed, 
with a long range field of fire, a line according to the proffle 
of the reinforced rifle trench, constructed in most cases by 
far separated groups. Differing from the French and the 
Russians, we rejected any formation in depth. The center 
of gravity lay in the most advanced trenches, beyond which 
a few shallow trenches for sentries were located. The de- 
mand, to defend only one line, was based on the proposed 
conduct of the defense which anticipated that the attacker 
would be held up by the fire fight of the defender, and the 
decision sought by the counter-attack by the reserves. There 
was no time for false works. There were but very few 
routes of approach and covered trenches. Shelters were 
to provide security at the most against straight hits of 
field artillery, otherwise only against shrapnel fire. The im- 
portance of flanking from open trenches was not generally 
appreciated. Fortification by squads or groups led to the 
reestablishment of the supporting points (which had been 
omitted in the ''Field Pioneer Service Regulations for all 
Arms'O which, constructed in the shape of ''kraals,'' allowed 
deployment of the firing line up to 400 meters, but resulted 
in violating of the general rule of the line being defended 
only by weak forces. Obstacles were to be placed so far 
from the trenches, that they could just be watched from the 
latter (about 50 meters). "If they are located close to a 
position, they will be within the effective zone of dispersion 
of the hostile artillery fire laid on the position ; and in addi- 
tion they will not provide sufficient cover for the garrison 
against hostile hand grenades'' (Field Pioneer Regulations, 
848). 



POSITION WARFARE IN THE WEST, 1914-1917 68 

The applied forms of fortification were in absolute con- 
sonance with the Resrulations of the Field Pioneer Service, 
in the preparation of which we might have reckoned with 
the possibility that the enemy would quickly bring to the 
front heavy batteries employing high-angle fire, and that 
we would also have to encounter heavy guns of large cali- 
ber using flat-trajectory fire. In any case the 16-cm. cali- 
bers were soon outclassed by heavier ones. Also new weap- 
ons, hand grenades in the hands of the infantry, minenwer- 
fers, gas shells and gas projectors, made their importance 
felt which caused a complete change in the forms of the 
fortifications. 

It was very soon seen that the attacking artillery, if 
in sufficient strength and liberally supplied with anmiuni- 
tion, would be capable with increased volume of fire (drum 
fire) to annihilate any position that it could observe; that 
thereby narrow, deep skirmish trenches would be trans- 
formed into flat depressions in which the artificial works 
disappeared beyond recognition, and that shelters and cov- 
ered trenches with insufiiciently strong roofs were in danger 
of being filled up. A remedy therefor was had by increas- 
ing the strength of cover first, then by constructing more 
covered and connecting trenches, whereby it became possi- 
ble to construct numerous shelters, distributed in breadth 
and depth, so that in the rear of the front trenches several 
lines were created, to which the garrison that was being 
heavily fired on could go for shelter. Where the ground 
was firm, very narrow and deep trenches were preferred. 
Such were not fit for defense at all points without additional 
preparation. They required special arrangements for fir- 
ing and for drainage ; could not be used indefinitely without 
proper waU supports; and finally, the rescue of the men 
who had l)een covered up by tr^iches caving in, was made 
more difiicult. We rightly asked ourselves if the extensive 
labor connected with the construction of such trenches was 
justified by the protection which they could offer the garri- 
son after a continued ''drum'' fire. If we assume that any 
position within reach of the hostile artiUery can be anni- 
hilated by the enemy, then second and third positions, 
against which the hostile artillery would have to deploy 



64 DEVELOPMENT OP TACTICS— WORLD WAR 

again, gained special importance; thus we discarded the 
view: ''as a general rule only one defensive position will 
be selected and fortified with aU available means/' (Pio- 
neer Regulations, 216.) 

Very properly aU regulations pointed out the necessity 
of making the parapet as low as was possible from the 
nature of the terrain, the character of growth covering the 
ground, and condition of the ground. But following the 
directions of regulations alone did not suffice as soon as hos- 
tile artillery could observe the trenches, which could not 
be hidden from aerial observation, and which betrayed them- 
selves by the obstacles. False works offered only a minor 
remedy. Thus in the first months of 1915 we took a step 
farther and abandoned the long range field of fire by draw- 
ing back the firing line from the forward slope of the pla- 
teau, in many cases even to the rear edge of the plateau, 
and merely observing the forward slope of the plateau by 
sentries, who frequently were posted in false positions. The 
advantage of being able to defend positions having a good 
field of fire with only a small number of skirmishers became 
of small consideration as a powerful hostile artillery would 
prevent skirmishers from using their rifies. It is certain, 
however, we mainly secured protection against hostile ar- 
tillery fire by this method, but as a matter of fact, we lim- 
ited the co-operation of our own artillery and gave the enemy 
increased facilities for finding the range. In a short time 
the hostile artillery could bring its observers to the aban- 
doned, or only weakly held heights ; and then, from a cov- 
ered position and unhindered, could without difficulty smash 
our infantry by its fire. 

It is true that the observation stations of our artillery 
could be placed for temporary purposes beyond our own 
position, but for the decisive battle they had to be located 
within the position. On account of the immense dust and 
smoke clouds caused by the drum fire, observation from 
flank positions and from positions in rear gained in impor- 
tance. It became more and more evident that the location 
of the observation was of the utmost importance in the 
selection of the fighting line. 

The non-successful battles of the Entente— in Flanders 
and in the Champagne starting with the 16th of February, 



POSITION WARFARE IN THE WEST, 1914-1917 66 

1915 — differed from the prior attacks only by the stronger 
artillery preparation that had been increased to ''drum'' 
fire. This preparation was to enable the attacking infantry, 
advancing in dense skirmish lines and f oUowed at 100 meters 
distance by companies and battalions in columns, to take 
possession of the battered position without fighting. It is 
said that in the Champagne 18 to 20 rounds were thrown 
on each lineal meter of positions, but the success did not 
accord with the expenditure of ammunition because, ac- 
cording to the views of the leaders, the assaulting mass was 
not broad and deep enough and there were no reserves. 

''Experiences in war concerning field fortifications'' de- 
manded that a portion of the infantry should be trained in 
throwing and manufacturing suitable hand grenades. The 
ball (time-fuse) hand grenade, the disk (percussion-fuse) 
and the steel handled grenades (time and percussion fuses) 
were all employed. Only the steel handled grenades (time- 
fuse) were found to be serviceable and also a lighter egg- 
shaped hand grenade with throwing ranges of 30 and 40 
meters. It was recommended that in stubborn defense, at 
least two positions, one in rear of the other, be taken up, 
each consisting of several lines (60 to 100 meters distance) , 
with organization in depth to prevent a rolling, up after the 
enemy had broken in. Not much emphasis was laid on the 
value of a large field of fire, but it was recommended that 
machine guns and light field pieces be placed in the forward 
trenches and thoroughly covered at the most important 
points ; that the trenches be arranged for defense toward the 
rear as well as toward the front ("closed rifle trench," sep- 
arating the two trenches by 50 meters) ; and that they be 
surrounded by obstacles. Construction of loopholes was 
left to the discretion of the defender. Smaller shelters were 
provided in the front trench; larger ones holding as much 
as one platoon in the rear trench. Head-cover was not re- 
quired in rifle trenches, except in observation stations, be- 
cause it interfered with the use of the rifle and bayonet. If, 
in exceptional cases, head-cover was desired, one with a 
continuous horizontal loophole was considered best. Closed 
works were of no use ; straight lines rather were pref erred^ 
as it was not possible to depend on securing flank fire. When 



DEVELOPMENT OF TACTICS-WORLD WAR 

FoBTincAnoNS Pn« to thc Adtiiiin 

Dngovti (foK-lMdM, for about 4 nw dtttnc) 





GoBuiiunication Trencb 



POSITION WARFARE IN THE WEST, 1914-1917 

BATTLB in the CUAMFAOm 



67 



Trench 




r^^Side tnek, for MHdm t» 
Q) the riidit 



Riflemen's nichsi 

Shelters on larser scale (worminff places, 
€ Coomiand stations, rest stations, dressing 
stations) 

^ Route of approach 




O.&O . 
Reinforced trench as infantnr position. To be oonstmeted In 

one niffht in favorable groond 



Loopholes, reinforced with lumber (profile) 

I * * ' ' /f/ :RaBgfeSg!fl B8^w Sod or sand-ba«i 




1 Concrete head cover with 1 meter thickness offer protection 

acpainst 8h<d]s fired from mortars 

2 Trenches in the shape of reinforced trenches. Approach 

trenches 2i meters from floor to ceiling 



68 DEVELOPMENT OF TACTICS— WORLD WAR 

the flanks of a position have to be drawn back, it should be 
done by echeloning the rifle trenches. Machine guns had 
better be posted on the flanks and, if possible, under cover. 
They should not expose themselves prematurely in order to 
avoid early destruction by artillery fire. In many cases the 
front trench was designated the fighting trench, the rear 
trench (12 meters distance), the communication trench. 
The strength of cover against light field howitzer projec- 
tiles was given as 2 meters (using alternate layers of stones, 
logs, etc., of 1.5 meter) . Using concrete the cover was 0.6 
to 1 meter thick. The necessity of having two exits from 
shelters was not sufi&ciently emphasized. Obstacles were to 
be constructed not farther than from 20 to 30 meters in front 
of the trenches, and to be in 2 to 3 bands, each 6 to 10 meters 
wide. They were to protect against sudden attack and to 
force the attacker into the spaces swept by our fire. Sev- 
eral bands of obstacles proved themselves just as good as 
obstacles placed in depth. High obstacles betrayed them- 
selves clearly on aerial photos. C!lose co-operation between 
infantry and artillery was absolutely necessary. 

Posting the artillery at a designated time and place was 
absolutely required for attack as well as for defense. To- 
day infantry cannot fight unless the artillery has fully pre- 
pared the way. But the artillery effort is in vain, if it is 
not utilized by the infantry at the correct time. The effect 
of the artillery fire supplements that of the infantry and 
vice versa, and both must work in close conjunction. 

The French positions, exclusive of those in unimpor- 
tant sectors, as a general rule consisted of three lines of 
which the first line — strongly constructed but weakly held, 
— ^was designed for security. The second line, 150 to 200 
meters distant, was the main line of resistance, in rear of 
which were located a series of supporting points prepared 
for all around defense. 400 to 500 meters in rear of the line 
of main resistance, the reserves were held in secure shel- 
ters for use in the counter-attack. Special consideration 
was given to carefully planned flanking works for all lines. 
Sector reserves, labor and park battalions, were employed 
in constructing a second position. 

French generals ascribed the cause of the success of 
the German attacks to the fact that the front lines of the 



POSITION WARFARE IN THE WEST, 1914-1917 69 

positions were always too strongly occupied, though Joffre 
as early as January 5, 1915, had cautioned (to reduce the 
loss) against crowding defenders into the front line. The 
men in the front line were merely to report the assault and 
delay it; the decision was to be sought in the counter-at- 
tack by the reserves. This created the need of a position 
system, with carefully planned formation in depth in which 
strong obstacles and a powerful flank defense played an es- 
pecial role. The difficulty in gathering together the widely 
distributed garrison of the front line for attack purposes, 
led to the construction of covered assembly places. Num- 
erous captured documents show that the French were slow 
to accept the very correct views of Joffre. Similar views, 
as we will show later, encountered stubborn resistance in 
the German Army. 

The British, not caring much for a long range field of 
fire, tried to hide the fortifications from the view of the 
enemy. In many instances they were constructed on re- 
verse slopes and in the cover of hedges. Their position con- 
sisted of three lines, connected by approach trenches. The 
front rifle trench was as narrow as possible (45-cm.) and 
90-cm. (or more) deep, with shelters and breastworks. 15 
yards in rear was the communicating trench, 45 to 60 meters 
long and 1.8 to 2.1 meters deep. 25 yards in rear of the 
communicating trench was the cover trench with a breadth 
of 60 cm. and with a depth as great as practicable, under 
special conditions, 4.8 meters. The earth dug out was scat- 
tered and smoothed off or used for the construction of false 
works. For protection against explosive sheUs, a parados 
was required. Simple arrangements were provided for 
fixing the rifles with proper sight elevations, for use in 
night attacks. 

The experiences gathered and lessons gained by the 
Allies in the Winter engagements were arranged and com- 
piled by Greneral Joffre into the Regulations of April 16, 
1915, entitled : ''Aims and Prerequisites of the General Of- 
fensive,'' which shortly after publication fell into German 
hands. The procedure established therein was that the 
troops, after a concentrated heavy artillery fire of all cali- 
bers ("drum fire") lasting for from four to five hours and 



60 



^SO^ZOOm 



DEVELOPMENT OF TACTICS— WOBLD WAB 
Abranobiixnt op ths Fibld of Attack 

Hostile positloB 

^ ^ ^ P P 

i 





rT Dressing Station 



o- 



Dressiniir Station 



POSITION WABFARE IN THE WEST, 1914-1917 61 

... AOOQBDINO TO JomB'S PLAN 



LaoBND: 

J^ m Advanced trenches 

O * Exit trench (Paralle de Depart) 1 meter broad 

AT • Approaching: trench 

/t IB Trench of the 1st Line, 1 meter broad 

Approach trenches, 1 meter broad 
two for each company 

Assembly stations of the 1st Line 
two each for one platoon 

D =^ Communicating trench, 2 meters broad 

F. « Approach trench, 2 meters broad, one for each company 

F » Evacuation trench 

G ^ Side tracks 

n ■■ Assembly stations of the 2d Line 

J ■■ Connecting trenches, 2 meters broad 

|S| j^«B Headquarters location (Bn, regt, Brig., Div.) 
" Observation Stations of Artillery 

Position in Readiness to Attack 

German Front (about 600 rifles) 

fStOOm 



rso'^ 



900^ 



H companieB of each of the Ist resrixnentB in a single dense skirmish line— 2400 



lifles 



JOO^ 



**6 companies of the 2d resriments in same formation as those of the 1st reeiments — 
1200 rifles 

Distance variable, at head of field of attack. 6tii Co., 2d Begiment 

Second Brigade at rest» but in readiness 

^•♦v'^«nj?**?5* is made in the start by the Ist Brisade only with 18 companies. Le.. 
with 8600 nfles, placed m readiness on a stretch of ground 122 m. broad and 160 m. 



62 DEVELOPMENT OF TACTICS— WORLD WAR 

concentrated on a very limited space with great rapidity of 
fire were to be brought up under cover to assault distance 
(160 to 200 meters from the enemy) and placed in readiness 
under cover and in the assembly trenches. During the 
night preceding the attack a line of departure for the at- 
tack was to be constructed. If the fire preparation proved 
insufficient to annihilate the materiel and moral resistance 
of the defender, then the fire was to be extended to several 
days. The attack was to commence with the assault. 

It was intended to overrun the hostile line on a broad 
front (1200 to 1500 meters for each division) and in strong 
formation in depth, column of brigades and regiments. By 
continuous advance of successive echelons the movement 
was to be kept up to a designated attack objective deep in 
the hostile position. The attacking waves followed each 
other with little distance and consisted of entire companies 
deployed with intervals of only half a pace between skirmish- 
ers, so that 3600 men were together on a space of 1200 me- 
ters extent and 170 meters deep. The attack was initiated 
only by the 1st Brigade, while the 2d Brigade remained on 
the alert in its quarters. ^Placing the regiments in column 
led to a complete mixing of units and prevented the exer- 
cise of command. Penetrating a position is impossible by 
successive waves, as the ones following in rear cannot find 
any room to pass through the halting masses of the leading 
lines. It is evident that there is danger that the supports 
will be hung up in the first line on account of the small in- 
tervals. Jeffreys procedure changed the division into a 
single unwieldy phalanx which, once started moving could 
only advance straight ahead or turn back in complete disor- 
der. In this illogical scheme no opportunity for leadership 
or exercise of initiative was allowed. A minor portion of 
the artillery was to hold down the hostile artillery, one-third 
of the field artillery being assigned to destroy the obstacles 
and two-thirds to prepare the hostile trenches for the as- 
sault. In this, to each 200 meters two 75-cm. batteries were 
assigned with 80 to 100 rounds per piece. Of the heavy cali- 
bers 40 to 50 rounds for each 220-mm., 50 to 60 for each 
150-mm., and 60 to 80 for each 120-mm. piece were allotted. 
This plan of attack precluded all surprise, though as late as 



POSITION WARFARE IN THE WEST, 1914-1917 6S 

March Ist the army had seen in surprise fhe ^^dispensable 
factor of success/' The plan of attack was based on the 
effect of superior expenditure of ammunition and the weight 
of the assault by superior infantry that had been saved there- 
for. It demanded the time-consuming removal of enormous 
quantities of earth and forced the command to attack only 
where the two opponents were at assault distance. To de- 
feat such an attack the German measures were : Great dis^ 
tance between the rear lines protected by obstacles and sup- 
plied with numerous shelters which had to be occupied sim- 
ultaneously with the front line ; full utilization of the flank- 
ing effect of machine guns and single field pieces ; tiie ene- 
my's penetration to be answered by an immediate counter- 
attack; hostile batteries that could be seen, to be engaged 
prior to the attack ; demolition of the hostile assault works 
by heavy guns of high angle fire and minenwerf ers for the 
purpose of delaying the attack ; during the attack the heavy 
artiUery to keep up its fire on the hostile front trenches to 
cut off the hostile supports. 

Joffre's army orders of September 14, 1915, designated 
''the present moment as especially favorable for the offen- 
sive.'' Found by the Germans, these orders pointed out 
the danger, while at the same time the results of the Ger- 
man ground and aerial reconnaissances as well as the state- 
ment of prisoners and deserters furnished valuable infor- 
mation of attack preparations against the line Auberive — 
Ville sur Tourbe, that was then held by only four divisions 
(33 kilometers) . General von Castelnau led the attack. On 
September 27, at 7 :00 A.M., reinforced artillery fire opened 
which the defender sought to reduce by regular firing on the 
hostile batteries. Gas was liberally used by the attacker. 
French airplanes ascertained the effect of the fire, and 
sought to interrupt traffic in the hostile rear by dropping 
bombs. The French attack divisions had been in the front 
line a short time to get familiar with the terrain and the 
position, and had then been taken back to the vicinity of 
Chalons for rest. Thus it was that as early as September 
22d we knew 18 of the 30 hostile attack divisions. Com- 
pletely fresh troops, that had the advantage of the super- 
iority in numbers and materiel were to make the attack 



64 DEVELOPMENT OP TACTICS— WORLD WAR 

against the weak German first line divisions which for the 
most part had been in the trenches for several months. The 
preparations for the attack were completed by j;he 22d. Thor- 
ough preparation to the minutest detail was absolutely 
essential for success. Improvement of the approach and 
assembly trenches was demanded. Even preparations for 
concealed assembly places for and provision for advancing 
the cavalry divisions in columns of twos, had been made. 
The morning of the 24th the drum fire was suspended for 
a short time, so that reconnoitering detachments could as- 
certain its effect. The grouping of the French, different 
from Joffre's attack method was the same along the entire 
front. In each front line division 3 regiments were placed 
abreast with the 4th in reserve. In the regiments the bat- 
talions were in column, each in dense skirmish lines at 50 
meters distance. The first wave was preceded by hand gren- 
ade squads and followed by detachments to clear the 
trenches (nettayeurs) and detachments to salvage captured 
material, to break any surviving resistance and to bring 
back the prisoners. 

Some battalions were accompanied by 65-mm. moun- 
tain guns to fire on- machine guns. In order to have the ar- 
tillery distinguish their own troops the men wore white 
cloths sewed on their backs. 

Along the 33-kilometer battle front the French had de- 
ployed 1285 light and 650 heavy guns, or 40 light and 20 
heavy guns per kilometer. Prior to the attack the 50th In- 
fantry Division (German) had only 12 gun batteries, one 
Russian 15-cm. gun battery, and 2 heavy field howitzer bat- 
teries; the 15th Reserve Division (German) had only 6 
gun batteries, one 10-cm. gun battery, 4 heavy field howitzer 
batteries and 4 mortar batteries. Thus the French super- 
iority in artillery was ten to one. 

During the night of the 24-25th, trench work on a large 
scale took place to provide assault positions. Routes of ap- 
proach permitted reserves to be brought up under cover to 
within 4 to 5 kilometers. In order to have the entire day for 
development of the success, and after a drum fire that had 
lasted 72 hours, at 9:15 A.M. the attack started with 22 
divisions in the first line and eight divisions in the second, 



POSITION WARFARE IN THE WEST, 1914-1917 66 

along the front Massiges — ^Auberive (88 Idlometeni) against 
the position held by four German divisions after the artil- 
lery lire about 7:00 A.M. had again been increased to 
maximum volume. 

The French assault positions were about 80 to 400 me- 
ters from the foremost German trenches. Generally the 
distance was about 200 meters, which was still more de- 
creased by the leading French waves which in order to avoid 
the defensive barrage, had, shortly before the attack, taken 
position lying down in the open ground in front of tibeir 
trenches. Considering this short distance, the defender 
could not man the fire positions at all points at the proper 
time ; but the resistance was not broken thereby. Each squad, 
each individual kept on fighting. The 53d Infantry Regi- 
ment (50th Infantry Division) which had furnished 4 com- 
panies for the corps reserve, held a sector of 2250 meters 
front with only 3 companies in front, 8 companies in rear 
and 2 companies in second line. The attack of 4 French regi- 
ments with 8 battalions was directed at that point, these 
battalions close alongside each other. The front lines were 
overrun by the attacker, it is true ; but the stubborn resis- 
tance of the 53d Regiment was broken by envelopment and 
attack in the rear. All honor to this brave regiment, which 
here succumbed to superiority in numbers, but the remains 
of which, reinforced by a few reserves held their place until 
the evening of October 1st. From September 22d to October 
10th the regiment lost 56 officers and 2588 men. In conse- 
quence of the superiority in men and materiel and of the 
stubborn resistance of the front line, heavy losses in pris- 
oners could not be avoided. The French army conmiunique 
of the evening of the 26th, reports for the entire front 16,000 
unwounded prisoners, 200 of whom were officers and 24 guns 
captured. The evening of the 29th they reported 23,000 
prisoners and 121 guns captured. 

The excellent conduct of that regiment stands by no 
means alone. I shall cite only the stubborn resistance of 
the 2d Battalion, 65th Infant^, under an efficient cavalry 
captain on Hill 196 at Le Mesnil, in a reverse slope positioii 
which could not be reached by the flat-trajectory guns. The 
troops, though attacked in flank and rear, held their posi- 



66 DEVELOPMENT OP TACTICS— WORLD WAR 

tion; gained time by counter-attacks; and by the evening 
of the first day of the battle had captured 300 prisoners and 
2 machine guns from the enemy, they themselves losing 8 
officers and 180 men (only 60 killed) and 5 prisoners. 

We cannot emphasize too much in peace training that 
threatening and envelopment of the flanks are of themselves 
no justification for withdrawing; that a well constructed 
trench system can be held even under very unfavorable con- 
ditions, while retreat across an open terrain will almost al- 
ways lead to annihilation. On the French side the main 
cause for the failure of the attack was attribtited to the 
efficient employment of the German machine guns. 

It appears that the main plan was to break through 
along the four main roads leading northward. The objec- 
tive of the attack was to be the Vouzieres — ^Rethel road, 
about 24 kilometers distant. We will not go into details 
here. However, a deep penetration succeeded only on both 
sides of the Souain — Somme road for a width of 17 kilome^ 
ters and a depth of 3.5 kilometers. At another point the 
French had penetrated only to a depth of 1600 to 2000 
meters. The German defense was by means of barrage fire. 
Though this fire did not break up the first waves, it still had 
the effect of delaying the succeeding waves ; and in many 
cases stopped them. Where the penetration had been suc^ 
cessful, advancing reserves strove to gain the fruits of 
that success, regardless of the losses they suffered. 

This resulted in the mixing of units and in confusion 
which was increased by the French fire and the stubborn 
defense of the intermediate ground by the German troops 
held in readiness. In the evening the battle came to a halt 
in front of the second position, which had been located, after 
the experiences of the Winter battles, on the rear slope of 
the hilL Attempts by the cavalry to exploit the success 
met with a bloody defeat. The attack preparations had 
been thoroughly planned up to the penetration and regulated 
to the minutest details; and from there on the command 
left everything to the initiative of the subordinates, who 
failed. 

The next few days passed in minor actions with no re- 
sults. The artillery required till October 4, 1916, to d^ptoy 



POSITION WARFABE IN THE WEST, 1914-1917 67 

in new positions. The attack, resumed then, gained no suc- 
cess, as by that time sufficient German reserves had ar- 
rived. The battle died out by the end of the month. 

Prior to the attack it had been presumed that the Ger- 
man resistance would be completely broken by the artillery 
fire. This presumption did not materialize. The artillery 
support, in connection with support of airplanes, had been 
excellent; the ^Mrum fire'' had annihilated all works that 
could be seen; but after the penetration, connection with 
the artillery was entirely lost. The skill of the individual 
French infantrjrman in clinging to the ground and in 
rapidly and skillfully strengthening the terrain, showed up 
advantageously ; but less successful was the leadership from 
the company commanders up, for the leaders did not under- 
stand how to keep their men in hand. The effect of the ar- 
tillery fire was disappointing, for in spite of the enormous 
expenditure of ammunition, it did not have the power of 
annihilating a very numerically inferior opponent. 

It was deduced from the Autumn battle, as far as posi- 
tion warfare is concerned, that each defensive system must 
consist of at least two positions, which should be so far 
separated that the enemy cannot overcome by his artillery 
fire the second from the first ; and that the main purpose ct 
the second position is to prevent the attacker moving by the 
flank after penetrating the first position in order to extend 
his success. Each position should consist of two lines, sep- 
arated by a distance of 50 to 100 meters and connected by 
numerous trenches capable of being defended. The most 
advanced line, in which the larger portion of the machine 
guns should be, should form the main battle line. It was 
found that flanking fire along the front of the line was es- 
pecially important. Obstacles should be placed in several 
rows rather than in a single row. Although it is desirable 
to assume formation in depth to decrease the losses, this 
should not lead to splitting up commands. A trench garri- 
son, attacked from all sides, must hold out at all costs in 
tlie expectation of relief by a counter-attack that is certain 
to come. Narrow, deep r^e trenches, if well revetted, en- 
dured on quiet fronts. Deep and broad trenches lasted bet- 
tar under artillery fire. Loopholes were also used. Dugouts 



6S DEVELOPMENT OP TACTICS— WORLD WAR 

had to be proof against the continual high-angle fire of the 
15-cm. pieces. The entire garrison of the first and second 
line had to be given protection against artillery fire. Ma- 
chine guns shbuld be placed in casemate-like shelters and 
sited for flanking fire. The advantages of taking position 
on the rear slope, just behind the crest of a hill, had been 
clearly shown. The decisive factor in the selection of the 
position continued to be primarily the location of the ar- 
tillery observation stations, from which it should be pos- 
sible to observe to a considerable depth the terrain over 
which the enemy would advance, and secondarily the pos- 
sibility of organizing our own artillery positions in sufiSicient 
depth. Part of the artillery was assigned for the general 
defense and part for closed defense against the assaulting 
troops. The latter, distributed by platoons or pieces, was to 
be kept hidden from hostile observation up to the time of 
the assault ; but in most instances it betrayed itself by the 
size of its cover ; and being placed in the front line, it was 
always sacrificed during the preparatory drum fire. The 
generally weak artillery of the defense did not justify such 
a detachment on a part of the artillery. 

The first German attempt to break the position warfare 
was the attack on Verdun, which had originally been 
planned for February 12, 1916; but was postponed to the 
21st on account of bad weather; and executed after a drum 
fire lasting for 24 hours on the bank of the Meuse. Whether 
this attempt should not have been made earlier and in an 
effectual manner, we shall not discuss here. Of course sur- 
prise had not been fully safeguarded. Nevertheless, the 
enemy lacked time to prepare counter measures. The ad- 
vance commenced (according to French reports by the 3d, 
18th and 7th Army Corps) on the 22d on the 12-kilometer 
front, Consenvoye — ^Azannes. The French positions were 
overrun on a 10-kilometer front for a depth of 3 kilometers. 
Then a second advance started from fhe direction of Etain. 
On the 25th Fort Douaumont and the village of Vaux were 
taken. From the middle of March, the attack was also con- 
ducted on the west bank of the Meuse. Unfortunately army 
headquarters had neglected to place reserves at the dis- 
posal of the attack. Toward the end of March the front 



/ 



POSITION WARFARE IN THE WEST, 1914-1917 69 

of attack had been decreased, by pushing in the arc, from 
72 to 55 kilometers. The advance on that front was car- 
ried to a depth of 8 kilometers. Early in April, on the east 
bank of the Meuse, 6 German corps and on the west bank, 
9 German corps carried on the offensive while the French 
gradually placed in the line more than 80 corps. The 
French command attempted in vain to throw back the Ger- 
mans. In the shifting of the location of the attack lay the 
possibility of surprise for either opponent. By the middle 
of July, the battles came to a standstill on the recommen- 
dation of the Crown Prince, on account of the enormous ex- 
penditure of forces. The French command had conducted 
the battle as offensively as possible ; but the emplojonent of 
artillery and the power of the German infantry proved 
superior to the French defense in a battle terrain that had 
been prepared in time of peace and during 18 months of 
war and was densely covered with woods in many places. 
The forces available were insufficient to carry on ilie de- 
fensive battle on the Somme and at the same time force 
a decision on the Meuse. The lessons learned by all arms 
were of special importance in subsequent attacks. They 
pointed out before all else the necessity of the closest co- 
operation with the artillery, pioneers, and aerial forces. 
Hinenwerf ers proved themselves to be an auxiliary means 
especially well suited for the attack. As large a number 
of mien as possible should be trained in the employment of 
machine guns and hand grenades. Everywhere it was 
learned that best method of advance of the infantry in as- 
sault was on a broad front and in waves following closely 
llie preparation by the artillery and minenwerf ers, and ac- 
companied by machine guns, pioneer detachments, and ar- 
tillery. Intervals between skirmishers should be about 3 
meters; distance between waves, about SO to 50 meters in 
order to pass through the barriage fire. Every advantage 
that offers itself, such as gaps in obstacles and favorably 
situated shell-holes, must be utilized in the attack. The 
stubborn French defense from pill-boxes and dugouts located 
in woods and the failure of the attack under the fire posi- 
tions arranged for close defense but unknown to and unob- 
served by the attackers, compelled the employment of 



70 DEVELOPMENT OF TACTICS—WORLD WAR 

'*€LB8ault groups*' of a strength of 1 to 3 inf antiy squads in 
conjunction with a pione^ squad, machine guns, minenwer- 
fers and flame throwers. The ^'assault groups'' were spe- 
cially trained; and in the attack remained in rear of the 
front line at the disposal of the company eomnumders. In 
place of the rigid skirmish line which was too easily stopped 
by organized resistance, the employment of series of smaU 
assault detachments was considered, in which the personali- 
ty of the leader, on which everything depended, was brought 
into greater use. Adopting this method^ on July 7, 1915, 
the Bavarian Ersatz Division captured a position of 1800 
meters front to the 4th line within twenty-five minutes 
after an artillery preparation of only seventy-five minutes, 
the same procedure having had satisfkctory .results .on 
May 6th and 14th. The one liile governing the attack ia 
rapid advance, utilization of the smallest success, and ad- 
vance as far as the effect of our own artillery, the tooken 
morale and the losses of the enemy will permit. Artificial 
construction of an attack forces us, in attacks with a 
objective, not to cross the established line. From an 
le^r viewpoint, the defense was organized on barrage fire 
lasting initially three or four minutes and then decreasing 
and counter-off ^isive preparation in which stretches of hos- 
tile trenches are fired upon and which secures relief for our 
heavily pressed infantry. The realization of the necessity 
for offensive employment of the artillery, even in def epse, 
had not yet become general. In the attack, the hostile 
artillery was to be held down effectively, while the inf antiy 
was to work its way up close to the place where our artil- 
lery projectiles fell. ''Our infantry must absolutdy rid itself 
of all nervousness when our own artillery fire strikes close to 
them. Co-operation between heavy and field artillery should 
be improved. Our artillery fire should be supplemented 
by minenwerfer fire.'' The superiority of the enemy in 
the matter of heavy fiat-trajectory guns was sorely felt 
by us. Aside from consideration of the kind of gun and 
organization of mixed artillery groups proved uniformly 
advantageous in the solution of tasks and facilitated co- 
operation with the infantry. For support of the infantry 
only the common tasks, frontid and flanking effect combined* 



POSITION WARFARE IN THE WEST, 1914.1917 71 

as well as conditions for observation and communication, 
but not the location of the batteries, are decisive. With this 
understanding the issuance of clear orders and satisfactory 
liaison are gained, and personnd (observers, artillery liai- 
son officers, telephone operators) and implements are hus- 
banded. To facilitate co-operation with the infantry it 
became necessary to locate the command posts of artillery 
grcHip and infantry regimental commanders as close to- 
gether as possible. These requirements in the defense led 
to a well advanced long-range artillery group taking the 
hostile artillery as its principal target and a dose range 
group, kept f arthar in rear, to fire on the hostile infantry. 
All artillery groups were under the orders of the artillery 
commander. Whether that artillery commander should be 
under the orders of the division or corps commander was 
a matter of discussion for a long time. Finally, differing 
from the French policy which placed the main portion of 
the artillery under orders of the corps commander, our 
entire artillery, except the heaviest artillery, was placed 
under the orders of the division conmnander. 

LESSONS OF THE BATTLE ON THE SOMME AND THE 
BATTLES IN FRONT OF VERDUN IN THE 

LATE FALL OF 1916. 



After the Autumn battle the Allies, while still employed 
in defense against the German attacks on Verdun, started 
preparations for a new battle on both banks oof the Somme, 
which was to be a joint operation by British and French; 
but considerations in the matter of training the British 
troops compelled the attack to be postponed to the end of 
June. The Russian attack in Volhynia occurred in the 
meantime. 

It may have been thought that the cause of the failure 
in the Autumn battle was insufficient ammunition supi^, 
but now, in addition to a pronounced superiorly in the air 
whidi made its impcHrtance felt in the observation of fire 
and in the participation by airplanes in the ground battle, 
such an amount of ammunition was to be supplied that ao-> 
cording to all human calculations the resistance of the de- 
fender should be completely broken. Thus was created 



72 DEVELOPMENT OF TACTICS— WORLD WAR 

the ^'Battle of MaterieF' which could be fought to a finish 
only by timely and maximum exertion of all munition in- 
dustries. Delays at home reacted bitterly on the troops 
that had to face that storm. The anmiunition brought 
from home in the first weeks was hardly sufficient for the 
barrage fire and for the absolutely essential support of the 
infantry. There was no ammunition for counter-battery 
work. In addition the attention of the supreme command 
was fixed completely on the battles around Verdun when 
the first signs of another serious battle on the Somme be- 
came noticeable. Thus the German defense was from the 
very beginning under very unfavorable conditions; but 
what did it not achieve in spite of these ! 

According to French reports the leading German posi- 
tion consisted of 3 main lines of a total depth of 500 to 
1000 meters, between which there were underground shel- 
ters and machine gun nests. A second position, consisting 
of 1 or 2 main lines and protected by excellent obstacles, 
was 2 to 4 kilometers distant from the first, so that the 
artillery of the attack would be forced to change position. 
Between the two positions preparation had been made for 
conduct of the battle by ^^ock'' trenches. 

In April, a systematic artillery fire took place on the 
British wing; the road net was extended; positions rein- 
forced; mine warfare started; and plenty of ammunition 
and subsistence stored. By the end of April the German 
command no longer doubted that an attack was to be ex- 
pected here. On June 22, 1916, the artillery preparation 
began. During the concentration of the hostile artillery, 
however, the German artillery had been able to cause heavy 
losses. The number of allied guns was materially larger 
than in the Autumn battle in the Champagne ; the method 
of the bombardment differed by the fire being more con- 
centrated against single stretches of trenches; and strong 
gas bombardment was interspersed 'from June 24, 1916, 
to July 1, 1916, on a front of 25 kilometers in forty differ- 
ent places." (Haig.) Our drum fire, fire against the Somme- 
bridges and gassing the ground in rear, occurred from 
June 25 to June 30. Hostile reconnaissance operations, 
started in the last week prior to the attack, proved excel- 



POSITION WARFARE IN THE WEST, 1914-1917 78 

lent. They furnished important information concerning 
the effect of the fire on trenches and batteries. On June 
26th battle airplanes made an attack against German cap- 
tive baUoons, destroying 9 of them. The attack, which was 
to start on June 30th according to the statement of a de- 
serter, started at 7 :30 A.M., July 1st south of the Somme, 
and two hours later, north of the river. It was apparent 
that thereby the attention of the German reserves was to 
be distracted ; but that two hours' space of time served only 
to preclude surprise on that part of the battlefield. 

Joffre's. systematic conduct of attack, reckoning only 
on mechanical means and merely frontal pressure by masses, 
declining the use of cover offered by the terrain and favor- 
ing the mixing of all units, was changed in this Somme bat- 
tle. Front line units stood alongside each other with great 
formation in depth. Thus in each regiment 6 companies no 
longer were formed abreast in a dense skirmish line; but 
only 8 half-platoons from 4 companies were formed abreast 
in light skirmish lines, with intervals from 4 to 5 paces. The 
regular construction of trench system had been abandoned. 
More attention had been paid to surprise. Stress was laid 
on the necessity of the artillery accompanying the infantry 
by fire in such manner that the fire would be lifted to posi- 
tions in rear only immediately preceding the penetration. 
Airplanes were employed as ''infantry observation planes'' 
to locate the positions of the infantry in the terrain. Joff re^s 
demand of a penetration deep into the enemy's lines, to con- 
nect the operative penetration with the tactical penetration, 
could not be carried out in the Autumn battle and the 
French attacks came to a stand-still in front of our weakly 
held second positions. In the Battle of the Somme this pro- 
cedure was followed as a fixed rule. The enemy contented 
himself with attacks against a fixed objective, renewing 
the same procedure after renewed artillery preparation. 
This ''eating through" the entire position system was char- 
acteristic of the Somme battle. The French attack carried 
the first position, but then encountered serious resistance. 
The British attack, preceded by explosion of mines and ac- 
companied by smoke clouds, did not initially have the same 
success, though its objectives were only 6 kilometers dis- 



POSITION WARFARE IN THE WEST, 1014-1017 76 

tant. The successes of the center and the right of the at- 
tack were exploited by the employment of reserves, so that 
finally the left also made progress. According to Marshal 
Haig's report a night attack on a large scale took place on 
July 14th. After a march covering 1200 meters a position 
was taken, at a distance of 250 to 400 meters from the 
enemy, under cover of security detachments on a f nmt of 
6 kilometers and the attack started at 3:26 A.M. against 
the Trones woods, Longueval and Bazentin le Petit. After 
the artillery preparation that had started on July 11th, the 
attack resulted in the cloture of the Trones woods. It is 
stated that cavalry performed excellent service in this at» 
tack. On July 1st the defense had, n<Hrth of the Somme, 
five German divisions on a 8&kilometer front; and south 
of the river, four divisions on a SS-kilometer front. From 
July 19th on, the defensive front was formed into 2 armies 
in place of 2 army corps as previously, separated by the 
Somme. On the average, 1 German division held 7 to 8 
kilometers, against which the enemy brought 3 to 4 divi* 
sions. Placing fresh troops into line and thereby changing 
the sector limits did not prove advantageous to the battle 
command. 

On September 15th tanks were used for the first time. 
In the beginning they gained success by surprise, which 
however did not last long. Far more annojring was the 
firing on the trenches by low flying battl^ airplanes equipped 
with machine guns. The fear of being seen by the 
airplanes and then annihilated by the hostile artillery fire 
started by the information given by the airplanes, prevented 
the infantry for some time from taking countermeasures 
until they became convinced that this fear was groundless. 
Automatic rifles of the allies played a very important and 
effective role during the attack. Locally they proved of 
great value in the defense, wherever buildings with cel- 
lars were available for defense. Thanks to its ^'catacombs'* 
Combles was held from September 16th to 26th, when it 
was voluntarily evacuated by the few German defenders. 
An oppcntunity offered during the evening to the cavalry 
held in readiness in an advance by way of Sailly eastward, 
was not noticed by the British. 



76 DEVELOPMENT OF TACTICS— WORLD WAR 

The Somme battles concluded with a heavy attack on 
November llth« The penetration had not been accom- 
plished; but the object of the British conmiand in the at- 
tack, of holding the German forces had been accomplished. 

For their success, the Allies had to thank the enormous 
expenditure of anununition and their superiority in the air, 
which could only be offset gradually. A numerically inferior 
artillery can in the end hold down a superior artillery if it 
has superior aerial observation facilities. But an artillery 
superiority should make itself doubly felt if the enemy, as 
was the case on the Somme, is superior both in number of 
guns and aerial forces. But in spite thereof the Allies had not 
been able to destroy the weak German artillery, but only 
lessened its effectiveness. In place of the destruction of 
the hostile materiel, the attack demanded annihilation of 
men or even only the reduction of their power of resistance 
for the decisive moment of the assault. This led to gassing 
shelters of the defense. The German infantry, though in- 
ferior in numbers, understood how to evade the hostile fire 
by moving forward or sideways; and was not shaken to 
the extent the enemy had expected. It is true that the 
trenches were in a very short time changed into a field of 
shell-holes; but the deep shelters with overhead cover six 
meters thick, hdd out and were found useful as long as the 
entrances were kept open. The artillery defense consisted 
mainly in firing a barrage which however was ineffective 
when the signals from the front line were not observed or 
when the larger portion of the guns had been destroyed by 
the hostile artillery preparation. This caused ceaseless com- 
plaints concerning the lateness of the barrage or its lack 
of density. The infantry, trusting entirely too much to 
the barrage, easily failed to defend itself independently. 
Orders from higher headquarters at this time compelled the 
artillery defense to be conducted offensively before the at- 
tack proper started. The exhausted German divisions needed 
relief. Though the quickly exhausted infantry could be 
withdrawn the relief of the weak artillery could not be ac- 
complished. Thus units were disrupted. As a result the 
commanders of the Infantry and the artillery were not per- 
sonally known to each other any longer. Fresh divisions ar- 



POSITION WARFARE IN THE WEST, 1914-1917 77 

rivin^r after the loss of the first position found very bad and 
weak fortifications, which contained nothing necessary for 
a successful defense. Labor within the zone of the hostile 
fire was impossible. Still those troops held out bravdy, 
cheered by the example of their cheers. It is no wonder 
that the infantry during these heavy battles in which it 
had to suffer patiently under tlte heaviest kind of fire, de- 
teriorated in battle efficiency. Only timely relief^ thorough 
training during the rest periods and offensive conduct of the 
defense could return to the German infantry its former 
superiority over the enemy. The hostile airplanes found 
special difficulties in locating for their artillery the German 
lines in that shell-torn ground, so that the latter had to be 
content with merely sweeping certain portions of the ter- 
rain. Losses for the defense were thereby decreased. Posi- 
tions, that could be definitely determined by ihe airplanes, 
were destroyed and obliterated in a very short time. This 
explains why our men were much opposed to converting 
shell-holes into trenches. At any rate, the losses increased 
as soon as the trenches appeared on the photos taken by the 
airplanes. Trenches thus rapidly constructed could of 
course, not offer the same protection as would trenches on 
which weeks of labor had been expended. The troops soon 
adopted the expedient of digging trenches in rear of the 
shell-holes that were to be defended. This diverted the 
fire from the occupied shell-holes; and during lulls in the 
battle and at night, offered protection against weather. It 
was a question whether on the whole it would not be better 
to abandon the time-consuming construction of trenches 
and confine ourselves to the mere construction of obstacles 
and shelters. Though this might have been justified in case 
of the first construction of works in rear, it was on the 
other hand, extremely dangerous to abandon trench con- 
struction completely. Command, subsistence of the men, 
care of the wounded, issuance of orders, and the service of 
information would have been made more difficult. In addi- 
tion, there was the rapid consumption of man power in the 
damp Autumn weather. Ip, the shell-holes the men were 
in no way protected against the weather and often stood 
in water up to their waists. As obstacles were lacking in 



78 DEVELOPMENT OF TACTICS— WORLD WAR 

the sheU-tom terrain the dispersed troops were open to at- 
tack by any hostile patrol and it would have been diffietdt 
to organize a counter-attack. The difficulty of giving our 
artillery the exact location of our men» decreased the possi- 
bility of laying the barrage exactly in front of our line. In 
addition in a position in the shell torn terrain there was 
continuous movement caused by relief and by the inclina- 
tion of the men to evade striking projectiles. This caused 
loss of contact and co-operation with adjacent troops. As 
we have learned, contact by eye alone with the neighboring 
units does not suffice in general. The commanders must 
consult and agree on combined action. ThK>ps holding their 
place stubbornly in a frontal attack pay no attention to 
the flanks, if they know supporting troops are there. If an 
adjacent unit gives way and the attacker follows him up 
and turns toward the flanks, the troops holding their ground 
will find themselves in a precarious situation, which they 
can overcome only by withdrawing (defense of Combles) 
or by a counter-attack. The points of contact of two units 
become weak points, against which the enemy preferably 
directs his attack and which the defense meets by placing 
in readiness a special connecting detachment. The neces- 
sity of reducing the effect of the hostile artillery led to a dis- 
tribution of all works in depth. It was found best to post 
the machine guns, withdrawn from air observation, between 
the first and second positions, especially when it was possi- 
ble by constructing obstacles, to guide the attack into defin- 
ite directions. In the intermediate terrain nothing was 
shown which would draw the artillery fire ; and on the other 
hand, false works may have a good effect in diverting the 
hostile fire. The artillery also had to form its observation 
stations and battery positions in depth with numerous alter- 
native positions, so that the hostile infantry could not re- 
duce them in one effort. In this disposition, of course, we 
had to reckon on the loss of the foremost batteries. 

There is no hard and fast rule governing the strength 
with which a position should be occupied. At Gommecourt 
the defender believed he would be able to get along in 
trenches without shelters emplojring 1 rifle to every S 
meters, whereas prior to the attack of July 1, 1916, in the 



POSITION WARFARE IN THE WEST, 1914-1917 79 

well constructed sector of Fricourt, there were 800 rifles 
and 8 machine sruns for each 1000 meters and this was be- 
lieved instiiiioient; The more the offensive def Mise is pre- 
pared in advance, and the more the hostile fire strikes only 
the leading trenches, the smaller the garrison of the lead- 
ing trenches may be. In the 12th Infantry Division (Ger- 
man) it was found sufficient to have on a front of 6 Idlome- 
ttfs, 1 battalion per 500 meters. 2 ccnnpanies of each bat- 
talion were in the front trenches, each company having 1 
platoon in the front line, li platoons in the second line, and 
i platoon in the third line. 1 company of each battalion was 
in battalion and 1 in regimental reserve. 

On the Grerman side, it had been sufficiently perceived 
that a serious attack, involving a penetration, could be 
stopped only by a position of great depth, so that it was im- 
material to absolutely hold the first trenches requiring dense 
skirmish lines and resulting in increased losses. It was 
seen that it would be far more advantageous to conduct the 
defense offensively by giving way under the hostile fire, 
and then seeking a decision by an independently organized 
counter-attack by the troops in readiness and by the posi- 
tion reserves, so that at the end of the battle the trenches 
would be again in the hands of the defender. At the 
Somme there was an absence of reserves at many points, 
but wherever a counter-attack struck the enemy imme- 
diately after his entry into the position, success was com- 
plete and nullified the entire artillery preparation of the at- 
tack (Fricourt, July 1, 1916). If the counter-attacks exe- 
cuted by the troops in readiness immediately behind the 
position within about 400 meters, are not successful, then 
commanders of all rank must immediately do everything 
they can to prevent the extension of the penetration. Ex- 
tension in depth must be prevented by using the lines in 
rear, extension toward the flanks by the use of approach 
trenches ; and around the enemy's nest we must immediately 
create a certain ^'locking*' position, the most threatened 
comers requiring special reinforcement. The commander 
must be at once notified of the creation of such a nest. Fre- 
quently, because of human reasons the extension of the 
penetration is reported as a minor one, and that fact leads 



80 DEVELOPMENT OF TACTICS-WORLD WAR 

to arranging a coanter-attack with insufficient forces. And 
just those very 'Trench nests'* that had been created hy 
penetration facilitated the subsequent hostile attack very 
materially, because they were made points of departure of 
new operations. 

It was found urgently necessary by army headquarters 
to have fresh divisional and army artillery in readiness for 
insertion. Only thus was it possible to withdraw the infan- 
try and artillery simultaneously from the fighting line and 
to give them some rest for recuperation, for replacement of 
men and for training prior to putting them in line along a 
quieter front. During the Somme battle this was impossi- 
ble and had its effect on the morale. 

The Battle of the Somme led to a complete change of 
views concerning the conduct of the defense. The passive 
endurance of the hostile fire ceased. A fresh, offensive 
spirit was incorporated in the conduct of the defense. On 
August 29, 1916, Field Marshal von Hindenburg, who had 
been promoted from Colonel-General was assigned to the 
supreme command and General von Ludendorff was ap- 
pointed First Quartermaster General. At once, in a new 
regulation ''the Defensive Battle,^' the new method of 
fighting was presented : ''The object of battle defense con- 
sists in letting the attacker wear himself out, bleeding him- 
self white, but saving our own forces. The defense will be 
conducted not by putting in line large numbers of men, but 
mainly by inserting machines (artillery, minenwerfers, ma- 
chine guns, etc.) . At the same time, in addition to the ques- 
tion of numbers, organization, employment and co-ordina- 
tion, all bear an influence. In the distribution of forces, the 
fundamental consideration is saving of personnel.'' The 
troops only reluctantly accepted the general rule of weakly 
holding the front line because it was feared that that line 
could be retaken only with difiiculty. But with the very 
short assault distance over which the attacker passed, it 
happened only too frequently that the strong garrisons 
could not leave their shelters in time and were captured. 
In addition to the thin occupation of the front line, narrow 
sectors (260 meters for a company of 160 men, on quiet 
fronts 600 meters) and great depth formation were or- 
dered to facilitate a stubborn resistance. 



POSITION WARFARE IN THE WEST, 1914-1917 81 

The Aatumn Battles Around Verdun became of special 
importance. The enormous German attack in February 
was followed by French counter-attacks, which made great 
progress only when General NiveUe, a colonel, commanding 
an artillery regiment at the outbreak of war, was assigned 
to the command of the army group at Verdun. The battles 
of October 24 and December 15, 1916, showed a very pecu- 
liar method of attack, by thoroughly discarding the Joffre 
method based on a syst^natic procedure, declining the use 
of the cover of the terrain and by assigning deeper objec^ 
tives to the attacks then had been the case in the battle of 
the Somme. The first attack-shock was to end at least in 
the hostile artillery position. Difficult points of attack were 
to be covered sparingly by infantry but thoroughly covered 
by artillery. Weak pointe, especially in defiles and depres- 
sions were to be penetrated by strong forces which, paying 
no attention to what happened on the right or left, took 
possession of important sections of the terrain. Everjrthing 
that stiU held on the flank of or in rear of these shock col- 
umns which had pushed far ahead, had to fall by flank and 
rear attacks. ''If we have once opened a door in a wall that 
has to be crossed, then nobody would possibly think of climb- 
ing over the wall.'' From this resulted quite naturally the 
organization of the units into assault units and reserves, 
corresponding to the points to be attacked. The front of 
the assault columns was dependent on the battle task ; the 
deeper the troops were to penetrate into the hostile battle 
position, the narrower the front and the greater the depth. 
The zone of attack of a ''divtsian de breehe'' of 9 battalions 
was to be 2000 to 2400 meters wide, according to Jeffreys 
views only 1200 to 1500 meters in case of 12 battalions. The 
division might be formed either in column of regiments or 
with re^^ents abreast, the latter in column of battalions 
separated by a distance of 600 meters. The attack pro- 
ceeded automatically so that the leading battalion halted on 
reaching its objective, the succeeding battalion passed 
through to a more distant objective. Hie reserve divisions 
followed at distance of only 2000 meters. In the uninter- 
rupted advance by the troops lay the best help for adjacent 
troops, should the latter not advance rapidly enoui^ or be 



82 DEVELOPMENT OF TACTICS— WORLD WAR 

held up. As the initial penetration was almost always easy, 
the determining factor lay with the reserves. It was their 
duty to conduct the subsequent attack. They must follow 
up as rapidly as possible and deepen the penetration. Nivelle 
counted on the fact that second positions are not complete^ 
occupied, and that in all cases gaps would be found through 
whidi detachments under skilled leaders could work. As 
the attack was made on a broad fronts an abundance of ar- 
tillery was necessary. On the one hand, artillery protec- 
tion for the infantry was attained by a rolling barrage^ auto- 
matically advancing 100 meters every four minutes, and on 
the other hand, by forming a mobile artillery reserve, which 
was taken from divisions not designated for the attack. 
Success of the attack rested primarily on surprise, and then 
on artillery effect. Surprise required that we abandon all 
thoughts of time-consuming earthworks, such as routes, of 
approach, bomb-proof battery emplacements and reserve 
trenches, and especially the position from which the attack 
is to depart be not advanced. It is better to make the at- 
tack across an extended open stretch under protection of 
the artillery, than to lose the advantage of surprise. For 
that reason the air forces should not be increased prema- 
turely on the attacking front. A planned attack against the 
hostile captive balloons should not be started earlier than 
one day prior to the artillery preparation. Counter-battery 
work is of particular importance. The hostile artillery must 
be neutralized before the artillery preparation proper be- 
gins ; then the hostile artillery must still be kept under fire. 
Nivelle planned an artillery preparation, dep^ident on bat- 
tle conditions, of 6 days. An increase in the volume of fire 
prior to the attack was not to take place. Special emphasis 
was laid on the necessity of using fresh troops in the attack, 
which had been trained in the rest area behind the front. 
Weak points are encountered in every position. If the at- 
tack finds them, a penetration of the position system is possi- 
ble in case the position lacks sufficient depth. This shows 
the necessity of special protection for the artillery of the 
defense. The batteries should at least have some machine 
gun protection in order to give the artillery an opportunity 
to participate in the fighting for the intervening terrain 



POSITION WARFABS IN THE WEST, 1914-1917 88 

( witii poirttioiis for observation dose to the batteries) . Thus 
too, is evident the necessity of previously prepared plans for 
strong and timely counter-attacks, which, however, can 
be effective only it arrangements for prompt transmission 
of information have been perfected. 

Prior to the December attack the 4 French Divisions 
(87th, 38th, 126th, 188d) had been trained from 5 to 6 weeks, 
commencing with the school of the soldier and ending with 
division maneuvers, in a terrain similar to that over which 
the attack was to be made. The attack was to be made 
from the east bank of the Meuse to the village of Vaux in 
a northerly direction towards Louvemont, Hill 878 and Be- 
zonvaux on a A*ont of 9 kilometers. On December 11, 1916, 
the assault divisions were brought up by auto trucks. The 
artillery preparation began at the same time. On the night 
of December 13-14th the divisions took their places in the 
line; and early on the 16th, all were in position. The re- 
serve divisions which were to relieve them later were in 
readiness in rear. The left division (126th) was to pene- 
trate the German position only 500 meters, the other to 
penetrate up to about 3000 meters. At 12 o'clock (10 
o'clock French time) the penetration, without increasing 
the artillery preparation, was to start either in skirmish 
lines or smaU assault columns. The surprise succeeded, so 
that the Grerman barrage came too late, and men in the deep 
shelters were taken completely by surprise. The turning 
off by the penetrating troops to the flanks proved correct. 
Special units had been placed opposite the division sector 
limits and in depressions, as the least resistance was ex- 
pected there, with the mission of advancing against the 
flanks and rear of hostile units still holding out. It was left 
to the initiative of all subordinate commanders to exploit 
the success should the troops encounter little or no resis- 
tance. Greneral Nivelle undoubtedly gained a great victory. 
The main cause of the failure of the defense was probably 
the lack of proper advance preparation of the intermediate 
terrain, and of preparation for co-operation between infan- 
try and artillery in the event of the loss of the leading ob- 
servation positions. In addition the defense was conducted 
completely passive, though four divisions were in readiness 



84 DEVELOPMENT OF TACTICS— WORLD WAR 



behind the front. It was shown plamly that a single forti- 
fied line was not sufficient to stop an attack formed in depth. 
Fortified areas in great depth have to be arranged which will 
enable us to hold the terrain even if separate portions are 
lost. In this organization the different field works must 
support each other, and we must also be able to prevent 
the enemy from extending towards the flanks after he has 
penetrated. 

In December, 1916, Nivelle was appointed conmiander- 
in-chief of the troops in France* His instructions for at- 
tack dated December 16, 1916, were captured two months 
later by German troops in the successful operation against 
the heights south of Ripont on February 16, 1917. 



The XVI Army Corps had become engaged in stubborn 
battles in the Argonnes. Mountainous, broken forest ter- 
rain and a tenacious enemy, superior in numbers, allowed 
but slow progress so that even mine warfare was again 
resorted to and minenwerf ers played a great role. Attacks 
with limited objective started. The smaller attacks proved 
less successful than attacks on a larger scale. 

Thus, positions could be taken only after thorough pre- 
paration by artillery and minenwerfers. Requirements 
were that the positions for departure of attacks should be 
as close to the enemy as would permit firing on the trenches 
by minenwerfers without at the same time firing on our 
own. A charge across 30 to 40 meters ground in open ter- 
rain was accompanied by extraordinary high losses. In 
most instances, the time for the assault was chosen so that 
the hostile trenches could be taken before dark in order 
that the troops organizing the position for occupation and 
defense might be hidden from the hostile fire. Preceding 
the attack, the assault detachments of infantry and pioneers 
underwent careful practice behind the front, in position and 
open warfare, which included accurate synchronization of 
watches. If the enemy had been thoroughly shaken the 
assault generally succeeded easily and quickly. If we al- 
lowed even a few minutes for the enemy to recover, the 
hope of success faded; and even if the attack succeeded, it 



POSITION WARFARE IN THE WEST, 1914-1917 86 



was only at top great a cost. The assault itself demanded 
accurate co-operation between infantry and artillery. The 
assaulting troops must not deviate from the procedure 
agreed upcm. Otherwise they would run into our own artil- 
lery fire. The main weapon of the assault was the hand 
grenade. The troops were to pay no attention to connec- 
tion with neighboring troops. They were to charge through 
to the designated objective and then seek to help tiie less 
fortunate neighbor. The assault against the hidden pill- 
box was very difficult. Their resistance was most quickly 
overcome by the use of flame throwers. 

UTILIZING THE EXPERIENCES GAINED 

On November 18, 1916, the Battle of the Sonune was 
concluded with a final British attack ; the battle had lasted 
four and a half months. The Training Regutatiana far Foot 
Troops (German) of January, 1917, published the exper- 
iences of aU for the benefit of aU. General introduction of 
hand grenades, automatic rifies and steel helmets, facili- 
tated the command in the infantry. Light minenwerfers, 
flame and grenade throwers were assigned as auxiliary 
weapons. The winter battles in the Champagne demanded 
wen trained men to handle these weapons. 

For support in specially heavy attacks, and also for 
use as shock troops, special assault battalions were formed 
which, armed with the auxiliary weapons (machine guns, 
flame throwers and infantry guns) , were attached by groups 
or platoons to the infantry. These shock battalions were 
basically different from the Russian Scout Detachments 
which were formed from the best men of the companies. 
These scout detachments in consequence of their special 
training, performed valuable services. They operated under 
the direct orders of the regimental commander. There can 
be no doubt of their usefulness in difficult terrain and occa- 
sionally in minor warfare; but in large battles, there was 
a lack of the necessary maneuvering space for their em- 
ployment. Combining the best elements of a company into 
a special detachment appears always dangerous, for the 
troops cannot do without their best men to replace fallen 
leaders and to serve as the backbone of the company in dif- 



86 DEVELOPMENT OF TACTICS— WORLD WAR 

fieult situations. In the Boer war the battle value of the 
British infantry decreased rapidly through the continual 
detachment of the best men for duty with mounted infan- 
try. A wen trained infantry unit should be able, undw its 
own commanders to perform the same services as a scout 
detachment. On the other hand, the assault troops remained 
with their companies where they were useful as leaders in 
the fight, and for particular missions they may also be 
temporarily employed away from their companies. The 
''assault'' battalions were complete units, which get their 
recruits from all arms and are about equal to J&get bat- 
talions. In France also, about January, 1917, assault groups 
were organized (groupes francs) or (grenadiers d'elite). 
In the Russian army (August 24, 1917), a shock battalion 
of four companies (each of 3 platoons) was formed in each 
division and formed into a technical unit. 

The shock squads (1 noncommissioned officer and 6 to 
8 men) ''lead the infantry at difficult points to open the 
points of entry into the hostile position, roll up hostile 
trenches, capture hostile machine guns and pill boxes and 
support the infantry in consolidating the captured position." 
(Field Training Regulations, I, 889.) The method of the 
trench fighting taught progressively on the training ground 
should be participated in also by the infantry shock squads. 
''The company possesses in its assault squads, composed of 
the best men, suitable leaders in the fight who should be 
placed at points where the strongest hostile resistance is 
expected" (244). Arrangements should be made for the 
participation of auxiliary weapons where artillery is lack- 
ing for strong fire preparation. Assault squads advance 
during an assault with the leading wave which is to pene- 
trate to the farthest objective. 

Machine guns are attached to the assaulting infantry, 
and also distributed to the waves in rear, for taking strong 
points, for flank protection and to neutralize hostile machine 
guns. In the defense, the automatic rifles form the main 
weapon of defense against the assault, while the machine 
guns will find their most effective employment in rear of the 
front line and are placed in the front line only, in numbers 
absolutely necessary, to furnish harassing and barrage fire. 



POSITION WARFARE IN THE WEST, 1914-1917 87 

Defilacted and concealed positions are essential for their 
effective employment. As 'Veapons of the intervening ter- 
rain'' they facilitate stubborn resistance and also find em- 
idoyinent in. defense agiiinst airplanes. 

The needs of the defense govern the dispositions of 
troops for the first line. The regiments are placed abreast 
and are generally divided into position battalions, reserve 
battalions, and rest battalions. The first are the front line 
battalions with attached auxiliary arms and artillery ob- 
servers. The second are the battalions which, reinforced 
l^ security detachments, support the front line battalions 
in their most important task, by emergency garrisons and 
^joint" detachments reinforced by machine guns ; and which 
independently eject the enemy from the leading trenches 
he may have taken. 

The Regulations gave special attention to the offensive 
defense. The trench garrison must hold out at aU costs, 
even though in danger of being surrounded by the attacker. 
'The holding of the position then depends on the success of 
counter-attacks. Until the latter start, every man of the 
garrison and every gun must hold his or its place'' (227). 
Counter-attacks should be made at once and without await- 
ing orders, not alone in our own but also into a lost adjacent 
sector. 

Based on subsequent experiences and after the lapse 
of one year, the Regulations for Field Training II appeared. 

''Thorough training in the use of the rifle and machine 
gun as well as efficient handling of hand grenades enable 
the infantry to successfully carry out its task. The rifle 
and hand grenade, however, can only achieve their very 
best effect by supplementing each other" (177). The last 
sentence is of special importance, as the troops were in- 
clined (and even our opponents complained of it) to neglect 
the rifle in favor of the hand grenade. The first thing to 
do is to stop the enemy with the rifle and to resort to the 
hand grenade only when the enemy is within throwing dis- 
tance or when he can no longer be fired on with the rifle 
( 192) . A new note in these regulations was the demand that 
"all noncommissioned officers and men must be trained in 
the shock-squad procedure," and that shock squads for spe- 



g8 DEVELOPMENT OP TACTICS— WORLD WAR 

cial tasks be formed 'from specially suited men from time 
to time'' (226). 

The directions for fighting against tanks (416-419) and 
against low flying battle airplanes (420-^9) were also new. 
Each machine gun, that is not occupied in the battle against 
ground targets, must direct its fire on low flying airplanes 
(423). 

The general directions for the conduct of position bat- 
tles were proven correct throughout. The machine guns 
had been specially useful as arms in the intermediate ter- 
rain, especially in nests of one to three guns under infantry 
protection. These nests were again combined under one 
leader into defense groups, to which dispersed men were 
to hasten. Skillfully nested in, and hidden from hostile ter- 
restrial and air observation, the machine guns opened fire, 
only at the moment of the first hostile assault, by surprise, 
and with flanking effect whenever possible (406) . 

The battalion commander of the front line battalion 
was the ^'fighting troops" commander. Heavier emphasis 
than in the first Regulations was laid on the offensive de- 
fense and the necessity of cutting x)ut security detach- 
ments. For the counter-attack the commander of the bat- 
talions in readiness must not await request for counter- 
attack. To his companies battle tasks and definite battle 
sectors can be assigned before the start, in which they are 
responsible for timely and independent counter-attacks 
(445). The battalion commander was responsible that the 
counter-attack hit the enemy before he has had a chance 
to consolidate the captured position. 

THE BATTLES IN 1917 

In the Spring of 1917 the German command reckoned 
on a repetition of larger attacks in the Champagne and in 
Picardy. Valuable basis for the conduct of the def ensfe was 
furnished by Nivelle's Regulations in conjunction with the 
experience gained at Verdun. It was seen that the best 
counter-means for the new attack procedure was a weak 
garrison of, and even temporary abandonment, of unfavor- 
ably situated first lines. Absolute abandonment of the fore- 
most trenches had given the usual French strong reconnids- 



POSITION WARFARE IN THE WEST, 1914-1917 89 

sance shocks prior to the attack a new point for starting 
the assault, and had not made it necessary to insert stronger 
forces. If the penetration succeeded, then the assaulting 
troops were to be attacked by the garrison holding the in- 
termediate terrain from all sides with fire and held, so that 
the French infantry would lose the protection of the auto- 
matically advancing fire wave, and that was the prerequisite 
for the subsequent offensive defense by all arms from the 
depth, and by reserves brought up. The method of the French 
fire distribution left no attentive defender in doubt concern- 
ing the selected point of entry, so that the reserves could 
be placed in position in time. The immediately started coun- 
ter-shock — ^into the uncertain — ^was the finger mark of the 
new leadership. Thus was created the elastic defense, tem- 
porary evasion of the annihilation fire to the front — ^which 
however found its limits at the obstacles — and then the 
assault-shock of the reserves. This method required also 
special artillery preparation : locating the observation posi- 
tions in direct proximity to the batteries, and special ar- 
rangements for communication. Of importance were the 
directions that pauses in the hostile fire must not lead the 
defensive artillery into inactivity, as those very pauses 
allowed the enemy to bring up ammunition, to coinplete his 
attack preparations and to bring up his assault troops. 
From these facts we can now judge why the German army 
leadership decided, in the face of the progressing attack of 
the enemy, to evacuate a large projecting arc of the West 
front, and to go back to the ''Siegfried Position," under the 
protection of independent detachments which had been left 
behind, and after the execution of extensive destructions. 
The object had been attained, time had been gained, and the 
enemy had been compelled to abandon his plan of attack. 
The Allies were forced to far separated attacks at Arras, on 
the Aisne and in the Champagne. 

Possession of the terrain plays a very minor role, 
though it had long been considered, especially at home, as 
the means of determining whether a situation is favorable 
or a little less favorable, and though the enemy credited 
himself with success in taking possession of a piece of ter- 
rain that had been voluntarily abandoned by his opponent. 



90 DEVELOPMENT OP TACTICS— WORLD WAR 

Thereafter the German command husbanded its strength 
and held terrain only as long as it was warranted by the 
means that had to be used therefor. 

On the side of the Allies also no time was lost in with- 
drawing their attack divisions temporarily to the rear in 
order to practice in the open terrain the attack against the 
German trench system (4 to 5 kilometers depth). This 
practice began with the platoon. In addition to the battle 
and march practice proper and to training in the service of 
special arms, rifle firing and physical training was carried 
on, with and without packs. To hasten the advance, move- 
ment across country as well as by approach trenches was 
directed, thus minimizing the effect of the German machine 
gun fire. Intervals between skirmishers were 4 to 6 paces 
(3 to 3.75 meters) . In terrain devoid of trenches all column 
of fours were to be discarded, the platoons to advance in 
open order, as units or in lozenge formation (see plate) , to 
lessen the effect of hostile artillery fire. 

COMPANY IN LOZENGE FORMATION 

1. 

II 

8. S. 

II II 

4.' 

II 

In forests the troops assumed single formation. Cap- 
ture of villages was left to the tanks. Woods were entered 
only after the artillery had fired upon them for at least 20 
minutes. 

In the attack within the division, 1 regiment was gen* 
erally placed as a shock-detachment with 3 battalions (each 
760 meters, companies 250 meters front) abreast on a front 
of 2400 meters. 2 regiments followed in reserve. The com- 
pany was formed into 2 waves, following each other at 80 
meters distance. The diistance of the battalion reserve was 
400 meters. 



POSITION WARFARE IN THE WEST, 1914-1917 91 

The preparations of the British Army for the Battle ct 
Arras were made in the same manner ; it appears that spe- 
cial weight was attached to increasing the artillery materiel 
In a sector north of Arras oi| a 5-kilometer front 456 light, 
240 heavy and heaviest guns and 68 minenwerfers were 
counted. On April 9, 1917, after an enormous expenditure 
of ammunition, the British infantry forced its way into the 
German position under protection of a rolling barrage and 
tanks on a front of 17 kilometers. Between Gavrelle and 
Roeux (5 kilometers) a depth of penetration of 6 kilometers 
was attained. By the well laid and exceedingly heavy bar- 
rage portions of the German garrisons in the front trenches 
were completely cut off from their reinforcements. Tlirougfa 
the loss of dominating ridges in the leading line, which had 
been boxed in by the British artillery, the hostile fire com- 
pelled a withdrawal at some points of 4 kilometers. Gas 
was employed on a very large scale. The guns that had to 
be abandoned to the advancing enemy were made unservice- 
able and blown up. A British a1i;empt to penetrate at 
Roeux with a strong force of cavalry was defeated by strong 
fire. On the 23d the front of the penetration was increased to 
34 kilometers ; any further increase or advance was made im- 
possible by the energetic counter-attacks of the German 
reserves. 

After artillery preparation lasting for days the British 
divisions attacked with the assistance of tanks along a divi- 
sion front of 1500 to 2000 meters, mostly with 2 brigades 
(4 battalions) in four waves. These were to pass through 
each other in such manner that the leading wave halted on 
the first objective, where the second wave passed through 
the first wave to the second objective ; and similarly followed 
by the third and fourth wave ('^eap-frog system'')* In 
this manner an organized front was always to be offered to 
the German counter-shock. (See plate.) 



92 DEVELOPMENT OP TACTICS— WORLD WAR 

FORMATION OF A BATTALION 

(1917) 

Of the 9th Inf. Div. 

B A 
2 platoons 



60 

1 platoon 



60 

1 platoon 



Of the 8l8t Inf. 
C B 


Div 


A 










86 


m 






1 


46 


m 



60 m 90 



60 m D 



60 



Note: — ^In En^^and the ocnnpanies are designated by letters. 

The infantry was formed into waves or followed in nar- 
row, short columns, which were pref erred^ in order to avoid 
losses, from the creeping barrage closely preceding the 
troops, from shorts or from fragments thrown back. Ev- 
ersrthing had been very minutely arranged in orders and 
nothing was left to chance or to the initiative of the com- 
manders. In order to avoid the standing barrage of the de- 
fender, the first two lines (the first in very open skirmish 
line, 4 to 6 meters intervals between skirmishers, the sec- 
ond often in ishort narrow single columns) were sent before 
the assault beyond the leading trench, while the third line 
occupied that trench and the fourth immediately formed 
in rear of it. At the start of the assault the artillery laid 
its fire on the front to be penetrated and moved it forward 
slowly (for instance 90 meters in four minutes) first the 
heavy calibers, followed by the field artillery with the ma- 
chine guns between the two. In many cases advance was 
made under protection of smoke clouds. The assaulting 



POSrriON WARFARE IN THE WEST, 1914-1917 98 

infantry followed the creeping barrage immediately. Losses 
by its own artillery fire had to be reckoned with. Halts 
were made to prevent the creeping barrage running away 
from the infantry. The barrage was then designated a 
standing barrage. It was intended to produce an effect 
on any counter-attack and gave the infantry the oppor- 
tunity to entrench in expectation of a counter-attack and 
to reorganize. The first wave remained in that position, 
while the rest continued the advance in the rear of the 
creeping barrage. It is true that the necessity of taking 
advantage of the initial success was theoretically acknowl- 
edged, but exploiting the success was left to the subordinate 
commanders up to the battalion commander, who, being 
insufiiciently trained, completely failed. In every instance 
it was proven a mistake to put in cavalry masses without 
attaching infantry to them as ''pace-makers'' (to clear 

PLATOON IN FORMATION UNDER HOSTILE 

ARTILLERY FIRE 

1 squad hand grenade throwers (HG) ; 1 squad automatie rifles 

(AR); 1 squad riflemen (RM); 1 squad rifle grenade 

throwers (RG); total 10 officers, 40 men. 

< 100 meters > 

RM 



/ 



•\ 



<^ 



EG HG "*»20« 

II II X 

' ' " 12 to 46 m 

16 to 26 m 



i 



Platoon commander with 4 messengers. 

Platoon in Two Watis 



^ 



RM HG 

y\ 

12 to 20 m 
RG AR 
\y 



Trench moppers«up, 12 m distance. 



94 DEVELOPMENT OF TACTICS— WORLD WAR 

the way) . Employing divisions to gather the fruits of 
the victory^ was not resorted to tor reasons which may 
probably be accounted for by insufficient training of officers 
and men. Thus, in the Arras battle an entrance into the 
position was made, which, however, was not followed up 
by a penetration because the British command did not ex- 
ploit initial successes. The attacking troops were pre- 
maturely relieved while they still possessed sufficient fight- 
ing power. The troops were trained entirely too theoretir 
cally or according to a hard and fast scheme for the attack, 
and everything depended on the success of the creeping bar- 
rage. They reasoned that any attack that came to a halt had 
better be at once discontinued, and tried at some other 
point. In that case we must hope to strike some portion of 
the front that is weaker, or occupied with more or less worn 
out troops. 

The extensive French preparations for attack in the 
Western Champagne could not be hidden from the German 
airplanes, and had been materially corroborated by state- 
ments of prisoners. Nivelle's main idea of avoiding portions 
of the front lines that were difficult to attack and taking them 
subsequently by flank and rear attacks, was to be carried 
out on a large scale. Such portions, or supporting points, 
were the Brimont and Berry sectors in the Western Cham- 
pagne. An army was to attack beyond the Aisne against 
Laon, to secure the left flank of the main attack and then to 
take up the pursuit. The main attack, in the Champagne, was 
planned so that a stretch of 20 kilometers was to be left free 
in the center at and east of Reims between Bethany and 
Prouvais. On the left the 5th Army (as had been gleaned 
from an order captured April 4, 1917, at Sapigneul) with the 
32d Corps was to push forward towards Brienne and Prou-^ 
vais ; with the next division (37th) , to Suippe ; and with the 
next division (14th), supported by the Russians, to the Bri- 
mont. On the right wing a corresponding advance was then 
to be made by way of Auberive. As strong forces were also 
concentrated around Chalons, it is not improbable that here 
a strong group was placed in readiness for attack, probably 
in the direction of Vouziers and for furnishing flank protec- 
tion. If the German center made a stand, its fate would be 



POSITION WARFARE IN THE WEST, 1914-1917 95 

sealed, provided the wings succeeded. Thus it was a Cannae 
battle in position warfare 1 

According to the captured orders, prematurely issued, 
the attack of the 20th Corps, to be started at 6:00 A.M*, 
April 16th, was to reach the village of Lierval situated about 
16 kilometers north of the Aisne and to occupy the flats 
west thereof as far as the canal in the vicinity of the vil- 
lage of Uzel. Thus the rear of the new Siegfried position 
would be gained and the attack would work towards the 
British. The 6th Corps had the task of working its way 
up west of the 20th (on the Soissons — ^Laon road) and to 
occupy that village as well as the terrain to the west. Then 
the pursuing army was to push through this gap. This or^ 
der assigned to the 89th Infantry Division of the 20th Corps 
for its first objective the German position on the Chemin dee 
Dames; for its second, the Aillette; for its third, Lierval 
and the flats to the Aisne — Oise Canal. In this operation 
the 168d Regiment of the 89th Infantry Division had the 
task of advancing to the attack in column of battalions. 
The 8d Battalion was to take the first objective; the 2d Bat* 
talion, the second objective ; and the 1st Battalion, the third 
objective. The orders contained similar instructions con- 
cerning the adjoining regiments (146th and 166th). The 
132d and 106th Regiments of the 56th Infantry Division of 
the 6th Army Corps were ordered to advance as far as 
Chavignon. The 182d Regiment was to advance alongside 
the canal, the 106th Regiment to ascend the Bovette Hill 
throufl^ the Bois des Ciomites d'Or and then push forward 
in the direction of La Croix-sans-Tete. 

NiveUe intended to overrun three or four GSerman posi- 
tions and expected an advance of 10 kilometers in 8 hours, 
taking the Brimont the first forenoon and Laon the morning 
of the second day. On the second and third day of the at- 
tack the advance was to be more rapid and was to be made 
in the direction of Sedan and Namur to the southern comer 
of Holland. Nivelle held the view that the penetration of 
the (Serman front could be made within 24 hours, or at the 
most 48 hours. Otherwise the battle could not be won. 

The opening of the battle was fixed for the British on 
April 9th, for the French left wing at first on the 15th (later 



96 DEVELOPMENT OF TACTICS— WORLD WAR 

on the 16th), and for- the French right wing on the 17th. 
The blowing up of the Aisne— Champagne front line sector 
and the accompanying fan-like rolling up to the east and 
north showed a plan, well thought out and on large lines, 
which, however, would be successful only if the penetra- 
tion could be accomplished on the first day, or not later than 
the second day and to the depth ordered. If the attacks 
did not succeed one after the other in rapid succession, the 
plan had to fall to pieces of itself. It was unfortunate for 
the attack that at the very start the above mentioned at- 
tack orders for the infantry of the 6th Army Corps in min- 
ute detail should fall into German hands, lliough this fact 
was known to the French headquarters, the preparations 
for the attack were not changed. Thus the advantage of 
surprise was lost. 

On the German side. Headquarters of the 1st Army 
(General of Infantry — ^Fritz von Below) was placed between 
the 7th and 3d Armies. These headquarters had been able 
to gather valuable experience on the defensive in the battle 
of the Somme. As each army (the 7th and 3d) detached the 
corps on its interior wing continuity in the issuance of or- 
ders was assured on the battlefield. Nevertheless disad- 
vantages were connected therewith. Headquarters of the 
1st Army assumed command over the two corps while the 
attack preparations of the French were in full swing; and 
conduct of the defense was made more difiicult by the later 
insertion of fresh divisions. It was found inadvisable to 
decrease division fronts by putting new divisions into the 
line. It would have been better to attach a fourth regiment 
to the divisions already in line and to retain the remainder 
of the new divisions as interference divisions (in reserve). 
Besides it was incorrect to make the division sectors too 
small, as then it became difiicult to post the artillery and 
facilitate changes in its position. One division with an inter- 
ference (reserve) division in its rear could perfectly weU 
hold a sector of 4 to 5 kilometers front, although the front 
of a division should not be extended beyond 3 kilometers. 

Tlie attack was commenced by heavy artillery fire reach- 
ing fiur into the hostile rear. ^ Towards evening of April 6th, 
near Reims, an attack by artillery fire on a large scale, 



POSITION WABFARE in THE WEST, 1914*1917 97 

asramst the German captive balloons and planes, was started ; 
but was unsuccessful, presumably on account of gassing 
by German defensive guns. 

Differing from the Somme battle, in the ''Easter'' bat- 
tle the Germans had a superiority in the air even prior to 
the French attack, which fact became of special use to the 
artillery counter-preparation by enabling the defense to 
adjust its fire on the hostile batteries and other targets far 
into the hostile rear. If a destructive fire was specially di- 
rected against any part of the German line, the hostile line 
opposite was immediately taken under a destructive fire. 

In the French 5th Army, the course of the attack was 
about as follows: April 6th to 11th an artillery prepara- 
tion including, on the first day, attack against the German 
air forces. On the 12th, a decrease of the artillery fire, which 
in the evening again increased to ''drum'' fire. On the 12th 
and 13th, reconnaissance attacks (raids). On the 13th, a 
diminution of the fire, presumably assault troops going into 
position. On the 14th and 16th resumption of heavy fire 
preparation, directed methodically on the second position. 
Night firing on the routes of approach and shelters, prepara- 
tions for bringing up artillery. Early on the 16th "drum" 
fire again. At 7 :00 A.M., the assault. 

The German trenches, that could in most cases be easily 
observed, had be^ changed into a stretch of shell-holes by 
the strong artiUary fire. Attempts to repair trenches under 
this fire, merely increased the losses. On the other hand, 
it was possible to construct in the shell-hole terrain in rear 
of theselectedlin'eof resistance a continuous trench, which in 
daytime had the effect of a false position, and which at night 
offered shdter from the bad weather, to the garrison of 
the front trenches. Now the advantages and disadvantages 
of advanced elements of the position came into considera- 
tion; and emphasis was laid on the necessity of declining to 
hold or recapture unfavorable portions of the terrain in place 
of accepting the heavy losses necessary to hold them. The 
German conduct of the defense was well known to Genwat 
Nivelle. It was mainly : weak occupation of the first line ; 
eontttiding for the first trenches by automatic counter-at- 
tack of the troops in readiness, thus delaying the assault 



'\ 



98 DEVELOPMENT OF TACTICS— WORLD WAR 

while the creeping barrage continued on its way ; defense of 
the intermediate terrain by machine guns ; and then decis- 
ive counter-attack by the reserves starting at the same 
time the attack commenced. General Nivelle demanded raids 
on a broad front, prior to the conunencement of the assault, 
in order to induce the machine guns in the intermediate ter- 
rain to open fire prematurely and to cause the reserves to 
be committed to the counter-attack. Captured trenches 
were to be held as positions of departure for the further ad- 
vance. Special weight was laid on itfie pfotnpt forward dis- 
placement of the artillery immediately after the initial 
success. ''If we require two months to prepare a new at- 
tack, thus permitting the enemy to dig in on a new front, 
we simply play into his hands. The time of preparation 
must be limited to the time necessary to bring forward the 
artillery. The period of two months must be diminished to 
but a few days, even if the question is one of attack against 
fortified positions." The battle confirmed the correctness 
of this German conception concerning the conduct of de- 
fense ; but also showed its weakness which lay at the junc- 
tion of two sectors. Only too easily could differences in the 
estimate of the situation concerning the main defensive 
position arise in the different sectors ; and in that case co- 
operation in conducting the counter-attacks was not as- 
sured and liaison with the neighboring artillery was likely 
to be interrupted. "Joint" (meaning junction, where sec- 
tors adjoin) detachments could offset these disadvantages 
only in part. The danger of occupying large tunnels im- 
mediately in rear of the leading line came into prominence. 
This procedure of course held the large garrison thereof 
firmly in the hands of the commander ; but on the other hand 
the garrison was in danger of being shut in by a caving-in 
and of falling helplessly into the enemy's hands. According 
to French reports, in the Comillet Tunnel the garrison, num- 
bering several hundred men^ was killed by gases of explo- 
sions. Permanent evacuation of portions of the position 
was made only by direct orders from higher headquarters. 
This was of great importance in the conduct of counter-at- 
tacks. Instructions that portions of rifle trenches should 
be prepared so as to provide all round defense appeared nee- 



POSITION WARFARE IN THE WEST, 1914-1917 99 

essary. The known French method of conducting the at- 
tack insisted that each unit assure itself of connection on 
its flanks. The commander was not merely to ascertain 
that troops were there but he had also to ascertain how and 
in what line those troops were to fight. The battle confirmed 
the frequent previous experiences: that portions of the 
position which had been observed by the enemy were quickly 
annihilated ; that routes of approach observed in quiet per- 
iods very soon became unserviceable during battle; that 
troops in the open suffered less losses than those remaining 
in known trenches, which, after evacuation, served well as 
false positions. Counter-attacks conducted by even weak 
detachments had surprising success if they were started 
immediately after the enemy's entry. The enemy must be 
counter-attacked as long as he has not overcome the resis- 
tance of the position, and as long as he has not reorganized ; 
and therefore the counter-attack must be started under en- 
tirely unknown conditions even under the danger of being 
started- too early. The new conception of the counter-at- 
tack gained more and more in value ; it was not only to off- 
set a local defeat but the intention was to destroy the enemy. 
This required the introduction of larger infantry forces; 
and, as we could not absolutely count on the co-operation 
of the artillery, light horse-drawn batteries had to be at- 
tached to the infantry. The "interference" (reserve) divi- 
sions were designated for independent counter-attack. This 
employment of the reserves made it necessary to keep posi- 
tions in rear garrisoned by weak security detachments. But 
if counter-attack could Hot take place within a short time 
after the penetration, it became necessary to make specially 
careful preparation for a systematic counter-attack. Great 
difficulties were encountered in taking up a position under 
cover close to the attack objective. Reorganization of the 
formation in depth ^ter isuccess also presented difficulties. 
The more time the troops were given, the more certain was 
success. 

Early on April 18, 1917, the 5th and 6th Infantry Divi- 
sions with weak artillery, arrived about 8 kilometers north 
of the ridge from Cornillet to Hochberk and Pdhlberg, 
which since the 17th had been in French hands ;,they moved 



M 1 6r 



/ 



100 DEVELOPliENT OF TACnCS-^WORLD WAft 

forward oii the 19th in ccmjundjon With a detaehment of 
the 1461h Infantry Regiment on their right and with the 
28d Infantry Division on their left. They intrenched them- 
selves on the ridge, but did not recapture the Fiehtel hills 
and the first positions at that time, as the enemy was 
ready for defense in a prepared position and had brought 
up Us artillery. A counter-attack started on Ifay 27th 
against the Pdhlberg was successful at first but the cap- 
tured terrain could not be held. A few days later the Hoch- 
berg was also recaptured. Lack of forces prevented a shn- 
ultaneous attack along the entire front, which surdy would 
have had more success, than several partial attacks follow- 
ing each other, rqieating the same attack procedure. 

-The French main attack had gone to pieces with bloody 
loss. The tanks had not measured up at all to what had 
been expected from them. For political reasons Nivdle^s 
lesignation was demanded ; and in his place stepped^ Gen- 
eral Petain, who had come into prominence. The main 
causes for the defeat lay principally in that surprise was 
not attained — ^which is easily explained in such a large at- 
tadc; that the German artillery had shown a tenacity and 
power beyond what had been expected, and that the very 
dastic d^ense was doubtless a great surprise. The French 
cmnmand renounced large attacks for the present. Efuring 
the subsequent battles of the year the French satisfied them- 
sdves, as the British did also, with limiting the objectives 
in breadth and depth, and in most cases the object of flie 
attack was merely the capture of a single position by strong 
forces with great formation in depth (the division on a 
1200^meter front) . After thorough artillery preparatioB — 
including strong gas concentrations on the attack objective 
and after arrangements had been made for the emplosrment 
of tanks, the dense skirmish lines started. If they sue* 
ceeded in their attack, ever:ything was prepared to meet tiie 
counter-attack. As the attacks were of shallow depth such 
counter-attacks were successful only when they struck the 
enemy iHifle still moving. The attack could be carried be- 
yond the objective only by direct approval of the army com- 
nandor. A position of departure for the attadc 500 meters 
from the enemy, and never closer than 200 meters, was re- 






POSmON WARFARE IN THE WEST, 1914-1917 101 

commwded. The assault was to advance under proteetion 
of a creeping barrase; and if the target objective was dis- 
tant» intermediate halts were to be made. If intennediate 
objectives were designated frequently a change would be 
made in the assaulting troops for the subsequent continua- 
tion of the attack. The attack was ccanposed of a series of 
successive (rffensive-shocks, which finally led to a penetra^ 
tion. 

The attack against the Laffaux Comer (October 28, 
1917) the base line of which was closed by the Aisne Canal, 
was a specially good example of the new French tactics. 
Hie 7th Army had no doubts whatever of the coming attack. 
Even the day and hour had become known. As the attack 
had to be accepted under conditions very favorable to the de- 
fense, it had to be decided whether it would be more desirable 
to consider the Laif aux Comer as a foreground zone and to 
execute the defense pnver north of the CanaL Army head- 
quarters, however, insisted on stubborn defense of the lead- 
ing positions in front of the Canal contrary to the wishes 
of the troops and to the views of Imperial headquarters. 
The French intended to penetrate at Pinon and Fort Ifa^ 
maison with the support of tanks, then to turn in in rear 
pf the German position and to cut off the projecting comer. 
October 16, 1917, the artillery preparation commenced with 
greatly superior artillery (for instance, in the 14th Infantry 
Division 125 against 82 German batteries), which very soon 
reduced the effectiveness of the German artillery. The crea- 
tion of a strong gas barrage in the canal lowland of such ex* 
tent and density that bringing up supplies and reserves to 
the front line was absolutely impossible was employed for 
the first time. The interference divisions posted in rear 
were distributed by battalions to the troops in the first line 
of the defense. Shelters that had been located from aerial 
photos were systematically destroyed. The Fr^ich were 
satisfied with tihe line they reached on October 28d, and did 
not even interfere with the German retreat across the 
canaL The simultaneous battles at Verdun show similar 
events. It was specially noticeable that the French under- 
stood how to overcome garrisons by attacking them from 
flank and rear. The advance of the assault troops, ddayed 



102 DEVELOPMENT OP TACTICS— WORLD WAR 

by the defense of the intermediate terrain, was very slow. 
In many instances two hours were consumed in covering one 
kilometer. The movements of the defenders under artillery 
fire were not much quicker, when gassed spaces under heavy 
fire had to be croisised. There were no more attempts at 
imiteid shocks by entire interference divisions. The diffi- 
culties of bringing the troops through the shell-torn and 
covered terrain were great but where counter-attacks suc- 
ceeded, they always had fine results. 

The British Army attached more weight than the 
French to the change in the conduct of the attack. 

Th6 British attack was marked by careful preparation 
and utilization of the different offensive means. Against 
the Wytschaete Arc, the extension of which projected about 
16 Idlonieters beyoiid the line Double Hill 60 at Zillebeke and 
Wameton, the attack preparations had so far progressed 
that by the eiid of April we had to reckon with an attack. 
In the beginning of May the artillery preparations coni- 
inenced,' then a pause occurred from May 16 to 21, 1917, on 
the 22d of May a second preparation, increasing continually 
in volume, commenced so that there was no doubt whatever 
(without considering the numerous raids and statements of 
a fiyer, captured on June 6th) that the assault was imminent. 
On the German side, the means were insufficient for bringing 
up to the defensive troops (35th, 2d, 40th Infantry Divi- 
sions) reinforcements in troops and materiel. In addition, 
the defensive line was situated on the front slope of the 
hill. This, the British engineers had skillfully taken ad- 
vantage of since 1914. The statements of infantry that 
the noise of mining could be heard, were said by our pioneers 
to be incorrect. The standing barrage of the defense was 
too weak. The heavy batteries brought up a few days be- 
fore had been destroyed, except a few guns. Field artil- 
lery, that had arrived shortly before the attack, had gone 
into position in rear of the third position and remained there 
unmolested by the hostile artillery. The infantry had also 
Suffered so much, that the battalions kept in readiness and 
resting had already beien drawn up into the first line. 

On June 7th, at 4:00 A.M., enormous explosions oc- 
curred, followed again by heavy ''drum'' fire, gassing of the 



POSITION WARFARE IN THE WEST, 1914-1917 103 

terrain in rear and employment of a new kind of bomins 
mine. At 5:40 AM. 11 British Divisions of the 2d Army 
(Plummer) started on a front of 16 kilometers. They con- 
tented themselves mainly with reaching an assault position 
in front of the German ''Sehne" position. The resisting power 
of the defender was broken when the assatdt began, but 
everyone that could possibly fight, held out. Tlie remnants 
of the 2d Infantry Division retreated only when they were 
enveloped on both wings. A counter-attack against Mes- 
sines started with portions of the Bavarian 8d and the 1st 
Guard Reserve Division, made good progress at first, but 
was attacked in the right flank. Naturally the German 
losses were very heavy. As in all these battles probably a 
renewed attack with fresh forces would have encountered 
but little resistance. Apparently the British did not think 
of a prompt exploitation of the success. Their reserves were 
used for relieving the troops in the first line, but not for in- 
creasing the entry to a penetration. 

Even the subsequent battles show the same : Destroy- 
ing by "drum" fire a relatively small sector of the front line 
through enormous artillery effect; the attacking infantry 
satisfied with assaults of minor depth; the advantages of 
surprise, not utilized. 

The attack at Cambrai on November 20, 1917, against 
a quiet, weakly held front to a depth of 12 kilometers: 
against the 20th Landwehr Division and the 54th Infantry 
Division brought from Russia, shows an entirely different 
picture. It seems this conduct of attack can be used only 
against such a front. Everything was based on surprise, 
which was rendered more probable on account of the bad 
flying weather. The broad field of obstacles, undestroyed, 
30 meters and at some points 100 meters broad, made it 
appear that any immediate attack was impossible. 

In this instance the obstacles were not to be destroyed 
by a long protracted artillery fire but were to be broken down 
by tanks followed by the infantry. The position had been 
excellently constructed and had not suffered in any manner 
whatever. Statements of prisoners on November 19th about 
the impending attack, were not believed. The British orders 
for secrecy had been strictly complied with. Large move- 



104 DEVELOPMENT OF TACTICS—WORLD WAB 

ments of troops and vehicles were permitted only at night 
and without Ughts. The attacking troops did not come into 
any contact with the front line troops already known to us, 
so that prisoners taken by us were only from the latter 
troops, who naturally could not know or report anything 
about the attack. Strict orders were issued not to fire from 
new positions of the front line artillery prior to the opening 
of the assault. The front line artillery kept up its normal 
fire. As late as the 17th the 2d Gennan Army reported 
that hostile attacks on a large scale were probably not to be 
expected and that there were no indications thereof on 
the northern half of the army zone as indicated by railroad 
traffic observed. A dense fog, artificially reinforced, hid 
the position in readiness. Bursts of destructive fire drowned 
the noise of the tanks, so that when the waves of tanks ap- 
peared the German security detachments were completely 
surprised. After an imme<fiately inserted standing barrage 
they disappeared in the fog and appeared by surprise in 
front of the German batteries. 

The penetration was made by 6 divisions on a front of 6 
kilometers in a first and a second line. 4 cavalry divisions 
were to work through the German position and, turning off 
to the north, to cut off the sectors north of the front. To 
each British division 1 tank battalion had been attached, 
consisting of 8 companies, each of 4 platoons, each of 4 
tanks. The attack was preceded by 6 tanks to ride down the 
obstacles. These were followed by 2 lines, of 18 and 12 
tanks respectively, with 80 and 50 meters distance and a 
rate of advance of 8.2 kilometers per hour. The infantry 
followed at 100 to 800 meters distance. Hie attack was 
preceded as every other attack b|y a creeping barrage. 
Thanks to the surprise and the new method of fighting, the 
penetration succeeded the first day to a deptib of 8.5 kilo- 
meters; but had sufficient artillery of the defense been at 
hand, the attack would have gone to pieces. 

Tanks probably can take a position, but cannot hold 
it. They take the place of accompansring batteries and pro- 
duce an effect on the morale of the defenders who feel so 
helpless. A difference is made, according to armamoit, be- 
tween male and female tanks, the former having 2 light 



POSITION WABFABE IN THE WEST, 1914-1917 106 

field guns and 4 automatic rifles, the latter 6 machine guns. 
Older tanks are used for carrying supplies. Tanks are to 
open a road for the infantry, destrojing the obstacles or 
Iqr direct support of the infantry in beating down machine 
gun nests and supporting points. In attacking villages 
tanks were frequently used. Considering their limited 
height and slow velocity, the employment ot tanks against 
artillery is precluded. The capacity of tanks is smalL In- 
fantry carriers can haul 20 to SO men and 6 machine guns. 
Mobility in the terrain is very good, the tanks can traverse 
slopes of 1 to 2, and can cross ditches 2.6 meters wide. 
Serious obstacles are offered to tanks by swampy ground, 
brooks, and watercourses about 60-cm. deep. The '^passive 
defense'' against tanks consists in such obstacles, construc- 
tion of automatic mines, which will permit passage of in- 
dividuals, but wiU explode on the passage of heavy weight 
vehicles. Bridges of great bearing capacity had better be 
destroyed. 'Tank traps'' on roads are of minor value, as 
the tanks can run across country. As the garrison within 
the tanks is deafened by the noise of the tank in motion,- 
and as the narrow loopholes permit only a very limited view 
of the battlefield, it is necessary to assign men to accompany 
the tanks to inform the garrison of the tank where the 
esn&axy still holds out. In any case the tanks never fight in- 
dividually but always in mass; and never unacccnnpanied ; 
but, like the accompanying batteries, whose place they take, 
in dose conjunction with infantry. The tanks rapidly use 
up their power. Therefore reli^ and formation in d^th 
are absolutely required. 

The gun remains the best defensive weapon against 
tanks. If the troops of the defense have once become used 
to the effect and impression created by tanks, then light 
tanks can be disposed of by machine guns using armor- 
piercing bullets; the heavy tanks by use of increased 
charges, or by direct flat trajectory fire of the light minen- 
werf ers or by the fire of specially constructed anti-tank guns. 
Even larger caliber (18-mm.) rifles with special ammuni- 
tion has been found effective. The more the fire of the 
diBeresat arms of the defense is concentrated on the assault- 
ing tanks, the greater will be the losses to the accompanying 



/^ 



^ 



106 DEVELOPMENT OF TACTICS— WORLD WAR 

inf antry» and the more will that infantry lose the desire to 
accompany such attacks. We cannot deny that the tanks, 
at first underestimated, have proven themselves to be a 
very effective means of attack. The snocess mmt mit be 
accredited to the new tactics proper, but only to the change 
in the conduct of the attack, and thus must be credited to 
surprise. But if the defender is once prepared for this 
change in tactics and if he has abundant defensive means, 
the hopes of success diminish. 

Before the end of the year, November 30th, the German 
position, lost on the 20th, was recaptured by a brilliant 
counter-attack on a large scale, based on surprise, against 
the enemy's flank and rear. It was a good example of the 
importance of surprise as the enemy did not expect any 
attack. We succeeded in guarding the secret, so that the 
enemy did not observe the artillery and minenwerf ers going 
into position. In one division sector, 38 batteries had to be 
placed in position in 7 days; and in 3 days, 60 heavy and 
medium minenwerfers. For the first time, accompanying 
batteries were assigned to the infantry, and to each of these 
batteries 1 platoon of pioneers and 1 automatic rifle was at- 
tached. The success justified this measure. After an ar- 
tillery preparation lasting only one hour, the infantry broke 
into the hostile position with ease. The initial penetration 
was made without loss. The second wave and the attached 
accompanying guns finished the British machine guns. It 
was found of great use to signal the progress of the infan- 
try across each line by light signals of different kind and 
color agreed upon in advance. All participants emphasized 
the importance of keeping the advance moving or avoiding 
any halt under any circumstances. It was believed that 
rapidity of advance was of more importance than formation, 
which latter would soon be lost in the maze of trenches. It 
was shown again, that the advance in a number of waves 
corresponding to the lines of resistance was advantageous 
and that the leading and strongest wave should be assigned 
the most distant objective. Differing from this, the British 
have their single waves pass through each other, which takes 
more time. Skirmish lines have not always been found 
best, but a wave consisting of shock squads was proven the 



POSITION WARFABE IN THE WEST, 1914-1917 107 

best formation in workin^r through a position system. Shock 
squads were inserted against points where they could, pre- 
sumably, make the best progress. In any case this exper- 
ience brought forward the idea of forming flexible battle 
lines. Fighting in the intermediate terrain was speciaUy 
bloody. Fighting machine gun nests consumed very much 
time, as co-operation of the different auxiliary arms had to 
be arranged first. This took much time and thus the in- 
fantry lost the help of the creeping barrage, which kept 
moving on. A suggestion was made that assault groups be 
organized, for such tasks in each battalion and regiment, 
consisting of infantry, light minenwerf ers and machine 
guns accompanied by artillery which, for thorough co-op- 
eration, were to be trained in advance for the capture of 
supporting points and 'machine gun nests. The light minen- 
werf ers were indispensable in the assault preparation; but 
were at that time still too heavy to accompany the infantry 
attack. Here a change had to be made. The depth of the 
penetration did not suffice. Most of the divisions penetrated 
not more than four kilometers into the British position sys- 
tem. This was mainly due to the fact that the reserves 
were not close enough. ''The deep penetration," reports a 
division that had participated in the attack, ''would not 
have succeeded, had not the artillery followed the infantry 
directly. In the afternoon of the first day and especially 
on the second day, counter-attacks by tank battalions would 
not have been defeated with such success had not the artil- 

lery been immediately in rear of the infantry Two 

cavalry charges were defeated, in which a portion of the 
batteries had to face about. Unfortunately the loss in offi- 
cers in the division in the first line were specially high, 
which influenced the conduct of battle." But in any event 
the command had gained the conviction that a penetration 
on the West front was possible, especially against the Brit- 
ish. Thus, after a long period, a brilliant attack had suc- 
ceeded, an excellent omen for the future! 



IV. The War in the East and in Italy 

RUSSIA 

The Russian army entered the thoroughly popular 
World War excellently equipped. The Regulations pub- 
lished in 1912 were complete in all respects; they prescribed 
the attack in accordance with Russian war intentionsi but 
had paid little attention to the peculiarities of the Russian 
people ; th^ had not yet become the very life of the troops 
when war opened with the Central Powers. The leaders 
had their men firmly in hand. The higher officers were 
the weakest part, they were too much inclined to wait and 
be guided by events as they arose. The high command 
sank into inactivily when its own measures were met by 
the firm will of its opponent; in attack it endeavored to 
reach its objective by brute force and an inconsiderate sacri- 
fice of men. The general staff was theoretically weU 
trained and worked better in the execution of this war than 
in prior ones. There was more seriousness, more critical 
supervision, more self-restraint in the staffs, more under- 
standing of minor tactical work on the part of subordinates 
and troops, than ever before. The infantry — ^the Siberian 
troops were especially good — ^was brave in attack and tena- 
cious in defense, but was handicapped by insufficient fire 
training (it was inclined to long range fire), poor march- 
ing ability, and also by a mass-training which had done 
away with all independence. The conduct of rear guard 
fighting was brilliant, the troops were skillful in gaining 
positions and in constructing field fortifications. The cav- 
alry was very aggressive ; it performed excellent service in 
reconnltissance and in dismounted fighting; and sought op- 
portunities to attack the hostile infantry. Though the ar- 
tillery preferred firing at long ranges it had made great 
progress in technical firing and training, in comparison with 
the other arms since the Russo-Japanese war, and it formed 
the backbone of the army in the first months of the war. 

Taken as a whole, the Russian was superior to our al- 
lies, and was also a menace by reason of his numerical 

108 



WAR IN THE EAST AND IN ITALY IM 

fltrengtfa. '^e cm flght down fhe Russian onljr if we en* 
ergeticaltjr utilize all the advantages caused by our better 
annament, our excellent human material and our iron die* 
dpline, and if we train each individual to a sense of abso- 
lute superiority over his opponent; inoreasing his military 
knowledge and aptitude, and steeling his Arm and unshak- 
able will to gain victory/' 

During the course of the campaign the experiences in 
actual war were found advantageous in the training and 
the conduct of the Russian troops, but the great losses in 
men and material had a retarding influence on their battle 
activity. 

In a handbook compiled by Captain Ignatowitsch of the 
General Staff, based on experiences gained in war, it was 
recommended that the attack, adhere as long as practicable 
to close formation, from which deployment in depth and 
breadth, or in half companies could be made, and to take up 
the battle foitnation only at 1500 paces (1066 meters). 
The advance was then to be made in long and broad rushes, 
from firing position to firing position, under fire protec- 
tion of neighboring detachments, and without firing from 
intermediate positions. Intrenching during attacks was no 
longer mentioned. The closer the troops came to the ene- 
my, the shorter became the rushes. Charging distance was 
to be 50 to 100 paces (85 to 70 meters). Entry of the posi- 
tion was to be facilitated by grenade throwers. Very ef- 
fective was a simultaneous charge against flank and rear, 
even if conducted by only a few men. In night attacks the 
rifle should also be used; and the advance made in close 
order formation with increased intervals between ranks. 
Prior to contact hand grenades were to be used. Should 
the troops encounter an undestroyed obstacle, they were to 
lie down at 100 to 200 paces in front of it, fire rapidly for 8 
to 5 minutes, and allow two wire cutting squads to cut two 
roads through for each company. It was not stated if the 
charge was to be made with or without shouting. 

In the construction of positions, narrow and deep 
trenches similar to our reinforced ones, which in most cases 
were to be covered with shrapnel proof roofs, were re- 
commended; defensive fire through loopholes. Shelter for 



110 DEVELOPMENT OP TACTICS-rWOBLD WAR 

about 10 men each were to be constructed in rear at a dis- 
tance not more than one-half of the distance between the 
leading trenches and the enemy. Defense by rifle fire with 
regulation sight elevation (600 paces — 420 meters) and by 
advancing to the counter-attack up to 50 paces (80 meters) 
distance was advocated. The counter-attack was not re- 
commended if tiie enemy had already made a penetration. 
At night the troops protected themselves by cossack posts, 
which were advanced 200 to 400 paces, each company of 
the leading line securing itself by 2 double posts sent out 
80 to 70 paces, and for 2 companies 1 field picket also to 
be sent out. ()ne part of the defensive force r^nained in 
reserve; the strength of this post depended on the prob- 
ability of attack and the discipline of the troops and might 
amount to two-thirds of the force. The rest of the men 
were to sleep; distribution of the men should provide that 
the wakened and sleeping men were equally divided behind 
the works. Counter-attack was recommended for the de- 
fense. 'The morale of the troops is more valuable than 
numerical superiority.'' 

The high losses in men forced the adoption of position 
warfare in Russia as well as on the west front. In the 
West, excellent communication arrangements, and a highly 
developed industry had been able to supply half the world 
with war implements of all sorts in almost unlimited quan- 
tities, while a highly trained people furnished the men to 
work the machines. In Russia financial difficulties, a limi- 
ted number of ice-free harbors and greater distances pre- 
vented the utilization t)f these auxiliary means. Thus, in 
Russia machines had to be replaced by men, which we en- 
deavored to save in the West. The Central Powers could 
not compete with the Russians in numbers of troops, any 
more than the Western Powers could compete in the matter 
of material and ammunition ; therefore, in the East, the di- 
visions scantily supplied with artillery, had to be assigned 
to relatively broad fronts (in the Winter of 1916-17 sectors 
of from 20 to 30 kilometers), m which two-thirds of the 
troops were placed in the front lines* The disadvantage 
of broad stationary barrage fire had to be taken into con- 
sideration, and less depth in formation was possible. 



WAR IN THE EAST AND IN ITALY 111 

In contrast with the West, a minor warfare resulted in 
which we learned that the Russian was our equal and fre- 
quently our superior. It was supported by a brilliantly 
carried out espionage system, but the German infantry 
knew how to get the best of that. After 1916 the Russians 
employed the "drum'' fire, but this did not reach the same 
development as on the Western front, so that only in excep- 
tional cases did the front trenches become a sheU-holed ter- 
rain. This naturally led to a stronger occupation of the 
front line trenches. The artillery fire also was not strong 
enough to destroy the wire entanglements, and the hostile 
charges went to pieces on it. Important places were cov- 
ered by strongly constructed supporting points in the front 
line, surrounded by obstacles. The reckless Russian high 
command sought to gain an entrance by a mass attack. It 
is true, in some instances that the enormous masses suc- 
ceeded in entering a position at some points, but they were 
ejected by the counter-attack of fresh troops. 

The Russian tactics were strongly dominated by Joffre's 
views, but the Russians were unable to obtain the enormous 
artillery effect used in the West, and which was a material 
factor in the success of the French method of attack. In 
the Spring of 1916 the difficult situation of the French 
troops in front of Verdun forced the Russian Allies to a 
partial offensive by the Ehvert Army Group, though the 
Russian Army had not yet completed its preparations. This 
probably also was the reason the Army Group of Brussilow 
did not simultaneously start the attack between the Polish 
and Rumanian frontier. 

In an order issued by General Ewert, Wilna and the 
Wilna-Dunaberg railroad were designated as the main ob- 
jective. Points of entry were to be gained by the weight of 
a crushing mass. Minute details concerning their tasks 
were prescribed for the troops and for headquarters. The 
assaulting troops had been made to believe that the artil- 
lery alone would perform all the work. At the same time 
it was thought necessary to portray to the troops the hor- 
rible tortures they would suffer if taken prisoners. ''After 
capturing the trenches of the enemy, the attackers must 
not allow themselves to be stopped. The penetration must 



112 DEVELOPMENT OF TACTICS— WORX^D WAR 

suffer no interruption . • The artillery must take all 
measures to avoid firing on its own troops. Herein lies 
the real strength of co-operati<m of infantry with artillery/' 
This was a peculiar confession of lack of tactical co-operation 
and conception of the principles of fighting of both arms. 
Peculiar for the reascm that paragraph 7 of the or- 
der almost demanded certain preparations for firing on its 
own troops : 'It must be drilled into the troops that the re- 
serves and the artillery will open fire on th^cn in case the 
attackers show any inclination to let themselves be cap- 
tured." Cossacks also were detailed to drive forward with 
knouts troops that vacillated in the charge. 

The Summer offensive undertaken by General Brus- 
silow was also without success; it started about the time 
the Somme battles began in the West, and had some hope 
of success in the beginning. Officers and men had been 
trained during a long rest. Shock troops had been formed 
from picked men, artillery and ammunition had been brought 
up in plenty; and thus nothing seemed forgotten that could 
achieve success. The orders issued by Brussilow for the 
battle are of special interest, as the Russians also, per- 
ceived the correctness of the view that attacks had to be 
made on a broad front to make it impossible for the enemy 
to shift his reserves. ''The simultaneous attack on all 
fronts must be conducted so that in each army, and in each 
corps a persevering attack is made against a definite sector 
of the hostile position. At a few points along tihie line 
the attack will be made, and for that purpose I shall con- 
centrate reserves from the other fronts of the armies.'' 
The attack was to be conducted in such breadth that the de- 
fender could not fire on the troops from the fianks. This 
would require a breadth of penetration of at least 10 kilo- 
meters and of not more than 30. In a normal 15 kilome- 
ters attacking front of an army two corps were to be em- 
ployed, each reinforced by one brigade, which then at- 
tacked with one-half of their force in four waves — skir- 
mish interval of 2 paces — awhile the other half formed the 
shock-reserve. For a division of 16 battalions the Rus- 
sians reckoned on an attacking extension of 2 kilometers. 
Brussilow was opposed to the extensive use of "drum'' fire 
and did not mention gassing. 



WAB IN THE EAST AND IN ITALY 118 

Co-operation of the artillery with the infantry was 
deemed a prereqniaite of any success. For this purpose the 
batteries under the orders of the commander of attacking 
infantry went within 2 kilometers, the heavy batteries with- 
in 8 to 4 kilometers of the enemy. Stress was laid on the 
opening of ''roads" through the wire obstacles — several 
roads required, so that no crowding would ensue-— on firing 
on machine guns and on flanking works. Hie infantry at- 
tack immediately followed the artillery preparation. On the 
the other hand, the infantry was not to demand endless 
firing on the sectors to be attacked. The infantry attack 
was to be made in not less than S to 4 waves, following each 
other at 160 to 200 paces distance, machine guns with the 
third and fourth wave. Squad columns were to be in rear 
of the wings for flank protection, reserves in rear of the 
shock troops for immediate repetition of unsuccessful at- 
tacks. Assault distance about 300 paces from the line to 
be attacked ; arrangement of the starting point as with the 
Joffre attack procedure— one wave for each hostile trench. 
Capture of the first position by one attack with reliable 
flank protection was demanded. The German counter- 
attack was to be defeated by fire; if that succeeded then a 
trial was to be made immediately to take the next positions 
by foUowing attacks. ''After successful penetrations, a 
promising field of activity opens to the cavalry, supported 
by its riflemen and its artillery; I order that our numer- 
ous cavalry be made as much use of as possible.'^ 

Emphasis was laid on the fact that by means of con- 
tinuous insertion of strong infantry forces the entrance 
could be forced even if the first troops had no success. This 
led to mass attacks, in which the troops — ^being cautioned 
that in case th^ stopped they would be fired on by Russian 
machine guns and field guns — ^were driven forward even un- 
der the knout. In consequoice of this lack of considera- 
tion and the tactical inefficiency of the subordinate com- 
manders the charges were executed in even denser and more 
vulnerable formation than had been planned by the leaders. 
Lying down during pauses was even forbidden. The losses 
were enormously increased. In working up to the assault 
position, the infantry showed itself very adept and seldom 



114 DEVELOPMENT OF TACTICS— WORLD WAR 

more than two or three men could be seen ; but the exten- 
sive earthworks and the systematic ''road-firing'' of the ar* 
tiUery swept away all doubts concerning the points to be at- 
tacked* When the attack did not succeed in gaimag a fooU 
hold at assault distances, attacks were made at longer dis- 
tances, 300 to 1100 meters, on the run with interspersed 
walking periods. Large losses resulted in crowding through 
the ''roads'' the artillery had opened in the wire entangle- 
ments, and which could not be made sufficiently wide. What- 
ever was lacking in artillery preparation (for time for pre- 
paration was between 2 and 5 hours, and 10 to 16 rounds were 
reckoned for the running meter of the position) was at- 
tained by the mass insertion of troops. The Grerman ar- 
tillery positions were bombarded only when their location 
had been accurately ascertained. The artillery often de- 
ployed in 2 and 3 lines in rear of each other and plentiful 
use was made of false works and developatnent of smoke 
clouds. Special weight was attached to flanking fire, there 
was no hesitation in bringing batteries forward into posi- 
tions where they were not at all expected (for instance in 
the sand dunes in the swampy terrain) and which had been 
considered impassable even for the infantry. Single field 
pieces were brought into the most advanced lines to fight 
the machine guns. The Russian attached little importance 
to surprise ; it expected to be victorious by force of numbers ; 
terrain obstacles had no terror for Russian troops in attack ; 
the Russian infantry even attacked going through swamp 
and water up to the neck. The activity of the artillery was 
excellent; however, lack of matmel and ammunition pre- 
vented it from displaying its power to the fullest extent. 

In 1917 the (jerman high conmiand calculated on a con- 
tinuation of the Russian offensive taking advantage of the 
experiences in the West. In May the signs became more 
numerous that an offensive was about to start against the 
South Army, with minor attacks in the direction of Sloczow 
.'^nd Stanislau, and then against Krewo-Smorgen and Duna- 
burg. Everything pointed to the fact that this time the 
Russian offensive would assume Hie character of a "materiel 
battle," which was the more required as the disinclination 
of the Russian troops for a repetition of Brussilow's bloody 



WAR IN THE EAST AND IN ITALY 116 

attacks had resulted in open mutiny. After the strongest 
artillery preparation, which swept obstacles away, which 
changed tibe front battle positions into sheU-holes, the at- 
tack commenced on June 30, 1917, against the South Army 
and against the Sloczow sector; on July 6th also against 
the Austrian 3d Army at Stanislau. This latter attack, 
originally considered a minor one, had an unhoped-for suc- 
cess (loss of the heights on the Bystrzyca and the Lomnica 
position in rear and capture of Halusch), which was skill- 
fully utilized by the Russian high command in bringing up 
fresh forces. At Koniuchy the position of the Austrian 9th 
C!orps was also entered to an extent of an area 20 kilometers 
broad and 4.8 kilometers deep. In expectation that forces 
had been taken from the fronts that had not yet been at- 
tacked and brought into Galicia, an attack was made on 
July 21st at Krewo-Smorgon, and on the 23d at Dunaburg. 
The Russian high command prepared the first part of the 
attack most excellently. All plans had been prepared to 
insure a successful attack, if only the tools had corres- 
ponded to the demands of the commander. One main cause 
for its non-success was the inefficiency of the subordinate 
leaders, whose power of decision and independence did not 
meet expectations. The attack went to pieces on the firm 
stand of the German infantry, which was supported by an 
inferior artillery; the Russian batteries knew how to pre- 
pare the attack but not how to support it. Lack of am- 
munition compelled firing to cease altogether at night and 
the fire did not reach as far into the terrain in rear as it 
did in the West. Differing from the attack in 1916 the 
artillery also was fired on, but without effect due to lack 
of artillery airplanes. The scheduled gassing had no suc- 
cess. Tanks, inserted in insufiScient numbers, had no efifect. 
The Russian infantry attacked, after heavy "drum" fire, 
in dense skirmish lines by divisions with their brigades 
alongside each other, and at an extension of 1.5 kilometers. 
In the brigade the regiments were formed in rear of each 
other, 3 battalions with 550 men each in first line, battal- 
ions with 4 companies divided into half-companies and 
formed in depth with 50 paces distance. The newly or- 
ganized regimental pioneer companies were distributed to 



116 DEVELOPMENT OF TACTICS— WORLD WAR 

the battalions. The 4th Battalion followed at 100 paeeB» 
tiira followed at the same distance, the 2d Regiment of the 
brigade. In the assault the officers were in front. Ac- 
companying cannon and trench guns that had kept silent 
up to the assault fired over the heads of the infantry. The 
attack was to be carried within the German artillery posi- 
tion. 

The German high command had early pointed to a pos- 
sibility of a ^^materieF' battle, so that as a matter of course 
the same changes were made here that were made in the 
West. In place of a defense by a densely garrisoned front 
line, the infantry was organized into weaker garrisons of 
the front trenches with stronger troops in readiness in rear 
and reserve, which carried on the battle offensively around 
the front trenches. The counter-attacks immediately 
started were successful ; more power was gained by the de- 
fence through strongly fortified but weakly occupied sup- 
porting points in the intermediate terrain. 

The battle in Eastern Galicia found a brilliant finish 
in the transition from the defensive to the offensive, which 
demanded great exertions, especially from the troops of the 
South Army, which joined the advance of the Group 
''Sloczow'' without preparation. 

The penetration executed with 10 divisions on a fron- 
tal extension of 89 kilometers, one flank on the Sereth, was 
a complete success and the Russian lines far into Rumania 
were terribly shaken, and the Russian offensive was forced 
to stop. The German attack proved a far better means 
of relieving the front that had been attacked, than direct 
support by the Austrian Sd Army could have been. Such 
a counter offensive is at all times the best means to defeat 
a penetration attempted at another point, which, starting a 
short time after the commencement of the infantry attack on 
a neighboring front, will presumably strike troops weaker 
and probably already worn out. The German troops proved 
themselves equal to the demands of battle even if the 
lack of practice in marching, the difiiculties of transporta- 
tion, and insufficient supplies for headquarters and troops 
for mobile warfare weighed heavily against theuL The 
troops going from position to mobile warfare had to get 



WAR IK THE BAST AND IN ITALY 117 

used not only to marching, but to other battle conditions ; 
th^ had to releam to attadc, to advance again wiHidut hav* 
ing close touch with comrades, and to conduct attacks 
without the support of large artillery masses. The insufll* 
cient training of the subordinate commanders, who in most 
instances knew mobile warfare only theoretically, was sore- 
ly felt. The experience gained at this time was especially 
valuable to us in the subsequent battles on the Western 
theater of war. 

A brilliantly prepared penetration of the bridgehead at 
Jakobstadt took place on September 21, 1917. In the early 
morning hours of September 1st a crossing of the Duna was 
accomplished. The troops then turned iilpng the roads 
leading from Riga northeastward, which caused the imme^ 
diate evacuation of the bridgehead on the left bank of the 
river. The Russian high command firmly believed in an 
impending attack against the soutiieast front of the bridge- 
head and did not believe the reports coming in concerning 
preparations for a crossing above Riga. This is a good 
example of the value of so-called ^'reliable reports.*' 

In all battles the Russian infantry showed an enviable 
skill in overcoming difficult terrain, considered by us in 
most cases as impassable. 

BATTLES IN NORTHERN ITALY 

May 23, 1915, the Italian Army joined our enemies, 
after it had adopted as early as 1911, contrary to its prior 
custom, a gun of French manufacture (Deport, same cali- 
ber as the French field gun) and modeled its Regulations 
for the Higher Command and Staff on the French Regula- 
tions. In Austria they had well learned the method of 
battle peculiar to the Italians. ^ 

The Italian commander-in-chief. General Cadoma, had 
net^ected at the opening of the war to bring the superiority 
of his army into immediate account by an offensive against 
tiie weak Austrian frontier guard troops ; and as the latter 
could not count for the present on reinforcements, a posi- 
tion warfare resulted on the Isonzo that shows the same 
characteristics as the one in France. Not until June 29, 
1915, was commenced the first 'Isonso Battle,'^ which was 



118 DEVELOPMENT OP TACTICS— WORLD WAR 

without result; and the subsequent battles (the tenth 
fought from Ifay 6, to June 7, 1917, after Joffre's methods) , 
showed only a minor gain in territory, in spite of strong 
artiUery preparation and mass insertion of Italian infantry. 
Based on the experiences of the tenth Isonzo battle, Gen- 
eral Capello, commander-in-chief of the Second Army, is- 
sued a general ''Maxims for the Employment of Infantry 
and Artillery in Attack/' 

The infantry was not to carry the special training too 
far; attack was to be executed in waves with great depth 
(double-barreled Fiat-Revelli pistols — corresponding to the 
light minenwerf ers — ^in the first wave, machine guns in the 
succeeding waves) and up to the final attack objectives. 
The importance of machine guns for flanking fire was pointed 
out; the troops were to endeavor to avoid the trench war- 
fare proper, which consumed too much time. Similar to 
the methods in Eastern Galicia, on October 24, 1917 — aft^r 
the 11th Isonzo battle — ^the counter-attack of the 14th Army 
(Otto von Below) with the adjoining Austrian 1st Isonzo 
Army was started against superior Italian forces. The at- 
tack, oblique to the Italian front, conducted from about the 
line Flitch-Tolmein in the general direction of Cividale 
turned into a penetration. Through Hiis penetration the 
adjoining south front was drawn into the defeat of the 
Italian Second Army. The direction of the attack across 
the Bainsizza plateau and by Flitch-Tolmein had become 
known to the Italians as well as the time the attack was to 
be made. The unfavorable condition of our forces did not 
permit us at this time to utilize the favorable situation in 
Southern Tyrol. 

On September 8, 1917, the first orders for the offensive 
were issued. The attacking troops of the 14th Army had 
been assembled in the basin of ICrainburg and north of the 
Karwanken. There the preparations for the war in the 
mountains were completed. The 'Tolmein Group" had to 
be concentrated in the narrow spaces of the basin near Flitch 
and Tolmein. Along two narrow mountain passes, 60 kilo- 
meters distance, with heavy grades, several divisions had 
to be brought forward behind each other (as weU as half 
the artillery and minenwerfer groups) in the face of dom- 



WAR IN THE EAST AND IN ITALY 119 

inating hostile positions allowing an extensive view. This 
was accomplished in such manner, that the Groups Flitch 
and Tolmein stood in readiness with 5 divisions in the first 
Hat^ and 3 divisioi^ in eeek of the second and thkd lilies^ 
In spite of the unfavorable weather the march proceeded 
without friction. The concentration of the artillery and 
minenwerfers took from September 27th to October 17th. 

The experiences of the Carpathian mountains, of the 
Vosges, and Alpine battles, pointed out the road to the 
offensive tactics; rapid advance in uninterrupted attack 
across the connecting land bridges from height to height, 
utilization of the valley roads to bring up the artillery re- 
serves and supplies; forcing dominating ridge positions 
by enveloping attack in flank and rear, and by marching 
around them ; and mutual support between the frontal and 
enveloping groups. 

With this in mind, the following operations were start- 
ed with a view to the simultaneous penetration at several 
points : The Group Krauss as right wing of the 14th Army 
by Saga towards the Stol ridge ; a strong left column was to 
open, after the capture of the Vrsik, the basin of Karf reit 
by taking direction on Ravna — ^Karf reit and roll up the Stol 
ridge from the East. The Groups Stein was started against 
Mt. Matajur. The capture of this area opened the Karfreit 
basin from the Southeast and thus supported the simul- 
taneous operation of the Group Krauss having the same 
mission. The Group Berrer was started against the Hu- 
mand for simultaneous advance via Drenchia against the 
San Martino in conjunction with the Group Stein. The 
Group Scotti received the task of assisting the Isonzo Army 
in its crossing of the Isonzo by attacking with the objec- 
tive Globocak — ^Kostanjewitza. By the middle of October 
the offensive preparations had make such progress that the 
general attack could be ordered to start the third week in 
October. 

The plan had been worked out down to the minutest 
details. After a gas bombardment from 2 to 6 A.M., gen- 
eral fire was to be opened at daybreak. After a short but 
steadily increasing bombardment the entry was to be made 
into the positions along the entire Flitch-Selo front. How- 



120 DEVELOPMENT OF TACTIGS-WOBLD WAB 

ever, unfavorable weather delayed the attack till the morn- 
ing of October 24» 1917. This day was foggy» which waa 
favorable to the advancing troops. 

The Italian 2d Army under Lieutenant General Capello 
awaited the shock of the Allies in strong, apparently invin- 
cible mountain positions in the Julian Alps. The prepara- 
tions for the attack, the filtering through of our divisions 
along the narrow, distant but visible valley roads» and the 
attack formations could not remain hidden. Inexplicable 
seems the small efforts put forth by the enemy during the 
final days prior to the attack. In the district from Rombon 
to the Km the Italian 4th Corps (50th, 4Sd and 45th In- 
fantry Divisions) blocked the valley road of Saga and the 
basin of Earf reit. Forces from the 17th Army Corps could 
keep the Tolmein basin under fire from the line Eolovrat 
Ridge to vicinity of the Auzza Canal. The narrow frontal 
sector of the plateau of Bainsizza — ^Heiligegekit was de- 
fended by the 26th, 60th and 30th Infantry Divisions (14th 
Army Corps) in strong positions. North of the Monte San 
Gabriele the line was prolonged by the 2d Corps (23d, 53d 
and 8th Infantry Divisions). At, and north of Grorz, stood 
the 4th Army Corps (68th, 24th and 48th Infantry Divi- 
sions)- as the south wing of the Second Army joining the 
Third Army under the Duke of Aosta. Not less than 56 
brigades and 4 Alpine groups, a total of possibly 350 bat- 
talions of the Second Army were to stop the attack. 

Cadoma was fully confident and reported : 'The oppo- 
nent, strongly supported by German troops and war muni- 
tions, has assembled material forces for an offensive on our 
front The hostile attack finds us firm and well prepared.'' 
At 2:00 A.M., October 24th, the gas cannonade opened 
against the hostile batteries. At daylight, an annihilating 
fire for effect was started by the entire artillery and minen- 
werf ers. Instructions for the forward displacement of ar- 
tillery for adjustment, and for fire for effect were conformed 
to and the roads were opened to the infantry for the at- 
tack. At 8:00 A.M. the infantry started the attack be- 
tween Tolmein and Flitch. By 2 :00 P.M. the Group Erauss 
had stormed the positions on the Rombon upon the right 
wing of the army. In an intrepid charge Austrian infantry 



WAR IN THE EAST AND IN ITALY 121 

advanced from the right wing of the Group Stein and took 
the positions on the line Em — ^western slope of the MrzlL 
In the meantime the German Division Lequis pushed for- 
ward along the valley road of Tolmein. It is true that the 
enemy held on the left and right the dominating hiU posi- 
tions. But fog banks prevented a view to the far horizon 
and into the valley. The Italians did not suppose that deep 
below German infantry had pushed its way through to E^- 
freit, Oiat already El Kammo, and shortly after 2 P.M., 
Ideersko on the Isonzo was reached. The left wing of the 
Group Stein, after overruning the front vall^ position 
southwest of Tolmein, attacked in conjunction with the 
Group Berrer, the hostile main position on the Kolovrat 
ridge. Attacking infantry, under hostile fire, crept up the 
steep slopes from the vidley for more than 1000 meters. 
The crest of the Kolovrat formed the key point of the entire 
system in the strongly fortified position on Hill 1114. In 
the afternoon the attack against Hill 1114 and the neighbor- 
ing massive Hafnik could be started. Farther south the 
position on the Jeza block went to pieces under the attack 
of the Group Berrer. Without stopping, the south wing of 
the army (Group Scotti) pushed its way across the Hrad 
Vrh against the Blobocak chain. 

By the evening of October 24th, the Flitch basin was 
opened to immediately east of Saga. The southern slopes 
of the Km had been stormed. Karf reit in the Isonzo valley 
and the ridges west and southwest of Tolmein had been 
taken. No addition to or change in the army orders was 
piade. The attack proceeded. 

The gas bombardment shortly before the attack un- 
doubtedly played a great role in its success, as the Italian 
gas masks did not offer any protection against it. The 
infantry, though well trained for fighting in the higher 
mountains, up to this time was acquainted with only the 
lowland and the intermediate mountains, yet performed its 
very best. The superiority of troops equipped for and 
trained in high mountain warfare, was plainly discernible. 
The Italian infantry could not cope with such troops. Up 
to the evening of the 26th the Furttemberg Mountain Bat- 
captured over 100 ofilcenr, among them 2 regimental 



122 DEVELOPMENT OF TACTICS-WORLD WAR 

commanders, 4600 men, 60 machine sruns, 20 guns, and 25 
caissons. Its own losses were but 7 killed and 28 wounded. 
The skill of our riflemen in moving in the mountains, and 
the skill of the subordinate commanders saved the battalion 
from heavier losses. Another thing had its influence ; the 
Italian infantry had no experience in defense; till then it 
had always attacked ; its front trenches were densely occu- 
pied; active defense of the intermediate terrain had not 
been prepared in advance. The Italians had correctly per- 
ceived that in the mountains, possession of a certain ridge 
line depends on the holding of a few dominating points. 
These points, against which the fire of the attacker was 
principally directed, were occupied by the Italians entirely 
too strongly, when they should have attempted to hold them 
from the side. The attacker's fire had an annihilating 
effect. No plans had been made for using the reserves, their 
place should have been immediately in rear of the dominat- 
ing points which the attacker was forced to take ; the roads 
the attacker had to take were also well defined, and a thor- 
ough study of the terrain should have furnished exact 
knowledge where to employ the reserves. The Italian ar- 
tillery occupied positions from which it could have well sup- 
ported an attack, but from which it could hardly carry on 
the defense ; positions for defense should have been selected 
which commanded the ascent on both slopes of a mountain 
ridge. Flanking fire is most effective and can be employed 
almost everywhere ; frontal defense had best be left to the 
minenwerfers. 

The penetration was immediately followed by a bril- 
liant pursuit by the German troops to across the Taglia- 
mento ; the fleeing masses and their army impedimenta were 
congested at the crossings of Codroipo and Latisana, and 
only the delay of the slowly following Isonzo army prevented 
a greater capture of prisoners and supplies. An attempt 
to take possession of the Tagliamento crossing at Dignano 
by infantry carried on auto-trucks miscarried. 

MOUNTAIN WARFARE 

No one could have foreseen in time of peace that Ger- 
man troops would ever be called on for mountain warfare ; 
we shared the opinion of our neighbors that it required 



WAR IN THE EAST AND IN ITALY 128 

special training for mountain warfare, that recruits, natives 
of the lowlands, would take a long time to get accustomed 
to mountains. Only well trained mountain troops who in 
their sports do not forget the objects of war, are suited for 
this special work. We will point out that fostering snow- 
sports and ''Alpinism'' is the best foundation for carrying 
on war in mountains, as was the case in the Alps, in the 
Carpathians and in the ice and snow covered lower moun- 
tain warfare, but troops that had no such training met the 
requirements after some time. The German troops received 
aid from their Allies, who had gathered rich experiences 
concerning equipment, march capacities, and methods of 
fighting. In the ignorance of high mountains, every inex- 
perience must be paid for bitterly ; the first adverse effects 
win be noticed on the pack animals, and through them the 
supply of ammunition and subsistence will suffer. Tacti- 
cal measures must conform to the peculiarities of the ter- 
rain, high mountains (Alps) clear of forests, demand dif- 
ferent measures than snow or forest covered lower moun- 
tains (Vosges, Carpathians). Terrains with and without 
glaciers, summer and winter conditions, continually con- 
front the leader with changing situations. During the win- 
ter battles in the mountains demand the greatest possi- 
ble exertions. Losses by frost bite and death by freez- 
ing can assume the same proportion as the hostile effect 
of arms. ''The main thing for troops in the mountains 
is mobility and endurance. Troops will then have the best 
success if they overcome more and greater difficulties than 
the enemy believed possible, and when they know how to 
work their way through apparently impassable terrain." 
(Training Regulations for the Austrian Army, 1911.) The 
accomplishment of these demands presupposes special train- 
ing. 

The battle of man against nature becomes more pro- 
nounced in mountain warfare. Troops must become accus- 
tomed to the impressions and dangers of fog, of the break- 
ing off of projections, of snowslides. Fortunately service 
in the mountains also offers moments that increase the 
capacity of the troops and have a favorable effect on their 
spirit and morale. First comes the danger, which undoubt- 
edly has a high educational value. The troops find them- 



124 DEVBLOPMEMT OP TACTIGS— WORLD WAS 

•ehrM not only in diiBeolt» bat also in dangerous sitoationa, 
in wUch it becomes dear to every one that only strict obe- 
dience and onity of effort will avert the periL If that suc- 
ceeds, trust in the leader increases. The qiirit of sdf ^pre- 
servation drives many a weakling forward on the march 
who otherwise would have reached the limits of his endnr* 
aace. The troops have to learn to march and breathe dif- 
ferently, for the rules of the lower altitudes do not apply. 

As a general rule, single file will be the method of 
marching in mountains. This requires special training and 
an entirely different method of issuing orders. To pass 
orders throu^^ a ccdumn of files will frequently be entirely 
impossible, and even messengers will have difficulty in get- 
ting to the rear. The deep column in sini^ file, or in c<d- 
umn of twos, increases most surprisingly in length and is 
beyond any calculation. This is specially annojring whoi 
starting the march and when deploying stronger colimms 
for fighting. Keeping the same pace and cadence is very 
difficult especially when part are ascending and part de- 
scending. A battalion of 800 men will occupy a depth of 
L6 to 2 kilometers, which corresponds to about one hour of 
deployment. Depth formation makes division into single 
detadmients unnecessary, the sending ahead of one rein- 
forced security detachment wiU generaUy suffice. Columns 
of more than 1 battalion can be employed only if the troops 
are used to mountains, as otherwise they get too clumsy. 
The change from teams to pack Miimals, hard to handte, 
the equipment of the troops with snow spurs and . snow- 
shoes requires far reaching changes. Marching difficulties 
are especially fdt in shifting of troops, in counter-attacks, 
and in bringing up reinforcements; we can count on the 
troops only that are in, or immediately in rear of the fight- 
ing line. In comparison with earlier wars the developmrat 
of the use of high explosives, the service of communication 
{wirdess stations, visual signal means of all kinds) and 
methods of bringing up supplies has made the conduct of 
mountain warfare easier, but all our fights have merdy con- 
firmed the old rule that in mountain fighting the value of 
and not the number of troops is decisive. A decMye, active 
battle leadership can make for success in mountains even 
if the enemy is greatly superior in numbers. Flirting in 



< 



-WAB IN THE EAST AND IN ITALY 126 

iMuntains is especially diiBcult for large bodies of troops. 
Decisive battles are infrequent, more often the action will 
result in forcing deployment, or preventing deployment, out 
of the mountains. Placing light, mobile forces along the 
roads through passes gives opportunities for gathering the 
fruits of success in mountain fighting. In the mountains 
themselves the possession of the highest ridge is decisive; 
on account of the far view, the domination of the important 
Valley roads that have to be used to bring up supplies, and 
on account of difficulties of ascent for the attacking army. 
Considering the limited number of troops available, not 
ev^ry ridge can be occupied, but points on the flanks must 
be 4ftcttred from which tlie domination of the position and 
connections With the vall^ is possible. The defender bar- 
ricades the points he considers passable. Therefore moun* 
tain troops must understand how to work up through appar- 
ently impassable terrain, in order to strike the oiemy's 
flanks and rear^ Frequently the more appearance of minor 
detachments on heights, difficult of ascent, will produce a 
great morale effect, to which probably a good fire effect can 
be added^ since good observation is generally available. The 
Italians had probably considered this point well in the de- 
fense on October 24, 1917, but they had occupied the pcnnts 
too strongly on the possession of which d^i^nded posses- 
sion of the crest line, instead of defending them by flanking 
fire and fire on the two slopeB leading up to these points. 
Good mountain troops will always find a way out of serious 
^Hfficulties, while troops unused to same easily allow them- 
sdves to be taken prisoners when their retreat is endan- 
gered. As machine guns display a great fire power in a 
narrow space, they are an especially preferred arm in moun- 
tain warfare when but few troops can be employed. The 
arm proper in such warfare is the infantry. CSavaliy serves 
mainly for the purpose of message, connection, and securi- 
ty duties, along the pass roads. Emidoyment of airplanes 
is niade m<Mre difficult by the frequency of fogs, the diffi- 
culties of rapidly gidning military altitudes and of finding 
suitable landing places. Artillery and minenwerf ers inay 
be broui^t into position almost every whare^ of course it 
takes time and labor, but the fire efficiency suffers und^ the 



126 DEVELOPMENT OF TACTICS— WORLD WAR 

difficulties of influence of altitude and weather. Minenwer- 
fen are of especial use because of their curved trajectory. 

In defense, possession of heights gives freedom of ac- 
tion and possession of the valley ; therefore forward slopes 
of the valley are kept occupied as long as the troops have 
the reverse slope securely in hand. There is nothing more 
unfavorable than to have the troops surprised by fire dur- 
ing the ascent; therefore the advance is made by bounds 
from crest to crest. Heights dominating the line of march 
must be taken possession of as early as possible — ^frequently 
a day ahead if on the flanks and in front — ^by detachments 
sent ahead and largely supplied with mountain artillery. 
The organization ordered at the commencement of the 
daily operation remains unchanged for that day, as shift- 
ing consumes time. Division into several columns decreases 
in a desirable manner the march depth, which, however, 
the defender can take advantage of only under favorable 
conditions. In starting the attack flank effect and threaten- 
ing the line of retreat gains greatly in importance as the 
enemy in most cases can give ground only in a definite di- 
rection. The decisions in forces fighting separately will be 
materially independent of each other, and in the case of 
larger units will vary greatly in time ; frequently the pos- 
session of a portion of the position wiU bring about the final 
success. It is the task of the highest leader to perceive the 
key to the position and to utilize the partial success by the 
pressure of pursuit and by starting more troops against 
those hostile portions that have not yet been driven off. 
Caution is necessary in following up a retreating enemy; 
with systematic leadership he will attempt to find oppor- 
tunities to gain success by fire surprise and unexpected coun- 
ter-attacks. 

The defender fights under more unfavorable condi- 
tions than the attacker, who can select time and place of 
attack, and does not suffer under the fire of the defender in 
the same measure as he does in the lowlands, but is re- 
stricted in the employment of his reserves by the mountains. 
On the other hand, the defender aided by the difficulties the 
attack encounters in the terrain, can delay the attacker for 
a longer time with but weak forces. The construction of 



WAR IN THE EAST AND IN ITALY 127 

a position presents sreat difficulties, trenches can be con- 
structed in many cases only by application of explosives; 
losses through iSying stones (against injury from which 
the steel helmets offer good protection) are frequent. In 
the foothills and intermediate mountains the conditions are 
not so bad as in the higher mountains. One of the inher- 
ent qualities of mountain warfare is the favorable oppor- 
tunity which the defender has in starting a counter-attack 
when the attacker is exhausted by a hard climb and ap- 
proaches the crest in disorder. It is only necessary that the 
reserves be present at the decisive point, which can fre- 
quently be ascertained from a study of the map. On Octo- 
ber 24, 1917, in the Tolmein penetration, the Italians lacked 
these very reserves at the decisive point. The decision lies 
in this attack brought by the reserves, which if made by 
fresh troops — ^if possible down hill and against the flanks — 
has excellent chances for success against an attacker who 
is exhausted by the ascent. Even when the position is lost 
a stubborn defense, based on rocks and ravines may still 
promise success. Only troops, well trained and fresh, can 
achieve permanent success in mountain defense, if they 
endeavor to weakly occupy long fronts (but with plenty of 
machine guns), keeping strong reserves available and re- 
maining prepared for aggressive action. 

Superior force plays a smaller role than in the rolling 
country. All measures are good which prevent the attack 
of positions from higher points, from the flank or by envel- 
oping, supported by superior artillery. Making the ascent 
more difficult to the enemy is always advantageous; every 
opportunity to take the long columns of the enemy under 
a cross fire must be utilized. This method leads to forma- 
tion in squads and platoons in the infantry and its auxiliary 
arms. Utilization of advanced positions on main lines of 
resistance on the rear edge of a crest with a limited field 
of fire will offer themselves. We must caution against the 
danger of covering everjrthing; the defender must block 
the hostile routes of approach by advanced detachments, 
must delay the attacker and seek the decision by means of 
reserves kept concentrated in rear. The terrain difficulties 
must positively invite the attack; at no place are cordon 
positions more dangerous than in mountains. 



! 



128 DEVELOPMENT OF TACTIC&-WORLD WAR 

The Austrian defense of the Southern Tsrrol in 1866 
always remain an excellent pattern. 

As early as the winter of 1914-16 mountain formations 
were adopted for employment in the Vosges, after the 
French had forced us to adopting snowshoes in that region. 
The Bavarian Greneral v. Hoehn demanded that the following 
be equipped with snowshoes: 

^1. Infantry in strenirth from patrols to company; suited for 
rooonnaisaanoe, seciirity and figfatinff, mesaengera, signal and tale- 
phone troops, litter beurera, and machine gun platoons. 

**2. Artillery headquarters, artillery offlcePs iMitrols for target 
reconnaissance, fire control and observation for, artillery. 

''8. Commanderst general staff officers, personnel of tlie staff, 
telephone squads.^ 

Snowshoe runners can be employed also for tramping 
down roads for detachments in close order. 

The very useful drawing of two to three snowshoe runr 
ners by troopers was not adopted; but the transportation 
of machine guns on snowshoe sleds proved advantageous. 

In the Masurian battles snowshoe runners were em- 
ployed, but thawing weather setting in, their usefulness 
soon ceased. Snowshoe runners of the German 2d Snowshoe 
Battalion were used by the German South Army for minor 
operations successfully, for instance, to gain flanking fire 



V. Technique In War 

• 

Utilization of all the peace time technique for pur- 
poses of war conforms to the nature of a people's army. 
Only with the start of the position warfare could the pro- 
per basis be secured which this technique required for the 
full development of its capacity. In time of peace, this is 
easy but it is very difficult in mobile warfare. Thus, the 
technique will either accomplish something brilliant, or its 
tools and implements will become mere ballast. Nothing 
would be worse than to place our reliance principally on 
technical means. The moral forces in the breast of the 
commander and in the soul of the entire people are the 
qualities which have finally turned the scales in war. ''Only 
the psychically qualified may chain chance to fortune^' 
(Goethe) or, as Moltke expressed it ''Only the efficient has 
fortune permanently.'' Possibly even more than at the 
opening of the 19th Century, the words written by Scham- 
horst in April, 1806, have importance: "We have started to 
appraise the art of war higher than military virtues. This 
has caused the collapse of peopled in all times. Never do 
the moral qualities rest, they fall, as soon as they cease to 
strive for improvement." 

In a more confined sense the military technique had 
been prepared by peace establishments and was gradually 
increased to the enormous extent demanded by the World 
War. The utilization of our highly developed industries and 
chemistry by the military departments was difficult as there 
was a lack of a central governing authority. Recommenda- 
tion and control by some bureau was of decisive importance. 
In addition to the utilization of industrial establishments 
and raw materials captured in the enemy's country we will 
merely mention: Concrete construction, light and water 
supply, etc. In Germany the lack of raw materials was of 
special importance. It is possible that more could have 
been achieved, and more rapidly, by the use of digging ma- 
chines (especially in construction of cable trenches) and 
with drill machines in the construction of positions. Con- 

129 



130 DEVELOPMENT OF TACnca— WORLD WAR 

struction of positions finally demanded a closer study of 
'Var geology/' In what follows we will treat only of the 
war technique in a closer sense. 

The principal question was to replace the animal trains 
by mechanical trains through adoption of auto trucks for 
more rapid transportation of men, supplies and materieL 
In Italy, in France, and finally also with us auto trucks were 
used on a large scale for shifting reserves, and with suc^ 
cess. The disadvantages of disrupting units, the separation 
of the dismounted men from their horses and vehicles could 
not be avoided. A further link in the chain was the intro- 
duction of ''tanks'' for traveling across country. The pos- 
sibility of demolishing obstacles by their means, was a minor 
result. They were in the first line an attack weapon, under- 
estimated by us at the start, the more so as they, appearing 
in small number, could be quickly put out of commission; 
toward the end of the war, however, our defensive means 
were insufikient to meet the unexpected mass employment. 
Our defensive measures could not keep pace with the in- 
crease of the fighting means, though our industries achieved 
the almost impossible. ''The best arms against tanks are 
the nerves, discipline and intrepidity. Only with the de- 
cline of discipline and weakening of the fighting i)ower of 
our infantry did the tanks in their mass employment and 
in conjunction with artificial fog gain a dangerous influ- 
ence on the course of military events." (Ludendorif .) 

The French had heavy tanks (40 to 60-ton) under con- 
struction, possessed medium (24 to 80-ton) and light tanks 
(model Renault, 6.5 ton). Medium tanks were organiased 
by fours, in groups (3 batteries) and sur-groups (3 groups 
each). Small tanks were formed into platoons (5 tanks), 
companies (15 tanks) and battalions (45 tanks). Marching 
on roads was to be avoided because of the tracks left and 
the damage done to the roads. 

For attack each army corps received 1 "tank regiment," 
consisting of several battalions light tanks and sur-groups 
of heavy tanks. The light tanks were allotted to the divi- 
sions participating in the attack, in general, 1 tank bat- 
talion (46 battle tanks) to the division. The heavy tanks, 
on tli^e other hand, were a unit attached to that attack sec- 



TECHNIQUE IN WAR 181 

tor in which the greatest resistance was expected. In or- 
der to keep close connection between infantry and tanks, 
the tank units were placed under direct orders of the com- 
manders of the attacking troops down to the battalion com- 
mander. 

The medium tanks had to be directly^ accompanied by 
special troops in the fight, who helped them across espe- 
cially difficult obstacles. One infantry company was required 
for this for each group of medium tanks (12 tanks). That 
company, which was usually taken from some unit not 
designated to participate in the battle, received special train- 
ing for 14 days. It usually was divided into 2 detachments ; 
1 advanced with th6 first infantry wave and executed the 
heaviest repair work (ditch and trench crossings, alleys 
through entanglements, foot bridges). The other detach- 
ment furnished 3 selected men for each tank who followed 
the tanks directly, furnished connections between the men 
in the tank and the exterior world and served as supports 
in possible accidents. Removal of obstacles which can delay 
the march must be finished at the latest by evening of the 
day before the battle. The most dangerous enemy of the 
tank is the artillery, which must be destroyed, or engaged 
by airplanes. Tanks quickly give out, so depth formation 
is necessary. 

In attack the heavy tanks preceded the infantry and 
the light tanks, to open the road for the infantry and to 
cover it during the cleaning up of the hostile position. The 
light tanks advanced with the infantry. The light tank 
platoon as a rule fought in close order against definite tar- 
gets. As soon as success was attained, the platoon was 
again at the disposal of the infantry commander to whom 
it reported for orders, and followed the infantry on recon- 
noitered roads. 

The infantry must quickly perceive and utilize the 
freedom of action given it by the tanks. It must keep close 
to the tank3 ; it must be a point of honor with the infantry 
never to let a tank fall into the enemy's hands. In no case 
must the infantry retreat if it sees the tanks traveling to 
the rear, as they have probably turned around only to fight 
against nests reappearing in rear, or to assemble. 



182 DEVELOPMENT OF TACTICS— WORLD WAR 

If at some points a block occurs in the forward move- 
ment through fire from some trench, or from a machine gun 
nest, the tanks fight down the resistance. The parts of 
the infantry not struck by the fire must not let themselves 
be stopped in the forward movement. If that piece of ter- 
rain has been cleaned up, the troops that for a time re- 
mained behind advance again with the tanks. 

In England 8 battalions (86 tanks in 8 companies) 
were formed into a brigade. In addition to the new large 
battle tank (86-ton) and the light (Whippet) tanks there 
were transport tanks for infantry (20 to 80 men with 5 
machine guns) and for guns (12.7 CM.). Each division had 
1 tank battalion (86 tanks), which was formed into 2 waves 
and its first wave traveled 80 paces in rear of the creeping 
barrage, which was intensified by smoke and fog shells. 
Infantry remained with platoons deployed in squads up to 
ISO meters. Inmiediately behind these infantry platoons 
1 tank of each tank platoon remained as reserve to fill up 
resulting gaps in the line and to fight down reappearing 
resistance. Then followed the mass of the infantry, which 
overcoming the stationary barrage, closed up and gained 
the necessary depth in the further advance. The distance 
of the second tank wave from the first differed. It was the 
task of the second wave (which also was followed by inf an- 

A BRITISH TANK ATTACK, 1918 (POSITION IN READI- 
NESS—ADVANCE GUARD) 

A A A A 



fOOm 



11 A A A ^-^'^ A 

:: M II ri ii 

Sj li II 11 II 



II 



II II M II n 

IJ It II II M 

A second battalion follows in reserve 



TECHNIQUE IN WAR 188 

try) to fight down the resistance nests that remained in the 
intermediate terrain, and attack any new target. Some- 
times a third wave was formed which had the same tasks 
as the second. If a tank became disabled and could not pro- 
ceed, the crew had orders to descend with 1 or 2 machine 
guns and gain a foothold in the terrain until the infantry 
came up. In crossing watercourses without stable bridges 
a tank may be used as a bridge tank. 

The infantry must fight as if no tanks were present. 
It must make full use of its auxiliary arms and must in no 
case halt if tanks fall out or remain stationary. 

In fighting for villages the infantry at the start pushed 
forward to the edge of the village and nested itself there. 
In the meantime the tanks drove to the side and outer edge 
of the village and then broke through the houses into the 
interior. 

Because of the enormous demands on the German in- 
dustry for supplies of all kinds, ammunition, and artillery 
materiel, we were prevented from introducing tanks in suffi- 
cient numbers, as the main point was to keep up our artil- 
lery, the aerial fighting forces and the submarines. Only by 
limiting construction of these essentials could we have man- 
ufactured tanks in sufficient numbers. The United States 
of America has finally adopted the last word in transporta- 
tion of guns in difficult terrain by introducing '^otor mules," 
which are not limited to the hard surface roads, like the 
draggers of the heaviest guns. The advantage consists 
principally in the greater economy (driving power for 80 
kilometers at only 25% of the cost of the teams), lessening 
the train parks, shortening the march columns (in conse- 
quence of greater load carriage) and increased capacity. 

The necessity of securing to the higher conmiand in 
position warfare quietness for work, brought about a draw- 
ing back of division headquarters from the front lines to 
about 10 kilometers. From there, according to the needs, 
officers were sent forward to the fighting points for obser- 
vation and as representatives of the commander. As the 
leader no longer could personally view the terrain, facilities 
for making reliable maps on a large scale (in the West 
1:80,000, 26,000, 10,000 and 6,000) was necessary. These 



134 DEVELOPMENT OF TACTICS— WORLD WAR 

were executed with the assistance of photographs taken 
from airplanes and reproduced in large quantities in three 
colors, the different points being designated by numbers. 

For connection with the places of commanders,, tele- 
phone nets were established with other points, with defense 
guns, infantry and artillery. Up to May, 1917, there were 
515,000 kilometers (322,000 miles) of lines in the West and 
349,000 (220,000 miles) in the East. The average monthly 
expenditure on main battle front in the West was in 1917 
1.7 kg of Imm wire, 19 kg of 2mm, 21 kg of 3mm and 20 kg 
of 4mm wire per kilometer of front (about 250 lbs. i>er nule 
of front). The telephone had the advantage of direct per- 
sonal exchange of thoughts and thereby became an unre- 
placeable auxiliary means of leadership, even if there was 
danger of interference of higher headquarters and the dan- 
ger of attempts to influence the decisions of the higher leader 
by direct communication between adjutants and general 
staff officers. In spite of all efforts we were not able to 
overcome many disadvantages. The traffic left no visible 
traces behind like the telegraph instrument with writing 
attachment. It is true that the efficiency of the wire cir- 
cuits was influenced by gun fire and by the weather, still 
we succeeded in keeping the telephone in commission under 
fire by seeking out stretches where the fire was not so hot, 
by avoiding places open to direct bombardment, by laying 
the wires in cables in trenches from 0.5 to 0.8 meters deep 
and by employing numerous ''trouble" squads. The dan- 
ger of listening-in (by the enemy) was grave; this we at 
first underestimated; the enemy did listen-in at ranges of 
3 kilometers ; we tried to overcome that by construction of 
double lines and by the strictest supervision of conversation, 
by installing special ''centrals'' (also by use of code, by lim- 
iting conversation in the danger zone to the utmost neces- 
sary, using a tone of voice strange to the enemy; and by us- 
ing a foreign language or dialect). Wireless telephones 
with a range of 16 kilometers (10 miles) were used by the 
British toward the end of the war for airplane communica- 
tion. 

As a supplement to, or to take the place of, the tele- 
phone, the slow relay chain of messengers or mounted mes- 



TECHNIQUE IN WAR 186 

sengers proved reliable under heavy fire. We soon aban- 
doned the very slow wig-wag signaling of peace time, while 
on the other hand, utilization of carrier pigeons found favor 
(each division sector in the West was aUowed about 4 cotes, 
situated at least 16 kilometers behind the front) ; these 
contained 80 to 120 pigeons, which were freed hy twos or 
more. Long changes of location of the cotes should be 
avoided, but minor changes up to 8 kilometers did no harm ; 
it took about two weeks' training for the pigeons in their 
new home when the change was longer. The pigeon was 
little influenced by hostile fire or gas and the altitude at 
which they are released from airplanes (up to 8500 meters) 
did not matter; but the pigeons were dependent on clear 
weather. The flying velocity was about one minute per 
kilometer. The disadvantage remained that the pigeon 
should be used only in (me flying direction, from the front 
to the rear. It was a reliable means of information coming 
from the first line and could be also advantageously em* 
ployed during the range finding by the artillery. 

The war dog (message dog) proved very good up to a 
distance of 2 kilometers after previous training (12 dogs 
per regiment) between two well known points occupied by 
commanders whom the dogs knew and who knew the dogs. 
The service in such cases was exceUent. We will not dis- 
cuss the sentry dogs the French loved to employ ; it is not 
known if their ^'attack'' hounds were a success. The dogs 
found their road even in very difficult terrain ; but we must 
avoid demanding anything beyond their abilities. Only a 
very small percentage of the dogs employed were failures. 
Some dogs carried more than SO messages per day. Occar 
sionally their work was decreased by their being fire-shy. 
To overcome this was a matter of training. Frequently 
under very heavy fire dogs sought cover in shelters and 
bombproof s and continued their run carelessly as soon as 
the fire became weaker. Prerequisite for employment was 
good treatment by both dog masters, and that no one else 
interfered with the dog. However, in case of disabilily of 
one of the dog masters, the dog was useless. Carrier pig- 
eons and message dogs frequently were the only means of 
communication between the fighting troops and the com- 
Mmders. 



186 DEVELOPMENT OF TAGTIGa— WORLD WAR 

Technically, the means of information were mtpfde- 
mented by message projectiles, by the message throvring 
grenade fired from the rifle at a range of 1000 meters and 
Iqr the light trench mortar with a range of ISOO meters. 
Daring the course of the campaign the light signal appara- 
tus, that had been neglected during peace, was improved 
and information troops were equipped with the large flash- 
light apparatus. The infantry, artOlery, pioneers and min- 
enwerf ers were equipped with the medium and small flash- 
light apparatus. Flash-light No. 17 served for c<mmiunica- 
tion with airplanes. The source of light was an dectric 
lamp; communication was by means of the Morse code. 
Division and regimental headquarters (infantry and artil* 
lery), and fidid artillery battalions had 4 medium signal 
apparatus. No. 16, and brigade headquarters 2 No. 16, while 
infantry battalions, and cavalry rifle regiments had eight, 
and machine gun companies two small signal apparatus No. 
16. 

Flash-light stations were quickly arranged, but they 
were dependent on the intermediate terrain and could not 
be entirely hidden either in the terrain nor on the aerial pho- 
tograph. They were influenced in their use by unfavorable 
weather conditions (fog, rain, snow) , and the transmission 
took time, as all messages had to be sent in code as the 
en^my could also read the signals. 

Special progress was made in equipping the divisions 
with a wireless battalion of two platoons. This battalion 
had a wheeled wireless station with a range up to 100 kilo- 
meters ; two large, portable wireless outfits for transmitting 
4 to 6 kilometers, five medium for transmitting 2 to 3 kilo- 
meters, and 6 smaller portable for transmitting 0.5 to 1 
kilometers range. Bullet-proof establishments were re- 
quired for employment in position warfare. By wireless 
communication with the airplanes, separated detachments 
were kept in touch regardless of the terrain, but we had to 
reckon of course with the fact that the enemy could listen 
in and so might gain information as to our cUstribution of 
forces. The French claim that by such listening-in that 
they received information of movements of Grerman range- 
finding airplanes and were able to notify the pursuit squad- 



TECHNIQUE IN WAR 187 

rons to interf ^e with the hostile aetivi^. Listening-in 
was also an aid in the location of hostile artillery* The fact 
that sudi listening or cutting in is possible in wire and 
wireless communication offers the possibili^ of misleading 
the enemy. The advantages of wireless are its independence 
of terrain and fire effect, and that short messages can be 
rapidly transmitted. Interruptions by traffic occurred to our 
own and hostile stations, so that placing of many stations 
in a confined area was impossible. Wireless can be com- 
pletely cut out by aerial electrical currents. 

Of special importance was the ground telegraph, by 
which we mean the transmission without wire of Morse 
signals, where geological conditions were favorable, up to 
2 kilometers distance, when communication over telephone 
lines was impossible. As the enemy could listen-in, mes- 
sages had to be coded; telephone and ground telegraph 
could not be used simultaneously alongside each other; 
ground noises, thunder and strong currents interfered. 
Ground telegraph instruments performed valuable service 
during advance and in attack, and, at a distance from the 
message assembly stations, and frequently were the only 
means of communication with the flanks of the divisions. 
It proved valuable in solving the communication question 
from company to battalion; from battalion with ground 
telegraph, flash-light, or small wireless equipment to divi- 
sion battle observing station and to the division message 
center; and from the artillery observer to the guide bat^ 
tery to start barrage fire by agreed signals. 

False signals sent out and screening of the service of 
radio communication should in each instance be carefully 
considered, for measures that are too crude fail in their 
object. Measures for receipt should be the same in all dis- 
tricts under hostile observation. To screen this service of 
communication it may become necessary to stop the divi- 
sion in the second line from using wireless. Calls should 
never be sent out in the clear (for instance, one Italian army 
called all its units in sequence prior to our offensive in Octo- 
ber, 1917). Wireless silence and then a sudden renewed 
wireless activity are almost always a sure sign of change in 
the tactical situation and should therefore be avoided. 



138 DEVELOPMENT OF TACTICS— WORLD WAR 

The fifth arm, the aerial fighting forces owe their re- 
markable development to the World War. Airships (dirigi- 
bles) were replaced by airplanes due to the developments in 
size and capabilities of the latter. 

The primary superiority of the French in the matter 
of planes did not last long. 

At the opening of the World War airplanes were only 
used for reconnaissance. They were poorly equipped with 
photographic apparatus and machine guns were not mounted 
nor wireless equipment. 

At the beginning of the war, airships (dirigibles) op- 
erated at an altitude of 2400 meters and attained a velocity 
of from 54 to 87 kilometers per hour. Airplanes operated 
at an altitude of approximately 1000 meters with a velocity 
of 70 to 120 kilometers per hour, and an effective range of 
350 kilometers. Dropping bombs was still in the incipient 
stage. At the most, only 4 bombs weighing 10 kg each could 
be carried. The employment of improved anti-aircraft ar- 
tillery continually forced the planes to seek higher operat- 
ing altitudes. 

Aerial battles were not thought of at the opening of 
the war, airplanes avoided each other, unless some fired with 
revolvers at others. The mounting of machine guns on 
airplanes was first done by France. This forced all other 
states to follow suit. An excellent solution of firing from 
an airplane was found by synchronizing the machine guns 
to fire through the revolving propellor without interfering 
with it. The mounting of 37-cm. guns on airplanes inter- 
fered with the rate of speed, so that procedure was aban- 
doned. Bomb dropping was consistently developed. The 
Allies dropped bombs on open cities at the commencement 
of the war and made no secret of their intentions to attack 
the hostile capital. Wireless communication from airplanes 
(adopted by the Allies in December, 1915) proved of incal- 
culable value for transmission of messages and exceeded in 
reliability all heretofore applied means of communication. 

Captive balloons received little attention at the begin- 
ning of the war. They were at the mercy of hostile artil- 
lery and planes. The gases of the balloons easily caught 
fire, and the protection afforded by the anti-aircraft de- 



TECHNIQUE IN WAR 189 

f ense guns was seldom sufficient. Batteries in action were 
seen from balloons at distances of 14 kilometers and march- 
ing troops and columns at 25 kilometers. Hostile artillery 
fire forced captive balloons to be kept about 8 kilometers in 
rear of the front lines. The balloon observer at 1000 meters 
altitude could see from 80 to 40 kilometers. He also had a 
stationary position with good permanent ground communi- 
cation. These advantages were so great that captive bal- 
loons were universally employed for observation of artillery 
fire and the battlefield. By means of the balloon and tele- 
phone, headquarters received reliable information on both 
the hostile and friendly situation. 

Dirigibles took a back-seat in the war ; they offered too 
large a target, demanded special arrangements for landing, 
skillful construction, required an enormous amount of ma- 
terial and took a long time for construction. They also 
required a large suitable landing place, and an expensive 
hangar construction which took more time. Tliese disad- 
vantages could not be offset by the fact that the dirigi- 
ble could carry large quantities of explosives and traffic sup- 
plies. Employment of dirigibles was confined to naval war- 
fare. 

With the development in the capabilities of airplanes 
grew the manifoldness of their tasks, which naturally led 
to dividing the airplane organizations into: pursuit squad- 
rons for attacking hostile aerial fighting forces ; protective 
squadrons for offensive protection for observation airplanes ; 
artillery squadrons for adjusting the fire of the artillery; 
attack squadrons for fighting ground targets with machine 
guns ; and infantry squadrons for the requirements of the 
infantry in their battle on the ground. Bomb-dropping was 
retained. The photographic work of the reconnaissance 
flights steadily increased in importance. The participation 
of planes in an attack in best illustrated by the attack on the 
Chaume forest (Omes sector in front of Verdun) on Sep- 
tember 24, 1917, by the 18th Reserve Division, to which 5 
plane detachments, 8 protective and 8 pursuit squadrons 
were attached. Route of attack and attack objective were 
photographed. Based on these photographs, maps were re- 
produced showing the lines to be reached by the attacking 



140 DEVELOPMENT OF TACTICS— WORLD WAB 

troops and the desired route to be taken by the attack squad- 
rons, and these maps were liberally distributed to headquar- 
terSy troops, and airplane' organizations. At 5 :40 A.M. the 
infantry started the attack, at 5:66 A.M. the first attack 
squadron with 6 planes appeared and in four attacks fired on 
hostile reserves from an altitude of 100 to 400 meters; a 
second squadron repeated the procedure and dropped bombs. 
Reconnaissance planes that had timely perceived the start 
of the hostile counter-attack, called for barrage fire and 
caused a repetition of the attacks by the attack squadrons. 
Pursuit squadrons prevented hostile planes from attacking 
during the entire action. 

The necessity to sweep away the counter effect and to 
reach the objective in spite of losses suffered, led to the 
employment of airplanes in formations ; in chains (of 3 to 
4), in swarms (2 to 3 chains), or as pursuit echelons. Sev- 
eral of the latter, assigned to a troop unit, formed a pur- 
suit squadron, several squadrons a group. Thus aerial bat- 
tle units in close formation of 50 to 70 airplanes, had their 
inception. . The unit was under the orders of the comman- 
der, who, however, had authority to employ the unit in echel- 
ons or squadrons, or as a complete unit. Captain Boelke 
was the originator of the organization. The third year of 
the war saw squadron fights in the air. Giant airplanes 
served for the purpose of dropping bombs on important tar- 
gets. A prerequisite for the solution of tasks given to all 
airplane organizations, is, supremacy of the air — ^by re- 
lentless attack, never by defensive tactics. If supremacy 
of the air is once lost, it is difficult to regain. Screening is 
possible only after the enemy has been thoroughly fright- 
ened. 

The airplanes found support in the anti-aircraft guns 
which compelled the hostile airplanes to use greater caution 
and to operate at higher altitudes. Toward the end of the 
war the number of airplanes brought down by anti-aircraft 
defense guns increased materially. Machine gun fire with 
some tracer ammunition is the best means of attacking low 
flying airplanes. 

How the fighting conditions developed during the war 
is shown by the following table : 



TECHNIQUE IN WAB 141 

The foUowing airplanes were lost: 

In S0pUmber, 1915 : In SepUmber, 1918 : 

IvVffllVOW A^^WM^ Ww^^9m^ ^^'^^vHSiW ^vHMP 

In aerial battte 8 4 11 107 S62 

Sf firing fiom ground 2 14 126 

iasing 2 8 7 186 

Airplane defense searchlights proved to be a valuable 
means of airplane defense at night for the protection of 
important works, villages and roads. They forced airplanes 
to seek higher altitudes and thus abandon the chance of 
finding the marks on the ground so necessary for correct 
orientation. Blinding the airplane pilot by the direct rays 
of the searchlight deprived them of their assurance in the 
flight, so that they were induced to drop their bombs pre- 
maturely or unaimed. 

The attack sections, or echelons (6 airplanes), organ- 
ized in the fourth year of the war from protective echelons, 
produced a moral effect by diving from high to low altitudes, 
and participating in the fight with machine guns, bombs 
and hand grenades at the moment of the assault. Their 
assault was directed on the leading infantry lines or against 
the artillery which was laying down a barrage. In the de- 
fense, in the pursuit, and also in the retreat (holding up 
advancing cavalry and artillery) attack sections found far 
reaching employment. 

Infantry airplanes watched permanently over the bat- 
tlefield and took over the battle reconnaissance, transmitted 
signals from the front to the rear, and supplied the front 
line, when required, by dropping orders, subsistence, ammu- 
nition and hand-to-hand fighting means. Details of our, 
or the enemy's position, were noted on maps or photo- 
graphed. Of special importance was the information, 
whether or not the hostile trenches were filled, where the 
reserves were and location of hostile tanks. To be able to 
perceive all positions in readiness ih hostile trenches, the 
airplane had to descend to an altitude of 300 meters. Infan- 
try airplanes could solve their task only if the infantry, on 
demand, showed the location of its advance line with either 
white or red cloths, according to the color of ground, or by 
displaying light signals. It took a long time for infantry 
to lose its fear in thus supporting the airplanes. The infantry 



142 DEVELOPMENT OP TACTICS— WORLD WAR 

airplane frequently was the only means of communication 
between the fighting troops and headquarters and thus fur- 
nished the first basis for battle conduct. The certainty of 
being recognized by its airplanes strengthened the moral 
feeling of the infantry and induced it to hold out, even 
if surrounded by the enemy. Effective firing against the 
enemy offered the infantry airplane the best protection. 

The development of photographic reproduction by air- 
planes forced troops to pay more attention to the screening 
of all works because success of the attack was primarily 
based on surprise. The rule was to construct the works so 
exact in color and shade effect that they could not be distin- 
guished from the surrounding ground. 

Screening the position, which must absolutely begin 
even before construction of the position, must not only con- 
ceal important works from view (for instance gun embras- 
ures, entrances to shelters), but must also not attract at- 
tention to the point screened. The point screened must not 
distinguish itself by any means from its prior appearance 
and must never appear unusual. On March 11, 1918, the 
British 5th Army demonstrated that the screening should 
even be in conformity with the approaching spring colors. 
For instance, the battery commanders had to state the color 
and peculiarity of the ground where their batteries stood. 

Camouflage was especially improved on by our enemies. 
"The Frenchman suggests hiding from sight any work on 
the ground, even before starting it, by cover and to carry 
on the work under that* cover. The Englishman does not 
recommend that expressly. The Frenchman believes sow- 
ing and setting plants on artificial works to be especially 
effective ; and besides this he uses opaque screening or grates 
covered with sod or painted canvas. The Englishman loves 
to use nets to the meshes of which colored strips of cloth 
are attached. The Frenchman hides his batteries prefera- 
bly in old or not yet completed, infantry works ; the English- 
man likes to make battery positions unperceivable by avoid- 
ing the shadow effect and to then camouflage them on a 
large scale to correspond to the entire surrounding (decep- 
tion of square fields, etc.) or to give them a very irregular 
appearance defying detection. For the rest, both French- 



TECHNIQUE IN WAR 148 

man and Englishman place the same value on the fact of 
having the terrain weU guarded by airplanes prior to and 
during the work. 

^'The French have within their airplane units certain 
'camouflage detachments' which are charged with advising 
other troops concerning the camouflage. The Englishmen, 
on the other hand, have a special camouflage officer at each 
corps headquarters. In addition they have special camou- 
flage depots; both showing that they carry on camouflage 
on a large scale. Even the pioneer parks keep camouflage 
material in readiness. As such thev serve special nets with 
attached (to the meshes) strips of material made of gunny 
sack and scrim, which are colored corresponding to the sea- 
son." 

In the last months of the war the airplanes were not 
only means of reconnaissance to evade battle but they parti- 
cipated in the battle in many ways, such as described in the 
preceding paragraphs. Aerial fighting was their principal 
duty. Tlie intention to prevent the enemy from accom- 
plishing its manifold aerial tasks was merely a means to the 
object. '"There can be no doubts whatever, that supremacy 
of the air will be of decisive importance in future wars. The 
one that has gained supremacy of the air will see the victory 
on the ground fall into his lap like a ripe fruit. Wars will 
be of short duration, if they are not smothered in their in- 
ception. That party which gains a complete victory in the 
air will prevent the hostile mobilization, transportation and 
concentration, or at least greatly interfere with them. The 
hostile centers of industry, the hostile capital will be anni- 
hilated, no matter how far they are from the frontier. No 
longer are there protective distances in the face of modem 
^rplanes with their radius of action. Protection by dark- 
ness is cut off as well, as parachutes with magnesium lights 
can, change night into day. Even low flying clouds do not 
interfere with the activity of airplanes, as was proven by 
the British aerial escadrilles in the September and October 
days of 1918. All defensive measures such as anti-aircraft, 
machine guns, searchlights, etc., have been unable to ma- 
terially interfere with the hostile flying activity, have not 
prevented the hostile airplanes from successfully dropping 
their bombs, in diving down to within 10 meters of the 



144 DEVELOPMENT OF TACTIGS— WORLD WAR 

ground and attacking march columns and skirmishers with 
machine guns. If no absolute defense means is found, the 
aerial arms will become decisive. This knowledge our ene- 
mies took into account in the peace conditions, by prohibit- 
ing our keeping war airplanes, in place of which they allowed 
us to keep a relatively strong cavalry.'' 

The Hague Conference of July 21, 1899, prohibited, 
against the representations of England and the United 
States, 'i;he employment of projectiles the sole purpose of 
which is to spread choking and poisonous gases,'' — ^this with 
the intention of preventing the adoption of a form of mass 
killing which it would be impossible for anyone to evade. 
Attempts to increase the effect of arms by the use of gas 
originated with our enemies. 

Attempts to use gas weapons were begun in France as 
early as September, 1914, and, according to the 'Tall Mall 
Gazette," these weapons were employed the same month at 
the front. A French War Ministry Order, dated May 21, 
1916, contained directions concerning deadly gas projectiles 
(26-mm. rifle grenades) . On the steamer '^Lusitania," which 
was sunk on May 7, 1915, there were 2500 hundredweigl^t 
of tetrachloride intended for the production of poisonous 
gases. The French began to use stupifying gases March 1, 
1915, and subsequently used them in the battles of Suippes 
(April 10-17, 1915,) and Verdun, and the British at Ypres. 
We made the first gas attack, according to British reports, 
on April 22, 1915, in front of Ypres. The effects of this 
attack were the more marked because of the lack, on the 
part of the enemy, of gas defense equipment A gas bom- 
bardment of the 9th Army, at Bolimow, on May 2, 1915, was 
without result. 

In view of our lack of raw materials for manufacturing 
ammunition, we were compelled to follow the example set 
by our enemies, and the highly developed chemical industry 
in Germany was called upon to furnish us the preponder- 
ance in the production of the new offensive and defensive 
materials. At first, chlorine and certain gases which, with- 
out being poisonous, affected the membranes, were used. 
These were later replaced by phosgene, and finally, in 1918, 
an especially effective fighting gas was found in the 'Tellow 



J 



TECHNIQUE IN WAR 146 

Cross'' (called '^mustard'' gas by the English, and Yperite 
by the French) . Poisonous gases are the more effective the 
less they can be detected by the eye (i.e. when colorless and 
invisible) and the nose ; and when their effects make them- 
selves f dt only after some time. Gases may be made effec- 
tive through clouds released from gas cylinders or through 
the use of gas projectiles. The cylinder discharge is depen- 
dent on wind, weather, and terrain. Only in rare cases can 
these prerequisites be brought into consonance with the in- 
tentions of the commander. Gas clouds can be employed 
more effectively for generally damaging the enemy. When 
the method was first tried out, the disadvantage became evi- 
dent that in a change of the wind our own troops must be 
endangered, and frequently the cylinders had to be held in 
readiness for a long time awaiting favorable winds. Par- 
ticularly wide gas attacks are required when it is desired 
to extend the effects so as to nullify hostile flank fire. As 
a rule the troops underestimate the effects of a gas attack, 
as shown by later statements of prisoners. In all cases 
the execution of a general attack as a follow-up of a gas 
cloud was a matter of chance, dependent on the wind. We 
therefore followed the lead of the French and British, who 
utilized gas clouds, not for introducing an attack, but for 
the general damaging of the enemy, thereby obviating 
the necessity of assuring the difiicult matter of cooperation 
with the infantry. 

More independent of the wind is the employment of 
minenwerfer and artillery gas shell, which, however, can 
be used only against reserves and artillery positions, par- 
ticularly for neutralizing barrage batteries, but not against 
troops in the front line. Against the use of the heavy 
minenwerfer shell, there is to be considered the slow rate 
of fire and the limited range of this weapon, as well as the 
great weight of the projectile; its advantage lies in inde- 
pendence of the terrain. In case heavy minenwerfers are 
installed for other purposes, however, they can be utilized 
for gas bombardments at any time by bringing up the 
special gas ammunition. The effects of gases depend on 
the terrain (depressions, cornfields, and low-lying roads 
are especially favorable) and on the weather (rain during 



146 DEVELOPMENT OF TACTICS— WORLD WAR 

gas firing diminishes its effect) ; in the matter of time, 
the first morning hours have proven best. Gas bombard- 
ments have effect only when large areas are gassed, with 
much ammunition and for long periods of time, as other- 
wise the troops can too easily avoid the bombarded area. 
If the gas is to be effective, it must appear as a surprise, 
or else, by its long duration, exhaust the efficiency of the 
gas-mask canister. 

If we fire gas in connection with an attack, the enemy 
does not know how long this will continue or what kind of 
gas we are using, and he will be forced to wear his mask. 
This facilitates the hand-to-hand fighting for us, if our 
infantry has been fully informed and instructed concerning 
the gas fire; the small possibility of incidental damage by 
gas to our own troops must be borne just as is the case 
with occasional loss from our own high-explosive fire. If 
it becomes necessary, our infantry can advance with masks 
on across short stretches where they may happen to en- 
counter any remaining gas. 

A particularly valuable improvement was brought about 
by the introduction of gas projectors. These were used 
first by the British (April, 1917, in front of the Sixth Army) , 
and were afterwards employed by us with advantage. They 
were dug in, in the style of earth mortars, in a number of 
rows, one behind the other. Fired by electricity, a burst 
of 900 projectors could be fired with a range up to 1800 me- 
ters (time of flight, 25 seconds) . Each projectile contained 
13.5 kgs., of liquid phosgene or chlorpicrin, so that a great 
concentration of gas could be produced on a limited area. 
Projector emplacements could be detected on airplane pho- 
tographs, so that by timely bombardments the emplacements 
could be destroyed. Detection of the projector discharge 
could not be prevented, but the sudden mass strike of the 
projectiles nevertheless came as such a surprise as to be 
effective before the men could put on their masks. All regu- 
larity in the employment of projectors was avoided. A pro- 
jector bombardment was accompanied by a burst of other 
fire, or was made independently, or simultaneously with 
an attack, thereby befogging the point of penetration. With 
gas projectors one can attain a far greater gas density than 



TECHNIQUE IN WAR 147 

with cylinders, and, in addition, the gas is released directly 
at the target. 

The matter of gas protection became one of particular 
importance. Each unit had its gas defense officers and non- 
commissioned officers, who were charged with the mainten- 
ance and careful fitting of masks in the gas chamber, under 
the full responsibility of their commanders. Gas masks were 
also used for horses and messenger dogs. Rapid adjustment 
of masks ; removal of the same only by order of an officer ; 
regulation of the gas alarm ; practice in carrying the mask 
at rest and while in motion; and training in wearing the 
mask at exercises, trench work, and ammunition work, are 
necessary to avoid losses. Practice with masks should in- 
clude passing orders from mouth to mouth» giving com- 
mands, estimating distances, aiming, and target practice, 
throwing of hand grenades, serving the telephone, repair 
of telephone lines, and first aid to wounded. In addition to 
the ordinary mask, certain self-contained oxygen breathing 
sets were used for protection against mine gases, for which 
the regular gas mask offered no protection. Toward the 
end of the war the use of gas increased continually, so that 
gas defense appliances became equally as important as the 
battle weapons themselves. 

Of other projectiles, there are still to be mentioned 
hand grenades (filled with irritating and smoke charges), 
and incendiary and smoke shell. The British were the first 
to make use of smoke materials, for the purpose of protect- 
ing the flanks of advancing infantry; to obscure the view 
of an observation station or machine gun; for deception, 
to divert the attention from important points; to simulate 
a gas attack ; to screen troop assemblies in the open, move- 
ments along roads and the flash of guns ; and, finally for the 
purpose of hiding troop concentrations and movements, bat- 
teries, and the flash of guns from the view of airplane ob- 
servers. Batteries were completely helpless in the pres- 
ence of smoke. One could not see 10 meters. Progress could 
therefore be made from gun to gun only by clinging to guide 
ropes. Registration points could not be seen. Security 
sentries sent out could not find their way back. An enemy 
attacking from the flank would not be noticed until he 



t 



148 DEVELOPMENT OF TACTICS— WORLD WAR 

reached the flank piece. Terrain and troope in rear of our 
own smoke screen were in general poorly protected against 
aerial observation. In the interior of the smoke cloud, the 
men, of course, could not be seen, but were, nevertheless, 
subjected to the fire directed against the smoke cloud. For 
this reason troops generally kept 400 meters in rear of the 
smoke cloud. By frequent repetition of the smoke screens, 
the enemy may be made to believe that he is being de- 
ceived, merely to draw his fire. If this has been done a few 
times an attack following a smoke cloud may have success. 
Troops advancing through a smoke cloud easily lose direc- 
tion. 

The employment of minenwerfers had been foreseen 
in Germany in time of peace for use in fortress warfare for 
the destruction of wire entanglements. In position warfare, 
the need was soon felt of supplementing the fire of the high 
angle artillery by some additional means for accurately pro- 
jecting high-explosive bombs at the shorter ranges. Aux- 
iliary weapons were thus introduced in the form of ''La- 
dungswerf er*' and smooth-bore minenwerfers. Even if min- 
enwerfers could not compete with artillery in the matter 
of range, improvements nevertheless seemed desirable in 
order to assure greater freedom in the selection of emplace- 
ments. 

Minenwerfers were set up on base plates, which did not 
become firmly emplaced, however, until after the first few 
rounds were fired. The light minenwerfer bomb had the 
explosive force of the field artillery shell, but, naturally, 
with less force of impact, due to the smaller initial velocity. 

With like weight projectile, the propellant charge was 
less, the explosive charge larger, and the dispersion smaller 
than is the case with artillery. The construction was sim- 
pler and the weight of the barrel smaller than for guns of 
the same caliber; consequently it was possible to produce a 
larger material and moral effect with less weapons and am- 
munition. But the gun had the advantage of a wider field 
of utilization ; it was also difiicult to conceal the fire of the 
minenwerfer, the bringing up of ammunition was difficult, 
and the rate of fire was far less than that of field guns. 
Iifinenwerfers took the place of and augmented the use of 



TECHNIQUE IN WAR 148 

field guns, and in their method of employment conformed to 
that laid down in artillery regulations. 

The problem as to which arm of the service the minen- 
werfers — ^which are placed in position and served by pio- 
neers ; intended to augment the fire of the artiUery, and for 
larger battle control assigned tactically the same ; and which 
must be employed in co-operation witii and emplaced in the 
area on the infantry — ^were to be assigned, was a much- 
mooted question that was not answered conclusively in the 
World War. Probably it would have been the simplest thing 
to have assigned them from the beginning to the infantry. 

At first each division was furnished a minenwerf er com- 
pany by the pioneers, in which all 3 kinds of minenwerfers 
(3 heavy, 6 medium, and 12 light) were united, and which 
was under the orders of the pioneer commander. Finally 
these companies included 4 heavy and 8 medium minen- 
werfers. This arrangement was not satisfactory. Then 
each battalion during the position warfare, until May, 1917, 
was equipped with 4 light minenwerfers, which were served 
by the infantry, and which were employed, in defense, for 
barrage fire and against trenches ; in attack they were em- 
ployed especially against machine guns. The instructions 
which were issued to combine the minenwerfers of a regi- 
ment for the purpose of securing uniformity of training and 
employment, prepared the way for the organization of regi- 
mental minenwerfer companies. Distributing the minen- 
werfers to companies in the battalion was forbidden. The 
difficulties of ammunition transport by means of infantry 
carrying parties necessitated the adoption of minenwerfer 
carts. With the introduction of a wheeled, flat-trajectory 
mount, the possibility of employing minenwerfers against 
tanks and machine guns was very much increased. Later, 
the High Command held a large number of minenwerfer 
units at its disposition. On September 1, 1918, the division 
minenwerfer companies were disbanded and their, place 
taken by regimental minenwerfer companies, with 9 light 
minenwerfers, horse-drawn (1-horse) ; and with 3 medium 
minenwerfers, unhorsed, to serve as reserve material. The 
order to organize a company consisting of medium and heavy 
minenwerfers came to naught because of the difficulties of 
the draft replacement situation. 



160 DEVELOPMENT OF TACTICS— WORLD WAR 

In large battles, the tactical employment of minenwer- 
f ers was in accordance with the orders of the artillery com- 
mander. The minenwerfers adopted the methods of fire 
of the artillery (destraction, harassing, annihilation and 
barrage fire) • This was especiaUy applicable to the minen- 
werf er battalions of General Headquarters. But as the 
minenwerf er companies were integral units of the infan- 
try regiments, they had of course to revert to the regiments 
at the proper time. Heavy and medium minenwerfers w^re 
unsuited for barrage fire, because of their low rate of fire. 
They found employment, however, in destruction fire and 
also in gas bombardments. They were quite effective against 
obstacles. When equipped with flat-trajectory mount, the 
light minenwerfers were employed especiaUy against live 
targets, agafinst machine gun nests, and in flat-trajectory 
fire against tanks. They performed valuable services also 
in mountain warfare. To facilitate fire control, the minen- 
werfers were formed in groups according to target and ter- 
rain. It was estimated that, for assault preparation, 2 med- 
ium or 1 heavy minenwerf er bombs were required for each 
meter of front of the hostile position, so that in order not 
to extend the duration of the preparation too much, 1 med- 
ium or heavy and 1 light minenwerf er were employed for 
every 50 meters of front. 



VL The Defensive Battle in Position Warfare 

The battles during the year 1917 had brought opinion 
concerning the defense to a final conclusion. All battles 
confirmed the experience, that any penetration of a posi- 
tion can be stopped only by a properly prepared depth zone ; 
this of course detmanded enormous labor of the troops. No 
matter how desirable it was to have the troops construct 
the position in its entire extent of breadth and depth the 
necessity of having the troops enjoy complete rest at times 
operated against this. A division front in tranquil sectors 
could be 6 kilometers and more, but in an active sector that 
breadth could not be more than 8 to 4 kilometers. It was 
difficult to determine the exact time when the division sec- 
tors could be reduced by the insertion of fresh divisions ; if 
reinforcements came too late, the fresh troops did not have 
time to get accustomed to their new surroundings. As 
knowledge of the terrain is one of the principal advantages 
the defender has over the attacker, it is better to hold the 
fresh troops back than to insert them too late. Even the 
strongest construction went to pieces in the course of time 
under heavy concentrated fire. The trench system was 
changed into a shell-torn field. Obstacles were swept away, 
well known points or targets destroyed or obliterated in a 
very short time, while inconspicuous works, that did not 
betray themselves on the airplane photographs by tracks 
and shadows, had an unusually long life. Artfllery positions 
in the open protected from aerial observation lost far less 
than well entrenched batteries. Artillery effect can be di- 
minished only by extending the targets in breadth and depth. 
The greater is the distribution of works, the less they are 
perceived by the enemy, and the more difficult it will be from 
a question of ammunition supply, to destroy them. 

It is important to have the main line of resistance and 
observation posts withdrawn from direct terrestrial obser- 
vation. Of less importance is the extent of the field of fire, 
which is frequently best met by a ''Rear-slope position'' with 

161 



Ifi2 DEVELOPMENT OF TACnCS— WOBLD WAB 

direct observation. On nearly all parts of the fronts 
attack preparations were easily perceived. Surprise, as a 
prerequisite for success^ gained increased importance. 

In tranquil times, earthworks, with shelters and ob- 
stacles, are important principally as they force Ihe attacker 
to start extensive preparations. His artiUery fire will aid 
in penetration, and therefore will be directed mainly upon 
the front line trenches; these must be weakly occupied as 
''securitjr" trenches and equipped with only a few small shel- 
ters. In the 2d and 3d line trenches the rest of the garri- 
son is distributed with their light minenwerfers, while the 
heavy minenwerfers are imbedded, under infantry protec- 
tion, in the shape of a checkerboard, and as inconspicuously 
as practicable, to serve as ^'strong points'' in the interme- 
diate terrain. It is the task of the troops in readiness be- 
hind the first line to eject the enemy by an immediate coun- 
ter-shock. More and more did the importance of a deeply 
formed gallery system become evident. There was a lack 
of labor to meet this demand. Skillfully constructed ap- 
proach trenches, arranged for fire preparation, proved es- 
pecially suited to "lock-in" an enemy that had entered. 

Battle zones must extend sufficiently in depth to force 
the attacker to move his artillery forward. Every oppor- 
tunity for defense between the lines and fighting zones 
must be utilized for strong points. 

Looking back we will once more emphasize, that, at the 
beginning of the position warfare, the front lines (without 
regard to the terrain) were where our attacks had stopped. 
For reasons of morale abandonment of terrain was rejected ; 
the battle was conducted on a single line with the resisting 
power increased by all possible means (Autumn battle in 
the Champagne). Even in favorable cases hostile entry 
into our positions was prevented only with heavy losses. 
In the Somme battle the attackers succeeded in smashing 
with their artillery our front line, so that the troops in 
readiness could not start the counter-attack. That had to 
be left to troops farther in rear. The smashed trenches in 
front could have been temporarily abandoned, if, at the 
conclusion of the battle, the terrain was still in our hands. 
We fought for the possession of the front trench at a heavy 



DEFENSIVE BATTLE IN POSITION WARFARE 168 

loss. The necessity to camouflage all works from air and 
ground reconnaissance, in order to deceive the hostile ob- 
server by color and form, assumed prime importance both 
to us and our enemies. It was shown in 1917 that in the 
face of the increased artillery preparation the leading 
trenches were always lost. The more the defense insisted 
on holding the leading trenches, the more severe was the 
loss in men and materiel besides the loss of terrain. Loss 
of men could be avoided only by a distribution in depth. 
Difficulty of replacing our losses demanded husbanding our 
men. 'The object of defensive battle is no longer clinging 
to possession of the terrain, but rather the utmost infliction 
of damage to the enemy while preserving our own forces.*' 
If possible, unfavorable terrain should be left in the hands 
of the enemy. It is better to voluntarily abandon unfavor- 
able, shell plowed and torn positions and to lose terrain only, 
not men and valuable materiel in addition. 'The defender 
is not rigidly bound to his place, he is far rather justified 
to fight mobile on the battlefield, that is to advance or give 
way according to need. But the maxim, that at the conclu- 
sion of the battle the entire battle terrain shall be in pos- 
session of the defender unless direct orders are received to 
the contrary must be at heavy cost in men.'' The defender 
gave way under the hostile fire, the battle line responded to 
the pressure and laid itself like an elastic band around the 
attacker. For the support of this mobile defense "strong 
points" were created in the intermediateterrain interspersed 
with machine guns and infantry groups. Through skillful 
adaption to the terrain these were hidden from view, and 
could not be annihilated by the artillery preparation. It 
was their task to stop the flood that rolled forward after 
the shattering of the front trenches, and to disorganize it 
prior to its reaching our rear. Only with the introduction 
of tanks was it possible to hold down our machine gun fire 
during the attack. Mobility of the reserves was increased, 
regular mobile battles ensued within a position, in which 
horse artillery participated. 

That was the substance of the elastic defense. Thus, 
from the battle for possession of the front trench, developed 
a battle in a depth zone. 



154 DEVELOPMENT OF TACTICS— WORLD WAR 

By the fact that the opposing positions were very 
close to each other, frequently considerably less than 100 
meters (believed impossible prior to the war), it happened 
that the front line was exposed to hostile minenwerf ers. 
This proximity of front lines enabled the attacker to quickly 
pass the stationary barrage, and be in the hostile trenches, 
before the ganison could leave its shelters. Almost with- 
out exception the weak defender, working gradually out of 
his shelters, fought against superior numbers. Tliis fact 
accounts for the large number of prisoners taken. Creat- 
ing more favorable conditions for the defender could be 
obtained only by avoiding crowding the front trenches, and 
prolonging the distances the attacker had to cover. This 
led to the establishment of the zone defense. Troops in 
the front line in the zone defense are charged with the same 
tasks that in the mobile war, falls to the outposts. 

The outpost position is for the purpose of preventing 
the enemy gaining, without great losses, a point of attack 
in the immediate vicinity of our main line of resistance. 
Resistance should increase the deeper the enemy penetrates. 
Deceiving the enemy concerning the location of our main 
line of resistance and creating time for the troops in rear 
to be ready for action, is the duty of the outpost position. 
Whether the main battle runs its course on the outpost line, 
or in rear is the business of the highest commander, and 
not of the troops. The troops offer a stubborn resistance 
at the point designated, until different orders are received. 
The outpost position is occupied in force only by direct or- 
ders of the commander when the attack has been perceived. 
In tranquil times, the outpost position is held lightly but 
with sufficient force to drive off hostile patrols, or eject at 
once a hostile raid. The effort to establish two fortified 
positions, keep them up, and supply them with means of 
information and communication, requires too many troops 
for the labor purposes. 

In tranquil times, we will frequently quarter a portion 
of the garrison in the outpost position, and withdraw them 
at the opening of the artillery duel, or at other unmistakable 
signs that an attack is about to commence. To deceive the 
enemy as to strength of garrison is difficult at all times. 



DEFENSIVE BATTLE IN POSITION WARFARE 156 

Diminution of troops and the traffic as well as the up-keep 
of the position will soon leave traces on aerial photographs. 
The method of conducting the battle cannot be kept away 
from the troops, and may be brought to the knowledge of 
the enemy by prisoners. In any case the outpost position 
must have sufficient troops to compel the attacker to deploy 
his full force, so that they can hold the hostile advanced 
troops with the means at hand and give way, according to 
plan only when it is clearly seen that the main attack is im- 
minent, and our main defensive forces have been alarmed. 
Only when we strictly insist on the outpost position not be- 
ing evacuated without orders, can we overcome the disad- 
vantages connected with giving way to the rear. The out- 
post position may also be given the mission of offering as 
much resistance as possible to the attack, in order to create 
favorable conditions for the counter-attack. It is better to 
sacrifice the garrison than to abandon the original battle 
plan. If the attacker succeeds with his surprise, then the 
foreground will become a false position. In such case we 
can count on a decrease of the hostile artillery fire. But 
if the defender himself intends to carry out extensive 
artillery fire on the hostile back areas, there is nothing 
left for him but to leave a portion of his guns in the fore- 
ground ; it will be questionable in case of a heavy attack if 
the guns can be brought back in time. The outpost battle 
when conducted with small detachments with many machine 
guns firing from all sides will delay the attacker. Infantry 
in the outpost acts primarily as security detachments for 
machine guns and advanced artillery. Reduction of the in- 
fantry force in a sector leaves no doubt in the minds of 
the troops concerning the intention of our high command, 
and natuarlly does not increase the feeling of security of 
the garrison. Consideration of plans for artillery in de- 
fense requires a decision as to whether the falling-back to 
its final position is to be made in one move or whether a 
temporary halt wiU first be made in the intermediate ter- 
rain. The longer we delay the enemy in front of our main 
battle position the better, but this should never lead to the 
outpost troops being cut off. Frequently the necessity of 
keeping possession of important observation posts leads to 



IM DEVELOPMENT OF TACTICS— WORLD WAR 

the decision to hold the outpost line. In this case the main 
line of resistance would coincide with the outpost defensive 
line, unless terrain in front could be gained by an attack. 
The primary mission of artillery is protection of the main 
battle position and it must in certain cases decline the im- 
portant artillery duel during the hostile attack preparation. 
Some artillery must be in advanced positions, and stationary 
barrage and annihilating fire must be assured in front of 
the main line. Detached pieces (advanced guns) which 
can deceive the enemy as to their number by rapid fire from 
several different positions are efficacious. As we learned* 
bringing back the artillery was successful beyond all ex- 
pectations. The main target for the artillery will always 
be the hostile infantry. It must be beaten down before it 
reaches the main battle position. This requires powerful 
and accurate fire concentration from the artillery's own and 
neighboring sectors against definite points of the foreground. 
The different tasks of the artillery in the fight for posses- 
sion of the outpost zone of the main battle position can be in 
most cases solved only from many different positions. The 
most careful preparatory work in tranquil times covering 
supply of ammunition, arranging observation posts, laying 
lines of communication, etc., is essential Regard for se- 
crecy must never mislead the highest commander to aban- 
don range finding for the primary barrages and the annihil- 
ating fire directed in front of our main battle position. Nor^ 
mal fire in tranquil times may be laid on zones far in rear 
of the enemy. With increasing activity of the hostile ar- 
tillery there arises increased difficulties in distinguishing 
between our and the enemy's hits, a temporary evacuation of 
portions of our positions may be ordered, and firing data 
carefully collected in quiet times must be available. The 
main point is to hide the range-finding from the attention 
of the enemy by proper distribution over a period of time. 
For the purpose of deception it was found well to use for 
range-finding shells with time-fuses, or fire by salvos of 
three pieces against the enemy, while the fourth piece found 
the desired range. A few trial shots sufiSced for the ne- 
cessary firing data. 



DEFENSIVE BATTLE IN POSITION WARFARE 167 

Theoretically, the idea is that the outpost should be 
evacuated only at a well understood signal of the commander 
of the troops, but errors in this are very easy. How is this 
order or signal to be made known to neighboring and our 
own troops? If the departure is to be made without fric- 
tion, the subordinate commanders and troops must have a 
clear idea concerning the course of the retreat fight, and this 
knowledge will betray the intention of the commander. 
Probably the first intimation given to the subordinate com- 
manders that the position will be evacuated will be when 
the artillery is drawn back, which requires at least one or 
two nights. The later the orders are issued, the more diffi- 
cult will it be to transmit them through the increased hostile 
fire. Before evacuation begins the enemy should at least 
be compelled to resort to extensive enveloping work, heavy 
losses, and expenditure of a large amount of ammunition. 
There are no rules governing the exact time these orders 
should be issued ; to properly determine that time is one of 
the most difficult tasks of leadership. Raids, increased gass- 
ing, and the arrival of deserters, are sure signs an attack 
is imminent. The right to order the evacuation had best be 
left to the Army or C!orps, not to the different division com- 
manders. There may be cases when the commander of the 
outpost troops should have authority to independently order 
the evacuation. Orders for covering fire should be issued by 
the same headquarters at the same time evacuation orders 
are issued. The possibility of fully utilizing the advantages 
of the outpost terrain depends above all else on the efficiency 
of the means of communication, which must assure battle 
control. Increased means of information are necessary. 
As a last means there remains communication by simple 
signals with our airplanes. Comtmunication between troops 
on the ground with airplanes is difficult. It is probably too 
much to demand the troops keep one eye on the enemy and 
one in the air. Every lull in the hostile fire should be 
taken advantage of to move troops. 

In general the depth of an outpost zone will be about 
500 meters and never more than 1000. In greater depth, 
means of communication are insecure. Barrage fire to the 
fiank on portions of the position which are holding out be- 



168 DEVELOPMENT OF TACTICS— WORLD WAR 

come ineffective and personal observation which the com- 
mander should have is difficult. If any foreground is less 
than 100 meters, the advantage of compelling the opponent 
to cross a fire swept zone is lost. In any case, the weaker 
our artillery is, and the less time available to prepare the 
position, the greater will be its depth. A prime requisite 
for success is concealment. 

After an attack has been stopped the troops, by coun- 
ter-attack, at once regain possession of the foreground. If 
this is not done, we will finally be forced to evacuate our 
position. But the danger that the decisive battle may take 
place too far in advance must be called to attention ; if we 
desire to hazard that, it must be done with firm decision and 
not left to the subordinate commanders. 

The strength of the outpost garrison is generally fixed 
at 1 to 2 infantry squads and one machine gun squad for 
each company; this is in general better than to place an 
entire company on a broad sector. We generally disliked 
to do without the light minenwerf ers of the infantry ; me- 
dium and heavy minenwerfers, considering their limited 
range, have to be placed far to the front. These can 
hardly be brought back in time because of their poor mo- 
bility, but they are excellent means of misleading the ene- 
my, and if they succeed in deceiving and infiicting loss 
on the enemy their own loss has been well paid for. The 
outpost zone has proved its value in the battles in Flanders. 
The events at Verdun speak less in its favor. 

The infantry soon felt that strong occupation of the 
front trenches, unnecessarily increased the losses. When 
the Allies concentrated their heavy fire more and more on 
the trenches in rear, it was quite natural that all eluded the 
fire by moving forward and thus again made more dense the 
occupation of the front trenches (October, 1917, in Flan- 
ders) ; large losses were the result. 

The demand came from the troops themselves not to 
occupy the front trenches systematically. This procedure 
undoubtedly would have been advantageous at the time of 
the hostile artillery preparation in attacks on a large scale. 

There must be sufficient troops to stop an attack and 
the individual must have a neighbor in his vicinity to in- 
crease his self-confidence. 



DEFENSIVE BATTLE IN POSITION WARFARE 169 

The losses among the defenders of the front line can 
not be avoided ; the commander however, must refrain from 
continually reinforcing that line, for the losses then would 
be unjustifiable. The reserves must seek the decision. The 
general rule within the division was to place the infantry 
regiments abreast with battalions in column. These were 
known as fighting, readiness, and rest battalions, and were 
changed about every two weeks. In the matter of forma- 
tion it was advised that from the front line battalion of 4 
companies to every 200 meters front there should be one-half 
of each company occupying the main position, one-sixth 
of each cdmpany with their automatic rifles 200 meters in 
advance for security purposes and to resist weak attacks 
by raiding parties, about one-third of each company with 2 
automatic rifles forward as a support. Before a threatened 
attack the support was withdrawn to the main line of re- 
sistance. The supporting battalion deployed on a depth 
of 500 meters was brought up to within 200 meters, and 
in case of an attack the 2 companies automatically took 
the place vacated by the support while the 2 rear companies 
take the place of the 2 front companies. The regiment had 
a front of 800 meters with both its forward battalions hav- 
ing a depth of about 1000 meters. Proper consideration 
was given to holding only weakly the front line,, only 80 
men with 12 automatic rifles, while in the main line of re- 
sistance were 240 men with 12 automatic rifles and 8 ma- 
chine guns, which were reinforced by 160 men with 4 ma- 
chine guns, while for a counter-attack 240 men with 12 
automatic rifles and the Machine Gun company were in 
readiness. This was the arrangement made for strong 
counter-attacks. Against hostile attacks, artillery counter- 
offensive preparation fire proved specially effective, whereas 
barrage fire was always a makeshift as in most cases it 
came too late, and on account of the intensive artillery 
counter preparation was undertaken by fewer guns than 
had originally been intended. The ^'stationary barrage" 
required material augmentation by light minenwerf ers and 
by the fire of machine guns withheld until the moment of 
the assault. It was intended to stop the hostile reinforce- 
ment by placing barrage fire on critical points, to delay by 



\ 



160 DEVELOPMENT OF TACTICS— WORLD WAR 

tenacious resistance within the position the progress of the 
hostile infantry, at least enough to prevent tiieir having 
the protection of the creeping barrage. Only artillery which 
co-operates fully with infantry and which observes every 
nook and comer of the battlefield is equal to the demands 
of the defensive battle. In the forces held in readiness in 
rear lies the decision. 

In the penetration planned by General Nivelle certain 
troops held in readiness were to be thrown in if the attack 
had passed over the trench system. It was realized that 
the deeper the penetration the weaker would be the leading 
detachments, the more insufficient the artillery support and 
the deeper the flank. To counteract the effect of the pene- 
tration that might cross our trench system, an '^interfer- 
ence''* division was placed in readiness behind the trench 
system. This division could be used if the attacker endeav- 
ored to reach far off objectives beyond the main position. 
But when General Petain perceived the German method of 
defense he contented himself with limited objectives at 
close range aided by a systematic barrage of shell and gas. 
Nothing therefore remained for us but to execute the coun- 
ter-attack with troops consisting of all arms in tiie position 
system. If the counter-attack is not launched immediately 
the reserve troops remain inactive and are finally destroyed 
by fire. The commander of the defense has to reckon with 
a deep penetration and also with an attack against a limited 
objective. The task of the interference troops is difllcult; 
they must be at the proper point and beyond hostile artil- 
lery fire. They must not be frittered away by piece-meal, 
but must be thrown as a unit into the counter-attack. 

The local counter-attack and the general counter-at- 
tack differ only in point of time when they become effective. 
The local counter-attack acts automatically, imjnediately 
when the enemy enters. The disorder among the hostile 
troops, the impossibility of support by artillery facilitate 
the counter-attack and explain the success achieved by even 
weak, but well led detachments. 

The general counter-attack is a systematic attack, miade 
after thorough artillery preparation unless the situation per- 



*Interferenee troops axe those held in zeadinees f6r eoimter^ttack. 



DEFENSIVE BATTLE IN POSITION WARFARE 161 

mits surprise. ''Local counter-attacks launched too late 
do not differ from overhasty general counter-attacks ; they 
are useless and must be prohibited/' 

In the general counter-attack preparations can be made 
only after a definite decision has been reached. The time 
element is immaterial. General counter-attacks have been 
successful when the leadership understood how to wait and 
not endeavor to recover every loss of ground. It is an er- 
ror to be led into making any attack insufficiently prepared. 
Frequently the measures taken by the enemy are insufficient 
to retain what he has gained. 

In the local counter-attack all preparations must have 
been completed beforehand. If the enemy has entered, 
that is the signal for starting the local counter-attack; if 
we delay all hopes of success will fade. In tranquil times 
local and general reserves must utilize the available time 
for practicing counter-attacks. Any counter-attack comes 
too late if the enemy has succeeded in arranging his de- 
fense, that is, if he has regulated his depth formation, is^ 
sued his orders and prepared artillery barrage, and coun- 
ter preparation fire. The better we have prepared our de- 
fensive zone the more effectively our machine guns can 
participate from all sides, the more decisive the counter- 
attack will be executed, and the more stubborn will strong 
points hold out. The !more time leading waves can be held 
up in overcoming strong points and machine guns the less 
will the attacker profit from the protection of his steadily 
advancing barrage. Only in this manner can we account 
for the surprising success gained by small detachments. If, 
however, the troops hesitate only a short time and allow the 
attacker to reorganize the terrain it is not improbable that 
the shock of even a prepared detachment will not be suc- 
cessful. 

A prime requirement for success is that the commander 
of the troops in readiness carefully observes conditions of 
the front line, that he does not allow himself to be drawn 
into making a jump in the dark, or an uncertain start. 
Interference troops held in rear of the defensive zone are 
reserves in the hands of the higher commander. They 
form a support to the fighting front. They may relieve 



162 DEVELOPMENT OF TACTICS— WORLD WAR 

front line divisions by furnishing security detachments, 
connecting groups and Anally, battle reserves, so that the 
divisions in the front line can insert their last man without 
fear in holding the position. When the battle reserves of 
the division in the front line are inserted, portions of the 
interference division may take their place as battle reserves, 
their own places being then taken by troops from the rear. 
Through this systematic closing up of the reserves, the 
sharp and rapid counter-attack is assured. Places of readi- 
ness for troops depend on the task and the possibility of 
shelter. In tranquil times it will be advisable to advance 
ifrom each division, what is called an interference third (8 
battalions, 8 batteries, pioneer company and telephone de- 
tachment) , to the edge of the zone of hostile fire. ' Attach- 
ing labor parties to the interference third to construct 
covered trenches is desirable. The remainder of the troops 
of the interference division remain in their quarters in rear 
and in their sectors complete their training. For bringing 
up troops and supplies, horses are useful in tranquil times. 
The message service perfoitned by troopers, proved very 
satisfactory in Flanders and at Verdun. Only thus can we 
avoid the regimental units becoming lost or losing touch 
with their conmianders. From the base of attack forward, 
the employment of horses, however, will be impracticable, 
from there on the message dog and above all the efficient 
battle runner, starts his work. 

Reinforcing the front line almost always proceeds under 
the pressure of events and under heavy hostile fire. Only 
prior knowledge of the battle terrain, of its shelter, com- 
munication, organization, and plan of defense, may procure 
favorable conditions for the success of the interfering 
troops. It is necessary that the plan be so completely pre- 
pared that it can be started at a given work or signal. 

As a matter of experience^ the division in the front 
line had best keep control of its interference troops ; that 
division knows the progress of the battle information con- 
ditions, and the terrain, and has the greatest interest in 
holding its fighting zone. Only in this way may we assure 
unity in battle action. For the same reason it is advisable 
to have the headquarters of the division in the front line 



DEFENSIVE BATTLE IN POSITION WARFARE laS 

and the interference division at one and the same command 
post. If the interference division covers several division 
sectors, then personal conferences should take place prior 
to the battle and it will be advisable to send liaison officers 
to the divisions in the front line. In the same manner, 
connection between the interference regiments and the regi- 
ments in the front line will be required. Headquarters 
should be at one and the same point if for nothing else 
than mutual exchange of experiences and utilizing the shel- 
ter facilities that have already been prepared. 

Rank of the commanders must not be allowed to inter- 
fere. By placing the interference division under the or- 
ders of the division in the front line, tiiere is danger that 
the latter will absorb gradually the battalions of the for- 
mer for mere defensive purposes. This absorption has no 
bearing on the final result of the battle, and is absolutely 
dangerous. In nearly all battles, portions, sometimes strong 
forces, of the interference division remain immobile. In 
most cases the commanders of the troops in the front posi- 
tion have, without good reason, requested that units should 
remain in a certain place in order to reinforce some endan- 
gered part of the position. No attention must be paid to 
such requests. The commander who declines will take the 
responsibility for declining. Orders for starting local 
counter-attacks can be given only by the headquarters of 
the division in the front line. Headquarters of interfer- 
ence divisions knows too little about the situation, and its 
orders in any case would arrive too late. The conunander 
of the interfering troops had better orient himself with his 
own means at hand, concerning the situation, and indepen- 
dently order the general counter-attack. Army Headquar- 
ters ordered, and not without justification : 'It is strictly 
forbidden to delay local counter-attacks while pennission of 
next higher headquarters is requested.'' That is indeed 
very correct: such a question merely shows that the com- 
mander is lacking in decision and energy, and in waiting for 
an answer so much time is lost that the shock would come 
too latQ. This holds good equally for small and large en- 
gagements. The question, when the division counter-attack 
shall be startedr is of very great importance. At Verdun it 



164 DEVELOPMENT OF TACTICS— WORLD WAR 

took in favorable cases, from li to 2 hours after the divi- 
sion orders were issued until the troops started. In heavily 
shelled terrain we can count, in favorable cases, on 45 min- 
utes for covering 1 kilometer, and frequently it is even more 
than double that time. Not only the hostile fire, but the 
difficult passage over the shell-torn terrain, obstacles, and 
the necessity of making detours to avoid being seen, increase 
the period of time. Haste can be made only by an indomi- 
table will and by the example of the leaders. A premature 
start is, in any case, a lesser fault than to wait too long. 
The main point is not to come too late! A premature start 
may result in a jump in the dark, but delayed too long, the 
counter-attack leads to a costly defeat, or in the most fav- 
orable case, to a mere densi^ring of the troops in the front 
line. It is wrong, therefore, to wait for the results of re- 
connaissance. Considering the incalculable slowness of the 
service of communication, and the march, we will surely 
come too. late and lose the best opportunity. Therefore, 
without question, we must start without the usual battle 
reconnaissance. 

The lack of reconnaissance also increases the difficulties 
of the flanking counter-attack, which presupposes that the 
commander knows the exact location of the enemy. It is 
wrong to move forward and then change direction to gain 
a favorable attack position. Great losses will be the con- 
sequence, troops will be used up before they gain contact 
and will probably arrive too late. Such an attack had there- 
fore best be executed by troops in a neighboring sector. 
It would be unjustifiable independence unless a very favor- 
able opportunity offered, to push a flanking attack into a 
neighboring sector without orders from higher headquar- 
ters, which can see the whole situation. Only commanders 
of connecting groups may act in this manner. 

Flanking counter-attacks demand reconnaissance of the 
terrain, so that obstacles will not be encountered unexpect- 
edly. They further require special protection on the flank 
towards the enemy. The attacking troops should not be 
bothered with their own flanlc protection. Flanking coun- 
ter-attacks have the advantage of forcing the enemy to 
execute a change of front and further of avoiding the hos- 



1 
I 



DEFENSIVE BATTLE IN POSITION WARFABE 165 

tile box barrage. Finally, in the flanking counter-attack 
the occupation of the captured position is by no means sim- 
ple, but if the attack succeeds, the prize in prisoners and 
supplies is large. 

Carrying the counter-attack rapidly forward, is the 
surest guarantee for success and for lessening the losses. 
But an attack can be carried through rapidly only if it is 
supported by the necessary fire. This is not the business 
of the assaulting troops, but of the attached automatic rifle, 
machine guns, light minenwerfers, infantry guns, and ar- 
tillery. If we must transfer the fire preparation to the in- 
f antiy, we must then form strong firing lines at long ranges, 
and not employ open skirmish waves, and thus suffer greater 
losses. The open skirmish waves are not for fighting but 
a movement formation. The advance without firing can 
be made as soon as the hostile infantry is held in their shel- 
ters by our machine gun, and artillery fire. Only when 
that has not been accomplished must our infantry work it- 
self up to the hostile position under protection of its own 
fire. Only the skill of the subordinate commanders and 
the efficiency of the troops can find the right road to suc- 
cess. We may say, in general, that in such cases our old 
attack, confonning to the drill regulations, promises the 
best success. Working up by rushes, with fire support by 
all available machine guns and with sufficient support by 
the accompanying artillery using direct fire against the hos- 
tile machine guns. 

The general rule, to close rapidly with tjne enemy, is 
frequently infiuenced by the endeavor to keep the proper 
direction. The direction is the enemy. Therefore it is the 
general maxim never to wait for one's neighbor ; if we wait, 
there is no assurance that the troops will get to the enemy. 
The less open the terrain, the more depressions and woods 
prevent a view on the points of direction, the more neces- 
sary is it that a 'leading'' company precede with carefully 
sought out intermediate points, straight along the compass 
direction. Companies following in echelon may utilize 
without hesitancy any cover the terrain affords if they al- 
ways regain their proper position in regard to the ''leading" 
company. 



lee DEVELOPMENT OP TACTICS— WORLD WAR 

Whenever possible, our old line of trenches must be re- 
taken in the counter-«ttack. If that is not done, absolute 
connection with the artillery must be arranged so that the 
latter can regulate its fire accurately. It is advisable to 
attach liaison officers from supporting artillery to the at- 
tacking troops. Frequently at the most important moments 
reconnaissance to the flanks and connection with the rear 
is neglected. If the line is reached all available weapons 
must be fully utilized in pursuing fire, and steps taken at 
once to reorganize the position. Of importance is depth 
formation. Rapid re-establishment of depth formation 
must therefore be practiced in advance, like the counter- 
attack itself. In most cases establishment of depth forma- 
tion can be done only at night. Immediate withdrawing of 
detachments may be considered by weak minds as a sign 
of an intended retreat. 

In taking the measures for the counter-attack, the high 
command must always consider that the troops may lack 
the necessary material for constructing and holding the 
position after their success. To lay out panels is possible 
only after the action has been brought to a close and this 
had best be done by selected detachments, when requested 
by the infantry airplanes. Fog and smoke increase the 
difficulties of observation, the position has been changed 
into a shell-torn terrain and infantry airplanes have to 
descend very low to distinguish by its uniforms their own 
infantry. 

Co-operation of artillery and infantry has been men- 
tioned several times. Grenerally one artillery battalion of 
two gun and two howitzer batteries is attached to each in- 
fantry regiment as accompanying artillery. It has the task 
of sweeping obstacles out of the path of the infantry, and 
especially of destroying hostile machine guns. After the 
assault, the accompanying artillery gives the necessary sta- 
bility to the victorious infantry, and supports it during the 
first critical hours. This is impossible from positions far 
off and under cover. Accompanying the infantry attack is 
necessary also for moral reasons. It is said that the artil- 
lery loses time in moving forward but that is much better 
than to keep silent in positions in rear, because in critical 
situations it may fear to fire on its own infantry. 



DEFENSIVE BATTLE IN POSITION WARFARE 167 

The fire of the defense artillery must stop the enemy 
who has entered, then annihilate him. Daring a hostile 
attack with limited objectives the artillery in defense is 
uncertain as to what is taking place behind the rolling bar- 
rage. The questions : How far did the attacker push for- 
ward? Do our troops holdout? Are local counter-attacks 
being made? cannot be answered with any degree of cer- 
tainty. One of the duties of defense artillery is '^ocking-in 
the enemy that has entered/' that is cutting off by fire hos- 
tile reinforcements as well as keeping up the artillery pro- 
tection for the nighboring sector. Only where the enemy can 
be absolutely distinguished in our position can he be im- 
mediately fired on. Systematic locking-in to the rear pre- 
pares in the best way our counter-attack. C!ounter pre- 
paration fire which should be laid by all available batteries 
on the space in rear of the barrage can then be of decisive 
effect. When the enemy steps into the open, under the pro- 
tection of his creeping barrage, moving slowly to the front, 
the task becomes easier. The situation is llien dear and 
locking-in takes a back seat. 

The aerial fighting forces are of great importance in 
the execution of the counter-attack. Infantry airplanes 
should not be given too many tasks. The commander must 
be moderate. He should demand only: 

1. Where is our front line infantry? 

2. Where is the point of entry? 

3. Conduct of ^e enemy (advancing in what direction, rolling tip» 
etc.?). 

Pursuit squadrons did excellent service; their appear- 
ance at low altitude over the battlefield and their participa- 
tion with machine guns and bomb never failed to create a 
deep impression on friend and foe. Especially effective is 
their participation at the moment of the assault. In coun- 
ter-attack the leading hostile infantry is the most impor- 
tant objective. The main point is to confuse it, to hold it 
down, and to defeat it. Hostile machine guns, minenwer- 
fer positions, as well as batteries firing on our infantry's 
advance are proper targets for our machine gun fire. By 
the help of panels staked out at the headquarters it is pos- 
sible to give the battle airplanes information of the direc- 



168 DEVELOPMENT OF TACTICS— WORLD WAR 

tion of the hostile penetration and to demand a repetition of 
flights and complete reports. 

When well prepared counter-attacks did not have good 
results as at Verdun and in Flanders, the reason alwajrs lay 
in the piecemeal manner of making the attacks, general^ 
too late. Giving missions that are not clear, insufficient 
co-operation between troops in the front line and supporting 
artillery, and transfer of reserve to various commanders 
should be avoided. 



VIL Gennan Attacks with Limited Objectives 

Attacks with limited objectives occurred along the en- 
tire front for the purpose of securing local advantages such 
as obBervation positions. These generally succeeded after 
aiidUery preparation with relatively small loss. Frequently 
more material losses occurred in holding what had been 
gained. These operations proved to be effective means of 
increasing the morale of the troops. The desire for the exe- 
cution of such operations was expressed by the troops them- 
selves and they were encouraged even though the expendi- 
ture of ammunition and the losses suffered were not always 
commensurate with the success attained. The narrower the 
attacking front, the easier can the defender concentrate 
his artiUery fire on it. In the winter of 1916-1917 a large 
number of operations took place on the Somme (9th Army 
Corps Maisonnette, October 29, 1916; Pierre Vaast Forest, 
November 20, 1916), in the Champagne (Hill 185, south of 
Ripont, February 15, 1917) and around Verdun (18th In- 
fantry Division on Hill 804, January 15, 1917) and others, 
which offered valuable lessons in attacks with limited ob- 
jectives. The selection of the places to be entered and of 
the attack objective has to be made so that a sector shall 
be captured in the hostile trench system which cannot be 
easily flanked by the enemy himself and which can be easily 
incorporated into our trench system. Through mining and 
entrenching, flank protection and routes of approach must 
be prepared in advance. 

Conditions of the ground also have to be considered. 
Heavy rains frequently render the assault terrain too slip- 
pery and soft. Progress across No Man's Land will be very 
slow. Under certain conditions the operation may have to 
be delayed. 

The plan for the attack, made by division headquarters, 
is submitted to the Corps. It is important in such matters 
that Corps Headquarters should not discourage the initia- 
tive of the division commanders by unimportant objections. 
Any order that is sound is better than a technically correct 

160 



170 DEVELOPMENT OF TACTIOS— WORLD WAR 

order that is carelessly carried out. Battalion and company 
commanders issue their orders in the form of sketches and 
tables which should include position in readiness, attack 
objectives, roads, formations, composition of the waves, 
time and place of issue of equipment, etc. The artillery com- 
mander, assures proper range finding, and in hid order covers 
the artillery fire activity and co-operation with the minen- 
werfers and the infantry. 

An important element for any attack is the correct de- 
signation of time and watches must be repeatedly compared. 
Orders must be in the hands of the troops at least 24 hours 
prior to the attack and should be repeated back to assure 
correctness. 

Surprise is one of the decisive elements in the success 
of the attack. Every repetition of a former method of at- 
tack makes success questionable. False operations (insert- 
ing flame throwers, exploding mines, sudden artillery fire 
at different points and cessation after 5 minutes) will mis- 
lead the enemy. There should also be pauses in the artil- 
lery preparation, increase in the volume of fire, which the 
enemy will in most cases reply to with stationary barrages, 
so that at the assault proper the final fire is delivered with- 
out increasing the volume and with shells with time fuses. 
The assault wiU hardly be noticed by the enemy, and this 
will make it easier for our infantry closely f oUowing the 
barrage. A slow forward creeping barrage in front of l^e 
assaulting troops has been found very effective. The in- 
fantry must endeavor to push forward under the hostile 
barrage in close order and take depth formation subse- 
quently. The less the distance between starting point and 
final objective the greater the possibilities of success. Lift- 
ing of the barrage and the sudden start of the assault is 
made by watches that have been accurately synchronized* 

The time of attack had best be set shortly before dark, 
so that the troops wiU have about an hour^s daylight to find 
their way into the position, to strengthen it and to regulate 
the barrage. The longer it remains light, the more time 
will the enemy have to direct counter preparation fire on 
the position ; a strong gassing of the hostile batteries after 
the assault is launched will benefit our infantry, while gass- 



ATTACKS WITH LIMITED OBJECTIVES 171 

ing prior to the assault merely attracts the enemy's atten- 
tion. Darkness also facilitates the use of more troops for 
labor, but these must be withdrawn before daylight to lessen 
the losses. 

We must absolutely adhere to the hour and minute fixed 
for the start of the assault. New troops generally underes- 
timate the time absolutely required for issuing ordns and 
believe that it is possible to change an attack at the last 
hour. That is impossible and leads only to losses and fail- 
ure. 

The leader must decide whether he can base his success 
on surprise, or whether a longer artillery preparation will 
be required. A mixing up of both methods almost always 
leads to defeat. Along battle fronts troops are on the qui 
vive and a surprise wiU be difficult. We select that method, 
nevertheless, when there is no chance of assembling the 
assaulting troops under cover. The better the construction 
of the hostile position, the deeper the attack is to penetrate, 
the more thorough must be the artillery preparation. This 
will be more effective the more it is crowded into a short 
space of time. In about 3 hours a heavy field howitzer bat- 
tery can batter down a trench of 100 meters with from 400 
to 600 rounds. For light howitzers it would take one-half 
more of that amount of ammunition, and for mortars one- 
half less. Destruction of hostile obstacles may be an addi- 
tional artillery task during the preparation, or this mission 
may be assigned to the minenwerf ers during the artillery 
preparation. It may be best to assign to the artillery the 
task of counter battery fire and to make the preparation 
on hostile trenches by minenwerfers, concentrated with 
maximum fire for about 5 minutes. Heavy and sudden min- 
enwerf er fire, in any case, has a great moral effect. About 
20 medium and heavy minenwerfers, per 1000 meters of 
trench will be needed for this five-minute preparation. The 
preparation has to be so regulated that, even if no view can 
be had, the artillery fire runs like clockwork. Besides mak- 
ing the position ripe for assault, our artiUery must fire on 
hostile headquarters and places where it is Imown reserves 
are held in readiness as well as on routes of approach. The 
hostile artillery must be neutralized prior to our attack. As 



172 DEVELOPMENT OF TACTICS— WORLD WAR 

to whether or not the hostile artiUery should be gassed dur- 
ing our fire on the trenches depends on available time and 
whether we are willing to abandon the surprise element. 
In addition, the artillery must prevent the approach of the 
hostile reserves, and annihilate visible machine gun posi- 
tions and flanking works. The assault troops had best be 
placed in readiness during darkness in order to prevent hos- 
tile counter preparation fire. Passage through our own ob- 
stacles should be made as late as practicable. This is best 
accomplished by explosives (gas containers, filled with 
explosives, open passage ways from 4 to 5 meters broad). 
Air protection is secured by cautious use of airplanes which 
must not attract the enemy's attention (they should not 
arrive until 10 minutes prior to H hour) and by taking the 
necessary defensive means against hostile airplanes. 

Assurance of success lies in employing fresh troops 
that have been trained behind the lines and practiced in 
the formation to be adopted for the particular assault and 
who adhere closely to the exact course the attack is to take. 
These troops should be relieved as soon as possible after 
their success. A sufficient number is about 3 battalions, 3 
minenwerfer companies (each of 4 heavy and 8 light minen- 
werf ers) , and 2 pioneer companies, per 1000 meters of front. 
More important than numbers is the quality of the troops ! 

The assault detachment is divided into attack waves, 
to which machine guns, pioneers, artillery runners and sig- 
nal squads are distributed ; shock-troops go ahead to be em- 
ployed against resistance points noted by the leader. Car- 
rier squads with sandbags, quick obstacles, and means for 
hand-to-hand fighting follow. Whether the attacking waves 
advance in skirmish lines or in single columns, must be con- 
sidered in each case. Distance to pass beyond the hostile 
barrage at the same time to prevent the attack from be- 
coming disrupted should be as short as possible. Each wave 
must reach its designated objective independently and with- 
out regard to the other waves. The guide is the objective. 

If we assign the task of demolishing the hostile front 
line trenches to the medium and heavy minenwerfers, the 
artillery has the task of protecting the operation against 
the hostile works lying farther in rear, and throwing a bar- 



of a iMittaUon in attack from tmch to trmeh, picreiiiff depth into tkm 
that is to be held aboat 600 meten. Strrasth of out infiuitiy 

8 officezB (off. aspiranto), 22 iQoad leaden and MG eondza. 
116 men and 6 antixnatie machine suns 



ipanjr; 






i 



2. 



i 



ohodc Xfodpa 



i 



t 



I 



I 
I 



/. — 
2. — 

3-.- 



-•I- — 



4 



¥ 



tihtSm 



i&^H* ^ 



3. •{••h'h 



^••H* f 



^m, 




BirttaUoB 






+.J. 

Flank protection 



Ha 




t 



in f ^ravii 



* 



Plane defenee 



Fire support tlim flrinc over h ea d 

6 maeh. suns and 4 It. mineawwrfers 



I. 

z. 

3. 
4. 
5. 



strength of a company: 
Wa^t; 



n 



Officers 3 

« 4 

— 4 
3 LM.6.3 



1 
I 
I 



Sonad and MO leaders 



99 









Flank protection 



+ 



I 
i 
I 



Musieian 



Miaielaa 



24 


Men 


24 


»i 


22 


•I 


.20, 


•» 



99 



22 



• automat MG 

Am a rule companies with less than 120 men ara formed into two platoons. S au- 
tomatle maeh. suns with 12 men and cartaridces for each gun are in Stfa Wave. 

The necessary 24 litter-bearers have to be furnished by the bearer company eon- 
siderins the small strength of the flghtlng companiea. The grenade tiirower squad is 
also in the 6th Wave. 



174 DEVELOPMENt OF TACTICS— WORLD WAR 

rage on the hostile routes of approach as well as counter 
battery fire. Range adjustment must be done without at- 
tracting attention and with the aid of balloon observation. 
Considering the smaU distance between the opposing lines 
hostile front line fire by our artillery can hardly be accom- 
plished without endangering our own infantry. Early artil- 
lery fire against the hostile artillery betrays the coming 
assault. The barrage fire of the artillery increased by the 
fire of the minenwerf ers starts simultaneously with the in- 
fantry. After the successful assault it remains, gradually 
lessening, on a spot beyond the objective and thus forms the 
stationary barrage for the new position. 

The entrance of the infantry in the hostile trenches 
following directly on the bursting of the last shell has al- 
ways proven comparatively easy ; it takes some time before 
the opponent regains his senses and perceives the situation ; 
that condition the attacker must take full advantage of by 
rapidly pushing forward. The deeper he pushes his way 
into the enemy's position the larger is the gain in prisoners 
and the less will be his own losses. 

Success of the attack depends on rapidity and on sys- 
tematic co-operation between infantry and artillery; this 
requires that after reaching the objective, the infantry does 
not advance beyond it ; otherwise the infantry will be anni- 
hilated by its own artillery fire. Careful arrangements must 
be made as to the action to be taken in the event that the 
assault miscarries, or is perceived by the enemy ahead of 
time. There is no hope of sending orders to the first line ; 
the time consumed is incalculable, and it would only lead 
to a portion of the troops assaulting without artillery sup- 
port and they would surely be annihilated. 

When the objective has been reached it must be clearly 
made visible to our artillery and, on demand, to our air- 
planes, and communication to the rear must be promptly 
established. The artillery immediately starts range ad- 
justment for barrage. 

More difiScult than taking, is the holding of hostile 
trenches. In most cases counter-attacks accompanied by 
artillery will be started. The subordinate commanders must 
be prepared to combat the carelessness and the disinclina*- 



/ 



ATTACKS WITH LIMITED OBJECTIVES 17S 

tion for labor that so easily takes hold of troops after a 
success. 

Occupation must permit active defense in depth; all 
crowding is to be avoided; approach and communication 
trenches towards the enemy must be blocked by fire or oth* 
erwise, as the first counter-attacks will come from these. 
Experience has shown that after a successful assault there 
generally is a lack of hand grenades and means of illumina- 
tion. 

Raids are undertaken along the same lines ; the main 
question is that of returning to the exit point after a short 
stay in the hostile trenches. The necessity of these recon- 
naissances in force is for the capture of prisoners; In these 
operations also we may achieve much with weak detach- 
ments. 



VIII. Machine Guns 

MACHINE GUN EQUIPMENT AT THE OUTBREAK OF 

THE WORLD WAR 




176 



MACHINE GUNS 177 

Based on the experiences of the Russo-Japanese War 
all armies had gradually come to about the same number of 
machine guns, viz., 2 guns per battalion and about 2 guns per 
cavalry brigade. In Germany and Russia the machine gun 
platoons had been consolidated into machine gun companies 
of 8 guns, while France and England preferred the platoon 
formation and regulated the tactical consolidation of pla- 
toons according to requirements. In any case, the company 
was enabled to more easily cover greater spaces in depth 
with its 3 platoons than could have been done by single sep- 
arated platoons. Doubtless all experiences gained in war 
speak for tactical employment by platoon, while for admin- 
istration the company formation is better. While Germany 
had decided to transport the guns on wagons, almost all 
other countries favored transportation on pack animals 
(load of pack horse 130 to 150 kg) . The latter method ena- 
bled the machine guns to follow the troops everywhere and 
offered smaller targets, but the amount of ammunition that 
can be carried is limited, and getting into action is retarded 
as gun and tripod have first to be assembled. Should a pack 
animal fall the gun may not get into action at all. Saddle 
sores cannot be avoided no matter what care is taken. And 
finaUy the question depends on the number of pack animals 
available in the country. But in spite of all these obvious 
disadvantages, on the battlefield it is probably preferable 
to transport machine guns on pack animals instead of 
wagons. 

In a technical relation, there were differences in the 
various armies, first in the mounts, second in the manner of 
cooling the barrel by air, or by enclosing the barrel with a 
heavy waterjacket and, third, in the manner of ammunition 
supply either by clip of 25 rounds, or by means of belts hold- 
ing up to 250 rounds. The endeavor of the manufacturers 
had been directed to simplify the arm (System Schwarzlose 
has only one spring, while the breech of the Maxim has 22 
parts) and to decrease the weight so much that it would be 
possible to equip companies and troops with several guns 
carried by the men. At the opening of the World War ex- 
periments had not been completed in use of machine guns in 
airplanes. The question of employment of bicycles had been 



178 DEVELOPMENT OP TACTICS— WORLD WAR 

sufficiently cleared, for instance in Austria in 1913, the cy- 
clist company of 4 J&ger battalions received a fourth pla- 
toon of 4 machine guns. The question of ammunition sup- 
ply had not been solved ; if we did not desire to adopt loading 
clips, as in France, we had to fall back on the time consum- 
ing loading belts with cartridges. It is absolutely necessary 
that the rifle and the machine gun should fire the same am- 
munition, and also use the same way of packing, so that the 
loading clips of the rifle, attached to each other niay be 
readily used by the machine guns. The machine gun gave 
the opportunity of developing the strongest possible infan- 
try fire effect on a narrow space in the shortest time, es- 
peciaUy against upright targets. Considering the possi- 
bility of jamming, the rapid expenditure of ammunition, 
and the difficulties of anmiunition supply the arm was con- 
sidered suited in the first line only for taking advantage of 
short, momentary periods for flank and defensive fire but 
by no means for carrying on a long fire fight. The cone of 
dispersion, itself very narrow, either has no effect at all, or 
a brilliant moral and material effect. This opinion caused 
the machine guns to be posted in dominating or flank posi- 
tions, allowed them to fire through gaps in the fighting front, 
or under strict limitation, to fire over the heads of the front 
lines. If placed in the front lines, machine guns required 
a protective shield to enable them to await the proper time 
for going into action, but the size and shape of the shield 
soon drew the hostile artillery fire. Test showed that even 
on tihe level ground at a range of 1500 meters, with 100-meter 
dispersion in depth, it was practicable to fire over our ad- 
vancing infantry. In firing over the infantry we must re- 
member that in continuous fire dispersion increases and 
finally, when the barrel gets hot, the bullets no longer main- 
tain the same trajectory. It was an acute disadvantage tiiat 
the various positions of the machine guns naturally attracted 
the hostile artillery fire and that information between infan- 
try and machine guns was always difficult. All these condi- 
tions were materially simplified when the machine guns 
were placed in the firing line. In mobile warfare the Ger- 
man machine gun companies adhered to the general rule to 
''support the infantry fight directly'' with all means at 
hand, and they abandoned the maxim that the arm ''is not 



MACHINE GUNS 179 

enabled to carry on a protracted fire fight." They were 
crowded prematurely into the firing line, where the heavy 
weight of the sleds was very troublesome when the mission 
was to keep up with the advancing infantry. Regulations 
requiring them to keep in place during the assault until the 
hostile position had been taken was felt to be onerous, the 
more so as reverses were attributed to their belated arrival. 

In France the light mount was fully utilized for the 
emplo3anent of the gun from effective points, for instance, 
from trees and houses. French tactics did not advocate 
their use' in the firing line, and favored employing the guns 
as a '^nerveless firearm" for defense at close range. Objec- 
tion to air cooling was emphasized on the ground that the 
barrel became red hot after only a j^w hundred rounds, 
making handling and transportation more difficult; that 
heat waves rose above the barrel and prevented aiming, and 
that the dispersion increased continually. We have not heard 
of any cases where the cartridges exploded of themselves ; 
to prevent that orders were issued that no more than 200 
rounds should be fired per minute. The British machine 
guns proved especially serviceable in all respects and fre- 
quently fired over the heads of the troops. In Russia also 
it appears that the machine guns were in the beginning 
looked on as a defensive arm at short range; they found 
little emplo3rment in the offensive, while the regulations in 
defense were that positions could be held by machine guns 
alone. However the idea of using them in separated ''nests" 
soon arose. 

In position warfare the machine gun soon gained special 
value as a flanking arm for the purpose of defeating an 
assault. As it is difficult to hide them in a frontal fight from 
sight and fire, and as this fact was relalized there grew up 
masked, bullet-proof shelters in checkerboard formation, 
where machine guns were concealed prepared to fire only 
when necessity demanded. During the attacks on a large 
scale by the Allies in 1915, placing the machine guns in 
well constructed positions in the front line, even in con- 
crete positions, did not prove of advantage, as the guns gen- 
erally were put out of the fight by heavy artillery drum fire 
even before the attack actually started. From shelter in 



180 DEVELOPMENT OF TACTICS— WORLD WAR 

deep chambers below the ground the guns could not be ra- 
pidly brought into the firing position in time to be of assis- 
tance considering the short distance the attack had to cover. 
After the Autumn battles in the Champagne, the conclusion 
was reached that machine guns had best not be placed in 
well constructed positions in the front line — except a few 
guns designated for flank fire— but should be placed in in- 
conspicuous positions in rear of the front line* The unex- 
pected opening of fire of a single machine gun under effi- 
cient leadership, even if served by only a few cool men, has 
several times been the decisive factor in victory and defeat. 
Good training must overcome malfunctioning of the gun. 
No matter how excited the man may be, the constant readi- 
ness for fire depends on his training. He should be able to 
repair stoppages in about 20 seconds as these are annoying 
and hard to perceive. Only continual practice and training 
can insure that the men quickly perceive the cause of the 
trouble and rapidly repair it. 

The importance of the machine guns in battle was pro- 
perly estimated by friend and foe in position warfare. In 
Grermany single machine gun platoons of 3 guns each were 
organized as early as 1914, and soon combined into com- 
panies. In 1915 each infantry regiment received a second, 
in September 1916, a third machine gun company. Machine 
gun sharpshooter battalions, used by army headquarters at 
the most important battle fronts, formed a material rein- 
forcement of the fighting power; they were posted from 1 
to 3 kilometers from the front line and principally under- 
took the protection of the artillery. In the Spring of 1917 
the number of guns of a company was increased from 6 to 
9 and later to 12 ; but for mobile warfare, companies of 6 
gHns proved better. 

In France, in 1916, considering the difiiculties of bring- 
ing up supplies, battalions of 3 infantry companies and 1 
machine gun company of 8 guns were formed. In addition 
each company received a reinforcement of 8 automatic rifles. 

In England the machine guns, first attached to the bat- 
talions, had been organized into a brigade machine gun com- 
pany of 16 guns. In its place the battalions received 8 to 
12 automatic rifles. The Lewis type gun weighing 12.5 kg, 



BiACHINE GUNS 181 

drum magazine, air-cooled, fire rapidity 300 rounds per min- 
ute, being materially lighter could accompany the troops 
on any kind of terrain. A mobile reserve was formed by 
the machine gun batteries transported on motorcycles. In 
the battle of the Somme the Allies attempted to gain the 
decision by an enormous use of material. A new methed 
was the liberal use of the automatic rifles by the Allies. 
These accompanied the front waves of the attacking troops 
and offered them a rallying point in counter-attacks. We 
could have done the same in Germany, had we been willing 
to do without the sled and utilize the sandbag mount. Num- 
erous automatic rifles were an absolute necessity. Issue of 
these was delayed for a long time, so that the troops in 
many cases used the ones that had been captured. At first 
2 ''Bergmann" guns (model 08.15) were issued to each com- 
pany, these were gradually increased to 6 per company in 
the fighting line. The weight was about 15 kg. A disadvan- 
tage was that it was impossible to fire over high grain or 
grass fields. We must not forget that the automatic rifle 
can perform good service only at ranges from 600 to 800 
meters, provided the troops are weU drilled in handling the 
piece, and that the arm is not suited for firing over the heads 
of our infantry. Firing while in motion may be of advan- 
tage. 

The infantry found it very annoying in the Battle of 
the Somme to be exposed to the machine gun fire of the low 
flying hostile airplanes, against which they were helpless at 
the beginning. The actual effect was not very material, 
but the infantry exaggerated it at first and erroneously as- 
sumed that each airplane was able to get its artillery to im- 
mediately open fire on any desired target, and that the 
pilot could observe any movement on the ground even that 
of an individual. They soon learned their fears were ground- 
less and that they could, with machine guns, fight these 
airplanes successfully. It was especially beneficial that in- 
fantry officers by making trips in the airplane could per- 
sonally convince themselves that the observer could by no 
means see everything, much less hit everything. In any 
case, the danger of attracting the artillery fire was less than 
giving the hostile airplane the opportunity to carry on its 
business undisturbed. During the Winter of 1916-17 the 



182 DEVELOPMENT OF TACTIGS— WORLD WAR 

airplane defense was materially improved by constructing 
anti-aircraft mounts and by the construction of a circular 
front sight for the rifle, which in a very simple manner, in- 
dicated the range at the arrival and departure of the air- 
plane from view (arranged for 800 meters range and 160 
kilometers per hour airplane velocity). Machine gun fire 
is especially annoying to airplanes. They cannot evade it. 
A firing machine gun can be perceived only at a low alti- 
tude, and can be heard only at 400 meters, with the airplane 
motor cut oflF. The activity of the hostile airplanes forced 
us to equip batteries and columns with 2 machine guns each ; 
these guns were also for use at short range fighting. The 
rule for airplane protection is that up to the commencement 
of the infantry battle proper anti-aircraft machine guns fire 
on every airplane within range, while the remainder take 
cover in order not to betray themselves. At the commence- 
ment of the infantry fight the number of hostile airplanes 
will increase, so when each machine gun, not employed 
against the hostile infantry attack, must take part in the 
anti-aircraft defense. 

According to battle experiences the effectiveness of 
machine guns in the first line was a failure as the available 
shelters were unable to withstand the heavy artillery bom- 
bardment« Far better service was performed by the machine 
guns when dispersed in the intermediate terrain, protected 
by weak infantry ; these guns found protection in shell-holes 
or behind the remnants of destroyed trenches where they 
could best be hidden from the airplanes. Thus they evaded 
the systematic artillery fire and had to be destroyed singly 
by the guns accompanying the infantry of the attackers. 

For fighting detached German machine guns the Brit- 
ish Regulations laid down (April 17, 1917) : '^Cautious ad- 
vance under protection of covering detachments which lo- 
cate machine gun nests and call for artiUery iire against 
them. Our own machine guns are to cover the advance of 
the light mortars and infantry. It will be found advisable 
frequently to attract the attention of the German machine 
guns by opening fire with a Lewis gun in a certain direction. 
Attack from the flanks, especially with rifle and smoke shells 
usually leads to the desired residt/' 



MACHINE GUNS IBS 

The emplo3rment of machine guns in embrasures re- 
quired thorough practice in firing over the infantry in the 
front trenches and then gradually lifting the fire over troops 
advancing to the attack in depth formation. In peace time 
the machine gun firing regulations had mitde this procedure 
dependent on a number of limitations, but in the face of the 
enemy we had to proceed with fewer restrictions. Under the 
conditions of the position warfare, firing over the heads of 
our infantry could become a very important factor in attack 
operations. New barrels, that had not yet fired 5000 rounds, 
must be used and water must be replenished after every 500 
rounds. Very close connection with the troops to be fired 
over has to be maintained, and the wishes of the troops must 
be rapidly transmitted. Bullets that strike twigs, grass, etc., 
will be easily deflected and endanger our own men. 

Employment of reserve machine guns leads quite na- 
turally to the employment of indirect fire. Machine gun- 
ners at first satisfied themselves with the simple means of 
'liidden rifie fire,'' with the range ascertained by trial shots, 
from a position out of view of the enemy. Technique of 
machine gun fire soon improved; instead of aiming on tar- 
gets hard to distinguish, we employed auxiliary targets that 
could easily be seen, thereby simplifying fire control and in- 
creasing the effect. As the gun itself was hidden, it was 
not exposed to the hostile artillery fire. 

In the defense it may be advisable to conduct the fight 
with machine guns with small infantry protection, and to 
hold the infantry compames back for counter-attacks. 'Tirst 
let machines produce effect, and only then men.'' If we ex- 
pected in the defense, heavy hostile artillery preparatory 
fire, we would decline the defense of lines, and form the 
machine guns in depth. The battles of 1917 showed the 
necessity of separating tasks for automatic rifies and ma- 
chine guns. For the battle in the firing line the automatic 
rifies were employed as well as for fighting in the interme- 
diate terrain ; the machine guns were for &ing over the at- 
tacking waves against reserves. Depth formation of machine 
guns is best arranged by machine gun nests, surrounded by 
obstacles, with infantry guard under specially selected lead- 
ers, and in close connection with observation wd command 



184 DEVELOPMENT OF TACTIGB— WORLD WAR 

stations of the artiUery. Distributed generally by platoons, 
in checkerboard formation, they offer the possibilily of 
keeping the front position and every point in the interme- 
diate terrain under fire. As fire comes from two or more 
directions the enemy cannot immediately locate it Special 
attention must be given points which our artillery cannot 
reach. Flanking fire was employed as much as possible. 
When these detached machine guns should open fire the 
conunander alone can decide. Not every attack is planned 
for a penetration ; in many instanceis reconnaissance attacks 
on a broad front are undertaken prior to the attack proper 
in order to draw the fire of the machine guns. The effective- 
ness of the machine guns depends on remaining unobserved 
until the main assault. The machine gun nests formed the 
natural raUying points for the defender and facilitated coun- 
ter-attacks with their fire. But their participation was effec- 
tive only if they did not betray themselves on the aerial pho- 
tographs; attention was attracted to them especially by 
their obstacles and through tracks left by the daily move- 
ments. They are best assumed to be in the proximity of 
trenches, ditches, and shelters, in single clumps of woods, 
projecting angles of villages, etc. The more inconspicuous 
the position, the longer life will the machine gun have. Sur- 
prised from all sides by machine gun fire, even well prepared 
attacks make progress with difficulty after the primary suc- 
cess. In many cases it was found the attacker turned about 
or threw himself down, and was finally overcome by the 
counter-attack. The first dangerous enemy to the machine 
gun was the tank. All machine guns have their definite 
tasks (barrage, defense of the zone in depth, support of 
their neighbor, protection of the artillery), which they must 
solve from several prepared positions. 

Machine guns drawn back are by no means to be used 
for the purpose of serving as replacement for guns disabled 
in the front line. These latter must be replaced from a re- 
serve held out by the commander from the beginning. We 
allowed for each machine gun about 8000 cartridges in 
belts and 2 water jackets ; a liberal supply of spare parts and 
grease, in expectation of a gas attack, decreased the danger 
of jams. It is very advisable to construct a shelter for the 
belt filling squad in a position in rear. 



MACHINE GUNS 185 

The fire of the attacker compels the defender to seek 
cover so that the latter no longer offers a target for infan- 
try fire. It is therefore well to work up as close as possible 
to the enemy under the protection of artillery and machine 
gun fire. Machine guns sweeping across the cover of the de- 
fender, keeps him below ground. 

According to ''Training Regulations for Foot Troops'' 
the automatic rifles were distributed to platoons and bore 
the brunt of the fire fight. They wore down the enemy. 
If that was accomplished then, and then only our costly 
infantry materiel was to be employed for a rapid advance. 
It was very favorable if the first firing position of the in- 
fantry was so close to the enemy that from it the assault 
could be delivered. 

In the offensive battle a certain depth formation can- 
not be avoided ; when the automatic rifles had tasks under 
500 meters range, the rifles, model '08, had targets beyond 
that range. The difficulty in reaching favorable firing posi- 
tions in open terrain forced the machine guns to be em- 
ployed at extreme ranges. In many cases nothing remained 
but to put the guns in action early and let them go forward 
with the skirmish lines. If a gun was once inserted in the 
skirmish line, it suffered in changing position in the ad- 
vance by rushes. If the infantry delayed the rush, for the 
slower machine guns, favorable moments were lost. In 
many cases infantry frequently missed the supporting fire 
of the machine guns at the moment of the rush, and the 
guns reached the new firing position later than the skir- 
mishers and thus betrayed themselves. 

An attack method may consist in the infantry working 
up, under the fire protection of machine guns, together with 
automatic rifles to a short distance from the enemy, the 
latter facilitating that working up from advanced ''attack 
nests." 

Frequently from elevated positions in conjunction with 
the artillery, machine guns facilitate the deployment of the 
troops, the advance to the first fire position, or the assault. 
Time for coordination and timely reconnaissance of the 
next firing positions are necessary, It may be well to send 
ahead supplies, water jackets and ammunition even during 



186 DEVELOPMENT OF TACTICS— WORLD WAR 

the flre fight. A portion of the guns may be designated anti- 
aircraft defense. In the attack from trench to trench ma- 
chine guns take over the preparatory fire as soon as the ar- 
tiUery and minenwerf era lift then* fire farther to the front 
They remain in action as long as possible without endanger- 
ing their infantry moving forward to the assault. Direct 
fin is the general rule ; indirect fire may be employed against 
terrain in rear of the enemy. As soon as practicable the 
machine gun companies follow in platoon echelons and assist 
with the fire protection of the infantry. From elevated posi- 
tions, they break up counter-attacks and assist neighboring 
sectora by rapid flanking fire. ''The final purpose of the 
machine guns remains always the same: To assist the 
relentless pushing forward of our infantry by the shock 
power of their fire." 

In open warfare the fire of the machine guns assists 
the infantry, together with the automatic rifles, in reach- 
ing the first firing position. ''Machine gun squads of 1 gun 
with 8 men, a very mobile fire power, form the skeleton of 
the infantry attack.'' Liaison between squads is necessary 
for mutual co-operation and increase of fire power. In the 
attack from trench to trench, automatic rifles are sent ahead 
during the artillery preparatory fire as "attack nests/' At 
the cessation of tibe artillery fire, or when that fire lifts, 
they take over the fire fight and cease only when the firat 
waves of infantry overrun the enemy. There are also auto- 
matic rifles in the first wave which fire at short range dur- 
ing the advance. If guns in the leading waves are disabled, 
they are replaced from the rear. 

The advance of infantry under fire protection demands 
much initiative on the part of all subordinate machine gun 
commandera. They must understand how to concentrate 
fire against points where the enemy still offers resistance, 
and after the successful assault to take over the fire pro- 
tection of the troops that are reestablishing the position. 

In the advance the skirmish squads are in groups or in 
single file ; the gaps resulting can be used to fire through by 
machine guns in rear. Automatic rifles go with their skir- 
mish squads. The line upon which the general fire fight is 
to be taken up can seldom be designated in advance ; if it is 
known, the machine guns kept back must work up to it. 



MACHINE GUNS 



187 



^/ ^7 ^5 ^S ^9 P3 pi p^ 



I 



% 












>8 » 



I 
I 



^ 1^ GvB eoouBMidar 



i9* 






Ghm fominaiidar 



An important element for success is that the machine 
guns succeed in reaching the firing position unperceived; 
machine guns that have been seen are easily annihilated by 
an alert enemy. In the fire fight the automatic rifles will 
fire short bursts, say 5 to 10 rounds, and save the duration 



188 DEVELOPMENT OF TACTICS— WORLD WAR 

fire for specially favorable moments. For the assault the 
fire on the move gains special importance, although the ac- 
tual effect of this fire was bdow expectations. 

At the time of the large offensive in March, 1918, equip- 
ment of divisions in machine-guns, were as follows: 

Germany: 9 battalions to the division with 9 madiine gon eom- 
l»niei of 6 irons (12 ganB) a total of 64 (lOS) machine gona and 144 
antomatie nflee. 

France: 9 battalione to the diyision, with 9 madiine gon com- 
IMaiet of 8 (12) guns, a total of 72 (108) machine gans and 216 
automatic riiks, 8 (12) per company. 

Great Britain: 12 battalions to the division; 8 Brigade machine 
gun companies of 16 gons— 48 and 1 division machine gon company 
with 16 gons a total of 64 machine gons and 192 aotomatic rifles, 4 
per company. 

Italy: 12 battalions to the division. Each battalion had 1 madiine 
gon company of 6 gons, a total of 72 machine gons (additions had 
been arranged for), each battalion 1 to 2 plato<ms, 24 to 48 aotomatic 
rifles. 

United States: 8 battalions to the division. Each regiment of 
infantry. 1 machine gon company of 12 gons; each brigade 1 bat- 
talimi of 4 machine gon companies and the division, 1 battalion of 2 
motoriaed machine gon companies with 12 gons each. This makes a 
total of 168 machine gons and 768 aotomatic rifles. 16 aotomatic rifles 






IX. The In^try Attack in Open Warfare 

REQUIREMENTS 

Fighting methods are governed by the technical pro- 
gress of arms and armament and their resulting moral ef- 
fect on the troops. Both have to be considered in training 
troops and in the conduct of battle, if we want to be equal 
to the demands of war and safeguard ourselves against dis- 
appointments. France underestimated the effect of the 
German infantry and machine gun fire, while we were sur- 
prised by the effect of the French field artillery. At the 
opening of the World War the effect of the flat-trajectory 
fire of all arms had been increased to its fullest eictent, but 
the development of high-angle fire had been neglected. Only 
Germany and England had sufficient high-angle fire artil- 
lery, while gas projectiles were unknown. At the conclu- 
sion of the World War we see the demand that infantry 
and artillery must be equipped for high-angle and flat- 
trajectory fire, to force the enemy into his shelters and still 
reach him at short range. 

The French infantry was the first to realize the pos- 
sibilities of short range, high-angle fire; in 1916 it intro- 
duced the automatic rifle and the rifle grenade, but confined 
the equipment with hand grenades to but a few individuals, 
while we equipped aU our men witii both hand and rifle 
grenades. Since the Franco-Prussian War the number of 
batteries of an infantry division has at least trebled and 
the other arms have materially increased, it is clear that 
the influence of firearms on the fighting of infantry is, at 
the present time, far different than was the case in the 
campaign against France in 1870-71. This appeared most 
plainly during attacks in position warfare. Divisions had 
more artillery at their disposal than in mobile warfare and 
the question was not one of covering with fire a wide front, 
but only of a narrow sector with the fire lifted from trench 
to trench. When the infantry attacks in mobile warfare, 
its road is opened by artillery and machine guns which 

189 



190 DEVELOPMENT OF TACTICS— WORLD WAR 

force fhe enemy to take cover. The result is the attacking 
infantry sees no target. Prior to the war, this matter had 
been carefully considered in France; there appeared doubt 
that fire superiority could be attained as French artillery 
had no high-angle guns, and the hostile infantry, in trenches, 
could not of course be reached by rifle and shrapnel. On 
the other hand it was believed that their field artillery could 
overcome any halt forced on the troops by hostile fire. It 
was held that the infantry would not fire except when ab- 
solutely necessary; it was to advance under the protection 
of its artillery without firing a shot, keep on the move and 
close with the enemy. In Germany, the infantry was cau- 
tioned to never wait for our own fire superiority, which as 
had been learned no one could perceive. The infantry 
should far rather compel the enemy, by its advance, to 
occupy his position thereby offering shrapnel targets to our 
artillery. This did not happen at all times, however, as 
the main role in defense was not left to the infantry but to 
the artillery. The attacking artillery must finally lift its 
fire from the defender so as not to endanger its own ad- 
vancing infantry. In Germany we attached special impor- 
tance to ''carrying forward the fire to the enemy,'' in case 
of need to the closest possible range. Thereby sanguinary 
hand-to-hand fights resulted before a final decision, in which 
the German infantry finally gathered the fruits of its thor- 
ough training on the rifie range, though under heavy losses. 
On the other hand, it suffered out of all proportion, if it 
attempted to gain the assault position without waiting for 
fire preparation. 

The Russian infantry gained the same experiences in 
the Russo-Japanese War. The bayonet training preached 
by Dragomiroff had its roots in the belief of the moral in- 
fiuence of the bayonet, as well as in the knowledge of the 
poor marksmanship of the troops ; thus it came about that 
(with the intention of getting up to the enemy as rapidly 
as possible) the infantry pushed forward witiiout proper 
fire preparation. Fire success attained by infantry and ar- 
tillery, are due to better target practice, fire control, fire 
discipline, better cover, and the concentration of a larger 
number of projectiles on the decisive point. Thus, the ene- 



; 



INFANTRY ATTACK IN OPEN WARFARE 191 

xn/s resistance is finally overcome, he no longer dares to 
raise his head above ground, even if the fire temporarily 
abates, and our movements are no longer interrupted by 
fire. With the defender, individual men attempt to escape, 
followed by entire squads, while with the attacker, active 
groups encouraged by small losses, attempt by quick rushes 
to get ahead. We then speak of the fire superiority of the 
attacker. 

The attacker should always feel out the fire superiori- 
ty when he attempts to advance, and ascertain if the ene- 
my is able to prevent his advance by its fire. Fire super- 
iority can generally be gained only through a costly fire 
fight lasting for hours. Machine guns hurl, with less 
guns and less men, the sione number of bullets upon the 
enemy in shorter time and with greater accuracy, while the 
number of casualties decrease due to fewer men in the fir- 
ing line. Thus, we can replace the infantry fire at the 
shortest ranges by automatic rifles, and at mid and long 
ranges by machine guns. For instance, one infantry regi- 
ment could insert on a 1,000-meter front 20 automatic rifles 
and 18 machine guns throwing about 10,000 projectiles per 
minute. The main fire power of the infantry platoon lies 
in the automatic rifle, its shock power in the skirmish squads. 
The endeavor to preserve the power of the skirmish squads 
as long as possible for the decisive shock, leads to the fire 
fight being conducted principally by automatic rifles and 
machine guns. As a general rule we must gain a point as 
close to the enemy as possble before opening Are. We must 
demand of well trained troops that they open fire only when 
special effect can be secured from use of rifle and machine 
gun. This will be the case in open terrain at ranges less 
than 1,000 meters. Our new tactics for the attack seeks 
rapid decision, with a minimum use of shock-tactics. It 
seeks to save men. Maximum use of machines at the start 
before throwing into the battle the very costly human ele- 
ment, so hard to replace, is the rule. But the decision pro- 
per, as always, lies in the hands of the infantry, whose phy- 
sical endurance and moral power must be increased to the 
highest possible point by all available means. This battle 
method has been prov^ correct. It demands thorough 
training in the co-operation of all arms. 



192 DEVELOPMENT OF TACTICS— WORLD WAR 

Besides our own fire effect utilization of the terrain has 
a special importance in working up to the enemy. Battle 
reconnaissance must ascertain the difficulties of deplojonent 
from any cover in the terrain. Utilization of cover pre- 
supposes flexibility of the skirmish line, both at a halt and 
moving forward. The whole idea is wrong if the demand is 
made to cover the entire terrain equally with skirmishers 
under any and all conditions. 

PROCEDURE OF ATTACK 

Our training regulations distinguish between rencontre 
attacks, against an enemy deployed for defense, and against 
a fortified position. We shall hereafter adhere to these 
distinctions. Air reconnaissance may be able to procure 
information as to what degree the defender has made pre- 
paration for defense. We may alwasns be sure that trenches, 
and false works, have been established in a greater or lesser 
degree of completeness. The construction of extensive ob- 
stacles and the presence of numerous sheltering works 
(which it takes weeks to construct) made a ''fortified posi- 
tion,'' the capture of which demands reinforcement and in- 
crease of artillay and technical troops. The German Army 
has al wasrs preferred the attack knowing that by the attack 
alone can a victory be won. Even situations which strate- 
gically force defense upon the troops can be solved tacticaUy, 
in many cases, better by attack than by defense. The de- 
cision whether the commander shall attack or defend does 
not, as is the case so frequently in peace maneuvers, lie in 
the relation of strength of the two forces. We hardly ever 
know at the moment of having to make a decision, anything 
but the general situation and the mission. The enemy is in 
the same uncertainty as we. A sharp quick attack may 
convince him that we are stronger. Frederick the Great 
cautioned his troops, which in almost all instances had to 
fight against superior numbers : "Attaquez done toujaursJ' 

Troops trained in the will to attack will always be able 
to attack. Not the number of rifles is necessary for the 
success, but the fighting power gained in training, the 
equipment, the skill of leaders and troops, the co-operation 



INFANTRY ATTACK IN OPEN WARFARE 108 

of the arms, the utilization of the terrain, and rapidity and 
decision in the fight. Attacks, without fire support, are 
useful only in surprise situations or occasionally at night. 
The normal method consists in combining fire and move- 
ment. It is the task of regulations to point out the correct 
relation between fire tactics and shock tactics. Infantry 
regulations state : 'The best means for working up to the 
enemy is by securing fire superiority,'' which is made known 
by the decrease of the hostile fire or the enemy's bullets 
flying high and wild. Field Training Regulations presents 
another viewpoint based on experiences in war. ''Rifles 
and machine guns bear the brunt of the fire fight; the de- 
cision is brought about by the shock power of the infantry," 
and "Opening and carrying on the fire fight will generally 
be left to machine guns; tiiey must make the enemy take 
cover, and then only will our infantry appear in the intre- 
pid attack." First, the effect of machines, then insert men. 

Par. 324, I. D. R.: "The attadc consists in carrying 
forward the fire." It is demanded of well trained troops 
that they open fire, even in terrain devoid of cover, only 
at mid range (800 to 1200 meters). Today, when all 
means of fire have been improved and augmented, we must 
demand more. The infantry must work up without firing 
a shot so close to the enemy that the fire will be opened as 
late as possible — ^in any case less than 1,000 meters. The 
time to open fire is when the troops can work forward under 
the hostile fire only with great loss. The front of a divi- 
sion sector will hardly ever be less than 2 kilometers, which 
will, of course, not be covered equally by skirmishers. In- 
fantrymen may be replaced without loss of fire power by 
machines (guns, machine guns) , etc. An attack that is to 
be carried deep into the enemy requires formation in depth. 
This cannot be secured if the front is too wide. If too wide 
nothing remains but to make a portion of the division sec- 
tor strong, place the center of gravity there and penetrate 
deep into the enemy, while other portions of the sector have 
to be satisfied with near objectives. The enemy which was 
threatened only, and not directly attacked, will later on f aU 
upon the flanks of the penetrating attack. 



194 DEVELOPMENT OF TACTICS— WORLD WAR 

NEAR RECONNAISSANCE 

Troops approach in march column the effective zone of 
the hostile fire, and without having to face lonsr range bat- 
teries, we must count on the route being fired upon when 
within 10 kilometers of the front. This requires the forming 
of the march column into separate elements, ''Deployment/' 
In view of the modem long range guns it may easily hap- 
pen that the enemy's projectiles will arrive sooner than the 
first reports of the reconnoitering units ; thus, the infantry 
will have to feel its way after it reaches the fighting zone. 
In no case must troops enter the hostile zone of fire in close 
order. 

All arms participate in battle reconnaissance. As 
shown by experience, after the first reports have been re- 
ceived concerning the enemy, there will be for a time a lack 
of information which may be explained by the cavalry giv- 
ing way towards a wing in the face of the enemy, and by the 
other arms pushing out reconnoitering parties. We must 
also not forget that a column marching along a road is more 
easily distinguished than troops advancing deployed. While 
patrols can overlook entire hostile columns advancing along 
roads, they are when observing deployed troops confined to 
only small portions, and this observation will be made more 
difficult by the opponent's cavalry which after deployment, 
can protect his fianks more thoroughly than was possible 
with the long march column. Should the patrols neverttie- 
less succeed in getting through the opponent's cavalry to 
their infantry they will in most cases only be able to make a 
hasty long range reconnaissance as they will now be afraid 
of the hostile infantry and of being discovered in a precar- 
ious situation. What happens between the wings of the op- 
ponent they can neither see nor report. Natural cover will 
increase the difficulties of view. Observation of the imme- 
diate front can of course furnish local information only. Pa- 
trols may ride far enough to the front to be fired on, and 
report that fire, but what hostile troops fired and the exact 
direction from which the fire came they will, in most cases, 
be unable to correctly ascertain, as the troops firing are 
hidden and the smokeless powder conceals the direction from 
which the shots came. 



INFANTRY ATTACK IN OPEN WARFARE 195 

There must be no interruption in the reconnaissance, 
and it must be carried on with great care : not however, to 
excess so that it may delay the battle action, and thus ren- 
der questionable the final success. Only cavalry can deter- 
mine the position of the enemy along extensive lines, ascer- 
tain extension of wings, location and movement of reserves, 
and approach of probable reinforcements. This while im- 
portant lacks the details so necessary for the conduct of the 
infantry battle. Details of the hostile position can be gained 
only if the enemy is careless or unskillful; false and ad- 
vanced positions can be discovered only during the course of 
the battle. Terrain reconnaissance will include the fol- 
lowing points : Up to what point may the troops move un- 
der cover and in what formation? Where does the terrain 
allow cover for portions of the troops only ? Where will be 
the first firing position and what is its nature ? Where are 
the supporting points for the attack located? Where are 
dead angles in front of the line to be attacked? What are 
the difficulties to be encountered in passing over the field of 
battle? 

A simple sketch showing the above points will help the 
leader to think himself into the situation. Concerning the 
actual view of the hostile position a good sketch will be val- 
uable. Airplanes will generally have much other work to 
do to meet the requirements of the high command. Artillery 
patrols, usually sent out only when definite reports have 
been received of the presence or advance of the enemy, must 
ascertain what is necessary to know for the artillery. 
Strength, deployment and location of the hostile artillery, 
as was demanded at the opening of the war, they will be 
able to ascertain only in exceptional cases. This had best 
be left to aerial reconnaissance. Location of hostile obser- 
vation positions is especially valuable. It is not correct to 
charge artillery patrols with reconnoitering roads and firing 
positions ; these patrols must not allow themselves to be di- 
verted into such general reconnoitering tasks, or to help 
other arms. On the other hand, artillery patrols must es- 
tablish connection with the reconnoitering units of the 
other arms and exchange information with them. 

The infantry must be called on in good time for near 
reconnaissance, as well as for the usual battle reconnaia- 



196 DEVELOPMENT OF TACTICS— WORLD WAR ' 

sance. The main difficulty for the infantry lies in the fact 
that their cyclists are bound to the roads, that the mounted 
officers must be temporarily withdrawn from their proper 
duties, and that there is a lack of men to accompany the 
officer's patrols to send back messages. But this should not 
preclude (it must even be demanded) that officers ride to 
points permitting a good view and observe through field 
glasses. But if they personally ride back to report, the re- 
connaissance is interrupted, a great disadvantage. Infan- 
try patrols move slowly and send in their reports more 
slowly which, before reaching the desired point, may have 
been overtaken by events. But, on the other hand, with 
correct use of cover they can get very close to the enemy un- 
observed. At the same time they possess a certain fighting 
power especially when machine guns are attached. 

Near reconnaissance at the present time is of more 
importance than ever before. Changes of front with de- 
ployed lines is impossible and troops coming under hostile 
fire in close order, suffer losses amounting almost to anni- 
hilation. The reconnaissance becomes more difficult as we 
have to reckon with the enemy's covering forces and we may 
be sure he wiU not voluntarily permit us to get a view from 
high points. Weak infantry can neither break the hostile 
covering forces nor force them back sufficiently for the 
leader to reconnoiter personally. In many cases strong 
forces will be required. From these reconnaissance patrols 
we come to the reconnaissances in force. Unless we threaten 
an attack, the defender will not uncover his positions. Re- 
connaissances in force must be resorted to, because we can 
immediately make use of its results. 

Infantry reconnaissance must be carried on with great 
care, proper arrangements must be made during the march, 
distribution of tasks, and plans for sending messages back 
must be thought out. If we for instance send infantry pa- 
trols with automatic rifles along the march route and along 
neighboring roads as soon as the first cavalry reports of 
the enemy are received, these patrols will offer a certain 
stability to our cavalry patrols, while they increase the 
enemy's difficulties of reconnaissance. The leader had best 
reconnoiter personally and not rely entirely on the tactical 



INFANTRY ATTACK IN OPEN WABFABE 197 



knowledge of his subordinates. He must, under ii 
protection, see for himself and then he can the better use 
his troops. Sending out company commanders for reconr 
naissance which is much done in peace maneuvers, should be 
avoided. Exaggerated thoroughness, especially on short 
winter days, may delay the reconnaissance so much that 
the success of the attack will become questionable. It would 
be entirely wrong and our attacking power would be weak- 
ened if we delayed the decision for the attack until we have 
full knowledge concerning the enemy's strength and group- 
ing. Clearness in this regard can be attained only after the 
battle. The general situation in the first line decides the 
question of attack. The situation may force immediate ac- 
tion without awaiting the results of the reconnaissance* 
Conditions of the terrain and degree of known preparations 
of the enemy will govern the leader in deciding whether to 
attack at once, to await darkness to bring up the troops, or 
to maneuver the opponent out of his position. 

DEPLOYMENT AND DEVELOPMENT 

The near reconnaissance is supplemented by systematic 
observation. From favorable observation points, by means 
of good field glasses, the distribution of the enemy and his 
routes of approach can often be observed. The results of 
such observations are collected first at the regiments and 
form the basis for battle orders. 

Within the division regiments receive a battle sector 
of from 600 to 1000 meteris. The regimental commander 
equally divides his sector into battalion sectors of about 300 
to 600 meters. Each battalion receives with its battle sec- 
tor its fighting mission. Regimental orders cover, connec- 
tion with the artillery, arrangement of means of informa- 
tion, orders for the combat train, and for the regimental aid 
station. 

Generally the battalion will assign 2 companies for the 
front line, each on a 200 meters front, the other 2 follow in 
the second line. The machine gun company remains, as a 
general rule, with the battalion, but may temporarily be 
employed otherwise: ''As support for special tasks, fiank 
protection, independent support of the firing line from dom- 
inating positions, etc.'* 



198 DEVELOPMENT OF TACTIGS— WORLD WAR 

In addition the regiment disposes of its minenwerf er 
comiMmy and its information detachment. The liberal sup- 
ply of means of information furnished during the Worid War 
permitted tdephone connection from division headquarters 
to the division artQlery commander, from there to the ar- 
tillery hattalions, and from division headquarters via bri- 
gade headquarters to the infantry regiments and from the 
latter to their battalions. The possibility of division head- 
quarters conversing directly with different battalions must 
never lead to its interference with the detailed conduct of 
the same. 

The development of the attacker must be made under 
the assumption that heavy artillery fire may start at any 
moment, and the larger his force the sooner will the enemy 
open fire. The utilization of the cover offered by terrain, 
assuming formations calculated to lessen the effect of fire, 
timely removal of all vehicles from the column, and Uieir 
movement from cover to cover, are the best means of avoid- 
ing the hostUe artillery fire. Of course, this covered ap- 
proach takes time, but it saves loss. In a rencontre battle 
such a loss of time may be fatal, but sometlung will have 
to be hazarded. In the attack against an enemy deployed for 
defense, time consumed has not ttie same importance as loss 
in men. 

Narrow formations — (the French even reject march 
column of fours on the battlefield)— of little depth, irregu- 
larly dispersed over the terrain, have proved best in cross- 
ing ground swept by the hostile fijre. French and British 
almost always divided their platoons into 4 separate squads 
and lessened the losses. In general we adhere to division of 
platoons into half -platoons. However, when the hostile fire 
becomes effective a further division should be made, with 
due consideration of loss of time and control. But if the 
advance is not to be completely disrupted, a line has to be 
designated. This line should be under cover and at close 
range. The class of artillery fire will in each case demand 
a special procedure. If the hostile artiUery seeks to find 
the range to a unit with some short or over shorts, that unit 
will at once, lessen the effect of the expected fire by deploy- 
ing into smaller and narrower units, dispersing over the 



INFANTRY ATTACK IN OPEN WABFARE 199 

terrain, or by rapid backward or forward movements. In 
many cases the enemy artiUery will find the range to some 
point in the terrain, one task of the near reconnaissance is 
to locate those points and warn troops not to stop near them, 
but pass quickly by without offering large targets. Sweep- 
ing fire is best avoided after exact observation of the spaces 
that are comparatively free from the fire, and those spaces 
crossed in close order, the commander keeping his troops well 
in hand. Much depends on luck. It is far more difficult to 
move forward and so observe counter preparation fire ; a bat- 
talion that is struck by such fire can do nothing better than 
to lie down and wait, fail to give the enemy any guide for 
regulating his fire and avoiding conspicuous appearances 
such as horses, standing (rfficers, etc. Meanwhile units more 
favorably situated continue the forward movement. Assum- 
ing that the hostile batteries are ready for action and await- 
ing the appearance of favorable targets, we must never offer 
large targets but attempt to gain ground by squads or in- 
dividuals. Detours are frequently correct. While in the 
next cover the troops are reassembled and an attempt made 
to continue the movement. By paying due regard to these 
experiences we have been able to pass through heavy artil- 
lery fire without material losses. 

ARTILLERY OF THE ATTACK 

Simultaneously with the infantry advance the attack- 
ing artillery from covered positions fire at a range of about 
3000 to 4000 meters. Its fire no longer com/es from long, 
inflexible lines, which make the firing comparatively easy, 
but from groups, hugging the conformation of the ground 
which can be located by airplanes only. This echelonment 
of guns has an advantage for infantry and artillery in that 
infantry no longer crosses the line of its artillery support 
at one time. A consequence of the old method of lines of 
batteries was that the defender succeeded in a very short 
time in ascertaining the position of the hostile artillery, and 
could apply sweeping fire within relatively small zones thus 
effectively carrying on the artillery duel. One saw from the 
point where noise of cannon came» that thunder suddenly 
cease, and at the same time skirmishers appeared working 



200 DEVELOPlfENT OF TACTICS— WORLD WAR 

forward by rushes, or several columns with exact intervals 
came over the rise and the hostile artillery again resumed 
the fire. 

Even though valuable time may be lost and another 
unit's battle zone temporarily invaded it is better to go round 
than to cross a fire swept zone. 

A portion of the artillery will hold down the hostile 
artiUery, while as many guns as possible direct their fire on 
the hostile infantry. The danger must never be underesti- 
mated that at long ranges there is doubt concerning the 
situation of the hostile position, which can be cleared up 
gradually only by infantry observers and the artillery liai- 
son oflScers who have gone ahead with the infantry. 

Defensive artillery should not reply to the guns of the 
attack. Its target is the hostile attacking infantry; only 
when that target cannot be fired on will it fire on the hostUe 



Whether the attacking artillery succeed in holding 
down the defender's artillery depends on how exactly the 
batteries of the defender have been located. The outcome 
of the artillery duel is doubtful, but it is absolutely neces- 
sary for the success of the battle. Thus, the underlsring 
teachings of earlier regulations to seek with all means at 
hand to gain the fire superiority, before starting the infan- 
try to the attack could not be discarded. Even if the artil- 
lery duel has not been absolutely decided in our favor, the 
infantry must not decline the execution of the attack as an 
impossibility. 

How many batteries of the attacker may fire on the de- 
fensive artillery depends upon the effect of the latter. If 
the defender's artillery can materially increase the difficul- 
ties of advance for the attacker, then he must place a por- 
tion of his guns, which have heretofore fired on the enemy's 
infantry, against the artillery ; the task of such guns may 
be then taken over by machine guns. It cannot be avoided 
at times that a portion of the defensive artillery is not fired 
on and can direct its fire on the advancing skirmishers. 

WORKING UP TO THE ENEMY 

We must lay down the following maxims for the at- 
tack: 



INFANTBY ATTACK IN OPEN WABFABE 201 

1. The attacking infantry advances under the protection of 
its artillery and machine gvai fire as dose to the enemy as the 
hostile fire permits; it would be simpler if the assault could be 
taken up from the first halt, as was laid down in Troop Training, 
but which was found to be not always feasible. 

2. The defensive artillery attempts to drive off the skirmishers 
of the attack; the attacking artillery fires on the d^ensive ar- 
tillery only with as many guns as is necessary to hold down effec- 
tively the latter's fire. 

8. Attacking artillery will finallv be forced to enter into a 
time and ammunition consuming artiUery duel (to relieve the at> 
tackin^r infantry from the defender's artillery fire) and the 
attackmg infantry will be forced into an inf ant^r fire fight 

The main fire power of the infantry platoon lies in the 
light minenwerfers. We have thoroughly proven during 
the war that the machine gun is an excellent arm of oppor- 
tunity but not suited for a long continued fire fight. The 
increase of machine guns in all armies, has made move- 
ments in long, regular, dense skirmish lines impossible; 
thereby finally and definitely answering the question whether 
it is correct to make long and broad or short and narrow 
rushes. The former are far too costly under machine gun 
fire, not even considering the fact that a platoon in a thin 
skirmish line cannot be controlled any longer by its com- 
mander. It is different in situations where we are not 
directly under machine gun fire, for if the troops can get 
rapidly forward, long and broad rushes are in order. Short 
rushes by no means shorten the time during which troops 
are visible; they only increase the enemy's difficulties in 
directing a well aimed fire on the skirmishers. Under pro- 
tection of our own effective fire we can make each rush long, 
provided we do not prefer to advance at the walk. For train- 
ing reasons we will generally adhere to long rushes with 
great breadth; they are executed as soon as the effect of 
the fire permits. 

We, as well as our enemies, saw ourselves in the World 
War forced to make our skirmish lines thinner and thin- 
ner. The 176th Infantry Regiment adopted the interval 
between skirmishers at two meters, while, based on ex- 
periences of the Boer War, the British adopted in the fire 
deployment an interval ranging between 5 and 15 meters^ 
which shortly before the assault could be diminished to 
0.90 meters. In Russia, based on the Experiences of ilie 
Russo-Japanese War, it was thought that one rifle to every 



202 DEVELOPMENT OF TACTICS— WORLD WAR 

4 meters (2.6 to 3 meters interval) was sufficient. Train- 
ing Regulations required an interval of 6 paces, which could 
be diminished according to need to one pace. France perma- 
nently favored the formation of dense skirmish lines. 

The more the intervals were enlarged, the more diffi- 
cult became control, as the individual skirmishers were left 
more to themselves. But even the thin skirmish lines were 
exposed to heavy loss when under machine gun fire, and the 
machine gun carriers in the skirmish lines could be easily 
distinguished. It was therefore found advisable to make 
the lines irregular especially shortly before the opening of 
fire. Advance by squads in single file with intervals or by 
skirmish squads with little intervals, but with gaps between 
them and neighboring squads, offer many advantages. The 
squad leader can make his infiuence felt on his men from 
the time fire is opened. Gaps between squads permits ma- 
chine guns to fire through them, disruption of squads in 
joining the line is avoided, and the squad leader fights with 
the men whom he has trained for battle. Small, irregular 
targets permit the infantry to get forward without material 
loss. 

We should not demand regular formations within the 
platoon : One portion is more favored, the other less. The 
formation must be such that it will not unnecessarily iii- 
crease the losses if hostile fire is suddenly encountered. To 
take up the desired formation only at that moment is en- 
tirely too late and will lead to annihilation. In terrain 
where a good view cannot be had, scouts must be sent ahead, 
and even in open terrain it will often be advantageous for 
the platoon commander to go far to the front with a few 
scouts. The advance is made at the walk as long as prac- 
ticable, with the firm will to open fire as late as possible. If 
broad stretches, without cover, have to be crossed the hos- 
tile fire ^ect may compel formation into the smallest units 
(squads) , which gain ground either by thin skirmish lines 
following each other in waves, or single file formation 
abreast with large intervals. This formation must how- 
ever not interfere with combined leadership. In addition, 
direction of attack and guide must be definitely laid down 
by the platoon commander and strictly adhered to by all 
units. Formation and method of movement are left to the 



EXAMPLE OF THE FORMATION OF Olj 

OPENING OF THE FIB] 

2d Platoon 



f 



I 









05# 






! 









Musician' 



••••1 



* 



OJV». 



Mus 



1. 






! 



3(1 Platoon 



iXmm«»* ••••• 






Company Commander 

Platoon commander 

Squad leader 

Supernumerary squad leader 

Musician 

Automatic machine gim 

(1) The interval may be materially 1 



204 

di£ 
mj 
co- 
st? 
at( 
off 

by 
mi 

an- 

Ee 

bU 

if 

fai 

wl 

po 
8h< 
mi 
for 
mi 

SU} 

po 

ini 

foJ 

th 

au' 

inf; 

Un 

gr< 

for* 

ca] 

hii 

po 

inft 

by 

to 

na' 

uni 

th< 

vie 



■ I 



DEVELOPMENT OF TACTICS— WORLD WAR 



scretion of the squad leaders. Only effective hostile fire 
(ist induce the individual soldier to work forward from 
ver to cover. The troops do not like to crawl for longer 
retches and the expenditure of energy is not commensur- 
3 with the lessening of losses, as the crawling skirmisher 
ers a larger target than the man lying down. To cover 

crawling 100 meters in favorable terrain will take 2i 
nutes, and in addition, it is difficult to preserve order 
d cohesion and to keep to the general march direction, 
itirdy different is the occasional use of crawling, for in- 
ince, in entering a position. The advance will be aided 
we succeed in getting a portion of the machine guns into 
irorable, inconspicuous firing positions, ^'attack nests'' from 
dch they facilitate the advance at the start and then sup- 
rt it by firing over the heads of the troops. The infantry 
Duld never lack the support of machine guns, and nothing 
ist be allowed to interfere with bringing the machine guns 
n^ard. It is therefore advisable to let a portion of the 
u^hine guns at the very start follow the infantry for fire 
pport at short range and to move the other part up from 
sition to position. Closely hugging the ground, appear- . 
X suddenly, skillfully evading the hostile artillery fire, they 
•low the fighting line in echelon. They also keep in mind 
3 task of facilitating the advance of the skirmishers and 
^tomatic rifles. We must avoid, as long as possible, plac- 
t machine guns in the skirmish line. They crowd that 
e badly, and increase the losses and in addition, it is a 
3at disadvantage to have to take much time to get ready 
r a rush. Guns must be unloaded first, and as the carriers 
inot rush forward like the skirmishers ; they remain he- 
ld and as a result draw the hostile fire on this most im- 
rtant fighting arm. 

The fire fight is conducted by the skirmish line accord- 
g to well known rules. The opening of fire will be ordered 

the platoon commander only when there is no other way 
get forward. Target and sight elevation will be desig- 
ted by the platoon commander as long as possible, but 
tder heavy fire the control will slip from his hands and 
sn the squad leader takes his place and finally the indi- 
lual soldier. Lioss of fire control does not matter very 



•\ 



INFANTRY ATTACK IN OPEN WARFARE 206 

much at short range if through training the skirmisher care- 
fully observes his fire effect, husbands his ammunition, in- 
creases his rate of fire, when the target is favorable, de- 
creases it when the target becomes unfavorable or entirely 
stops firing when the target disappears. The command ''fire 
faster," or 'fire slower," only emphasizes the inattention or 
bad training of skirmishers. 

Light minenwerf ers must work their way forward in 
the same independent manner as machine guns. Positions 
should be selected within 600 meters of the hostile position, 
so that they can use their short range weapons effectively 
during the assault. Prematurely taking up a position is 
correct only in a reverse ; support had best be left to the ar- 
tillery. Their mission is annihilation of observed machine 
guns or guns placed in readiness to defeat the assault. It 
will almost always be advisable to attach them by pairs to 
the leading companies. When going into position and when 
ammunition is being brought up for them they offer large 
targets; therefore cover for them gains in importance. 

Co-operation between infantry and artillery previously 
arranged for is put to the test on the demand of the infan- 
try when entering battle. Machine guns seen by the infan- 
try are best and more quickly annihilated by artiUery, but 
it will not be easy to designate the target to the artillery in 
rear. If the advance of the infantry is delayed or prevented 
by hostile fire, it can be resumed best by waves. The infan- 
try must be trained to inmiediately utilize fire support of 
the sister arm, which means infantry must immediately ad- 
vance when it sees our shells bursting on the enemy. The 
advance is not to be made in rushes but at a walk, when 
the enemy is blinded by dust, smoke or iron. The infantry 
has often been able to push on when single guns or artil- 
lery platoons came up close and supported its advance, pay- 
ing no attention to losses. There is no reason to expect espe- 
cial danger to these accompanying guns, as the opponent in 
defeating an assault has more important things to attend 
to than to bother about them. Accompanying guns should 
always be under the orders of the infantry commander. Ar- 
tillery follows up the infantry after the capture of the posi- 
tion, and secure its possession. A well planned, energetic 



206 DEVELOPMENT OF TACTICS-WORLD WAR 

counter-attack made after the capture of the position will 
hardly go to pieces under artillery fire which comes from 
far in rear, and the observation of which by the advance 
troops is very difficult, especially in close country. The ar- 
tillery at long range, limbering up at that very moment 
leaves to the infantry the entire difficulties of meeting the 
eountar-attack. Retreating infantry finds stability on^ 
at its first guns and is enabled to midce a stand there. We 
must not underestimate the moral effect which is made on 
the infantry by guns following up dose on its heels. 

Hie infantry guns either Iqr piece or by platoon, sup- 
port the troops directiy at an effective range of 1200 to 
1800 meters. These guns fire point Uank which demands 
that every shot be observed; the difficulty of observation 
requires employment of single guns. The battery is merely 
the unit for supply, discipline, etc. 

During the entire war the question of employing the 
spade in the attack was hotly discussed. During the Russo- 
Japanese War, the Japanese as a general rule fortified their 
first fire position, and resorted to the spade again in the 
subsequent advance. The cause is probably to be found in 
the fact that the Russian fire was of littie account of itself, 
that the Japanese probably could have advanced further, 
but their disinclination to suffer loss caused them to ad- 
vance slower. The Russians did not follow the Japanese ex- 
ample. In the first regulations, based on the experiences 
of the war, the Russians demanded intrenching in every 
firing position, even in the assault position, while the Jap- 
anese army was warned against the too frequent use of the 
spade. The German infantry never relished the use of the 
spade, realizing that gaining time is more advantageous to 
the defender than to the attacker. The difficulty of bring- 
ing a skirmish line that has just dug itself in under effective 
lire by hard labor, to again resume the advance is very diffi- 
cult, hence caution is necessary in the use of the spade. 
Never must construction of cover weaken the will to attack. 
Troops that entrench during an attack must lay aside one- 
half of their rifles ; even fire superiority, difficult to attain, 
cannot offset this loss. The man at work entrenching offers 
a larger target, the freshly turned ground is an easy aim- 



w 

i 



INFANTRY ATTACK IN OPEN WARFARE 207 

ing mark, and the cover obtained is insufficient. U the 
cover is such as to provide actual protection, much time is 
required, which the opponent will use to good advantage. 
Thus a fire superiority that has been gained with difficulty, 
can easily be lost. We must not forget that the steel hel- 
met offers protection at long range against rifle fire and 
thus lessens losses better than earthwork. If the hostile 
fire is actually so powerful that there can be no thought of 
a further advance, it is, in any case, better for the troops to 
defend themselves at all hazards with their rifles, than to 
endeavor to intrench. On the other hand if the enemy is so 
shaken that the loss of half the rifles is of no moment, we do 
not need to remain lying, but can, in most cases, close with 
him. Filled sandbags will, in most instances, give better 
protection than hastily constructed trenches. Carrying 
along filled sandbags, which retard all movements niay be 
justifiable in position warfare, when the ground is rocky, or 
when it is frozen hard. 

The spade is used in attack only when the question is 
to hold captured terrain, when diminishing hostile fire per- 
mits labor and when the situation or the available fighting 
means do not permit an immediate continuation of the ad- 
vance. Thus in former times as now, the use of the spade 
in an attack was a makeshift, which delayed the course of 
the battle in order to offset the lack of sufficient fire sup<- 
port. Troops that have fought successfully will decline the 
use of the spade, while troops, insufficiently supported by 
artillery, which have worked themselves up to the enemy, 
will employ the spade in situations in which it does not ap- 
pear at all necessary. We must absolutely discountenance 
a general use of the spade in every attack, as the best pro- 
tection against the hostile fire lies always in our own fire. 

THE ASSAULT 

The World War has shown clearly that the assault is 
by no means the spectacular performance of timea gone by. 
Now as ever '^he assault with cold steel seals the defeat of 
the opponent The skirmish line is made denser, the fire 
of rifles and machine guns increased. Of the greatest im- 
portance is fire protection, in advancing to the charge, few 



208 DEVELOPMENT OF TACTICS-WORLD WAB 

orders can be issued then ; much must be left to the initia- 
tive of the subordinate commanders, who must be thor- 
oughly imbued with the importance of fire protection. 

Charging distance cannot be lajd down by regulations ; 
in peace time we calculated it to be 100 meters. In any case 
we must, with one breath and one rush, enter the hostile 
position* In the last Russian Regulations before the World 
War it was recommended that the start of the assault was 
the length of a bayonet — ^this of course was figurative only. 
In war it has been shown only too often that assaults started 
at long range broke down under the hostile fire. The shorter 
our charging distance, the surer the success. On the other 
hand, however, infantry desirous of attacking will start the 
charge at long distances when there are clear indications 
that the enemy will completely abandon resistance. 

The decision for the assault either originates in the 
firing line with the subordinate commanders, or the com- 
mander of the attack gives the impetus by engaging his re- 
serves. The latter without doubt is the better way. ''If 
the decision to assault has its origin with the leader in 
rear, the signal 'fix bayonet' is hoisted as a sign for the 
assault, which must be obeyed by all troops concerned. At 
this signal all skirmishers increase their fire to the utmost 
limit; the portions of the firing line farther back work up 
as rapidly as possible to close distance; all reinforcements 
in rear hasten straight to the front. As soon as the lead- 
ing line is to start the charge, all trumpeters sound 'charge,' 
all drummers beat the general, and all portions of the liiie 
throw themselves with the utmost decision upon the enemy. 
It is a matter of honor on the part of the skirmishers never 
to let the supports overtake them before they have reached 
the hostile position. Immediately in front of the enemy 
pieces are brought to the 'charge bayonet' and the position 
entered under loud shouts of 'hurrah !' " Thus it read in 
the old Regulations. In the future also we shall not be able 
to do without "fix bayonets." 

It is very difficult for the commander of the attack to 
determine the correct time for starting the assault as he 
is forced, especially in the case of larger units, to remain 
far from the attacking line, so that he can follow the course 



INFANTRY ATTACK IN OPEN WARFARE 209 

of the attack only along general lines. Whether the fire 
power of the enemy is entirely broken or not he can deter- 
mine only by the progress made in the advance of his own 
lines. Still more seldom can he correctly perceive when 
the resistance of the enemy breaks at some one point, and 
he may be entirely unable to issue orders in time to take 
prompt advantage of such an event. 

For these reasons the firing line will have to give the 
impetus for the assault, probably in every instance. It is 
especially necessary, when the enemy vacates his position^ 
that the firing line, quickly taking advantage of that oppor- 
tunity, proceeds to the charge. If it were to wait, in this 
event, until the reserves come up, much valuable time would 
be lost, the opponent might recover from his temporary un- 
favorable situation and again face about, and even receive 
reinforcements. If the impetus for the charge originates 
with the firing line, there is danger that instead of a gen- 
eral attack only a partial attack will take place. It is im- 
possible for the entire skirmish line to judge simultane- 
ously that the proper time for the assault has arrived, only 
a portion can do this. Such partial attacks are dangerous 
and have a chance of success only in covered terrain. In 
most cases such an attack, made by a single battalion or 
even by one company, will draw upon itself the fire of a 
large portion of the hostile line and naturally go to pieces 
under it. Even if a bold charging detachment is fortunate 
enough to enter the hostile position at some one point, it will 
be at once attacked by the defender with superior forces 
and forced to flee under heavy loss. Usually the defeat is 
not confined to a single detachment for the troops on the 
right and left will as soon as they see the others charging 
start forward on the run without having gotten close enough 
to the enemy in their front to have sufiiciently shaken him. 
Thus, in most cases the decision for the charge, arrived at 
in the skirmish line, will turn into a series of defeats but, 
will, nevertheless, bring fresh troops closer and closer to the 
enemy until finaUy the great moment of the preponderance 
of men and fire is attained. 

The execution of the assault portrayed above demands 
still another special preparation, namely advancing the ar- 



210 DEVELOPHENT OF TACTICS— WORLD WAR 

tillery and minenwerf er fire. As long as that stays in one 
position, any assault is impossible. If telephone or flash 
light connection still exists between the artillery liaison 
officer and the batteries, it is possible of execution, other- 
wise it will be necessary to come to an agreement prior to 
starting the assault, to the effect that when the ''artillery 
signal flags'' are waved from side to side, or when light 
pistols are fired with certain colored lights, that signal is to 
be understood by the artillery to indicate the commence- 
ment of the charge. But if this decision is confined to but 
a portion of the line, errors are unavoidable, which will 
probably be paid for in blood. A more favorable way is the 
execution of the forward advance of the fire of the artillery, 
when the commander decides to assault. The signal ''fix 
bayonet" acts like a preparatory warning, and then, by 
a previously agreed-on signal, the fire can be advanced. 

In many instances our enemies met us with a counter- 
attack in our assault. In such an emergency the assaulting 
troops must throw themselves down and defeat the counter- 
attack by fire, and resume the forward movement as soon as 
possible. The guiding thought in an assault must always 
be : to reach the designated objective, regardless of whether 
that consists in breaking the resistance or irresistible pur- 
suit. 

In preparing the assault we generally favored the em- 
ployment of hand grenades in such manner that shortly be- 
fore the closing 3 hand grenades were thrown by the men 
lying down, who then charged with loud shouts. The rule, 
that in a charge in daytime rifies must be loaded appears 
not to be superfluous as the bullet is the most effective at- 
tacking means at close range, although our infantry, as weU 
as our enemies, overestimated the effect of the hand gren- 
ade as compared with the rifle. The charge requires special 
training, in which the hand to hand struggle with pickax, 
sharpened spades, rifle with bayonet fixed, or even with 
bare hands must be practiced. Shouting hurrahs shortly 
before closing in, and shouted with full lung power, hurts 
the morale of the enemy, however premature shouting pre- 
vents surprise. In exceptional cases it may be well to charge 
without shouting so as not to attract the attention of neigh- 



INFANTRY ATTACK IN OPEN WARFARE 211 

boring forces of the enemy. ''As soon as the shout is taken 
up, all trumpeters sound the 'charge'/' The platoon com- 
mander charges ahead of his men ; his example is of decis- 
sive importance; a charge once started must lead into the 
midst of the enemy. Every hesitancy spells annihilation. 
The charge must find the commanders of the machine guns 
in the front line. The minenwerf er fire can be of great use 
at short ranges. The main point is that the machine gun 
squad leader gives to the best of his abilities such fire sup- 
port that the skirmish squads reach the enemy with as little 
loss as possiUe. If the entrance succeeds only at a few 
points, those points are to be immediately changed into ma- 
chine gun nests from which the hostile machine gun fire is 
held down, and from which the enemy can be enfiladed. 

If the intention is to penetrate the position, we must 
avoid stopping within the position. The charging troops 
must not pay any attention to either taking prisoners or 
materiel and attempts by the hostile supports in the matter 
of counter-attacks are to be smothered in their inception 
by relentless pushing forward. The retreating enemy must 
be kept on the run, thus preventing his lines in rear from 
firing and causing them to vacillate. When the attack ob- 
jective has been reached, a moment of weakness sets in, 
which must be quickly overcome to avoid reverses. The con- 
trol, lost by pushing into the position, must be at once rees- 
tablished with severity. Detachments no matter how small, 
are to be cut out as reserves. These troops in "readiness'' 
with their machine guns must at all times be able to defeat 
counter or flank attacks. Machine guns and light minenwer- 
f ers held in rear must come up quickly, but the highest com- 
mander must at once prevent excessive crowding in the posi- 
tion, on which the hostile artillery fire will soon be directed. 
The importance of the quick arrival of all commanders, from 
the regiment up, cannot be emphasized enough. 

The inclination to pursue the enemy along the entire 
line with the bayonet instead of halting and utilizing the 
fire power of the rifle has shown itself in all battles. The 
measures the attacker has to take after a successful attack 
are, to use the victorious infantry for pursuing fire as soon 
as a field of fire presents itself, under protection of which 



212 DEVELOPMENT OF TACTICS— WORLD WAR 

fire, fresh detachments start from the wings in pursuit. In 
selecting the points from which the pursuing fire is to be 
delivered due attention must be paid to the observers who 
have gone ahead and the batteries that are to arrive shortly. 
Reconnaissance on the wings is necessary during the pur- 
suing fire, as is also locating our position for the artillery 
by means of fiags and to the airplanes by spreading out 
panels, as well as bringing up hand to hand fighting material. 

If the enemy has fled beyond point blank or short range, 
all subordinate commanders will independently follow with- 
out waiting for orders from above. Assembly of the troops 
in close order, leading off prisoners, occupying the position, 
replenishment of ammunition must come after the more 
important mission of starting the pursuit. Only later when 
the enemy permits, are units correctly reorganized. 

Means for defeating a counter-attack are especially 
important, and we have to count on counter-attacks in all 
cases. Our main support lies in the machine guns and light 
minenwerf ers that have been brought quickly into position. 
If possible the first minenwerf er men must arrive with the 
charging infantry. These men select and mark the posi- 
tions, while the regiments send forward ammunition. Or- 
ders for this must have been issued even before the decision 
for the assault is arrived at. The artillery will endeavor to 
regulate its stationary barrage for defeat of counter-attacks 
in case our infantry does not go beyond the position, and if 
possible, verify its correctness by airplane observation; 
however if the attack is to be carried further a change of 
position of our artillery will generally be necessary for pro- 
per fire support. 

If the attack miscarries, it is the duty of all comman- 
ders to bring the men, retreating under the hostile fire, to 
a halt. As a general rule the halt is made at our artillery. 
But a halt probably cannot be thought of if the hostile fire 
is very effective. Only at the nearest cover, or where there 
is no cover, only at long range, when fatigue brings the 
retreat to an end or when the hostile fire is no longer effec- 
tive, can a halt be made. This distance is the greater, the 
longer the enemy, not being interfered with by our fire, can 
make full use of the ballistic qualities of his arms, unless 



INFANTRY ATTACK IN OPEN WARFARE 218 

the attacker's artillery prevents him from doing so. The 
general rule must be to give up as little as possible of the 
terrain that has been once captured. Wherever the troops 
come to a halt they again form, and intrench if the hostile 
fire effect permits. They must then try their very best to 
attain, under protection of darkness, what could not be at- 
tained during daylight. An attack in the late evening, after 
the decisive fight has not been quite completed, has always 
had success. 

INFLUENCE OF FOG 

Examination of a large number of operations shows 
that bad weather, fog, snow storms, heavy rains accom- 
panied by thunder, beating into the enemy's face has never 
under various excuses been fully utilized by troops in war, 
to carry out an attack. An exception is the dust storm on 
March 9, 1905, at Mukden which facilitated the attacks of 
the Japanese wherever they were attempted. Fog is never 
cited as a motive for attack, but almost without exception as 
an accuse for inactivity or failure. 

The difficulties connected with executing attacks in 
fog and carrying them to completion are well known; the 
direction is easily lost, and frequently detachments will fire 
on each other. The fear of running into an ambush creates 
a feeling of insecurity. On September 9, 1917, we started 
a well prepared attack against Hill 344 at Samogneux (Ver- 
dun). The attack went to pieces in the fog, though the 
troops were excellently trained in executing night attacks. 
The assaulting troops lost the direction and the connection 
with a successful column could not be kept up as no signals 
or signs could pierce the fog. We must calculate in an ad- 
vance in fog or on the fog rising or dispersing suddenly. 
While close order formations are preferred in an advance un- 
der fog and all distances are to be shortened and special 
measures taken for keeping up connection, all subordinate 
commanders must, upon the dispersion of the fog, indepen- 
dently take all measures that are necessary, as in an advance 
without cover, viz., formations with more intervals and dis- 
tance. Fog is a valuable ally for well trained troops. 



214 DEVELOPMENT OF TACTICS— WORLD WAR 

FIGHTING IN WOODS 

Woods have different influence on the attack as well 
as on the defense depending upon their extent^ density, con- 
formation of the ground, and season. In all cases they per- 
mit placing troops in readiness and shifting them, unob- 
served by the enemy's terrestrial and aerial observation. 
Possibilities for surprise always exist, as reconnaissance is 
difficult. During protracted bombardment the trees grad- 
ually disappear and the view is facilitated. In case of large 
woods the artillery has at the start little possibility of pro- 
ducing much effect, but this effect will be materially in- 
creased by application of aU technical means— especially 
if the battle is protracted. 

If woods prevent the enemy's gaining a view, they will 
prove very valuable in the matter of supporting points and 
flanking works, and in many cases will become the critical 
points of the battle. If, on the other hand,, they can be 
seen from afar, they easily attract hostile artfllery fire and 
soon become useless as cover against observation. Woods 
facilitate the defense by an inferior numerical force, favor 
the delaying battle by permitting troops to be disengaged, 
and deceit as to strength. They generally compel the at- 
tacker to employ relatively large forces, and therefore he 
wiU generaUy seek to avoid them. 

Experiences in war have taught us that only minor 
importance is to be attached to infantry fire in woods, that 
the flight of the bullet is easily deflected by striking twigs, 
etc., and that not the bullet but cold steel is the decisive 
weapon in forest fighting. Anyone engaging in protracted 
fire fighting in woods will only lose time. The defender seeks 
to gain time, while the attacker must seek to bring the for- 
est fight to an end as soon as possible. Hand grenades in 
forests are generaUy more dangerous to the thrower than 
to the enemy. 

From the above we find: 

1. Close order is discontinued only if the condition of the 
woods demand it; the denser tiiie woods, the closer the troops; 
only thus can the oflScer retain control over his men. 

2. Dense woods must not be circumvent^; the troops must 
work their way through them, unless they want to run the risk 
of losing marchndirection. 



INFANTRY ATTACK IN OPEN WARFARE 216 

3. The less noise troops make passins: throurh woods, the 
better; all shouting, and signaling must cease. Omy in clearings 
and at spots where the woods are very clear and devoid of brush 
can firearms be resorted to, otherwise the decision is sought by 
sharp attack with the bayonet. 

4. Experiences have taught that a decision may be gained 
in forest battles by exhaustion of the enemy, but that the decision 
can be more <]|uickly reached by making attacks with fresh de- 
tachments agamst the hostile flanks. 

5. There is special danger in the infantry attack of units 
becoming dispersed and the consequent loss of control by officers, 
thus frittering away the attacking power. 

The French, having wired the edges of the woods, pre- 
ferred the system of posting small posts and machine guns 
there, and arranging a disconnected defensive position, with 
low blockhouses connected with wire, in the interior. Ob- 
stacles were placed from the edge into the interior so as to 
lead the attacker into blind alleys dominated by machine 
guns. Sharpshooters and machine guns lodged in trees kept 
paths and roads under fire. Much use was made of sound 
detectors which gave notice of the approach of the enemy. 
In the Vosges the Alpine Chasseurs, lodged in trees, gave 
the alarm by imitating the cries of birds. The defense was 
made offensively (ambush), the principal intention being 
to fall on the flank of the careless pursuer. The artillery 
fire at the start was barrage fire, with ravines, crossroads, 
paths, and roads through the woods systematically kept un- 
der fire. 

If woods are of small extent we will push through them 
rapidly, but such procedure has its danger if the woods are 
large. In the latter case it was found better to push into 
the woods some 50 meters, then halt and establish order. 
Pursuing patrols advance along all roads, combat patrols 
are stationed at the edge and outside the woods with orders 
to keep up connection with the troops on the roads leading 
into the forest. After fixing the march direction and prop- 
erly securing the service of information, the advance is 
made slowly under protection of patrols. Clearings and 
roads, where hostile fire may be met, are avoided, shock 
troops with flame throwers are held back by battalion and 
company commanders, machine guna are posted at roads 
and clearings to be ready to fire along them. Hand grenades 
and light minenwerf era are but seldom suitable in forest 
fighting. 



216 DEVELOPMENT OF TACnCS— WOBLD WAR 

It depends on the kind of forest in what formation the 
troops must traverse it. Forests with no underbrush per- 
mit advance in skirmish f ormation, but in dense woods, skir- 
mish lines are not easy to control and have little power of 
resistance, so that a detachment advancing, in close order, 
can overrun them. 

In forests, detachments will encounter each other at 
ranges hardly more than 100 meters, so that rapid fire, and 
energetic attack with cold steel wiU secure to the troops 
all the advantages resulting from surprise, and consequently 
success. The reason for this may be found in the fact that 
the opponent, unable to see, allows himself to be frightened 
by the attack, overestimates the hostile strength, and, hear- 
ing loud shouts from aU sides and confused by the rever- 
berating sound of rifle fire, turns to hasly flight, believing 
his line of retreat cut off. He has not been given a chance 
to think of offering any serious resistance. 

Attack with cold steel must become second nature to 
the troops; it requires a formation that allows rapid fire 
and a rush in close formation ; such a formation might be 
the company in line, but as line formations on the move are 
difiicult in woods, and as men in skirmish line can easily 
evade the supervision of their officers, and f urthermore^ as 
the skirmish line has no power of resistance against troops 
in close order neither line nor skirmish formations seems 
suitable for the purpose. 

For traversing dense woods platoons of the company in 
line abreast of each other, or echeloned, each platoon formed 
in double column of squads is recommended. It is not so 
good to advance in double column by platoon or half-pla- 
toons abreast of each other, as the columns are too long, 
deployment too difficult, and intervals hard to maintain. 
Squad columns on the other hand will be near enough to 
keep in sight of each other, and they can more easily work 
through the forest than can detachments in close order. De- 
ployment for purpose of delivering fire and for the assault 
can be easily accomplished. A disadvantage is their lack 
of resistance against a flank attack or against an attack com-^ 
ing obliquely ; in that event we need support of flank echel- 
ons. Columns of greater depth which require a longer time 



INFANTRY ATTACK IN OPEN WARFARE 217 

for deployment, have the disadvantage that intervals may 
be lost in the endeavor to keep proper direction. Front and 
flanks will have to be protected by patrols and by detach- 
ments in echelon, so that any hostile flank attack is itself 
flanked by detachments moving to the front. In rear of 
this first line follows a second line in close order, where prac- 
ticable along roads. The patrols must not be so far away 
that they cannot see the troops they are protecting. A 
stop is made at all crossroads, when direction and order is 
reestablished. Of course, this formation, halting and re- 
establishing order, takes time, which however is offset by 
the increase in order and security. To prevent connection 
with the retreating enemy from being lost at the edge of the 
woods, strong pursuit patrols must be started along roads 
after the enemy. 

It is advantageous to signal the progress of the infan- 
try to the artillery and to the commander by flares, which, 
by the way, easily remain hanging on the tops of trees. 

In an attack upon a forest we do without the stationary 
barrage. At the start we throw shells on the edge of the 
forest for a depth of some 200 meters, then, at the moment 
of the charge, the flre is lifted to 400 meters and the infan- 
try advanced to the edge of that flre, then the fire again 
lifted some 400 meters, and so on. All available means of 
communication must be utilized for connection with the ar- 
tillery. For communication within the battalion, we will 
have to be satisfied with runners. 

It has always been of advantage to mark the lines 
along which battalion and regimental headquarters moved 
forward, when entering woods, by posting runners. 

Woods are of advantage to the defense in that they 
conceal to a large extent the works, prevent aerial observa- 
tion and increase the difiiculty of the enemy's artillery locat- 
ing them. Surprise works, such as blockhouses, offer in 
woods good opportunities for the counter-attack against 
an enemy that has entered the position, as the defender 
knows the ground well. Obstacles from branches of trees 
are poor, they soon dry out and can be easily set on fire by 
hostile patrols or artillery fire. They suffer much in a 
bombardment. Men familiar with woodcraft, such as for- 



218 DEVELOPMENT OF TACTICS— WORLD WAR 

esten and wood cutters, should be employed on listeninsr 
posts. By observing many indications, such for instance, 
as the conduct of birds they can ascertain the location of 
observation posts in trees or the approach of hostile patrols. 

In the retreat woods facilitate disengaging troops from 
the enemy. Even sharply pursuing detachments can be held 
up by weak but skillfully handled rear guards until the 
main body is marching in the desired direction undisturbed, 
or has made its preparations for a counter-attack. Rear 
guards generally have no trouble in drawing the enemy into 
a direction favorable for the counter-attack. 

VILLAGE FIGHTING 

Localities such as villages at all times during the war 
exerted a fatal magnetic power on all troops. In vain were 
all precepts and regulations of all armies to occupy vil- 
lages, for the defense, only with weak forces and to defend 
villages only when their occupation offered tactical advan- 
tages ; to avoid in any attack village fighting, but to seek the 
decision beyond their limits, in the open terrain. Aspem, 
Ligny, Bazeilles, and Le Bourget are merely examples of 
how little the teachings helped. Beaune la Rolando with 
1610 meters extension was held by 2300 men with a loss 
of only 6.6%. Frequently the defender gave way to the 
conflagration rather than the attacker. The hope enter- 
tained prior to the World War that the increased artillery 
effect would make village fighting superfluous, was incor- 
rect. 

The World War can cite many instances of tenacious 
village fights carried on around complete ruins that once 
were villages, but which still were capable of being defended. 
I shall mention only Combles and Sailly in the Somme battle, 
and Mondidier in the Sunmier of 1918. On the Eastern 
theaters of war a number of tenacious village fights could 
be cited. In the Masurian battle a bitter house to house 
fight took place in Johannisburg held by two Russian regi- 
ments ; the place fell into the hands of troops of the 2d In^ 
f antry Division brought up from three sides. 

Par. 436, L D. R. pointed out that tenaciously defended, 
massive villages might become the key point of tiie battle, 



INFANTRY ATTACK IN OPEN WARFARE 219 

and occupation and arrangement of the edge of the village 
occasionally facilitated the rapid creation of a defensive 
position and an increase in the enemy's difficulties in per- 
ceiving the strength of the garrison. The disadvantage of 
the ease with which the hostile artillery could find the range 
should be well considered also, so that it would often be bet- 
ter to place the defensive line outside the village. Fortifi- 
cation of the flanks of villages is not to be neglected and 
the rear edge should be opened so far that the counter-at- 
tack by reserves would not be made more difficult. 

Riflemen's positions when the village is within the range 
of effective hostile artillery fire, should be in massive build- 
ings or in houses that are covered by a rise in f ront, or by 
other portions of the village. For this reason the French 
recommended that the defensive line be withdrawn into the 
interior of the village, using the portions lying between the 
village and the enemy's position as a protective cover against 
the hostile artillery fire. As early ag the Campaign of 1870 
it appears that, the defense of Bazeilles was planned with 
this intention. ^'Defensive arrangements in the interior of 
the village must be made so that each sector, and each en- 
closure can be defended. Broad streets, open places, or 
brooks, running through the village parallel to the defen- 
sive line favor defense by sectors." (Pioneer D.R., par. 
344.) 

In the World War the difficulties of village defense 
were increased by aerial reconnaissance, by the increased 
size of projectiles of the attacking artillery, and by the em- 
ployment of flame throwers. On the other hand the defense 
of a village was increased by the introduction of machine 
guns which permitted the replacing of numerous men by 
machines. Utilization of minenwerf ers favors the attacker 
as well as defender. In the fights for the fortress-like Ar- 
mentiers on April 9, 1918, the German infantry declined to 
make a charge against the strongly occupied village, which 
had been covered by a hail of shells and gas bombs. The vil- 
lage finally fell (the British falsely state that they voluntar- 
ily evacuated it) when taken in flank and rear. The spoils 
were 50 officers and 3000 men as well as 45 guns, and a cloth- 
ing and ammunition depot. The place had cover, toward the 



220 DEVELOPMENT OF TAGTICa— WORLD WAR 



of the enemy, so that the artillery, which was given ihe 
mission of throwing flanking fire came from neighboring sec^ 
tors. In many cases the British understood how to post ma^ 
chine guns in buildings in the vicinily of and visible to our 
artillery positions. These were so hidden they could be 
located by our airplanes only with difficulty, and by frequent 
change of position they were able to even escape our flash 
and sound ranging squads. 

A peculiar village fight happened on April 24, 1918, 
around Villers-Bretonneux, east of Amiens, which was held 
by the British. After an artillery preparation lasting two 
hours and a half the German infantry supported by tanks 
attacked the village. The first attacking wave of infantry 
stopped in front of the village, which was still under fire, 
while infantry, and tanks, gained ground in the open terrain 
to one side of the village. Tanks cleared a supporting point 
at the southeast comer of the village for the assault and 
then opened fire on the viUage street, later they turned off 
to the north edge of the village and from that point produced 
good effect against its interior. Still later the tanks ad^ 
vanced towards the railroad station and church of Villers- 
Bretonneux, where the British in vain endeavored to hold 
out. 

These examples sketched in mere outline, show that 
a tenacious defender cannot be driven out by artillery fire 
from a well prepared village, if it has been supplied with 
deep shelters and that an artillery bombardment does not 
assure the possibility of avoiding a bitter fight in the in- 
terior. Fights in streets and houses consume forces and 
time. We wiU cite the capture of Messines on October Slst, 
of Wytschaete the same day and on November 1st and 2d, 
1914. A later study will have to show, if with all the heroic 
valor of the troops it would not have been possible to shorten 
these fights. If that is impossible, nothing remains except 
to give troops flame throwers and accompanying artillery. 

The surest means for the attacker to take a village, is 
in locking the place in by fire and troops. Villages com- 
pletely surrounded by troops are almost always taken 
quickly, with rich spoils in prisoners and with little loss on 
' the part of the attacker. Rapid penetration of the village. 



INFANTRY ATTACK IN OPEN WARFARE 221 

immediatdy in rear of a rolling barrage frequency breaks 
resistance without losses. If a house to house fight results^ 
the attacker must gain a foothold just as the defender^ in 
houses and ruins. He must avoid the streets, work his way 
through gardens and yards, and thus slowly but surely 
break the enemy's resistance. 

Defensive arrangements of a village take much time 
and troops. Works that can be fired on by artillery will be 
destroyed in a short time and do not pay for the trouble of 
constructing them. But the ruins still offer, to the stub- 
bom defender, a possibility of continuing the battle. Battle 
positions, in front of the village, are hard to hide from the 
view and camera of the airplane, they require special works 
for bomb-proof shelter, while connecting roads with the vfl- 
lage take much labor, and invariably will betray the posi- 
tion. Thus it is an open question if the defense of the edge 
of the village would not be better. Cellars are numerous, 
and even if they require reinforcement, the labor connected 
therewith is slight in comparison with constructing bomb- 
proofs and connecting trenches. Only works that are hid- 
den from view of artillery and airplanes are a reinforce- 
ment to the defense. Thus it may well be accounted for 
how the tenacious resistance happened on November 11, 
1914, by the left wing of the Guard Division in the Chateau 
Veldhoek in the forest of Herenthage. Positions, hidden 
from hostile fire and aerial reconnaissance, from which a 
fight can be advantageously conducted, require especially 
strong garrisons and materiel. They turn into critical points 
of the battle, on which the entire defensive system may be 
constructed. Obstacles and bomb-proofs have precedence 
over earthworks, which in a bombardment, are generally 
annihilated beyond recognition in a very short time. Sup- 
porting points are for the purpose of holding out, even if 
the surrounding country has already fallen into the enemy's 
hands, but whether or not completely surrounding them with 
obstacles is correct, requires special consideration in each 
instance. Probably it will always be well to do it I 



X. Cavalry 

CAVALRY PRIOR TO THE WORLD WAR 

After aU wars of modem i^imes — ^f rom Bliicher's sharp 
condemnation of the activity of the Prussian cavahy during 
the Wars of Liberation down to the reproach for the break- 
ing down of the Russian cavahy in Eastern Asia — ^the same 
complaints were heard concerning the small achievements 
of the cavalry in the face of the enemy. Besides the increase 
in fire power, we must seek for the cause in the continually 
increasing discrepancy in the strength of cavalry and infan- 
try. The only exception of this is shown in the American 
War of Secession ; unfortunately we delayed making use of 
the experiences of Stuart, Sheridan, and Forrest; in any case 
we could have arranged for similar achievements as early 
as the Franco-Prussian War by timely change in armament 
and training. We were quite justified in concluding from 
the Russo-Japanese War that only first-class cavalry pays, 
that cavalry has to be trained for mounted and dismounted 
fighting, that it must by no means be allowed to sink to the 
level of mounted infantry, which means poorly riding rifle- 
men like the British in South Africa. Cavalry has. to be 
able to fire especially well. The Russian cavalry in Eastern 
Asia, except a very small fraction thereof, did not come up 
to the standard in training, equipment and armament of 
modem cavalry; it did not seek dismounted fighting; it fell 
into inactivity. Cavalry that can be employed in war can- 
not be improvised, it must be systematically trained during 
long peace years. Then, however, the army leadership must 
set cavalry clearly defined tasks and pitilessly demand defin- 
ite results. Field Marshal Oyama stated after the Battle of 
Mukden : 'If I had had but two or three cavalry divisions, 
the Russians would never have gotten northward after the 
defeat of Mukden, or at least their right wing would have 
fallen victim to the pursuit of the army cavalry .'' In Ger- 
many and Fraiice we considered the dismounted action only 
a makeshift and sought employment on the battlefield by 
the charge against all arms. Little attention was paid that 



CAVALRY 228 

the Turkish cavahry, trained according to German principles, 
had sought in the Balkan War of 1912-13 the attack, but 
had to be satisfied with very successful participation with 
fire arms in the actions of the other arms« 

Our conception was confirmed by the fact that the 
British cavalry, after some hesitation, then the Russian 
cavalry after the Russo-Japanese War, adopted the belief 
that the days of the cavalry charge were by no means past ; 
that cavalry rather, was clearly destined for mounted fight- 
ing; that the attached arms in many ■ instances relieved the 
cavalry from the necessity of fighting on foot. The German 
cavalry driU regulations with the same maxims were adopted 
by Japan in 1895. 

If the cavalry has not sufficiently achieved success in 
the face of the enemy there are two good reasons. First, a 
lack of individual commanders capable of requiring tte 
higher efforts of cavalry under favorable conditions without 
regard to probable losses, and second, the supply of horses 
on the battlefield, by no means fulfilled the requir^nents 
clearly foreseen in time of peace. 

The extension of the battlefield increased the difficulties 
of reaching the flanks and rear of the enemy. But if cav- 
alry desires, through its charges, to accomplish a decisive 
influence on the course of the battle, nothing remains, in 
consideration of the growth of modem armies, but to attack 
in mass. However, the increased fire power of rifle and gun 
increased the difficulties of placing cavalry in readiness and 
bringing up larger units. It is very probable that charges 
in regimental and brigade formation in favorable terrain 
are still possible, but attacks in division formation are out 
of the question. In the machine gun the infantry has re- 
ceived an arm which is absolutely annihilating against tall 
and broad targets. In addition, increasing cultivation, and 
covering the terrain with industrial works, fencing the 
fields with barbed wire, are increasing the obstacles to the 
mounted use of cavalry. 

VIEWS ON CAVALRY ATTACKS 

Regulations of ^11 armies lay stress on the necessity in 
time of peace of cavalry employing the arme blanche in ac- 



224 DEVELOPMENT OF TACTICS— WORLD WAR 

tion to drive off the hostile cavahy and to open a road for 
our participation. Thus, in the World War we ought to have 
seen numerous cavalry charges, but the very opposite hai>- 
pened. The Russian cavalry — ^without success — several 
times attacked the German infantry; but in the West the 
cavalry declined to attack. Equipment with the lance, and 
our well trained horses, undoubtedly gave our cavalry a 
superiority and our opponents well knew the capability of 
our well trained cavalry in movements and maneuvering 
in larger units. Minor contacts showed the difficulties, due 
to the increased fire power, of keeping the cavalry close 
enough to the firing line, to be able to fully and quickly 
utilize rapidly passing opportunities for participation. The 
rapid wearing out of the trained horses, hard to replace, 
worked against a probable costly employment on the battle- 
fidd. Replacement of trained horses will probably be diffi- 
cult in the future, but by no means impossible. As long as 
wars are conducted with impressionable humans, so long 
win a suddenly appearing cavalry mass have its effect on 
infantry that has suffered physically and in morale, and 
where training has not given it the proper stability. On 
the other hand, the organization of infantry favors success 
of a charge. Arms become effective only in the hands of 
men; but these men appear more than ever susceptible to 
moral infiuences on account of the increase of urban popu- 
lation. Because favorable opportunities were not grasped/ 
or because cavalry was not just at the right place, or be- 
cause the commander had been deceived by the 'parade'* 
aspect of the enemy, or because the conmiander was unable 
to see that the infantry . opposed to him was ripe for a 
charge, we have no justification for concluding that the at- 
tack was impossible. Uncertainty, which leads to inactivity, 
taken in conjunction with the difficulties in reaching the 
decisive place at the correct time, has frequently saved, 
broken and demoralized infantry from the charge. 

Charges against marching batteries, surprise attacks 
against flank and rear of artiUery in action, or even against 
the front of artillery firing from under cover, against artil- 
lery that is not fighting in connection with infantry, can fre- 
quently have great success. Frequently it is advantageous 
to silence batteries if only for a short time. Successes 



CAVALRY 

against ammunition columns are also to be valued. Pres- 
ent day cavalry must therefore be able to fight mounted as 
well as dismounted ; it is entirely too costly to sink to the 
level of rapidly moving infantry. 

The moral impression produced by charging cavalry 
remains undisputed (Custozza, Mars la Tour) , and even its 
mere presence has a weakening effect on the enemy. 

The leader will have to consider that the losses in a 
charge of cavalry against cavalry will probably be very 
small, and against other arms extraordinarily high, while 
the losses inflicted on the enemy are generally relatively 
smaU. The leader must consider this in arriving at a deci- 
sion whether to resort to the charge or to dismounted fight- 
ing, especially when, by fighting on foot there is a possi- 
bility of causing the enemy great loss. But in no case must 
the fact that a dismounted action can quickly and easily be 
broken off, be permitted to affect the decision to make a 
mounted charge, when it would cause a quick and decisive 
result. Prerequisite for a favorable attack lies in sudden 
appearance. In any case we will probably have to prepare 
the success of a mounted attack by the firefight of carbines 
and machine guns. Tanks also can perform valuable service 
here. Frequently battle airplane echelons and escadrilles 
will have the same effect as a mounted attack. The mounted 
attack, which utilizes celerity and shock power to ride down 
the enemy, has lost in importance in modem war, though 
plenty of opportunities will offer for successful charges. 
The larger the cavalry unit, the greater the difficulties of 
coordinated action and the more will cavalry have to seek 
opportunity to participate by fighting dismounted in the 
battle, to which the cavalry must be trained, and without 
paying any attention to the neighboring troops to solve in- 
dependently its task in any terrain and at any hour of the 
day. Due to increased attachment of fire arms of all kinds, 
in conjunction with its mobility cavalry wiU more than ever 
be assigned to operate against the enemy's flank and rear, 
in which hostile cavalry must be quickly swept aside, if pos- 
sible by mounted attack, and not by the slower method of 
fighting on foot. 

As the hostile cavalry will have the same tasks, both 
parties will naturally endeavor to bring about a rapid deci- 



226 DEVELOPMENT OF TACnCS— WORLD WAR 

sion, to uninterruptedly continue their tasks. Thus cavahy 
engagements wiD result ; and only the inferior cavalry will 
resort to dismounted fire fight to accomplish its task. This 
resistance has to be overcome by dismounted offensive at- 
tack. The cavalry has to be ready day and night for the 
^'offensive fire fight/' to be able, after the victory over the 
hostile cavalry, to move against the flank of the enemy, 
against the line of communication, or against cavalry. 

The cavalry must never wait for opportunities to at- 
tack, it must not delay its activity until it is let loose for 
the pursuit, but must participate with all means at hand in 
gaining the victory and carry its strong fire power rapidly 
to the weak points of the enemy's line. This advantage of 
being able to rapidly cover short distances to occupy impor- 
tant points, was not properly valued prior to the war. Ef- 
ficient cavalry can always mislead the enemy and can also 
thereby make him afraid. The long continued battles and 
the dependence of the armies on uninterrupted supplies of 
ammunition and provisions, mostiy brought up by rail, make 
operations against communications to the rear of decisive 
importance. Without rail connection modem armies rapidly 
lose their fighting power and freedom of movements. The 
raids, condemned before the war by C. D. R. 527, gain there- 
by special value. 

It is probably to the point to here point out the raid, 
undertaken with strong cavalry in September, 1919, by Gen- 
eral Mamontoff. The end of August he pierced the Bolshe- 
vist line south of Moscow, was enabled to live, without sup- 
plies from home, in rear of the 8th Bolshevist Army, off 
the country, and supported the advance from Moscow against 
Orel. 

EMPLOYMENT OF CAVALRY IN WAR 

Let us try and sketch the tasks of cavalry in war. The 
army cavalry performs all the duties that an aimy deniands ; 
the generally weaker divuional cavalry , frequentiy rein- 
forced or totally replaced by cyclists, performs the duties 
required by smaller troop units. In conjunction with the 
aerial observation forces the cavalry carries on reconnais- 
sance, screens movements of the army, and secures the con- 
nection between separated portions of the army. After the 



CAVALRY 827 

contact of the armies the cavalry seeks employment in flank 
or rear of the enemy, against important railroad junctions, 
prevents the enemy from bringing up supplies and attempts 
to keep approaching reinforcements from the battlefield. 
The development of wireless and of the airships make it 
possible to supply distant cavalry with orders and facili- 
tates co-operation with the armies in retreat and in pursuit. 
The cavalry participates with all arms in battle and finds 
special employment in retreat and in pursuit. 

Reconnaissance 

Aerial reconnaissance, in so far as it is not influenced 
by weather, supplements the cavalry reconnaissance and 
indicates the directions to the cavalry in which it is to re- 
connoiter. Cavalry reconnaissance always has the advan- 
tage of gaining by personal contact (prisoners) important 
points for the leader's decisions. The cavalry reconnoiters 
with weaker detachments (officers' patrols) which are fol- 
lowed in support by reconnoitering squadrons; reconnais- 
sance with stronger forces ("patrol rides by cavalry divi- 
sions") leads to frittering away the forces. 

The general rules laid down in peace time for the ser- 
vice of reconnaissance, for far, near, and battle reconnais- 
sance, have stood the test. We reckoned only too little with 
the fact that the Russians as a general rule appeared with 
stronger cavalry patrols and made more use of deceptions 
than we foresaw. The strength of a patrol is governed by 
its task and the probable number of messages to be sent 
back. The weaker the patrol, the easier can its road be 
blocked and the quicker will it use up its forces. It was 
proven wise to attack hostile patrols and reconnoitering de- 
tachments whenever the situation permitted. Armament 
with the lance gave the troops superiorily and all reports 
prove that the intrepid attack of our reconnoitering de- 
tachments always had good success. Of course, heavy losses 
resulted sometimes by overhasty action. The leader will 
solve his task mounted if practicable, otherwise dismounted. 
This requires a certain strength of the patrol. It is advis- 
able in close reconnaissance — ^to save the horses — ^to move 
the patrols forward in waves from sector to sector. The 
relieved patrols assemble first as message centers along the 



228 DEVELOPMENT OF TACnCS— WORLD WAR 

route of mBich and finally join fhe advance guard cavalry. 
In fhia manner the sending back of meBsages is best assured. 
Good reconnaissance is the first requirement of security. 

For the purpose of furnishing ralljring points and sup- 
ports for distant patrols, reconnoitering squadrons were 
of advantage in 10 to 20 kilometers broad reconnaissance 
strips. light radio stations performed the service of com- 
munication ; we cannot caution too strongly against sending 
messages uncoded. We found in Haubeuge and Namur 
copies of our own wireless messages which the enemy had 
picked up with his wireless. In order to overcome hostile 
resistance the reconnoitering squadron requires a certain 
fighting power, and that is increased by the assignment of 
I, machfaie guns, and even single field pieces. 



SCBEENING THE MOVEMENTS OF THE ARMY 

At the opening of the World War screening the move- 
ments of the army was so successfully performed by both 
cavalry corps that the advance of the 1st and 2d Armies had 
a completely surprising effect on the allied leadership, in 
spite of the fact that participation in war in Belgium by the 
population increased the tasks of the cavalry. 

Similarly in October, 1914, the Cavalry Corps Frommel 
(5th, 8th and Austrian 4th Cavalry Divisions) succeeded in 
screening the march to the left of the Army under Hinden- 
burg to the vicinity of Thorn. 

Employment Against Flank and Rear 

In July, 1915, the 6th Army Cavalry Corps accompanied, 
with its 2d, 6th and 8th Cavalry Divisions, the attack of 
the North Corps of the Niemen Army against the right 
flank of the Russian Army in position on the Dubissa and 
upper Windau as far as the vicinity of Popeljany. The ad- 
vance of the German cavalry against the Russian flank was 
decisive. In no case could infantry have operated at such 
a distance from its main force. The greater the distance, 
the greater the effect 

In September, 1914, during the ''race to the coast" there 
were concentrated on the German north wing— on a nar- 
row space: liie 2d Cavalry Corps, v.d. Marwitz, consisting 



CAVALRY 229 

of the 7th and 9th Cavafay Divisions ; the Ist Cavafafy Ciorpe, 
V. Richthofen, consisting of the Guard and 41h Cavafafy 
Division, and the 4th Cavahy Corps, v. HoUen, consisting <tf 
the Sd, 6th and Bavarian Cavahry Division. 

At the present writing it cannot be stated whether the 
cavahy could have caused more delay to the inarch of the 
British or could have pushed into the hostile communica- 
tions zone. Lack of infantry compelled the army leadership 
to employ this brilliant cavalry in position warfare. 

What will ever remain as a brilliant cavalry achieve- 
ment, was the employment of the Cavalry Corps Gamier 
(the 1st, 8d, 4th, 6th and 9th Cavalry Divisions took part) 
after the f aU of Eovno against Ihe rear of the Russian army 
at Wilna in September, 1916, Smorgon September 15th, Solg 
16th, and Wiledd 16ih), all of which were taken in dis- 
mounted fighting in the face of continually reinforced Rus- 
sian infantry, which was being brought up by raiL In the 
face of this infantry the cavalry could not take Molodeczno, 
so that finally the Russians were able to also break the resis- 
tance of the rapidly brought up 76th and 116th Reserve Divi- 
sions. In a captured Russian order appeared the f oUowing 
praise: ''The cavalry must take example by the energetic, 
brave and rapid action of the German cavalry ; I consider it 
sufficient to call back to mind to the cavalry, especially to the 
Cossacks and their commanders, the former valor of their 
predecessors — definite, bold reconnaissance under the nose of 
the enemy, especially in his rear, f uU freedom to raise havoc 
among his batteries and columns, to fall on his tired infan- 
try — ^which is history with which each commander must be 
familiar in order to cite examples of the Russian cavalry, 
which the German cavalry now so eagerly emulates." 

If cavalry has once reached the rear of the hostile 
army, then only relentless action without any regard to 
losses can attain large success. 

CoNNEscnoN Between Separated Parts of the 

Army in Retreat 

After the first battle of the Mame the gap between the 
1st and the 2d Army was closed by German cavalry. 



280 development of tactics— world wab 

Mounted Participation in Battle and as 

MoRiLE Foot Troops 

Cavalry, by surprise, can bring its strong fire power 
into eflfect. 

CAVALRY DIVISIONS 

Tlie cavalry division is the fighting unit of the army 
cavalry ; its organization and composition must be in accor- 
dance with the new tasks. It is true that the formation of 
8 brigades of 2 regiments, taken over from the Franco-Prus- 
sian War — ^principally designed for mounted action and for 
reconnaissance— has proven itself excellent, but the regi- 
ment nevertheless requires another machine gun troop of 6 
guns besides the 8 to 12 automatic rifles assigned each troop. 
If cavalry desires to solve its problem, it requires an abso- 
lutely strong equipment with automatic arms, as the horses 
rapidly decrease — as we have learned. In present day employ- 
ment of cavalry after deducting reconnoitering troops and 
far patrols, 1 regiment will not be able to insert more than 
200 carbines, 1 cavalry division no more than 6 troops. Each 
trooper must be an excellent marksman ; the small number 
of carbines must be increased by automatic arms. For this 
employment we have too many headquarters (staffs) and 
it should be considered if the demands of dismounted action 
cannot be better met by organizing brigades of 3 regiments 
and to increase the number of troops in the regiment to 6. 
Attaching good marching infantry to the division and not 
to the cavalry corps, say about 1 battalion to the brigade 
with necessary autos to carry the knapsacks, is recom- 
mended; in addition too, infantry cyclists are indispensaUe 
even on an Eastern theater of war. 

AU men that cannot be mounted temporarily because of 
loss of horses, are to be formed within each Inigade into 
bicycle detachments. As has been learned, bicycles are al- 
ways available. In Italy every cavalry regiment had 1 cycle^ 
platoon. The cyclist companies can occasionally also be or- 
ganized into battalions. 

In the attachment of artillery, the lack of light field 
howitzers was sorely felt, and also the lack of long range 
guns. Attachment of heavy field howitzers and lO-cm. can- 
nons, frequently demanded, may occasionally be of advan- 



CAVALRY 231 

tage, but would seriously affect the mobility of the troops. 
It is also wrong to carry along more than two calibers. At- 
tachment of anti-aircraft guns was demanded with justifi- 
cation. 

In the matter of mounted pioneers (supplied with ex- 
plosives carried on pack animals), 1 company with cavalry 
bridge equipment was found sufficient for the division. It 
was found best to mount the pioneers, in place of equipping 
them with bicycles, 6n account of the explosives they had to 
carry along. Information detachments should be equipped 
with light radio stations for the reconnoitering troops, heavy 
radio stations for the brigade and division ; this is absolutely 
necessary. The service of information in a regiment had 
best be performed by one platoon. Subsistence and ammu- 
nition columns should be changed from horse-drawn to mo- 
tor-drawn. 

CYCLISTS 

In organizing cyclist companies (collapsible wheels) 
with the Jftger battalions on October 1, 1918, the dislike to 
the employment of cyclists in Germany had been overcome. 
At the opening of the World War Great Britain had cyclist 
battalions (coast guard), Austria (with single Jftger bat- 
talions) , Switzerland (to reinforce the weak cavalry) . Italy 
and France also had cyclist detachments. Preparations had 
been made in Russia to supply the cavalry divisions with 
cyclist detachments. At the dose of the World War each 
Italian cavalry division had 1 cyclist battalion of 3 com- 
panies and in addition each cavalry regiment had a cyclist 
platoon of 80 men. Each British army corps had 1 cyclist 
battalion (280 men) of 8 companies with 6 automatic rifles 
carried on cycles. In mobile warfare each division was to 
receive 1 bicycle company, as there was no longer any divi- 
sional cavalry toward the close of the campaign — only corps 
cavklry stiU existed (8 troops). The French caval^ divi- 
sion had a groupe eydiate of 8 peUtans of 3 officers, 15 non- 
commissioned officers, and 90 men each. 

In Grermany the cyclist companies were doubled shortly 
after the commencement of the war, and then ersatz com- 
panies were organized in addition, which were brought into 
the field as mobile troops. Cyclist companies were enabled 



282 DEVELOPMENT OF TACTIGS— WOBLD WAB 

to atiUse fheir rapidHy to good advantage during the quick 
advance tiirough Belgium and Northern France, and because 
the road net was exceDent, they could perform tasks inde- 
pendently, and did excellent service in the holding of impor- 
tant sectors, especially in the retreat from the Mame, and 
offered a rallying point to the cavalry far in front. The fear 
that through the permanent assignment to cavalry corps 
and cavalry divisions their movements would be retarded, 
did not prove correct as long as the roads were good and the 
road net f avoraUe. In the East cyclist detachments also 
performed excellent service. During the course of the cam- 
paign additional cyclist companies were organized based on 
the good results so far shown« The cyclist companies not 
attached to the cavalry corps and cavalry divisions were 
oaployed for duty at army headquarters, army corps and 
single divisions, and in part also formed into cyclist bat- 
talions of from 6 to 7 companies and 1 machine gun company 
of 6 guns, model '08, on auto trucks. 6 of such battalions 
(4 Prussian, 1 Bavarian, 1 Saxon) later on formed the 2d 
bifantry (cyclist) Brigade, which performed most excel- 
lent services at different times, the last time in the Autumn 
of 1918, on the retreat in the West. 

While in time of peace a cyclist company consisted of 
only 8 officers and about 110 noncommissioned officers and 
men, its strength in war--corresponding to the demand — 
was increased to 3 officers and 150 noncommissioned officers 
and men and each company was assigned 1 passenger and 2 
supply autos as well as 3 wagons. Such a strength was nec- 
essary as the company in mobile war in encountering the 
enemy was left to itself and because we had to count on de- 
tachments for patrols and messengers and personnel for 
the trains, as well as on losses in battle. 

In cyclist troops the desire was expressed to have ma^ 
chine guns to increase their fighting power, and in the sum- 
mer of 1918 a portion of the cyclist companies were sup- 
plied with 3 automatic rifles, which with spare parts and 
ammunition were carried along on cycles. 

The strength of a cycle detachment is dependent on 
the march depth and on the fact that in mobile warfare 
half of the men are designated to bring along the wheels 



CAVALRY 288 

of the skirmishers. The folding wheel did not show up well 
with us. If we calculate, in addition, on detachments for 
supply trains, subsistence and headquarters service, wheels 
getting out of order, etc., it will be seen tiiat it is not cor- 
rect to attach to a cavalry unit less than 1 company of cy- 
clists of 2 platoons, of a strength of at least 2 officers, 22 
noncommissioned officers and 160 men. The baggage neces- 
sary for daily use is carried on the wheel, the remainder on 
wagons. If the cavalry division wants to make effective 
use of its cyclists, at least 1 battalion of from 8 to 4 cyclist 
companies and 1 machine gun company should be attached. 
Units of this strength proved themselves to be effective. 
Marching in the West usually was in column of threes, in 
the East generally in column of ffles. Method of move- 
ment and rate of speed require that the cyclists be taken out 
of the march column and moved from sector to sector. In 
the battle cyclists are to reinforce the fire power of the 
cavalry and to reUeve the cavalry as much as possible from 
the fire fight. 

FRENCH VIEWS CONCERNING EMPLOYMENT 

OF CAVALRY 

At the opening of the World War French cavalry divi- 
sions were not equal to the German cavalry divisions be- 
cause of the latter's superiority in machine guns and at- 
tached Jftger battalions. Men's equipment, horse equip- 
ment, and means of communication, were in favor of the 
German troopers. In the Autunm of 1914 the French cav- 
alry officers' corps was rejuvenated, infantry was attached 
to the divisions, and in November portable intrenching equip- 
ment and bayonets were supplied. In the Spring of 1916-17 
Cuirassier regiments were changed into ''Cuirassiers dis- 
mounted'' organization and strength of foot troops, and in 
addition dismounted groupes legeres were formed from 
other regiments. Grenade rifles ( Vivien Bessieres) were in- 
troduced, the number of cartridges carried by individuals 
was increased, and the number of automatic riflc»9 increased 
first to 8 and later to 6 per troop. Automatic rifles were 
given to combat patrols as a general rule. Their employ- 
ment in attack and in reconnaissance wafi specially empha- 



284 DEVELOPMENT OP TACTICS— WORLD WAR 

sized. To each regiment was attached one 37-mm. cannon 
with tripod mount (range 1500 to 2000 meters) . Attached 
w^re also motor guns and motor machine ^ns ; and in the 
Spring of 1918, also light tanks. The auto guns were not 
to be considered as fire reserve, but were to be energetically 
inserted with utilization of their mobility and their protec- 
tion against infantry and shrapnel fire, to precede the de- 
tachments, to cut off the enemy's retreat, and to be employed 
in attack and in defense for sudden fire from flank posi- 
tions. In effective fire zones they were to keep up connec- 
tion between commander and advanced detachments. At 
night, motor guns and machine guns were not to be used. 
Employment of long range heavy cannon had also been con- 
sidered. The object was to create quickly moving fire 
groups which were to make use of the success attained by 
the other arms in trenchless terrain. While for mounted 
employment the requirements of C.D.R. of 1912 were fully 
to be adhered to, three regulations were published consecu- 
tively in 1916, which looked on the dismounted fight as the 
normal employment of cavalry. The final precept was the 
regulations dated May 26, 1918: ''Celerity, mobility and 
ability to maneuver are the inherent qualities of cavalry, 
as long as it is a question of marching and maneuvering. 
But to reach objectives, which are assigned the cavalry, to 
solve its tasks, the cavalry will have to fight. In most in- 
stances these fights will have to be carried on dismounted, 
for there is no longer any German cavalry on the Western 
theater of war Cavalry tactics will have to corres- 
pond to the fire power of modern battle, and its organiza- 
tion and armament give it the possibility to f uUy use the 
fire power. Therefore cavalry, supported by artillery, must 
be able to fight on foot. But nevertheless the inounted fight 
also has to be considered and prepared for. For that oppor- 
tunity is offered against cavalry which seeks to attack or 
accept attack, against infantry surprised or disheartened in 
terrain devoid of trenches or ditches, against artillery on the 
move, or when it can be attacked in position from the flank 
or rear. The cavalry is a fragile arm, its reconstruction is 
slow and far from easy. It therefore must not be sacri- 
ficed to the impatience of a leader under conditions in which 
its inherent qualities cannot be utilized/' 



CAVALRY 235 

After December, 1916, the cavalry troop consisted of 8 
platoons of 3 squads each ; the dismounted platoon, like in- 
fantry platoons, suitable for a protracted fire fight accord- 
ing to armament and organization. In the Regulations of 
1918 it is stated: '^nder present day requirements for 
battle the dismounted fight is the normal procedure for cav- 
alry. It is important that, if the cavalry dismounts to 
fight on foot, it forms each time when possible units cor- 
responding to infantry formations, as those possess all 
means for the fire fight, for movements and for connection 
in consonance with well known experiences/' Thus, the 
cavalry platoon dismounted was to correspond to the infan- 
try section, the dismounted troop to the infantry platoon or 
half-company, the dismounted regiment to 2 infantry com- 
panies. The men trained for special purposes (rifle gren- 
ade throwers, automatic riflemen) were distributed equally 
to the platoons. 1 platoon inserted 14 skirmishers, the re- 
mainder serving as horseholders and for reconnaissance. 1 
brigade formed a foot battalion with 1 machine gun com- 
pany. The led horses of a troop were under charge of 1 
ofScer; he has 2 noncommissioned officers and 32 troopers, 
including 6 litter bearers. The led horses of a half-regi- 
ment with the combat train of the machine gun platoon were 
conducted by the captain second in rank, and the led horses 
of the troop were generally posted separately. The follow- 
ing points were laid down for the tactical insertion: ''In 
the dismounted fight the celerity and mobility of the cav- 
alry must come to the top, to fully utilize the surprise and 
the possibility of turning against flank and rear of the 
enemy. VHiere the enemy makes a stand, he will be held 
in front by fire, and mounted portions of the command press 
forward in the direction our task lies. The troops are 
formed into dismounted detachments, led horses with com- 
bat train, mounted reserve (up to one-fourth of the total). 
The mounted reserve secures the flanks and maintains con- 
nection with neighboring detachments, gathers the fruits 
of success by rapidly throwing forward detachments for 
the fire fight against flank and rear of the enemy or pursues, 
mounted, the enemy, covers the now mounting detachment 
that has fought on foot, and informs it of everything im- 
portant for the continuation of its task.'' 



286 DEVELOPMENT OF TACTICS— WORLD WAR 

Hie cori>8 cavalry and the divisional cavalry secure in 
fhe offensive battle the attainment of success, and the army 
cavalry secures the fruits of success. In the defensive bat- 
tle cavalry is enabled to materially lessen the effect of a 
penetration. Corps and divisional cavalry participate in 
gaining success in battie; if that is attained, they keep 
touch with the enemy, reconnoiter and cover the exit of fhe 
troops units into the terrain that is free from trenches. 
The following tasks are set the cavalry : To attack retreat- 
ing infantry and artillery ; to increase the point of entry by 
rapidly bringing up fire echelons with automatic arms 
against the flanks of detachments that still make a stand. 
'It will be the general rule to bring the frontal fight of 
dismounted detachments into consonance with the appear- 
ance of detachments, that are especially strong in machine 
arms, against the flank and rear. In the defense these cav- 
alry detachments will secure the approach of reserves and 
occupy important points in advance. The army cavalry will 
overtake the gathering of the fruits of victory, will increase 
the ^ect of a tactical and operative surprise, and will 
secure the movement of armies. Celerity, mobility and its 
fire power give the army cavalry an opportunity to solve 
tasks which infantry divisions cannot solve with the same 
rapidity, Bud which corps and divisional cavalry cannot 
solve with the same power. Such tasks are after a penetra- 
tion: Threatening the rear of the hostile troops, rear at- 
tacks, against parts that still make a stand, reconnaissance 
and attack on approaching reinforcements and rear guard 
positions, to prevent the enemy to again make front in a 
position in rear, to hold positions until the arrival of the in- 
fantry, to execute important demolitions, to take possession 
of provision uid ammunition depots. Assignment of air- 
planes, means of communication, light tanks, artillery, in- 
fantry units and labor troops will f requentiy be necessary. 
Again and again emphasis is laid on the necessity to not 
allow itself to become entangled into serious 'frontal fights, 
but to turn with fire arms against flank and rear of the 
enemy, but stiU adhering to the general march direction 
and then to inform the commander of the situation by using 
all observation and communication means. The cavalry divi- 
sion will divide itself into reconnaissance detachments, into 



CAVALRY 287 

a detachment d^expUntatian for pursuit, into the portions 
designated to utilize its mobility, and into infantry sup- 
port. In the defense the larger units are mainly designated 
in accordance with a kind of mobile reserve, to close up a 
gap should the enemy have succeeded in entering the posi- 
tion. The cavalry divisions have the advantage of being 
composed only of rapidly moving portions, so that on their 
arrival on the battle field they can participate in full num- 
erical strength with their guns, their automatic arms, and 
their ammunition. In that lies the value and the justifica- 
tion of existence of cavalry divisions.'' In this rapid inser- 
tion as a unit lies the advantage of cavalry divisions as con- 
trary to infantry divisions, the hasty insertion of which is 
possibly only gradually, foot troops and vehicles separated. 
The French cavalry, without entirely doing away with 
the mounted attack, turned into mounted infantry, which 
sought to solve its main task on the battlefield in participa- 
tion in the battle. The corps cavalry and the divisional cav- 
alry was to fight also as a general rule. Generally frontal 
fire fight was to be connected with an advance against the 
enemy's fliank and rear, so that artillery and automatic arms 
could be inserted against the most sensitive points of the 
enemy. 



XL TkeAitilleiy 

ORGANIZATION AND COMBAT PRINCIPLES 

Infantry and artflleiy tactics, in so far as concerns the 
enqdoyment of lii^t and heavy artiUery cannot be separated ; 
hat as has been already pointed out, we can here only treat 
of them together along general lines and we cannot discuss 
special questions. 

Before the World War, when only open warfare had to 
be considered, the quantity of artillery for one army corps 
operating as a part of a burger unit, was determined by the 
number of guns that could be placed into position in the 
corps zone of action, which amounted to 24 batteries to each 
6000 meters front. The deployment of the artillery, ex- 
cept the light howitzer batteries, was to be in a single line. 
As a result of experience it was believed in marches that, 
in relation with the infantry, the proportion of artillery 
could not be too great The march length of the artillery 
columns, with the light munition trains included, was at the 
beginning of the World War, 6600 meters for a division, as 
against a march depth of the infantry at normal strength, 
of 4800 meters. Before the campaigns started there were 
for each 12 battalions of infantry 12 batteries, that is for 
full war strength there was for each 1000 men one 6-gun 
battery, as against 2 guns to 1000 men, the amount provided 
in 1870. As campaigns proceeded this proportion had 
changed; for example, the 10th Army Corps had at Vion- 
ville 4.16 guns per 1000 men ; at the time of the capture of 
Metz this proportion was 6.8; at Beaune-la-Roland 6.4; on 
December 8d (1870), 8.8 and on the 9th of December, 11.4. 

The Second Army at the beginning of the World War 
in 1914 had 3 active corps ; 3 reserve corps ; an army cavalry 
corps of 2 divisions ; 2 Landwehr brigades for lines of com- 
munication duty; 4 battalions of mortars; 1 battalion 4.2 
guns ; 2 heavy coast mortar battalions, and 2 pioneer regi- 
ments. Each active army corps, 24 battalions strong, had 
at the beginning of the World War in each division 12 in- 
fantry battalions, 12 light batteries, of which 3 were how- 

238 - 



ARTILLERY 239 

itzers, besides 4 heavy howitzer battalions belonging to the 
corps artillery. The assigning of only 4 heavy batteries to 
the army corps was entirely too small in view of the super- 
iority of the French light artillery gun ; each division should 
really have had heavy artillery, in order to itself solve its 
own battle missions, especially when the question to be solved 
was to deploy the guns under the protection of the heavy 
artillery. The necessity for employing heavy artillery in 
the attack soon became apparent, although the regulations 
had assigned them only a mission in the defense. In for- 
mer wars, the fact that inferior infantry required stronger 
artillery was not given consideration. We can never foresee 
that during the development of war the inferior troops, such 
as the Landwehr and Landstrum organizations may not ob- 
tain the more dilBicult missions. At the beginning of the 
war the reserve corps, for example had no heavy howitzers 
and only 12 gun batteries. This was an entirely too weak 
artillery strength for any task. For lines of communication 
purposes the Landwehr brigades had a strength of 6 to 8 
battalions with 1 or 2 batteries mostly of an old type, but 
they were often called upon to fight alongside active corps 
equipped with entirely different material. In any case, re- 
serve and provisional organizations should have rather a 
stronger than a weaker proportion of artillery. The re- 
serve corps in October brought into the field, divisions with 
9 batteries, of which 3 were light howitzers. 

In the rapid and noteworthy increase of the army it 
soon became necessary to adopt the 4 gun battery, without 
any disadvantages appearing, such as had been feared in 
time of peace, and then also to adopt the 8 unit organiza- 
tions for the division infantry and to assign to each infan- 
try 6 light gun, and 8 light howitzer batteries. Finally, 
when position warfare commenced, it became necessary to 
assign high angle fire artillery to the divisions, and each 
division tiierefore received 1 heavy battalion of 2 heavy 
howitzer batteries, and 1 10-cm. battery. 

These formed the basic strength of the division and 
allowed exhausted divisions to be replaced by fresh troops. 
But it was desirable that the heavy batteries while under 
division control be assigned to the sector and left there 



240 DEVELOPMENT OF TACTICS— WORLD WAR 

unto relieved by the army artillery. In the development of 
oar tactics the experiences on the west battlefields had a 
major influence. In mobile warfare the battle was fought 
from concealed positions, exposed positions being excep- 
tional. An artillery fighting principally from defiladed po- 
sitions could not be neutralized, and only partially damaged. 
At the beginning shrapnel was the principal projectile. Its 
eflfect against vertical targets was annihilating, especially 
when these were in unsuitable formations, while the effect 
against skirmish lines taking skillful advantage of cover was 
remarkaUy low. The French artUlery distinguished itsdf 
by close co-operation between the artiUery and the infantry 
and by emplojring their guns up to the limits of their range. 
Hie German principles concerning combat proved to be cor- 
rect; but we soon felt the superiority of the hostile artil- 
lery, which could not be offset by the use of our high angle 
guns. 

Soon the strength of the hostile artillery forced us to 
organize an advanced guard artiUery with the usual dis- 
advantage of meeting the enemy suddenly and having to 
take unfavorable positions. For security on the march and 
during an advance against an enemy in position, single 
guns with the advance guard were sufficient, but not when 
a meeting engagement took place. It was necessary to co- 
ordinate the action of the infantry and the artillery. In 
reconnaissance it was first necessary to select observation 
posts, the gun positions being regulated on these. Often 
the caution had to be given that reconnaissance must not 
be so made as to draw fire upon our troops. In general the 
artillery had to fire over the heads of the infantry. It 
could not always be avoided that, at increased ranges, our 
troops were fired on by our own artillery, because of inat- 
tention or inability to determine the location of the dispersed 
formations. This bad feature could never be overcome in 
spite of all efforts. Our enemy also complained of the same 
trouble. 

In the attack the principal factor was to neutralize 
the enemy's artillery, and to place our fire with full strength 
on the hostile infantry, in order to permit our individual 
skirmishers to arrive within short range of the enemy's 
position without firing. Artillery fire to accompany the 



ARTILLERY 241 

infantry attack was indispensable ; caution had to be given 
as to carelessly taking exposed positions, as this, in the 
West often caused us heavy losses. In the defense it was 
advisable to decrease the effectiveness of the hostile fire 
and the possibility of being seen from the air by closely hug- 
ging the ground. The deplojrment of the artillery in a sin- 
gle line, to which we had become accustomed during peace 
time, quickly disappeared. The employment of single en- 
trenched platoons or sections in flanking positions, was soon 
adopted. 

The further development of the German artillery took 
place under the requirements of position warfare, shrapnel 
was replaced by sheU and also the range of the guns was in- 
creased with employment of high-angle fire artiUery; also 
the effect of the shdls was increased by the use of gas pro- 
jectiles. Germany paid especial attention to the develop- 
ment of high-angle fire artillery, whereas the Allies laid 
preponderance upon heavy, flat-trajectory guns. In addi- 
tion both sides took up the employment of railroad artillery 
on special railroad mounts, either for standard gauge or 
narrow gauge railroads, to be fired from specially con- 
structed sidings, for very long ranges. 

The effects of artillery fire was constantly increased 
by co-operation with the air service on observation of fire, 
reconnaissance and report of shots and by reducing and 
finaUy suppressing the emplojonent of trial shots, which 
were replaced by theoretic^ calculations based upon daily 
atmospheric conditions. This method was fully employed 
only in the last year of the war, and our enemy had a start 
over us in this respect. A prime requirement was an ac- 
curate map of the targets and the batteries. By the intro- 
duction of artillery position finding troops (flash and sound 
registry), who plotted observations of hostile batteries on 
a plane table (captive balloons could also be used to give 
data) results of observations free from all daily factors 
could be rapidly obtained. 

The trigonometrical determination of positions of bat- 
teries and infantry positions was the foundation for this 
work. Sound registering troops, who were independent of 
the terrain, could distinguish between actual aad f^lse posi- 
tions, although both sides sought to mislead such observa- 



242 DEVELOPMENT OF TACTICS— WORLD WAR 

tions. Also in open warfare, the obBervation of reports of 
sin^rle shots by sound registering and flash observation troops 
of the light position finding troops, could be counted upon 
within three to six hours. It was necessary for these troops 
to have a reliable information service and the trigonometri- 
cal position of hostile batteries and observation posts, 
promptly transmitted to the proper headquarters. 

In the defense timely steps could be taken to trigonome- 
trically determine suitable points along lines in the battle 
zone leading towards the rear, from which points measure- 
ments could later be made. 

CO-OPERATION OF INFANTRY AND ARTILLERY 

It was found that co-operation between infantry and 
artillery was very important. During the Franco-Prussian 
War, it was a matter of honor for our artillery to have a 
proper understanding with the infantry based on visual ob- 
servations. Such support was possible as the battles took 
place at short ranges and as both sides could readily be dis- 
tinguished by their uniforms, which, in spite of the clouds 
of smoke that lay over the firing lines, were clearly visible 
to the batteries fighting in the open. AD our pre-war regu- 
lations required co-operation between the arms but it did 
not explain the means or methods by which this was to be 
accomplished ; the infantry relied on the artillery, and the 
latter believed that it was the duty of the infantry to call 
upon its sister arm at the proper time for assistance against 
a designated objective. This question had received special 
consideration by our enemies and a correct system had been 
arranged to secure co-operation between the two arms in 
which nothing was to be left to chance. In supporting the 
infantry the artillery must keep down the fire of all hostile 
troops, both infantry and artillery, directed on its own in- 
fantry. To be practicable such support had to be given at 
particular moments, which only the attacker could deter- 
mine, and also on particular points, which could be partly 
designated by the attacker but which, on the other hand, 
were partly dependent upon the movements of the enemy. 
On the defensive, the infantry must dsignate exactly the 
targets on which its artillery is to fire. A good system of 



ARTILLERY 24S 

liaison is therefore required. The foundation for such a 
system is obtained through orders and battle plans which 
designates the task and place for co-operation between 
designated infantry and its supporting artillery. In gen- 
eral the method of issuing orders remains unchanged, but 
exceptionally designated artillery units may be placed, in 
close terrain, under the infantry, and always so for accom- 
panying artillery on special tasks such as, village fights and 
suppression of machine gun nests. In the absence of a 
battle plan, co-operation between infantry and artillery is 
seldom possible even if both parties do their best to obtain 
it. In such cases, in vigorous rivalry, each arm looks after 
itself, leading to a lack of co-operation. 

The basis for combined tactics is the battle plan; co- 
operation between the two arms proceeds by mutual agree- 
ment between subordinate commanders through which the 
nature and location of targets is designated, and by direct 
requests from the battle line to obtain artillery fire at a 
designated point at a definite moment. If artillery fire con- 
trol is not to be completely dropped, then only previously 
designated batteries should receive such requests, not from 
each infantry platoon commander but from the higher in- 
fantry commanders. These agreements must be made be- 
tween the two headquarters. In no case must the infantry 
allow itself to be misled, by not taking advantage of f avora-* 
ble situations when there is no artillery support available. 
But the artillery must depend on its own observations, even 
when the infantry does not request special support. Thus 
battle plans and observations supplement and support one 
another. The battle plans operate only for prearranged 
tasks. Even with the assistance of the best means of com- 
munication, co-operation between the two arms can be at- 
tained only when the commanders have thoroughly studied 
the method thereof, and have determined the best method 
through combined exercises. Without special preparations, 
co-operation of the arms depends entirely on the good will 
of participants, and this always leads to disillusionments 
and losses of time. 

The infantry can only be properly supported when it 
can make its actual battle lines known to the artillery be- 



244 DEVELOPlfENT OF TACTIGS— WORLD WAR 

yond any doubt As to location of the front line by the Air 
Service, a buminff of colored liffhta and the exhibiting of in- 
fantry panels upon demand of the infantry airplane, as well 
as at pre-determined hours, has proven valuable; photo- 
graphs can be utilized only after a longer time. Good re- 
sults in marking front lines has been given by flags about 
60 cm. square, white, black, red, red-yellow or ydlow-red, 
but the side of the flag facing the enemy must be of some 
neutral protective color. The English prescribed that these 
flags were to be paid attention to only when they were 
moved back and forth, or when shown at pre-determined 
times; it was feared, not without good reason, that flags 
left standing by the enemy might lead to errors. As a mat- 
ter of course, the front battle Une only was to be so marked. 
It was found advisable upon occasions to bum bengal flares 
when the line designated in orders had been reached. 

In the autumn offensive in Champagne in 1916, the 
French infantry wore linen cloths on their backs, to prevent 
their own artiUery from flring on them. This method was 
not successful, as in spite of it, the infantry were heavily 
fired on by their own artillery during the advance. Mistakes 
can never altogether be avoided; we can only attempt to 
reduce them to a minimum and for this, information as to 
the plan of attack and good observation of the battlefield is 
necessary. Often enough, in close terrain, single detach- 
ments coming to the rear will be taken for enemy detach- 
ments advancing. Besides signals for ''fire in front of us," 
which should be different for different units of the army, 
there should be the signal ''We are fired on by our own ar- 
tiUery !'' Often will the infantry mistake hostile shell frag- 
ments flying to their rear, or shells coming from a flank, 
for bad markmanship of their own artillery. To avoid such 
errors during position warfare, artillery trench patrols 
were sent forward, and even better was the sending of bat- 
tery commanders or battery officers to the infantry trenches. 
They here had opportunity to verify their own observations, 
to discuss matters personally with the infantry; to hear 
their requests and complaints, and to exchange information 
with them. Thus it was easy to secure co-operation of the 
arms, the infantry learned to meet the artillery half way 
in providing liaison, to take care to have good observation 



ABTILLEBT 246 

positions, and to assist observers materiaUy by giving them 
their own observations. In any case an artiUery observer, 
posted with general and superficial instructions, and to 
whom no one thereafter pays much attention cannot accom- 
plish much in supporting the infantry. When a critical sit- 
uation arises, where the infantry does not find the expected 
support, the infantry should not blame the artillery, but 
should seek the reason for the failure in their own conduct- 
In this manner artillery liaison officers sent to the front 
can perform good reconnaissance service. 

Co-operation between the artiUery and the infantry 
rests on artillery liaison officers who should be liberally sup- 
plied with maps and means of communication. The prdimin- 
ary instructions given them should refer mainly to the battle 
missions of all batteries which are the barrage batteries 
for the sector, what batteries are available for fire within 
the sector, and finally how these batteries can be reached. 
BInowledge of the capability of the various guns is naturally 
necessary. As these officers have knowledge of the orders 
and information received by the infantry, they are in a posi- 
tion to recommend proper artillery support, inform their 
own commands as to the general situation, so that they may 
quickly prepare for coming missions. Their duty is espe- 
ciaUy to orient both arms in mutually informing them as 
to the whole situation. Frequently, it is desirable to com- 
municate directly with the f orwaid artillery observers, es- 
pecially when targets are to be fired upon on request of the 
infantry. 

The duties of observation and liaison officers are sharidy 
separated, even when observation officers are in a position 
to transmit requests from infantry for artillery support. 
It is well to attach to each infantry battalion headquarters 
an artillery liaison officer, and supply him with necessary 
information. Never can the artillery count on using the 
infantry lines of communication which are always over- 
worked. In position warfare the liaison officer is detailed 
permanently to a battalion, and lives permanently with the 
battalion staff ; in quiet sectors he will often be charged with 
the service of observation also within the battalion sector. 
The heavy artillery will have one artillery liaison officer de- 
tailed to each regiment; it is very desirable that artillery 



246 DEVELOPMENT OP TACTICS— WORLD WAR 

groups should conform to infantry subsectors^ the co-opera- 
tion of arms is much enhanced thereby. If the respective 
artillery commanders are in the vicinity of the headquarters 
to which they are attached, this wiU facilitate personal con- 
sultation of staffs and co-operation between the arms. Even 
when it is expected to have to meet an attack, timely consul- 
tation between the respective infantry regiments and the 
corresponding artillery units is thereby facilitated. 

CO-OPERATION BETWEEN ARTILLERY AND 

AIR SERVICE 

Close co-operation between the air service and artillery 
in order to locate targets, and in order to observe fire, is 
always a prime requirement for success. It is always very 
desirable to attach air service officers to the staffs of artil- 
lery commanders. Artillery airplanes supplement the ser- 
vice of the flash observation and sound registering troops ; 
air photographs make it possible to definitely determine the 
effect of fire, by proving in a short flight the accuracy of 
the artillery adjustments and the results of the fire. Of 
the greatest importance is information of those hostile bat- 
teries which have been firing most or have been silenced, as 
weU as good or poor fire by our artillery. Only those air- 
planes which are equipped with radio sets can remain in un- 
interrupted communication with the^ artillery without stop- 
ping their flight. Firing for effect with aerial observation 
requires a previous understanding and good communication 
between the observer and the battery commander. Previous 
discussions cannot cover long engagements and in such 
cases each observation airplane must, through its. radio, 
call on batteries for fire. Artillery equipped with radio for 
receiving information, and the air service must have a firm 
determination to co-operate ; in such cases all difficulties will 
be overcome. On the defensive it is necessary to designate 
certain batteries to receive information from the observa- 
tion airplanes, which batteries should have particularly good 
communication. The positions of these batteries must be 
such that it will be possible for them to cover as much as 
possible of the battlefield, their range should be such that 
they will be independent of the movement of the infantry 



ARTILLERY 247 

in order to avoid frequent changes of position and arrange- 
ments must be made with the air service at each station un- 
der all kinds of weather, so that quick decoding of radio 
messages may be made and transmitted to the proper head- 
quarters or directly to designated batteries. Should no in- 
telligence messages be received for his own command the 
radio officer must listen in on messages from airplanes to 
neighboring units for his own orientation. 

THE DECISIVE BATTLE IN POSITION WARFARE 

• 

In the first months of the war, an insufficient supply of 
ammunition made difficult the systematic employment of the 
artillery in the attack. The artillery on the defensive was 
divided into two classes ; guns in concealed battle positions 
and forward guns ; the latter were often in the first line in 
masked or covered positions and were to fire only when the 
hostile assault took place. These guns were generally de- 
stroyed early and a belief arose in an automatic defensive 
barrage, which was to be brought down by means of bengal 
light signals. The front of a defensive barrage for 1 bat- 
tery was not to exceed 200 meters ; but such a width was 
generally exceptional. This proceeding misled the infantry 
into relying too greatly on the artillery support, and to call 
too frequently for this support. Through the loss of guns 
during the enemy's artillery preparation, the defensive bar- 
rage was never as strong as intended, often came too late, 
and without endangering the defender could not be brought 
close to the front line trenches. Even when defensive bar- 
rages lasted only 2 to 3 minutes, it increased the expenditure 
of ammunition. In the same way reprisal fire did not serve 
any good purpose. 'It was designed to square the account 
with the hostile artillery by annihilating fire on the re? 
spective sectors of the opposing hostile lines and so to pre- 
vent him from deploying and advancing to his attack posi- 
tion.^' This was the method we stUl adhered to in the 
battle of the Somme. This procedure was of no value. We 
found out that air superiority and the efficiency of the posi- 
tion finding troops was essential in the artillery duel. New 
regulations prevented reprisal fire, substituted annihilating 
fire for defensive barrages, and prescribed for each kilo- 



248 DEVELOPMENT OF TACTICS— WORLD WAR 

meter front an artillery strength of 6 to 7 barrage batteries, 
4 to 6 high angle batteries, and 1 to 8 heavy gun batteries. 
The strength of the artillery defense does not lie in 
making the barrage and annihilating fire denser, but in at- 
tempting to foresee the enemy's movements. In a new edi- 
tion of the Regulations dated September 1, 1917, this idea 
was further discussed, and no mention was made of a sys- 
tematic calculation of the artillery strength. It was re- 
quired that timdy preparations be made, when the artillery 
deployment is undertaken, as to the artillery battle, without 
regard to the damaging of guns during the artillery battle, 
because such a battle would at least delay and weaken, possi- 
bly decisivdy, the enemy. The difficulties lay in the nature 
of the targets; well concealed batteries can withstand a 
heavy bombardment ; in the greater number of and mobility 
of targets, the enemy made more frequent changes of his 
battery position and of his artillery reserves than we did; 
finally there was the difficulty of observation. As terres- 
trial observation is in most cases impracticable, artillery 
superiority is closely connected with air superiority. Very 
naturally two groups arose, one for long range fire into the 
enemy's back areas and the other a close defense group, 
just close enough to take under fire the enemy's battle 
trenches. These designations were not well named, the 
close defense group was as a matter of fact the furthest to 
the rear. Grouping of the artillery in several lines was 
recommended, the reinforcing batteries, which arrived gen- 
erally only during the progress of the battle, being placed 
in the second line. Orders must, from time to time, be given 
as to how long the artillery battle is to continue against a 
threatened attack, and where and when the annihilating 
fire is to be laid down. In any case the general principle is 
that the artillery fire should be laid down as late as possible. 
It was much to be desired that our artillery should be strong 
enough to lay down annihilating fire against the hostile in- 
fantry and at the same time, with a part of it continue the 
artillery duel. For such a purpose there was needed heavy 
gun batteries, from 18-cm. caliber upwards, which deployed 
in depth, could with assistance of special balloons and air- 
planes, undertake to neutralize distant hostile batteries. By 
separating the artillery into a distant and close defense 



ABTILLEBY 249 

group there was a natural depth formation. Against a pene- 
tration this greatly decreased the effect of surprise, as well 
as furnishing flank support for adjacent sectors. The great 
diflSculty in the transmission of orders and information soon 
became very clear. 

Fire against the hostile infantry is the second mission 
of the artillery in the defensive battle. In this we must be 
free from the thought that it is possible, or necessary, to 
destroy all of the hostile infantry positions. The question 
is, during the development of the hostile attack, to select 
the tactically most important targets such as infantry, mor- 
tars, command and observation posts, support trenches, 
bomb-proof dugouts ; to fire either in order to destroy them 
or else prevent their occupation. Daily bombardment of 
unoccupied trenches is useless. 

The artillery defense against assault is by pre-arranged 
mass fire, in more or less rigid form, annihilating fire, or 
barrage fire. Annihilating fire is directed against the rear 
echelons of the attack, if possible against supports of the 
assault troops ; the barrage is laid down on the enemy's ap- 
proach lines to strike down the advancing infantry. With- 
drawing barrage fire to our own front line should 1^ avoided 
due to the fact that our infantry, at the moment of the hos- 
tile assault, must be saved from the demoralizing effect of 
short shots coming from their own artiUery. Where the 
hostile trenches are only a short distance away and when 
artillery cannot cover dead angles in front of our trenches, 
the artillery fire is replaced by trench mortars, and rifle or 
hand grenades. In these cases, the artillery directs the mass 
of its fire in rear of the foremost hostile lines. Trench mor^ 
tars are under the command of the artillery to insure pro- 
per regulation of the barrage. When the opposing trenches 
are far apart, the higher commanders prescribe the line in 
No Man's Land on which the barrage fire is laid ; it is best 
to have this line where natural obstacles will delay the 
enemy. It must be possible to lay down the barrage on all 
parts of our front and in front of adjacent sectors. Should 
the enemy penetrate into our position, supporting the in- 
fantry fight, is difiicult. It will often only be possible to 
limit the amount of penetration and by firing gas shells at 



250 DEVELOPMENT OF TACTICS— WORLD WAR 

the hofltUe artillery prevent tbem from assisting the at- 
tack. A mobile artillery reserve is of great advantage; it 
is brought up and employed as prescribed for open warfare. 

For calculating the strength of the artillery we have 
but few rules to go by. The number of guns required is 
naturally dependent on the enemy's artillery strength, but 
this can only be determined by its effects. It is then usually 
too late to provide reinforcements. In such cases, we would 
have to renounce the superiority in the artillery fight. The 
strength of the field fortifications also has a material effect 
as to the strength of the artillery. The weaker our position 
may be the greater should be the proportion of guns. The 
commander must start from a definite basis ; it is never pos- 
sible to be too strong in artillery. As a guide it was pre- 
scribed that for each 400 meters of hostile trench, 1 high 
angle fire battery for annihilating fire was to be provided, 
and that the barrage fire was to be laid down by machine 
guns, light trench mortars and the infantry. For destroy- 
ing 1 battery, according to Par. 254 Battle Regulations for 
Artillery, 300 rounds from heavy howitzers or 200 rounds 
from mortars were required ; and by Par. 268, same Regula- 
tions, for each division sector at the commencement of the 
attack, 1 to 2 mortar batteries ; 2 to 3 heavy howitzer bat- 
teries and 1 to 2 heavy gun batteries were required for the 
artillery battle. Light artillery was used against casual 
targets. The difficult situation of our troops compelled us 
to operate generaUy with materially smaller forces than 
our enemy. Usually in quiet sectors for each kilometer of 
front, each division besides its 9 light batteries had 2 to 4 
heavy howitzer batteries, 1 mortar battery and 2 heavy 
gun batteries. This was not enough. It did not suffice at 
all and besides, the division sectors were generally notice- 
ably wider than given above. 

No steps had been taken in Germany to prescribe a 
rule for the occupation of a battle front by the artillery. 
This was determined by the strength of the hostile artillery 
and the method of conducting the battle. In active sectors, 
when an attack was foreseen, there was provided for about 
each 4 kilometers front (when a counter-attack division was 
provided for each 2 division sectors whose artillery was 



ARTILLERY 261 

available) the following artillery was provided for each kilo- 
meter of front : 

8.5 to 5 light fiTon and 1 light howitser batteries, 
1.5 to 2.5 heavy howitser batteries, 
1 to 1.5 10-cm. gun batteries. 

This made a total for a 4 kilometer sector of 14 to 20 
light batteries, 6 to 10 heavy howitzer batteries, and 4 to 6 
heavy gun batteries. In addition, it is desirable to have an 
artillery reserve, and a mobile artillery repair shop in rear 
of the front. 

THE OFFENSIVE 

The first attacks were, to be true, erroneously based on 
destro3ring the hostile artillery in preparation lasting many 
hours, surprise being unobtainable. Until we came to the 
point of neutralizing the hostile artillery at the moment of 
the assault, when we decided to do without bomb-proof cover 
for the battery positions, and when we were able to avoid 
range finding and trial shots by substituting therefor theo- 
retical calculations based on daily weather conditions and 
accurate measurement of distances between battery and tar- 
get, then only could the high command count on the possi- 
bility of obtaining surprise for the decisive action. This was 
made the easier by the number of battery positions that 
had been constructed along the entire front during the 
years of war, and subsequently by the introduction of 
tanks. 

Batteries designated for counter battery work can be 
kept back; the artillery employed to support the infantry 
assault has, however, to be held close in hand, and these 
should, as long as possible, utilize the full possibilities of 
range and not make a change of position, especially when 
a counter-attack may require artillery support for the in- 
fantry. Only effective support will keep the assault going; 
as a matter of experience the infantry only gets as far as 
its artillery fire reaches. As a general rule it must be held 
that the artillery preparation should be as short as possi- 
ble ; with one blow the resistance of the enemy must be over- 
come. Almost always will there be the accessory mission of 
destroying the hostile liaison and intelligence communica- 
tions. 



262 DEVELOPMENT OF TACUGS— WOBLD WAB 

Gm dieDsy whieh at the dose of the war were loaded 
wtth st rcmg er bunting chargee, were the more effective aa 
thqr were the lees perceived by the senses (being odoriesa 
and colorless). After a ''green cross'' bombardment Ihe 
terrain can again be entered within two hours. The effect 
of '^yeDow cross" gas shell is noted gradually for hours and 
days, the effectiveness continues a long time in good weather, 
especially in villages and woods. 

In a large attack for penetration, the following missions 
fUl to the artillery: 

1. Dcstmetioii and nestnllmtkni of bostils mrUSkirj and 
trtDch BMirtan. 

2. Neatralixatioii of the hostile trcneh and suppcnrtiiig poai- 
timis. 

2. Firing on w a e ivoa and rear conmimieations, on obaerrap 
tion and eoounand poots, parick lialloons, transportation and 
mimition eentera in rear beyond tno objeetiTe of the attadc 

4. Supporting tiie aanvlt by a rolling bamge. 



6. Acoompanying tiie infantry aesanlt witn infantry gmis 
light arfiUenr to !m ' 
direct and doee sheQ iire. 



and light arfiUery to hrealc np local centers of resistance 1^ 



S. Protecting the infantry after reaching the dbjectives by a 
standinff barrage. 

7. Repulsing hostile eoimter^ttacks, and distant approach 
of 



For counter battery work at the beginning of the at- 
tacks there should be provided for each division sector for 
observed destructive fire, 1 to 2 mortar batteries, 2 to 3 heavy 
howitzer batteries, and 1 to 2 heavy gun batteries; this 
number must be reinforced for neutralization of the hos- 
tile artillery in order to obtain a quicker success. Only sel- 
dom will the general situation justify a delay. It is better 
to employ over the entire terrain gas shell and high ex- 
plosive shells to neutralize the enemy's activities. In gass- 
ing areas the quick fire guns of the light artillery are more 
suitable than the heavier projectile from the bigger guns. 
In order to neutralize 1 hectare, by gas, the f oUowing ammu- 
nition is required : 

100 rounds light artillery, or 50 rounds light howitzer, or 

10-cm. guns, 
25 rounds heavy howitzers, 
10 rounds mortars. 

In order to neutralize by gas 1 square kilometer from 
1 to 2 hours, there is required 10 light batteries, 17 10-cm. 
batteries, 12 heavy howitzer batteries and 10 mortar bat- 
teries. 11 the enemy's artillery is contained in a zone 2000 



ARTILLERT 258 

meters wide and 800 meters deep, or in area of 1.6 square 
kilometers there would be required for neutralizing by use 
of gas: 

12 light iMitteriM (of whieh S tettoriai hare not proviouily 

fired), 
8 lijriit howitier batteries, 

2 IfheuL suns liatteriee (neither hayiw prenonsly fired), 
6 heavy howitwr batteries (1 to open tm for surprise effeet 

later). 

At the moment of the assault these batteries neutralize 
hostile artillery with gas and high explosive shells. In this 
mission attack airplanes can greatly assist by attacking hos- 
tile batteries with machine gun fire, and by bombing. Single 
batteries remain available for the second period as surprise 
batteries, waiting for the attack to develop in order not to 
be previously known to hostile barrage batteries. The En- 
glish Regulations for the defensive against a strong attack 
provided for special ''silent*' batteries ; the French similarly 
provided ''dumb'' batteries. 

To prepare a position for assault for each 100 meters 
of front penetration, 1 high-angle fire battery is needed. 
Places which are not to be assaulted are neutralized by the 
smaller caliber gun. A part of the preparation is taken over 
by the trench mortars with 1 heavy or 1 medium trench 
mortar for each 40 meters front. For 8 division trench 
mortar companies, containing all kinds and including 12 
heavy and 24 medium trench mortars, with 10% of spare 
mortars, a front for penetration of 1600 meters can be as- 
signed, and in 1 hour they can fire 280 heavy and 900 medium 
trench mortar bombs. As for the artillery, where the second 
line must cover neighboring sectors, they must cover with 
their fire 1200 to 1500 meters of the front to be penetrated. 
Details cannot be given here without considering a definite 
case of attack. Par. 267, Battle Regulations for Artillery, 
prescribes the following ammunition as required for the 
artillery preparation preparatory to an assault: 
In about 8 hours a position can be prepared : 



Ml 



One battery One line of trenehee Number of 

width in meters rounds 

Light Howitzer . . 100 800 

Heavy Howitser, Model 18 160 600 

Heavy Howitzer, Model 02 100 400 

Heavy Howitier, Model 96 75 800 

Mortar (8 gmuB battery) 100 226 



254 DEVELOPMENT OF TACTICS— WORLD WAR 

'The above ammunition includes that required for the 
destruction of bomb-proofs ; also for the destruction of ob- 
stacles which are not more than 30 meters from the line 
of trenches fired at. Against more distant obstacles, special 
anununition and time must be allowed. In a contrary 
case their destruction will be accomplished through the 
dispersion of shells directed at the trenches. By use of 
flank fire, which also always gives a moral effect, 'the 
amount of ammunition prescribed above may be reduced one- 
third. Whether all the lines of defense of a position are neu- 
tralized simultaneously or successively will depend upon the 
number of available batteries.'' 

There wiU be required for example : 

8 light batteries each covering 100 meters front firing 800 

rounds each, giving a total of 800 meters front with 2400 

rounds. 
7 heavjr howitzer batteries each covering 100 meters front 

firmg 700 rounds each, giving a total of 700 meters front 

and 4000 rounds. 
4 mortar batteries each covering 100 meters front firing 400 

rounds each, giving a total of 400 meters front and 900 

rounds. 

However, this number of batteries alone is not sufficient 
to prepare one position for assault. Adjacent sectors must 
also as far as possible be neutralized by use of gas or smoke 
in order to prevent them from using enfilade fire or making 
counter-attacks. A secondary mission is the firing on the 
positions in readiness of the reserves, lines of approach, 
command and observation i>osts, munition dumps, balloon 
ground stations. 

As the main point of defense rests not in the less easily 
found intermediate lines but in the less conspicuous built 
machine gun nests between lines, General Nivelle in 1916 
introduced the rolling barrage. This after the first jump 
was to permanently precede the infantry at the rate of 100 
meters in about 4 minutes and was to halt for a time be- 
yond designated attack objectives as a standing barrage. 
In this manner, all defensive positions in the entire terrain, 
such as machine gun nests are effectively neutralized until 
the infantry with their own weapons can attack them them- 
selves. Frontal counter-attacks should be broken up by the 
barrage fire. But it is necessary that the infantry should 



ARTILLERY 266 

closely follow the rolling barrage, that the infantry should 
not stop its advance, which would result in the rear lines 
over-running the front lines, nor that the advance should 
proceed too rapidly. The latter case is the most unfavora- 
ble, as then the positions being attacked would be free from 
fire too soon, and machine guns and reserves could then 
undertake a counter-attack. The limit for the rolling bar- 
rage is the effective fire of the guns which is about 4000 
meters. This requires the artillery to be posted immediately 
in rear of the line of departure, and then that the batteries 
advance rapidly to the position penetrated. If the rolling 
barrage is to provide effective protection for advancing in- 
fantry, the latter must adhere exactly to the prescribed 
program and closely foUow the points of burst of the shells. 
Consequently, it follows that in the first line, gun batteries 
are most suitable, that howitzer shells may be employed as 
a pace marker but cannot be used in the barrage on account 
of the backward flight of the splinters of their shells. In- 
fantry can be taught to overcome the fear of closely follow- 
ing the barrage by thorough training, in which at the start 
they follow the barrage to one side. The closer the infantry 
follows the better. The best formation to follow is small 
narrow columns, so as to reduce losses from back flying 
shell splinters. Attempt to regulate the advance of the 
rolling barrage by the advance of the infantry has not been 
successful. 

It was found that often in variable terrain the rolling 
barrage had been set at too fast a speed, so that it got away 
from the infantry and could not be stopped. It therefore 
was determined that the barrage would advance slowly from 
one line to another. In general 1 light battery is needed 
for 100 meters front. 

The infantry must never rely on rolling barrage alone, 
but must advance through the strength of their own weap- 
ons. EspeciaUy must they depend on the infantry accom- 
panying guns. Each regiment of infantry has at least 1 
accompanjdng battery assigned to it, which is entirely de^ 
tached from the artillery command, and which is distributed 
among battalions by sections or platoons, each gun having 
2 caissons filled with shells. It is to be noted that the guns 



266 DEVELOPMENT OF TACTICS— WORLD WAR 

of reserve battalions are given 2 extra caissons per gun for 
replenishing anununition. These form an ammunition sec- 
tion. Considering the great demands, selected teams must 
be provided, and for each gun and each caisson at least 6 
extra horses are necessary. 

The possibility of watering the horses during the ad- 
vance must not be lost sight of. The quickest method for 
crossing trenches is to fill them in. The battery comman- 
der remains with the infantry regimental staff, but may 
temporarily join his guns. Sending junior officers to in- 
fantry regimental staffs is never advisable, as they may not 
receive the same respect, and may have less tactical train- 
ing and experience. The battery commander will recom- 
mend to the regiment to have necessary roads marked by 
streamers, by signs or by paper markers. The attacking 
infantry cannot, by any means, be called on to furnish labor 
detachments, but the division should attach at least 1 pio- 
neer squad to each gun and to each caisson. For each man 
1 long handled shovel or 1 pick axe is required. The in- 
trenching tools on wagons serve as a reserve. In addition 
materiel is carried along for camouflage against airplanes 
(wire netting is too heavy and too rigid to shape) . By us- 
ing natural cover (as for example shell-holes for the teams) , 
and guns protected by camouflage nets against airplanes, 
they are ready to have teams quickly hitched and accom- 
pany the attacking infantry. Occupying a position in readi- 
ness can ordinarily be done at night only after the roads 
have been carefully reconnoitered, so as to avoid unnecessary 
fatigue for the teams. About 15 minutes before the concen- 
tration or assault the teams are hitched up, otherwise the 
guns will only be left behind. Crossings over our own 
trenches (in this case it is best to utilize bridges provided 
in advance) are to be made during the artillery prepara- 
tion. Crossing places must be watched by details and kept 
in proper repair. The difficult movement across shell torn 
terrain while at the same time keeping up with the infantry 
behind the barrage, is hard for the artillery. 

The gun commander (to be generally an officer when 
single guns are detached to battalions) remains with his 
vehicles, and sends a nonconmiissioned officer forward to 



ABTILLERT 267 

the battalion to prepare timely information through patrols 
of machine gun nests to be fired at, as these are the princi- 
pal targets. Whether it is practicaUe to fire over the heads 
of the infantry in front depends on the situation ; often the 
infantry on pre-arranged signals will have to clear the field 
of fire of the guns. Fire is opened only when the infantry 
can no longer advance or when hostile counter-attack is 
noted, or when tanks start to attack. Generally the fire wiU 
be at short ranges and from open or partly covered posi- 
tions. Positions from an elevation above that of the in- 
fantry which allow of direct fire without danger to the in- 
fantry are especially valuable. A caisson is unlimbered 
near each gun, but the ammunition is not removed there- 
from so that no loss of time occurs when the advance is re- 
sumed. The following dispositions of the battery are gen- 
erally suitable : 

1. The battery commander with his special detail remains 
with the infantry regimental commander. 

2. 1 officer and 1 noncommissioned officer is with each gon, 
and 2 caissons and 18 extra horses witii each battalion acting as 
accompanying: artillery in the front line. Battle missions are: to 
support the infantry in the first line against machine gans and 
tanks. 

8. With the reserve infantry battalion: 2 guns and 6 cais- 
sons. Battle missions: advance from one covered position to 
another prepared to defend against counter-attacks. 

The light ammunition colunm must plainly mark its 
position; agents sent to the rear by batteries to obtain 
ammunition are to be directed to guide caissons to their 
batteries. 

The losses in personnel of accompanying batteries have 
been relatively small, while the loss in animals has been 
often very great, without, however (thanks to providing 
sufficient reserve horses), interfering with the mobility of 
the artillery vehicles. 

The inconvenience during the advance of the infantry 
in the rolling barrage becoming gradually thinner, can only 
be overcome by the timely advance of the artillery. This 
then gives a second artillery approach march having the 
following features: early special reconnaissance and air 
observation, bringing forward and installing the artillery 
position finding troops and providing artillery orientation, 
observation service, firing positions and the quick construe- 



268 DEVELOPMENT OF TACTICS— WORLD WAR 

tion of lines of communication, the re-establishment of the 
battery stations, orienting battery positions and aiming 
points, selecting targets, and determining ranges and organ- 
ising the meteorological service for the artillery. So far as 
concerns a new prepared attack: drawing up the orders 
for the development and advance of the batteries, the am- 
munition supply and location, the battle missions for single 
guns, barrage tables, etc. 

AU the foregoing tasks are supervised by the division 
in accordance with instructions from higher authority and 
the preparations are to be completed before the first pene- 
tration. The carrying out of the order must proceed as 
soon as the first line stops, without waiting for information 
and decisions as to further intentions. If reconnaissance, 
range finding, etc., proves to be premature, or unnecessary 
on account of the attack continuing forward or because the 
attack has been discontinued, the fact that the artillery for- 
mation has been changed, etc., is of no importance; it is 
better to start these preparations early rather than too 
late. The systematic carrying out of missions must not be 
interfered with by unnecessary rapidity of movements. 

Altogether the artiUery for an attack on a division 
sector whose zone of action was 2 kilometers wide is as fol- 
lows: 



I ^ i 



i I i i S 

3 S ^ :; ^ ;^ 

Neutralizing hostile artil- 
lery with gas 6 8 8 

Surprise batteries (may 
also be used for firing 

gas) 6 2 1 

Standing barrage fire 4 2 

Artillery for preparatory fire 5 7 4 

Accompansring artillery 8 10 

Total 16 9 6 2 16 4 

We thus have for each division sector 24 light bat- 
teries, 8 heavy howitzers, and 2 mortar battalions with 26 

batteries. Altogether 62 batteries. Before the World War 



ARTILLERT 269 

no one of us thought of the amount of ammunition. Orders 
for the deplo]rment of artillery, for its occupation of posi- 
tions, and for furnishing ammunition could be developed 
only gradually. In 1917 the English for 1 division sector, 
with a zone of action, 2 kilometers wide, had artillery of 8 
divisions, 24 light batteries and 33 heavy batteries. Luden- 
dorff provided for each kilometer front of attack 20 to 80 
batteries, that is for a division zone of action 40 to 60 bat- 
teries. Grenerally in most battles 3 divisions were posted, 
one in rear of the other, each provided with 13 to 20 bat- 
teries. The question as to the proportion of the artillery 
is no longer: ^'How many guns for each thousand men 
should be provided ?" but far rather : 'llow much infantry 
will be required to utilize the success of the fire of the ar- 
tOlery?'' But is not the infantry stiU the principal arm as 
heretofore? It is no use to argue this matter. General 
Ludendorff writes in his memoirs: 'The artillery sufifered 
as much as the infantry. The longer the war lasted, the 
greater were its losses in the defense as well as in the at- 
tack ; it became more and more the bearer of the burdens of 

the battlefield and the hold-fast of the front It 

is right not to dispute the question that the infantry is the 
main arm. There are no longer principal arms. Each arm 
has its use, all are necessary.'' 



XII. The Year 1918 

TRANSITION FROM POSITION TO 
MOBILE WARFARE 

The year 1917 finally brought the campaign in Russia 
to an end. The troops relieved there furnished the means 
for an offensive in the West, in which the experiences gath- 
ered concerning the penetration battles in Eastern Galicia, 
on the DUna, in Roumania and in Upper Italy found excel- 
lent use. The general situation demanded the offensive, 
which the troops also greatly desired. 'In the same meas- 
ure in which the defense depressed the troops, the offensive 
raised their morale. In defense the army had finally to suc- 
cumb to the hostile superiority in men and munitions of 
war, while the offensive had always been the strongest role 
of the army.*' The attack is the most powerful form of war, 
and brings the decision; and a modem defensive battle is 
far more costly than the attack. With a superiority of 
from 25 to 30 divisions over the enemy, the offensive could 
be started. 

When thus the situation of the German armies de- 
manded the offensive, the principal question could, however, 
be only that of a penetration, considering the extensive of 
the hostile front, from which action there could follow later 
on, in case of a favorable decision, im enveloping attack of 
the separated hostile wing. 

The penetrating attack had been about rejected in Ger- 
many prior to the war as, of course, it was harder to start 
and harder to execute, and certainly furnished smaller re- 
sults than an enveloping battle as Count Schlieffen had de- 
scribed it in his "Cannae" procedure. From the ill-fated pene- 
tration at Liaoyan it was concluded : "The ill-success only 
furnishes proof that with modem fire effect and modem field 
fortifications tactical penetration has little hope of success.'' 

The question as to whether in the selection of the point 
of attack strategic demands should be placed above tactical 
requirements, — ^the afiirmative being demanded by the 

260 



THE BATTLES IN 1918 261 

"Criticism of the World War" but denied by Ludendorflf— 
cannot be discussed here. I agree with Ludendorff in his 
statement: 'Tactics should have been placed above mere 
strategy. Without tactical success strategy could not be 
accomplished. Strategy which does not think of tactical 
success, is condemned in the very start to failure. Numer- 
ous examples for this were furnished by the attacks of the 
Entente in the first three years of the war." 

In the penetration we could attempt to break in at a 
favorable point for maneuver, to annihilate the reserves 
streaming together here, and to bring to an end through this 
victory the operation in the quickest possible way; or, we 
could, through operations intended to deceive, draw the re- 
serves away from the decisive point, and utilize our super- 
iority in numbers to enlarge the tactical entry and to con- 
duct the final battle against the gradually appearing re- 
serves. Brussilow adopted this procedure in Eastern Asia, 
his false operations extending to the lower Diina, his pene- 
tration starting on the right wing of the attack zone and 
then continuing south. A German counter-attack broke 
the Russian offensive. 

The German leadership selected the first, and shorter 
road, as it hoped ; it had learned by experience how very diffi- 
cult it is, in view of the great development of aerial fighting 
forces to deceive the hostile leadership by trying to execute 
double attacks separated in time and space. Preparations 
for further attacks took place along the entire front, as well 
as feints on a smaller scale. The German leadership had 
made its decision ; all it had to do now was to see that the 
attacks would not come to a standstill at strong hostile posi- 
tions. Did the forces suffice to have a second attack imme- 
diately follow the first ? "It will be a giant struggle," writes 
Ludendorff, "commencing at one point, continuing at another 
and which will take a long time." 

However, the infantry had a far different value than 
had the infantry with which we executed the offensive at 
the opening of the World War. Battalions and companies 
were commanded by young, for the greater part by very 
young, officers who probably had had by then experiences 
in the field, but who still lacked those qualities which had 



/ 
/ 



262 DEVELOPMENT OF TACTICS— WORLD WAR 

formed the strength of our officers in peace. Thoroughly 
trained to meet all situations, indefatigable in their care f(»* 
the subordinates, an example in danger as well as in endur- 
ing fatigue and hardships, our old officers' corps was the 
best proof of the correctness of our peace training. The 
difficulties of filling up the gaps in our officers forced us to 
husband our trained officers and to cut out a 'deader reserve'' 
prior to each battle. But, nevertheless, there still was ex- 
cellent spirit in the ranks, and the men followed their lead- 
ers in full confidence. Of course, the training was no longer 
the same, as we could not accomplish in six months what had 
taken two years in time of peace, which was especially true 
of the use of the rifle as a firearm and as a shock arm. The 
troops were younger, and added to the lack of training was 
the fact that the youthful recruits were neither fully de- 
veloped physically nor trained. The youthful recruits, who 
had grown up without the parental influence because their 
fathers were in the field, were spoiled by high wages and 
manner of living, and were lacking in morale as well. 

METHODS OF HOSTILE DEFENSE AND 
GERMAN OFFENSIVE 

On the Allied side an attack on a grand scale was ex- 
pected — ^but by no nieans before April. Just as was the 
case with us, the decrease in moral value of the troops was 
felt by the Allies. In Regulations for the Defense the En- 
tente sought development of the then accepted German 
views in adopting the mobile defense. The leader was to 
designate : 

1. Whether {he attack was to be accepted in the first or the 
second position, or, 

2. Whether the attack was to be avoided, in order to launch 
a well-prepared powerful counter-attack during the enemy's ad- 
vance, and, at the time that he would not have the support of his 
artiller y, or, 

3. Whether voluntary retreat should take place to a position 
in rear, to profit, at some other point, by the delay the enemjr's 
attack would encounter in such a movement. 

The condition of all works, badly shot to pieces, was 
accepted as a fact that could not be helped, and thus we 
declined to keep up a defense of ^'trunk'' lines, which the 
troops up to this time regarded as a prerequisite for a sue- 



THE BATTLES IN 1918 268 

cessful defense. After a protracted drum fire the only 
question could be of tenaciously holding a few critical points ; 
these the commanders attempted to establish inconspicu- 
ously under mutual fire support; the trenches situated be- 
tween them, with strong obstacles, were to divert the atten- 
tion of the enemy from the combat groups, which, grouped 
behind or beside each other, were consolidated into strong 
points in company sectors, and these latter again into bat- 
talion centers of resistance. The trenches were only for 
the purpose of facilitating communication in quieter times. 
A battle group, usually of the strength of half a platoon, 
consisted of 1 machine gun with some artillery observers, 
riflemen, and hand and rifle grenade throwers^ In this 
manner formation in depth and an intermediate terrain 
defense had been prepared, the importance of which we, 
as attackers, found out to our cost. Special weight was 
laid on the counter-attack and flanking fire was specially 
recommended for the reason that it would in this manner 
be possible to save rifles. Only one general difference from 
the German Regulations was adopted, in this case the me- 
dium and heavy flat-trajectory guns with the corps artil- 
lery and a few high-angle fire batteries remained under the 
orders of corps headquarters, which latter thus retained 
control of fighting the hostile artillery. The divisions kept 
only their field artillery, the stationary tank defense guns 
and a few high-angle fire batteries. We will point out here, 
that, on the part of the enemy, the flat-trajectory artillery — 
different with us since we had a preference for batteries — 
was preferred for tactical and technical reasons (minor 
weight of gun, less powder charge). 

The divisions occupied sectors of from 2 to 4 kilome- 
ters breadth ; first and second positions were separated by 
a distance of from 6 to 8 kilometers, so that in any case the 
attacker would be compelled to again deploy his artillery. 
The main fighting trench is the most important work of any 
defensive position ; security detachments are placed in front 
of it, and in rear lies the pardUele de soutien at a distance 
of about 200 meters as an exit for counter-attacks. Within 
the trench were bombproof shelters capable of defense, while 
in the main fighting trench there were small shelters. The 



264 DEVELOPMENT OF TACTICS— WORLD WAR 

construction of deeper and stronger bombproof s in the main 
fighting trench was prohibited. Besides the connecting 
trenches, inter-locking trenches and intermediate lines were 
constructed, as wdl as numerous rallying points indepen- 
dent of the trenches. In front of the batteries was an artil- 
lery protective position consisting of strong points connected 
by trenches. False positions and protection against aerial 
reconnaissance were specially valued. 

The fire activity had almost completely discarded the 
retaliation fire ; the stationary barrage fire was left mainly 
to the infantry auxiliary arms; greater importance was 
assigned to the daily destruction and annihilation fire (coun- 
ter preparation) in which the minenwerfers participated. 
High angle fire was directed against fortifications, fighting 
positions, observation positions, bombproof s and crossings 
of trenches ; batteries were attacked by fiat-trajectory guns ; 
routes of approach stopped by machine guns. The defen- 
sive method demanded tenacious resistance on the part of 
all detachments and also of the ones that had been dispersed, 
so that points of resistance sprang up which could cause the 
attack confusion and thereby prepare the success of counter- 
attacks. Reserves that were not designed for counter-at- 
tacks, according to the defensive plan, prepared themselves 
for defense just where they happened to be. 'It is impor- 
tant that the reserves remain in readiness for employment 
at the places originally designated.'' 

The Grerman commanders distinguished between at- 
tacks with a limited objective (for purpose of bettering the 
position, relieving the main attack through deceiving the 
enemy) , and the penetrating battle which could develop af- 
ter entry into the hostile position into the battle of maneu- 
ver. In spite of extensive preparations the penetrating at- 
tempts on the part of the enemy had come to a standstill 
after the relatively easy beginnings; the German penetra- 
tion was based on the maxim that the hostile artillery had 
to be taken the first day. This required an entry of at 
least 10 kilometers. Emphasis was laid on the great value 
of surprise. The extensive preparations for attack could, 
however, hardly be hidden from the hostile aerial observers 
and this information was augmented by statements of de- 
serters and prisoners (and for that reason, we, as well as 



THE BATTLES IN 1918 266 

our enemies, stopped all minor attack operations prior to the 
battle) . Though we received in this manner timely know- 
ledge of each hostile attack, we had to assume that the 
enemy received the same in regard to our intentions. To 
keep the operation secret as well as strict control of the 
troops in shelter, on the march, in deploying and in the ser- 
vice of communication, remained one of the most important 
tasks of leadership. 

The Battle of the Somme had furnished an illustration 
of the slow ''eating through'' process of a stabilized situa- 
tion ; the longer this took the more time the defender gained 
for his counter-measures. That had to be avoided at all 
cost. After the first entry, quick attack was essential so as 
not to allow the enemy to come to his senses. Tlie hostile 
attacks never were able to progress beyond the fire zone 
of their own artillery fire ; thus the attack fell through. A 
deeper entry could have been accomplished only by rapidly 
advancing the artillery, inserting reserves, relieving the 
shock troops from the care for the fianks and rear, and then 
the utilization of tanks. 

While the regulations of the Allies laid emphasis on 
the necessity of depth formation and narrow fronts in at- 
tack, we sought to find the proper mean between the de- 
mand for the pressure of fresh forces from the rear (nar- 
row front) and the possibility of advantageously posting 
the artillery, as well as the necessity for assigning to each 
division in the first line a good road for bringing up supplies 
(broad battle fronts). From this it resulted that it was 
not considered correct in a deep attack to obtain less than 

2 kilometers' breadth of a battle front for the division, while 
in an attack with a limited objective the breadth could be 

3 kilometers. Even with a decrease of numerical force a 
battle breadth of 2 kilometers was found not to be to large, 
as the power of fire arms had been materially increased and 
as the gun carriers in the first line were designated to util- 
ize fire effect. But if, on the other hand, the number of 
rifles decreased too much we had to limit ourselves to hold- 
ing the enemy and first gaining terrain supported by the 
progress of a stronger neighbor. The necessity was per- 
ceived furthermore of disregarding some single well forti- 
fied supporting points, around which, we had learned, pro- 



266 DEVELOPMENT OF TACTICS— WORLD WAR 

tracted fighting would ensue and holding them down for 
the present and taking them later on l^ an attack from 
flank or rear. As was the practice in Eastern Galicia and 
on the Isonzo, it was believed to be advantageous to conduct 
the attack so that preparations were made on a large scale 
in the earUest assignment of the troops for an envelop- 
ing attack against single supporting points. In the further 
development of this thought, not all parts of the hostile 
position were to be attacked equally, but the center of 
gravity was to be laid on such points the possession of which 
had to be decisive in the further conduct of the battle. 

Two to three division sectors were placed under one 
corps headquarters; in the second and third liile followed 
divisions for relief, to be inserted after the leading troojMS 
had expended their force. The thought of placing them 
under the orders of the leading division was rejected, as 
that method would merely have hastened their premature 
use. They were to remain as a reserve in the hands of the 
highest conmiander. Their insertion in any case was pro- 
per only where the attack made good progress. As we had 
learned, taking full advantage of a success best helped the 
troops that had remained behind or even had suffered re- 
verses. In any case they had to be kept close at hand, but 
did not need to reach the assault terrain until the very day 
of the assault. 

Based on our experiences it was deemed correct to have 
a strong artillery preparation lasting for about two hours, 
with a full utilization of gas. A shock-like fire preparation 
might well be suited for minor or medium operations, but 
attack without fire preparation could be successful only 
under specially favorable conditions. Carefully thought 
out distribution of tasks for the artillery and minenwerf ers 
as well as separation of targets was required. Minenwerf ers 
were placed under the orders of the artillery and, on account 
of their limited range, directed their fire only on the near- 
est trenches, especially on obstacles. 

The longer the artillery fight lasts, the harder will it 
be to accomplish surprise; the more difficult will it be to 
protect the assaulting troops in their places of readiness. 
The artillery fight commences at dark, so that the infantry 
can start the assault in the early morning hours after a 



THE BATTLES IN 1918 267 

powerful bombardment of the hostile position. We must 
prove in every case how few gans we require for the artil- 
lery action and how many can be employed to prepare the 
hostile positions for the assault. We must avoid creating 
spaces near the trenches which suffer nothing from the 
bombardment and into which the enemy may withdraw. The 
major battles have shown that the defensive power is by 
no means broken by changing the trenches into a field of 
shell-holes ; but the hostile morale is most surely destroyed 
by creating fire pauses, then advancing the fire and finaUy 
bringing a burst of fire back upon the point to be entered. 
It is entirely wrong (as the French and British did) to cre- 
ate the conviction in the infantry that through this fire it 
would be saved from close range fighting; this kind of 
fighting can only be shortened and made easier. 

After the position has been entered, the infantry de- 
mands : 

1. Artillery support in the position when overcoming the 
hostile resistance (machine gun nesttf and raUying points) by at- 
taching about 1 accompanying battery per regiment. 

2. Fire protection during the advance,^ — ^^Creeping barrai^e.'' 
A further success lies only in the rapid advance — commencmg 
with the entry— of an artillery wave^ with teams hitched, that 
has been kept back for tiiat purpose, and the entire force of 
teamed minenwerf ers. The difaculties of the groond have to be 
overcome and proper steps taken to prevent the roads from |pet- 
ting blocked up. It is luso necessary to bring forward, besides 
the artillery, mortars and long range 10-cm. guns. 

3. If the infantry has reached we objective of the attack, the 
creeping barrage changes into a stationary barrage (which gra- 
dually decreases and is again called for at places of counterat- 
tack). 

4. Until the forward movement is resumed| the fire activity 
of the artillery is to be regulated according to tne points of view 
of defense; advancing reserves are to be kept off and counter- 
attacks, as soon as perceived, are to be annihilated by timely 
action. 

The artillery receives a very material support by the 
attack squadrons and the observation airplanes of the aerial 
fighting forces. 

The deeper the entry is to be made, the earlier must 
the time for the assault be fixed. It depends on the time 
required for artillery preparation whether the infantry 
leaves the assault position prior to the artillery fire during 
the night or, in case a longer artillery preparation is planned, 
during the artillery fire. Losses by hostile artillery fire have 



2e8 DEVELOPMENT OF TACTICS— WORLD WAR 

to be reckoned with in the latter case. It is unfavorable if 
the preparations extend over several days; for, in spite of 
all destructive artillery fire, the enemy can recuperate dur- 
ing the night and shift his forces. The longer the artillery 
preparation, the more favorable for the attack. 'Tlacing 
the assault infantry in readiness is the first crisis of the 
attack.'' Danger lies in a too extensive crowding together 
of men, which increases the difiiculties of hiding all noises 
in movements and which increases the losses in sudden fire. 
Complete clearness must obtain concerning the procedure 
in case the enemy has perceived the position in readiness 
and attempts to destroy it. The commander must regulate 
exactly what equipment the troops are to carry. The 
troops are placed in readiness in very compact formation 
to rapidly run under the barrage fire, and assumes depth 
formation only during the advance. If infantry is placed 
in readiness in great depth formation the first lines must 
be made especially strong, as we have to reckon with the 
rear waves being cut off by sudden barrage fire, and the 
leading waves in that case must be strong enough to force 
the decision. Whether the infantry attack is to be executed 
in thin lines or in lines of squads must be carefully con- 
sidered in each instance. If the enemy's resistance is strong, 
each line of trenches will as a rule consume the attack 
power of one wave, so that the number of attack waves 
should be fixed by the number of lines of trenches to be 
taken; but, in no case, must the leading wave halt at the 
first trench; what can proceed forward^ must keep on till 
the attack objective is reached. As the opponent will no 
longer occupy his trenches in a dense battle line, we might 
•eliminate the entry in a dense skirmish line. In this man- 
ner the entry in close swarms, in "shock-squads" was de- 
veloped. But we cannot do at all without a skirmish line 
as a battle formation ; formation into shock squads is merely 
a makeshift. In the skirmish line, supervision is easier; 
the skirmish line facilitates the greatest deployment of 
fire power. The skirmish line must be formed prior to the 
time that the entrenched terrain is reached. More impor- 
tant than the formation of a definite attack procedure is 
the celerity of the advance. Infantry must become accus- 
tomed to having machine guns fire over its heads and also 



THE BATTLES IN 1918 269 

through gaps between detachments. But by this manner 
of advance, the question has also been answered whether 
troops are to advance in long or short rushes. As long as 
possible the walk is continued, and only when the enemy's 
fire becomes effective, are short, irregular rushes taken by 
squads and individuals. The farther off the attack objec- 
tive, the less attention will be paid to the trench garrison ; 
they are captured by the ^^mopping-up" detachments. It 
would be an error for the leading line to move toward a 
flank to assist a neighboring detachment ; that would merely 
promote the loss of the march direction. Only succeeding 
troops engage in action towards the flank, and minor de- 
tachments may be successful by a mere attempt to do so. 
The farther the attacking troops have gotten beyond the 
attack objective, the more effective has been their shock, 
and the less chance of meeting a new front. The number of 
machine guns must be brought into consonance with the 
number of rifles ; to 1 platoon of 40 men 2 automatic rifles 
would generally be attached, and a materiel reserve should 
be held out, in addition. ''The automatic rifle was and had 
to be (considering its fire power as against that of a rifle) , 
the main and increasing reliance of the infantry fire fight, 
especially as it was improved in design. This is not to say 
that the riflemen must not flre, — ^to the contrary, the great- 
est value was laid on such action.'' (Ludendorff.) Machine 
guns are to provide permanent fire protection for the in- 
fantry. A few machine gun companies are placed in ambush 
prior to the infantry start, while the others advance with 
the infantry to quickly reach important firing positions, 
which, in position warfare, can easily be reconnoitered in 
advance. Single machine guns serve for airplane defense. 
Light minenwerf ers which are made mobile for the attack 
and also for direct fire during the advance are attached to 
the battalions for attacking strong objectives. 

Simultaneous breaking forth of the assaulting infantry, 
entry immediately behind the last projectiles fired into the 
position — ^generally with time fuse — ^then in one rush pierc- 
ing to the attack objective, was always found effective. After 
reaching the attack objective measures must be immediately 
taken for warding off a counter-attack, and the infantry 



270 DEVELOPMENT OF TACnCS— WOBLD WAB 

given stability hy mohQe minenwerf en and artillery. Sapid 
attaek can freqaently eaoae poeitiona farther off to falL 

Regimental headqoarten moat make arrangements for 
bringing along the accompanying guns, moat arrange for 
locating the bearer sqoada, r^rolate the police service, sup- 
ply of ammunition fcnr machine guns (in belts), for rifles, 
grenade rifles, Bght minenwerf ers and hand grenades. It 
is the duty of division headquarters to regulate the move- 
ments of the combat trains. Telqdione communication to 
brigade and artillery headquarters remains of special im- 
portance, as well as wireless. Connection between regi- 
mental headquarters and the battalions is generally had 
by runners and dogs, but an attempt to establish communi- 
cation Iqr Ufl^t signals must not be abandoned. 

THE SPRING OFFENSIVE 

All conditions for the success of the planned offensive 
battle had been met. Information of the insertion of the 
18th Army (v. Hutier) between the 2d (v. d. Marwitz) and 
the 7th Army (v. Bohn) , the creation of numerous flying 
fields and troop camps in the vicinity of Laon had of course 
not escaped, since March 4, 1918, the observance of the hos- 
tile aerial reconnaissance, but had not been sufficiently 
evaluated by the hostile command. In reserve, the defender 
had 16 British and 35 French divisions. General Foch as 
president of the Supreme War Council made preparations 
for the organization of a reserve army of 30 to 60 divisions, 
but declined to get ahead of the German attack by an offen- 
sive on his part. There were probably 6 divisions of Ameri- 
cans present, but still under training. During the night of 
March 18-19th two Lorraine deserters went over to the 
French and, as was ascertained later, betrayed the German 
intentions. The evening of March 20th the concentration 
was completed. Though the fine spring weather in the sec- 
ond and third week of March had favored reconnaissance, 
the screening had not yet been successful. Great thanks 
is due the aerial fighting forces; they increased the diffi- 
culties of the hostile aerial reconnaissance while they them- 
selves took many photographs of positions in rear; and, 
continually watching over the traffic in the terrain in rear. 



THE BATTLES IN 1918 271 

could report their impression that the enemy did not expect 
an attack. The infantry could be instructed by means of 
the aerial photos of all details of the position to be assaulted, 
of each machine gun and of each minenwerf er location. In 
the second half of the month rainy weather softened the 
fields and the roads, and with sorrow did the assaulting 
troops watch the fog roll down during the night of March 
21st. But the high command insisted on the execution of 
the attack. 

The attack along the 70-kilometer front by the 17th 
Army (Otto v. Below), the 2d Army (v. d. Marwitz) and 
the 18th Army (v. Hutier) , started principally north of the 
junction between the French and the British, was, at that 
time and at that place a complete surprise, as an attack was 
not thought of before April in the terrain of the Somme bat- 
tle which offered such unfavorable conditions after the re- 
treat into the Siegfried position. The British commander- 
in-chief was home, in England. It was the task of the 
shock troops to pierce through the three main lines of the 
British position system to a depth of from 12 to 15 kilome- 
ters. As the 17th and 2d Armies advanced from the bend 
of the Cambrai arc in southwesterly direction, and the 18th 
Army, on the other hand, advanced westward, a concentric 
attack resulted, which was brought into a straight line dur- 
ing the course of the first movements by the 18th Army 
pushing sharply ahead and by the stubborn resistance of- 
fered by the British Army in the north. 

At 4:40 A.M. March 21st, — ^that is when it was still 
completely dark — ^the artillery battle opened along the 
entire front between CroisiUes and La Fere with gas and 
high explosive shells. At and after 6:40 A.M., the concen- 
trated fire of the short range battle groups swept the 3 
British positions. The failure to find a gradual weakening 
of the positions through days of effective fire had to be off- 
set by a doubled energy of the fire wave lasting for 3 hours. 
The incomparable shock-power of the infantry (which broke 
forth at 9 A.M.) was so instrumental in securing success 
that, in spite of fog and mud, the second hostile position was 
reached the evening of the first day of battle and partly cap- 
tured and partly over-run. The foremost trenches were 
taken comparatively easily by the deep-formed phalanx 



272 DEVELOPMENT OP TACTICS— WORLD WAR 

under cover of the fog ; at the most difficult points, such as 
the south exits of the city of St Quentin, flocks of tanks 
reinforced the momentum of the shock. Thereafter, how- 
ever, a net work of numerous supporting points had to be 
overcome with the help of machine guns, minenwerf ers and 
batteries. The fog, though it had materially favored sur- 
prise, now increased the difficulties of orientation, and slack- 
ened the rate of the attack. At many points in the after- 
noon, when the weather had cleared, and when our pursuit 
and battle squadrons threw themselves on the enemy, we 
had to wait for the arrival of the field artillery working its 
way under unbelievable difficulties through the bottomless 
mud of the terrain (which was merely a mass of shell-cra- 
ters) , in order to overcome stronger bulwarks. 

Most of the airplane detachments moved their flying 
fields to correspond with the progress of the action forward, 
some of them even to places heretofore used by the enemy's 
airplanes. The captive bajHooris, without being drawn down, 
foUowed the infantry at 4 to 6 kilometers distance and ccm- 
stantly reported to higher headquarters the course of the 
action. Anti-aircraft guns were brought close in rear of the 
foremost infantry line to fire on low flying battle airplanes, 
and there found opportunity to accomplish good results 
against the fleeing infantry and other ground targets at 
short range. In the first days the hostile battle airplanes 
found little opportunity for counter-attack. 

The shock between the Scarpe and the Oise struck the 
British 9th Division on the front line with the 8th Infantry 
Division and the Sd Cavalry Division behind it. At the 
time the British 3d Army in the north (Byng) made a stand 
in the face of the attack by the 17th Army, the resistance 
of the 5th Army (Gough) broke quickly and entirely un- 
expectedly. An unsuccessful counter-attack was started 
on the 22d by foot troops without artillery support, who were 
brought up by autos from the vicinity of Senlis and Com- 
piegne and who belonged to the French 9th and 10th In- 
fantry Divisions and 1st dismounted Cuirassier Division, 
supported by the 25th Infantry Division from the vicinity 
of Chauny. During the evening of the 22d, the Army of 
Hutier had broken through the third hostile position, while 
the 17tii Army (Otto v. Below) rested in front of the Brit- 



THE BATTLES IN 1918 273 

ish 3d position. The evening report of the 23d read : '^e 
are fighting in a line about north of Bapaume, P^ronne and 
Ham/' The penetration had been successful as well as the 
pursuit connected with it; and thus the dead point of all 
previous penetrating battles had been overcome. During 
the night of the 24th Bapaume fell ; the day after, Noyon ; 
in vain did the 55th Infantry Division, hurriedly brought 
from Paris, attempt to ward off fate there. Farther north, 
around Cambrai, the British troops also gave way. The 
French forced to insert their divisions as they came up, were 
thrown into the general retreat. '^The support of our bat- 
tle airplanes was specially effective during the infantry at- 
tacks on Bapaume and in the charge against Ferme la Mai- 
sonnette at P^ronne which had been attacked before. At 
Albert they attacked fleeing columns with bombs and fired 
more than 80,000 rounds from machine guns. Our bomb- 
ing squadrons continued night after night the fight against 
the lines of communications, the troop shelters, and the 
flying fields of the enemy. In the attack against railroad 
stations behind the front they frequently fiew at altitudes 
of less than 100 meters. Hits in the midst of railroad works 
and on trains and heavy explosions and conflagrations were 
the result of their audacity. In the three nights after 
March 25th a total of 100,000 kg. of explosives was thrown 
down." 

The most important event of the day was the final ap- 
pointment of Foch as commander-in-chief of the Allies. 
Countershocks by British divisions brought from Flanders 
made themselves felt. A counter-attack on a large scale 
started by Foch encountered a rapidly formed new German 
front Mondidier — ^Lassigny — ^Noyon. The French reserves 
had been held back to the fullest extent, since the German 
progress was naturally not so clearly defined as in the first 
days. The German front had approached within 15 kilome- 
ters of Amiens. The gap between Ancre and Oise was closed 
by British reserves and by the army under FayoUes brought 
from Flanders. A conthiuation of the attack which had 
be^n expected by the Allies did not take place, so that the 
Allies on April 4th advanced from Anliens to an attack, 
which, however, had no success. 



274 DEVELOPMENT OF TACTIGS— WORLD WAB 

In the very first days of the attack the Army Groiq[> 
commanded by the Crown Prince of Prussia which had been 
designated to carry forward the attack of the 2d and 17th 
Armies in the direction of Albert and Abbeville had 15 divi- 
sions, which were quite fresh yet, but of the second and 
third line. But the High Command held the view that to 
succeed quickest was to attack where the enemy had given 
way easiest. It was indisputably confirmed that the hostile 
resistance was stronger than our power. A battle of attri- 
tion ought not to have been fought. This was precluded by 
our strategical as well as tactical situation. The High Com- 
mand had to come to the exceedingly grave decision — in con- 
junction with all headquarters concerned — ^to definitely stop 
the attack on Amiens. At the same time, while the atten- 
tion of the Entente was completely engaged at Amiens, on 
the opposite wing, on April 6th, the 7th Army started tlie 
offensive and was enabled to bring its right wing to the 
Ailette and to the Aisne — Oise canal by the battles at Au- 
rigny and Coucy-le-Chateau. 

The indicated shifting of the British Flanders reserves 
and the expectation of finding along the Flanders front less 
efficient units (Portuguese) or even tired out divisions, was 
(in addition, the possibility of exerting a pressure on the 
British communications leading from the canal ports), the 
inducement for a new attack in Flanders. Unfortunately, 
we lacked fresh forces to execute the shock with full vigor. 
April 9th the 6th Army (v. Quast) attacked on the line 
Festubert — ^Armentieres, not bothering with the latter place 
in the start, which later on, on the 12th, fell into our hands. 
This time there was a short preparatory fire from 4:15 to 
8 :45 A.M., then excellent artillery support for the infantry, 
so that most of the enemy's machine guns were unable to 
get into action. As early as 10 :00 A.M., the third position 
of the enemy was taken. The difficulties encountered in 
the completely sodden, shell-cratered terrain, appeared al- 
most insurmountable, but we succeeded on the first day in 
bringing up several heavy guns, though after unbelievable 
labor. Early on the 10th, the 4th Army (Sixt v. Armin) 
commenced the attack under the same difficulties farther 
north between Armentieres and HoUebeke, while in the 
next few days the 6th Army gained more ground in the 



THE BATTLES IN 1918 276 

direction of Hazebrouck and HoUebeke, so that between the 
16th and 18th our successes in the Ypres arc made them- 
selves felt also on the right bank of the Yser. On April 21st 
the number of prisoners taken was 117,000, 1550 guns had 
been captured and more than 200 tanks. The amount of 
anmiunition, subsistence, supplies, and equipment captured 
could not even be calculated. In the course of 4 weeks the 
British not only lost what they had gained in the half-year 
battles on the Somme and the rest of their success at Cam- 
brai, but also about two-thirds of the terrain they had 
bought so dearly in 16 battles in Flanders. On April 25th 
the battles ceased by the well planned and brilliantly exe- 
cuted capture of Kemmel. Had headquarters arrived in time 
on the hill, continuation of the attack would have been pos- 
sible before the hostile reserves could come up. German pa- 
trols had reached without danger the Scherpenberg, while 
the troops occupied a very unfavorable line which they had 
been ordered to hold. Thus we contented ourself with the 
mere capture of a comer piUar of the important Flanders 
position. ''The attack fulfills its purpose as long as it causes 
the enemy more damage and loss than it does ourselves. 
That is always the case in the first stages of a successful 
surprise attack. Here the main point is to engage the enemy 
without fear of losses, but not neglecting fire support and 
fire preparation. The commander must determine whether 
the effect of the hostile reserves is becoming stronger. If 
that is the case, the attacks must gradually become more 
systematic ; strong artillery preparation with plenty of am- 
munition develops more and more until the transition to 
defense on the part of the victor becomes realized. By his 

I insufficiently prepared 'counter-shocks' and counter-at- 

tacks the opponent will suffer material losses in ihe face of 

, a skillful defense, while we save our forces. With us the 

question hardly ever is to gain terrain at any price. We 
must strike down the enemy ; we must save ourselves. Our 
military feeling must in this regard become «ven firmer; 

I we are inclined to attack in the later days of an attack with 

entirely too insufficient means and are inclined to fight for 

I the purpose of gaining ground, which is entirely immaterial 

^ -* to the whole situation.'' (Reference— High Command on 



276 DEVELOPMENT OF TACTICS— WORLD WAR 

June 9, 1918.) In these expressions lies the entire system 
of our battles in 1918. 

Our new method of attack had been still farther de- 
veloped. It consisted in surprise, by short, powerful artil- 
lery preparation directed on single points and in rapid ad- 
vance of the infantry with co-operations of its auxiliary 
arms. 

On March 21st the principal question was of an attack 
that had been planned by Great General Headquarters in 
quietness and with the utmost care; in its fortunate exe- 
cution the German troops struck newly organized fronts, or 
fronts that were still under construction. Every delay in 
time might bring the success of the undertaking into ques- 
tion; the subordinate commanders had to act quickly and 
independently, and General Headquarters had to bring the 
different operations into consonance with the whole. 3 to 5 
divisions had been placed under the orders of Corps Head- 
quarters; 2 to 3 of these divisions were to execute the at- 
tack in the first line, under proper artillery support, as long 
as possible. The attack frontage of 2 kilometers proved too 
narrow in most cases for properly leading the division and 
therefore an extension to 3 kilometers was recommended. 
Strong resistance was to be overcome by inserting more ar- 
tillery but not by increasing the density of the infantry, 
which would only increase the losses. 20 to 30 batteries 
were inserted per kilometer. Their effect was supplemented 
by medium and heavy minenwerf ers. The artillery attached 
to the divisions by the war organization tables was rein- 
forced in most cases by a second field artillery regiment, 
one or more heavy field howitzer battalions and 1 mortar 
battery. If the reinforcing artillery was taken away from 
some one division, then arrangements had to be made to have 
that artillery return at the proper time to its own division. 
The Group (army corps) had the disposition of a number of 
field batteries of the army artillery, heavy field howitzer 
battalion^, mortars, and heavy flat-trajectory batteries, 
which were attached to the division in the front line ac- 
cording to requirements. Assigning such artillery was to 
be kept within limits, so as not to create specially difficult 
conditions by the insertion of an unweildy mass. The num- 
ber of guns had to be brought into proper consonance with 



THE BATTLES IN 1918 277 

the increase of the ammunition supply. 'Too many gwoB 
in relation to the amount of ammunition is very bad. The 
fighting power of troops is lessened, not increased, since 
the difficulties of leadership are increased by the volum- 
inous carriages and vehicles of the numerous guns, which 
cannot be properly used. And probably what the troops 
need more is denied for that very reason.'' 

Success of the attack does not lie in the number of in- 
fantry units inserted, but in the manner of artillery pre- 
paration and the immediate utilization of the moral impres- 
sion created by our fire effect and in the use of the auxil- 
iary arms (machine guns, light minenwerfers, flame throw- 
ers and accompan3dng guns) that accompany the infantry. 
We had but very few tanks, and could do without their as- 
sistance. The German industries could have constructed 
a sufficient number of tanks by neglecting other means of 
fighting. We adopted a tank gun (13-mm.) which unfor- 
tunately was very heavy and required two men to handle. 
Important services were performed by battle airplanes. 
Technique now also had the same importance as tactics and 
psychology. 

The machine gun had become the main arm of infan- 
try and formed the skeleton of each battle formation. A 
fifth machine gun was now furnished the companies. The 
skirmish line takes a back seat now in the fire fight ; we had 
absolutely broken with the meaning of a '^werf ul skirmish 
line'' with which we entered war under other conditions. In 
the start great objection was put forth against this inovar 
tion ; for in the first battles the skirmish lines were still too 
dense. ^'Overcoming machine gun nests had caused diffi- 
culties beyond expectation at many points and delayed our 
attack. The independence of our infantry squad had been 
found wanting frequently, as well as its co-operation with 
the auxiliary arms. It had become especially difficult for 
the troops to form again for defense after the completion 
of an attack and especially to perceive that a continuation 
of the attack could have no further success." But, in the 
end, the number of men carrying rifles will determine the 
result. It is questionable if that number might not decrease 
too much ; ''the danger is the greater because a weaker unit 
requires the same administrative staff, same number of in- 



278 DEVELOPMENT OF TACTICS— WORLD WAR 

formation troops, and name train as that of a stronger unit, 
bat has only a smaller number of rifles. It was necessary to 
consider whether it would not be best to overcome this 
difficulty by consolidating two divisions/' Another diffi- 
culty was the new attack method. Heretofore the entire 
attacking space was ffiled with dense skirmish lines. Battle 
leadership, which had to pay attention only to the hostile 
skirmishers, or at most to a few machine guns, was rela- 
tively simple. Everybody went straight for the objective. 
At the present day, however, there can be no thought of 
an equal distribution of the infantry in the terrain. Hie 
troops divide according to the breadth and depth into sub- 
detachments and lose themselves in the terrain — ^but all 
must work toward one objective. The auxiliary arms also 
demand depth formation ; if they are once let out of hand, 
the infantry will soon be unable to get forward. The new 
method of attack puts especially high demands on the sub- 
ordinate leaders. If these give out, the attack stops ; here- 
tofore we could carry it forward again by inserting strong 
skirmish lines led as units. But this — and therein lies the 
danger — no longer is possible in the new method, which 
expresses itself in a slow forward movement in overcom- 
ing hostile centers of resistance. It will be the main task 
of battle training from now on to assure rapid, decisive 
co-operation between the separated portions within the 
whole of the battle action. 

Infantry advances in ffies, in swarms, as ^'shock squads'* 
or in skirmish line ; its fire power lies in the machine guns 
and light minenwerfers, the protection of which it assumes. 
Overcoming hostile machine gun nests takes much time. 
Frontal attack leads to nothing. In view of the checker- 
board like distribution of the nests and in consideration of 
the strength of the supporting points, attacks from flank 
and rear were but seldom possible ; the use of gas proved 
useful; the troops also soon learned to estimate the value 
of the rifle grenades adopted from the French system. Diffi- 
cult situations arose when machine gun nests were discov- 
ered too late, so that their attack could not be prepared un- 
der cover, and when the accompanying guns had heavy losses 
before they could fire even a single shot. The accompany- 
ing guns were able— even if only slowly — ^to follow up the 



THE BATTLES IN 1918 279 

infantry by the help of pioneers and laborers, being partially 
protected by gas masks. Proper attention also has to be 
paid to watering the animals during the advance. 

Finally the infantry storms the terrain still held by 
the enemy, but adheres to its depth formation until entry 
of the enemy's position and then seeks to re-establish that 
depth formation immediately after reaching the attack ob- 
jective — ^generally toward the front. It has not always 
been easy for infantry to regain this depth formation. De- 
tachments in rear take over the fire protection against coun- 
ter-attacks and secure the flanks or turn to the side to 
help the neighbor. 

The commander endeavors to achieve surprise with all 
means at hand. Though success is accomplished in hiding 
the attack preparations from the enemy by well thought 
out orders, strict discipline and development of oral exam- 
inations and instructions of all concerned, it is by no means 
assured that the enemy will not receive a warning of the 
storm about to break through statements of deserters. But 
the statements of single prisoners can hardly ever give a 
complete picture ; they are usually received so late, that the 
leader can utilize them but seldom, even disregarding the 
fact that false statements are mixed up with true state- 
ments, and that probably no credence will be given to the 
true ones. It was impossible to stop communication between 
the army and home, ''the mania for talking and boasting, 
so inherent to the Germans, brought the most important 
and most secret things to the knowledge of the press and 
consequently also to the knowledge of the enemy." 

If we must count absolutely on the defense of a ''fore- 
ground,'' it may be of advantage to take possession of that 
foreground even before the assault and to gas more effec- 
tively those hostile batteries that fire standing barrages dur- 
ing the advance. 

The infantry attacks under protection of the creeping 
barrage which advances in lifts of about 200 meters each, 
at the rate of 40 to 50 minutes per kilometer. In spite of 
the slow progress of the creeping barrage it has in many 
cases outstripped the infantry; it also proved to be too 
rigid, to correspond to the progress of the infantry. The 
danger was very great that in disruption of the barrage, 



280 DEVELOPMENT OF TACTICS-WORLD WAR 

some batteries would fire on our own inf antryl It was of 
course entirely out of the question to bring a creeping bar- 
rage back, as the location of the infantry could not be defin- 
itely shown to the artillery. If the infantry was able to 
follow close at the heels of the barrage, it could in most 
cases over-run the hostile machine gun nests and support- 
ing points before the latter could become effective. But if 
the barrage continued its way and outstripped the infantry, 
then the latter was obliged to break the stubborn resistance 
of the enemy with its own means — accompanying guns and 
teamed minenwerfers. The infantry must never come to 
believe that its very existence lies within the creeping bar- 
rage. The existence of infantry much rather rests in the 
fighting value of the troops and the skillful utilization of 
its auxiliary arms. The desire for more accompanying ar- 
tiUery was heard everywhere ; in many cases a 96-mm. gun 
battery (the 16-cm. was too heavy) and 1 platoon of light 
field howitzers were attached to the regiments, and single 
guns with several caissons to the battalions. Often heavy 
field howitzers were desired. It was quite natural that the 
regiments finally received in addition to their accompany- 
ing batteries an artillery group consisting of 2 96-mm. gun 
batteries and 1 light and 1 heavy field howitzer battery, 
while the rest of the artillery remained at the disposal of 
the divisions. This was correct as the regiments because 
of their gradually decreasing effective strength (companies 
frequently had but three men carrying rifles) were required 
to cover materially greater spaces with the number of rifles 
available. Due to these broad attack sectors, the battle 
generally resolved itself into partial fights which were con- 
ducted by "mixed" regiments in the front line. The divi- 
sion commander was able to bring his influence to bear on 
the battle only through his reserve regiment and his artil- 
lery. This dissolution of the attack into partial fights was 
an unavoidable and bad feature which had to be abandoned 
when the situation permitted systematic attack or defense. 
The first step in that direction, was to deprive the regiments 
of their artillery. 

After the position is entered, the divisional artillery 
is brought up ; and, to avoid congesting the roads all vehicles 



THE BATTLES IN 1918 281 

that can be spared, even the field kitchens, must be kept 
back. 

RESUMING THE OFFENSIVE 

The German offensive in Flanders had succeeded in at 
least one object; it had contained strong hostile forces, it 
had filled the enemy with fear for his canal ports and had 
aroused the last remnants of the Belgian state. It is pro- 
bable that it did have a farther reaching objective. Based 
on the recommendations of the Army Group conunanded by 
the Crown Prince of Prussia, Great General Headquarters 
first sought a success in the attack in front of the 1st and 
7th Armies, and then to again transfer the decisive attack 
there as soon as the first attack had had itis effect on the 
situation in Flanders. Though prisoners taken on May 25th 
and 26th had divulged everything they knew concerning the 
time and place of the attack, the attack of the 7th Army on 
June 27th at the Chemin des Dames and farther east to 
Brimont was very successful; in conjunction with the 1st 
Army (Fritz v. Below) it was able to accomplish the cap- 
ture of the heights of Reimes and Epemay. The French 
high command had ordered early readiness for gassing and 
had placed its reserves (13th Infantry Division) on the 
heights between the Aisne and Vesle. 

At 4 :40 A.M., on the 27th, after a short artillery pre- 
paration, the German troops broke forth to the attack from 
their positions between Landricourt and Brimont. The at- 
tack struck battle-worn British divisions, which lost about 
15,000 prisoners, and then encountered several fresh French 
divisions. As early as the first day of the attack a depth 
of 18 kilometers was reached inside of the hostile position by 
the irresistable center of the attack. In the next few days 
the wings of the attack could not keep up with the advance 
ing center. On the 3d day of battle, Soissons was taken, the 
number of prisoners increased to 35,000 ; on May 30th the 
Mame was reached south of La Fere en Tardenois, the 
number of prisoners being 45,000; and, on the 31st heavy 
battles took place near Soissons. 

The French leadership saw the cause of its defeat in 
the error of having deployed all its batteries in a single line, 
so that the artillery was annihilated in a short time; in 



282 DEVELOPMENT OF TACTICS— WORLD WAB 

addition, the French did not utilize the advantage to get 
ahead of the recognized German artiUery preparation by 
a timely counter-preparation. It was quite diflferent in the 
succeeding batties I The leaders abandoned the rigid hold- 
ing of terrain, and evacuated portions of the position when 
it was impossible to lay a barrage to 100 to 150 meters in 
front of its own line. Such positions show themselves to 
be untenable. Untenable positions are defended only when 
it is absolutely unavoidable. It is an advantege only if the 
new position is not exposed to the destructive fire of the 
light artillery and minenwerfers, so that a new deploy- 
ment of the artillery becomes necessary. The troops should 
fight in a position of their own selection, not in one selected 
by the enemy. 

From June 6th the two new front line positions of the 
Germans, Chateau Thierry — Villers Cotterets — ^Noyon and 
Chateau Thierry — ^Dormans — ^Reims were under the des- 
perate counter-attacks of the enemy but the situation did 
not change. The Army Group commanded by the German 
Crown Prince had teken up to June 6th; 55,000 prisoners 
(of these, 1500 officers) 650 guns and more than 2000 ma- 
chine guns. 

A fourth German offensive shock by the 18th Army 
with four corps (9th, v. Ottinger, 62d, v. Webem; 8th, v. 
Schdler; 28th Reserve Corps, Hojffman), led to the Buttle 
of the Matz between Mondidier and Noyon, with the deter- 
mined intention to take the forest terrain of Compiegne 
which possessed permanent danger for the' Germans. It 
was a rare coincidence that the commander-in-chief of the 
French 3d Army, Greneral Humbert, claims to have had in- 
formation as early as June 4th (the orders of the attacker 
were dictated only that day) . It is, of course true that the 
army had expected an attack here since the end of May 
and had sought by all means to prevent a surprise. A deep 
foreground had been ordered, defended by many machine 
guns, from which the guns were at the proper time with- 
drawn. The first position, the foreground, was said to be 
strongly defended by 8 divisions disregarding the existence 
of a second position. This first position was to be the exit 
point for counter-attecks of the corps reserves. The 10th 



THE BATTLES IN 1918 288 

Division of the anny reserve was to be brought with its 
leading elements as far as the second position. 

The German attack was to be started hy gas shell fire, 
but in 1 division of 1000 throwers only 150 could be placed 
into position because of strong destruction fire, and, of 
these, 70 were destroyed. 

The attack preparations commenced at 12 :50 A.M. with 
a surprise fire of gas and bursting projectiles lasting 10 
minutes against the main hostile batteries, observation and 
command positions (which however had been withdrawn 
during the night) and against the shelters and routes of 
approach. Between 1 and 2:30 A.M. a. stronger artillery 
preparation; between 2:30 and 4:10 A.M. destructive fire 
against the hostile positions in such manner that each time 
two lines were taken under fire, while the hostile artiUery 
remained under fire of the long range batteries. At 4:20 
A.M. a powerful fire shock was to give the signal for the 
assault, which was to be preceded by a creeping barrage 
(1st lift 300 meters, than 10 minutes for each 200 meters) 
up to the line Ecouvillers-Atteche Ferme. The infantry 
worked its way up to within 300 meters of the leading hos- 
tile trenches, ran under the barrage squads, formed within 
the hostile trenches for further advance and also advanced 
the artillery at once. The execution of the attack had been 
well planned; in spite of superior forces the Grerman divi- 
sions gained ground, and could gain ground everywhere. In 
this battle, the infantry naturally did not have the full 
support of its guns, while being exposed to the concentrated 
fire of the hostile artillery. Rapid deployment of the ar- 
tillery against this new line of resistance was the principal 
requirement for success, while the infantry had to be halted 
as soon as it reached the artillery protective position. The 
French counter-attack executed on June 10th, 11th, and 
12th, and supported by tanks and airplane squadrons, had 
no success worthy of the name. 

The German leadership would have been able to break 
the resistance prepared here only with heavy losses, and 
it therefore decided not to continue the attack at this point. 
It was correct in any case for both opponents to seek the 
further decision in the Champagne between Rheims and 
(Thalons. The Grerman attack, with 15 divisions in the first. 



284 DEVELOPMENT OF TAGTIGS-WOBLD WAR 

and 10 in the second line, started on July 15th against a 
folly prepared enemy. The 4th Army, Gouraud, claims to 
have made preparations as early as July 12th in the CSham- 
pagne, and to have occupied its leading positions only vnUi 
detachments, in rear of which were single strongly wired 
supporting points to catch the shock in the line Poumoy — 
Froenes — ^Perthes, while the decision was to be sought by 
a counter-attack on a large scale against the front Chateau 
Thierry — Soissons. The day, place and frontage of the at- 
tack had been known to the Entente since the 18th. Pris- 
oners taken on July 14th divulged all further details of 
the attack. At 12:10 A.M. July 15th the artillery fire was 
to start and the infantry was expected to start for the as- 
sault at 4:15 A.M. The French counter preparation started 
at 11 :00 P JH on the 14th. 

According to British statements, the outpost garrison 
had materially delayed the German attack, and the French 
artillery, in rear positions, brought heavy losses to the as- 
saulting troops and the batteries that had gone forward 
too soon. The 6th Army (Berthelot), connecting on the 
left, held the advanced line between Rheims and the Mame 
in the Montagues de Reims, but could not prevent the Ger- 
man 1st Army (Fritz v. Below) from taking the heights 
north of Venteuil and the Ardre valley to include Poury. On 
the German right wing the 7th Army had experienced 
greater successes and had been able to gain ground. On 
July 16th the German headquarters directed the cessation 
of the attack, at 5 :30 A.M. On the 18th, between the Aisne 
and the Mame a completely unexpected French counter- 
attack came from Belleau (near Chateau Thierry) — ^Fon- 
tenay (west of Soissons). This counter-attack was made in 
superior force, with great depths and directed against the 
right flank of the Chateau Thierry — ^Reims front, and was 
executed by the Army Group of Fayolle (6th and 10th 
Army) • It was preceded by a short fire preparation which 
immediately continued in the shape of a creeping barrage 
accompanied by numerous fast tanks. The tanks, favored 
by the tall standing crops in which our machine guns could 
produce no effect, plowed through our lines, unloaded their 
passengers with machine guns in our rear to form machine 
gun nests, and then the tanks returned to bring up rein- 



\ 



THE BATTLES IN 1918 285 

f orcements. Loss of ground was unavoidable as our troops 
were battle worn, but, nevertheless, the scheduled flank at- 
tack was frustrated. Reserves rapidly brought up with 
auto trucks re-established the situation by a strong attack 
(20th Infantry Division). We will here again point out 
the difficulties encountered when the foot troops are sep- 
arated from their guns and vehicles while using auto trucks. 
The French did not utilize their initial success and thus 
saved us from a heavy defeat. The German offensive had 
passed its zenith ; the enormous losses in leaders and trained 
men were not to be replaced as quickly as the situation de- 
manded, and the value of recruits fell greatly. Even the 
best army cannot be victorious if it is not supported from 
home. 

The Grerman conduct of the offensive up to then was 
based on the assumption of surprising the enemy and that 
the enemy would remain in the terrain dominated by the 
German artillery. Any attack method that is correctly 
formed breaks the enemy's resistance. Surprise makes 
this easier. The attack must disintegrate if the enemy de- 
fends an outpost zone of some kilometers with weak forces, 
and takes a position in readiness with strong forces out- 
side the effective fire zone of our artillery, like Humbert 
did for the first time on the Matz and the Army under 
Gouraud in the Champagne. Now, could not the preparation 
for a change of the battle plans have been ascertained by 
aerial reconnaissance? Against such a battle formation a 
long artillery preparation merely meant squandering am- 
munition and allowed the inf anlxy without sufficient artil- 
lery preparation to enter into the unbroken artillery fire of 
the defender. . From these reflections result the fact that 
the hostile outpost zone should have been taken after a 
short artillery preparation (possibly only by medium and 
heavy minenwerfers) in the afternoon before the attack 
on a broad front (for instance 3 to 4 army fronts) and 
that only at the assault points should the assaulting troops 
be placed immediately in front of the main line of resistance 
during darkness. Prior to the infantry attack the second 
artillery deployment into the hostile outpost zone should 
commence. This will not be possible without resorting to 
the mechanical truck trains (caterpiUar trucks). Only a 



286 DEVELOPMENT OF TACTICS— WORLD WAR 

portioii of the artillery with the required amount of ammu- 
nition should be placed in readiness in our outpost zone. 
Simultaneously with the infantry attack the gassing of Hie 
hostile artillery commences ; almost the entire artillery par- 
ticipates in this; the infantry has to break the hostile re- 
sistance with its own accompanjdng arms; this becomes 
more necessary since it will hardly be possible to start an 
effective creeping barrage. Obstacles in the positions in 
rear have to be flattened and torn down by tanks. The at- 
tack has become more difficult through this battle method. 
The defender can materially delay the second deployment of 
our artiUery through harassing fire and gas attacks ; he can 
bring to naught the preparations for the attack by nightly 
sorties on a large scale, and can himself take the offensive 
at other points. 

We are obliged to admit that the French Army has 
shown a greatly increased offensive power in these battles 
through mass insertion of airplanes, anununition and tanks. 
As large forces of American troops had already arrived, the 
supreme command decided to evacuate the Mame to hold 
for the present the Aisne — ^Vesle line. 

THE LAST DEFENSIVE BATTLE 

After the defeat of the German July offensive, Marshal 
Foch had given the task to the British Army which on July 
28th the 1st French army (Debeney) was subordinated; 
i.e., to press in the projecting arc east of Amiens. For 
this attack Marshal Haig designated the 4th Army (Raw- 
linson) which, with 7 divisions in the first line and 4 divi- 
sions in the second line (starting on a 17-kilometer front 
from the line Hangard — ^Morlancourt) , was to reach Le 
Quesnil (9 km.) — ^Mericourt sur Soinme (6 km.) as its first 
objective. 1 cavalry corps (3d Cavalry Division) brought 
up by night marches, reinforced by 1 cyclist battalion and 
2 motor machine gun battalions, was to push forward along 
the Amiens — ^Roye road. The French 1st Army, connect- 
ing on the south, was to attack about one hour after the 
effect of the British attack made itself felt. Marshal Foch 
retained at his disposition the French 3d Army farther to 
the south. By means of well planned false movements and 



\ 



THE BATTLES IN 1918 287 

wirdess messages sent out to mislead the enemy, it was 
the plan to create the belief that the attack would take place 
in Flanders. It appeared that an operation of the German 
27th Infantry Division on August 6th south of Morlancourt 
did not clear up the situation for us. Favored by a heavy 
fog that had, in addition, been artificially thickened, the 
artillery fire suddenly opened on August 8th; this fire soon 
changed into a creeping barrage, immediately followed by 
tanks and infantry. Surprise against the leading division 
of the 2d Army had succeeded. The entry nearly reached 
the objective, that is, between Albert and Pierre-pont on 
the Avre, a depth of penetration of 14 kilometers had been 
reached, though, with the exception of organizing the posi- 
tion, everything that could be done for defense was done, 
including narrow division sectors with excellent depth for- 
mation and plenty of ammunition. The hostile tanks annir 
hilated our machine gun nests, and, foUowed by cavalry, 
entered deep into our position system spreading disorder 
and confusion — even headquarters were surprised in their 
quarters. Valuable documents were lost. 7 German divi- 
sions, considered absolutely fit for fighting, were annihil- 
ated ; the enemy had but little loss ; but the increasing bad 
morale of our poorly trained recruits became a decisive 
fact. Under such conditions, there could be no longer any 
mention of tactics. ''The 8th of August is the black day 
of the German Army in the History of this War." It was 
very fortunate that the attack of the Allies was not con- 
tinued on the 9th in the same strength. 

Further attacks between Ancre and Avre pushed back 
our lines without disrupting them. Gassing in conjunction 
with fog, mass insertion of tanks, and insertion of strong 
airplane battle squadrons were the causes of British suc- 
cess. Based on the estimate of the situation General Lu- 
dendorff perceived that in the face of the relative strength 
of both sides no victory could possibly be gained by the Cen- 
tral Powers. The Government of the Empire was obliged 
to start negotiations for peace, the army high command was 
forced to try and prevent a penetration of the German lines, 
in order to gain time, and to avoid unnecessary losses, and 
to save the living power of the army for subsequent tasks, 
which could be solved farther in rear on a shorter line. Thus 



288 DEVELOPMENT OF TACTICS— WORLD WAR 

the trooiNS were forced to hold the enemy, to cover the bring- 
ing away of the materiel, and to gain the time required for 
preparing a new position. Tenacious defense of the exten- 
sive foreground delayed the enemy's advance. These fore- 
ground battles gave the defender time for his preparations 
and furnished important viewpoints for estimating the of- 
fensive intentions of the enemy. In many cases it could 
not be prevented, having the fights extremely stubborn in 
the foreground zone. Foreground and mobile defense with 
countershock proved excellent in the defense. A British 
attack undertaken on August 21st at 6 :00 A.M., with a 16- 
kilometer front between Croisilles and the Ancre, against 
the 17th Army, according to the arrangements of August 
8th, was driven off. 104 tanks alone had been inserted on 
the Bucquoy — ^Achiet le Petit road. On the 22d an attack 
on a larger scale succeeded, and then the British continued 
their attacks the next few days. ''The characteristics of 
the attacks were narrow, but deep, entries by tanks after 
a short but exceedingly heavy artillery preparation, in con- 
junction with artificial fog. Mass insertion of tanks and 
artificial fog remained our most dangerous enemies in the 
future. They became more and more dangerous the more 
morale became lowered and the more tired and weaker our 
divisions became. The depth of the entry, but not its en- 
tire extent, were soon seen. Correctly inserted counter- 
attacks by the reserves generally offset these entries. But 
there was danger that the local command employed its 
troops in an overhasty manner and not in close order .'* 

The decisive battle was from now on carried on in the 
mobile attack method only where success was assured. If 
deep indentures occurred in a position, then the entire por- 
tion of the front was taken back rather than exposing the 
troops to heavy loss by flanking or envelopment. 

Frequently, the enemy was deceived concerning the 
evacuation of a position, which fact led him to attack the 
void left by the abandoned position. The differences be- 
tween empty positions and positions only temporarily de- 
fended by rear guards offering serious resistance, frequently 
misled the attacker into rash attacks, so that he subse- 
quently suddenly attacked fully occupied fire fronts. In 



I 

\ 



THE BATTLES IN 1918 289 

Army Reports of July 21, 1918, it is stated : ''On the south 
bank of the Mame the enemy conducted attacks against the 
positions evacuated by us during the preceding night (19th- 
20th) ; these attacks were made yesterday (20th) in the 
forenoon after an artillery preparation lasting four hours, 
and were executed under heavy fire protection and accom- 
panied by numerous tanks; the latter attacked but smoke 
in the empty positions. Our fire from the north bank, in 
part flanking, caused the enemy heavy losses/' "Our task'* 
said General Ludendorff after the army high command 
had ordered the retreat to the Siegfried position in July, 
1918, "is not for the purpose of gaining terrain nor to hold 
it at any cost, but to reduce the fighting power of the 
enemy/' Possession of the terrain is in itself a visible sign 
of success, of course, but was greatly overestimated at home. 
But possession of the terrain can never justify remaining in 
a tactically unfavorable situation ; however, timely evacua- 
tion of unfavorable portions of terrain may well offset the 
gain of time if it compels the enemy to undertake a new 
deployment or to an advance across difficult terrain under 
our uninterrupted fire. However, we must not fail to re- 
cognize the moral gain to the enemy when he attributes the 
voluntary retreat to the effect of his own measures. But, 
if we could hold or abandon terrain without material disad- 
vantage, conditions were different for the enemy. The re- 
captured terrain was to him native soil, the reoccupied lo- 
calities gave some measure of the extension of his power. 
The German defensive method was something of an inno- 
vation which was demanded by considerations of expediency. 
The "mobile defense" makes enormous demands on the 
troops. A retreat never raises the self-confidence of troops, 
for whatever remains behind in men, materiel and arms, 
falls into the enemy's hands. 

On September 18th attacks by large units on the part 
of the British were begun ; the weak German troops still 
proved their tenacity and desire to attack; heavy fighting 
took place around the debris and ruins of Gauzeaucourt and 
Epehy. In spite of the use of all available means, no deep 
entry was made on September 21st. But the German losses 
had been very heavy. The troops had executed a task that 
was superhuman. The hostile attacks were based upon the 



2d0 DEVELOPMENT OP TACTICS— WORLD WAR 

employment of tanks, followed by infantry with great depth, 
in many cases with hesitation. The artillery prepara- 
tion consisted principally of a short powerful fire shock, 
which changed at the appearance of the infantry into a 
creeping barrage. The appearance of tanks in large masses 
demanded a formation for tank defense arms in great 
depth, as we could not expect that all tanks could be de- 
feated by a single line, especially when their advance was 
also screened by fog. Thus, a certain ''tank defense zone" 
was created between the main line of resistance and the 
artillery protective position in such manner that, in front, 
machine guns and ''tank guns" were formed in groups, far- 
ther in rear light minenwerf ers and field guns which pos- 
sibly gave opportunity for effect towards both flanks. 

Directions from Great General Headquarters of August 
8th pointed out the probability of surprise attacks. Pro- 
tection against those was not to be sought by making the 
defense more dense but by holding forces, in rear, in readi- 
ness. "Their location must be determined with a view to 
meeting hostile surprise attacks. The activity of the units 
placed in the front line together with their decisive conduct 
in the battle zone, must facilitate timely readiness for ac- 
tion on the part of the forces held in rear; these latter 
forces must attack the enemy with a countershock during 
his sudden entry and in conjunction with the divisions in 
the front line, drive him back beyond the battle zone." The 
question of the foreground also was made clearer: "In 
open terrain, when the view is good, the position can be rela- 
tively open. In close terrain and when there is a bad view 
(night fog) the garrison must move closer together within 
the foreground ; whether the movement is forward, toward 
the center, or toward the main line of resistance depending 
on the terrain. In all other portions of the foreground, pa- 
trols must suffice in such situations. The strength runs 
all the way from a simple chain of sentries to the deep, or- 
ganized outpost position, with artillery and minenwerfers. 
In battle the foreground garrison conducts itself like out- 
posts. It will drive off hostile patrols or capture them. 
The foreground garrison does not conduct this fighting rig- 
idly, but with mobility and by using every opportunity for i 
favorable movements. In hostile attacks (to this also belong t 



THE BATTLES IN 1918 291 

patrol operations on a larger scale prepared by artillery 
fire) the foreground garrison, as a general rule, falls back 
fighting — and in pursuance with a prearranged plan and 
according to instructions from the commander — on the main 
line of resistance — ^also laterally, towards the wings. K the 
hostile attack is driven off, the foreground will again be 
occupied as before unless the commander directs otherwise. 
The recapture of the foreground is conducted independently 
by the fighting troops, in so far as possible; but that de- 
mands (if the enemy has gained a strong foothold there) , 
a systematic counter-attack which must be ordered by the 
higher comimander and which requires stronger forces than 
the foreground garrison." 

It was quite correct to point out that, as soon as the 
morale of the troops will be endangered thereby, defense of 
the foreground should be abandoned as a general rule. 

The artillery was to conduct its fight in a formation 
with depth and as a mobile force ; it was cautioned against 
''gun fanaticism;" was informed that ammunition was of 
more importance than number of guns, and that concentra- 
ted fire against the important targets was of great impor- 
tance. Systematic defense by stationary barrage was dis- 
countenanced absolutely. ''Barrage fire is without any ef- 
fect, seldom is correct in its adjustment, is too thin, starts 
usually too late, requires a great deal of ammunition and 
materially endangers the infantry in a mobile battle. Sta- 
tionary barrages should in all cases be strictly forbidden 
in ordinary position warfare. Their place should be taken 
by a concentrated fire of annihilation." 

Again Great General Headquarters pointed out that 
losses could be materially lessened without endangering 
the battle objective by the creation of an extensive fore- 
ground, 1 to 8 kilometers deep, and by systematic, well 
planned depth formation. The designations ^'foreground 
zone" and "main battle zone" are primarily designations 
of command. It can only confuse troops if they have to 
differentiate between these designations. In speaking to 
troops the expression "foreground zone" and "main battle 
zone" should not be mentioned, but the troops should re- 
ceive directions in each ca39 a? tQ wtot zone they must hold. 



292 DEVELOPMENT OF TACTICS— WORLD WAR 

That is their ''battle zone/' Positions in rear thereof , are, 
for the troops, merely ''positions in rear." In the sense of 
an "elastic defense" the troops must free themselves from 
an overestimation of the possession of terrain. In any 
case it must be made clear to the troops in orders, whether 
they must give up ground and how much, and what terrain 
must be in their hands absolutely at the conclusion of a 
battle. Keeping up the fighting power and the offensive 
spirit are more important than possession of a few villages 
an shot to ruins, and of a few trenches ; for that reason, the 
higher and the subordinate commanders must very quickly 
decide clearly whether a counter fire shock or a counter- 
attack is required, and if the losses suffered in any attack 
are commensurate with what we hope to attain by gaining 
ground. We never succeeded at all points in organizing 
positions in rear in sufficient extension. The fighting troops 
required rest, and in addition they were required to fight, 
march and then intrench. It was quite comprehensible 
that the troops were only too willing to do without trenches, 
especially as the hostile aerial reconnaissance could not 
then ascertain without considerable trouble where the main 
resistance was to be made. Obstacles and bomb-proofs 
were of greater value than trenches. It was the old fight 
between shell crater positions and rifle trenches. The ad- 
vantages of a well constructed trench system however show 
themselves immediately as soon as the troops are forced to 
remain for a considerable time at one place. Again and 
again the high command was obliged to point out the im- 
portance of thoroughly constructed positions. Time for re- 
cuperation could not be granted to the tired troops during 
this fatal period ; the strength of the companies in many in- 
stances sank to about 40 men, and supplies failed to suffice 
for the demands either in numbers or value. 

I quote the following excerpts from a letter written in 
the field: 

"The man is supposed to work and fight. That, which 
he has created today, is lost the very next day. Supply 
from the rear is going to pieces, there is a shortage of ra- 
tions. Actual rest the troops enjoy but seldom. For three 
weeks training has been very fatiguing for the officers. The 



-y 



THE BATTLES IN 1918 293 

men on furlough do not return at the proper time; they 
loaf around railroad stations and around the cities and 
appear probably seven days after the expiration of their 
furlough. If they are punished therefor, we lose their 
work; if we let them off without punishment, discipline 
suffers. And then the traffic and theft ! The young officer 
is undoubtedly brave, but training is lacking in the officers' 
corps. He cannot treat his older men properly. The older 
gentlemen (officers) are for the greater part no longer equal 
to their tasks ; they break down, and the battalions and regi- 
ments pass from one hand to the other.'' 

And again from another letter prior to the revolution : 

''The army has been fighting for six weeks continually 
on the retreat, partly in dissolution, thrown out of one posi* 
tion into the other, the troops without relief, played out, 
airplane attacks by day and by night, firing on the quarters, 
lack of ammunition in consequence of the railroads going 
to pieces, — ^that was the picture offering itself to the eyes 
of the man behind the scenes. The men lived only in the 
thought : why should I let myself be shot to death, peace 
will soon come, and so they gave themselves up as prisoners. 
Our enemies of course soon perceived this state of affairs. 
Therefore their continual desire is to gain time, so as to com- 
pletely annihilate us. Would we have acted differently? 
But the answer of the enemy to our offer for an armistice 
has had a very different effect on the entire front than the 
enemies thought. The men perceived the hostile desire to 
annihilate us and had come to their senses. For the past 
eight days I believe that an entirely different sentiment has 
taken possession of the men. We have learned to organize 
the retreat, there are again troops present, and ammuni- 
tion, and everything runs smoothly along tranquil roads. 
The front holds out. Our positions are adapted to the at- 
tack methods of the enemy. The enemy wiU find that out. 
Our men again have confidence, and then we wiU be superior 
to every opponent." 

The much abused "militarism" misunderstood by its 
opponents at home but well known and feared by the 
enemy, has stood the test splendidly in these dark days and 
proved its justification. 



Gmciusion 

Tile World War found a premature end through politi- 
cal events in Germany, without additional battles having 
been able to prove the correctness of our new defensive 
method. Great General Headquarters had the fullest con- 
fidence in these general rules, though it did not underesti- 
mate the value of the hostile procedure of the offensive, the 
influence of superiority of numbers, and the importance of 
the hostile fighting means. The high command reckoned on 
success, or at least on gaining time, in the defense of the 
expects Antwerp— Meuse position and also in further bat- 
tles. But things happened differently ! The army at home 
knifed the undefeated field army in the back, like Hagen of 
old did to the unconquerable valiant hero, Siegfried. The 
lessons of history were not written for us, our peoples 
trusted to the ^'goodheartedness" of unforgiving enemies, 
who concealed their hatred under the banner of the League 
of Nations. Our peoples were to find out for themselves 
"Vae vicHar 

Supported by weak allies, the German army had op- 
posed for almost four and one-half years a world of enemies, 
who had at their disposition all means of industry and tech- 
nique, of foodstuffs and raw materials. If we seek for the 
criminal responsible for our fall, he should be sought, not 
in the army, but he should be sought in the ranks of the 
leaders of our political parties, in the ranks of the men who 
placed pursuit of their own ends above the weal and woe of 
Germany. In the territory of the enemy, the army has se- 
cured its greatest achievements, and was very near in reach- 
ing an advantageous peace. In the foregoing pages no one 
should seek for errors ; they are only for the purpose of in- 
formation of what we have done, so that the reader may 
proudly look back on the achievements of our people under 
arms. The ignominious peace of Versailles is for the sole 
purpose of making us helpless and consequently no longer 
dangerous. Can the annihilation of our peoples be God's 
will? We do not believe so. Our peoples have given too 

294 



CONCLUSION 295 

many proofs of their efficiency to permit a stroke of the 
pen of the Entente to annihilate them. In spite of aU diffl* 
culties piling up in front of us, I firmly believe in the future 
of our peoples. German power may be lamed for years, 
but the German spirit cannot be killed. The German Spirit 
will continue to live ! 

When later on a new aurora of history embellishes our 
days, then may our peoples remember also our heroes of 
the World War. Work, suffer, and fight like they did, and 
thus Germany's future will be secure. And then, also, the 
blood of our heroes who fell in the belief of Germany's vic- 
tory will not have been spilled in vain I 



I 



i^ \ 



4 



■ 

{ 



^A^ 



^k