Ohio History Journal




BRADDOCK'S DEFEAT

BRADDOCK'S DEFEAT

by JOHN SCHLEBECKER

Hiram College

One of the things with which the reader of early American

history is most impressed is the remarkable courage and wisdom

of the American soldier. The Europeans it seems just did not

understand the ins and outs of forest fighting. Only the American

seemed to know that effective fighting on the frontier was by neces-

sity done from behind trees. History books are full of pictures of

minute men shooting redcoats, but they do not have many pictures

of minute men running away from redcoats. History books do say

that Von Steuben came to America to teach Patriot soldiers how

to fight like Europeans, but they do not point out clearly enough

that fighting behind trees was in those days a rather silly way to

fight. And this last fact, of course, explains the first. Part of the

purpose of this paper is to show how and why fighting from behind

trees was in the early days a poor way to fight.

From this misconception of forest fighting has come some

peculiar reports of forest battles, the most peculiarly reported of

all being Braddock's defeat at the Monongahela. In this paper the

story of that battle will be retold and some sort of evaluation of it

will be made. The evaluation will not be the customary one.

In order that the evaluation of the battle have validity, there

will be a brief review of the arms, soldiers, and tactics of the period.

The period to be considered is the last half of the eighteenth cen-

tury; the date of the battle was July 9, 1755.

The chief weapon of the armies of this time was the musket.

This weapon was inaccurate at any range over one hundred yards

and even at that distance or less it was so inaccurate that individual

firing was relatively harmless.1

Of the other weapons of the period only two require mention

here. One was the familiar three-sided bayonet, and the other was

the light field cannon used to fire upon infantry formations in the

field.

1 Edward M. Earle, Makers of Modern Strategy (Princeton, 1943), 51.

171



172 OHIO ARCHAEOLOGICAL AND HISTORICAL QUARTERLY

172    OHIO ARCHAEOLOGICAL AND HISTORICAL QUARTERLY

 

The typical soldier of this period was the professional who en-

listed for long terms and who fought as a business.2   He was

chiefly interested in his pay; he fought without political passion.

If his food supply was reduced or if operations should prove to

be unduly strenuous his morale would drop.3 Under these circum-

stances he would be inclined to desert. So prevalent was desertion

that men were required to march in close order when going and

returning from bathing. They could not be trusted to forage for

food, to pursue an enemy, or even to take cover in a fight because

in these activities they were liable to get out of their officer's sight

and desert. Even camping near forests was dangerous because of

the temptation it offered to deserters.4

The Indian also was a mercenary soldier, with this difference:

his pay was chiefly in loot and so he was inclined to stick around

the battlefield a little longer to see what he could pick up.

Tribal warriors fought under their own chiefs and generally

near their homes. If a clan or tribe should be bored or should

think that the fight was not to their advantage they would take a

rest. They were liable to do this any time the idea occurred to

them.

Tribal warfare is different from civilized warfare in that all

men of a tribe are related, and therefore a casualty is a deep per-

sonal tragedy to all the men of that tribe.5 Thus any considerable

number of casualties in a tribe may cause that tribe to withdraw

sooner than a civilized platoon or battalion would withdraw.

The range and accuracy of the musket determined the tactics

used by Europeans. Since the musket was inaccurate and had a

short range the men advanced into battle shoulder to shoulder and

fire began at only a little over one hundred yards. Each battalion

fired as a unit at the word of command. The battalion was usually

three ranks deep and the fire from it was terrible and deadly.6 The

ultimate object of the fight was to close with the enemy on a

bayonet charge.7

 

2 Ibid., 50.

3 Ibid., 51.

4 Ibid., 55-57.

5 T. E. Lawrence, Seven Pillars of Wisdom (Garden City, 1938), 194.

6 Earle, op. cit., 51.

7 William Wood and Ralph H. Gabriel, The Winning of Freedom (The Pageant

of America, VI, New Haven, 1927), 197.



BRADDOCK'S DEFEAT 173

BRADDOCK'S DEFEAT                     173

 

It took time to draw the armies up into battle line, and thus

if one side decided to leave while the other was forming there

could be no real battle.8

The technique of destructive pursuit was undeveloped at this

time.9

Firing was done by alternating platoons or battalions. This

was done because the difficulty of reloading and the nearness of the

troops made bayonet charges particularly dangerous.10

When troops were marching through woods the tactics of the

day required that they march in platoon formation so that they

could be more easily maneuvered in case of action. In case the

vanguard did meet the enemy the main force was to be halted until

the commanding officer knew what was happening. This because

once the troops were formed it was difficult if not impossible to

rearrange the formation. In addition, a commanding officer should

know something of the size and disposition of the forces opposing

him before going into action.11

The tactics of the Indians and colonials was a hit and run kind

of warfare. The Indians probably used it because in tribal warfare

(aside from blood feuds) the object is to sustain minor losses, kill

as much as necessary, and steal everything that can be stolen.

Where loot rather than murder is the primary object the tactics are

liable to be so constructed that efficiency in killing is sacrificed for

safety in getting home with the bacon. The colonials probably

used this type of warfare because they were not numerous enough

to put massive armies in the field.12  Most historians hold that

massive armies would not have been effective anyway. This argu-

ment causes one to wonder why the colonials were always trying

to raise armies to fight the Indians if armies were not the answer.

The first thing that Braddock tried to do upon arrival in

America was to turn the colonial troops into real soldiers. He had

them drilled in the European manner by British officers and en-

forced an old regulation that colonial officers were to have no rank

 

8 Earle, op. cit., 51.

9 Ibid., 52.

10 Wood and Gabriel, op. cit., 196.

11 Stanley Pargellis, "Braddock's Defeat," American Historical Review, XLI

(1936), 264.

12 Wood and Gabriel, op. cit., 17.



174 OHIO ARCHAEOLOGICAL AND HISTORICAL QUARTERLY

174    OHIO ARCHAEOLOGICAL AND HISTORICAL QUARTERLY

 

while regular officers were in the field.13 These actions only made

the colonials bitter and caused a drop in morale.

Soon after setting out the troops were subjected to an epidemic

of fever and dysentery. This naturally caused a considerable lower-

ing of morale which a lack of sufficient food only made worse.

The line of march was over rough terrain which added to the

woes of the British soldiers. These became disheartened at their

hardships, while the colonials continued to be soured by army dis-

cipline. The whole army was in a poor psychological condition.

Engineers preceded the troops, clearing a roadway twelve feet

wide while squads of men were sent out in front and on all flanks

to guard against surprise.  These squads, however, were not

numerous enough to offer effective security. The entire length of

the line of march was about four miles.

In order to avoid dangerous defiles on the route, Braddock

decided to ford Turtle Creek and then ford the Monongahela a few

miles down. It was on the other side of this last ford that 1,300

British regulars and colonials met the French and Indians.

Contrecoeur, the commander of the French at Fort Duquesne,

was in favor of retreat. However, one of his captains, Beaujeu,

felt that if he could get the Indians to come with him he might stop

the English. Three times he tried to talk the Indians into accom-

panying him, and on the third try he was successful. His total

force included 230 French and Canadians (70 of whom were

French regulars), and 637 Indians, a total of 867 men.14 Of this

force three men were French officers: Beaujeu, Dumas, and

Ligneris.

The French plan of battle called for an ambush at the ford

which Braddock was to cross, but at the crucial moment the Indians

decided to leave. Beaujeu managed to talk them into continuing

the expedition and had barely managed to reassemble his men

when he met the advancing British. Despite this recent disaffection

the morale of the French and Indians was considerably higher

than that of the British.

Beaujeu, dressed as an Indian, came walking jauntily over the

13 William M. Sloane, The French War and the Revolution (New York, 1898),

42.

14.Ibid., 44.



BRADDOCK'S DEFEAT 175

BRADDOCK'S DEFEAT               175



176 OHIO ARCHAEOLOGICAL AND HISTORICAL QUARTERLY

176   OHIO ARCHAEOLOGICAL AND HISTORICAL QUARTERLY



BRADDOCK'S DEFEAT 177

BRADDOCK'S DEFEAT                     177

 

brow of a hill. On the other side a British engineering officer was

mapping the route. Both men were somewhat startled.

Beaujeu turned and with his hat signaled his forces which de-

ployed and opened fire. With the admirable precision and cool

courage of European soldiery the British advance column under

Gage wheeled into line and fired several volleys in the general

direction of the enemy. The British casualties up to this point were

insignificant. Gage's cannon opened fire quickly and the Canadians

and Indians fell back in confusion while the British infantry con-

tinued to fire. Beaujeu fell dead.

It was at this point that seventy French regulars, fighting in

European style, won the battle. With the courage born of despair

Dumas advanced with the French regulars and returned a fire which

was sharp enough to halt and surprise the British.15 It seems

probable that after all of the talk that they had heard about Indian

ambushes that the last thing the British had expected to meet was

effective European fighting. They stood it for about fifteen minutes

and then broke and ran.16 Meanwhile the Indians and Canadians,

taking heart at the stand of Dumas and the regulars, were gradually

reassembling behind cover and opening fire upon the British.17

As the vanguard of the British broke and fled it ran into the

main body of troops under Braddock which were advancing to the

front. At this time Braddock had no clear idea of what was happen-

ing and the troops were in a state of confusion because the officers,

including Braddock, were not exercising proper control.18 The sud-

den influx of the advance guard only increased the confusion. Com-

plicating this confusion was the fact that the main body was

split in two sections on either side of the wagon train. This reduced

the mobility of the troops and made it more difficult for them to

form for action.19 In any case, the troops under Braddock were

still in column of march rather than in battle formation because

"the order to form line of battle had either not been given or had

not been heard."20 This last fact is particularly important. It means

that the main body of the British at the time of the encounter was

 

15 Francis Parkman, Montcalm and Wolfe (2 vols., Boston, 1899), I, 223.

16 Wood and Gabriel, op. cit., 76.

17 Parkman, op. cit., 224.

18 Pargellis, loc. cit., 260.

19 Ibid., 269.

20 Wood and Gabriel, op. cit., 76.



178 OHIO ARCHAEOLOGICAL AND HISTORICAL QUARTERLY

178    OHIO ARCHAEOLOGICAL AND HISTORICAL QUARTERLY

 

unprepared to fight in any manner be it European or American. Too

late they attempted to perform this complicated maneuver on rough

terrain, but being in disorder themselves and at the same time meet-

ing the advance guard they never completed the formation. The con-

sequent panic resulted as much from the muddling confusion as

from the enemy fire.21

Dumas, noting the position of the British troops, seized the op-

portunity to send out flanking parties along both sides of the

enemy column. Had the British been in line of battle this maneuver

would probably have been impossible due to the extension of the

flanks and the fire power of the British. In any case, it would

not have been as effective, because the British could have launched

bayonet charges from the line of battle. As it was the British were

nearly defeated already.

Meanwhile the Virginians had dispersed behind trees and were

wasting their powder on marksmen as unseen as themselves.22 The

Virginians did not suffer many casualties, neither did they inflict

many. The total French and Indian losses were only forty officers

and men.

At this point of the battle the worst fire was coming from a

hill on the right. This hill should have been taken as soon as the

action began, or even before, but the command to take it was

not given until after the army was helplessly piled up.23 At that

time Braddock ordered Lt. Col. Burton to take the hill. The colonel

managed to get about one hundred men to follow him, but upon

being wounded himself his men retreated. Failure to take the hill

sooner shows Braddock's incompetency; failure to take the hill at

all shows the complete dependence of the soldier of the day on his

officers.

For three hours the French and Indians poured in fire while

the British held. They "formed in oblique angle lines, twelve or

more deep, while those in the rear remained in column formation."24

They were never able to form an effective line of battle. At the

same time the Virginians counseled the British to take cover, an

action which Braddock refused to permit. The British made excellent

 

21  Pargellis, loc. cit., 263.

22 Justin Winsor, The Mississippi Basin (Boston, 1895), 362.

23 Pargellis, loc. cit., 265.

24 Ibid., 261.



BRADDOCK'S DEFEAT 179

BRADDOCK'S DEFEAT                   179

 

targets, and at the same time they had long since ceased firing by

battalions. Their fire power was gone. After three hours Braddock

ordered a retreat, and soon thereafter he was mortally wounded. The

retreat turned into a rout, but the French and Indians did not

pursue. The battle was over.

The British opportunity for victory came with the advance

of the main body. With the terrific fire power of the battalions

(and the fire was tremendously effective and deadly),25 plus the

possible bayonet charges, the enemy could hardly have held their

ground. Through stupidity or bad luck, however, the British arrived

in column, and that was that.

Parkman holds that a bayonet charge would have been futile

against such a lurking enemy who would dodge from one tree to

another, but his reasoning hardly holds up.26 Later during the Revo-

lutionary War the British found that the only way to drive men

from cover was to charge them. The Americans, on the other hand,

when charged upon soon discovered that the only way to keep from

being killed in the rush was to stand like Europeans and fire by

battalions.27 They had to correct some of the foolish notions that

they had picked up from fighting Indians, who after all did not

use bayonet charges.

It seems probable that Braddock's greatest failure was not in

what he wanted to do but rather in what he did not do. If he had

formed his troops properly he might have done some effective

damage. His failure to form his troops might be attributed to the

fact that the murderous fire of an unseen enemy threw his troops

into confusion and prevented any useful action. This is undoubtedly

part of the answer, but it is entirely too pat an answer to be the

complete explanation. Added to the fire of the enemy was the fact

that the British officers " 'messed up' their formations and never

gave their soldiers a chance to demonstrate that Old World methods,

properly applied, might have won the day."28

Another error of Braddock was this: he did not observe the rules

of tactics of his own day in the matter of marching through forests.

Bland, the leading tactical writer of the period, stated that when

 

25 Earle, op. cit., 51.

26 Parkman, op. cit., 225.

27 Wood and Gabriel, op. cit., 197.

28 Pargellis, loc. cit., 253.



180 OHIO ARCHAEOLOGICAL AND HISTORICAL QUARTERLY

180    OHIO ARCHAEOLOGICAL AND HISTORICAL QUARTERLY

 

marching through woods the army should be divided into platoons

for easier maneuvering. Whatever formation Braddock's troops

might have been in, it was not a formation by platoons.29

When the action first started Braddock should have halted the

main body until he knew what was going on at the front. Instead

he went blindly forward.30

In conclusion it should be said that the error of Braddock was

not that he did not use American tactics, but rather that he did not

even use good European tactics.

In regard to this battle many historians make several explicit

or implied contentions which deserve some notice since they recur

in almost all accounts. Briefly stated they are five in number and

include the following ideas or notions:

1. The red-coated British in a huddled mass made excellent

targets. This is certainly true. They were like so many pool balls

racked up in the middle of the table. A shot in the general direction

of the mass was sure to hit something. It is remarkable that in

consideration of this fact some have attributed a great deal of

marksmanship to the French and Indians. They may have been

marksmen, but this action is hardly proof of it. The great losses

in British officers can be attributed not to sniper activity but rather

to the fact that in those days officers were accustomed to stand in

front of their troops in order to direct their movements and fire.

Under these circumstances the officers were most likely to be killed

by one side or the other.

2. The British volleys were generally wasted. Actually, the

first volleys of the vanguard were not wasted. In fact they dispersed

and terrorized the Indians and Canadians. After the retreat of the

vanguard the British can hardly have been said to have fired volleys

with any consistency. If the troops had been properly formed and

had fired by platoons it is not at all certain that the volleys

would have been wasted.

3. The British should have taken cover and fired independ-

ently. Usually this is followed by an implied corollary which is:

Had the British taken cover they would surely have won the battle.

This brings up the subject of relative and absolute standards. If a

 

29 Ibid., 264.

30 Ibid.



BRADDOCK'S DEFEAT 181

BRADDOCK'S DEFEAT                   181

man should bet ten dollars on a losing horse it might be said that

he would have been better off to have bet only five dollars. This

is obviously true. Relatively speaking, however, it would have been

better still if he had bet on the winning horse. It is hard to say what

the absolute good in this case might be, but it would probably be

best not to bet anything.

Now a battle is a gamble, so it is useless to discuss whether or

not it is right to take a chance in war, because one cannot help but

take a chance. However, there are intelligent bets and unintelligent

bets, and Braddock made some unintelligent bets. Relatively speak-

ing, it would have been much better if he had followed the advice

of the colonials and allowed his men to deploy. This does not

mean that he would have won the battle thereby, but only that his

losses in life might have been smaller. His army would probably

have disappeared in any case. There are several reasons for this.

In the first place the British soldiers were fighting for a living,

and after the first few minutes even the most stupid soldier could

have seen that there were easier ways of making money. Had Brad-

dock allowed his men to disperse they would have done just that.

Some may claim that this is mere speculation, but the history of the

times shows that it is well founded speculation. From Frederick the

Great to Washington the armies of the world were cursed with the

tendency of troops to leave the field when things got rough. The

extreme penalties which Frederick placed on desertion, and the

difficulties which he encountered toward the end of his reign in

keeping an army together are examples in point.31 Another ex-

ample might be the difficulties that the Americans experienced in the

Revolutionary War in keeping an army in the field. One of the

reasons for close order fighting of this period was that in close

order the men could be watched and if any one tried to desert he was

shot or run through on the spot. It would have been foolish to expect

the army of Braddock to act any differently from any other army of

 

31 Earle, op. cit., 55-57.



182 OHIO ARCHAEOLOGICAL AND HISTORICAL QUARTERLY

182     OHIO ARCHAEOLOGICAL AND HISTORICAL QUARTERLY

 

the time.32 Only the Americans seemed to be too dense to realize

this.

In the second place, if the British had deployed, the French and

Indians would have had no targets and the same might be said for

the British. The net result would have been another backwoods

skirmish with both sides retiring for a while, and then an ultimate

attack by the British on Fort Duquesne. All of this presupposes that

the British army would hold together, an unlikely occurrence. In any

case, the British by frontier methods of fighting could not hope for

a positive decision until they could manage to force the enemy into

a battle in the open. The British soldiers were not trained for fron-

tier fighting, neither were they armed for it. It took many years of

training to make a good frontier fighter, and even then the arms of

the period did not allow it to be very effective fighting. A conclusive

decision could only have been reached by the use of European

tactics.33

Another reason why Braddock was right in not following

Virginian instructions was that he had a superiority in numbers and

in discipline which he would be less than wise not to use. As long as

he had any chance at all to use this superiority he would be foolish

to dissipate it by dispersing his men.

From the foregoing statements it is possible to see that although

deployment would have been better than what was done, at least

in respect to loss of life, it was not a good solution.

What then would have been a good solution? Proper use of

European tactics, which tactics culminated in the bayonet charge.

There was nothing in America which would make the bayonet charge

ineffective. In fact, as has been pointed out, it was the only way of

dislodging men who were hiding behind trees. And as Revolutionary

experience showed, it was not only effective in that respect, but it

usually meant defeat for the deployed forces since at the beginning

of the charge most of the defenders would be inclined to fire. This

 

32 There are several reasons why these same factors did not affect the Virginians,

Canadians, or Indians in the same manner that they would have affected the British:

(1) these soldiers were not mercenaries in the same sense as were the British;

(2) in their cases a certain amount of patriotism was involved; (3) they were used to

this type of fighting; fighting where every man was a general; (4) they were not

suffering many casualties; (5) on the French side at least they were winning. On

the other hand, the Indians and Canadians did start to leave at the beginning of the

fight and there is no sure way of determining how many if any of these fighters

(including the Virginians) did desert during the battle.

33 Pargellis, loc. cit., 265.



BRADDOCK'S DEFEAT 183

BRADDOCK'S DEFEAT                        183

 

would leave them with empty guns, foolish expressions, and a pair

of legs for running.

4. The colonial militia were the better fighters. They almost

saved the day. The Virginians disrupted the British discipline

further and managed to save their own skins. There is no reason

to suppose that they did much else. They did not hold off a charge

by the French and Indians, because these forces could not effectively

charge from their extended positions, and besides they were having

too much fun as it was. The colonials may have covered the retreat

of the British, but this was no great service since the French and

Indians did not pursue. After all, the French, Canadians, and Indians

were no great exception to their times; all of them fought for a

living as a rule. There was no reason why they should pursue an

enemy when there was so much loot lying about on the bodies of

the dead who were left behind. Besides, the idea of pursuit was un-

developed at this time. If the Virginians covered the retreat, all that

that means is that they were not quite as fast as the British in leav-

ing the field.

These are the only claims to glory that the Virginians have,

and it is just as well to leave it at this.

5. The British were surprised.34 For that matter so were the

French. During the first few volleys both sides were surprised. The

French were surprised not only at meeting the British but also at the

fierceness of the British fire. The British were surprised on exactly

the same counts. A quotation from Clausewitz will help explain why

with this double surprise the side with the fewer forces won the

battle. As Clausewitz said: "If one side through a general moral

superiority is able to intimidate and outdo the other, then it will be

able to use the surprise with greater success, and even achieve good

results where properly it should come to ruin."35

That is just about the essence of the matter; the French and

Indians had moral superiority and they intimidated the British.

British and colonial monkeyshines made things easier for them, but

the victory went to the side with the greater confidence and the

 

34 "We are not now speaking of the actual raid, which belongs to the chapter on

attack, but of the endeavor by measures in general, and especially by the distribution

of forces, to surprise the enemy, which is just as conceivable in a defense, and which

in tactical defense is particularly a chief point."  Karl von Clausewitz, On War

(Modern Library ed., New York, 1943), 142. Italics are the author's.

35 Ibid., 645.



184 OHIO ARCHAEOLOGICAL AND HISTORICAL QUARTERLY

184    OHIO ARCHAEOLOGICAL AND HISTORICAL QUARTERLY

 

greater daring. It is apparent, then, that the following conclusions

are justified:

1. During the first few minutes of the fight it was the French

regulars who saved the French from defeat. The French counterpart

of the Virginians, that is, the Canadians, were among the first to leave

the field when the British opened fire. This fact does not do much to

prove the vaunted superiority of American militia over European

regulars.

2. Before going forward Braddock should have found out

what was going on. Then he should have arrived at the scene of

action in good order and in line of battle.

3. If Braddock had wanted to win the battle he should have

formed his troops and eventually ordered a charge.

4. If he had wanted to lose the battle or at best gain a stale-

mate, he could either have done what he did do or else what the

Virginians wanted him to do.

5. The colonial tactics were not good tactics in an absolute

sense. They were the tactics of men who could not put a real army

into the field.

6. The tactics used by Braddock were not good European

tactics.

7. There were two crucial points in the battle. The stand of the

French regulars and the arrival of Braddock in line of march. All of

the errors of the British stemmed from these two points, the genius

of the French consisted in utilizing them.

8. The morale of the French and Indians was better than that

of the British.

9. And finally, if the proverb is true, if desperate strength

does indeed make a majority, then Braddock can be pardoned for his

defeat. He was outnumbered.