Table of contents
Introduction
A
Prisons
Introduction
It has often been said that, apart from the military aspect
consisting in the war against the British, there was also
a civil war facet consisting in the conflict between Patriots
and Loyalists. For historians who wish to get an
understanding which goes beyond the enumeration of battles
and skirmishes the purely military features are not the most
interesting part because they mainly follow the rules of warfare
of that time: How many guns, cavalry and infantry on each side?
How to manage the procurement of powder, shells, carriages,
food, uniforms? How to recruit and pay the troops? How to reduce
desertions both in the militia and in the Continental Army.
All European countries in their endless wars were confronted
to these same challenges.
The last question is of course strongly connected with the
civil war aspects. Each side tried to recruit troops and
to identify the spies and recruiting officers sent by the other
side. We think that these aspects need to be studied
more fully and more systematically. This is particularly true
for the judiciary aspects, i.e. number of civilians
in custody, number of trials, number of banishments,
number of death sentences, number of pardons and executions.
It is the lack of global official judiciary data which
makes the task of historians exceedingly difficult.
For the French Revolution there are official data of the number
of people arrested, tried, sentenced, executed. For instance,
in a book by Donald Greer there is a map of France with the
numbers of executions in each departement. It appears clearly
that these numbers are strongly correlated with the insurrections
of Royalists, the latter being
supported by France's neighbors and particularly
by England; see on the map the region around Toulon in the south
and the
region of the Vendee in the west and see also our
discussion below about the landing in Brittany.
In the Thirteen colonies there are no official statistics whatsoever.
Even for important data such as sentences of civil courts
or courts martial, no annual data were published.
In contrast the British have recorded data of the number
of prisoners of war in British custody
which allowed American historians to
estimate the number of fatalities on prison ships.
Prison ships have also existed on the Patriot side but for them
we do not have any data and therefore no death rates
can be estimated.
In several states suspected Loyalists were given the choice
between taking the oath of allegeance or staying in prison until
a possible trial several months later.
Only the
richest Loyalists could afford to pay 100 or 200 dollars
(or even more) for their security bond.
Naturally, nobody would expect the numbers of executions in America
to be in any way comparable to those of the year 1793 (called
Terror time) in France.
To begin with, there are two major differences.
These are certainly important factors but they are not
sufficient to explain the lack of data.
It seems that
in many cases the relevant data were lost. For instance,
we were told by archivists that in New York state
there are no remaining archives for the sessions of the
court of Oyer and Terminer (a court particularly
in charge of state trials). Often existing series stop in
1776: that is the case of the "American Archives" collected and
published by Peter Force. In Demond (1940,p.262) one reads:
"The state of North Carolina is almost destitute of source material
for the period of the Revolution" (except for what is found
in the Colonial and State Records which is limited).
So far we have under our eyes what can be called
a "patchwork" picture of the conflict between Patriots
and Loyalists. We use this word not only because
most of the studies are at state level but even at that
level one rarely finds data for key variables
(e.g. numbers of prisoners or trials) that are
global state level estimates.
Summary
Let us illustrate these three levels through a few examples.
With respect to mob violence.
(a) would consist in describing 3 or 4 events. (b) would
consist in recording ALL episodes of mob violence at least
in a limited area as documented in newspapers.
In (c) the events of mob violence would be recorded by police
or other state agencies (e.g. in present days, the FBI)
In his history of the Loyalists in South Carolina
Demond (1964) has a chapter entitled "Suffering of the Loyalists".
He uses the patchwork method described in (a).
He is well aware that this method is not satisfactory for
he tells his reader (p.119): "Throughout the latter part of the
war the district courts were most active in indicting the Tories
for treason and an indictment usually resulted in conviction."
An account based on all these trials would be close to the (b)
level. Certainly Demond would have tried this way if transcribed
records had been available. Later on in the book, Demond repeats the
same plea (p.152). "In 1782 when Faning [a loyalist leader]
saw his followers convicted of treason and hanged, in retaliation
he put to death his prisoners". However, this statement is not
supported by any evidence. How then does Demond know that?
Here our plan is to present those judicial data that we can
find and to attract attention on existing gaps.
Our study will be focused on few aspects but
whenever possible we try to propose
a fairly comprehensive and systematic view.
This is a plan which is neither easy nor comfortable.
It is not easy because usually the data that we need
are not available. It is not comfortable in the sense that we
will have to harvest our cases one by one with in addition the
unpleasant conviction
that many additional cases will not be caught in our net.
For instance it will take us much time and effort to
make a list of Loyalists who were executed whereas that information
should have been collected by the Ministry of Justice or,
in the case of court martials, by General Washington's headquarters.
To explain what we have in mind,
let us consider the question of the
mob violence against Loyalists or
officials of the British government.
It is known to have occurred repeatedly in northern colonies
and in fact it started well ahead of the American
Revolution.
Usually, historians limit themselves to the description
of a few cases.
Clearly, this cannot give a global view.
There is a world between
isolated attacks occurring say once a month and relentless
attacks several times a week. In the first case it may be just
an expression of popular discontent in the form of
practical jokes, whereas in the second it would create
an atmosphere of terror among the persons who are targeted.
An historical parallel would be the
attacks against opponents in fascist Italy.
In other words, depending on the frequency, the picture changes
completely.
The same observation holds also for many other
types of events. For instance, desertions do not attract much
attention unless they reach a degree which threatens the
very fighting capability of the army.
It is well known that there were collective trials
of Loyalists suspected of siding with the enemy. Courts of
Oyer and Terminer which had been commonly used in the British judicial
system for the trials of state prisoners had the same function
in the thirteen colonies and, after 1776, in the young republic.
There is so far no agreement among American historians
as to the frequency of such trials. Some hold that they
were rare, others describe a substantial
set of cases but add that anyway after being sentenced
all defendants were pardoned, especially when they had
been sentenced to death.
A third group of scholars wonders if the cases already known is not
just the tip of a large iceberg.
What makes such a study challenging is the
fact that so far judicial archives have been largely neglected by
historians, precisely perhaps because the question of dissent
was somewhat sidelined. If you search the index pages
of classical accounts of the Revolution you will not often
find the entry "Treason trials" or "Oyer and Terminate courts".
This
can be seen fairly clearly in the archives of Virginia.
Whereas in 1778-1779 one sees almost only issues
concerning military procurement,
in the early 1780s the part devoted to treason trials surged.
Probably there were also treason trials earlier but they
did not find their way into the records kept in county or
state archives.
In the following documents we try to shed some light on these
issues. The specific questions on which we will focus were
selected because we were able to find appropriate data.
In other words, our investigation was driven by data availability.
A
Prisons
The first document is a list of 50 prisons in use in the 13 colonies.
Why is such a list important?
Thus, if they were available, data giving the total population of
inmates would represent useful estimates of the strength
of the Loyalist population. Unfortunately, almost no
jail occupation data have been transcribed and published in
the vast movement that took place in the second half of the
19th century when the 13 states funded the printing and publication
of many historical documents. Prisoners of war have attracted much
attention but state prisoners did not. Naturally, it is
understandable that the very notion that there were state
prisoners, whose only crime was to be faithful to their
king, was perhaps not palatable for a nation
in which freedom is so highly praised.
B
Repression against Loyalists
In the same line of thought it is hardly surprising that
in American historiography
the repression against Loyalists is belittled or overlooked.
In this section we present two documents which
illustrate this bias. The first document is a 3-page article
(p.40-42) published in the "Canadian Loyalist Gazette".
Incidentally, it is worthwhile to notice that the illustrations
were chosen by the publisher of the Gazette, not by the authors.
The picture on the top of p.41 represents a person (meant to be a
Loyalist) hanging from a tree. At first sight this appears
to be in contradiction with the common view that mob violence
did not result in any death of Loyalists. Here, the fact that
the person is hanging from a tree obviously refers to mob violence
because in a lawful execution there would be a gallow.
However, this picture can be found on Internet and there
it has a caption
that says that it is not a person but an effigy. It was indeed
common to express hostility against someone by hanging his effigy.
Yet, that is not the end of the story because in Moore (1860,p359)
there is an episode in which a dissident minister is hanged by a
mob.
The account gives fairly detailed information.
The location is Charleston, the date is 1 December 1776, the name
is John Roberts. However the primary source remains unclear.
The following article is an extended version of the previous
paper.
C
Draft of "Forging consensus"
including a list of some 250 legal executions
The following document (220 p.)
It contains also an investigation of the fate of
Union citizens during the Civil War
who were in favor of a negotiated peace
with the Confederation. Like the Loyalists, these people
were considered as domestic enemies and often accused
of treason.
All these cases illustrate policies aimed at
forging consensus. Below we indicate a number of issues
considered in the manuscript. Once we have received some
feedback the manuscript will be completely rewritten.
Sorry for its present "untidy" shape.
What makes the American Revolution unique
List of executions (p.78) with indication of:
Name, Date of execution, State, Source.
Court martial sentences
Elusive consensus in the Civil War
Consensus forging in the First World War
References: archives, books, articles
The following file contains accounts of court-martials
in the Continental army. The source is a database and search
engine for letters exchanged by the Founders of the young
republic. It contains also "General orders" issued
at Washington's headquarters.
This database was set up by the National Archives
and the Library of Congress. The present compilation
is a very provisional version.
It can be noted that the army had also local headquarters
at state level or for theaters of war comprising several
states. Pardons (or reprievels)
could be issued by General Washington
but also by field Major Generals, particularly after 1778.
Death sentences issued by court-martials
E
Low key control of suspected Loyalists
How did Patriot investigations by various committees impact
the daily life of American citizens? Note that here we do not
restrict the question to avowed Loyalists for it will be seen
that even citizens who wanted to stay "neutral" were ordered
to appear before the committees for questioning.
For historians the challenge is to find a source describing
the daily interaction between the population and the
investigations conducted by the Patriots. Here the important
word is "daily". The records of the various committees give
some information about their work but most often what is
reported is a selection of the activity of the committee.
Incidentally the term "minutes" that appears in the titles
of such archives is a misnomer for they are rather
summaries.
However what we are looking for seems to
exist in the archives of Albany county. Victor Hugo
Palsits, the scholar who edited these archives, tells us
in his introduction that in all other counties of
New York State such archives may have existed but were
lost or destroyed. His own words are as follows.
For Albany county there are 4 volumes totaling 1,071 pages.
The first two cover the period: 17 Oct 1776 - 12 Oct 1777
whereas the last two cover: 15 Apr 1778 - 13 Aug 1781.
The information provided by these volumes was compiled and
arranged into a standardized form comprising the following
columns:
This standardization will enable us to count the
interactions between the Committee and the citizens
and to see its evolution in the course of time. It will
be seen that altogether in Albany County almost 50% of
adult males came in contact with the Committee for
one reason or another.
As more and more Loyalists were imprisoned,
moved to New York or went abroad
one would of course expect the "density" of Loyalists
in the free population to decline and this is indeed
what is observed.
The number of monthly "collisions" decreases from 170
in 1776 to about 50 in 1780. Note however that there is a bias
in the sense that in the course of the war the conflict
moved from the northern states to the south. If a similar
study could be done in the south it might reveal an
increase rather than a decrease.
Naturally, it is not surprising that the events occuring
in Albany County also reflect the major events of the conflict.
A clear illustration is the fact that in Oct 1776
some 120 prisoners were sent from New York State to the
town Exeter in the state of New Hampshire. This suggests
that Exeter had a prisoner camp with a high capacity.
F
Number of imprisoned Loyalists
Estimate of the number of imprisoned individuals derived
from Loyalist compensation claims
As already observed despite the fact that there were
over 50 prisons in the northern states (see above)
almost no prisoner list records are available. As there were
town jails, county gaols and provincial prisons such archives
should be available at various levels. It is possible that
lists of prisoners are still hidden in the archives just waiting
for historians to get interested in them. An alternative
explanation would be that, for some reason, they were
destroyed. Actually, the two explanations do not
exclude each other. Lists of prisoners (or merely counts) would
constitute direct evidence. Lacking that one needs to resort
to indirect evidence. Such an attempt was conducted in the
following paper by one of the few American historians who took
interest in this issue:
O'Keefe (K.J.) 2021: Mass incarceration as Revolutionary
policy: the imprisonment of Hudson Valley Loyalists.
Early American Studies pp.495-527.
Here we try another method in order to see if it
leads us to an estimate that is consistent with
the result obtained by Prof. O'Keefe.
During and after the war, many individuals who had been
denounced as "having assisted the enemies" became
refugees who settled abroad. Starting in May 1778,
long lists of suspected
Loyalists had been published. In Pennsylvania, for instance,
they comprised nearly 500 names. More details can be found in
the "Forging consensus" manuscript mentioned above.
Among these refugees those who felt that they they had really been
faithful Loyalists (and could prove it) presented demands to the
British Claim Commission. A compilation of many (but not all) of these
claims was published in 1980 in the following book:
G
South Carolina Loyalists
Needless to say, even before independence,
the Loyalists in northern states were submitted to considerable
pressure. Many cases are reported of Loyalists being pushed
to departure by mob violence.
Then, in the face of a threatening British invasion
but before the city was duly occupied,
it made sense to remove from New York
all those suspected of being "disaffected".
In the following months
all farmers whose property was considered too easily accessible by sea
were relocated farther away from the seaside.
The same scenario was repeated when British forces threatened
and eventually invaded Philadelphia, then the largest American
city.
What then made the invasion of South Carolina so different?
Many answers could be proposed but one obvious difference is
that in the north the British troops remained confined in
the cities and did not try to support the Loyalists in the few counties
surrounding New York (or Philadelphia) where they were a majority.
In SC, on the contrary, the British and Loyalist troops tried to
hold the whole state. In order to ensure local visibility they
needed to create militia companies formed of Loyalists.
Thanks to the presence of their troops not far away the British
could enlist even reluctant Loyalists. In the north,
recruiters were also sent from New York to neighboring counties,
but in SC this was done on a much larger scale and that is how
the confrontation really became a civil war.
For instance, we
are told that at the battle of Kings Mountain
(7 Oct 1780) there were only few British troops, most of the fighting
force consisted in Loyalists. The same observation holds for the
battle of Kettle Creek (14 Feb 1779).
Nothing similar can be seen in the north for the obvious
reason that it took time to organize, arm and train companies
of Loyalists.
In short, from a military perspective, in northern states
the Loyalists were never a real threat.
Between 1775 and 1779 there were already SC Loyalists but they were
weak and exposed to Patriot repression. With the perspective of
the arrival of British forces (including Loyalist regiments)
by sea as well as by land, the perspective changed completely.
A parallel with the French Revolution may be useful.
In 1792 the first violent protests in the region of Vendee
were triggered by the introduction of
conscription which itself was in
response to Prussian and Austrian armies threatening the
north east of France.
At that point the situation
was still manageable. Protests against conscription are
not something unusual. After all, the protesters had no cannons
and probably only few (fairly outdated) guns.
The situation changed
completely when, thanks to English support, the protesters
got weapons and were able to form military units led by experienced
officers from the nobility.
In addition, the British fleet was never far away allowing
English occupation of a number of offshore places (e.g. Noirmoutier,
Belle Ile) and threatening to land
an expeditionary force of rebels, which indeed happened in June 1795.
(see below).
H
Hunting down Loyalists
Hunting down Loyalists in South Carolina:
the logic which led to the treason trials
According to present knowledge, there have been only relatively
few treason trials in the northern states. A number of them
were listed and described earlier. We made a distinction between
civilian courts and court martials. Following British rules,
treason trials were tried at courts of Oyer and Terminer (O&T).
There were courts of O&T in Delaware, Georgia, New Jersey, New York,
and Pennsylvania. In the other states there were similar courts,
e.g. Supreme Courts of Judicature or Courts of General Session.
It seems that almost all the archives of the courts of O&T have
disappeared. An archivist of the New York state archives told us:
"The only documents produced by Courts of O&T were minute books and
none survived for the period of the Revolution (1775-1782)".
We got similar answers in several other states.
I
Parallel between SC and the Vendee
Drawing a parallel between the war in the Vendee (1792-1793 and 1795)
and the war of independence in South Carolina is not absurd for
the following reasons.
(1) Militarily, the big event was the landing of June
1795 in Quiberon
which involved some 26,000 troops on each side; this was a
close parallel of the invasion of Charleston
Moreover, this
action was preceded by the Vendee
guerrilla warfare which was similar, albeit less
violent, to the repression against the SC Loyalists in the years
1775-1779.
(2) In SC the landmark event was the siege and taking of Charleston
by the British
in May 1780. It involved some 6,000 troops on each side.
The British invasion of Quiberon in Brittany consisted in two
successive landings separated by an interval of one week.
A description of the landing force gives an idea of its strength.
Consisting in 60 transport ships protected by 9 warships, it
brought military equipment for about twice as many soldiers than those
on board. The intent was to deliver this equipment to
the insurgents who would join the rebellion after its first victories.
(3) In SC to change allegiance was not uncommon, both among the
population and among the troops. This raised the question
of whether someone should be seen as a Patriot, a Loyalist,
a traitor or a POW.
It is known that hundreds of defeated troops were tried by
makeshift Patriot courts and that a number of them were
sentenced and hanged.
The defeat of the expeditionary force in Quiberon raised the
same question.
Apart from the Royalist emigres, French soldiers held as prisoners
of war in England had been incorporated into the invading force.
After its defeat they could have been treated as traitors
but in fact most were pardoned.
Many of the Royalist emigres were taken back to
England by the fleet but for those who became prisoners, it was
another matter.
Some 300 of them were tried as traitors, sentenced and executed
by firing squads. At that time such a move was
not unusual as attested by the similarly bloody
repressions which followed the Monmouth rebellion in 1685
(after the defeat at Sedgemoor)
or the Jacobite rebellion in 1746 (after the defeat at Culloden).
J
British goal in the Brittany landing
Contrary to superficial accounts, the primary purpose of the
landing of a British expedionary force in the south of Brittany
was not just to help the Royalists and Chouans
(i.e. disaffected).
The real goal was to trigger a mass
insurrection which would take the city of Rennes
and occupy the north-western part of France. That, it was hoped,
would bring down the Republican government and permit the return
to power of Louis XVIII. This thesis is confirmed by the fact
that the 60 transport ships brought equipment for many more
soldiers than those taking part in the landing, for instance 60,000
shoes, 35,000 muskets, 17,000 infantry uniforms and
5000 cavalry uniforms (Champagnac 1989)
Another indication in the same direction is the participation of
British troops. Three infantry regiments (90th, 19th, 27th)
representing some 4,000 troops were on board with the first wave,
although it does not seem that they took part in the action.
Had the action in south Brittany been successful, the plan was
to make a second landing with 10,000 British troops
in north Brittany in the intention
of taking the port of Saint Malo.
K
Civil war features of the Revolution in South
Carolina
It is often said that in South Carolina the conflict between
Patriots and Loyalists took the form of a civil war.
However in historical accounts one finds only few facts which support
this assertion. One key feature of SC was the the existence of
a fairly strong minority of Loyalists which, quite naturally,
was amplified by the British occupation of Charleston.
A comparison with Massachusetts may be useful because
in contrast this state had one of the weakest Loyalist population.
It is said (see Maas 1989) that only 2% of the adult population
refused to sign the Association Act of 1776. Nevertheless in the
same book by David Maas there is a chapter entitled: "Legislative
efforts to purge the Tories in Boston". This was of course the
objective pursued by the Patriots in all states, but a policy
which may work against a tiny opposition may not be suitable
in states where the Loyalists formed a substantial opposition.
In Massachusetts, one half of the Loyalists departed when the
British forces evacuated Boston on 17 March 1776.
Yet, as soon as 26 March a list of 85
so-called "inimical" persons was set up and sent to the General
Court where 7 justices of the peace would be in charge of
interrogating the suspected Loyalists. Their decision would
imply release, confined on bond or held in jail until a possible
trial that might take place several months later. The threat of
being held in prison was real as shown by the fact that of
16 persons brought to trial by the militia
in April 1776, 5 (Charles and Miles Whitworth,
William Perry, Benjamin Davis, Thomas Edward) were placed in
close confinement in Boston jail (Maas p.181). As an example of
the verdicts deivered at the trials one can mention the case
of William Gardiner who was sentenced to banishment (Maas p.256),
a sentence that the legislature changed into one year in jail.
In SC between 1775 and 1778 the Loyalists had suffered under
Patriot repression but as soon as British forces
(which included regiments of Loyalists) drifted toward
the Carolinas the Loyalists started to form local militias.
This had two consequences.
(i) It was no longer possible
to bring suspected Loyalists to prison or to trial.
(ii) Once armed, Loyalists could oppose force to force..
From a mere police operation, the conflict became a real civil war
with the result that the Patriots themselves felt threatened.
Fear generated harsher actions.
When taken prisoner armed Loyalists were often summarily executed or
tried by "special" courts often composed of officers (as
in the trials following the battle of Kings Mountain) but which
were not regular court martials.
This trasformation raised also a major difficulty for historians
that we explain in the following section.
How to narrate civil wars?
The narration of civil wars raises the same difficulty as already
mentioned for mob violence. In a large country like the United
States there are at any time incidents in which police
officers or federal agents are killed. However, nobody would
call that a civil war situation. On the other hand the American
Civil War did not immediately start with large battles; at
first there were skirmishes. In North and South Carolina
the War of Independence consisted in fairly small battles.
The fall of Charleston which was certainly the largest involved only
some 3,000 Patriots; in addition the attackers were British troops
rather than Loyalists. The largest involment of Loyalists was
probably at the battle of
Kings Mountain. Although it was assuredly an important battle,
there were only a few hundred soldiers on each side.
Is that enough for calling it a civil war?
Below we examine how well known historians have narrated
the civil war aspects of the war in the Southern states.
L
Epuration of the Loyalists
This process of radicalisation was marked by the steps described
below
Special court created to try Loyalists
On 20 Feb 1779,
reacting to the threatening British
invasion, the SC General Assembly passed an Act which gave 40 days
to the citizens who had joined the enemy for surrendering unless
being liable to the death penalty. Manned by
Thomas Heyward and John Mathews, a special court was set up to
try all persons charged with sedition. (Lambert p.81)
Execution of spies in Charleston
It seems that the first application of the new law was made
during the siege of
Charleston. Thanks to the fact that Thomas Heyward was one of the
signers of the Declaration of Independence its biography can be
found on Internet and there we read that he presided at the trial
and condemnation of some Loyalists charged with treasonable
correspondance with the enemy. They were executed in sight of the
enemy lines. The biographty does not tell us how many were
hanged nor does it give the date(s) of the hangings.
Somewhat independently one learns that
on 17 March 1779 also in Charlestown William Tweed
and Andrew Groundwater were hanged. Was their trial
part of those presided by judge Heyward?
Trial and executions at Ninety Six
Robert Lambert (p.83) tells us that in March 1780
no less than 150 prisoners were waiting for their trial
by the newly created special court.
A number of them were Loyalists taken prisoners at the battle
of Kettle Creek in Georgia and who had been marched in chains to Augusta,
a distance of about 200km. The information available about the
trial is quite sketchy. There was a clerk (we even know his name)
but the records did not survive.
The trial lasted
21 days from 22 March 1779 to 12 April which means that on average
20 persons were tried daily (including on Sundays). Eventually,
all except 5 were released or reprieved in the sense that if caught
a second time on the British side they would be executed
immediately.
It turns out that holding trials at Ninety Six was not uncommon.
It was a village of only few houses but which had a jail ans a
courthouse. According to Robert Davis (p.175), in 1778 there
had been trials
of Loyalists and "at least one execution, that of a man named
Allen". One suspects that such a cryptic sentence does not
rely on an appropriate primary source for otherwise
more information would be available (date, complete name of the;
person named Allen.
Other trials of Loyalists were held in Salisbury, North Carolina
on 15 Sept 1779 (Davis p.180). We are told that several [how many?]
Loyalists were sentenced to death and that all were granted reprieve
except two, namely Captain Samuel Richardson and Lieutenant William
Armstrong who were hanged on (or around) 5 November 1779.
Summary trial and executions after the Patriot victory at Kings
Mountain
Following the American victory at Kings Mountain against a militia
of Loyalists, there was a summary trial in which 30 were sentenced
to be hanged but only 9 were actually executed.
Death march after the Patriot victory at Kings
Mountain
Excerpt of the Royal Gazette of 24 February 1781
The executions are reported
in all accounts. In contrast the march which followed did not
attract the attention of historians to the same degree.
As a matter of comparison one can consider the "Bataan Death March"
in April 1942 for this is a case for which there are some data.
Some 60,000 Filipino and 15,000 American prisoners of war
had to march from Bataan to their prisoner camp some 100km
away. The number of American deaths is fairly well known and is
around 500 which represents a rate of 3.3%.
The Filipino deaths are more uncertain (between 5,000 and 18,000);
if we take 10,000 as a rough estimate, the death rate is 17%.
It should be added that the march took place after a siege of
3 months. For the 700 POWs of Kings Mountain a death rate
of 3% would have led to 20 deaths. Naturally the conditions
differed greatly but at least this number gives an order
of magnitude in the sense that one can be sure that the
real number is not 10 times smaller or 10 times larger.
According to Allaire, the rebel officers would often came among
the prisoners, draw their swords and wound "those whom their
wicked mind prompted".
Lt Allaire reports that on 17 oct 1780 three prisoners attempted
to make their escape. Two succeeded but the third "was shot through
the body".
In a general way suspected Loyalists were often led on long
marches. For instance in November-December 1776 some 230 Loyalists
of New York state had to march 300km to Exeter in New Hampshire.
Other NY Loyalists were sent to Massachusetts, Connecticut and
Pennsylvania. As 35km/day can be considered as
a reasonable assumption of average velocity a march of 300 km
would last some 9 days. In the mid of winter, probably with
little food and rudimentary shelters during the night,
one would like to know the death rate for weak and
elderly persons.
In all American
accounts of the long marches faced by British prisoners
of war
there is a marked tendency to attribute melting numbers to escapes
rather than to deaths. The fate of the Convention army
(consisting in prisonrers made at Saratoga) is another
illustration.
The so-called "Trials of Tears" of native American Indians
come to mind. In the 1830s they were led from various locations
to the state of Oklahoma, i.e. west of the Mississippi. Often
such marches took also place in winter time. One reads
that this was the request of the Indians themselves,
supposedly in order to avoid the risk of diseases. Whether it
was a wise decision is to be seen.
Later on, in the 1880s and 1890,
the western Indians were marched to their reservations often
after being kept for several months in stockade camps in bleak
conditions. It was not uncommon that they were transferred
from one temporary reservation to another before being eventually
settled in a more permanent place.
M
The spiral of retaliations
In December 1780 General Cornwallis declared: "Rebel parties
have so terrified my people that I can get nobody to venture
far enough out to ascertain anything".
The reason given by Robert Lambert (p.200) is so uncertain that it makes
the reader uneasy. He says: After Cornwallis left South Carolina
many of the British posts in which Loyalists served were taken
by the Patriots. After the Loyalists were confined,
it happened that some of them were set aside and hanged
or shot individually or in small groups. This is of course a very
serious assertion. If Lambert had some supporting evidence he
should give it to his readers. If he had none it would be better
to omit such a statement.
It is true that in the subsequent pages he cites some
cases, but in small number. Here are two illustrations.
When Fort Motte surrendered Tory Lieutenant George Fulker and
John Jackson were hanged.
After Patriot General Thomas Sumter captured Orangeburg 14 prisoners
who were being escorted to the camp of General Greene were
executed. The shooting became known because one of the victims,
militiaman Joseph Cooper, who was presumed dead in fact had survived
(Lambert p.201)
Contrary to the previous one which concerned only individuals,
this episode is a mass shooting of the kind described above.
However the event is described in two lines only and
without indication of date. It is reported that the shooting
occurred in the
Fisher regiment but this was not of great interest because
Lt-Col John Fisher was in charge of all the troops in Orangeburg.
N
List of names of 300 Loyalist victims in SC
A petition dated 19 April 1782 was sent by American Loyalists
to Lord Germain who was the minister of George III in charge
of the American war. The petition was signed by 11 officers from
the Ninety Six and Camden districts.
In an appendix the petition contained 300 names of Loyalists
said to have been "massacred" or "murdered" (depending on the version)
in the district of Ninety Six.
The terms "massacred" or "murdered" mean that they were not
killed in regular battles. The list is available on Internet
but one should pay attention to the fact that there are several
versions. Although the names are the same, the comments which
introduce and follow the list may be somewhat different.
O
Names of Loyalists in the NY state counties of
Albany and Queens
P
Evolution of desertion
The fall in the number of desertions is impressive but before we
try to find how it should be interpreted we must answer the two
following questions:
(i) Do our data concern the militia or the Continental Army?
(ii) How were these estimates obtained?
Answering the first question is not easy because the
articles mentioning the desertions indicate only rarely
the origin of the deserter. The desertion estimates
were obtained thanks to the fact that
when a private deserted, one of his officers would ask a newspaper
to publish a search notice offering a reward to any person
giving information that may facilitate his arrest.
In Boyle (2009) the author has collected all such notices. Together
with their date of publication these articles
give a chronological list of the desertions. This list can then be
explored in two ways (i) By counting the number of pages
for a given time interval. This leads to the curve in blue.
The first point corresponds to a moment when the Patriot army
was only coming into existence. After this initial point the desertions
jumped to a high level after which they declined steadily.
Many reasons can be invoked, some may seem more plausible
than others but plausibility is a fairly subjective notion.
It would be useful to know whether the "bounty jumpers"
(i.e. those who enlisted, got their bounty, deserted, enlisted
again in another regiment, got a second bounty, and so on)
represented a substantial proportion of the deserters.
If so a better identification of those who enlisted could
prevent bounty jumping.
Q
Widows whose husbands were executed
The claims presented by emigrated Loyalists to the Claims
Commission set up by the British government
for the purpose of awarding compensations to Loyalists
are an interesting
source because it is completely independent from
American sources. Most of the claims describe losses incurred
through the expropriation of Loyalists. However, some 10%
of the claims were presented by widows or descendants and refer
in a broad way
to deaths resulting from being a Loyalists. Among such causes of
death one can distinguish:
In the following file some twenty cases are described.
Widows whose husbands were executed
The previous list has two important limitations.
We have used two sources: firstly a partial compilation published
in book format by P.W. Coldham (1980); secondly an Internet
source which reports mostly Canadian cases.
R
Executions of civilians in New York
In New York
between 1777 and 1783 there were some 21 death sentences issued by
courts martial against civilians. The source is Wiener (1967).
Below is a list of their names.
ABRAHAMS (Moses),
S
Anti-Loyalist mob actions in New York
(June 1776)
References
Allaire (A.) 1781: Diary of Lt. Anthony Allaire in the form of a
letter dated Charlestown, 30 January 1781
and published in the "Royal Gazette" of 24 February 1781.
Allaire (A.) 1881: Diary of Lt. Anthony Allaire. published
in "King' Mountain and it's heroes" edited and
published by Lyman Copeland Draper.
During the march the men were obliged to give 35
Continental dollars for a single ear of Indian corn, and 40 for
a drink of water, they were not allowed to drink when fording a
river.
Several of the [Loyalist] militia
that were worn out with fatigue, and not
being able to keep up, were cut down, and
trodden to death in the mire.
The Rebel officers would often go in amongst the prisoners,
draw their swords, cut down and wound those whom their
wicked and savage minds prompted.
Without these sentences it would appear to be an ordinary march.
When these sentences are added it seems rather a terror march.
We are told that many prisoners tried to escape; while some were
shot, a substantial number succeeded.
Allaire (A.) 1968: Diary of Lt. Anthony Allaire of Ferguson Corps.
Arno Press.
Coldham (P.W.) 1980: Almerican Loyalist claims. Published by
the "National Genealogical Society", Washington D.C.
Davis (R.S.) 1979: The Loyalist trials at Ninety Six in 1779.
South Carolina Historical Magazine p.172-181.
Jones (T.) 1879: History of New York during the Revolutionary War
and of the leading events in the other colonies at that period.
Edited and notes by Edward Floyd de Lancey.
Vol.1: 872p. Vol.2: 826p.
Printed for the New York Historical Society.
Greer (D.) 1934,2014: Incidence of the Terror during the French
Revolution. A statistical information.
Harvard University Press.
Maas (D.E.) 1989: The return of the Massachusetts Loyalists.
Garland Publishing, New York.
[Despite its title only two chapters in 11 are devoted to the return
of some of the Loyalists]
Moore (F.) 1860: Diary of the American Revolution. From newspapers
and original documents. New York.
[The author was a 19th century journalist and compiler. The reference
that he gives for the hanging of John Roberts is "Clift's Diary",
but we could not find it.]
O'Keefe (K.J.) 2011: Mass incarceration as Revolutionary policy.
The imprisonment of the Hudson Valley Loyalists.
Early American Studies 19,3,495-527.
Sanderson (J.) 1823: Biographies of the signers of the Declaration
of Independence. 4 vols. The biography of James Hamilton Jr Heyward
is in vol.4.
Wiener (F.B.) 1967: Civilians under military justice. Especially
in North America during the Revolution.
University of Chicago Press.
Wikipedia: article in English entitled: "Invasion of France (1795)"
B
Death sentences in state trials of Loyalists
C
Draft of "Forging consensus" with a list of executions
D
Death sentences by court martials
E
Control of suspected Loyalists in Albany county
F
Number of imprisoned Loyalists
(to be completed)
G
South Carolina Loyalists
H
Hunting down Loyalists in special courts and O&T courts
I
Parallel between SC and the French Vendee
J
British goal in Brittany landing
K
Civil war features of the Revolution in SC
L
Epuration of the Loyalists
M
The spiral of retaliations
N
List of names of 300 Loyalist victims in SC
O
Names of about 4,000 Loyalists in NY state
P
Evolution of desertion (1776-1782)
References
(i) Firstly, there was a great difference in population:
some 20 millions in France against some 2 millions (without the
slaves) in America.
(ii) Secondly, whereas in France religion was a major factor
of division within the French population, in America almost
all Anglican ministers were Loyalists.
In other words, religion
was not a distinct additional factor of division.
With respect to historiography the situation can be summarized
as follows.
In terms of increasing accuracy there are three levels of
description.
(a) Minimal anecdotal description.
(b) Exhaustive anecdotal description.
(c) Global description based on reliable global data.
List of 50 prisons
Jailing suspicious characters was the easiest and most
common way to get rid of the threat they may have represented.
Depending upon the outrage, it could be for a term
of one, six, nine, 12 months or more. In practice the term
was conditioned by the ability of the person to
provide a security, that is to say a guaranty in the form of
a sum of money. Often it was a joint guaranty in the sense
that a part of the money was provided by a friend.
Elusive Loyalists (p.40-42)
Mass-trials, sentences and pardons
Forging consensus.
The American Revolution in Loyalist perspective
contains a very provisional
version of the manuscript of a book in which the American
Revolution is considered from a Loyalist perspective.
As a third case we consider also the fate
of the persons (often German immigrates) who did not
wish the United States to take part in the First World War.
The fight against communism provides still another example of
how to eliminate dissenters. This part has yet to be written.
The Quakers in the Revolution
Former uprisings in several of the thirteen colonies
Tar-and-feather episodes
British judges and officials left at the mercy of the mobs
Imprisonments
Death sentences and executions
Identification of pardons
The numbers per year are as follows (provisional):
1776:1 1777:51 1778:50 1779:58 1780:56 1781:38 1782:20 1783:2
Total for 1776-1783: 276 executions
Acts of attainder and their consequences
Putting New York afire
The draft uprising in New York
During the war confiscation of German assets in America
Appendix: Civilian Patriots against Loyalists (based on Peter Force)
It is well-nigh inexplicable that so little remains
of the vast amount of
minutes and related records of this State body, existent in 7
counties and operative during about 5 years of the war. The several
boards were obligated by statute to keep accurate minutes of all of
their proceedings, subject to review by higher authority.
(FAMILY-NAME Given-name) page (month-year) Event description
These standardized events can be found in the following files:
First Commission, vol.1
Sometimes, after being arrested for no other reason than
the complaint of a neighbor, they were released after
one or two days; of course after having taken the oath
and having given security.
Usually, those who did not wish to take the oath
were imprisoned until they were willing to take it.
These individuals can be labelled as being real
Loyalists.
Coldham (P.W.) 1980: American Loyalist claims. National
Genealogical Society, Washington D.C. (616p.)
The main part of these claims consists in an enumeration of
forfeited property and confiscated estates. However, the accounts
also shortly mention possible stays in prison as proofs
of Loyalist status. By counting the number of imprisonment mentions
it becomes possible to estimate the number of imprisoned
Loyalists.
It can be called a death march for
according to the testimony of Anthony Allaire, a British Lieutenant,
those who
were unable "to keep up were trodden to death in the mire".
How many died is not revealed. One should remember that there were
some 700 prisoners and that 160 of them had been injured in
the fighting.
The number of deaths during the march may have surpassed
the number of those who were executed but as no data were
recorded or kept we will never know.
So far, we have never seen any official death
rate data for such episodes. Naturally, on weakened organisms, both
disease and exertion had deadly effects.
What led to this situation?
This method is very easy but it relies on the
assumption that the number of desertions by page remains
fairly uniform. As the index gives the page numbers for all
individual deserters one can also count the number of deserters in each
time interval. This leads to the curve in red. The fact that the
two curves are fairly parallel shows that the first method was
also fairly satisfactory.
(i) deaths due to difficult imprisonment conditions.
(ii) killings by Patriot mobs
(iii) legal executions.
(i) The claims give little information about the specific
circumstances of the death. For instance, very often the
year of the death is not given.
(ii) At the time of writing (16 June 2024) it seems that
no
exhaustive compilation of the claims has been published.
Whereas for property losses this limitation is not crucial
for our present project it is a major defect.
"inh" means inhabitant whereas "ref" means refugee.
ALGER (Isaac) ref,
CHANDLER (Solomon),
CABROL (David) inh,
COLLINS (Richard),
DONNALY (Cornelius),
FARREN (John),
FERNE (John) inh,
FOGWELL (George),
GARRETSON (John),
GELLIN (John) inh,
GUFFIE (James),
HARDING (Jesse),
MASON (John),
McINTIRE (John),
McNEAL (Mary) inh,
NORMAN (Isaac),
PARKER (Nathaniel) ref,
PURDY (John),
RANDEN (Thomas),
ROACH (James) inh.
[Lt. Allaire was a British officer.
His testimony covers
the period from 1 September to 30 November 1780 which includes
the battle of Kings Mountain (7 Oct), a surprise attack
which was a Patriot victory and led to some 700
Loyalist prisoners of war.
In the weeks which followed the battle the prisoners
were marched to a stockade.
There is another version of this testimony which was published
one century later; see the next reference.
[Lyman Draper was an American historian who collected
documents or testimonies by individuals who played
a role in historic events. Sometimes, when the persons
were no longer living, he recorded the testimonies
of their sons or daughters.]
This is a rephrased version of the one of 1781.
We do not know how it was set up but it can be observed that
a number of crucial sentences have been omitted among which
are the following:
On the morning of the 15th October, Col. Campbell [Patriot leader]
had intelligence
that Col. Tarleton was approaching. He gave orders to his
men, that should he come up with them, they were
immediately to fire on Capt. Depeyster [Loyalist leader]
and his officers, who were
in the front, and then a second volley on the men.
[This publication reproduced the sanitized text of 1881.]
Boyle (J.L.) 2009: "He loves a good deal of rum ...".
Military desertion during the American Revolution.
Vol.1: 1775-1776, Vol.2: 1777-1783. Genealogical Publishing Company.
Baltimore.
[The term "Loyalist" does not appear in the title, but
Judge Thomas Jones was a prominent Loyalist. He wrote this
history between 1783 and 1787 while being a refugee in England.
It was only published about a century later.
Apart from general historical events, the book describes also
events of particular concern for the Jones and de Lancey families.]
Lambert (R.S) 1987: South Carolina Loyalists in the American
Revolution. University of South Carolina Press. Columbia (SC).
[A very lucid and well documentated account but the reader
has often the impression that the author knows more
than he wishes to say. Therefore some episodes which would
be of the highest interest remain shrouded in uncertainty.]