![Ridiculous as this negro’s idea of honour must appear to us, it bears
a sufficient resemblance to other notions of the same kind that have
passed current in the world at different times to satisfy us of the
extreme variability of the sentiment in question. Cæsar built with
difficulty a bridge across the Rhine, chiefly because he held it
beneath his own dignity, or the Roman people’s, for his army to
cross it in boats. The Celts of old thought it as ignominious to fly
from an inundation, or from a burning or falling house, as to retreat
from an enemy. The Spartans considered it inglorious to pursue a
flying foe, or to be killed in storming a besieged city. The same Gauls
who gloried in broadsword-wounds would almost go mad with
shame if wounded by an arrow or other missile that only left an
imperceptible mark. The use of letters was once thought
dishonourable by all the European nations. Marshal Montluc, in the
sixteenth century, considered it a sign of abnormal overbookishness
for a man to prefer to spend a night in his study than to spend it in
the trenches, though, now, a contrary taste would be thought by
most men the mark of a fool.
Such are some of the curious ideas of honour that have prevailed at
different times. Wherein we seem to recognise not merely change
but advance; one chief difference between the savage and civilised
state lying in the different estimates entertained in either of martial
prowess and of military honour. We laugh nowadays at the ancient
Britons who believed that the souls of all who had followed any
other pursuit than that of arms, after a despised life and an
unlamented death, hovered perforce over fens and marshes, unfit to
mingle with those of warriors in the higher and brighter regions; or
at the horsemen who used before death to wound themselves with
their spears, in order to obtain that admission to Walhalla which was
denied to all who failed to die upon a battle-field; or at the
Spaniards, who, when Cato disarmed them, preferred a voluntary
death to a life destined to be spent without arms.[204] No civilised
warrior would pride himself, as Fijian warriors did, on being generally
known as the ‘Waster’ or ‘Devastator’ of such-and-such a district; the
most he would look for would be a title and perhaps a perpetual](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-26-320.jpg)
![pension for his descendants. We have nothing like the custom of the
North American tribes, among whom different marks on a warrior’s
robe told at a glance whether his fame rested on the slaughter of a
man or a woman, or only on that of a boy or a girl. We are inferior in
this respect to the Dacota tribes, among whom an eagle’s feather
with a red spot on it denoted simply the slaughter of an enemy, the
same feather with a notch and the sides painted red, that the said
enemy had had his throat cut, whilst according as the notches were
on one side or on both, or the feather partly denuded, anyone could
tell after how many others the hero had succeeded in touching the
dead body of a fallen foe. The stride is clearly a great one from
Pyrrhus, the Epirot king, who, when asked which of two musicians
he thought the better, only deigned to reply that Polysperchon was
the general, to Napoleon, the French emperor, who conferred the
cross of the Legion of Honour on Crescentini the singer.
And as the pursuit of arms comes with advancing civilisation to
occupy a lower level as compared with the arts of peace, so the
belief is the mark of a more polished people that the rapacity and
cruelty which belong to the war customs of a more backward nation,
or of an earlier time, are absent from their own. They invent the
expression civilised warfare to emphasise a distinction they would
fain think inherent in the nature of things; and look, by its help, even
on the mode of killing an enemy, with a moral vision that is absurdly
distorted. How few of us, for example, but see the utmost barbarity
in sticking a man with an assegai, yet none whatever in doing so
with a bayonet? And why should we pride ourselves on not
mutilating the dead, while we have no scruples as to the extent to
which we mutilate the living? We are shocked at the mention of
barbarian tribes who poison their arrows, or barb their darts, yet
ourselves think nothing of the frightful gangrenes caused by the
copper cap in the Minié rifle-ball, and reject, on the score of the
expense of the change, the proposal that bullets of soft lead, which
cause needless pain, should no longer be used among the civilised
Powers for small-arm ammunition.[205]](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-27-320.jpg)
![But whilst the difference in these respects between barbarism and
civilisation is thus one that rather touches the surface than the
substance of war, the result is inevitably in either state a different
code of military etiquette and sentiment, though the difference is far
less than in any other points of comparison between them. When
the nations of Christendom therefore came in contact with unknown
and savage races, whose customs seemed different from their own
and little worthy of attention, they assumed that the latter
recognised no laws of war, much as some of the earlier travellers
denied the possession or faculty of speech to people whose
language they could not interpret. From which assumption the
practical inference followed, that the restraints which were held
sacred between enemies who inherited the same traditions of
military honour had no need to be observed in hostilities with the
heathen world. It is worth while, therefore, to show how baseless
was the primary assumption, and how laws of war, in no way
dissimilar to those of Europe, may be detected in the military usages
of barbarism.
To spare the weak and helpless was and is a common rule in the
warfare of the less civilised races. The Guanches of the Canary
Islands, says an old Spanish writer, ‘held it as base and mean to
molest or injure the women and children of the enemy, considering
them as weak and helpless, therefore improper objects of their
resentment; neither did they throw down or damage houses of
worship.’[206] The Samoans considered it cowardly to kill a woman:
[207] and in America the Sioux Indians and Winnebagoes, though
barbarous enough in other respects, are said to have shown the
conventional respect to the weaker sex.[208] The Basutos of South
Africa, whatever may be their customs now, are declared by Casalis,
one of the first French Protestant missionaries to their country, to
have respected in their wars the persons of women, children, and
travellers, and to have spared all prisoners who surrendered,
granting them their liberty on the payment of ransom.[209]](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-28-320.jpg)
![being equivalent to saying, that a district full of the habitations of
men should shortly be turned into a pasture for sheep.[210]
The Fijians used to spare their enemy’s fruit trees; the Tongan
islanders held it as sacrilege to fight within the precincts of the burial
place of a chief, where the greatest enemies were obliged to meet
as friends.
Most of the lower races recognise the inviolability of ambassadors
and heralds, and have well-established emblems of a truce or
armistice. The wish for peace which the Zulu king in vain sought
from his English invaders by the symbol of an elephant’s tusk
(1879), was conveyed in the Fiji Islands by a whale’s tooth, in the
Sandwich by a young plantain tree or green branch of the ti plant,
and among most North American tribes by a white flag of skin or
bark. The Samoan symbol for an act of submission in deprecation of
further hostilities conveys some indication of the possible origin of
these pacific symbols. The conquered Samoan would carry to his
victor some bamboo sticks, some firewood, and some small stones;
for as a piece of split bamboo was the original Samoan knife, and
small stones and firewood were used for the purpose of roasting
pigs, this symbol of submission was equivalent to saying: ‘Here we
are, your pigs, to be cooked if you please, and here are the
materials wherewith to do it.’[211] In the same way the elephant’s
tusk or the whale’s tooth may be a short way of saying to the victor:
‘Yours is the strength of the elephant or the whale; we recognise the
uselessness of fighting with you.’
In the same way many savage tribes take the greatest pains to
impress the terms of treaties as vividly as possible on the memory of
the contracting parties by striking and intelligible ceremonies. In the
Sandwich Islands a wreath woven conjointly by the leaders of either
side and placed in a temple was the chief symbol of peace. On the
Fiji Islands, the combatant forces would meet and throw down their
weapons at one another’s feet. The Tahitians wove a wreath of
green boughs, furnished by each side; exchanged two young dogs;](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-30-320.jpg)
![and having also made a band of cloth together, deposited the wreath
and the band in the temple, with imprecations on the side which
should first violate so solemn a treaty of peace.[212] On the Hervey
Islands, the token of the cessation of war was the breaking of a
number of spears against a large chestnut tree; the almost
imperishable coral tree was planted in the valleys to signify the hope
that the peace might last as long as the tree; and after the drum of
peace had been solemnly beaten round the island, it was unlawful
for any man to carry a weapon, or to cut down any iron-wood, which
he might turn into an implement of destruction.
Even our custom of proclaiming that a war is not undertaken against
a people but against its rulers is not unknown in savage life. The
Ashantee army used to strew leaves on their march, to signify that
their hostility was not with the country they passed through but only
with the instigators of the war; they told the Fantees that they had
no war with them collectively, but only with some of them.[213] How
common a military custom this appeal to the treason of an enemy is,
notwithstanding the rarity of its success, everybody knows. When,
for instance, the Anglo-Zulu war began, it was solemnly proclaimed
that the British Government had no quarrel with the Zulu people; it
was a war against the Zulu king, not against the Zulu nation. (Jan.
11, 1879.) So were the Ashantees told by the English invading force;
so were the Afghans; so were the Egyptians; and so were the
French by the Emperor William before his merciless hordes laid
waste and desolate some of the fairest provinces of France; so, no
doubt, will be told the Soudan Arabs. And yet this appeal to treason,
this premium on a people’s disloyalty, is the regular precursor of
wars, wherein destruction for its own sake, the burning of grain and
villages for the mere pleasure of the flames, forms almost invariably
the most prominent feature. The military view always prevails over
the civil, of the meaning of hostilities that have no reference to a
population but only to its government. In the Zulu war, for instance,
in spite of the above proclamation, the lieutenant-general ordered
raids to be made into Zululand for the express purpose of burning](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-31-320.jpg)
![empty kraals or villages; defending such procedure by the usual
military logic, that the more the natives at large felt the strain of the
war, the more anxious they would be to see it concluded; and it was
quite in vain for the Lieutenant-Governor of Natal to argue that the
burning of empty kraals would neither do much harm to the Zulus
nor good to the English; and that whereas the war had been begun
on the ground that it was waged against the Zulu king and not
against his nation, such conduct was calculated to alienate from the
invaders the whole of the Zulu people, including those who were
well disposed to them. Such arguments hardly ever prevail over that
passion for wanton destruction and for often quite unnecessary
slaughter, which finds a ready and comprehensive shelter under the
wing of military exigencies.
The assumption, therefore, that savage races are ignorant of all laws
of war, or incapable of learning them, would seem to be based
rather on our indifference about their customs than on the realities
of the case, seeing that the preceding evidence to the contrary
results from the most cursory inquiry. But whatever value there may
be in our own laws of war, as helping to constitute a real difference
between savage and civilised warfare, the best way to spread the
blessing of a knowledge of them would clearly be for the more
civilised races to adhere to them strictly in all wars waged with their
less advanced neighbours. An English commander, for instance,
should no more set fire to the capital of Ashantee or Zululand for so
paltry a pretext as the display of British power than he would set fire
to Paris or Berlin; he should no more have villages or granaries burnt
in Africa or Afghanistan than he would in Normandy; and he should
no more keep a Zulu envoy or truce-bearer in chains[214] than he
would so deal with the bearer of a white flag from a Russian or
Italian enemy.
The reverse principle, which is yet in vogue, that with barbarians you
must or may be barbarous, leads to some curious illustrations of
civilised warfare when it comes in conflict with the less civilised
races. In one of the Franco-Italian wars of the sixteenth century,](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-32-320.jpg)
![more than 2,000 women and children took refuge in a large
mountain cavern, and were there suffocated by a party of French
soldiers, who set fire to a quantity of wood, straw, and hay, which
they stacked at the mouth of the cave; but it was considered so
shameful an act, that the Chevalier Bayard had two of the
ringleaders hung at the cavern’s mouth.[215] Yet when the French
General Pélissier in this century suffocated the unresisting Algerians
in their caves, it was even defended as no worse than the shelling of
a fortress; and there is evidence that gun-cotton was not
unfrequently used to blast the entrance to caves in Zululand in which
men, women, and children had hoped to find shelter against an
army which professed only to be warring with their king.[216]
The following description of the way in which, in the Ashantee war,
the English forces obtained native carriers for their transport service
is not without its instruction in this respect:—
‘We took to kidnapping upon a grand scale. Raids were made on all
the Assin villages within reach of the line of march, and the men,
and sometimes the women, carried off and sent up the country
under guard, with cases of provisions. Lieutenant Bolton, of the 1st
West India Regiment, rendered immense service in this way. Having
been for some time commandant of Accra, he knew the coast and
many of the chiefs; and having a man-of-war placed at his disposal,
he went up and down the coast, landing continually, having
interviews with chiefs, and obtaining from them large numbers of
men and women; or when this failed, landing at night with a party of
soldiers, surrounding villages, and sweeping off the adult population,
leaving only a few women to look after the children. In this way, in
the course of a month, he obtained several thousands of
carriers.’[217]
And then a certain school of writers talks of the love and respect for
the British Empire which these exhibitions of our might are
calculated to win from the inferior races! The Ashantees are
disgraced by the practice of human sacrifices, and the Zulus have](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-33-320.jpg)
![barbarise the civilised than to civilise the barbarous population. It is
well worthy of notice and reflection that the European battle-fields
became distinctly more barbarous after habits of greater ferocity had
been acquired in wars beyond the Atlantic, in which the customary
restraints were forgotten, and the ties of a common human nature
dissolved by the differences of religion and race.
The same effect resulted in Roman history, when the extended
dominion of the Republic brought its armies into contact with foes
beyond the sea. The Roman annalists bear witness to the
deterioration that ensued both in their modes of waging war and in
the national character.[218] It is in an Asiatic war that we first hear of
a Roman general poisoning the springs;[219] in a war for the
possession of Crete that the Cretan captives preferred to poison
themselves rather than suffer the cruelties inflicted on them by
Metellus;[220] in the Thracian war that the Romans cut off their
prisoners’ hands, as Cæsar afterwards did those of the Gauls.[221]
And we should remember that a practical English statesman like
Cobden foresaw, as a possible evil result of the closer relations
between England and the East, a similar deterioration in the national
character of his countrymen. ‘With another war or two,’ he wrote, ‘in
India and China, the English people would have an appetite for bull-
fights if not for gladiators.’[222]
Nor is there often any compensation for such results in the improved
condition of the tribes whom it is sought to civilise after the method
recommended by Sepulveda. The happiest fate of the populations he
wished to see civilised by the sword was where they anticipated
their extermination or slavery by a sort of voluntary suicide. In Cuba,
we are told that ‘they put themselves to death, whole families doing
so together, and villages inviting other villages to join them in a
departure from a world that was no longer tolerable.’[223] And so it
was in the other hemisphere; the Ladrone islanders, reduced by the
sword and the diseases of the Spaniards, took measures
intentionally to diminish their numbers and to check population,](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-35-320.jpg)
![preferring voluntary extinction to the foul mercies of the Jesuits: till
now a lepers’ hospital is the only building left on what was once one
of the most populous of their islands.
It must, however, be admitted in justice to the Spaniards, that the
principles which governed their dealings with heathen races infected
more or less the conduct of colonists of all nationalities. A real or
more often a pretended zeal for the welfare of native tribes came
among all Christian nations to co-exist with the doctrine, that in case
of conflict with them the common restraints of war might be put in
abeyance. What, for instance, can be worse than this, told of the
early English settlers in America by one of themselves? ‘The
Plymouth men came in the mean time to Weymouth, and there
pretended to feast the savages of those parts, bringing with them
forks and things for the purpose, which they set before the savages.
They ate thereof without any suspicion of any mischief, who were
taken upon a watchword given, and with their own knives hanging
about their necks were by the Plymouth planters stabbed and
slain.’[224]
Among the early English settlers it soon came to be thought, says
Mather, a religious act to kill an Indian. In the latter half of the
seventeenth century both the French and English authorities
adopted the custom of scalping and of offering rewards for the
scalps of their Indian enemies. In 1690 the most healthy and
vigorous Indians taken by the French ‘were sold in Canada, the
weaker were sacrificed and scalped, and for every scalp they had a
premium.’[225] Caleb Lyman, who afterwards became an elder of a
church at Boston, left an account of the way in which he himself and
five Indians surprised a wigwam, and scalped six of the seven
persons inside, so that each might receive the promised reward. On
their petition to the great and general court they received 30l. each,
and Penhallow says not only that they probably expected eight times
as much, but that at the time of writing the province would have
readily paid a sum of 800l. for a similar service.[226] Captain
Lovewell, says the same contemporary eulogist of the war that](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-36-320.jpg)
![lasted from July 1722 to December 1725, ‘from Dunstable with thirty
volunteers went northward, who marching several miles up country
came on a wigwam where were two Indians, one of whom they
killed and the other took, for which they received the promised
bounty of 100l. a scalp, and two shillings and sixpence a day
besides.’ (December 19, 1724.)[227] At the surprise of Norridjwock
‘the number of dead which we scalped were 26, besides Mr. Rasle
the Jesuit, who was a bloody incendiary.’[228] It is evident that these
very liberal rewards must have operated as a frequent cause of
Indian wars, and made the colonists open-eared to tales of native
outrages; indeed the whites sometimes disguised themselves like
Indians, and robbed like Indians, in order, it would appear, the more
effectually to raise the war-cry against them.[229]
Since the Spaniards first trained bloodhounds in Cuba to hunt the
Indians, the alliance between soldiers and dogs has been a favourite
one in barbarian warfare. The Portuguese used them in Brazil when
they hunted the natives for slaves.[230] And an English officer in a
treatise he wrote in the last century as a sort of military guide to
Indian warfare suggested coolly: ‘Every light horseman ought to be
provided with a bloodhound, which would be useful to find out the
enemy’s ambushes and to follow their tracks. They would seize the
naked savages, and at least give time to the horsemen to come up
with them.’[231] In the Molucca Islands the use of two bloodhounds
against a native chief was the cause of a great confederacy between
all the islands to shake off the Spanish and Portuguese yoke.[232]
And even in the war waged by the United States in Florida from
1838 to 1840, General Taylor was authorised to send to Cuba for
bloodhounds to scent out the Indians; nor, according to one account,
was their aid resorted to in vain.[233]
Poison too has been called in aid. Speaking of the Yuta Indians, a
traveller assures us that ‘as in Australia, arsenic and corrosive
sublimate in springs and provisions have diminished their](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-37-320.jpg)
![number.’[234] And in the same way ‘poisoned rum helped to
exterminate the Tasmanians.’[235]
But there is worse yet in this direction. The Portuguese in Brazil,
when the importation of slaves from Africa rendered the capture of
the natives less desirable than their extermination, left the clothes of
persons who had died of small-pox or scarlet fever to be found by
them in the woods.[236] And the caravan traders from the Missouri
to Santa Fé are said by the same method or in presents of tobacco
to have communicated the small-pox to the Indian tribes of that
district in 1831.[237] The enormous depopulation of most tribes by
the small-pox since their acquaintance with the whites is one of the
most remarkable results in the history of their mutual connection;
nor is it likely ever to be known to what extent the coincidence was
accidental.
It is pleasant to turn from these practical illustrations of the theory
that no laws of war need be regarded in hostilities with savage tribes
to the only recorded trial of a contrary system, and to find, not only
that it is associated with one of the greatest names in English
history, but also that the success it met with fully justifies the
suspicion and disfavour with which the commoner usage is
beginning to be regarded. The Indians with whom Penn made his
famous treaty in 1682 (of which Voltaire said that it was the only
treaty that was never ratified by an oath, and the only treaty that
was never broken), were of the same Algonquin race with whom the
Dutch had scarcely ever kept at peace, and against whom they had
warred in the customary ruthless fashion of those times. The treaty
was based on the principle of an adjustment of differences by a
tribunal of an equal number of Red men and of White. ‘Penn,’ says
the historian, ‘came without arms; he declared his purpose to
abstain from violence, he had no message but peace, and not one
drop of Quaker blood was ever shed by an Indian’[238] For more
than seventy years, from 1682 to 1754, when the French war broke
out, in short, during the whole time that the Quakers had the](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-38-320.jpg)
![principal share in the government of Pennsylvania, the history of the
Indians and Whites in that province was free from the tale of
murders and hostilities that was so common in other districts; so
that the single instance in which the experiment of equal laws and
forbearance has been patiently persevered in, can at least boast of a
success that in support of the contrary system it were very difficult
to find for an equal number of years in any other part of the world.
It may also be said against Sepulveda’s doctrine, that the habits of a
higher civilisation, where they are really worth spreading, spread
more easily and with more permanent effect among barbarous
neighbours by the mere contagion of a better example than by the
teaching of fire and sword. Some of the Dyak tribes in Borneo are
said to have given up human sacrifices from the better influences of
the Malays on the coast district.[239] The Peruvians, according to
Prescott, spread their civilisation among their ruder neighbours more
by example than by force. ‘Far from provoking hostilities, they
allowed time for the salutary example of their own institutions to
work its effect, trusting that their less civilised neighbours would
submit to their sceptre from a conviction of the blessings it would
secure to them.’ They exhorted them to lay aside their cannibalism,
their human sacrifices, and their other barbarities; they employed
negotiation, conciliatory treatment, and presents to leading men
among the tribes; and only if all these means failed did they resort
to war, but to war which at every stage was readily open to
propositions of peace, and in which any unnecessary outrage on the
persons or property of their enemy was punished with death.
Something will have been done for the cause of this better method
of civilising the lower races, if we forewarn and forearm ourselves
against the symptoms of hostilities with them by a thorough
understanding of the conditions which render such hostilities
probable. For as an outbreak of fever is to some extent preventable
by a knowledge of the conditions which make for fevers, so may the
outbreak of war be averted by a knowledge of the laws which
govern their appearance. The experience which we owe to history in](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-39-320.jpg)
![affects to tremble from a panic of invasion, and vials of virtue are
expended on the vices of the barbarians which have been for years
tolerated with equanimity or indifference. Their customs are painted
in the blackest colours; the details of savage usages are raked up
from old books of travel; rumours of massacres and injuries are
sedulously propagated; and the whole country is represented as in
such a state of anarchy, that the majority of the population, in their
longing for deliverance from their own rulers, would gladly welcome
even a foreign conqueror. In short, a war against them comes
speedily to be regarded as a war in their behalf, as the last word of
philanthropy and beneficence; and the atrocities that subsequently
ensue are professedly undertaken, not against the unfortunate
people who endure them, but to liberate them from the ruler of their
choice or sufferance, in whose behalf however they fight to the
death.
To every country, therefore, which would fain be spared from these
discreditable wars with barbarian tribes on the borders of its
colonies, it is clear that the greatest caution is necessary against the
abuses of missionary propagandism. The almost absolute failure of
missions in recent centuries, and more especially in the nineteenth,
is intimately associated with the greater political importance which
the improved facilities of travel and intercourse have conferred upon
them. Everyone has heard how Catholicism was persecuted in Japan,
till at last the very profession of Christianity was made a capital
crime in that part of the world. But a traveller, who knew the East
intimately at the time, explains how it was that the Jesuits’ labours
resulted so disastrously. On the outbreak of civil dissensions in
Japan, ‘the Christian priests thought it a proper time for them to
settle their religion on the same foundation that Mahomet did his, by
establishing it in blood. Their thoughts ran on nothing less than
extirpating the heathen out of the land, and they framed a
conspiracy of raising an army of 50,000 Christians to murder their
countrymen, that so the whole island might be illuminated by
Christianity such as it was then.’[240] And in the same way, a modern
writer, speaking of the very limited success of missions in India, has](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-41-320.jpg)
![asserted frankly that ‘in despair many Christians in India are driven
to wish and pray that some one, or some way, may arise for
converting the Indians by the sword.’[241]
Nor are the heathen themselves blind to the political dangers which
are involved in the presence of missionaries among them. All over
the world conversion is from the native point of view the same thing
as disaffection, and war is dreaded as the certain consequence of
the adoption of Christianity. The French bishop, Lefebvre, when
asked by the mandarins of Cochin China, in 1847, the purpose of his
visit, said that he read in their faces that they suspected him ‘of
having come to excite some outbreak among the neophytes, and
perhaps prepare the way for an European army;’ and the king was
‘afraid to see Christians multiply in his kingdom, and in case of war
with European Powers, combine with his enemies.’[242] How right
events have proved him to have been!
The story is the same in Africa. ‘Not long after I entered the country,’
said the missionary, Mr. Calderwood, of Caffraria, ‘a leading chief
once said to me, “When my people become Christians, they cease to
be my people.”’[243] The Norwegian missionaries were for twenty
years in Zululand without making any converts but a few destitute
children, many of whom had been given to them out of pity by the
chiefs,[244] and their failure was actually ascribed by the Zulu king to
their having taught the incompatibility of Christianity with allegiance
to a heathen ruler.[245] In 1877, a Zulu of authority expressed the
prevalent native reasoning on this point in language which supplies
the key to disappointments that extend much further than Zululand:
‘We will not allow the Zulus to become so-called Christians. It is not
the king says so, but every man in Zululand. If a Zulu does anything
wrong, he at once goes to a mission-station, and says he wants to
become a Christian; if he wants to run away with a girl, he becomes
a Christian; if he wishes to be exempt from serving the king, he puts
on clothes, and is a Christian; if a man is an umtagati (evil-doer), he
becomes a Christian.’[246]](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-42-320.jpg)
![It is on this account that in wars with savage nations the destruction
of mission-stations has always been so constant an episode. Nor can
we wonder at this when we recollect that in the Caffre war of 1851,
for instance, it was a subject of boast with the missionaries that it
was Caffres trained on the mission-stations who had preserved the
English posts along the frontiers, carried the English despatches, and
fought against their own countrymen for the preservation and
defence of the colony.[247] It is rather a poor result of all the money
and labour that has been spent in the attempt to Christianise South
Africa, that the Wesleyan mission-station at Edendale should have
contributed an efficient force of cavalry to fight against their
countrymen in the Zulu campaign; and we may hesitate whether
most to despise the missionaries who count such a result as a
triumph of their efforts, or the converts whom they reward with tea
and cake for military service with the enemies of their countrymen.
[248]
It needs no great strain of intelligence to perceive that this use of
mission-stations as military training-schools scarcely tends to
enhance the advantages of conversion in the minds of the heathen
among whom they are planted.
For these reasons, and because it is becoming daily more apparent
that wars are less a necessary evil than an optional misery of human
life, the principal measure for a country which would fain improve,
and live at peace with, the less civilised races which touch the
numerous borders of its empire, would be the legal restraint or
prevention of missionary enterprise: a proposal that will appear less
startling if we reflect that in no quarter of the globe can that method
of civilising barbarism point to more than local or ephemeral success.
The Protestant missions of this century are in process of failure, as
fatal and decided as that which befel the Catholic missions of the
French, Portuguese, or Spanish, in the sixteenth and seventeenth
centuries, and very much from the same causes. The English wars in
South Africa, with which the Protestant missionaries have been so
closely connected, have frustrated all attempts to Christianise that](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-43-320.jpg)
![region, just as ‘the fearful wars occasioned directly or indirectly by
the missionaries’ sent by the Portuguese to the kingdoms of Congo
and Angola in the sixteenth century rendered futile similar attempts
on the West Coast.[249]
The same process of depopulation under Protestant influences may
now be observed in the Sandwich Islands or New Zealand that
reduced the population of Hispaniola, under Spanish Christianity,
from a million to 14,000 in a quarter of a century.[250] No Protestant
missionary ever laboured with more zeal than Eliot did in America in
the seventeenth century, but the tribes he taught have long since
been extinct: ‘like one of their own forest trees, they have withered
from core to bark;’[251] and, in short, the history of both Catholic
and Protestant missions alike may be summed up in this one general
statement: either they have failed altogether of results on a
sufficient scale to be worthy of notice, or the impartial page of
history unfolds to us one uniform tale of civil war, persecution,
conquest, and extirpation in whatever regions they can boast of
more at least of the semblance of success.
Another measure in the interests of peace would be the organisation
of a class of well-paid officials whose duty it should be to examine
on the spot into the truth of all rumours of outrages or atrocities
which are circulated from time to time, in order to set the tide of
public opinion in favour of hostile measures. Such rumours may, of
course, have some foundation, but in nine cases out of ten they are
false. So lately as the year 1882, the Times and other English papers
were so far deceived as to give their readers a horrible account of
the sacrifice of 200 young girls to the spirits of the dead in
Ashantee; and people were beginning to ask themselves whether
such things could be suffered within reach of an English army, when
it was happily discovered that the whole story was fictitious. Stories
of this sort are what the Germans call Tendenzlügen, or lies invented
to produce a certain effect. Their effect in rousing the war-spirit is
undeniable; and, although the healthy scepticism which has of
recent years been born of experience affords us some protection, no](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-44-320.jpg)
![about the swords being turned into ploughshares as in part fulfilled,
because his co-religionists, who in times past had killed one another,
did not then know how to fight even with their enemies. The charge
made by Celsus against the Christians, that they refused to bear
arms even in case of necessity, was admitted by Origen, but justified
on the ground of the unlawfulness of war. ‘We indeed,’ he says, ‘fight
in a special way on the king’s behalf, but we do not go on campaigns
with him, even should he press us to do so; we do battle on his
behalf as a peculiar army of piety, prevailing by our prayers to God
for him.’ And again: ‘We no longer take up the sword against people,
nor learn to make war any more, having become through Jesus, who
is our general, sons of peace.’ Nothing could be clearer nor more
conclusive than this language; and the same attitude towards war
was expressed or implied by the following Fathers in chronological
order: Justin Martyr, Tatian, Clement of Alexandria, Tertullian,
Cyprian, Lactantius, Archelaus, Ambrose, Chrysostom, Jerome, and
Cyril. Eusebius says that many Christians in the third century laid
aside the military life rather than abjure their religion. Of 10,050
pagan inscriptions that have been collected, 545 were found to
belong to pagan soldiers, while of 4,734 Christian inscriptions of the
same period, only 27 were those of soldiers; from which it seems
rather absurd to infer, as a French writer has inferred, not that there
was a great disproportion of Christian to pagan soldiers in the
imperial armies, but that most Christian soldiers being soldiers of
Christ did not like to have it recorded on their epitaphs that they had
been in the service of any man.[252]
On the other hand, there were certainly always some Christians who
remained in the ranks after their conversion, in spite of the military
oath in the names of the pagan deities and the quasi-worship of the
standards which constituted some part of the early Christian
antipathy to war. This is implied in the remarks of Tertullian, and
stands in no need of the support of such legends as the Thundering
Legion of Christians, whose prayers obtained rain, or of the Theban
legion of 6,000 Christians martyred under Maximian. It was left as a
matter of individual conscience. In the story of the martyr](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-51-320.jpg)
![Maximilian, when Dion the proconsul reminded him that there were
Christian soldiers among the life-guards of the Emperors, the former
replied, ‘They know what is best for them to do; but I am a Christian
and cannot fight.’ Marcellus, the converted centurion, threw down
his belt at the head of his legion, and suffered death rather than
continue in the service; and the annals of the early Church abound
in similar martyrdoms. Nor can there be much doubt but that a love
of peace and dislike of bloodshed were the principal causes of this
early Christian attitude towards the military profession, and that the
idolatry and other pagan rites connected with it only acted as minor
and secondary deterrents. Thus, in the Greek Church St. Basil would
have excluded from communion for three years any one who had
shed an enemy’s blood; and a similar feeling explains Theodosius’
refusal to partake of the Eucharist after his great victory over
Eugenius. The canons of the Church excluded from ordination all
who had served in an army after baptism; and in the fifth century
Innocent I. blamed the Spanish churches for their laxity in admitting
such persons into holy orders.[253]
The anti-military tendency of opinion in the early period of
Christianity appears therefore indisputable, and Tertullian would
probably have smiled at the prophet who should have predicted that
Christians would have ceased to keep slaves long before they should
have ceased to commit murder and robbery under the fiction of
hostilities. But it proves the strength of the original impetus, that
Ulphilas, the first apostle to the Goths, should purposely, in his
translation of the Scriptures, have omitted the Books of Kings, as too
stimulative of a love of war.
How utterly in this matter Christianity came to forsake its earlier
ideal is known to all. This resulted partly from the frequent use of
the sword for the purpose of conversion, and partly from the rise of
the Mahometan power, which made wars with the infidel appear in
the light of acts of faith, and changed the whole of Christendom into
a kind of vast standing military order. But it resulted still more from
that compromise effected in the fourth century between paganism](https://image.slidesharecdn.com/2357537-250528113527-77c12b31/85/Constraints-In-Discourse-2-Peter-Khnlein-Ed-Anton-Benz-Ed-52-320.jpg)
Constraints In Discourse 2 Peter Khnlein Ed Anton Benz Ed
- 1. Constraints In Discourse 2 Peter Khnlein Ed Anton Benz Ed download https://ebookbell.com/product/constraints-in-discourse-2-peter- khnlein-ed-anton-benz-ed-4715074 Explore and download more ebooks at ebookbell.com
- 2. Here are some recommended products that we believe you will be interested in. You can click the link to download. Constraints In Discourse 2 Peter Khnlein Ed Anton Benz Ed https://ebookbell.com/product/constraints-in-discourse-2-peter- khnlein-ed-anton-benz-ed-1892588 Constraints In Discourse 172th Edition Anton Benz Ed Peter Khnlein Ed https://ebookbell.com/product/constraints-in-discourse-172th-edition- anton-benz-ed-peter-khnlein-ed-2010442 Constraints In Discourse 3 Representing And Inferring Discourse Structure Anton Benz Ed https://ebookbell.com/product/constraints-in-discourse-3-representing- and-inferring-discourse-structure-anton-benz-ed-4688972 Constraints In Computational Logics Theory And Applications International Summer School Ccl 99 Gifsuryvette France September 58 1999 Revised Lectures 1st Edition Jeanpierre Jouannaud https://ebookbell.com/product/constraints-in-computational-logics- theory-and-applications-international-summer-school- ccl-99-gifsuryvette-france-september-58-1999-revised-lectures-1st- edition-jeanpierre-jouannaud-4198638
- 3. Constraints In Phonological Acquisition 1st Edition Rene Kager https://ebookbell.com/product/constraints-in-phonological- acquisition-1st-edition-rene-kager-1715176 Functional Constraints In Grammar On The Unergativeunaccusative Distinction Susumu Kuno https://ebookbell.com/product/functional-constraints-in-grammar-on- the-unergativeunaccusative-distinction-susumu-kuno-1715714 Accounting For Constraints In Delay Systems Giorgio Valmorbida https://ebookbell.com/product/accounting-for-constraints-in-delay- systems-giorgio-valmorbida-47165958 Choices And Constraints In Family Life 2nd Edition Maureen Baker https://ebookbell.com/product/choices-and-constraints-in-family- life-2nd-edition-maureen-baker-34720864 Moral Animals Ideals And Constraints In Moral Theory Catherine Wilson https://ebookbell.com/product/moral-animals-ideals-and-constraints-in- moral-theory-catherine-wilson-1470702
- 6. Volume 194 Constraints in Discourse 2 Edited by Peter Kühnlein,Anton Benz and Candace L. Sidner Founding Editors Jacob L. Mey University of Southern Denmark Herman Parret Belgian National Science Foundation, Universities of Louvain and Antwerp Jef Verschueren Belgian National Science Foundation, University of Antwerp Pragmatics & Beyond New Series is a continuation of Pragmatics & Beyond and its Companion Series. The New Series offers a selection of high quality work covering the full richness of Pragmatics as an interdisciplinary field, within language sciences. Pragmatics & Beyond New Series (P&BNS) Editorial Board Robyn Carston University College London Thorstein Fretheim University of Trondheim John C. Heritage University of California at Los Angeles Susan C. Herring Indiana University Masako K. Hiraga St. Paul’s (Rikkyo) University Sachiko Ide Japan Women’s University Kuniyoshi Kataoka Aichi University Miriam A. Locher Universität Basel Sophia S.A. Marmaridou University of Athens Srikant Sarangi Cardiff University Marina Sbisà University of Trieste Deborah Schiffrin Georgetown University Paul Osamu Takahara Kobe City University of Foreign Studies Sandra A. Thompson University of California at Santa Barbara Teun A. van Dijk Universitat Pompeu Fabra, Barcelona Yunxia Zhu The University of Queensland Editor Anita Fetzer University of Würzburg Associate Editor Andreas H. Jucker University of Zurich
- 7. Constraints in Discourse 2 Edited by Peter Kühnlein function2form Anton Benz Centre for General Linguistics, Berlin Candace L. Sidner Worcester Polytechnic Institute John Benjamins Publishing Company Amsterdam/Philadelphia
- 8. Library of Congress Cataloging-in-Publication Data Constraints in discourse 2 / edited by Peter Kühnlein,Anton Benz and Candace L. Sidner. p. cm. (Pragmatics & Beyond New Series, issn 0922-842X ; v. 194) Includes bibliographical references and index. 1. Discourse analysis.2. Constraints (Linguistics) I.Kühnlein,Peter.II.Benz,Anton,1965- III. Sidner, C. L. IV. Title: Constraints in discourse two. P302.28.C67 2010 401’.41--dc22 2009047143 isbn 978 90 272 5438 2 (Hb ; alk. paper) isbn 978 90 272 8854 7 (Eb) © 2010 – John Benjamins B.V. No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other means, without written permission from the publisher. John Benjamins Publishing Co. · P.O. Box 36224 · 1020 me Amsterdam · The Netherlands John Benjamins North America · P.O. Box 27519 · Philadelphia pa 19118-0519 · usa The paper used in this publication meets the minimum requirements of American National Standard for Information Sciences – Permanence of Paper for Printed Library Materials, ansi z39.48-1984. 8 TM
- 9. Table of contents Rhetorical structure: An introduction 1 Peter Kühnlein Clause-internal coherence 15 Jerry R. Hobbs Optimal interpretation for rhetorical relations 35 Henk Zeevat Modelling discourse relations by topics and implicatures: The elaboration default 61 Ekatarina Jasinskaja The role of logical and generic document structure in relational discourse analysis 81 Maja Bärenfänger, Harald Lüngen, Mirco Hilbert Henning Lobin Obligatory presupposition in discourse 105 Pascal Amsili Claire Beyssade Conventionalized speech act formulae: From corpus findings to formalization 125 Ann Copestake Marina Terkourafi Constraints on metalinguistic anaphora 141 Philippe De Brabanter Appositive Relative Clauses and their prosodic realization in spoken discourse: A corpus study of phonetic aspects in British English 163 Cyril Auran Rudy Loock Index 179
- 11. Rhetorical structure An introduction Peter Kühnlein 0.1 General remarks Texts, and in general types of discourse, vary along a multitude of dimensions. Discourse can be spoken or written, monological or an exchange between a number of participants, it can be employed to inform, persuade (and serve many more or even mixed functions), it can take place in various settings and be arbitrarily extensive. However, some characteristics are shared by all kinds of texts. One of those shared properties is that text, and discourse in general, is struc- tured, and they in turn are so in a multitude of ways: classical written text as the present, e.g., typically has logical and graphical structuring into paragraphs, sections, chapters etc. depending on the type of text or the genre, but it is by nature monological. Spoken dialogue, at the other end of the spectrum, inter alia is characterized by assignment of and changes in roles participants assume in the exchange, stretches of overlapping speech, repairs and many more phenomena that are not regularly observed in written text (notwithstanding chats on the internet and the like) and which give rise to completely different types of structure. All of these forms of communication fall under the common denominator discourse; we will keep using this cover term here to refer to them. The different kinds of structures in discourse have been object toresearch for a considerable time. One type of structure has been of special interests for researchers working in more formal paradigms and has been hotly discussed ever since: it is what is called the rhetorical or coherence structure. Rhetorical structure is built by applying rhetorical relations recursively to elementary units of discourse. This kind of structure is to be distinguished from, e.g., cohesive structure that comes to existence by means of, e.g., coreference in various forms. The present collection comprises papers that give a wide variety of perspec- tives on the constraints governing discourse structure, and primarily rhetorical structure: various ways of thinking of constitutive units of discourse along with a variety of conceptions of rhetorical relations are presented, and the issue of which kind of structure is right for the description of the rhetorical make-up of discourse is tackled from different points of view.
- 12. Peter Kühnlein Accordingly, this introductory chapter is intended to provide the necessary background to understand the discussions by sketching as briefly as possible the state of the art; the reader is referred to the individual chapters in the volume where appropriate. As the previous volume, Constraints in Discourse (Benz Kühnlein, 2008), the present one is the result of selecting and compiling papers that are extended versions of presentations at a workshop in the series “Constraints in Discourse.” The second of these workshops was held in Maynooth, Ireland, and organized by Candace Sidner (chair), Anton Benz, John Harpur and Peter Kühnlein. All the authors who contribute to the present volume submitted their re-worked and substantially extended papers to a peer reviewing process, where each author had to review two other authors’ papers. In addition, John Benjamins conducted an own reviewing process before agreeing to publish the collection. This two-stage reviewing process is intended to secure high quality of the contributions. 0.2 Elementary units Just as in any formal description of structures, one basic step in describing rhetorical structure of a discourse is to identify the elementary units. Due to the multitude of dimensions along which discourse can vary and due to differ- ences in theoretical assumptions, there is no consensus on what to count as an elementary unit. Exemplarily, there is a divide between proposals for different domains: a proposal for spoken discourse can refer to intonational features as an important criterion for segment status, whereas a proposal made specifically for written dis- course can’t. On the other hand, a proposal set up for written discourse can make reference to punctuation and syntactic units, whereas the first is absent and the the latter are not reliably correct in spoken discourse. Research in prosodic features of discourse and its segments reaches far back: Butterworth (1975) reports that speech rate changes during discourse segments, being higher at the end of a segment than at the beginning. Chafe (1980) observes that pause lengths are varying at segment boundaries too. Much corpus based and computer linguistic research in this area was conducted by Julia Hirschberg with various collaborators, e.g., Hirschberg and Pierrehumbert (1986), Grosz and Hirschberg (1992), Hirschberg and Pierrehumbert (1992), or Hirschberg and Nakatani (1996). One of the most detailed empirical inquiries into the relation between discourse segmentation and prosody is given in (Hirschberg Nakatani, 1996),
- 13. Rhetorical structure and the methodology employed there deserves a little closer description. The authors set up a corpus of directives, where subjects had to give route descriptions of varying complexity through Boston. The first series of descriptions was given spontanously by the subjects, recorded and then transcribed. In a second series, the same subjects read the corrected (i.e., freed of false starts etc.) transcripts of their route descriptions, and again the speech was recorded and transcribed. The data obtained from one of the speakers were prosodically transcribed using the ToBI standard, split into intermediate phrases, pause lengths were measured and fundamental frequencies (F0) and energy (RMS) calculated. Then two groups of annotators marked up the texts with segment boundaries: one group was given the transcripts only, the other group was given transcripts plus recorded speech. The theory that served as background to segmenting the texts was that of Grosz and Sidner (1986). Their account is potentially indepen- dent from domain, i.e., applicable to both spoken and written discourse; Grosz and Sidner claim that discourse structure actually consists of three distinct, but interacting levels. The most central of these levels is the intentional one: for every coherent discourse, that is the claim, one can identify an overarching discourse purpose the initiating participant seeks to pursue. The segments of discourse according to this theory correspond to sub-purposes, the so-called discourse segment purposes. Elementary units in this theory correspond to single purposes. The two other levels, attention and linguistic realization, concern which objects are in the center of discourse and how the discourse is actually realized using cue-phrases and special markers. The results obtained by (Hirschberg Nakatani, 1996) in their study on spoken discourse confirm previous findings and reveal much more detail than, e.g., the work by Butterworth (1975); Chafe (1980) reports: both F0 and energy are higher at the beginnings of discourse segments than at their ends. Speech rate on the other hand increases towards the end of a segment, and pause lengths during a segment are shorter than before a segment beginning and after a segment end. So segment boundaries as judged according to purposes indeed seem to be correlated with measurable changes in the speech signal. These, and similar, findings seem to indicate good mutual support between the intention-based theory of discourse structure developed by Grosz and Sidner (1986) and the claim that discourse segment boundaries are marked intonationally in spoken discourse. Considerably more work than on spoken discourse has traditionally been devoted to written text than to spoken discourse phenomena. The pioneering work dates back to the 80s, and the cited work by Grosz and Sidner (1986) is among that. An account of discourse structure that had comparable impact at that time was developed by Mann and Thompson (1987a, 1988b). This account,
- 14. Peter Kühnlein known as Rhetorical Structure Theory (rst) was explicitly developed as a means to capture analysts’ judgements about writers’ intentions while composing texts. Thus, rst is devoted to the analysis of written text, but the analysis is not primar- ily guided by linguistic surface structure: according to its founders, it is rather “pre-realizational” in that it aims to describe the function of (the interplay of) constituents in abstraction from linguistic realization. Mann and Thompson (1987a, 1988b) claim that the base case for linguis- tic expressions conveying intentions are clauses of certain types: main clauses, non-restrictive relative clauses are of the right variety, whereas, e.g., restrictive relative clauses and complement clauses (e.g., in subject or object position in a matrix clause) are not counted as minimal units. It turned out in the develop- ment of rst since its inception that narrowing down the type of constructions that express writers’ intentions to clauses poses problems in multi-lingual applications: what is expressed in a clause in one language might more suitably be expressed in a different construction in another language. This case was made especially by Rösner and Stede (1992) and Carlson and Marc (2001) who consequently proposed extensions of the set of minimal units. The motivation for the inclusion of certain constructions (or the exclusion of others) is not always readily understand- able.So,researcherscomprisingCarlsonandMarcu(2001)andLüngenetal.(2006) (cf. also paper 0.5) working in the rst paradigm, but likewise Wolf and Gibson (2005,2006),optforincludingcertainpre-posed ppslike“OnMonday,”inthelistof minimal units, whereas temporal adverbs that potentially convey the same infor- mation (“Yesterday,”) are not included. One reason for this decision might be that those researchers are working on corpora based on journal texts, where temporal expressions preferably are of the variety they include in the list; so the decision to include one type of expression but not the other might be rather pragmatic than theory-driven. Other work has been less intention-oriented than that of Grosz and Sidner and that in the rst paradigm, and consequently employed a different reasoning to select units as elementary discourse units. One line of research that also dates back to the mid-80s of the last century seeks to understand coherence in more general terms than tied up with linguistics. In (Hobbs, 1985) and work that can be seen in its tradition, like (Kehler, 2002), it is argued that coherence in text is by and large a product of the capability of rational agents to understand the world as being coherent. In this tradition, in its roots at least dating back to Hume (cf. (MacCormack Calkins, 1913)), what is related by the rational mind are events or states of affairs. Consequently, what counts as a minimal unit in these accounts are expressions that can serve to convey states or events, or in short eventualities. A first class citizen here is the clause again, and once again with suitable restrictions excluding, e.g., restrictive relative clauses. For different reasons, Asher and Lascarides (2003)
- 15. Rhetorical structure and Reese et al. (2006, 2007) working in the Segmented Discourse Representation Theory (sdrt) paradigm too consider the expression of eventualities to be the decisive criterion for individuating minimal units. As is well known from work in Montague grammar and elsewhere, it is all too easy to coerce the type of expressions to that of an eventuality. In fact, the paper by Jerry Hobbs in this collection (see 0.5) focuses on going below the clause level as minimal units. Hobbs there points out that he does think that even single words potentially express eventualities. It seems there is a thin line between raising types to that of an eventuality too easily and missing out on sub-clausal constituents that in fact are rhetorically interesting. Yet another line of research different from the intention-oriented and the eventuality-based ones can be seen in processing-based accounts. One of the first candidates there, and also rooted in the mid-1980s is the work of Polanyi (1986, 1988); but also the work by Webber (2004) can be seen in that tradition. Polanyi’s ldm most closely mirrors the incremental nature of text processing in that a discourse tree is built by adding sentences as elementary units to the existing representation of the text so far perceived. Sentences obviously are larger units than clauses since they potentially consist of multiple clauses (matrix, relative clauses, complement clauses). Both the ldm and Webber’s d-ltag suggest exten- sions to sentential syntax to model discourse structure, and thus there seems to be no need for sub-sentential segmentation. Both ldm and d-ltag — at least as concerns segmentation — thus seem to follow the opposite strategy than that pursued by Hobbs in his contribution in the present volume and reserve the rhetorical importance to larger units. As a summary to the above approaches to segmentation, it seems that all accounts agree on a core set of units (main clauses that form sentences) that should be treated as elementary units, whereas there is large disagreement as to what else should be considered a unit in discourse. 0.3 Rhetorical relations In Section 0.2 various views on how to split up discourse were reported. The present section is concerned with putting Humpty Dumpty together again: it is agreed among linguists that coherent discourse should be represented as a connected structure where each segment is connected to the rest by rhetorical relations. Islands in the representation of the analysis of a text are dispreferred and viewed as either a sign of incoherence of the discourse under analysis, faulty analysis itself or some lack in descriptive power in the inventory of rhetorical relations.
- 16. Peter Kühnlein In what follows in this section, the accounts used to introduce segmentation strategies in the order chosen in Section 0.2 will be taken up in turn again and the rhetorical relations employed by those accounts will be sketched. On the intention-oriented side, Grosz and Sidner (1986) employ a surpris- ingly small set of rhetorical relations. In their seminal paper they mention only two of them, one being dominance (the dominated discourse unit serves to achieve the goal of the super-ordinate) and the other satisfaction precedence (the preceding goal has to be achieved before the next can be achieved). The authors are aware of the fact that in, e.g., the work by Mann Thompson (ultimately published in Mann and Thompson (1987a), but circulating in various grey versions before- hand) a much larger number of rhetorical relations are discussed. However, since for Grosz and Sidner primacy is on intentions and their relations to each other rather than on textual realizations of intentions, they can claim that dominance and satisfaction-precedence are sufficient for the description of rhetorical struc- ture and the specific relations between textual units derivable from structure and intention content. Grosz and Sidner (1986) explicitly set up their account for construction dialogues; given the goal orientation of that dialogue type, it seems that the inven- tory consisting of dominance and satisfaction-precedence suffices to describe the intentional structure of dialogues from that domain. This might be questioned, however, in a more general domain, where a putative task structure (if there is any) might be not as tightly bound to discourse structure. On the other hand, it has to be said that Grosz and Sidner (1986) do not deny the existence of more relations between discourse purposes. The claim, it seems, is just that the set suffices for the analysis of the given type of discourse. As mentioned, Mann and Thompson (1987a, 1988b) posit a much larger set of discourse relations that should be used to describe the functional role of elementary units as recognized by the analyst: according to that classifica- tion, one unit might, e.g., elaborate on another, or units might form a list. There are two main divides within the class of relations: the first divide concerns the functional classification of relations: part of them are subject-matter relations (reporting about facts), another part is presentational, employed to influence the readers’ stance towards the (main or local) discourse topic. Both of these types of relations can be realized in either of two ways (giving the second divide) — connecting a less important part of discourse (a satellite) to a more important one (called nucleus), or connecting nuclei to nuclei. The first type along the latter divide is called mono-nuclear relation (or nucleus-satellite relation), the second multi-nuclear. One of the tests for nuclearity of discourse units is an elimination test: elimi- nating nuclei from a text tends to render it incoherent, while eliminating satellites
- 17. Rhetorical structure tends to leave coherence intact. This property of nuclei has led Marcu (1996) to posit the nuclearity principle for rst, claiming that spans of texts are connected by a relation iff their nuclei are. (A consequence of that principle for the representa- tion of discourse structure will be discussed in 0.4.) According to Matthiessen and Thompson (1988), there is another indicator for nuclearity: they observed that there is a high correlation between the status of being a nucleus in a text and of being realized in a main clause just in case a rhetorical relation is present between two syntactically related clauses. Syntactically subordinate clauses tend to realize satellites in turn. As Matthiessen and Thompson (1988) warn, this is not a hard and fast rule, but rather a tendency, and counter examples abound. There even seem to be language specific discourse connectives that trigger an inversion in nuclearity, like the dutch connective zodat (so that) which in a majority of cases syntactically subordinates a nucleus to a satellite. Bateman and Rondhuis (1994, 1997), and recently Stede (2008), systemati- cally investigate rhetorical relations across different discourse theories and, for rst’s nuclearity, propose not to tie the assignment of nuclearity to the presence of certain rhetorical relations (i.e., to drop the divide between mono-nuclear and multi-nuclear relations) and to view assignment of nuclearity as an effect of the presence of other factors. This seems to be in line with the findings by Asher and Vieu (2005) who claim something similar for an analogous divide among rela- tions in sdrt. The insight that certain relations can be viewed as connecting nuclei to satellites or satellites to other satellites seems also to be the rationale behind the explosion of number of relations in the rst-flavor proposed by Carlson and Marcu (2001), where multi-nuclear versions of relations formerly categorized as mono-nuclear abound. The discussion about the “right” relations for rst doesn’t seem to be settled nor does it seem it has to be: Taboada and Mann (2006a) in their recent overview over developments in rst propose that researchers in the paradigm should tailor their own relations according to their specific needs for specific purposes. The situation is different in sdrt, which, as mentioned, knows a similar divide as the nucleus-satellite distinction in rst. sdrt knows thorough axiomatizations of the discourse relations that are employed. These relations take the semantic representations of the minimal units and join them in either of two ways: by coordinating a unit to a preceding one, or by subordinating one unit to another. The nature of the relation involved (subordinating or coordinating) has influence on the possibilities where subsequent units can be attached: if the last relation involved was coordinating, then the constituent to which the last unit was related by it is blocked for attachment. If the last relation, on the other hand, was subordi- nating, then both the last unit and the one to which it was attached are available. These constraints on attachment points for new discourse units give rise to what is
- 18. Peter Kühnlein called the Right Frontier Constraint (rfc), first postulated by (Polanyi, 1986). One effect of the rfc is to limit anaphoric accessibility: antecedents are said to be only available for (pronominal) anaphoric uptake if they occur in a unit that is on the right frontier. Asher and Vieu (2005) re-examine the distinction between the two classes of relations and suggest that the question whether, e.g., a cause-relation is subordinat- ing or coordinating depends in part on the information structure exhibited by the units that are related. Certain information structural configurations in the units can lead to anaphoric accessibility of discourse referents whereas truth semanti- cally equivalent variants of that information structure makes them inaccessible. This fact can be accounted for if it is assumed that the information structure at least in part can change the way a unit is attached to preceding discourse. sdrt draws another distinction between discourse relations that resembles the distinction between presentational and subject-matter relations in rst: many of the discourse relations are content-level relations which are similar to subject- matter relations, whereas other relations bear more resemblance to presentational relations, like text-structuring, cognitive-level discourse relations or metatalk rela- tions. Interestingly, the so-called satisfaction scheme for veridical relations holds for some, but not all, relations of either variety. The latter scheme tells that two (representations of) discourse units connected by a relation are part of the inter- pretation of a discourse just in case the interpretations of the units are and the interpretation of the relation is. Whereas a veridicality criterion like that is to be expected for content-level relations, it is not so clear that a relation like parallel (a text-structuring relation) should have that property. None of the cognitive-level relations are veridical, though. Just like sdrt and the account of Grosz and Sidner (1986), but unlike rst, the account of discourse relations given by Hobbs (1985); Hobbs et al. (1993) and, more or less based on it, Kehler (2002) is an attempt to give a principled way to define the ways units of discourse are combined to form larger units. Hobbs et al. (1993) distinguishes four classes of discourse relations: some that are inferred to hold because the units that are connected are about events in the world (like casual relations), others that relate what was said to an overall goal of the discourse, again others that relate a unit to the recipient’s prior knowledge (e.g., background) and finally “expansion” relations (like contrast). 0.4 Structures and their properties So, both what counts as minimal units and in which ways they can be combined by rhetorical relations are matters of dispute in discourse theory. Given this situation,
- 19. Rhetorical structure it can be expected that there is also no consensus on which structures discourse can be expected to have. The expectation is confirmed by the literature. Whereas many researchers — e.g., Polanyi (1986, 1988, 2001), Grosz and Sidner (1986, 1998), Mann and Thompson (1987b, 1988a), Taboada and Mann (2006b) — assume that trees suffice to model the rhetorical structure of dialogue, there is an increasing number of theorists that doubt this assumption for a variety of reasons. Prominent among the latter are Wolf and Gibson (2005, 2006) who argue for a much less constrained type of graphs for the description of rhetorical structure. The chain graphs they postulate as adequate for the description of rhetorical struc- ture feature all kinds of violations of tree structure: they posit nodes with multiple parents, crossing edges, and in general graphs without a distinguished root node. Their strongest constraint on structures seems to be connectedness and acyclicity. These graphs are capable of describing all kinds of relations between elementary units; a closer look at their annotation manual and the set of relations they employ reveals that this seeming strength is a real weakness too: the set of relations Wolf and Gibson (2005, 2006) employ is a mixed bag, mostly taken from (Hobbs, 1985; Hobbs et al. 1993), but considerably modified and enriched with some relations from (Carlson Marcu, 2001). The annotation manual requires analysts to anno- tate not only rhetorical relations used for combining minimal units, but also coref- erence relations and other cohesive devices that can be present within minimal units as well. Their first step of analysis, grouping, actually consists in connecting units that are related by cohesive links. Only after that very step rhetorical relations are applied to the units — alas not to the units connected by the first step. Thus, it is no wonder that crossing dependencies and nodes with multiple parents abound in the analyses presented in Wolf and Gibson (2005, 2006). Knott (2007) questions the statistics Wolf and Gibson (2005, 2006) perform on their data, claiming that the small percentage of tree violations that can be tied to the special relations intro- duced by Wolf and Gibson in their evaluation of their data is implausible. I don’t think so: rather, Knott’s critique seems to set in too late. The true reason for the high amount of tree violations does not lie in the special relations, but in the conflation of levels of analysis. Another line of attack on tree structures as the adequate description for rhe- torical structures can be found in (Danlos, 2004, 2008). Danlos compares the generative capacity of a comparatively unrestricted formalism (an extension of Mel’c̆uk’s (Roberge, 1979) dependency syntax to discourse) with those of rst and sdrt. She derives all the structures that can be generated by either formalism and tries to find discourses that exhibit the respective structure. Her benchmark for- malism generates directed acyclic graphs (dags), whereas rst and sdrt generate trees. According to Danlos analysis, rst undergenerates (is not complete), her
- 20. Peter Kühnlein benchmark formalism overgenerates (is not correct), and sdrt is closest to being bothcompleteandcorrect, with theexception being a few structures that can not be described as sdrt-trees. Danlos argument to my mind has two drawbacks, though it is admirably ingenious. First, it should be evaluated on corpus data instead of relying on constructed discourse for confirmation. This is mainly a precaution against an overreliance on intuitions, of course. The second point is a bit stronger: the reconstruction of rst mainly — based on (Marcu, 1996) and (Carlson Marcu, 2001) — seems to contain too strong an interpretation of the nuclearity principle that leads to the assumption of graphs that are not in accordance with most other work in rst. So, rather than raising an argument against general rst assumptions, her attack is directed against a very idiosyncratic version of rst. But both the account of Wolf and Gibson (2005, 2006) and of Danlos (2004, 2008) are under active discussion, and until there is conclusive evidence to the contrary, it has to be assumed that there are strong arguments against general tree- hood of discourse structure. A weaker warning comes from Webber (2001) and Lee et al. (2008): these authors caution that although most discourse can be mod- elled as trees, there might be certain cases where a departure from tree structures is required. So the warning would be to give up the general claim in favour of a rule of thumb with dened exceptions. It seems that the question of how rich a structure has to be assumed for the description of discourse these days is more hottly debated than ever. The papers in the present volume will help to solve or focus in this debate by contributing insights in the fundamental questions that have to be answered. 0.5 About the papers Jerry R. Hobbs: Clause-Internal Coherence As was discussed in Section 0.2, there is no unanimity about the size or gen- eral characterization of elementary discourse units, just as there is no agreement on the definitions of relations between them. In his contribution to this volume, Hobbs extends his account, e.g., from (Hobbs, 1985), to cover coherence at a sub-clausal level. Henk Zeevat: Optimal Interpretation for Rhetorical Relations Zeevat argues that rhetorical relations can be reconstructed from general optimality theoretic (ot) assumptions. He gives a comprehensive introduction to ot, with special emphasis on the constraints *new, relevance, faith and plausi- ble. He then continues to demonstrate how a range of rhetorical relations can be derived from a certain ordering of these constraints; most importantly, *new and
- 21. Rhetorical structure relevance tend to introduce rhetorical structure defaults, with plausible being a filter over the generated relations. This account of coherence relations is in marked contrast to accounts such as that of Hobbs (1985) or Asher and Lascarides (2003). Ekatarina Jasinskaja: Modelling Discourse Relations by Topics and Implicatures: The Elaboration default Jasinskaja argues in her paper for the position that discourse relations can be inferred by utilising underlying pragmatic principles such as topic continuation and exhaustive interpretation as defaults. In the absence of linguistic markers that make one more inclined to infer a different relation, she opts for Elaboration as one of the default relations, since it best obeyes both principles, i.e., does not induce topic shifts and at the same time add information to the topic at hand. Maja Bärenfänger, Harald Lüngen, Mirco Hilbert Henning Lobin: The role of logical and generic document structure in discourse analysis The authors of this contribution propose to add two descriptive levels to the local rhetorical analysis of discourse structure: the logical structure (like title, paragraph etc.) and the genre specific structure (introduction, method). Structure at these levels is usually not explicitly signalled, yet conventionalized, and can thus be used to guide (automatic) parsing of texts. The authors strive to clarify which cues and constraints can be observed at these levels of discourse and demonstrate their utility for automatic text processing. Pascal Amsili Claire Beyssade: Obligatory Presupposition in Discourse Presupposition triggers have been considered obligatory under certain condi- tions by a variety of authors. One of the conditions that was deemed necessary in previous work was that the triggers are additive particles, like too. Amsili Beyssade argue that this condition is not a necessary one, but that obligatoriness is the case for triggers that have no asserted content. (Too being but one of them.) They give a general explanation for the apparent sensitivity of this class of triggers to discourse relations and provide a formalization in terms of an sdrt update mechanism, building on Asher and Lascarides (1998). Ann Copestake Marina Terkourafi: Conventionalized speech act formulae — from corpus findings to formalization Copestake Terkourafi present an account to the semantics and pragmatics of conventionalized speech acts which renders the contribution of the illocution- ary force as an addition to the compositional semantics of the utterance. They motivate their account with examples from a corpus of Cypriot Greek and formalize it within the framework of hpsg. They show how their account leaves the possibility for a literal interpretation of conventionalized speech act formulae open, thus opening the possibility to react to them in a variety of ways.
- 22. Peter Kühnlein Philippe De Brabanter: Constraints on metalinguistic anaphora De Brabanter reports the results of his research into a specific kind of referring expressions, where the referent itself is a linguistic object (like in “‘Boston’ is disyl- labic”). He draws a number of distinctions among those expressions, arguing that the class of metalinguistic anaphora referring back to expressions that themselves do have non-linguistic referents are especially interesting for a number of reasons. Cyril Auran Rudy Loock: Appositive Relative Clauses and their Prosodic Realization in Spoken Discourse: a Corpus Study of Phonetic Aspects in British English In their contribution to the volume, Auran Loock argue that differences in pragmatic functions fulfilled by appositive relative clauses are correlated to differ- ences both in morphosyntactic and semantic characteristics and prosodic features. The latter mainly concern intonation, rhythm and intensity. The data they use are extracted from a corpus of spoken British English. Bibliography Asher, N. Lascarides, A. (1998). The semantics and pragmatics of presupposition. Journal of Semantics, 15: 239–99. Asher, N. Lascarides, A. (2003). Logics of Conversation. Cambridge University Press. Asher, N. Vieu, L. (2005). Subordinating and coordinating discourse relations. Lingua, 115: 591–610. Bateman, J. Rondhuis, K.J. (1994). Coherence relations: analysis and specification. Technical Report R1.1.2: a,b, DANDELION Esprit Basic Research Project 6665. Bateman, J. Rondhuis, K.J. (1997). Coherence relations: towards a general specification. Discourse Processes, 24: 3–49. Benz, A. Kühnlein, P., editors (2008). Constraints in Discourse, volume 172 of Pragmatics and Beyond new series. John Benjamins. Butterworth, B. (1975). Hesitations and semantic planning in speech. Journal of Psycholinguistic Research, 4: 75–87. Carlson, L. Marcu, D. (2001). Discourse tagging manual. Technical Report ISI-TR-545, ISI. http://www.isi.edu/marcu/discourse/tagging-ref-manual.pdf. Chafe, W.L. (1980). The Pear Stories, volume 3 of Advances in Discourse Processes, chapter The Deployment of Consciousness in the Production of a Narrative, pages 9–50. Ablex. Danlos, L. (2004). Discourse dependency structures as constrained dags. In Strube, M. Sidner, C., editors, Proceedings of the 5th SIGdial Workshop on Discourse and Dialogue, pages 127–135. Danlos, L. (2008). Strong generative capacity of rst, sdrt and discourse dependency dags. In Benz, A. Kühnlein, P., editors, Constraints in Discourse, pages 69–96. John Benjamins. Grosz, B. Hirschberg, J. (1992). Some intonational characteristics of discourse structure. In Proceedings of the International Conference on Spoken Language Processing, pages 429–32.
- 23. Rhetorical structure Grosz, B.J. Sidner, C. (1986). Attention, intentions, and the structure of discourse. Computa- tional Linguistics, 12(3): 175–204. Grosz,B.J.Sidner,C.(1998).LostIntuitionsandForgottenIntentions.InWalker,M.A.,Joshi,A.K., Prince, E.F., editors, Centering Theory in Discourse, pages 39–51. Clarendon Press. Hirschberg, J. Nakatani, C.H. (1996). A prosodic analysis of discourse segments in direction-giving monologues.InProceedingsoftheAnnualMeetingoftheAssociationforComputationalLinguistics, pages 286–93, Santa Cruz. Hirschberg, J. Pierrehumbert, J. (1986). The Intonational Structuring of Discourse. In Proceedings of the 24th Annual Meeting of the Association for Computational Linguistics, pages 136–44. Hirschberg, J. Pierrehumbert, J. (1992). The intonational structuring of discourse. In Proceedings of the 24th Annual Meeting of the Association for Computational Linguistics, pages 136–44. ACL. Hobbs, J. (1985). On the coherence and structure of discourse. Technical Report 85-37, CSLI. Hobbs, J., Stickel, M., Appelt, D., Martin, P. (1993). Interpretation as abduction. Technical report, SRI International. Kehler, A. (2002). Coherence, Reference and the Theory of Grammar. CSLI. Kempen, G., editor (1987). Natural Language Generation. Number 135 in NATO Advanced Science Institutes—Applied Sciences. Martinus Nijhoff Publishers. Knott, A. (2007). Review of “Coherence in Natural Language: Data Structures and Applications” by Florian Wolf Edward Gibson. Computational Linguistics, 33(4): 591–5. Lee, A., Prasad, R., Joshi, A., Webber, B. (2008). Departures from tree structures in discourse: Shared arguments in the penn discourse treebank. In Benz, A., Kühnlein, P., Stede, M., editors, Proceedings of CID III. Lüngen, H., Puskás, C., Bärenfänger, M., Hilbert, M., Lobin, H. (2006). Discourse Segmentation of German Written Texts. In Advances in Natural Language Processing, volume 4139 of Lecture Notes in Computer Science, pages 245–56. Springer. MacCormack, T.J. Calkins, M.W., editors (1913). Hume, David: An enquiry concerning human understanding and selections from A treatise of human nature, volume 7 of Bibliotheca philosophorum. Meiner. Mann, W.C. Thompson, S.A. (1987a). Rhetorical Structure Theory: A Theory of Text Organization. Technical Report RS-87-190, Information Sciences Institute, Los Angeles, CA. Mann, W.C. Thompson, S.A. (1987b). Rhetorical Structure Theory: Description and Construction of Text Structures. in: (Kempen, 1987). pp. 85–95. Mann, W.C. Thompson, S.A. (1988a). Dialogue Games: Towards a functional theory of text organization. Text, 8(3): 243–81. Mann, W.C. Thompson, S.A. (1988b). Rhetorical Structure Theory: Toward a functional theory of text organization. Text, 8(3): 243–81. Marcu, D. (1996). Building Up Rhetorical Structure Trees. In The Proceedings of the Thirteenth National Conference on Artificial Intelligence, pages 1069–74, Portland, Oregon. Matthiessen, C. Thompson, S.A. (1988). The structure of discourse and ‘subordination’. In Haimann, J. Thompson, S.A., editors, Clause Combining in Grammar and Discourse, volume 18 of Typological Studies in Language. John Benjamins Publishing Company. Polanyi, L. (1986). The linguistic discourse model: Towards a formal theory of discourse structure. Techn. Report TR-6409, BBN Laboratories Inccap. Polanyi, L. (1988). A formal model of the structure of discourse. Journal of Pragmatics, 12: 601–38.
- 24. Peter Kühnlein Polanyi,L.(2001).TheLinguisticStructureofDiscourse.InSchiffrin,D.,Tannen,D.,Hamilton, H.E., editors, Handbook of Discourse Analysis. Blackwell. Reese, B., Denis, P., Asher, N., Baldridge, J., Hunter, J. (2006). Reference manual for the analysis and annotation of rhetorical structure. Technical report, Discor, Univ. of Texas, Austin. Reese, B., Hunter, J., Asher, N., Denis, P., Baldridge, J. (2007). Reference manual for the analysis and annotation of rhetorical structure (version 1.0). Technical report, Discor, Univ. of Texas, Austin. http://comp.ling.utexas.edu/discor/manual.pdf. Roberge, R.T., editor (1979). Studies in Dependency Syntax — Igor A. Mel’ uk. Karoma, Ann Arbor. Rösner, D. Stede, M. (1992). Customizing rst for the automatic production of technical manuals. In Proceedings of the 6th International Workshop on Natural Language Generation, pages 199–214, London, UK. Springer-Verlag. Stede, M. (2008). RST revisited: disentagling nuclearity. In Fabricius-Hansen, C. Ramm, W., editors, ’Subordination’ versus ‘coordination’ in sentence and text. John Benjamins. Taboada, M. Mann, W.C. (2006a). Rhetorical structure theory: looking back and moving ahead. Discourse Studies, 8(3): 423–59. Taboada, M. Mann, W.C. (2006b). Rhetorical Structure Theory: looking back and moving ahead. Discourse Studies, 8(3): 423–59. Webber, B. (2004). D-LTAG: extending lexicalized TAG to discourse. Cognitive Science, 28: 751–79. Webber, B.L. (2001). Computational Perspectives on Discourse and Dialogue. In Schiffrin, D., Tannen, D., Hamilton, H., editors, The Handbook of Discourse Analysis. Blackwell. Wolf, F. Gibson, E. (2005). Representing Discourse Coherence: A Corpus-Based Study. Computational Linguistics, 31(2). Wolf, F. Gibson, E. (2006). Coherence in Natural Language: Data Structures and Applications. MIT Press. c̆
- 25. Another Random Document on Scribd Without Any Related Topics
- 26. Ridiculous as this negro’s idea of honour must appear to us, it bears a sufficient resemblance to other notions of the same kind that have passed current in the world at different times to satisfy us of the extreme variability of the sentiment in question. Cæsar built with difficulty a bridge across the Rhine, chiefly because he held it beneath his own dignity, or the Roman people’s, for his army to cross it in boats. The Celts of old thought it as ignominious to fly from an inundation, or from a burning or falling house, as to retreat from an enemy. The Spartans considered it inglorious to pursue a flying foe, or to be killed in storming a besieged city. The same Gauls who gloried in broadsword-wounds would almost go mad with shame if wounded by an arrow or other missile that only left an imperceptible mark. The use of letters was once thought dishonourable by all the European nations. Marshal Montluc, in the sixteenth century, considered it a sign of abnormal overbookishness for a man to prefer to spend a night in his study than to spend it in the trenches, though, now, a contrary taste would be thought by most men the mark of a fool. Such are some of the curious ideas of honour that have prevailed at different times. Wherein we seem to recognise not merely change but advance; one chief difference between the savage and civilised state lying in the different estimates entertained in either of martial prowess and of military honour. We laugh nowadays at the ancient Britons who believed that the souls of all who had followed any other pursuit than that of arms, after a despised life and an unlamented death, hovered perforce over fens and marshes, unfit to mingle with those of warriors in the higher and brighter regions; or at the horsemen who used before death to wound themselves with their spears, in order to obtain that admission to Walhalla which was denied to all who failed to die upon a battle-field; or at the Spaniards, who, when Cato disarmed them, preferred a voluntary death to a life destined to be spent without arms.[204] No civilised warrior would pride himself, as Fijian warriors did, on being generally known as the ‘Waster’ or ‘Devastator’ of such-and-such a district; the most he would look for would be a title and perhaps a perpetual
- 27. pension for his descendants. We have nothing like the custom of the North American tribes, among whom different marks on a warrior’s robe told at a glance whether his fame rested on the slaughter of a man or a woman, or only on that of a boy or a girl. We are inferior in this respect to the Dacota tribes, among whom an eagle’s feather with a red spot on it denoted simply the slaughter of an enemy, the same feather with a notch and the sides painted red, that the said enemy had had his throat cut, whilst according as the notches were on one side or on both, or the feather partly denuded, anyone could tell after how many others the hero had succeeded in touching the dead body of a fallen foe. The stride is clearly a great one from Pyrrhus, the Epirot king, who, when asked which of two musicians he thought the better, only deigned to reply that Polysperchon was the general, to Napoleon, the French emperor, who conferred the cross of the Legion of Honour on Crescentini the singer. And as the pursuit of arms comes with advancing civilisation to occupy a lower level as compared with the arts of peace, so the belief is the mark of a more polished people that the rapacity and cruelty which belong to the war customs of a more backward nation, or of an earlier time, are absent from their own. They invent the expression civilised warfare to emphasise a distinction they would fain think inherent in the nature of things; and look, by its help, even on the mode of killing an enemy, with a moral vision that is absurdly distorted. How few of us, for example, but see the utmost barbarity in sticking a man with an assegai, yet none whatever in doing so with a bayonet? And why should we pride ourselves on not mutilating the dead, while we have no scruples as to the extent to which we mutilate the living? We are shocked at the mention of barbarian tribes who poison their arrows, or barb their darts, yet ourselves think nothing of the frightful gangrenes caused by the copper cap in the Minié rifle-ball, and reject, on the score of the expense of the change, the proposal that bullets of soft lead, which cause needless pain, should no longer be used among the civilised Powers for small-arm ammunition.[205]
- 28. But whilst the difference in these respects between barbarism and civilisation is thus one that rather touches the surface than the substance of war, the result is inevitably in either state a different code of military etiquette and sentiment, though the difference is far less than in any other points of comparison between them. When the nations of Christendom therefore came in contact with unknown and savage races, whose customs seemed different from their own and little worthy of attention, they assumed that the latter recognised no laws of war, much as some of the earlier travellers denied the possession or faculty of speech to people whose language they could not interpret. From which assumption the practical inference followed, that the restraints which were held sacred between enemies who inherited the same traditions of military honour had no need to be observed in hostilities with the heathen world. It is worth while, therefore, to show how baseless was the primary assumption, and how laws of war, in no way dissimilar to those of Europe, may be detected in the military usages of barbarism. To spare the weak and helpless was and is a common rule in the warfare of the less civilised races. The Guanches of the Canary Islands, says an old Spanish writer, ‘held it as base and mean to molest or injure the women and children of the enemy, considering them as weak and helpless, therefore improper objects of their resentment; neither did they throw down or damage houses of worship.’[206] The Samoans considered it cowardly to kill a woman: [207] and in America the Sioux Indians and Winnebagoes, though barbarous enough in other respects, are said to have shown the conventional respect to the weaker sex.[208] The Basutos of South Africa, whatever may be their customs now, are declared by Casalis, one of the first French Protestant missionaries to their country, to have respected in their wars the persons of women, children, and travellers, and to have spared all prisoners who surrendered, granting them their liberty on the payment of ransom.[209]
- 29. Few savage races were of a wilder type than the Abipones of South America; yet Dobritzhoffer, the Jesuit missionary, assures us not only that they thought it unworthy of them to mangle the bodies of dead Spaniards, as other savages did, but that they generally spared the unwarlike, and carried away boys and girls uninjured. The Spaniards, Indians, negroes, or mulattoes whom they took in war they did not treat like captives, but with kindness and indulgence like children. Dobritzhoffer never saw a prisoner punished by so much as a word or a blow, but he bears testimony to the compassion and confidence often displayed to captives by their conquerors. It is common to read of the cruelty of the Red Indians to their captives; but Loskiel, another missionary, declares that prisoners were often adopted by the victors to supply the place of the slain, and that even Europeans, when it came to an exchange of prisoners, sometimes refused to return to their own countrymen. In Virginia notice was sent before war to the enemy, that in the event of their defeat, the lives of all should be spared who should submit within two days’ time. Loskiel gives some other rather curious testimony about the Red Indians. ‘When war was in contemplation they used to admonish each other to hearken to the good and not to the evil spirits, the former always recommending peace. They seem,’ he adds with surprise, ‘to have had no idea of the devil as the prince of darkness before the Europeans came into the country.’ The symbol of peace was the burial of the hatchet or war-club in the ground; and when the tribes renewed their covenants of peace, they exchanged certain belts of friendship which were singularly expressive. The principal belt was white, with black streaks down each side and a black spot at each end: the black spots represented the two people, and the white streak between them signified, that the road between them was now clear of all trees, brambles, and stones, and that every hindrance was therefore removed from the way of perfect harmony. The Athenians used the same language of symbolism when they declared war by letting a lamb loose into the enemy’s country: this
- 30. being equivalent to saying, that a district full of the habitations of men should shortly be turned into a pasture for sheep.[210] The Fijians used to spare their enemy’s fruit trees; the Tongan islanders held it as sacrilege to fight within the precincts of the burial place of a chief, where the greatest enemies were obliged to meet as friends. Most of the lower races recognise the inviolability of ambassadors and heralds, and have well-established emblems of a truce or armistice. The wish for peace which the Zulu king in vain sought from his English invaders by the symbol of an elephant’s tusk (1879), was conveyed in the Fiji Islands by a whale’s tooth, in the Sandwich by a young plantain tree or green branch of the ti plant, and among most North American tribes by a white flag of skin or bark. The Samoan symbol for an act of submission in deprecation of further hostilities conveys some indication of the possible origin of these pacific symbols. The conquered Samoan would carry to his victor some bamboo sticks, some firewood, and some small stones; for as a piece of split bamboo was the original Samoan knife, and small stones and firewood were used for the purpose of roasting pigs, this symbol of submission was equivalent to saying: ‘Here we are, your pigs, to be cooked if you please, and here are the materials wherewith to do it.’[211] In the same way the elephant’s tusk or the whale’s tooth may be a short way of saying to the victor: ‘Yours is the strength of the elephant or the whale; we recognise the uselessness of fighting with you.’ In the same way many savage tribes take the greatest pains to impress the terms of treaties as vividly as possible on the memory of the contracting parties by striking and intelligible ceremonies. In the Sandwich Islands a wreath woven conjointly by the leaders of either side and placed in a temple was the chief symbol of peace. On the Fiji Islands, the combatant forces would meet and throw down their weapons at one another’s feet. The Tahitians wove a wreath of green boughs, furnished by each side; exchanged two young dogs;
- 31. and having also made a band of cloth together, deposited the wreath and the band in the temple, with imprecations on the side which should first violate so solemn a treaty of peace.[212] On the Hervey Islands, the token of the cessation of war was the breaking of a number of spears against a large chestnut tree; the almost imperishable coral tree was planted in the valleys to signify the hope that the peace might last as long as the tree; and after the drum of peace had been solemnly beaten round the island, it was unlawful for any man to carry a weapon, or to cut down any iron-wood, which he might turn into an implement of destruction. Even our custom of proclaiming that a war is not undertaken against a people but against its rulers is not unknown in savage life. The Ashantee army used to strew leaves on their march, to signify that their hostility was not with the country they passed through but only with the instigators of the war; they told the Fantees that they had no war with them collectively, but only with some of them.[213] How common a military custom this appeal to the treason of an enemy is, notwithstanding the rarity of its success, everybody knows. When, for instance, the Anglo-Zulu war began, it was solemnly proclaimed that the British Government had no quarrel with the Zulu people; it was a war against the Zulu king, not against the Zulu nation. (Jan. 11, 1879.) So were the Ashantees told by the English invading force; so were the Afghans; so were the Egyptians; and so were the French by the Emperor William before his merciless hordes laid waste and desolate some of the fairest provinces of France; so, no doubt, will be told the Soudan Arabs. And yet this appeal to treason, this premium on a people’s disloyalty, is the regular precursor of wars, wherein destruction for its own sake, the burning of grain and villages for the mere pleasure of the flames, forms almost invariably the most prominent feature. The military view always prevails over the civil, of the meaning of hostilities that have no reference to a population but only to its government. In the Zulu war, for instance, in spite of the above proclamation, the lieutenant-general ordered raids to be made into Zululand for the express purpose of burning
- 32. empty kraals or villages; defending such procedure by the usual military logic, that the more the natives at large felt the strain of the war, the more anxious they would be to see it concluded; and it was quite in vain for the Lieutenant-Governor of Natal to argue that the burning of empty kraals would neither do much harm to the Zulus nor good to the English; and that whereas the war had been begun on the ground that it was waged against the Zulu king and not against his nation, such conduct was calculated to alienate from the invaders the whole of the Zulu people, including those who were well disposed to them. Such arguments hardly ever prevail over that passion for wanton destruction and for often quite unnecessary slaughter, which finds a ready and comprehensive shelter under the wing of military exigencies. The assumption, therefore, that savage races are ignorant of all laws of war, or incapable of learning them, would seem to be based rather on our indifference about their customs than on the realities of the case, seeing that the preceding evidence to the contrary results from the most cursory inquiry. But whatever value there may be in our own laws of war, as helping to constitute a real difference between savage and civilised warfare, the best way to spread the blessing of a knowledge of them would clearly be for the more civilised races to adhere to them strictly in all wars waged with their less advanced neighbours. An English commander, for instance, should no more set fire to the capital of Ashantee or Zululand for so paltry a pretext as the display of British power than he would set fire to Paris or Berlin; he should no more have villages or granaries burnt in Africa or Afghanistan than he would in Normandy; and he should no more keep a Zulu envoy or truce-bearer in chains[214] than he would so deal with the bearer of a white flag from a Russian or Italian enemy. The reverse principle, which is yet in vogue, that with barbarians you must or may be barbarous, leads to some curious illustrations of civilised warfare when it comes in conflict with the less civilised races. In one of the Franco-Italian wars of the sixteenth century,
- 33. more than 2,000 women and children took refuge in a large mountain cavern, and were there suffocated by a party of French soldiers, who set fire to a quantity of wood, straw, and hay, which they stacked at the mouth of the cave; but it was considered so shameful an act, that the Chevalier Bayard had two of the ringleaders hung at the cavern’s mouth.[215] Yet when the French General Pélissier in this century suffocated the unresisting Algerians in their caves, it was even defended as no worse than the shelling of a fortress; and there is evidence that gun-cotton was not unfrequently used to blast the entrance to caves in Zululand in which men, women, and children had hoped to find shelter against an army which professed only to be warring with their king.[216] The following description of the way in which, in the Ashantee war, the English forces obtained native carriers for their transport service is not without its instruction in this respect:— ‘We took to kidnapping upon a grand scale. Raids were made on all the Assin villages within reach of the line of march, and the men, and sometimes the women, carried off and sent up the country under guard, with cases of provisions. Lieutenant Bolton, of the 1st West India Regiment, rendered immense service in this way. Having been for some time commandant of Accra, he knew the coast and many of the chiefs; and having a man-of-war placed at his disposal, he went up and down the coast, landing continually, having interviews with chiefs, and obtaining from them large numbers of men and women; or when this failed, landing at night with a party of soldiers, surrounding villages, and sweeping off the adult population, leaving only a few women to look after the children. In this way, in the course of a month, he obtained several thousands of carriers.’[217] And then a certain school of writers talks of the love and respect for the British Empire which these exhibitions of our might are calculated to win from the inferior races! The Ashantees are disgraced by the practice of human sacrifices, and the Zulus have
- 34. many a barbarous usage; but no amount of righteous indignation on that account justifies such dealings with them as those above described. If it does, we can no longer condemn the proceedings of the Spaniards in the New World. For we have to remember that it was not only the Christianity of the Inquisition, or Spanish commerce that they wished to spread; not mere gold nor new lands that they coveted, but that they also strove for such humanitarian objects as the abolition of barbarous customs like the Mexican human sacrifices. ‘The Spaniards that saw these cruel sacrifices,’ wrote a contemporary, the Jesuit Acosta, ‘resolved with all their power to abolish so detestable and cursed a butchery of men.’ The Spaniards of the sixteenth century were in intention or expression every whit as humane as we English of the nineteenth. Yet their actions have been a reproach to their name ever since. Cortes subjected Guatamozin, king of Mexico, to torture. Pizarro had the Inca of Peru strangled at the stake. Alvarado invited a number of Mexicans to a festival, and made it an opportunity to massacre them. Sandoval had 60 caziques and 400 nobles burnt at one time, and compelled their relations and children to witness their punishment. The Pope Paul had very soon (1537) to issue a bull, to the effect that the Indians were really men and not brutes, as the Spaniards soon affected to regard them. The whole question was, moreover, argued out at that time between Las Casas and Sepulveda, historiographer to the Emperor Charles V. Sepulveda contended that more could be effected against barbarism by a month of war than by 100 years of preaching; and in his famous dispute with Las Casas at Valladolid in 1550, defended the justice of all wars undertaken against the natives of the New World, either on the ground of the latter’s sin and wickedness, or on the plea of protecting them from the cruelties of their own fellow- countrymen; the latter plea being one to which in recent English wars a prominent place has been always given. Las Casas replied— and his reply is unanswerable—that even human sacrifices are a smaller evil than indiscriminate warfare. He might have added that military contact between people unequally civilised does more to
- 35. barbarise the civilised than to civilise the barbarous population. It is well worthy of notice and reflection that the European battle-fields became distinctly more barbarous after habits of greater ferocity had been acquired in wars beyond the Atlantic, in which the customary restraints were forgotten, and the ties of a common human nature dissolved by the differences of religion and race. The same effect resulted in Roman history, when the extended dominion of the Republic brought its armies into contact with foes beyond the sea. The Roman annalists bear witness to the deterioration that ensued both in their modes of waging war and in the national character.[218] It is in an Asiatic war that we first hear of a Roman general poisoning the springs;[219] in a war for the possession of Crete that the Cretan captives preferred to poison themselves rather than suffer the cruelties inflicted on them by Metellus;[220] in the Thracian war that the Romans cut off their prisoners’ hands, as Cæsar afterwards did those of the Gauls.[221] And we should remember that a practical English statesman like Cobden foresaw, as a possible evil result of the closer relations between England and the East, a similar deterioration in the national character of his countrymen. ‘With another war or two,’ he wrote, ‘in India and China, the English people would have an appetite for bull- fights if not for gladiators.’[222] Nor is there often any compensation for such results in the improved condition of the tribes whom it is sought to civilise after the method recommended by Sepulveda. The happiest fate of the populations he wished to see civilised by the sword was where they anticipated their extermination or slavery by a sort of voluntary suicide. In Cuba, we are told that ‘they put themselves to death, whole families doing so together, and villages inviting other villages to join them in a departure from a world that was no longer tolerable.’[223] And so it was in the other hemisphere; the Ladrone islanders, reduced by the sword and the diseases of the Spaniards, took measures intentionally to diminish their numbers and to check population,
- 36. preferring voluntary extinction to the foul mercies of the Jesuits: till now a lepers’ hospital is the only building left on what was once one of the most populous of their islands. It must, however, be admitted in justice to the Spaniards, that the principles which governed their dealings with heathen races infected more or less the conduct of colonists of all nationalities. A real or more often a pretended zeal for the welfare of native tribes came among all Christian nations to co-exist with the doctrine, that in case of conflict with them the common restraints of war might be put in abeyance. What, for instance, can be worse than this, told of the early English settlers in America by one of themselves? ‘The Plymouth men came in the mean time to Weymouth, and there pretended to feast the savages of those parts, bringing with them forks and things for the purpose, which they set before the savages. They ate thereof without any suspicion of any mischief, who were taken upon a watchword given, and with their own knives hanging about their necks were by the Plymouth planters stabbed and slain.’[224] Among the early English settlers it soon came to be thought, says Mather, a religious act to kill an Indian. In the latter half of the seventeenth century both the French and English authorities adopted the custom of scalping and of offering rewards for the scalps of their Indian enemies. In 1690 the most healthy and vigorous Indians taken by the French ‘were sold in Canada, the weaker were sacrificed and scalped, and for every scalp they had a premium.’[225] Caleb Lyman, who afterwards became an elder of a church at Boston, left an account of the way in which he himself and five Indians surprised a wigwam, and scalped six of the seven persons inside, so that each might receive the promised reward. On their petition to the great and general court they received 30l. each, and Penhallow says not only that they probably expected eight times as much, but that at the time of writing the province would have readily paid a sum of 800l. for a similar service.[226] Captain Lovewell, says the same contemporary eulogist of the war that
- 37. lasted from July 1722 to December 1725, ‘from Dunstable with thirty volunteers went northward, who marching several miles up country came on a wigwam where were two Indians, one of whom they killed and the other took, for which they received the promised bounty of 100l. a scalp, and two shillings and sixpence a day besides.’ (December 19, 1724.)[227] At the surprise of Norridjwock ‘the number of dead which we scalped were 26, besides Mr. Rasle the Jesuit, who was a bloody incendiary.’[228] It is evident that these very liberal rewards must have operated as a frequent cause of Indian wars, and made the colonists open-eared to tales of native outrages; indeed the whites sometimes disguised themselves like Indians, and robbed like Indians, in order, it would appear, the more effectually to raise the war-cry against them.[229] Since the Spaniards first trained bloodhounds in Cuba to hunt the Indians, the alliance between soldiers and dogs has been a favourite one in barbarian warfare. The Portuguese used them in Brazil when they hunted the natives for slaves.[230] And an English officer in a treatise he wrote in the last century as a sort of military guide to Indian warfare suggested coolly: ‘Every light horseman ought to be provided with a bloodhound, which would be useful to find out the enemy’s ambushes and to follow their tracks. They would seize the naked savages, and at least give time to the horsemen to come up with them.’[231] In the Molucca Islands the use of two bloodhounds against a native chief was the cause of a great confederacy between all the islands to shake off the Spanish and Portuguese yoke.[232] And even in the war waged by the United States in Florida from 1838 to 1840, General Taylor was authorised to send to Cuba for bloodhounds to scent out the Indians; nor, according to one account, was their aid resorted to in vain.[233] Poison too has been called in aid. Speaking of the Yuta Indians, a traveller assures us that ‘as in Australia, arsenic and corrosive sublimate in springs and provisions have diminished their
- 38. number.’[234] And in the same way ‘poisoned rum helped to exterminate the Tasmanians.’[235] But there is worse yet in this direction. The Portuguese in Brazil, when the importation of slaves from Africa rendered the capture of the natives less desirable than their extermination, left the clothes of persons who had died of small-pox or scarlet fever to be found by them in the woods.[236] And the caravan traders from the Missouri to Santa Fé are said by the same method or in presents of tobacco to have communicated the small-pox to the Indian tribes of that district in 1831.[237] The enormous depopulation of most tribes by the small-pox since their acquaintance with the whites is one of the most remarkable results in the history of their mutual connection; nor is it likely ever to be known to what extent the coincidence was accidental. It is pleasant to turn from these practical illustrations of the theory that no laws of war need be regarded in hostilities with savage tribes to the only recorded trial of a contrary system, and to find, not only that it is associated with one of the greatest names in English history, but also that the success it met with fully justifies the suspicion and disfavour with which the commoner usage is beginning to be regarded. The Indians with whom Penn made his famous treaty in 1682 (of which Voltaire said that it was the only treaty that was never ratified by an oath, and the only treaty that was never broken), were of the same Algonquin race with whom the Dutch had scarcely ever kept at peace, and against whom they had warred in the customary ruthless fashion of those times. The treaty was based on the principle of an adjustment of differences by a tribunal of an equal number of Red men and of White. ‘Penn,’ says the historian, ‘came without arms; he declared his purpose to abstain from violence, he had no message but peace, and not one drop of Quaker blood was ever shed by an Indian’[238] For more than seventy years, from 1682 to 1754, when the French war broke out, in short, during the whole time that the Quakers had the
- 39. principal share in the government of Pennsylvania, the history of the Indians and Whites in that province was free from the tale of murders and hostilities that was so common in other districts; so that the single instance in which the experiment of equal laws and forbearance has been patiently persevered in, can at least boast of a success that in support of the contrary system it were very difficult to find for an equal number of years in any other part of the world. It may also be said against Sepulveda’s doctrine, that the habits of a higher civilisation, where they are really worth spreading, spread more easily and with more permanent effect among barbarous neighbours by the mere contagion of a better example than by the teaching of fire and sword. Some of the Dyak tribes in Borneo are said to have given up human sacrifices from the better influences of the Malays on the coast district.[239] The Peruvians, according to Prescott, spread their civilisation among their ruder neighbours more by example than by force. ‘Far from provoking hostilities, they allowed time for the salutary example of their own institutions to work its effect, trusting that their less civilised neighbours would submit to their sceptre from a conviction of the blessings it would secure to them.’ They exhorted them to lay aside their cannibalism, their human sacrifices, and their other barbarities; they employed negotiation, conciliatory treatment, and presents to leading men among the tribes; and only if all these means failed did they resort to war, but to war which at every stage was readily open to propositions of peace, and in which any unnecessary outrage on the persons or property of their enemy was punished with death. Something will have been done for the cause of this better method of civilising the lower races, if we forewarn and forearm ourselves against the symptoms of hostilities with them by a thorough understanding of the conditions which render such hostilities probable. For as an outbreak of fever is to some extent preventable by a knowledge of the conditions which make for fevers, so may the outbreak of war be averted by a knowledge of the laws which govern their appearance. The experience which we owe to history in
- 40. this respect is amply sufficient to enable us to generalise with some degree of confidence and certainty as to the causes or steps which produce wars or precede them; and from the remembrance of our dealings with the savage races of South Africa we may forecast with some misgivings the probable course of our connection with a country like New Guinea. A colony of Europeans in proximity with barbarian neighbours naturally desires before long an increase of territory at the expense of the latter. The first sign of such a desire is the expedition of missionaries into the country, who not only serve to spy it out for the benefit of the colony, but invariably weaken the native political force by the creation of a division of feeling, and of an opposition between the love of old traditions and the temptation of novel customs and ideas. The innovating party, being at first the smaller, consisting of the feeblest and poorest members of the community, and of those who gladly flock to the mission-stations for refuge from their offences against tribal law, the missionaries soon perceive the impossibility of further success without the help of some external aid. The help of a friendly force can alone turn the balance of influence in their favour, and they soon learn to contemplate with complacency the advantages of a military conquest of the natives by the colony or mother-country. The evils of war are cancelled, in their eyes, by the delusive visions of ultimate benefit, and, in accordance with a not uncommon perversion of the moral sense, an end that is assumed to be religious is made to justify measures that are the reverse. When the views and interests of the colonial settlers and of the missionaries have thus, inevitably but without design, fallen into harmony, a war is certain to be not far distant. Apparently accidental, it is in reality as certain as the production of green from a mixture of blue and yellow. Some dispute about boundaries, some passing act of violence, will serve for a reason of quarrel, which will presently be supported by a fixed array of collateral pretexts. The Press readily lends its aid; and in a week the colony trembles or
- 41. affects to tremble from a panic of invasion, and vials of virtue are expended on the vices of the barbarians which have been for years tolerated with equanimity or indifference. Their customs are painted in the blackest colours; the details of savage usages are raked up from old books of travel; rumours of massacres and injuries are sedulously propagated; and the whole country is represented as in such a state of anarchy, that the majority of the population, in their longing for deliverance from their own rulers, would gladly welcome even a foreign conqueror. In short, a war against them comes speedily to be regarded as a war in their behalf, as the last word of philanthropy and beneficence; and the atrocities that subsequently ensue are professedly undertaken, not against the unfortunate people who endure them, but to liberate them from the ruler of their choice or sufferance, in whose behalf however they fight to the death. To every country, therefore, which would fain be spared from these discreditable wars with barbarian tribes on the borders of its colonies, it is clear that the greatest caution is necessary against the abuses of missionary propagandism. The almost absolute failure of missions in recent centuries, and more especially in the nineteenth, is intimately associated with the greater political importance which the improved facilities of travel and intercourse have conferred upon them. Everyone has heard how Catholicism was persecuted in Japan, till at last the very profession of Christianity was made a capital crime in that part of the world. But a traveller, who knew the East intimately at the time, explains how it was that the Jesuits’ labours resulted so disastrously. On the outbreak of civil dissensions in Japan, ‘the Christian priests thought it a proper time for them to settle their religion on the same foundation that Mahomet did his, by establishing it in blood. Their thoughts ran on nothing less than extirpating the heathen out of the land, and they framed a conspiracy of raising an army of 50,000 Christians to murder their countrymen, that so the whole island might be illuminated by Christianity such as it was then.’[240] And in the same way, a modern writer, speaking of the very limited success of missions in India, has
- 42. asserted frankly that ‘in despair many Christians in India are driven to wish and pray that some one, or some way, may arise for converting the Indians by the sword.’[241] Nor are the heathen themselves blind to the political dangers which are involved in the presence of missionaries among them. All over the world conversion is from the native point of view the same thing as disaffection, and war is dreaded as the certain consequence of the adoption of Christianity. The French bishop, Lefebvre, when asked by the mandarins of Cochin China, in 1847, the purpose of his visit, said that he read in their faces that they suspected him ‘of having come to excite some outbreak among the neophytes, and perhaps prepare the way for an European army;’ and the king was ‘afraid to see Christians multiply in his kingdom, and in case of war with European Powers, combine with his enemies.’[242] How right events have proved him to have been! The story is the same in Africa. ‘Not long after I entered the country,’ said the missionary, Mr. Calderwood, of Caffraria, ‘a leading chief once said to me, “When my people become Christians, they cease to be my people.”’[243] The Norwegian missionaries were for twenty years in Zululand without making any converts but a few destitute children, many of whom had been given to them out of pity by the chiefs,[244] and their failure was actually ascribed by the Zulu king to their having taught the incompatibility of Christianity with allegiance to a heathen ruler.[245] In 1877, a Zulu of authority expressed the prevalent native reasoning on this point in language which supplies the key to disappointments that extend much further than Zululand: ‘We will not allow the Zulus to become so-called Christians. It is not the king says so, but every man in Zululand. If a Zulu does anything wrong, he at once goes to a mission-station, and says he wants to become a Christian; if he wants to run away with a girl, he becomes a Christian; if he wishes to be exempt from serving the king, he puts on clothes, and is a Christian; if a man is an umtagati (evil-doer), he becomes a Christian.’[246]
- 43. It is on this account that in wars with savage nations the destruction of mission-stations has always been so constant an episode. Nor can we wonder at this when we recollect that in the Caffre war of 1851, for instance, it was a subject of boast with the missionaries that it was Caffres trained on the mission-stations who had preserved the English posts along the frontiers, carried the English despatches, and fought against their own countrymen for the preservation and defence of the colony.[247] It is rather a poor result of all the money and labour that has been spent in the attempt to Christianise South Africa, that the Wesleyan mission-station at Edendale should have contributed an efficient force of cavalry to fight against their countrymen in the Zulu campaign; and we may hesitate whether most to despise the missionaries who count such a result as a triumph of their efforts, or the converts whom they reward with tea and cake for military service with the enemies of their countrymen. [248] It needs no great strain of intelligence to perceive that this use of mission-stations as military training-schools scarcely tends to enhance the advantages of conversion in the minds of the heathen among whom they are planted. For these reasons, and because it is becoming daily more apparent that wars are less a necessary evil than an optional misery of human life, the principal measure for a country which would fain improve, and live at peace with, the less civilised races which touch the numerous borders of its empire, would be the legal restraint or prevention of missionary enterprise: a proposal that will appear less startling if we reflect that in no quarter of the globe can that method of civilising barbarism point to more than local or ephemeral success. The Protestant missions of this century are in process of failure, as fatal and decided as that which befel the Catholic missions of the French, Portuguese, or Spanish, in the sixteenth and seventeenth centuries, and very much from the same causes. The English wars in South Africa, with which the Protestant missionaries have been so closely connected, have frustrated all attempts to Christianise that
- 44. region, just as ‘the fearful wars occasioned directly or indirectly by the missionaries’ sent by the Portuguese to the kingdoms of Congo and Angola in the sixteenth century rendered futile similar attempts on the West Coast.[249] The same process of depopulation under Protestant influences may now be observed in the Sandwich Islands or New Zealand that reduced the population of Hispaniola, under Spanish Christianity, from a million to 14,000 in a quarter of a century.[250] No Protestant missionary ever laboured with more zeal than Eliot did in America in the seventeenth century, but the tribes he taught have long since been extinct: ‘like one of their own forest trees, they have withered from core to bark;’[251] and, in short, the history of both Catholic and Protestant missions alike may be summed up in this one general statement: either they have failed altogether of results on a sufficient scale to be worthy of notice, or the impartial page of history unfolds to us one uniform tale of civil war, persecution, conquest, and extirpation in whatever regions they can boast of more at least of the semblance of success. Another measure in the interests of peace would be the organisation of a class of well-paid officials whose duty it should be to examine on the spot into the truth of all rumours of outrages or atrocities which are circulated from time to time, in order to set the tide of public opinion in favour of hostile measures. Such rumours may, of course, have some foundation, but in nine cases out of ten they are false. So lately as the year 1882, the Times and other English papers were so far deceived as to give their readers a horrible account of the sacrifice of 200 young girls to the spirits of the dead in Ashantee; and people were beginning to ask themselves whether such things could be suffered within reach of an English army, when it was happily discovered that the whole story was fictitious. Stories of this sort are what the Germans call Tendenzlügen, or lies invented to produce a certain effect. Their effect in rousing the war-spirit is undeniable; and, although the healthy scepticism which has of recent years been born of experience affords us some protection, no
- 45. expenditure could be more economical than one which should aim at rendering them powerless by neutralising them at the fountain-head. In the preceding historical survey of the relations in war between communities standing on different levels of civilisation, the allusion, among some of the rudest tribes, to laws of war very similar to those supposed to be binding between more polished nations tends to discredit the distinction between civilised and barbarian warfare. The progress of knowledge threatens the overthrow of the distinction, just as it has already reduced that between organic and inorganic matter, or between animal and vegetable life, to a distinction founded rather on human thought than on the nature of things. And it is probable that the more the military side of savage life is studied, the fewer will be found to be the lines of demarcation which are thought to establish a difference in kind in the conduct of war by belligerents in different stages of progress. The difference in this respect is chiefly one of weapons, of strategy, and of tactics; and it would seem that whatever superiority the more civilised community may claim in its rules of war is more than compensated in savage life both by the less frequent occurrence of wars and by their far less fatal character. But, however much the frequency and ferocity of the wars waged by barbarian races as compared with those waged by civilised nations has been exaggerated, there is no doubt but that in warfare, more than in anything else, there is most in common between civilisation and savagery, and that the distinction between them most nearly disappears. In art and knowledge and religion the distinction between the two is so wide that the evolution of one from the other seems still to many minds incredible; but in war, and the thoughts which relate to it, the points of analogy cannot fail to strike the most indifferent. We see still in either condition, the same notions of the glory of fighting, the same belief in war as the only source of strength and honour, the same hope from it of personal advancement, the same readiness to seize any pretext for resorting to it, the same foolish sentiment that it is mean to live without it.
- 46. Then only will the distinction between the two be final, complete, and real, when all fighting is relegated to barbarism, and regarded as unworthy of civilised humanity; when the enlightenment of opinion, which has freed us already from such curses as slavery, the torture-chamber, or duelling, shall demand instinctively the settlement of all causes of quarrel by peaceful arbitration, and leave to the lower races and the lower creation the old-fashioned resort to a trial of violence and might, to competition in fraud and ferocity.
- 47. CHAPTER VII. WAR AND CHRISTIANITY. Etsi adierant milites ad Joannem et formam observationis acceperant, si etiam centurio crediderat, omnem postea militem Dominus in Petro exarmando discinxit.—Tertullian. The war question at the time of the Reformation—The remonstrances of Erasmus against the custom—Influence of Grotius on the side of war—The war question in the early Church—The Fathers against the lawfulness of war—Causes of the changed views of the Church—The clergy as active combatants for over one thousand years—Fighting Bishops— Bravery in war and ecclesiastical preferment—Pope Julius II. at the siege of Mirandola—The last fighting Bishop—Origin and meaning of the declaration of war—Superstition in the naming of weapons, ships, c.—The custom of kissing the earth before a charge—Connection between religious and military ideas—The Church as a pacific agency—Her efforts to set limits to reprisals —The altered attitude of the modern Church—Early reformers only sanctioned just wars—Voltaire’s reproach against the Church—Canon Mozley’s sermon on war—The answer to his apology. Whether military service was lawful for a Christian at all was at the time of the Reformation one of the most keenly debated questions; and considering the force of opinion arrayed on the negative side, its ultimate decision in the affirmative is a matter of more wonder than is generally given to it. Sir Thomas More charges Luther and his disciples with carrying the doctrines of peace to the extreme limits of non-resistance; and the views on this subject of the Mennonites and
- 48. Quakers were but what at one time seemed not unlikely to have been those of the Reformed Church generally. By far the foremost champion on the negative side was Erasmus, who being at Rome at the time when the League of Cambray, under the auspices of Julius II., was meditating war against the Republic of Venice, wrote a book to the Pope, entitled ‘Antipolemus,’ which, though never completed, probably exists in part in his tract known under the title of ‘Dulce Bellum inexpertis,’ and printed among his ‘Adagia.’ In it he complained, as one might complain still, that the custom of war was so recognised as an incident of life that men wondered there should be any to whom it was displeasing; and likewise so approved of generally, that to find any fault with it savoured not only of impiety, but of actual heresy. To speak of it, therefore, as he did in the following passage, required some courage: ‘If there be anything in the affairs of mortals which it is the interest of men not only to attack, but which ought by every possible means to be avoided, condemned, and abolished, it is of all things war, than which nothing is more impious, more calamitous, more widely pernicious, more inveterate, more base, or in sum more unworthy of a man, not to say of a Christian.’ In a letter to Francis I. on the same subject, he noticed as an astonishing fact, that out of such a multitude of abbots, bishops, archbishops, and cardinals as existed in the world, not one of them should step forward to do what he could, even at the risk of his life, to put an end to so deplorable a practice. The failure of this view of the custom of war, which is in its essence more opposed to Christianity than the custom of selling men for slaves or sacrificing them to idols, to take any root in men’s minds, is a misfortune on which the whole history of Europe since Erasmus forms a sufficient commentary. That failure is partly due to the unlucky accident which led Grotius in this matter to throw all his weight into the opposite scale. For this famous jurist, entering at much length into the question of the compatibility of war with the profession of Christianity (thereby proving the importance which in
- 49. his day still attached to it), came to conclusions in favour of the received opinion, which are curiously characteristic both of the writer and his time. His general argument was, that if a sovereign was justified in putting his own subjects to death for crimes, much more was he justified in using the sword against people who were not his subjects, but strangers to him. And this absurd argument was enforced by considerations as feeble as the following: that laws of war were laid down in the Book of Deuteronomy; that John the Baptist did not bid the soldiers, who consulted him, to forsake their calling, but to abstain from extortion and be content with their wages; that Cornelius the centurion, whom St. Peter baptized, neither gave up his military life, nor was exhorted by the apostle to do so; that the Emperor Constantine had many Christians in his armies, and the name of Christ inscribed upon his banners; and that the military oath after his time was taken in the name of the Three Persons of the Trinity. One single reflection will suffice to display the utter shallowness of this reasoning, which was after all only borrowed from St. Augustine. For if Biblical texts are a justification of war, they are clearly a justification of slavery; whilst, on the other hand, the general spirit of the Christian religion, to say nothing of several positive passages, is at least equally opposed to one custom as to the other. If then the abolition of slavery is one of the services for which Christianity as an influence in history claims a large share of the credit, its failure to abolish the other custom must in fairness be set against it; for it were easier to defend slave-holding out of the language of the New Testament than to defend military service, far more being actually said there to inculcate the duty of peace than to inculcate the principles of social equality: and the same may be said of the writings of the Fathers. The different attitude of the Church towards these two customs in modern times, her vehement condemnation of the one, and her tolerance or encouragement of the other, appears all the more surprising when we remember that in the early centuries of our era
- 50. her attitude was exactly the reverse, and that, whilst slavery was permitted, the unlawfulness of war was denounced with no uncertain or wavering voice. When Tertullian wrote his treatise ‘De Corona’ (201) concerning the right of Christian soldiers to wear laurel crowns, he used words on this subject which, even if at variance with some of his statements made in his ‘Apology’ thirty years earlier, may be taken to express his maturer judgment. ‘Shall the son of peace’ (that is, a Christian), he asks, ‘act in battle when it will not befit him even to go to law? Shall he administer bonds and imprisonments and tortures and punishments who may not avenge even his own injuries?... The very transference of his enrolment from the army of light to that of darkness is sin.’ And again: ‘What if the soldiers did go to John and receive the rule of their service, and what if the Centurion did believe; the Lord by his disarming of Peter disarmed every soldier from that time forward.’ Tertullian made an exception in favour of soldiers whose conversion was subsequent to their enrolment (as was implied in discussing their duty with regard to the laurel- wreath), though insisting even in their case that they ought either to leave the service, as many did, or to refuse participation in its acts, which were inconsistent with their Christian profession. So that at that time Christian opinion was clearly not only averse to a military life being entered upon after baptism (of which there are no instances on record), but in favour of its being forsaken, if the enrolment preceded the baptism. The Christians who served in the armies of Rome were not men who were converts or Christians at the time of enrolling, but men who remained with the colours after their conversion. If it is certain that some Christians remained in the army, it appears equally certain that no Christian at that time thought of entering it. This seems the best solution of the much-debated question, to what extent Christians served at all in the early centuries. Irenæus speaks of the Christians in the second century as not knowing how to fight, and Justin Martyr, his contemporary, considered Isaiah’s prophecy
- 51. about the swords being turned into ploughshares as in part fulfilled, because his co-religionists, who in times past had killed one another, did not then know how to fight even with their enemies. The charge made by Celsus against the Christians, that they refused to bear arms even in case of necessity, was admitted by Origen, but justified on the ground of the unlawfulness of war. ‘We indeed,’ he says, ‘fight in a special way on the king’s behalf, but we do not go on campaigns with him, even should he press us to do so; we do battle on his behalf as a peculiar army of piety, prevailing by our prayers to God for him.’ And again: ‘We no longer take up the sword against people, nor learn to make war any more, having become through Jesus, who is our general, sons of peace.’ Nothing could be clearer nor more conclusive than this language; and the same attitude towards war was expressed or implied by the following Fathers in chronological order: Justin Martyr, Tatian, Clement of Alexandria, Tertullian, Cyprian, Lactantius, Archelaus, Ambrose, Chrysostom, Jerome, and Cyril. Eusebius says that many Christians in the third century laid aside the military life rather than abjure their religion. Of 10,050 pagan inscriptions that have been collected, 545 were found to belong to pagan soldiers, while of 4,734 Christian inscriptions of the same period, only 27 were those of soldiers; from which it seems rather absurd to infer, as a French writer has inferred, not that there was a great disproportion of Christian to pagan soldiers in the imperial armies, but that most Christian soldiers being soldiers of Christ did not like to have it recorded on their epitaphs that they had been in the service of any man.[252] On the other hand, there were certainly always some Christians who remained in the ranks after their conversion, in spite of the military oath in the names of the pagan deities and the quasi-worship of the standards which constituted some part of the early Christian antipathy to war. This is implied in the remarks of Tertullian, and stands in no need of the support of such legends as the Thundering Legion of Christians, whose prayers obtained rain, or of the Theban legion of 6,000 Christians martyred under Maximian. It was left as a matter of individual conscience. In the story of the martyr
- 52. Maximilian, when Dion the proconsul reminded him that there were Christian soldiers among the life-guards of the Emperors, the former replied, ‘They know what is best for them to do; but I am a Christian and cannot fight.’ Marcellus, the converted centurion, threw down his belt at the head of his legion, and suffered death rather than continue in the service; and the annals of the early Church abound in similar martyrdoms. Nor can there be much doubt but that a love of peace and dislike of bloodshed were the principal causes of this early Christian attitude towards the military profession, and that the idolatry and other pagan rites connected with it only acted as minor and secondary deterrents. Thus, in the Greek Church St. Basil would have excluded from communion for three years any one who had shed an enemy’s blood; and a similar feeling explains Theodosius’ refusal to partake of the Eucharist after his great victory over Eugenius. The canons of the Church excluded from ordination all who had served in an army after baptism; and in the fifth century Innocent I. blamed the Spanish churches for their laxity in admitting such persons into holy orders.[253] The anti-military tendency of opinion in the early period of Christianity appears therefore indisputable, and Tertullian would probably have smiled at the prophet who should have predicted that Christians would have ceased to keep slaves long before they should have ceased to commit murder and robbery under the fiction of hostilities. But it proves the strength of the original impetus, that Ulphilas, the first apostle to the Goths, should purposely, in his translation of the Scriptures, have omitted the Books of Kings, as too stimulative of a love of war. How utterly in this matter Christianity came to forsake its earlier ideal is known to all. This resulted partly from the frequent use of the sword for the purpose of conversion, and partly from the rise of the Mahometan power, which made wars with the infidel appear in the light of acts of faith, and changed the whole of Christendom into a kind of vast standing military order. But it resulted still more from that compromise effected in the fourth century between paganism
- 53. Welcome to our website – the perfect destination for book lovers and knowledge seekers. We believe that every book holds a new world, offering opportunities for learning, discovery, and personal growth. That’s why we are dedicated to bringing you a diverse collection of books, ranging from classic literature and specialized publications to self-development guides and children's books. More than just a book-buying platform, we strive to be a bridge connecting you with timeless cultural and intellectual values. With an elegant, user-friendly interface and a smart search system, you can quickly find the books that best suit your interests. Additionally, our special promotions and home delivery services help you save time and fully enjoy the joy of reading. Join us on a journey of knowledge exploration, passion nurturing, and personal growth every day! ebookbell.com