truth becomes undeserving of its name." "These people are no less
light in mind than in body, and by no means to be relied on. They are
easily urged to undertake any action, and as easily checked from
prosecuting it.... They never scruple at taking a false oath for the
sake of any temporary advantage.... Above all other peoples they are
given to removing their neighbours' landmarks. Hence arise quarrels,
murders, conflagrations, and frequent fratricides. It is remarkable
that brothers show more affection to each other when dead than when
living; for they persecute the living even unto death, but avenge the
dead with all their power."
Finally, as a scientific observer of politics, he discusses how Wales
may be conquered and governed, and how the Welsh may resist.
A prince who would subdue this people must give his whole energies to
the task for at least a whole year. He must divide their strength, and
by bribes and promises endeavour to stir up one against the other,
knowing the spirit of hatred and envy which generally prevails among
them. He must cut off supplies, build castles, and use light-armed
troops and plenty of them; for though many English mercenaries perish
in a battle, money will procure as many more; but to the Welsh the
loss is for the time irreparable. He recommends that all the English
inhabitants of the Marches should be trained to arms; for the Welsh
fight for liberty and only a free people can subdue them. His advice
to the Welsh is: Unite. "If they would be inseparable, they would be
insuperable, being assisted by these three circumstances--a country
well defended by nature, a people contented to live upon little, a
community whose nobles and commoners alike are trained in the use of
arms; and especially as the English fight for power, the Welsh for
liberty; the English hirelings for money, the Welsh patriots for their
country."
I hope I may persuade some who do not yet know Gerald to make his
acquaintance, and to read either his works on Ireland and Wales,
translated in Bohn's library, or Mr. Henry Owen's brilliant and
delightful volume, "Gerald the Welshman," my indebtedness to which I
wish to acknowledge. Gerald tells us many miracles; but he has himself
performed a miracle as wonderful as any he relates; he has kept all
the charm and freshness of youth for more than seven hundred years.
[Illustration: CASTLES & RELIGIOUS HOUSES. (12th & 13th Centuries)]
IV
CASTLES
Wales is pre-eminently the land of castles. There are between thirty
and forty in Glamorgan alone. The accompanying map, though it is by no
means exhaustive, shows the general lie of the castles, which may be
divided into three groups, having as their respective bases Chester,
Shrewsbury, and Gloucester. But though there is some evidence of an
organised plan for the conquest of Wales in the time of William Rufus,
it is useless to look for any great and general system of offence or
defence, because most of the castles were not built by a centralised
government with any such object in view, but by individuals to guard
their own territories and protect their independence against either
their neighbours or the English king. The great age of castle-building
was between 1100 and 1300. Castles play a very small part in the
fighting in Wales till the end of the eleventh century. Before that
time indeed there were few stone castles anywhere; the usual type, even
of the early Norman castles, was a moated mound surrounded by wooden
palisades. One hears for instance of a castle being built by William
the Conqueror in eight days. An example of this early type of fortress
was Pembroke Castle at the end of the eleventh century, "a slender
fortress of stakes and turf," which had the good fortune to be in
charge of Gerald of Windsor, grandfather of Giraldus Cambrensis. It
stood several sieges, which shows that the siege engines of the Welsh
were of a very poor and primitive type. One of these sieges was turned
into a blockade, and the garrison was nearly reduced by starvation. The
constable had recourse to a time-honoured ruse. "With great prudence he
caused four hogs which still remained to be cut into small pieces and
thrown down among the enemy. The next day he had recourse to a more
refined stratagem: he contrived that a letter from him should fall
into the hands of the enemy stating that there was no need for
assistance for the next four months." The besiegers were taken in and
dispersed to their homes.
The characteristic types of castles in the twelfth century were the
rectangular keep and the shell keep; in the thirteenth the concentric
castle. Of the two last we have splendid examples in Cardiff and
Caerphilly. Of rectangular keeps there are very few in Wales--Chepstow
is the only important one--though there are several on the borders,
notably Ludlow. The square keep seems to us most characteristic of
Norman military architecture; the Tower of London, Rochester,
Newcastle, Castle Rising, are well-known examples, and there are many
more in a good state of preservation; there are many more solid square
keeps than shell keeps well preserved, but this is simply due to the
greater solidity of the former; the shell keeps were far more numerous
in the twelfth century; and the reasons for this are obvious--the
rectangular keep was much more expensive to build, and it was too
heavy to erect on the artificial mounds on which the Norman
architects generally founded their castles.
The keep of Cardiff Castle is one of the most perfect shell keeps in
existence. It is built on a round artificial mound, surrounded by a
wide and deep moat--the mound and moat being, of course, complements
of each other. Such mounds and moats are common in all parts of
England, and in Normandy. They are not Roman, nor British, nor are
they, as Mr. G. T. Clark maintained, characteristic of Anglo-Saxon
work. They are essentially Norman, and a good representation of the
making of such a mound may be seen in the Bayeux Tapestry, under the
heading--'He orders them to dig a castle.' When was the Cardiff mound
made? Perhaps the short entry in the Brut gives the answer: "1080, the
building of Cardiff began." It would then be surrounded by wooden
palisades, and surmounted by a timber structure, as a newly made mound
would not stand the masonry. The shell keep was probably built by
Robert of Gloucester, and it was probably in the gate-house of this
keep, that Robert of Normandy was imprisoned. A shell keep was a ring
wall eight or ten feet thick, about thirty feet high, not covered in,
and enclosing an open courtyard, round which were placed the
buildings--light structures, often wooden sheds, abutting on the ring
wall--such as one may see now in the courtyard of Castell Coch. The
shell keep was the centre of Robert's castle, but not the whole. From
this time dated the great outer walls on the south and west--walls
forty feet high and ten feet thick and solid throughout. The north and
east and part of the south sides of the castle precincts are enclosed
by banks of earth, beneath which, the walls of a Roman camp have
recently been discovered. These banks were capped by a slight
embattled wall. Outside along the north, south and east fronts was a
moat, formerly fed by the Taff through the Mill leat stream which ran
along the west front. The present lodgings, or habitable part of the
castle built on either side of the great west wall, date mostly from
the fifteenth century. The earlier lodgings were, perhaps, on the same
site--though only inside the wall; a great lord did not as a rule
live in the keep, except in times of danger.
The area of the enclosure is about ten acres--more suited to a Roman
garrison than to a lord marcher of the twelfth century. That the
castle was difficult to guard is shown by the success of Ivor Bach's
bold dash, _c._ 1153-1158. Ivor ap Meyric was Lord of Senghenydd,
holding it of William of Gloucester, the Lord of Glamorgan, and,
perhaps, had his headquarters in the fortress above the present
Castell Coch. "He was," says Giraldus Cambrensis, "after the manner of
the Welsh, owner of a tract of mountain land, of which the earl was
trying to deprive him. At that time the Castle of Cardiff was
surrounded with high walls, guarded by 120 men at arms, a numerous
body of archers and a strong watch. Yet in defiance of all this, Ivor,
in the dead of night secretly scaled the walls, seized the earl and
countess and their only son, and carried them off to the woods; and
did not release them till he had recovered all that had been unjustly
taken from him," and a goodly ransom in addition. Perhaps the most
permanent result of this episode was the building of a wall 30 feet
high between the keep and the Black Tower--dividing the castle
enclosure into two parts and forming an inner or middle ward of less
extent, and less liable to danger from such sudden raids.
Cardiff Castle was much more than a place of defence; it was the seat
of government. The bailiff of the Castle was _ex officio_ mayor of the
town in the Middle Ages. The Castle was also the head and centre of
the Lordship of Glamorgan. This was divided into two parts--the shire
fee or body, and the members. The shire fee was the southern part;
under a sheriff appointed by the chief Lord: the chief landowners owed
suit and service--_i.e._, they attended and were under the
jurisdiction of the shire court held monthly in the castle enclosure,
and each owed a fixed amount of military service--especially the duty
of "castle-guard"--supplying the garrison and keeping the castle in
repair. There are indications of the work of the shire court in some
of the castle accounts published in the Cardiff Records, _e.g._, in
1316, an official accounts for 1d., the price of "a cord bought for
the hanging of thieves adjudged in the county court: stipend of one
man hanging those thieves 4d." The "members" consisted of ten
lordships (several of which were in the hands of Welsh nobles): these
were much more independent; each had its own court (with powers of
life and death), from which an appeal lay to the Lord's court at
Cardiff: generally they owed no definite service to the Lord (except
homage, and sometimes a heriot at death), but on failure of heirs the
estate lapsed to the chief Lord. At Cardiff Castle the Lord had his
chancery, like the royal chancery on a small scale--issuing writs,
recording services and grants of privileges, and legal decisions:
practically the whole of these records have been lost--and our
knowledge of the organisation of the Lordship is mainly derived from
the royal records at times, when owing to minority or escheat, the
Lordship was under royal administration. The Lord of Glamorgan owed
homage, but no service to the king; and (though this was sometimes
disputed by his tenants and the royal lawyers), no appeal lay from his
courts to the king's court. The machinery of government was probably
more complete and elaborate in Glamorgan than in any other Marcher
Lordship.
Caerphilly Castle had not the political importance of Cardiff, but far
surpasses it as a fortress. By the strength and position of
Caerphilly, one may measure the power of Llywelyn ap Gruffydd after
the Barons' War and before the accession of Edward I. The Prince of
Wales had extended his sway down as far as Brecon, and Welshmen
everywhere were looking to him as the restorer of their country's
independence. Among them was the Welsh Lord of Senghenydd, one of the
chief "members" of Glamorgan, and his overlord probably saw reason to
suspect his loyalty. An alliance between him and Llywelyn would open
the lower Taff Valley to the Welsh prince and give him command of the
hill country north of Cardiff. It was on the lands of the lord of
Senghenydd that Gilbert de Clare, Earl of Gloucester, built Castell
Coch and Caerphilly.
[Illustration: CARDIFF CASTLE. (12th Century)]
[Illustration: CAERPHILLY CASTLE. (13th Century)]
Caerphilly is described by the latest historian of the Art of War as
the grandest specimen of its class; it represents the high-water mark
of mediaeval military architecture in this country, and was the model
of Edward I.'s great castles in the north. It illustrates the
influence of the Crusades on Western Europe, being an instance of the
"concentric" system of defences, of which the walls of Constantinople
afford the most magnificent example, and which the Crusaders adopted
in many of their great fortresses in the East.
Caerphilly Castle consists of three lines of defences, and the way in
which these supplement each other shows that the work in all
essentials was designed as a great whole; it did not grow up bit by
bit. There are of course many evidences of alterations and rebuilding
at later times; the buildings in the middle ward, on the south side,
seem to be later additions; the hall appears to have been enlarged,
and the tracery of the windows suggests the fourteenth century; the
state-rooms to the west of the hall have been much altered; but such
alterations as appear are confined to the habitable part of the
castle, and do not affect it as a military work. It has been suggested
that the castle may have been greatly enlarged in the latter years of
Edward II., when it played an important part in connection with the
division of the Gloucester inheritance and the younger Despenser's
ambitions. There are a number of notices of the castle in the
chronicles and public records of that time, but apparently no
references to any building operations. And the unity of plan is
evidence that the whole dated from the same time.
The castle is built on a tongue of gravel nearly surrounded by low,
marshy land, forming a sort of peninsula; a stream on the south
running eastwards to the Rhymny; and two springs on the north. By
damming these waters and cutting through the tongue of gravel an
artificial island was secured for the site of the castle. The inner
ward, or central part of the castle, consists of a quadrangle with a
large round tower at each corner: in the centre of the east and west
side are massive gate-houses defended by portcullises; from the
projecting corner towers all the intervening wall was commanded. The
gateways communicate with the second line of defence or middle ward.
This completely encircles the inner ward, on a much lower level; it is
a narrow space bounded by a wall, with low, semi-circular bastions at
the corners; it is commanded at every point from the inner ward; the
narrowness of the space would prevent the concentration of large
bodies of assailants or the use of battering-rams, and communication
is at several points stopped by walls or buildings jutting out from
the inner ward. The middle ward had strong gate-houses at the east and
west ends, and was completely surrounded by water--east and west by a
moat, north and south the moat widens into lakes: note how on the
north a narrow ridge of gravel has been used to ensure a water moat on
that side, in case there was not enough water to flood the whole lake.
These lakes form part of the third line of defence or outer ward,
which includes also on the west the "horn-work" and on the east the
grand front. The horn-work is about three acres in extent, surrounded
by a wall 15 feet high, which is of the nature of an escarpment, the
ground rising above it. It is entirely surrounded by a moat, and
connected with the middle ward on one side and the mainland on the
other by drawbridges. It would probably be used for grazing purposes,
and thus would be of great value to the garrison; but so far as the
actual defences of the castle are concerned, a lake would have been
much more effective; the nature of the ground would however have
prevented this. The horn-work was intended to cover the only side upon
which the castle was open to an attack from level ground, and to
occupy what would otherwise have been a dangerous platform.
The eastern side of the outer ward--the grand front--is a most
imposing structure. It is a wall about 250 yards long, and in some
parts 60 feet high, furnished with buttresses and projecting towers
from which the intervening spaces are easily commanded, culminating
in the great gate-house near the centre, and terminating at both ends
in clusters of towers which protect the sally-ports. On the outside is
a moat spanned by a double drawbridge. The northern part of this
front, which was probably occupied by stables, would in dry weather be
the least defensible part of the castle; but it was cut off from the
rest by an embattled wall running from the gate-house to the inner
moat and pierced only by one small and portcullised gate. The southern
half was more important and stronger. It crossed the stream at the
dam, the walls being 15 feet thick where subjected to the pressure of
the water, and the strong group of towers at the end--on the other
side of the stream--guarded the dam on which the safety of the castle
largely depended; the wall and towers here form a semicircle, curving
back into the edge of the lake, so as to avoid the danger of being
outflanked.
On the inside of the grand front were various buildings, such as the
mill. This eastern line was divided from the middle ward by a moat 45
feet wide--a space which is too wide to be spanned by a single
drawbridge, and as there are no signs of the foundations of a central
pier, it seems probable that the bridge rested on a wooden support,
which could be removed when necessary, and the assailants plunged into
the moat below.
There are a large number of interesting details connected with both
the military functions of the castle and its domestic economy. There
were at least four exits (not counting the two water-gates); this
would give the garrison opportunities of harassing assailants by
sallies, and would make a much larger army necessary in order to
blockade the castle; contrast the single narrow entrance to the Norman
keep--high up in the wall and visible to all outside. The water-gates
are worth studying, especially the methods of protecting the eastern
water-gate--two grates with a shoot above and between them. One should
notice, too, the "splaying" of the outer wall, by which missiles from
the top would be projected outwards; and also the use of the
mill-stream to carry away the refuse of the garderobe tower. And there
are many other points, to which one would like to call attention, if
time allowed.
The history of Caerphilly in the Middle Ages need not detain us long.
It was besieged by Llywelyn in 1271, while it was being built.
Llywelyn declared he could have taken it in three days if he had not
been persuaded to submit the dispute to the arbitration of the king.
It is clear that the castle was not finished; shortly after this
Gilbert de Clare obtained license from the king to "enditch" the
castle: such license was not, as a rule, required in the Marches (as
it was in England) and was only necessary now because the king was
acting as arbitrator. The Earl of Gloucester kept possession. We next
hear of it in 1315, when it resisted the attack of Llywelyn Bren. It
was then in the hands of the king, pending the division of the
Gloucester inheritance among the three co-heiresses. In 1318
Caerphilly, with the rest of Glamorgan, was granted to the younger
Despenser, who perhaps enlarged the hall and made the other
alterations referred to above. Edward II. was there for a few days
when flying for his life; had he trusted to Caerphilly, instead of
fleeing further through South Wales, he might have saved his head and
his crown; at any rate, there would have been a great siege to add to
the history of mediaeval warfare. The king's adherents held out in
Caerphilly for months, and only surrendered when, the king being dead,
there was nothing more to fight for, and they were allowed to go free.
Happy is the castle which has no history. The perfection of Caerphilly
as a fortress saved it from serious attacks.
In conclusion, I will give two illustrations of the relations between
the garrison of a castle and those outside. The first refers to
Swansea. There is a curious Charter of King John to the good men of
Swansea, in which he releases them from the "custom of eating" forced
on them by the men of the castle. This would be a solid variation of
the liquid scot-ales or free drinks which officials and garrisons were
in the habit of exacting from their neighbours, and which were among
the most persistent grievances in the Middle Ages.
The second concerns Builth, and is taken from the Patent Rolls of
Edward II. in 1315. Builth was then in the hands of the king, to whom
the townsfolk appeal for redress of grievances. The community complain
that, though they are only bound to carry timber to the castle twice a
week, they are often forced to carry it three times a week and more,
and victuals too; and the men of the castle compel them to plough
their lands and cut their corn, and hold them to ransom if they
refuse; and they carry away from the houses of the said complainants
divers kind of victuals--lambs, geese, hens, &c.--and pay only one
quarter of their value, or nothing at all; and though the complainants
gave the keeper of the castle L120 that they might be free from such
oppressions, he took the money and oppresses them just the same.
Further, the courts which the people have to attend are multiplied;
and recently the court was held at a time when so great a flood had
happened that neither horsemen nor footmen could approach the court,
and so thirty-six men and women, fearing the cruelty of the bailiffs,
entered a boat and were overwhelmed in the rush of the river. And one
night men of the castle, maliciously seeking occasion against the
commonalty of the town, went out of the castle and pretended to
besiege it and shot arrows at it; and then secretly re-entered the
castle and declared the townsfolk had been attacking the castle. And
on this account many burgesses were imprisoned in the castle and
ill-treated, and their swine maliciously killed. And things are so
intolerable that many of the greater burgesses have left the country,
and the residue, without speedy remedy, cannot remain.
Life was evidently dull in a castle: one had to play practical jokes
to relieve the monotony; and life was anything but pleasant outside a
castle. The castles of Wales are much more attractive to us to-day
than they were to those who lived in them or round them six or seven
hundred years ago.
V
RELIGIOUS HOUSES
In speaking of the Religious Houses in Wales I shall deal with those
which flourished in the twelfth and thirteenth centuries--the period
we have hitherto been studying--though it is tempting to go back to
the glories of the old Welsh monasteries of the sixth century, such as
Llantwit Major and Bangor Iscoed, whose dim memories must always
exercise a strong fascination. The monasteries of this early type had
fallen on evil days in Wales, as in Ireland and elsewhere, before the
twelfth century, many had been wiped out by the Danes; and those that
remained seem to have lost the spirit of life (save in a few distant
islands or inaccessible mountains), and made no struggle for existence
against the vigorous invasion of the new monasticism.