Chapter VI: England and the Barons (2)

Oxford was, from its geographical situation, one of the most important strategical posts in England. It stood at the very centre and crowning-point of the valley of the Thames, the great high-way which led from the eastern sea and the capital into the western shires, through the very heart of the land. So long as it remained loyal to Stephen, he was master of the whole Thames valley, and the Angevins, however complete might be their triumph in the west, were cut off from all direct communication with eastern England and even with the capital itself. The surrender of Oxford castle to Matilda in the summer of 1141 had reversed this position of affairs. It probably helped to determine--it was at any rate soon followed by--the surrender of London; and even when London was again lost to the Empress, her possession of Oxford still gave her command over the upper part of the river-valley and thus secured her main line of communication with her brother’s territories in the west, while Stephen in his turn was almost prisoned in the eastern half of his realm. For nearly eleven months he had seen her defying him from her father’s palace of Beaumont or from the impregnable stronghold of the castle, where the first Robert of Oilly, not content with raising a shell-keep on the old English mound, had built another tall square tower which still stands, on the western side of the enclosure, directly above the river.[941] Not until her brother had left her did the king venture to take up the challenge which her very presence there implied; then indeed he felt that the hour had come. Matilda, as if in expectation of his attack, had been employing her followers on the construction of a chain of forts intended to protect and keep open her communications with the west.[942] One by one Stephen broke the links of the chain--Cirencester, Bampton, Ratcot[943]--and from this last place, a little village in the midst of a marsh, half-way between Bampton and Farringdon, he led his host across the Isis and round by the meadows on its southern shore to the ford below S. Frideswide’s from which the city took its name. Matilda’s partizans no sooner discovered his approach--three days before Michaelmas[944]--than they streamed down to the bank of the river, across which they greeted him first with a torrent of abuse and then with a flight of arrows. The vanguard of the royal host, with Stephen himself at their head, sprang into the water, swam rather than waded across the well-known and time-honoured ford,[945] and by the fury of their onset drove their insulting enemies back to the city gates. The rest of the army quickly followed; Matilda’s adherents fled through the open gate, their pursuers rushed in after them, entered the town without difficulty, set it on fire, captured and slew all on whom they could lay their hands, and drove the rest to take shelter in the castle with their Lady.[946]

[941] See above, p. 42, note 2{113}.

[942] _Gesta Steph._ (Sewell), pp. 87, 88.

[943] _Ib._ p. 88. “Apud viculum Ratrotam fluctibus inaccesse
et paludibus obseptum.” _Ib._ p. 87. _Ratcot_ is Anthony Wood’s
rendering.

[944] Will. Malm. _Hist. Nov._, l. iii. c. 71 (Hardy, p. 766).

[945] “Præmonstrato antiquo sed eximiæ profunditatis vado.”
_Gesta Steph._ (Sewell), p. 89.

[946] _Ibid._

Stephen had doubtless not braved S. Frideswide’s wrath by entering Oxford, so to say, under her very eyes. His troops had won the city; his task was to win the castle, and that task he vowed never to abandon till both fortress and Empress should be in his hands. For nearly three months he blockaded the place, till its inhabitants were on the verge of starvation. The barons who had sworn to protect Matilda, bitterly ashamed of their failure, gathered at Wallingford ready to meet Stephen if he should chance to offer them battle; but he had no such intention, and they dared not attack him where he was.[947] At last a gleam of hope came with Earl Robert’s return, quickened, it seems, by tidings of his sister’s danger. Landing at Wareham with a force of some three or four hundred Normans, he regained the port and the village without difficulty, and as his force was too small to effect Matilda’s relief directly, he laid siege to the castle, hoping by this means to make a diversion in her favour.[948] The garrison of Wareham did in fact send a message to Stephen beseeching him to come and relieve them before a certain day, as if he did not, they must give up the place.[949] But the king was not to be drawn from his prey; he left Wareham to its fate, and after a three weeks’ siege it surrendered. Robert went on to Portland and Lulworth, took them both, and then summoned all the friends of the Empress to meet him at Cirencester, thence to set out with their united forces for the rescue of Matilda herself.[950] In Oxford castle the provisions were all but exhausted; the Lady despaired of succour.[951] Her faithful friend the lord of the castle, Robert of Oilly, had died a fortnight before the siege began.[952] Christmas was close at hand; the snow lay thick on the ground; the river was frozen fast. From the top of D’Oilly’s tall tower nothing was to be seen but one vast sheet of cold, dead white, broken only by the dark masses of Stephen’s host encamped round about upon the frozen meadows:--a dreary outlook, but the prospect within was drearier still. Matilda had gone through too many adventures to shrink from the risk of one more. One night four white-robed figures[953] dropped down by a rope[954] over the castle-wall upon the frozen river at its foot; they crossed dry-shod over the stream whose waters, a little lower down, had been almost over the heads of their enemies three months before; their footsteps fell noiseless upon the fresh snow, their white garments reflected its gleams and deceived the eyes of Stephen’s sentinels; in the stillness of the night, broken only by the bugle-call and the watchman’s cry, they stole through the besieging lines and across the very sleeping-quarters of the king--never caught, never discovered save by one man in all the host; and he, whether taking them for ghosts, or in chivalrous sympathy for their desperate venture, let them pass unchallenged and kept his story till the morrow.[955] Five miles they fled on foot “over snow and ice, over ditch and dale”; at Abingdon they took horse, and before the morning broke the Empress Matilda and her faithful comrades were safe under the protection of Brian Fitz-Count in his great fortress of Wallingford.[956]

[947] Will. Malm. as above.

[948] _Ib._ cc. 72, 73 (Hardy, pp. 767, 768). _Gesta Steph._
(Sewell), p. 91. Gerv. Cant. (Stubbs), vol. i. p. 124.

[949] Will. Malm. _Hist. Nov._, l. iii. c. 73 (Hardy, p. 768).

[950] _Ib._ c. 74 (p. 768). Gerv. Cant. (Stubbs), vol. i. pp.
124, 125.

[951] _Gesta Steph._ (Sewell), p. 90.

[952] Ann. Osen. a. 1142 (Luard, _Ann. Monast._, vol. iv. p.
24).

[953] _Gesta Steph._ as above. Gerv. Cant. (Stubbs), vol. i. p.
124, makes them six.

[954] Eng. Chron. a. 1140. Gerv. Cant. (as above) says “per
posticium.”

[955] _Gesta Steph._ as above.

[956] _Gesta Steph._ (Sewell), p. 90. Gerv. Cant. (Stubbs),
vol. i. pp. 124, 125. Hen. Hunt., l. viii. c. 20 (Arnold, p.
276). Will. Newb., l. i. c. 10 (Howlett, vol. i. p. 43).

At Wallingford her brother came to meet her, accompanied not by her husband but by her son, a child nine years old whom Geoffrey, now absorbed in the conquest of Normandy, had sent to England in his stead.[957] The escape from Oxford was Matilda’s last exploit. The castle surrendered to Stephen as soon as she had left it;[958] she returned to her old quarters at Bristol or Gloucester; and thenceforth she ceased to figure prominently in the war which dragged languidly on for five more years. A battle between Stephen and Earl Robert near Wilton, on July 1st, 1143, in which the king was utterly routed and only escaped being made prisoner a second time by taking to headlong flight,[959] was the last real success of the Angevin party. The year closed with a severe blow to the Empress, in the death of her trusted friend Miles of Hereford, who was slain on Christmas Eve, not in fight, but by a chance shot in hunting.[960] Early in the next year Ralf of Chester again seized Lincoln castle;[961] but Ralf fought for his own hand rather than for the Empress; and so, too, did Hugh Bigod, Turgis of Avranches and Geoffrey of Mandeville, who kept all eastern England in ceaseless commotion.[962] Stephen’s energies were absorbed in a vain endeavour to reduce them to order, while Robert struggled almost as vainly against the anarchy of the western shires; in the north Ralf of Chester now ruled supreme from the Witham to the Dee; and the upper valley of the Thames was at the mercy of William of Dover, who had built a castle at Cricklade, from which he ravaged the whole country between Oxford and Malmesbury.[963]

[957] Will. Malm. _Hist. Nov._, l. iii. c. 70 (Hardy, p. 765).
Rob. Torigni, a. 1142. Gerv. Cant. (as above), p. 125.

[958] Will. Malm. as above, c. 74 (p. 769. At this point he
ends). _Gesta Steph._ (Sewell), p. 91. Hen. Hunt. as above.

[959] _Gesta Steph._ (Sewell), p. 92. Gerv. Cant. (as above),
pp. 125, 126. Will. Newb. as above (p. 42).

[960] _Gesta Steph._ (Sewell), p. 101. Joh. Hexh. (Raine), p.
146.

[961] Hen. Hunt., l. viii. c. 22 (Arnold, p. 277).

[962] On Hugh Bigod and Turgis see _Gesta Steph._ (Sewell),
pp. 109–111; on Geoffrey of Mandeville, _ib._ pp. 101–104;
Will. Newb., l. i. c. 11 (Howlett, vol. i. pp. 44–46); and Hen.
Hunt., l. viii. c. 21 (Arnold, pp. 276, 277).

[963] _Gesta Steph._ (Sewell), pp. 106, 107, 111.

Suddenly, after capturing the commandant of Malmesbury and sending him as a great prize to the Empress, the lord of Cricklade threw aside his evil work and went off to die for a nobler cause in Palestine.[964] Geoffrey de Mandeville, the worst of all the troublers of the land, who had accepted titles and honours from both the rival sovereigns and had never for one moment been true to either, met his death in the same summer of 1144 in a skirmish with the king’s troops; his fellow-sinner Robert of Marmion was soon afterwards slain by the earl of Chester’s men at the gates of the abbey of Bath which he had desecrated.[965] For a moment it seemed as if the cry which had long been going up from all the desolated sanctuaries of England--“Up, Lord, why sleepest Thou?”--had been heard and answered at last.[966] Philip of Gloucester, Earl Robert’s son, who had taken William of Dover’s place at Cricklade, was so hard pressed by the garrison of Oxford[967] that he called his father to his aid; Robert built a great castle at Farringdon, but the king besieged it with such vigour that its defenders were compelled to surrender.[968] From that moment the Angevin party fell rapidly to pieces. Young Philip of Gloucester himself went over to Stephen and turned his arms against his own father.[969] The earl of Chester came to meet the king at Stamford,[970] humbly apologized for his rebellion, and sought to prove the sincerity of his repentance by regaining Bedford for Stephen, by constantly accompanying him with a band of three hundred picked knights, and by helping him to build a fortress at Crowmarsh to keep the garrison of Wallingford in check.[971] As, however, he still refused to give up the castles which he had seized and to pay his dues to the royal treasury, he was naturally regarded with suspicion by the other barons and by the king himself.[972] In the summer of 1146 their mutual distrust came to a crisis at Northampton. Ralf besought Stephen’s help against the Welsh; the barons persuaded Stephen to let them answer in his name that he would not give it unless Ralf surrendered his castles and gave hostages for his fidelity; he refused indignantly; they accused him of plotting treason, laid hands upon him with one accord, and gave him in charge to the royal guards, by whom he was flung into prison.[973] As in the case of the seizure of the bishops, it is difficult to say how far Stephen was responsible, and how much justification he had, for this arrest. We can hardly get nearer to the truth than the English chronicler: “The king took him in Hamton through wicked rede, and did him in prison; and soon after he let him out again through worse rede, with the precaution that he swore on the halidom and found hostages that he should give up all his castles; some he gave up and some gave he not, and did then worse than before.”[974] But among the castles which Ralf did give up for the sake of regaining his freedom was that which Stephen valued most--Lincoln.[975] Then at last the king felt that his enemies were at his feet; and he resolved that the city which had beheld his worst overthrow should also behold his highest triumph. In defiance of an old superstition which forbade any English king to appear in regal state within the walls of Lincoln, he kept his midwinter feast there with a splendour which had been unknown for years, and wore his crown at high mass in the minster on Christmas-day.[976]

[964] _Ib._ p. 111.

[965] Will. Newb., l. i. cc. 11, 12 (Howlett, vol. i. pp.
46–48). _Gesta Steph._ (Sewell), p. 104. Hen. Hunt., l. viii.
c. 22 (Arnold, p. 277).

[966] “Dicebaturque a laborantibus piis ‘Exsurge, quare
obdormis, Domine?’ At postquam ... ‘excitatus est,’ ut ait
propheta, ‘tanquam dormiens Dominus, et percussit inimicos Suos
in posteriora.’” Will. Newb., l. i. c. 11 (Howlett, vol. i.
p. 45). “Quia igitur improbi dixerunt Deum dormire, excitatus
est Deus.” Hen. Hunt., l. viii. c. 22 (Arnold, p. 227)--two
different interpretations of the Chronicler’s phrase, “men said
openly that Christ slept, and His hallows.”

[967] Under William of Chamai, “civitatis Oxenefordiæ præses,
regalisque militiæ dux et assignator.” _Gesta Steph._ (Sewell),
p. 112. This seems to mean that he was the king’s constable--an
office which had apparently gone with the command of Oxford
castle ever since the Norman conquest.

[968] _Gesta Steph._ (Sewell), pp. 112–114. Hen. Hunt.,
l. viii. c. 23 (Arnold, p. 278). Will. Newb., l. i. c. 13
(Howlett, vol. i. p. 48).

[969] _Gesta Steph._ (Sewell), p. 116.

[970] Eng. Chron. a. 1140. The real date must be 1146, as given
by Hen. Hunt., l. viii. c. 24 (Arnold, p. 279).

[971] _Gesta Steph._ (Sewell), p. 115. Hen. Hunt. as above.
Gerv. Cant. (Stubbs), vol. i. pp. 129, 130.

[972] _Gesta Steph._ (Sewell), pp. 115, 116.

[973] _Ib._ pp. 121–123. Cf. Hen. Hunt. as above.

[974] Eng. Chron. a. 1140.

[975] _Gesta Steph._ (Sewell), pp. 123, 124. Hen. Hunt. as
above. Will. Newb., l. i. c. 13 (Howlett, vol. i. p. 49).

[976] Hen. Hunt., l. viii. c. 25 (Arnold, p. 279). Will. Newb.,
l. i. c. 18 (Howlett, vol. i. p. 57). Compare the different
tone of the two writers.

The hour of Stephen’s exultation over Matilda in England was the hour of her husband’s complete triumph on the other side of the Channel. In the seven years which had gone by since they parted, the count of Anjou had really achieved far more than his wife. As soon as he heard of Stephen’s capture, early in 1141, Geoffrey again summoned the Norman barons to give up their castles and submit to his authority in peace. They held a meeting at Mortagne in the middle of Lent to consider their answer; despairing of Stephen, yet still unwilling to accept Geoffrey, they fell back upon their original scheme and once more besought Theobald of Blois to come and take possession of both duchy and kingdom. Theobald refused the impossible task; but, thinking like every one else that all was over with Stephen, he undertook to arrange terms with Geoffrey for the pacification of both countries. Stephen’s claims, as king and duke, were to be given up to the Angevins on condition that they should set him at liberty and secure to him and his heirs the honours which he had held during his uncle’s lifetime; while to Theobald, as the price of his services in negotiating this settlement, Geoffrey was to restore the county of Tours.[977] The treaty however remained a dead letter; for one of the contracting parties had reckoned without his brother and the other without his wife, both of whom refused their consent. But it served Geoffrey’s purpose nevertheless. The twin earls of Meulan and Leicester, hitherto Stephen’s most active partizans, and the former of whom was after Robert of Gloucester the most influential man in Normandy, at once accepted the proposed terms as final and made their peace with Anjou.[978] Nearly a third part of the duchy followed their example. Mortagne had submitted already; Verneuil and Nonancourt soon did the like; in the last week of Lent Lisieux was surrendered by its bishop;[979] Falaise yielded shortly after;[980] and in a few weeks more the whole Roumois--that is, the district between the Seine and the Rille--except the capital itself, acknowledged Geoffrey as its master.[981]

[977] Ord. Vit. (Duchesne, _Hist. Norm. Scriptt._), p. 923.

[978] Ord. Vit. (Duchesne, _Hist. Norm. Scriptt._), p. 923. Cf.
Rob. Torigni, a. 1141.

[979] Ord. Vit. as above. At this point we lose him.

[980] Chronn. S. Albin. and S. Serg. a. 1141 (Marchegay,
_Eglises_, pp. 34, 145).

[981] Rob. Torigni, a. 1141.

All this happened while the Empress was in full career of success in England. There, however, as we have seen, summer and autumn undid the work of spring; the news of Matilda’s triumph were quickly followed by those of her fall, of her brother’s capture, of his release in exchange for Stephen, and finally, at Whitsuntide 1142, by the visit of Earl Robert himself to entreat that Geoffrey would come and help his wife to reconquer her father’s kingdom. Geoffrey’s views of statecraft were perhaps neither very wide nor very lofty; but his political instinct was quicker and more practical than that of either his wife or her brother. He saw that they had lost their hold upon England; he knew that he had at last secured a hold upon Normandy; and he resolved that no temptation from over sea should induce him to let it go. Instead of helping Robert to conquer the kingdom, he determined to make Robert help him to conquer the duchy. He represented that it was impossible for him to leave matters there in their present unsatisfactory condition; if the earl really wanted him in England, he must first help him in bringing Normandy to order. Thereupon Robert, finding that he could get no other answer, agreed to join his brother-in-law in a campaign which occupied them both until the end of the year.[982] The central part of Normandy, from Nonancourt and Lisieux on the east to a line marked by the course of the Orne on the west, and from the Cenomannian border up to Caen, was already in Geoffrey’s power; he had in fact inserted a big wedge into the middle of the duchy. To gain its western side was the object of the present expedition. The brothers-in-law seem to have started from Robert’s native Caen, and their first success was probably the taking of Bastebourg--Bastebourg above the ford of Varaville, whose name recalls an earlier time and another Geoffrey of Anjou. Then the expedition moved south-westward from Caen through the diocese of Bayeux and up the left bank of the Orne to Villers, Aunay, Plessis and Vire, till it reached and won the already historic site of Tinchebray, on the north-eastern frontier of Stephen’s old county of Mortain.[983] The town and castle of Mortain, and the whole county, with the fortresses of Le Teilleul and St.-Hilaire, were speedily won.[984] Geoffrey marched on to Pontorson, the south-western outpost of the Norman duchy, close upon the Breton frontier, at the bottom of a sandy bay guarded by the Mont-St.-Michel; warned by the general experience, the whole population, men and women, townsfolk and garrison, streamed out to welcome the conqueror as soon as he made his appearance. Thence he turned northward again, to Cérences in the Avranchin; and this place, too, surrendered without striking a blow.[985]

[982] Will. Malm. _Hist. Nov._, l. iii. c. 70 (Hardy, p. 765).

[983] The story of this campaign, as told by the historians of
the time, is little more than a list of the places taken, put
together evidently at random, just as the names happened to
come into the writer’s mind. Its real order must however have
been somewhat as suggested above. The fullest list is in Will.
Malm. _Hist. Nov._, l. iii. c. 70 (Hardy, p. 765): Tinchebray,
St. Hilaire, “Brichesart,” Aunay, Bastebourg, “Trivères,”
Vire, “Plaiseiz,” Villers, Mortain. Bastebourg lies quite
apart from all the rest, and must have been the object of a
distinct expedition from Caen. The other places would follow in
geographical order. “Plaiseiz” may be either Plessis-Grimoult
or Placy; “Brichesart” and “Trivères” are still to be accounted
for. There is a Trévières about half-way between Bayeux and
Isigny, but this is even farther away from all the other
places than Bastebourg, and in an opposite direction. From
Rob. Torigni (a. 1142) we get another list: Aunay, Mortain,
Tinchebray, Cérences, Le Teilleul, all in the county of
Mortain. The _Hist. Gaufr. Ducis_ (Marchegay, _Comtes_, p.
295) names only Mortain and St. Hilaire. The Chronn. S. Albin.
and S. Serg., a. 1142 (Marchegay, _Eglises_, pp. 35, 145), say
Geoffrey won “castella plurima,” but specify only Mortain.

[984] _Hist. Gaufr. Ducis_ (Marchegay, _Comtes_), pp. 295, 296.
Rob. Torigni, a. 1142.

[985] _Hist. Gaufr. Ducis_ (as above), pp. 296–298. The
last-named place appears in Rob. Torigni, a. 1142, as
“Cerences.” In the _Hist. Gaufr. Ducis_, as printed by M.
Marchegay (p. 298), it is “Cerentias”; in the old editions it
was “Carentias,” which the editors of _Rer. Gall. Scriptt._
rendered “Carentan.” “Cérences” is the rendering of M.
Delisle (_Rob. Torigni_, vol. i. p. 226, note 2). It lies
about half-way between Avranches and Coutances. There is a
“Chérencé-le-Roussel” a few miles north-west of Mortain.

At this point the campaign of the count and the earl seems to have been interrupted by tidings of Stephen’s success and Matilda’s danger at Oxford. That Robert must go at once was clear; but that it would be wise for Geoffrey to accompany him was even more doubtful now than it had been six months ago. A substitute was found in the person of little Henry Fitz-Empress, who, if he could do nothing practically to help his mother’s cause and his own, at least ran no risk of damaging it by raising such a storm of ill-feeling as would probably have greeted the count of Anjou himself. While Robert and Henry sailed for England together, Geoffrey remained to finish his work in Normandy. Avranches, the next place which he threatened, made a ready submission; he took up his abode in the castle, and summoned the lords of all the fortresses in the Avranchin to come and do him homage, one after another. When they had all obeyed, he set himself to win the Cotentin. St.-Lô, which had been strongly fortified by the bishop of the diocese, surrendered after a three days’ siege. The victor advanced straight upon Coutances; the bishop was absent; no one else dared to offer resistance; Geoffrey simply marched into the city and took it. Thither, as at Avranches, he summoned the barons of the county to perform their homage, and they all obeyed except two brothers, Ralf and Richard of La Haye. Ralf was soon brought to submission; Richard flung himself with some two hundred knights into Cherbourg, a mighty fortress on a foundation of solid rock, guarded on one side by a belt of woodland full of wild beasts, and on the other by a bay whose advantages as a naval station have only been put to their full use in much later times. A siege of Cherbourg was likely to be a lengthy, troublesome and costly undertaking. But such a siege was of all military operations that in which Geoffrey most excelled and most delighted. He had little sympathy with the downright hand-to-hand fighting by which Fulk Nerra had won his spurs at Conquereux, or Fulk V. had repulsed Theobald and Stephen before Alençon, or Stephen had put his very captors to shame beneath the walls of Lincoln. Engineering was Geoffrey’s favourite science; in its developement he spared neither labour nor expense; and he now brought up against Cherbourg such a formidable array of machines that Richard thought it prudent to slip away by sea, intending to go to England and ask help of King Stephen. He was however overtaken by pirates and carried away “among strange peoples”; and a rumour of his fate reaching the garrison whom he had left behind, they lost heart and made submission to the Angevin.[986] The whole duchy south and west of the Seine was now his,[987] except the one town of Vaudreuil; before the close of the year this, too, was won, and the Angevin power even advanced beyond the river, for “Walter Giffard and all the people of the _Pays de Caux_ made agreement with Count Geoffrey.”[988] The Norman capital now stood out alone against the Angevin conqueror of Normandy, as Tours had once stood out alone against the conqueror of Touraine. In January 1144 Geoffrey crossed the Seine at Vernon and pitched his camp at La Trinité-du-Mont, close to the walls of Rouen.[989] Next day the citizens opened their gates, and conducted him in solemn procession to the cathedral church.[990] The castle was still held against him by some followers of the earl of Warren;[991] the barons, headed by Waleran of Meulan, came to help him in besieging it, but neither their valour nor his machines were of any avail, and it was not till a three months’ blockade had reduced the garrison to the last straits of hunger that the citadel of Rouen was given up on S. George’s day.[992]

[986] _Hist. Gaufr. Ducis_ (Marchegay, _Comtes_), pp. 298–301.
The year, 1143, is given by Rob. Torigni.

[987] Chronn. S. Serg. and S. Albin. a. 1143 (Marchegay,
_Eglises_, pp. 35, 146). The Chron. S. Flor. Salm. (_ib._ p.
191) ventures to say in 1142: “Goffredus Comes totam Normanniam
adquirit hoc anno, iii. octabarum Paschæ, x. kalendas maii.”
This is the true date for the Wednesday in Easter week, 1142,
but the fact is placed two years too early.

[988] Rob. Torigni, a. 1143.

[989] _Ib._ a. 1144.

[990] Chron. Rotom. a. 1144 (_Rer. Gall. Scriptt._, vol. xii.
p. 785); Rob. Torigni, a. 1144. The former makes the day
January 19; the latter, January 20.

[991] Rob. Torigni, as above.

[992] Chron. Rotom. a. 1144 (_Rer. Gall. Scriptt._, vol. xii.
p. 785); Rob. Torigni, a. 1144.

Allies offered themselves readily now to help in the little that remained to be done; foremost among them was the overlord of Normandy, the young King Louis VII. of France. All was changed since the days when his father, Louis VI., had granted the investiture of Normandy to Stephen’s little son. The inveterate enmity between the house of Blois and the French Crown had broken out afresh, in a new and most disastrous form, between Count Theobald and the young king; Louis fell back upon the traditional policy of his forefathers and gladly embraced the Angevin alliance against all the branches of the house of Blois on both sides of the sea. Thus when Geoffrey, after composing matters as well as he could at Rouen, mustered his forces to subdue the few still outstanding castles, he was joined at once by his own brother-in-law Theodoric of Flanders and by the king of France. Driencourt was the first place won by their united hosts; then Lions-la-Forêt--the old hunting-seat where King Henry had died--was given up by Hugh of Gournay;[993] the rest of the castles beyond Seine were quickly won, and then Geoffrey was master of the whole Norman duchy,[994] save one fortress, Arques, which a Fleming called William the Monk held so pertinaciously for Stephen that the Angevin was obliged to leave a body of troops before the place and go home without waiting to finish the siege in person.[995] Next summer the “monk” was shot dead by a chance arrow, and the surrender of Arques completed Geoffrey’s conquest of Normandy.[996] He made no pretence of holding it in the name of either his wife or his son; it was his own by right of conquest, and that right was formally acknowledged by the king of France. Before they parted in 1144 Louis granted to Geoffrey the investiture of the whole Norman duchy, save one spot which he claimed as the price of his favour:--the old bone of contention, Gisors.[997]

[993] Rob. Torigni, a. 1144. Driencourt is now known as
Neufchâtel-en-Bray.

[994] Chronn. S. Albin. and S. Serg. a. 1144 (Marchegay,
_Eglises_, pp. 35, 146); Chronn. S. Michael. and S. Steph.
Cadom. a. 1144 (_Rer. Gall. Scriptt._, vol. xii. pp. 773, 780).

[995] Rob. Torigni, a. 1144. “Willermus Monachus
Flandrensis”--can he have been really a monk?

[996] Rob. Torigni, a. 1145.

[997] _Hist. Gaufr. Ducis_ (Marchegay, _Comtes_), p. 282.

The Angevin conqueror had been called home by a revolt among his own barons.[998] The leader was, as before, Robert of Sablé;[999] but there was worse to come. Geoffrey’s brother Elias was persuaded by the rebels to put forth a claim to the county of Maine and uphold his pretension by force of arms. Geoffrey defeated him, took him prisoner, and put him in ward at Tours,[1000] where he remained five years, and whence he was released only to die of the effects of his imprisonment.[1001] The revolt failed as all previous revolts against Geoffrey had failed; the count swooped down upon Robert and his accomplices with such irresistible energy that they were utterly confounded and made submission at once.[1002] Undisputed master from the Poitevin border to the English Channel, Geoffrey once more cast his eyes across the sea, not with any thought of joining his wife in her desperate venture, but with an uneasy longing to get his heir safe out of the entanglement of a losing cause and bring him home to share in his own triumph. He therefore sent envoys to Earl Robert, begging that Henry might be allowed to come and see him, if only for a short time. The request was at once granted, and by Ascension-tide 1147 the boy was again at his father’s side.[1003] His uncle the earl of Gloucester had escorted him as far as Wareham;[1004] there they parted, as it turned out, for the last time. Robert caught a fever and died at Bristol early in the following November.[1005] Then at last the Empress herself felt that all was lost. Her last faint chance had expired with the wise and valiant brother whose patient devotion she had never fully appreciated until it was too late. In the early spring of 1148 she gave up the struggle and followed her son back to Normandy, to live thenceforth in peace by her husband’s side;[1006] while the knot which the sword had failed to cut was left to be slowly disentangled by more skilful hands which had long been preparing for their task.

[998] Chronn. S. Albin. and S. Serg. a. 1145 (Marchegay,
_Eglises_, pp. 35, 146).

[999] _Hist. Gaufr. Ducis_ (as above), p. 269.

[1000] _Ibid._ _Gesta Cons._ (_ibid._), p. 155.

[1001] _Gesta Cons._ as above. The Chron. Vindoc. (Marchegay,
_Eglises_, p. 173), gives the date, 1150. Cf. Chron. Tur. Magn.
a. 1110 (Salmon, _Chron. Touraine_, p. 131).

[1002] _Hist. Gaufr. Ducis_ (Marchegay, _Comtes_), pp.
270–272. It is here that the writer places the building of
Châteauneuf-sur-Sarthe (see above, p. 267). In connexion with
this affair he gives an amusing reason for the warlike habits
of the Angevins: “Antiquitus nempe Andegavenses præliandi
consuetudinem habebant, forsan, ut puto, a Deo sibi permissum,
ne per otium pejoribus inimicis expugnarentur, moribus scilicet
vitiosis.” _Hist. Gaufr. Ducis_ (as above), pp. 270, 271.

[1003] Gerv. Cant. (Stubbs), vol. i. p. 131. Rob. Torigni, a.
1147.

[1004] Gerv. Cant. as above.

[1005] Gerv. Cant. (Stubbs), vol. i. p. 131. Gervase is not
clear about the year, which we learn from Ann. Tewkesb. a. 1147
(Luard, _Ann. Monast._ vol. i. p. 47), and from Ann. Cantuar.
a. 1147 (Liebermann, _Geschichtsquellen_, p. 6). The place is
given in _Gesta Steph._ (Sewell), p. 132.

[1006] Gerv. Cant. (as above), p. 133--dated a year too early.

NOTE.

THE TOPOGRAPHY OF THE BATTLE OF LINCOLN.

The topography of the battle of Lincoln is a very puzzling matter. We have two sources of information, and it seems impossible to make them agree. The questions to be solved are two: 1. Which way did Robert and Ralf approach the city? 2. Where was the battle actually fought?

1. The first question lies between William of Malmesbury and Henry of Huntingdon. William (_Hist. Nov._, l. iii. cc. 39, 40; Hardy, p. 741) says distinctly that the main army started from Gloucester; that Ralf and his troops joined them somewhere on the road; that Stephen, hearing of their approach, left off besieging the castle and went forth to meet them; and that on Candlemas day they arrived “ad flumen quod inter duos exercitus præterfluebat, Trenta nomine, quod et ortu suo et pluviarum profluvio tam magnum fuerat ut nullatenus vado transitum præberet.” He then gives the story of the crossing. Henry of Huntingdon (l. viii. c. 13; Arnold, p. 268) describes the crossing much in the same way, except that the “consul audacissimus” to whom he attributes the first plunge seems to be Ralf, whereas in William’s version Robert is the hero. But Henry makes no mention of the Trent; in his story the plunge is into “paludem pœne intransibilem.”

For both these versions there is something to be said. The authority of the two witnesses is very evenly balanced. Chronologically, both are equally near to their subject. Geographically, the archdeacon of Huntingdon is nearer than the librarian of Malmesbury; but he is not a whit more likely to have been personally present; and if Henry may have got his information from Bishop Alexander, William may just as probably have got his from Earl Robert himself. The question therefore becomes one of the intrinsic probability of the two stories. Here again there is something to be said for William; for although the most direct and obvious road from Gloucester to Lincoln would undoubtedly be the Foss-Way, along the eastern side of the Trent valley, yet it is possible that the earls might have chosen a more unusual route along its western side, just because it would seem less likely to their enemies. Yet we can hardly accept William’s version; for the fording of the Trent, especially in winter, and when its waters were--as he himself tells us--swollen with heavy rains, would be little short of a physical impossibility. At the origin of his mistake (or of Earl Robert’s, for it must surely have been Robert who told him the story) we may perhaps be able to guess. The writer of the _Gesta Stephani_ (Sewell, p. 71) says nothing of either river or marsh; the only thing which he mentions is a ford, of whose whereabouts he gives no indication whatever. “Cumque fortissimam ... [Stephanus] præmississet cohortem in exitu cujusdam vadi eis ad obsistendum, illi ... cum violentiâ in ipsos irruentes vadum occupaverunt.” Now, if the earls had followed the Foss-Way quite up to Lincoln, it would have brought them not to any ford, but to the bridge over the Witham, leading directly into the city by the south gate. But the city was bitterly hostile to them; had they attempted to pass through it to reach the castle, they must have cut their way through a crowd of enemies. There was however another and a much more practicable route open to them. Some little distance to westward of the bridge, the Witham at its junction with the Foss-Dyke expands into a broad sheet of water known by the name of Brayford. The kindness of the Rev. Precentor Venables has enabled me to ascertain that half way between the bridge and Brayford Head (_i.e._ the eastern end of this sheet of water) there still exists in the bed of the river a well-paved ford road, probably of Roman origin. By this ford the army could cross the river and advance towards the castle without entering the town at all; and I feel little doubt that this was the ford at which Stephen posted the guard mentioned by his biographer, and across which the two earls swam with their followers. In that case William of Malmesbury’s mistake as to the name of the river is not surprising. The Foss-Dyke unites the Witham and the Trent; a medieval geographer could hardly be expected to know accurately where the one ended and the other began. Out of the three names so closely connected, he not unnaturally chose the one most generally known, and concluded the whole water-way under the comprehensive name of Trent; while on the other hand, the overflowing of dyke and river may quite sufficiently account for Henry of Huntingdon having described them and the flooded ground on each side of them all together as an “almost impassable marsh.”

2. Local tradition persists in asserting that the battle was fought to the north of the city, somewhere beyond the New Port. If this was so, Stephen must have led his troops out of the city by the old Roman way--the Ermine Street--through the New Port, and drawn them up on the plateau formed by the top of the range of hills whose southern extremity is occupied by the city itself; and his enemies, after crossing the water, must have marched all round the south-western foot of the hill, below the castle, and then climbed the western slope to meet Stephen on the top. Such a manœuvre is doubtless possible; but it hardly seems to agree with the indications--provokingly few and slight though they are--given us by the historians. None of them indeed tells us which way Stephen went forth; the nearest approach to a clear statement is that of his own biographer, who says “extra civitatem obvius eis audacter occurrit” (_Gesta Steph._ as above). Now marching up northward can hardly be called “going forth boldly to meet” an enemy who was coming from the south-west. The tradition in fact is in itself very improbable, and has no evidence to support it. In 1881 I made two attempts at a personal examination of the topography, with the help of indications kindly furnished me by Precentor Venables. The result was as follows: The western wall of the castle-enclosure does not stretch to the extreme edge of the hill; beyond it lies a part of the plateau, now occupied by the County Asylum, and marked by Stukeley as the site of Stephen’s encampment. Stukeley was probably misled by the circumstance that an adjoining bit of ground was called “Battle-piece”--a name which is now known to have been derived not from any battle fought there, but from the place having been set apart for trials by battle. But farther to the west there lies at the foot of the ridge a tract of comparatively level ground, rising slightly on the one side to join the slope of the hill, and on the other gradually sinking into the lower land which spreads to the bank of the Trent. This tract--part of it is now a race-course--seems to be really the only place in which it is possible for the two armies to have met. The ground immediately south of the castle, between its outer wall and the northern bank of the Foss-Dyke, is too steep to allow of anything like a pitched battle between two formally-arrayed armies. The earls after crossing the ford could hardly do anything but lead their troops round the foot of the hill, to draw them up at last on the western side of the level tract above described. Stephen, on the other hand, could hardly have chosen a better post for defence than its eastern side, with the ridge of the hill at his back.