The Castle mill is mentioned in the Domesday Survey. The present mill no doubt occupies the same site; its foundations may preserve some of the same masonry as that which is thus recorded to have existed hereabouts before the Conquest.
You will notice that the Castle occupies almost the lowest position in the town, and remembering all the other Norman castles you have seen, Windsor or Durham, Lincoln or William the Bastard’s own birth-place at Falaise, the Oxford site may well give you pause, till you remember that the position of the old tenth-century fort had been chosen as the one which best commanded the streams against the Danes, whose incursions were mainly made by means of the rivers. If Carfax had been clear, D’Oigli would have built his castle at Carfax; but it was covered with houses and S. Martin’s; and, shrinking from the expense that would have been involved, and the outcry that would have been raised, if he had cleared the high central point of the town, he was content to modify and strengthen the old fort. But as the descent of Queen Street from Carfax threatened the Castle, if the town were taken, there was no regular communication made between the Castle and the town. A wooden drawbridge across the deep ditches that defended the Castle led to the town, somewhere near Castle Street. This would be destroyed in time of danger. No other entrance to the town was allowed on this side. “All persons coming across the meadows from the West and all the goods disembarked at the Hythe from the barges and boats would have to be taken in at the North Gate of the town, the road passing along the North bank of the City ditch and following, probably, exactly the same course as that followed by George Street to-day” (Parker). And round about the Castle itself an open space was preserved by the policy of the Castellan, and known as the Bailly (ballium, outer court). The Church of S. Peter le Bailly recalls the fact.
Study the history of most cathedrals and you will discover that, like Chartres or Durham, “half house of God, half castle ‘gainst the Scot,” they have served and were intended to serve at some period of their career as fortresses as well as churches.
When Bishop Remigius removed the see from Dorchester to Lincoln, as he did at this time (1070), Henry of Huntingdon writes: “He built a church to the Virgin of Virgins, strong in a strong position, fair in a fair spot, which was agreeable to those who serve God and also, as was needful at the time, impregnable to an enemy.” The tower of S. Michael’s at North Gate is a good example of this mingling of the sacred with the profane, and the architectural feature of it is that it combines the qualities of a campanile with those of the tower of the Castle. It was a detached tower, and not part and parcel of the church which stood at the North Gate, as it is now. In the fifteenth century the city wall was extended northwards so as to include the church.
The tower is placed just where we should expect to find that the need of fortification was felt. South and East, Oxford was now protected by the Thames and the Cherwell as well as by her “vallum,” and on the west was the Castle. But the North Gate needed protection, and D’Oigli built the tower of S. Michael’s to give it, spiritual and temporal both. At a later date there was erected a chapel, also dedicated to S. Michael, near the South Gate, and with reference to this church and chapel and the Churches of S. Peter in the East and in the West, there is a mediæval couplet which runs as follows:
“Invigilat portæ australi boreæque Michael,
Exortum solem Petrus regit atque cadentem.”
“At North Gate and at South Gate too S. Michael guards the way,
While o’er the East and o’er the West S. Peter holds his sway.”
The military character of S. Michael’s tower is marked by that round-headed doorway, which you may perceive some thirty feet from the ground on the north side. Just as the blocked-up archways at the top of the Castle tower once gave access to the wooden galleries which projected from the wall, so this doorway opened on to a lower gallery which guarded the approach to the adjoining gateway. On the south side of the tower you will find traces of another doorway, the base of which was about twelve feet from the level of the ground. It is reasonable to suppose that the tower projected from the north side of the rampart, and that this doorway was the means of communication between them. The other doorway, on the west side, level with the street, gave access from the road to the basement story of the tower.
Architecturally the tower may be said to be a connecting link between the romanesque and Norman styles. The system of rubble, with long-and-short work at the angles, has not yet given place to that of surface ashlar masonry throughout, and the eight pilaster windows, it should be observed, of rude stone-work carved with the axe, present the plain, pierced arches, with mid-wall shafts, which preceded the splayed Norman window and arches with orders duly recessed. The church itself adjoining the tower is of various periods, chiefly fourteenth century. It was, together with S. Mildred’s, united (in 1429) to All Saint’s Church, which then was made a collegiate parish church by the foundation of Lincoln College adjoining.