Mediaeval roof construction explained
In the construction of roofing, a problem to be overcome is the longitudinal instability of the rafters holding the physical roof and its tiles in place. Today, most roofs are stabilised by purlins which run along the inside of the rafters which stop them racking when struck by a side force (for example a wind or indeed ground or building movement). Yet this solution was not common in the early middle ages.
Instead, in the period up to that of the Tudors, a common solution in larger buildings was the use of collars connecting each pair of rafters towards the top of the roof space and then a central collar purlin which processed down the centre of the roof on top of which these collars rested. Each collar would have been joined to the rafters and then resting on the collar purlin to provide a rigid structure.
The diagram below (top two drawings) shows the relationship between the rafters, collars, collar purlin and the crown post. When I describe the roof at St James' I will refer to the relevant parts of the roof shown in this diagram by highlighting them in bold and in blue.
The length of the roof space at St James' means that it would be very unlikely to find a beam of wood long enough to act as a collar purlin for the whole roof space.
Consequently, the collar purlin would most likely comprise several separate beams joined together at each end - perhaps by a scarf joint as seen at nearby Place House in Ware - and then held up by one or more crown posts along its length.
The scarf joint is one of the triumphs of mediaeval historic carpentry, enabling a beam to be joined to another unsupported and yet retaining its rigidity. The scheme is sbown in this diagram (right):
What is currently visible at St James'?
As can be seen in the photograph at the top of this page, St James' has its own crown post roof, which is immediately visible once the visitor enters the nave and looks towards the altar.
As we look that way, we see above our heads a series of large oak beams which tie together each side of the nave and are known as tie beams. These are also fundamental in allowing a location for the crown posts; four of the tie beams are topped with their own crown post.
However, apart from the four posts which remain, it is difficult for the visitor to see the complex roofing problem which the posts resolve because the bulk of the carpentry is hidden within the lath and plaster work in the ceiling vaults.
Yet, if you look closely at the photograph, you can see that the plaster does not form an apex but instead is flat towards the top of the roof space. It is above this flat area which the collars and the collar purlin lie.
A secret untold - was there ever a third crown post and what happened to it?
Having said all this, the roof space does have another interesting story to reveal - why does the tie beam at the east end of the nave not have its own crown post?
It would seem at first glance that the tie beam nearest the altar does not have a corresponding crown post; without venturing up there, I am unable to determine whether there is a locating mortise for the tenon of a corresponding post or whether indeed there was never a post in the first place.
Interestingly, however, if we look towards the altar we see that the last two tie beams are linked by timber wall plates which run along the top of the wall and form a rectangle of wood. This feature is not replicated at the western (font) end of the nave nearest the door. More, if we take a closer look we can see that this frame appears to run below the tie beams at this end of the church, whereas at the western end the tie beams appear to be jointed into a frame which is currently only visible in terms of the plaster work which has been put in place over it.
We might conclude therefore that the additional wooden frame at the altar end may reflect a reinforcement put in place at the time of the construction of the Baeshe Chapel in the 1570s (photo left shows window of the chapel with dating detail). Indeed, the frame may have formed part of a temporary support structure to hold the roof as the north wall was cut through and then left in place.
If there was indeed a further crown post, is it possible that during this period it was removed with, instead, a greater reliance being placed on a reinforced wall plate in conjunction with the gable end and the next crown post being sufficient to maintain rigidity without going to the expense of re-cutting and fitting a new crown post?
We can only now imagine the nave and chancel once full of workmen making these decisions in those ancient days many summers ago.