Buildings with stone foundations and timber superstructures elevated above ground level would therefore have a significantly longer life span. Composite construction never accounted for a substantial number o f the total buildings in the sample (only 77 out o f the 1040 or 7.40%).
However, the patterns found indicate a growing use of this technique over time.
T a b le 3 .2 : B u ild in g s w ith sto n e f o u n d a tio n s a n d tim b e r s u p e r s tr u c tu r e s .
P eriod S to n e /T im b e r S to n e / W a ttle and D aub
T o ta ls
A D 4 3 -1 0 0 4 3 7
A D 100-150 6 3 9
A D 150-250 14 7 21
A D 2 5 0 -3 5 0 15 2 17
A D 3 5 0 -4 5 0 23 0 23
T o ta ls 62 15 77
■ Stone/Wattle and Daub
□ Stone/Timber
AD 43-100 AD 100-150 AD 150-250 AD 250-350 AD 350-450
Fig. 3.3: The number of stone-founded buildings with a timber or wattle and daub superstructure
There were another 72 (6.92% ) stone-founded buildings where excavators could not determ ine the superstructure (Table 3.3 and Fig. 3.4). W hereas the superstructure could be assum ed in the tim ber-founded buildings w ith unknow n superstructures, stone founded
buildings present m uch more o f a problem . Since many buildings had a stone foundation and a w ooden superstructure, it would be w rong m erely to assum e that all stone foundations were indicative o f stone superstructures. W hile the pattern reveals one sim ilar to m asonry buildings, it is worth considering what these w ould do to the pattern o f com posite buildings. Thus Figure
3.4 illustrates w hat these buildings w ith flimsy stone foundations would do to the overall pattern o f com posite structures if added together.
AD 43-100 AD 100-150 AD 150-250 AD 250-350 AD 350-450
Fig. 3.4: The number of stone-founded buildings with an unknown superstructure
U nlike tim ber and stone structures, no site developed a long-standing tradition o f com posite construction. D espite its advantages to entirely tim ber buildings, no tow n had a dom inant num ber o f com posite for more than one period. In the late first-century, Richborough had 75 percent o f its buildings constructed in this fashion. In Sapperton and B ourton-on-the- W ater over 60 percent o f their total buildings w ere com posite construction in the late third and early fourth centuries. W anborough and H ilbaldstow both had over 75 percent o f their
buildings o f com posite construction in the late fourth and early fifth centuries (see A ppendix B).
These numbers should be taken with som e caution since the total sample is relatively small.
H ow ever, taken in com parison w ith other patterns, they have some correlation.
Even here patterns begin to em erge in the use o f com posite construction. The b rief use o f com posite construction at R ichborough led to a long standing tradition o f stone m asonry im m ediately thereafter. B ourton-on-the W ater also followed its period o f com posite dom inance w ith a period o f dom inance by stone, though it is at the end o f the Rom an period in Britain and is one o f the few sites to increase its use o f stone in the late fourth and early fifth centuries.
Both o f these sites seem to indicate that com posite construction was used in a progression to stone masonry. Conversely, at W anborough and H ilbaldstow , both tow ns had predom inantly stone buildings the century im m ediate before the heavy use o f com posite construction (see Table 3.5 for an exam ination o f W anborough). These sites seem to revert to com posite construction during the econom ic m alaise o f the late Rom an period. Sapperton had too few buildings prior to and after its period o f com posite dom inance to m ake a determ ination o f its importance.
4 5 - 4 0 - 35 30 25 2 0- 1 5 - 1 0-
5 - 0-
AD 43-100 AD 100-150 AD 150-250 AD 250-350 AD 350-450
Fig. 3.5: The number of composite buildings in addition to those with unknown superstructures
□ Stone/Wattle and Daub
■ Stone/Unknown
□ Stone/Timber
C. Stone Buildings
O f the 1040 building sam ples, 543 (52.21% ) were o f stone construction (see Table 3.4 and Fig. 3.6). Stone architecture m ade a m arked ju m p in use during the late second and early third-century w ith over a three-fold increase from the previous period. H ow ever, this was relatively short lived as by the late fourth and early fifth-century the num ber o f purely stone buildings was reduced by over 54 percent from the previous period.
Table 3.4: Sum m ary o f stone constructed buildings by time period.
P E R IO D A D 4 3 -1 0 0
S T O N E /S T O N E 20
A D 100-150 48
A D 1 50-250 165
A D 2 5 0 -3 5 0 2 16 A D 3 5 0 -4 5 0 94
T o ta ls 543
0
AD 43-100 AD 100-150 AD 150-250 AD 250-350
Fig. 3.6: The number of entirely stone buildings
AD 350-450
Purely stone buildings reached their height when the timber buildings were declining in the late third and early fourth centuries (see above). However, in the period when timber buildings were at their height in the late second and early third centuries, there were still a greater number o f stone buildings (99 timber to 165 stone buildings). This, as mentioned above, was possibly due to the destruction o f ephemeral traces left by timber buildings. It is important therefore to remember the patterns that are revealed are more important than the actual numbers.
Again, this sudden jump in the use o f the stone construction techniques corresponds with the general observation that the major population centers were undergoing a decline in economic importance as the smaller towns in the rural areas were increasing in their importance (Millett 1990, 133; Hingley 1985, 85; Dark and Dark 1997, 70). Like the timber constructed buildings, there is a marked decline in the late fourth and early fifth centuries that corresponded with the economic, political, and military problems facing the western Empire as it neared its collapse.
As with timber, some sites developed long-standing masonry traditions. In this case more certain geographic connections can be made. From the conquest period to the mid-second century stone buildings were few in number but popular in certain regions (see Maps 3.7-3.11).
South and west o f Cirencester on the lower Fosse Way stone buildings were very common, and from Springhead north to Godmanchester stone buildings are found in increasing numbers but still on par with timber buildings in the same areas (compare with Maps 3.1 and 3.7). Stone was also used exclusively at the port town o f Richborough, likely due to government influences.