If you have any feedback or new information about this record, please email the Southampton HER (her@southampton.gov.uk).
Data derived from Heritage Gateway will not be acceptable for any use associated with development proposals and other planning matters.

For important guidance on the use of this record, please click here . © Southampton City Council

Summary

Historical sources show that there was once a circuit wall around the area enclosed by the inner ditch at Bitterne Manor. Speed (in 1770) and Englefield (in 1800) described various lengths of wall still visible above ground, and other structures along the part of the wall just inside the inner ditch. The sources give conflicting information as to the locations of these above-ground remains. In 1804/5, the rampart bank inside the inner ditch was levelled, and Englefield reported that foundations of "the whole of the ancient eastern wall", along with various structures, including towers, along its route were revealed. Parts of the wall seem to have been traceable into the 20th century, but the only part surviving above ground today is just west of Bitterne Manor House. The latter was investigated during archaeological excavations in 1936/1938 ( SOU 215); the dating evidence obtained suggested that the wall was built in the 4th century AD. Two excavations in 1951 (SOU 202 and SOU 203) extended across the area of the demolished eastern wall. On each site, a broad band of postholes, aligned north-south, was found west of the Inner Ditch. This was interpreted by the excavator as a stockade dating to AD150+. However others have suggested the postholes were piles for a stone wall, and a 4th century date for the postholes can be argued.
[This record is not yet complete. Further details can be found in the sources listed and elsewhere.]

Protected Status: None recorded

Other References/Statuses

• HER backup file (new series): MSH336  Shared with MSH302, 316, 338, 452.
• Old Southampton SMR No/Backup file: SU 4313 SW 10  

Monument Type(s):

• CURTAIN WALL (Roman - 300 AD? to 409 AD?)
• TOWN WALL (Roman - 300 AD? to 409 AD?)

Full description

[1] (1770): Speed describes the "present Remains of Clausentum", including some stone structures. The structures included various segments of wall around the area enclosed by the inner ditch, and the following wall segments are specifically mentioned. "At the north end of the Barn are still to be seen some remains of the foundation of the old Wall of the Fort, about four feet in thickness"; the barn was at the north end of the inner bank (see separate record). There was a segment of wall between the stable and square tower (see separate records), and "near this Tower there still remain foundations of Flint for a considerable space, and many fragments of the same appear all round the Ground within the inner Ditch and on the Bank of that Ditch; but many of the materials were remov'd in Queen Elizabeth's time . . . ".

[3] (1770): The key of this plan of Bitterne (reproduced as [2] and originally included with the manuscript copy of [1]) refers to "the remains of the foundations of a wall which in some places is about four feet thick", all around the area west of the inner ditch. The wall is not shown on the plan itself.

[4] (1770): This rough plan of the Bitterne Manor peninsular, shows various features including some structures on the western edge of the inner ditch, identifiable as the barn, gateway and segments of wall. Two upstanding segments of wall appear to be shown, one identifiable as the segment north of the barn in [1]. There is also a presumably conjectural representation of the original length and height of the wall by the inner ditch. (IP 4/9/03).

[5] (1798): A plan indicates "Thick Walls of Masonry, in some part projecting as much as 20 feet" along the bank inside the inner ditch to the south of the new road (now Bitterne Road - IP) and around the shoreline of much, but not all, of the area inside the inner ditch. The text only gives the following information: "The walls towards the water were in some places below the present high water mark . .".

[6] (1800) (reprinted in [7]): Englefield noted that the Roman wall (see below - IP) was about 9 feet thick, built of flint, with a course of tile ("large flat bricks") and faced "very roughly with square small stones". The wall "has no foundations whatever, but is literally set down on the surface of the ground, and is therefore undermined by the waters of the Itchen, which only reach it at spring tides. A large bank of earth is thrown up against it on the inner side, and, in the only place where I have been able to examine its interior construction, it seems as if, at a distance of about nine feet within the outer wall, another wall of about two feet thick has been erected, seemingly as a sort of strengthening to the rampart of earth. (A section* of the wall and bank is illustrated - said to be drawing No 8, but probably Plate V.)
* IP 12/1/04: The location of this section is unclear. [6] refers to the 1798 plan in [5], and the section could be along the bank/wall inside the inner ditch or the wall along the shore (part of which seems to be shown as two parallel walls) shown on the 1798 plan.

[7]: A plan (dated ?1802/?1804) shows walls at various places around the area west of the inner ditch. A "wall and bank" is shown just inside the inner ditch, and "wall" is shown along the south shore between the inner ditch and the new road, and in two places along the northwest/north shore west of the inner ditch. The spring tides came up to the wall along the shore, and filled the inner ditch (see accompanying note on pages 145/6).
IP 8/9/03: The plan in [7], although "taken in great part" from the plan in [5], differs from the earlier plan in several respects with regard to the location of surviving walls west of the inner ditch; this is not explained in the text of [7].

(See other records for individual features observed along the east wall.)

[8] (1805): In 1804/5, "in the course of levelling and filling the inner fosse, north of the road, for the purpose of making a kitchen garden", "the whole of the ancient eastern wall", along with various structures, including towers, along its route were revealed.
IP 10/9/03: The information given in [8] is contradictory. It is not clear whether "the whole of the eastern wall" north of the road (Bitterne Road), or across the entire peninsular, was revealed. Map evidence may support the latter interpretation. See event record (ESH1369) for full discussion.

[9] (1858): Kell states that the foundations of the wall around the inner island at Clausentum "are still traceable - composed of large flint stones, cemented by mortar now nearly as hard as the stone itself." (Then gives details that must be from Englefield [8].)

[10] (1900): Refering to the wall around the inner area of Clausentum, when Haverfield visited the site in 1898, "a small piece of exposed masonry could be seen west of the house, and the course of the wall could be traced some little way." He "thought to detect the north-west corner of the inner walled area a little west of the house, and most of the line of the west wall". (Haverfield cites Englefield from [6], [7] and [8], amalgamating these texts to give an unambiguous account, not necessarily justified by the original.) ("Lines of Roman Walls" are shown on Fig 29, but this is conjectural.)

IP 24/9/03: The only part of the circuit wall surviving above ground today is a length about 20m long just west of Bitterne Manor House, at SU43331337, for which see child record. This is shown on current OS maps.

SOU 215 (Roman Wall excavation in the Bitterne Manor House Grounds 1936 - 1939, Waterman Site F):
The above-ground length of Roman wall just west of Bitterne Manor House was investigated during archaeological excavations in 1936/1938 ( SOU 215). The dating evidence from these excavations suggested the wall was built in the 4th century AD (two possible dates were suggested). See child record for details. (IP 12/1/04)

SOU 218 (Watching brief on south corner of Hawkeswood Road and Bitterne Road in 1939):
[13]: Roman occupation layers were revealed, suggesting the line of the Roman town wall lay SW of the site.

Two archaeological excavations in 1951 (SOU 202 and SOU 203) extended across the area covered by the wall recorded by Englefield. The relevant evidence from these excavations is given below, as presented in the report [11], followed by a discussion of the interpretation of the evidence. (IP 12/1/04)

SOU 202 (Excavation in Steuart Road in 1951 - Cotton Site B):
[11]: A trench was dug across the inner ditch of Clausentum, extending as far as possible either side of the ditch. (See trench location plan figure 8, and trench sections and detailed plan on figure 7.) (See MSH 1742 for full details and results for SOU 202.) 103 postholes were found, which together formed a north-south alignment, 9' wide east-west and 48' west of the inner ditch. These postholes cut an old turf line and natural gravel to a depth of up to 1' 6". Other large postholes to the east and another pit or posthole may be associated with this group. On figure 7, one of the postholes in the group is shown cutting layer 3W, although this relationship is not mentioned in the text; the date given for layer 3W is AD300 - 350. It was not possible to date the postholes. The group of 103 postholes and other ?associated postholes were sealed/overlain by an "unstratified" "sub-humus" layer. (See below for discussion.)

SOU 203 (Excavation in the Grounds of Bitterne Manor House in 1951 - Cotton Site C2):
[11]: A trench was dug just inside the inner ditch of Clausentum. (See figure 10 for trench sections and detailed plan.) (See MSH1684 for all results of SOU 203.)
A broad band of 43 postholes, forming a north-south alignment (about 5.5 ft across), was found on SOU203. Another group of 7 postholes was found cutting a "turf line" across the top of a backfilled ditch. This group of 7 postholes was thought to have no connection with the larger group, and the larger group were assumed to be contemporary with the ditch. The fill of the ditch was dated to 120 - 170/180 AD (pottery evidence). The group of 43 postholes and the ditch all cut a gravel surface (dating evidence was a single 1st century sherd), below which were 1st century Roman or undated Roman deposits. Both groups of postholes and the ditch were all covered/sealed by a single, thick "debris" level (relationship referred to in text and shown, perhaps rather ambigously, on section drawing - IP). This debris layer was said to date to 170/180 - 350 AD, reflecting the range of pottery it contained.
[HK 1995: The cuts of the main group of 43 postholes and ditch are in the same stratigraphic phase and so COULD be contemporary, as suggested in [11]. However, if the ditch was earlier than the main group of postholes, both groups of postholes could be contemporay; Richmond in [14] suggested that this could be the case. It seems to have been assumed in [11] that the debris layer followed immediately after the ditch fill, hence the date suggested for the debris layer of 170/180 - 350 AD in [11]. However, the second group of postholes indicate post-ditch activity, while the main group of postholes, would not need to be out of use because the ditch was. From the pottery evidence, the debris layer could date to the first half of the 4th century. (The formation of a suggested "turf line" over the top of the ditch would also support the latter interpretation. IP 2003)] (See below for discussion.)

SOU 202 and SOU 203 Interpretation:
HK 1995/IP 2003: [11] is the report for the results of these excavations, although the results and interpretations presented in that source are problematic. On each site, a broad band of postholes was located west of the line of the Inner Ditch. In [11], Cotton interpreted the band of postholes as a stockade, rejecting the alternative interpretation (followed by many later commentators, for instance [12][14]), that the postholes were pilings for the footings of a stone wall. On SOU 202 it was not possible to date the "stockade" so SOU 203 was excavated to find dating evidence. On SOU 203, a date of AD150+ for the stockade was suggested (p40), but this is based on a questionable interpretation of the evidence (see above); a date any time between 120 AD and the first half of the 4th century might be possible based on the alternative interpretation suggested above. On SOU 202, layer 3W (which was cut by at least one posthole) was dated to AD300 - 350, which conflicts with the date for the stockade given in [11]. On SOU 203 the posthole fills were apparently sealed by the debris layer dating to no later that the first half of the 4th century; this argues against the postholes being piling for a wall, unless the wall was perhaps reconstructed cAD300 - 350. However, how safe is the assertion in [11] that the debris layer "covered"/"sealed" the postholes? Further work with the site archives and further excavation work is required to resolve these problems. The width of the alignment on SOU 202 (but not SOU 203) exactly corresponds to the width of the stone wall described by Englefield in [6].

GIS: approximate location of wall drawn using information from various historic maps. IP 24/9/03.

Sources / Further Reading

Associated Finds: None recorded

Associated Events

• ESH1366 - Rough Field Survey of the Roman Remains at Bitterne Manor in 1770
• ESH1367 - Observations at Bitterne Manor during the construction of Bitterne Road in 1798/1800
• ESH1368 - Surveys and Descriptions of the Bitterne Manor peninsular between 1798 and 1800
• ESH1369 - Observations at Bitterne Manor in 1804 and 1805
• ESH1374 - Excavation in the Grounds of Bitterne Manor House in 1951 (Cotton Site C2) (Ref: SOU 203)
• ESH202 - Excavation in Steuart Road in 1951 (Cotton Site B) (Ref: SOU 202)

Related records

Associated Links: None recorded

If you have any feedback or new information about this record, please email the Southampton HER (her@southampton.gov.uk).

Search results generated by the HBSMR Gateway from exeGesIS SDM Ltd.