NHER 1012 (Monument record) - Spong Hill Early Saxon cemetery and multi-period finds and features

EARLY DISCOVERIES AND EXCAVATIONS

1711. Discovery of Early Saxon cremation cemetery.
At least 150 urns of blue and yellow colour were found when the site was investigated.
See drawing (S57) for an Early Saxon urn with a fake Roman inscription.

1746. Further urns excavated.

1762. Further excavations.
Finds included pairs of tweezers. See sketches (S58).

1853. Cremations excavated.

See sketch (S59) of urn found sometime before 1901 and now held by the British Museum (ex Sloane Collection). This appears to be the British Museum urn that is also illustrated in The Victoria History of Norfolk (S60) and (S73).
See drawing (S66) of urn found pre 1900 that was donated to the Norwich Castle Museum in 1954 (NWHCM : 1954.150). This illustration is reproduced in (S72).

1929. Urn excavated.
A single urn was found at TF 9816 1942.

Before 1952.
Unprovenanced and largely complete Early Saxon cremation urn.
This urn was purchased from an antiques shop in Fakenham before 1952. The provenance of this urn is unknown but it was suggested by [2] to be a missing pot from this site. However, A. Rogerson (HES) believes it is more likely from an Early Saxon cremation cemetery at Kettlestone (NHER 44367) (A. Rogerson (HES), 23 September 2011).
See find document (S69), photographs (S70) and publication off-print (S71).
A. Beckham (HES), 4 May 2017.

1954. Found while ploughing.
A cremation urn was ploughed up at TF 9817 1944.

1954-1955. Excavation by Dereham & District Archaeological Society.
11 urns were found and contained cremations.
Reported in (S68).

Late 1965. Stray finds.
In September [1] noted a number of worked flints at TF 9814 1939 (Context 4), including scrapers cores and other flints. In 1999 the Norwich Castle Museum retrospectively accessioned a collection of flints that had been collected at various sites by [1]. These include a number of finds from Spong Hill that are recorded as found in late 1965:
5 Mesolithic/Early Neolithic blade-like flakes (NWHCM : 1999.328.6.2; TF 9810 1967).
1 Neolithic scraper/knife combination tool, 10 end scrapers, 1 borer, 3 flakes and 2 blade cores (NWHCM : 1999.328.6.1; TF 9820 1960).
3 undatable prehistoric flint flakes (NWHCM : 1999.328.6.3; TF 984 182).
Amended by P. Watkins (HES), 31 August 2014.

1968. Fieldwalking and Trial Excavations.
By 1968 Spong Hill was being deep-ploughed, bringing a scatter of pottery sherds to the surface. Around this time the western and eastern margins of the site were also threatened by development. As a result of these threats a programme of fieldwalking and trial trenching was undertaken in order to assess the extent and degree of preservation of the Early Saxon cemetery. The trial excavations saught to establish the western limits of the cemetery, with the two trenches revealing over 50 complete and fragmentary cremation burial. This work suggested that the cemetery was both well-preserved and extensive; covering an area of about 1ha on the top and southern slope of the hill.

Grave-goods were found in thirty-one urns, the most unique object recovered was part of a bronze applied saucer-brooch the nearest parallel to which is a silver disc from Waben in the north-east of France. This was found together with some fused glass beads and an iron clip in a small biconical urn, Other grave-goods include fragments of four cruciform brooches, three iron brooch-pins, broken and sometimes fragmentary triangular bone combs, bone gaming-pieces, bone rings (one at least cut from the rose of a deer antler), bone beads, fragments of bag rings of elephant ivory, glass beads, iron knives, iron shears, bronze and iron tweezers and fragments of sheet bronze. See (S38) for short summary of this work. See drawings (S63) for urns found during this excavation. See (S92) for lists of grave goods recovered.

THE SPONG HILL PROJECT - EXCAVATIONS 1972-1981

Due to the threats facing the Saxon cemetery it was decided that total excavation was the only viable course of action. This work started in 1972 and annual excavations were undertaken until 1981. A final phase of excavation was completed in 1984. Geophysical surveys were also carried out by the Ancient Monuments Laboratory in 1979 and 1986.

The primary aim of this extensive work was the complete excavation of the Early Saxon cemetery, the survival and extent of which had been established in 1968. The excavation also uncovered extensive evidence for prehistoric, Iron Age and Roman period activity, as well as the remains of a settlement associated with the cemetery itself.

See file for early records, excavation notes, photographs (S2), rubbings (S61)-(S62), drawings (S63) and (S74)-(S88), correspondence (S4) and press cuttings with index in file. For detailed photographs of urns, see (S37). Brief summaries of the results of each excavation season appeared in the journal Medieval Archaeology (S39)-(S48).

The collection of Anglo-Saxon artefacts recovered during the excavation was acquired by Norfolk Museums Service in 1988 (NWHCM : 1994.192). See unpublished document (S91).

THE RESULTS OF THE EXCAVATIONS

The results of the excavations (including the 1968 evaluations) have been published as a series of volumes in the East Anglian Archaeology Series, with the burials from the Early Saxon cemetery published in a series of catalogues (Parts I to V); (S1), (S49), (S50), (S51) and (S52). The first volume (S1) also included a useful summary of the earlier discoveries. Separate volumes have also been produced detailing the evidence for prehistoric occupation (S6), the Iron Age, Roman and Early Saxon settlements (S7) and the human and animal bone from the burials (S53). Spong Hill was also one of four sites included in a recently published analysis of Anglo-Saxon inhumation burial in East Anglia (S54). It is anticipated that a final synthetic volume on the Spong Hill cemetery will also be produced.
E. Rose (NAU), 30 January 1991 amended by M. Langham-Lopez (HES), 02 May 2013, P. Watkins (HES), 20 August 2013 and A. Beckham (HES), 4 June 2021.

The results of the 1968-1984 work at Spong Hill is summarised below, by period.

PREHISTORIC ACTIVITY

Evidence for Palaeolithic activity:
A very rolled flake from the excavations was identified as potentially Palaeolithic by J. J. Wymer. This object is described and sketched on (S67). Aparently two further flakes in a similar condition were more dubious.

Evidence for Mesolithic activity:
The earliest activity on the site appears to have occurred during the Mesolithic, with worked flints of this date widespead over the excavated area. The diagnostic Mesolithic worked flints included simple obliquely-blunted points, (later) geometric microliths, 'truncated pieces', at least one microburin and a tranchet axe. It is likely that at least some of the debitage and less-diagnostic retouched pieces found in unstratified and later contexts were also Mesolithic. It was however difficult to positively identify other Mesolithic pieces in the assemblage, due largely to the presence of a significant Early Neolithic component. The nature of this activity is inevitably somewhat uncertain, the majority of the lithic material being either found residual in later features or recovered from unstratified contexts. A notable exception was a small assemblage of Mesolithic flint (including microliths and debitage) recovered from a natural hollow that also contained large quantities of pine charcoal and evidence for in situ burning. A radiocarbon determination confirmed that this debris was associated with Mesolithic activity (HAR-7063 8230+/-80 BP). Two further radiocarbon determinations also gave similar dates; one taken from an otherwise undated possible hearth (HAR-2903 8150+/-100 BP) and one for charcoal that was presumed to be residual in a pit of probable Neolithic date (HAR-7025 8250+/-90 BP). The environmental evidence suggests that at this time the hillside was covered mainly with pine woodland that was growing on the hummocks and hollows left after partial infilling of periglacial features. It is not clear whether the evidence for burning represents deliberate clearance or domestic hearths. It is possible the activity at this time was only sporadic, particularly as the microlith assemblage appear to include piece typical of both the earlier and later Mesolithic.
See (S6) for further details. These finds are also noted in (S56).
Amended by P. Watkins (HES), 20 August 2013.

Neolithic and Bronze Age activity.
Radiocarbon dates have confirmed that Neolithic occupation had begun in Spong Hill by the early third millennium BC, and the site appears to have been visited periodically throughout the Early Neolithic. The occupation continued into the second millennium BC, but far less subsoil features are recorded for this period, reflecting patterns throughout East Anglia.

Only a small proportion of the excavated features could be confidently assigned a pre-Iron Age date. These included five discrete groups of pits and post holes that were almost certainly associated with Early Neolithic activity, these containing substantial quantities of Mildenhall Ware pottery and contemporary worked flint. Similar, but smaller, assemblages of material were also recovered from a number of elongated hollows thought to be either periglacial formations or root-holes. Only two of the main feature groups contained features identified as post holes. One group possibly formed a roughly rectangular setting, whilst in another location a irregular row of seven or eight post holes was identified. It was however noted that many of the features recorded as post holes were not entirely convincing. A small group of potentially contemporary Late Neolithic/Early Bronze Age features was also identified. These features contained charcoal, burnt flint and a range of pottery sherds, including several from a Food Vessel and a 'pygmy' cup. A number of worked flints were also recovered. Two radiocarbon determinations confirmed a Late Neolithic/Early Bronze Age date for this feature group. It addition to these more readily datable features at least 19 features of indeterminate prehistoric date were also identified. These were assigned to this phase of activity on the basic of either their stratigraphic associations or the present of finds (such as worked flint) that suggested a general prehistoric date. At least two were possible hearths and a number produced burnt flint.

It is likely that the majority of the 7839 worked flints recovered (including material found during the earlier fieldwalking) were associated with later prehistoric activity. The majority of the stratified pieces were recovered from Early Neolithic features and it is reasonable to assume that much of the unstratified material was of a similar date, particularly as several of the concentration observeds coincide with Early Neolithic feature groups. Retouched tools recovered from Early Neolithic features include leaf arrowheads, scrapers, borers, denticulate/serrated pieces and a flaked or ground axe. The Late Neolithic/Early Bronze Age features did not produce much lithic material although at least some of the unstratified material was probably associated with this phase of activity. Notable later prehistoric pieces found in unstratified contexts include several addition flaked/ground axes, chisel, oblique and barbed and tanged arrowheads and several unfinished arrowheads. The remainder of the flint assemblage comprised cores, debitage, other retouched pieces and a large number of flakes and blades.

See (S6) for further details. The Early Neolithic pits from this site were also included in Garrow's analysis of Neolithic pit groups in East Anglia (S55).
H. Hamilton (NLA), 7 February 2008. Amended by P. Watkins (HES), 20 August 2013.

IRON AGE OCCUPATION
Iron Age activity dates from the 3rd or 2nd centuries BC. The excavation appears to have located the edge of the main settlement area, and occupation appears to have been confined to the highest part of the hill.

Excavated features included a few elongated pits, grouped together and found to contain burnt material, either from domestic or industrial activity nearby. One isolated circular pit was more like a typical storage pit, although environmental analysis did not show that any of these pits had been used for crop processing or storage. Some time in the 1st century BC or slightly earlier a large ditch was constructed with a bank on its western side, revetted by timberwork. This was aligned roughly north-south along the eastern edge of the ridge of high ground, and has been interpreted as a territorial boundary, defining the eastern edge of the Iron Age settlement area. Another much shallower ditch was dug down the western slope at the end of the ridge, perhaps delineating pasture or arable land on the slopes of the hill.

In the late Iron Age (probably the 30s or 40s AD) an enclosure was built on top of Spong Hill, at the edge of the ridge and just to the west of the large boundary. No traces of any internal structures were recorded within the enclosure. The pottery from its ditch shows a gradual transition to Romanised wares, all presumably of local origin. A concentration of Roman pottery consisting of large sherds along a stretch of the southern ditch of the enclosure may indicate settlement nearby, perhaps within the enclosure itself. It is not known whether the enclosure was a communal refuge, a stock enclosure or one containing the permanent residence of an important family.

ROMAN OCCUPATION
Occupation continued without a break into the Roman period. The late 1st century saw a complete remodelling of the settlement boundaries and layout, with new shallow ditches creating a system of paddocks with an associated trackway and enclosure. At the same time, a pottery kiln was established, producing medium-sized jars and possibly flagons, dishes and large jars. This kiln may have served a local civilian or military market, but it is notable that no coins belonging to this phase were recovered during the excavations. There seems to have been a shift of site away from the Iron Age settlement area towards an enclosure on the hillslope by the early to mid 2nd century AD. This seems to have remained the centre of occupation for the rest of the Roman period. Within this area was a droveway, small enclosures containing buildings, pits and possibly a garden or small field along with stock enclosures and fences. In the later Roman period this area appears to have been used for crop processing. This farmstead enclosure was remodelled several times during successive phases of occupation. By the 4th century the ditches within the enclosure were no longer being cleaned out and by the end of the 4th century these were being used as dumps for domestic rubbish. Iron smelting and smithing occurred at Spong Hill during the Roman period, but only the slag from these activities was found during the excavations. There was also plenty of evidence for crop processing and stock rearing at the farmstead.
A. Cattermole (NLA), 15 February 2008.

SAXON OCCUPATION.
In total up to 2,200 cremations dating from the early 5th to the mid-6th century were recorded and excavated. Some of the cremations in this southern half of the site had been grouped together, including a group of twenty urns which seem to have been buried at the same time. Grave goods found with the cremations include an early cruciform brooch of Aberg type I and a small bronze bell. One cremation contained substantial parts of an equal-armed brooch, a pair of undamaged saucer brooches with spiral scroll pattern, silver bound iron rings and silver finger rings as well as the more normal bone combs, and ivory and glass beads.
Previously unknown inhumation cemetery was also discovered immediately north-east of the cremations. The inhumations ranged from the mid to the late 6th century, slightly later than the main cremations cemetery, and did include a small number of cremations as well. It included a chamber grave that was surrounded by a ring ditch on which three inhumations had been aligned on it. It has been suggested that it was the earliest inhumations which were buried in this area around the chamber grave and away from the main cremation cemetery, later inhumations continued to be placed in the area.
Other features and structures uncovered during the excavation are clear evidence that there was a settlement on the hilltop which was contemporary with the cemetery. This includes a Grubenhaus located north-west the cremation cemetery which is one of the first evidence of settlement in the area. Other structures associated with the cemetery include two post-hole buildings and further buildings in the north-west corner, which has a shallow boundary-ditch which runs along the north side. This excavation also uncovered part of the previously excavated enclosure ditch which had a small quantity of rectangular post-holes which were of Saxon date. There was also a pit which was found within one structure which contained remains of annular loom-weights and packed vertically. A notable Early Saxon find recovered from a rabbit burrow during the excavation was 'Spong Man'; a pot lid in the form of a seated figure (S89 and (S90).

THE CROPMARK EVIDENCE

November 2007. Norfolk NMP
Cropmarks relating to the Iron Age to Roman farmstead and Saxon cemetery at Spong Hill are visible on aerial photographs (S8-S35). Cropmarks of undated ditches and post medieval field boundaries are also present. Due to the complexity of the site it has not been fully possible separate the features into separate periods and all phases have been mapped by the Norfolk NMP. Cropmarks of three ring ditches of Bronze Age date are the only features to have been separated from the other cropmarks and they are now recorded as NHER 50704. Interpretation and, in particular, dating of the cropmark features draws heavily on the results of the excavations at this site (S6, S7). Despite a very large number of aerial photographs showing the site being available (only a selection of the best of these are listed as sources below), mapping of the cropmarks was not straightforward. This was partly due to the topography of the site, which made accurate photo rectification difficult. Not all of the features shown on published cropmark plots (S6, Fig.3 and S7, Fig.2) were plotted by the NMP. Some of these were not convincingly visible on the available aerial photographs, whilst others were considered to be of probable natural origin. Conversely new features were mapped, particularly away from the area of the excavation. Although the excavation results enabled cropmark features in the southern part of the site to be assigned an accurate date, they also obscured part of the site, preventing cropmarks from being recorded in those areas. Some features were visible in open excavation areas and were mapped in this form rather than from cropmarks.

The earliest phase of datable cropmarks mapped by the NMP are a group of rectilinear enclosures located at the southern end of the site. These features were partly exposed during the excavations (S6-S7) and were shown to be of Iron Age to Roman date. The plan of these enclosures as revealed by the cropmarks is inevitably not as complete as that produced by the excavation. A large incomplete enclosure, measuring at least 190m by 120m externally was located at the southern end of the group. Adjoining its northern side were a series of smaller rectangular enclosures. The most complete of these, in terms of the cropmark evidence, has internal dimensions of 65m by 67m. Defining the eastern side of some of these enclosures was a roughly north to south aligned linear ditch that extended for at least 375m. Beyond the limits of the excavation area it is difficult to be certain which, if any, of the cropmark features are of Iron Age to Roman date. This is partly due to post medieval field boundaries having a similar alignment (see below).

Only two archaeological features relating to the Anglo Saxon phase of activity were mapped. These comprised a complete and half ring ditch located at TF 9823 1947. These small ring ditches, both with external diameters of 8m, surrounded individual graves. These features lay within the excavated area and they were mapped from aerial photographs of the excavation rather than from cropmarks. No features relating to the Anglo Saxon cemetery were visible as cropmarks. Seven grubenhauser belonging to the Anglo Saxon settlement were revealed in the excavation but none of these was visible as cropmarks. This lack of cropmark evidence for grubenhauser is a phenomenon that has been noticed elsewhere (e.g. at Witton in north-east Norfolk (NHER 1009)).

Cropmarks of post medieval field boundaries were visible to the north and west of the excavated area. Some of these correspond to boundaries marked on an 1829 plan of Elmham parish (S36). These include three curving ditches that extend westwards from TF 9831 1961, TF 9832 1967 and TF 9832 1974. A double ring ditch is located at the western end of the most northerly of these three ditches. This ring ditch (NHER 50704) forms the meeting point of three field boundaries and it is likely that it survived as an earthwork into the post medieval period. A rectangular enclosure cropmark, previously recorded as NHER 17659, is present in the northern part of the site at TF 9826 1995. The cropmark appears to correspond to field boundaries shown on the 1829 plan of Elmham (S36) and is likely to be a field boundary of post medieval date. Other boundaries marked on this map are also visible as cropmarks. Ditch cropmarks of unknown and possible Iron Age to Roman date are also present across the wholes site.
J. Albone (NMP), 28 November 2007.

RECENT WORK

Autumn 1987. Watching brief on pipeline.
No archaeological finds were recovered, but several features including pits and ditches were observed.
See notes in file.
A. Cattermole (NLA), 5 February 2007.

13 March 1990. Test Pitting.
Series of test pits excavated by [1] on land surrounding the excavated site. This work was monitored by NAU staff. No archaeological features were observed, although it was noted that the machine bucket often 'smeared' the sections of the pits and as a result minor features would probably have been obscured. It does however seem reasonably likely that no major features such as ditches or deep pits were present. See file for full details and location plan.
P. Watkins (HES), 20 August 2013.