Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

There is no national or regional record for World War II heritage in France. This is of particular concern for Normandy where, in terms of history, the impact and significance of World War II arguably equals that of its Norman heritage. The camp at La Glacerie was rediscovered by accident during an archaeological evaluation in advance of a housing development (Tournier 2006). The site was originally investigated for likely Gallo-Romain remains, but significant archaeological features that contained artifacts dating to World War II were found. Discussions with local land owners revealed that the site was indeed a PoW camp, which prompted the Service Regional de Archéologie to commission an additional historic study that confirmed broadly its identification. As part of this study an aerial photograph was identified in the French national archive (Institut Geographique National (IGN), 15 aout 1947, Mission F1110-1310, cliché no.298) that showed features belonging to a significant site that had been dismantled and returned to pasture. An archaeological excavation covering an area of 1.6 hectares was placed where the most significant archaeological remains were identified during evaluation. Oxford Archaeology was the only organization in France that expressed an interest in the research, demonstrating that World War II archaeology and specifically the archaeology of PoW camps is not yet recognized widely in the French archaeological community as a research objective.

The research objectives were to characterize the archaeological remains over the 1.6 hectare site and to comment on their value compared to other available sources of evidence. This research was being conducted within the restricted environment of contract archaeology (see also Chap. 3), and given the limited experience of such research in a European context it is timely to assess the contribution that such studies can make to our understanding of PoW archaeology.

Aerial Photography

Four relevant aerial photographs have been identified dating from January 1944 to August 1947; three were located at Keele University, England and one in the IGN archives in France. Two photographs dated January 5 and May 7, 1944 confirm that the location of the camp was not in use and the area was then a green-field site made up of enclosed land parcels delineated by hedgerows and roads. When compared to the later photographs, it is apparent that these earlier land divisions dictated the layout of the camp, where internal divisions respect earlier field boundaries. The most significant aerial photographs are dated August 3, 1945 (106G/L1B.218.3/4157 and 4158) and August 15, 1947 (as cited above).

The former shows an extensive and occupied PoW camp and is, to date, the only historic record of the original layout at La Glacerie (Fig. 6.1). The latter is a view of the dismantled site (Fig. 6.2).

An Analysis of the August 1945 Photograph

This vertical photograph shows that the “main” camp was developed to the south of a road aligned east–west that is still in use today and which descends to the west into Cherbourg. The camp had a formal and ordered layout of administrative buildings, barrack blocks, recreational features, and tents (Fig. 6.1). Other associated areas, can be found to the north (Area E), northeast (Area D), and west of the main camp (Area B). The areas of excavation, that were defined prior to the identification of this photograph, lie within a further rectangular area (Area A) subdivided into 16 compounds, 12 of which contain rows of smaller structures (Figs. 6.2 and 6.3). Further developments extend in an ad hoc fashion to the southwest, where there is no noticeable site boundary.

Fig. 6.1
figure 1

Aerial photograph of La Glacerie camp in use August 3 1945

Full size image
Fig. 6.2
figure 2

Aerial photograph of demolished La Glacerie camp, August 15 1947, with excavation areas indicated

Full size image
Fig. 6.3
figure 3

La Glacerie excavation site plan

Full size image

A series of structures, possibly watchtowers or control points, can be observed along the boundary of the main camp adjacent to the main road. The main entrance to the site is also clearly visible and gives onto a formal garden. The camp is subdivided into rectangular blocks separated by defined paths or walls. Each block contains identical buildings with rectangular footprints, and on close inspection these have pitched roofs. To the southwest of the main area, narrow rectangular compounds that contain parallel rows of rectangular buildings can be observed, elsewhere uniform rows of buildings run along the sides of square blocks enclosing open areas that in some instances have large structures in their center. Towards the southeast of the main camp, situated centrally and close to the formal garden, there is a block that exhibits different characteristics. The structures visible here tend to be larger and their arrangement less ordered. This is possibly an administrative control, in contrast to the other blocks that may have been used for accommodation, where central buildings served communal functions such as sanitary or dining blocks.

A compound with six rows of Nissan huts and a number of rectangular structures of varying size is positioned north of the main road, adjacent to the camp (Area E). The well-structured nature of this compound compared to the southwestern development of the site might suggest that it is contemporary with the primary phase of the camp’s development. Adjacent and to the east of the main camp there is an area of less well-structured development. The most prominent features appear to be the square tents observed elsewhere, organized in rows and sometimes positioned around central open areas. Also visible are probable Nissan huts and larger rectangular structures, some with pitched roofs, and open recreational areas comprising small gardens, yards, football pitch-sized areas, and a baseball pitch for sporting activities.

A distinct southeast to northwest orientated rectangular area is adjacent to the southwestern boundary of the main camp (Area A). It consists of four approximately square blocks, each divided into four linear zones, making up 16 compounds. In the ten most northerly compounds, four parallel rows of tightly organized structures are divided in two by a central path. The structures here are insubstantial when compared to those observed in the main camp and are most numerous in the northwest half. Approximate counts of 328, 361, 361, and 140 structures can be made for the most northwestern compounds. Each occupied compound has what looks like a substantial rectangular structure, possibly a communal building, at either its northwestern or southeastern end. This is separated from the rows of structures by an open area, possibly recreational space.

The four compounds located at the southeast end are empty. There is no evidence of paths or former structures, but there is what looks like a singular building similar to the “communal structures” located to the northwest. This suggests that this area had not reached its full capacity by August 3, 1945 and that logistically compounds were occupied from the northwest to the southeast as shown by the decreased number of structures. Towards the southwest end, one compound houses two rows of identically square pyramidal structures. These are identical in form with other examples in the photograph, outside the site to the west and east, and are possibly standard issue tents.

To the southwest, a less well-organized area can be observed with no noticeable boundary (Area B). As seen elsewhere, square tents and Nissan huts often situated around central open areas occupy the area. There are a number of paths and roads visible that connect to major national roads. Most significantly, three access points to the adjacent compounds to the northeast can be identified. These are the only clearly recognizable entry points into the compounds, perhaps suggesting that they were organized and managed from the southwestern area rather than the main camp.

August 1947 Photograph

The later vertical aerial photograph depicts a now redundant site (Fig. 6.2). The organization of structures and roads within the main camp is still evident as features, however to the southwest much of the area has returned to grassland, suggesting that it had been abandoned some time previously, and was originally largely accommodated by tents or makeshift structures. The basic plan of the camp does not seem to have extended since August 3, 1945. It is noteworthy that very few features can be observed within the area that has been subsequently subject to excavation. Although the archaeological evaluation only included the southwest corner of the main camp, its results indicate that more archaeological features and artifacts extent to the south.

Archaeological Excavation

Two rectangular areas (105  ×  90 and 158  ×  45 m) were stripped by machine to reveal the underlying archaeological footprint (Fig. 6.3). The ground dropped steeply from south to north, and in general preservation was best at the top of the slope, although this might simply relate to post-camp development. In the northwestern half of both zones, 15 rows of southwest to northeast oriented rectangular features (typical dimensions of 5  ×  3 m) were partially or completely revealed. In total 180 of these were found. Along the southern side of Zone 1 and through the center of Zone 2 were two lines of postholes defining a double fence line; an iron gate was identified close to this alignment in Zone 1. An area of open space was observed in between the fence line and the rows of rectangular features. This was dissected by two parallel ditches 4 m apart that ran northeast–southwest through both zones and possibly represent the remains of a track way with parallel drainage ditches. The remains of three stone-lined ovens and two possibly related soil marks were found close to the double fence line in Zone 1.

The archaeological “footprint” was significant, particularly for such a short period of occupation, and matched quite closely the evidence presented on the aerial photographs. However, only those features that were totally or partially negative left any evidence. The rectangular features can be confirmed as evidence for structures located in five separate compounds. The fence lines and the position of the track way that are also visible on the 1947 aerial photograph (Fig. 6.2) seem to reinforce the spatial organization visible on the aerial photographs. Interestingly, no evidence was found for the feature interpreted as one of the “large communal structures” seen in the compounds on the 1945 aerial photograph (Fig. 6.1). This may suggest that that they represent either a prefabricated pier-supported structure or perhaps were placed on a hard standing that was removed following the decommissioning of the camp.

Structures

In total, 98 of the 180 structures were excavated. The 180 structures on the photographs were typically between 2 to 2.3 m wide and 5 m long, but varied significantly, particularly in length. Subsequent excavation revealed sunken-featured structures (similar to medieval sunken-featured buildings) of variable depth, set in regimented rows. The strategy adopted during excavation attempted to characterize the structures in each compound to identify whether or not these had different functions and possibly accommodated different groups of occupants. The majority of structures could be confirmed as dwellings, but excavation demonstrated that there was no uniformity either within particular compounds or across the whole excavated area. While most sunken features had a consistent depth, there were some examples where only part of the floor area had been lowered, leaving a step within the structure. The archaeological footprint often represented only the negative feature within the buildings; elsewhere often little archaeological evidence survived. Most, but not all, sunken areas showed signs of badly burnt timber floors with overlying deposits of burnt soil. This indicates that they were burnt before the camp was decommissioned and returned to pasture. The general lack of structural and artifactual evidence suggests that material of value had been either purposefully removed or salvaged from the structures before their destruction.

Flooring material was regularly identified within the sunken areas and consisted of whatever material was available—metal sheets, roofing felt, wooden planks, or a combination of these. Regularly sized suspended wooden floors with floorboards nailed across joists were also identified (Fig. 6.4). The artifactual record from these structures provides no conclusive evidence of whether or not they had specific functions; however, the quality of the materials used, compared to the majority of structures, might be an indicator of status or rank. Only two structures had postholes outside of their sunken features that could have supported a roof structure. Internally, small postholes and stakeholes were found, which possibly provide evidence of internal partitions or fittings. Where sunken features were significantly deep, wood plank shuttering was used. This was most common at the entrance to structures where erosion of the pit would be more likely. Other structural remains found include fragments of window glass, part of a window frame, and a set of wooden steps. A number of stoves were found within structures, and a majority has been identified as the US Army issue—Stove, Tent, M-1941—made of press steel that came as a kit for assembly (Fig. 6.5). Their distribution was almost exclusively limited to the uppermost compound (Fig. 6.3 cages 1 and 2). In contrast, cast iron flue pipes that would have attached to the stoves were more widely found, perhaps suggesting that stoves might have been more widely distributed.

Fig. 6.4
figure 4

La Glacerie structure 2108, cage 8, with preserved floor joists

Full size image
Fig. 6.5
figure 5

La Glacerie structure 1037, cage 1, with US Army issue stove in situ

Full size image

The average area of each structure has been estimated at about 10 m2. The arrangement of interior space was evident in a number of structures, but this is not consistent. Evidence for partition walls was generally rare, but when identified it was represented by small postholes or sill beams. More commonly, divisions of space are represented by a change in floor level (Fig. 6.5). One structure had three floor levels, where the highest level (the shallowest part of the structure) was interpreted as an entrance lobby/vestibule. This was the only area of floor with a roof felt covering and its entrance would presumably have given onto the main track way through the compound to the east. A small storage pit and the setting for a stove were identified in the middle, at a slightly lower level, and this is possibly shared living space; concentrations of coffee and lemonade sachets were found in this area. The lower level, at the west end, constitutes the “deepest space” in the structure in all senses. Here, the remains of a drawer were found in between the possible location for bunks.

Other Features

The double fence line comprised over 50 postholes regularly spaced and averaging 0.25 m in diameter. Apart from the iron gate mentioned above, little additional evidence associated with the camp’s boundary was found. Scraps of barbed wire were recovered from the overburden and backfill, but presumably reusable elements of the fence were salvaged. Test excavations were carried out along the northeastern and southwestern perimeter of the camp but no ditches or further structural evidence were recovered. It is likely that the southeastern entrances to the compounds, as identified on the 1945 aerial photograph (Fig. 6.1), were outside the area of the excavations.

The remains of ovens were identified adjacent to the fence line and would have served the top two compounds (Fig. 6.3, cage 1 and 2) in Zone 1. These were in differing states of repair, with the best preserved having extant traces of limestone flue walls with a brick and tile floor. Heavily burnt natural at its northeast end suggests the setting for the fire, whereas a rectangular pit at the opposite end is the likely location for an oven chamber and chimney. Another oven was quite different in character, and the lack of surviving building material might indicate that its main body was made from metal that was salvaged when the camp was cleared. Lengths of metal pipe 70 mm in diameter, found with a crude trench-built structure with associated burning, might indicate water heating in close proximity to the ovens. There was possibly a network of pipes in this area, either very shallowly buried, or that ran on the surface; most of these would have been removed during site clearance.

The central track way running through Zones 1 and 2 (Fig. 6.3) was defined by its flanking U-shaped ditches that were 0.4 m deep and 4 m apart. Although no definitive dating evidence has been identified, it is likely that these were contemporary with the camp. Numerous bent and buckled “Marston Mats” were identified within the fills of features and it is likely that these had been used as surfaces for main communication routes within the camp.

Artifacts

Almost 4,000 finds have been recovered from the excavations at La Glacerie. These include almost 1,700 metal objects, mainly of iron but including those of copper, aluminum, and lead. Over 1,200 pieces of glass, including 587 pieces of window glass and over 400 bottles or sherds from bottles were also recovered. The excavations yielded 444 pieces of plastic, mainly fragments of sheeting. Other materials include ceramics, leather, and fabrics, including pieces of uniform and other garments, as well as those of rubber, paper, and wood. The majority of the assemblage is characterized by material that had no saleable value when the camp closed and had been purposely discarded. Nevertheless, some items recovered show the ingenuity of the prisoners in reusing materials and in some instances hoarding potentially useful items for the future. The provenance of artifacts is, as expected, attributed to either the American or German Army and Navy or from the local community where relationships between prisoners and local inhabitants established during occupation were perhaps still continued from the camp.

Clothing and Buttons

A majority of the clothing items identified were pieces of fabric, buttons, and footwear. A hoard of leather pieces and rubber heels and soles salvaged from disassembled elements from American Army boots was located, where boots, presumably worn out, seem to have been carefully taken apart to recover the leather. These were perhaps collected by a cobbler or a prisoner designated this function in the camp. Little evidence of German military boots was found, suggesting prisoners were issued with American boots. Fragments of wool, which could have come from a civilian jacket or a Kriegsmarine uniform, were found as well as fragments of herringbone twill from a pullover possibly of American origin. A majority of the clothing fragments found were khaki waterproofs used for either raincoats or ponchos. The abundance of this material possibly relates to its reuse in dwellings either as flooring or to patch up the roofs. One hundred and forty five buttons were found, most of which were either of the United States or German origin; these included six US General Service and eight Kriegsmarine buttons. The distribution of clothing over the excavation does not show any marked pattern beyond the fact that items of clothing fabric are found in a limited number of locations. By contrast buttons are widely, if thinly, spread across many structures.

Bottles and Ceramics

A significant collection of bottles, jars, ampoules, and other glass vessels have been confidently identified as specifically for medical or pharmaceutical uses, although a small distinct group had contained grooming products. In addition, a wide range of drinking bottles of American, English, and French origin were found. These include champagne, gin, whisky, mineral, and soda water bottles that did not necessarily reach the site with their contents intact, although it is impossible to be certain on this matter. More certainly interpreted are the 28 inkbottles of the United States origin that were found widely distributed (no structure has produced more than three bottles) and demonstrate that prisoners were writing correspondence on camp. In marked contrast to the glass assemblage, only 168 sherds of ceramic and two complete vessels were found. The two complete vessels included a cup printed with the badge of the US Army Medical Department and a small handless cup of beaker with white glaze and no decoration. In contrast, and of no practical use, the base of a pair of blue and white Delft figurines was found in one of the structures. The quantity of ceramic is small and has no significant distribution; only four structures produced more than 10 sherds. It can be concluded that very little china and earthenware was available to the prisoners and that personal mess kits and army issue cooking equipment were widely in use.

Metal and Plastic Items

A variety of metal objects, mainly of sheet metal, have been recovered in varying states of repair. The interpretation is skewed by the larger objects that are better preserved and the noncorroding items. The assemblage comprises buckets, jerry cans, tools (picks, shovels, pliers) circular trays, a sink with central drainage hole, numerous tin cans, a number of small aluminum dishes, and the United States and German Army issue cutlery. A significant proportion of objects have been salvaged and reused. A most interesting example is an oil drum, which has been converted into a wood-fired stove for heating a hut; rectangular openings for the stokehole and ashtray have been cut in the front of the drum and at the back a circular hole was pierced for the flue. Other examples include two small hand-made metal boxes, reused aluminum army field dressing tins and water bottles (containing a white substance either soap or wax), and a jug or mug made from a tall tin found inside a large can and presumably used to extract water. Materials collected and saved by the prisoners were found, including quantities of wire and cable, off-cuts of metal sheet, piles of used nails, and a large pile of tin cans. With the exception of plastic sheeting that was widespread, and used in the dwellings perhaps as wall linings or floor coverings, and in the case of clear plastic possibly to cover window openings, only two other plastic or Bakelite items were found.

Personal Hygiene and Medical Items

Items relating to personal hygiene and grooming are widely distributed across the excavated compounds and comprise 23 toothbrushes and 14 combs, and smaller numbers of jars or bottles containing hair care products, a part of a safety razor, a tube of shaving foam, and Colgate toothpaste. The ownership and use of these items would have been highly valued and, as an example, excavation revealed that a bottle of hair care tonic had been deliberately concealed beneath the wooden floor in one structure. Medical items are generally more numerous and show a distinct concentration in one structure in Compound 1 (Fig. 6.3). Here, a number of fragments of single dose ampoules or phials (possibly 22 phials) made from narrow glass and containing unidentified white powder were found, together with brown and amber glass medicine bottles. Elsewhere, a head of a small hospital bed, parts of four stretchers and numerous medicine bottles of the United States, French, and English origin were found across the camp. Of specific note is a tube that was part of a prophylactic kit issued by the US Army for treatment of venereal disease.

Identification Tags

Five rectangular prisoner identification tags were found made from an aluminum alloy and of American manufacture. Text was punched into the tags on two or three lines and provided the following information: camp name and its identification number/status/name/matriculation number. Two of the prisoners had the same matriculation numbers, which was presumably an error on registration. One other had an additional tag that gave his family address. Three of the tags were found together in the same structure and are numbered in sequence. Since the identification numbers would have been attributed in order of arrival, it seems likely that these prisoners had been registered at the same time and placed together in the same structure or tent. Communication with Deutsche Dienststelle Wehrmachtsauskunftstelle für Kriegerverluste und Kriegsgefangene revealed that two of the named prisoners were still alive, though neither chose to communicate their experience when invited.

Documentary History

There is an abundance of historic documentation for PoW camps situated in Normandy, and those relevant to this study are not under represented. But discussion here concentrates principally on the historic documentation that is relevant to the excavation. In this context, it is important to recognize that events happened quickly within the relatively short lifetime of the camps. Some of the documentation captures very specific information that does not necessarily provide a representative background for camps that were dynamically evolving in response to an ever changing military situation.

Events happened during the history of the camps that would have stimulated change in their organization and function, as well as affecting the mood and moral within. Locally, there are recorded changes in Commandant that produced more relaxed conditions, but conversely documentation suggests that the takeover of the camp by French authorities in the autumn of 1945 created resentment. In the testimony of one prisoner (see below), they believed that they had been sold to the French. In the later life of the camp under French authority, when fewer prisoners were interned, it is not clear whether more space would have been available or whether, as is more likely to be the case, prisoners were regrouped into fewer compounds.

Internationally, the end of the war created a “feel good factor” and arguably boosted the moral of all at the camp; however, the allied discovery of concentration camps is widely acknowledged to have affected the treatment of PoWs, as is attested in the region by Red Cross images. Despite the relatively short period of time that the camp was occupied, it is important to calibrate the timeline of events and not to overgeneralize from any single source information. What is apparent is that the camp occupants changed, and the camp evolved quickly. Archaeological evidence within a multidisciplinary approach should aim to identify and address changes that happened that directly relate to the experiences of those that occupied the camps. However, new approaches and techniques may be needed to identify the subtle traces of evidence of these events in the archaeology record.

Camps for PoWs in the region of La Glacerie are referred to by several names in documentation, and this has confused attempts to cross-reference evidence from excavation, aerial photography, and documentary research. The five identification tags found during excavation have been key in confirming that the discussed archaeological record relates to POWLE 112 (Prisoner of War Labour Camp Enclosure 112). It is now clear that there were two adjacent camps at La Glacerie—a “transit camp” and a “labor camp.”

The “transit camp” was established by the American authorities on August 1, 1944 and was officially referred to as CCPWE (Central Continental Prisoners of War Enclosure) n° 10 Cherbourg-Haute-Gringor. Its initial function was to register new prisoners that were to be relocated to the UK and America. However, with the expulsion of the German forces from Normandy and the allies’ continued success pushing east, it was decided shortly after the summer of 1944 that prisoners would no longer be evacuated abroad, and were best used in mainland Europe, where they could be put to work. The official name for the “transit camp” changed in December 1944 to CCPWE n° 20 Cherbourg-Haute-Gringor, presumably as a new official numbering system was applied to cater for additional camps constructed in the region. On August 11, 1945 the camp was transferred to the French Army and was renamed Dépôt de PGA (prisonniers de guerre allemands, ou prisonniers de guerre de l’axe) n° 301. The PoW camp referred to here was the formal and ordered camp identified south of an east to west aligned road on the 1945 aerial photograph (Fig. 6.1).

The excavation was carried out on the site of the “labor camp” referred to as Labor Service Center 137 112A and Camp de travail 137, L.S.C. APO 562 (ex 112A) in Red Cross reports. It was the distinct southeast- to northwest-orientated rectangular area made up of 16 compounds that was adjacent to the southwestern boundary of the “transit camp” (Figs. 6.1 and 6.2). References made to this camp indicate that it was managed and administered separately and, at least initially, performed a different function to the adjacent camp. This information corresponds well with evidence from the 1945 aerial photograph that suggests that the camp was managed from the southwestern area (Fig. 6.1).

The Transit Camp

Le Comite International de la Croix Rouge (CICR) visited this camp on May 5, 1945, February 19, 1946, and September 9, 1946 (Amsler 1945a, b, c; Courvoisier 1946a, b). The information that follows has been extracted from Red Cross visitor reports. During the May 5 visit, it was recorded that 17,567 prisoners were interned, 16,236 of whom were in transit. At this time, the maximum capacity of the camp was 25,000 prisoners. The camp comprised 18 compounds, though this number had reduced to 14 by February 19, 1946. Each compound contained on average 30 standard issue American tents that were shared by 35–40 prisoners. During the May 1945 visit, the camp was undergoing expansion to cater for what was by then a regular flow of prisoners. Materials were being conserved for construction and prisoners were either sleeping on blankets on the floor or thick cardboard at the time of the visit. In total, 3,000 beds had been reserved for the hospital. Tents are reported to be heated in the winter by stoves but only the kitchen, workshops, and hospital had electricity. Kitchens were constructed in each compound and were referred to as in très bon état in the February 1946 report. There was a good supply of food and a boulangerie 20 km from camp that employed 500 prisoners. All the prisoners had their own uniforms that were replaced by American issue if worn out.

The Camp was under French control during the visits on the February 1946 and the September 1946. At this time only 7,097 prisoners were interned and were accommodated in tents and barracks. Of these, 5,158 prisoners were divided up into units that worked on reconstruction and mine clearing 8 h a day (mine clearing is recorded to have claimed 3–4 deaths every 2 weeks). It is evident that the camp was no longer functioning solely as a transit camp at this time. During the September 1946 visit, 7,200 prisoners are recorded, although prisoners are now being repatriated regularly. Those in the camp were still being used for mine clearing, rebuilding, and agricultural work at this time.

The Labor Camp

Red Cross visits were made on May 7 and September 8, 1945; the camp had a capacity of between 7,500 and 8,000, but only 4,017 and 4,480 prisoners were recorded during the May and September visits, respectively. The camp’s perimeter was enclosed by a barbed wire fence and its access points were controlled by watchtowers. It was divided into 15 compounds that each interned 500 prisoners; each compound was further bounded by barbed wire fences. Structures were arranged in rows separated into two by a central walkway; the September visitor notes that walkways were covered with Landinstrip d’aerodrome (Marston Mats). Each compound had an area reserved for exercise, and a football was regularly passed around the camp. Toilets and two kitchens were reported in each compound. The excavations identified remains of the kitchens that were described as comprising a tank for boiling water, a wood cooking range with two hotplates and two ovens. However, no evidence of toilets, reported to be at the end of each compound, was found, and it is likely that these were outside the area of excavation. No communal structure is mentioned and prisoners are reported to have eaten outside or in their accommodation.

Possibly, the most significant finding in the Red Cross Reports is a comprehensive description of the structures within each compound, supported by photographs taken in 1944 and August 1945 (Figs. 6.6 and 6.7). This important evidence was only directly linked to our site by the discovery of the camp identification number on five identification tags.

Fig. 6.6
figure 6

Red Cross photograph taken at the La Glacerie camp, August 1945

Full size image
Fig. 6.7
figure 7

Red Cross photograph taken at the La Glacerie camp, August 1945

Full size image

Les habitations pour 4 à 6 personnes ont été construites par les habitants avec du matériel de fortune et du bois mis a leur disposition. Toutes ces maisonnettes sont creusées dans le sol à une profondeur de 1m50 environ avec une toiture dont la partie basse fait corps avec la terre remblayée. La fantaisie la plus grande a préside à la construction de ce village. Chacun semble avoir mis une pointe d’orgueil à posséder une <<villa>> plus belle que celle de son voisin et il en résulte une grande diversité très picturesque (Amsler 1945a, 1).

This account helps to explain the atypical form of the structures excavated. At the creation of the camp in October 1944 the prisoners had to sleep in small tents supplied by the American Army, referred to as “pup tents.” With the severe winter of 1944/1945 and their prolonged detention, prisoners did their utmost to improve their living conditions. In time, and with a sense of pride and possibly competing with one another, prisoners endeavored to create their perfect “villa.” Photographic evidence demonstrates that the individual characters of the prisoners have contributed to the reinvention of their confined space. Unlike the neighboring camp, a sense of place was created that reinforced group identity and in cases made strong references to the prisoner’s homeland.

No photographic evidence was identified that depicted the interior of the ­dwellings, but this account below highlights the ingenuity of the individuals that created them.

L’intérieur de ces habitations de laisse pas d’être à la mesure de l’extérieur. Il est marque au coin du bon goût et de l’ingéniosité des occupants. Le sol est recouvert de bois et des châlits simples ont été construits avec les moyens qui s’offraient, tout comme les divers meubles (tables, chairs, <<cosy corners>>, armoires) qui garnissent la pièce. Partout des fourneaux. L’aération par les petits fenêtres est excellent (Amsler 1945a, 1).

It is the archaeological record that provides detail for the organization of interior space. It is particularly interesting that structures rapidly evolved from tent to “villa.” The “causal factures” (Barrett 1994: 88) of their development rapidly shift from environmental and technical necessity towards a social and possibly symbolic use of space. These interiors may, also, have made reference to the prisoners’ homeland where the tradition of semi-sunken buildings, either as dwellings or workshops has persisted in northwestern, central, and eastern Europe well into the twentieth century (Hamerow 2002: 34–35).

Evidence from a Prisoner

In 1957, the German government began to interview former PoWs and the results were published almost a decade later (Maschke 1986). A publication based on these findings has been identified that is a testimony of a PoW in Labor Service Center 137 112 (Schröder 1997). Extracts have been chosen here to cross-reference against findings made from the other areas of research. This is clearly a significant source of evidence and other accounts may also exist.

The general layout of the camp was described in brief, confirming that it was split into compounds and that each had between 25 and 30 tents and held between 150 and 180 prisoners. Compounds were enclosed by 2 m high barbed wire fences with barbed wire rolls on top. As is known from the system of administration in other PoW camps (Waters 2004: 22), one PoW per compound was in charge and reported to a US guard. When prisoners were sent to the camp, six randomly picked men were selected to live together. They were given a tent canvas and poles that they had to erect themselves; there was no ground sheet. They were also given a US military coat with PW written on the back. This was used as a blanket at night. The account recalls that the tents were not waterproof and that there was a sleeping rotation system in practice, as those that slept by the sides of the tent often became wet at night when the tent touched them. Within a short time prisoners made floors from flattened corned beef tins covered in cardboard. Tins were initially one of the only building materials available to them. Using tin cans as shovels under the cover of darkness, prisoners began to dig inside their tents. Excavated soil was used to create walkways. Only sometime later were prisoners allowed to build larger structures, but in the account it is mentioned that only cardboard was allowed as a building material.

Most of the prisoners worked in the harbor. The day started at 5 o’clock, breakfast was at 5 30 and at 6 o’clock between 4,000 and 5,000 PoWs were driven down to the harbor. It is mentioned that during the unloading of ships, the US troop rations were stolen. Stolen tools were used to construct their dwellings, stolen cigarettes were exchanged for chickens and prostitutes, smuggled into camp wearing the US army issue PW overcoats, were also paid with stolen goods. There was a special compound for those caught trying to flee the camp, where the account recalls that prisoners were nearly starved to death and were only given thin clothes even during winter months.

It is clear from this testimony that the prisoner was one of the first arrivals at the camp and possibly left before the camp closed. It is particularly interesting to have an account of how structures rapidly evolved from tent to a more solid construction, although this is not as positive and vivid as the descriptions in the Red Cross visitor report. These two archives cross-referenced against the archaeological record represent different subjective narratives that can be used to recount life and experience inside and outside the camp at different times of its development.

Oral History

An oral history project ran parallel with the fieldwork program. Of 20 volunteers identified to share their memories, only 5 were found to have a true knowledge of either camp. The interviewees were young when the camp was built, the oldest was 16 years old, and consequently have retained childhood memories that are often vague and lack detail. None had visited the labor camp or passed the main entrance to the transit camp. The narratives produced, therefore, focus on personal relationships that were formed with their families and represent an “outsiders” view into the camps.

According to several accounts, the local inhabitants had a good rapport with the German prisoners and the American guards. Some of the interviewees had worked alongside prisoners in either local factories or farms and recount positive experiences. One recalls how he had met a prisoner in the local factory where four prisoners worked. He recounts how he would have liked to have seen him again after his presumed return to Germany suggesting a good relationship had developed. Another remembered a 17-year-old prisoner who was escorted back and forth to work from the camp to his farm by his brother, as part of an agricultural displacement scheme that operated at the camp. His family eventually invited the prisoner to stay at the farm, as did many others, according to records. One interviewee’s family knew one of the Commandants of the camp, a French Canadian, whom they meet when he was looking for a farm to take kitchen waste from the camp. They developed a strong relationship with him and he often visited them in the evenings. The Commandant’s family was successfully contacted during the 60th anniversary of the D-day landings.

As would be expected, very little detail regarding the plan and layout of the camps was found in this source of information. There was mention of barbed wire fences, tents, barracks, watchtowers, workshops, kitchens, toilets, and hospitals, but not sufficient detail to securely place these elements, in either of the camps, within a meaningful and reliable context. Accounts were often non-corroboratory and more reliable information is found in the Red Cross reports, personal testimony, and archaeological record. Of interest, however, is an account of fields being transformed into a PoW camp equipped with barrack blocks and occupied by prisoners in a matter of 15 days. A reference was also made to a general sale that occurred once the camp had been dismantled. This perhaps explains the general lack of diagnostic material (such as fencing, barbed wire, etc.) identified during excavation. Another account mentions that hardcore was transported from recently surfaced roads that led to a V1 and V2 German launch site to surface tracks within the camp during periods of wet weather. It was also stated that the Americans did not mix with the prisoners and had their own adjacent camp. This is presumably a reference to the distinct areas located on the 1945 aerial photograph to the southwest of the labor camp and/or northeast of the transit camp.

These living memories remind us that the influx of both prisoners and Americans after the creation of the PoW camp at La Glacerie had a significant impact on the personal histories of members of the local community. There was an active dialogue between prisoners, Americans, and locals that was also implied by the diverse artifactual assemblage recovered during excavation. Together with the testimony of a prisoner (see above), the oral history accounts undoubtedly remind us of the importance of the wider significance of PoW camps, providing an outsider’s perspective as well as the inside view. Economically, these camps would have had significance in the local landscape and made lasting contributions, at some considerable sacrifice, to the rebuilding of postwar France.

Conclusion

Considering the accidental discovery of the PoW camps at La Glacerie, the subsequent pluridisciplinary approach to this project has been surprisingly informative in identifying a series of quite distinct narratives and perspectives to a unique period in the history of Normandy. It is revealing, however, that within living memory a well-documented site with a substantial archaeological footprint that in history had a significant impact on its local community can be lost in the memories of a generation. While our own points of reference to this difficult period in history change and the historic record is interrogated to provide new interpretation, it is important to remember the significance of the archaeological record. During this study, we have found that military documentation as a single source is imprecise, and arguably misleading when researching a specific site, and that oral history, soon no longer available for this period, often provides subjective and vague information. It has been the archaeological evidence for this site that has provided us with an objective record that in turn binds together our other sources of evidence.

It is significant that La Glacerie is the first PoW camp to be excavated in Normandy of many documented (Schneider in prep.; see also Chap. 7). The PoW camps in this region were created at a defining period of international history and as a group is distinct from other PoW camps located around the world. It is arguable, therefore, that they should be recognized together as heritage of regional significance and as important components of France’s war and postwar history. The excavation and research at La Glacerie will raise awareness of these sites’ importance in this region. It is hoped that archéologie préventive will continue to recognize the need to undertake PoW camp research.

In terms of the archaeological methods employed, it has been interesting to note that the machine stripping of such a large area, as is commonly adopted in modern archéologie préventive, has been particularly valuable in assessing spatial organization, particularly if we consider that aerial photographs may not always be available for other sites. However, it should be noted that archaeological practice often provides evidence for occupation that has occurred over large time periods, particularly in prehistory. It is, therefore, necessary to rescale methodologies for PoW camp archaeology where events significant to our research happened quickly within the relatively short lifetime of the camps. Consequently, more detailed excavation of the structures could have resulted in more useful information regarding their evolution from tent to structure, as well as providing valuable information regarding individual’s use of space and personal experience at the camp. From the outside and as viewed from the aerial photograph, the military design intent of the transit, and arguably the labor camp, is clear and as one would expect at a PoW camp. However, at the labor camp more detailed analysis demonstrates human interaction with the planned design, where individually shaped structures outwardly express a distinct character that connects us to their architects and occupiers. It is noteworthy that there is such a contrast in neighboring camps, which is presumably a result of transient occupation in one and more permanent occupation in the other. It would be interesting to explore this evidence for more relaxed governance, with other similar sites in the region and elsewhere. The dense archaeological footprint identified was unexpected particularly when considering the evidence from the 1947 aerial photograph (Fig. 6.2). It will be interesting to see how excavations of contemporary archaeological sites that are associated with good documentary sources could be used to appraise archaeological interpretation for prehistoric sites and others that do not have a documentary source. The sunken features at La Glacerie are reminiscent of Anglo-Saxon sunken-featured buildings, grübenhauser (Hamerow 2011). The known short period of occupation here could raise questions regarding interpretations of temporary and permanent settlement elsewhere in earlier periods.

Making sense of the abundant documentary sources in the contemporary period creates new challenges, particularly for archaeologists unused to such a wealth of material (Shanks et al. 2004). There are clearly more historic resources, including additional prisoner testimonies, which could be collected and researched relating to the camps at La Glacerie. Well-defined research objectives in PoW camp studies will be required to ensure that the collation of this wealth of information contributes significantly to research. Full analysis of the archaeological record will provide its own objective narrative and is arguably best used in PoW camp studies to cross-reference and contextualize the historic archive.