Kongens Nytorv, KBM 3829

Transcription

1 KØBENHAVNS MUSEUM/ MUSEUM OF COPENHAGEN/ ARCHAEOLOGICAL REPORT Kongens Nytorv, KBM 3829 Cultural Historical Report, Metro Cityring Excavation Morten Steineke and Jane Jark Jensen

2 Museum of Copenhagen Vesterbrogade København V Telefon: Fax: museum@kff.kk.dk Cover picture: The Kongens Nytorv excavation. The eastern gate building looking SW. Photo: Museum of Copenhagen Museum of Copenhagen 2017 Museum of Copenhagen 2

3 Foreword The excavations in connection with the Metro Cityring project have been the largest ever archaeological project in the history of Denmark. The results have provided brand new and extensive knowledge about the history and development of the capital of Copenhagen and the former inhabitants. The excavations at the site of Kongens Nytorv carried out between the years was one of the three main sites and were located partly within the borders of the medieval city. At Kongens Nytorv the archaeological results not only confirmed the knowledge from previous excavations, but also brought to light very significant new information. The most eye-catching features from the excavations were the discovery of the fortifications and the well preserved remains of the eastern gate to the city. The uncovering of boundary ditches, clay lined pits, workshops and domestic buildings dating from the Early medieval period revealed that the area was part of a settlement that was important enough to fortify, however, it was not until the Late medieval/post medieval period that this area developed greatly. The finds material is unique and due to the anaerobic conditions found in the moats, the well-preserved organic artefactual reveal the more diverse side of the lives of the inhabitants of historical Copenhagen. The results are presented in a technical report containing the most important discoveries and a comprehensive appendix with the descriptions of all archaeological features follows this report. In addition a cultural historical report based on the technical report has been produced with the focus on the results in a cultural historical perspective. The expectations are that the reports will be read both by professionals and the general public, but also used for research and comparable material for many years to come. Lars Ewald Jensen, Archaeological Leader, Museum of Copenhagen Museum of Copenhagen 3

4 Contents Cultural Historical Report 9 Introduction to the excavation 10 The location 10 The excavation 11 Excavation areas 12 Regional geology and topography 13 The Metro excavation Before the fortification AD 16 Medieval fortification AD 18 Post medieval fortification AD 21 Christian the 4th's fortification c AD 22 Other activities and buildings behind the fortification line 23 Time Phases at Kongens Nytorv 24 Phases 24 Chronological development 25 Finds 27 Type of finds 27 The finds 28 General reflections on the finds material 34 Moraine, salt marshes and other Quaternary observations 35 Results 35 Prehistoric finds and features 37 Results 37 Stone Age finds from central Copenhagen 39 A Late Mesolithic activity spot or camp site at Kongens Nytorv 40 Early medieval activities AD 41 Results 41 Museum of Copenhagen 4

5 Ditches 43 Clay lined pits and a so-called Grumbod 44 Pits 51 Pit house 52 Post- and stakeholes 55 Levelling and activity layers 55 Boundary ditches and the establishment of plots in the eastern part of the city 56 Østergård and Byens Tårn 59 Herring fishery and clay lined pits 61 High medieval fortification AD 64 Results 64 Rampart 65 Bulwark building material and construction details 67 Fence lines 72 Wattle fence in the 13th century moat 73 Stockade or part of Byens Planker 74 The High medieval moat 75 Østervold and the High medieval fortification of Copenhagen 76 Rampart 76 Bulwark 80 Byens Planker 82 Late medieval city wall AD 84 Results 84 City wall building material and construction details 84 The construction of the city wall 90 Maintenance work and final demolition 96 Eastern gate building AD 98 Results 98 Museum of Copenhagen 5

6 Eastern gate building building and construction details 98 Revetment and road surfaces outside the gate building 114 Customs- or guard building next to the gate building 118 Østerport the eastern entrance to the city 120 Late medieval and Renaissance fortification AD 127 Results 127 Bulwark 128 The Late medieval moat construction, usage and deconstruction 129 Work and modification on the fortification in the 16th century 132 High and Late medieval settlement and activities AD 136 Results 136 Buildings 137 Boundaries, ditches and fences 143 Streets and getting around in the area 150 A well, possible wells and a bucket 152 Pits, postholes and different layers 155 Postholes 156 Eastern gate building and Post medieval fortification AD 157 Results 157 Østerport additions, activities and deconstruction 158 Bulwark and the demolition of the gate building 162 Roads and pathways along Østervold 162 Bastion, curtain, bridge and moat 163 Dam with barrier tower 190 Temporary bridge in the Late medieval moat 198 Christian the 4th and the rebuilding of the new fortification 203 Østerport 203 Østervold 205 Museum of Copenhagen 6

7 Settlement and activities behind the rampart AD 211 Results 211 Buildings in the area around Hviids Vinstue 212 Buildings in Lille Kongensgade 215 Other building activities 217 Street and pavements 218 Pits, postholes and stakes 219 Ditches and a water pipe 219 Barrel and wooden box 221 Activity, dump and levelling layers 223 Outside the moat. Settlement and activities AD 224 Results 224 Buildings and stone structures 225 Roads and wheel ruts 246 Fence- and stake lines 248 Pits, ditches and boulder 248 The suburbs outside the city gate 249 Buildings, courtyards and intermediate streets 249 Kongens Nytorv AD 254 Results 254 Lille Gjethus 254 Hviids Vinstue 255 Other building remains 256 Road and street surfaces 260 Wooden water pipes 260 Kongens Nytorv from defences to the city s new square 263 Buildings 263 Kongens Nytorv and Krinsen 265 Museum of Copenhagen 7

8 Roads and streets 266 The city water supply 266 The Scania market 268 Urbanisation in the eastern part of Copenhagen/the area of Kongens Nytorv 271 Medieval period 271 Post medieval period 275 Conclusion 278 The eastern gate and fortification through time 280 The medieval gate building AD 280 The medieval fortification AD 283 Rampart and bulwark 283 The city planks 286 The Late medieval city wall 288 Late medieval fortification building operations and maintenance work 292 Christian the 4th s new fortification and gate building 294 The new gate building 294 Østervold 295 Assessment of results 301 Site specific questions 301 Objectives of the project in the light of the project aims 302 Background, organization, direction and characterization of urbanization 302 Economic and demographic fluctuations 302 Cultural and social implications and consequences of city life 303 Future potential 304 Bibliography 305 Literature 305 Archaeological reports and archive material 309 Internet references 314 Museum of Copenhagen 8

9 Verbal information 314 Abbreviations 314 Museum of Copenhagen 9

10 Cultural Historical Report The purpose of this report is to place the results of the excavations at Kongens Nytorv in a broader cultural historical context within the wider frame of the historical development of Copenhagen with focus on Kongens Nytorv, and to do so in a format accessible to the general public. In this report the main results will be presented and a cultural historical perspective will be added to the results. During the medieval period Kongens Nytorv was situated on the eastern outskirts of the medieval city but after expansion of the city in the 17 th century, the square came to play a central role in the city; a role it continues today. Archaeological observations have been undertaken at Kongens Nytorv over the last 100 years and after planning of the first metro station occurred a full excavation began in 1996/1997. The excavations revealed that whilst fortifications were the dominant feature, remains such as boundary ditches were uncovered providing settlement evidence of the area before the fortification. This led to high expectations for the excavations for the discovery of both medieval and post medieval fortification as well as traces of earlier settlement prior to the fortification. An area of nearly 7000 m 2 was excavated as part of the recent Metro Cityring Kongens Nytorv project and despite the previous knowledge of archaeological remains in the area, the recent excavations exceeded expectations. The most important new knowledge obtained from the excavation was the uncovering of the city s eastern gate building (Østerport), which was found to date from the early part of the 13 th century. Important findings also include the discovery of boundary ditches from the early medieval period, which could help define the settlement and crafts in the area from the early medieval period until present day. Also, the uncovering of settlement evidence outside the moats from the early 16 th century came as an unexpected revelation, portraying life outside the city boundaries. This cultural historical report is based on the main excavation report that conforms to specific KUAS guidelines (Steineke and Jensen 2017). The main report complies with statements in Danish Museum law (Bekendtgørelse af museumsloven nr. 1505). The main author of the report is Field Leader Morten Steineke (all Chapters except Chapter 1, 2, 6, 7.10, 8, 9, 17, 19 and 22) which was written by Excavation Leader Jane Jark Jensen). The final editing and other report related work has been done by Morten Steineke. Responsible for figure production has been Karen Green Therkelsen. The texts have been modified and shortened to a more readable form. This report contains many chapters split into various themes of the archaeological results both chronologically and thematically. Two of the main chapters focus on the Scanian Market (Morten Steineke) and the urbanization of the Kongens Nytorv area of Copenhagen (Jane Jark Jensen) and the other one on the medieval and Post medieval fortifications (Morten Steineke). Lastly it should be mentioned that a chapter on the first Metro excavation at Kongens Nytorv is applied to supplement the knowledge about the area. Museum of Copenhagen 10

11 Introduction to the excavation The excavation at Kongens Nytorv took place in advance of the establishing of a new Metro station. This is a part of the new Metro Cityring system that will provide a new transportation system to the areas of the city surrounding the city core (Fig. 1). The Cityring is eventually going to be connected to the existing Metro system. Kongens Nytorv will be one of two already existing stations where the new metro line will meet the old one. The establishing of an adjoining Metro station at Kongens Nytorv would affect high potential archaeological remains and rescue excavations were necessary in accordance with the Consolidated Act on Museums. Fig. 1. Plan of the Metro Cityring in Copenhagen with location of Kongens Nytorv marked. The excavation of the main area started in December 2009 and ended in September The main area of investigation was the location of the future access to the new metro station and in a high potential area of archaeological interest. In 2012, 2013 and 2014 the excavation of the Station Box was conducted under archaeological surveillance. In 2016 an access from the Royal Theatre was established and again this was conducted under archaeological surveillance. The Museum of Copenhagen is responsible for the archaeological work carried out in Copenhagen and Frederiksberg municipalities and therefore conducted the archaeological work on this site. The work was carried out under Part 8 of the Consolidated Act on Museums. This means that the contractor must finance the cost of the archaeological work (registration, observation and documentation) in connection with destruction of archaeological remains. The location Kongens Nytorv (translates into the King s New Square) is today a central square of the inner city, with Krinsen as the central area. This was not always the case; the eastern city gate of the medieval fortification was placed here on the Museum of Copenhagen 11

12 first brick built fortification surrounding Copenhagen, but neither written nor do archaeological sources so far seem to indicate that the area was used to its potential in terms of close settlement or activities. All knowledge up to the present day seems to indicate an area on the outskirts of medieval Copenhagen with limited activities. The square of Kongens Nytorv was originally founded in 1688 and placed just east of the medieval city and just west to the New-Copenhagen which was laid out by King Christian the 4 th when he expanded the city that outgrew its medieval boundaries in the beginning of the 17 th century. The square was laid out as a baroque garden (named Krinsen) by King Christian the 5 th with an equestrian statue of himself as a central figure and the square was lined with mansions. Two of these 17 th century buildings still stand today and the square is also home of the Royal Theatre and the oldest existing pub in Denmark Hviids Vinstue, founded in For a period between 1749 and around a hundred years after the garden was closed down and the square functioned as a military drill ground. Krinsen was reconstructed back to its original baroque style in 1908 after the original square. With the large mansions surrounding the square, today s Kongens Nytorv has preserved its original characteristics from the 17 th century (Fig. 2). Fig. 2. Krinsen and Kongens Nytorv before the excavations in October Photo: Morten Steineke. The excavation A prioritization of the large scale excavation area at Kongens Nytorv was necessary in order to keep the archaeological budget at a reasonable level, but also according to the different types of archaeological features, that were expected in the area, that would demand different kinds of excavation methods. On the basis of what was known about the area from both archaeological and written sources, a plan of prioritization was laid out (see below). Some knowledge Museum of Copenhagen 12

13 of Kongens Nytorv was gained from earlier archaeological excavations; in particular the excavation for the first Metro station at the square excavated in (see Chapter below). From the results of this excavation the museum knew that there would be potentially quite significant remains of the fortification surrounding Copenhagen in the medieval and Renaissance period. At this excavation also remains from the 11 th century were retrieved. Excavation areas In order to prioritize the archaeological work the investigations were separated in four different areas with different potential based on what was known about the area from historical and archaeological sources (Fig. 3). This resulted in the following prioritizing with type 1 and 2 as high potential areas and 3 and 4 as representing less potential. Type 1 The high priority area with the highest expectations of complex archaeological findings. This was the area that was behind the medieval fortification and which was expected to contain remains from the early medieval period and maybe even Viking Age. This would include settlements, ditches, wells, etc. This area was fully excavated to the geological layers and mainly by hand. Type 2 The medieval fortification and the area just outside. This would include walls, embankments, moats, etc. Less complicated archaeology was expected in this area due to larger and more recognizable archaeological remains. The archaeological fieldwork in this area was often assisted by machine power due to less complex layers. Type 3 Mostly watching briefs including the Station Box. As well as in the type 2 area, larger archaeological findings were expected and also of primarily post medieval dating. In these areas the excavation was done by machine to the depth needed by the contractor and surveyed by archaeologists. When archaeological remains were detected the archaeologists registered the remains at a fairly high speed and in less detail. Type 4 Watching brief areas where mostly less complex and larger features like levelling layers were expected. Most of the type 4 area was on the outskirts of the Kongens Nytorv area and expected to be of a relatively young age. When archaeological remains were detected the archaeologists registered the remains at a fairly high speed and in less detail. Fig. 3. Map of Kongens Nytorv with the type 1-4 areas. Museum of Copenhagen 13

14 Regional geology and topography Copenhagen is situated on ground moraine gently undulating from 0 to 30 metres above sea-level. To the north the region is hilly and partially wooded with several large dead-ice lakes. Late-glacial gravel and sand outwash deposits cover the hilltops to the north. To the south the region is dominated by a low open clay-till plain called Heden. Here the region is generally flatter with fewer moraine ridges, Valby Bakke being a notable exception. Geotechnical surveys around Kongens Nytorv have recorded the stratigraphy to a depth of 56 m. Danien chalk and bryozoan limestone form the base of the sequence up to a depth of about 13 m. This is overlain by about 10 m of alternating layers of clay till and well sorted coarse outwash sand. The post-glacial deposits are comprised of marine silt and/or sand overlain by detritus gyttje and/or humified peat. In some sequences it is unclear whether the silt and fine sand layers originated during the Littorina Transgressions or as Late-glacial Diluvium (Crone & Koch 1965:3; Bahnson 1973:2-3). The landscape around Copenhagen has risen since the last Ice Age leaving the area as flat and low lying moraine with scattered hills. In the Middle Ages the area consisted of a natural coastal bay sheltered by the island of Amager and the islets of Slotsholmen and Bremerholm. The Littorina Sea existed around BP and included as many as four transgression and regression cycles producing four relative stationary shorelines during high-stands in the Late-Atlantic and Early-Sub-boreal periods. The determination of the shoreline displacement curve and the Littorina Transgressions are based on dated material from prehistoric settlements and graves, stray finds of flint tools together with pollen analysis, diatom analysis 1 and radiocarbon dating 2 of material within and beneath the Littorina deposits. However there is still some uncertainty about the isostatic uplift and the eustatic ocean fluctuations. 3 The maximum high-stand during the late-atlantic Transgression at 4300 BC in the Copenhagen area has been estimated to lie between kote +3.2 and +4.1 m (Christensen 1995). During this transgression maximum, Copenhagen was divided by a narrow strait from Øresund that extended through the "Lake Valley" formed in the present location of Sortedamssøen, Peblingesøen and Skt. Jørgens' defensive reservoirs. This strait connected to Kalvebod Strand near the preset location of Vesterbro Torv in the south and near the present position of Frihavnen to the north. The paleo shoreline lie approximately where Kronprinsessegade is today. Refshaleholmen, Plantholm (Mågeøen), Bremerholm, Strandholm (Slotsholmen), Skarnholmen and all other islets between Zealand and Amager were all covered by water (Christensen 1963:2-3). After the Littorina period the sea retreated gradually to the coastline we know from historical times. According to estimations made by geologists, the shoreline around 1000 AD is believed to have been at kote +2.0 to +2.5 m with a course just east of the Royal Theatre, northeast of Charlottenborg and through the Amalienborg area (Fig. 4). It can not be excluded that the shoreline may have stretched to the north of Vingårdsstræde near Lille Kongensgade (Crone & Koch 1965:2; Fabricius 1999:Fig. 33b), though as seen on Fig. 4 this suggestion is uncertain. It must be pointed out that shallow marshy areas with low relief coast, such as Copenhagen, were frequently flooded. These floods produced a variable shoreline. 1 Diatoms are a major group of algae and among the most common types of phytoplankton. Diatoms are good indicators of a range of water quality variables and indicators of water salinity. 2 Radiocarbon dating (C-14 dating) is a method for determining the age of an object containing organic material by using the properties of radiocarbon ( 14 C), a radioactive isotope of carbon. 3 Land and sea level changes. Museum of Copenhagen 14

15 Fig. 4. Estimated shoreline around 1000 AD and previous geological and archaeological observations of peat layers (former salt marshes) around Kongens Nytorv before the excavations in The shoreline of the Early Middle Ages has not been completely identified as hardly any excavations have revealed information on this matter. The early 13 th century shoreline has been ascertained from Løngangsstræde in the west end to Højbro Plads in the east (Skaarup 1999:81). It is generally thought that marshy, shallow areas existed in many parts of Copenhagen in these centuries, and that these may have flooded frequently, resulting in a fluctuating shoreline. Analyses of pollen and macrofossil samples 4 suggest that the Copenhagen area during the 11th century was dominated by meadows with reed swamps and wet meadows. This environment in the upper coastal intertidal zone between land and salt water or brackish water produced peat layers of varying thickness from 0.2 m to over 1.0 m, depending on local depressions, stream courses etc. (Christensen 1963:3 et seq.). Earlier Metro investigations proved traces of this wet meadow vegetation survived in the southern part of the excavation area, which indicated that the land had been flooded sometime during part of the 13 th century (Kristiansen 1998:41). Copenhagen's original topography has changed much during the last 1000 years. Refuse from households, livestock and industrial production has been dumped to fill and grade the coastal terrain. Debris from large fires in 1728, 1795 and 1807, produced when the remaining ruins were demolished to provide room for new buildings, has contributed to the construction of thick cultural layers. The city grew from a small medieval settlement on the Kalveboden strait, 4 Preserved organic remains. Museum of Copenhagen 15

16 sheltered by the island of Amager and former Slotsholmen and Bremerholm islets. Also, islets and small bays have been capped and filled, and the town has broadened far beyond the original shoreline (Christophersen 1985:69). Museum of Copenhagen 16

17 The Metro excavation During the first phase of the Metro development in the parking east of Magasin towards Holmens Kanal (c m 2 ) was excavated including large parts of the fortifications and early settlement (Kristiansen 1998; 1999a-c and Fig. 5). The dominant feature was the city's fortification, but it also uncovered a number of different features that can be linked to the settlement in the area before and after the fortification. It was possible to divide the constructions and structures into five main phases. Fig. 5. Overview of Metro excavation in , facing west. Photo: Museum of Copenhagen. Before the fortification AD Except for a C14-dating 5 from a natural deposit dated to the Stone Age, the earliest activities were represented by several north-south orientated boundary ditches in the northernmost and western parts of the excavation area and about 10 m east of Magasin's department store facade. The ditches probably formed boundaries between a property and the adjacent salt marshes. Oldest were three ditches which replaced each other. After use the latest ditch was followed by an at least 11.5 m long and m high wicker fence and a feature interpreted as some form of boundary bank to the west, at least 0.6 m high and maximum at 5.0 m wide. The stratigraphical relations between these two features were unclear. In the middle and southern parts of the excavation area only one single ditch was documented. 5 Radiocarbon dating (C-14 dating) is a method for determining the age of an object containing organic material by using the properties of radiocarbon ( 14 C), a radioactive isotope of carbon. Museum of Copenhagen 17

18 The ditches were orientated as good as parallel with the current plot boundary at Magasin s eastern facade with a varied length between m, width m and a depth between m, the latter based on the bottom measurements. The sequences of the ditches profiles changed, being slightly rounded to the north and wide to flatbottomed with gently sloping sides to the south. South of the oldest ditch, and the extension of this, a series of elongated pits were documented without further determination. The fence consisted of horizontal wicker. Some of the vertical poles were still in place, but the majority had been drawn up and the fence tipped over to the east (Fig. 6). Fig. 6. Excavating north-south running wicker fence. Photo: Museum of Copenhagen. A total of nine radiocarbon datings were completed in connection with the boundary ditches. Seed from fill in the oldest ditch was C14-dated to AD 6. Seed from one of the elongated pits was C14-dated to AD and gyttje from the bottom of a ditch was C14-dated to AD. Twigs and a wood artifact was C14-dated to AD, but the date is highly likely to be second half of the 12 th century/early 13 th century, partly due to CBM (Ceramic Building Material) in a structure stratigrapically older than the ditch and the reservoir effect on the gyttje sample. Traces of flax retting and animal bone in a ditch could be dated to late 12 th century/early 13 th century. A bundle of flax from the same ditch was C14-dated to AD and animal bones to AD. Collected ceramics were 6 The following calibrated radiocarbon dates are presented using the 2-sigma values which account for 95.4% of the probability of the date falling within that particular range. All dating from previous excavations has been calibrated using OxCal version 4.2. Museum of Copenhagen 18

19 dated to AD, AD, AD and as late as AD and AD. A later contamination could be the case since several of the ditch fills consisted of red bricks, but without further description or information. A branch and wood from the wicker fence were C14-dated to AD and AD, where the latter dating is most likely. Twigs and wood chips in a layer of seaweed were C14-dated to AD. The boundary ditches and the wicker fence were disused in the last decades of the 13 th century, and then covered by salt marshes and a faeces layer from humans and animals, where wood and twigs were C14-dated to AD, AD, AD and AD. A handle made of deer antler with animal ornamentation recovered from this rubbish layer could be dated to the 11 th century (Fig. 7). After this the area was covered by the facilities of the city's fortification. Fig. 7. Handle of deer antler found at the first Metro excavation at Kongens Nytorv (KBM 1410) and dated to the 11 th century. Photo: Museum of Copenhagen. Due to the excavation s orientation, the supposed settlement in the west was affected only to a lesser degree by the survey. Some layers and observations in the NW part could be interpreted as remains of buildings consisting of floor layers and imprints of a north-south orientated foundation belonging to a building dated to the s, a northsouth orientated base plate, a fireplace, pits and trenches. Three of these pits were interpreted either as postholes, board- or erosion holes (?). Activity layers registered against Magasin s facade in the same way as the levelling and rubbish layers over the boundary ditches indicated different types of activities from the 11 th century, perhaps as early as the first half of the century, though this suggestion is based on only one C14-dating and the Viking Age handle. Medieval fortification AD The medieval rampart's oldest range was documented in the western part of the excavation area approximately m east of the current facade of Magasin. Due to the orientation of the excavation area, the rampart and city wall were predominantly documented using profiles where different and interpreted phases and construction details were compared. Only smaller areas were excavated contextually. The High medieval rampart was divided into 14 more or less identical and chronologically contemporary sequences and could be followed for at least 75 metres with a minimum height of 1.7 m, where the top had been dug away by modern truncations. The foot of the structure was recorded from 4.5 to 6.0 m west of the foundation of the city wall, to the east the distance was 2.6 to 3.6 m to the edge of the moat, which made the width between m. Museum of Copenhagen 19

20 No reliable growth horizons were recorded either archaeologically or through macrofossil analysis and the building material consisted of sub-surface material (moraine) mixed with organic material (peat and topsoil), with the exception of one deposit that to a greater extent consisted of brick rubble. Only a few finds were collected consisting of ceramics dated to late 13 th century, 13 th 15 th century and 14 th (15 th ) 17 th century. The city wall was interpreted to be contemporary with the oldest rampart, and not a later addition. The courses and the rampart layers had been built step by step, where the purpose of the rampart had been to stabilize the boulders. No part of the brick wall was preserved. The foundation of this medieval structure could in a more or less fragmented extent be recorded in a large part of the excavation area. The city wall was preserved as three foundation courses and boulders consisting of two rows of stones with a width between m and a height of at least 1.4 m. The top of the foundation was not preserved and no mortar was recorded on the stones. Part of the city wall, consisting of a continuous remnant of a brick wall at least 1.8 x 1.2 m with medieval course was documented in the moat together with bricks and limestone ashlars. The remnant was interpreted as part of the parapet. The wall had been 0.42 m thick with unfilled joints. No reused bricks were identified in this part of the brick wall. Three thermoluminescence dates 7 from mortar connected to the remnant were dated to 1227±60 AD, 1255±60 AD and 1230±60 AD. Inclusions of smaller 16 th century bricks together with medieval bricks ( munkesten ) in the decomposition layers in the Late medieval moat and within the new rampart in the southern part of the excavation area, were an indication of repairs and extensions of the wall. The interpretation was also that the city wall mainly had been demolished in connection with Valkendorf s work on the fortification in the late 1500s. The High and Late medieval moat could be followed the entire length of the excavation immediately east of the city wall foundation (Fig. 8). This could, however only with certainty be separated from a later post medieval moat phase in the moat s westerly part located under the rampart (see above). The moat had been cleaned up through the Middle Ages and estimated to have been at least 1.4 m deep. 7 Thermoluminescence (TL) dating is the determination, by means of measuring the accumulated radiation dose, of the time elapsed since material containing crystalline minerals was either heated (lava, ceramics) or exposed to sunlight (sediments). Museum of Copenhagen 20

21 Fig. 8. Profile through Late medieval moat, facing SW. Photo: Museum of Copenhagen. Archaeobotanical analysis 8 showed that the environment had consisted of brackish, stagnant water. A weak salt effect showed that the mud had been formed before the filling up of Dybet between the coast and Bremerholm in the s. At the excavation a north-south orientated palisade used as alternative rampart reinforcement at the edge of the moat was documented. This consisted of several pits and postholes in a row at an estimated length of either 42.0 or 64.5 m. A bigger structure was recorded orientated across the rampart and consisting of two to three courses of foundation stones and a 6.7 m long, 1.0 m wide and 1.7 m high double faced wall of edge set courses, coated with lime mortar and with intermediate fill of mortar. Mortar from the northern side was C14-dated to AD, bricks from the lower course thermoluminescence dated to 1140±70 AD and animal bones from the same context C14-dated to AD. The construction could not be statigraphically related to the rampart and it is unknown if this was contemporary with the medieval fortification or a later structure. The feature was, however interpreted as part of the medieval tower Smørhætten, described in written sources in 1496 and Other post-rampart activities in the area consisted of a brick kiln built in several phases (kiln, brick floor, two brick benches, six heating channels and brickwork of so-called munkesten preserved in six courses. The sides were remarkably built with lime mortar. The heating room was 6.0 m wide and 3.2 m deep (inside dimensions). In front of 8 Archaeobotanical analysis is the study of remains of plants cultivated or used by man in ancient times, which have survived in archaeological contexts. Museum of Copenhagen 21

22 the kiln, to the south, a fire pit with charcoal was recorded. The suggestion was that the brick kiln had been used for brick production connected to the construction of the city wall and covered by a later phase of the rampart. TL-dates of bricks in the kiln were dated to 1210±60 AD and 1310±60 AD. Charcoal from the fire pit was C14-dated to AD. The rampart street covered the foot of interpreted rampart phase and consisted of a regular and compact layer of pebbles with inclusions of animal bones. Several wheel ruts were recorded. The street had been covered by a later rampart sequence of unknown date. Animal bones collected were C14-dated to AD. Two structures consisted of a north-south orientated flat based wood reinforced ditch cut into a rampart layer and under another rampart layer. The ditch had a length of 10.5 m, but could also have been as long as 60.0 m. The total width was 5.8 m. The fill consisted of kitchen and stable waste: bones of cattle, domestic pig, sheep/goat and fish together with plant remains such as bog myrtle, hops and figs. Dendrochronological analysis 9 dated the wood to winter 1298/1299 AD and a repair to c AD. Post medieval fortification AD Under the rule of Frederik the 2 nd, the fortification underwent a major modernization along its length headed by Christoffer Valkendorf in The rebuilding of the fortification could be seen as a substantial stone structure consisting of two to three courses immediately east of the foundation of the city wall, although the relationship between this development phase and older features was unclear at several places within the excavation area (Fig. 9). Fig. 9. Interpreted addition to the fortification in the 1580s conducted under Valkendorf s management consisting of a large foundation of boulders, facing north. Photo: Museum of Copenhagen. 9 Dendrochronology is the scientific method of dating tree rings to the exact year they were formed in order to analyze atmospheric conditions during different periods in history. Museum of Copenhagen 22

23 The suggestion of a new city wall was based on the orientation and traces of mortar fringe after a masonry with 0.42 m thickness on the foundation stones, demolition material and looting ditches, but it is worth mentioning that this interpretation is rather uncertain. The oldest sequence of the new city wall was recorded as courses of foundation stones together with demolition material from the original brick wall. Since there were no traces of the brick wall itself, this was explained by the fact that this could have been standing on higher placed sill courses. Suggestions that the city wall was restored at some time could also be explained by functional differences within the new structure. The medieval city wall had probably been torn down, whereupon the demolition material of limestone ashlars, rubble (both medieval bricks and smaller 16 th century bricks), larger pieces of the city wall combined with other materials had been re-used in a new foundation front. The rampart had partially been made of peat and made taller and wider, covering at least 1.4 m of the wall base. Over the rampart a straw layer was documented, interpreted as remains of a growth horizon on the rampart s surface. In the new phase the embankment was pushed approximate 4.0 m to 5.5 m beyond the medieval moat. The post medieval moat could not be separated from the earlier moat, but had been approximately 10.0 m wide, where the bottom kote lay between -1,0 to -0,5 m. The depth had not been more than 1.0 m. The sedimentation in the moat consisted of natural gyttje deposited in freshwater and swamp peat with vegetation that thrived in low water, the latter at kote +0,4 to +0,6, which proves that the connection to the sea at this moment was interrupted. Some of the layers recorded were interpreted as cleanup layers from the moat when still in use. Ditches connected to the foundations suggested a looting of the wall in the early 17 th century, but the traces were not clear. Christian the 4 th 's fortification c AD The excavation also came to affect the city fortification from the years AD. A new fortification was constructed around the city, consisting of small bastions and the long intervening stretches of ramparts (curtains). The excavation area included the rampart between the new Østerport and its bastion and the so-called Bastionen ved Iskulen. The medieval moat was filled up, while the rest of the medieval fortifications were incorporated in the new defences. All of the work on this part of Copenhagen's fortifications was completed around 1617 AD. Only a few layers could be interpreted as remains of the curtain. A row of stone imprints were suggested as the western part of the curtain foundation. At the excavation a 40 metre long bulwark of boulders of at least three courses, orientated in a north-south direction was recorded, possibly a western reinforcement or bulwark of the new harbor. Through the curtain there was a gate, the so-called Vandporten, which served as a transportation route into the city for the goods that were unloaded from the ships in Holmens Kanal (Fig. 10). Museum of Copenhagen 23

24 Fig. 10. Vandporten and later sewer pipe, facing east. Photo: Museum of Copenhagen. The gate had been 11.6 m long with a west facade 0.8 m thick. The gate room was 5.8 m wide, the sidewalls 1.9 m wide with the foundations slightly wider. The preserved height varied from 1.78 m to 2.75 m of an estimated total height of 6.0 m. The foundation of the facade was 13.0 m long and stretched respectively 1.8 m and 1.5 m outside the rest of the structure s dimensions. In relation to the gate room s width the facade was twice as wide based on the total length of the structure. The foundation was partially made of reused medieval bricks and this had survived up to street level. TL-dating of reused bricks from the gate s facade gave the dates 1200±60 AD, 1170±60 AD and 1270±60 AD. Placed between the sidewalls there was a 14.8 m long brick sewer with walled barrel vault built with small yellow bricks and an associated wooden water pipe. This sewer was somewhat misleading, TL-dated to 1400±60 AD. In the gate room s north-eastern part traces of the paving was recorded placed c m under the current street level. Other activities and buildings behind the fortification line Building remains that may be identified in the written records mentioning buildings behind Christian the 4 th 's rampart were also identified. One feature consisted of the remains of a brick chimney C14-dated to AD. Beside this, a foundation of boulders, looting ditches connected to the former city wall, ditches and pits with unknown function, rows of sill stones and floor layers and wooden water pipes were recorded. Remnants of the demolition of the fortification consisted of looting ditches, 20 wooden water pipes and one pipeline trench mainly orientated in a north-south direction and where two of the pipes were dendrochronologically dated to AD and AD. The existing moat had gradually been filled up with different material. Museum of Copenhagen 24

25 Time Phases at Kongens Nytorv The following chapters are a description of all time phases and an introduction to the finds excavated at Kongens Nytorv from (Fig. 11). The chapters will describe the most distinct archaeological features of every time phase and discuss and interpret the results in a cultural historical setting. In some of the time phases there will also be paragraphs on features historical or archaeological that characterize or describe the time period in question. A full listing of all archaeological features/groups in each time phase is presented in the excavation report and the appendices (Steineke and Jark Jensen 2017). Fig. 11. An overview of all archaeological features (marked in black) measured with total station or GPS and registrered. Phases The archaeological results are presented chronologically, but in order to provide an adaptable overview, the time phases have been sorted in areas that make sense in both time and an archaeological frame. The main excavation areas were split into subareas (Fig. 12) and these subarea namings will be used when relevant to locate the archaeological findings. The presentation of the time phases has been split up into three main areas; behind the fortification, the fortification and outside the fortification as follows: Moraine, salt marshes and other Quarternary observations Museum of Copenhagen 25

26 Prehistoric finds and features Early medieval activities AD High medieval fortification AD Late medieval city wall AD Eastern gate building AD Late medieval and Renaissance fortification AD High and Late medieval settlement and activities AD Eastern gate building and Post medieval fortification AD Settlement and activities behind the rampart AD Outside the moat. Settlement and activities AD Kongens Nytorv AD Fig. 12. Excavation at Kongens Nytorv with type 1 and type 2 areas with subareas named phases during excavation. Chronological development As an introduction to the next chapters a brief review of the site of Kongens Nytorv is presented here in chronological order. Before Kongens Nytorv was inhabited by people, the area was dominated by salt marshes with the original coastal meadows that were documented archaeologically as a dark layer on top of the natural sand and clay/natural ground. There are only a few traces of cultural influence and the area had probably been used for grazing of livestock. The prehistoric period is represented by finds of worked flint and even though most flints occurred in secondary contexts, Museum of Copenhagen 26

27 it points to the presence of a fragmented near-shore settlement at Kongens Nytorv. The dating of the collected finds points to the Late Mesolithic and Neolithic periods ( BC). The first clear human presence documented at Kongens Nytorv is from the early medieval periods before 1200 AD. Ditches parallel with the present north-south line interpreted as plot markers and pits that might have been used for storage were excavated. The most interesting features in this period are the clay lined pits and they are placed in the mid 12 th and early 13 th century based on AMS-dating 10. Their function is not clear, but similar pits documented on the coastline in Scania indicate that they have been used for storage or preparation of herrings in connection with fishing in Øresund. The first phase of the city fortification can be dated to the early 13 th century. This consisted of a rampart built using material dug up in connection with establishing the moat. Traces of a probable stockade Byens Planker were registered in one of the two earliest phases of the rampart as well as 17 meters of bulwark (dated dendrochronologically to the early 13 th century) to support the rampart on the edge of the moat. The key feature in the fortification was the eastern gate (Østerport), of which original parts date to the time of the first phase of the fortification in the early 13 th century. The building was maintained and remodeled several times until it was demolished in Through the gateway several road layers with wheel ruts were preserved. Road layers were also documented in connection with the inner gate building both in connection with the rampart street and outside the late medieval moat. Just south and in connection with the eastern gate the foundations of a customs- or guard building was excavated and the dating of this falls in the 14 th century. Following the interpreted outline of the Byens Planker a brick built city wall was erected which is difficult to date but is likely to originate around 1350 AD. The city wall could be traced for a length of 19 metres. Remnants of the wall found in other parts of the excavation area suggest that the city wall had merlons. In the early 17 th century the eastern gate and the guard building are renewed and expanded in the exact same spot as the medieval gate. The entire medieval fortification is replaced with a new and larger fortification and part of the bastion, the curtain running along the new 17 th century moat and five of the bridge pillars across the moat were excavated. This fortification does not last long, and by the middle of the 17 th century it is abandoned. Behind the medieval fortification settlement and boundary ditches can be followed continuously from the 13 th century onwards till the present day. There is evidence of permanent housing, streets, wells, pits and other traces of urban environment. Also the remains of a forge were excavated. The findings indicate that the area was relatively modestly used, when compared with other findings in urban Copenhagen. This can also be a result of modern disturbances. By the middle of the 16 th century and until the final destruction of the Post medieval fortification, a settlement existed outside of the moat. At least two buildings were excavated and also traces of possible gardens, alleys, boundary fences and roads with wheel ruts were documented. From the middle of the 17 th century and to the present day observations of predecessors to standing buildings around Kongens Nytorv were observed. Also building remains from the Lille Gjethus (the cannon foundry) from 1698 were excavated. 10 AMS (Accelerator Mass Spectrometry) radiocarbon dating is a way to obtain radiocarbon dates from samples that are far tinier than that needed for standard radiocarbon dating. Museum of Copenhagen 27

28 Finds Finds in an archaeological excavation are of great importance for dating layers and they can also tell something about the different kinds of activities in different areas or constructions. It is of vital importance to note that almost all finds from an archaeological excavation come from secondary deposited layers. This means layers that have been moved to their location from somewhere else. Only a very small percentage of the finds come from primary layers like for instance floor layers or road layers layers that are created on the spot. This is also the case with the finds from Kongens Nytorv, where a lot of layers are secondary deposits on site. These being dump layers, levelling layers, fills and garbage layers in pits or deconstructed postholes. In some cases it is more likely that the secondary layers come from the nearest surroundings. These being fill in pits, postholes and maybe also the moats. Larger levelling layers and fills in boundary ditches can be moved from places further away since this kind of work would demand a larger volume of soil or dump material. This means that using the finds for dating layers or identifying activity has to be carefully approached. The moats generated the greatest amount of finds. Due to gradually filling up, the moats have been cleaned at several times while in function, and this means that the finds to some degree have been moved around in the moats. The finds have most likely been dumped in the moats some of them one at a time for instance a broken toy or a drinking vessel. But it is also most likely that citizens of the city would dump larger amounts of garbage in the moats even though this was probably forbidden by law. In general the finds at Kongens Nytorv were in excellent condition when excavated. The moist layers in the area caused by the nearness to the former coast line and present water canals/moats made the circumstances excellent for finds to be preserved in the layers. Of course finds and primarily the organic finds were in fantastic condition from the moats. The finds were prioritized for conservation, but since organic finds of bone, wood, textile and leather in such a good state of preservation are rare and the fact that they represent unique findings (one of a kind), many of these were prioritized for conservation. Also metal finds were fairly well preserved and therefore prioritized for conservation, but many of these were nails or unidentifiable artifacts. Quite a lot of these were still unidentifiable after X-raying, but some of these have been kept for research or future identification. Type of finds The finds have been sorted into subclasses and classified by function using function types. The table below shows the different kinds of categories of finds and the percentage distribution (Tab. 1). Finds Subclass Examples of types of finds Percentage of total Medieval pottery Pottery dating from medieval periods from c AD 4% ( AD) Post medieval pottery Pottery dating from the Renaissance period ( AD) and 54% ( AD) later Post medieval period ( AD) Stove tiles Decorated and undecorated tiles from stoves 5% Leather Various types of clothes (non military), shoes, slippers 5% Textile From clothing to household furnishings, i.e. covers, curtains, etc. 1% Glass All types of glass vessel and window glass 5% Slag and other All by-products of metalworking 1% metallurgical waste Wall tiles Decorated and undecorated tiles from structure Under 1% Household equipment A selection containing wooden house furnishings to metal storage items, barrels, buckets and cooking and non ceramic storage 2% Museum of Copenhagen 28

29 objects Arms and armour Military weapons, projectiles and armour Under 1% Coins and tokens Coins and various trading tokens Under 1% Personal items A broad category representing finds linked to the individual, i.e. Under 1% jewelry, badges, brooches, religious, buttons, etc. Combs and comb Combs and comb cases from all periods from all different materials Under 1% cases Toys and games An assemblage of all types of toys, games, gaming boards and Under 1% gaming piece fragments Horse equipment All types of equipment for the horse, i.e. horseshoes, nails Under 1% Tools All various craft tools and equipment linked to industry 1% Ships and fishing Finds covering the broad spectrum of maritime archaeology and Under 1% equipment fishing equipment Security equipment Archaeological equipment comprising various types of keys, locks Under 1% and draw bars Textile production tools Crafts persons equipment linked to textile production and repair, i.e. pins, needles, thimbles, needle cases, etc. Under 1% Trading equipment and cloth seals and other seals Cloth seals, trading stamps, various trading objects outside of glass and ceramic finds objects Under 1% Writing equipment Slates, pens, stylus, stylus cases, wax slates Under 1% Knives and cutlery Domestic knives and knife and handle fragments, table knives, Under 1% spoons and forks Flint Flint tools, gun flint and percussion flint 1% Clay pipe Clay pipe fragments, pipe production equipment and figurines 5% Rope Rope fragments Under 1% Building materials Roof-, floor-, ridge tiles, bricks and other building related materials. 6% Decorated and undecorated. Statues and figurines Unidentifiable objects Various finds, either unidentifiable or corroded through soil 4% conditions and not subclasssed Other Off-cuts, nails, other organic, skeleton 2% Tab. 1. Kongens Nytorv. Finds subclasses, types of finds and approx. percentage of total. The finds By far the largest finds category is ceramics (and at most archaeological sites in general), and in order to say something general about the finds at Kongens Nytorv and socioeconomic status of the area it is an obvious first place to start. The medieval ceramics date from around 1100 AD and up until the start of the 14 th century there is an absence of imported pottery that is very common in this period in other urban societies in Denmark and southern Scandinavia. Imported ceramics is a sign of status and foreign connections like trade. The largest amount of medieval ceramic ( AD) dates to AD which might indicate a higher level of activity in the area, since ceramics dating before 1200 are very few (Fig. 13). From 1350 there seems to be an increase in imported ceramics that indicate a rise in status of the area suggested by the increasing more mercantile influence on the material. Museum of Copenhagen 29

30 Fig. 13. Baltic ware with a date frame c AD (FO203807). Photo: Museum of Copenhagen. 45% of the total amount of the ceramics found on the site dates to AD and there is evidence of local production from 1500 onwards, but also a larger amount of imports from the Netherlands, Germany, France and Italy (Fig ). The area seems at this point to be on a higher socioeconomic level with a larger diversity in the ceramic material. Even Chinese porcelain is registered in the material before the official trade with China in That said the ceramic material was still dominated by kitchen vessels and also trays and jugs. Things that are attached to everyday life in all social groups. Fig. 14. Stoneware sherd from Cologne dated to 1566 AD (FO206929). Photo: Museum of Copenhagen. Museum of Copenhagen 30

31 Fig. 15. A so-called Stjert-pot that was a very popular cooking wessel in the Late medieval and Post medieval period. It had three legs, a handle (a stjert) and glazed on the inside (FO213022). Dating frame AD. Photo: Museum of Copenhagen. Fig. 16. Dutch faience with the year 1648 written on it (FO202729). Photo: Museum of Copenhagen. The mercantile influence traced in the later medieval ceramic material can only be found in a more modest degree in other material like glass, coins and cloth seals (Fig. 17). Museum of Copenhagen 31

32 Fig th century Italian glass. This is the winged goblet or Flügelglas. A wineglass with a twisted stem on which blue wings have been applied (FO227015). Photo: Museum of Copenhagen. Trade objects like seals, scales and other related objects are all of a post medieval date (Fig. 18). This does not mean that trade was none existent in the medieval period, but it is not to be proven by findings of directly related objects. Museum of Copenhagen 32

33 Fig. 18. An Early post medieval cloth seal made of lead. The A is the initial for the German town of Augsburg. The damage reverse of the seal has probably shown a pine-cone, the heraldic badge of the town. Augsburg was well known for its cloths, which were mixed linen-warp and cotton weft fabrics used as a substitute for velvet. Photo: Museum of Copenhagen. Everyday life around the area of Kongens Nytorv is obvious in the finds material. As mentioned above pottery is the number 1 finds category and also a great representative of daily life. Also personal items as part of daily life were found; like buckles, combs, shoes, buttons, pins, bags, etc. (Fig ). Fig. 19. Late medieval spring catch purse (FO503885) from AD. The leather purse (200 x 140 mm) had traces of textile decoration and traces of wood as stabilizer for the bottom. The frame still appears to be in situ around the edges and there are metal fittings along the top edge. Photo: Museum of Copenhagen. Museum of Copenhagen 33

34 Fig. 20. Iron key dating from the Late post medieval period (FO501843). The shape and size indicates it has been used for a furniture or casket. Photo: Museum of Copenhagen. Fig. 21. X-ray photo of the same key. Some details are more obvious using this method. Photo: National Museum of Denmark. Bevaring og naturvidenskab, Brede. A collection of children s toys were also excavated from the moat, such as the little toy boat seen in figure 22 below, indicating that everyday life also played a major role around the fortification. The figure below shows a boat that would have been the obvious toy for children playing by the moat. Fig. 22. Wooden toy boat (FO200948). Photo: Museum of Copenhagen. Museum of Copenhagen 34

35 The ever-present fortification does leave some traces in the finds material (cf. Fig. 23). Traces of musket balls and iron cannonballs were found, some of them had actually been used and hit a target. The conclusion is that despite these finds, the amount of similar finds would be expected to have been larger but this might be due to cleaning of the moat in several stages. Fig. 23. Early post medieval, possibly German rapier. Quillon (FO501816). Photo: Museum of Copenhagen. Fig. 24. Early post medieval decorated bone plaque (FO207364). Probably from a weapon s handle. Photo: Museum of Copenhagen. General reflections on the finds material In general the finds from Kongens Nytorv are what to be expected from most (larger) medieval and Post medieval urban contexts. There are signs of imported goods and artifacts, but considering the fact that Copenhagen in the Post medieval/renaissance period was a fortified trade town, the amount of finds in general from a city of this status, as mentioned above, is rather few. There is also a limited amount of prestige finds which would have been owned by the more prominent citizens like the bourgeoisie and courtiers that are present in the city from the Late medieval period. Instead there are traces of craft work like the finding of a medieval forge building, finds of slags and a great amount of household equipment. This points in the direction of a more modest part of Copenhagen with workshops and general everyday life of differing social status mostly reflected through the ceramic material. This might also be confirmed by the finds of military objects like gambling pieces and plates with soldiers on (see Fig. 24 above). This could indicate an area with a more concentrated presence of soldiers, which would be obvious in connection with the fortification being in the area, but soldiers were often quite poorly paid and would be in need of a inexpensive place to live. The finds of used musket balls and cannonballs in the Post medieval moat could indicate a somewhat unsteady area to live in at times. Museum of Copenhagen 35

36 Moraine, salt marshes and other Quaternary observations Results The Ice Age deposits in the Copenhagen area consist of two layers of moraine separated by transgression layers (Bahnson 1973). The excavation area at Kongens Nytorv is placed on the transition between areas with respectively moraine and sand at the top of the sequence. Most of the excavation area consisted of subsoil of uniform transgression layers in places with a regular layering of thin layers of slightly coarser sand and in some places more compact layers of gravel and flint of different size. In the transition zone between the moraine and the transgression layers there were locally very irregular stratifications in the moraine that do not represent truncations, but may be due to processes associated with moving material or so-called cryoturbations. As shown in the figure below the transgression layers and salt marshes are mainly registered in Lille Kongensgade, in connection with the fortification, where these layers have been protected by the medieval rampart, as well as in a wider area to the east outside the 17 th century moat (Fig. 25). Fig. 25. Overview showing traces of transgression layers and salt marshes at Kongens Nytorv. The concentration of salt marshes in the western part is directly connected to the overlying and protecting rampart. The rest of the excavation area has been truncated by later activities, both historical and modern. In the watching brief trenches these natural layers were not measured due to limited depth and difficult survey conditions. A salt marsh is an area of marshy ground that is intermittently inundated with salt water or that retains pools or rivulets of salt or brackish water, together with its characteristic halophytic vegetation. The salt marshes consisted of Museum of Copenhagen 36

37 water deposited sand with very decomposed organic material with varying contents of plant parts, leaves, stems, roots and rhizomes, branches, buds, etc. where this represents a growth horizon formed in a moist, brackish or fresh environment occasionally flooded by the sea (Steen Henriksen 1998:7-8). Limited cultural influences can be seen within these layers and no indications of human activity could be seen among the archaeobotanical material but this is connected with the sampling and lack of further analysis. The original coastal meadows were documented as a 0.2 m thick, dark layer on top of the natural sand and clay. According to estimations made by geologists and recent archaeological investigations the shoreline around 1000 AD is believed to have been going straight beside the contemporary Royal Theatre, northeast of Charlottenborg and then through the later Amalienborg area. It could not be excluded that the shoreline may have stretched to the north of Vingårdsstræde near Lille Kongensgade, but nothing in the main excavation area or nearby trenches supported such a suggestion. Test drilling in Østergade has also determined that the area was within the medieval coastline, but influenced by the previous Littorina Sea Transgression (Sørensen 2002). It must be pointed out that the marshy area in the low-lying landscape of Copenhagen would have been frequently flooded at high tide and during storm situations, which must have resulted in a varying shoreline. Analyses of pollen and macrofossil samples suggest that the Copenhagen area during the 11 th century was dominated by meadows with reed swamps. The coast, where Copenhagen was founded, was characterized by wet meadows. This environment in the upper coastal intertidal zone between land and salt or brackish water produced peat layers of varying thickness, depending on local depressions, stream courses, etc. At the Metro investigations during there were traces of this meadow vegetation in the southern part of the excavation area, which indicates that the area during part of the 13 th century had been washed over by the sea. The boundary ditches and a wicker fence had been abandoned in the last decades of the 13 th century, and then covered by the salt marshes. Pollen and macrofossil analysis has shown that this natural, dense reed swamp was with time changed to pasture land (Kristiansen 1998). No further pollen analysis was conducted on the salt marsh layers, but the area has probably been used for grazing, where the most extensive settlements from the Iron Age, with some exceptions, mainly existed in the interior. This type of settlement localisation is also familiar from the other side of Øresund (Björhem & Magnusson Staaf 2006:195 et seq.). However, C14-dates from the Viking Age could represent fishing activities on a more seasonal basis. Museum of Copenhagen 37

38 Prehistoric finds and features Results Despite the occurrence of most flints in secondary contexts, the evidence points to the presence at Kongens Nytorv of a fragmented, near-shore settlement or activity spot dating to the Late Mesolithic and to the Neolithic ( BC). Most of the material consists of non-diagnostic debitage. Notwithstanding these limitations, the assemblage affords important evidence of prehistoric settlement activity within the limits of Copenhagen. Also, patterns can be teased out of this material, which have chronological import and consequently corroborates inferences drawn on the basis of the formal tool types. The assemblage suggests that the majority of the finds derive from the later part of the Mesolithic (Ertebølle period), but Middle and Late Neolithic material is also present. The assemblage consists of 113 flaked lithics from a variety of stratigraphic contexts and areas across the excavation. The principal findings of this study on the lithic material can be summarised in the following statements: A small amount of primary debitage. Few formal tool types. The presence of indirect percussion technique for blade production. A distinct lack of diagnostic Early Mesolithic components. It is possible to identify some of the flints as likely deriving from the coast by the presence of a marine affected cortex (white patinated) and indications of rolling. There are 48 blades in the assemblage, of which 31 can be categorized as irregular blades. The remaining 17 regular blades are mostly plein débitage (i.e. prismatic blades with no cortex and transversal scars) struck with indirect percussion. They generally exhibit evidence of overhang removal and have distinct bulbs or bulbs and lips. Examination of butts, impact rings, eraillure scars, bulbs and striking lips shows that almost all regular blades were struck off the core by indirect soft-hammer percussion, while most irregular blades and flakes were struck by direct percussion, most likely with a hammerstone. Hard-hammer direct percussion, used to detach blades from a core, results in generally thicker blades and is strongly associated with the Late Mesolithic Ertebølle period, although some Ertebølle sites have a higher use of indirect soft-hammer technology. Platform butts on both blades and flakes are with a few exceptions flat, but in 34 cases (of which 22 are blades) there is evidence of preparation of the front of the core, while such preparation was absent in 37 cases (seven were blades). Edge preparation work is done between striking off flakes or blades by gently tapping or rubbing the sharp, curving lip of the core platform with the hammer stone to trim it back. This technique was used to assure an exact blow was delivered to the striking area. Four flakes and one coarse blade (4.4%) have more than 50% cortex 11 present, and can therefore be described as primary flakes, the first ones removed from corticated flint nodules. Thirty pieces had smaller amounts of cortex present (1-50%) The cortex was described as either fresh (n=12) or worn (n=21) and was in all cases thin (c. 1 mm). Fifteen pieces displayed evidence of retouch or shaping, and included two flake scrapers, a Neolithic sickle blade and a polished Neolithic axe. In addition to these specific tool types there were also three truncated blades, one laterally 11 The outer "skin" of a flint nodule or spall. Museum of Copenhagen 38

39 retouched blade and five miscellaneous retouch flakes (MRF), these are pieces that have retouch, but do not resemble any specific tool form. In total, 13.3% of the flint assemblage was attributed to a tool form or identified as an MRF. A strike-a-light is made on a very thick blade of dark, grey Danien-type flint with flakes struck off from the distal end. Two or three broad, shallow flakes have been struck off (coincidentally?) from the blade s steep sides. Six unretouched blades exhibit usewear and two of these have heavy usewear. The most notable piece in the assemblage is a complete, Middle Neolithic thick-butted axe of the Lindø-type, dated BC (Fig. 26). The sides of the axe are partially polished, except the third towards the neck. Fig. 26. Middle Neolithic thick-butted axe of Lindø-type. Photo: Museum of Copenhagen. The cutting edge is chipped and also appears rather dull in its present state. The axe is white-patinated but there are also areas where the patination is light grey. A few rust-spots can also be observed. It was found embedded in the natural sand subsoil and could have been lost on the former beach in the Neolithic. It is a quite common find in settlements of the period. A circular flake scraper is made with inverse, short retouch on what appears to be a blade core platform rejuvenation flake. The scars from detached blades and evidence of overhang trimming are clearly visible on the platform butt of the scraper. It is probably of Neolithic date. Another scraper is a flake end scraper with linear retouch on the distal end. There is also a flake with distal, oblique retouch and a flake with lateral retouch. A flake has some distal, steep retouch and a possible notch on one distal lateral. A retouched flake with linear, steep retouch is a possible scraper. Finally, there is a blade-like flake with lateral retouch. Of special mention is a bifacial, crescent-shaped sickle with a high, arched back. These are typical for the Late Neolithic period ( BC), when there was a large-scale production of flint sickles of different forms. It was a tool type that remained in use until the end of the Early Bronze Age. Unpatinated flints (n=35) exhibit a limited colour range, with shades or mixtures of grey and brown being the dominant colours. The remainder are a mixture of the above colours, with faint shades of yellow, green and blue included. There are two heat-affected lithics exhibiting colour change and hairline cracks. Burning probably occurred Museum of Copenhagen 39

40 prior to deposition. A majority of the lithics (72%) have been subject to some degree of patination resulting in a distinctive, chalky white colour. While most are still sharp-edged, a few appear to have experienced rolling in a marine environment, which has smoothed and rounded their edges through erosion. Stone Age finds from central Copenhagen The Copenhagen area comprises a rich cultural landscape with traces of human activity and habitation dating back to BC. The vast majority of these places are situated beyond the old city and stem from the Palaeolithic (before 7000 BC) to the Bronze Age. Stone Age settlements, including submarine settlements from Kongemose culture are known at Frihavnen and beyond the coast of Amager. There are also Ertebølle settlements by Vedbæk, in Ordrup Mose, which was then a fjord, outside Kastrup at Amager, at Fredriksberggade and Fredriksholms quarters at Sydhavnen (Ramsing 1940, Vol. I:38 et seq.; Kjersgaard 1980:15; Christophersen 1985:12 et seq.). Some scattered archaeological evidence of prehistoric settlements is also to be found within the old city area. The occasional stray find has also been unearthed in this area. Caused by the Littorina Sea s distribution the prerequisites for Neolithic remains are extremely limited. Archaeological investigations of the oldest historical Copenhagen has for many years held a naturally dominant position among archaeologists working within the city. In contrast, the study and observations of the city's prehistoric finds have been left significantly in the background. The discovery of a flint arrowhead from Rigshospitalet (the National Hospital) testifies to the fact that people have frequented the area for more than years, but actual settlements seem to appear only from the later part of the Mesolithic period nearly years ago with the finds from Frihavnen occupying a prominent position (Rosenkjær 1893; 1896; Andersen 1985). In 1928 an article appeared with the first full description of finds from the prehistoric period in Copenhagen, Under Asfalt og Brostene, in which Kjær summarized the state of research at the time (Kjær 1928). In 1939, Troels-Smith discussed in some detail a number of Mesolithic sites on Amager and their relationship to shoreline displacement (Troels-Smith 1939). The following year, Ramsing could refer in his three-volume work Københavns Historie og Topografi to the new discovery of a worked antler from Middlegrunden, which for many years was considered the oldest find in Copenhagen (Ramsing 1940, Vol. I:38 et seq.). Ramsing s works continued to be central for an understanding of the city's prehistoric development and it is only in 1985 that Knud Andersen threw new light on the findings from Frihavnen (Andersen 1985:42 et seq.). The investigation and subsequent fieldwork in showed that there had been at least two settlements from the second half of the Mesolithic period. In 2004, trial excavations at Amager Strandpark revealed remains of a large Mesolithic settlement (Dencker 2006). Most recently Stensager has summarized the current state of knowledge regarding Mesolithic finds in Copenhagen (Stensager 2004). More than 110 Stone Age find spots are currently known in the municipality of Copenhagen of which around 30 are find spots in the central part of the city. While most are single finds, there seems to be at least one Ertebølle locality at Nytorv/Frederiksberggade, 550 m SW of Kongens Nytorv (Københavns Sogn, Sokkelund Herred, Københavns Amt, SB nr. 14). In 1907, a blade core and two blades were found here at a depth of approximately 3.7 metres. In connection with archaeological excavations of the old town hall (also Nytorv) by Chr. Axel Jensen in 1937, 71 pieces of flint were collected, mainly larger blades and flakes and an edge flake from a core axe. Unfortunately, finds information is sparse and it is likely that the flints were merely casually collected during the course of the excavation. The existence of a settlement site at Nytorv was confirmed after the construction of bunkers just opposite the court house in On this occasion, 88 blades and flakes were retrieved. Although none of these works documented the exact position and context of the finds, there is little doubt that a settlement site was located on the beach under and around the later town hall. The flints are reported as patinated. The flints would be part of a larger complex which was lost in connection with the many conversions of the square. Other assumed settlement remains from central Copenhagen (the medieval parts) involving lithics have been retrieved at Amaliegade 13, Farvergade 15, Gammeltorvs Apotek, Kattesundet and Slotsholmen (Kjær 1928: ; Stensager 2004). Outside central Copenhagen, a significant assemblage from the Ertebølle period and the Neolithic period has been found at Frederiksholms Teglværk in Museum of Copenhagen 40

41 Sydhavnen (Larsen 1947). The Ertebølle flint was white-patinated while this was not the case for the Neolithic material (Stensager 2004:4). A Late Mesolithic activity spot or camp site at Kongens Nytorv Although it is not anticipated that there are significant remains of prehistoric date at Kongens Nytorv, there are low levels of activity in the form of residual artefacts. There may also be evidence of occupation as the site lies in an area surrounded by wetlands/marshlands. Locations similar to these can be places of preferential activity during the prehistoric period as they provide ample resources for hunting, fishing and foraging. Such satellite locations are well known from Mesolithic contexts previously. The lack of a distinct Early Mesolithic component, such as microblades, microblade cores, microburins and microliths, suggests that most or all of the assemblage from Kongens Nytorv perhaps should be dated to the later part of the Mesolithic and to the Middle and Late Neolithic. Finds of flint tools and debitage have indeed shown that current central Copenhagen was visited or lived on, on several occasions during the Stone Age. The first visitors may have arrived in the area from around 6000 BC as attested, for instance, by the find of a rhomboid core axe from Fredericiagade, transverse arrowheads from Rådhuspladsen and the finds from Frihavnen. How the findings should be interpreted, we can only guess. Many recovered flints are stray finds and there is often uncertainty about the circumstances in which they have been found. This is the reason why potential settlement indicators, as fragmented as the material may be, are so significant when found within the urban environment. Based on the current evidence, it seems certain that hunters moved along the former Øresund coastline and perhaps stayed at small settlements for short periods to extract food and other resources from the coastal environment, and to knap flint. Late Mesolithic find spots show a maritime preference both in terms of elevation and location for nearshore locations. Early and Later Neolithic find spots are located in the same areas, but generally these sites are found at higher elevations and more inland in relation to the coastline. The displacement of the coastline became an important aspect of life during the Late Mesolithic and would also have affected the locality at Kongens Nytorv. It is unclear whether the flint was deposited in a settlement area with a subsequent light rearrangement during a sea level rise or whether the flint was transported within sediment transfers to Kongens Nytorv from a different location. A few lithics from the excavation have been found in geological layers, with one identified as a transgression layer. Lithics from these contexts are clearly patinated, which lends support to the suggestion that at least some, if not most, lithics were originally deposited around Kongens Nytorv, despite the fact that many of the lithics later were redeposited into secondary contexts. There are no discernible differences in form and technology between the patinated and unpatinated flints. While no evidence of actual habitation sites belonging to the Mesolithic have been found in the area, the flints marine-affected patina points to an origin within the zone affected by the sea level rise. The current evidence therefore seems to support the hypothesis that a Late Mesolithic activity spot or camp site proper at Kongens Nytorv was inundated and disturbed on one or several occasions by marine transgression (cf. Christensen 1995:15 et seq.). The Neolithic site's location in the beach zone along with the lack of burnt flint or charcoal argues that the activities do not represent a main settlement and the vast majority of the worked flint belongs to this activity. Museum of Copenhagen 41

42 Early medieval activities AD Results The presentation of the remains of Early medieval origin will be given from two perspectives. First there will be an account of the different feature types ditches, pits, buildings, postholes, stakeholes and different surfaces (Fig. 27 and 28). After the overall description the features are placed in a structural and historical context. Fig. 27. Early medieval features at Kongens Nytorv. Museum of Copenhagen 42

43 Fig. 28. Early medieval features in the western part of Kongens Nytorv. Museum of Copenhagen 43

44 All structures were documented as truncations in the natural moraine/transgression layers under the salt marshes, mixed with this or below features belonging to the medieval rampart or gate building from the early 13 th century. Besides stratigraphical observations some of the deposits or features have been grouped in this phase based on finds and AMS-datings. Ditches The ditches were orientated as good as parallel with the current plot boundary on former restaurant A Porta s eastern facade (Fig. 29). Profiles of the sequences of ditches changed from being slightly rounded to the north, to wide to flatbottomed with gently sloping sides in the south. North of the ditches a series of elongated pits were documented without further determination. Fig. 29. Boundary ditch in the middle, facing west. Photo: Museum of Copenhagen. The usage fill in the ditches consisted of decomposed buried soil representing a growth horizon and secondary natural moraine where the sedimentation must have happened incrementally. Inclusions of CBM in the latter could perhaps be connected to the construction of the gate building and/or the city wall. The mud deposition in one of the features showed that the ditches had been used over several years and regularly cleaned. The occurrence of silt and fresh/brackish water plants also shows that the ditches had not been directly connected with the sea, but flooded at regular intervals. In relation to ditch SG-333 soil excavated from the ditch could be seen next to the ditch towards the east and may have represented some sort of boundary bank. AMS-analysis of material from the deconstruction and usage phase dates the ditches to the Early Middle Ages. The features represent drainage and boundary ditches, to separate a certain area, and at the same time used the local topography. However, the structures were too small to judge the slope, etc. Museum of Copenhagen 44

45 Clay lined pits and a so-called Grumbod A total of four clay lined pits and an interpreted Grumbod were investigated at Kongens Nytorv. SG-332 represents a rectangular NW-SE orientated clay lined pit (c. 7.0 x 1.6 x 0.4 m) with straight, steep sides and a flat base (Fig. 30). Fig. 30. Clay lined pit with modern truncation (removed concrete box) in the middle, facing NW. Photo: Museum of Copenhagen. The clay lining following the construction cut, consisted of firm and uniform mid brownish grey and yellowish green clay with occasional inclusions of pebbles. The clay lining was approximately 0.05 m thick, at some places as much as 0.15 m, but it is unclear if this really belongs to the construction phase or is part of later activities/repair work within the pit. The activity layers consisted of uniform dark greyish black clay with inclusions of fish bones, where the inclusions of stones, CBM and lime fragments in probably represent material from the deconstruction layers being pressed into the underlying layer. SG-366 was a north-south orientated and rectangular clay lined pit (c. 3.4 x 2.0 m), 0.3 m deep with straight, steep sides and a flat base (Fig. 31). The southern and western parts were not as deep as the northern and eastern parts. The western part of the cut and particularly the northern part were a bit diffuse, where the feature was truncated by the modern shoring. The cut was dug through a second clay lined pit and it cut one of the rampart phases, but this latter interpretation has later been re-evaluated based on AMS-datings and other stratigraphical observations. Museum of Copenhagen 45

46 Fig. 31. Clay lined pit (SG-366), facing north. Photo: Museum of Copenhagen. The clay lining, following the construction cut, consisted of firm and uniform mid bluish yellow and light greenish yellow clay with frequent inclusions of fish bones and occasional inclusions of sand and charcoal (Fig. 32). At certain points it looked like the clay lining had been repaired with clay, since there were big clay lumps on top of the actual lining. The clay lining was thicker to the east than to the west; from 1.5 to 10.0 cm thick. Fig. 32. Clay lining seen in section, facing south. Observe how the pit has cut through an older clay lined pit to the west. Photo: Museum of Copenhagen. Museum of Copenhagen 46

47 The finds consisted of bones (horse, mammals unspecified, herring, plaice/flounder/dab and fish unspecified). The activity layer consisted of a thin layer of uniform dark reddish brown and orange-blue organic and spongy material with frequent inclusions of fish bones and occasional inclusions of twig, covering most of the bottom of the pit. In a few places both the brown and the orange/blue layers were separated/layered by yellow clay. The finds consisted of worked wood and bones (mammals unspecified, cod, herring and fish unspecified). Almost all the identified bones could be identified to herring, as only 10 out of 687 identified fish bones belong to other species such as bream, cod and plaice/flounder/dab. No bird bones and only 55 mammal bones constitute the remaining waste material. The mammal bones belong to domestic species. The herrings recovered in a collected clay block sample seem to be represented by all the elements belonging to the herring skeleton. The herrings have thus not been processed. The herring bones in general represent all parts of the skeleton and complete individuals seem to be represented in the material. The small elements from the gills were not counted. Pelvic bones are small and might have been overlooked (Bødker Enghoff and Magnussen 2015). SG-370 was a north-south orientated rectangular clay lined pit (c. 2.7 x 2.7 m), 0.2 m deep with straight, steep sides and a flat base (see Fig. 33 below). The feature was truncated by the modern shoring to the north and by a later clay lined pit (SG-366) in the middle. The easternmost side of the cut and backfills was not as clear as the western part, without clay lining, and only documented in section. This is probably due to later activities in the area (medieval rampart), but the contexts could also represent remains of a third clay lined pit, although this can not be confirmed with certainty. The clay lining (3 cm thick) consisted of uniform mid bluish grey clay, more brownish yellow at the sides and with moderate inclusions of charcoal specks. The usage layer consisted of uniform dark brown sand, 1 cm thick. As in (SG- 366) there was a small area with a different crisp layer of other organic material underneath the brown deposit. Finds consists of bones (herring and fish unspecified). Unfortunately the retrieved material from this clay pit was too little for any elaborate discussions and conclusions. The bones seem to confirm the impression based on the much larger quantify of material found in SG-366, where herring bones were present in the activity layer, and the two bones found in the deconstruction fill seem to belong to some of the most commonly found domestic mammals (Bødker Enghoff and Magnussen 2015). Group (SG-657) represents a clay lined pit backfilled with dumped material. There seemed to be no usage deposition within the pit. Most of the north-south orientated pit had been heavily truncated by the modern piling, but the original shape was rectangular; c. 2.3 m long, 0.3 m deep with moderate, concave sides and an irregular base (Fig. 33). Museum of Copenhagen 47

48 Fig. 33. Clay lined pit (SG-657), facing south. Photo: Museum of Copenhagen. The clay lining, following the construction cut, consisted of firm and uniform mid greenish yellow clay with occasional inclusions of pebbles, stones, lime fragments and fish bones (herring), where the building material originated from the deconstruction layer. All the animal bones collected in this subgroup belong to species which are commonly eaten. Only one element, a metacarpus, was identified to goat. It is not possible to conclude whether this element represents waste from a meal or waste from a workshop. Only one herring bone and no other fish bones were found a result probably related to the fact that the fills were not wet sieved (Bødker Enghoff and Magnussen 2015). None of the clay lined pits were complete due to later truncations or limits of the excavation, though recorded length vary from m and width from m (Fig. 34). Museum of Copenhagen 48

49 Fig. 34. Clay lined pits and interpreted so-called Grumbod. Usage layers with fish remains were observed in all pits with the exception of SG-657. Such organic layers occur in the clay lined pits where preservation conditions are favourable. However, no clear traces of planks, wooden lids, stakeor postholes were observed in connection to the pits, with the exception of worked wood in pit (SG-366). Nor were there any traces of robbed clay from the pits, though traces of repair and reuse in pit (SG-332) were seen in the section (see Fig. 30 above). Museum of Copenhagen 49

50 The features were aligned fairly parallel (N-S and NW-SE) over a distance of 37 m. Based on the AMS-dates to the mid 12 th century and early 13 th century the pits represent two generations of clay lined pits in the area an argument that is also proved by the fact that SG-366 was overlapping clay lined pit G-370. These two phases also correspond well with earlier age determinations for the pits (Cardell 1995; 2005; Ersgård 2006) and two of the fishing periods mentioned in the written sources (see Chapter The Scania market below). The pits did not contain any repair work in the form of complementary clay layers or lenses of eroded material. In clay lined pit SG-366 in general all parts of the herring skeleton seem to be represented (Bødker Enghoff and Magnussen 2015) (Fig. 35). Since there are elements from the hyoid arch and pectoral girdle, it is possible to conclude that the herrings had not been gilled (cf. Bødker Enghoff 1996; 1999). However, some of the nine accumulations of fish bones studied separately include no vertebrae, and it cannot be excluded that some of the herring accumulations represent only heads of herrings. This interpretation is supported by the finds of two præoperculae showing cutmarks which might indicate cutting off the heads. Another explanation of the missing vertebrae might be that the small vertebrae lying behind the head bones were overlooked during the excavation and only parts of the accumulations were sampled. Many scales have also been found pressed firmly together in cakelike aggregations. Herring scales very easily detach, and the many scales might just indicate that herrings have been lying in the pit. Fig. 35. Preserved herrings from clay lined pit, SG-366. Photo: Museum of Copenhagen. Thin section and ICP-analysis (Inductively Coupled Plasma) from SG-366 and SG-370 show sorted coarse clays with high levels of both sand and silt, which were not fine enough to hold different types of liquids, etc. Neither was there salt in the pits and the content of Na was normal in the analysed samples. These results correspond well with other clay lined pits investigated in Malmö, Scania (Brorsson 2006; 2012). Five chemical samples from both clay lining and usage layers show no traces of lipids 12 (Glanstrup 2013) a fact that is explained by refuse degraded into water-soluble products which later have diffused into the soil (Christensen 2015a; 12 Substances of biological origin that are soluble in nonpolar solvents and comprises a group of naturally occurring molecules that include fats, waxes, sterols, etc. Museum of Copenhagen 50

51 2015b). Another theory is that most of the herring were removed quickly from the pits after sorting during which no lipids were deposited (Cardell 2015b). This argument can also be seen in connection with the fact that some of the clay lined pits investigated do not contain any fish bones (Cardell 2005:102). AMS-analysis on material from the clay lined pits dates the activities to the 12 th century and early 13 th century. The irregular shallow pit (G-656; approximately 4.0 m 2 and 0.26 m deep) was recorded below the High medieval rampart and had been truncated on three sides by the modern piling. The sides were sharp/steep with an almost 90 degree angle between side and base. The base was flat (Fig. 36). Fig. 36. Construction cut of the Grumbod, facing north. Photo: Museum of Copenhagen. A lot of fish bones were observed in the pit during excavation (though not collected) and the feature has been interpreted as part of a Grumbod a type of booth mentioned in the written sources. According to Late medieval sources it was forbidden to throw the waste into the sea and it had to be removed within two days (Eriksson 1980:37). After the herring had been gilled the fish waste was brought to specific booths in Falsterbo (Grumhöjebackarna) where fish oil was separated from the rotten fish to be used as fuel in oil lamps, etc. Since no Grumbod has ever been excavated (Ersgård 2015) this suggestion is very tentative. A small assemblage of animal bones was found in the fill of this pit. They seem to be a random selection of elements from commonly eaten species of fish and a few bones from dog, pig and cattle (Bødker Enghoff and Magnussen 2015). Museum of Copenhagen 51

52 Pits Some of the other pits were dated to the Viking Age and the 12 th century, where the oldest pit probably is Early medieval based on the surrounding activities and material analysed. The larger pits were between 2.5 m and 2.0 m in diameter and up to 1.5 m deep (see Fig. 28 above). Since the original edge of the cut rarely was preserved, it is difficult to know the true depth of the pits. The sides of the pits were typically vertical or concave and the base was flat. The basal deposits were typically quite clean, and contained little or no finds, while the secondary backfills generally were mottled and rich in finds. This suggests that the usage of the pits did not leave many material traces, while the later backfills show they were used for disposing of household and production waste. Some pits contained deposits which indicated that they could have been left open for some time before the backfilling continued (or in between usage and backfilling?). Due to most material from the pits being related to their backfilling phase, their original function(s) is not quite clear. The regular shape of most of them could suggest they have been used for storage purposes. Also the inclusions from the backfill deposits point to the area being a combined household/production environment with bones, slag, clay spots, charcoal and chalk mixed with the soil. In the Early medieval household, pits were generally used as storage for food, and they could be placed outside houses or inside, dug into the floor. They could also have been functioning as containers for something which needed a controlled atmosphere, perhaps in connection with craft or production. Among the larger pits G-451 should be mentioned. The feature consisted of a cesspit and an SW-NE running ditch (Fig. 37). The two parts (ditch and pit) were interpreted as being part of the same feature. Fig. 37. Post-excavation of pit south of High medieval annex (SG-248), facing NW. Photo: Museum of Copenhagen. The pit might originally have been dug to extract clay, as the substrate here in places was quite clayish (bluish green clay), or used for some specialized craft, etc., and secondarily used over a period of time to deposit domestic, organic waste (bones). The small, linear ditch truncated by the modern piling, seemed to have been cut to lead either waste material (liquid?) into the pit or perhaps functioned as some sort of air channel. No traces of burning were Museum of Copenhagen 52

53 documented in relation to the feature, nor any flax in the completed archeobotanical analysis. AMS-analysis of material from the deconstruction fill dates the pit to the Early Middle Ages. Pit house Group G-330 represents a square pit with a stepped cut, vertical sides and flat base together with a barrel casing in the bottom of the cut (Fig. 38). The barrel was resting on horizontal planks, which in turn were resting on the basal deposit of the well cut. Fig. 38. Pit house. Upper and base of construction cut with top of barrel, facing north. Photo: Museum of Copenhagen. The barrel consisted of 15 staves with the remains of wooden hoops. The planks were from m wide and m long. All the staves in the barrel had survived, but the bottom was missing (Fig. 39). Museum of Copenhagen 53

54 Fig. 39. Recording and removing barrel, facing west. Photo: Museum of Copenhagen. Underlying timber consisted of nine horizontal planks; six oriented NE-SW and three oriented NW-SE (Fig. 40). The latter could be part of the missing base or a lid, but were not placed directly under the barrel. The wood had probably been used as "stepping planks" to avoid sinking into the mud, alternatively the planks had been used for stabilization while the barrel was positioned. Museum of Copenhagen 54

55 Fig. 40. Interpreted pit house with barrel and a N-S running pathway. Fill in the barrel and deposit above were both rapidly deposited backfills of dumped material of light reddish grey clayish sand and mid brownish grey clayish sand with some bones. No dating or environmental material was recovered from the deposits, although the barrel can be dated to AD based on dendrochronological analysis, and underlying clay lined pit (G-332) and was probably connected to the fishing activities on site. Though no postholes were recorded in the connection with the larger cut the structure could be part of a pit house with a sump in the middle and contemporary with pit G-353 placed to the north, as they were both sealed by the same deposit. Nothing indicates that the backfills represent excrement so the barrel had not been used as a latrine, nor for storing liquid since the base was missing. The pit house is dated to c AD and should be connected to the fishing activities on site represented by the clay lined pits. Since the barrel had no proper bottom the function could be a simple sump, etc. Museum of Copenhagen 55

56 Post- and stakeholes The postholes did not form any clear structure and further interpretations are difficult, but these probably represent part of a roof bearing construction of temporary character in connection to N-S running fence line(s) SG-508 (Fig. 41). Fig. 41. Post-excavation. Concentration of stakeholes (SG-508), facing north. Photo: Museum of Copenhagen. The fence consisted of horizontal wicker, though no vertical poles were documented, so these had been drawn up when the structure went out of use. Disturbances made it impossible to reconstruct the height or width of the fence line. AMS-analysis of material from one of the backfills dates the postholes to the 12 th century. Levelling and activity layers The levelling and activity layers in this time phase probably represent some sort of water deposited sand together with very decomposed organic material of plant parts also documented at the Metro excavation in (Moltsen og Steen Henriksen 1998). Museum of Copenhagen 56

57 Only a few finds were collected, consisting of iron nails, slag, leather, flint debitage and -tools together with ceramics (Undefined Greyware; AD and Early redware; AD). The bones mainly represent typical household waste and the fish is mainly represented by herring (Bødker Enghoff and Magnussen 2015). Boundary ditches and the establishment of plots in the eastern part of the city The objectives of the Kongens Nytorv investigations were to clarify land use, at least the relationship between streets, plots and buildings. It was also important to clarify the oldest activity in the area and prove pre-urban remains. Settlement remains had until the excavation was initiated mostly been found in the western areas in conjunction with the medieval city (Fabricius 1999). Former C14-datings and findings from Kongens Nytorv in relation to the so-called Clemensstaden 13 date the activities in the area to the second half of the 11 th century. This information is based on only one C14-dating to AD from the fills in one boundary ditch and an antler handle with animal ornament from the 11 th century (Kristiansen 1998:11 et seq.; 1999a:108 et seq.). Sometime in the early 1200s parts of the beach became regulated and developed. The implication of this is that the area was divided into plots with more permanent activities. The regular and irregular ditches represent boundaries between a property and the adjacent salt marshes (Kristiansen 1998:42 et seq.; 1999a:104). One of the ditches in the former excavation was filled with blue-grey clay not naturally found in the local moraine, where the sedimentation must have happened incrementally and a clear boundary marks the subsequent intentional backfilling a construction detail which was also seen in the later investigation at Kongens Nytorv (SG ). The mud deposition shows that the ditches had been used over several years and regularly cleaned. The occurrence of silt and stagnant brackish water also proved that the ditches had not been in direct connection with the sea, but flooded at regular intervals (Moltsen 1998). The boundary ditches can be seen with connection to other Early medieval activities in the area: Investigations in 1998 identified peat layers, cultural layers and an east-west orientated and at least 7 meter wide drainage (?) ditch with High Medieval finds, C14-dated to the 13 th century, another ditch, and postholes (Poulsen 2000; Moltsen 2001). The bottom layers contained large amounts of fish bones and one of these deposits was C14- dated to the 12 th 13 th century. The High medieval cultural layers were observed directly over natural substrates. At the establishment of the tunnel between Magasin's basement and the Metro station, cultural layers, paving, foundations, ditches, a cellar and other types of pits were recorded. The finds could be dated from the 13 th 14 th century onwards and elder from one of the features was C14-dated to the 11 th 13 th century (Jark Jensen 2001; Grootes 2002). The ditches and fence line can be interpreted as allotment fences for dividing the land behind the former rampart and a spatial continuation of the Metro investigations in which revealed several north-south orientated boundary ditches in the northern and western parts of the excavation area and about 10 m east of Magasin's department store facade (Kristiansen 1998:41 et seq.; 1999a). The oldest activities were three ditches that replaced each other and after disuse the latest ditch was followed by a wicker fence and a feature interpreted as some form of boundary bank. In the middle and southern part of the excavation area only one single ditch was documented (Kristiansen 1998:41-62; 1999a: ). The investigation at Kongens Nytorv shows that at least one of the boundary ditches recorded at the Metro investigations in continues to the north. No evidence, due to modern truncations, could be put forward to determine if this consisted of several phases or was a single, broad boundary ditch (Fig. 42 and 43). 13 The supposed oldest settlement in Copenhagen. Museum of Copenhagen 57

58 Fig. 42. Boundary ditches and pits at Kongens Nytorv; Metro excavations and Metro Cityring Some of the ditches represent east-west cadastrals similar to the ones that were investigated in the basement of Magasin in , delimited by an east-west aligned ditch (Kristiansen 1998:42 et seq.; 1999a:105). The former group includes ditches whose primarily purpose was to improve the growing conditions by redistributing the water, and also ditches whose primary function was to divert unwanted water from specific land areas, such as a house plot. Museum of Copenhagen 58

59 Fig. 43. North-south running ditch investigated at the Metro excavations in , facing north. Photo: Museum of Copenhagen. A total of 6 radiocarbon dates were completed dating the ditches and pits to AD. These AMS-datings correspond well with earlier dates from the Metro investigations in (Kristiansen 1998; 1999a:113 and 114), where different types of material from fills in the ditches and pits were C14-dated to between AD (Kristiansen 1998:65; 1999a:113 et seq.), with the exception of animal bones dated to AD. The collected ceramics were dated to AD, AD, AD and as late as AD and AD. A later contamination could be the case since several of the ditch fills contained red bricks, but without further description or information. The fence line is not AMS-dated based on the lack of datable wood and macrofossils, but a branch and wood from the wicker fence from the former Metro excavations were AMS-dated to AD and AD, where the latter dating is most likely. Twigs and wood chips in a layer of seaweed were dated to AD (Kristiansen 1998:65). The boundary ditches and the wicker fence were abandoned in the last decades of the 13 th century, and then covered by salt marshes and a faeces layer from humans and animals, where wood and twigs were AMS-dated to AD, AD, AD and AD, however there is no explanation for these late dates and why they diverge so much in time. A handle made of deer antler with animal ornamentation recovered from this rubbish layer could be dated to the 11 th century. After this the area was covered by the facilities of the city's fortification (Kristiansen 1998:11, 43 et seq., 65, 86; 1999a:101 et seq.; 1999b:158 and 159; Rud og Heinemeier 1998b). There is also an alternative interpretation for the boundary ditches and their function in this case should be seen in relation to the fishing activities in the area. The ditches could be compared with the ditches and clay lined pits at Kv. Liljan in Malmö (Larsson & Balic 2006: ), which like Kongens Nytorv had a clear regulation of the area in terms of several ditches located together, perpendicular to the beach and parallel with the pits lengthwise. All the pits at Museum of Copenhagen 59

60 Kongens Nytorv are located west of these boundary markings. Pits east of this marker can unfortunately not be determined with certainty, since the natural subsoil here largely was destroyed by the Late medieval moat and the later fortification from the 1600s. No clear traces of Early Middle Age settlement in the area west of the boundary ditches were recorded as probably had been seen at the former Metro excavations. Here some layers and observations in the NW part of the excavation could be interpreted as remains of buildings consisting of floor layers and imprints from a north-south orientated foundation belonging to a building dated to s, a north-south orientated base plate, a fireplace, pits and trenches. Three of these pits were interpreted either as postholes, board- or erosion holes (?) (Kristiansen 1998:48 and 55; 1999a:109 and 110). Activity layers recorded against Magasin s facade in the same way as the levelling and rubbish layers over the boundary ditches, indicated different types of activities from the 11 th century, perhaps as early as the first half of the century (Kristiansen 1998:11). In Early medieval Lund in Scania boundaries consisting of wickerwork, poles and planks have been investigated (Andrén 1976:21 et seq.). Similar facilities have most likely existed in other Danish cities as provided for in the Danish provincial laws. Among the mentioned ditches there are those which are interpreted as boundaries for cabbage farms or enclosures, limits of "activity areas" and plots within the city. The fences were of course necessary for practical reasons, for example to prevent domestic animals from entering the cabbage- and apple farms, etc. (Jönsson & Kockum 2005:9 et seq.). Palle Siemen s investigations in the Esbjerg area have shown that thwart ditches can be traced back to the 11 th century (Siemen 1994:52 et seq.). In connection with investigations in Tårnby, south of Copenhagen, regulation of plots in the form of trenches can be seen from the 1100s, although older, but indistinct boundary markers were excavated in the form of single trenches (Kristiansen 1999d:105). Østergård and Byens Tårn One of the objectives in the Method Statement 14 was to clarify whether any traces survived of an Early medieval property, entitled Østergård, reputedly located in close proximity to the excavation site at Kongens Nytorv. No traces of this farm, through high-status buildings or activities possibly affiliated to the bishops estate were recorded in the excavation and the existence of Østergård is therefore still questionable since it is not firmly located (Frederickson 1999:64). Though one archaeological observation should be mentioned: the fact that the High medieval rampart is placed on the same N-S line as the boundary ditches (cf. Fig. 42 above and Fig. 58 below), could indicate an already existing boundary and plot either specified by the king or his bailiff responsible for the fishing activities in the area. The same argument applies for the existence, location and dating of the so-called "Byens Tårn" and Kringelen. Freestanding towers are already known in the 12 th century, in Denmark, Scania and Gotland (Sprogø, Aose in Åhus, Flynderborg in Helsingør and Malmö (?) and Kruttårnet in Visby). Before Archbishop Absalon fortified Slotsholmen there is information about a tower at the harbour, built on the shore near the old ferry landing on the islet of Bremerholm. A tower near the shoreline is also mentioned in Johannes Krags Stadsret from 1294 and from two later sources in 1343 as "turrim ville" and in 1380 (KD I:84; Christophersen 1985:70). The tower should have been on Østergård's plot (see Østergård above), reputedly located on the corner of Vingårdsstræde and Holmens Kanal. In 1496 a tower called Kringelen is mentioned lying south of Østerport and east of the big ground where department store Magasin is today or "paa det Hjørne sønden for Østerport og østen for Her Benediktes Gaard" (on the corner south of the eastern gate and east of Mr. Benedict's farm) (Vingården) (KD I:234 and 334). 14 The work method statement for the archaeological excavation at Kongens Nytorv. Museum of Copenhagen 60

61 Ramsing suggest that "turrim ville", the tower, which lay on the beach, and Kringelen are one and the same tower (Ramsing 1940, Vol. III:19). Fabricius argues that the Late medieval tower Kringelen may have been placed south of Vingårdsstræde (Fabricius 1999:126). Both investigated a feature which he interpreted as a tower found under the current Magasin's south-eastern corner in The foundations were 120 degrees of a circle lying on the moraine and had an estimated diameter of 32.5 m. The wall thickness was of m (Both 1873:4 et seq. and Fig. 44). Ramsing writes that on the corner of Laxegade/Vingårdsstræde he observed the outer part of a circular grave outside this so-called tower, which continued through Vingårdsstræde (Ramsing 1940, Vol. III:12). Hans Stiesdal, museum inspector at the National Museum of Denmark, has since rejected the structure as a tower, and suggests that the masonry stems from a mantle wall around a core tower which theoretically may be older (Stiesdal 1975:2). Fig. 44. Both s interpreted tower (M) found under the current Magasin's south-eastern corner. From Both The circular construction encountered by archaeological investigations under Magasin could not verify Both's observations, lacking traces of masonry, stone foundations and ditches in the sub-surface. In connection with the excavation in the same area of possible looting pits destroying this structure, boulders, peat layers and m thick rubbish and cultural layers dated to the 13 th and 14 th century were documented (Skaarup 1997). It should be noted that only one of the archeological trenches affected the curtain wall s interpreted location. Further investigations in 1998 identified a foundation of boulders at the location and a curvature interpreted as Both s curtain wall, however dated to the 17 th or 18 th century, and thus not of medieval origin. High medieval cultural layers were observed directly over the natural substrate (Poulsen 2000; Moltsen 2001). Considering the limited excavations conducted with their uncertain interpretations the question of "Byens Tårn s" existence and the significance of Østergård is still unanswered, but one should nevertheless not disregard the fact that Museum of Copenhagen 61