MEROVINGIAN QUERN STONES FROM MAYEN
Investigating the distribution of tephrite quern stones to the
Netherlands in the Merovingian period
Max Koopman
MA Thesis
Merovingian quern stones from Mayen: Investigating the distribution of tephrite quern stones to the Netherlands in the Merovingian period
Max Koopman
Student no.: 10447296 MA Thesis
Archaeology of North-Western Europe University of Amsterdam
Supervisor: Dr. A.A.A. Verhoeven Second reader: Dr. M.F.P. Dijkstra 03-04-2018 (Final version)
‘Do not take a pair of millstones—not even the upper one—
as security for a debt, because that would be taking a
person’s livelihood as security’
-
Deuteronomy 24:6
1. Introduction 6
2. Production and use of tephrite quern stones 8
2.1 Geologic context 8
2.2 Production of tephrite quern stones 10
2.3 Use of tephrite quern stones 11
3.1 Historical overview of the Mayen quern stone production center 14
3.2 A decrease of production and exportation during the Merovingian period? 17
4. Mayen as a ceramics production center 19
4.1 Historical overview of the Mayen ceramics production center 19
4.2 The distribution of Mayen ceramics 20
5.1 The political situation and the organization of trade in the Netherlands during the Merovingian period 24
5.1 Political situation 24
5.2 Trade and exchange during the Merovingian period 28
5.3 Agents within the production and distribution of Mayen querns 31
6. The distribution of tephrite quern stones to the Netherlands during the Merovingian period 33
6.1 Methods 33
6.2 Problems 36
6.3 Distribution map 38
7. Distribution to the riverine area and Rhine estuary 40
7.1 The riverine area and Rhine estuary during the Merovingian period 40
7.2 Sites 41
7.3 Possible sites 43
7.4 Conclusion 45
8. Distribution to the western part of the Netherlands 46
8.1 The western part of the Netherlands during the Merovingian period 46
8.2 Sites 47
8.3 Possible sites 47
8.4 Conclusion 49
9. Distribution to the northern part of the Netherlands 50
9.1 The northern part of the Netherlands during the Merovingian period 50
9.2 Sites 51
9.3 Possible sites 52
9.4 Conclusion 52
10. Distribution to the eastern part of the Netherlands 54
10.1 The eastern part of the Netherlands during the Merovingian period 54
10.2 Sites 55
10.3 Possible sites 57
10.4 Conclusion 59
11. Distribution to the southern part of the Netherlands 60
11.1 The southern part of the Netherlands during the Merovingian period 60
11.2 Sites 61
11.3 Possible sites 62
12. Geographical, political and economic factors in the distribution of querns 64
13. Conclusion 68
Literature 70
List of images 84
Merovingian quern stones from Mayen: Investigating the distribution of tephrite quern stones to the Netherlands in the Merovingian period
1. Introduction
For many centuries, the area around Mayen in Germany functioned as an important center for the production of ceramics. From prehistoric times onwards, the area around Mayen also produced quern stones. These querns were made out of tephrite, also known as basaltic lava. These quern stones were often exported over large distances. The exportation of ceramic wares from Mayen to the area that is now known as the Netherlands seems to have decreased in the 6th and 7th centuries. An interesting 1
question concerns the supply of Mayen quern stones during the same period: did this also decline or did it remain constant? The distribution of these quern stones will be mapped in order to make a comparison with the presumed decline of ceramic imports from Mayen in this period. The results of this research will be placed within a historical and economic context. The outcome of this thesis will therefore not only contribute to the knowledge about the supply of these specific quern stones but also to the knowledge about trade and exchange in the Dutch areas during this period.
First, an overview of the production and use of tephrite quern stones is presented in chapter 2. This part will briefly explain the physical geographic processes which formed the tephrite sources of the Eifel region and how this material was worked to create quern stones. It will also clarify how these quern stones were used in daily life.
Second, the existing literature about the Mayen area will be used in order to investigate what is already known about both the quern stone and ceramic production centers of this region. Third, archaeological publications about known Merovingian settlements within the Netherlands are inventoried and studied for presence of tephrite quern stones in order to create a map that offers an image of the distribution of these quern stones during the Merovingian period. The distribution of these quern stones will be compared to the supply of ceramics from the Mayen area to the Netherlands during this period in order to answer the question if the supply of ceramics and the supply of quern stones from the Mayen area are related or if they are independent. The research will also investigate if the distribution can be explained by the geographical location of the Merovingian settlements in which quern stones have been found. A more dense distribution in settlements close to the Rhine would seem likely.
Verhoeven 2014, 13 1
Besides the evaluation of the distribution and use, the results will be placed within a historical and economic context. A brief overview of the current knowledge about the historical background of this area during the Merovingian period will be presented. The political situation in this area during the Merovingian period will be outlined and the way in which political factors might have been of influence on the distribution of tephrite quern stones will be investigated. It will also outline how trade and exchange was organized during this period.
2. Production and use of tephrite quern stones
The quern stones from Mayen were made out of tephrite. Different names are used for this material within the existing literature, such as basaltic lava, Mayener lava, Niedermieniger lava or Aldernacher lava. This material was favored for the production of quern stones 2
because of the fact that its composition makes it very suitable for grinding. When the 3
stone is used it will not become smooth very fast and it will keep its coarse surface, which is needed for milling grain. This part will briefly explain the physical geographic processes which formed the tephrite sources of the Eifel region and how this tephrite was used in order to make quern stones. It will also clarify how the quern stones were used in daily life.
2.1 Geologic context
The Eifel area contains large amounts of tephrite sources and it is now generally accepted that almost all of the tephrite quern stones found in North-Western Europe originate from the Eifel region. The name tephrite is derived from the Greek word tephra, which means 4
ashes. It is a rather porous stone with many cavities of which the edges are very sharp. It is mainly composed of clinopyroxene, calcic plagioclase, analcime, leucite and nepheline. 5
The tephrite sources are a result of former volcanism in the area. The oldest volcanic activity in the Eifel dates back to the Tertiary Period and the youngest volcanic activity took place approximately 10.900 years ago. The Eifel contains two volcanic fields; the West- und Hocheifeler Vulkanfeld and the Osteifeler Vulkanfeld, the latter being located closer to the Rhine. These two areas together contain 117 known sites where tephrite was quarried for the production of quern stones, dating from the Neolithic period to the modern era. Prehistoric quern stone production has been proven for 22 of these sites. The Eastern 6
and Western Eifel volcanic fields are two of the very few Quaternary volcanic fields in Central Europe. The Western Eifel volcanic field is the largest with a surface area of 600 square kilometers and contains around 240 extinct volcanoes, while the eastern volcanic field has a surface area of 400 square kilometers containing around 100 extinct
Parkhouse 1997, 99 2 Pohl 2010, 148-149 3 Parkhouse 1997, 3 4 Bowes 1989, 557 5 Mangartz 2006, 25 6
volcanoes. Because it is generally accepted within the existing literature that most of the 7
tephrite quern stones that are found in North-Western Europe must have their origin in the Mayen area, this chapter will only focus on the geophysical properties of the volcanic field in the eastern Eifel. The center of this part of the Eifel is the Laacher See area, which lies in between the Rhine, the Nette and the Brohl rivers. All of the volcanism within this area is of Pleistocene age. Here, the volcanoes have produced a large number of lava streams with a basaltic to highly unsaturated, nepheline containing composition. Due to high
proportions of built up gasses, volcanic eruptions in this area were so heavy that liquid and solid volcanic material were blown into the air, together with large amounts of other types of rocks. As a result, large parts of the Laacher See area are covered with thick deposits of volcanic tuffs. During the eruptions of the magma, which was rich in gas, streams of liquid 8
lava broke out and eventually cooled down, creating solid layers of volcanic material. A good example of this phenomenon is the Bellerberg volcano, which is located near Mayen. Eruptions produced valuable deposits of volcanic materials, such as basaltic lava (tephrite) which was used for quern stones, and tuff stone which was extensively used as building material. The lava streams have produced huge formations of basalt pillars. When lava 9
flows cool down relatively fast, forces of contraction will be build up. When the lava flows shrink in a horizontal direction, it can result in an enormous network of fractures that will form these basalt columns.10
According to Kars (1983) the choice for the Eifel region for the intense exploitation of tephrite can be explained by different geologic and geographic aspects of this area. The first is the above mentioned composition of tephrite. The surface of other types of stones of which quern stones were made in the past, such as granite and sandstone, can become smooth during the milling process. The surface of a tephrite quern stone makes it more suitable for grinding. The stone has many cavities of which the edges are very sharp. The stone will not become smooth after use: instead, new cavities will be opened, which will preserve the rough surface of a tephrite quern stone. If the surface of the quern stone eventually became too smooth after long-term use, it was quite easy to make it rough again, for instance by making grooves or pits in the surface of the stone. At the same time,
Gluhak 2010, 39-41 7
Kars 1983, 113 8
Hunold & Schaaff 2010,177-180 9
Smalley 1966, 110-113 10
tephrite is a quite fine crystalline and this makes it a viscous and hard wearing material. It is therefore unlikely that a fragment of the quern stone will end up in the flower.
Another explanation for the popularity of the Mayen area was the fact that the tephrite sources are relatively easy to access in this area. The tephrite sources are located close to the surface, which makes it easy to reach them. Also, the above mentioned column
formation, which is characteristic for many volcanic rocks, makes it easier to form the tephrite into round quern stones. Lastly, its location close to the Rhine is an important 11
reason for the exploitation of tephrite within this area, because the river allows the easy exportation of quern stones to more remote areas. The relatively low density of tephrite makes the querns not really heavy so that they are easy to transport and handle.
Fig 1. Volcanic fields in the Eifel; the Mayen tephrite sources are marked with a square (Kars 1983, 111).
2.2 Production of tephrite quern stones
The basaltic lava pillars that form the sources for the quern stone production are one of the main reasons why the production took place in the Mayen area. Although production
techniques varied over time, as will be outlined in the next chapter, the main principle of extracting the raw material and forming them into the desired shape stayed quite the same
Kars 1983, 117-118 11
over time. These pillars will form a good start for the production process of a quern stone because the producer will only have to take of a ‘slice’ of the basalt pillar and form it into the desired shape. At the same time, the column formations can be seen as a restricting factor for the production of quern stones: the diameter of a basalt pillar will form the maximum diameter of the quern stone. The composition of the stone makes it easy to 12
work. During almost the entire history of the production center the production within the Mayen area seems to have been carried out by groups of people using tools and pack animals. A ‘slice’ of tephrite was first taken of a basalt pillar with the aid of a wedge and a hammer. Then, the piece of tephrite was formed into the desired shape by refining it with a hammer and chisel. The querns were not always finished within the Mayen area. During 13
the Roman period it seems that several workshops existed around Mayen in which the querns were formed into the desired shape. During the Early Medieval period, however, evidence from trading settlements such as Dorestad and Haithabu suggests that (a part of) of the querns left the Mayen area in a semifinished form. They were formed into the
desired shape within the trading settlements.14
2.3 Use of tephrite quern stones
Quern stones are tools that are used for grinding different kinds of material, mostly cereals. The first known quern stones from the Netherlands can be found in the southern part of Limburg. These were probably introduced by agrarian colonizers, which belonged to the so-called Linear Pottery culture, during the fifth millennium BC. These quern stones do not show the typical round form of the Roman and medieval quern stones that are made out of tephrite. These Neolithic flat quern stones often have a rounded rectangular form and are often made out of a coarse-grained stone, such as arkose. The first quern 15
stones from Mayen seem to have appeared in the Netherlands during the Early Iron Age. Kars (1983) mentions that there are indications that tephrite quern stones already have been used in the western part of the Netherlands during the Late Bronze Age.16
Kars 1983, 118 12 Mangartz 2006, 26-31 13 Parkhouse 1997, 7-9 14 Harsema 1979, 7 15 Kars 1983, 114 16
The Iron Age querns often have the so-called Napoleons hat shape. The bedstones of these querns are not round and flat. Instead they are pointy at the bottom. This point was most likely placed in the ground to make the quern more stable during the milling process. During the Late Iron Age there seems to be a sudden switch to another kind of quern: the rotary hand quern, which consists of two flat round stones. First, there is the lower stone on which the material that has to be ground is placed, called the ‘bedstone’. The bedstone will remain in a fixed position during the milling process. Second, there is the upper stone which is called the ‘runner’. The upper stone will be brought in motion in order to grind 17
the material that is placed in between the two stones. A hole was drilled in the middle of the two stones. A wooden shaft was placed in the bedstone and the runner could turn around this shaft freely. At the same time, there was enough space to pour in grain from the top to the grinding level. Another hole was drilled at the edge of the runner, which was most likely intended for a rope. This rope attached the runner to a stick which was used in order to rotate the quern by hand. The same principle can still be seen in use by
contemporary societies in North Africa. 18
Fig 2: Rotary hand quern stones used in contemporary Tunisia. (Harsema 1979, 21)
Some scholars have argued in the past that hand querns from Mayen were not used in a rotary motion: instead, the runner would continuously be turned 180 degrees clockwise and counterclockwise. This idea would support the symmetrical polishing of the runner
Feveile 2010, 133-134 17
Harsema 1979, 20-21 18
stones. This hypothesis resulted in the German name Pendelmühle for this type of quern stone. This would mean, however, that the runner has to be stopped continuously in order to change the movement of the stone. According to Harsema this is an unproductive and energy wasting technique. Contemporary societies in North Africa also use the rotating technique.19
Besides the production of these hand querns, the Mayen area also produced querns for use within larger power mills. The stones that were used within Roman watermills are well known examples. These stones are usually much thicker and have a larger diameter.20
Although the composition of tephrite quern stones does not allow the grinding surface to become smooth very fast, it is sometimes necessary to make it more rough. This can be done by means of ‘dressing’. One technique for doing this is called ‘grooving’ whereby many short parallel lines are carved in the surface of the quern stone with the aid of a hammer and a chisel. Another technique is called ‘pitting’ where the same is done by means of covering the surface with a very dense pattern of pits.21
Harsema 1979, 23-25 19 Hörter 1994, 33-35 20 Feveile 2003, 135 21
3. Mayen as a quern stone production center
The Mayen area functioned as a quern stone production center since the Neolithic period. This chapter will first give a brief overview of the history of this area in relation to quern stone production from prehistoric times to the modern era. After that, it will investigate what information the existing literature has to offer about the production and exportation of tephrite querns from the Mayen area during the Merovingian period.
3.1 Historical overview of the Mayen quern stone production center
The earliest production of querns is thought to have taken place around 5000 BC. This production took place at the edges of the Eifel tephrite sources, where the material is relatively easy to extract. The quern stones from this period have a bread-like form and are usually 40 to 50 centimeters in length. Because these querns have the same form as the 22
natural stones that occur in the Eifel area, semifinished products from this period are often hard to recognize. Stone tools from this period have been found near the tephrite sources. These tools are made of hard basalt. There are some indications that the quern stones were extracted with the aid of fire. By heating the stone, existing cracks will widen and pieces of tephrite become separated from the pillar. The exportation of querns in this period was limited to areas within a 20 kilometer radius of the production center.
During the Late Bronze Age (1200-750 BC), the querns still seem to have had the bread-like shape and the tools and techniques remained the same. In this period, a total of 12.500 cubic meters of stone was extracted, which was enough for creating 62.500 querns. The first exportation of querns to the Netherlands seems to have taken place 23
during this period, as quern stones that are found within settlements in the western part of the country might date from this period.24
During the Early Iron Age the production was intensified. The quern stones were larger and took the shape of a boat. Hard basalt hammers were now used which made it possible to make the first stone quarries which were usually several meters deep. During the period from 750 to 450 BC a total of 150.000 cubic meters was quarried which was enough for the production of 750.000 quern stones. Examples of quern stones from this period have
Hörter 1994, 17-19 22 Mangartz 2006, 26 23 Van Heeringen 1985, 378 24
been found within settlements along the Dutch North-Sea coast. In the later part of the 25
Iron Age the production was again intensified and the stone quarries became larger. The so-called Napoleons hat quern stones, which are characteristic for first part of this period, are larger and are pointy at the bottom. Iron tools were now used for the production of 26
querns. During this period the first hand rotary querns started to appear. Because these querns have a hole in the middle, they could only be produced with iron tools. Both the Napoleons hat and the hand rotary querns were finished within the production center. During the period from 450BC to 0 AD, 480.000 cubic meters of stone were quarried which was enough for the production of around 2,4 million hand rotary querns. Examples of these querns are known from settlements in the northern part of the Netherlands. These were most likely transported over sea after they had reached the Rhine estuary. 27
Fig 3: Neolithic/Bronze Age bread-shaped quern stones (left) and Iron Age ‘Napoleons hat’ quern stones (right). (Hörter 1994, 15)
The Mayen area continued to produce quern stones during the transition from the Late Iron Age to the Roman period. A wide variety of tools was used for the production process; stone tools were no longer used. Iron wedges were used for extracting the stones and zoological remains indicate the use of pack animals for the transportation of the finished querns to the Rhine. There are also indications that the miners used marking and
measuring methods for a systematical exploitation of the tephrite sources. Besides hand rotary querns, there were also larger quern stones produced during the Roman period which were intended for use within larger constructions such as the famous Roman water
Mangartz 2006, 27 25 Harsema 1979, 17 26 Mangartz 2006, 28 27
mills. During the Roman period (0-450) an enormous amount of 3.412.500 cubic meters 28
of stone was quarried, which was enough to create 17.062.500 querns. There are
indications that the port of Andernach was enlarged in order to cope with this increase of export of quern stones. Several workshops existed around Mayen in the Roman period, in which the semifinished querns were made into functional querns. The querns were
exported to many areas of the western part of the Roman Empire, including the British Isles. Hand rotary querns from Mayen have also been found in the unconquered parts of Germania. No Mayen querns seem to have reached Denmark during this period.29 30
Several scholars write about a sudden drop in or even a stop of export of querns during the Merovingian period, especially during the sixth and seventh centuries (this will be outlined in the next part of this chapter). Semifinished querns are found near the quarries that are thought to have been in use during the transition from the Roman period to the Merovingian period, but a lack of other types of finds makes it hard to date them properly. Many of these stones are simply dated as ‘early medieval’. The production and labor 31
efficiency seem to have declined to pre-Roman levels in the Early Medieval period. 32
During a large part of the Early Middle Ages, at least a part of the quern stones that left the Mayen area were most likely unfinished products: round tephrite stones without a hole in the middle. These semifinished querns were transported to trading posts, the so-called emporiae. They were finished in workshops within these emporiae. From there, the
finished querns were shipped to their final users. Evidence in the form of unfinished querns and remains of workshops in trading posts such as Dorestad near Wijk bij Duurstede and Haithabu support this idea. During the entire Early Middle Ages, 450.000 cubic meters of 33
stone were quarried which was enough for creating 2.250.000 querns. The Mayen area continued to produce querns until the end of the nineteenth century. From 1500 onwards, underground mines were constructed.34
Hörter 1994, 32-37 28
Mangartz 2006, 29-30 29
Hauken & Anderson 2014, 59 30 Hörter 1994, 40 31 Mangartz 2006, 31 32 Parkhouse 1997, 7-9 33 Mangartz 2006, 31 34
Fig. 4: Theoretical number of hand querns produced each year based on the amount of tephrite that was extracted in each period (After Mangartz, 2006)
3.2 A decrease of production and exportation during the Merovingian period?
What becomes clear from the above is that the production decreased during the Early Middle Ages or perhaps already during the Late Roman period. What remains unclear is the situation during the Merovingian period, especially during the sixth and seventh centuries. Also unclear is the volume of quern stone production during the Late Roman period (350-450), a period at which the manufacturing of ceramics reached its peak in Mayen. The question rises if the production made a sudden drop or even stopped for a 35
certain period of time after the collapse of the Roman Empire or if the production of Mayen quern stones continued without interruption. More important is the question whether or not the markets for these quern stones remained the same or if certain areas were not
supplied with querns after the fall of the Roman Empire. Several scholars write about a
Grunwald 2015a, 195 35 0 10000 20000 30000 40000
Late Bronze Age 750-450BC 450BC-0 AD Roman period Early Middle Ages Annual production of (hand) querns
sudden decrease of production and exportation after the Roman period. Harsema (1979) states that there was most likely a serious decrease of export, but he does not refer to any sources on which he based these hypotheses. As noted above, Hörter (1994) states that 36
a lack of other types of finds near the quarries makes it hard to draw any conclusions about the production during the Merovingian period. Mangartz (2006) states that the 37
export via the Elbe and Weser suddenly stopped in the fifth century until the seventh century, but he does not report anything about the exportation via the Rhine during this period. 38
An inventory of quern stones, which date from the Iron Age to the Late Medieval Period, found within four counties in the southeast of Britain, shows some interesting results. During the Roman period many tephrite quern stones from Mayen were used in these counties, especially within ‘Romanized’ urban and rural settlements, such as villas. Querns made out of puddingstone, which originates from Britain itself, were also used during this period. The Roman tephrite querns can be found within settlements dating up to the fifth century AD. The presence of Mayen querns in fourth and fifth century Britain is an indication of the continuity of production during the Late Roman period. No tephrite and puddingstone quern stones dating from the sixth and seventh centuries were found in this survey. This might indicate that both the local production of puddingstone querns and the importation of tephrite querns from Mayen ceased during this period. The earliest medieval tephrite querns analyzed within this survey date only from the eight century.39
The existing literature raises an image of a decrease in production and exportation of Mayen querns during the Merovingian period. At the same time, a presumed decline in population during and after the collapse of the Roman empire means that the consumer market for the Mayen quern stones was seriously downscaled. The distribution to the areas that are nowadays known as the Netherlands remains unclear.
Harsema 1979, 27 36 Hörter 1994, 40 37 Mangartz 2006, 31 38 King 1986, 95, 119 39
4. Mayen as a ceramics production center
Mayen was also an important center for the production of ceramics from the Roman period onwards. This part will first give a brief overview of the history of this area in relation to ceramics production in the Roman and medieval periods. After that, it will investigate what knowledge the existing literature has to offer about the production and exportation of ceramics from the Mayen area during the Merovingian period. It should be noted that the picture of distribution of ceramics might be incomplete due to a lack of (documented) archaeological finds.
4.1 Historical overview of the Mayen ceramics production center
Mayen was a producer of ceramics from the first to the fourteenth century. All of the ceramics that have been produced here in both the Roman and medieval period are wheel-thrown. The ceramics from this area are often described as ‘Mayener Ware’,
‘Eifelware’ or ‘Eifelkeramik’. Mark Redknap (1987, 1988, 1999) has studied the production and consumption of Mayen ceramics. It should be noted, however, that these studies are for a large part based on the archaeological evidence available in the 1980s. Redknap mainly studied the archaeological material from the pottery kilns in Mayen itself and paid less attention to the distribution of the ceramics.
The provenance of ceramic wares from Mayen can often be attested with the aid of petrology. The clay of the Mayen wares is characterized by inclusions in the form of small fragments of volcanic material, such as augite and sanidine. As they are quite distinctive 40
and different from ceramics from other production centers in the Eifel, ‘Mayener Ware’ is probably the most suitable name for this group of ceramics. The inclusions in the clay and the firing methods made the ceramic wares from Mayen to be very hard and it made them heat resistant. These characteristics of the Mayen wares, combined with the location close to the river Rhine, explain the popularity of the Mayen pottery on the market for many centuries.41
The first production of ceramics is presumed to have taken place in the middle of the first century AD. This Roman production likely continued in the second century and the
production increased significantly around 350 AD. Mayen can be seen as a vicus during 42
Redknap 1988, 4-5 40 Grunwald 2009, 150 41 Grunwald 2015a, 195 42
the Roman period, which had a favorable location next to the road that led from Trier to Aldernach, where ports for long distance exportation were located. The main products during this period were coarse wares, which were intended for daily use. There are also indications for the production of finer ceramics and the so-called terra nigra wares, which are seen as luxury goods. Several kilns dating from different periods were found during archaeological excavations in Mayen. The Late Roman kilns in Mayen were situated on the northern bank of the Nette river. The eight kilns that have been found here all have the same rectangular shape. The structures have single flues and are semi-sunken. The oven floors are raised. None of these Roman kilns were excavated with the aid of modern 43
techniques. Descriptions and photos from 1922 show that the Roman kilns consisted of stable structures made out of basalt and lava stone.44
During the Early Medieval period, the production center shifted to the other side of the river Nette. Near the Genovevaburg, several kilns have been found, along with production waste material. Some kilns show the same updraught principle as the Roman kilns but they show a greater variety of plans. There are also some crossdraught examples of medieval kilns found in Mayen. Although the production and settlement shifted to the 45
south, the ceramics (Mayen ware MD) were still for a large part being produced within a specific Roman tradition. Both the ceramic forms and production techniques were still the same during the early sixth century. This would suggest an undisrupted continuity of 46
production of ceramics after the end of the Roman empire. Most of the early medieval wares from Mayen show similarities with their Roman predecessors until the eight century. A change occurred in the tempering material in the Carolingian period. While the Roman and Merovingian wares were mainly tempered with feldspar, the Carolingian wares were tempered with quartz sand. During the end of the eight century Mayen started to produce hard-fired, watertight vessels, classified by Redknap as ‘Mayen Ware ME’.
4.2 The distribution of Mayen ceramics
Late Roman coarse Mayen wares can be found in many parts of the Roman empire, both up and down the Rhine. It is found in both military and civilian sites, reaching from
Redknap 1987, 88 43 Redknap 1999, 39-40 44 Redknap 1988, 4-6 45 Grunwald 2012, 144 46
Switzerland to Britain. The proportion of Late Roman Mayen wares within downstream 47
settlements is often as high as within settlements in the vicinity of the production center. The ceramic assemblage of Maastricht during the period 350-450 AD, for instance,
consists for up to 75 percent of Mayen wares. These Late Roman wares continued to be 48
exported to areas both up- and downstream the Rhine and via the Mosel and Meuse until the late fifth century or the early sixth century.49
Although the above suggests a continuity of production from the Roman period to the later Medieval period, there are some indications that the consumer markets shrank during the Merovingian period. This idea is mainly based on the rise of local ceramic production centers during this period in certain areas which were previously supplied with Mayen wares. The question rises if these local industries must be seen as a result or a cause of 50
a decline of exportation of Mayen ceramics. In the case of the Netherlands, Merovingian kilns were discovered in Maastricht which can be seen as an example of such a local workshop. As noted above, the ceramics spectrum of Maastricht consisted for up to 75 percent of Mayen wares in the early fifth century. Excavations in 1991 in Maastricht-Wyck unearthed the remains of four pear-shaped kilns from the Merovingian period. These kilns, together with two postholes, were the only Merovingian structures that were found during this excavation, but it is clear that other kilns and workshops existed in this area. Inside the kilns large quantities of sherds were found, which must be seen as production waste. The kilns were probably filled with this production waste after they were abandoned. All the structures contained sherds from different kilns and no distinctive layers of waste could be identified. This shows that al the kilns were abandoned simultaneously and filled with waste in the same moment. Radiocarbon dating of charcoal found within the kilns provided dates between 540-660 AD. The production probably exceeded the needs of the
settlement, so (a part of) the ceramics that were produced here were intended for exportation. Unfortunately there are no Merovingian sites in the vicinity of Maastricht to corroborate this hypothesis. Apart from Maastricht, several other production centers are known from this period in the Meuse valley.51
Brulet et al. 2010, 420-422 47 Redknap 1999, 133-134 48 Grunwald 2012, 143 49 Grundwald 2015b, 455, Redknap 1999, 350 50 Van Wersch 2004, 19-31 51
Different studies can be used in order to investigate the distribution of Mayen ceramics during the Merovingian period. However, it should be noted that distribution maps of ceramics are as much an inventory of archaeological excavations as they are surveys of archaeological artifacts. In other words: these distribution maps are dependent on the available archaeological sources and they provide an incomplete picture. The studies of Redknap (1999) show that the distribution of the Mayen ceramics to upstream areas has decreased during the Merovingian period, when compared with the distribution during the Roman period. The Mayen ware MD is distributed in small amounts to Trier and to areas near the upper Rhine and the Main. The ceramics assemblages of upstream settlements near the Rhine, near the Main estuary and in Rheinhessen consist for 15 to 30 percent of Mayen coarse wares. The decrease of imports within Trier is explained by the emergence of a local industry in the period 600-670. Mayen ware MD is more frequent in downstream areas during this period, especially in the Rhine estuary. Mayen coarse wares, especially the so-called Wölbwandtöpfe, are present in multiple settlements and burial sites in the vicinity of the Rhine and Meuse estuaries. The ceramic assemblages of Merovingian 52
settlements within the Dutch province of South-Holland, where there are no indications for local production, consist for 80 to 100 percent of wheel-thrown import ceramics. A part of these ceramics are thought to have been produced in Mayen. Grunwald (2015) states 53
that there was a decline of exportation of ceramics from the first half of the seventh century until the end of the seventh century or the beginning of the eight century. According to Grunwald, the decrease in the seventh century cannot be explained by a collapse of trade relations: he again sees the rise of local production centers as the main reason for a decrease of exportation. It can be concluded that the distribution of Mayen ceramics 54
continued in the Merovingian period, most likely on a reduced scale. At the same time, wether as a cause or a result of this decrease of distribution of ceramics, local ceramic industries emerge in former outlet areas during the Merovingian period.
Later, the export of Mayen ceramics to the consuming markets of North-Western Europe increased again in the Carolingian period. This is corroborated by the research in 55
Leiderdorp. The main Carolingian products were Kugeltöpfe and Tüllenkannen, which 56
Redknap 1999, 135-137, 350 52 Dijkstra 2011, 309 53 Grunwald 2015b, 455 54 Grunwald 2015a, 204 55
Dijkstra et al. in prep., 130 56
were exported to many parts of North-Western Europe during the ninth century. In the 57
Low Countries the supply of ware displays a marked drop around 875 from which it never recovered. Mayen ceramics from the later Medieval period are only incidentally found within settlements in the Low Countries. The production continued during the period from 58
the tenth to the twelfth century. According to Grunwald (2012) Mayen was still a major center for the production of ceramics intended for exportation during that period.59
Grunwald 2015a, 197-199 57
A.A.A. Verhoeven, personal communication, March 2, 2018 58
Grunwald 2012, 152-153 59
5.1 The political situation and the organization of trade in the Netherlands during the Merovingian period
This chapter provides an overview of the political and economic scene of the Netherlands during the Merovingian period (ca. 500-750). It will outline to what extent the area that is now known as the Netherlands was incorporated in the political sphere of the Frankish Empire and how this might have influenced the trade and exchange of goods. When studying the exchange of goods, a differentiation might be made between luxury or prestige goods, such as gold, and ‘common’ or staple goods (commodities) such as ceramics. The quern stones from Mayen can be viewed as commodities, since they are often found within settlements. A central topic within this chapter will be the role of elites within the trade networks of the Merovingian period. Although the exchange of luxury goods most likely happened within elite spheres, the degree of elite control of the trade of commodities is under debate. This chapter concludes with outlining the role of Mayen quern stones within these trade networks and the way in which querns were distributed to their consumers. It will try to determine which agents were responsible for both the
production and distribution of querns. Were these exchanged under elite control or by ‘free entrepreneurs’?
5.1 Political situation
The current knowledge about the political situation during the Merovingian period is for the largest part based on written sources. It should be noted that from the Merovingian period little to no written sources about the political situation in the Netherlands are available. Archaeology is often of little use for the reconstruction of the rather complex political situation of this period. The Frankish kings belonged to the dynasty of Merovingians. The Merovingians ruled the Frankish empire, of which the ironic fate seems to have been that it alternately became united and then fell apart again during this period. The political heart of the Frankish empire was situated in North-Eastern Gaul. The influence of the Frankish 60
empire on the Low Countries varied over time. It seems that Clovis already had ambitions to expand his empire in northern direction. In 491 he conquered the area around
Tongeren. After this event, the historical sources become silent. During the sixth century 61
Noord-Brabant and the areas along the Rhine and Meuse were not yet incorporated in the
Wood 1994, 55-59 60
Bachrach 1970, 24 61
Merovingian kingdom, although there might have been some Merovingian aristocratic presence in the areas in and around Nijmegen. This does not mean that there were no contacts between the elites of these areas and the Frankish empire during the early
decades of the Merovingian period. After the death of Chlotarius I the Frankish empire was divided into three smaller kingdoms: Austrasia, Neustria and Burgundy. As can be seen on the map below, the southern part of the Netherlands was incorporated into Austrasia. The Mayen quern stone and ceramic industries were also situated in this kingdom.
Fig. 5: Map of the Frankish empire during the Merovingian period (After Shepherd, 1911)
To what extent the Frankish kingdom reached to the north is not entirely clear. There might have been some nominal control over Noord-Brabant and the Scheldt estuary, but the amount of direct political influence remains uncertain. Around 630, the Frankish king 62
Dagobert visited the Kromme Rijn area and founded a church in the former Roman castellum, which he donated to the bishop of Cologne. This means that there was, at 63
least in the middle of the seventh century, some Merovingian political interest in these areas. Influence of the Franks did only took place incidentally on a royal level. Because of
Van Es 1994, 83-85 62
Van Es 1994, 90-91 63
the power vacuum in the northern part of Austrasia, representatives of new Frankish elites could settle in these areas. These elites formed social networks in which the power was exercised. 64
Next to the Franks, two other groups of peoples were present in the Netherlands during this period: the Frisians in the northern and western coastal areas and the Saxons in the eastern part of the Netherlands. For the Merovingian period, no written sources exist from Frisia. The result is that there are many different theories about the way in which this area was politically organized. These theories are often based on a combination of
archaeological evidence and historical evidence dating from the Carolingian period or later, such as the Lex Frisionum. The main topic of debate seems to be if Frisia must be seen 65
as one unified kingdom or if it was compromised of multiple smaller kingdoms during the Early Medieval period. Halbertsma (2000), for instance, advocated the idea that the coastal areas reaching from the Scheldt estuary to the Weser, sometimes combined with the areas around the Old Rhine were one single political unity (‘Frisia magna’) ruled by a single king. The power basis of this kingdom was situated in the modern-day province of 66
Friesland. In recent years this image seems to have been largely abandoned, although Nicolay (2014) has incorporated many aspects of this theory in his recent publication. 67
The political scene of sixth century Frisia is now often viewed as one which is
characterized by warlords, instead of kings, which based their power on the loyalty of their followers (Gefolgschaften). These slowly evolved into small kingdoms and they
transformed into bigger political unities during the seventh century. Around the time of the Frisian king Radbod (ca. 680-720) the Frisian territories probably became more unified since Radbod apparently had access to a large quantity of soldiers during his battles against the Franks. The power base of Frisia is thought to have been situated in the 68
region of Westergo. One of the main arguments for this theory is the high concentration of gold finds in this area, which would indicate an early medieval power center. This idea has been criticized by Bazelmans, Gerrets and Pol (1998). They argue that one should take
Mostert 2009, 103 64 Dijkstra 2011, 359-362 65 Halbertsma 2000, 298-299 66 Nicolay 2014, 363-366 67 Heidinga 1999, 8-9, Dijkstra 2011, 363 68
post-depositional factors into account. The high concentration of golden finds, with a total 69
weight of 1.5 kilograms, might also be the result of relatively favorable conditions for metal detectorists and the fact that in this region 60 percent of the ‘terp’ settlements were
excavated from the second half of the nineteenth century onwards for their fertile soils. The large quantity of gold depositions in the Frisian areas is often seen as an indicator for an unstable social-political scenario in which multiple kings or warlords tried to assert or maintain their power. Bazelmans, Dijkstra and De Koning (2004) have argued that the 70
center of the Frisian power was located in the riverine area in the modern-day provinces of South-Holland and Utrecht. Historical sources mention that Frisian kings such as Aldgisl and Radbod have resided in these areas. Again, they see the high amount of gold depositions in the northern part of the Netherlands not necessarily as an indicator of a process towards higher political complexity. Instead, it might be an indication for intensified political contacts with the middle part of the Netherlands caused by special social-political developments that took place there. The above shows that it is hard to reconstruct the 71
social-political situation of Merovingian Frisia based on the archaeological record since this archaeological record might be interpreted in different ways.
The third group of people are the Saxons, who inhabited the eastern part of the
Netherlands. These Saxons, originating from the North-Sea coast would have expanded their habitation area in the centuries after the Roman empire. The Saxon cultural area is, however, hard to identify on the basis of the available archaeological material. The
theories about the social-political character of the society of the Saxons is for a large part based on historical sources, mainly written from Frankish perspectives. Their society is often viewed as a confederacy of democracies, without a central (royal) authority, that only was united under one leader in times of war. The lack of elite burials in the Saxon area was used in order to support this idea. Recently this idea has been challenged and greater archaeological emphasis is now being placed on the formation of power and the search for ‘central places’. Examples of Saxon central places might be the fortresses found in the German Münsterland, although these date from later periods. The lack of archaeological 72
visibility of elites might also have something to do with depositional factors. 73
Bazelmans et al. 1998, 21 69 Dijkstra 2011, 364, Nicolay 2014, 231-232 70 Bazelmans et al. 2004, 8-9 71
Van der Velde 2011, 161-162 72
Nicolay 2014, 27-30 73
5.2 Trade and exchange during the Merovingian period
As stated above, a distinction can be made between luxury or prestige goods and
commodities. A second distinction can be made between the ways in which goods could change of owner during the Early Medieval period. Trade was just one way in which goods could be exchanged. A lot of goods were acquired as a (reciprocal) gift, by means of taxation or even were robbed from another person or group. The exchange of luxury goods most likely happened within elite spheres. The degree of elite control over trade of commodities is under debate. A third distinction must be made between regional trade and long-distance trade. The quern stones from Mayen are, like Mayen ceramic wares , 74
viewed here as commodities that were exchanged mainly by means of trade. Since many remote areas were consuming Mayen querns (at least in the Roman and Carolingian periods) they should not be viewed only in the light of a regional economy but also as products that were part of long-distance trade. It should noted that the way in which trade and exchange were organized might have varied over time during the Merovingian period. During the sixth century, the economy of the Roman Empire might, despite a collapse of the imperial society, still have functioned in large parts of Europe. It is, however, unlikely 75
that this was the case in the Netherlands. The seventh century and certainly the early eight century witnessed the emergence of emporiae all over North-Western Europe which can be seen as the result or even a cause of a changing economic scene.
Early Medieval sources are often silent about trade and exchange, especially when it comes to commodities. In the case of Mayen querns, an exception might be a letter from Charlemagne to the king of Mercia written around 796 in which he mentions ‘black
stones’. However, it is more likely that Charlemagne meant building materials instead of 76
querns in his letter. Especially for the Merovingian period the knowledge about trade and exchange is for the largest part based on archaeology. Although artifacts can tell
something about trading contacts between different areas, the way in which these contacts were organized and the role of (elite) agents will not always be archaeologically visible. This makes it hard to apply a single system of exchange to the Merovingian economy. Instead, it might be viewed as an economy that was organized along ‘diffuse and fuzzy lines’ , containing elements of different types of economies in which various forms of
Wickham 1998, 283 74 Hodges 1998, 6-7 75 Peacock 1997, 709 76
exchange coexisted. The Rhine was one of the main transport routes of North-Western 77
Europe during the Early Medieval period and fulfilled an important role in the interregional and international trade. It linked the German Rhineland and the middle Meuse region with the areas around the North-Sea, Anglo-Saxon England and the Scandinavian world. The excavations of settlements found along this river in the Netherlands, especially those along the Old Rhine such as Oegstgeest and Katwijk Zanderij, show that these settlements had access to a wide variety of goods from different areas. Although these settlements show 78
imported goods, harboring facilities and crafting activities, they are often not classified as trading settlements. It is often hard to identify trading settlements from the Merovingian period. The fact that Merovingian settlements contain a high amount of imports does not contribute to the current knowledge about how the distribution of imports took place. 79
Based on the archaeological record of early medieval sites, some of the main products that were shipped via the Rhine were wine (in wooden barrels), grain, metal, ceramic wares and last but not least, quern stones from Mayen. 80
These goods were unquestionably transported with ships. Unfortunately few Merovingian shipwrecks are known from the Netherlands; shipwrecks from the period 300-600 have not yet been found. From the seventh century, ships have been found mainly along the Rhine and along the coast. Except for canoes, many of these ships only survive in small wooden fragments. Also, many rivets which were used in shipbuilding have been found within Merovingian sites. The wood of many of these ship fragments, such as the ones found in Oegstgeest and Naaldwijk originates from the German Rhineland. This is another 81
indicator for an extensive trade network that stretched along the river and beyond.
Although no Merovingian wrecks containing tephrite querns are known, they were probably shipped in Rhine barges that would have had similarities with those that are known from the Roman period. A ship dating from the ninth century that was discovered at Lüttingen contained both unfinished querns and ceramics from Mayen. This supports the idea that 82
there was, at least in the Carolingian period, an association between both Mayen
Theuws 2012, 43-44 77 Jezeer 2011, 113 78 Dijkstra 2011, 375-376 79 Dijkstra 2011, 308-310, 316 80 Brouwers et al. 2015, 6-22 81
Hörter 1994, 40-41, Grunwald 2015a, 204 82
industries, as Redknap (1987) has proposed earlier. A special role within the trade 83
networks of the Early Medieval period is often attributed to Frisian sailors. They were the most prominent traders in North-Western Europe during this period. Historical sources dating from the seventh to the ninth century indicate that they were present in many
countries. They might have controlled the Rhine trade from the late fifth century, but their 84
role likely started in the late sixth or early seventh century. It should be noted that Frisian 85
traders belonged to a small professional group that had experience in sailing over rivers, over sea or perhaps both and that not every ‘Frisian with a boat’ took part in this trade. 86
Especially during the Carolingian period Frisian sailors played an important role in the distribution of Mayen querns to Britain and Scandinavia. However, between the late fifth 87
and the early eight century exchange was mainly focused on trade on the continent via the Rhine. Only a very small amount of ceramics from the German Rhineland seem to have left the continent during this period. As shown in chapter 3, archaeological surveys seem 88
to indicate that no tephrite querns have reached Britain during the same period. 89
Although the above might suggest that trade was mainly based on a ‘free market’ in which Frisian shippers made grateful use of the opportunities that were available for them as they functioned as intermediaries in the trade between different areas, it is likely that there was some kind of elite control during this period. The elite should not necessarily be
viewed as agents that tried to get a grip on the trade networks by means of taxation or tolls but may also be viewed as driving factors behind this trade, besides other agents as the peasantry and the church. The trade networks must perhaps be viewed as complex networks of close relationships. Because of the demand of the elite for prestige goods, some communities were encouraged to engage in long-distance trade in order to meet this demand. As this trade with ‘outsiders’ often took place beyond political boundaries, as has been shown above, long-distance exchange networks had their own rules, which did not necessarily follow the existing political structures. The elite became aware of the
significance of these long-distance trade networks and tolls and customs were established Redknap 1987, 90 83 Lebecq 1983, 269-274 84 Hodges 1982, 87-88 85 Dijkstra 2011, 320 86 Hodges 1982, 124-125 87 Dijkstra 2011, 318 88 King 1986, 95, 119 89
in order to control this exchange. A few sources from the late sixth and early seventh 90
century indicate that there was a toll (teloneum) system within the Frankish empire that was managed by royal authorities. The teloneum was a tax on transport of goods and possibly also on the sale of goods, which was paid at toll stations. Ecclesiastical
institutions could be exempt from paying toll but could also profit from the toll system as they were often granted the right on the interest of a certain toll station by the royal elite. It was also possible that the king granted an ecclesiastical institution the right to charge tolls itself. No written sources from the Merovingian period seem to support the existence of a 91
toll system within the Rhine estuary. In the case of Zuid-Holland, the elite might have played different roles within trade and exchange. There may have been system in which the elite gave goods to their followers in order to maintain their power. Another possibility is that the agents that had access to commodities had to pay a proportion of their wares by means of toll. Several historical sources seem to suggest that trade during this period 92
was first agreed by means of treaties between members of the elite. It is likely that Mayen products were traded in the Netherlands for raw materials such as wool, hides or salt.93
5.3 Agents within the production and distribution of Mayen querns
What becomes clear from the above is that the Rhine functioned as one of the main transport arteries within the trade networks of Europe during the Early Medieval period. The Frisians seem to have fulfilled an important role in distribution of commodities that came from the Rhineland. The elite might have fulfilled different roles within trade and exchange, both as initiators and controllers. What does not become clear, however, is on whose initiative tephrite querns were produced and distributed in order to meet the demands of the consumer markets. Early Medieval written sources are silent about this topic. The Mayen quern production was intensified during the Roman period. Military orders are often seen as the most important factor in the distribution of querns during this period. However, Roman civil settlements also often contain tephrite querns. The querns left Mayen as finished products during this period. As stated previously, from the late 94
Hodges 2012, 126-127 90 Ganshof 1958, 5-7, 14-20 91 Dijkstra 2011, 376 92
Hodges & Whitehouse 1983, 93-94 93
Mangartz 2006, 29 94
seventh century at least a part of the querns left Mayen in a semi-finished form and were shipped to emporiae in which they were finished. This makes it likely that there were contacts between the agents of the Mayen quarries and those within the emporiae during this period. As stated previously, it is uncertain if the querns were finished in Mayen during the sixth and early seventh century or if they left Mayen in an unfinished form during this period.
Redknap (1999) imagines that the Mayen ceramic production center could have enjoyed aristocratic or monastic protection but he also thinks it is possible that it was an
independent industry that could rely on its raw materials and its reputation regardless of outside help. As it is likely that the two industries of Mayen were associated, the same 95
might have been true for the quern quarries. The fact that Mayen querns were often shipped over rather long distances during the Early Medieval period makes it almost inevitable that this trade happened under elite control. The same can be said about the production process. Although the number workers at the Mayen quarries during the Early Middle Ages was smaller than the number of workers the Roman period, which is
estimated at 585 persons, there was probably still a large group of skilled workers present here during the Merovingian period. This makes it likely that there was some kind of 96
organization, rather than that the quarrying was performed by different ‘free entrepreneurs’.
Redknap 1999, 135 95
Mangartz 2006, 31 96
6. The distribution of tephrite quern stones to the Netherlands during the Merovingian period
There are indications that the production of querns decreased in the Mayen area and that the markets shrank. As a result of a reduction of population, we have no clear image of the consumer market for quern stones or ceramics originating from Mayen. In order to get an idea about the supply of Mayen querns to the Netherlands during the Merovingian period, finds of tephrite within Merovingian contexts in the Netherlands will be plotted on a map.
Parkhouse (1997) already tried to produce a map of the distribution of tephrite querns during the Early Medieval period in North-Western Europe. However, his map can now be seen as outdated and incomplete because it only includes the data of pre-1997
excavations. The map shows finds dating from the entire Early Middle Ages and is therefore of little use for investigating the distribution during the Merovingian period. Parkhouse admits that due to this lack of chronological precision the map fails to demonstrate any significant changes in the distribution, for instance between the Merovingian and the Carolingian period. 97
One could ask oneself what this archaeological survey of querns in the Netherlands can contribute to the knowledge of Early Medieval trade and exchange. The area now known as the Netherlands was far from a political, ethnical or economic unity during the
Merovingian period. Focussing only on sites within the Netherlands, inevitably an incomplete picture of the pattern of exchange of tephrite querns will emerge. However, given the fact that the Netherlands can be seen as an area of interplay between different political or ethnic groups, it might contribute to the knowledge about the trade and
exchange between these groups.
6.1 Methods
Different steps were taken in order to create an overview of the distribution of tephrite querns in the Merovingian period. Merovingian archaeology has traditionally been an archaeology of grave fields. However, in the last few decades more and more Merovingian settlements have been excavated. First, an inventory of known Merovingian settlements within the Netherlands was made by using different kinds of sources. Publications that
Parkhouse 1997, 4 97
cover the settlement histories of certain areas, such as South-Holland and Noord-98
Brabant , combined with a dataset provided by my supervisor and a list of Merovingian 99
settlements created by the Rijksmuseum van Oudheden100, formed the basis for this inventory. It should be noted, however, that a large part of these settlements have habitation phases that do not only date from the Merovingian period; many settlements already existed in the Roman period or continued into Carolingian period. Secondly, excavation reports of these settlements were studied for finds of tephrite, both in the form of (semi) complete querns and in the form of smaller fragments. Most of the excavation reports of (recent) excavations were available within the digital repository for Dutch archaeology.101
Archis, the database for archaeological finds in the Netherlands, was used in order to get an overview of Merovingian sites within the Netherlands. It was also used to create a list of early medieval querns. Many of the querns within this database, however, are
inadequately dated. A large part of the reported querns within Archis are dated to a period reaching from 350.000 BC to 1999 AD! This makes the database useless for chronological research on this topic. Also, the material of the querns in the database is sometimes
unclear; some records do not mention if one is dealing with a tephrite quern stone or a quern stone that is made out of another kind of material.
The third step was plotting the results on a map in order to create an overview of the distribution of querns in this period. Different colors were used on the map: a green dot for a settlements that contains tephrite within a Merovingian context and a red dot for
Merovingian settlements that do not contain any tephrite. Lastly, yellow dots were used for settlements that contain tephrite, but of which it remains uncertain if it dates from the Merovingian period or not. The map was created by using the open-source geographic information system application QGIS.
Now that the political situation and the organization of trade and exchange have been outlined in chapter 5, the next chapters will look into further detail what the pattern of distribution of tephrite querns looks like in different parts of the Netherlands. This research will investigate five regions. This division in different regions is loosely based on the
Dijkstra 2011 98
Kimenai & Huijbers 2016, 257-304 99
Langbroek & Van Leeuwen 2014 100
https://easy.dans.knaw.nl/ 101
‘archeo-regions’ that are used within Dutch archaeology. These archaeological regions are areas in which there is a lank between landscape and settlement history.102
Fig. 6: The five different regions of the Netherlands that have been investigated for presence of tephrite quern stones within settlements. 1. Riverine area and Rhine estuary, 2. Western (coastal) part of the Netherlands, 3. Northern part of the Netherlands, 4. Eastern part of the Netherlands, 5. Southern part of the Netherlands.
The first region will be the riverine area and Rhine estuary, which would be the most logical region to encounter tephrite querns due to the favorable location of the settlements next to the Rhine. The second region will be the western part of the Netherlands, with a focus on settlements in the coastal region. Coastal settlements are sometimes seen as special players within the trade networks of the Early Medieval period. Due to their favorable location near the North-Sea coast, these ‘marginal’ coastal settlements would have had access to different imported commodities and therefore they fulfilled an imported role in the exchange of these goods.103 The third region is the northern part of the
Netherlands, especially the ‘terp’ area of Frisia. These areas are seen as having cultural and political links with the western coastal area. They are, however, also often viewed as
https://archeologieinnederland.nl/bronnen-en-kaarten/archeoregios-0 102
Loveluck & Tys 2006, 140-169 103
part of a so-called North-Sea culture and therefore they might have had more links to Scandinavia, Denmark and Northern Germany instead of with the rest of the Netherlands during the Merovingian period.104 The fourth region that will be investigated is the eastern part of the Netherlands. During a large part of the Merovingian period, the Frankish empire had little influence on this area. It is generally accepted that this part of the Netherlands was lying in the western ‘periphery’ of the Saxon cultural area during the sixth and seventh centuries.105The final region that will be investigated is the southern part of the
Netherlands, corresponding to the modern-day provinces of Noord-Brabant and Limburg. These areas, especially the city of Maastricht, were situated near the core of the Frankish empire and therefore more cultural ties are expected with Belgium and the northern part of France. The Maastricht-Tongeren region, for instance, is often viewed as an important territory of early christianity, with bishop Servatius as a binding factor.106
6.2 Problems
Although a distribution map will provide a general image of the distribution of tephrite querns during this period, some problems regarding this approach must be taken into account before general conclusions can be made. Already mentioned above, the first point of critique is the fact that distribution maps of archaeological artifacts will almost always provide an incomplete picture. Merovingian archaeology has traditionally been an
archaeology of grave fields and only in recent decennia the attention has shifted towards settlements. The fact that no Merovingian tephrite querns were found in a certain part of the Netherlands does not necessarily mean that no tephrite was distributed to these areas: absence of evidence is not evidence of absence. It might also have something to do with limited archaeological activity in these areas. Distribution maps of artifacts are maybe as much an inventory of archaeological excavations as they are a survey of archaeological finds. At the same time a concentration of dots on the distribution map does not
necessarily mean that this area had better access to tephrite querns; a concentration might also mean that there has been more archaeological activity in this area.
The second problem with investigating the distribution of tephrite during the Merovingian period is dating the querns. As shown above, minor typological differences in quern shape
Heidinga, 1999, 7-13 104 Van Beek 2011, 95 105 Willemsen 2014, 133-135 106
and size might be identified between querns from different periods. This would make it possible to date them on the base of their morphological features. However, the shapes and sizes of quern stones that are found within settlements are not always described within excavation reports. Moreover, a typology is often of very little use since querns are almost always broken or worn-out which makes it often impossible to classify them. The distribution map does not make a difference between a settlement were multiple (semi) complete querns are found and a settlement were only a small fragment of tephrite is found. Querns within early medieval settlements seem to be broken more often than
querns within Roman settlements. Querns within Roman settlements often seem to be in a more complete state and still have diagnostic characteristics. It remains unclear why this is the case. Parkhouse (1997) states that this might have something to do with the fact that Roman quarries extracted stone that was less likely to become fractured. However, this image might also be the result of selective recovery of material during archaeological excavations.107 The fact that in early medieval settlements often only fragments of these querns survive makes it very hard to date them based on their morphological features. In this case they can only be dated based on their archaeological contexts. If no other (datable) artifacts, such as pottery or metal objects, are found within these contexts, their period of consumption remains uncertain.
When a (fragment of a) tephrite quern is found within a Merovingian context, especially when dealing with small fragments, is does not necessarily mean that it is of a Merovingian production date. Formation processes of the archeological record should be taken into account. In some cases, quern stones are not found within their primary contexts and they might therefore be viewed as residual finds. A tephrite quern that was distributed to a settlement, consumed and then finally discarded in the Roman period might have lied on the surface in fragments as litter waste during the Merovingian period. A fragment of tephrite is then excavated within a Merovingian context, but this does not mean that is of a Merovingian production date. The problems mentioned above make it necessary to briefly outline the contexts in which the tephrite querns were found; the following chapters will therefore examine these contexts for each region of the Netherlands into further detail. To makes things worse, quern stones that were used during the Merovingian period do not necessarily have to be produced during the Merovingian period. This idea is supported by archaeological evidence from England. Studies on West Stow, an Early Anglo-Saxon settlement in eastern England, show a possible reuse of Roman querns. Of the seventy
Parkhouse 1997, 5-6 107
