Skip to content. | Skip to navigation

Personal tools

The Role of Fishing in the Late Mesolithic (Ertebølle Culture) in Northern Europe

— filed under:

Dr. Ken C. Ritchie

The centrality of marine resources to the Ertebølle cultural adaptation has been hypothesized since the First Kitchen-midden Committee recognized the anthropogenic nature of some of the large shell heaps found in Denmark back in the mid-nineteenth century, and recent work with stable isotopes has confirmed that the sea was indeed, the source of much of the food consumed during this period (e.g., Fischer et al. 2008; Price et al. 2007; Richards et al. 2003; Tauber 1981). Archaeological evidence in the form of the location of sites at good fishing localities, procurement technology (i.e., fishhooks, fish traps and fences, leister prongs, and nets), and vast quantities of fish bone has confirmed that fish constituted a major part of the marine resource base (Andersen 1995; Enghoff 1994; Fischer 1993; Pedersen 1995). The Ertebølle fisheries – an important, if not the most important part of Late Mesolithic subsistence – must be recognized and described if we are to advance in our understanding of cultural variability during the Ertebølle period.

Water-screening for fish bones at Asnaes Havnemark

Looking at the fish bone record, I examine the relative representation of the different types of fish at various sites, the skeletal elements present, the sizes of the fish, and any associated technological remains to identify the seasonality and methods of procurement for this resource. Methodological issues of preservation and recovery are also addressed. In addition to previously analyzed Danish assemblages (Enghoff 1983, 1986, 1989, 1991, 1994, 2009; Jensen 2001; Larsen 2005; Richter and Noe-Nygaard 2003; Rosenlund 1976), my project includes fish bones I have identified from eight sites in eastern and western Denmark (Asnæs Havnemark, Bøgebjerg, Dragsholm, Fårevejle, Havnø, Jesholm, Lollikhuse, and Nederst).

From a methodological standpoint, project results indicate the impact of screen-size and excavator skill to the final results. First, excavation techniques create a significant issue for the analysis of archaeological fish assemblages, in that the method of recovery has a strong impact on the results. While bone sizes vary between individual fish, consistent differences exist between species.
 Fish bone from Dragsholm

The decision to use sieving, whether it is conducted with or without water, and the size of the mesh selected all influence which bones are recovered. Remains from larger-sized fish, such as those of the cod family, are recovered in large numbers even with 4 mm mesh screens, while the bones of smaller fish pass through. In fact, there appears to be a statistically significant correlation between the mesh sizes used during excavation and the percentage of codfish in the results: the larger the screen size, the higher the percentage of cod. This is especially problematic in light of the fact that all of the assemblages from eastern Denmark were excavated using mesh sizes of 3.5 mm or larger, while almost all those from western Denmark were excavated using mesh sizes of 2 mm or less. Clearly this variable must be accounted for before results from the two areas can be compared. To this end, a series of bulk samples were taken during the excavations at Dragsholm and Asnæs Havnemark. These samples were screened through nested geologic screens with mesh sizes of 4, 2, and 1 mm. Identification of the remains allows a quantitative assessment of how many bones are potentially lost during excavation and the effect of such differential loss on species representation. Results show that while smaller fish (e.g., eel and herring) are recovered in greater numbers with smaller screens, and some fishes are missed entirely with larger screens (e.g., three-spined stickleback), the general impression of these two sites that were excavated with 4mm mesh-size screens was accurately captured. The bias introduced by differing skill levels of the excavators is a thornier issue, but at least its impact on the Ertebølle fish assemblages has been recognized and described. Less experienced excavators recover fewer remains of smaller fish and are less likely to recover non-vertebra elements.

Using the primary data generated by my own analysis of fish bones as well as data available from the published literature, I can confirm that there are differences in the assemblages between eastern and western Denmark. On the island of Zealand, codfish make up the majority of most assemblages, while in Jutland eel, flatfish, codfish, or cyprinids can all be the predominant species. The Jutland assemblages generally have a higher diversity, defined as having a more equal distribution of specimens across species. Environmental factors surely played a role in what fish were available, and differences do exist in bathymetry, bottom type, and other marine characteristics in eastern versus western Denmark; nevertheless, these results point to differing emphasis on various procurement methods in the different regions (and may also suggest different seasonality of maximum fishing success). Stationary fishing structures such as fish fences and traps might have been favored in western Denmark, while more active methods, such as hook-and-line fishing or spearing, were preferred in the East. The high numbers of codfish in eastern Denmark (and the large size of some individuals) suggest that fishing was conducted throughout the year, whereas in western Denmark the evidence suggests a more seasonal (warm-weather) focus. These data agree well with other differences in material culture between eastern and western Denmark (e.g., distribution of T-shaped antler axes, Limhamn axes, shoe-last axes, bone combs and rings, pottery form and decoration) that demonstrate a division centered on the Storebælt (Andersen 2001; Fischer 1981; Petersen 1984; Zvelebil and Rowley-Conwy 1984). An additional finding of this project that is of considerable interest is the intra-site variability in many fish bone assemblages. Where stratigraphic layers represent different periods in time, they often display different assemblage characteristics (i.e. relative abundance of fishes or sizes). At the shell midden site of Nederst, there is high variability both vertically and horizontally. Ertebølle fisheries were, in many instances, a dynamic phenomenon. By identifying regionally patterned and temporal variability in something as fundamental to a society as its subsistence regime, this project contributes to a broader understanding of the nature of the Ertebølle cultural adaptation.

Document Actions