Académique Documents
Professionnel Documents
Culture Documents
Knowledge
&
Power
The Social Aspects of Flint-Dagger Technology
in Scandinavia 2350-1500 cal BC
Jan Apel
Uppsala
200l
ISSN 1404-1251
ISBN 91-973674-2-7
Jan Apel 2001
PhD dissertation 2001
Coast to coast-book 3
Series editor: Helena Knutsson
Design and layout: martin hgvall
Cover design: Lotta Sundberg and martin hgvall
Cover drawing: Errett Callahan
Revision of the English text: Neil Tomkinson
Published and distributed by: Coast to coast project
Department of Archaeology and Ancient History
Uppsala University
St Eriks torg 5, SE-753 10 Uppsala, Sweden
Department of Archaeology
University of Gothenburg
Published with aid from The Bank of Sweden Tercentenary Foundation
Printed by Wikstrms, Uppsala 2001
341 051
Contents
Preface .............................................................................................. 5
1 Introduction ............................................................................... 7
The theoretical stance ....................................................................... 7
Scandinavia, 2350-1500 cal BC .................................................... 10
Transitions and material culture .................................................... 13
Outline of the dissertation ............................................................. 14
Jan Apel
Contents
Preface
Jan Apel
a peculiar interest in Danish flint something that a customs officer at
the port of Gothenburg was once made aware of.
Donald Broady and Thomas Hkansson commented on drafts and
contributed to the content of the book. I have had interesting discussions
with Debbie Olausson and Anders Hgberg throughout the years.
Berndt Andersson generously placed his collection of Neolithic artefacts
from sterunda at my disposal. Wladyslaw Duczko helped me to
translate Polish texts and Neil Tomkinson revised my English. Kalle
Lindholm helped me to make distribution maps, Martin Hgvall put
the book together and Lotta Sundberg and Alicja Grenberger made
drawings. High Voltage fuelled my spirit.
I am especially grateful to Carola and Ellinor for being there, for all
their support and for putting up with me, especially during the last
couple of months. We shall spend this summer together all four of us.
A significant part of my time as a full-time PhD-student has been
spent on museum collections, studying flint daggers and flint flakes.
The staffs at the following museums have helped me out with practical
details: the Archaeological Museum in Stavanger, the Archaeological
part of the Gothenburg City Museum, the Gvleborg County Museum,
the Halland County Museum, the Historical Museum in Bergen, the
Historical Museum in Stockholm, the Historical Research Centre at
Lejre, the Jmtland County Museum, the Kalmar County Museum,
the Lund University Historical Museum, the Moesgrd Museum in
rhus, the Gustavianum Museum at Uppsala, the Museum of Natural
History and Archaeology in Trondheim, the National Museum in
Copenhagen, the Oldsagsamling of Oslo University, the Uppland
County Museum, the Vsterbotten County Museum and the Vstmanland County Museum.
During my years as a PhD student, I received funding from the
following foundations: the Berit Wallenberg Foundation, the Birger
Nerman Foundation, the Knut Stjerna Foundation, the Mrten
Stenberger Foundation and the Vallsgrde Foundation.
Jan Apel
Berthga, 26 Mars 2001
Chapter 1
Introduction
This book concerns two main subjects. On the one hand, it deals with
the ways in which the division of labour and the use of prestige technologies are related to social complexity in traditional societies. This
discussion concerns the theoretical aspects, as well as the questions and
observations raised by my involvement in technological experiments in
flint. On the other hand, it also deals with a cultural-historical problem
the nature of the social organisation of the Late Neolithic agrarian
communities in Scandinavia. The social interpretation presented in Chapter 10, springs from a dialectical research strategy in which the theoretical issues of Chapters 2-4 are confronted with the results of the studies
of the flint industries in Chapters 5-9 here seen as a prestige technology
in a stratified society.
This introduction is divided into three sections. It begins with a
presentation of the chosen theoretical stance. This section is followed
by a brief presentation of the Late Neolithic and the Early Bronze Age
in Scandinavia. A third section, in which the subject of each chapter in
the book is presented, follows.
Jan Apel
agency, maintained in modern social theories in general, cannot be applied
uncritically to questions regarding traditional societies. Due to current
politics, individual agency is considered to be a far more important
factor in social change than it ever was in traditional societies, and
probably more important than it is in modern societies as well. This
does not mean that we should abandon modern social theory for more
dogmatic, classical, social theories. However, we have to consider the
specific characteristics of traditional societies and relate them to the
modern actor/structure-based theories.
In this context, traditional societies is a general term referring to all
prehistoric and historically known societies that did not adhere to
modern economic principles. In Chapter 3, the holistic perspective on
culture reproduction of the French sociologist Emile Durkheim is used
to argue for the necessity to adjust modern social theory in discussing
traditional societies. Since important aspects of Durkheims sociology
have been incorporated into Pierre Bourdieus theory of practice,
Bourdieus ideas will play an important role in this book as well. I do
not believe that the recognition of the fact that individual agents, through
strategic actions, can improve the quality of life for themselves, as well
as ruining it for others, is the greatest achievement of the social sciences.
We all know that this is true and we all feel that we have a considerable
scope in which to act according to our wishes on many occasions in our
social life. Thus, as I see it, the scholars that today, in their polemics
with the proponents of holism and structuralism, have brought these
ideas into the scientific debate are only stating the obvious. No, the
greatest intellectual achievement in the history of the social sciences is
the recognition that important aspects of the ways in which we act are
governed by conditions that lie outside ourselves, in the form of norms,
institutions and social facts that we are seldom aware of. In fact, quite
often these cultural dispositions are difficult to recognise and appear to
be natural. I feel that these achievements are neglected in the current
archaeological debate and that they therefore need to be re-introduced
once again. Chapters 2 to 4 will be devoted almost exclusively to an
attempt to understand this relationship, and certain aspects of this
problem will be returned to in the other chapters as well.
In this dissertation, I reject the idea of the existence of a fundamental cultural relativism that is suggested in some quarters in the
archaeological debate of today. I also argue that archaeology, like any
8
Introduction
other contemporary social phenomenon, is based on the experience of
the object that is given by the senses. However, since this realism is one
of the two pillars of positivism (von Wright 1993 [1957]:145), I must
add that I do not think that archaeologists must adhere to the second
pillar of positivism, namely that the experience given by the senses should
be used only to formulate general, cross-cultural laws. Even though
such laws can be useful tools in trying to understand certain aspects of
prehistory, an archaeology that concerns itself only with such work is
sterile. An outside, modernistic perspective must incorporate an emic,
inside perspective if the social aspects of prehistory are to be discussed.
Thus, I strive in this book to establish a dialectic relationship between
experience and a more formal approach. In Chapters 2 to 4, a theoretical
preconception of Late Neolithic society is established by references to
social theory and anthropological parallels. The conclusions of these
chapters are confronted with different types of flint materials in Chapters
5 to 9.
I also propose that an external reality is able to communicate itself
through scientific principles. This is my personal view of how the world
works and it is based on the ideas of scholars like John Searle, Steven
Lukes, Maurice Bloch and others (see Chapter 3). They maintain that
there are cognitive faculties of perception that are common to all humans.
However, the important discussion of social archaeology should not
revolve around whether or not there is an external reality. A
constructivistic approach to this matter, for instance, that presented by
Pierre Bourdieu, also takes the social aspects seriously, even though it is
based on entirely different principles. The important thing that should
be acknowledged is that sociality, present in the external world or constructed can be objectively studied through concepts like symbolic capital,
habitus and production fraternity. In fact, I suspect that the material
culture of any age may be studied within the framework of Bourdieus
economy of symbolic assets. It has been appreciated that, while studies
of material culture diminished during the 20th century within academic
disciplines such as ethnology and ethnography, we have become more
and more dependent on artefacts in our everyday life (Miller 1987:3).
This state of affairs actually justifies archaeological-artefact studies
because, in the long run, they may contribute to a better understanding
of our own society.
Jan Apel
Introduction
8). All through LN I, northern Jutland remained a core area for the
production of lancet-shaped, flint daggers (Rasmussen 1990:37,
Vandkilde 1996:13, Apel 2000a & b:141, see also Chapter 8). Since
these were the first flint daggers produced in Scandinavia and since my
thesis uses daggers to discuss the development of social complexity as
shown by related prestige technologies it was necessary discuss extensively
the conditions in northern Jutland during LN I in this book.
The battle-axe and pitted-ware traditions of Denmark, the southern
and central parts of Sweden and the south-eastern parts of Norway
merged into a unified material expression in the first half of LN I. In
the second half, a Late Neolithic tradition with several common traits
embraced the central and southern parts of Scandinavia. It included the
production and consumption of characteristic flint artefacts, such as
bifacial sickles and arrowheads, daggers and spoon-shaped scrapers. In
addition, objects such as shaft-hole axes in stone, slate pendants, bone
needles and bronze axes also characterise this period. The netice area
in central Europe influenced this tradition during LN II. The singleburial tradition and the production of fishtail-flint daggers and elaborately
shaped, bone needles, obvious copies of metal artefacts, are cases in
point.
A collective burial tradition was practised in the southern and central
parts of Scandinavia during LN I. In northern Jutland, a type of gallery
grave with a single chamber was common; it was initially intended for
a single burial, but was later used as a collective grave (see Fabech 1988).
A type of gallery grave with several chambers, sometimes with round
portholes made of stone, was at the same time erected mainly in the
central parts of Sweden and in the south-eastern parts of Norway. This
tradition apparently had its roots in northern France (Johansson 1961).
While these collective graves were used, a proportion of the population
was buried in inhumation graves in flat ground in Skne (Strmberg
1975) and certainly elsewhere too, although they are not common in
the archaeological record.
The Late Neolithic is characterised by evidence of a considerable
landnam that is interpreted as the result of an increased population.
Thus, at the transition between MN A and B, c. 2800 BP cal, a landnam
took place in areas that earlier had few indications of Neolithic
populations, such as the central part of Jutland (Kristiansen 1982:263).
Eventually, a landnam also affected the southern and central parts of
11
Jan Apel
Scandinavia during the Late Neolithic (Kristiansen 1982:263). There is
today consensus among archaeologists that the population of Late
Neolithic Scandinavia practised an economy that included crop growth
and animal husbandry (Hedeager & Kristiansen 1988, Welinder 1998,
Ethelberg 2000). This was also true of large areas of Scandinavia with
no previous indications of Neolithic technologies, such as the Norwegian
west coast (Prescott 1995:175), the remote, forested areas of Swedish
provinces such as Smland (Lagers 2000) and Vrmland (Heiman 2000)
and along the Swedish north-eastern coast (Knigsson 1989, Welinder
1998). However, in the inland areas of Sweden, north of the Mlaren
valley in the east and the province of Vrmland in the west, huntergatherer groups were still present and would be so for a long period to
come.
The population of southern and central Scandinavia lived in long
houses during the Late Neolithic. Two-aisled, long houses have been
recovered from the Norwegian west coast and eastern central Sweden
in the north (Bakka 1976, Johnson & Prescott 1993, Andersson &
Hjrthner Holdar 1989) to Bornholm and the southern parts of Jutland in the south (Nielsen & Nielsen 1985, Ethelberg 2000). At Myrhj
in the province of Himmerland in Jutland (Jensen 1973), Fosie in the
south-western part of Skne (Bjrhem & Sfvestad 1989) and Stngby
in western Skne (Artursson 2000), several, Late Neolithic, long houses
that may have been in use at the same time have been excavated. Twelve
house structures of a slightly different kind, dated to the Late Neolithic
and Early Bronze Age, have been recovered at the Hagestad 44 site in
south-eastern Skne (Strmberg 1992:61). Although the excavator was
reluctant to consider these houses as forming a small hamlet (Strmberg
1992:88), such an interpretation must at least be considered, especially
since none of the houses overlapped one another. Consequently, it may
be assumed that several domestic units lived together in permanent long
houses forming small hamlets during the Scandinavian Late Neolithic
and Early Bronze Age. During the latter part of the Late Neolithic, the
size of the houses increased, particularly in central and eastern Denmark, and continued to do so in the Early Bronze Age as well (Vandkilde
1996:285). Helle Vandkilde (1996:285) has suggested that this trend
should be connected with the concentration of metal objects that this
region displays.
It has been pointed out that few social interpretations of the Late
12
Introduction
Neolithic have been made, compared with the Early Bronze Age. Most
of the recently published descriptions of Early Bronze Age society in
Scandinavia acknowledge the institutionalised, hereditary, inequality.
As regards the Late Neolithic, however, consensus has not been reached.
Some scholars regard the period as socially undifferentiated as while
others regard it as socially differentiated (see Vandkilde 1996:259, Fig.
278). Vandkilde suggests that Late Neolithic society was group-oriented
and that the emerging production and consumption of bronze objects
were collectively controlled. During the first period of the Bronze Age,
larger quantities of metal became available and this contributed to the
disintegration of the collective structure (Vandkilde 1996:306).
Jan Apel
phase in Jutland. There are no signs of bifacial sickles and spoon-shaped
scrapers on sites dated to the early part of LN I and consequently they
should probably be dated to the later part of LN I. These sickles were
produced well into the Early Bronze Age. Daggers with lancet-shaped
blades date to LN I (types I-III, according to Lomborg 1973), while
daggers with fishtail handles date to LN II (types IV and V according to
Lomborg 1973). Daggers of Lomborgs type VI, as well as bifacial strikea-lights (Lomborg 1959), belong to the Early Bronze Age. Thus, the
Late Neolithic archaeological material stems from differences of local,
regional and chronological origin.
14
Introduction
of the pathways to power in transegalitarian societies and Pierre
Bourdieus discussion of the economy of symbolic assets are also
introduced. The chapter also includes an ethno-archaeological example,
based on Jean Browns close study of traditional iron-working in Kenya,
in which the theoretical concepts are brought to bear. The chapter ends
with a description of the social role that the flint-dagger technology
might have played during the Late Neolithic and the Early Bronze Age
in Scandinavia.
In Chapters 5 to 9, the ideas discussed in Chapters 2 to 4 are
confronted with archaeological materials. Thus, Chapter 5 consists of a
research history where previous attempts to define bifacial-reduction
sequences in the flint production debitage are presented. The main goal
of this chapter is to create analytical tools that can be used in the
following investigations of flint debitage from an area close to natural
flint resources (Chapter 6) and flint debitage from an area without
natural flint sources (Chapter 7). In each of these two chapters, the flint
technologies of each area are related to the previous theoretical discussion.
Thus, these chapters contain descriptions of several Late Neolithic/Early
Bronze Age sites with flint artefacts that have been analysed by myself.
In Chapter 8, the flint-dagger typology is discussed in relation to the
previous results concerning the organisation of the production. I hereby
deal almost exclusively with previous research on the Late Neolithic
flint daggers. The main idea here is that the typology does not only
reflect a chronological sequence but also certain recipes of action related
to controlled production in workshops. Here, the system used to classify
the flint daggers referred to in the following chapter is presented. This
chapter also necessarily investigates the chronological significance of the
flint-dagger typology, proposed by Lomborg (1973) to be able to define
contemporaneous dagger types and thus workshops.
In Chapter 9, the Swedish flint-dagger material is presented and
related to materials from other parts of Scandinavia, as well as northern
Europe. In this I investigate whether the geographical distribution of
flint daggers, understood as expressing exchange networks, can be used
in order to understand the character of the political organisation of the
Late Neolithic and Early Bronze Age. I also discuss the possible presence
of workshops as shown by the spatial organisation of dagger types.
Finally, in Chapter 10, the results from the other chapters are
15
Jan Apel
summarised and a detailed interpretation of the social structure of the
Late Neolithic/Early Bronze Age, agrarian societies in Scandinavia is
presented.
16
Chapter 2
The aim of this chapter and the two following chapters is to present a
theoretical framework that can be used to understand the way in which
the Late Neolithic societies in southern Scandinavia organised and used
their flint technology, especially how it was used as a vehicle for social
reproduction. It will initially proceed from an evaluation of the flintdagger technology based on experiments (Callahan & Apel, ms). Important issues to be discussed are the degree of skill needed to make
daggers and, more importantly, how this craftsmanship may have been
reproduced through the generations. The purpose of this investigation
is to delimit the production context of the flint daggers, both geographically and socially. This initial discussion ends with an interpretation of the reproduction strategies involved in the production of flint
daggers, which is fairly tentative in its nature. In order to formalise this
interpretation, it will be related to a school of thought concerning the
question of social reproduction established by Emile Durkheim and
further developed and modified by Marcel Mauss, A. R. RadcliffeBrown, Raymond Firth, Maurice Bloch and others. This theoretical
discussion will be presented in Chapter 3, and in Chapter 4 it will be
related to the complexity debate within archaeology and anthropology
and to an explicit theory of practice suggested by Pierre Bourdieu. Subsequently, terms such as symbolic capital and habitus will be used to give
social meaning to the reproduction of the technology and its role,
actively and as a metaphor, in the reproduction of the Late Neolithic
society.
Jan Apel
classical philosophy, the term techne embraced what we today call craftsmanship and artistic know-how (Lbcke 1988:540, Bernstein
1987:205). A definition of techne, in relation to Late Neolithic flintknapping, is anticipated and discussed in the following section of this
chapter. It will in turn be used to delimit areas of dagger production
within a larger consumption area, but, more importantly, to try to delimit
those segments of society that made and used flint daggers. Yet, in order
to get further understanding of the Late Neolithic societies in
Scandinavia, it is not sufficient to define techne. The social aspects of
dagger technology from an inside perspective, its fronesis, i.e. the ethical
knowledge involved in the process (Bernstein 1987:204 f.), or the good
and valuable aspects of the technology in a culture-specific sense, have
also to be considered. Since a proper understanding of technology can
be found only in the dialectic (or hermeneutic) relation between techne
and fronesis, Chapter 3 is aimed at gaining an understanding of the social
aspects of technology.
18
The distinction between the modern western view and other views
on this matter can be traced to the dichotomy between formal and
informal ways of learning. It has been suggested that formal, institutionalised learning, for instance, in the form of the western educational
system, differs in a radical way from learning by doing, which is the
way in which traditional apprenticeship systems supposedly worked.
To support the relevance of this dichotomy, differences, such as the use
of language versus activity as the major vehicle of instruction and the
creative versus conservative outcomes of formal versus informal teaching,
have been put forward (Pelissier 1991:87 f.). However, recent studies
indicate that so-called informal apprenticeship systems include a fair
proportion of theoretical learning. This has led Catherine Pelissier to
suggest that the dichotomy between formal and informal learning should
be avoided (Pelissier 1991:88 ff.). This conclusion may be part of a
tendency in the social sciences of today to avoid the terms hinting at
cultural evolution that pervaded earlier writings on the subject. Earlier
informal ways of learning based on apprenticeship were, implicitly or
explicitly, regarded as a primitive form of education that had been
replaced by formal means of learning in developed countries. It is
interesting to note here that Pierre Bourdieu, at least in his earlier writings,
stresses the differences between institutionalised education versus learning by doing in traditional societies (Bourdieu 1977:87), especially as
regards the form of asset which is referred to as cultural capital. However, recently Bourdieu seems to have moderated his opinion on this
subject in a way that makes his view more compatible with Pigeots.
I shall now proceed to unfold technology and try to pinpoint its
ingredients in relation to the character of learning discussed above. Two
19
Jan Apel
ideas on how technology can be studied in archaeological contexts will
be referred. These two concepts, (1) technological knowledge and (2)
the operational chain, differ to some extent from each other, but I find
that the following discussion will profit from including both these
approaches and relating them to each other.
Technological knowledge
The term technological knowledge (Schiffer & Skibo 1987) embraces
the theoretical and practical knowledge that defines a technology and
has three essential components: (1) recipes for action, (2) teaching frameworks, and (3) techno-science. A recipe for action consists of the theoretical knowledge and all the formal rules needed to make an artefact
from the acquisition of raw material through the different stages of
production (Schiffer & Skibo 1987:597). A recipe for action includes a
list of the raw materials used, the tools and facilities employed, and a
description of the sequence of specific actions carried out in the technological process. The term also embraces the rules used to solve any
problem that may arise. Accordingly, a recipe for action will guarantee
that the theoretical knowledge that is involved in a technology can be
transmitted from generation to generation. A teaching framework will
allow the practical knowledge involved in a technology to be reproduced.
This is accomplished through imitation, verbal instructions, practical
demonstrations and self-teaching by trial and error. In a way, it can be
said that the understanding of a technology is related to the apprenticeship that guarantees its survival (Pigeot 1990:136). The teaching frameworks are not solely involved in the transmission of practical knowledge. They are also concerned with the legitimacy behind a technology.
This form of ideology can be based on anything from purely rational
arguments to arguments like we have always manufactured arrowheads
in this way. These types of explanations can be regarded as rationales
that give the master or teacher authority during the class. It is also likely
that this kind of knowledge, especially when it takes on an esoteric
appearance, can be used in order to control the practical know-how
that is the foundation of a technology (Hodder 1990:156). In other
words, as we shall see more clearly in the following chapters, those who
formulate the rationales of a technology also control it.
The third component of a technology is, according to Schiffer and
Skibo, techno-science. In this part of technology, the scientific founda20
Jan Apel
main problem is the lack of living informants. In archaeological, technoscientific studies, living informants are simply replaced by, for instance,
a petrographic study and it is anticipated that the result of this study
will reveal patterns that will explain certain aspects of the selection of
raw materials in prehistoric times.
22
In the following text the term gesture refers to the kind of socially constituted gestures that
Mauss referred to as technical acts.
Jan Apel
different stages and events of a technology and to comprehend their
operational chain (Schlanger 1994:144).
The concept was introduced into archaeology and further developed
by the French archaeologist Andr Leroi-Gourhan (Leroi-Gourhan
1964, 1965, Schlanger 1994:144 f.). In his work, important aspects of
Mausss ideas were incorporated in the development of a functionalistic
view of techniques. It was pointed out that archaeologists had to be
aware of the structural and functional aspects of techniques and thereby,
much in line with Durkheims functionalistic interpretation of social
institutions and practices, techniques were analogous to biological
organisms. In a similar way to that in which the limb or organ is
structured, serves a purpose and operates in a certain manner, LeroiGourhan regarded the technical element, gesture or procedure. An important conclusion from this idea was that archaeologists were urged to
document the elementary means of matter of a technology, because
they were considered to be the building stones that, put together in a
sequence, would reveal the technical aspects of a technology. In the case
of lithic technologies, the elementary means of matter was regarded
as the recognition of different techniques, such as percussion, abrasion,
etc. (Schlanger 1994:145). It also deserves to be mentioned in this
context, since it shows Leroi-Gourhans close attachment to the
Durkheimian tradition, that he was the first archaeologist to use
structuralistic principles explicitly to explain an archaeological material
the Palaeolithic cave art of western Europe (Leroi-Gourhan 1968).
Jan Apel
ting of the waste material from the production of a prehistoric flaked
artefact, the different production stages must be defined. Each of these
definitions must correspond to an end product or a series of complete
or fragmented flakes that can be distinguished from other flake types
(Geneste 1989:443, Sellet 1993:108). An important consequence of
this is that, even if the experimenting flint-knapper intuitively feels that
he or she has moved from one production stage to the next, because of
a changed mental template, this stage definition cannot be used until it
has been empirically defined in the lithic debitage. The lack of prehistoric
informants ensures that the credibility of the experimenting flint-knapper
is based on his or her ability to replicate all the conceivable aspects of
the prehistoric artefacts in question. This means that a true replication
of a prehistoric artefact demands an initial analysis in which the production has been defined in detail in archaeological materials. Thus, a
true replica, as opposed to a simulation, is defined as a copy that is alike
in all attributes, while a simulation is alike in only some attributes
(Callahan, personal communication). Because of the fact that there are
few circumstances in which the entire production process can be defined in archaeological contexts by refitting, true replicas are seldom
made by flint-knappers today. Instead, unknown stages, for example,
in the making of flint daggers, can be simulated by the knapper and the
debitage can be defined. These definitions can then in turn be confronted with a prehistoric debitage to see where the newly defined characteristic traits occur. In this way, the work of the archaeologist/lithic
analysts and the flint-knapper proceed in a hermeneutic fashion.
A classic example of how such a emic research strategy works is
Kjel Knutssons investigation of knapped quartz assemblages at the
Bjurselet site in Vsterbotten, northern Sweden (Knutsson 1988b). The
main goal of this investigation was to understand the categorisation of
lithic materials that was used by the prehistoric populations of northern
Sweden, i.e. to identify emic categories in a prehistoric material
(Knutsson 1988b:11). First of all, the quartz material from the Bjurselet
site was classified into basic formal categories, such as cores, scrapers,
debitage, etc (Knutsson 1990:18). At random, a sample of the number
of quartz flakes recovered during the excavations of the Bjurselet site
were subjected to a use-wear analysis. A certain proportion of the flakes
displayed use-wear. These were then tentatively divided into several emic
categories based essentially on the character of the edge. Knutsson then
26
Jan Apel
Knowledge
Know-how
Explaining
Acting
Explicit memory
Unconscious memory
Communicative
Intuitive
Theoretic memory
Muscle memory
Jan Apel
No artefacts will be produced until the flint-knapper has learned to
control the repertoire of gestures involved in the production. In relation
to Schiffer and Skibos terms previously discussed, one may say that, in
order to make remarkable objects, it is not sufficient to have access to
the recipe for action; one must also take part in the teaching framework
(in which practical know-how is transmitted from generation to generation). A modern example can be used to illustrate how knowledge
and know-how interact in a systemic context. When someone is learning how to ride a bicycle, theoretical knowledge in the form of the
positions to be adopted, where to put hands and feet, and so on, can be
provided. However, information of this kind is not sufficient. Practical
training will be a vital part of learning how to ride a bicycle, since this
act involves the co-ordination of different gestures and muscles (Hodder
1992:206).
The geographical distribution of a technology will ultimately be
dependent on whether or not different groups of people choose to incorporate it or not. But, independently of free choice, the spatial distribution of a technology will be influenced by at least three other factors
that in turn are related to each other: knowledge, know-how and rawmaterial availability. Individuals or groups can control each of these
factors. The choice of whether or not to embrace a new technology is
open only to those who accept the requirements of these factors. Theoretical knowledge can be controlled by keeping secret the recipes for
action that lie behind the technology. The level of know-how that is
needed to be able to perform a technique can be controlled by introducing exclusive apprenticeship systems open only to members of certain
segments of society. Raw material can be controlled in different ways
by monopolising the sources of raw material.
This conclusion leads to the idea that an advanced technology, i.e. a
technology which demands specialists, will occur only in societies in
which time and labour can be invested for its development and
reproduction. In such a scenario, specialists will have to be subsistentially
supported by others and this will in turn demand a redistribution system in which economic and administrative power is organised beyond
the family unit (Sahlins 1972). On the other hand, one may also argue
that advanced technologies, such as the Scandinavian flint-dagger technology, may appear in communities with a surplus of food that in turn
enables certain persons to become full- or part-time specialists. Such a
30
Raw material
The majority of the Scandinavian flint daggers were made of highquality Senonian flint. This type of flint occurs naturally in chalk cliffs
on the Danish islands (south-eastern Zealand, Mn, Lolland and Falster)
and around the Limfjord in northern Jutland (Fig. 2:2, Becker 1988:46).
In south-western Skne, ice-transported chalk with Senonian flint has
been deposited at Sallerup and Kvarnby outside Malm and on the
western part of Jravallen, an ancient shoreline outside Malm (Becker
31
Jan Apel
Fig. 2:1. A Scandinavian, type I, flint dagger (left) dated to LN I (2350-1950 cal
BC)(from Lomborg 1973) and Solutrean laurel-leaf-shaped point in GrandPressigny flint (right), dated to the period between 20 000 and 17 000 BP
(Drawing made by Errett Callahan).
1988:47). Even though there are rare examples of simpler daggers produced from other raw materials, such as Danien flint, Kristianstad flint
and even quartzite, it is obvious that the overwhelming majority were
made of Senonian flint. It is therefore likely that big pieces of Senonian
flint were a prerequisite for the flint-dagger technology. Accordingly,
this raw material was used in our experiments on order to understand
the production process.
Fig. 2:2. Primary flint sources in southern Scandinavia according to Becker (1993). (A)
Kristianstad flint. (B) Senon flint. (C) Danien flint. (D) Senon flint from Sjlland, primary
and large, secondary sources. (E) Senon flint of the Hov variant. (F) Senon flint from
Jutland with characteristic white spots. (G) Plain senon flint from Jutland.
Jan Apel
of the flakes found on a production site. For instance, all flakes created
during platform preparation will belong to this category, since such
flakes are produced during every stage of the production of a bifacial
tool. A consequence of this is that only a portion of the flakes from the
production of a lithic artefact will be assignable to the different production stages.
In this case, stages relevant to the production of flint daggers have
been formulated after production experiments conducted by Callahan
and documented by myself. In order to be able to use this information
on prehistoric materials, a classification system that will tie the
production debitage to the production stages has to be drawn up. In
this context, it is not possible to give an account of the classification
system used on the experimental material. However, several gestures
that in turn generate different types of diagnostic flakes have been
identified during the experiments. These results are based on a dialogue
between the experimenters and an analysis of the experimental debitage
and original flint daggers and preforms. If one considers these results as
valid, they can also be used in interpretations of the prehistoric making
of flint daggers. The following judgements regarding the degree of craftsmanship are based mainly on Callahans vast experience of making flint
daggers, but also to a certain extent on my own limited experience in
making bifacial flint tools.
34
Stage 2
ROUGH OUT
Stage 3
PRIMARY
PREFORM
Stage 4
SECONDARY
PREFORM
Stage 5
FINAL
PREFORM
Stage 6
FLAKED
IMPLEMENT
Stage 7
RETOUCHED
IMPLEMENT
35
Jan Apel
The purpose of the first production stage (blank) is to acquire a
suitable piece of raw material. This stage involves a certain degree of
theoretical knowledge and a low degree of practical know-how. It is
important that the nodule does not have internal cracks and that it has
the correct proportions: it must be big enough and have the right
thickness in relation to its width and length. After receiving verbal
instructions and participating in collecting potential flint-dagger nodules,
it was possible to collect suitable nodules from shores, fields and chalk
mines.
The purpose of the second stage (rough-out) is to create a centered
edge around the preform. It is possible to create an edge by soft and
hard, direct techniques. I was able to learn this stage after a few days of
practice and it is obvious that the average archaeology student can grasp
this stage quickly, at least on small preforms. It may be concluded that
this stage does not require any great degree of practical know-how.
During the third stage (primary preform), the most important aim
is to see that the preform is given a generous, lenticular cross-section.
This will be achieved by applying a soft, direct technique, for instance,
using an antler billet. If the lenticular cross-section is not achieved, it
will be impossible to make the preform smaller without losing width.
This has to be avoided, since the later stages will be concentrated upon
obtaining a fairly large, width-thickness ratio. The lenticular cross-section
is therefore needed to allow flakes during later stages to reach over the
middle zone of the biface, thereby thinning it down in the process.
During the fourth stage (secondary preform), a dagger preform with
symmetrical contours and with a blade that has a width-thickness ratio
of between 4:1 and 5:1 is demanded. During a considerable part of my
apprenticeship period, I was stuck in the transition between the third
and the fourth stage. I could not thin the preform down without making
it smaller at the same time. In spite of the fact that, through Callahan,
I had access to all the theoretical knowledge needed, in the shape of
what tools to use, striking angles, positions, etc., I was not able to
complete the stage. It became apparent that somehow I lacked a great
deal of the practical know-how needed to cope with the process. This is
a problem that all flint-knappers will encounter to some degree in their
learning processes. This threshold regarding lithic ability and skill can
also be recognised in the evolutionary perspective. It took several hundred
thousand years for Homo to develop the first bifacial handaxes
36
Jan Apel
$
! %&&'&
&
#
!" #,'
# ,'
!"
& ()
&
(* )
!+&
#.
#
!" #'
#'
!" ' '
#
#
#' #
!" '
#.
#
#
!" # -#,
!" -,
& )
& ( #)
Jan Apel
that both prestigious and simpler daggers of each defined type were
produced during the Late Neolithic. Callahans description is concerned
with the prestigious type IV daggers:the degree of skill varies considerably from type to type and up and down the staging. Type IV is tops,
type I C (with parallel retouched blade) next... Type I D is also very
skillful...but is more attainful as its percussion only (2 less stages)
(Callahan, unpublished letter, 17 November 1997).
Callahan is of the opinion that the first three stages contain a relatively low degree of practical know-how. Thus, they can be carried out
by relatively inexperienced flint-knappers after they have acquired the
proper theoretical knowledge. It is tempting to consider the idea that
the early stages and the later grinding stage were undergone by youths
in an apprenticeship system on their way to becoming artisans. As we
have seen earlier, the fourth production stage demands a greater deal of
practical know-how. According to Callahan, this stage is 8-10 times
more difficult to complete than the previous stage. This means that
beginners were not likely to be able to succeed. It is likely that the
fourth stage was unxdergone by journeymen with a great deal of theoretical knowledge and practical know-how, but not yet masters. The
gap in required skill between the fourth and the fifth stages is as great as
that between the third and the fourth stages. Consequently, the fifth
stage (final preform), according to Callahan, is 8-10 times more difficult
than the previous stage. Any mistakes that the flint-knapper makes
during this stage will be difficult to repair: its not just a matter of
skill, as I said, but of perception, observation, and intelligence. Rectify
by looking at the originals (I keep the Hindsgavl cast at hand as I make
my daggers. It's my greatest aid). One must duplicate the angles, flake
scar types, et., as said above. Locking a 3-dimensional template in your
brain isnt easy. It comes with know-how (Callahan, unpublished letter,
17 November 1997).
The fifth stage, according to Callahan, is the most difficult stage to
master. During the production of certain dagger types, for example,
subtype I D with a percussion, flake-scar finish this is the last
stage before the final retouch. If the fifth stage is conducted without
flaws, the following stages, i.e. grinding, parallel retouch and final retouch, will be completed fairly easily in comparison. The following
stages, with the grinding stages as an exception, demand a certain degree
of practical know-how, but they are not as demanding as the fifth stage.
40
Jan Apel
1 2 G
Apprentice/youth
Journeyman/young adult
7 6
Master/Old adult
Fig. 2:5. The seven production stages defined by Callahan (and a grinding stage = G),
graded according to the degree of practical know-how.
1 G
Learning by observation,
simple recipe
4 5
Moderate recipe
7 6
Complex recipe
Fig. 2:6. The seven production stages defined by Callahan (and a grinding stage = G),
graded according to the degree of theoretical knowledge.
43
Chapter 3
Apprenticeship as a Social
Institution
Jan Apel
logical and sociological texts. It will involve a discussion of social and
cultural reproduction that is essential to an archaeology that aims at
considering the meaningful aspects of prehistoric technologies in relation
to social structure. Some key terms used to shed light on the social
structure of the Late Neolithic societies in Scandinavia will be defined.
Certain issues that have been discussed in the complexity debate in
archaeology and anthropology will be recapitulated and related to the
tentative interpretations of the Late Neolithic societies in Scandinavia
presented above. Last, but not least, concepts such as technology, specialisation and division of labour will be discussed in an attempt to
understand their roles in the strategies of gaining and maintaining power
in a stratified society, strategies that, according to the interpretative framework used, led to the emergence and growth of institutionalised
inequality in Scandinavia during the Neolithic and Early Bronze Ages.
The theoretical stance that is proposed in this section originates from
a Durkheimian tradition seen chiefly in sociology and anthropology.
As we have seen earlier, the theories of Durkheim concerning the
reproduction of social facts inspired the development of the operativechain method in archaeology. This method is explicitly used in this
book, as well as in previous articles, to discover the complexity and
geographical distribution of the flint-dagger technology (Apel 2000a,
2000b, Callahan & Apel, ms). Durkheims view of social reproduction
was also instrumental in the development of Bourdieus general theory
of practice. This theory will be used below to elucidate how a technology
in itself, and not only its products, form a symbolic capital (see the
discussion on Bourdieu and his relation to the Durkheimian tradition
of sociology in Broady 1991:100 ff.). Thus, it seems appropriate to
start off by presenting the main ideas of this school of thought and to
follow its development towards the theories of human practice and
individual versus collective agency presented by Bourdieu (1977, 1999a).
Jan Apel
peoples habits, traditions and common ways of living and took the
forms of institutions, laws, moral values and ideologies (Durkheim 1978
[1895]:21 f.). A socialisation process resulted, accordingly, in feelings
of respect and a sense of duty towards the values of social norms. These
values and norms, that to the child appear as compelling, are through
the socialisation process, including upbringing, internalised into the
individual, i.e. they are integrated into the individual personality and
thereby become part of a persons self-identity (Guneriussen 1997:311).
The individual actor will, generally speaking, try to establish a balance
between himself and the community to which he belongs. Attitudes,
values and feelings among individual actors of a tightly knit group of
people will thereby, through time, tend to merge into a common
consciousness (Durkheim 1993 [1897]:86). Durkheim referred to such
embodied moral styles as ethos and, as we shall see, this term was later
used in the early works of Bourdieu as almost synonymous with habitus
(Broady & Palme 1986:12) . Bourdieu later confirmed that ethos, as
well as other classical concepts, such as eidos and hexis, all formed parts
in his habitus concept. The term habitus was initially used by medieval
scholars as synonymous with Aristotles term hexis, which means
posture or gesture (Bourdieu 1991:147). Apparently both Durkheim
and, and as we have seen, Mauss, also used the habitus concept in their
work. Durkheim used it to explain how the early Christian schools
were forced to use a pagan habitus in order to solve the problems that a
change of religion actually meant for the population (Bourdieu
1991:148).
Accordingly, the reproduction of society stemmed from the fact that
cognition was determined by social factors (Bloch 1989:1). In this
respect, the Durkheimian view of social reproduction has been seriously
challenged only by proponents of the idea that culture is the extrasomatic means of human adaptations to nature within the subject of
archaeology (Preucel & Hodder 1996:205). As we shall see, the fact
that Durkheim believed that social and cultural variability should be
assigned to social factors, and not be regarded as adjustments to natural
constraints, did not prevent him from making analogies to Darwins
evolutionary biology. However, in contrast to the evolutionary biologists, who at that time regarded change as a result of natural selection
on the individual level (Persson 1999:672), he was convinced that the
mechanisms of cultural change operated mainly on the collective level.
48
Jan Apel
ideal social order; it was the worship of values created by society, or
rather society worshipping itself. Durkheim believed that he had
demonstrated that religion was a primitive form of social organisation
simply because totemism, as such, was inseparable from the clan-based
social organisation of the Australian Aborigines (Durkheim 1995
[1912]:169). The members of a single clan were united first and foremost
because they had the same name and shared the same emblem. Since
clan members were not necessarily consanguineous and were often
scattered throughout the tribal territory, this seemed to be a logical
conclusion. According to Durkheim, clan unity sprang from a collective
belief that all members had the same relation with the same categories
of things, and from practising the same rites in other words, from
the fact that they commune in the same totemic cult (Durkheim
1995:169). Since Durkheim argued that the role of religion in the
reproduction of the social order was gradually replaced by other
institutions, he had an evolutionary perspective on cultural change, in
the Darwinian sense of the word, that is (see the further discussion on
this subject in Chapter 4).
Thus, religious rituals were regarded as one way of transmitting
knowledge of the social order to the individuals and that orderly social
life among humans presupposed certain sentiments, shared by all
members, that controlled the behaviour of the individual in relation to
the collective. One of the important functions of rituals was therefore
to regulate, maintain and hand down such sentiments from one
generation to another. This functionalistic approach to religion and ritual
may appear blatant and one-dimensional to modern scholars but, as
Morris (1987:111) has pointed out, throughout history there are
examples of how the social role of religion has been acknowledged by
at least certain segments of the population. The following quotation
may indicate that the ruling aristocracy in ancient China was aware that
rituals served social functions: Ceremonies are the bond that hold the
multitudes together, and if the bond be removed, those multitudes fall
into confusion (Morris 1987:111; translation of a quotation from the
ancient Chinese Book of Rites from the third century B.C.). As society
became more and more complex and thereby gradually moved from
mechanical to organic solidarity, religion lost its importance as a vehicle
of social and cultural reproduction in favour of other kinds of institutions
(see Chapter 4). It is important to remember that Durkheim, as op50
Jan Apel
forward, because the idea of social facts is intimately connected with
such questions.
Jan Apel
since all living humans share the same biological faculties of perception.
The relativistic, theoretical stance can be further illustrated by the
following statement, accompanied by a clarification by Bloch: If lions
could speak we could not understand them (Wittgenstein). In other
words, that communications with creatures with a fundamentally
different system of ideas and life is not possible, and surely people with
a different concept of time would in this respect be like lions, since
everyone agrees about the particularly fundamental nature of this proposition. (Bloch 1989:8). Richard Rorty has gone very far in this respect
and sees everything as the result of linguistic principles (Rorty 1989).
He maintains that not only the reality around us but also our personalities and our roles in society are constituted by linguistic praxis. According to this view, humans are shaped and constituted by the same linguistic praxis that creates and reproduces the external reality
(Guneriussen 1997:206).
As regards anthropology, the linguistic turn eventually divided the
field into two groups that both adhered to cultural relativism but based
their assumptions on two fundamentally different principles. On the
one hand the social anthropologists, functionalists and structuralists such
as Spencer, Durkheim, Radcliffe-Brown, Levi-Strauss and Edmund
Leach, emphasised how long-lived, external structures formed human
cognition. On the other hand, cultural anthropologists, from Boas to
Clifford Geertz, maintained that culture is something that is always
negotiable and is created on an individual level. In short, social aspects
are constructed and studied from an outside perspective, while cultural
aspects require an inside perspective: Culture is the fabric of meaning
in terms of which human beings interpret their experience and guide
their action; social structure is the form that action takes (Geertz
1957:33 f.). Perhaps it is not entirely unfair to claim that cultural
anthropologists today are empirically orientated and will have a hard
time seeing the forest for all the trees, while social anthropologists
may be seen as having a mechanical view of society that neglects
individuality. Personally, I prefer the outside view, although I accept
that it was constructed by an anthropologist and is not readily available
for study. As I see it, in line with Durkheim & Mauss (1963) the world
has to be classified in order to be understood and it is not something
that will reveal itself through empirical studies. However, as we shall
see, the cognitive faculties involved in this process cannot be solely
54
Jan Apel
world, archaeologists can safely assume that it is enough to define and
construct the prehistoric object through our senses.
Jan Apel
Hegardt (1997), have argued that Durkheims functional and positivistic
approach made him a forerunner of the processual school of thought in
archaeology. However, as I see it, Durkheims ideas of social facts and
the way in which they function as reproducers of society have very little
in common with, for instance, Leslie Whites or Lewis Binfords view
of culture as mans extra-somatic means of adaptation to nature. On
the contrary, it has been pointed out that these two ways of regarding
culture as determined by nature or by society are two traditions of
thought which exclude each other (Preucel & Hodder 1996). However,
as we shall see below, Binfords view on the emergency of inequality
does have certain aspects in common with Durkheims theory explaining
the principles behind the division of labour in society.
While Tilley claimed that Durkheims holistic view of society influenced processual archaeology, he was also of the opinion that Weber
greatly influenced post-processual archaeologists. Hodder has also
maintained that Webers ideas were neglected by conventional archaeologists and that the latters emphasis on subjective meaning and historical approaches in interpreting social and cultural reproduction
deserved to be introduced into the archaeological-theory debate (Hodder
1986:81 ff.). One of the main ingredients in Webers theory of practice
was that it is not possible to understand the complex set of relations
that make up society without a thorough knowledge of the subjective
aspects of the members of society in their internal affairs (Campbell
1986:203). For Weber, it was vitally important that the positivist search
for causal explanations should be combined with an understanding of
the meaningful individual act, i.e. the meaning that the social act had to
the individual actor (Campbell 1986:203). As we shall see below, Webers
hostility towards holistic, over-individualistic ideas concerning society
can perhaps be traced to his upbringing and political views. Weber observed the world from the perspective of an ontological individualist,
i.e. he regarded everything that exists as phenomena of an individual
nature and he regarded collective phenomena, such as society, group,
class as constructions in the minds of individuals (Andersson & Andersson
1991:7). Thus, Weber maintained that social facts had to be preceded
by an idea that had been formed by an individual agent. Or, as Edmund
Leach has put it, the idea of, for instance, a city must precede the actual
forming of the earliest cities in the Near East (Leach 1973).
This conflict between processual and post-processual archaeologists
58
Jan Apel
abilities are inherited within different social segments and that such
abilities are used in social strategies. This conclusion is even more striking if we consider the fact that these problematic tendencies have been
actively confronted and opposed within the Swedish educational system
during the last 40 years. It is difficult for any individual to actively
change traditional patterns and this may be due to the fact that important aspects of our social abilities are reproduced at an early age within
the family unit (Bourdieu 1999a:120).
Consequently, one of the important theoretical aims of this thesis is
to demonstrate the importance of the social structure in the reproduction of society. My field of inquiry is the flint-dagger technology of the
Late Neolithic period in Scandinavia and I shall try to uncover its social
facts in a Durkheimian sense. I want to discuss the principles which
determined who should be involved in the production of these daggers
and who should be excluded. In my opinion, it is not sufficient to
regard technological talent (an individualistic view) as something that
individuals are born with, i.e. a form of natural talent. On the contrary,
technological talent, just like other social facts, is, according to this
view, mainly reproduced within social contexts. If not, we shall inevitably
maintain a charismatic ideology in which the importance of individual
agency is allowed to completely overshadow the importance of structure,
as regards the social positions of individuals (Featherstone 1994:75). I
shall return to and elaborate on this crucial subject in discussing
Bourdieus theory of practice in Chapter 4.
Jan Apel
Type of intentionality
Person 1
Person 2
Individual intentionality
Collective intentionality
Table 3.1. An example of how the ideas of individual and collective intentionality may
differ in a concrete situation.
Jan Apel
duced the importance of the active individual that Durkheim denied in
regarding collective representations as springing from society. (3)
Through the field concept, Bourdieu can delimit smaller social units.
This allows him to regard the social world as more differentiated than
Durkheim, through his concepts, was able to do. I shall return to
Bourdieus theory of practice in the next chapter.
An alternative view of the relationship between structure and agency
which I find appealing has been put forward by Maurice Bloch (1989). It
will be recapitulated below. However, before proceeding any further, I
would like to suggest, in line with the general aim of this chapter, that the
differences in tradition and upbringing, i.e. structure, might have guided
Weber and Durkheim to opposed views on cultural reproduction.
Webers father was a member of a family that had prospered in the
German linen industry. He was a lawyer and a successful, liberal-minded,
nationalistic politician effective in official circumstances and as a
father very authoritative. His mother was interested in arts and culture,
was strictly religious and concerned herself with charity, something that
was despised by her husband. On the emotional level, Weber took his
mothers side, even though he did not support her religious views. He
came to following his fathers footsteps in that he studied the law and
showed an interest in public matters, but he also followed his mother
in that he cherished cultural refinement and spirituality (Campbell
1986:205). As a political writer, he was a defender of bourgeois ideals
and much of his research can be regarded as an indirect critique of
Marxism (Andersson & Andersson 1991:7) and other holistic traditions
in the social sciences.
Durkheim was the son of a rabbi and grew up in an orthodox Jewish
family in a long-established Jewish community in Alsace-Lorraine. He
was himself intended to become a rabbi and studied for some time in a
rabbinical school until, while still a school boy, he decided not to follow
the family tradition (Lukes 1973b:39). His background marked him
with several lasting traits: Scorn for the inclination to conceal effort,
disdain for success unachieved by effort, horror for everything that is
not positively grounded: the life of the individual within the framework
of the group, truths through their rationally established implications,
conduct by its moral regulation (Davy 1960:15, Lukes 1973b:40).
Lukes further writes: From the time of his childhood, he retained an
exacting sense of duty and a serious, indeed austere, view of life; he
64
Jan Apel
Paris. Apparently there is a connection between a form of socially created ethos and a form of intellectual moral that is, in the case of the
former group, demonstrated in the urge to re-instate science and scientific
research (Broady 1991:47, note 77).
A social situation of a similar kind appeared in archaeology during
the 1960s. It has been pointed out that the emerging processual school
in American archaeology during the early 1960s was developed in a,
middle class, mid-western context (University of Chicago) far away from
the prestigious universities of the east coast (Trigger 1993). It is also a
fact that its foremost proponent, Lewis Binford, with his claims to a
scientific, anti-authoritative, positivistic archaeology, never received a
chair in any of the prestigious universities. In fact, whatever deficiencies
the early processual archaeology may have had (and they have all been
severely scrutinised during the archaeological debate over the last 20
years), its aims were to develop objective methods that would work
independently of the individual researchers personal positions, views
and history. It is clear that the claims of such a research strategy include
the ideas that research should be valued according to the accuracy of the
results and not by the reputation of the individual scholar.
Realism or relativism
Lukes has pointed out that the urge to regard what counts as truth or
valid reasoning as relative to cultural systems of thought has its roots in
assumptions as to how reality is constituted (Lukes 1973a:231, 235
f.). For example, it has been argued that since human perception, and
therefore also the human understanding of the world, is theorydependent, there is no such thing as an theory-independent, external
world outside the senses (Lukes 1973a:235). But just because ones
understanding of how human perception is constituted must be theorydependent, since humans cannot be positive on anything outside their
senses, it does not automatically follow that there are no theory-independent objects of perception and understanding (Lukes 1973a:236).
In order to establish this realism, Lukes is obliged to present social facts
that can be regarded as universal. Thus, he continues to illustrate his
critique with a linguistic example in which the main argument is that
the existence of a common reality is a necessary precondition in order
to learn and understand foreign languages something that we know
is possible. The argument is concluded with the proposition that if
66
Jan Apel
simply the frame of references that humans receive from an outside
source. As we shall see, Bourdieus use of the term habitus is to an
important degree equivalent to Searles use of the word background. It
is important to note that Searles discussion does not verify the actual
presence of an external reality outside our senses. It does, however, demonstrate that all languages, and thereby also the representations that are
made in them, presuppose the presence of an external reality.
A critique of relativism and anti-realism is often dismissed by proponents of these schools of thought simply because it comes from the
outside. The urge to view and study in a distant perspective, the
cornerstone of all positivist/modernist research, is regarded as useless,
since it cannot account for the personal opinions of individual actors. If
cognition is a relative affair, every culture and field of enquiry will have
its own standards, as regards truth and valid reasoning. Critique from
an outside source will appear strange and of lesser value, because it is
not operating within the proper discourse. Post-modern proponents
will therefore tend to regard all systems based on the principles of the
Enlightenment as oppressive (Lyotard 1984). In line with this, a project
aiming at creating universal standards, such as positivism, is considered
to be dogmatic, because it stipulate rules for what is true and what is
not (Hegardt 1997:193). However, post-modernist claims to the discovery of real historical discontinuity are not unproblematic (Featherstone 1997:17 f.). The post-modern critique of universality and totality, the post-modern feeling of being beyond traditional, normative
and modernistic class analysis and the emphasis on discontinuity, written
texts, paradoxes, ironies and the critique of the Enlightenment and the
modernists way of thought may just not be the end of history after all.
On the contrary, some suggests that the post-modern project may be a
necessary break with the dominance of 19th-century, orthodox social
theories but that the relevance of large-scale studies of social change will
prevail (Giddens 1987:41). It is therefore maintained that the necessary
critique of orthodox social theories would be more aptly regarded as
high modernity rather than as post-modernism (Giddens 1990). In archaeology, however, this debate is of less concern, since post-processual
archaeologists in general work in a conventional way but with a new
jargon.
The Durkheimian view of the nature of social reproduction, however tempting and convincing it may seem, is problematic for those
68
Social structure
In the British functionalistic tradition of anthropology, Durkheims ideas
became especially important. A. R. Radcliffe-Brown was heavily
influenced by the French functionalism and his definition of social structure shares important similarities with the meaning of social facts and
collective representations. According to Radcliffe-Brown, structure
should be regarded as an ordered arrangement of components (RadcliffeBrown 1952:9). The components of social structures were individual
human beings. Social structures were created as a result of the fact that
observable social phenomena within a society were not the immediate
result of the nature of the human beings (Radcliffe-Brown 1952:190).
Thus, social structures appeared as soon as a group of individuals structured their internal relationships in some other way than those based on
flock behaviour: There is continuity of the structure, just as a human
body, of which the components are molecules, preserves a continuity
of structure though the actual molecules, of which the body consists,
are continually changing. In the political structure of the United States
69
Jan Apel
there must always be a president; at one time it is Herbert Hoover, at
another time Franklin Roosevelt, but the structure as an arrangement
remains continuous (Radcliffe-Brown 1952:10).
It is important to recognise that the term social structure was created
under the influence of Durkheims idea of society as something that
encompasses the individuals within a social group, especially since the
term is frequently used in the archaeological literature. Radcliffe-Brown
regarded society as a homogeneous, organised and self-reproducing entity
and thereby shared Durkheims holistic view of cognition as originating
from society (Bloch 1989:3). According to Radcliffe-Brown, persons
could not be studied except in terms of their social structure and, conversely, social structure could not be studied except in terms of the
persons of which it was composed (Radcliffe-Brown 1952:194). However, the social structure was part of a social system that also included
certain functional aspects (Hatch 1988:224). Thus, a social system was
seen as consisting of three aspects: (a) the social structure, (b) the totality
of all social usages, and (c) the special modes of thinking and feeling
which we can infer or assume (from behaviour and speech) to be related
to the social usage and the social relations that make up the structure
(Radcliffe-Brown 1937:152). This implies that there may be room for
cultural change and generalisations regarding different societies in his
theory. The main problem with Radcliffe-Browns definition is that it
includes all social relations between persons. Or, as Firth has put it,
When I knock into someone on the street and apologise, we have a
social relation. But is this to be reckoned as part of the social structure?
(Firth 1964:37). In this context, it must be stressed that the (b) and (c)
aspects are always dependent on the shape of (a), i.e. the social structure.
Radcliffe-Brown thereby maintained Durkheims idea that cognition
in general has a social origin. This also means that he shared a relativistic
view, in that he regarded each cultural system as a unique, closed system
and rejected generalisations. In order to argue this point, he used an
example from the Australian kinship systems that were known from
anthropological field studies. The following quotation illustrates his
theory: It was impossible for a man to have any social relations with
anyone who is not his relative because there is no standard by which
two persons in this position can regulate their conduct towards one
another(Radcliffe-Brown 1913:157). The fact that Radcliffe-Brown
used the Australian Aborigines, the people who at the time were seen as
70
Jan Apel
Durkheimian view of cultural reproduction, that Firth failed to produce
a sufficient explanation of the question of cultural change. Even though
Radcliffe-Brown and Firth tried to overcome the problem by introducing
a practical level of cognition, operating in a dialectic process at a
theoretical and ideal level, they failed in their attempts to produce a
satisfactory explanation. According to Bloch, this failure can be traced
to the fact that the second level descends from the first. That is, the
different cognitive levels belong to the same cognitive system, which in
turn is unique to a certain cultural system. Consequently, one of the
main problems with the Durkheimian view of cultural reproduction is
that the organic model of society also leads to the assumption that
societies can be regarded as closed systems in equilibrium (Leach
1973:165). In the systems theory laid out by, for instance, biologist
Ludwig von Bertalanffy (1969) or within archaeology by Lewis Binford
(1972) a distinction is made between open and closed social systems. In
order to be fully effective, a pre-condition of the Durkheimian view of
cultural reproduction seems to have been the idea of a closed social
system in equilibrium. It was suggested that social systems in balance
(equilibrium) should be anticipated in social studies for analytical reasons
(Parsons 1951, Guneriussen 1997:75). However, even if closed social
systems exist, they are, just as biological systems, never in equilibrium
but quite frequently exhibit the characteristics of equifinality (Leach
1973:164). In general systems theory, the term equilibrium refers to
the condition that every closed system strives towards, namely maximum
entropy, i.e. the dissolution of all differentiation. In order to be consistent
with general systems theory, Parsons and, for that matter Binford, should
have used the term homeostatic open systems, i.e. systems that strive to
maintain a high degree of organisation (Guneriussen 1997:76). It does
not take a genius to figure out that social systems tend to be open in the
respect that they are in contact with other social systems. This means
that the members of a social system will always be aware of other ways
of living, and inspiration to change the social structure may very well
come from the outside. What should be considered in the study of
human societies is the degree to which they are open. For instance, the
archaeological remains from the Late Neolithic period in Scandinavia
reveal evidence of contacts over large areas, so in this context we obviously
need to operate with a fairly open social system in mind.
72
Jan Apel
pastoralists and farmers and that their economy was integrated with
these societies. The revisionists urged anthropologists and archaeologists to regard the traditional societies in Africa as formed by historical
and social factors and not as the last remnants of past evolutionary stages
(Wilmsen & Denbow 1990).
Regardless of these views, it may even be suggested that, if we return
to the anthropological descriptions of these societies today, they display
signs that indicate that they have what Durkheim called a strong mechanical solidarity. Obligations, such as sharing, can be regarded as
social facts maintaining a rigid social structure, albeit without any
obvious religious undertones. If this view is accepted, the egalitarianism
of hunter-gatherers is ironically a social construction that may very well
have been created and maintained in order to prevent social stratification.
Bloch seems to be aware of this when he states that the difference between
hunter-gatherers in Australia and the Hazda lies not in their economy
but in their degree of social inequality (Bloch 1989:17). This view is
also consistent with Marxs opinion that ideology appears only in class
societies (Campbell 1986:152 f.).
It was not only the new ethnographic evidence touched upon above
that inspired Bloch to modify the Durkheimian/anthropological view
of social reproduction. Bloch aimed to reintroduce a non-ideological
level of cognition, an idea that was originally suggested by Marx and
Engels in their distinction between ideology and non-ideology. These
concepts were accordingly used to separate social categories and ethical
notions, such as the bourgeois notion of the law of property , that are
imposed on the individual by society, from a kind of cognition that
grows out of experience of the surrounding world. As we have seen, a
non-ideological aspect of cognition has been avoided by the anthropologists who followed in Durkheims footsteps. In order to find further
evidence for a deficiency in the Durkheimian view that cognition must
always preced action and therefore be created by society, Bloch refers to
the important progress made by psychologists studying the development
of children (Bloch 1989:113). Jean Piaget, for instance, suggested that
the child forms concepts as a result of its interaction with the environment and that consequently categories are constructed from the childs
experience of the world. The world from which this categorisation stems
can, according to Piaget, be reduced to the following three components:
(1) the childs own body, (2) the non-human environment and (3) other
74
Jan Apel
point of view. This has led cultural anthropologists, such as Geertz, to
suggest that they have a unique perception of time which cannot be
understood by outsiders and that their social life takes place in a
motionless present(Geertz 1973:404). Geertz article is a well formulated
justification for the cultural-anthropological view of social reproduction.
Accordingly, human thought is consummately social: social in its origin,
social in its functions, social in its forms, social in its applications (Geertz
1973:361). This is a huge claim, since it suggests that anthropology is
indeed the first science, and it is not surprising that this view has caught
on among anthropologists. Geertz is empirically orientated, in that he
denies any underlying structures of human thought; there is no need to
speculate on what goes on inside peoples heads; the meaning will be
apparent through everyday life. In this respect, the article should be
regarded as a critique of structuralism, as it was presented by Claude
Lvi-Strauss, in which the importance of a universal, underlying, social
structure in the reproduction of societies was emphasised. So, even if
Geertz follows Durkheim and Weber in regarding cognition as socially
determined, he parts from Durkheim when he adopts an extremely
individualistic and relativistic approach to cultural studies. It should
be kept clearly in mind that, both here and in the discussion to follow,
distinctions are formulated from the actors point of view, not from an
outside, third-person observer (Geertz 1973:366 f., note. 5). Here he
explicitly refers to actor-oriented (sometimes miscalled subjective) constructs in the social sciences, as they appear in Webers methodological
writings.
The theoretical stance that I take in this book follows Durkheim,
Lvi-Strauss and Bourdieu, in that an outside perspective, in which
theoretical concepts are constructed, is preferred. Bloch, in sharp contrast
to Geertz, claims that, if one chooses to study everyday life on Bali, for
instance, the agricultural work, one will find that the farmers share a
perception of time, formed by the constraints of nature, such as the
weather, the seasons, climate, geology, etc. This is an understanding
shared by all humans, since it is based on brute facts. It seems as if
cultural phenomena related to the past in the present have been preferred
by sociologists and anthropologists simply because it has been in their
interests to study particular aspects of culture. Even today, when there
is general agreement that there is a biological foundation for human
behaviour, there seems to be a consensus in the social sciences that humans
76
Jan Apel
social message (Olausson 2000, Hgberg 2000). Artefacts such as pencils, cars, gloves or trousers all contain these two aspects. It is therefore
important to recognise that the division between the reproduction of
the past in the present and the universal, everyday, practical behaviour in
the present is an academic reduction of reality. This does not mean that
the demarcation of these categories is useless. On the contrary, scientific
reductions are necessary in that they tend to reveal patterns that are not
obvious and therefore are difficult to appreciate by common sense. Bloch
has exemplified how these two modes of cognition may work within a
continuous process rather than as two opposites. In his article Symbols,
song and dance: Is religion an extreme form of traditional authority? he
argues that there is a continuum in which communication from
repetitive (formalised) communication to generative (everyday)
communication represents the two ends (Bloch 1989:22 ff.). In this
example, ritual communication is formalised and repetitive and belongs
near the Past-in-the-present end of the continuum, while everyday language stems from the present cognition. In Table 3:2, a few variables in
speech acts that separate the past in the present from the present are
listed.
It is important to remember that Bloch does not consider the contrasts displayed in Table 3:2 as a dichotomy of two types of speech but
as a continuum between two extreme forms of speech (Bloch 1989:25).
I have found it very interesting to note the circumstances under which
these two extremes are displayed in speech by myself and by people in
my own surroundings. There are many situations in which individuals
refer to the past in the present, using a formalised language in order to
gain audience confidence and support. For instance, at the beginning of
the 20th century, Max Weber pointed out that there was a low degree of
correlation between good scientists and popular academic speakers at
the German universities. He concluded that the popularity of certain
speakers was not necessarily based on the knowledge that they had of
their subject but rather on what were, from an academic point of view,
pretty shallow factors, such as temperament and the chosen tone of
voice (Weber 1991:14). I think that this observation can be seen as
recognising the importance of formalised speech acts in public contexts.
In his article, Bloch convincingly demonstrates how the repetitive and
formalised, ritual language is used in traditional societies to uphold a
form of traditional authority that is used as a means of creating and
78
Jan Apel
Everyday speech acts
Choice of loudness
Choice of intonation
Complete vocabulary
Partial vocabulary
parallels
Table 3:2. Aspects of speech that contrast formalised speech (the past in the
present) from everyday speech (the present), according to Bloch 1989:25).
try to formulate covering laws, i.e. cross-cultural laws. Such laws can be
established only by working on the generative aspects of, for instance,
technology and in this respect Binford is one of the few who have
accepted the consequences of his theoretical conviction.
This means that archaeological artefacts are always to some degree
material expressions of an ideal social structure. Some artefacts, for
instance, blades from conically-shaped flint cores in the Scandinavian
battle-axe context or Late Neolithic flint daggers, contain a large portion
of the past in the present. Other artefact categories, for instance, bipolar
flakes in quartz that has been used in an expedient manner, may contain
a smaller degree of the past in the present. However, since all archaeological
objects are social facts, in the sense that they are the result of what Searle
(1997) has defined as institutionalised facts, since meaning has to be
ascribed to them from an outside source (society), their original meanings
will inevitably be lost when the social structure collapses or undergoes
radical changes. Since the meanings of institutionalised facts, as opposed
to raw facts, are relative, the story behind prehistoric technologies with
a large degree of repetitive cognition (the past in the present) will appear
to be more difficult for us to understand than technologies with a large
degree of generative cognition (the present).
Even if we accept that every technology, as defined in this context,
includes both repetitive and generative aspects, it has proved to be fruitful
to sort technologies into two categories, defined according to their
inherent degree of these aspects. Binfords (1962) division of technology
into technomic and socionomic types or Haydens (1998) distinction
between practical and prestige technologies reflect the need to separate
80
Jan Apel
produced in an exclusive way near the flint source and distributed to be
used as harvest tools and then to be deposited as grave gifts. It is,
according to Knutsson, not a coincidence that this ritualised form of
technology was used in a highly formalised, grave ritual and by a people
that accepted and adhered to the Neolithic package (Knutsson 2000:30).
On the other hand, the production of the Pitted-Ware blades was
apparently not restricted, since evidence of production is found on the
settlements. These blades were also used in a seemingly profane, hunting
activity. Knutsson suggests that these differences corresponded to two
types of mentalities. One ritual version was formed by a sedentary,
Neolithic lifestyle, in which there was a strong need to reproduce what
Bloch would refer to as the past in the present. The other day-to-day
version was formed by a mobile, hunter-gatherer way of life, in which
there was little need for reproducing the past in the present through a
prestige technology or through other ritual activities.
The prestigious blades in evidence in Scandinavian Boat-Axe contexts are the result of an old tradition that can be traced over large areas
in northern, central and southern Europe, as well as the western part of
Asia (H. Knutsson 2000:18 ff.). Knutsson suggests that the grave ritual
in which the technology is embedded should be related to a ritual motif
common to a large region covering most parts of Europe and western
Asia, which was probably rooted in the emergent Neolithic way of life
in western Asia (H. Knutsson 2000:29). These blades, and the
technological process by which they were produced, obviously served
as vehicles for the reproduction of the past in the present in the Neolithic
societies of western Asia and Europe between 8000 B.C. and 2000
B.C. These objects served as a form of insignia for a certain segment of
these societies and their symbolic meaning, i.e. the story that they told,
was widely recognised in the early farming communities. It is interesting
to note that in this case it is not the technological complexity or the
normative style that separates these two blade technologies. Rather it is
the way in which the objects were produced and deposited that gives
the interpretation credibility. Thus, in this case, the interpretation springs
from the archaeological contexts. A parallel situation to these circumstances, concerning the production and consumption of bifacial arrowheads in eastern central Sweden during the Late Neolithic and Early
Bronze Age, has been published (Apel 2000c) and is presented in Chapter
5.
82
Conclusion
In this chapter, it has been suggested that the Durkheimian view of the
reproduction of social facts and collective representations can be used
to understand the reproduction of prehistoric technologies, such as that
of the Late Neolithic, flint-dagger technology over an 800-year period.
However, while Durkheim tended to regard all categories of thought as
social phenomena, I have followed Maurice Blochs suggestion that
cognition basically takes two fundamentally different forms. The past
in the present refers to a type of cognition that is of a social origin. It can
be regarded as analogous to Durkheims social facts and may serve to
reproduce the ideal structure of a society from generation to generation.
It is therefore imposed on the individual, through imitation, upbringing and education by society. The present, according to Bloch, refers to
a cognition that springs from individual experience of the surrounding
world. This form of cognition is not social in its nature, in the sense
that it is handed down through the generations, but is grounded in
human biology. This type of cognition makes it possible to reach an
understanding over cultural borders and over time. The present-abilities
make themselves known in an everyday perception of time, as well as in
the universal way in which the basic structure is identical in all known
human languages.
The past in the present in its purest form is displayed in rites, myths
and religion. Bloch has demonstrated how the formalised language used
in rituals serves as an extreme form of traditional authority by which a
idealistic notion of society is imposed on its members. It has been
suggested, in this chapter, that the past in the present is also a vital
ingredient in aspects of social life that were earlier regarded as purely
functional in their nature, such as technologies. Since the purpose of
the-past-in-the-present strategies is to reproduce an idealistic notion of
the society itself, it follows that technologies that are based on normative
rules and formalised sequences of gestures also serve as vehicles, perhaps
through metaphors, by which ideal notions of the social structure are
reproduced.
It has been demonstrated, by experiments, that the reproduction of
the flint-dagger technology required theoretical knowledge, in the form
of a complicated recipe for action, and practical know-how, involving a
fixed sequencing of gestures. Thus, it follows that the organisation of
83
Jan Apel
the flint-dagger technology, in addition to the finished daggers, should
be regarded as a vital ingredient in the reproduction of the social structures in the agricultural Scandinavia of the Late Neolithic. Since social
facts of this kind are always used in competitions for prestige and power,
the following chapter is directed at an understanding of how technologies function in society.
I hope that the discussion above has shown that the people who
used and made flint daggers were not only the creators of tradition but
that their values and expressions were also formed by tradition. As I see
it, the main point about the dagger tradition was that it should be
reproducible. It was only by being transmittable from generation to
generation that the technology was able to serve its purpose, i.e. to be a
vehicle of the dominant social values of the community.
84
Chapter 4
In the previous chapters it was argued that technologies not only meet
functional needs but also play important roles in the reproduction of
the social structure in a community. In the present chapter this knowledge will be related to certain ideas, on how institutionalised inequality
emerges and grows, that have been put forward in the complexity debate. This will lead over to a discussion of the relation between dagger
technology and social structure during the Late Neolithic and the Early
Bronze Age that will be continued in Chapter 10. Thus, in this chapter
it will be argued that the production of large quantities of high-quality
flint daggers distributed over several generations required a hereditary
system that could be upheld only in a fairly stratified society with
institutionalised heredity.
Some definitions
All human societies display certain aspects of inequality (Price 1995:130).
It is therefore not sufficient, in this context, to define egalitarian societies as societies in which there is no inequality between their members.
An acceptable definition might be that an egalitarian society has no
division of labour based on other principles than sex or age groupings.
In other words, these societies do not have permanent social institutions that can be achieved, or inherited, by persons with ambitions and
a cultural capital that allows them to do so. Since status differentiation
is present among primates and in other animal societies, Price has pointed
out that it is not an issue that concerns anthropologists and archaeologists.
A discussion of the development of hierarchical societies should aim at
an understanding of how ascribed and institutionalised status is introduced, maintained and developed (Price 1995:130).
Since the complexity debate has been focused on the growth of hier85
Jan Apel
archies from egalitarian to complex, a definition of the latter term is
also called for in this discussion. In this context, complex societies are
defined as possessing social and labour relationships in which leaders
have sustained or on-demand control over non-kin labour and social
differentiation is hereditary (Arnold 1996:78). The most complex form
of society that concerns us in this book is the chiefdoms that, according
to Service (1962) bridge the gap between tribal societies and archaic
states. Chiefdoms are politically stratified societies with settlement hierarchies, institutionalised inequality based on hereditary principles and
lite control over domestic labour (Earle 1991:1, Arnold 1993:77).
After Marshall Sahlins elaboration of a theory developed by Karl Polanyi
(1944), chiefdoms have been connected with concepts such as redistribution and pooling (Sahlins 1958, 1972). In such situations, surplus
commodities are collected by a political lite (as taxes or tributes) and
then redistributed downward in order to provide for warriors, specialists, etc. In recent years, mainly as the result of Timothy Earles work on
Hawaian chiefdoms (1977, 1978), the importance of redistribution
and pooling in the formation of complex chiefdoms has diminished.
In most cases chiefs will indeed receive tributes, but these tributes are
not the driving force in the development of complexity, even in cases
when they are actually used to provide specialists and warriors. Instead,
the incitement behind surplus production seems to be the need to create
and acquire prestige goods that are used to establish exchange networks
and thereby form the basis of a prestige economy (Earle 1991). It has
also been suggested that the development of segmentary tribes into
chiefdoms is not the result of a linear process but may take on many
different forms (Kristiansen 1991). In chiefdoms, external contacts with
lite fractions in other communities, often situated quite far away, make
up the foundation for the emergence of the organised production of
prestige items that can be exchanged. In economic terms, the prestige
items will not be as valuable where they are made as they are further
away from the production area. A thorough, cross-cultural, ethnographic
analysis has demonstrated this principle in traditional societies, i.e. statusor wealth-indicating ornamental substances are more often non-local
than local (Taffinder 1998:31 and Fig. 2:5). As regards the Scandinavian
Neolithic period, Jaqueline Taffinder concludes that there is evidence
of prestige exchange, according to her model, in Early and Late Neolithic
contexts but not to Middle Neolithic ones. Especially during the Late
86
Jan Apel
29
46
27
35
38
9
36
24
25
12 39
8
3 11
21
34
49
33
48
10
6
28
16
19
41
37
5 47
42
32
18
23
30
14
17
43
26
2
40
44
7
20
45
13
22
31
Fig. 4:1. The geographical distribution of historically known or present-day societies with
a social structure of tribal complexity (after Sahlins 1968).
15
Jan Apel
regard the division of labour as something that concerned only
economic life it affected all types of institutions (Durkheim 1984
[1893]:2).
During the late 19th century, there was general agreement among
social scientists that one of the most important social aspects of industrialisation was that it seemed to free the individual from the social
restraints of traditional societies. According to the utilitarians, the older
type of social solidarity was simply replaced by the contractual agreement, which, as it was based on intellectual decisions, would regulate
the behaviour of humans towards one another. Durkheim, on the other
hand, was convinced that an unfettered individualism would inevitably
lead to social decay and that the modern cult of the individual had to
be based on a more fundamental principle to secure social solidarity.
Durkheim was a passionate believer in the autonomy of the individual
and the main question that he strove to answer was how this autonomy
could be reconciled with the necessary regulations and discipline needed
in order to maintain and reinforce the bonds that held society together
(Coser 1984:xiii). In his doctoral thesis, he aimed to understand how
social solidarity was maintained as the degree of the division of labour
increased. In this respect, he worked in opposition to the leading German
social scientists at the time, such as Gustav Tnnies, who argued that
the only true solidarity was to be found in the village communities of
the past and that modernity served to undermine this Gemeinschaft
(Coser 1984:xiv). Durkheim agreed that an all-embracing, moral
consensus characterised traditional societies with a low degree of the
division of labour. However, he objected that, even if an increased
division of labour does not result in an increased, all-embracing, moral
consensus, it does not follow that there is a lack of social solidarity in
modern societies with a high degree of the division of labour. Since
laws and sanctions are expressions of the character of the moral codes
within a certain society, Durkheim believed that a classification of the
sanctions practised in different types of societies would shed light on
the relationship between social solidarity and the degree of division of
labour. In comparing societies with a high degree of division of labour
with societies with a low degree of division of labour, Durkheim noticed
a pattern. In traditional societies, with a low degree of division of labour,
a penal law with repressive sanctions served to protect collective and
strongly held, moral beliefs (Giddens 1978:25). In a repressive legal
90
Jan Apel
is always strong. It is interesting to see that Bourdieu, in discussing
these questions, refers to a saying from his own background amongst
the farmers of Barn in south-western France (Bourdieu 1986:43, note
17). It is apparent that Bourdieu, like Durkheim, has first-hand
knowledge of how mechanical solidarity tends to shape individual agency
in accordance with the collective consciousness.
An increasing division of labour seemed to promote the autonomy
of the individual. Since groups with different forms of collective consciousness had to function together, the collective consciousness of any
of these groups could not serve as the sole foundation of law. In order
to maintain social solidarity and thereby to unite people, a new form of
social solidarity emerged: organic solidarity. In such a system, each
specialised group of people (field) functioned together like organs in a
body, since they were dependent on each other. As a result, repressive
sanctions, which, in order to protect society have to rely on a shared
value system, were gradually replaced by restitutive, formalised law.
Hence, a greater complexity in the division of labour correlates with a
progressive transition from mechanical to organic solidarity (Giddens
1978:25).
Durkheim did not believe in unilinear evolution. On the contrary, a
growing complexity, as measured by the degree of the division of labour
in society (or, as in this context, by the relative number of institutions)
was not something that inevitably had to occur in all societies. In order
to understand why some societies developed complex social systems,
Durkheim employed a perspective based on the theory of evolutionary
biology. As we have seen earlier, the term division of labour was introduced by Adam Smith in order to describe a social phenomenon
that became apparent during the early phase of the Industrial Revolution. It was later adopted by evolutionary biologists to explain the
principles of diversity in nature. It is therefore, perhaps, not surprising
that Durkheim regarded the division of labour, not as springing from
the intelligence and will of humans, as the utilitarians would have put
it, but as a general biological phenomenon (Durkheim 1984:3).
For Durkheim, the crucial point to be explained in order to understand the mechanisms behind the growth of complex societies was the
moment at which the segmentary type of society was dissolved as a
result of an increasing division of labour. According to Durkheim, an
advanced division of labour was not possible in segmental societies and,
92
Jan Apel
those who chose this strategy (Darwin 1868). It is ironic that Durkheim
and Binford drew analogies with Darwinian biology in order to avoid
intellectualistic explanations of the emergence of inequality, while
Darwin himself proposed an intellectualistic explanation for the
emergence of Neolithic technologies.
It is important to remember that Durkheims explanation of the
emergence of the division of labour is fundamentally anti-intellectualistic, since it excluded the free will of individuals as an important part
of the process. Durkheim regarded segmental societies, and probably
all types of social formations, as conservative in their nature. He therefore
concluded that any fundamental change in the organisation of society,
such as the abandonment of the segmental principle, most probably
had to be provoked from a source outside society. In this respect,
Durkheims view of social change radically differs from the Marxist
view. The view that Durkheim proposed accordingly provoked a critique
from Marxist scholars, since they regarded it as conservative (Giddens
1978:105 f.). The segmental principle would not be abandoned if population pressure occurred in a society in which surplus of resources and
space was available the problem would simply be solved by letting
new domestic units move to new territories. So, as long as there was no
external phenomenon creating a moral density that forced the society
to abandon the segmental principle, it would, owing to the conservatism of any social organisation, maintain the same principle.
There are ethnographic examples that support the idea that groups
of people, when forced to move to new territories because of economic
stress, will strive to keep their traditional social structure in the new
environment. In the northern Tanganyika region (present day Tanzania), groups of people from two different traditions gradually inhabited the slopes of Mount Meru during historical times. Both the Meru
(mountain farmers) and the Arusha (pastoral Maasai) settled in this
environment during the 17th-19th century as a result of population
pressure within both groups (Spear 1997). Initially the population grew
faster among the Arusha because they were successful in stealing their
neighbours cattle. Thus, successful age-sets rapidly colonized the slopes
as each generation of warriors retired and settled down to clear new
land to farm. The Meru population, however, expanded slowly because the lineages slowly increased in numbers, filled the land and forced
younger members to move. Spear concludes his findings in the follow94
95
Jan Apel
96
Jan Apel
after it has been invented, cannot solely be explained by this theory, it is
interesting to bear it in mind in discussing the evidences for technological
specialisation that occur in the Late Neolithic communities of southern
Scandinavia. In Chapter 10, I discuss whether the archaeological evidence
from Jutland, as regards the emerging flint-dagger technology at the
transition between the Middle and Late Neolithic periods, can be interpreted as the result of a marginalization process.
In later writings, Binford modified his opinions but maintained the
anti-intellectualistic proposition that the origins of agriculture, and
complexity in general, should be regarded as the result of population
pressure. He now argued that the reason why population pressure occurred all over the world almost simultaneously was that humans are
most reproductive in the temperate zone at the mean earth bio-temperature of 14.4 degrees Celsius. From a biological point of view, this
makes sense, since the main regulatory mechanism seems to be mortality in the tropics and fertility in the Arctic, the two coming together
in the temperate zone (Binford 1983:209). This means that when
Homo sapiens reached the temperate zone, it was only a matter of time
before density problem would occur that in turn would provoke new
economic strategies, such as husbandry and agriculture. This process
evolved especially quickly on the American continent, since the population there had gone through a disease filter, and were no longer
threatened by native disease organisms, which allowed a population
explosion (Binford 1983:209 f.).
Binford maintained that the role of craft specialisation in the development towards greater complexity in traditional societies should not
be overemphasised (Binford 1983:221). The implications of the antiintellectualistic ideas presented by Durkheim and Binford suggest that
it is problematic to explain specialisation in traditional societies by references to modern economic theories. It is, in fact, rather common in
traditional societies for specialists to be low-class or outcast individuals
or families with no real political influence. This implies that the occurrence of craft specialisation is a secondary effect of hierarchical differentiation. In such a scenario, specialists are people or families who, as a
result of, for example, population pressure, are forced to move into
low-production areas and to earn their livings as part- or full-time
specialists. Dean Arnold followed this line of reasoning in explaining
individuals or families who have turned to ceramic craft specialisation
98
Jan Apel
maintained by limiting the number of children, for instance by abortion, contraception, abstinence or infanticide (Binford 1972:437). This
may indicate that the social problems following from increased population density are known in these societies. In more open social systems,
such as transegalitarian societies, a steady population density is maintained by letting surplus families, or individuals, move to new areas. In
such societies, Binford suggests that during a period of population
growth, the increase in the size of the local family or extended family
will tend to split into equal units, as long as there is space enough to
allow the new family to act as a unit of production. When this is no
longer an option, the segmentary principle will break down and groups
of people will become marginalized, while others will keep or develop
their privileges (Binford 1983:2).
The main problem with the population pressure theory is that several studies of mobile hunter-gatherers published during the 1960s demonstrated that these societies never seemed to allow the population to
grow above the carrying capacity of the land. It has also been suggested
that larger, more complex societies, such as the Iroquois, in some cases
strive towards maintaining egalitarian structure in villages with several
thousands of inhabitants (Trigger 1990). So the main question one
should ask, as regards specialisation and the emergence of complexity is
perhaps why the Neolithic way of life was ever considered, or allowed,
in the first place. The crucial point that needs to be explained is why
hunter-gatherers are forced to become sedentary (Binford 1983:208).
Since sedentism is regarded as the ideal way of living in western societies, this question has seldom been asked before. It has simply been
taken for granted that humans will prefer a sedentary way of life before
mobility, if they can find ways to support themselves in one place.
Binford, however, presents several examples that indicate that mobile
hunter-gatherers never consider a sedentary way of life because all their
knowledge of the world and how to use aspects of it in order to live a
good life is based on mobility. In other words, they regard mobility as
the ideal way of living. Binford does not accept Childes and Braidwoods
intellectualistic explanations of the emergence of the Neolithic technologies which are based on the assumption that humans will become
sedentary when there is an abundance of available resources in one place.
On the contrary, he maintains that, when present day hunter-gatherers
(he explicitly refers to the Nunamiut Eskimos in Alaska and to Aus100
Pathways to power
From an empirical point of view, Hayden has tried to answer the questions about the emergence of inequality. He has conducted a crosscultural analysis of anthropologically documented, transegalitarian societies and has divided them into sub-groups according to their relative
degrees of complexity. These sub-groups are defined by the different
strategies that they use to gain and maintain power. Hayden discusses
only communities with hunter-gathering and/or agriculture in their
economies and does not consider pastoral groups. In Scandinavia, pollen evidence speak in favour of an economy that included cultivation
during the Late Neolithic (Hedeager & Kristiansen 1988, Welinder
1998, Heiman 2000, Ethelberg 2000, Lagers 2000). An increasing
number of excavated sites with houses and large structures implying
sedentary hamlets in both the southern and central parts of Scandinavia
are strong arguments for this case (Jensen 1973, Bjrhem & Sfvestad
1989, Strmberg 1992, Artursson 2000). According to Hayden, the
main reason for the growth of social inequality was the continuous and
increasing production of surpluses and a social order in which aggrandizers were encouraged to use these surpluses in strategies to achieve
power. Hayden maintains that the aggrandizer concept covers a human
type that has an aggressive and acquisitive nature and is present in all
human populations.
Initially, Hayden distinguishes between transegalitarian communities in which the social structure is reproduced within independent
households and those in which it is reproduced within a larger corpo101
Jan Apel
rate group consisting of several households that share the exclusive control of an important subsistence resource (Hayden 1995:18 f.). Within
these two types of communities, the strategies used to accumulate power
are different in their natures. In the case of the Late Neolithic societies
in southern Scandinavia, we shall have to consider both of these possibilities. It is plausible that a corporate community had exclusive control of the raw-material sources for dagger production and that the
apprenticeship was open to all children belonging to this group. A
situation like that would indeed have certain advantages for the reproduction of the technology. It would, for instance, mean that a larger
base was available to choose apprentices from. On the other hand, it is
also plausible that the flint-dagger technology was organised by families in control of the knowledge and the know-how inherited in the
technology rather than of the flint source. Such a scenario would also
guarantee the exclusiveness of the technology to a certain segment of
the community and would in fact also make it possible for communities with interesting recipes for action to use them in social strategies. I
shall return to these questions below.
In both communities based on independent households and those
based on corporate groups, three different levels of complexity have
been recognised and described: Despot, Reciprocator and Entrepreneur
societies (Hayden 1995:19, Fig. 2). The developments of these forms
relates to how an increasing amount of surplus was used by aggrandizers in strategies to maintain and increase their power (Hayden 1995:25).
Haydens approach to an explanation of the occurrence of institutionalised inequality is explicitly materialistic in its nature. This is revealing
in many ways, especially since it makes the argumentation logical within
its own realm. However, Haydens ideas are, as I see it, compatible with
the general theory of human practice and symbolic capital that has been
developed by Bourdieu. The incorporation of some key concepts from
Bourdieus theory of praxis into the discussion will demonstrate that a
purely materialistic approach to the problem can be moderated (see
below).
In Despot communities, the strategies used by aggrandizers to maintain and increase control include provoked warfare and reciprocal exchange (Tab. 4:1). In Reciprocator communities, provoked warfare and
reciprocal exchange are still used, but bride-wealth and child-growth
payments, and to a certain degree also different rituals, are also intro102
103
Jan Apel
(2)Initiation rites (into secret societies).
(3)Special formal training (in a certain technology, often involving esoteric knowledge of some kind).
If a family is able to let children participate in such rituals or training, their symbolic capital will increase and the symbolic capital can
then be used in marriage negotiations. Those wanting to marry a child
who has undergone an expensive ceremony, or initiation rite or a special
form of apprenticeship have to be able to afford it, i.e. they will have to
have an equal amount of symbolic capital.
It may now be suggested that one of the main functions of the flintdagger apprenticeship system was to maintain and enhance the symbolic capital that this craft gave, through the generations. It is not necessary to believe that the flint smiths belonged to the lite of their
communities in order to suggest this. The ontological security that the
membership of a production fraternity of this kind gave its members
is, in itself, a sufficient explanation of its reproduction. I shall now
proceed to relate Haydens important ideas to the theory of symbolic
economics (in the widest meaning of the word) presented by Bourdieu.
Egalitarian
Despots
Reciprocators
Entrepeneurs
Chiefdoms
..xxxxxxxxxxxx
xXXXXXXXXXXXXXXXXXXXxxxxxxxxxxx..
..xxxxxxxxxxxxxxxxXXXXXXXXXXXXX lite/Nonelite
xxxxxxxxXXXXXXXXXXxxxxx.
Child growth
.xxxxxxxxxxXXXXXXXXXXXXXXlite/Nonelite
Investment
.xxxxXXXXXXXXXxx
Feasts
- Solidarity
- Reciprocal Exchange
- Competitive
Cults: Ancestors and Others
XXXXXXXXXXXxxxxxxxx.
xxxxxxxXXXXXXXXXXXXXXXxxxxxxxxxxxxxxx
.XXXXXXxxxxxx(lite only)
..xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxXXXXXXX
Jan Apel
empirical experience, and the rejection of intuition as a useful method
in sociology, are things that Bourdieu shares with the Durkheimians
and makes him a proponent of a kind of applied rationalism. Broady
claims that the classical opposition between explanation and understanding is transgressed by this dialectic relationship between the hypothesis
and the statistical verification (Broady 1991:397). The fact that Bourdieu
claims that sociology should construct its own subject in this way means
that he promotes a continuous dialogue between closed rationalism, on
the one hand, and naive empiricism, on the other. This applied
rationalism is thus not deductive or inductive but is best described as
constructivistic in its character (Broady 1991). In archaeology, a similar, theoretical, approach has been presented by Kjel Knutsson and
successfully applied during archaeological excavations in eastern central
Sweden (see Apel et al. 1994, 1996 & 1997, Hallgren et al. 1996 &
1997, Knutsson & Lekberg 1996). As we shall see in Chapter 5, an
archaeology that concerns itself only with rational arguments that can
be verified in formal ways will tend to be very sterile. On the other
hand, an archaeology that is not firmly embedded in a scientific way of
thinking but concerns itself only with learned speculations is also, at
least to my mind, not worth the trouble.
One of the cornerstones in Bourdieus theory of practice is actually
the idea that social complexity is the result of a certain kind of diversification process. The main idea being that all institutions derive from a
common ancestor institution. As we have seen, Durkheim believed
that religion was an institution of this kind. Bourdieu considers the
domestic unit, the family, as the fundamental human institution
(Bourdieu 1999a). The domestic unit is one of the few institutions
that actually practice a positive reciprocity (Sahlins 1972). It is also in
the families that important aspects of habitus are created. According to
Bourdieu, the strategies for social reproduction proceed from the fact
that families are corporate bodies that aspire to maintain their social
being and accordingly preserve all their possibilities and privileges. This
is achieved through strategies for childbirth, marriage and rules of inheritance, economic strategies and strategies of education (Bourdieu
1999a:31). Among transegalitarian societies, strategies for concentrating power include warfare, bride-wealth, child-growth payments, membership of corporate lineage groups, and claims to supernatural power
(Hayden 1995:25). In this respect, Bourdieu and Hayden share funda106
Symbolic capital
According to Bourdieu, different assets, in the form of different kinds
of capital, i.e. physical, economic, cultural or social capital, are transformed into symbolic capital when they are perceived by social agents
with categories of perception that allow them to recognise them, acknowledge them and appreciate their value (Bourdieu 1999a:97). Broady
(1991:169) has defined Bourdieus term capital as values, assets and
resources that are symbolic or economic in their nature. Bourdieu has
been especially interested in symbolic capital, and the shortest definition of symbolic capital is what is recognised as valuable. In this sense,
symbolic capital is a very general term, but it is important to recognise
that it does not concerns single individuals; it solidly rests on the
common consciousness of groups of people (Broady 1991:169). It is
therefore not enough to have a certain possession or skill; it is also required
that this asset shall be accepted as valuable by a group of people. In
107
Jan Apel
relation to the subject of this book, one could argue that exceptional
flint-working skills cannot, in themselves, be regarded as a form of
symbolic capital until they are recognised by the collective as such. One
important consequence of Broadys definition of symbolic capital is
that economics, which in certain schools of economic thought has been
regarded as an independent field of research following its own logic,
can in this context be regarded only as a variant of symbolic capital.
Cultural capital
This form of symbolic capital is, according to Bourdieu, a very important form of capital in western societies, such as France. It is connected
with the kind of prestige that knowledge of the fine arts, literature and
classical music gives persons that possess such knowledge. It can also be
the kind of prestige that a certain dialect gives its speaker. Bourdieu
(1991:138) has pointed out that those born in Arrondissement 7 in
Paris will gain linguistic prestige as soon as they speak. In order to make
his point, he concludes that the overwhelming majority of those who
led France at this particular point in time, 1978, spoke this dialect.
Degrees from prestigious schools are also to be counted as cultural
capital. It is important to recognise that, as opposed to symbolic capital,
which may be specific to certain segments of society, cultural capital is
a form of capital that is acknowledged by society as a whole. Symbolic
capital, on the other hand, also incorporates assets that are only acknowledged within certain fields of society (Broady 1991).
It has been suggested that the acknowledgement of cultural capital
emerged as the result of specific historical circumstances that led to the
possibility of more or less permanently stored, symbolic capital in the
form of degrees, rules and regulations, written documents, etc. (Broady
1991:171). According to this view, cultural capital could not have been
important as a dominant form of capital before the introduction of
written language, printing techniques and the development of formal
institutions for education that can provide printed degree certificates.
However, as we have seen earlier in this chapter, Haydens generalisations regarding the more complex transegalitarian societies indicate that
different kinds of cultural capital are recognised within traditional societies as well. Families that can afford to let their children go through
initiation rites or let them participate in time-consuming education,
such as certain apprenticeship systems, can use the value that these
108
Social capital
This form of capital, according to Bourdieu, can be regarded as assets
that take the form of personal relations based on kinship, on ties between old school friends or on personal contacts that have been established during work. Personal relations of this kind can obviously be
used as a form of capital in different forms of social strategy (Broady
1991:169). As regards traditional societies, Bourdieu has noted that
certain Arabic-speaking tribes in Algeria have recourse to a certain capital
initial that is attached to the kinship with certain mythological ancestors, a capital which can be converted into power and prestige (Bourdieu
1958:85, Broady 1991:183). We may therefore assume that certain
kinship ties were used in prehistory as well. As we have seen in Chapter
3, this was actually suggested by Helena Knutsson when she connected
the use of a certain blade type as grave goods with the communication
of mythical, ancestral relations. In a similar way, it is tempting to regard
the Late Neolithic flint dagger as an insignia that referred to a kinship
with mythological ancestors.
It is only the complexity of the social structure of the society in
question that sets the limits to the number of types of capital that can
be defined. Thus, in the modern western world an infinite number of
such types can be defined, while, in traditional societies, the number of
types of capital that are acknowledged as such within the communities
may be considerably fewer.
Habitus
Habitus is a system of arrangements that will enable people to act, think
and orient themselves in the social world and, as such, it can be regarded as a form of embodied symbolic capital (Broady 1991:17). In
the habitus concept, Bourdieu has demonstrated how individual and
collective intentionality merges into the individual actor. The habitus
of a certain individual is formed by the life that he or she has lived and
109
Jan Apel
that will guide his or her imagination and practice and thereby, in most
cases, contribute to the recreation of the social world. However, in
situations in which the habitus in fundamental ways does not coincide
with the structure, it may help to change the social world (Broady
1991:225).
Habitus can under certain circumstances be regarded as form of symbolic capital, although it differs from, for instance, cultural capital, in
that it does not take the form of degrees or titles, etc. Habitus is embodied symbolic capital; it takes the form of a certain dialect, the capacity
to distinguish good taste in literature, theatre, film, etc. I did not put
quotation marks around the word good in the preceding sentence
since I referred to the objective form of good taste in society the
taste of the dominant segments of society and consequently the kind of
taste that is rewarded in social strategies. Someone who speaks a country
dialect and has simple tastes as regards, for instance, literature but is a
good motor mechanic will most probably only be able to use his or her
habitus as capital to promote his or her status and influence in a limited
circle of people. This means that peoples habitus can be related to the
constitution of their inherited or acquired capital (Broady 1991:228).
Any given habitus will facilitate the composition of a certain register of
strategies that will define and limit the freedom to act of the person in
question. It is very difficult to change ones habitus. Accordingly,
Bourdieu has pointed out that different dispositions that are contained
within the habitus concept will remain long after the social situations in
which they were created have ceased to be. A persons habitus is primarily
formed in his or her family (Bourdieu 1999a). A person in a familiar
environment will intuitively know what to do and what to say if the
habitus is in accordance with the surrounding milieu. Such a situation
will make the person feel at ease with the situation and the surroundings will appear to be natural. Durkheim referred to these types of
feelings as springing from a form of social instinct (Durkheim
1993:118).
In traditional societies, the habitus can sometimes take on a very
tangible appearance, for instance the way in which the degree of fatness
corresponds to prestige on certain islands in the Pacific Ocean. Certain
initiation rites that will result in characteristic, physical traits in the
individual who undergoes them are also encountered in the ethnographic
literature. In some communities, girls are subjected to long stays in110
Social field
A social field is a system of relations between positions that are held by
people or institutions that argue and negotiate on something that they
have in common (Broady 1991:9 f.). Usually, what unites the participants in a field is certain common factors of their habitus. The existence
of a field presupposes specialists, institutions and value hierarchies that
are recognised by the participants in the field. A field is also characterised by different agents having different positions. This results in a
continuous power struggle between the dominant and the dominated
fraction of the field. The members of a field will share certain characteristics in the habitus with each other. This means that they will have a
deeper understanding of each other and each others actions than is usually the case among individuals in society in general. Most archaeology
students will, for instance, realise this when they think back to the feelings
of mutual interest that occurred when they started to socialise with
other archaeology students. However, there are also differences in the
dispositions that make up the habitus of the members of a field. These
differences stem from the fact that the members have different social
backgrounds. Some members of the field will have backgrounds that
have given them dispositions that are more in accordance with the
objective structure of the field than other members have. For such
111
Jan Apel
persons, the objective order of the field will simply appear to be in a
natural state. In intellectual fields, such as archaeology, it is not unusual
for persons who have a high degree of their habitus in accordance with
the ideal structure of the field to regard themselves as being smart and
being talented. Members with a lesser part of their habitus in accordance with the structure cannot avoid reflecting on the fact that parts of
the objective structures of the field make them feel uncomfortable.
They will use their influence to change the structure of the field if they
are given the opportunity. It is the dynamics, formed by the objective
structure of the field and the variations of the constitutions of the members habitus, which promote continuous change in modern production
fields.
As regards prehistoric institutions that produce symbolic capital, for
instance, in the form of dagger production, the field concept may not
be an appropriate term to use. In traditional communities, in which
crafts are transmitted vertically through the generations within lineage
groups, clans or castes, the members of the guilds will, owing to their
similar social backgrounds, have certain almost identical traits in their
habitus. Thus, the dynamics that exist between the objective structure
and the variations in the different habitus of the members that are present
in a modern production field will be present to a much smaller extent
in a craft guild in a traditional society. In other words, in traditional
societies, in which almost everybody acknowledges the same type of
symbolic capital there is less chances of internal change since there is less
variation in the habitus of the members of any congregation. Instead,
social structures will tend to be very conservative. Therefore, it may not
be appropriate to use the term production field in discussing prehistoric guilds. In this context, the term production fraternities will be
used to describe the traditional form of Bourdieus term production
field, i.e. it is a group of people that in some way produces symbolic
capital that is distributed unequally between its members. The similar
traits in the habitus of the memebers of a production fraternity will
make it far more conservative in its nature than the modern production
fields. A historical example of a production fraternity in which such a
conservative nature eventually led to a internal collapse as the result of
outside pressure is the production of Swedish iron works in Bergslagen
from medieval times up to the 19th century. In this almost secluded
community, the different crafts were passed down from father to son
112
Apprenticeship systems
The most commonly known type of apprentice organisation is probably the medieval guild. The complexity of the medieval societies, as
regarded production and consumption, meant that artisans became specialised in producing different types of items that were needed for everyday life. Hence, bakers, butchers, tailors, blacksmiths, etc. were in
evidence in every little town (Pirenne 1936:179). In order to protect
both consumers and producers, it became necessary to create craft guilds
that guaranteed quality and controlled prices. It has been suggested that
this new type of organisation was a development of the fraternities that
appeared among associated craftsmen of different trades during the 11th
century (Pirenne 1936:181). Even though these fraternities had a charitable and pious aspect, they must also have functioned in a protectionistic
way (Pirenne 1936:181). It is important to note that these fraternities
were based on free association and that the introduction of the guilds
meant a strict legislation for the crafts. The guilds, as opposed to the
fraternities, had legal authority to force any artisan either to join the
guild or to close down his workshop. I shall not dwell on the subject of
the medieval guilds any further. However, they are good examples of
traditions becoming institutions as a result of an increasing moral
density. So even though the medieval guilds were political creations in
a fashion unlikely to appear in prehistorical contexts, they stemmed
from a tradition with roots far back in prehistory and well into preRoman times.
There are at least three different ways in which apprenticeship systems are known to have been organised in traditional societies: as guilds,
castes or as kin association groups (Clark & Parry 1990:334). The anthropological use of the term guild means considerably less than it
did in the history of the medieval guilds but seems to have more in
common with their predecessors, the craft fraternities. In anthropology
a guild is defined as an association of craftsmen that is not necessarily
based on kinship ties or on a causal association based on residential
groups (Clark & Parry 1990:334). Castes and kin association groups
differ from guilds in that they base recruitment on kinship ties. The
113
Jan Apel
degree of kinship differs for various groups it can be lineage- or clanbased, but caste organisations are distinguished from kin association
organisations by the fact that they are endogamous (Clark & Parry
1990:334). In some cases, when the craft is based on a limited rawmaterial source, the corporate forms of kinship-based, apprenticeship
systems may appear. These may consist of the people who are living
near the raw material and can control it.
Now, how may a prehistoric craft fraternity have been organised? It
has been suggested that the segment or caste, of the society that controlled the flint resources and the dagger technology formed a kind of
corporate group based on close kinship. The reproduction of the
technology took the form of an apprenticeship system and the selection of apprentices was based on an institutionalised, hereditary system.
In order to illustrate how an apprenticeship system may be organised in a traditional context, I have chosen the example of traditional
metal-working in Kenya. This tradition has been studied comprehensively and the following descriptions will focus on questions regarding
the importance of kinship in the reproduction of the technology.
Jan Apel
A smith whose master is still alive will often obtain his blessing
before taking on apprentices from outside of the family. Yet this practice varies from tribe to tribe. Sometimes it is prophesised that a woman
will bear a son destined to become a smith or that a young man with
distant smith ancestry has been chosen to be a smith. These apprentices
are more expensive than ordinary ones, since they have to be purified in
a ceremony in which they are obliged to provide animals, often a goat,
to be sacrificed. Their relatives are obliged to build a smithy, to which
objects from the ancestral smithy are brought, where they can be taught
in privacy. They will also have to pay a higher fee than an apprentice in
the close family does.
Usually, a smiths son starts to appear in the smithy when he stops
sleeping with his mother around the age of seven or eight years. However, serious training does not begin until the voice starts to break. If
the apprenticeship starts at this age, it will usually last eight to ten years.
When an apprentice is brought in from the outside, he is generally much
older and will receive intense training, lasting between three and six
months, provided that he is quick to learn and can pay the necessary
fees. If he is not good with his hands he may have to stay as an apprentice
for one to three years in order to acquire the necessary skills. It is
interesting to note that the apprenticeship system is flexible enough to
teach the skills to people with very different qualifications. In view of
the theoretical considerations presented in this chapter, one would suspect
that a person growing up around the smithy, other things being equal,
would appear to have a larger degree of natural ability. Tutoring fees are
usually demanded from all apprentices. Sons will pay smaller fees than
apprentices from outside the immediate family. The fees are paid in the
form of sheep, cows, hens, baskets of grain and honey beer. An apprentice
from outside the family may start his apprenticeship by giving the smith
a goat or sheep and liquor. He will then pay a hen during the first stage
of tuition, a goat or sheep during each of the second and third stages
and finally one or two animals and liquor during the ceremony of
initiation to the craft.
Stages of apprenticeship
In order to become a smith, the apprentice must go through three stages
of apprenticeship. Each of these stages includes certain taboos that must
not be broken. Usually, the apprentice deliberately overstep a taboo in
116
Jan Apel
and have children. Since smiths are often classed as elders and are expected
to function as counsellors, it is believed that only married men with
children can command the necessary respect.
The details of the initiation rite will vary from tribe to tribe but
usually follow the same general pattern. All the smiths in the neighbourhood are supposed to attend the initiation rite. In some cases smiths
will travel up to thirty miles in order to participate in the ceremony.
Usually, the rite starts with a practical demonstration of smelting and
forging by the master smith. He will then describe the progress of his
pupil. The apprentice will then demonstrate his skills in smelting and
forging, in order to show that he is worthy to join the assembly. The
highpoint of the ceremony is when the apprentice is presented with his
tools and particularly his hammer, which is the symbol of the craft.
The apprentice is then taught to curse and the remaining innermost
secrets of the trade, which he swears never to disclose. He is informed
that he will die if he does and it is understood that he must remain a
smith for the rest of his life. The oldest smith in the congregation blesses
the new smith. The sacrifice is made and must be eaten during the
ceremony. When a man is initiated as a smith the mystical power of
his smith ancestors, derived ultimately from god, is passes on to him to
enable him to smelt and forge iron successfully (Brown 1995:122).
The most powerful curses in the traditional societies of Kenya are
those uttered by a smith. Regardless of their status, which may vary
from tribe to tribe, smiths are always feared and respected. They are
regarded as being different from everybody else in society because of
the mystical powers they have inherited from their ancestors and from
god. Close contact with smiths or their products is believed to cause
ritual pollution as a result of innate and automatic curses.
Jan Apel
recognise the close connection between Durkheims view and the structuralism proposed by Claude Lvi-Strauss.
According to Lvy-Strauss, many cultural phenomena, such as rituals and myths, rules for kinship and descent, and settlement patterns,
are external expressions of the way in which the human mind structures
data. Accordingly, if this unconscious, general structuring of data can
be recognised beneath the obvious structure of a specific institution,
such as iron-working, it can be used to explain other institutions and
customs (Levy-Strauss 1958:28). In the following discussion, I shall
avoid picking out specific aspects of traditional iron-working in order
to use them as direct analogies with the Late Neolithic flint technologies of Scandinavia. Instead, I shall try to pin-point some principles of
technology which can be assumed to be more general in their character.
However, if Durkheim and Lvi-Strauss were correct in their assumption that there is no fundamental difference between primitive and
modern ways of thinking, the obvious implication is that ethnographic
analogy does not have any more explanatory value than a modern,
western analogy. So why not use modern, western examples instead of
traditional metal-working in Africa? As I see it, there are two reasons
for doing this. First of all, the scientific principle that guides all
conclusions based on analogy must be put forward. This principle simply states that the more common features that can be established between the object of analogy and the object of study increase the possibility that there are further similarities between them. Therefore it makes
sense to compare societies that share similarities in, for instance economy
and climate in order to understand the social structure. As we have seen
earlier in this chapter, Hayden has successfully used this strategy. Secondly,
I believe that the outside perspective that the ethnographic examples
provide gives huge advantages just because the archaeologists and
anthropologists are not born in the system and thereby stand a better
chance of understanding the underlying principles of the activity studied.
The detour over an unfamiliar tradition is necessary in order to break
up the deceivingly well-known relationship that we have to our own
tradition (Bourdieu 1999b:13). In confronting a foreign tradition, the
anthropologist cannot take anything for granted and the judgements
will to a lesser extent be based on what Searle refers to as the background
of this particular society. Bourdieu uses analogies to Berber communities to understand aspects of the French societies because he feels that
120
Heredity
As we have seen rigid hereditary principles are followed concerning those
who are allowed to serve apprenticeships, as opposed to those who are
not. Some remarks indicate that these hereditary rules may be transgressed
if the person who is supposed to learn the craft does not show an aptitude
to do so. The fact that apprentices from outside the immediate family
have to pay higher fees also underlines the importance of heredity. In
iron-working, the question of talent is not that important, probably
since almost anyone will be able to master the practical know-how and
gain the theoretical knowledge involved in the technology. I hereby
suggest that a technology that includes a great deal of practical knowhow will be much more dependent on the motor abilities of the
apprentices. In such situations, the apprenticeship will last longer, since
motor skills are established in the body through repetition As I see, it
we shall have to accept that any natural ability, in the form of biological
heritage, will be distributed evenly between the members of any social
group. It has been suggested that natural abilities may have been decisive,
as regarded who would be considered as a flint-knapper and who would
not (Olausson 1998). This reasoning is appealing, since it is impossible
to avoid seeing that certain individuals find flint knapping easier than
others, especially while teaching students the basics of flint knapping.
However, I would like to add that, in most traditional crafts, hereditary
principles determine who is to follow the family tradition and who is
not. It may be difficult to judge whether, what we consider to be natural
ability, is dependent more on biological than on the social aspects and
vice versa. In most traditional cases, the combination of biological
heritage and social encouragement within family units is decisive, as
regards becoming talented artisans. This conclusion is actually in line
with ethnographic investigations, such as Barbro Santillo Frizells (2000)
study of the craftsmen that build the traditional, Apulian, dome-shaped
houses in southern Italy. This architectural tradition, based mainly on
the silent science of the hands (i.e. on know-how, as it was defined in
121
Jan Apel
Chapter 2), is transmitted vertically through the generations according
to hereditary principles. However, in cases in which the eldest son does
not show the appropriate interest or aptitude, another son will be chosen
to become an apprentice (Santillo Frizell 2000:75 f.). In a broader
context, it is tempting to regard prehistoric apprenticeship systems,
organised according to hereditary principles, as emerging because of a
strong need to keep natural ability within families, who in turn could
use this symbolic capital in order to gain and maintain social power and
prestige. Apprenticeship systems were thus created in order to guarantee
that as many as possible of those who were supposed to learn the craft
were able to do so, regardless of their intellectual and motor abilities. It
is important to remember that, in referring to natural ability one should
pay heed to the present social situation.
Jan Apel
reader that there is no fundamental difference between traditional Ironworking in Africa and modern technologies today.
124
125
Chapter 5
Previous Research on
the Determination of Debitage
from Bifacial-reduction Sequences
In Chapters 6 and 7, Late Neolithic sites in southern Scandinavia and
eastern central Sweden with flint will be studied and discussed in relation to my main concern, i.e. to see whether it is possible to recognise
different production stages between and within sites. Such findings
would shed some light on the ways in which daggers and other artefacts
in southern Scandinavian flint were distributed, these finding could
also be used to understand the character of the Late Neolithic political
landscape and the organisation and complexity of the social structure.
The Late Neolithic, flint technologies of southern Scandinavia are mainly
based on bifacial-reduction strategies. The presence of bifacial-reduction debitage on sites in southern Scandinavia habitually used to date
them to the Late Neolithic period. Hence, it is appropriate, as a starting-point, in this context, to describe the formal ways in which production debitage from bifacial-reduction sequences can be distinguished
from other flake types in archaeological contexts. These kinds of flakes
are actually not difficult to recognise; any experienced archaeologist or
lithic analyst will immediately recognise flakes from bifacial-reduction
sequences when such flakes are recovered in the field. In spite of this,
the formal ways in which these flakes can be distinguished from other
types of flakes, such as flakes from unifacial platform cores or the production of square-shaped axes needs to be evaluated. In my specific
discussion of the bifacial technology, in relation to the idea of craft
specialisation that will follow later on, there is a need to be able to
distinguish flakes from different stages in the production of flint daggers. If this can be done, it will be possible to understand and designate
the production debitage from different production stages between and
within sites. Consequently, I am not interested solely in identifying and
separating bifacial flakes from other types of flakes but also in being
127
Jan Apel
able to classify them to different stages of the production. Thus, in this
chapter, in preparation for the analysis of original production debitage
in Chapters 6 and 7, earlier attempts to classify debitage from bifacialproduction sequences are referred to and discussed.
The following paragraph summarises some elementary concepts in
the art of flint-knapping that the reader may find useful to know in
reading the rest of this chapter. During flint-knapping, flakes are struck
from larger flint pieces (cores) with the help of hammerstones or antler,
wooden or metal billets. This technique is in this context referred to as
percussion technique. In some cases, a punch, in the form of an antler
or copper billet, is placed on the platform of the original core and is
positioned between the hammerstone and the core. This technique is
referred to as indirect percussion. If the percussion hammer is of a harder
material than flint, for instance, a harder variant of quartzite or metals
such as copper or steel, the technique is referred to as hard percussion
technique. If the percussion hammer is made of a material that is softer
than flint, such as sand and limestone, wood or antler, the technique is
referred to as soft percussion technique. Flakes can also be removed by
pressing a bone, antler or copper-tipped, pressure flaker on the edge of
the platform. This technique is referred to as pressure technique (Pelegrin
1991, Madsen 1996, K. Knutsson 1988a).
Lithic-production stages
In 1972, Don E. Crabtree defined the term method in flint-knapping
as ... the logic manner of systematic and orderly flaking process, or the
preconceived plan of chipping action based on rules, mechanics, order
and procedure (Crabtree 1972:2). Among flint workers and lithic analysts, method is the phenomenon that verifies the idea that the production of knapped artefacts is not haphazard but carefully planned and
answers to specific needs formulated before the actual flint-knapping
begins. In other words, The shape, length, width, thickness, form and
technique of applied force to fashion the tool was predetermined by
the toolmaker before the initial fracture of the raw material (Crabtree
1972:2). The preconception of the goal (a flake, core or tool with characteristics) that the flint worker has in mind before the actual knapping
begins can be referred to as a form of mental template.
Technique is the way in which the force is applied to the core during
the removal of each flake. It may be direct or indirect, done with a hard
128
Jan Apel
change has to be great enough to affect the appearance of the debitage
in a recognisable way. That is, a number of variables that define and
isolate flake types from different production stages have to be present
in the debitage. Some authors argue that a natural stage occurs when a
stage defined in this way coincides with the completion of the mental
template that is needed to secure a continued reduction (Callahan 1979).
Other authors prefer to work with definitions that are independent of
any technological changes and instead are based on an artificial division
of the reduction sequence, i.e. experimental materials are divided into
four, equally large sections. The former way of working is demonstrated
in the following examples.
Fig. 5:1. Outline drawing of marginal (on the edge) and non-marginal (behind
the edge) flaking and typical flakes.
Jan Apel
analytical tool developed by modern flint workers, in order to formalise
the learning process, and by lithic analysts, in order to tidy up the debris
on prehistoric knapping floors for the purpose of studying cognition or
technological specialisation (K. Knutsson, in prep., 15 ff.). In this
context, the discussion as to whether or not every flint worker is aware
of production stages is not relevant. Stages, like typologies, are often
partly subjectively constructed in order to solve specific questions.
In my experimental study of the type-IV daggers, it has been of vital
importance to evaluate how lithic-production stages relate to the production debitage. If this relationship can be established, it will open up
possibilities of studying potential production sites and discussing the
degree of technological specialisation and organisation of dagger production in relation to the organisation of society in general: ... the
flintknapping techniques are culturally learned behaviors. They are highly
variable through time and space, since specific lithic reduction processes
are affected by different variables in response to cultural needs (Yerkes
& Kardulias 1993:93).
Earlier attempts
The idea that flint tools were produced in natural stages is old and has
been used in archaeological interpretations since the late 19th century.
The American archaeologist W. H. Holmes conducted flint-working
experiments in order to find ways of distinguishing between implements (finished tools) and preforms in the Virginia Tidewater quarries
(Johnson 1978:340). Holmes used five criteria for the identification of
implements: (1) degree of elaboration, (2) indications of specialisation,
(3) signs of use, (4) manner of occurrence and (5) association of other
articles (Holmes 1894:123). These criteria had to be taken into consideration before an artefact could be regarded as an implement rather
than as some kind of preform.
Holmes suggested that, in order to evaluate prehistoric stone tools,
especially those found in quarries, their history had to be studied through
the process employed in their manufacture (Johnson 1978:341). In
fact, as early as 1894, he had argued for an approach similar to the
operational-chain studies of today (Holmes 1894:122, Sellet 1991).
He was able to demonstrate that there were significant similarities between crude implements and early preforms for more elaborate tools
and presented different production stages from blank to finished pro132
Jan Apel
order to convert a complicated mental template into flint.
In 1975, Bradley (1975:7) argued that carefully controlled, experimental analogies were probably the most useful way of gaining understanding of lithic-reduction sequences for different, prehistoric-tool types.
Such an experimental procedure would accordingly involve a careful
examination of the production debitage, the techniques and methods
used, and the sequence of reduction. Bradley suggested that, once the
reduction sequence had been reconstructed, it could be divided into
stages and steps of manufacture (ibid.). These stages were (1) preliminary
modification, (2) blank and (3) preform. He maintained that the most
difficult task for the analyst was to demonstrate intention, and suggested
that the modified, non-implement, non-core category was to be used
in this process, i.e. the production debitage.
Within this category there may be typological groups identifiable by
clustering of morphological and technological attributes. Once these
groups are defined, the nonmaterial evidence (recovery associations)
may be examined to determine whether or not there is a high correlation between the two. In an archaeological context this is most likely to
be encountered in the form of caches or closely associated (spatially)
groupings. If there is a statistically valid correlation, it may be inferred
that these groups represent intended stages. These stages may then be
classified as preliminary modification, blank, or preform (Bradley
1975:7).
Bradleys suggestion, that it was possible to demonstrate the intention of the prehistoric flint worker if the right variables were documented in the production debitage, was followed by several archaeologists during the following years. Many of these analyses concerned
worked, bifacial-reduction sequences, and I shall recapitulate some of
these in the following pages.
Recent attempts
Skilled flint workers and lithic analysts can recognise and differentiate
between bifacial-thinning flakes and flakes produced during unifacial
reductions. It is also possible to argue that certain flakes stem from
early or late production stages. This way of classifying debitage stems
from experience and is thereby intuitive and not formalised. It can be
considered as emanating from an emic approach in which reason is based
134
135
Jan Apel
Alternate Flake
Interior Flakes
Late
Pressure Flakes
Newcomer 1971
In 1971, the results of quantitative experiments in Palaeolithic, handaxe
manufacture were presented by Newcomer in World Archaeology. One
of the goals of these experiments was to discuss the morphology of the
waste flakes produced during handaxe manufacture in relation to stages.
Initially, Newcomer subjectively defined three different production
stages:
1 A rough-out stage in which the nodule is bifacially flaked in order to
obtain an edge all round. This is achieved by using the flake scar
136
Alternate flake
Edge-preparation flake
Margin-removal flake
Early-stage, pressure flake
Definition
Produced during the initial removal of cortex. These flakes have
cortex over their entire dorsal surfaces.
Produced during final stages of cortex removal. Cortex is present on
part of their dorsal surfaces.
Flakes removed from the interior of the core or cobble. There is no
cortex on their surface.
Produced by percussion during the primary reduction of a biface (as
a result of making the biface symmetrical). They have few dorsalsurface scars, are slightly curved or twisted in longitudinal section,
have single-faceted or multi-faceted platforms and are usually the
largest thinning flakes produced during biface manufacture.
Produced by percussion during the final stages of roughing out a
biface. They have numerous scars on their dorsal surfaces are almost
flat in longitudinal section, usually exhibit a feathered termination
and have multifaceted platforms.
Produced as a result of the creation of a bifacial edge from a square
edge on a piece of stone. They are much wider than they are long and
are triangular in cross-section.
Removed from the edge of a flake blank in order to prepare the blank
for further reduction. The dorsal surface of the flake blank serves as
the platform for these flakes.
Half-moon-shaped fragments of bifacial edge produced when the
biface is struck too hard and too far from the margin.
The first series of pressure flakes removed when bifaces are being
finished. These small flakes have multiple scars on their dorsal
surfaces and are twisted in longitudinal section, and their platforms
form an oblique angle with the long axis of the flake.
Produced during the final episode of pressure flaking on a tool. They
are small, parallel-sided and slightly twisted in longitudinal section
and have multifaceted, abraded platforms and a single dorsal arris.
Table 5:1.Verbal description of some technological-debitage categories concerning bifacial reduction developed by Jeffrey Flenniken at the Flintknapping
Field School (after Yerkes & Kardulias 1993:97, Table 1).
Jan Apel
Flakes from the second production stage were regarded as characteristic, and Newcomer also pointed out that flakes like these are known
in numerous examples from sites belonging to Acheulian and Mousterian
traditions. They were described as typically thin with all edges feathered. Any undulations on the ventral surfaces are poorly marked, and
they also display negative flake scars of other, similar flakes on their
dorsal surfaces. Newcomer also suggests that flakes from this stage, if
they were removed with deer antlers, would display a lip on the ventral
side of the platform remnant (Newcomer 1971:88 f.).
Flakes from the third production stage closely resemble flakes from
the second stage, but are smaller. Newcomer points out that the flakes
from the second and third stages have often been lumped together as
hand-axe finishing flakes or eclats de taille in earlier literature. This
type of flake has been referred to as thinning flakes (Crabtree 1972) or
bifacial-thinning flakes (Flenniken 1985). It seems as if Newcomer was
of the opinion that the flakes from the second and the third stages
could not be distinguished in any other way than by size. This implies
that the change in manner is not significant enough to show in the
qualitative variables that he used. It should be emphasised that these
variables are not accounted for in the article. In other words, there is no
way of judging their credibility.
During the experiments, three handaxes were made and each flake
removal was weighed. This data was processed and the results revealed
that flakes gradually became smaller as the reduction sequence continued. Newcomer also concludes: As indicated in the figures, these curves
seem to be divided into three distinct areas, each roughly corresponding
to one of the stages of manufacture subjectively defined above
(Newcomer 1971:93).
The results of the experiments suggested that the three different stages
inherent in Newcomers production of handaxes could be defined in
the debitage in two ways: by qualitative classifications of flake attributes,
and by quantifying the weight of each principal flake. In this context, a
principal flake is a flake that is the result of planned action, i.e. cases in
which the knapper knows beforehand where to strike and what will be
achieved by this stroke. Edge-preparation flakes can seldom be considered
as principal flakes, since they are not predicted in the same exact way as
principal flakes are. However, it should be stressed once again that the
flakes from stages 2 and 3 could not be distinguished by any other
138
Jan Apel
The tests were apparently guided by earlier empirical observations made
by the authors. They knew that flake sizes were repeated within stages,
not stage to stage, and we proposed that it would be possible to
predictively assign individual flakes to stages by their sizes, and to
characterise stages by the redundancy of these flakes (Nichols &
Callahan 1980:13).
The tests suggested that there were statistically significant differences
between the mean values of the length and weight variable from the
different stages. A decreasing variability per reduction stage was also
discovered, i.e. the flakes tended to be more and more regular in size
(and shape) per reduction stage. It was also concluded that there was a
non-repetition of flake size within the experiments, i.e. the flakes became
smaller.
141
Jan Apel
Flake form: (a) platform flakes, (b) no-platform flakes and (c) others.
Flake width
Flake thickness
Cortex flakes: (a) primary cortex flakes, (b) secondary cortex flakes
and (c) no-cortex flakes.
5 Platform length
6 Platform width
7 Lip: presence or absence of lip on platform remnant.
Holm concludes that a clear pattern is discernible in the material
from Grsvattnet VI (Holm 1985:307 ff.). The width-thickness relationship between flakes from the upper part of the stratigraphic sequence (the A-horizon), on the one hand, and the lower part of the
stratigraphic sequence (the cultural layer), on the other, differs consid-
142
Jan Apel
division are not presented in the article; however, it is obvious that
these stages do not reflect the intention of the knapper. It is an arbitrary
division and consequently not related to intention. This can arguably
be considered as a major weakness in the study, since changes in the
production debitage as a reflection of a change in manner cannot be
studied. On the other hand, owing to the fact that the subjective element of the experienced knapper or analyst has largely been omitted,
this kind of study will perhaps be the most suitable method of approaching an archaeological material. Of course, this method will have
to be accompanied by a limited technological evaluation of the material that will indeed be subjective and that will stem from folk knowledge. In order to work, this evaluation will have to include a wellargued distinction between flakes removed in bifacial-reduction sequences
and flakes removed in other kinds of reduction sequences.
In the study, flakes with cortex-covered, dorsal surfaces were divided
into two sub-categories: primary decortication flakes (51 - 100% dorsal
cortex cover) and secondary decortication flakes (0 - 50% dorsal cortex
cover). The results of the initial statistical treatment of the cortical flakes
revealed that primary decortication flakes were significantly more
frequent in the first stage than expected and that they were significantly
less frequent in the last production stage. In the middle stages, no
significant differences were discovered. It was concluded that the cortex
appeared to have been removed rapidly and that the variable was
diagnostic only in the earliest stage of reduction (Amick et al. 1988:28).
Only a small number of flakes possessed more than two dorsal-flake
scars. The dorsal-directionality variable displayed a similar distribution.
This suggested that the importance of these variables in an analysis aiming
at recognising bifacial-production stages might be limited. It should be
noted that the experimental reduction sequence was being taken through
only to a point in the reduction that would be somewhat equal to
secondary thinning. Because these variables were closely correlated, it
was decided that dorsal directionality should be dropped at this point
of the investigation. In the following univariate analysis, it was concluded that flakes with higher than expected, dorsal-scar counts tended
to occur late in the reduction sequence, while flakes with lower than
expected values tended to occur early in the reduction. It was also
concluded that the intermediate- and late-stage debris exhibited similar
distributions, while only the first stage was distinct.
144
Description
Event
Sequence
Hammer
Percussor type: 1=large hard, 2=small hard, 3=large soft, 4=small soft
Cortex
Scars
Direction
The number of directions, recorded on an 8-coordinate system, which have scars present on the
dorsal surface*
Weight
Length
Mid-width
Max-width
Distance
Distance from the platform to the maximum width recorded for complete flakes only.
Measurement to the nearest 0.1 cm
P-width
Platform width. Recorded to the nearest 0.1 cm on all platform-remnant-bearing flakes (PRBs)
P-thick
The authors stressed that the different hammer types had a significant effect on the morphology of the flakes. This is demonstrated by
the different morphological traits displayed by the flakes produced with
different hammers (Amick et al. 1988:29). Here, the authors face
another crucial problem. The only way to evaluate whether a certain
hammer type is responsible for certain morphological traits in the production debitage is by conducting experiments in which all possible
variables, except the type of tool used, are held reasonably constant.
This is not done in the article and it means that all the differences that
Amick et al. present as due to different hammers may be regarded as the
result of other phenomena. In fact, it is likely that these differences may
be the result of significant changes in the knappers manner of organising the flaking. Earlier studies have indeed demonstrated that the
significance of the hammer types related to traits in the production
debitage has been slightly exaggerated (Patterson & Sollberger 1978:107
f.). However, it is concluded that the large hammer flakes are the largest
in all dimensions, except distance and length, where soft hammer flakes
are largest. This demonstrates, according to the authors, that large
hammer flakes are thicker relative to their length than soft hammer
flakes (Amick et al. 1988:29). It is hereby suggested that these differences may also be the result of on-the-edge or behind-the-edge flaking
strategies.
145
Jan Apel
Even though weight seems to be correlated with almost all the other
variables, it is suggested that the variable has considerable value in
debitage studies (Amick et al. 1988:29). It is further concluded that
flake weight reflects flake thickness, that distance is a function of length,
and that maximum width and medium width are closely correlated
variables.
After conducting a multivariate analysis, based on the variables described above, it was concluded that the reduction stages could be reliably predicted after edge-trimming and platform-preparation flakes had
been identified and omitted from the investigation. These kinds of flakes
are manufactured throughout the whole reduction sequence and change
little in morphology and, because of this, they tend to confuse attempts
to define relative reduction sequences within the debitage (Amick et al.
1988:32). It was further concluded that the counting of dorsal scars
and cortex-cover percentages could be used to define the early stages of
the production. However, it proved difficult to define the intermediate
stages on the basis of the chosen variables. These stages are very important in the production of a biface, since a large proportion of the flakes
from a production will be debris produced during bifacial trimming
and thinning. The authors suggest that alternative variables may improve the chances of recognising the intermediate stages (Amick et al.
1988:34).
Odell 1989
In 1989, Odell published a paper in which he sets out to investigate
three questions regarding un-retouched debris:
1 Is it possible to accurately distinguish flake-core and bifacial technologies?
2 Is it possible to accurately distinguish the various stages of a bifacial
technology?
3 Is it possible to discern chronological differences in the debitage?
To answer these questions, Odell conducted a set of controlled experiments, together with a competent flint worker, Greg Thomas. We
had determined that the metric and discrete attributes of reduction flakes
appeared to contain sufficient internal variation and intrasite homogeneity
to warrant trying to discriminate among technologies and stages by
recording and analyzing attributes (Odell 1989:165). Odell stressed
146
Description
1. Weight
2. Striking-platform width
3. Striking-platform thickness
4. Maximum length
5. Maximum Width
Description/comment
1. Bulbar scar
2. Distal termination
3. Flake twist
4. Lipping
Table 5:4. The nominal and ordinal variables used in Odell (1989).
147
Jan Apel
the importance of an objective way of recording the variables. This was
done to ensure that any variable used had to be replicable by another
person to 95% reliability. In the study each production stage was defined
as a sequence in which the knapper mentioned that he was progressing
to a different phase of operation (Odell 1989:164). Thus, a form of
natural staging was used in the experiments, but since an explicit,
technological definition of each stage is missing, this should be
considered to be a methodological weakness in the study.
The experiments included a flake core knapped in one stage, a large
biface knapped in one stage, the first two stages of both a Snyders and
a Hardin, bifacial-projectile point, stages 3 and 4 of the Snyders sequence, and stages 3 to 6 of the Hardin sequence (Odell 1989:170).
The variables that were recorded in the production debitage from these
stages are presented in Tables 5:3 and 5:4.
The data obtained from the variables were subjected to several series
of bi- and multivariate statistical tests. It was regarded as important to
analyse the nominal, ordinal and interval variables in separate groups, in
order to be able to use the methods most appropriate to each group
(Odell 1989:170). This procedure must be regarded as crucial, especially since multivariate statistics are often used uncritically in archaeology.
In order to answer the first question whether there was a possibility of distinguishing between cores of biface debitage the five nominal variables were evaluated in a series of chi-square tests, in which the
debitage from both core-reduction and different, biface-reduction stages
participated. The result of this test was that flake-core debitage was not
easily distinguished by with this data but that it displayed significantly
more bulbar scars than the other stages. Odell suggests that the relative
lack of platform preparation in the reduction of the flake core is responsible for this difference. It was further concluded that one of the
ordinal variables was better suited to discriminate between debitage from
flake-core and biface-reduction sequences than any nominal variable.
This was scarring on the dorsal perimeter of the striking platform
(Odell 1989:172). Odell was able to demonstrate that the flake-corereduction debitage had substantially less platform-preparation damage
than the flakes from the bifacial stages. As Odell himself points out,
this is not a surprising conclusion, but he concludes that ... it is
comforting that, at least in this data set, dorsal perimeter scarring can be
148
Jan Apel
H gberg 1998
In order to create a reference collection of flint flakes from the production of different kinds of Neolithic flint artefacts, a flint-knapping session in which several axes and one flint dagger were made was organised
by the Malm Museum in Malm, Sweden, in the summer of 1997.
An experienced flint-knapper, Thorbjrn Petersen, made the artefacts,
and the debitage from the experiments was analysed by Anders Hgberg,
of the Malm Museum (Hgberg 1998 and 1999). One of the objects
of this investigation was to see whether axe and dagger production could
be distinguished from each other in the production debitage (Hgberg
1998:1). A large number (26) of conceivable attributes were documented
on the flakes. According to Hgberg, none of the variables was discriminatory by itself. Although a certain attribute may be considerably
over-represented within the production of one artefact category, it will
150
Concluding discussion
In Chapters 6 and 7, flakes have been defined, besides my own experiments and observations, in accordance with the findings of this chapter.
As I see it, those who have worked with emic categories based on
generalisation from experiments (for instance, Flenniken and Hgberg)
have constructed flake categories that are easy to use but may be hard to
151
Jan Apel
distinguish for those who lack experience of materials like these.
However, in the cases in which processual archaeologists have been able
to define flake categories in an independent way, these definitions tend
to confirm the obvious. Thus, my own classification system is based
mainly on experience, although I have tried to formalise it whenever it
has been possible to do so. The following flake categories will be used
in order to classify the flint debitage on sites in southern Scandinavia
and eastern central Sweden in Chapters 6 and 7.
(Sub-category 1.) On-the-edge flakes removed by a soft technique. These
flakes are often thin and display a carefully performed, platform
preparation that is necessary in order to prevent the platform from
breaking during impact. Bifacial-thinning flakes, which, in this context, are the most common flake type within this category on Late
Neolithic sites in Scandinavia, often have a platform angle of c. 45
degrees.
(Sub-category 2.) Behind-the-edge flakes removed by a soft technique.
These flakes tend to be thicker than flakes of the former category.
The striking platforms are wide and deep have not been prepared.
This means, as has been suggested by several of the authors presented earlier in this chapter, that the relative thickness may be a
variable that can be used to separate on-the-edge flakes from behind-the-edge flakes. It has also been pointed out that the relative
thickness of different flake categories may reflect weight differences
as well. Consequently, in Chapter 6, the flakes from the Myrhj
site, which have been classified as Sub-category 1 and 2, have been
weighed in order to see whether weight is an independent variable
that can be used to distinguish these two flake types or not.
(Sub-category 3.) Bipolar flakes. This flake type has not been discussed
earlier in this chapter and it is very rare in the settlements that will be
discussed in the following chapters.
(Sub-category 4.) Behind-the-edge flakes removed by a hard technique.
(Sub-category 5.) Pressure flakes. I have used Flennikens definition, although these flakes are not divided further into primary and secondary pressure flakes.
(Sub-category 6.) Flakes that have been difficult to classify. In the Myrhj
investigation, presented in Chapter 6, this sub-category contains larger,
152
Fig. 5:4. Typical flake from squareshaped axe productio (upper flake) and
dagger production (lower flake) (from
Hgberg 1998:16 and 1999:85 f.).
153
Jan Apel
fragmented flakes without preserved striking platforms. In the investigations carried out in Chapter 7, the sub-category will contain
splinters and other fragments of flakes.
Our knowledge of the ability to infer production stages in the debitage
from bifacial-reduction sequences has increased dramatically during the
last 20 years. The summarised studies that have demonstrated that variables such as weight and length are useful analytical units. As we have
seen, the following variable has been successfully used in order to separate flakes from unifacial- or quadrifacial-reduction sequences from those
produced during bifacial-reduction sequences.
1. Flake curvature: bifacial flakes have curved profiles while core and
axe flakes tend to have straight profiles.
2. Bulb profile: Axe and core flakes tend to have more pronounced
bulbs than bifacial flakes.
3. Distal termination: bifacial flakes tend to have feathered distal terminations while the higher frequency of axe and core flakes have
abrupt distal terminations.
4. Flake thickness: Core and axe flakes tend to be thicker than bifacial
flakes.
5. Flake weight: The relative weight of bifacial flakes tend to be lower
than the weight of core and axe flakes. Flake thickness and flake
weight is most probably dependent variables.
6. Striking platforms width: Axe and core flakes tend to have wider
striking platforms than bifacial flakes.
7. Platform preparation (dorsal perimeter scarring): Bifacial flakes generally display signs of extensive platform preparation while core and
axe flakes generally are in devoid of any platform preparation.
8. Type of impact point: If the point of impact has been positioned
behind the edge or on the edge. Bifacial flakes are on-the-edge flakes
while core and axe flakes are behind-the edge-flakes.
9. Dorsal flaking angle: Bifacial flakes will have dorsal flaking angle of
approximately 45 degrees while axe flakes will tend to have a dorsal
flaking angle of 90 degrees. Most platform core flakes will have a
dorsal flaking angle that fall somewhere in between.
154
155
Chapter 6
Jan Apel
3
2
4
8 7
11
9
10
13
1514
16
12
Fig. 6:1. The geographical distribution of the sites discussed in this chapter.
generally lie in the east-west direction and have two aisles divided by a
row of roof-bearing posts. The eastern part of the houses often has a
sunken floor that reveals itself as a darker area during the excavations. In
connection with these depressions, post-holes from roof- and wallbearing posts are regularly found. The depressions vary in depth between
0.25 and 0.6 m and in size between 9 and 14 m in length and between
5 and 7 m in width (Simonsen 1985:86).
This architectural tradition has its roots in the Single-grave Culture
in Jutland, where several houses of this type have been recovered and
dated to MN B (see, for instance, Hvass 1977, 1986). The excavation
of a house belonging to the Swedish-Norwegian Battle-Axe Culture in
eastern central Sweden indicated that this tradition was widely spread
even as early as the later part of the Middle Neolithic period (Hallgren
2000:19 ff.). The tradition was in use in virtually all Scandinavian areas
where agriculture was practised through the Late Neolithic and well
158
Gug (1)
The first Late Neolithic house to be recovered in Jutland was the traces
of a building at Gug, south of the town of Aalborg, in the province of
Himmerland, near the Limfjord (Fig. 6:1). It was found by two amateur
archaeologists in the early 1950s and has never been properly published
(Simonsen 1982:86). A large number of flint implements were collected
during the investigation, including preforms, nearly finished and
complete flint daggers (Brndsted 1957:311). The house was found
when the remains of a sunken floor measuring 4 m x 2 m and filled
with flint waste, were discovered. The fact that large amounts of flint
were found led Brndsted to interpret the house as a flint-workers hut
(Brndsted 1957:312). Among the flint debitage covering the floor,
the following artefacts were found: 10 dagger and spearhead preforms,
25 sickle preforms, four arrowheads, one flake scraper, one drill, one
blade scraper and pottery sherds (Brndsted 1957:312).
In 1982, Simonsen pointed out that the flint debitage filling up the
depression probably represented a secondary filling of the pit (Simonsen
1982:88). From the ethnographical point of view this interpretation
makes sense, since it has been demonstrated that natural depressions,
abandoned structures and pits often became disposal locations in all
types of settlements (Schiffer 1987:61 f.). As regards contexts like these,
which are likely to have been used over a long period, it is almost
impossible to distinguish between primary and secondary depositions
(OConnell 1987). Different methods are needed to allow for alternative
interpretations, and it is likely that a commonsense inference based on
empirical observations alone will be wrong. It is important to use
formation theory based on first-hand observations formalised, in order
to allow for general application (as an example dealing with square-axe
production, see Hansen & Madsen 1983; as for daggers, see Callahan
& Apel, ms). Until this crucial step is taken, different views on how to
interpret materials like these can only be vague hypotheses. In the case
of the house at Gug, the less specific interpretation presented by
Simonsen seems more likely than Brndsteds more specific
interpretation.
159
Jan Apel
Fig. 6:2. Excavation plan of the Myrhj site (from Jensen 1972).
160
Myrh j (2)
The largest known settlement dating from the early Late Neolithic period
northern Jutland is Myrhj. The site is located only a couple of
kilometres from the shore of the Limfjord (Fig. 6:1) and was discovered by archaeologists from Aalborg Historical Museum during the
investigation of a row of stones dated to the Late Bronze Age. The site
was eminently well preserved, with sealed layers without archaeological
materials from earlier or later periods (Jensen 1972:74). Features
forming three houses (Fig. 6:2) were recovered, together with numerous
artefacts, many of which were typical of the early part of LN I. A
proportion of the pottery consisted of Bell-Beaker types. This is regarded
as a strong indication that western-European influences affected northern
Jutland during the late Single-grave Culture and the early Late Neolithic
period (Jensen 1973, Lomborg 1977, Vandkilde 1996). The site was
C14-dated to 3860 BP, which, according to the calibration curve
presented by Pearson and Stuiver (1993), gives an interval of 24642142 cal BP (1 sigma).
The Myrhj settlement and the LN-I period in Scandinavia are
contemporary with the Dutch Veluwe-Beaker phase. Several sites dated
to this period have produced Beaker pottery, especially in the Limfjord
area (Vandkilde 1996:281, Fig. 289). It deserves to be mentioned that
only a small percentage of the pottery on these sites is of typical Bell161
Jan Apel
Beaker forms. Thus, ordinary types of pottery, often plain and coarse,
predominate. Accordingly the occurrence of fine Beakers in settlements
as well as burials suggests that Beaker pottery served as a kind of table
ware, which was used for feasting and drinking rather than ordinary
household work (Vandkilde 1998). On the continent as well as in the
British Isles, the Beakers of this phase are regularly associated with
triangular, bifacially-shaped arrowheads in flint and stone wristguards
with two or four holes (Harrison 1982:26). A fragment of a stone
wristguard was indeed found at the Myrhj site and all three houses
contained triangular, bifacial arrowheads in flint. In addition, ten
arrowhead preforms were also recovered, indicating that arrowheads
were actually produced at the site (Jensen 1973:74 ff.). Type-I daggers
are also frequently associated with Beaker pottery (Vandkilde 1998).
Thus, in the early phase of LN I, there is a close connection between
northern Jutland and the late, western, Bell-Beaker horizons, as they are
manifested in the Rhine delta and in Wessex. As we shall see in Chapter
9, a considerable number of Scandinavian flint daggers have been found
in the Netherlands (Bloemers 1969), but only a few examples have
been recovered in association with bell Beakers. Harrison (1982:26),
for instance, mentions one context only in which a Scandinavian flint
dagger was recovered, together with a bell Beaker: a barbed-wire Beaker
from the Dutch Early Bronze Age, which is contemporary with LN II
in Scandinavia. Unfortunately, Harrison does not present any
information on what dagger type it was. The Veluwe phase represents
the first archaeological indications of metalworking in the Rhine delta
(Harrison 1982:27). The emergence of local coppersmiths may explain
why imported Scandinavian daggers were not considered to be grave
goods and were thus not found in this period. In earlier Beaker phases
in western Europe, for example, in the AOO (All-Over-Ornamented)
Beaker phase, flint daggers of imported, honey-coloured GrandPressigny flint were frequently used as grave goods (Harrison 1982:21).
Seed impressions in the pottery and the presence of harvesting knives
made of blades struck from conically-shaped cores suggests that crop
agriculture was practised by the inhabitants of Myrhj. Animal teeth
from domestic cattle were also recovered. Concentrations of mussel
shells were present in all three houses, indicating that mussels were
collected on the shores of the Limfjord 15 km to the west. It is likely
that this work was done at locations such as Rnbjerg Strandvolde, a
162
Jan Apel
10 cm
Fig. 6:3. Fragments with negative-flake scars displaying a parallel-flaking
strategy from Myrhj (drawings by Lotta Sundberg).
and danian flint might not be correct from the geological point of view
(see Thomsen 2000). However, from the flint knappers point of view,
these terms are relavant.
Most of the flint nodules recovered at Myrhj were obviously
collected as beach pebbles and not mined. This conclusion is further
substantiated by the fact that more than one-third of the flakes display
cortex on their dorsal sides (Jensen 1973:80). In this context, it is
suggested that these pebbles and flakes represent a local and domestic
production of everyday items that is probably present in all the Late
Neolithic hamlets in the area. It is likely that this domestic production
was aimed at producing scrapers, since a large quantity of scrapers was
recovered in each of the houses. Of these, 80% display cortex,
something that indicates that they were made from beach pebbles and
that a large part of the flakes from the site should be related to the
production of scrapers. The debitage stemming from the production
164
Jan Apel
aim of this thesis, it is suitable to start this investigation here.
The flakes recovered at Myrhj belong to four main discrete contexts; flakes were recovered in the depressions of all three houses found
on the site (designated D, EAB and GAB), as well as in an area without
any structures located immediately to the east of house D, designated
VfD. The basis of the following analysis is a database consisting of
1 896 documented flakes. These flakes make up 13% of the total
number of flakes from the site. However, flakes were sampled from
only three of the four discrete contexts. Since house D was the first Late
Neolithic structure to be found during the excavation, it was not as
extensively documented as the other houses. It was also stated in the
report that only 26.8% of the flakes from house D were so-called wide
flakes, presumably stemming from the production of core artefacts,
such as axes and daggers, while a higher proportion of the flakes from
the other houses were flakes of this kind (Jensen 1973:80). For these
two reasons, in combination with the fact that I had a limited amount
of time at my disposal, the investigation was restricted to flakes from
EAB, GAB and VfD. In Table 6:1, the actual relation between the total
number of flakes from the contexts and the sampled numbers is
presented.
The ratio between the black- and grey-flint variants in the total flake
material is about 2:1. In each of the three contexts that concern us here,
the relationship between these two flint types is expressed in Table 6:2.
It is apparent that, at least as regards this aspect, the variability of the
original contexts has been covered in EAB and GAB where the
relationship between the sample and the total amount coincide perfectly.
In VfD, however, there is not such a good correlation. In order to
understand this discrepancy, a discussion of the sampling strategy, in
relation to the anticipated characters of the different contexts, is needed.
In the museum collection, the flakes from each square metre are contained
in brown-paper bags. During the classification of the flakes, I simply
worked my way through the bags, from each investigated context, in a
random fashion. This means that a formalised sampling strategy, aimed
at covering a certain area, was not used in this case. For practical purposes,
this way of working was convenient, since bags containing flakes from
certain square metres did not have to be located. This somewhat
primitive, sampling strategy was used because it was anticipated that
166
D
5856
0
0
fD
1203
414
34%
EAB
5290
1109
21%
GAB
1838
507
28%
Table 6:1. Total number of flakes in each of the four main contexts at Myrhj
in relation to the sample size.
F lin t ty pe
S en o n ian
D an ian
V fD
77
23
Sample
69
31
EAB
70
30
Sample
68
32
GAB
55
45
Table 6:2. Percentages of the two flint types in the relevant contexts in
relation to the samples.
167
Jan Apel
4. Behind-the-edge flakes removed by a hard technique.
5. Pressure flakes.
6. Larger, fragmented flakes without preserved platform remnants.
In this context, small-sized, production debitage, such as small flake
fragments and splinters, will not be discussed. However, samples
including such flake fragments were also documented and I hope to be
able to present this data in the near future. In order to give a more
detailed, technological meaning to the flake categories presented above,
some generalising arguments about the categories that will be used to
separate axe-production debitage from dagger-production debitage will
be presented. Sub-category 1 flakes are consistently produced during
the thinning stages of bifacial-reduction strategies. These flakes have a
combination of attributes that are consistent with Flennikens or
Hgbergs definitions of bifacial-thinning flakes and dagger production
debitage that were presented in Chapter 5 (compare the flakes in Figs.
5:3 and 5:5 with the flakes of sub-category 1 in Fig. 6:4). A large amount
of the middle-stage, production debitage from dagger-making will be
of this type. However, these flakes will also be produced in some stages
of square-axe production, mainly during the construction of the cutting
edge, when on-the-edge flakes will be removed from the edge and
into the interior parts of the axe. For instance, it has been estimated that
these flakes make up only c. 1 % of the total amount of flakes from the
production of squared-shaped axes in stone (Sundstrm & Apel
1998:160 ff.). The investigation of an axe-production site in southwestern Skne reveals that 6 % (nine of 150) of the flakes from the
production of a Neolithic, square-sectioned axe in flint were of a typical
bifacial character (Hgberg 1999:86, Fig. 86). Generally speaking, the
flakes produced during the shaping and preparation of the cutting edge
on an axe will tend to be smaller than flakes from the thinning stages in
the production of bifacial artefacts, if the size is the same. This is due to
the fact that the former flakes are struck in order to form the edge and
do not have to reach into the central area of the axe side. Bifacial-thinning
flakes, on the other hand, must reach into the central area of the artefact
in order to be overlapped by flakes struck from the opposite margin,
thereby making the blade thinner. It must also be mentioned that onthe-edge flakes, with prepared platforms, will be produced during the
168
10 cm
Fig. 6:4. Examples of sub-category-1 flakes from the Myrhj site. These flakes
are the result of typical bifacial-reduction strategies. In this context, this flake
category is interpreted as production debitage from the later stages in the
production of flint daggers. Drawings by Lotta Sundberg.
169
Jan Apel
10 cm
Fig. 6:5. Examples of sub-category-2 flakes from Myrhj. In this context, this
flake category is interpreted as being related to the production of squareshaped, flint axes. Drawings by Lotta Sundberg.
170
10 cm
Fig. 6:6. Examples of sub-category-5 flakes from the Myrhj site. In this
context, this flake category is interpreted as being related to the parallelflaking of flint daggers of type I C. Drawings by Lotta Sundberg.
Jan Apel
70
60
50
40
30
20
10
0
00, 0,0
00
0
1- 09
0, 0,0
00
0
2- 19
0, 0,0
00
02
39
0, 0,0
00
0
4- 39
0, 0,0
00
0
5- 49
0, 0,0
00
0
6- 59
0, 0,0
00
06
79
0, 0,0
00
0
8- 79
0, 0,0
00
0
9- 89
0,
00
99
0,
01
-
0,
01
0,
00
0-
0,
00
1- 09
0,
0,
00
00
19
20,
0,
00
00
29
30,
0,
00
00
39
40,
0,
00
00
49
50,
0,
00
00
59
60,
0,
00
00
69
70,
0,
00
00
79
80,
0,
00
00
89
90,
00
99
70
60
50
40
30
20
10
0
00, 0,0
00
0
1- 09
0, 0,0
01
00
9
20, 0,0
0
00
3- 29
0, 0,0
0
00
4- 39
0, 0,0
0
00
5- 49
0, 0,0
05
00
9
60, 0,0
06
00
9
70, 0,0
07
00
9
80, 0,0
08
00
9
90,
00
99
0,
01
-
00, 0,0
00
0
1- 09
0, 0,0
00
0
2- 19
0, 0,0
00
02
39
0, 0,0
00
0
4- 39
0, 0,0
00
0
5- 49
0, 0,0
00
0
6- 59
0, 0,0
00
06
79
0, 0,0
00
0
8- 79
0, 0,0
00
0
9- 89
0,
00
99
0,
01
-
70
60
50
40
30
20
10
0
Fig. 6:7. The distribution of the density (measured as the average weight per mm2) of flakes
from sub-categories 1, 2, 4 and 5.
No cortex
85%
95%
68%
84%
58%
73%
96%
95%
< 50 %
14%
5%
25%
11%
27%
20%
3%
5%
> 50 %
1%
0%
8%
5%
15%
7%
1%
0%
Table 6:3. The degree of cortex coverage on the dorsal faces of the flake subcategories presented in the text (A = senonian, B = danian flint).
173
Jan Apel
19%
49%
51%
81%
EAB flakes of flint quality b
6%
27%
73%
GAB flakes of flint quality a
94%
GAB flakes of flint quality b
13%
38%
62%
87%
WfD flakes of flint quality a
Fig. 6:8. The relationship between sub-category-1 flakes (lighter area) and subcategoryc-2 flakes (darker area). The first flakes are produced mainly during
dagger making while the latter flakes are produced during square-axe production.
generally shaped when the body has reached its squared form. This means
that there is less of a chance that any cortex will still be present on the
axe body when the flakes are detached. During bifacial-reduction
strategies, however, the knapper will try to avoid detaching flakes from
areas that are flat or concave, since such flakes will tend to be abruptly
terminated in hinge fractures. Consequently, in areas such as these, cortex
is liable to be untouched well into the middle stages of the production.
If the bifacial, subcategory-1 flakes in danian flint actually are production
174
Fig. 6:9. The width (horisontal axis)-length (vertical axis) relationship of the sc1 flakes from Myrhj. Danian flint (circles), senonian flint (dots).
debitage from the making of square axes, they will tend to be shorter
than the bifacial flakes created during the production of, for instance,
daggers. In the diagram presented in Fig. 6:9, we can see that this was
actually the case on Myrhj.
The investigation of the flakes from the Myrhj site suggests that a
large part of the material was the result of the production of everyday
items, especially scrapers. However, traces of more sophisticated
knapping strategies are also present in the flakes. The presence of
subcategory-2 flakes, together with a large number of preforms, indicate that thick-butted axes were made on the site. These axes were preferably made of grey, calceous, danian flint, even though senonian flint,
to a lesser extent, was also used for this purpose. Bifacial-thinning flakes
and large pressure flakes in senonian flint reveal that the later stages in
the making of type-I flint daggers were also performed on the site.
Since large, early-stage flakes, from axe as well as dagger production, are
lacking on Myrhj, it may very well be that these early stages were
performed in more secluded places near the raw-material sources.
As we have seen in Chapters 2 and 4, production stages that require
a large degree of know-how are sometimes performed on full display in
tribal societies, in order to bring a message of authority to the members
of the community. In Fig. 6:10, a flow chart of different aspects of the
flint industry at Myrhj in relation to their inherent degrees of the
175
Jan Apel
Fig. 6:10. Flow chart of the flint industry at Myrhj related to the theoretical concepts of
the present and the past in the present presented in Chapter 4.
theoretical terms the present andthe past in the present (see Chapter 3) is
presented.
Stendis (4)
The settlement at Stendis in north-western Jutland resembles Myrhj
in many ways. The house construction is similar to those of the three
houses at Myrhj. Bell-Beaker pottery and large amounts of flint were
recovered at both sites. At Stendis, 1 445 pieces of flint were found. Of
these, 24 were secondarily worked (Skov 1982:40). There was a general
tendency for the un-retouched flakes to be short and wide, just like the
Myrhj flakes. In his report, Skov does not mention anything about
the platforms on the flakes, and, as we have seen above, this makes it
difficult to give a technological interpretation of the material on the
basis of the written report. However, as at Myrhj, some of these flakes
may very well have been debitage from bifacial-production sequences.
No lancet-shaped daggers were found at the site, but three bifacial
arrowheads and one preform for an arrowhead were recovered (Skov
1982:40).
Skov mentions three hammerstones in quartzite with battered ends,
which, he suggests, were used for finer flint work, such as the making
of bifacial arrowheads and daggers (Skov 1982:40). However, it seems
unlikely that the stones were used for this purpose. Quartzite would
make a hard hammer and this in turn would indicate behind-the-edge
flaking, which requires a thick platform (Callahan 1979:67). This
technique, if used at all in biface work, would be preferred only in the
early stages of production. On the other hand, if the flakes at the site
stem from square-axe production, the hammerstones may be related to
this activity.
Tastum (5)
At Tastum, another typical, Late Neolithic, house construction with a
sunken floor was recovered. Bell-Beaker pottery was also found here. A
layer in the house contained a fragment of a dagger handle and a bifacialarrowhead preform (Simonsen 1985:85). This layer is interpreted as a
secondary filling of the floor pit that was filled in after the house had
177
Jan Apel
been abandoned (Simonsen 1985:82). Simonsen writes that the material
confirms the exclusion of a dating of the house to the period previous
to LN (Simonsen 1985:87). This is a curious statement, since the
artefacts that were actually found in the depression were interpreted as
secondary filling. However, by analogy with the sites mentioned above,
it seems likely that the site should be dated to an early part of the Late
Neolithic. Apparently, it was not possible to classify the dagger fragments
according to any of Lomborgs types (Simonsen 1985:86).
Jan Apel
Category
Flakes
Cores
Preforms
Retouched frost flakes
Hammerstones
Burnt flint
Scrapers
Blades
Burins
Borers
Knives
Strike-a-lights
Arrowheads
Axes
Daggers
Sickles
Thy 2758
1211
8
5
16
0
242
5
0
2
0
2
0
3
1
1
1
Thy 2756
91
0
0
0
0
30
0
0
0
0
0
0
0
0
0
0
Thy 2922
470
1
1
0
0
0
4
10
0
0
0
0
0
0
2
0
Table 6:4. Flint artefacts recovered in plough-zone samples taken at Late Neolithic sites
investigated by the Thy project (from Steinberg 1997:144 ff.).
m2 and seven plough-zone samples distributed over the site area were
sieved. Only 91 flakes and 30 pieces of burnt flint were recovered (Table
6:4). Fortunately, the site could be dated to the Late Neolithic period
on account of a dagger found during field-walking, an axe of a Late
Neolithic type found in the excavation backfill and a bifacial arrowhead
found during the excavation of the cultural layer (Steinberg 1997:150).
181
Jan Apel
sites fairly homogeneous, while the Bronze Age sites were heterogeneous.
Consequently, Steinberg suggests that there was no intra-site
specialisation during the Early and Late Neolithic and that specialisation
occurred in the Bronze Age. This investigation thus support Kristiansens
view that specialisation, and therefore also chiefdom complexity, occurred
during the Bronze Age. Unfortunately, I have not had time to visit the
Thy museum in order to investigate the archaeological materials from
the Thy project personally. However, after consulting Jens-Henrik Bech,
the head of the Thy museum, by e-mail, it became apparent that no
dagger-production sites had been investigated in the Thy region. The
only specialised production sites that were recovered in the project were
Bronze Age sites with evidence of bifacial-sickle production. The fact
that no specialised Late Neolithic production site was investigated within
the Thy project was not, as I see it, a result of the fact that there is no
technological specialisation was present during this time. On the contrary,
it is a strong indication of intra-site specialisation during the Late
Neolithic. The geographical scale of the Thy project was simply not
large enough to contain the complexity of Late Neolithic craft
specialisation in southern Scandinavia. It is quite possible that the results
of Steinbergs investigation would have been different if the surveys
had been carried out in the parishes of Thy, such as Thisted, where Late
Neolithic flint mines have been found. It is likely that more specialised
production sites were located in this area. Steinbergs conclusion would
certainly have been different if he had considered the Late Neolithic
sites in Himmerland and Djursland, such as Myrhj, that are presented
in this chapter. It seems as if the investigation of the Late Neolithic sites
in Snderh parish managed to map only the domestic aspects of flint
work, simply because the specialised production took place elsewhere.
As will be evident in the conclusion of this chapter, my opinion is
that it is impossible to understand the political landscape of the Thy
region without considering an area covering, at least, the central and
northern parts of Jutland. The specialisation of the Late Neolithic period,
that Steinberg (1997) and Earle et al. (1998) failed to recognise, is
evident on considering a larger geographical area, including sites with
evidence of flint extraction and dagger production. When Steinberg
proposes that the flint daggers were unproductive artefacts in the exchange
networks of prestige objects, as is implied by the following quote, he is
simply ignoring plain facts: The Late Neolithic is characterized by a
182
Egeh j (7)
At Egehj in Djursland, three houses with rows of single, roof-bearing,
post holes were recovered not far from the Egehj barrow. This barrow
has been excavated and contained a burial dated to the Early Bronze
Age (Boas 1983:90). One hundred and twenty kilograms of flint was
found at the site. More than 20 000 waste flakes and many artefacts
were excavated. Bifacial arrowheads with hollowed bases make up the
biggest find category, a total of 90 examples being found. Twenty-one
daggers and dagger preforms were recovered (Boas 1983:95). Two types
of daggers are represented in the material Lomborgs type V and
type VI. The type VI daggers are all under 13 cm, which, according to
Lomborgs classification, would make them strike-a-lights of type A
or B (Lomborg 1973:27 f., 1959:160). Three of these dagger/strike-alights are complete and show no wear pattern and this may mean that
they were produced on the site. Apart from the fact that one would not
expect used tools on a production site, Lomborgs length definition is
hard to accept. Further down the chain of consumption, only distinct
wear patterns or contexts (hearths, etc.) would permit a strike-a-light
183
Jan Apel
interpretation before a dagger interpretation. However, it seems to be
clear that these specimens should be dated to the Early Bronze Age.
The flint at Egehj is of a very good quality. Only 18 % of the
scrapers at Egehj have traces of cortex on their dorsal surfaces. This
should be compared with the Myrhj site, where 80 % showed signs of
cortex. Only 1 % of the flint debitage consists of short and wide flakes.
Boas suggests that this kind of flake resulted from axe manufacture and
that the higher figure at Myrhj (26 %) is explained by the large number
of axes that were found there (Boas 1983:95). Axes were not found at
the Egehj site. This reasoning implies that Boas was acquainted with
axe-production technology and was able to recognise the morphological
attributes of its debitage. If this was the case, it is not apparent in the
site report. One way of confirming this hypothesis would be to reexamine the Myrhj material, classify the flakes by flint and see in which
category the majority of the flakes fall. Good-quality flint would imply
cutting tools and poor quality flint would imply chopping tools.
Another way to verify the hypothesis would be to make a technological
analysis of the flakes from Myrhj and compare the material with the
results of square-axe debitage (Hansen & Madsen 1984) and with the
flint-dagger debitage, as it is presented in this volume. It is important
to remember that other reduction methods than square-axe production
produce short and wide flakes. Flakes with attributes like these are
produced in the making of most bifacial-tool types and, as was the case
with the Myrhj material, one would have to consider a range of
technological information, i.e. types of platforms, etc., before a
judgement could be made.
The distribution of the flint material allows Boas to make a detailed
interpretation of different activity areas and deposits in the settlement.
He divides the flint material into three categories in order to examine
specific activity areas (Boas 1983:97): (1) everyday items such as simple
flake tools, (2) finished pressure-flaked tools, and (3) production
debitage and preforms in flint (Boas 1983:97). The last group is seen as
indicating specific activity areas connected with the production of bifacial
arrowheads. Boas concludes that this, therefore, seems to be an example
of specialized arrowhead production, with rough shaping taking place
near the first roof-bearing post, finer working between this post and
the gable wall, and the apparently rejected examples of rough-outs being
184
Jan Apel
Boas was able to establish a chronology of the houses that stretches
from the Middle Neolithic to the Late Bronze Age. This scheme also
includes one of the houses at Egehj (Boas 1983). It is interesting to
note that the earliest Late Neolithic house, House III at Hemmed
plantation, contained fragments of type-I daggers, that the latest Late
Neolithic house, House I at Hemmed plantation, contained the hilt of
a type-IV dagger and that the earliest Bronze Age house, House III at
Hemmed church, contained a type-V dagger (Boas 1991:125 ff.). This
confirms that lancet-shaped daggers date to LN I and that daggers with
fishtailed handles date to LN II or the Early Bronze Age. In fact, these
datings also support Lomborgs chronological distinction between types
IV and V. For a thorough discussion of the chronological value of the
dagger typology, see Chapter 8.
Most of the artefacts at Hemmed church were recovered in house
III. This applies to the flint waste as well. Boas does not interpret any
activity area in connection with the waste and implements of flint. One
would suspect that a technological examination of the waste material in
house III would give us information as to which flint-knapping was
carried on and, if that was the case, what reduction strategy was in use.
However, This has not been done.
Jan Apel
Vejlby (10)
This site is located on a slope in the Eg valley at Vejlby parish in the
Hasle district, north of the town of rhus (Fig. 6:1). At the site, a pit
(4.5 m x 4 m) was excavated and was interpreted as the sunken, eastern
end of a Late Neolithic/Early Bronze Age house, such as the ones
discussed above (Jeppesen 1984:101 f.). Four post holes were also
revealed and they seemed to constitute part of the slightly rounded,
western end of the house (Jeppesen 1984:102). If the house
interpretation is correct, Jeppesen suggests that we are dealing with a
house at least 9.5 m long and 5 m wide. The pit itself contained charcoal,
numerous small fragments of fire-cracked stone, flint objects, flint waste
and pottery (Jeppesen 1984:99). I have personally inspected the flint
material from the excavation and there are no signs of the production
of bifacial tools on the site.
Forn s (11)
Most of the dagger-production sites that have been discussed so far
have been found in settlement contexts. There are, however, also
production sites with no evidence of structures, such as houses or huts.
Forns in Djursland is such a place. The site is located on the beach just
beside Sangstrups Klint, a white limestone rock that dives into the
Kattegatt. The beach is littered with nodules that have fallen down
from the cliff (Glob 1951:24). There are three types of raw flint material
at Forns; the first is dark and the second is lighter (white-brownish
or white-grey). The Late Neolithic flint-knappers used both of these
types extensively. The third flint type is grey and was used almost exclusively for axe preforms (Glob 1951:26). These flint types correspond
to the senonian and danian flint types mentioned earlier. As was the
case at Myrhj, the better-quality, senonian flint was used for more
complicated flint work, such as dagger production, while the dryer,
danian flint was used for axe production. This may reflect a convention. It seems clear that axes were seldom made of the finest quality
flint, not even at places where the raw material was abundant. During
the Late Neolithic period, large nodules of finer quality flint were exclusively used in the production of bifaces. This coincides with
observations that Errett Callahan has made in Danish museum
collections: Danish flint daggers are almost exclusively made of senonian
flint. This is, however, a rule with exceptions. Flint daggers were indeed
188
Jan Apel
Artefact categories
Axe preforms, complete
fragments
Bifacial preforms, whole
fragments
Sickle Preforms, whole
fragments
Axes
Daggers, whole
handles
blades
cracked
Sickles, whole
fragments
Spearheads
Arrowheads
Strike-a-lights
Blade scrapers
Borer
Hammerstones, flint
stone
Black flint
17
16
66
72
27
51
6
2
38
15
5
15
31
3
4
8
17
2
W/G flint
20
17
1
4
4
1
7
7
1
1
1
1
8
11
7
3
2
1
Total
69
63
76
90
33
52
8
2
46
24
6
16
32
4
12
21
25
5
6
65
Table 6:5. Breakdown of totals of flint objects at the Forns site (from Glob
1951:30, Fig. 8).
Jan Apel
Tegelbarg (12)
Not far from the town of Flensburg in northern Germany on a former
shoreline of the Baltic Sea, a Late Neolithic site was excavated by Volker
Arnold at the beginning of the seventies. Considerable flint waste was
recovered at the site. All artefacts were plotted within a quarter-metre
grid, and this accuracy made it possible to distinguish between
concentrations of flint waste that could be analysed. A large proportion
(75 %) of flint artefacts and production debitage recovered on the site
was found in a two square meter area. Arnold is of the opinion that the
heap did not represent a primary deposition but that it should be
regarded as a secondary deposition of material from an as yet unrecognised
knapping site (Arnold 1990). In cases like this, in which a secondary
deposition has been found, it may be possible to recognise a nearby
primary deposition by taking samples with a probe every metre to detect
micro debitage that, owing to its size, missed the maintenance of the
activity area (af Geijerstam 1996). Of course, if a blanket or some other
kind of foundation had been in use during the production, this would
not have produced such results, but then the micro-debitage could have
been detected in the secondary deposition (Thorsberg 1984:47 ff.,
Schiffer 1987:63). If so, one way to define an activity area of this kind
would be to make a spatial interpretation of the documented scatter
pattern. In Callahan & Apel (ms), the scatter patterns documented during dagger-production experiments will be presented and discussed.
From the method of refitting, Arnold concluded that the heap
contained debitage from the production of at least four different flintknives. Among them, two seemed to have been lancet-shaped daggers
and one a rough-out, perhaps for another dagger. Only a few flakes
could be assigned to the fourth knife (Arnold 1990). The waste of the
other three specimens consisted of 80 flakes from a broken, unfinished
Lancet-shaped dagger, 87 flakes from a dagger rough-out, and 155 flakes
that revealed an almost complete, dagger-reduction sequence. The original
blank for this piece could be completely reconstructed (Arnold 1990).
The refitted part is like a cast, and the lancet-shaped-dagger interpretation
is based on the shape of the missing piece in the middle of the blank.
The produced dagger was not recovered at the site.
The successful refitting of this dagger allowed for some interesting
technological conclusions. Early in the production sequence, hardhammer percussion had been used. This activity is interpreted as having
192
Jan Apel
Fig. 6:11. Two flint daggers found in the separate layers at Drenges. On the
left, a type-III A dagger from the earlier layer and, on the right, a broken
preform of a type-VI dagger from the later layer (from Kempfner-Jrgensen &
Liversage 1985:23 f., Figs. 8 & 10).
Jan Apel
flaked. Judging from the small pressure flakes, some of the arrowheads
may have been produced on the site. In general, a lot more flint tools
were found at Rjle Mose than at Lindebjerg.
Str by (16)
In 1934, Therkel Mathiassen published an article on the Neolithic flint
trade. The article was initiated by numerous flint artefacts discovered in
an area surrounding Strby Ladeplads at Stevns on eastern Zealand (Fig.
6:1). Here, collectors had found hundreds of artefacts, mainly preforms
for different tool types, made when the area was exploited. Mathiassen
suggested that the prehistoric peoples of this area exported preforms for
artefacts that were finished in other places (Mathiassen 1934:22). Most
of the finished artefacts that were recovered were axes or axe preforms,
but some of the square-shaped preforms were so thin that they were
more likely dagger preforms. If this was the case, there seem to have
been at least two fundamentally different methods of creating preforms
suitable for daggers, the bifacially almond-shaped preform and the square
preform. Maybe these two methods could be spatially distinguished
from each other (Jutland/Zealand?). In any event, a finished flint dagger
has been discovered among the preforms, but Mathiassen does not discuss its morphology (Mathiassen 1934:20).
The danian flint at Stevns is tannish-grey and calceous and is not as
brittle and easily worked as the senonian flints of Stevns, Falster/Lolland
or northern Jutland. Thus, it is not quite suitable for advanced dagger
production (Callahan, personal communication). As we have seen above,
the dagger-production centre moved from northern Jutland to the Danish
isles in the transition between LN I and LN II (Vankilde 1989). The
not-yet-found production sites of large, prestigious, type-IV daggers
are therefore likely to be found on Lolland/Falster and perhaps Zealand.
It is possible that they may be found as the result of a specific survey
aimed at detecting sites like these. The uniformity that these daggers
show indicates that they may have been the products of a single production site with quality control (Callahan, personal communication).
This is obviously not the case with the earlier types. An increased specialisation that occurred during LN II may therefore explain why production sites of this phase are scarce.
196
Sk ne
Skne, the southernmost part of Sweden, should be mentioned in this
context. This part of Sweden has more in common with Denmark
than with other parts of the country. For instance, this is the only area
in Sweden where flint is abundant. In 1951, Mrta Strmberg wrote
that there were approximately 60 known settlements with material from
the Late Neolithic period in Skne. However, almost all these sites were
mixed settlements with material from other periods (Strmberg
1951:72). The Late Neolithic artefacts on these sites were few, compared
to the materials from other periods.
On some of these sites, especially around the Barsebck area and in
the Malm area, numerous flint daggers have been recovered (Althin
1954: sites 55, 56 and 56a). It is likely that there are many production
contexts to be found in these parts of Skne. However, for the time
being, as regards dagger-production sites, I shall restrict myself mainly
to the data I have gathered from Denmark.
197
Jan Apel
(Becker 1993:111). At Skovbakken, the flint was mined in two different ways. In cases in which the flint was available close to the surface,
pits were dug into the chalk from the ground surface (Rudebeck, in
press). When the flint layer was situated at a deeper level, shafts were
dug sometimes up to 7 m into the chalk. From these shafts, galleries
were dug in several directions when the flint layers were reached (Becker
1993: 111 f., Rudebeck, in press). In 1993, Becker pointed out that no
Late Neolithic settlements had been found in the vicinity of the
Skovbakken mines (Becker 1993:112). However, during the last couple
of years, an excavation conducted by the Aalborg Historical Museum at
Bejsebakken, located c. 500 m from the mines at Skovbakken, has
produced several Late Neolithic houses and large numbers of formal
artefacts and debitage in flint. It is likely that the population that used
the flint mines lived here. The flint nodules that were extracted from
the Skovbakken mine-shafts were quite small. Consequently, only daggers of up to 20-25 cm could be made from these nodules (Becker
1993:112).
At Hov, near Hillerslev in the province of Thy in the western part of
the Limfjord area, smaller mines, dated to the Late Neolithic, were
recovered in 1957 (Becker 1980, Rudebeck, in the press). The mining
seems to have been carried on in a similar way to the mining at
Skovbakken. Circular shafts with a diameter of about 4 m were dug
into the ground. When a flint layer was found, galleries were extended
in different directions.
As was the case at Skovbakken, the Late Neolithic mines at Hov did
not, according to Becker, produce large nodules. Consequently, it was
concluded that the large daggers of type I C, which undoubtedly
originated from the Limfjord area, must have been made of flint from
mines that had not yet been recovered (Becker 1993:121). I have
personally collected large flint nodules at a present day chalk mine at
Hillerslev near Hov and must agree with Becker on this point. However,
I do not agree with his casual remarks on the origins of some of the
other dagger types. Becker does not acknowledge, as is suggested in this
book, that certain dagger types may have been produced almost
exclusively on the eastern Danish islands. Even though Becker was aware
of the fact that the majority of the caches of certain types of daggers, for
instance, types I D, II and III, were located on the eastern Danish islands,
198
Fig. 6:12. Cross-section and horizontal plan of three Late Neolithic mines at Skovbakken
near Aalborg (from Becker 1993:112, Fig. 1).
Conclusion
Some of the investigated sites display production debitage that contain
a high proportion of knapping errors. Interestingly, these sites were
located to secluded places near the extraction area (for instance Forns
and Drenges). On the settlement sites however, such as Gug and
Myrhj, evidence of high quality knapping as well as knapping of lower
quality has been recognised. A closer examination of the flint production
debitage from Myrhj revealed that debitage from both the production
of everyday items and debitage from the production of more elaborate
artefacts, such as thick-butted axes and daggers, are present on the site.
It is suggested in this context that the range in the quality of the flint
industry at the Gug, investigated by Olausson (1997 & 2000), may be
explained in a similar way.
199
Jan Apel
Thus, the most important result in this chapter is that the production of flint daggers usually took place on at least three different types
of sites. In the following paragraphs, I shall relate these sites to some
theoretical issues, regarding the social organisation of complex
technologies in traditional societies, discussed in Chapter 4. First of all,
the flint was extracted at certain locations where the raw material was
readily available. Obviously, the flint mines at Hov and Skovbakken
represent important links in the chain of production. However, certain
beaches, such as those at Forns and Drenges, where flint nodules were
washed clean by the water, were probably important as extraction sites
as well. In fact, since the Late Neolithic farming methods were based
on the use of ards, there is no reason to believe that flint nodules were
ploughed up to the surface in the way that happens with modern farming
techniques. Thus, on Lolland and Falster, where large flint nodules of
good quality can be picked up in the ploughed fields of today, suitable
nodules would have had to be prospected for in prehistory, just as chunks
of obsidian are located with iron rods and dug up by hand at Flint
Butes in present-day Oregon (Callahan, personal communication).
Alternatively, nodules could be collected on the beaches, where they
were washed clean from soil and were readily available for collection. It
has been suggested that the limited availability of high-quality flint
nodules for the production of prestige artefacts, such as axes and daggers,
during the Neolithic might have contributed to the fact that a large
part of the production can often be found outside the settlements
(Knarrstrm 2000a:163). However, in the following conclusion, I shall
focus mainly on the relation between the intra-site, spatial distribution
of the dagger technology and the theoretical issues discussed in Chapters
2-4.
The early stages of the dagger production were, owing to a relatively
low degree of know-how, conducted at secluded and private places,
preferably near the natural flint resources. Just as regards traditional ironworking in Kenya, where the smelting is surrounded by prohibitions
and taboos, these stages were shared with other people apart from those
directly involved in the production. The low degree of know-how did
not guarantee that the know-how could be kept exclusive by within the
group if the technology was conducted openly. The symbolic capital
that was connected with the craft could only remain a valuable asset as
long as the tricks involved in the production could be kept secret.
200
201
Chapter 7
Flint Technology
in Eastern Central Sweden
Jan Apel
perspective, it will then not be possible to incorporate the static archaeological material into a systemic context (Binford 1983:19 f.).
In the following pages, flint artefacts from 16 Late Neolithic and
Early Bronze Age sites in eastern central Sweden are discussed (Fig. 7:1).
In this context, eastern central Sweden consists of the provinces of
Uppland, Vstmanland and Sdermanland. The majority of the sites
have been interpreted as settlements. However, in order to broaden and
deepen the context of the production and consumption of flint in the
area, the flint materials from a Late Neolithic mortuary (Skillsta 2:15,
Skogs-Tibble parish in the province of Uppland), a ritual deposition of
thin-edged, broad-bladed, flint axes (Granath Zilln 1999) and four
Gallery-Graves have also been studied. The geographical distributions
of the sites and the flint artefacts are presented in Fig. 4:1 and Table 4:1.
The main idea behind this study is to establish how flint, in the form
of flakes and more elaborate objects, was used in social strategies to gain
power and prestige, similar to those used in the Late Neolithic
communities of southern Scandinavia.
The sites included in this chapter are not the only ones known
from the period in this part of Sweden. For an extensive presentation of
the archaeological sites dating to the Late Neolithic period and excavated
by the National Heritage Board (UV-mitt and UV-Uppsala) see Holm
et al. (1997). For a presentation of the Gallery-Graves of this region,
dating to the Late Neolithic and Early Bronze Age periods, see Apel
(1991 & 1992) and Holm et al. (1997). However, in my opinion the
information to be gained from the flint materials in these contexts would
not change the conclusions of this chapter in any significant way. On
the contrary, I believe that the 16 sites included in this study are representative of the flint industry in the currently known, Late Neolithic
occupation of this region. This is indicated by the repeated occurrence
of certain features of the flint industry on the sampled sites. The Late
Neolithic, flint materials in eastern central Sweden have been discussed
briefly in Holm et al. who concluded that there are few traces of a local
reduction in the settlements and that there is no evidence of a bifacial
technology. This conclusion is not entirely consistent with my own
experience of excavating the Late Neolithic sites of Fgelbacken in the
province of Vstmanland (Lekberg 1996) and Lugnet in the province
of Uppland (Lekberg & Apel, ms). I believe that it is possible to extend
this first impression given by Holm et al. in a way that may lead to a
204
Fig. 7:1. Late Neolithic sites in eastern central Sweden mentioned in the text.
Settlements are marked by circles, gallery graves by triangles and the ritual
deposition by a square. The small dots mark the parishes where flint daggers
have been found. The sites are projected on a map showing the coastline at
3600 BP (c. 2000 B.C. cal.), made by Leif Andersson at the Swedish Geological
Survey (SGU) in Uppsala.
deeper knowledge of the flint industry and its context and thus of the
role that flint artefacts may have played in the formation and
reproduction of the social order in eastern central Sweden during the
Late Neolithic period.
The sites
In this section, a selection of the Late Neolithic and Early Bronze Age
settlement sites with flint material treated in the following analysis is
briefly presented in order to inform the reader of the contexts in which
flint is to be found. Some of the sites included in this study have not
been properly published. This is mainly due to the fact that they have
205
Jan Apel
been excavated recently. In these cases, the excavators have been kind
enough to let me present the flint materials, but any further presentation will have to wait until the post-excavation work is finished and
published. However, there is one exception. The Piparberget site was
excavated in 1998 and the excavators have kindly provided me with
additional data prior to the publishing of the report, thus making it
possible to include a brief description of the site in this context.
F gelbacken, ra 163, 160 and 162 (Hubbo parish in the province of V stmanland)
These three sites were discovered during salvage-excavations of a
prehistoric complex with several archaeological sites dating from the
Early Neolithic to the Viking Age (Lekberg 1997).
Ra 163 was located on a sandy plateau on the western shore of a
bog, c. 43 m above the present sea-level. Only two features were recovered during the investigation, a hearth and a pit with fire-cracked stone.
The artefacts consisted of pottery, a hammerstone, a grindstone and
some stone flakes, a core and a scraper in quartz. Twenty-seven flint
artefacts were recovered. Nine of these were complete flakes, while 16
were fragments of complete flakes. The formal artefacts consisted of
the tip of a bifacial arrowhead and a fragment of the handle of a type
IIA flint dagger (Fig. 7:2). Two samples, charcoal from the hearth and a
burnt hazelnut shell, were C14-dated to 2140 70 BP (Ua-10 000)
and 4420 90 BP (Ua-4091) respectively (1 sigma, Stuiver & Pearson
1986). Thus, the Late Neolithic dating of the site was based solely on
typology. The flint dagger seemed to be of an older type; the site was
therefore dated to LN I (2350-1950 cal BC).
206
Jan Apel
chronologically separated horizons: an Early Iron Age phase and a
Neolithic phase. The latter horizon, which concerns us here, consisted
of potsherds from at least 35 identified vessels that were classified as
Battle-Axe pottery of Malmers types Ad H and J (Lekberg 1996:
Chapter 9, Malmer 1962:8 ff. & 1975:26). These types are usually
dates to the later part of the Middle Neolithic period and the flint
material from this context would therefore appear to be of less interest
in this study. However, Battle-Axe pottery of the J type, now and then,
occurs in Late Neolithic contexts (M. Olausson 1995, Holm et al.
1997) and there is a strong possibility that this pottery was also used
during the early part of the Late Neolithic. Thus, the flint from ra 162
is also included in the following analysis.
The three sites at Fgelbacken were tentatively interpreted as making
up different parts of the periphery of a large, Late Neolithic, settlement
complex only partly affected by the salvage excavation (Lekberg 1996).
Thus, the flint material is treated as coming from a single site. During
the Late Neolithic period, when the sea-level was situated about 25-30
m above the present one, the Late Neolithic parts of the Fgelbacken
complex were on an inland site.
Number
54
148
205
6
4
20
116
1
163
236
3
1
Weight (g)
45,4
2027,3
34,8
66,4
2,2
42,5
1238,2
1,1
164,6
351,6
71
0,9
Table 7:1. The find material recovered at the Lugnet site in the province of
Uppland.
with an offset beneath the rim date the site to Late Neolithic or Early
Bronze Age (Lekberg & Apel, ms). Finds of later dates were also found
(Table 7:1).
All but five pieces of the flint artefacts were unmodified, complete
or fragmented flakes. However, two fragmented arrowheads, two flakes
with retouches (interpreted as arrowhead preforms) and one gun flint
were also found at the site.
Jan Apel
According to the report, most of the remains belonged to two, separate,
chronological phases: one was the Late Neolithic and the other was
oldest phase of the Iron Age (Appelgren et al. 2000:4). During the
excavation, 232 features were revealed. Of these, 22 were hearths and
numerous post holes were also documented. Two areas with post holes
and cultural layers were tentatively interpreted as the remains either of
pit houses from the Iron Age or of depressions belonging to Late
Neolithic long houses (see Appelgren et al. 2000:20 f.).
Numerous artefacts of different kinds were recovered during the
investigation. This material consisted of, among other things, several
stone axes and axe fragments, grindstones for the processing of vegetables or crops, pottery, worked quartz and burned and unburned bone
(Appelgren et al. 2000:13). The bones of sheep or goats, pigs, horses
and cattle were recovered on the site (Appelgren et al. 2000:17). The
flint material from the site is presented in Table 7:2.
Conclusion
Eastern central Sweden has been severely affected by land uplift in the
wake of the latest ice age. As a result, the sea-level during the Late
Neolithic period was approximately 25-30 m above the present sealevel. Among the settlements presented above, there are both inland
and coastal sites. Annelund and Grndal (both located c. 27 meters
above present sea level) were situated on the coast, while, for instance,
Lugnet and the Fgelbacken sites, located between 40 and 50 m above
the present sea-level, were inland sites. Several of the sites are mixed, in
the sense that they contain artefacts or ecofacts from other periods. In
some cases, such as Fgelbacken, older materials are present, while in
most cases Iron Age materials predominate.
Jan Apel
The material was initially divided into two groups: formally defined artefacts and production debitage (Fig. 7:4). The first group consists
of the following artefact categories:
(a) daggers
(b) sickles
(c) bifacial arrowheads
(d) square-shaped axes/chisels
(e) unspecified bifacial objects
(f) scrapers
(g) strike-a-lights
(h) retouched flakes
The second group, the production debitage, consists of what appear
to be unmodified, complete or fragmented flakes and cores. Since this
group must be understood in detail if the aim of this paper is to be
fulfilled, the material was subjected to a detailed analysis.
Consequently, the production debitage was related to a couple of
important analytical concepts in the field of lithic analysis, namely those
of technique and method (see Chapter 5). These two concepts, together
with the natural constraints of the flint, determine the appearance of a
flake. Technique refers to the way in which the power is transmitted
from the knapper to the raw material. Method, on the other hand,
refers to the choices of techniques, impact angles and working positions,
as well as the intentions of the knapper during the production process.
The relative importance of these concepts, as regards the appearance of
the production debitage, has been extensively discussed in the literature
(Bradley 1978, Hayden & Hutchings 1989, Dibble & Pelcin 1995,
Pelcin 1997 and Bradbury & Carr 1999). In recent years, it has become
apparent that the choice of method plays a greater role in this process
than previously realized. It has been acknowledged that experienced flint
knappers can produce flakes with similar attributes but by different
techniques (Callahan, personal information). The obvious conclusion
of these findings is that the intentions of the flint knapper, which are a
vital element in the method concept, will determine the morphology
of the flakes (Pelcin 1997, Bradbury & Carr 1999). I shall not discuss
this problem any further in this paper. It will be sufficient to conclude
212
sc4
sc5
sc3
h
c
10 cm
Fig. 7:4. Drawings of different types of production debitage and formally
defined artefacts in flint recovered at Late Neolithic sites in eastern central
Sweden. The artefacts have been classified according to the system described
in the text. Drawings by Alicja Grenberger.
213
Jan Apel
that the production debitage from the Late Neolithic sites of eastern
central Sweden has been assigned to categories that are based on choices
of technique and/or method (see Chapter 5).
The production debitage was primarily divided into the following
categories, which are discussed in more detail in Chapter 5:
1. Complete or fragmented flakes with lips on the inside of the platform
remnant.
These flakes lack signs of crushing on the platform remnant and
pronounced bulbs on the inside.
2. Complete or fragmented flakes with pronounced bulbs on the inside
and no lipping at the edge of the platform remnant.
These flakes often bear signs of crushing on the platform and/or on
the distal end and sometimes display lateral fractures from the platform remnant downwards.
3. Small (less than 15 mm in long), complete or fragmented flakes
produced with lipped platform remnants.
These flakes are distinguished by their small size as well as by the
fact that the platform remnant is in mint condition.
The three categories coincide in some parts coincide with the terminology based on the technique that was formerly used to classify lithicproduction debitage, i.e. the flakes would have been regarded as detached with soft, hard or pressure technique respectively.
Remaining flakes
Pieces of flint that cannot be assigned to any of the categories proposed
above have been classified in sc6.
Conclusions
In Table 7:2, all formal objects and production debitage in flint from
the sites included in the investigation are presented according to the
described classification system. From these facts, it may be concluded
that complete, formally defined artefacts, with the exception of the
retouched flakes, were not deposited in settlements if they were not
broken. Accordingly, it seems as if formally defined artefacts were
215
Jan Apel
primarily deposited in graves (Table 7:2). However, there are some
minor exceptions to this rule. In the settlement at Annelund/Stenvreten,
two complete, bifacial arrowheads were found, and at the Grndal
settlement several, complete, flint artefacts were recovered. In the latter
case, a source-critical explanation can be put forward. Grndal is a large
site and it has been extensively excavated by hand in a way in which no
other settlement site included in the study has been excavated. This
procedure may have had the effect that scarce find categories not usually
found in settlements were recovered here (Appelgren et al. 2000). This
knowledge should influence the choice of excavation methods used on
Late Neolithic settlement in the future, at least if the artefact assemblage
and its composition are expected to offer important contributions to
the interpretation of the sites. From the general point of view, it must
be mentioned that almost all the settlement sites included in this study
are problematic in the source-critical perspective. In many cases, the
sites are of mixed ages, and structures, in the form of houses etc., are
unusual. In this context, I shall not discuss this problem further but
may return to this subject in the future.
Production debitage is not usually found in graves but predominates in the settlements. When production debitage is recovered in graves,
it is usually in the form of disc-shaped flakes that are excellent as blanks
for bifacial arrowheads. Two of the four gallery-graves included in this
study contained production debitage in flint. In the gallery-grave
at Sjtorpet, one flake was found and in the Annelund gallery-grave,
three (Fig. 7:5). Two of the flakes in the Annelund grave were produced
by a bifacial-reduction method and were thereby classified as sc1-flakes.
It is likely that they were deposited as arrowhead blanks. This interpretation is further enhanced by the fact that some of the retouched flakes
found in the Annelund grave should most probably be regarded as
arrowhead preforms. Curiously, a bipolar flint core was also found in
the Annelund grave. A retouched, disc-shaped flake was also recovered
in the gallery-grave at Dragby. This flake is a good example of a preform
for a bifacial arrowhead. The flake found in the gallery-grave at
Sjtorpet has not been studied by me personally and is therefore assigned
to sub-category 6. However, it would not surprise me if this flake was
in fact disc-shaped and stemmed from a bifacial-reduction sequence.
216
217
Jan Apel
Formal artefacts
Site
0
0
0
0
0
0
0
0
0
0
0
0
2
11
2
1
0
0
0
2
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
12
0
0
0
0
1
18
16
3
4
3
2
0
1
1
1
0
0
7
2
17
3
1
10
1
7
0
6
2
8
10
17
38
5
11
11
7
13
25
46
56
122
1
0
1
0
0
1
0
1
1
1
1
2
0
0
2
0
0
2
0
1
1
1
3
7
0
1
1
2
0
0
4
0
0
2
0
0
0
0
5
3
1
3
0
0
0
0
3
0
2
5
0
0
1
0
12
5
2
13
0
0
0
0
1
3
3
20
26
0
1
1
5
25
12
49
32
6
4
3
37
45
57
24
18
25
90
46
139
458
Production debitage
d e
tot.
21 10 3
Table 7:2. The flint material grouped by site and by the classification system used in the text.
Formal artefacts, except retouched flakes (cat. h) that are complete, are marked by bold type.
The first five sites consist of the four gallery graves and a ritual deposit, while the last ten
sites have been interpreted as settlements. Further information on the sites can be found in
the following literature: Andersson 2001 (Alby), Andersson & Hjrthner-Holdar 1989,
Andersson, Fagerlund & Hamilton 1990, Fagerlund & Hamilton 1995 (Annelund/Stenvreten),
Borenius-Jrpeland MS (Piparberget), Eklf 1999 (stertlje 1:15), Appelgren et al. 2000
(Grndal), Gejvall 1963 (Dragby), Granath Zilln 1999 (Rssberga), Hallesj & Hamilton
1996 (Maren), Jaanusson 1969 (Bjurhovda), Jaanusson & Silvn 1962 (Dragby), Lekberg (ed.)
1996 (Fgelbacken), Lekberg & Apel MS (Lugnet), Olausson 1995 (Sjtorpet), Strucke ms
(Mrby), Strucke 1998 (Ribby).
218
10 cm
Fig. 7:5. Retouched flake (designated scraper in the report) from the
Annelund gallery grave. This flake is, in this context, regarded as a preform for
a bifacial arrowhead. Drawing by Lotta Sundberg.
materials available in this area that would have been suitable for the
production of bifacial-core artefacts. For instance hlleflinta is likely to
work fairly well, but for some reason the population of eastern central
Sweden preferred to make their arrowheads of flint from southern Scandinavia and thereby also relied on a simpler production method. In this
way, values and ideas from the southern parts of Scandinavia were
enhanced and expressed in contrast to those of southern Norrland and
Dalarna. This is a pattern that is also reflected in the choice of gallerygraves in which to bury the dead and in the architecture of the settlements.
Local raw materials were used to produce bifacial arrowheads in northern
Sweden and Dalarna (personal observations at the National Museum in
Stockholm, the Gvleborgs County Museum in Gvle and the
Jmtlands County Museum in stersund). A glance through
Lannerbros catalogue of archaeological materials from Dalarna reveals
only two arrowheads in southern Scandinavian flint, while there are
numbers of arrowheads made of local, plastic, rock types (Lannerbro
1991 & 1992). On the sites discussed in this paper, only one bifacial
arrowhead not made in flint was recovered: an arrowhead of blybergs
219
Jan Apel
Jan Apel
probably hard to come by in eastern central Sweden. The bifacial-thinning
flakes stem from the middle and late stages in the production of large
bifaces, such as daggers and sickles, and they had to be imported into
eastern central Sweden. It is therefore likely that they were considered
to be valuable and even precious. Large, bifacial-thinning flakes with
initial pressure scars around the margins found in Late Neolithic gallerygraves, such as those at Annelund and Dragby, suggest that this was the
case. I have personally noticed, during visits in the Archaeological
Department of the Gothenburg City Museum (GAM) and the Lund
University Historical Museum (LUHM), that large, bifacial-thinning
flakes, often with traces of some preliminary shaping in the form of
pressure-flake scars, are common as grave goods in Gallery-Graves in
the western and southern parts of Sweden as well. However, they have
generally been classified as retouched flakes or scrapers. In this context,
it may be appropriate to mention a deposit from Fjlkestad in Skne,
containing 16 flint artefacts and 8 of these were preforms for heartshaped arrowheads (Malmer 1955:6 ff., Karsten 1994:95). On account
of contextual evidence, this deposition was interpreted as an offering.
This interpretation is further strengthened by the fact that at least two
preforms for bifacial arrowheads have been found in wetlands in Skne.
According to Karsten, the fact that arrowhead preforms have been
recovered in wetland contexts may indicate that arrowheads in general
that are found in wetlands can be regarded as offerings and not as arrowheads lost during hunting trips (Karsten 1994:95).
So one way of producing arrowheads in eastern central Sweden was
to import suitable blanks from the southern parts of Scandinavia. However, simpler flake types used for the same purpose were produced locally.
These flakes had to be relatively wide and thin in order to serve as
blanks. To achieve these formal criteria, the flakes had to be removed
from slightly convex, almost flat surfaces without pronounced ridges.
One way of making such flakes is to use the so-called combewa-method
in which flakes are removed from the ventral surfaces of larger flakes.
Combewa flakes tend to be wide and thin just because they are detached
from slightly convex and flat surfaces. In Sweden, this strategy was first
observed on the Middle Neolithic site at Kusmark in the province of
Vsterbotten (K. Knutsson 1986). At Kusmark, the method was used
to produce blanks for transverse arrowheads in flint. Since this is a simple
but effective method, it should theoretically have been suitable for areas
222
Fig. 7:7. The distribution of flint artefacts at the Lugnet site in Uppland
(from Lekberg & Apel, ms).
Jan Apel
high-speed fractures. It is obvious that the local methods used to produce
blanks for arrowheads also produced large amounts of debitage.
The shaping of the blank into an arrowhead was performed in the
simplest way. A pressure flaker, presumably with a handle of wood and
a tip of antler, bone, copper or flint, was used to press small, nail-shaped
flakes (sub-category 5-flakes) from the edges on both sides of the blank.
If a bifacial-thinning flake was used as a blank, the tip of the arrowhead
often coincided with the platform remnant of the flake, in order to
take advantage of the natural shape of the flake. The majority of the
bifacial arrowheads from eastern central Sweden exhibit the original
surface of the flake in their central parts on both sides (see the arrowheads in Fig. 7:2). The pressure flaking was never used to thin the arrowhead down by detaching flakes from two sides that overlap at the
centre a procedure that is common in the production of similar
arrowheads in Norrland. In eastern central Sweden, the pressure technique was used only to shape the arrowhead according to formal conventions. Experimental work revealed that all the stages in the production of bifacial arrowheads belong to the lowest grade of difficulty
involved in the production of daggers (Callahan, personal information).
The most difficult stage of the production was to get hold of or to
make blanks.
Jan Apel
flake was recovered at stertlje and several such flakes were found at
Skillsta (Ekmyr-Westman 1991:18). It has been suggested that the
production of square-shaped axes in flint diminished at the transition
between LN I and II (1950 cal BC), when a local production of flanged
bronze axes was introduced into southern Scandinavia (Vandkilde
1996:270 ff.). Thus, it is possible that sites with axe flakes in eastern
central Sweden should be dated to LN I (2350-1950 cal BC).
In contrast to the situation in Denmark, flint flakes were fairly
common as grave gifts in Sweden during the Late Neolithic period
(Spetz 1994:45 ff.). The frequency of flint flakes with retouched edges
in Late Neolithic graves, in relation to all flint artefacts deposited in
graves, tends to increase along the south-north axis. A study of unretouched flakes in Late Neolithic graves reveals two distinct areas. In
the western area, consisting of Halland, Smland, Vstergtland,
Bohusln and Dalsland, 59% of the flint objects are un-retouched flakes;
in the eastern area, Nrke, stergtland, land, Gotland and Blekinge,
only 34% of the flint objects are un-retouched flakes (Spetz 1994:49).
It is likely that the larger amount of un-retouched flakes in the graves of
western Sweden is related to the presence of the hoarded flakes in the
same area.
According to Ebbesen, the flakes that were deposited as offerings
stemmed from the reduction of a core that might have been reduced on
the site. However, I do not agree with this conclusion. It is my impression that the Norwegian hoards, for instance, the one at Hauske on
Karmya in Rogaland (inv. no. S. 3 513, Archaeological Museum in
Stavanger), consist of flakes selected on the grounds of their formal
qualities and not all the flakes from one or several reduction sequences.
The flakes may in some cases come from a single reduction sequence,
but I maintain that they do not originate from platform cores. In my
opinion, the flakes are debitage from the production of large and
regularly shaped, core artefacts, such as daggers and sickles and in some
cases square-shaped axes. Recently, similar flakes were reported to have
been found in the Oslo area in Norway as well (Berg 1998).
I have previously suggested that the production of flint daggers in
southern Scandinavia was organized in regional apprenticeship systems
(see Chapter 2 and Apel 2000a & b), i.e. kinship-based systems, in
which the craft was passed on from one generation to the next. This
reasoning implies that the division of labour was institutionalised during
226
Jan Apel
duction of bifacial arrowheads and daggers for that matter, for instance,
the Myrhj site in Jutland, C14-dated to 2350 cal BC (Jensen 1973).
Bifacial arrowheads in flint or in other plastic rock types have also
been recovered in graves all over Europe from the Middle Neolithic
period up to the Bronze Age, for instance at Mycanae, where 29 and 17
bifacial arrowheads of flint and obsidian were found in two of the grave
shafts (Mylonas 1973: Plate 72, 123). Since these graves are dated to
the interval 1700-1500 BC, they are contemporary with the first period
of the Scandinavian Bronze Age.
Another parallel to the treatment of disc-shaped, flint flakes in eastern
central Sweden during the Neolithic and Early Bronze Age has been
presented by Helena Knutsson (see Chapter 3 and H. Knutsson 1996,
1999 & 2000). She has been studying a certain kind of flint blades that
are recovered from Battle-Axe graves in Scandinavia. She relates these
findings to Haydens theory of prestige technologies (Hayden 1995:68;
Olausson 1998:90). In this context, a prestige technology does not, or
not only, have a practical function, but is also used in power games by
aggrandizers in order to achieve and maintain their social positions
through their connections with the blades and their makers (H. Knutsson
2000). This type of technology can be traced over large regions encompassing areas in northern, central and southern Europe, as well as the
western parts of Asia (Knutsson 2000). It is likely that the production
and distribution of disc-shaped flakes were organized in a smaller but
similar fashion.
Conclusions
Large flint artefacts, such as flint daggers, sickles and axes, were imported from areas with natural flint resources, but they were re-sharpened in eastern central Sweden. The flint-knapping on the sites studied
in this paper was directed to the sharpening of large imported, edged
tools but also to the production of bifacial arrowheads. Thin, discshaped flakes were used as blanks and they were either imported or
made locally. The ideal blanks for the production of bifacial arrowheads were bifacial-thinning flakes and such flakes were subsequently
imported into eastern central Sweden, together with large artefacts. The
imported and locally produced blanks were then shaped into arrowheads by a simple, pressure technique. Disc-shaped flakes of a poorer
quality were produced locally by simpler methods.
228
229
Chapter 8
Jan Apel
1996:17). This means that they are partly discovered and partly invented
by the archaeologist (Willey & Phillips 1958:13). The following
quotation clarifies this view coherently: We can say as a general rule
that nature (or culture) creates modalities, but we, the typologists, choose
among them those that are informative for any particular purpose of
our own (Adams & Adams 1991:68). According to William and Ernest
Adams, it is the combinations of natural and artificial boundaries that
have made it difficult to end this debate (Adams & Adams 1991:68). It
seems as if we shall have to accept that each nomothetic typology in
itself is a combination of natural (or given) and artificial aspects (Grslund
1996:16 ff.). In this respect, types can be regarded as partly essential
and partly instrumental, since they are determined partly by what they
are and partly by what they do (Sokal 1977:187 ff.). This conclusion,
however, does not justify the fact that types sometimes are poorly defined in the archaeological literature (Malmer 1963).
As regards the Scandinavian flint-dagger types, it is apparent that
they were discovered rather than constructed. The boundaries between
the different main types were so sharp and intuitively obvious that some
of them may very well be a priori evidence of a prehistoric reality. It is
also worth considering the fact that the dagger as an object is still an
institutional fact today. We shall therefore recognise a dagger when we
see one if it fulfils the dagger norm. Ironically, since traditional
archaeology was concerned mainly with the chronological aspects, the
obvious patterning that the different dagger types displayed was
interpreted almost exclusively in chronological terms rather than as the
result of, for instance, different geographical distributions or as reflecting
different social strata. Ebbe Lomborgs thorough typological treatment
of the Danish dagger material was, in many respects, instrumental and
to some parts artificial in its attempt to formalise the dagger typology
to meet the typological standards of the 1970s. As we shall see in the
definitions of the different subtypes, Lomborg used the technical aspects,
the measurements and the design to formalise the typology. Through
classifying Danish and Norwegian flint daggers, I have been able to
compare my classifications with Lomborgs (1973) and Scheens (1979)
classifications and I did not encounter any important discrepancies in
our classifications. From this it may be concluded that the type
descriptions presented by Lomborg are sufficient.
The fact that the main, Scandinavian, flint-dagger types, with minor
232
Jan Apel
Fig. 8:1. Lomborgs six main dagger types (assembled from various pictures in
Lomborg 1973).
234
two cutting edges. Considering the first type, however, this definition
has to be complemented, since the majority of the type-I daggers do
not display pronounced handles (see below).
Type I
Lancet-shaped daggers without handles, or with marginally marked
handles that display a lenticular cross-section (Fig. 8:1). There is no
difference in thickness between blade and handle. The type is distinguished from other oblong and thin, bifacial objects, such as lance- and
spearheads, by the fact that the edges of the handle taper towards the
base and that the base tends to be convex or sometimes even pointed.
The edges of spearheads, for instance, do not taper towards the base;
instead they tend to have straight or concave bases (see Mller 1902,
Lomborg 1973).
Type I A
Percussion-flaked, type-I daggers with their greatest width nearer the
tip than the middle of the dagger (Fig. 8:2). The blades of these daggers
tend to appear as very wide, an impression that is enhanced by the fact
that the widest part of the dagger is found near the tip. In Denmark,
this subtype is most common around the Limfjord in northern Jutland
(Lomborg 1973:36, Fig. 12).
235
Jan Apel
Type I B
Percussion-flaked, type-I daggers with their greatest width at the middle or below the middle of the dagger, but always nearer the middle
than the tip (Fig. 8:2). In Denmark, this subtype, like the previous
type, is most common in the Limfjord area in northern Jutland
(Lomborg 1973:36, Fig. 13). In cases in which the blade has been extensively re-sharpened, in a way that has distorted its original outline,
the two previous subtypes cannot be distinguished from each other.
Daggers of this kind are classified as type I A/B.
Type I C
Parallel-flaked, type-I daggers with symmetrically curved and concaveshaped, handle edges (Fig. 8:2). In Denmark, this subtype is almost
exclusively concentrated to the Limfjord area in northern Jutland
(Lomborg 1973:37, Fig. 14). To a greater extent than the other type-Idagger subtypes, it is common in LN I graves in this area (Rasmussen
1990:34 ff.), and the type has been connected with the west European
Beaker influences that affect this area during the first half of LN I
(Vandkilde 1996:281).
Type I D
Type-I daggers with concave-shaped, handle sides; a trait that enhances
the impression of these daggers as having pronounced handles (Fig. 8:2).
This type is reminiscent of type I B but has a more pronounced handle
than the other type-I daggers. On certain examples, the handles tend to
be somewhat thicker than the blades and, according to Lomborg, the
transition from type I D to type II A is indeed continuous (Lomborg
1973:41, 79). In Denmark, the hoards, as well as single finds containing
this type, are concentrated to eastern Danish islands (Lomborg 1973:37,
Fig. 15).
Type I E
A type-I dagger with systematically flaked notches on the edges at the
transition between blade and handle. This is a problematic type, since
the notches were most likely created in order to keep the shaft or handle in place. Subsequently, the retouched notches may not have had
anything to do with the production of the dagger but may have been
the solution of a practical problem that occurred at a later stage when
the handle was being adjusted to the dagger base. This is also indicated
236
Type I x
Extensively re-sharpened, type-I daggers that, on this account, cannot
be assigned to any of the subtypes defined above. Curiously, Lomborg
includes dagger preforms in this subtype as well, provided that they can
be distinguished as type-I preforms (Lomborg 1973:44). In my classification of the Swedish daggers that will be presented in the next chapter, I have not included preforms in this category. Preforms are instead
discussed as a separate category.
Type II
Daggers with lancet-shaped blades and pronounced handles that are at
least 2 millimetres thicker than the blades. The handles display oval or
lenticular cross-sections (Fig. 8:1). This type is divided into two subtypes,
defined according to the width/thickness relationship of the handle. In
fact, as we shall see, it is only this relationship that so far makes it
possible to differentiate between daggers of types I D, II A and B, and
III A. I have usually defined these types impressionistically by sight,
which means that there may be some in-consistencies involved in the
classification. However, as we shall see below, in many of the analyses
that will be made in Chapter 9, these types, together with the other
type-III daggers, are regarded as a homogenous group and will in most
cases be merged. This is due to the fact that I regard the boundaries
between these types as a construction by the archaeologist and not relevant to any prehistoric norm, just as we have seen that Lomborg himself did (Lomborg 1973: 44 ff.). These daggers represent one and the
same regional and temporal tradition that is rooted on the eastern Danish
islands and in the south-western parts of Skne in Sweden. Arguments
in support of this opinion will be presented later on in this chapter.
Type II A
Type-II daggers with handles that display a width/thickness relationship of between 2:1 and 3:2 (Fig. 8.3).
Type II B
Type-II daggers with handles that display a width/thickness relationship of between 3:2 and 1:1 (Fig. 8:3).
237
Jan Apel
Type III
Daggers with small handles with parallel sides, rhombic, roundedrhombic or rectangular cross-sections and evenly shaped outlines (Fig.
8:1). This type, like the earlier types, has lancet-shaped blades. With the
notable exception of subtype A, type-III daggers regularly exhibit pressure-flaked seams on the edges and sides of the handles. The majority
238
Type III A
Type-III daggers whose handles have a width/thickness relationship of
1:1 and square cross-sections (Fig. 8:4). This subtype does not have
pressure-flaked seams on either side of the handle. Occasionally, though,
a pressure-flaked seam is visible on the handle edges.
Type III B
Type-III daggers with handles that have a width/thickness relationship
of 1:1 and square cross-sections (Fig. 8:4). This subtype has pressureflaked seams on both sides, as well as on the edges of the handles.
Type III C
Type-III daggers with handles that have a width/thickness relationship
of 1:1 and square cross-sections (Fig. 8:4). This subtype has pressureflaked seams on one side, as well as on the edges of the handle.
Type III D
Type-III daggers whose handles have a width/thickness relationship of
1:1 and square cross-sections (Fig. 8:4). However, in comparison with
the other type-III daggers, the handle is turned through 45 degrees, so
that two sides of the squared handle are parallel to the blade. This subtype
has four pressure-flaked seams. Only three examples of this subtype are
known in Scandinavia (Montelius 1917:421, Mller 1902:129,
Lomborg 1973:49).
Type III E
Type-III daggers with pressure-flaked seams on both sides, as well as on
the edges of the handles (Fig. 8:4). The width of the handles is always
greater than the thickness and, consequently, they display rectangular
cross-sections.
239
Jan Apel
240
Type III F
Type III daggers with pressure-flaked seams on one side and on both
edges of the handle. The width of the handles is always greater than
their thickness and they have rectangular cross-sections.
Type IV
Daggers with fishtail-shaped handles with rhombic, rounded, triangular or rectangular, cross-sections (Fig. 8:1). All the subtypes, except F,
display pressure-flaked seams in the middle of the handles on at least
one side (Fig. 8:1). Subtypes A and B have blades that are almost lancet-shaped but are often slightly wider nearer the blade-handle transition than in the middle of the blade. Subtypes D-F display blades with
a rounded, triangular outline with their greatest width nearer the bladehandle transition than the blade (Fig. 8:5). In Denmark, the geographical distribution of the type-IV daggers varies between the subtypes.
However, the four known hoards containing this dagger type were found
on the Danish islands, three of them on Fyn (Lomborg 1973:55, Figs.
30-34). This may indicate that the type has predominantly, an eastern
Danish and south-western Swedish origin. A microscopic investigation
of the pressure-flaked seams of nine type-IV daggers revealed traces of
what appeared to be metal on the stitched seam (Stafford 1998:342).
This may mean that copper-tipped pressure flakers were needed in order
to make these small and elaborate seams.
Most of the parallel-flaked blades on the type-IV daggers were
ground before the final pressure-flaking commenced. However, there
are examples of large, parallel-flaked, type-IV daggers that were not, for
example two daggers from Igellsa parish in the province of Smland
(Fig. 8:6) (Hofrn 1925:32 f., Gurstad-Nilsson 1999:191). These were
found together in a square-shaped, small, construction made of sandstone slabs that had been dug into the clay.
Type IV A
Type-IV daggers with handles that are thick and display rhombic crosssections and have straight edges that broaden evenly towards the base
(Fig. 8:5). The handles thicken towards the base and display pressureflaked seams on the back and front sides, as well as on the edges. The
type can be distinguished from type III E only in that the handles become wider towards the base. The blades are lancet-shaped, although
they tend to be widest near the blade-handle transition. In Denmark,
241
Jan Apel
242
Fig. 8:6. Two type-IV flint daggers (32 and 30 cm long) from Igellsa parish in
the province of Smland. These daggers were not grinded before parallel
flaked (photo: the Kalmar County Museum).
243
Jan Apel
their distribution is concentrated to the eastern Danish islands, mainly
to Sjlland, and the sole known Danish hoard containing daggers of
this type was also recovered here (Lomborg 1973:54, Fig. 30).
Type IV B
Type-IV daggers that have thick handles with rhombic cross-sections
and concave edges that broaden towards the base (Fig. 8:5). The handles thicken towards the base and display pressure-flaked seams on the
back and front sides, as well as on the edges. The blades are lancetshaped, although they tend to be widest near the blade-handle transition. In Denmark, this type is evenly distributed over the northern and
eastern parts of Jutland, as well as over the eastern Danish islands
(Lomborg 1973: 54, fig. 31).
Type IV C
Type-IV daggers with pressure-flaked seams on the back and front of
the handle (Fig. 8:5). The handles have flat, rhombic cross-sections and,
in contrast to the two preceding types, they are evenly thick. The blades
are relatively wider than the blades of the two preceding types. The
blades in type IV C (as well as in other, later, dagger types) usually
display a rounded, triangular outline, widest near the blade-handle transition. In Denmark, the daggers of this type found singly seem to be
evenly distributed over the northern and western parts of Jutland, as
well as over the islands. However, the two hoards containing this type
were recovered on Fyn and Sjlland respectively (Lomborg 1973:55,
Fig. 32).
Type IV D
Type-IV daggers with pressure-flaked seams on the front, but not on
the back, of the handles (Fig. 8:5). The back of the handle and its base
are slightly curved. The blade on these daggers is usually rounded, triangular, with the widest part near the blade-handle transition. In Denmark, this type is concentrated mainly to the southern parts of Jutland
and to the island of Fyn (Lomborg 1973:55, Fig. 33).
Type IV E
Type-IV daggers with pressure-flaked seams on the front, but not on
the back, of the handles (Fig. 8:5). The back of the handle is flat or
rounded and the back of the base is straight. The blades usually have a
triangular outline, with the edges rounded, and they are widest near the
244
blade-handle transition. In Denmark, this type is common in the southwestern parts of Jutland, and, to some degree, on Fyn (Lomborg
1973:57, Fig. 34).
Type IV F
Type-IV daggers with handles that have rectangular cross-sections with
the broad sides positioned parallel to the blade (Fig. 8:5). Two pressureflaked seams are positioned on the edges of the handle. However, no
seam is present in the middle of the blade. This type is very unusual;
Lomborg lists only five examples from Denmark and only two of them
have information on provenience (Lomborg 1973:58).
Type V
Daggers with handles that widen at the base and display lenticular crosssections (Fig. 8:1). The handles do not have pressure-flaked seams on
either side but may display seams on the edges and on the base. In
Denmark, the type-V daggers are evenly distributed and, of the six
known hoards, three were recovered in the northern parts of Jutland
and three on the eastern Danish islands one on Fyn and two on Sjlland
(Lomborg 1973:59, Figs. 36 & 37). The type is divided into two
subtypes according to the form of the base of the handles.
245
Jan Apel
246
Type V A
Type-V daggers with straight or curved bases similar to the bases of the
type-III and IV daggers (Fig. 8:7).
Type V B
Type-V daggers on which pointed ridges run across the base (Fig 8:7).
Type VI
Daggers with handles that have straight, parallel-sided edges (Fig. 8:7).
The handles display lenticular or oval cross-sections and are often unevenly shaped. However, the base is never the widest part of the handle.
These daggers, according to Lomborgs definition, always have
percussion-finished blades. In Denmark, the type-VI daggers found singly are evenly distributed. Of the five hoards containing daggers of typeVI A, two were recovered in Jutland, north of the Limfjord, one on
Fyn and two on Sjlland. The two hoards containing type VI B daggers
were both found in Jutland (Lomborg 1973).
247
Jan Apel
Type VI A
Type-VI daggers with handles that are thicker than the blades (Fig. 8:8).
Type VI B
Type-VI daggers with handles that are not thicker than the blades (Fig.
8:8).
Type VI C
Type-VI daggers with details that have obviously been copied from
daggers and swords in metal (Fig. 8:8).
Jan Apel
250
251
Jan Apel
Fig. 8:11. Mller and Forssanders flint-dagger typologies (from Becker 1964).
Jan Apel
the later dagger types (i.e. those with a pronounced handle) were exclusive to Scandinavia, while the earlier types were common European
forms (Mller 1902:131). We can assume that in this context he was
referring to the resemblance with daggers made of the French honey
coloured Grand-Pressigny flint found in Bell Beaker graves in Western
Europe and flint daggers made on the British Isles. Mller also points
out that the positioning of the earlier dagger types in Late Neolithic
graves confirms the dagger interpretation; they are often found beside
the hip as if they had been attached to a belt. Some 30 years later Mllers
idea was confirmed when a Type-I dagger in a leather sheath and belt
was found in a bog in Wiepenkathen near Hamburg in Germany (Cassau
1935). A wooden handle covered the base of the dagger and, in order to
make it fit, a piece of cloth had been inserted between the handle and
the base. This little cloth, by the way, is the earliest evidence of wool in
northern Europe (Jrgensen 1992). Microscopic use-wear investigations
of flint daggers from late Bell Beaker and Beaker contexts in Britain
have strengthened the dagger interpretation further. These daggers have
revealed a patina created by the fact that a form of handle had covered
the base and that the daggers were kept in leather sheaths (Green et al.
1982, Grace 1990:11). According to Green et al. (1982:495), the
repeated sheathing and unsheathing of a flint dagger would tend to
remove any micro-wear polishes left by daily use, simply because dry
hide/leather polish is a very abrasive type of micro-wear polish.
Jan Apel
Fig. 8:12. Photographs showing the Skogsbo gallery grave during excavation in
1913 (Photos: ATA, Stockholm).
257
Jan Apel
Bronze Age. He presents descriptions of graves from northern Germany
containing type-VI daggers that confirm the dating of this type to the
Early Bronze Age (Forssander 1936:127 f.). His sole argument for the
chronological order of the other types is based on stratigraphic evidence
from a Late Neolithic Gallery-Grave at Skogsbo in the province of
Vstergtland (Fig. 8:12) (Forssander 1936:121). It was excavated by
Sune Lindqvist in 1913 and contained 17 flint daggers of different types.
Later, the typology was criticised by archaeologists (Strmberg 1975).
It is obvious that those who had the greatest problem with the practical
application of the typology were working with materials recovered in
areas far from the flint source. The typology was regarded as less useful
in these areas, because the flint daggers that are found often show signs
of extensive re-sharpening. In order to test the typology on an
archaeological material, Lili Kaelas classified flint daggers from two
Swedish provinces, Vstergtland and Bohusln, kept in the collection
of the Archaeological Museum in Gothenburg (GAM). She concluded
that only 28% of the daggers from Bohusln and 38% of the daggers
from Vstergtland could be classified according to Forssanders typology (Kaelas 1964:135). Unfortunately, this statement is rather difficult
to judge, since all daggers with extensively re-sharpened blades were
sorted out, because they were regarded as impossible to classify. This is
rather curious, since, in most cases, there is sufficient evidence to classify
the dagger type by handle characteristics only. In 1975, Ebbe Lomborg
stated that he had no problem in classifying daggers that had been found
far away from the production area according to his revised version of
the Mller-Forssander typology (Lomborg 1975:116). My personal
opinion is that the typology can be used on daggers from northern and
central Sweden without any major problems, as long as the shape of the
handle is allowed to guide the classification. However, a more serious
problem with Kaelas investigation is that she used a metrical definition
of type-I daggers. Only those examples with lengths over 18 cm were
considered to be daggers, while shorter examples were regarded as arrowand spearheads (1964:145, note 16). This mixture of type definition,
on the one hand, and her personal interpretation of what the object was
actually used for during prehistory, on the other, confuses the
argumentation.
In her article, Kaelas also comments on the fact that Forssanders
idea of the chronological significance of his typology rests solely on his
258
Jan Apel
examples that are 13 cm or longer are regarded as daggers (Lomborg
1973:27 f.). This is a very unfortunate definition, since daggers that are
found far from the production area have more frequently been resharpened and consequently tend to be shorter (Scheen 1979:58). In
cases like these, it is important to classify the objects according to context
and/or wear patterns. If artificial boundaries like these are used, the
defined types will risk reflecting only groups of more or less used daggers.
Lomborg was of the opinion that the six main types formed a chronological sequence. Thus, types I-V represented five, separate, chronological phases of the Late Neolithic, and type VI was still fixed firmly in the
Early Bronze Age. However, according to Lomborg, it would not be
meaningful to divide the Late Neolithic period into five phases, because
there were no other Late Neolithic artefacts that could be classified in a
way that would support such a detailed chronology. Instead, he chose
to divide the Late Neolithic into three phases designated LN A, LN B
and LN C. Daggers of type I define LN A, types II and III define LN B,
and types IV and V define LN C (Lomborg 1973:158). Evidence, in
the form of stratigraphic observations and find-combination data, was
presented to support this tripartition. These studies give, according to
Lomborg, ein berraschend eindeutiges bild. The diagram in Fig. 8:13
shows that none of the main types has been found underneath an older
type. However, it is worth mentioning that 8 of the 67 examples in the
diagram are the results of what Lomborg labels horizontal stratigraphy;
i.e. when two or more daggers have been found in a passage grave, for
instance, those nearest the entrance are considered to be later than those
in the chamber. It is unwise to use these observations in the way that
Lomborg does and he subsequently received criticism for this reasoning.
There are no facts that justify the claim that a dagger found near the
entrance belong to a later burial. On the contrary, it could be argued
that these daggers were older and that they were cleared out of the chamber
in order to make room for later burials. The diagram in Fig. 8:14 shows
that (almost) all daggers found in find combinations are combined either with the same type or with the nearest type (Lomborg 1973:67).
The two cases that do not fit the picture are explained away as having
been recovered during earlier, poorly conducted excavations and therefore not safe to trust. In this context, one must ask whether Lomborg
was suspicious also of old excavation reports in which his chronology
was confirmed?
260
Jan Apel
The typology has been used to classify Scandinavian flint daggers all
over northern Europe (see Chapter 9). During the 1970s and 1980s,
Kaelas initial critique of Forssanders typology was complemented with
a critique of Lomborgs refined typology. As we have seen earlier,
Lomborg does not trust a purely typological reasoning to sustain his
chronological results and therefore tries to strengthen his typology with
reference to find-combination data and stratigraphic observations. A
great deal of the criticism that has been published has concentrated on
this issue.
Jan Apel
technology would, in itself, have created new technological opportunities that opened doors to a greater exploitation of natural resources. In
this context, Ebbesen pointed out that the most important tools to
study would be flint sickles and other practical tools and not flint daggers. If I have understood Ebbesen correctly, he maintained that the
bifacial sickles had technological advantages that in some way would
improve, for instance, harvest capacity or the collecting of leaves as
animal fodder. Perhaps one should look here for the causes of the sudden
disappearance of the old cultures, just as this circumstance may be behind
the enormous economic growth in population in the heart of South
Scandinavia together with a great expansion to the north (Ebbesen
1975:111). In my opinion this is an interesting idea but it needs to be
confirmed with experiments.
Lomborgs reply regarding the inconsistent typology was that he
had just followed old conventions regarding typology and that it is
perfectly clear from the type descriptions whether a type is regarded as
reflecting a chronological sequence or regional patterning (Lomborg
1975). When it came to the question of the regional differences in the
material, Lomborg avoided the principal discussion by referring to his
distribution maps.
Jan Apel
Fig. 8.15. Daggers and spearheads in flint from the Skogsbo gallery grave
(From Forssander 1936: Taf. XXIII-XXIV).
266
Jan Apel
ing closed hoards in seriation tables based on find combinations. If the
hoards represent workshop stocks or, for that matter, offerings deposited near the workshop, their distribution would reflect the production
of a single workshop. The result of this is that the hoard content will be
homogenous and the chronological differences, compared with other
dagger hoards, may be greater than expected (Madsen 1978:57).
Lisbeth Wincentz Rasmussen draws the obvious conclusions from
Madsens revised dagger chronology (Rasmussen 1990). She presents a
new division of the Late Neolithic period into an earlier and a later
phase, instead of Lomborgs and Madsens tripartition. In this new
division, daggers of types I and II represent the early phase and daggers
of types III-V the later phase. Daggers of type I A-C are assigned to a
delimited period in northern Jutland when this part of Scandinavia was
influenced by the western European, Bell-Beaker and Beaker traditions.
The area around the Limfjord was, according to Rasmussen, the centre
of the dagger production during the early phase. Daggers of types I D
and II are seen as local variants from the eastern Danish islands. This
area, in turn, shows signs of contact with the netice area during the
early phase (Rasmussen 1990:38).
The Beaker influences can be noted on the sites of the period in the
form of typical pottery from settlements and in graves in north-western Jutland. There are, for example, seven graves in which a type-I dagger
has been found with Bell-Beaker pottery (Rasmussen 1990:41, note
16). There are several graves in this area in which type-I daggers are
combined with typical, Bell-Beaker, artefact categories, such as V-incised,
conically-shaped buttons in amber and heart-shaped, bifacial arrowheads
in flint (Rasmussen 1990:35). A fragment of a wrist guard in slate has
also been found in the early Late Neolithic settlement of Myrhj in
northern Jutland (see Jensen 1973 and Chapter 6 in this book). This
artefact category is typical of the western European Bell-Beaker tradition
(Rasmussen 1990:37). Fragments of type-I daggers were also found on
the site (Jensen 1973:88) and a considerable proportion of the flint
debitage suggests that the later stages dagger-production was actually
conducted on the site (Chapter 6). Rasmussen concludes that the
population on the eastern Danish islands, where daggers of types I D, II
and III were made, had contacts with the netice area in central Europe.
This is due to the fact that metal prototypes of needles and other artefacts in bone found in eastern Danish graves were made in the netice
268
Fig. 8:16. (a) The relative chronology of the flint-dagger types, based on the
numbers of daggers, the regional differences and the find combinations. (b)
The Late Neolithic phases in accordance with a. (c) The Late Neolithic phases
in accordance with Lomborg (1975) (from Madsen 1978).
269
Jan Apel
area (Rasmussen 1990:37). According to Rasmussen, the regional differences between western and eastern Denmark ceased when the production of type-III daggers started. This conclusion, as we have seen
earlier, was not drawn on account of any major changes in the distribution of hoards, since the type-III hoards, like the type-II hoards, are
concentrated to the eastern Danish islands (Lomborg 1973:50 f., Figs.
25-28). Instead, Rasmussen argues that the general distribution of daggers
implies that they were more widely consumed. At the transition
between the early and the late period, the core area of the flint-dagger
production was moved from northern Jutland to the islands. This state
of affairs was strengthened when the production of dagger with fishtail
handles commenced (Rasmussen 1990:37).
In the late 1980s, Helle Vandkilde presented a division of the Late
Neolithic into two phases that, in some minor but important details,
differ from Rasmussens (Vandkilde 1989 & 1996). Here, Lomborgs
SN A and SN B are incorporated into an earlier Late Neolithic phase
(LN I) while Lomborgs LN C (that corresponds chronologically to
the classical phase of the netice Culture) is renamed LN II. Thus,
Vandkilde, as opposed to Rasmussen, consider that the daggers of type
III belong to the earlier phase. In the previous section of this chapter,
the argumentation behind this decision has been explicitly described.
Nevertheless, it may also be appropriate to mention that Vandkildes
division does not proceed primarily from the dagger typology. It is
based on a thorough, typological investigation of the metal hoards from
the Late Neolithic and the Early Bronze Age in Scandinavia (Vandkilde
1996). Vandkilde presents a typology of the bronze axes in the hoards
and is able to define two phases in this material. A seriation diagram
based on find combinations of different artefact categories shows that
an early metal horizon coincided with the later Late Neolithic period,
LN II, and that two metal horizons coincided with the first two periods of the Early Bronze Age. These results are formalised by the application of a statistical method (correspondence analysis) that confirms
the seriation. The development through time in the chemical composition of the early metal objects and their alloying techniques also confirms
the chronology (Vandkilde 1996:136 ff.). Vandkildes results, as regards the production of flint daggers, are presented in Fig. 8:17. The
local production of metal artefacts replaced the production of flint daggers during the latter part of the Late Neolithic. The technologically
270
advanced type-IV daggers may, seen from this angle, have been the result
of a last effort by the flint knappers to challenge the new metal
technology.
An alternative interpretation of the flint dagger has been presented
by Gundela Lindman (1987) and has been further developed by Eva
Weiler (1994). On the basis of Kaelas (1964) and Segerbergs critique
of the dagger typology, they suggest that it lacks chronological value
(Lindman 1987:128). According to Lindman and Weiler, the typology
primarily reflects social hierarchies in the Late Neolithic population.
Lindman claims that the type-IV daggers form one group and all the
other dagger types form another group. It is further stated that the
type-IV daggers signified high status and therefore belonged to chiefs.
The other daggers are, on the other hand, regarded as common types.
Lindman admits that, although type-IV daggers signified high status,
there are indications that this type was produced during a relatively
short period. However, Lindman avoids this possibility in the following paragraph: But this reasoning creates problems. First one has to
explain why the leaders all appeared during a very short time period in
the later part of Late Neolithic. Where did they dwell during the other
271
Jan Apel
parts of the period? (Lindman 1987:133 f.). The chronological discussion, earlier in this chapter, indicate that it might pose great sourcecritical problems to regard daggers from LN I and II as contemporary
types in social analyses. However, these ideas may be worth discussing
in areas far away from the natural flint sources, where older types might
have been in circulation for greater lengths of time (see Weiler 1994).
Conclusion
The Scandinavian flint daggers appear in distinct, morphological groups.
This fact has tempted archaeologists, since the 19th century, to formulate flint typologies that would reflect a chronological sequence. During the last two decades, a revision of the detailed chronological results
that Lomborg presented in 1973 has been conducted. The chronological relevance of the flint-dagger typology has received critique from
two directions. Some scholars have argued that the closed-find contexts
and stratigraphic observations that make up the empirical basis of
Lomborgs chronology can be questioned (Segerberg 1978). Likewise,
it has been maintained that Mller and Forssanders typology was difficult to use on materials far from the production areas, owing to the
extensive re-sharpening these daggers had been subjected to (Kaelas
1964). Others accept the data but are of the opinion that the typology
does not constitute a chronological sequence in which each type follows the other (Ebbesen 1975, Madsen 1978, Rasmussen 1990). This
critique is based on the regional distribution of hoards containing different dagger types and on the assumption that the dagger hoards reflect the work of local or regional workshops.
In 1989, a scheme (Fig. 8:18) was published that has resolved a
major part of the discussion on the Late Neolithic chronology (Vandkilde
1989). This scheme is based on almost 300 c14-samples from southern
Scandinavia and a central and western Europe. For the first time, relative
chronologies from different areas in this large region can be compared.
The beginning of LN I is set at 2350 BC, the transition to LN II is set
at 1950 BC and the transition to the Early Bronze Age at 1700 BC
(Vandkilde 1989:29). These results make the flint-dagger typology
superfluous as a detailed chronological tool.
If we accept the critique of Lomborgs chronological conclusions, it
seems that the daggers were produced in two core areas during LN I: a
272
western area around the Limfjord in Jutland and an eastern area including the eastern Danish islands and south-western parts of Skne in
Sweden. Type-I daggers, except for subtype I D, were produced in the
western area, and subtype I D, type II and III daggers were produced in
the eastern area (Fig. 8:19)
The Late Neolithic period in Scandinavia was about 650 years long.
Flint daggers were produced for at least 800 years. We know that the
daggers with lancet-shaped blades were produced during LN I and that
the core production area during this phase was the Limfjord area in
northern Jutland. The impulses that triggered the production seem to
have come from western Europe, most probably from the British Isles.
It is likely that the prototypes of the lancet-shaped daggers were tanged
copper daggers found in western European, Bell-Beaker contexts and
British flint daggers found in Beaker contexts contemporary with the
Late Single Grave phase on Jutland. During LN II (250 years), daggers
273
Jan Apel
Fig. 8:19. The distribution of Danish hoards with daggers from LN I. Type-I
daggers, except subtype I D (circles), subtype I D, type II and type III daggers
(triangles), based on information from Lomborg (1973).
275
Chapter 9
Jan Apel
Scandinavian flint-dagger materials of the northern and central parts of
Europe parts of the exchange network are also included in the
study (Clark 1932, Meinander 1954, Bloemers 1969, Wyszomirski
1974, Wojciechowski 1976, Khn 1979, Bantelmann 1982, Agthe
1989a & b, Trnka 1991 & MS, Rassmann 1993, Beuker & Drenth
1995).
Only daggers with at least parish provenience are used in this study.
Presumably, several thousands of daggers have been excluded because
they lack information on where they were found. Not all museum
collections, and no private collections, have been visited during the
registration of the data.1 In all, 4 454 daggers have been registered. This
may seem like a small number, especially since Malmer registered 6 515
flint daggers with known find locations in Skne only (Malmer
1962:699). This is partly explained by the fact that Malmer visited
more museums and private collections than I have done. However, it
may also to some extent be due to the fact that we used different
typological criteria. Malmer did not make a distinction between lanceand spearheads and palstaves in flint, on the one hand, and flint daggers,
on the other (Malmer, personal communication). Even though it is
highly unlikely that these differences make up for the discrepancy in
the numbers of daggers from Skne, they certainly contributed. It is
assumed that the collected material is representative of the flint daggers
that were deposited in prehistory. This is likely, since the discerned
pattern, as regards the frequencies and different types, coincides with
investigations conducted with a smaller material base (see Apel 2000a
& b). Thus, it is anticipated that the present material will be sufficient
to reveal important and general aspects of the production and
consumption of flint daggers in Sweden.
1 Flint daggers from the following Swedish museum collections are included in this study: the
Gvleborg County Museum, the Archaeological section of the Gothenburg City Museum, the
Halland County Museum, the Jmtland County Museum, the Lund University Historical
Museum, Gustavianum Museum at Uppsala University, the National Museum in Stockholm,
the Uppland County Museum and the Vstmanland County museum.
278
Representativity
Archaeological investigations based on stray finds are subject to severe
source-critical problems. However, this should not discourage
archaeologists from working with stray finds. As I see it, those who
choose to do so have two options. Either they may set out to record
every conceivable problem related to the use of stray finds in an
archaeological analysis or they may collect a material that is large enough
to allow of general and large-scale interpretations, in order to limit the
effect of any source-critical problems. To discuss and evaluate every
possible source-critical problem is a never-ending task that, if it is to be
convincingly pursued, will demand at least a book in itself, such as the
books on archaeological-formation processes (Schiffer 1976 & 1987,
Kristiansen 1985). In this context, therefore, the former strategy has
been chosen. It is apparent that the distribution of the Scandinavian
flint daggers, as regards the frequencies of the different types, forms
patterns that cannot be explained by post-depositional, cultural and
non-cultural, formation processes but that reflect the prehistoric reality
(see Apel 2000a & b).
279
Jan Apel
Province
Blekinge
Bohusln
Norrland
Dalsland
Gotland
Halland
Nrke
Skne
Smland
Sdermanland
Uppland
Vrmland
Vstergtland
Vstmanland
land
stergtland
Total
Type I
10
89
9
81
9
71
15
262
11
12
12
1
140
5
6
22
755
Type II
12
60
1
36
8
63
11
372
22
9
7
4
66
6
7
24
708
Type IV
6
72
5
34
14
78
18
193
19
16
14
8
57
8
13
35
590
Unclassified
8
9%
60
14%
8
25%
95
23%
8
10%
70
15%
13
14%
192
12%
16
11%
9
11%
16
18%
7
21%
96
16%
8
16%
4
7%
25
14%
635
14%
Total
94
427
32
408
83
479
91
1594
147
81
90
34
609
51
58
176
4454
Table. 9:1. The Swedish flint-dagger material divided into Lomborgs main types and
registered by province.
281
Jan Apel
1
2-3
4-5
6Fig. 9:2. The geographical distribution
of type-I flint daggers
in Sweden and
Norway. Information
on the Norwegian
daggers has been
extracted from
Scheen (1979). The
daggers are projected
on a map showing
the coastline at 3600
BP (c. 2000 cal BC)
made by Leif
Andersson at the
Swedish Geological
Survey (SGU) in
Uppsala.
282
283
Jan Apel
1
2-3
4-5
Fig. 9:4. The geographical distribution
of subtype-I D flint
daggers in Sweden
and Norway. Information on the
Norwegian daggers
has been extracted
from Scheen (1979).
The daggers are
projected on a map
showing the coastline
at 3600 BP (c. 2000
cal BC) made by Leif
Andersson at the
Swedish Geological
Survey (SGU) in
Uppsala.
284
1
2-3
4-5
6Fig. 9:5. The geographical distribution
of type-II flint daggers
in Sweden and
Norway. Information
on the Norwegian
daggers has been
extracted from
Scheen (1979). The
daggers are projected
on a map showing
the coastline at 3600
BP (c. 2000 cal BC)
made by Leif
Andersson at the
Swedish Geological
Survey (SGU) in
Uppsala.
285
Jan Apel
1
2-3
4-5
6Fig. 9:6. The geographical distribution
of type-III flint
daggers in Sweden
and Norway. Information on the
Norwegian daggers
has been extracted
from Scheen (1979).
The daggers are
projected on a map
showing the coastline
at 3600 BP (c. 2000
cal BC) made by Leif
Andersson at the
Swedish Geological
Survey (SGU) in
Uppsala.
286
1
2-3
4-5
Fig. 9:7. The geographical distribution
of type-IV flint
daggers in Sweden
and Norway. Information on the Norwegian daggers has
been extracted from
Scheen (1979). The
daggers are projected
on a map showing
the coastline at 3600
BP (c. 2000 cal BC)
made by Leif
Andersson at the
Swedish Geological
Survey (SGU) in
Uppsala.
287
Jan Apel
1
2-3
4-5
6Fig. 9:8. The geographical distribution
of type-V flint
daggers in Sweden
and Norway. Information on the
Norwegian daggers
has been extracted
from Scheen (1979).
The daggers are
projected on a map
showing the coastline
at 3600 BP (c. 2000
cal BC) made by Leif
Andersson at the
Swedish Geological
Survey (SGU) in
Uppsala.
288
1
2-3
4-5
6Fig. 9:9. The geographical distribution
of type-VI flint
daggers in Sweden
and Norway. Information on the
Norwegian daggers
has been extracted
from Scheen (1979).
The daggers are
projected on a map
showing the coastline
at 3600 BP (c. 2000
cal BC) made by Leif
Andersson at the
Swedish Geological
Survey (SGU) in
Uppsala.
289
Jan Apel
1
2-3
4-5
6Fig. 9:10. The
geographical distribution of flint-dagger
preforms in Sweden
and Norway. Information on the
Norwegian preforms
has been extracted
from Scheen (1979).
The daggers are
projected on a map
showing the coastline
at 3600 BP (c. 2000
cal BC) made by Leif
Andersson at the
Swedish Geological
Survey (SGU) in
Uppsala.
290
5
4
3
6
7
9
2
1
10
Fig. 9:11. Different regions of Scandinavia discussed in the text. (1) eastern
Denmark, (2) western Denmark, (3) eastern Norway, (4) western Norway, (5)
Norrland, (6) eastern central Sweden, (7) western central Sweden, (8) southeastern Sweden, (9) south-western Sweden and (10) Skne.
291
Jan Apel
share of the daggers found far away from the natural flint sources are
difficult to classify (Table 9:1). This may be explained by the fact that
the flint was more valuable in these areas.
Scandinavia
For practical purposes, Scandinavia will now be divided into regions on
the basis of the homogeneity of the flint-dagger material (Fig. 9:11).
Denmark is divided into an eastern and a western part because of the
two main sources of good-quality flint, regarded as being crucial for the
maintenance of the flint-dagger technology. The following parishes
belong to the eastern region: Bornholm, Copenhagen, Fredriksborg,
Holbk, Maribo, Odense, Prst, Sor and Svendborg and the
following belong to the western region: Haderslev, Hjrring, Randers,
Ribe, Ringkbing, Skanderborg, Snderborg, Thisted, Tnder, Vejle,
Viborg, benr, lborg and rhus. Thus, the eastern area comprises
the Danish islands east of Jutland, while the western region comprises
the Danish part of the Jutlandic peninsula (Fig. 9:11).
Norway has also been divided into an eastern and a western region.
This division is based on obvious differences in the Norwegian
archaeological materials dating from the Late Neolithic that are due to
both cultural and natural circumstances (Apel 2000a & b). The eastern
area consists of the provinces of Akershus, Aust-Agder, Buskerud,
Hedmark and Opland, Telemark, Vestfold and stfold and the western
area consists of Finnmark, Hordaland, Mre og Romsdal, Nordland,
Nord-Trndelag, Rogaland, Sogn og Fjorda, Sr-Trndelag, Troms
and Vest-Agder,
The following division of the southern and central parts of Sweden
is based on the data presented in Figs. 9:2 to 9:10 and in Table 9:1. This
area has been divided into five regions:
Eastern central Sweden, consisting of the provinces of Uppland,
Vstmanland, Nrke, Sdermanland and Vrmland.
Western central Sweden, consisting of the provinces of Bohusln,
Dalsland and Vstergtland,
South-eastern Sweden, consisting of the provinces of Gotland, Smland,
land and stergtland
South-western Sweden, consisting only of the province of Halland.
292
Jan Apel
smaller type-VI daggers may have been of a more general character,
maybe carried on at the household level.
Schleswig-Holstein, Germany
Over 2 000 flint daggers of Scandinavian types have been recovered in
Schleswig-Holstein in Germany (Table 9:2). This area consists of the
southern part of the Jutlandic peninsula and consequently borders on
Denmark. As we saw in Chapter 6, there are known production sites
with defined, bifacial-production debitage in the northern parts of this
area (Arnold 1990). The dagger preforms are concentrated mainly in
the eastern coastal area of Schleswig-Holstein (Khn 1979, Karte 2).
The large amount of preforms found here, in combination with the
fact that production debitage has been tied by refitting to daggerproduction sites, suggests that Schleswig-Holstein should be regarded
as a production rather than as a consumption area. However, this
production was probably not extensive. There are only three areas in
the northern parts of Germany where primary flint resources have been
located: on the islands of Helgoland off Holstein and Rgen off the
coast of the eastern part of Mecklenburg. The ice-transported flint in
northern Germany is not suitable as raw material for large tools and
blades (Hoika 1998:2). There is also a primary flint source further south
at Hemmoor in the Lneburg district of Germany (Beuker & Drenth
1998:8). Therefore we have to assume that the large, prestigious daggers
of Schleswig-Holstein were made of flint taken from the primary flint
sources of northern Jutland or the islands of Falster and Lolland, while
small daggers may have been produced in local flint. The majority of
the preforms that have been assigned to types have incipient handles
(18 examples) while only eight examples of type-I-like, lancet-shaped
preforms are present. Type-I daggers are not represented. The relative
distribution of dagger types (Fig. 9:12, Table 9:2) indicates that, from
this point of view, this area may have had closer contacts with the eastern
production area of southern Scandinavia than with the western
production area of northern Jutland. Khn argues that the flint-dagger
types of Schleswig-Holstein, like those of the Danish dagger material,
form a chronological sequence. However, the two diagrams presented
in favour of this idea do not constitute convincing evidence (Khn
1979:47, Abb. 7 & 8). For instance, of 12 closed-find combinations,
294
Austria
British Isles
Eastern central Sweden
Eastern Denmark
Eastern Norway
Finland
Mecklenburg
The Netherlands
Norrland
North Rhine-Westphalia
Thringen and Saxony
Poland
Schleswig-Holstein
Skne
South-western Sweden
South-eastern Sweden
Western central Sweden
Western Denmark
Western Norway
Total
%
Type I
Type II
Type III
Type IV
Type V
Type VI
Total
0
0
45
503
144
2
619
28
9
7
130
22
373
262
71
58
310
1511
456
4550
35
0
0
37
252
4
3
31
11
1
7
7
1
82
372
63
73
162
84
8
1198
9
0
0
71
252
111
0
214
17
6
3
21
5
393
315
116
112
351
210
118
2315
18
2
0
27
126
10
2
22
3
1
4
14
3
145
69
24
53
82
168
17
772
6
0
1
50
168
31
1
147
3
2
2
7
10
269
191
57
114
125
252
45
1475
11
3
4
64
252
145
0
235
8
5
1
34
6
807
193
78
87
163
420
353
2858
22
5
5
294
1553
445
8
1268
70
24
24
213
47
2069
1402
409
497
1193
2645
997
13168
100
Table 9:2. The Scandinavian, flint-dagger material of northern Europe divided into
Lomborgs main types.
only three combined two different dagger types: type V was once
combined with type IV and type V is twice combined with type VI.
Mecklenburg, Germany
Mecklenburg is situated immediately to the south-east of SchleswigHolstein along the Baltic shoreline. In the east, Mecklenburg borders
on Pomerania in the north-western part of Poland. In this area, 1 268
Scandinavian flint daggers have been registered (Rassmann 1993:283
ff.). The relative distribution of types in this material differs rather
curiously from the Schleswig-Holstein daggers. Here, type-I daggers
are predominant, while the other types are quite scarce (Table 9:2). In
this respect, the type profile of this area shows striking similarities
with the western production area in northern Jutland (Fig. 9:12).
According to, Rassmann, flint daggers were produced in this area too
(Rassmann 1993:68 ff.). However, there is no study in which flakes
from dagger-reduction sequences have been defined. The interpretation
rests solely on findings of dagger and bifacial-sickle preforms. Therefore,
295
Jan Apel
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%
0%
0%
T ype I
T ype II
Type I
Type II
Type III
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%
0%
Type IV
Type V
Type VI
Type IV
Type V
Type VI
Type IV
Type V
Type VI
Type V
Type VI
0%
Type I
Type II
Type III
Type IV
Type V
Type VI
Type I
Type II
Type III
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%
0%
0%
Type I
Type II
Type III
Type IV
Type V
Type VI
Type I
Type II
Type III
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%
0%
0%
Type I
Type II
Type III
Type IV
Type V
Type VI
Type I
Type II
Type III
Type IV
Fig. 9:12. The frequencies of Lomborgs six main dagger types in different
regions of northern Europe (in percentages).
296
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%
0%
0%
Type I
Type II
Type III
Type IV
Type V
Type VI
Type I
Type II
Type III
Type IV
Type V
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%
0%
Type VI
0%
Type I
Type II
Type III
Type IV
Type V
Type VI
Type I
Type II
Type III
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%
0%
Type IV
Type V
Type VI
Typ IV
Typ V
Typ VI
Type IV
Type V
Type VI
0%
Type I
Type II
Type III
Type IV
Type V
Type VI
Typ I
Typ II
Typ III
70%
70%
60%
60%
50%
50%
40%
40%
30%
30%
20%
20%
10%
10%
0%
0%
Type I
Type II
Type III
Type IV
Type V
Type VI
Type I
Type II
Type III
297
Jan Apel
Fig. 9:13. The distribution of Grand Pressigny blade daggers (black dots) and
locally made, pseudo Grand Pressigny daggers made in local flint types (from
Struve 1955: Tafel 33).
Poland
The distribution of Scandinavian flint daggers in Poland is almost
exclusively concentrated in Pomerania, i.e. the north-western province
of the country (Wojciechowski 1976: 72, Mapa. I). There are only a
few flint daggers in evidence east of the river Vistula (Wyszomirski
1974:77, Fig. 1) and only a handful in the south-western parts of the
country (Wojciechowski 1976: 72, Mapa. I). Wojciechowski points
out that it is difficult to determine whether some of the Polish flint
298
Jan Apel
Fig. 9:14. The geographical distribution of the netice culture (dots) and the
Scandinavian flint daggers (circles) in Bohemia (from Agthe 1989:50, Abb. 24).
Fig. 9:15. The geographical distribution of the proto-netice culture (dots) and
the proto-netice flint-dagger graves (circles with dot) in Moravia (Agthe
1989:51, Abb. 25).
300
301
Jan Apel
out that Bantelmanns form group 8, should most probably be regarded
as extensively re-sharpened daggers of the Scandinavian type I. In this
respect, they are not, as Bantelmann suggests (1982:44), flint copies of
tanged, bell-beaker daggers in bronze. They have gained their
characteristic form from the fact that the blade has been extensively resharpened in a way that makes it look like a tang. This is a problem that
has also been recognised by Beuker & Drenth (1999:33) in their reassessment of the flint-dagger material from the Drenthe province of
the Netherlands. In this context, I include these daggers in Bantelmanns
form group 5, which contains daggers of Lomborgs type I. Accordingly, the Scandinavian daggers in this area consist of 7 daggers of type
I, 7 of type II, 3 of type III, 4 of type IV, 2 of type V and 1 of type VI.
Only 24 Scandinavian daggers have been recovered in this area. For
obvious reasons, this makes it difficult to judge the type profile presented
in Fig. 9:12 and Table 9:2.
The Netherlands
The daggers of Scandinavian types in the Netherlands were published
1969 by Bloemers. In all, 99 daggers are included in his article. Of
these, 17 are regarded as uncertain and 12 are regarded as being of
Scandinavian origin but impossible to assign to any of Forssanders six
types. Bloemers also presents six examples of strike-a-lights. In this
context, they have been assigned to types V and VI, on account of the
published drawings and the published information. This means that 70
Scandinavian-type daggers are known from the Netherlands. All of
Forssanders (and Lomborgs) six main types are represented in this
material. The majority of the daggers are concentrated in the northeastern part of the country, especially in the province of Drenthe. Very
few daggers have been recovered south and west of Veluwe (Bloemers
1969:47, Abb. 2). There are no primary-flint sources in the Netherlands
and all flint is imported. Beuker & Drenth (1999:33) suggest that all
flint daggers arrived as finished products, since only one preform is
known. As regards the chronological aspects, the flint daggers of the
Netherlands cannot contribute to further understanding. Only one
dagger of Scandinavian type has been found in a grave (Harrison
1980:26). Therefore, the chronological knowledge of the Scandinavian
flint daggers in the Netherlands relies on evidence from Denmark and
302
Austria
Five, Scandinavian, flint daggers have been found in Austria (Trnka
1991 and ms). According to Gerhard Trnka, these are currently the
southernmost finds of Scandinavian flint daggers (Trnka, ms). Their
303
Jan Apel
geographical distribution is limited to the eastern part of Austria (Lower
and Upper Austria as well as Styria). The finest example is a type-IV
dagger from Gusen-Berglitz in Upper Austria. Apart from this nice
specimen, Trnka presents four fragmented daggers of Scandinavian types.
All these are examples of later types and it may be that the Scandinavian
daggers that reached as far south as the core area of netice culture, and
beyond, did so during the classical phase of the netice culture, 19501700 cal BC.
Jan Apel
306
307
Jan Apel
indication of such a down-the-line, distribution system is that the number
of artefacts declines as a function of the distance from the production
area. However, since this is a general pattern it may also apply to redistribution based on exchange, as well as to organised trade; it has to
be complemented. A more discriminating pattern for the existence of
down-the-line exchange is that artefacts found in areas far from where
the production took place would tend to circulate for a longer period
of time before being deposited than artefacts near the production area.
Thus, if an artefact type that has been continuously re-sharpened, such
as flint axes or daggers, is found far from the area of production, it
would tend to have been more re-sharpened than examples found nearer
the production area (Hodder & Lane 1982, Olausson 1983, Sundstrm 1992). Hence, the degree to which the average length varies
between the Swedish provinces, as well as the degrees to which daggers
from different provinces have been re-sharpened, will be investigated
below. If there is a significant decrease in the average length of the daggers
in relation to the distance from the production area, it is possible to
argue that the daggers were exchanged in a down-the-line network. If
the daggers are more frequently re-sharpened, the further away they are
from the production area, there will equally be a chance of finding
arguments for a down-the-line, exchange network. Such results will
point to a society without centralised power on the regional level.
In more complex traditional societies the exchange of prestige objects
was organised in a different way. Here, local aristocracies often
belonging to the lineage with the closest genealogical ties to the mythical
lineage ancestor had direct contact with other aristocracies and
participated in an exchange network in which prestige objects circulated.
These aristocracies had exclusive control over the influx of prestige objects
and other foreign materials into their sphere and, by redistributing these
objects to the other segments of society, they were able to collect returns
in the form of physical support, surplus food and daily goods.
Consequently, a prerequisite for a redistribution system of this kind is
the presence of regional, administrative centres from which foreign and
valuable objects are distributed locally (Renfrew & Bahn 1991:310).
The archaeological traits indicating a distribution organised in this way
are more difficult to comprehend. It has been suggested that significant
concentrations of artefacts, within the consumption area, may indicate
local re-distribution centres (Renfrew & Bahn 1991). However, as I see
308
309
Jan Apel
Degree of sharpening
There has been no systematic treatment of the problems related to the
degree to which flint daggers were sharpened. However, virtually all
those who have worked on flint daggers have pointed out the
methodological problems created by the continuous re-sharpening of
dagger blades in prehistoric times. For instance, and as we have seen
earlier, Lomborg introduced the subtypes I A/B and I x in order to be
able to classify extensively sharpened daggers that lacked a pronounced
handle. Regarding the rest of the dagger types, Lomborg was of the
opinion that the characteristics of handles would make it possible to
classify them according to the typology, even in cases in which the blade
had been sharpened to the point at which it lost its original outline.
All finished flint daggers have been sharpened at least once. Sharp
edges were apparently one of the characteristics that proper flint daggers
should have. The following description is based on information acquired
from Errett Callahan during our joint efforts to document the
production of type-IV daggers (Callahan & Apel, ms).
From the technical point of view, the sharpening of the edges of a
bifacial flint artefact, such as a flint dagger, will take place when the
blade has reached its intended outline during the production. The
sharpening process can be divided into two separate acts, each relating
to a repertoire of distinct gestures. (1) The entire edge of the blade is
retouched. Flakes are there evenly removed by pressure technique from
the blade-handle transition on one edge, down to the tip and to the
blade-handle transition on the other side. These flakes should be released
so that the flake scars penetrate well into the middle zone of the blade
(Callahan 1996: vii). This will give the blade the desired, evenly lenticular
cross-section. When the first stage of the retouch is finished, the edge
will display a denticulate outline. In fact, earlier on, bifacial sickles with
a denticulate edge were regarded as saw blades. Nowadays, we have to
accept that they may just as well be sickles that have not been subjected
to the second phase of the sharpening process. However, as regards certain
types of lance- and spearheads, the denticulate edges were no doubt
anticipated as the end result. (2) During the second phase of the sharpening process, the tip of each toothcreated during the first stage is
removed by a very light-handed, pressure technique performed with a
small pressure flaker (Callahan & Apel, ms). The edge that is created by
a properly conducted, sharpening process is almost knife-sharp and will
310
Jan Apel
display damage that happened during the discovery or in the
museums. Dagger blades of this category display a evenly curved,
lenticular cross-section.
2. Daggers with blades that have been re-sharpened by a person who
understood the principles of retouching bifacial edges. The scars of
the pressure flakes that indicate the first stage of the retouching process
run smoothly well into the blade, so that the edge angle is not
significantly decreased. Dagger blades of this category display a
lenticular cross-section (Fig 9:18). The blade surface on daggers resharpened in this way is clean and smooth, without major hinges
and other problems related to insufficient know-how.
3. Daggers with blades that have been re-sharpened by a person who
did not understand the principles of biface retouching. Usually, this
inexperience results in short, retouch, flake scars and, as a consequence
of this, higher edge angles, and a cross-section of such blades will
reveal an uneven, lenticular cross-section (Fig 9:18). The blade surface
on daggers of this kind will frequently exhibit abrupted, negative,
flake scars and hinges.
4. Daggers that have been turned into other kinds of artefacts by
additional knapping or grinding. Usually, it was broken daggers that
were transformed. It was for instance, quite common for handles of
types III and IV to be ground into axes and chisels. Dagger tips, on
the other hand, were turned into spear- and arrowheads.
In this context, it is important to acknowledge that the difference in
the degree of sharpening between categories 2 and 3 is qualitative in its
nature. This means that a dagger with a blade of category 2 may very
well have been extensively sharpened, so that it displays a significantly
altered, width-thickness relationship. However, the way in which this
sharpening has been executed will result in an evenly lenticular crosssection, as opposed to the uneven cross-sections of blades of category 3.
In certain cases, dagger blades of category 3 are not extensively sharpened,
although they were sharpened by someone who lacked the proper
knowledge and know-how needed to produce functional blades. This
means that the qualitative classification system presented above is needed
in order to complement any measurements taken on the daggers.
312
Measurements
Unfortunately, very few of those who have studied Scandinavian flint
daggers have bothered to publish data on the measurements. As we
shall see, measurements of daggers can indeed play an important role in
determining the social and political aspects of the Late Neolithic societies
in question. In this book, the measurements are also used in an
independent way, in discussing the degree of sharpening of daggers that
in some respect complements the qualitative classification system
described above. Consequently, the great majority of the Swedish flint
daggers included in my database have been measured. The three most
important measurements, used in Chapter 9, are as follows:
1. Maximum length, taken from the middle of the base to the tip of
the dagger.
2. Maximum width of blade, taken perpendicularly to the length on
the widest part of the blade.
3. Blade thickness, taken on the widest part of the blade.
313
Jan Apel
079
80
10 -99
01
12 19
01
14 3 9
01
16 59
01
18 79
01
20 99
02
22 19
02
24 39
02
26 59
02
28 79
029
9
30
0-
900
800
700
600
500
400
300
200
100
0
this material that relates directly to the distance from the production
areas. It should be noted that the dagger populations in some provinces,
such as Skne and Vstergtland, have higher standard deviations,
indicating that these materials are more heterogeneous, as regards the
length distribution, than others. However, this pattern cannot be related
to the distance from the production area. It can thus be concluded that
no fall-off pattern can be recognised from the length distribution of the
Swedish flint-dagger material. As regards length, the composition of
the dagger materials in the different provinces does not vary in a
significant way.
Now we have to bear in mind the specific way in which flint daggers
are re-sharpened. Since the edges are located mostly on the sides of the
daggers, from the theoretical point of view, a re-sharpened dagger may
not lose very much length, compared with an axe. This means that,
instead of investigating length, it may be fruitful to investigate whether
or not a fall-off tendency can be discerned, as regards the width of the
blades. In Fig 9:21, the distributions of the blade-width values of the
Swedish dagger material are presented. These distributions also take on
an almost bell-shaped form. In order to see if the dagger blades became
narrower, the further they were from the production area, the thickness
of the blade (mm) was divided by the maximum width of the blade
(mm). If the result was a relatively low value, it would mean that the
dagger blades are wide in relation to their thickness, while a relatively
high value would mean that they are narrow in relation to their thickness.
314
42
32
33
31
47
36
37
43
39
42
34
29
36
30
27
49
Province
Skne
Uppland
Gotland
Vstergtland
Blekinge
stergtland
Smland
Vrmland
Nrke
land
Sdermanland
Halland
Dalsland
Bohusln
0,17
0,11
0,14
0,13
0,11
0,11
0,11
0,12
0,11
0,15
0,10
0,12
0,09
0,10
Sdermanland
316
Uppland
26
0-
27
30
0-
0%
10%
0%
23
20%
10%
0-
30%
20%
22
40%
30%
40%
19
26
22
18
0-
27
23
19
30
0-
0-
0-
0%
0-
10%
0%
18
20%
10%
30%
20%
15
40%
30%
0-
40%
14
26
22
18
14
0-
27
23
19
15
30
0-
0-
0-
0-
0%
10%
0%
11
20%
10%
0-
30%
20%
10
40%
30%
40%
11
26
22
18
14
10
0-
27
23
19
15
11
30
0-
0-
0-
0-
0-
0%
0-
079
10%
0%
15
10
0-
27
23
19
15
30
0-
0-
0-
0-
11
20%
10%
0-
Skne
079
26
22
18
14
0-
30%
20%
0-
27
23
19
15
30
0-
0-
0-
0-
11
40%
30%
11
Halland
079
26
22
18
14
0-
10
079
40%
0-
0-
27
23
19
15
30
0-
0-
0-
0-
11
Dalsland
079
26
22
18
14
0-
10
079
Blekinge
14
0-
27
23
19
15
11
30
0-
0-
0-
0-
0-
10
079
30
0-
-2
79
-2
39
-1
99
-1
59
-1
19
26
0
22
0
18
0
14
0
10
0
079
79
10 -99
01
12 1 9
01
14 3 9
015
16 9
01
18 7 9
01
20 99
02
22 19
02
24 39
02
26 59
02
28 79
029
9
30
0-
0-
80
40%
35%
30%
25%
20%
15%
10%
5%
0%
10
26
22
18
14
10
079
Jan Apel
40%
30%
20%
10%
0%
Bohusln
Gotland
Nrke
Smland
4
3
3
2
2
1
1
Vstmanland
Vstergtland
40%
40%
30%
30%
20%
20%
10%
10%
0%
0%
200
150
100
50
317
26
0
22
0
18
0
14
0
0-
30
0-
-2
79
-2
39
-1
99
-1
59
-1
19
079
30
-2
79
-2
39
-1
99
-1
59
-1
19
0%
10
0
10
0
26
0
22
0
18
0
14
0
10
0
10%
88
0-
27
23
19
15
11
30
0-
0-
0-
0-
0-
079
20%
079
80
10 -99
01
12 19
01
14 39
01
16 59
017
18 9
01
20 99
02
22 19
02
24 39
025
26 9
02
28 79
029
9
30
0-
26
22
18
14
10
079
30%
81
079
80
10 -99
01
12 19
013
14 9
01
16 59
017
18 9
01
20 99
021
22 9
02
24 39
025
26 9
02
28 79
029
9
30
0-
40%
70
63
56
49
42
35
28
21
14
35%
30%
25%
20%
15%
10%
5%
0%
Vrmland
land
40%
30%
20%
10%
0%
stergtland
Fig. 9:20. The length distribution of the flint-dagger populations of the Swedish province.
Jan Apel
Provinces
Blekinge
%
Bohusln
%
Norrland
%
Dalsland
%
Gotland
%
Halland
%
Nrke
%
Smland
%
Sdermanland
%
Uppland
%
Vstergtland
%
Vrmland
%
stergtland
%
land
%
Skne
%
Total
Cat. 1
7
8%
48
13%
3
14%
29
8%
8
10%
60
13%
5
6%
10
7%
3
4%
4
5%
74
13%
2
6%
14
8%
8
15%
184
16%
459
Cat. 2
37
40%
151
41%
6
29%
128
37%
27
34%
186
41%
30
34%
65
46%
23
32%
20
25%
146
26%
7
22%
92
53%
26
47%
522
44%
1466
Cat. 3
47
51%
151
41%
11
52%
157
46%
41
51%
187
42%
46
53%
63
45%
42
58%
51
63%
276
50%
20
63%
64
37%
19
35%
437
37%
1612
Cat. 4
1
1%
15
4%
1
5%
28
8%
4
5%
16
4%
6
7%
3
2%
5
7%
6
7%
61
11%
3
9%
3
2%
2
4%
33
3%
172
Total
92
365
21
342
80
449
87
141
73
81
557
32
173
55
1176
3709
Table 9:5. The Swedish flint daggers divided into the four re-sharpening
categories defined in Chapter 8.
Regions
Eastern central Sweden
%
Western central Sweden
%
South-eastern Sweden
%
Skne
%
Halland
%
Norrland
%
Western Norway
%
Eastern Norway
%
Jutland (Moesgrd)
318
1
14
5%
151
12%
47
9%
184
16%
60
13%
3
14%
153
16%
40
17%
16
20%
2
80
29%
425
34%
247
46%
522
44%
186
41%
6
29%
295
31%
64
27%
47
59%
3
159
58%
584
47%
234
43%
437
37%
187
42%
11
52%
451
48%
131
55%
17
21%
4 Total
20 273
7%
89 1249
7%
13 541
2%
33 1176
3%
16 449
4%
1
21
5%
40 939
4%
4
239
2%
0
80
0%
nd
Sw
ed
en
c
rla
ay
N
or
N
or
Sw
c
ed
en
ay
w
en
.N
or
ed
en
e
Sw
S-
Sw
ed
Sk
n
e
Sw
Ju
tla
nd
90
80
70
60
50
40
30
20
10
0
Jan Apel
blades. In fact, the distribution of the width of the Swedish dagger
blades, in all Swedish parishes, also takes on a bell-shaped appearance.
These conclusions are supported by the fact that there is no fall-off
tendency in the standard deviations of either length or thickness-width
ratio. In these respects, the distribution of flint daggers is contrasted by
the distribution pattern of the thin-butted axes in the same area during
the Early Neolithic period (Sundstrm 1992). These differences reflect
serious changes in the scale of the political landscape from that of the
Early Neolithic TRB communities to that of the Late Neolithic
communities in the same area. In the latter case, the lack of evidence for
a down-the-line exchange may indicate that there were local centres,
based on hierarchical differentiation, conducting directional exchange
with other local centres. Bergljot Solberg (1994) has indeed presented
convincing evidence for directional exchange between the Thy district
in Jutland and central places, such as those at Jren and North Karmy,
on the Norwegian west coast during the Late Neolithic and the Early
Bronze Age. This conclusion is supported by my own observations in
Norwegian museum collections: the dagger materials from the
Norwegian west coast are quite homogeneous, as regards size and quality.
In Sweden studies have been made that indicate the presence of
centralised places of local or regional authority. A study of the Late
Neolithic and the Early Bronze Age in Vstergtland certainly suggests
hierarchical differentiation on the local level, among other things, as
reflected in the different qualities of grave goods in the numerous gallery
graves of this region (Weiler 1994:54 ff.). Investigations of the centralised
versus local production of different qualities of Late Neolithic, shafthole axes, utilized in different social contexts, also point to hierarchical
differentiation on the local level during the Late Neolithic (Lekberg
1997, 2000a, b & c). Studies conducted in the Mre region in the
province of Smland also suggest a political landscape that was organised
according to hierarchical principles. Here, the utilization of prestige
objects made of flint, in the Late Neolithic, and metal, in the Early
Bronze Age, indicates that similar structural principles were prevalent
in this period (Gurstad-Nilsson 1999:191). A spatial analysis of Late
Neolithic artefacts from two areas in the Mre region suggests that
there was an unequal distribution of valuable objects in the landscape,
indicating regional hierarchies (Hagberg 1999). This study is actually
320
321
Chapter 10
This chapter starts with a brief rsum of the previous chapters. Further
on, a more detailed interpretation of the agrarian Late Neolithic societies
in Scandinavia, based on the conclusions of the previous chapters, is
presented. An interpretation of the social setting in which the flintdagger technology was embedded will be presented and the disappearance
of the technology will be discussed in social terms.
Rsum
The aim of this dissertation was to investigate how far the organisation
of a traditional technology corresponds to the degree of social complexity
in a sedentary, agrarian society. Thus, I argue that the social structure of
the Late Neolithic societies in Scandinavia was more complex than has
previously been thought. An examination of the production of flint
daggers indicates the presence of formal apprenticeship systems based
on corporate descent groups. The selection of apprentices was based on
hereditary principles. In other words, the craft appears to have been
institutionalised, and this suggests a fairly complex, social structure.
This conclusion is related to the theory of symbolic assets presented by
Bourdieu and to the idea of the Aggrandizer presented by Hayden.
Accordingly, the main social function of the Late Neolithic, descent
groups in the southern and central parts of Scandinavia was to reproduce
their access to collective assets throughout the generations by following
strict, hereditary principles. In reaching these conclusions, I have concentrated on two main subjects:
(1) The ways in which the division of labour and the use of prestige
technologies were related to the social complexity in kin-ordered or
segmentary societies in general. This discussion includes theoretical issues,
323
Jan Apel
as well as the questions put and the observations made during controlled,
flint-working experiments.
(2) The role that the production of elaborate flint tools played in
the transformation of segmentary tribal societies into chiefdoms that
allegedly occurred during the Late Neolithic and Early Bronze Age in
Scandinavia. For this purpose, the contexts in which Late Neolithic,
Scandinavian, flint daggers were produced and consumed have been
investigated.
Technology is here defined as a coherent system of artefacts, behaviours and knowledge that can be handed down from one generation to
another. Since this definition concentrates on the reproduction of
technology, it recognises that technology is integrated with other aspects
of society. A holistic, supra-individualistic perspective is regarded as
especially important in this context, since flint daggers were made over
a period of c. 800 years.
My understanding of the flint technology is based on my collaboration with an experienced flint-knapper and archaeologist, Errett
Callahan. During the course of two summers, at the Historical Research
Centre in Lejre, Denmark, and at Uppsala University in Sweden, several
production stages were defined as the result of a dialectic research strategy
in which archaeological evidences and the experimental findings were
confronted and compared (Callahan & Apel, MS). This resulted in
some understanding of the degree of knowledge, skill and time that
had to be achieved in order to make prestigious flint daggers.
After being defined, the flint-dagger technology was subjected to an
operational-chain analysis. This method is rooted in Durkheimian
sociology and, consequently, gestures, even those that seem to reflect
the biological aspects of humanity, are regarded as social phenomena
that are learned in social contexts. This is an important observation,
since the implication is that the study of gestures concerns subjects such
as anthropology, archaeology and sociology. Two important concepts
form the basis of my investigation of the social aspects of the flintdagger technology: (1) knowledge (connaissance) and (2) know-how
(savoir-faire). These terms define two fundamental elements of a
distinctly, neuro-psychological nature that are involved in the execution
of all gestures. Knowledge has an explicit and declarative character and
can be communicated to others; it can be passed from teacher to pupil
by word of mouth, signs or written language. Know-how is an
324
Jan Apel
play in the last part of Chapter 4 in the form of an ethno-archaeological
example (traditional iron-working in Kenya). Here, strict hereditary
principles decide who will have the opportunity of becoming a smith
and who will not. It is suggested that those stages of iron-working that
are based mostly on knowledge (recipes for action) and do not demand
a great deal of know-how, such as smelting skills, are surrounded by
taboos and are conducted at secluded locations. However, stages that
require a high degree of know-how, such as forging, are conducted in
public, often at a central place of the hamlet. The point is made that the
recipes for action in iron-working are protected by the smiths at the
same time as they actively exhibit their practical know-how. The practical
know-how is exclusive, because it takes years to master, and thereby
gives the smiths both power and prestige when demonstrated in public.
Chapter 5 is concerned with a research history aimed at understanding
how flint debitage from bifacial-reduction sequences can be defined.
This work is needed in order to proceed with Chapters 6 and 7, in
which the flint debitage from Late Neolithic sites in eastern central
Sweden and Jutland is investigated and discussed.
In Chapter 6, the flint industry of southern Scandinavia is investigated. Several sites with evidence of flint-dagger production, mainly in
Jutland, form the basis of the investigation. These sites include
settlements, extraction sites and mines. It was concluded that the
different stages in the production of daggers took place in different
locations. The extraction sites of each production fraternity mines
and certain beaches were, for natural reasons, kept secret and were
guarded by taboos and esoteric regulations. No one from outside the
production fraternity was allowed to participate in the early production
stages, since this would threaten the exclusiveness of the technology.
The young, inexperienced apprentices collected suitable nodules for
dagger-making. It was also their task, together with the journeymen, to
make the early-stage preforms. This activity took place at secluded spots
near the raw-material sources and was also surrounded by esoteric
regulations. Thus, each corporate lineage had its own outcrop. The Late
Neolithic flint mines of Thisted and Himmerland are examples of such
raw-material sources. Places for the extraction of suitable raw material
have also been found, for instance, at the beaches of Fornaes and
Drenges. The flint debitage on these sites often displays obvious
knapping errors, and the sites are littered with broken preforms.
326
Jan Apel
The flint-working that was practised in the Late Neolithic settlements of eastern central Sweden was dictated by strict conventions
regarding the form and the choice of southern-Scandinavian flint as the
raw material. This resulted from the fact that the area could not maintain
the level of craftsmanship present in areas immediately to the north
where bifacial arrowheads were made in local raw materials at the same
time. From the purely practical point of view, it makes no sense to use
only flint as raw material in eastern central Sweden when there are local
raw materials, such as hlleflint , available. Consequently, we may expect
that this pattern will be explained by the fact that artefacts of southernScandinavian types and flint had a great symbolic value for the agrarian
population of eastern central Sweden.
In Chapters 8 and 9, aspects of the Scandinavian flint daggers are
investigated. It is concluded here that several of the different dagger
types and subtypes that have been defined by earlier scholars can be
regarded as reflecting regional, or even local, workshops. These regional
production areas distributed flint daggers to different consumption areas.
The general distribution of the Scandinavian flint daggers covered a
large region in northern Europe from the Rhine delta in the west to
Pomerania in the east, and from central Germany in the south to the
Norwegian north-west coast in the north. Outside this area, a small
number of daggers have been recovered in the British Isles, in Bohemia
and Moravia, in Austria, in Finland and the Baltic countries and in the
northern parts of Sweden.
Finally, an investigation of the distribution of flint daggers in
Scandinavia reveals that they were distributed by direct exchange.
Exchange systems of this kind require redistribution centres on the local
or regional level. This presupposes hierarchical settlement patterns on
the local and regional level and is an indication of chiefdoms. In this
context, it is appropriate to mention that institutionalised heredity, such
as has been proposed for the dagger production, is another indication
of a chiefdom level of society. It can thus be concluded that there are
strong archaeological arguments in favour of regarding the Late Neolithic
societies as having had a social structure at the chiefdom level (as defined
in Chapter 4).
328
329
Jan Apel
1900 B.C.
Establishing long-distance
Clans
alliances
1500 B.C.
land
1100 B.C.
Intensification of production.
Settlement clustering.
800 B.C.
500 B.C.
competition
Increasing degradation of
investment in ritual.
hierarchy.
houses.
Local migrations.
Fig. 10:1. Development and transformation of the southern Scandinavian societies from the
beginning of LN II to the end of the Bronze Age (after Kristiansen 1991:32, Fig. 2:4).
330
The formation
In this context, the formation of the dagger technology is related to the
anti-intellectualistic theory of culture change discussed in Chapter 4.
The main proposition, in line with the ideas presented by Durkheim
and Binford, was that technological specialisation and social complexity
may be the results of an increased moral density rather than of the
conscious decisions made by individual agents. One way of detecting
moral density in prehistory is to find arguments for an increased
population density creating a population pressure. Thus, it may be
possible to consider the appearance of the flint-dagger technology as
the result of a packing problem created when domestic units were
marginalized from rich areas and were thereby forced to develop new
technologies in order to maintain their traditional social structure.
What would be the archaeological consequences of a packing problem of the kind that Durkheim and Binford imagined? One of the
characteristic features of the Late Neolithic period in Scandinavia is
that there are signs of an increased population density. There is evidence
of a massive relocation of domestic units to remote areas where
previously there were no signs of Neolithic populations. This process
manifests itself, perhaps most clearly, in areas such as the provinces of
Vrmland, Dalsland and Smland in Sweden where gallery graves were
erected and Neolithic technologies, including cultivation, are introduced
during this period (Johansson 1961, Heimann 2000, Lagers 2000).
In eastern central Sweden settlements tended to be localized to the heavy
clay soils, rather than, as during earlier periods, to glacial eskers and
other well-drained soils. This means that, in this respect, the Late
Neolithic settlement pattern is similar to that of the Bronze and Early
Iron Ages. It is, in fact, more than likely that the Late Neolithic
settlements in these areas are hidden in the archaeological remains of
later periods. In areas with rich archaeological remains from later periods,
such as eastern central Sweden, Late Neolithic graves and settlements
tend to be covered by later settlements and graves (Holm et al. 1997:253,
Andersson 2001:81). The gallery graves of Alby, Annelund and Dragby,
for instance, have all been covered in later times. It has been suggested
that the Bronze Age population of eastern central Sweden, by covering
the earlier collective graves, buried an old, collective ideology that was
seen as a threat to the social values of the stratified, Bronze Age society
331
Jan Apel
(Andersson 2001:82). However, the forested areas of Smland and
Vrmland, which have few traces of agricultural practices after the Late
Neolithic/Bronze Age transition, can still be regarded as fossilised
remains of the Late Neolithic, cultural landscape, a landscape that was
never covered by Bronze Age graves and settlements. The tendency to
exploit the heavier clay soils during the Late Neolithic has, in earlier
research, been related to the land-upheaval process in eastern central
Sweden during the Neolithic. The main argument was that valleys with
clay soils were continuously exposed as the sea withdrew and that people
moved in because of the increased population pressure. However, the
tendency to use clay soils was not confined only to areas where landupheaval processes had changed the landscape. A similar tendency can
be discerned in Denmark as well (Douglas Price, personal
communication). Thus, it can be concluded that the choices of where
to live and where to bury the dead are cognitive aspects that the Late
Neolithic population of the southern and central parts of Scandinavia
shared with the peoples of later times.
As we have seen, there are archaeological signs indicating an increased
population density during the Late Neolithic period in the southern
and central parts of Scandinavia and this may have led to the exploitation
of new environments. But, in order to argue for an anti-intellectualistic
explanation of the emergence of new technologies related to craft
specialisation that is implied by the flint-dagger technology, we need to
gather more evidence. We need to investigate the area of the first
development of the Scandinavian dagger technology, i.e. the situation
in Jutland prior to the introduction of the flint dagger at the transition
between the MN B and the LN I around 2350 cal BC.
The Middle Neolithic period in southern Scandinavia is divided
into two distinct phases; MN A and B. In this context, we shall be
concerned only with the latter phase. In Jutland, MN B is defined by
the appearance of a new type of grave. The collective chamber-tombs
of the MN-A phase were replaced by individual or occasionally
double inhumation graves in small mounds. These graves make up
the majority of the archaeological remains from the Single-Grave
Culture of Jutland. The tripartition of MN B into the three sub-phases,
I, II and III, originates from Sophus Mllers definitions of the three
grave types commonly found in these mounds: lower, middle and upper
graves (Mller 1889, Damm 1991:83). Since these graves corresponded
332
Fig. 10:2. The geographical distribution of battle axes from the early phase of
MN B (a) and battle axes from the latest phase of MN B (b). The distribution
indicates the primary habitation area of the Single-Grave Culture during the
earliest and latest phases of MN B respectively (from Struve 1955).
to a stratigraphical sequence with chronological significance, the SingleGrave period was later divided into a lower, middle and upper subphase (Glob 1944) and this division corresponds to the MN B I, II and
III phases that are used today (Damm 1991:83). However, it has been
suggested that it would correspond better to the actual archaeological
material to divide MN B into an earlier and a later sub-phase (Damm
1991:81). The transition between MN B and LN I is set to the times
when daggers replaced battle axes in stone as the prestige gift in the
male single graves. Thus, it was within the context of the late SingleGrave-Culture in northern Jutland that the production of flint daggers
commenced at the transition between MN B and LN I.
The geographical distribution of the early phase of the Single-Grave
culture is illustrated in Fig. 10:2a. The grave mounds were situated in
groups of 10 to 20 along streams in the central part of Jutland. The
distribution of Single-Grave-Culture graves from the early MN B phase
333
Jan Apel
coincides with that of Jutlands most fertile soils light diluvial sand,
moraine clay or sandy moraine (Damm 1991:81). These kinds of soil
are suitable for crop agriculture with a wooden ard and for grazing as
well. Excavations of Single Grave sites on Jutland, such as those at
Hemmed on Djursland (Boas 1993) and Vorbasse near Esbjerg (Hvass
1977, 1986:327 ff.), have demonstrated that long houses no less
elaborate, as regarded construction details, than the houses of the FunnelBeaker Culture in Denmark and Sweden were built. The battle axes
from the early phase of MN B which reflect the distribution of
inhumation graves are concentrated to the central parts of Jutland
(Fig. 10:2a).
During the course of MN B the material manifestations of the SingleGraves culture expands to the north as well as to the Danish islands in
the east. This tendency is best illustrated by the distribution of the later
battle-axe types found in the upper graves (Fig. 10:2b). Thus, during
the final Middle Neolithic phase, the geographical distribution of battleaxes also included the provinces in the Limfjord area, such as
Himmerland and Thisted, and it was here that the first daggers were
produced somewhere around 2350 cal BC. It is also in this area that the
majority of the Danish, Late Neolithic settlements have been recovered
(Bjrhem & Sfvestad 1989:77, Fig. 73). However, this was apparently
not because the Limfjord area had a high population density at this
time (Damm 1991, Vandkilde 1996). The reason for the high frequency
of Late Neolithic settlements here is in my opinion, only because of
their archaeological visibility. That is to say, the flint industry of these
sites, including broken dagger preforms and thousands of easily
recognisable, flake types from bifacial-reduction sequences, tends to
amplify them and make them easy to find, as well as to date. While
Late Neolithic sites in other areas of Scandinavia have a tendency to get
lost in the archaeological remains of later periods, the sites near the
natural flint sources, are simply easier to spot. In fact, Jensens calculations
of the population density of Denmark, based mainly on grave
frequencies, clearly suggests that central Jutland was the most populated
area of Jutland during the Bronze Age (Jensen 1979). Thus, we may
assume that central Jutland had the highest population density in this
region during the Late Neolithic as well. It is therefore not unlikely
that the highly visible sites in the Limfjord area are the remains of a
population that was marginalised from central Jutland in the course of
334
Jan Apel
with the organised production of prestige goods. The connection between
outcrops and the mining of lithic materials, on the one hand, and the
production of prestige items, on the other, is a general pattern in the
greater part of Europe during the entire Neolithic period (see H.
Knutsson, in prep.).
The reproduction
The Middle Neolithic, Bell-Beaker communities of Europe were familiar with metallurgy, wore woollen clothes, practised horse-riding
and held ceremonies, including feasting and the drinking of alcohol
beverages. This is a context in which an idealised, male-warrior role was
formed and projected (Sherratt 1987:88, Vandkilde 1998). The copper
and flint daggers that appear during this period in large parts of Europe
form part of the symbolic language that was related to this new, warrior
identity. From now on, up to the Late Bronze Age, this new identity
characterised large areas of central and northern Europe. In Euro-Asia,
the dagger is traditionally thought of as a male weapon and symbol. In
the earliest, European, written sources, the daggers are part of a warrior
outfit dated to the Bronze Age. In areas of the Near East, such as Yemen,
metal daggers formed an important part of a respectable mans wardrobe
up to the present (Cammann 1977).
In Scandinavian prehistory, flint daggers seem to replace the battleaxe of stone as the male prestige gift in the upper graves of the Jutlandic
Single-Grave Culture. These daggers were made under influences from
western and central Europe. There are ethnographic examples of how
the spectacular outfits worn by warrior Age-Sets came to form the ideal
of a larger region, including several other ethnic groups. Among the
pastoral Masaais of Kenya, for instance, males were not allowed to
marry until they had passed their thirties. Because of this, the large
group of unmarried young males formed a powerful warrior class that
raided neighbouring territories for cattle and were generally feared and
respected. Eventually, the respect that this outfit gave its bearers tempted
the younger men of neighbouring tribes to adopt certain, typical, Masaaiwarrior features, such as the typical head-gear with feathers and the long
spears, thus forming their own warrior age-classes (Hkansson, personal
communication, see Muriuki 1979:98 f., Spear 1997:239). It is
intriguing to consider that in this example the young males from different
ethnic groups shared a similar interest in representing a warrior ideal. It
336
Jan Apel
right to wear a dagger was formally granted at the same time (Cammann
1977:33).
I have suggested that the dagger technology was complex and most
likely transmitted vertically within lineage groups or clans. It is likely
that the apprenticeship started at a fairly early age, since this would
simplify the learning process and thus make it easier to teach those who
were supposed to learn the craft independently of their natural abilities.
As we have seen, this interpretation also makes sense from the sociological
point of view. It would therefore have been convenient if the apprentice
was actually brought up with, or close to, the teacher. In the following
discussion, it is anticipated that the Scandinavian, Late Neolithic, flint
daggers were used as status insignia by an Age-Grade of warriors.
Among the traditional Kabyle farmers of Algeria, the rifle is a symbol
of nif (the personal qualities that define a respectable, honourable male).
The rifle functions as a form of symbolic embodiment of the nif within
the male agnatic group of these patrilinear farmers (Bourdieu 1977:61).
It is tempting to regard the flint dagger as functioning in similar fashion
as a symbol either of adult males or of an age-class of males in the Late
Neolithic communities of Scandinavia. This is actually very likely, since
the different qualities of bronze daggers and flint daggers differ radically,
thus making a purely functional interpretation highly unlikely (see
Chapter 9).
If the flint daggers were used as male prestige objects, it is not entirely
unlikely that the flint smiths who produced the flint daggers were males.
In the following discussion, it is thus anticipated that the flint-smiths
were males. It has been proposed that the flint-dagger technology was
most likely reproduced vertically within family units (see Chapters 24). It has also been suggested that the apprenticeship started at a fairly
early age, i.e. around puberty. However, even if the formal introduction
to the apprenticeship occurred at puberty, it would clearly be an
advantage if children intended to become apprentices were brought up
in an environment in which the craft was practised. In other words, it
would be a practical advantage if they were brought up among artisans
and master knappers. Such a scenario also makes sense if we consider
Haydens theory, in which the domestic unit tries to maintain and enhance
its status and power through child-growth strategies, including letting
children participate in costly education (see Chapter 4). Some form of
residence that maintained long contact between older males (masters)
338
Jan Apel
for instance, as the result of the death of the male elder. Such an event
will tend to provoke internal struggles between brothers or cousins, i.e.
the sons or grandsons of the elder. The more vulnerable a family, the
more nif it must possess to defend its sacred values. There are two reasons
why the organisation of a parallel-cousin marriage will maintain the
cohesion of the group. (1) The two brothers will, through the marriage
of their children, be closely united and thereby less inclined to engage in
a power struggle that would lead to a fraction within the lineage. (2)
The community as a whole will appreciate the occasion, since it honours
the tradition. This may be why the women in these communities are
less inclined to instigate parallel-cousin marriages; usually the elder male
within the lineage does it, often without consulting the women and
actually trying to arrange it over their heads. Usually, under circumstances in which the lineage is not threatened, women tend to actively
participate in matrimonial negotiations, since they are free to negotiate
without losing face. The matrimonial strategies used by women aim at
incorporating more women from their original groups through marriage,
since such strategies will increase their own social power (Bourdieu 1977).
This is actually a nice illustration of a continuous dialectic relationship
between social structure and practice. The parallel-cousin marriages are
necessary in order to maintain the cohesion of the lineage, while exogamy
is necessary, not only for purely biological reasons, but also because it
provides the necessary contacts used to construct exchange networks.
Since it is likely that the flint-dagger technology was organised
according to hereditary principles, i.e. that it was institutionalised, we
must also accept that other aspects of the society were also organised in
this fashion. What we are dealing with is a social structure that revolves
around a ranked and segmented, common-descent group.
The reformation
The dominant social structures of the Late Neolithic period were upheld
primarily by male elders and organised as unilineal descent groups. The
chiefly lineages had control over the most productive land and were,
through their close genealogical ties to the clan mythical ancestor and
through their external exchange contacts, distinguished from the general
populace. In marginal areas, lineage-groups specialised in the production
of prestige items in order to maintain their traditional way of life. These
prestige items circulated in an exchange network controlled by the male
340
Jan Apel
the producers of prestige items in flint could no longer maintain the
authority of the social structure. Certain lineages in the eastern production area solved the problem by increasing the quality of their product,
hence the production of the prestigious, fishtail, parallel-flaked daggers
of LN II. However, in the Limfjord area, the craft diminished.
Only agnatic relatives in the lineages were buried in the collective
gallery graves. Others, mostly women who had married into the lineage
but also males who were excluded from the right to inherit a position
in the corporate groups because of limited resources, were instead buried
in individual inhumation graves. This is supported by a significant
imbalance between the sexes in the Late Neolithic collective graves.
Thus, in the gallery graves of eastern central Sweden, there are 2,3 men
buried for every woman (Apel 1992) and from a larger material base, it
has been concluded that there are 1,7 buried men for every women
during the same period (Welinder 1979). The decline of the lineage
society is in fact signalled by the individual burial of the Early Bronze
Age, which should be connected with the single burials of women and
outcast men during LN II.
Within the consumption area, the daggers increased in value as a
function of the distance from the production areas. In general, they
were not very valuable in the production areas, simply because of the
large number of daggers in circulation. In these areas, it was the prestige
objects that were exchanged for daggers and that came from far away
that were valuable. In eastern central Sweden, the flint daggers, as well
as the debitage from the production stages that required a high degree
of know-how and were performed in public on the production sites,
were extremely valuable. As Mary Helms (1998) has pointed out,
excellent craftsmanship and long-distance trade are two strategies by
which tangible objects may become charged with intangible, esoteric
powers that are derived from cosmological realms and that give their
owners power and prestige. The chiefly lineages of eastern central Sweden
thereby actively used the flint artefacts and flakes that they obtained
through external contacts to maintain their power.
342
References
Adams, W. Y. & Adams, E. W., 1991. Archaeological Typology and Practical Reality. A
Dialectic Approach to Artifact Classification and Sorting. Cambridge University
Press. Cambridge.
Aghte, M., 1989. Bemerkungen zu Feuersteindolchen im nordwestlichen
Verbreitungsgebiet der Aunjetitzer kultur. Arbeits- u. Forschungsber. z. schs.
Bodenkmalpflege 33: 15-113.
1989. Bemerkungen zu Feuersteindolchen im nordwestlichen Verbreitungsgebeit
der Aunjetitzer Kultur. Praehistorica XV-XIV. Internationales symposium,
Univerzitas Karlova, Praha 1989: 305-309.
Ahlbeck, M., 1995. Kvartsiell Affasi. Den gropkeramiska boplatsen Bollbackens
stenmaterial ur ett processuellt perspektiv. BA-thesis, Department of Archaeology,
Uppsala University.
Althin, C. A., 1954. The Chronology of the Stone Age Settlements of Scania, Sweden, I.
Acta Archaeologica Lundensia. Series in 4, No 1. Lund.
Amick, D. S., Mauldin, P., and Tomka, S. A., 1988. An Evaluation of Debitage
Produced by Experimental Bifacial Core Reduction of a Georgetown Chert
Nodule. Lithic Technology, Vol. 17, No 1:26-36.
Andersson, A. & Andersson, S., 1991. Frord. I: Tre klassiska texter. Bokfrlaget
korpen. Gteborg.
Andersson, K. & Hjrthner-Holdar, E., 1989. Annelund - ett senneolitiskt
bebyggelsekomplex i sydvstra Uppland. Fornvnnen 1988/3:209-215.
Andersson, K., 2001. En hllkista vid Alby i Botkyrka socken, Sdermanland. I:
Bolin, H., Kaliff, A. & Zachrisson, T. (eds.), Mellan sten och brons.
Uppdragsarkeologi och forskning kring senneolitikum och bronslder. OPIA 27 &
Stockholm Archaeological Reports No. 39, 2001:74-84.
Andersson, K., Fagerlund, D. & Hamilton, J., 1990. Annelund - en senneolitisk
bondgrd. Arkeologi i Sverige 1987. Stockholm.
Apel, J., 1991. Hllkistor i stra Mellansverige. BA-thesis, Department of
Archaeology, Uppsala University.
1992. Aspekter p senneolitikum i norra Mlardalen eller slkten framfr allt
Fjlnir 11/1:5-21.
343
Jan Apel
2000a. Flint Daggers and Technological Knowledge. Production and
Consumption during LN I. In: Olausson, D. & Vandkilde, H. (eds.), Form,
Function and Context. Material culture studies in Scandinavian archaeology.
Almqvist & Wiksell International. Lund:135-154.
2000b. Kunskap, handlag och rmaterial en diskussion kring produktionen
och konsumtionen av senneolitiska flintdolkar i Skandinavien. Tor 30: 35-82.
2000c. Strict Conventions and Limited Ability Late Neolithic Flint
Technologies in Eastern Central Sweden. In: Hjrthner-Holdar, E. & Risberg, C.
(eds), Hantverkets roll i samhllet produktion och reproduktion. Rapport frn den
tredje workshopen om hantverksspecialisering Uppsala 18-19 mars 1999.
Riksantikvariembetet, UV GAL, Forskningrapport R0004: 3-16.
Apel, J., Bckstrm, Y., af Geijerstam, M., Hallgren, A.-L. and Kritz, A., 1996.
Valhalla, en mesolitisk lgerplats vid Lingbo, Ockelbo sn, Gstrikland. Slutrapporter
frn Arkeologikonsult AB 1996:6. Upplands Vsby.
Apel, J., Bckstrm, Y., Hallgren, F., Knutsson, K., Lekberg, P., Olsson, E., Steineke,
M. & Sundstrm, L., 1995. Fgelbacken och trattbgarsamhllet.
Samhllsorganisation och rituella samlingsplatser vid vergngen till en bofast
tillvaro i stra Mellansverige. Tor 27.
Apel, J., Hadevik, C. & Sundstrm, L., 1997. Burning down the house. The
transformational use of fire and other aspects of an Early Neolithic TRB site in
eastern central Sweden. Tor 29: 5-47.
Appelgren, K., Eklf, N. & Evanni, L., 2000. En senneolitisk boplats vid Grndal.
UV mitt, Dokumentation av fltarbetsfasen 2000:8. Arkeologisk underskning.
Riksantikvariembetet, Avdelningen fr arkeologiska underskningar. Stockholm.
Arnold, D. E., 1985. Ceramic Theory and Cultural Process. Cambridge University
Press. Cambridge.
Arnold, J., 1993. Labor and Rise of Complex Hunter-Gatherers. Journal of
Anthropological Archaeology 12:75-119.
1996. The Archaeology of Complex Hunter-Gatherers. Journal of Archaeological
Method and Theory, vol. 3, no. 2:77-126.
Arnold, V., 1990. Refitting of Waste Material from Dagger Production of Site
Tegelbarg. In: Cziesla, E., Eickhoff, N., Arts, N. & Winter, D. (eds.), The Big
Puzzle. International Symposium on Refitting Stone Artefacts, Mon Repos, 1987.
Holos, Bonn.
Artursson, M., (ed.) 2000. Stngby stationssamhlle. Boplats- och bebyggelselmningar
frn senneolitikum till yngre jrnlder Skne, Vallkrra sn, vg 930. UV syd,
rapport 2000:79. Arkeologisk frunderskning och underskning. Malm.
Asplund, J., 1970. Om undran infr samhllet. Argos. Uppsala.
Bakka, E., 1976. Fire vestnorske kystbuplaser fr yngre steinlder. Iskos 1:
Bantelmann, N., 1982. Endneolithische Funde im rheinisch-westflischen Raum. Karl
Wachholtz Verlag. Neumnster.
Barth, F., 1953. Principles of Social Organisation in Southern Kurdistan. Universitets
Ethnografiske Museum Bulletin, No. 7. Oslo.
1981. Process and Form in Social Life. Selected Essays of Fredrik Barth: Volume 1.
Routledge & Keegan Paul. London.
344
References
Becker, C.-J., 1951a. Flintgruberne ved Aalborg. Nationalmuseets Arbejdsmark 1951:
107-112.
1951b. Late-Neolithic Flint Mines at Aalborg. Acta Archaeologica, vol. XXII: 135152.
1952. Die Nordschwedischen flintdepots. Ein beitrag zur geschichte des
neolithischen fernhandels in Skandinavien. Acta Archaeologica XXIII: 31-79.
1957. De stortandade flint-spyd og deres alder. Aarbger for nordisk oldkyndighed
og historie 1956: 205-220.
1959. Flintmining in Neolithic Denmark. Antiquity, vol. 23: 87-92.
1980. Hov, Gem. Sennels, Amt Thisted, Jtland. In: Weisgerber, G. (ed.), 5000
Jahre Feuersteinbergbau. Deutsche Bergbau Museum, Bochum:457-464.
1988. Sydskandinavisk flint i nordsvenske fund fra mellem- og sen-neolitisk tid.
Iskos 7: 43-52.
1993. Flintminer og flintdistribution ved Limfjorden. Limfjordsprojektet. Rapporter
nr. 6: 111-134.
Berg, E., 1998. Intensjonelle nedlegginger av flintgjenstander i Rygge og Rde,
stfold. Beskrivelse og analyse av tre funn og funnomstndigheter. Universitetes
Oldsaksamling 1997/1998:43-62.
Bernstein, R. J., 1987., Bortom objektivism och relativism. Rda Bokfrlaget.
Gteborg.
Bertalanffy, L. von, 1968. General System Theory. New York.
Bertvall, C., 1985. En studie av flintdeperna i Vsterbotten. BA-thesis. Ume
University.
Beuker, J. R. & Drenth, E., 1998. Scandinavishe dolken in Drenth. Van
Rendierjager tot Ontginner, Nieuwe oudheidkundige ontdekkingen in Drenthe XLIV:
3-33.
Binford, L. R., 1968. Post-pleistocene adaptations. In: Binford, L. & Binford, S.
(eds.), New Perspectives in Archaeology. Chicago:313-341.
1972. An Archaeological Perspective. Academic Press. New York.
1982. Objectivity-Explanation-Archaeology-1981. In: Binford, L., 1983.
Working at Archaeology. Academic Press. New York, pp. 45-56.
1983a. In Pursuit of the Past. Decoding the Archaeological Record. Thames and
Hudson. New York.
1983b. Working at Archaeology. Academic Press. New York.
1989. Debating Archaeology. Academic Press. New York.
Bjrhem, N. & Sfvestad, U., 1989. Fosie IV. Byggnsadstradition och
bosttningsmnster under senneolitikum. Malmfynd 5. Malm.
Bloch, M., 1974. Symbols, song, dance and the features of articulation: is religion an
extreme form of traditionary authority? Archives Europeenes de Sociologie 15:55-81.
1989. Ritual, History and Power. Selected Papers in Anthropology. Monographs on
Social Anthropology 58. Th Athlone Press. London and Atlantic Highlands, N.J.
Bloemers, J. H. F., 1969. Flintdolche vom skandinavischen Typus in den
Niederlanden. Berichten van de Rijksdienst voor het Oudheidkundig
Bodemonderzoek 18: 47-100.
345
Jan Apel
Boas, F., 1955. Primitive Art. New York.
Boas, N. A., 1980. Egehjbopladsen fra aeldste bronzealder. Thrane, H., (ed.),
Bronzealderbebyggelsen i Norden. Odense:103-121.
1983. Egehj. A Settlement from the Early Bronze Age in East Jutland. Journal of
Danish Archaeology vol 8, 1989: 88-107
1986. Tidlige senneolitiske bosttelser p Djursland. I: C. Adamsen & K Ebbesen
(eds.), Stridsxetid i Sydskandinavien. Arkeologiske skrifter 1: 318-324.
Kpenhamn.
1991. Bronze Age Houses at Hemmed Church, East Jutland. Journal of Danish
Archaeology vol 8, 1989: 88-107
1993. Late Neolithic and Bronze Age Settlements at Hemmed Church and
Hemmed Plantation, East Jutland. Journal of Danish Archaeology vol 10, 1991:
119-135.
Bogucki, P. I., 1987. Forest Farmers and Stockherders. Early Agriculture and its
Consequences in North-Central Europe. New Studies in Archaeology. Cambridge.
Bolin, H., 1999. Kulturlandskapets korsvgar. Mellersta Norrland under de tv sista
rtusendena f. Kr. Stockholm Studies in Archaeology 19. Stockholm.
Bondesson, B., 1980. Sknska flintdolkar. En jmfrelse av det sknska och det danska
flintdolksmaterialet med utgngspunkt frn E Lomborgs Die Flintdolche Dnemarks.
BA-thesis, Univerity of Lund.
Bonnichsen, R, 1977. Models for Deriving Cultural Information from Stoone Tools.
National Museum of Man. Archaeological Survey of Canada. Mercury Series 60.
Ottawa.
Borenius-Jrpeland. ms. Piparberget. Rapportmanuskript frn Riksantikvariembete
UV Mitt. Stockholm.
Boserup, E., 1966. The Conditions of Agricultural growth. London.
Bourdieu, P., 1958. Sociologie de lArgerie. Oue sais-je?; 802. Paris.
1977. Outline of a Theory of Practice. Cambridge University Press. Cambridge.
1986. Kultursociologiska texter. Salamander frlag. Liding.
1991. Kultur och kritik. Daidalos. Gteborg.
1996. Homo Academicus. Brutus stlings Bokfrlag. Stockholm/Stehag.
1999a. Praktiskt frnuft. Bidrag till en handlingsteori. Daidalos. Gteborg.
1999b. Den manliga dominansen. Daidalos. Gteborg.
Bradbury, A. P. & Carr, P. J., 1999. Examining Stage and Continuum Models of
Flake Debris Analysis: An Experimental Approach. Journal of Archaeological
Science 26:105-116.
Bradley, B., 1975. Lithic Reduction Sequences: A Glossary and Discussion. In:
Swanson, E. (ed.), Lithic Technology: Making and Using Stone Tools. Mouton,
The Hague:5-14.
1978. Hard Hammer Soft Hammer: an Alternate Explanation. Flintknappers
Exchange 1 (2):8-10.
Braidwood, A. P. & Howe, B., 1960. Prehistoric Investigations in Iraqi Kurdistan.
Oriental Institute Studies in Ancient Oriental Civilization. Chicago.
346
References
Brew, J. O., 1946. Archaeology of Alkali Ridge, Southeastern Utah. Papers of the
Peabody Museum of Archaeology and Ethnology. Harvard University.
Broady, D., 1991. Sociologi och epistemologi. Om Pierre Bourdieus frfattarskap och den
historiska epistemologin. Andra korrigerade upplagan. HLS Frlag. Stockholm.
2000. Frord. In: Bourdieu, P., Konstens Regler. Det litterra fltets uppkomst och
struktur. Brutus stlings Bokfrlag. Stockholm/Stehag.
Broady, D. & Palme, M., 1986. Utgivarnas frord. I: Bourdieu, P., 1986.
Kultursociologiska texter. Salamander frlag. Liding.
Brown, J., 1995. Traditional Metalworking in Kenya. Oxbow Monographs 44.
Cambridge. Monographs in African Archaeology 38. Oxford.
Brumfield, E. M. & Earle, T. K. (eds.) 1987. Specialization, exchange, and complex
societies. Cambridge Universitey Press.
Brndstedt, J., 1957. Danmarks Oldtid I. Stenalderen. Kpenhavn.
Burstrm, M., 1999. Cultural Diversity in the Home Ground. How Archaeology
Can Make the World a Better Place. Current Swedish Archaeology vol. 7:21-26.
Butler, J. J., 1975. Comments on the Flint Daggers of Denmark. Norwegian
Archaeology Review vol. 8, no. 2:
Callahan, E., 1979. The Basic of Biface Knapping in the Eastern Fluted Point
Tradition. Archaeology of Eastern North America 7:1-180.
1987. An Evaluation of the Lithic Technology in Middle Sweden during the
Mesolithic and Neolithic. Aun 8. Uppsala.
1996. The Basics of Biface Knapping in the Eastern Fluted Point Tradition: A
Manual for Flintknappers and Lithic Analysts (second edition). Archaeology of
Eastern North America 7:1-180.
Callahan, E. & Apel, J., ms. The Production of the Prestigious Flint Daggers of Type IV
Flint Daggers.
Cammann, S. V. R., 1977. The Cult of the Jambiya: Dagger Wearing in Yemen.
Expedition. The magazine of Archaeology/Anthropology Vol. 19, No. 2: 27-34.
Campbell, T., 1986. Sju teorier om samhllet. Akademilitteratur. Stockholm.
Cassau, A., 1935. Ein Feuerstaindolch mit Holzgriff und Lederschiede aus
Wiepenkathen, Kreis Stade. Mannus XXVII:199-209.
Childe, V.G., 1928. The Most Ancient East: The Oriental Prelude to European
Prehistory. London.
Childs, T. S., 1999. After All, a Hoe Bought a Wife: The Social Dimensions of
Ironworking among the Toro of East Africa. In: Dobres, M-A. & Hoffman, C. R.
(eds.), The Social Dynamics of Technology. Smithsonian Institution Press.
Washington and London.
Clark, J. E. & Blake, M., 1996. The Power of Prestige: Competitive Generosity and
the Emergence of Rank Societies in Lowland Mesoamerica. In: Preucel, R. W. &
Hodder, I. (eds), Contemporary Archaeology in Theory - a reader. Blackwell. Oxford:
258-182.
Clark, J. E. & Parry, W. J., 1990. Craft Specialization and Cultural Complexity.
Research in Economic Anthropology, Vol. 12:289-346.
347
Jan Apel
Clark, J. E., 1986. Craft specialization and cultural complexity. Unpublished paper,
Department of Anthropology, University of Michigan.
Clark, J. G. D., 1932. Scandinavian Flint Daggers in Britain. Man 1932, no.
223:186-190.
Coser, L., 1984. Introduction. In: Durkheim, E., The Division of Labour in Society.
MacMillan Press ltd. London.
Crabtree, D. E., 1972. An Introduction to Flintworking. Occasional Papers of the
Idaho State University Museum, Number 20. Pocatello, Idaho.
Cullberg, C., 1960. Der Dolch von Ntraby. Meddelanden frn Lunds universitets
historiska museum 1960: 34-42.
Czerniak, L. & Kabacinski, J., 1997. The Erteblle Culture in the Southern Baltic
Coast. In: Krl, D. (ed.), The Built Environment of Coast Areas during the Stone
Age. A Symposium at the Centenary of Archaeological Excavations at Rzucewo 4-9
October 1994. Rzucewo-Gdansk.
Damm, C. B., 1991a. Continuity and Change. An Analysis of Social and Material
Patterns in the Danish Neolithic. Ph.D. Thesis. Dpt. of Archaeology, University of
Cambridge.
1991b. The Danish Single Grave Culture Etnic Migration or Social
Construction? Journal of Danish Archaeology, Vol. 10:199-205.
Danielsson, T., 2000. Norr om neoliticum? Neolitiska fynd i Dalarna. Unpublished
BA-thesis, Inst. fr Humaniora, Mitthgskolan. stersund.
Darwin, C. R., 1868. The Variation of Animals and Plants under Domestication.
Murray. London.
1976 [1859]. Om arternas uppkomst genom naturligt urval, eller de bst utrustade
rasernas bestn i kampen fr tillvaron. Natur och kultur. Stockholm.
Davy, G., 1960. . Durkheim. Revue francaise de sociologie I:3-24.
Deetz, J. & Dethlevsen, E., 1965. The Doppler Effect and Archaeology. A
Consideration of the Spatial Aspects of Seriation. Southwest Journal of Archaeology
vol. 21.
Dibble, H. L. & Pelcin, A., 1995. The effect of hammer mass and velocity on flake
mass. Journal of Archaeological Science 22: 429-439.
Divac, G. & Sedlcek, Z., 1999. Hortfund der Altbronzezeitlichen dolche von Praha 6Suchdol. Museum Nationale Pragae. Prag.
Dobres, M-A. & Hoffman, C. R. (eds.), 1999. The Social Dynamics of Technology.
Smithsonian Institution Press. Washington and London.
Domanski, M. & Webb, j. A., 2000. Flaking properties, petrology and use of Polish
flint. Antiquity 74:822-832.
Durkheim, E., 1978 [1895]. Sociologins metodregler. Bokfrlaget Korpen. Gteborg.
1984 [1893]. The Division of Labour in Society. Macmillan. London.
1993 [1897]. Sjlvmordet. Argos. Lund.
1995 [1912]. The Elementary Forms of Religious Life. The Free Press. New York.
Durkheim, E. & Mauss, M., 1963. Primitive Classification. The University of
Chicago Press. Chicago.
348
References
Earle, T., 1977. A reappraisal of redistribution: complex Hawaiian chiefdoms. In
Earle, T. K. & Ericson, J. (eds), Exchange Systems in Prehistory. Academic Press,
New York: 213-232.
1978. Economic and Social Organisation of a Complex Chiefdom: the Halelea
District, Kauai, Hawaii. Museum of Anthropology, University of Michigan,
Anthropological Paper 63. Ann Arbor.
1991. The Evolution of Chiefdoms. In: Earle, T. (ed.), 1991. Chiefdoms: Power
and Economy, and Ideology. Cambridge University Press. Cambridge.
(ed.), 1991. Chiefdoms: Power and Economy, and Ideology. Cambridge University
Press. Cambridge.
Earle, T., Bech, J-H., Kristiansen, K., Aperlo, P., Kelertas, K. and Steinberg, J., 1998.
The Political Economy of Late Neolithic and Early Bronze Age Society: the Thy
Archaeological Project. Norwegian Archaeological Review, Vol. 31, No. 1:1-28.
Ebbesen, K., 1975a. Comments on the flint daggers of Denmark. Norwegian
Archaeological Review vol. 8. nr 2:107-111.
1975b. Die jngere Trichterbecherkultur auf den dnischen inseln. Arkaeologiska
studier II. Kpenhavn.
1980. Flintafslag som offer. Kuml 1980:147-157.
1994. Tragtbaegarkulturens dolkstave. Aarbger for nordisk oldkyndighed og
historie 1992:103-136.
Eklf, N., 1999. stertlje 1:15. Rapport, Riksantikvariembetet UV mitt.
Stockholm.
Ekmyr-Westman, E., 1991. Upl. Skogs-Tibble sn, Skillsta 2:15, ra 138. Rapport
1991. Ume.
Elster, J., 1990. Vetenskapliga frklaringar. Bokfrlaget Korpen. Gteborg.
Englund, L-E., 2000. Smeder och makt. I: Hjrthner-Holdar, E. & Risberg, C.
(eds.), Hantverkets roll i samhllet produktion och reproduktion. Rapport frn den
tredje workshopen om hantverksspecialisering Uppsala 18-19 mars 1999.
Forskningsrapport R0004, Riksantikvariembete UV GAL. Uppsala.
Ericson, J. E. & Earle, T. K. (eds.) 1982. Contexts for Prehistoric Exchange. Academic
Press. New York.
Eriksen, T. H., 1993. Ethnicity and Nationalism. Anthropological Perspectives. Pluto
Press. London.
Ethelberg, G., 2000. Det snderjyske landbrugs historie: sten- og bronzealder . Haderslev
Museum: Historisk Samfund for Snderjylland. Haderslev.
Fabech, C., 1988. Storstenkisten fra Blaere. Kuml 1988: 45-75.
Fagerlund, D. & Hamilton, J., 1995. Arkeologi p vg. En hllkista och bebyggelse frn
frn senneolitikum och bronslder. Riksantikvariembetet UV Uppsala, rapport
1995:13. Uppsala.
Featherstone, M., 1994. Kultur, kropp och konsumtion. Symposion. Stockhom/Stehag.
Fields, K., 1995. Religion as an Eminently Social Thing. Translators Introduction in:
Durkheim, E., 1995. The Elementary Forms of Religious Life. The Free Press. New
York.
Firth, R., 1964. Essays on Social Organization and Values. The Athlone Press. London.
349
Jan Apel
Flenniken, J., 1985. Stoone Tool Reduction Techniques as Cultural Markers. In: Plew,
M. G., Woods, J. C. and Pavesic, M. G. (eds), Stone Tool Analysis: Essays in Honor
of Don E. Crabtree. Albuquerque: University of New Mexico Press:265-276.
Flenniken, J., Patterson, L. W. & Hayden, B., 1979. More on Staging. Flintknappers
Exchange 2 (1):26
Ford, J. A., 1954. The Type Concept Revisited. American Anthropologist 56 (1):4254.
Forsberg, L., 1985. Site Variability and Settlement Patterns. An Analysis of the
Hunther-Gatherer Settlement System in the Lule Valley, 1500 BC-BC/AD.
Archaeology and Environment 5. Ume.
1988. Bosttningsmnster vid Lule och Ume lv under bronslder och
frromersk jrnlder. Arkeologi i norr 1: 51-101.
1992. De norrlndska hllristningarna sociala kontext - Alternativa tolkningar.
Arkeologi i norr 3, s. 55-70.
Forssander, J. E., 1936. Der Ostskandinavische Norden whrend der ltesten Metallzeit
Europas. Skrifter utgivna av Kungl. Humanistiska Vetenskapssamfundet i Lund
XXII. Lund.
Geertz, C., 1957. Ritual and Social Change: A Javanese Example. American
Anthropologist 59:32-54.
1973. The Interpretation of Cultures. Selected Essays by Clifford Geertz. Hutchinson
of London. London.
Geijerstam, M. af, 1996. Mikrodebitageanalys som metod vid arkeologiska
underskningar. En inledande metodutvrdering med den neolitiska lokalen Ra
162 vid Fgelbacken som analysomrde. In: Lekberg, P. (ed.), 1996. Fgelbacken
- ett fornlmningskomplex i stra Vstmanland. Del II. Lmningar frn
mellanneolitikum, senneolitikum och jrnlder underskta 1994. Rapport till
lnsstyrelsen frn Arkeologikonsult AB i samarbete med Arkeologiska
institutionen vid Uppsala universitet.
Gejvall, N., 1963. Skelettmaterialet i Dragby hllkista. Tor IX: 85-122.
Geneste, J-M., 1989. Les Industries de la Grotte Vaufrey: Technologie du Dbitage,
Economie et Circulation de la Matiere Premire. In: Rigaud, J. P. (ed.), La Grotte
Vaufrey. Mmoires de la Socit Prhistorique Francaise, vol. 19:441-517.
Gerloff, S., 1975. The Early Bronze Age Daggers in Great Britain, and a
Reconsideration of the Wessex Culture. Prhistorische Bronzefunde. VI.2. Mnchen.
Giddens, A., 1978. Durkheim. Fontana. Glasgow.
1984. The Constitution of Society. Outline of the Theory of Structuration. Polity Press.
Cambridge.
1987. Social Theory and Modern Society. Polity Press. Cambridge.
1990. The Consequences of Modernity. Stanford university press. Stanford.
Glob, P V, 1951. En flintsmedie p Forns. Kuml 1951: 23-39
1945. Studier over den jyske Enkeltgravskultur. Aarbger for Nordisk oldkyndighed
og historie 1944: 1-23.
Godelier, M., 2000. Is Social Anthropology Still Worth the Trouble? A Response to
Some Echoes from America. Ethnos, vol. 65:3:301-316.
350
References
Goldhahn, J., 1999. Sagaholm. Hllristningar och gravritual. Studia Archaeologica
Umensis 11. Jnkpings Lns Museums Arkeologiska Rapportserie 41.
Jnkping.
Grace, R., 1990. The Limitations and Applications of Use Wear Analysis. In:
Grslund, B., Knutsson, H., Knutsson, K. & Taffinder, J. (eds.), The
interpretative possibilities of microwear studies. Proceedings of the International
conference on lithic use wear analysis, 15th-17th February 18989 in Uppsala,
Sweden. Aun 14, Uppsala.
Granath Zilln, G., 1999. En deposition i ett senneolitiskt landskap. I: Olausson, M.
(red.), Spiralens ga Tjugo artiklar kring aktuell bronsldersforskning.
Riksantikvariembetet, UV mitt. Stockholm:321-346.
Green, S., Houlder, C. H. & Keeley, L. W., 1982. A Flint Dagger from Ffair Rhos,
Ceredigion, Dyfed, Wales. Proceedings of the Prehistoric Society 48: 492-501.
Grimes, W. F., 1932. The Early Bronze Age Flint Dagger in England and Wales.
Proceedings of the Prehistoric Society East Anglia 6:340-355.
Grslund, B., 1974. Relativ datering. Om kronologisk metod i nordisk arekologi. Tor
XVI. Uppsala.
1996. Arkeologisk datering. Studentlitteratur. Lund.
Guneriussen, W., 1997. Aktr, handling och struktur. Studentlitteratur. Lund.
Gurstad-Nilsson, H., 1999. En bronslders framvxt och konsolidering Om
kontinuitet och social strategi i Mre under senneolitikum och bronslder. I:
Olausson, M. (ed.), Spiralens ga. Tjugo artiklar kring aktuell bronsldersforskning.
Riksantikvariembete, Avdelningen fr arkeologiska underskningar, Skrifter nr
25:183-214.
Hagberg, L., (1999). Grav, rit, boplats. Handlingssfrer i det senneolitiska
kulturlandskapet. BA Dissertation, Department of Archaeology, Uppsala
University. Uppsala.
Hallesj, M. & Hamilton, J., 1996. Maren. Underskningsrapport,
Riksantikvariembetet, UV mitt. Stockholm.
Hallgren, F., 1998. Etnicitet under stenldern i Mellansverige och sdra Norrland. I:
Johnsen, B. & Welinder, S. (red.). Etnicitet eller kultur. Mitthgskolan. stersund.
2000. Lmningar frn stridsyxekulturen p Fgelbacken, Hubbo sn,
Vstmanland. Tor 30:5-34.
Hallgren, F., Bergstrm, . and Larsson, ., 1995. Prlngsberget en kustboplats frn
vergngen mellan senmesolitikum och tidigneolitikum. Tryckta Rapporter frn
Arkeologikonsult AB, nr. 13. Slutunderskningsrapport. Upplands Vsby.
Hallgren, F., Djerw, U., af Geijerstam, M. and Steineke, M., 1997. Skogsmossen, an
Early neolithic settlement site and sacrificial fen in the northern borderland of the
Funnel-beaker Culture. Tor 29:49-112.
Hansen, F., 1929. Neolitiska paleolitica. Fornvnnen: 247-249
Hansen, M., 1986. Enkeltgravskulturens gravmateriale fra Sydvstjylland. I: C
Adamsen & K Ebbesen (ed:s), Stridsksetid i Sydskandinavien. Arkeologiska
skrifter 1: 77-83
351
Jan Apel
Hansen, P. Vemming & Madsen, B., 1983. Flint Axe manufacture in the Neolithic.
An Experimental Investigation of a Flint Axe Manufacture site at Hastrup
Vaenget, East Zealand. Journal of Danish Archaeology. Vol. 2: 43-59.
Harrison, R. J., 1980. The Beaker Folk. Copper Age archaeology in Western Europe.
Thames and Hudson. London.
Harrison, R. J., Jackson, R., & Napthan, M., 1999. A Rich Bell Beaker Burial From
Wellington Quarry, Marden; Hereforshire. Oxford Journal of Archaeology 18 (1): 116.
Hatch, E., 1973. Theories of Man and Culture. Columbia University Press. New
York.
Hayden, B. & Hutchings, W., 1989. Whither the billet flake? In: Amick, D. S. &
Maulding, R. P. (eds.), Experiments in Lithic Technology. BAR International series
528:235-258.
Hayden, B., 1990. Nimrods, Piscators, Pluckers, and Planters: The Emergence of
Food Production. Journal of Anthropological Archaeology 9:31-69.
1995. Pathways to Power. Principles for Creating Socioeconomic Inequalities. I:
Price, T. D. and Feinman, G. M. (red.), Foundations of Social Inequality, Plenum
Publishing. New York: 15-86.
1998. Practical and Prestige Technologies. The Evolution of Material Systems.
Journal of Archaeological Method and Theory, vol. 5, no. 1:1-55.
Hedeager, L. & Kristiansen, K., 1988. Oldtid O. 4000 f. Kr 1000 e. kr. I: Bjrn,
C. (ed.), Det danske landbrugs historie I. Oldtid og middelalder. Landbohistorisk
selskab. Odense.
Hegardt, J., 1997. Relativ betydelse. Individualitet och totalitet i arkeologisk kulturteori.
Occasional Papers in Archaeology 12. Uppsala.
Heidegger, M., 1972 [1927]. On Time and Being. New York.
Heimann, C., 2000. Neolitiska Landskap vid Stora Le en studie av
neolitiseringsprocessen in vrmlndsk stenlder. Sammanlggningsuppsats fr fil. licexamen i arkeologi, seminarieupplaga. Department of Archaeology. Gothenburg
University.
Helms, M., 1998. Access to origins. Affines, Ancestors and Aristocrats. University of
Texas Press. Austin.
Herbert, E. W., 1993. Iron, Gender, and Power. Rituals of Transformation in African
Societies. Indiana University Press. Bloomington.
Hodder, I., 1986. Reading the Past. Current Approaches to Interpretation in
Archaeology. Cambridge University Press. Cambridge.
1990. Commentary to Technology in the Humanities. Archaeological Review of
Cambridge vol. 9:1:154-157.
1992. Material Practice, Symbolism and Ideology. I: Hodder, I., Theory and
Practice in Archaeology. Routledge.
Hodder, I. & Lane, P., 1982. A Contextual Examination of Neolithic Axe
Distribution in Britain. In: Ericson, J. E., Earle, T. K. (eds.), Contexts for Prehistoric
Exchange. Academic Press, New York:213-233.
352
References
Hoika. J. A. J., 1998. Trade and Communication within the Neolithic Funnel Beaker
Culture (TRB) of the Baltic Region. Digital document: http://209.217.18.237/
HoikaSAA1998.htm.
Holm, J., Olsson, E. & Weiler, E., 1997. Kontinuitet och frndring i senneolitikum.
I: Larsson, M. & Olsson, E. (red), Regionalt och interregionalt.
Stenldersunderskningar i Syd- och Mellansverige. Riksantikvariembetet
Arkeologiska Underskningar. Skrifter nr 23. Stockholm:217-236.
Holm, L., 1985. Variations in Flake Material. Different Utilization of Stone material.
In: In Honorem Evert Badou. Archaeology and Environment 4: 303-311.
1991. The Use of Stone and Hunting of Reindeer. A Study of Stone Tool Manufacture
and Hunting of Large Mammals in the Central Scandes c. 6 000-1 BC.
Archaeology and Environment 12. Ume.
Holmes, W. H., 1894. Natural history of flaked stone implements. Memoirs of the
International Congress of Anthropology, Chicago, pp. 120-139.
Horton, R., 1973. Lvy-Bruhl, Durkheim and the Scientific Revolution. In: Horton,
R. & Finnegan, R. (eds), Modes of Thought. Essays on Thinking i Western and NonWestern Societies. Faber & Faber. London:249-305.
Hvass, S., 1977. A House of the Single Grave Culture Excavated at Vorbasse in
Central Jutland. Acta Archaeologica 48.
1986. En boplads fra enkeltgravskulturen i Vorbasse. I: Adamsen C. & Ebbesen,
K. (eds.), Stridsksetid i Sydskandinavien. Beretning fra et symposium 28.30.X.1985 i Vejle. Arkeologiske skrifter 1. Forhistorisk Arkaeologisk institut,
Kpenhamn Universitet.
Hkansson, T., 1998. Rulers and Rainmakers in Precolonial South Pare, Tanzania:
Exchange and Ritual Experts in Political Centralization. Ethnology vol. 37, no.
3:263-283.
Hgberg, A., 1998. Manual till referenssamling fr flinta. Analys av avslag frn
tillverkningen av fyrsidiga yxor och bifacialt dolkmne. Stadsantikvariska
avdelningen Malm Museer 1998. Malm.
1999. Child and Adult at a Knapping Area. A technological Flake Analysis of the
Manufacture of a Neolithic Square Sectioned Axe and a Childs Flintknapping
Activities on an Assemblage excavated as Part of the resund Fixed Link Project.
Acta Archaeologica vol. 70-1999: 79-106.
2000. En bok om freml. I: Hgberg, A. (ed.), Artefakter arkeologiska ting. En
bok om freml ur ett arkeologiskt perspektiv. University of Lund, Institute of
Archaeology, Report Series nr. 71/Stadsantikvariska Avdelningen, Kultur malm.
Malm.
Jaanusson, H., 1969. Hllkistan frn Bjurhovda i Vsters. Fornvnnen 1969: 149160.
Jaanusson, H. & Silvn, U., 1962. Underskningen av Dragbyrset nr 88.
Fornvnnen 1962: 5-44.
Jger, A & Laursen, J, 1983. Lindebjerg and Rjle Mose, Two Early Bronze Age
Settlements on Fyn. Journal of Danish Archaeology vol 2: 102-117
Jensen, J. A., 1973. Bopladsen Myrhj. 3 hustomter med klokkebgarkeramik. Kuml
1972: 61-114.
353
Jan Apel
Jensen, J., 1979. Oldtidens samfund. Tiden indtil r 800. Dansk social historie 1.
Gyldendal. Kpenhavn.
Jeppesen, J., 1984. A Late Neolithic/Early Bronze Age Settlement at Vejlby, East
Jutland. Journal of Danish Archaeology vol. 3:99-103.
Jevons, W. S., 1874. The Principles of Science Volume 2. Macmillan & Co. New York.
Johansson, B. O. H., 1961. Stenldershllkistor i svensk-norska grnslandskap. Tor
vol. XII: 5-52.
Johnson, L. L., 1978. A History of Flint-Knapping Experimentation, 1838-1976.
Current Anthropology Vol. 19, No. 2, pp. 337-372.
Johnson, T. & Prescott, C., 1993. Late Neolithic houses at Stokkset, Sande in
Sunnmre. Arkeologiske Skrifter, Historisk museum, Universitet i Bergen, No. 7:2270.
Jrgensen, L. B., 1992. North European Textiles until AD 1000. Aarhus University
Press. Aarhus.
Kaelas, L., 1964. Senneolitikum i Norden. Tor X:135-148
Karsten, P, 1994. Att kasta yxan i sjn. En studie ver rituell tradition och frndring
utifrn sknska neolitiska offerfynd. Acta Archaeologica Lundensia Series in 8,
No 23. Lund.
Kempfer-Jrgensen, L. & Liversage, D., 1985. Mere om Sejers forhistorie. En
stenaldersboplads, et kulthus og et flintvaerksted. Fra Holbaek Amt 1985:7-27.
Knarrstrm, B., 2000a. Flinta i sydvstra Skne. En diakron studie av rmaterial,
produktion och funktion med fokus p boplatskronologi och metalltida
flintutnyttjande. Acta Archaeologica Lundensia. Series in 8, No. 33. Lund.
Knarrstrm, B., 2000b. versiktlig analys av flintmaterialet och vrigt stenmaterial. I:
Artursson, M., (ed.) 2000. Stngby stationssamhlle. Boplats- och
bebyggelselmningar frn senneolitikum till yngre jrnlder Skne, Vallkrra sn, vg
930. UV syd, rapport 2000:79. Arkeologisk frunderskning och underskning.
Malm.
Knutsson, H., 1995. Slutvandrat? Aspekter p vergngen frn rrlig till bofast tillvaro.
Aun 20. Uppsala.
1999. Two Technologies Two Mentalities. Urgeschichtliche Materialhefte 14,
Tbingen 1999:37-58.
2000. Rituella och funktionella yxor. I: Hjrthner-Holdar, E. & Risberg, C. (eds.),
Hantverkets roll i samhllet - produktion och reproduktion. Rapport frn den tredje
workshopen om hantverksspecialisering Uppsala 18-19 mars 1999.
Riksantikvariembetet UV GAL, forskningsrapport R0004. Uppsala.
in prep. Traces in flint and the movement of the agricultural package through
Europe. To be published in Documenta Praehistorica XXVII, University of
Ljubljana.
Knutsson, K., 1986. Sammanfogning av flintorna p Kusmarkdepn: Kombewametoden belagd p de nordsvenska flintfyndplatserna. Fjlnir 5:2: 3-20.
1988a. Making and using stone tools. The analysis of the lithic assemblages from
Middle Neolithic sites with flint in Vsterbotten, northern Sweden. Aun 11.
Uppsala.
354
References
1988b. Patterns of tool use. Scanning electron microscopy of experimental quartz tools.
Aun 10. Uppsala.
1989. C-14datings and the relative chronolgy of the Bjurselet flake scatter units.
In: Christiansson, H. & Knutsson, K. (eds.), The Bjurselet Settlement III. Volume 1.
Occasional Papers in Archaeology, Societas Archaeologica Upsaliensis. Uppsala.
1990. A New Lithic Scene. The Archaeological Context of Used Tools. In:
Grslund, B., Knutsson, H., Knutsson, K. & Taffinder, J. (eds), The Interpretative
Possibilities of Microwear Studies. Proceedings of the international conference on lithic
use-wear analysis, 15th-17th February 1989 in Uppsala, Sweden. Aun 14.
Uppsala.
1998. Convention and lithic analysis. Holm, L. & Knutsson, K. (eds.), Proceedings
from the Third Flint Alternatives Conference at Uppsala, Sweden, October 18-20,
1996. Occasional Papers in Archaeology 16, Uppsala: 71-94.
in prep. En bok om stenar och mnniskor. Institutionen fr arkeologi. Uppsala
universitet.
Knutsson, K. & Lindgren, K., in press. Making Sense of Quartz.
Riksantikvariembetet, UV mitt, Stockholm.
Knutsson, K. & Lekberg, P., 1996. Utgrvningsmetod vid Fgelbacken 1993 och
1994. I: Lekberg, P. (red.), Fgelbacken ett fornlmningskomplex i stra
Vstmanland. Del I. Slutunderskningsrapport frn Arkeologikonsult AB.
Upplands Vsby.
Kristiansen, K., 1982. The Formation of Tribal Systems in later European Prehistory:
Northern Europe, 4000-500 B.C. In. Renfrew, C. & Rowlands, M. J. and
Segraves, B. A. (eds.) Theory and Explanation in Archaeology. Academic press. New
York: 241-280.
(ed.), 1985. Archaeological Formation Processes. The representativity of archaeological
remains from Danish Prehistory. Kpenhamn.
1987. From Stone to Bronze: The evolution of social complexity in northern
Europe, 2300-1200 B.C. In: Brumfiel, E. M. & Earle, T. K. (eds.),
Specialization, Exchange, and Complex Societies. Cambridge University press.
Cambridge.
1991. Chiefdoms, states, and systems of social evolution. In: Earle, T., (ed.),
Chiefdoms: Power, Economy, and Ideology. Cambridge University Press. Cambridge.
Khn, H. J., 1979. Das Sptneolithikum in Schlesvig-Holstein. Offa-Bcher, Band
40. Neumnster.
Knigsson, L.-K., 1989. Appendice 1. Traces of Neolithic Human Influence upon
the Lanscape Development at the Bjurselet Settlement, Vsterbotten, Northern
Sweden. In: Christiansson, H. & Knutsson, K. (eds), The Bjurselet Settlement III
volume 2. Societas Archaeologica Upsaliensis. Uppsala:133-141.
Lagers, P., 2000. Burial rituals inferred from palynological evidence: results from a
Late Neolithic stone cist in southern Sweden. Vegetation History and Archaeology
9: 169-173.
Lannerbro, R., 1991. Det sdra fngstlandet. Katalog, del I. Vann. Stockholm
Archaelogical Reports. Field Studies Nr 2, 1991. Stockholm.
355
Jan Apel
1992. Det sdra fngstlandet. Katalog, del II. Orelven. Stockholm Archaelogical
Reports. Field Studies Nr 3, 1992. Stockholm.
Lanting, J. N. & van der Waals, J. D., 1976. Beaker culture relations in the Lower
Rhine Basin. In: Lanting, J. N. & van der Waals, J. D. (eds.),
Glockenbechersymposion Oberried 1974. Bussum Haarlem:1-80.
Larsson, L., 2000. Axes and Fire Contacts with the Gods. In: Olausson, D. &
Vandkilde, H. (eds), Form, Function and Context. Material culture studies in
Scandinavian archaeology. Almqvist & Wiksell International. Lund:93-104.
Leach, E., 1973. A View From the Bridge. In: Spriggs, M. (ed.), Archaeology and
Anthropology: areas of mutual interest. British Archaeological Reports 5.19.
London: 161-176.
Lee, R. B., 1972. The Kung Bushmen of Botswana. In: Bicchieri, M. G. (ed.),
Hunters and Gatherers Today. New York.
Lekberg, P. & Apel, J., ms. Kontextualisering av senneolitiska lsfynd i sterunda sn,
Uppland. Rapport frn ett fltarkeologiskt forskningsprojekt. Manuskript,
Institutionen fr arkeologi och antik historia, Uppsala universitet.
(ed.), 1996. Fgelbacken - ett fornlmningskomplex i stra Vstmanland. Del II.
Lmningar frn mellanneolitikum, senneolitikum och jrnlder underskta 1994.
Rapport till lnsstyrelsen frn Arkeologikonsult AB i samarbete med Arkeologiska
institutionen vid Uppsala universitet.
1997. Enkla skafthlsyxor i gravar. Tor 29: 137-156.
2000a. The Lives and Lengths of Shaft-hole Axes. In: Olausson, D. & Vandkilde,
H. (eds.), Form, Function and Context. Material culture studies in Scandinavian
archaeology. Almqvist & Wiksell International. Lund:155-162.
2000b. Senneolitiska skafthlsyxor i dep- och offersammanhang. Tor 30: 83-106.
2000c. Rvara och teknologi. Ngra aspekter p utformningen av senneolitiska
skafthlsyxor. In: Hjrthner-Holdar, E. & Risberg, C. (eds.), Hantverkets roll i
samhllet produktion och reproduktion. Rapport frn den tredje workshopen om
hantverksspecialisering Uppsala 18-19 mars 1999. Riksantikvariembetet, UV
GAL, Forskningrapport R0004: 35-52.
2001. Den enkla skafthlsyxan i Sverige en ful ankunge i fyndldan. I: Bolin,
H., Kaliff, A. & Zachrisson, T. (eds.), Mellan sten och brons Uppdrags
Leroi-Gourhan, A., 1968. The Art of Prehistoric Man in Western Europe. Thames &
Hudson. London.
Lvi-Strauss, C., 1963. Structural Anthropology. New York.
Lindqvist, S., 1912. Frn Nerikes sten- och bronslder. Meddelanden frn rebro lns
museum V.
Liversage, D, 1988. Mortens Sande 2. A single Grave Camp Site in Nothwest
Jutland. Journal of Danish Archaeology vol. 6, 1987:101-124
Lomborg, E, 1959. Fladehuggede flintredskaber i gravfund fra ldre bronzealder.
Aarbger for Nordisk oldkyndighed og historie 1959:146-183.
1968. Den tidlige bronzealderens kronologi. Et forsg p att fastlaegge graensen
mellem perioderne I og II. Aabger for nordisk Oldkyndighed og Historie 1968:91132.
356
References
1969. Frhbronzezeitliche triangulre Metalldolche in Dnemark. Acta
Archaeologica XXXIX:219-239.
1973. Die flintdolche Dnemarks. Studien ber Chronologie und Kulturbeziehungen
des sdskandinavischen Sptneolithikums. Nordiske Fortidsminder, series B - in
quatro, Bind 1. Kpenhamn.
1975. The Flint Daggers of Denmark. Studies in Chronology and Cultural
Relations of the South Scandinavian Late Neolithic & Reply to Comments.
Norwegian Archaeological Review vol. 8, nr. 2:98-101 & 115-124.
1977. Klokkebaeger- og senere Beaker-indflydelser i Danmark. Et bidrag till
enkeltgravskulturens datering. Aarbger for nordisk Oldkyndighed og Historie
1975:20-41.
Lowenthal, D., 1985. The Past is a Foreign Country. Cambridge University Press.
Cambridge.
Lukes, S., 1973a. On the Social Determination of Truth. In: Horton, R. & Finnegan,
R. (eds.), Modes of Thought. Faber & Faber. London:230-248.
1973b. mile Durkheim His Life and Works. A Historical and Critical Study. Allen
Lane / Penguin Press. London.
Lbcke, P.(ed.), 1988. Filosofilexikonet.Forum. Stockholm.
Lyotard, J.-F., 1984. The postmodern condition: a report on knowledge. Manchester
University press. Manchester.
MacKenzie, D. & Wajcman, J. (eds.), 1985. The Social Shaping of Technology. Open
University Press. Buckingham, Philadelphia.
Madsen, A P., 1868. Afbildninger af danske Oldsager og Mindesmaerker. Stenalderen.
Madsen, B., 1986. Nogle taxonomiske og nomenklatoriske bemrkningar til studiet
af flintteknologi - experimentelt og arkeologisk. Fjlnir 5/1: 3-28.
Madsen, T., 1978. Perioder og periodovergange i neolitikum: Om forskellige
fundtypers egnehed til kronologisk opdelninger. Hikuin 4: 51-60.
Magne, M. & Pokotylo, D., 1981. A Pilot Study in Bifacial Lithic Reduction
Sequences. Lithic Technology 10 (2-3): 34-47
Malmer, M. P. 1955. Kultplats och gravplats. Utgrvningar i Rbelv och Bromlla.
Handlingar angende Villands Hrad 12.
1962. Jungneolitische Studien. Acta Archaeologica Lundesia Series 8.2. Lund.
1963. Metodproblem inom jrnlderns konsthistoria. Acta Archaeologica Lundensia,
Series in 8, No 3. Lund.
1975. Stridsyxekulturen i Sverige och Norge. Lund.
1988. Konstanter och variabler i det frhistoriska samhllet. Fornvnnen 83: 8997.
Maltn, A., 1981. Vad r kunskap?. Gleerups frlag. Malm.
Mathiassen, T, 1934. Flinthandel i Stenalderen. Fra Nationalmuseets Arbejdsmark
1934:18-22.
Mauss, M., 1979. The Notion of Body Techniques. In: Mauss, M., Sociology and
Psychology. Essays. Routledge & Kegan Paul. London.
Meinander, C. F., 1954. Die Kiukaiskultur. Finska Fornminnesfreningens tidskrift
53. Helsinki.
357
Jan Apel
Mestrovic, S. G., 1992. Durkheim and Postmodern Culture. Aldine de Gruyter. New
York.
Montelius, O., 1872-74. Sveriges forntid. Frsk till framstllning af den svenska
fornforskningens resultat. Atlas. Svenska fornsaker I. stenldern och brnsldern.
Stockholm.
1917. Minnen frn vr forntid I. Stenlder och bronslder - beskrivningar. Norstedt
& sners frlag. Stockholm.
1924. Dolkar, spjut- och pilspetsar av ben och sten. Antikvarisk Tidskrift fr
Sverige XXII (2): 1-90.
Morris, B., 1987. Anthropological Studies of Religion. An Introductory Text. Cambridge
Univercity Press. Cambridge.
Moucha, V., 1963. Die Periodisierung der netice Kultur in Bhmen. Sbornik CSSA
3:9-60.
Muriuki, G., 1979. A History of the Kikuyu 1500-1900. Oxford University Press.
Oxford.
Mller, S., 1888. Ordning av Danmarks oldsaker I, stenalderen.
1897. Vor oldtid. Copenhagen.
1898. De jydske enkeltgrave frn stenalderen. Aarbger for nordisk oldkyndighed og
historie:157-282.
1902. Flintdolkene i den nordiske stenalder. Nordiske Fortidsminder I.
Kpenhamn.
Mylonas, G. E., 1972. Ho taphikos nyklos ton Mykenon, 2. Athens.
Newcomer, M. H., 1971. Some quantitative experiments in handaxe manufacture.
World Archaeology vol. 3 (1):85-93.
Nichols, J., and Callahan, E., 1980. Staging as Demonstrated in the Quantitative
Analysis of Debitage from Three Projectile Point Replicas. Contract Abstracts 1.
Nielsen, P. O., 1981. Stenalderen. Bondestenalderen. Sesams Danmarkshistorie.
Kpenhavn.
Nielsen, F. O. & Nielsen, P. O., 1985. Middle and Late Neolithic Houses at
Limensgard, Bornholm. Journal of Danish Archaeology vol 4, 1985:101-114
Nisbet, R. A., 1963. Sociology as an Art Form. In: Stein, M. & Vidich, A. (eds.),
Sociology on Trial. Englewwod Cliffs, New Jersey.
OConnell, J., 1987. Alyawara Site Structure and its Archaeological Implications.
American Antiquity 52:
Odell, G. H., 1989. Experiments in Lithic Reduction. In: Amick, D. S. and
Mauldin, R. P. (eds.), Experiments in Lithic Technology. BAR International Series
528:163-198.
Olausson, D., 1983. Lithic Technological Analysis of the Thin-Butted Flint Axe. Acta
Archaeologica vol. 53-1982: 1-87.
1993. Report on an ongoing research project: Craft specialization and prehistoric
society. Fornvnnen 88:1-8.
358
References
1997. Craft Specialization as an Agent of Social Power in the South Scandinavian
Neolithic. In: Schild, R. & Sulgostowska, Z. (eds.), Man and Flint. porceedings of
the VIIth International Flint Symposium Warszawa - Ostrowiec Swietokrzyski
September 1995. Warszawa.
1998. Different Strokes for Different Folks. Possible Reasons for Variation in
Quality of Knapping. Lithic Technology vol. 23, no. 2:90-115.
2000. Talking axes, Social Daggers. In: Olausson, D. & Vandkilde, H. (eds.),
Form, Function and Context. Material culture studies in Scandinavian archaeology.
Almqvist & Wiksell International. Lund121-134.
Olausson, M., 1995. Det inneslutna rummet - om kultiska hgnader, fornborgar och
befsta grdar i Uppland frn 1300 f. Kr. till Kristi fdelse . Riksantikvariembetet,
Arkeologiska underskningar. Skrifter nr 9, Studier frn UV Stockholm.
Stockholm.
Olsen, B., 1997. Fra ting til tekst. Teoretiske perspektiv i arkeologisk forskning.
Universitetsforlaget. Oslo.
Owens, D. & Hayden, B., 1998. Prehistoric Rites of Passage: A Comparative Study
of Transegalitarian Hunter-Gatherers. Journal of Anthropological Archaeology
16:121-161.
Parsons, L. W., 1951. The Social System. London.
Patterson, L. W., 1982. Replication and Classification of Large Size Lithic Debitage.
Lithic Technology vol. 11, No. 3:50-58.
Patterson, L. W. & Sollberger, J. B., 1978. Replication and Classification of Small Size
Lithic Debitage. Plains Anthropologist 1978:103-111.
Pearson G. W. & Stuiver, M., 1993. Radiocarbon 35:25-33.
Pelcin, A., 1997. The Effect of Indentor type on Flake Attributes: Evidence from a
Controlled Experiment. Journal of Archaeological Science 24: 613-621.
Pelegrin, J., 1981. An Interview with Flintknapper Jaques Pelegrin, by Errett
Callahan. Contract Abstracts vol 3 (1):62-70
1990. Prehistoric Lithic Technology: Some Aspects of research. Archaeological
Review from Cambridge 9:1:116-125.
Pelissier, C., 1991. The Anthropology of Teaching and Learning. Annual Review of
Anthropology 20:75-95.
Persson, P., 1999. Arkeologi och evolutionism. In: Gustafsson, A. & Karlsson, H.
(eds), Glyfer och arkeologiska rum en vnbok till Jarl Nordbladh : 667-678.
Piaget, J., 1968. Le Structuralisme. Paris.
Pigeot, N., 1990. Technical and Social actors. Flintknapping Specialists and
Apprentices at Magdalenian Etiolles. Archaeological Review from Cambridge 9
(1):126-141.
Pirenne, H., 1936. Economic and Social History of Medieval Europe. Routledge &
Kegan Paul ltd. London.
Polanyi. K., 1944. The Great Transformation. Farrar and Rineheart. New York.
Prescott, C., 1995. Aspects of Early Pastoralism in Sogn, Norway. Acta Archaeologica,
vol. 66:163-190.
359
Jan Apel
Prescott, C. & Walderhaug, E., 1995. The Last Frontier? Processes of IndoEuropeanization in Northern Europe: The Norwegian Case. The Journal of IndoEuropean Studies vol. 23, No. 3 & 4:257-280.
Preucel, R. & Hodder, I., 1996. Process, Structure and History. In: Preucel, R. &
Hodder, I. (eds.), Contemporary Archaeology in Theory. Blackwell. Cambridge:
205-219.
Price, D., 1995. Social Inequality at the Origins of Agriculture. In: Price, D. &
Feinman, G. D. (eds.), Foundations of Social Inequality. Plenum Press. New York.
Price, D. & Brown, J. A., (eds.), 1985. Prehistoric Hunter-Gatherers. The Emergence of
Cultural Complexity. Academic Press. San Diego.
Radcliffe-Brown, A. R., 1913. Three Tribes of Western Australia. Journal of the Royal
Anthropological Institute 43:143-194.
1937. A Natural Science of Society. Free Press, Glencoe.
1952. Structure and Function in Primitive Society. The Free Press. New York.
Rasmussen, K. L., 1993. Radiocarbon Dates of Late Neolithic and Early Bronze Age
Settlements at Hemmed, Hjgard and Trappendal, Jutland, Denmark. Journal of
Danish Archaeology vol 10, 1991: 156-162.
Rasmussen, L. W., 1990. Dolkproduktion og -distribution i senneolitikum. Hikuin
16:31-42.
Rassmann, K., 1993. Sptneolithikum und frhe Bronzezeit im Flachland zwischen
Elbe und Oder. Lbstorf. Beitrge zur Ur- und Frgeschichte MecklenburgVorpommerns. Lbstorf.
Renfrew, C., 1973. Monuments, Mobilisation and Social Organisation in Neolithic
Wessex. In: Renfrew, C. (ed.), The Explanation of Culture Change: Models in
Prehistory. Duckworth. London:539-558.
Renfrew, C. & Bahn, P., 1991. Archaeology. Theories, Methods and Practice. Thames &
Hudson. London.
Rorty, R., 1989. Contingency, Irony, and Solidarity. Cambridge University Press.
Cambridge.
Rosenberg, G., 1929. Nya Jttestuefynd. Aarbger for nordisk Oldkyndighed og
Historie 1929:189-262.
Rudebeck, E., 1986. ngdala. Flintgruvor frn yngre stenldern, S Sallerup.
Utgrvningar 1977-81. Rapport nr 1, Malm museer, stadsantikvariska enheten.
Malm.
2000. Tilling Nature Harvesting Culture. Exploring Images of the Human Being in
the Transition to Agriculture. Almqvist & Wiksell International. Stockholm.
in press. Flint Extraction in Southern Scandinavia during the Neolithic. In: Apel,
J., Hgberg, A., Knutsson, K., Olausson, D. & Rudebeck, E., The Spread of Flint
Axes and Daggers in neolithic Scandinavia. Manuscript.
Rydbeck, O., 1918. Slutna mark- och mossfynd frn stenldern. Frn Lunds
universitets historiska museum 1918.
Sahlins, M., 1958. Social Stratification in Polynesia. University of Washington Press.
Seattle.
1968. Tribesmen. Prentice-Hall inc. Englewood Cliffs, New Jersey.
360
References
1972. Stone Age Economics. Polity Press. London.
Santillo Frizell, B., 2000. Hndernas tysta kunskap. Om apuliska kupolbyggares
yrkeskunnande och dess reproduktion. I: Hjrthner-Holdar, E. & Risberg, C.
(eds.), Hantverkets roll i samhllet produktion och reproduktion. Rapport frn den
tredje workshopen om hantverksspecialisering Uppsala 18-19 mars 1999.
Forskningsrapport R0004, Riksantikvariembete UV GAL. Uppsala.
Scheen, R., 1979. De norske flintdolkene. En typologisk-kronologisk analyse. MA-thesis
in Nordic archaeology at Oslo universitet. Stencil.
Schiffer, M. B., 1976. Behavioural Archaeology. Academic Press. New York.
1987. Formation Processes of the Archaeological Record. University of New Mexico
Press. Albuquerque.
Schiffer, M. B. & Skibo, J. M., 1987. Theory and Experiment in the study of
Technological Change. Current Anthropology vol. 28, No 5.
Schlanger, N., 1994. Mindful technology: unleashing the chane opratoire for an
archaeology of mind. In: Renfrew, C. & Zubrow, E. B. W. (eds.), The Ancient
Mind. Elements of Cognitive Archaeology. Cambridge University Press.
Cambridge:143-151.
Schwarz-Mackensen, G., 1985. Die frbandkeramische Siedlung bei Eitzum, Landkreis
Wolfenbttel. Verffentlichungen Braunschweig 45. Braunschweig.
Searle, J. R., 1997. Konstruktionen av den sociala verkligheten. Daidalos. Gteborg.
Segerberg, A, 1978. Den enkla skafthlsyxan av sten. Fyndfrhllanden och datering.
Tor XVII:159-218.
Sellet, F., 1993. Chane opratoire; The concept and its applications. Lithic Technology
vol. 18, no. 1&2:106-112.
Service, E., 1962. Primitive Social Organization. Random House. New York.
Shennan, S., 2000. Population, Culture History, and the Dynamics of Culture
Change. Current Anthropology Vol. 41, No. 5:811-835.
Siemen, P., 1981. De nordsvenske depotfynd En belysning avhandelsbegrebet i
arkeologin. Kontaktstencil 21. Kpenhamn.
Sievert, A. K., 1990. Postclassic Maya Ritual Behaviour: Microwear Analysis of Stone
Tools from Ceremonial Contexts. In: Grslund, B., Knutsson, H., Knutsson, K.
& Taffinder, J. (eds.), The Interpretative Possibilities of Microwear Studies.
Proceedings of the international conference on lithic use-wear analysis, 15th-17th
February 1989 in Uppsala, Sweden. Aun 14. Uppsala.147-158.
Simonsen, J., 1982. Bolig eller ddehus? MIV. Museer i Viborg amt 11. Viborg.
1985. A Late Neolithic House Site at Tastum. Journal of Danish Archaeology 1983
(2):81-89.
Skousen, H., 1998. Rnbjerg Strandvolde en kystboplads ved Limfjorden. Kuml
1997-98:29-73.
Skov, T, 1982. A Late Neolithic House Site with Bell Beaker Pottery at Stendis,
Northwestern Jutland. Journal of Danish Archaeology vol 1:39-44.
Smith, A., 1970 [1776]. The Wealth of Nations. Penguin. Harmondsworth.
Snyder, J. F., 1897. The Method of Making Stone Arrow-Points. The Antiquarian pl.
9:231-232.
361
Jan Apel
Sokal, R. R., 1977. Classification: purposes, principles, progress, prospects. In:
Johnson-Laird, P. N. (ed.), Thinking: Readings in Cognitive Science.
Cambridge:185-198.
Solberg, B., 1994. Exchange and the Role of Import to Western Norway in the Late
Neolithic and Early Bronze Age. Norwegian Archaeology Review, Vol. 27, No. 2:
111-126.
Solli, B., 1996. Fortiden er et annet sted. Om arkeologi og kulturminnen, rtter og
ftter. Norsk Antropologisk Tidskrift 2/1996:79-90.
Spaulding, A. C., 1982. Structure in Archaeological Data: Nominal Variables. In:
Whallon, R. & Brown, J. A. (eds.), Essays on Archaeological Typology. Center for
Americal Archaeology Press. Evanstone Il.: 1-20.
Spear, T., 1997. Mountain Farmers. Moral Economies of Land & Agricultural
Development in Arusha & Meru. University of California Press. Berkley & Los
Angeles.
Spetz, L., 1994. Den sista vilan? Arbetsredskap i gravar under senneolitikum och ldre
bronslder. BA-thesis, Department of Archaeology, Uppsala University. Uppsala.
Splieth, W., 1900. Inventar der Bronzalterfunde aus Schleswig-Holstein. Kiel.
Squire, I. R., 1986. Mechanisms of Memory. Science 232:1612-1619.
Stafford, M., 1998. In search of Hindsgavl: experiments in the production of
Neolithic Danish flint daggers. Antiquity 72:338-349.
Steinberg, J. M., 1997. The Economic Prehistory of Thy, Denmark: A Study of the
Changing Value of Flint Based on a Methodology of the Plowzone. Phd-dissertation,
University of California. Los Angeles.
Steineke, M., Ekenberg, A., Hansson, K. and Ifverson, P., in preparation. Sunnan
19A-F. Boplasslevninger fra mesolitikum til historisk tid mellom riksvei 11 og
banevollen. Arkeologiske Slutunderskelse. Malm Kulturmilj. Malm.
Stenskld, E., ms. unpublished manuscript distributed at a seminar at the
department of Archaeology in 1998, Stockholm University.
Sterum., N., 1974. Hellekister i den jyske enkeltgravskulturen. Aarbger for nordisk
Olkyndighed og Historie:61-69
Stjernqvist, B., 1950. En hllkista p Bjrkn, Sklleruds sn, Dalsland. Hembygden
Dalsands Fornminnes- och Hembygdsfrbund: 1-38
Strucke, U., 1998. Sklgropar och agrara lmningar vid Berga . Riksantikvariembetet,
UV Mitt, Rapport 1998:107. Stockholm.
ms. Mrby 5:1. Underskningsrapport, Riksantikvariembetet, UV mitt.
Stockholm.
Struve, K. W., 1955. Die Einzelgrabkultur in Schleswig-Holstein und ihre
kontinentalen Beziehungen. Karl Wachholtz Verlag. Neumnster.
Strmberg, M., 1951. Forntida bebyggelse i Vallberga. Sknes Hembygdsfrbunds
rsbok.
1975a. Studien zu einem grberfeld in Lderup. Grabsitte - kontinuitt sozialstruktur. Acta Archaeologica Lundensia, series 8, no 10. Lund.
1975b. Comments on the Flint Daggers of Denmark. Norwegian Archaeology
Review vol. 8, no. 2.
362
References
1992. A Concentration of Houses from the Late Neolithic/Early Bronze Age at
Hagestad. Meddelanden frn Lunds universitets historiska museum 1991-1992.
New series, Vol. 9:57-89.
Stuiver, M. & Pearson, G. W., 1986. High-precision calibration of the radiocarbon
time scale, AD 1950-500 BC. Radiocarbon 28.
Sundstrm, L., 1992. Trattbgarsamhllet och dess utbytessystem. BA-thesis,
Department of Archaeology, Uppsala university. Uppsala.
ms. The Need For an Objective Methology in Archaeology. Manuscript at the
Department of Archaeology, Uppsala University.
Sundstrm, L.& Apel, J., 1998. An Early Neolithic Axe production and Distribution
System within a Semi-sedentary Farming Society in Eastern Central Sweden, c.
3500 BC. In: Holm, L. & Knutsson, K. (eds.), Proceedings from the Third Flint
Alternatives Conference at Uppsala, Sweden, October 18-20, 1996. Occassional
papers in Archaeology 16. Uppsala:155-192.
Taffinder, J., 1998. The Allure of the Exotic. The social use of non-local raw materials
during the Stone Age in Sweden. Aund 25. Uppsala.
Thomas, D. H., 1986. Hunter-Gatherer Archaeology. I: Meltzer, D. J., Fowler, D. D.
& Sabloff, J. A. (eds.), American Archaeology, Past and Future. Smithsonian
Institution Press:247-251.
Thomsen, E., 2000. Flintens geologi og mineralogi. In: Eriksen, B. V. (ed.),
Flintstudier. En hndbok i systematiske analyser af flintinventarier. rhus:17-36.
Thorsberg, K., 1984. Analys av boplatser. teori och metoder med exempel frn den
senglaciala boplatsen Segebro. BA-thesis, Department of archaeology, Uppsala
University.
Tilley, C., 1982. Social formation, social structures and social change. In: Hodder, I.,
(ed.), Symbolic and Structural Archaeology. Cambridge University Press.
Cambridge.
Trigger, B., 1993. Arkeologins idhistoria. Brutus stlings bokfrlag Symposion.
Stockholm/Stenhag.
Trigger, B. G., 1990. Maintaining Economic Equality in Opposition: an Iroquoian
case study. In: Upham, S. (ed.), The Evolution of Political Systems. Social Politics in
Small Scale Sedentary Societies. Cambridge University Press. Cambridge:119-145.
Trnka, G., 1991. Nordische Flintdolchen in sterreichs. Archologie sterreichs 2/
2:4-10.
ms. Nochmals zu den Nordischen Flintdolchen sterreichs. To be published in:
Rulf, J. (ed.), Memorial of Jan Rulf.
Turnbull, C., 1961. The Forest People. Jonathan Cape. London.
Walker, P. L., 1978. Butchering and Stone Tool Function. American Antiquity 43:
710-715.
Vandkilde, H, 1989. Det ldste metalmilj i Danmark. I: J Poulsen (ed), Regionale
forhold i Nordisk bronzealder: 29-45
1996. From Stone to Bronze. The Metalwork of the Late Neolithic and Earliest
Bronze Age in Denmark. Jutland Archaeological Society Publications XXXII.
Aarhus.
363
Jan Apel
1998. Beaker Representations in the Danish Late Neolithic. In: Nicolis, F. (ed.),
Bell beakers Today. Pottery, People, Culture and Symbols in Prehistoric Europe.
Proceedings of the International Colloquium, Riva del Garda, Trento, Italy. May
1998.
Vang Petersen, P., 1993. Flint fra Danmarks oldtid. Hst & Sn. Copenhagen.
Weber, M., 1991. Om positivismen. I: Tre klassiska texter. Bokfrlaget korpen.
Gteborg.
Weiler, E., 1994. Innovationsmiljer i bronslderns samhlle och idvrld. Kring ny
teknologi och begravningsritual i Vstergtland. Studia Archeologica Universitatis
Umensis 5. Ume.
Welinder, S., 1979. Prehistoric Demography. Acta Archaeologica Lundensia, series in 8,
No. 10. Lund.
1998. Neoliticum Bronslder 3900-500 f.Kr. I: Myrdal, J. (ed.), Det svenska
jordbrukets historia del 1. Jordbrukets frsta femtusen r. Natur och kultur/LTs frlag.
Stockholm.
Whittaker, J. C., 1994. Flintknapping. Making and Understanding Stone Tools.
University of Texas Press. Austin.
Whorf, B. L., 1956. Language, Thought, and Reality. Cambridge, Mass.
Willemark, K., 1992. Boplatsvariation i ett fngstsamhlle. Mikroskadeanalys av
skrapor frn tre fngstboplatser, frn tiden 1500-500 fKr, i mellersta Norrland . BAthesis, Department of Archaeology, Uppsala University. Uppsala.
Willey, G. R. & Phillips, P., 1958. Method and Theory in American Archaeology.
University of Chicago press. Chicago.
Willey, G. R. & Sabloff, J. A., 1974. A History of American Archaeology. Freeman. San
Francisco.
Wilmsen, E. & Denbow, J., 1990. Paradigmatic history of San speaking peoples and
current attempts at revision. Current Anthropology vol. 31, no. 5.
Wojciechowski, W., 1976. Sztylety krzemienne typu Skandynawskiego w Polsce
zachodniej. Acta Universitatis Wratislaviensis No 253, Studia Archeologiczne VII.
Warszawa.
Woodburn, J., 1968. An introduction to Hadza residential groupings. In: Lee, R. B.
& DeVore, I. (eds.), Man the Hunter. Aldine, Chicago.
Worsaae, J. J. A., 1860. Om en ny Delning af Steen- og Broncealderen, og om Et
maerkeligt Fund fra den aeldre Steenalderen ved Engestofte paa Laaland. Saerskilt
aftryck af Oversigt 1869.
Wright, G. H. von, 1993 [1957]. Logik, filosofi, sprk. Nya Doxa. Nora.
Wyszomirski, M., 1974. Scandinavian flint daggers in the southern and eastern
region of the Baltic. Meddelanden frn Lunds Universitets Historiska Museum.
Lund.
1979. Indigenous or not? Fornvnnen 74:47-52.
Yerkes, R. W. & Kardulias, P. N., 1993. Recent Developments in the Analysis of
Lithic Artifacts. Journal of Archaeological Research vol.1:89-119.
364
366