Vladimir Salac: The density of archaeological finds in settlement features of the
La Tene period. 264. Martin Gojda: Archaeologists or aviators'! A contribution to ...
WHITHER
ARCHAEOLOGY?
PAPERS IN HONOUR OF
EVIEN NEUSTUPNY
EDITED BY MARTIN KUNA AND NATALIE VENCLOvA
PRAHA 1995
© 1995 Institute of Archaeology, Praha Published by Institute of Archaeology, Academy of Sciences of the Czech Republic, Letenska 4. 118 01 Praha 1, Czech Republic Technical editor Dagmar Dreslerova Cover and the graphies Dagmar Dreslerovd & lana 2emlickova ISBN 80-901934-0-4 Printed by IRES Copies of this volume and a catalogue of other publications by the Institute of Archaeology, Prague, can be obtained from: Institute of Archaeology Letenska 4 118 01 Praha 1 Czech Republic fax: 02-539361
CONTENTS Martin Kuna - Natalie Venclova: Introduction
7
Paradigms and prospects
,
Torsten Madsen: Archaeology between facts and fiction: the need for an explicit
methodology ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . John Bintliff: "Whither archaeology?" revisited Leo Klejn: Prehistory and archaeology Martin Kuna: Pre-historic prehistory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Michael Shanks: The archaeological imagination: creativity, rhetoric
and archaeological futures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stanislaw Tabaczynski: A future for the Marxist paradigm in Central European
archaeology'! The Polish case John Collis: Celts, power and politics: whither Czech archaeology'! . . . . . . . . . . . . . Jacek Lech: Social functions of archaeology in the 20th century Mark Beech: Whither arcbaeozoology? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Arkadiusz Marciniak: Archaeology and palaeodemography: expectations
and limitations
13
24
36
43
52
69
82
93
97
110
Environment and settlement Carole L. Crumley: Cultural implications of historic climatic change Jaromtr Benes: Erosion and accumulation processes in the Late Holocene
of Bohemia, in relation to prehistoric and medieval landscape occupation .... Dagmar Dreslerovd: A settlement-economic model fora prehistoric microregion:
settlement activities in the Vinof - stream basin during the Hallstatt period Natalie Venclova: Settlement area, production area and industrial zone
121
133
145
161
Archaeology and language
•
c,
Marek Zvelebil: Indo-European origins and the agricultural transition in Europe . .. 173
Pavel M. Dolukhanov: Archaeology and linguistics: theories old and new . . . . . . .. 204
Alexander Hausler: Vber Archaologie und Ursprung der Indogermanen 211
Ruth Megaw - Vineelll Megaw: The prehistoric Celts: identity and contexruality .. , 230
Methodological issues [an Hodder: Towards a contextual methodology Andreas Zimmermann: Some aspects of the application of correspondence
analysis in archaeology Vladimir Salac: The density of archaeological finds in settlement features
of the La Tene period Martin Gojda: Archaeologists or aviators'! A contribution to the discussion
on aerial archaeology in Britain
249
255
264
277
Social models Alasdair Wltlllle; The scale of difference and the nature of community;
reflections on Neolithic social relations Stephen Shennan: Diffusion revisited Jan Rulf Provinces. regions and sub-regions: the Labe/Elbc group of Linear Pottery Culture example
283 293 299
WHITHER ARCHAEOZOOLOGY?
Mark Beech
"Dirt is matter out of place."
(Mary Douglas, Purity and danger ..., London 1966)
Animal bones usually represent one of the most visible forms of bio-archaeological remains on excavations. Traditionally they have been utilised as key evidence in understanding such questions as subsistence, consumption and economic organisation. Archaeozoologists have therefore tended to concentrate on the modelling of herd management strategies by attempting to assess the relative importance of different animal products such as meat, milk, wool or traction, and even the industrial exploitation of animal by-products (e.g. horn, skins, etc.). The discipline of archaeozoology within Central and Eastern Europe (e.g. some of the work of Boessneck, von den Driesch, Bokonyi, etc.) has concentrated much of its energies upon the analysis of morphological and metrical data in order to obtain information concerning the nutritional status of animals, as well as the development of recognisable breeds. Very little attention has been paid in the case of Central European material, however, to the contextual analysis of animal bones from archaeological sites. It has been a common practice, for example, to group all material from a site together as a single unit of analysis when modelling the subsistence of a particular site. Recently I have suggested that perhaps the way forward for archaeozoology is to adopt a more contextual/interpretative approach (Beech 19930). This paper also questions the traditional use of archaeozoological data and suggests future possible directions which might be fruitfully explored. Reference will be made to a recent analysis carried out by the author on material from the well known La Tene period site of Msecke Zehrovice in Bohemia in order to illustrate a number of important points.
The taphonomy of bones: interpretative problems In 1981 Evien Neustupny published a paper in Archeologicke rozhledy, entitled "The destruction of bones in prehistoric sites". He discussed the problem of how osteological material found on central European sites from the Neolithic and Bronze Age periods represented a strongly reduced selection from the basic assemblage of bones of animals consumed by the original prehistoric populations. It was pointed out that this reductive process was of both a quantitative (only a few bones from each individual surviving) and structural form (bones of larger species and adults having a better chance of preservation). Various factors were discussed which contributed towards this phenomena: the manner of consumption (particular joints being preferred, as well as the absence of bones of young individuals), the effect of physical factors (trampling, humidity and temperature changes), as well as biological factors. In order to examine such factors, Neustupny examined faunal data from the Late Bronze Age (Knoviz culture) settlement of Kamenna Voda. It was observed that the remains of
98
M. Beech
individuals and the bones of less numerous species were in the surface habitation layers distributed at random (the outcome being Poisson's distribution of the number of individuals in respective pits; ci.Neustupny 1981, Tab.!). Cattle bones however, tended to be more clustered, giving a pattern similar to Neyman's distribution of type A (Neuslupnj 1981, Fig.I). It was pointed out that further reduction in the amount of bones took place once burial occurred, although it was suggested that this probably did not account for the main cause of their absence (as in different parts of Europe, with varying climatic and soil conditions, an approximately equal percentage of the bones of single individuals were preserved). Neustupny concluded that the utilisation of faunal analyses in the study of prehistoric economy was not obviously affected by this quantitative reduction of the total assemblage, but rather, by the differential state of preservation of different bones. This thereby had an effect on the estimation of the percentual representation of different animal species, the assessment of the minimum number of individuals and their relative age structures. Unfortunately, comparatively little attention appears to have been paid to this important paper by archaeozoologists working in Central Europe. Archaeologists also have generally continued to use faunal data in a crude manner in order to infer subsistence and economic organisation. Partly this may result from limited knowledge or access to literature published concerning the taphonomy of bones (e.g. Behrensmeyer 1975,1978; Behrensmeyer - Hill 1980; Binford 1981; Binford - Bertram 1977; Brain 1967, 1981; Gifford 1981; Noe-Nygaard 1977; Shipman 1981; Shipman - Rose 1983). It is now clear to most modem archaeozoologists that the taphonomic history of an assemblage has to be evaluated prior to the formulation of economic interpretations. In the following section I would like to examine the use of different quantification techniques in archaeozoology, with particular reference to the animal bones recovered from recent excavations at the site of Msecke Zchrovice. Various additional analytical approaches which are not normally utilised within Central European faunal studies will subsequently be introduced, as well as a discussion of their broader implications.
Quantification in archaeozoology: the art of "guesstimation" There has been much discussion in the literature concerning various means of quantification of vertebrate faunal data. More recently these have been summarised in Brewer (1992), Ringrose (1993) and Winder (1991). There is no need within this paper to reiterate all the various problems associated with various quantification techniques. A point which emerges from all of these discussions, however, is the value of expressing data in different forms, so that it can be analysed in a number of different ways. Undertaking comparisons and contrasts between results permits a clearer insight into the nature of the data. It is clear that it is not sufficient merely to present a single result which simply confirms one's previous prejudices. Four different commonly utilised quantification techniques were therefore applied to the animal bones recovered from Msecke Zehrovice (Beech 1993b). The site is well known for the discovery of the famous carved stone head of a Celtic warrior (Megaw - Megaw 1988; Beech 1992). Recent archaeological investigations at the site have identified a settlement with two main chronological developmental phases, dating to La Tene B2-C I and La Tene C2-D I (Venclova 1989, 1991).
99
Whither archaeozoology?
Table 1. Quantification of the major domestic species at MSecke 2ehrovice by chronological period (all features combined) using the NISP, DZF. MNI and weight methods.
LTB2-CI NISP
Cattle Pig Sheep/Goat
W
MNI
DZF
N
%
N
%
N
%
N
%
357 164 69
60.5 27.8 11.7
125 89
53.0 37.7 9.3
7 11 4
31.8 50.0 18.2
6357 1595 486
75.3 18.9 5.8
22
LT C2-DI NISP
Cattle Pig Sheep/Goat
DZF
W
MNI
N
%
N
%
N
%
N
%
450 459 133
43.2 44.0 12.8
249 187 65
49.7 37.3 13.0
16 31 8
29.1 56.4 14.5
13981 5290 1113
68.6 26.0 5.4
Quantification of the material was carried out separately for both major chronological phases of the site (grouping all features together). as well as separately for each individual feature within each particular phase. This was carried out in order to compare the variability of different methods and to examine the internal variability of the data throughout different features across the site. The four quantification techniques utilised were as follows: (I) NlSP (number of identified specimens). This is the simplest method of quantifying vertebrate faunal data. All bone fragments are simply assigned to taxon and are then counted. (2) DZF (diagnostic zone method). This method attempts to overcome some of the problems associated with the NISP method. particularly the problem of counting the same bone more than once. Only non-repeatable elements are counted to try and ensure that each bone is not counted more than once. The following units were counted: homcore (base), skull (occipital condyle only). maxilla (body), mandible (body), atlas and axis (centrum), scapula (glenoid). humerus and radius (proximal/distal), ulna (proximal), pelvis (acetabulum), femur, tibia, metacarpal and metatarsal (proximal/distal), astragalus, calcaneus (proximal), l st, 2nd and 3rd phalanges (proximal). (3) MNI (minimum number of individuals). The minimum number of individuals was calculated based on the most numerous non-reproducible element, with no matching for pairs. (4) W (weight in grammes). All bone fragments were individually weighed, totals being calculated for each taxon within each separate feature, as well as within each phase. Table 1 and Figure 1 illustrate the results of overall quantification of the major domestic species within the two main phases. It is worth pointing out that the NISP or number of identified specimens is the most common method utilised by Central European archaeozoologists. Most conventional analyses of the referred to dataset would therefore interpret the material from LT B2-Cl as largely consisting of cattle, followed by pig then sheep/goat, inferring some form of economic change in the later period (LT C2-Dl) due to a decrease in the number of cattle, and an apparent increase in pigs. In actual fact. the situation is not so clear. If we examine the data from the viewpoint of the other quantification methods (DZF, MNI and weight) there does not appear to be such a substantial change in
100
M. Beech
the proportions of the major species between the two phases. The principal difference appears to be that cattle bones are more fragmented in the earlier phase, thereby artificially inflating the NISP figure. In contrast, in the later phase, pig bones are generally more fragmentary, their relative proportion being less when compared with the DZF method. The effects of these two factors can clearly be seen when comparing NISP-DZF ratios. It is clear that only by comparing and contrasting the relative merits of different quantification techniques can we gain a more detailed insight into the true nature of animal
LTB2-C1
-
~
lOll
CotUo
00 110' _
o""
70
f:
~
Sh/Gt
l
•
f
20
ceee
90 roc
•
30
'0 0
110 70 60
Sh/Gt
50 40
30 20
10 0
-
-
LTC2-D1
10
NISP
DZF
MNI
Ouantificatlon melhixl
W
O.
NISP
DlF MNI Quantification method
W
Figure 1. Results of overall quantification of the major domestic species within the two main phases,
by NISP (number of identified specimens), DZF (diagnostic zone fragments), MNI (minimum number of individuals) and W (weight in grammes). N.B.: all features combined.
bone assemblages. Previously many authors have probably overinterpreted the meaning of such quantitative results. What I would like to suggest here below is that perhaps another direction may be fruitfully explored within archaeozoology.
"Dirt is matter out of place": spatial archaeozoology? Mary Douglas in her book, "Purity and Danger" (1966), developed a theory of pollution, in which she pointed out how things are often out of place in terms of the order which a society imposes upon itself and on the universe it occupies. One of the problems, however, is that things can be out of place in many ways and in many different dimensions. I would like to proceed by attempting to develop this idea here in relation to the distribution of animal bones from archaeological sites. Clearly the relationship between man and the animal world is not entirely a simple matter. There may exist complicated relations between dietary prohibitions of cert.ainanimals and the classification of animals (cf. Tambiah 1969). Clear rules may also exist concerning the spatial placement of animals. Unfortunately the interpretation of the distribution of animal bones from archaeological sites is not entirely a simple matter. The problem of taphonomy has already been partly touched upon earlier. Bone preservation conditions as well as recovery techniques may, of course, vary between localities making inter-site comparisons problematic. Seldom can one compare the distribution of material within different features across a site in order to examine which particular data groups are influencing the overall picture. There is a need to be more critical when comparing data from different sites, which of course may have undergone completely different taphonomic regimes.
101
Whither archaeozoology?
One valuable approach may be to examine the composition of material within different feature classes across a particular site. Such an analysis may even provide evidence of particular activity zones within the site, indicating regular disposal zones within the settlement area (cf. Halstead - Hodder - Jones 1978). Table 2. List of features within the two main chronological phases (LT B2-Cl and LT C2-Dl) at Msecke Zehrovice. LT B2-CI
Code 1W SM HI H2 H3 H4 H5 H6 Al A2 A3 PH SL PI P2 P3 P4 SP
Feature number
Interpretation iron metallurgyworkshop manufacturing feature
1{79 38186 2Af79 2A+2B{79 2B{19 7181 9181 11/82 PS2AB PS2B PS 7/81 6/80 SL 3/83 2B/83 6/86 8/86 15/86
house house house house house house plough soil above houses 2A+2B plough soil above house 2B plough soil above house 7/81
fence aroundhouse 7/81 settlement layer under the bank pit pit
pit pit storage(?) pit
LTC2-DI Code HI H2 LAH PI LAP P2 ABC LAC
Feature number 4/83 20/84 AH20/84 1/83 LA 1/83
2N83 ABC LAC
Interpretation house house layer above house 20/84 storage pit layer above pit 1/83 pit accumulation of burnt clay layer above accumulation of burnt clay
If we return to the example of Msecke Zehrovice, a number of distinct archaeological features could be discerned during the excavations (see Table 2 and Figure 2). For the purposes of this analysis, the aforementioned quantification techniques were applied to the bones from the three most common domesticates (cattle, pig and sheep/goal) separately within each feature and phase (see Tables 3-5 and Figures 3-4). It is clear that there is a considerable amount of variation between different types of features.
102
M. Beech
,. ...•.....
..._
.'
c
.[
*
=\~?~~kc~~ •
~ PlIl~~~~1 .....
.
.
..
H2/A1
~
,
·0 (-",
\.i
..
cr===
U·
IOHS
(QIH6
DB
DD
('
.. ;~ .... ...~
,
.~
..... :
"',.'
/"
.:
:
P4
.
P3
Cy
EB o :
~SP 20m [
Figure 2. Location of archaeologicalfeatures at thesite of Msecke Zehrovice. See Tab.2 for explanation of codes. White features: LT B2-Cl; black shaded features: LT C2-Dl.
Whither archaeozoology?
103
In the period of LT B2-Cl most material occurs within the houses. The different quantification methods mostly appear to correlate giving similar overall distribution patterns. The only major exception was as follows: the NISP appeared to be over-emphasised for houses 5 and 6 which had high numbers of cattle bone fragments. These peaks were not so noticeable when quantifying cattle bones by the other methods. This confirms the overall picture, as discussed previously, that cattle bones were generally more fragmented than other species within these particular contexts. Table 3. Quantification of the cattle bones from Msecke 'Lehrovice by chronological period within separate features using the NISP, DZF, MNI and weight methods.
LT B2·C1
DZF
NISP
Feature
N
%
IW SM HI H2 H3 H4 H5 H6 Al A2 A3 PH SL PI P2 P3 P4 SP
23 3 17 29 28 95 67 6 1 1 43 6 14 -
6.44
TOTAL
357
-
24
W
MNI
N
%
9
7.20
N
%
N
%
344 60 654 927 467 784 629 190
5.41 0.94 10.29 14.58 7.35 12.33 9.89 2.99
3
12.00
-
-
-
-
-
0.84 4.76 8.12 7.84 26.61 18.77 1.68 0.28 0.28 12.05 1.68 3.92 6.72
2 13 18 15 19 13 2 1 1 18 4 9 1
1.60 10.40 14.40 12.00 15.20 10.40 1.60 0.80 0.80 14.40 3.20 7.20 -
1 3 3 2 2 1 1
4.00 12.00 12.00 8.00 8.00 4.00 4.00 4.00 4.00 12.00 4.00 8.00
-
0.80
125
-'
1 1 3 1 2 1
-
4.00
25
-
-
15 48 1109 219 897
0.24 0.76 17.44 3.45 14.11 0.22
-
14 6357
LTC2·Dl
DZF
NISP
Feature
W
MNI
N
%
N
%
N
%
N
%
HI H2 LAH PI LAP P2 ABC LAC
141 73 21 65 29 53 101 94
3.11 16.22 4.67 14.44 6.44 11.78 22.44 20.89
7 38 11 58 14 22 53 46
2.81 15.26 4.42 23.29 5.62 8.84 21.29 18.47
21 3 2 6 3 2 4 6
7.14 10.71 7.14 21.43 10.71 7.14 14.29 21.43
4341 2226 591 3280 1046 999 2849 2556
3.10 15.92 4.23 23.46 7.48 7.15 20.38 18.28
TOTAL
450
249
28
13981
104
M. Beech
Table 4. Quantification of the pig bones from Msecke Zehrovice by chronological period within separate features using the NISP, DZF, MNI and weight methods.
LTB2-Cl NISP
Feature
DZF
N
%
IW SM HI H2 H3 H4 H5 H6 Al A2 A3 FH SL PI P2 P3 P4 SP
18 1 6 25 45 14 7 1 1 4
10.98 0.61 3.65 15.24 27.44 8.54 4.27 0.61 0.61 2.44 15.85 3.05 3.05 0.61 3.05
TOTAL
164
~
26 5 5 1 5 -
-
N 5 1 6 18 22 7 3 1 1 3 9 I 5 I 6 -
W
MNI %
N
%
N
%
5.62 1.12 6.74 20.22 24.72 7.87 3.37 1.12 1.12 3.37 -
2 1 2 4 5 1 1 1 1 1 3 1
7.41 3.70 7.41 14.81 18.52 3.70 3.70 3.70 3.70 3.70 11.11 3.70 3.70 3.70 7.41 -
115 11 80 239 621 68 37 17 3 23
7.21 0.69 5.02 14.98 38.93 4.26 2.32 1.07 0.19 1.44 17.74 0.81 4.08 0.19 1.07 -
-
10.11 1.12 5.62 1.12 6.74 -
89
I I
2 27
-
283 13 65 3 17 -
1595
LT C2-D! NISP
DZF
Feature
W
MNI
N
%
N
%
N
%
N
%
HI H2 LAH PI LAP P2 ABC LAC
23 121 IS 93 11 62 74 60
5.01 26.36 3.27 20.26 2.40 13.51 16.12 13.07
9 46 11 38 7 18 37
4.81 24.60 5.88 20.32 3.74 9.63 19.79 11.23
2 8 2 8 1 3 9 5
5.26 21.05 5.26 21.05 2.63 7.89 23.68 13.16
233 1857 137 1086 93 571 696 617
4.40 35.10 2.59 20.53 1.76 10.79 13.16 11.66
TOTAL
459
21
187
38
5290
If we compare the rank order of occurrence of the major species within features distributed across the site, one can make the following interesting observation. In houses 1-4 pig or sheep/goat is nearly always in first place with cattle bones generally being the least frequent. In contrast, in houses 5-6 cattle bones generally appear in first place, followed by sheep/goat and pig. It is perhaps worth noting that houses 1-4 are situated relatively close to one another in the NE corner of the site (see Figure 2), in contrast to houses 5-6 which are situated some distance away.
Whither archaeozoology?
105
Table 5. Quantificationof the sheep/goal bones from Msecke:Lehrovice by chronological period within separate features using the NISP, DZF, MNI and weight methods.
LTB2-Cl
DZF
NISP
Feature
N
%
IW SM HI H2 H3 H4 H5 H6 Al A2 A3 FH SL PI P2 P3 P4 SP
3 4 10 IS 12
4.35 5.80 14.49 21.74 17.39 14.49 1.45 2.90 1.45 -
TOTAL
69
10 1 2 1 -
7 I 3 -
-
10.14 1.45 4.35 -
-
N 1
I 2 7 2 2 I I 2 I I I -
W
MNI %
4.54 4.54 9.09 31.82 9.09 9.09 4.54 4.54 9.09 4.54 4.54 4.54 -
22
N
%
N
%
1
6.66 6.66 13.33 13.33 13.33 6.66 6.66 6.66 6.66 6.66 6.66 6.66
9 21 130 140 60 22 8 8 17 20 8 43 -
1.85 4.32 26.75 28.81 12.35 4.53 1.64 1.64 3.50 4.12 1.64 8.85 -
-
1 2 2 2 1 1 1 1 I I I -
-
15
-
486
LTC2-Dl NISP
Feature
DZF
W
MNI
N
%
N
%
N
%
N
%
ABC LAC
14 65 IS 17 2 6 13 1
10.53 48.87 11.28 12.78 1.50 4.51 9.78 0.75
9 29 9 6 I 2 8 I
13.84 44.62 13.84 9.23 1.54 3.08 12.31 1.54
2 4 I 2 I I 2 I
14.29 28.57 7.14 14.29 7.14 7.14 14.29 7.14
178 487 113 93 14 59 142 27
15.99 43.76 10.15 8.36 1.26 5.30 12.76 2.42
TOTAL
133
HI H2 LAH PI LAP
P2
65
14
1113
In LT C2-D!. most material came from house 2, which appeared 10 principally contain sheep/goat bones, followed by pig, with cattle in third place. A broadly similar pattern appeared to be true for the layer above house 2, as well as for house!. In contrast, within the majority of other features, cattle or pig predominated with sheep/goat generally being in third place. Again there appears to be an interesting contradiction between bones found in houses and bones found within other types of features. A possible explanation for the apparent patterning of the animal bones at Msecke Zehrovtce is an opposition between domestic consumption/table/kitchen waste as opposed to primary butchery waste. The small size of sheep/goat plus the fact that they may have
106
M. Beech
been killed at a young age means that they genecally had a greater chance of arriving on the table relatively intact. The size of cattle almost certainly meant that they would have been butchered at some distance from the cooking area. Discard areas for such bones may have therefore been prescribed to particular zones within the settlement.
NISP distribution LTB2-C1 4{)
I
DZF distribution LTB2-C1
