301
Documenta Praehistorica XLIII (2016)
Tools tell tales – climate trends
changing threads in the prehistoric Pannonian Plain
Ana Grabund/ija 1, Emmanuele Russo2
1 Institut für Prähistorische Archäologie, Freie Universität Berlin, Berlin, DE
ana.grabundzija@fu-berlin.de
2 Institut für Meteorologie, Freie Universität Berlin, Berlin, DE
emmanuele.russo@met.fu-berlin.de
Introduction
Reconstructing the resources
Textile fibres can be roughly divided into two main
categories, vegetal and animal. At least according to
the textile evidence from both Europe (Barber 1991;
Rast-Eicher 2005; Médard 2006; 2012) and the Near
East (Schick 1988; Alfaro 2002; Burnham 1964;
Helbaek 1963) flax (Linum usitatissimum L.) and
tree basts were the most frequent sources of textile
fibre during the Neolithic period. Although most of
the direct findings suggest that the early raw mate-
rials used in textile production were mainly of vege-
tal origin, their preponderance should be considered
with caution. Despite the fact that animal fibres are
made of protein and therefore are more resistant to
decay than vegetal ones, which are composed of cel-
lulose, one should keep in mind that linen preserves
better in alkaline conditions, while animal fibres such
as wool require acidic environments for preserva-
tion (Cybulska, Maik 2007). The fact that they are
not found in the contexts investigated might not ac-
count for a genuine absence, but rather for the fact
that the conditions for preservation were not met.
From the plant
The wetland settlements of the Alpine region facili-
tated the investigation of fibre from the flax plant
(Linum usitatissimum) and its use in textile pro-
duction, revealing details of both the plant’s devel-
opment and its dynamics. However, since animal
fibres cannot survive in the alkaline soils of circum-
Alpine dwellings, the possibility of wool use in these
rich contexts remains elusive. Flax has been defined
ABSTRACT – This study of prehistoric textile production on the Pannonian Plain is based on indirect
evidence dated to the period between the 5th and 2nd millennium BC; the study of technological trends
and changes that occurred in manufacturing traditions concentrates on fibre processing and pro-
duction. The functionality analysis of spindle-whorls served as a basis for comparing textile produc-
tion trends with the results of the climate change model. Climatic changes in the area were simulat-
ed by means of a moderate-resolution Global Circulation Model (GCM). The simulation covered the
mid-to-late Holocene, from 7000 years BP to the pre-industrial period.
IZVLE∞EK – Predstavljamo ∏tudijo izdelave tkanin v prazgodovini na obmo≠ju Panonske ni∫ine, ki
temelji na posrednih dokazih, ki datirajo v ≠as 5. in 2. tiso≠letja pr. n. ∏t.; pri ∏tudiju tehnolo∏kih tren-
dov in sprememb, ki so se zgodile v tradicijah izdelave tkanin, se osredoto≠amo predvsem na proce-
siranje in izdelavo vlaken. Analiza funkcionalnosti predilnih vretenc nam je slu∫ila za osnovno pri-
merjavo med proizvodnjo tkanin in rezultati modela klimatskih sprememb. Klimatske spremembe
na tem obmo≠ju smo simulirali s pomo≠jo svetovnega modela cirkulacije (ang. GCM) pri srednji re-
soluciji. Simulacija pokriva ≠as med srednjim in poznim holocenom, torej od 7000 let pred sedanjo-
stjo do ≠asa pred industrijsko revolucijo.
KEY WORDS – textile production; spindle-whorls; functional analysis; climate forcing; climate mod-
elling
DOI> 10.4312\dp.43.15 
Ana Grabund/ija, Emmanuele Russo
302
as one of the ‘founder crops’ that started agriculture
in the Near East (Zohary 1996). Its early and simul-
taneous use for linen fibre production already in the
8th millennium BC has been attested by textile re-
mains from Nahal Hemar cave in Israel (Schick
1988.31). In Central Europe, linseed specimens from
the Jevi∏ovice site at Hundssteig (3100–3010 cal BC)
in Lower (eastern) Austria were incidentally found
within a row of loom weights (evidence of the old-
est in situ weaving loom in Austria. Although there
is certainly a long way to go between seed-bearing
flax plants and the fibres woven on a loom, this co-
incidence might well point to the use of flax in tex-
tile production (Kohler-Schneider, Cannepele 2009.
67). The earliest surviving linen cords from the cir-
cum-Alpine area of Europe attest to its use for tex-
tiles in the 4th millennium (Barber 1991.11–15; Mé-
dard 2012.386–9; Rast-Eicher 2005.118–20), al-
though the earliest flax seeds in the same area pre-
date the textile finds by almost two millennia.
However, the importance of flax in Eneolithic con-
texts within the investigated area of South East Eu-
rope has also been well documented in more recent
reports (Reed 2016). In several studies based on dif-
ferences in linseed sizes, archaeobotanists have
made an effort to clarify the dynamics and patterns
of its exploitation (Maier, Schlichtherle 2011; Her-
big, Maier 2011). Since it is possible to distinguish
between domestic and wild flax seeds (Linum bien-
ne), mainly on the basis of features such as seed size
(Herbig, Maier 2011; Zohary et al. 2012.101), a si-
milar methodological principle is used to distinguish
between the two different varieties of domesticated
flax.
Ancient Near Eastern seeds have a length of approx.
4.0 to 4.7mm and a width of approx. 2.3 to 2.7mm,
their size remaining unchanged from 5000 BC to
1800 BC (Helbæk 1959.109, Tab. 1). Interestingly,
in Central Europe, flax seems to have split into two
varieties at an early stage, one with the large seeds
used for linseed oil, and another with smaller seeds
specialised for making textile fibres.
An extensive study of 4th and 3rd millennium wa-
terlogged flax seeds from a series of wetland settle-
ments in the Alpine region revealed that in the early
phase (4000–3700 BC) seeds were significantly lar-
ger in comparison to samples from the ensuing pe-
riods (Herbig, Maier 2011.529). This difference in
size cannot be correlated with natural or climatic va-
riations and, according to the research, two flax va-
rieties may be considered as possible proof of fibre
specialisation. In the early phase, a flax type with
larger seeds was cultivated for both oil and fibre pro-
duction, similarly to the Near East (Maier, Schlicht-
herle 2011.571). At the same time, already in the
middle of the 4th millennium BC (around 3400–3300
BC) a small seed fibre type appeared, which was in-
troduced and cultivated to fit the needs of textile pro-
duction (Herbig, Maier 2011.529).
The effects of osmotic stress on long fibre flax (Che-
micosova 2006; Dash et al. 2014) and its preference
for colder and wetter conditions could offer an expla-
nation as to why this new fibre variety of improved
length and quality became the primary resource in
Europe. Indeed, a central European ‘flax-boom’ (Her-
big, Maier 2011.531–532; Maier, Schlichtherle 2011.
571) occurred at the end of the 4th millennium BC,
when textile production intensified.
Spindle-whorls, spindles, loom weights, and finished
fragments of netting and fabric allowed for the re-
construction of textile techniques throughout the
wetland settlements. Well-preserved layers like those
at Arbon-Bleiche 3 and Pfyn-Breitenloo contain the
remains of flax processing, which provided valuable
information on its preparation and handling. There
is sufficient evidence to show that certain villages
focused on growing it and processing the fibres
(Schlichtherle 2009). Furthermore, the spatial dis-
tribution of textile tools within settlements points
to the organisation and specialisation of textile work
(Maier 2001; Lauzinger, Rast-Eicher 2011.539–
540). Archaeological and botanical evidence suggest
that in addition to the division of labour, certain so-
cioeconomic differences developed whereby goods
and economic factors, including flax, were not avail-
able to the same extent to all inhabitants (Schlicht-
herle et al. 2010).
To the animal
Although the intensified use of animal fibres, name-
ly wool, in textile production, was initially discussed
in the context of Sherratt’s model (Sherratt 1981),
new evidence for milk consumption (Evershed et al.
2008; Bartosiewicz 2007; Craig et al. 2005; Vigne,
Helmer 2007) and animal traction (Fabi∏ 2005; Isa-
akidou 2006; Johannsen 2005) does not support a
big time lapse between the initial domestication, fo-
cused mainly on meat consumption, and the later use
of secondary animal products. No such arguments
for the early use of wool have been established so
far, since the first woollen textiles are exceptionally
rare and appear only from the 4th millennium BC
(Rast-Eicher 2014; Shishlina et al. 2003). Further-
Tools tell tales – climate trends changing threads in the prehistoric Pannonian Plain
303
more, linguistic research suggests a common root
for the word ‘wool’ in Indo-European languages,
pointing out that the notion of its soft and spinnable
characteristics appeared prior to its departure from
west-central Asia around 4000 BC (Barber 1991;
Chantraine 2009).
Some attempts to place the beginnings of wool use
earlier (Rooijakkers 2012) focus primarily on the
Near East. New zooarchaeological investigations of
sheep husbandry in Northern Mesopotamia (Vila,
Helmer 2014) point to fleece exploitation shortly
after initial domestication (around 6500 BC), sug-
gesting that several breeds were involved in the
selection process. Thus multi-origins and trade are
considered important factors for the cross-breeding
process.
Although the first use of wool was suggested to be
parenthetical in the context of fibre production and
processing, at least as far as spinning is concerned,
it does account for the early knowledge both of the
resource and its properties. It is thus considered
plausible that sheep fleece developed new desired
properties gradually (Rast-Eicher, Bender Jørgen-
sen 2013), as a cumulative result of animal selection,
involving out-breeding and cross-breeding practices
by shepherds. Already developed and established
Mesopotamian wool economies in the 3rd millenni-
um BC, known from the numerous cuneiform sour-
ces (Charvát 2011) raise the argument that wool’s
first appearance should be dated much earlier.
An analysis of textile tools from Arslantepe (Lau-
rito et al. 2014) proposed household wool spinning
already in the 4th millennium BC, and ethnological
research in Greece showed that unspecialised herd-
ing can, and did, account for an ample amount of
spun thread (Rokou 1994). In addition, both ‘Neo-
lithic’ and ‘Bronze Age’ wool types occur in Iron Age
contexts. Sheepskins and textiles from Hallstatt in
Austria (dated to the 1st millennium BC) show a
range of fleece qualities, with a selection of finer
wool for weaving (Ryder 1990a; 1990b; 1992). This
direct evidence confirms that the development from
‘hair’ to coarse wool and finally to fine wool in sheep
was successive. Consequently, a definition of a sin-
gle temporal horizon for the introduction of wool-
ly sheep can be considered unconvincing (Halstead,
Isaakidou 2011). Considering that multiple genes
account for fleece development (Ryder 1982), Paul
Halstead and Valasia Isaakidou proposed that the
better fleece qualities might have been selected un-
intentionally and on diverse occasions (Halstead,
Isaakidou 2011. 67).
Moreover, an important argument that additionally
supports the idea of the multiple origin of the de-
sired traits was a consequence of the dynamic gene
flow – rather than of developed wool economies
focused on increased production – is based on the
fact that the better qualities of wool, regarding its
finesse, are often found in places and periods re-
mote from those interpreted as more specialised
and established1.
And ‘back to the roots’
The definite separation of farming and foraging has
been fairly criticised, and it is plausible that both
strategies were practiced simultaneously (Cum-
mings, Harris 2011). The use of foraged plant fibres
before, during and beyond the Neolithic is support-
ed by the substantial amount of direct evidence at-
testing to the common use of indigenous plants, in-
cluding reeds, grasses and bast fibres from trees
(Harris 2014). The best examples from the Panno-
nian Plain region were found at the Ljubljansko
Barje dwellings in Slovenia, with preserved fibres
of the grass family (Poaceae) (Pajagi≠-Bregar et al.
2009) and leaf fibres of lesser bulrush (Typha an-
gustifolia) (Greif 1997.41). Nonetheless, the result
of the plant macro remains study of the late 4th mil-
lennium Stare Gmajne lakeshore settlement at Ljub-
ljansko barje, should be mentioned, since they re-
vealed that, besides poppy (Papaver somniferum),
flax was the highest represented oil plant, which
could also have been used for its fibres (Tolar, Ve-
lu∏≠ek 2009; Tolar et al. 2010).
Fabienne Médard concluded that, overall, there are
more preserved cords, threads and cloth made from
tree bast fibres than those made from flax (Médard
2010.57). She noticed that tree bast fibres (predomi-
nantly lime) were used mainly for twined cloth,
1 Commenting on Bronze Age wool quality, Michael Ryder compared contemporary Scandinavian and Egyptian produce: “The wool
in North Europe was naturally brown, while that in Egypt was white. Surprisingly, however, the Egyptian wool was coarser.
This theme of finer wool in more primitive areas, or in earlier periods, is repeated throughout subsequent history“ (Ryder
1974.110). Subsequently, this was explained by experts (Rast-Eicher, Bender Jørgensen 2013) as indeed a finer wool quality from
‘Nordic’ sheep (genetically different from central European and Near Eastern sheep breeds). The Bronze Age ‘Nordic’ type wool of
superior fibre quality to that of contemporary Egyptian production thus provoked a revaluation of the ‘improvement through in-
tensification’ principle.
Ana Grabund/ija, Emmanuele Russo
304
whereas flax was used mainly for woven textiles.
(Médard 2012.368). This led to the conclusion that
most of the tree bast fibres from the Neolithic lake
dwellings in Switzerland were spun without a spin-
dle, a technique characteristic especially of spinning
long vegetal fibres (Médard 2006.99–102).
The use of both early wool and flax for textiles could
be considered in this category. Collected wool tufts
and wild flax fibres were probably competitive with
the aforementioned plant resources, along with
other materials, such as frequently disregarded goat
hair (Frangipane et al. 2009). Wild flax fibres were
identified in a series of Upper Paleolithic layers at
Dzudzuana Cave in Georgia (Kvavadze et al. 2009),
indicating that prehistoric hunter-gatherers were
already using its fibre to make thread. There was
certainly a substantial time lapse in sequential adap-
tations of fibre use before it become a primary re-
source for textile production.
Finally, if considered in a wide-scale process that in-
volved gradual modifications and a refinement of
practices, it is extremely difficult to decide when and
where these secondary products were first recog-
nised and exploited. Although Hans Helbæk sug-
gested that flax was initially used for its seeds and
then for its fibre (Helbæk 1959.104; McCorriston
1997), linen evidence is direct proof of its early use
for textiles. Preserved textiles confirm that flax was
used for its fibre already in the Early Neolithic in the
southern Levant (Schick 1988.31).
Alternatively, the quality of early wool could have
been controlled to a certain degree through fibre
preparation. Sampling and manual sorting of fleeces
and tufts could have resolved its hairiness and coar-
seness, resulting in a selection of spinnable material.
This selection practice has been demonstrated with
an analysis of Bronze Age skins and textiles from Hal-
lstatt (Rast-Eicher, Bender Jørgensen 2013), reveal-
ing how a skilled wool-sorter could identify and ca-
tegorise a range of wool qualities even with a flock
of highly mixed sheep.
Due to gradual specialisation and the secondary pro-
duct character ascribed to these fibres, it is extre-
mely difficult to investigate the origin of their use
and dynamics of their development. On the other
hand, following their importance and representation
in textile production, it is to some degree possible
to determine when, where and why they became a
primary resource, as their intensified exploitation
was more culturally and environmentally dependent.
Theory and time
Apart from investigating the biological origins and
developments of both wool and flax fibres, social
and technological changes connected to their inten-
sified exploitation could be approached through cul-
tural historical narratives and theoretical frame-
works. The economies of prehistoric communities
across the Pannonian Plain would alter with specia-
lisation in either of the resources, transforming fibre
material into a commercial good, possibly involved
in export and exchange. Both strategies, if intensi-
fied, would imply an increased anthropogenic fac-
tor. Maintaining large herds calls for opening the
primeval forest vegetation in parts of Europe, where
forest cover is not limited by climatic factors. The
high water and high-quality agricultural land requi-
rements of flax suggest direct competition with food
crops. Thus, the benefits of developing traditional
technologies and production as sub-branches of the
fragile and not fully formed economies of the period
seem even more reasonable.
The processual archaeology approach would argue
that this next step was determined by a successful
cultural adaptation to climatic and environmental
conditions (Binford 1968), while the opposing post-
processual approach would focus on explaining all
changes in the past as a result of human agency
(Hodder 1986). Regardless of the approach, the fact
remains that developments of the technological as-
pect were the main determinant of ‘success’.
Andrew Sherratt assembled different strands of evi-
dence from archaeology and zooarchaeology into a
consolidated model of the Secondary Products Re-
volution (Sherratt 1981). By suggesting that the
main by-product of domesticates (traction, riding,
milk and wool) occurred simultaneously, first in a
narrow time span in the 4th millennium BC, he con-
ceptualised them as economic achievements that de-
veloped several millennia after primary, meat-focus-
ed consumption. In this model, the main commod-
ity, namely meat, was the objective of Neolithic her-
ders, against the full range of exploitation, an achie-
vement of the Eneolithic. Consequently, the model
addressed the secondary products’ potential for cul-
tural evolution and social complexity in the transi-
tion from the 4th to the 3rd millennium BC. Inte-
restingly, it is not the main postulate of the theory
which is criticised, but the contemporaneous occur-
rence or intensification of different types of exploi-
tation (Vorsteen 1996; Halstead, Isaakidou 2011).
One of the biggest supporters of Sherratt’s model,
Haskel Greenfield, argued for a revision of the con-
Tools tell tales – climate trends changing threads in the prehistoric Pannonian Plain
305
temporaneity of its parts (Greenfield 2005), under-
lining the importance of intensification in scale, ra-
ther than origin of use. The initial critique of the
model contributed to further questions on causality:
were these ‘innovations’ a cause or a consequence
of social and economic developments, did they influ-
ence each other, and finally, how?
The economic importance of secondary products,
even if known for a long time, as proved in the case
of milk (Evershed et al. 2008; Bartosiewicz 2007;
Craig et al. 2005; Vigne, Helmer 2007), increases
during the 4th millennium BC, which further shaped
our research objectives. The ‘competition of fibres’
found its place in the paradigm due to the contem-
poraneity of the specialisation of both resources and
the influence on economic factors that accompany
their intensified use.
The wide time frame of our investigation, which en-
compasses two millennia, gave us the opportunity to
pattern technological trends and changes on a large
scale, rather than addressing culture-specific ques-
tions and particular inter-site dynamics. The focus
on 4th and 3rd millennium BC textile production in
the Pannonian Plain placed the research within the
context of three periods: the Middle Eneolithic, Late
Eneolithic and Early Bronze Age. The transforma-
tion from Middle to Late Eneolithic and from Late
Eneolithic to Early Bronze Age societies in the Pan-
nonian Plain from the final 5th to the ending of the
3rd millennium BC, and the role of local and exoge-
nous influences, remain controversial. Further, all
three periods are characterised by a number of part-
ly overlapping ‘cultures’ whose definition is based
primarily on pottery inventories and burial rites, so
they do not necessarily represent social or econom-
ic communities. The ‘Chronological framework for
the Copper Age cultures of the Carpathian Basin and
neighbouring territories’ outlined by Pál Raczky
(1995.60, Fig. 1) was chosen as the basis for the pe-
riodisation of investigated textile tool samples.
The chronological division we used placed spindle-
whorl samples from the 1st half of the 4th millen-
nium BC in the Middle Eneolithic and samples from
the 2nd half of the 4th and the very beginning of the
3rd millennium in the Late Eneolithic assemblage.
Samples from the advanced 3rd millennium contexts
thus entered the Early Bronze Age assemblage. The
classification of the investigated cultural-historical
groups within these three periods remains arbitrary,
especially due to the transitional character2 of the
particular ‘cultures’.
Changing threads
In the process of hand spinning, the spinner conti-
nuously rotates the spindle and the spun yarn at-
tached to it. As the tool rotates, the hands are used
to draft the prepared fibres, while the weight of the
tool pulls them, and its rotation twists them into a
yarn. Tension and speed are the two main factors af-
fecting the transformation from fibre to yarn. Smo-
other and shorter fibres call for a shorter, lighter
draft and more twist to enter the fibres fast enough,
while on the other hand, coarser and longer fibres
allow for longer and heavier down-pull and less twist
without causing a break.
As textile tools for fabricating yarn, spindle-whorls
are the most commonly preserved evidence of fibre
processing and production practices found in prehi-
storic contexts. The same proved to be the case with
the sites studied in regions of South East and East
Central Europe. Fortunately, they are also the most
efficacious of textile tools for studying the differ-
ences of fibres. It has been widely accepted within
the respective field of textile archaeology that the
morphological features of the spindle-whorl relate
to its functionality. A good number of reports from
experimental archaeology have supported this de-
pendency (Grömer 2005; Mårtensson et al. 2006;
Verhecken 2010). Nevertheless, the real impact and
conditioning of these determinants continue to be
questioned and investigated (Kania 2015; Laurito
et al. 2014).
As mentioned above, both whorl size and weight
account for the rotational properties, while deter-
mining the tool’s moment of inertia. Perforation dia-
meter and position are two other features that have
a significant influence on rotation, later also accoun-
2 Despite its integration with the Boleráz group (Stadler et al. 2001.543) and mainly due to its transitional character (Horváth 2001.
83), typological fusion with Lasinja and Furchenstich (Horváth 2009.105), its stylistic affiliation with the ‘stab and drag’ or ‘Fur-
chenstich’ decoration complex (Kalicz 2001) and absolute dates that cluster between 3700 and 3500 BC (Horváth 2009.104), the
Proto- Boleráz sample recorded from the Abony site was placed within the Middle Eneolithic assemblage together with Lasinja and
Furchenstich spindle-whorls. The Vu≠edol sample recorded from the Gomolava and Slav≠a-Nova Gradi∏ka sites was placed within
the Early Bronze Age assemblage together with Somogyvár-Vinkovci spindle-whorls. Mainly because the big sample recorded from
the Ljubljansko Barje-Ig site, which is only relatively dated (Koro∏ec, Koro∏ec 1969) to the 3rd millennium BC, did not permit finer
classification according to the particular cultural-historically defined phases.
Ana Grabund/ija, Emmanuele Russo
306
ting for the whorl’s stability during the process. Per-
foration is to some extent indicative of the spindle
on which the whorl was used, thus its dimensions
and properties have to be considered, since they
also influence rotation.
In the context of the proposed functionality, the mo-
ment of inertia and weight of the whorl are given
the central role, mainly because they are connected
with both the raw material (Verhecken 2010) and
the properties of the spun thread (Bohnsack 1981;
Crewe 1998). It is generally accepted that lighter
weights would be preferred for thinner and lighter
threads and heavier for thicker and heavier yarns
(Andersson 2015.47). This quality standard is thus
taken as the parameter which has the most influ-
ence on the end products and their characteristics.
The fact that the fibre flax plant would have pro-
vided fairly long fibres, especially compared with
the much shorter length of wool expected for the
early woolly sheep, is the main reason this distinc-
tion of raw materials resonated in the dichotomy be-
tween spindle-whorl types, with an emphasised con-
trast between big and heavy and small and light ca-
tegories.
Climate trends
In recent decades, the media have taken a great in-
terest in future climate change and its possible im-
pact on society. Adaptation and mitigation policies
have become main discussion topics in scientific re-
search. Analyses of past events are often referred to
for predictions of future developments, so archaeo-
logical and historical data substantially contribute to
our understanding of the effects of climate change
and its influence on culture and society. Most im-
portantly, the way civilisation reacted or adapted to
different environmental conditions in the past could
give us vital information on how to respond to the
effects of renewed climate stress (Riede 2014).
In order to interpret altering technologies by distin-
ctions between fibres and their reliance on cultural
and environmental contexts, our research analysed
Eneolithic and Early Bronze Age textile tools in the
light of the results of a climate simulation for the pe-
riod of interest. This comparison of the two sets of
data revealed that climate changes might have sig-
nificantly influenced the observed dynamics of pro-
duction traditions. Thus, our approach is a valuable
interdisciplinary case study that incorporates bi-uni-
vocal examples of the suitability of archaeological
analysis, firstly, regarding its use as a comparison
data frame for climate modelling results and, second-
ly, for evaluating the effects of past climate change
on culture, particularly innovation and technology.
The results of climate model simulations are parti-
cularly suitable for analysing cultural and technolo-
gical dynamics, since they reveal trends which can
explain certain continuities, as well as changes at a
higher temporal resolution. This means they enable
a wider understanding of the climate’s impact on pre-
historic societies, since their coverage builds an en-
vironmental context for the study of both adaptation
processes and their consequences.
Materials and methods
Scope of the study
In order to avoid the absence of contextual data and
uncertainties regarding ‘cultural’ and chronological
assignments, the choice and collection of material
used in this research focused as much as possible
on recent excavations, with the exception of several
sites3. A textile tool database was created that com-
prises spindle-whorls from over 20 archaeological
sites4, which gave a presentable transect from the
northern slopes of the Central Balkan Mountains in
the south and the Danube River in the east to the
foothills of the Carpathian Mountains and the South
East Alps in the north and west (Fig. 1).
The period of interest, beginning with the Middle
Eneolithic (Lasinja spindle-whorl sample represent-
ing the earliest tools) and ending with the Early
Bronze Age (Somogyvár-Vinkovci spindle-whorl sam-
ple representing the latest tools) falls roughly be-
tween the middle of the 5th and the end of the 3rd
millennium BC. Within the geographically defined
region, changing environmental conditions were in-
vestigated with the use of moderate-resolution cli-
mate simulations. In the investigation of change and
possible innovation in fibre production and pro-
cessing, both climate information and the recorded
archaeological evidence were considered, analysed,
compared and finally discussed for possible corre-
lations.
3 Ljubljansko Barje-Ig and Dobanovci were excavated during the 1880s and the 1960s, while Gomolava and Slav≠a-Nova Gradi∏ka
are the only systematically excavated sites included in the study.
4 Kostolac: two sites (172 tools), Classical Baden: eleven sites (147 tools), Boleráz group: two sites (32 tools), Furchenstich complex:
ten sites (205 tools), Lasinja: six sites (47 tools), Proto-Boleráz: one site (37 tools), Vu≠edol: two sites (38 tools), Somogyvár-Vin-
kovci: four sites (17 tools), Late Vu≠edol/Somogyvár-Vinkovci: one site (141 tool).
Tools tell tales – climate trends changing threads in the prehistoric Pannonian Plain
307
The results of the comparative analysis present the
impact of dynamic climatic conditions on two chan-
ges in textile production which influenced spinning
technology: firstly, through the adaptation of the
fibre material resources, and secondly, through the
modification of final product demand. Both trends,
occurring in the 4th and becoming more pronounced
in the 3rd millennium BC, were mainly quantitati-
vely observed and discussed as the adjustment of
tools’ morphological, and thus functional features.
Therefore, both their typological and functional va-
riability were addressed.
The recorded trends in tool assemblages presented
in this paper can be attributed to more than two sug-
gested factors that may be equally culturally and en-
vironmentally promoted. Nonetheless, from the tech-
nological point of view, they are the most frequent-
ly discussed and considered in the framework of
archaeological textile research as being the factors
responsible for determining a tool’s functionality
(Loughran-Delahunt 1996; Grömer 2005; Mårtens-
son et al. 2006; Verhecken 2010). The approach pre-
sented here draws attention to prehistoric textiles
in the archaeological record, which enables a closer
insight into manufacturing traditions and produc-
tion technologies that can benefit our understand-
ing of social and cultural realities and their changed
economic strategies.
Our main objectives were intended to:
❶ Explore and detect trends of change within the
Pannonian Plain textile tool sample.
❷ Elucidate the chronological, cultural-historical and,
to some extent, environmental context of these
trends.
❸ Model and interpret the main climate change
trends within the geographical and chronological
range of the research.
❹ Correlate both sets of trends and consider their
connection.
❺ Explore the idea of climate conditions within the
spatio-temporal context of the textile production
structures.
Spindle-whorl analysis
The metrical standards outlined above have to be
considered cautiously, because the spinning process
is influenced by a combination of many factors. In
fact, recent extensive spinning experiments (Kania
2015) designed to shed more light on how fibre and
yarn depend on the tool’s weight and moment of
inertia revealed how both of these variables have
little to no influence when compared to the human
Fig. 1. The distribution of the sampled sites.
Ana Grabund/ija, Emmanuele Russo
308
factor. In other words, a skilled spinner can spin a
variety of fibres with a variety of spindle-whorl
types and produce a proportional variety of differ-
ent yarns. In addition, Karina Grömer’s experiments
(Grömer 2005) showed that although very heavy
suspended spindle-whorls (120–140g) were not suit-
able for very fine yarns and light whorls (8–20g)
were not efficient for thicker yarns, a medium weight
(40g) would be suitable for almost every thickness
documented in prehistoric Europe.
Although the influence of the tool’s weight and its
moment of inertia can to a certain degree be over-
come by the spinner’s skill, the substantial differ-
ence in tensile strength between the two fibres in-
vestigated (fibre flax and early wool) should be dis-
tinguishable, if not in the distribution of medium
spindle-whorl types, then in the distribution of ex-
tremely different tool categories. The functionality
analysis was thus chosen as appropriate with regard
to the big size of the considered sample and big dif-
ference in tensile strengths of the raw materials in-
vestigated. The sampling method accounts for the ro-
bust patterning; its purpose was to reveal interpre-
table trends in textile technologies within a broad
chronological context. Although moderate spindle-
whorl types are especially difficult to interpret, the
significant difference between the distributions of
extremely different tool types should be considered
as an indication of substantial differences in spin-
ning technologies.
During the three-year study, 1152 spindle-whorls
were recorded in a textile tools database. Initially,
the study included tools from 34 archaeological sites,
but subsequently the final analysis was conducted
on a limited sample of 836 spindle-whorls5. Each in-
vestigated cultural-historical group was documented
at two or more sites, with the exception of the Proto-
Boleráz group. Middle Eneolithic tools account for
34.6% (289), the Late Eneolithic for 42.0% (351) and
Early Bronze Age for 23.4% (196) of the entire sam-
ple of spindle-whorls. The recording protocol was
based on four main measurements (whorl diameter,
perforation diameter, height and weight) and by cal-
culating weight/diameter and height/diameter ra-
tios. The ranges and mean values for all the record-
ed variables are given in the chart (Fig. 2).
The analysis segment of this paper reports on 328
whorls which were completely preserved, 163 whorls
preserved in half, 223 partially preserved whorls
and 122 whorls that had small fragments (less than
10%) missing6. All spindle-whorls in the database
were documented in order to provide three sepa-
rate categories of information: contextual, morpho-
logical and illustrative. Not all specimens provided
equal amounts of information, and for the purpose
of the functionality analysis a minimum criteria was
applied: reliable chronological assignment, typolo-
gical determination and the above-mentioned mea-
surement standard.
In order to investigate the change in the metrical
features of the spindle-whorls, the tools were divid-
ed into height-diameter and weight classes. This
kind of typology served as a basis for their functio-
nality analysis. The entire sample (836 spindle-
whorls) was divided into three main weight types
(light, medium, heavy) and height types (flat, high,
and steep), according to the distribution of the spin-
dle-whorl’s height-diameter ratio and weight values
(Fig. 3).
Additionally, due to the extremely high values, a
fourth weight type was defined. The weight values
5 The sample standard was set at a minimum of three recorded spindle-whorls with complete metric data and chronological place-
ment. Furthermore, the Neolithic spindle-whorl sample was not included in the study. No Early Eneolithic spindle-whorls were re-
corded in the database, thus no continuity or change in textile production could be thoroughly investigated. Also, the climate si-
mulations did not cover a substantial part of the period.
6 The weight values of the spindle-whorls in the sample were documented in four different reliability categories depending on their
preservation status. The weights of complete samples were documented in the complete weights category; weights of almost com-
plete samples with small fragments missing were documented in the estimated weights category (estimated weight ≈ weight if not
complete); weights of samples preserved in half were documented in the calculated weight category (calculated weight = weight
if not complete doubled), and finally, weights of partially preserved samples were documented in the reconstructed weight cate-
gory (reconstructed weight = density x volume). Volume and density variables were provided from virtual (three-dimensional)
models created for the partially preserved spindle-whorls.
Fig. 2. Table showing ranges and mean values for
all the recorded variables in millimetres and grams.
Valid Number 836 Min. Max. Mean
diameter 26.00 96.00 56.4011
height 5.00 77.60 25.5841
perforation diameter 3.50 17.00 7.9317
weight 6.39 317.70 68.3354
height-diameter ratio 0.10 1.32 0.4559
weight-diameter ratio 0.14 4.25 1.1400
Tools tell tales – climate trends changing threads in the prehistoric Pannonian Plain
309
of very heavy spindle-whorls are generally connect-
ed to substantially long fibres and plying (Barber
1991.52; Hochberg 1979.21) heavier thick yarns or
ropes (Vakirtzi 2014.53). Represented in 10% of
the entire sample, spindle-whorls weighing above
140g were considered appropriate for the addition
of the very heavy weight type, making this class as-
sociable with very long, presumably plant fibres and
plying (Fig. 4).
As already mentioned, early wool was expected to
have shorter fibre lengths, thus the 33g upper limit
for light spindle-whorl category was considered rea-
sonable (making this weight type associable with
wool)7.
Model simulations
To our knowledge, the results of climate simulations
have not been employed so far to investigate chan-
ges in climatic conditions for a chosen region and
determined period. The decision to adopt this par-
ticular approach as support for our analysis was
made for several reasons. Firstly, the model allows
full coverage of the area investigated for the entire
period in question. Furthermore, the temporal reso-
lution of such analysis allows us to focus on differ-
ent temporal scales, leading to the representation of
inter-annual to millennial climate variability. But,
most importantly, besides allowing the simulation of
possible changes of different climatic parameters,
the model also allows us to find plausible physical
interpretations for them. Nonetheless, as support for
the conclusions based on the model results, we ad-
ditionally address several studies which rely on proxy
reconstructions.
A transient continuous climate simulation was per-
formed with the coupled atmosphere-ocean circula-
tion model ECHO-G, composed by the ECHAM4
(Roeckner et al. 1996) and the ocean model HOPE
(Wolf et al. 1997), at a spectral resolution of T30
(_ 3:75o_3:75o). The model was driven by changes
in the Earth’s orbital configuration calculated by An-
dre Berger and M. Loutre (2002), changes in solar
activity as reconstructed by Sami K. Solanki et alii
(2004) and variations in GHGs concentrations de-
duced from air trapped in ice cores (Flückiger et al.
2002). Further information on the simulation reali-
sation is provided in Sebastian Wagner et alii
(2007). The results of the aforementioned simula-
tion have been used in different studies: within
these Wagner et alii (2007) used the model results
in order to investigate the Mid-Holocene hydrologi-
cal climate in Southern Patagonia. Bijan Fallah et alii
(2015), Jonas Berking et alii (2013) and Emannu-
ele Russo et alii (2016) used the model’s outputs in
order to run a time-slices experiment for different
areas and periods of study.
The region considered in our analysis extends from
longitude 14W to 21W and from latitude 42N to
48N, covering most of the Pannonian Plain region
from the southern borders of Serbia in the south to
Northern Hungary in the south. The analysis covered
the period between 7000 BP and 3000 BP, taking
into consideration the different variables that were
particularly relevant for plant growth, such as near
surface temperature, humidity balance (i.e. precip-
itation minus evaporation) and rising temperature
N 836 height\diam. weight
(mm) (g)
(%) 10 0.20 22.00
20 0.23 FLAT 27.40 LIGHT
30 0.30 33.00
40 0.35 41.30
50 0.43 HIGH 56.00 MEDIUM
60 0.52 68.00
70 0.60 82.00
80 0.68 STEEP 104.40 HEAVY
90 0.80 140.00 (πVERY
HEAVY)
Fig. 3. Table of height-diameter ratios and weight
classes with considered values.
Fig. 4. Late Eneolithic spindle-whorl (from the Ko-
stolac context) beside a Late Neolithic loom-weight
from the Sopot context found at Slav≠a-Nova Gra-
di∏ka, demonstrating the large size of the very
heavy spindle-whorl type.
7 Compared to wool, plant fibres are more often considered to be spun with heavier weights Barber (Barber 1991.25). Very heavy
whorls, respectively 100g and above, could be connected with full-length flax or long-staple wool of later specialised breeds, while
this amount of tension is supported only by very long fibres. For spinning medium to heavy wool more moderate weights would
be more adequate, at around 30g, respectively (Gleba 2008.103–106).
Ana Grabund/ija, Emmanuele Russo
310
of days with above 5 degrees (Masson et al. 1999).
Sea level pressure anomalies were also considered.
Results
Adapting technologies
Regarding the robust period to period comparison,
the observed differences in spindle-whorls’ weight
and diameter distributions pointed to a significant
increase through time in values for both variables.
The lowest weight variability is obvious in the ear-
liest (Middle Eneolithic) and the highest in the latest
(Early Bronze Age) assemblage. Nevertheless, in the
case of Early Bronze age tools, it is noticeable that
the previously wide distribution of the Late Eneoli-
thic values was limited to a noticeable degree in the
frequency of the large diameter sizes (Fig. 5).
A similar trend can be observed regarding the corre-
lated height-diameter and weight categories. In the
earliest, Middle Eneolithic assemblage, light and me-
dium weight types are convincingly predominant,
and appear to be clustered in only one height group:
the lowest category or the flat spindle-whorl type.
Almost the opposite distribution is evident in the
latest, Early Bronze Age assemblage. Here, again,
most of the tools cluster in one height group, but
this time in the highest category of the steep spin-
dle-whorl type. Although in this period all four
weight groups appear to be more evenly distributed
in this category, heavy and medium spindle-whorls
are the most represented. The Late Eneolithic assem-
blage seems to fit the transitional period between
these two extremes. First, because it shows the in-
crease in the amount of heavy weight types, a trend
which displayed continuity in regard to the Early
Bronze Age period. Second, because the category of
medium weight spindle-whorls revealed the most
balanced distribution across all three different height
groups, with the highest frequency occurring in the
moderate (high type) height group (Fig. 6).
When analysed in the context of shape variability,
the largest weights (heavy and very heavy spindle-
whorl types) are mostly represented in the biconical
and conical groups, those of the moderate size (me-
dium weight spindle-whorl types) in biconical, con-
ical, convex and discoid groups, while, finally, the
Fig. 5. The weight-diameter distribution for each period separately (a, b and c) and diameter-height di-
stribution chart shown for all respective periods together (d).
Tools tell tales – climate trends changing threads in the prehistoric Pannonian Plain
311
smallest weights (light weight spindle-whorl type)
appear to be characteristic of the discoid, convex,
and, again, biconical groups. Finally, it is obvious that
biconical and conical spindle-whorl shapes account
for the greatest weight variability in the sample (Fig.
6). Both types are most represented in the Late
Eneolithic sample, while the biconical type is the
single predominant type in the later, Early Bronze
Age assemblage. Convex, followed by discoid types
are convincingly the groups most typical of the ear-
lier Middle Eneolithic assemblage (Fig. 7).
Although the maximum diameter and perforation
diameter variables seem to behave quite proportion-
ally to one another, it is possible to distinguish some
preferences and tendencies when comparing peri-
ods. Smaller maximum diameters and smaller per-
foration diameters seem to be more typical of the
Middle Eneolithic period, while the Late Eneolithic
values cluster towards the higher spectrum in the
case of both variables. Early Bronze Age values show
the greatest variability in their distribution, but even
in this case, the inspected variables still correspond
proportionally (Fig. 7). This parameter was primar-
ily tested for samples with significantly non-propor-
tioned correlations, since they would influence the
ensuing assumptions, attributing a greater influence
of spindles to the tools’ rotation. These would sub-
stantially change generalisations regarding rotation-
al properties, which are based solely on the whorls’
morphological and weight parameters.
If we consider how the spindle-whorl’s rotational cha-
racteristics, particularly its moment of inertia, de-
pend on the tool’s geometry and weight, a heavier
whorl of significant height could have similar rota-
tional properties to a much lighter tool of smaller
height and bigger maximum diameter. The princi-
pals of this approximation, presented by Tomasz
Chmielewski and Leszek Gardyński (2010), were
used to compare widely defined height and weight
categories of tools, not specific objects and values8.
Fig. 6. The distribution of
weight classes according to
height type (a, b and c) and
shape type (d), shown for
each respective period sepa-
rately.
8 Corresponding correlations were suggested by Chmielewski and Gardyński (2010.876): “the height-diameter ratio corresponds
quite accurately with the modulation of the basic technical parameters of the tool that is its weight and angular mass.
Changing both of the qualities influences the functional characteristics of a given spindle-whorl. For the increase in the angu-
lar mass of the tool, in order to slow down its rotational speed, it is purposeful to increase its maximum diameter (whereas
this parameter will be characterised with the higher variability). On the other hand, if a high increase in the mass moment
of inertia is undesirable (due to the use of shorter fibres that require more twist), but is needed to enlarge the mass of the
spindle itself in order to create better yarn tension, the height of the spindle-whorl will increase rapidly, whereas diameter
variability will be reduced.”
Ana Grabund/ija, Emmanuele Russo
312
The differences in geometry among recorded spin-
dle-whorl shapes enabled only rough comparisons
and general assumptions. These were based on the
expectation that a substantial increase in the spin-
dle-whorl’s height (Fig. 8) would lower the tool’s
moment of inertia if its weight remained constant.
Acceptance of this relational principle allowed fur-
ther correlations between different weight and
height types (medium-steep with light-high type, me-
dium-flat with heavy-high type, heavy-steep with me-
dium-flat type and light-flat with medium-steep type),
based on their comparatively corresponding or ra-
ther similar rotational traits. If we consider these
assumptions in the fibre material context it becomes
obvious that the medium weight type from the mid-
dle-high height class is the most ‘flexible’ among all
groups of spindle-whorls, presumably affording a
comparable twist or tension performance with the
greatest number of different category types (Fig. 9).
This accounts for the Late Eneolithic transitional
character in the technological sense, since it shows
how the most ‘flexible’ spindle-whorl category is
highly represented in its assemblage when compared
to samples from the other two periods (Fig. 6). The
greatest height type variability in the Late Eneolithic
assemblage could be interpreted as the intentional
manipulation of the tool’s height properties, possi-
bly due to the avoidance of a drastic increase in the
spindle-whorl’s mass moment of inertia. This kind of
tool adaptation is desirable, for example, in process-
ing shorter and lighter fibres which require more
twist, while the enlargement of the mass of the spin-
dle itself is intended, for instance, to obtain better
yarn tension (Chmielewski, Gardyński 2010.876).
During both the Middle and Late Eneolithic period,
light steep types of spindle-whorls are under-repre-
sented, but in the case of the light high types, the
Middle Eneolithic shows an increase in frequency,
which could be associated with wool spinning. Ad-
ditionally, the recently reported results of a techno-
logical analysis of 4th and early 3rd millennium BC
textile tools from Arslantepe in Turkey (Laurito et
al. 2014) support this conclusion. Furthermore, the
above-mentioned spinning experiments were de-
signed to prove that the wider hemispherical and
heavier spindle-whorls (comparable to the light high
type) were the most efficient tools in terms of quan-
tity and quality of yarn. Also, this category proved
to be proficient in processing both vegetal and wool-
len fibre and suitable for all thread thicknesses9. In-
terestingly, the single bone spindle-whorl in our
sample10 closely resembles (in shape, method of
production, and recorded size and weight values)
the aforementioned 4th millennium bone tools from
Arslantepe, proposed for spinning wool (Fig. 10). It
Fig. 7. The distribution of spindle-whorl maximum diameter and perforation diameter values (a). Histo-
grams presenting their ratio variable distribution against period (b). The spindle-whorl shape type dis-
tribution for all three periods separately (c).
9 Surprisingly, it has been suggested that the smaller and lighter clay spindle-whorls from Arslantepe were unsatisfactory for spin-
ning wool. Furthermore, authors explain how their moment of inertia was too small for short animal fibres, pointing out that wool
used in the experiment had presumably longer fibre lengths than early wool. They do propose that metal spindles might have been
used to increase the moment of inertia of the small whorls, making them suitable for spinning animal fibres (Laurito et al. 2014).
10 Bone spindle-whorls are very rare in the Pannonian Plain during this period, and the particular find was initially published as
a bone pendant (∞ataj 2009.31).
Tools tell tales – climate trends changing threads in the prehistoric Pannonian Plain
313
should be mentioned how the results of another re-
cent analysis (Vakirtzi 2014) of North Aegean Early
Bronze Age tools from Thasos also suggest that this
particular spindle-whorl weight class was used for
spinning wool.
Both light high and light steep tool categories become
significantly more represented in the course of the
3rd millennium, as shown by the Early Bronze Age
assemblage. These types, including the medium steep
type (if we accept the approximated comparisons ex-
plained), which is more substantially represented in
the Late Eneolithic sample, and the second most fre-
quent type in the Early Bronze Age sample, could be
appropriate for spinning shorter fibres, presumably
wool (Fig. 6). These medium steep types, although
heavier, are expected to have a significantly lower
moment of inertia due to their geometry, or more
precisely, their higher height-diameter ratio.
On the other hand, heavy and very heavy types
could be connected to processing long plant fibres.
Although they have larger height-diameter ratio val-
ues and tend to cluster in the high and steep height
groups, their weight leads to significantly larger mo-
ment of inertia values (too big for processing short
or even moderately long fibres that call for more
twist). Additionally, their weights alone are too big
for the lower tensile strengths of animal fibres. Thus
they can be considered appropriate for spinning long
plant fibres such as full-length flax. These spindle-
whorl types become predominant in the late 4th mil-
lennium and continue to be highly represented in
the ensuing 3rd millennium assemblage, as shown
by the distribution charts for the Late Eneolithic and
Early Bronze Age samples (Fig. 6).
The technological development did not reveal a lin-
ear process leading ‘from plant to animal’ fibre pro-
duction. As shown by the trends in maximum height-
maximum diameter and maximum height-weight cor-
relations, the dynamics of the use of
fibre material was more complex (Fig.
11). The greatest height value variabili-
ty apparent in the Late Eneolithic peri-
od is lowered in the 3rd millennium BC
sample. Frequency of the lowest height
values is decreased, while the maximum
diameter distribution clusters towards
the lower values.
Climate conditioning
The climate of the Holocene was highly
variable, and multiple controls must
have been responsible for this variabili-
ty (Mayewski et al. 2004). According to
Heinz Wanner et alii (2009), the Holo-
Fig. 8. Reconstructed three-dimensional models of
a flat, high and steep spindle-whorl type, based on
published conical spindle-whorls found at the Ba-
latonoszod-reti dulo site (Horváth 2013).
Fig. 9. Diagram of corresponding height and
weight types.
Fig. 10. A convex spindle-whorl made from the head of a cattle
femur bone, weighing 24.6g, from a Furchenstich (Retz-Gajary)
context found at Josipovac Punitova≠ki-Veliko polje, dated to the
1st half of the 4th millennium BC (Middle Eneolithic period).
Ana Grabund/ija, Emmanuele Russo
314
cene climate was dominated by the influence of
summer season orbital forcing; our results are in
accordance with this interpretation. In particular,
our model simulations revealed that from the 7th
millennium BP, summer climatic conditions started
to deteriorate in the middle latitudes of the North-
ern Hemisphere (NH), mainly due to a decrease in
summer insolation, which reached its maximum at
around 9000 BP (not shown). For all the simulated
climatic parameters, a pronounced trend is evident
for the summer season, which resembles one of a
complete decrease in insolation (Fig. 12), although
the humidity balance reveals an opposite, increasing
trend, mainly due to the effect of evapotranspiration
as a consequence of the enhanced insolation in the
mid-Holocene.
Such conclusions are also supported by other stud-
ies based on proxy reconstructions for the investi-
gated area and its surroundings. In particular, by
analysing Holocene conditions through pollen recon-
structions from several lakes, in Southern to North-
ern Italy and the Alpine region, Michel Magny (Magny
et al. 2012) concluded that the mid-Holocene sum-
mer precipitation regimes appear to have been oppo-
site between the south- and north-central Mediterra-
nean: a maximum was present in Sicily and a mini-
mum in north-central Italy. Later, Magny et alii
(2013) analysed a time series of wetness for the
Lake Ledro basin in Northern Italy spanning the last
10 000 years: while the early and mid-Holocene pe-
riods show relatively low flood frequency, the late
Holocene, from c. 4500 cal BP onwards is characte-
rised by an increase, which is in full agreement with
the reconstructions from neighbouring Lake Iseo.
The pattern of contrasting mid-Holocene summer
precipitation regimes was confirmed by Odile Pey-
ron et al. (2013) on the basis of pollen-inferred quan-
titative estimates and a multi-method approach that
revealed minima (maxima) of summer precipitation
and lake levels to the north (south) of c. 40° N. Sum-
mer temperature showed a similar partition for the
mid-Holocene in the central Mediterranean, with
warmer (cooler) conditions to the north (south). Mó-
nika Tóth et alii (2015) presented a Holocene sum-
mer air temperature reconstruction based on fossil
chironomids from Lake Brazi, a shallow mountain
lake in the Southern Carpathians, with results sug-
gesting that from c. 8500 cal BP, chironomid-based
summer temperatures decreased in the area. In par-
ticular, the period between 6000 and 3000 cal BP
was characterised by cooler temperatures in compa-
rison to the earlier period.
Conversely, winter-time model results do not show
a clear trend and the data are characterised by sig-
nificant multi-decadal to inter-centennial fluctuations
(Fig. 13). For winter, the trend of insolation during
the mid-to-late Holocene at the mid-latitudes of the
NH is not particularly pronounced (Fig. 13). The be-
haviour of the simulated data is probably a result
of the system’s internal variability. In winter, clima-
tic conditions in Europe are mainly driven by chan-
ges in atmospheric circulation (Hurrel 1995; Hur-
rel et al. 2003). In particular, the first mode of win-
ter atmospheric variability over the North Atlantic is
represented by the North Atlantic Oscillation (NAO),
which explains one third of the total variance in the
sea-level pressure (SLP) field over the area (Hurrel
et al. 2003). A positive phase of the winter NAO is
associated with an increase in strength and a north-
ward shift in westerly winds. This results in moist
Fig. 11. The distribution of correlated: a maximum diameter and height values; b weight and height val-
ues, for the three period assemblages together.
Tools tell tales – climate trends changing threads in the prehistoric Pannonian Plain
315
and warm air advection from the Atlantic to North-
ern Europe (55°–70° N) and, consequently, in drier
and colder conditions over Southern Europe. Con-
versely, the negative phase of the NAO is accompa-
nied by weaker and southward shifted (35°–50° N)
westerlies, responsible for wetter and warmer win-
ter conditions over Southern Europe. The use of the
NAO index (Fig. 12.a), which is based on the prin-
cipal component time series of the leading Empirical
Orthogonal Function (EOF) of the winter SLP ano-
malies (Hurrel et al. 2003), allows us to speculate
that the oscillations of the simulated winter tempe-
rature and precipitation are caused by the variabil-
ity between positive and negative phases of this
atmospheric pattern. The simulated winter tempera-
ture, and particularly humidity, present an extre-
mely high anti-correlation (~ –0.7) with the calcu-
lated NAO index.
Different studies hypothesise that at around 6000 BP
the NAO was in a more pronounced positive phase
in comparison to present-day conditions (Bonfils et
al. 2004; Braconnot et al. 2007a; 2007b; Davis et
al. 2003; Mauri et al. 2014; Russo et al. 2016). This
is not directly evident in our results and seems to be
a problem characteristic of climate models. Never-
theless, the changes in the simulated winter condi-
tions can be considered physically consistent: the
model simulates a reasonable and internally consi-
stent climate (Goosse 2012; Von Storch 2004). Other
events often reported in studies that are based on
proxy-reconstructions are the so-called Rapid Climate
Changes (RCC) (Mayewski et al. 2004; Budja 2015;
Weninger et al. 2009; 2015; Wanner et al. 2011).
They are abrupt cooling events, lasting from a cen-
tury to millennia, for which large drifting ice debris
has been registered. The complexity of our model
Fig. 12. Graphs presenting: a summer temporal evolution of near-surface temperature regional mean
computed over the area of study; b time series of summer precipitation + evapotranspiration calculated
in mm/day; c insolation changes over the period of study; d time series of rising temperature days above
5 degrees. All the time series are smoothed using a 200-year running mean.
Ana Grabund/ija, Emmanuele Russo
316
does not permit a precise reproduction of such
events. Additionally, even if evidence supporting
them seems clear, their cyclic occurrence and caus-
es are still highly debated, as well as the impact they
had on a regional scale (Wanner et al. 2011). This
is one of the main reasons we focus our attention on
long-term trends and the relevance of abrupt oscil-
lations in the context of population adaptability.
Nonetheless, in our discussion, we take into consi-
deration one particular example of RCC and the pos-
sible impact it had on the life of Eneolithic commu-
nities inhabiting the area. Although in the case of
winter, no distinct trend is evident in our simula-
tions, and the model results do not display the same
level of correspondence with the other studies (as
they do in the case of summer, particularly tempera-
tures and moisture balance), the evident oscillations
are physically plausible. Thus, acknowledging the
possibility of the effects of RCCs during the simulat-
ed period does not contradict the model results at
all. On the contrary, it means that if those events
occurred, they only contributed to the evinced dete-
rioration in climatic conditions.
Discussion
The period between the mid-5th and the end of the
3rd millennium BC, in which the spread of both
fleece-bearing sheep husbandry and new fibre flax
in South East and Central Europe probably occurr-
ed, was characterised by profound social changes
and has been studied so far in terms of intensified
contacts and mobility (Anthony 2007; Leary 2014),
environmental change (Binford 1968; Barker 1985)
and economic transformation (Sherratt 1981; 1997;
2006; Greenfield 1988; 2005). This provided the
entire contextual framework and theoretical back-
ground for the present investigation of textile pro-
duction trends and their climate dependency in the
aforementioned regions. Apart from the extensive
research on textile production in Romania (Mazare
Fig. 13. Graphs presenting: a winter temporal evolution of near-surface temperature regional mean com-
puted over the area of study; b time series of winter precipitation + evapotranspiration calculated in
mm/day; c insolation changes over the period of study; d DJF NAO index over the period of study. All the
time series are smoothed using a 200-year running mean.
Tools tell tales – climate trends changing threads in the prehistoric Pannonian Plain
317
2014) and Bulgaria (Petrova 2011), which took into
consideration indirect archaeological evidence from
larger and confined geographical sections, the tex-
tile tools and production of the period have not yet
been systematically approached in the Pannonian
Plain contexts. Narrowing the focus on the particu-
lar region enabled us to raise specific questions re-
garding textile fibre raw materials and final product
demand through their dependence on the changing
climate. The declining summer trend in tempera-
tures and the corresponding rising trend in moisture
content might have caused the cultural-historically
defined groups of the Post-Neolithic inhabiting the
area to be more sensitive to possible climatic fluctu-
ations like the simulated winter oscillations. Such
events could have acted as a trigger for shifts in fibre
procurement practices, resulting in a broadening of
resources and, especially, in intensified selection in
favour of their fitness.
Evidence has shown that the period between 6000
and 5000 BP, normally referred as a Rapid Climate
Change event, was generally characterised by glob-
ally colder conditions (Weninger et al. 2009; 2011;
Mayewski et al. 2004). In this case, the effects of the
proposed scenario of a deterioration in climatic con-
ditions would eventually have been amplified, and
abrupt changes could have had even more drastic ef-
fects on the ‘cultural groups’ under consideration.
Through a complex causal network, this could have
led to substantial perturbations within human soci-
eties, as suggested by changes in the archaeological
material in Central Europe (Arbogast et al. 1996;
Berglund 2003; Magny 2004). Weninger et alii
(2009) found that at different times in the 6–5ka
RCC period, there were abrupt abandonments of va-
rious settlements in Romania, Bulgaria, and Greece.
During the same period in the Alpine region, abrupt
oscillations on the scale of decades or a century
seem to have caused changes in hunting and farm-
ing patterns (Schibler 1997).
The high-quality organic preservation at Sipplingen,
an Alpine lakeshore settlement on the northern
shore of Lake Constance in south-west Germany, en-
abled comprehensive dendroarchaeological, archaeo-
botanical and zooarchaeological investigations that
revealed the adaptive character of the settlement
and its subsistence strategies in response to chan-
ges in climate, environment and cultural influences
over time (4000–2800 BC) (Styring et al. 2016.103).
Obvious economic changes that occurred during
the 4th millennium BC, which have been attributed
to climatic deterioration (Schibler et al. 1997), re-
veal a dramatic cycle of establishment, expansion
and reduction, or even abandonment of settlements
around Lake Constance (Billamboz, Köninger 2008).
A great deal of evidence of flax cultivation has been
found from lake dwellings north of the Alps, (Jaco-
met 2009). In particular, the find densities of flax
rise considerably during the second half of the 4th
millennium BC (Brombacher, Jacomet 1997.249).
It is most likely that during the Middle Eneolithic
period, which marked the turn and the first centu-
ries of the 4th millennium BC, animal products, pre-
sumably including fibres, started to be more impor-
tant as a result of deteriorating climate conditions.
All the scarce textile evidence favours wild plant
species, but the functional analysis of the Furchen-
stich and Lasinja spindle-whorls suggests that dur-
ing this, apparently, intensive foraging period, ani-
mal fibres should not be disregarded. Faced with
an environmental challenge due to the whimsical
climate in the following centuries of the 4th millen-
nium BC, in the context of intensified textile pro-
duction, foraged plant material and possibly the old
variety of flax might have been partially substitut-
ed with a new fibre fax in order to meet the grow-
ing demands of textile production. It is premature to
conclude this, but the above-mentioned archaeobota-
nical research supports this idea.
The earliest direct evidence of wool, respectively
woollen finds from Germany (Rast-Eicher 2014.16)
and the North Caucasus (Shishlina et al. 2003), when
observed in the framework of changed herding stra-
tegies revealed in the contemporary contexts of
South East and Central Europe11, intensified sheep
husbandry recorded at some sites12 and the spin-
11 Regarding Central Europe, analysed faunal evidence from the Bronocice site in South Eastern Poland (Pipes et al. 2014) suggest
that sheep rearing intensified during the second half of the 4th millennium BC (Funnel Beaker horizon). Indeed, the increased
herd sizes reported from the site revealed that the majority of animals were slaughtered as adults. In the case of South Eastern
Europe, analysis of Late Eneolithic faunal samples from the Bubanj and Mokranjske Stene-potkapina sites in Serbia suggests the
predominance of sheep husbandry (Bulatovi≤ 2012; Bulatovi≤, Milo∏evi≤ 2015). Furthermore, analysis of the age profiles from
the Mokranjske Stene-potkapina sample attributed to the Kostolac-Cotofeni horizon (late 4th to early 3rd millennium BC) revealed
that the adult age group was the most represented.
12 Intensified sheep husbandry was recorded at the Middle Eneolithic Furchenstich site at Pod Kotom – jug pri Krogu I/II (πavel
2009), at the Middle to Late Eneolithic Proto-Boleráz site at Abony (Fábián 2008; 2012), and at Late Eneolithic Boleráz/Baden
sites at Balatonőszöd-Temetői dűlő (Horváth 2010) and Balatonkeresztúr-Réti-dűlő (Fábián 2014).
Ana Grabund/ija, Emmanuele Russo
318
dle-whorl analysis results, push the presence of wool
in the context of fibre production already into the
4th millennium BC.
Following the results of our analysis and according to
the aforementioned studies, it is very likely that the
middle of the 4th millennium BC (characterised by
the appearance and spread of Baden culture in the
Pannonian Plain) was a period of major climatic
stress. Such pressure probably promoted further ma-
stering of both fibre resources, thus influencing their
specialisation. Intensified sheepherding is usually con-
nected to this cultural group (Maran 1998.514–516),
and many authors propose its acquaintance with
wool (Horvath 2010; Bondar 2012; Struhar et al.
2014). On the other hand, the level of flax pollen in
samples from lake-shore settlements in Switzerland
indicates intensive flax production, more recognisable
in the frequent textile remains (Capitani et al. 2002.
115–120). This allows us to speculate that, during
the course of the second half of the 4th millennium
BC, both fibre materials gained in importance.
By the beginning of the 3rd millennium BC, such
strategies had already been mastered and spread,
since textile production does not seem to have slow-
ed down at this point. On the contrary, judging by
the increase in the amount of spindle-whorls from
the end of the Eneolithic period, it seems it even in-
tensified. More organised fibre material resources
had to have been acquired and established by that
period. Both the changed environmental and cultur-
al conditions resonated further in the new demands
for specific types of textile products, which quite
possibly corresponded to the intensified use of trans-
port and traction. The predominance of very large
tools, probably used for plying plant fibres or their
filaments in order to make thick and strong threads
and ropes which could have been used for harnes-
sing animals, would support this scenario. Mobility
and the transport of goods would also have pro-
moted the trend. The increased variability of Late
Eneolithic spindle-whorl assemblages in favour of a
balanced distribution of moderate weight types be-
tween three main height categories is something
that might reflect this kind of broadened fibre as-
sortment, both in regard to the raw material resour-
ces and demand for end products, although the ob-
vious predominance of heavy and very heavy types
of tool, particularly in high and steep height catego-
ries, greatly accounts for longer plant fibre prefer-
ences. This trend is particularly pronounced at the
turn of the millennium, and continues in the Vu≠e-
dol and Somogyvár-Vinkovci phase. There is a pos-
sibility that this reflects the appearance of specia-
lised fibre flax and its intensified use in the plying
tradition of the late 4th and the following 3rd mil-
lennium BC. Impressed plied cord decoration (Grö-
mer, Kern 2010; Leghissa 2015) representative of
the Early Bronze Age Corded Wear complex (wide-
spread in the Pannonian Plain) could attest to the
growing importance of plied cordage in the course
of the following centuries.
Conclusions
Three main trends are evident in the spindle-whorl
sample. First, the Middle Eneolithic assemblages sug-
gest more intensive animal husbandry, which led
to the first exploitation of animal fibres. Second, on
the basis of the Late Eneolithic tools, continued use
of wool may be argued, which was substantially ac-
companied by the intensified production of long
plant fibre, presumably flax. And third, the results
of the analysis of Early Bronze Age spindle-whorls
reveal a specialisation of fibres, most probably due
to the refinement of both textile fibre resources and
demand for final products.
Despite the large textile tool sample on which this
research is based and regardless of the climate con-
ditions that support the outlined timeline for the
introduction of both sheep wool and the new fibre
flax, already during the Eneolithic in the Pannonian
Plain, further zooarchaeological and archaeobota-
nical evidence is required to understand and explain
the dynamics of these processes in finer detail. The
exploration of the new fibre resources occurred si-
multaneously in remote contexts and could have
been provoked by both cultural and environmental
conditions.
Numerous interacting factors were probably res-
ponsible for the intensification of their exploitation
and, to some degree, their further development,
which might not have been as linear as we thought.
Thus their evolutional character is contrary to the
revolutionary scenario. They gradually developed
from ordinary foraged fibres into occasional and
then into common secondary products. Due to the
intensification of their use, they finally became pri-
mary and specialised resources for developed textile
production. It is plausible that, due to different fac-
tors in different areas and periods, similar reasons
promoted fibre production as those that later con-
tributed to their specialisations. Although the earli-
est direct evidence for wool in the studied region
dates to the 2nd millennium BC (Grömer et al. 2013;
Tools tell tales – climate trends changing threads in the prehistoric Pannonian Plain
319
Rast-Eicher, Bender Jørgensen 2013), its much ear-
lier presence must be considered.
There are two possible reasons why in Europe the
scant textile evidence sample from the 4th and 3rd
millennium lacks woollen examples: first, its under-
represented use in comparison to flax and other
plant fibres, and second, the fact that the European
climate is not conducive to its preservation condition-
ing (Ryder 1964). Regarding resources for fibre pro-
duction, the Middle Eneolithic could mark their ini-
tial exploitation phase, while the Late Eneolithic, with
evident technological change which gradually result-
ed in greater tool standardisation in the Early Bronze
Age, marks their second exploitation phase. The ex-
ploration phase for both resources might have al-
ready started in previous periods, and so future tex-
tile research of Early Eneolithic tools might clarify
this dynamic. The results of the analysis suggest that
in the studied region of South East and Central Eu-
rope, wool was probably already being exploited in
the 4th and certainly the 3rd millennium BC, although
on a smaller scale, first in comparison to flax, and
also in comparison to the established wool econo-
mies of contemporary urban centres in the Near East.
The continuous deterioration in summer climatic
conditions in the Pannonian Plain region during
these two millennia would still have provided con-
ditions congenial to fibre flax. Areas with less fa-
vourable conditions for cultivating it might have
focused their developing textile production on other
resources, causing them to specialise in alternatives.
This is a question of difference in fibre specialisa-
tions, which so far have been interpreted as indi-
cations of use, creating an unacceptable dichotomy
in the period of intensified mobility and contacts.
A growing demand for raw materials, and metal in
particular, resulted in intensified trade that could
support this kind of fibre competition through divi-
sion in specialisation during the 4th millennium,
while textiles, and presumably fibres, may have been
substantially involved in both short- and long-dis-
tance trade. Metal objects were certainly accountable
for far-reaching cultural contacts, possibly facilitating
breed mobility, which then accelerated the gene
flow and the evolution of both fibres. Copper axes
from burial mounds at Moravian TRB-Boleráz sites
were wrapped in flax textiles, which could account
for a combination of two luxury goods (Baldia et al.
2008.264–265). Sherratt (Sherratt 1997; 2006) con-
nected both the animal traction and wool with social
stratification, explaining them as symbols of elites.
Quite possibly, this connection should be to some
extent related to inequalities between animals and
plants reflected in the people who had privilege to
select among them. Although wool’s appearance and
origin were for a long time considered within the
textile economies of the urban revolution package
in Bronze Age Mesopotamia (McCorriston 1997;
Kimbrough 2006), the oldest finds, genetic evidence,
and the analysis of the prehistoric textile tools push-
ed its early appearance on the fibre scene into a new
and much wider framework regarding its theoretical
(SPR), geographical (Fertile Crescent), chronological
(Bronze Age) and social (stratified organised sys-
tems) context.
On the other hand, archaeobotanical research on
the flax plant, taken in consideration with the pro-
cessed textile evidence from circum-Alpine pile dwel-
lings, place the fibre evolution in the same period
of the thriving copper trade in the 4th millennium
BC, suggesting that both resources were exploited
in the production of Eneolithic textiles. Although
both scenarios were connected to many cultural and
environmental trends, climate change was certainly
accountable, if not for their evolution, then for in-
fluencing the dynamics of their exploitation.
During the late 4th and early 3rd millennium BC, an
interesting example of a technological shift occurr-
ed, one which accommodated to both new options
of presumed raw material resources, and new de-
mands for a seemingly greater variety of end prod-
ucts. The moderate weight types of spindle-whorls
significantly represented in all height classes during
this technological phase suggest that all of the fibres
considered could have been used. This was evident-
ly accomplished through a gradual change in tool
properties, while assemblages were consecutively
adjusted.
Although the boom in heavy and very heavy types
of spindle-whorls is more obvious and easier to con-
nect with long plant fibres, and the appearance of
smaller sizes in the Early Bronze Age is something
we search for when we expect wool use, the modest,
ordinary, average, hard-to- interpret type of tool re-
vealed the biggest evolution: craftsmens’ slow and
gradual accommodation to the world around them.
Technology’s potential to be modified is mainly re-
flected in the outlined tools’ flexibility. Although its
moderate and gradual transformation does not ac-
count for a high degree of specialisation, it does re-
veal our ability to combine tradition and innovation
in altering and, more importantly, challenging cir-
cumstances.
Alfaro C. 2002. Étoffes cordées du site néolithique de Tell
Halula (Syrie – VIIIe millénaire avant J.-C.). Bulletin du
CIETA 79: 17–25.
Andersson E. 2015. Textile Production at Birka. House-
hold Needs or Organised Workshops? In Northern Archa-
eological Textiles: NESAT VII. Textile Symposium in
Edinburgh, 5th–7th May 1999. Oxbow. Oxford: 44–50.
Anthony D. W. 2007. The Horse, the Wheel, and Lan-
guage: How Bronze-Age Riders from the Eurasian Step-
pes Shaped the Modern World. Princeton University
Press. Princeton.
Arbogast R., Jacomet S., Magny M. and Schibler J. 2006.
The significance of climate fluctuations for lake level
changes and shifts in subsistence economy during the
Late Neolithic (4300–2400 cal B.C.) in Central Europe.
Vegetation History and Archaeobotany 15: 403–418.
Barker G. 1985. Prehistoric farming in Europe. Cam-
bridge University Press. Cambridge.
Baldia M. O., Frink D. S. and Boulanger M. T. 2008. The
Earthen Long-Barrow of D∫bán, Moravia, Czech Republic
and its Implications for the Interaction between the Nor-
dic Funnel Beaker and the Southern Baden Culture. In
M. Furholt, M. Szmyt and A. Zastawny (eds.), The Baden
Complex and the Outside World. Proceedings of the 12th
Annual Meeting of the EAA in Cracow 19–24th September,
2006. Studien zur Archäologie in Ostmitteleuropa 4. Bonn:
263–289.
Barber E. J. W. 1991. Prehistoric Textiles. The Develop-
ment of Cloth in the Neolithic and Bronze Ages. Prin-
ceton University Press. Princeton.
Bartosiewicz L. 2007. Making a living: further technicali-
ties. In A. Whittle (ed.), The Early Neolithic on the Great
Hungarian Plain: investigations of the Körös culture site
of Ecsegfalva 23, County Békés II. Varia Archaeologica
Hungarica 21: 733–742.
Berger A., Loutre M. F. 2002. An Exceptionally Long Inter-
glacial Ahead? Science 297: 1287.
Berglund B. 2003. Human impact and climate changes-
synchronous events and a causal link? Quaternary In-
ternational 105: 7–12.
Berking J., Körper J., Wagner S., Cubasch U. and Schütt B.
2013. Heavy rainfalls in a desert(ed) city: A climate-ar-
chaeological case study from Sudan. Climate, Landscapes
and Civilizations, Geophysical Monograph Series 198:
163–168.
Billamboz A., Köninger J. 2008. Dendroarchäologische
Untersuchungen zur Besiedlungs- und Landschaftsentwi-
cklung im Neolithikum des westlichen Bodenseegebietes.
In W. Dörfler, J. Müller (eds.), Umwelt–Wirtschaft–Sied-
lungen im dritten vorchristlichen Jahrtausend Mitteleu-
ropas und Südskandinaviens. Wachholz Verlag Neumün-
ster. Kiel: 317–334.
Binford L. R. 1968. Archaeological perspectives. In S. R.
Binford, L. R. Binford (eds.), New perspectives in archa-
eology. Aldine. Chicago: 5–32.
Bondár M. 2012. Prehistoric Wagon Models in the Carpa-
thian Basin (3500–1500 BC). Archaeolingua. Budapest.
Bonfils C., de Noblet N., Guiot J. and Bartlein P. 2004.
Some Mechanisms of mid-Holocene climate change in
Europe, inferred from comparing PMIP models to data.
Climate Dynamics 23: 79–98.
Braconnot P. and 19 co-authors 2007a. Results of PMIP2
coupled simulations of the Mid-Holocene and Last Glacial
Maximum – Part 1: experiments and large-scale features.
Climate of the Past 3: 261–277.
Braconnot P. and 19 co-authors. 2007b. Results of PMIP2
coupled simulations of the Mid-Holocene and Last Glacial
Maximum – Part 2: feedbacks with emphasis on the loca-
tion of the ITCZ and mid- and high latitudes heat bud-
get. Climate of the Past 3: 279–296.
Brombacher C., Jacomet S. 1997. Ackerbau, Sammelwirt-
schaft und Umwelt: Ergebnisse archäobotanischer Un-
tersuchungen. In J. Schibler, H. Huüster-Plogmann, S. Jaco-
met, C. Brombacher, E. Gross-Klee and A. Rast-Eicher
(eds.), Ökonomie und Ökologie neolithischer und bron-
Ana Grabund/ija, Emmanuele Russo
320
We would like to thank the TOPOI Excellence Cluster for making this research and cooperation possible, especially
our A4 – TEXTILE REVOLUTION and A2 – THE POLITICAL ECOLOGY OF NON-SEDENTARY COMMUNITIES research
groups. Finally, we express our gratitude to the many colleagues who contributed with their data and advised us
during the analysis.
ACKNOWLEDGEMENTS
References
∴
Tools tell tales – climate trends changing threads in the prehistoric Pannonian Plain
321
zezeitlicher Ufersiedlungen am Zürichsee. Ergebnisse
der Ausgrabungen Mozartstrasse, Kanalisationssanierun-
gen Seefeld, AKAD/Pressehaus und Mythenschloss in
Zürich. Monographien der Kantonsarchäologie Zürich vol.
20. Zürich: 220–279.
Burnham H. 1964. The World’s Oldest Textiles. Meeting
Place Journal of the Royal Ontario Museum 1: 105-108.
1965. Çatal Hüyük – the Textiles and Twined Fabrics.
Anatolian Studies 15: 169–174.
Bohnsack A. 1981. Spinnen und Weben. Entwicklung von
Technik und Arbeit im Textilgewerbe. Rowohlt. Reinbek.
Budja M. 2015. Archaeology and rapid climate changes:
from the collapse concept to a panarchy interpretative
model. Documenta Praehistorica 42: 171–184.
Bulatovi≤ J. 2012. Exploitation of Domestic Animals in
the Late Neolithic and Eneolithic in the Central Balkans.
Case Studies: Vitkovo and Bubanj. Journal of Serbian Ar-
chaeological Society 28: 279–300.
Bulatovi≤ J., Milo∏evi≤ S. 2015. Animal Remains from the
Mokranjske Stene-potkapina Site. In A. H. Kapuran, A. T.
Bulatovi≤ (eds.), Mokranjske Stene, Kulturno nasle∂e Ne-
gotinske Krjine. Muzej Krajine Negotin. Negotin: 41–55.
Chemikosova S. B., Pavlencheva N. V., Gur’yanov O. P.
and Gorshkova T. A. 2006. The Effect of Soil Drought on
the Phloem Fiber Development in Long-Fiber Flax. Fizio-
logiya Rastenii 53(5): 739–746.
Craig O. E., Chapman J., Heron C., Willis L. H., Bartosie-
wicz L., Taylor G., Whittle A. and Collins M. 2005. Diary
food production in Europe. Antiquity 79: 882–894.
Cummings V., Harris, O. 2011. Animals, People and Pla-
ces: The Continuity of Hunting and Gathering Practices
across the Mesolithic-Neolithic Transition in Britain. Euro-
pean Journal of Archaeology 14(3): 361–382.
Cybulska M., Maik J. 2007. Archaeological textiles – a need
of new methods of analysis and reconstruction. Fibres
and Textiles in Eastern Europe 18(3): 100–105.
Charvát P. 2011. On Sheep, Sumerians and the Early
State. In L. Vacín (ed.), U4du11-ga-nisá mu-ni-ib-du11.
Ancient Near Eastern Studies in Memory of Blahoslav
Hru∏ka. Dresden: 49–60.
Chmielewski T., Gardyński L. 2010. New frames of archa-
eometrical description of spindle whorls: a case study of
the late eneolithic spindle-whorls from the 1C site in Grö-
dek, District of Hrubieszów, Poland. Archaeometry 52(5):
869–881.
Chantraine P. 2009. Dictionnaire étymologique de la
langue grecque. Paris. Dictionnaire étymologique de la
langue grecque: histoire des mots. Klinksieck. Paris.
Crewe L. 1998. Spindle Whorls. A study of form, func-
tion and decoration in prehistoric Bronze Age Cyprus.
Paul Aström Förlag. Jonsered.
∞ataj L. 2009. Josipovac Punitova≠ki – Veliko Polje I.
Eneoliti≠ko, bron≠anodobno i srednjovjekovno naselje.
(Eneolithic, Bronze Age and Medieval Settlement). Hrvat-
ski Restauratorsi zavod. Zagreb.
Dash P. K. and 16 co-authors. 2014. Genome-wide analy-
sis of drought induced gene expression changes in flax (Li-
num usitatissimum). GM Crops & Food 5(2): 106–119.
Davis B., Brewer S., Stevenson A. and Guiot J. 2003. The
temperature of Europe during the Holocene reconstruct-
ed from pollen data. Quaternary Sciences Reviews 22:
1701–1716.
De Capitani A., Deschler-Erb S., Leuzinger U., Marti-Grä-
del E. and Schibler E. 2002. Die jungsteinzeitliche Sied-
lung Arbon-Bleiche 3. Funde. Archäologie im Thurgau
11. Frauenfeld.
Evershed R. P. and 21 co-authors 2008. Earliest date for
milk use in the Near East and southeastern Europe linked
to cattle herding. Nature 455: 528–531.
Fábián Sz., Marton T. and Serlegi G. 2008. Késő rézkori
település Abonyból (Turjányos-dűlő). In M. Gyöngyössy
(ed.), Képek a múltból. Az elmúlt évek ásatásaiból Pest
megyében. Szentendre: 24–25.
Fábián S. 2014. A badeni kultúra településtörténete a
dél-balatoni régióban az újabb kutatási eredmények
alapján. Unpublished PhD thesis. Eötvös Loránd Univer-
sity – Faculty of Humanities. Budapest.
Fábián Sz., Gábor S. and Daróczi-Szabó M. (in press). Be-
gräbnis, Massengrab oder Opferstelle? Salzmünde – rule
or exception? Proceedings of the Interdisciplinary Sym-
posium 18th–20th of October in Halle, Saale, 2012.
Fabi∏ M. 2005. Pathological alteration of cattle skeletons,
evidence for the draught exploitation of animals? In J. Da-
vies, M. Fabi∏, I. Mainland, M. Richards and R. Thomas
(eds.), Diet and Health in Past Animal Populations: Cur-
rent Research and Future Directions. Oxbow Books. Ox-
ford: 58–62.
Fallah B., Sodoudi S. and Cubasch U. 2015. Westerly jet
stream and past millennium climate change in Arid Cen-
tral Asia simulated by COSMO-CLM. Theoretical and Ap-
plied Climatology. On line: http://link.springer.com/article/
10.1007%2Fs00704-015-1479-x#/page-1
Ana Grabund/ija, Emmanuele Russo
322
Flückiger J., Monnin E., Stauer B., Schwander J., Stocker T.
F., Chappellaz J., Raynaud D. and Barnola J. M. 2002. High
resolution Holocene N2O ice core record and its relation-
ship with CH4 and CO2. Global Biogeochem Cycles 16:
10–29.
Frangipane M., Andersson E., Nosch M. L., Laurito R., Rast-
Eicher A., Möller-Wiering S. and Wisti Lassen A. 2009. Ars-
lantepe (Turkey): Textiles, Tools and Imprints of Fabrics
from the 4th to the 2nd millennium BC. Paléorient 35:
5–30.
Goosse H., Crespin E., Dubinkina S., Loutre M. F., Mann
M. E., Renssen H., Sallaz-Damaz Y. and Shindell D. 2012.
The Role of Forcing and internal Dynamics in explaining
the “Medeival Climate Anomaly”. Climate Dynamics 39:
2847–2866.
Gratten J., Wilson A. J., McRae A. F., Beraldi D., Visscher
P. M., Pemberton J. M. and Slate J. 2008. A localized nega-
tive genetic correlation constrains microevolution of coat
color in wild sheep. Science 319: 318–20.
Greif T. 1997. Prazgodovinska koli∏≠a na Ljubljanskem
barju. Arheolo∏ka interpretacija in poskus rekonstruk-
cije na≠ina ∫ivljenja / The prehistoric pile dwellings in
the Ljubljansko barje. An archaeological interpretation
and attempted reconstruction of the way of life. Arheo
18. Ljubljana.
Greenfield H. J. 1988. The Origins of Milk and Wool Pro-
duction in the Old World. Current Anthropology 29: 573–
748.
2005. A Reconsideration of the Secondary Products Re-
volution in South-Eastern Europe. On the Origins and
Use of Domestic Animals for Milk, Wool, and Traction in
the Central Balkans”. In J. Mulville, A. K. Outram (eds.),
The Zooarchaeology of Fats, Oils, Milk and Dairying.
Proceedings of the 9th Conference of the International
Council of Archaeozoology. Durham, August 2002. Ox-
bow Books. Oxford: 14–31.
Grömer K. 2005. Efficiency and technique – experiments
with original spindle whorls (Leistung und Technik – Ex-
perimente zum Spinnen mit Originalspinwirteln). In P.
Bichler, K. Grömer, R. Hofmann-de Keijzer, A. Kern and H.
Reschreiter (eds.), “Hallstatt Textiles”. Technical Analy-
sis, Scientific Investigation and Experiment on Iron Age
Textiles. British Archaeological Reports IS 1351. Archaeo-
press. Oxford: 107–116.
Grömer K., Kern D. 2010. Technical data and experiments
on corded ware. Journal of Archaeological Science 37:
3136–3145.
Grömer K., Rösel-Mautendorfer H. and Bender Jørgensen
L. 2013: Visions of Dress. Recreating Bronze Age Clothing
from the Danube Region. Textile Journal of Cloth and
Culture 11(3): 218–241.
Harris S. 2014. Flax fibre: Innovation and Change in the
Early Neolithic. A Technological and Material Perspective.
Textile Society of America Symposium Proceedings. Pa-
per 913. On line: http://digitalcommons.unl. edu/tsaconf/
913
Helbæk H. 1959. Notes on the Evolution and History of
Linum. KUML: årbog for Jysk Arkaeologisk Selskab.
Stiftsbogtrykkeri. Aarhus: 103–129.
1963. Tetxiles from Çatal Hüyük. Archaeology 67: 39–
46.
Halstead P., Isaakidou V. 2011. Revolutionary secondary
products: the development and significance of milking,
animal-traction and wool-gathering in later prehistoric
Europe and the Near East. In T. Wilkinson, S. Sherratt and
J. Bennet, J. (eds.), Interweaving Worlds: Systemic Inter-
actions in Eurasia, 7th to 1st Millennia BC. Oxbow Books.
Oxford: 61–76.
Herbig C., Maier U. 2011. Flax for oil or fibre? Morpho-
metric analysis of flax seeds and new aspects of flax cul-
tivation in Late Neolithic wetland settlements in south-
west Germany. Vegetation History and Archaeobotany
20(6): 527–533.
Hochberg B. 1979. Spin, Span, Spun: Fact and folklore
for Spinners. B. and B. Hochberg. Santa Cruz. California.
Hodder I. 1986. Reading the past: current approaches
to interpretation in archaeology. Cambridge University
Press. Cambridge.
Horváth L. A. 2001. Interpretationsmöglichkeiten der ur-
zeitlichen Kultgruben (Archäologische und religionsge-
schichtliche Analyse Aufgrund einer kupferzeitlichen Kult-
stätte) – Az őskori kultuszgödrök értelmezésének lehet-
őségei (Egy rézkori kultuszhely régészeti és vallástörté-
neti elemzése). MFMÉ–Studia Archaeologica 7: 43–89.
Horváth T. 2009. The intercultural connections of the Ba-
den “culture”. In G. Ilon (ed.), MΩMOΣ VI. Őskoros kuta-
tók VI. Összejövetele. Szombathely: 101–149.
2010. Manifestation des Transcendenten in der Bade-
ner Siedlung von Balatonőszöd-Temetői dűlő – Kultge-
genstände. Praehistorische Zeitschrift 85: 79–119.
2013 A Balatonőszöd–Temetői dűlő (M7/S10) lelőhely
őskori településrészei. Digitális kiadás. Budapest.
Hurrell J. W. 1995. Decadal trends in the North Atlantic
oscillation: Regional temperatures and precipitation. Sci-
ence 269: 676–679.
Tools tell tales – climate trends changing threads in the prehistoric Pannonian Plain
323
Hurrell J. W., Kushnir Y., Ottersen G. and Visbeck M.
2003. An overview of the North Atlantic Oscillation. Geo-
physical Monograph Series 134: 1–35.
Isaakidou V. 2006. Ploughing with cows: Knossos and the
‘secondary products revolution’. In D. Serjeantson, D.
Field (eds.), Animals in the Neolithic of Britain and Eu-
rope. Oxbow Books. Oxford: 95–112.
Jacomet S. 2009. Plant economies and village life in Neo-
lithic lake dwellings at the time of the Alpine Iceman. Ve-
getation History and Archaeobotany 18: 47–59.
Johannsen N. 2005. Palaeopathology and Neolithic cat-
tle traction: methodological issues and archaeological per-
spectives. In Davies, J., Fabis, M., Mainland, I., Richards,
M. and Thomas, R. (eds.), Diet and Health in Past Ani-
mal Populations: Current Research and Future Direc-
tions. Oxbow Books. Oxford: 39–51.
Kalicz N. 2001. Die Protoboleráz-Phase an der Grenze
von zwei Epochen. In P. Roman, S. Diamandi (eds.), Cer-
navodă III. – Boleráz – Ein vorgeschichtliches Phäno-
men zwischen dem Oberrhein und der unteren Donau.
Symposium Mangalia/Neptun, 18–24. Oktober 1999. Stu-
dia Danubiana, ser. Symp. 2. Bucuresti: 385–435.
Kania K. 2015. Soft yarns, hard facts? Evaluating the re-
sults of a large scale hand-spinning experiment. Journal
of Archaeological and Anthropological Sciences 7: 113–
130.
Kimbrough C. K. 2006. Spindle Whorls, Ethnoarchaeo-
logy and the Study of Textile Production in Third Mil-
lennium BCE Northern Mesopotamia: A Methodological
Approach. Unpublished PhD thesis. York University –
Department of Anthropology. New York.
Kohler-Schneider M., Cannepele A. 2009. Late Neolithic
agriculture in eastern Austria–archaeobotanical results
from sites of the Baden and Jevi∏ovice cultures (3600–
2800 B.C.). Vegetation History and Archaeobotany 18:
61–74.
Koro∏ec P., Koro∏ec J. 1969. Najdbe s koli∏≠arskih nasel-
bin pri Igu na Ljubljanskem barju / Fundgut der Pfahl-
bausiedlungen bei Ig am Laibacher Moor. Arheolo∏ki
katalogi Slovenije 3. Narodni muzej. Ljubljana.
Kvavadze E., Bar-Yosef O., Belfer-Cohen A., Boaretto E., Ja-
keli N., Matskevich Z. and Meshveliani T. 2009. 30,000-
Year-Old Wild Flax Fibers. Science 325: 1359.
Laurito R., Lemorini C. and Perilli A. 2014. Making Textiles
at Arslantepe, Turkey, in the 4th and 3rd Millennia BC. Ar-
chaeological Data and Experimental Archaeology. In C.
Breniquet, C. Michel (eds.), Wool Economy in the An-
cient Near East and the Aegean. Ancient Textile Series 17.
Oxbow Books. Oxford: 151–168.
Leary J. 2014. Past Mobilities: Archaeological Approaches
to Movement and Mobility. Ashgate. Burlington.
Leghissa E. 2015. Na≠in okra∏evanja keramike ljubljanske
kulture in pramenaste keramike – eksperimentalna arheo-
logija. Arheolo∏ki vestnik 66: 275–292.
Leuzinger U., Rast-Eicher A. 2011. Flax processing in the
Neolithic and Bronze Age pile-dwelling settlements of east-
ern Switzerland. Vegetation History and Archaeobotany
20(6): 535–542.
Loughran-Delahunt I. 1996. Functional analysis of North-
west coast spindle whorls. Western Washington Univer-
sity. Bellingham.
Magny M. 2004. Holocene climatic variability as reflected
by mid-European lake-level fluctuations and its probable
impact on prehistoric human settlements. Quaternary In-
ternational 113: 65–80.
Magny M., Peyron O., Sadori L., Ortu E., Zanchetta G., Van-
niere B. and Tinner W. 2012. Contrasting Patterns of Pre-
cipitation Seasonality during the Holocene in the South
and North-central Mediterranean. Journal of Quaternary
Sciences 27: 494–502.
Magny M., Combourieu-Nebout N. 2013. Holocene chan-
ges in environment and climate in the central Mediterra-
nean as reflected by lake and marine records. Climate of
the Past 9: 1447–1454.
Maier U., Schlichtherle H. 2011. Flax cultivation and tex-
tile production in Neolithic wetland settlements on Lake
Constance and in Upper Swabia (south-west Germany).
Vegetation History and Archaeobotany 20(6): 567–578.
Maier U. 2001. Archäobotanische Untersuchungen in der
neolithischen Ufersiedlung Hornstaad-Hörnle IA am Bo-
densee. In U. Maier, R. Vogt (eds.), Siedlungsarchäologie
im Alpenvorland VI. Botanische und pedologische Un-
tersuchungen zur Ufersiedlung Hornstaad-Hörnle IA.
Forschungen und Berichte zur Vor-und Frühgeschichte in
Baden-Württemberg 74: 9–384.
Maran J. 1998. Die Badener Kultur und der ägäisch-anato-
lische Bereich. Germania 76(2): 497–525.
Mårtensson L., Andersson E., Nosch M. L. and Batzer A.
2006. Technical report – Experimental archaeology –
Part 1, 2005–2006. Centre for Textile Research. Copen-
hagen. On line: http://ctr.hum.ku.dk/tools/Technical_ re
port_1_experimental_archaeology.pdf
Masson V., Cheddadi R., Braconnot P., Joussaume S., Tex-
ier D. and PMIP participants 1999. Mid-Holocene climate
in Europe: what can we infer from PMIP model-data com-
parisons? Climate Dynamics 15: 163–182.
Ana Grabund/ija, Emmanuele Russo
324
Mauri A., Davis B., Collins P. and Kaplan J. 2014. The in-
fluence of atmospheric circulation on the mid-Holocene
climate of Europe: a data-model comparison. Climate of
the Past 10: 1925–1938.
Mayewski P. A. and 15 co-authors. 2004. Holocene Cli-
mate Variability. Quaternary Research 62: 243–255.
Mazăre P. 2014. Investigating Neolithic and Copper Age
Textile Production in Transylvania (Romania). Applied
Methods and Results. In M. Harlow, C. Michel and M. L.
Nosch, (eds.), Prehistoric, Ancient Near Eastern and
Aegean Textiles and Dress. An interdisciplinary antho-
logy. Oxbow Books. Oxford and Philadelphia: 1–42.
Médard F. 2006. Les activités de filage sur les sites néo-
lithiques du plateau suisse. Système technique de pro-
duction du fil dans son contexte économique et social.
Monographies du Centre de Recherches Archéologiques
28. Paris.
2010. L’art du tissage au néolithique: IVe–IIIe millé-
naires avant J.-C. en Suisse. Le Centre national de la
recherche scientifique. Paris.
2012. Switzerland: Neolithic Period. In M. Gleba, U.
Mannering (eds.), Textiles & textile production in Eu-
rope: From Prehistory to AD 400, Vol. 11. Oxbow
Books. Oxford: 367–377.
McCorriston J. 1997. The fiber revolution: textile exten-
sification, alienation, and social stratification in ancient
Mesopotamia. Current Anthropology 38: 517–49.
Pajagi≠-Bregar G., Velu∏≠ek A., Tolar T., Strli≠ M., Buko∏ek
V., Kolar J. and Ravbar I. 2009. Raziskave in konserviranje
preje z Ljubljanskega barja / Analysis and conservation of
the Ljubljansko barje yarn. In A. Velu∏≠ek (ed.), Koli∏≠ar-
ska naselbina Stare gmajne in njen ≠as. Ljubljansko ba-
rje v 2. polovici 4. tiso≠letja pr. Kr. / Stare gmajne Pile-
Dwelling Settlement and its era. The Ljubljansko barje
in the 2nd half of the 4th millennium BC. Opera Instituti
Archaeologici Sloveniae 16. ZRC SAZU Publishing. Ljub-
ljana: 309–318.
Petrova V. N. 2011. Textile Production in the Bronze
and Early Iron Age in the North Balkan Region. Unpub-
lished PhD thesis. Sofia University. Sofia.
Peyron O., Magny M., Goring S., Joannin S., de Beaulieu J.
L., Brugiapaglia E., Sadori L., Garfi G., Kouli K., Ioakim C.
and Combourieu-Nebout N. 2013. Contrasting patterns of
climatic changes during the Holocene across the Italian
Peninsula reconstructed from pollen data. Climate of the
Past 9: 1233–1252.
Pipes M.-L., Kruk J. and Milisauskas S. 2014. Assessing the
archaeological data for wool-bearing sheep during the
middle to late Neolithic at Bronocice, Poland. In H. J.
Greenfield (ed.), Animal Secondary Products: Domestic
Animal Exploitation in Prehistoric Europe, the Near
East and the Far East. Oxbow Books. Oxford: 80–102.
Raczky P. 1995. New data on the absolute chronology of
the Copper Age in the Carpathian Basin. In T. Kovács (ed.),
Neuere Daten zur Siedlungsgeschichte und Chronolo-
gie der Kupferzeit des Karpatenbeckens. Magyar Nemzeti
Múzeum. Budapest: 51–60.
Rast-Eicher A. 2005. Bast before Wool: the first textiles. In
P. Bichler, K. Grömer, R. Hofmann-de Keijzer, A. Kern and
H. Reschreiter (eds.), “Hallstatt Textiles”. Technical Ana-
lysis, Scientific Investigation and Experiment on Iron
Age Textiles. British Archaeological Reports IS 1351. Ar-
chaeopress. Oxford: 117–131.
Rast-Eicher A., Bender Jørgensen L. 2013. Sheep wool in
Bronze Age and Iron Age Europe. Journal of Archaeolo-
gical Science 40: 1224–1241.
Rast-Eicher A. 2014. Bronze and Iron Age Wools in Eu-
rope. In C. Breniquet, C. Michel (eds.), Wool Economy
in the Ancient Near East and the Aegean. Ancient Tex-
tile Series vol. 17. Oxbow Books. Oxford: 12–21.
Reed K. 2016. Agricultural change in Copper Age Croatia
(ca. 4500–2500 cal B.C)? Archaeological and Anthro-
pological Sciences. doi: 10.1007/s12520-016-0330-3.
Riede F. 2014. Climate Models: Use archaeology record.
Nature 513: 315.
Roeckner E., Arpe K., Bengtsson L., Christoph M., Claussen
M., Dumenil L., Esch M., Giorgetta M., Schlese U. and Schulz-
weida U. 1996. The atmospheric general circulation
model ECHAM4: model description and simulation of
present-day climate. Technical Report. Max Planck Insti-
tut fr Meteorologie. Hamburg.
Rooijakkers C. T. 2012. Spinning Animal Fibres at Late
Neolithic Tell Sabi Abyad, Syria? Paléorient 38: 93–103.
Russo E., Cubasch U. 2016. Mid-to-late Holocene Tempera-
ture Evolution and Atmospheric Dynamics over Europe
in Regional Model Simulations. Climate of the Past 12:
1645–1662.
Ryder M. L. 1964. The history of Sheep Breeds in Britain.
Agricultural History Review 12: 1–12; 65–82.
1974. Wools from Antiquity. Textile History 5: 100–110.
1982. Sheep – Hilzheimer 45 years on. Antiquity 56:
15–23.
1990a. Skin, and wool-textile remains from Hallstatt,
Austria. Oxford Journal of Archaeology 9(1): 37–49.
Tools tell tales – climate trends changing threads in the prehistoric Pannonian Plain
325
1990b. Wool remains from Scythian burials in Siberia.
Oxford Journal of Archaeology 9(3): 313–321.
1992. Iron Age, haired, animal skins from Hallstatt, Au-
stria. Oxford Journal of Archaeology 11(1): 55–67.
Schibler J., Jacomet S., Hüster-Plogmann H. and Bromba-
cher C. 1997. Economic crash in the 37th and 36th centu-
ries cal. BC in Neolithic lake shore sites in Switzerland.
Anthropozoologica 25: 553–570.
Schick T. 1988. Nahal Hemar Cave: Cordage, Basketry and
Fabrics. Atiqot 18: 31–43.
Schlichtherle H. 2009. Eine neue Siedlungskammer im
westlichen Federseeriedund, ihre Bedeutung für das Ver-
ständnis neolithischer Siedelsysteme. In J. Biel, J. Heilig-
mann and D. Krausse (eds.), Landesarchäologie. Fest-
schrift für Dieter Planck zum 65. Geburtstag. Forschun-
gen und Berichte zur Vor- und Frühgeschichte in Baden-
Württemberg 64. Theiss. Stuttgart: 61–86.
Schlichtherle H., Bleicher N., Dufraisse A., Kieselbach P.,
Maier U., Schmidt E., Stephan E. and Vogt R. 2010. Bad
Buchau – Torwiesen II: Baustrukturen und Siedlungsab-
fälle als Indizien der Sozialstruktur und Wirtschaftsweise
einer endneolithischen Siedlung am Federsee. In E. Cla-
ßen, T. Doppler and B. Ramminger (eds.), Familie – Ver-
wandtschaft – Sozialstrukturen: Sozialarchäologische
Forschungen zu neolithischen Befunden. Fokus Jung-
steinzeit. Berichte der AG – Neolithikum 1. Kerpen, Loogh:
157–178.
Sherratt A. G. 1981. Plough and pastoralism: aspects of
the secondary products revolution. In I. Hodder, G. Isaac
and N. Hammond (eds.), Pattern of the Past: Studies in
honour of David Clarke. Cambridge University Press.
Cambridge: 261–305.
1997. Economy and Society in Prehistoric Europe:
Changing Perspectives. Edinburgh University Press.
Edinburgh.
2006. La traction animale et la transformation de l’Eu-
rope néolithique. In P. Pétrequin, R.-M. Arbogast, A.-M.
Pétrequin, S. van Willigen and M. Bailly (eds.), Pre-
miers chariots, premiers araires. La diffusion de la
traction animale en Europe pendant les IVe et IIIe
millénaires avant notre ère. Monograph du Centre de
Recherches Archéologiques 29. Paris: 329–360.
Shishlina N. I., Orfinskaya O. V. and Golikov V. P. 2003.
Bronze Age Textiles from North Caucasus: new evidence
of fourth millennium BC fibers and fabrics. Oxford Jour-
nal of Archaeology 22(4): 331–344.
Solanki S. K., Usoskin I. G., Kromer B. and Schüssler Beer
J. 2004. Unusual Activity of the Sun during recent decades
compared to the previous 11000 years. Nature 431:
1084–1087.
Stadler P., Draxler S., Friesinger H., Kutschera W., Priller
A., Rom W., Steirer P. and Wild E. M. 2001. Absolute chro-
nology for early civilizations in Austria and Central Europe
using 14C dating with accelerator mass spectrometry with
special results for the absolute chronology of the Baden
culture. In P. Roman, S. Diamandi (eds.), Cernavodă III.
– Boleráz – Ein vorgeschichtliches Phänomen zwischen
dem Oberrhein und der unteren Donau. Symposium
Mangalia/Neptun, 18–24. Oktober 1999. Studia Danubia-
na, ser. Symp. 2. Bucuresti: 541–562.
Struhár V., Soják M. and Cheben I.2014. The Baden cul-
ture hilltop settlements in Northern Slovakia and their
“Socio-symbolic” importance. In A. Zastawny (ed.), The
Baden culture around the western Carpathians. Via ar-
chaeologica. Kraków: 237–272.
Styring A., Maier U., Stephan E., Schlichtherle H. and Bo-
gaard A. 2016. Cultivation of choice: new insights into far-
ming practices at Neolithic lakeshore sites. Antiquity 90:
95–110.
πavel I. 2009. Pod Kotom – jug pri Krogu I/II. Arheolo-
gija na avtocestah Slovenije. Zavod za varstvo kulturne
dedi∏≠ine Slovenije. Ljubljana.
Tolar T., Velu∏≠ek A. 2009. Discovery of flax (Linum usi-
tatissimum) at Ljubljansko barje, Slovenia. Histria Anti-
qua 18: 187–194.
Tolar T., Jacomet S., Velu∏≠ek A. and ∞ufar K. 2011. Plant
economy at a late Neolithic lake dwelling site in Slovenia
at the time of the Alpine Iceman. Vegetation History and
Archaeobotany 20: 207–222.
Tóth M., Magyari E. K., Buczkó K., Braun M., Panagioto-
poulos K. and Heiri O. 2015. Chironomid-inferred Holo-
cene temperature changes in the South Carpathians. The
Holocene 25(4): 569–582.
Wagner S., Widmann M., Jones J., Haberzettl T., Lücke A.,
Mayr C., Ohlendorf C., Schäbitz F., Zolitschka B. 2007.
Transient simulations, empirical reconstructions and forc-
ing mechanisms for the mid-holocene hydrological climate
in Southern Patagonia. Climate Dynamics 29: 333–355.
Wanner H., Solomina O., Grosjean M., Ritz S. P. and Jetel
M. 2011. Structure and origin of Holocene cold events.
Quaternary Science Reviews 30: 3109–3123.
Weninger B. and 18 co-authors 2009. The Impact of Ra-
pid Climate Change on prehistoric societies during the
Holocene in the Eastern Mediterranean. Documenta Prae-
historica 36: 551–583.
Ana Grabund/ija, Emmanuele Russo
326
Weninger B., Harper T. 2015. The Geographic Corridor
for Rapid Climate Change in Southeast Europe and Ukra-
ine. In E. Schultze (ed.), Neolithic and Copper age be-
tween the Carpathians and the Aegean. Archeologie in
Eurasien 31.  Deutsches Archäologisches Institut. Berlin:
475–505.
Wolff J., Maier-Reimer E. and Legutke S. 1997. The Ham-
burg Primitive Equation Model HOPE. Technical Report.
German Climate Computer Service. Hamburg.
Verhecken A. 2010. The moment of inertia: a parameter
for the functional classification of worldwide spindle-
whorls. In E. Andersson Strand, M. Gleba, U. Mannering,
C. Munkholt and M. Ringgaard (eds.), North European
symposium for archaeological textiles X. Oxbow Books.
Oxford – Oakville: 257–270.
Vakirtzi S., Koukouli-Chrysanthaki C. and Papadopoulos S.
2014. Spindle whorls from two prehistoric sites on Thas-
sos. In M. Harlow, C. Michel and M. L. Nosch (eds.), Pre-
historic, Ancient Near Eastern and Aegean Textiles and
Dress. An interdisciplinary anthology. Oxbow Books.
Oxford – Philadelphia: 43–57.
Vigne J. D., Helmer D. 2007. Was milk a “secondary pro-
duct” in the old world Neolithization process? Its role in
the domestication of cattle, sheep and goats. Anthropo-
zoologica. 42 (2): 9–40.
Vila E., Helmer D. 2014. The Expansion of Sheep Herding
and the Development of Wool Production in the Ancient
Near East: An Archaeozoological and Iconographical Ap-
proach. In C. Breniquet, C. Michel (eds.), Wool Economy
in the Ancient Near East and the Aegean, Ancient Tex-
tile Series vol. 17. Oxbow Books. Oxford: 22–40.
Von Storch H., Zorita E., Jones J., Dimitriev Y., Gonzalez-
Rouco F. and Tett S. 2004. Reconstructing Past Climate
from Noisy Data. Science 306: 679–682.
Vosteen M. 1996. Unter die Räder gekommen. Untersu-
chungen zu Sherratts ‘Secondary Products Revolution’.
Archäologische Berichte 7. Deutsche Gesellschaft für Ur-
und Frühgeschichte. Habelt. Bonn.
Zohary D. 1996. The mode of domestication of the foun-
der crops of Southwest Asian agriculture. In D. Harris
(ed.), The Origins and Spread of Agriculture and Pasto-
ralism in Eurasia. University College London Press. Lon-
don: 142–158.