Content uploaded by Sarah B Mcclure
Author content
All content in this area was uploaded by Sarah B Mcclure on Mar 30, 2015
Content may be subject to copyright.
AMS 14C CHRONOLOGY AND CERAMIC SEQUENCES OF EARLY FARMERS IN THE
EASTERN ADRIATIC
Sarah B McClure1,2 • Emil Podrug3 • Andrew M T Moore4 • Brendan J Culleton1 • Douglas J Kennett1
ABSTRACT. The eastern Adriatic is a key area for understanding the mechanisms and effects of the spread of agriculture.
This article presents an accelerator mass spectrometry (AMS) radiocarbon chronology for the introduction and subsequent
development of farming villages on the eastern shore of the Adriatic (~6000–4700 cal BC) and evaluates this in comparison
with the established pottery chronology based on stylistic data from Pokrovnik (Drniš) on the Dalmatian coast of Croatia.
Models for the spread of agriculture rely heavily on changing pottery styles to dene cultural groups and trace geographic
relationships. Based on AMS 14C dates presented here, Impressed Wares rst appear in central Dalmatia by 6000 cal BC and
persist until 5300 cal BC, well into what is generally termed the Middle Neolithic. Similarly, a typical Middle Neolithic ware,
gulina, appeared earlier than anticipated. These ndings stand in contrast to cave and rockshelter assemblages in the eastern
Adriatic, but mirror assemblages from farming villages on the Italian Adriatic coast. This study argues that the similarities in
ceramic assemblage composition and change through time may have less to do with direct contacts between areas, but more
with the nature of ceramic production and consumption at village sites in general. These data shed light on the limitations of
regional ceramic chronologies in the eastern Adriatic and highlight the necessity for systematic expansion of 14C chronolo-
gies to address the social, economic, and ecological relevance of early farming in the Adriatic for the spread of agriculture
in Europe and the Mediterranean.
INTRODUCTION
The spread of agriculture into Europe is an important case study for understanding the dispersion
of food production in general and the timing, tempo, and nature of its underlying processes. Early
farming in the Mediterranean region of Europe is characterized by the manufacture of pottery in
addition to an economic reliance on domesticated species of plants and animals. Chronologically
and regionally distinctive pottery styles provide the temporal framework for economic change, agri-
cultural intensication, population movements, interaction, and exchange. This article presents the
accelerator mass spectrometry (AMS) radiocarbon chronology for the introduction and subsequent
development of farming communities on the eastern shore of the Adriatic and the pottery chronol-
ogy of the period through ceramic stylistic data from Pokrovnik (Drniš) on the Dalmatian coast of
Croatia (Figure 1). Recent research on the northern Adriatic coast highlights 14C dates and broad-
scale trends in pottery production for early farming populations documented largely in cave and
rockshelter sites in that region (Forenbaher et al. 2013). In contrast, our work focuses on open-air
farming villages spanning most of the Neolithic in central Dalmatia and more ne-grained analyses
of chronological shifts in pottery styles.
Neolithic pottery in southern Europe is of particular interest because it is one of the few clearly dis-
tinguishing material features of early farming groups. Models for the spread of agriculture through-
out the Mediterranean and Europe rely heavily on pottery style to dene different cultural groups
and trace geographic relationships (see Özdoğan 2011; Rowley-Conwy 2011). This is particularly
true in the Balkans, where differences in ceramic style have been used to model distinct pathways
for the spread of food production, linking some areas more closely to Mediterranean groups and
others to central Europe. Discussions of agricultural intensication and associated demographic
shifts are often related to a widespread phenomenon of ceramic diversication and regionalization
(e.g. Price 2000; Rowley-Conwy 2011).
Early Neolithic pottery from the central Dalmatian coast of Croatia suggests a coastal focus with
close ties to other peoples in the Adriatic region, as opposed to interior groups on the Balkan Penin-
1. Department of Anthropology, The Pennsylvania State University, University Park, Pennsylvania 16802, USA.
2. Corresponding author. Email: sbm19@psu.edu.
3. Šibenik City Museum, Šibenik, Croatia.
4. Rochester Institute of Technology, Rochester, New York 14623, USA.
Radiocarbon, Vol 56, Nr 3, 2014, p 1019–1038 DOI: 10.2458/56.17918
© 2014 by the Arizona Board of Regents on behalf of the University of Arizona
1020 S B McClure et al.
sula. Impresso or Impressed Wares, typical of the Early Neolithic in the central and western Medi-
terranean, dominate Dalmatian assemblages. In contrast, the Middle Neolithic saw a diversication
of pottery styles that became regionally distinct, following trends throughout the Adriatic, while
evidencing contact with cultural groups in the Balkans. The Late Neolithic remains more enigmatic,
with stylistic shifts in pottery decoration but little information on daily economic activities.
This article presents an AMS 14C chronology for central Dalmatia for the Early Neolithic (Impressed
Ware) to the beginning of the Late Neolithic (Early Hvar), including new dates from a series of
open-air farming villages and previously published dates, and results of an extensive pottery anal-
ysis at Pokrovnik, a key site in the region. In order to discuss these results within a larger cultural
context, we rst briey outline Neolithic settlement and economy in central Dalmatia and then focus
on Neolithic pottery in the region, particularly issues of style, sourcing, exchange, and chronology.
We then present the AMS 14C dates and results of pottery analysis from Pokrovnik, and discuss these
within the broader cultural context of Neolithic developments in Dalmatia.
BACKGROUND
Dalmatia is a geographically dened region of the Republic of Croatia, bounded to the west and
south by the Adriatic Sea, and to the north and east by the Dinaric Alps that separate it from the
rest of the Balkan Peninsula. It consists of a typical karst landscape with rows of relatively low hills
(up to 500 m elevation) that divide small, narrow, elongated fertile valleys. Given the nature of the
karst limestone landscape, water resources are more difcult to identify and consist of some rivers,
Figure 1 Neolithic sites in central Dalmatia mentioned in the text
1021
AMS 14C Chronology of Early Farmers in Eastern Adriatic
underground or seasonal streams, springs, and ponds that provided farmers in the past enough water
for agropastoral food production (Moore et al. 2007a,b).
There are over 50 documented Neolithic sites in Dalmatia, both open-air settlements and cave sites
(temporary pastoralist shelters or possibly cult/ritual places), but only a few have been excavated
with modern archaeological methods (for recently excavated open-air settlements see e.g. Moore
et al. 2007a,b; Marijanović 2009; Podrug 2010; for cave sites see e.g. Čečuk and Radić 2005; Mar-
ijanović 2005; Forenbaher and Kaiser 2008). Neolithic peoples lived in aboveground (wattle-and-
daub constructed) houses, planted a range of domesticated crops, and managed several domestic an-
imal species (Moore et al. 2007a,b; Marijanović 2009; Legge and Moore 2011). Hunting decreased
to a minimum and is not well represented in many sites, although the faunal material at Crno Vrilo
showed a greater diversity of wild fauna than at other open-air Neolithic sites in the region (Radović
2009; Legge and Moore 2011).
Based on the available data, the basic economy and the assemblages of non-pottery material culture
(primarily stone and bone tools) did not change much during the Neolithic (Batović 1979). As a
result, pottery remains one of the most important sources of information for a more detailed picture
of cultural change. Furthermore, until recently the 14C record for this period has been limited and
pottery style has served as the primary chronological marker for the Neolithic in the region (see
also Forenbaher et al. 2013 for other parts of the eastern Adriatic). The chronology of Neolithic
Dalmatia is traditionally divided into three phases: (1) Early Neolithic, Impresso or Impressed Ware
(~6000–5500 cal BC); (2) Middle Neolithic, Danilo (~5500–4900 cal BC); and (3) Late Neolithic,
Hvar (~4900–4000 cal BC), based largely on the stylistic sequences of pottery from excavated con-
texts (Figure 2; Batović 1979; see also Forenbaher et al. 2004; Čečuk and Radić 2005; Marijanović
2005).
Figure 2 Typical Neolithic pottery from central Dalmatia: A. Impressed Ware. B–D. Danilo Ware: B.
Danilo smudged wares; C. Figulina; D. Rhyton; E. Hvar Ware. Photos courtesy of Šibenik City Museum.
1022 S B McClure et al.
The Impressed Ware phase is dened by the appearance of farming communities with a pottery as-
semblage consisting of pots and bowls decorated with impressed motifs using various tools such as
shell edges (e.g. Cardial), animal teeth, ngernails, bone tools, int tools, roulette, and combs, among
others. The transition to the Neolithic in the Adriatic is documented by diverse cultural shifts that
have been explained as colonization by farmers, adoption of farming by hunter-gatherers, or combi-
nations of the two models (Mlekuž 2003, 2005; Forenbaher and Miracle 2005; Miracle and Foren-
baher 2006; Marijanović 2007, 2009; Moore et al. 2007b; Legge and Moore 2011; Forenbaher et al.
2013). The similarity of Dalmatian Impressed Ware to the early pottery in other parts of the Adriatic
and the western Mediterranean highlights the centrality of ceramic analyses for addressing questions
of the timing and spread of agriculture regionally (Gheorghiu 2008). However, Impressed Ware
pottery varies in shape and decoration by region and its chronology spans 1000 yr (Spataro 2002;
McClure 2011). Several researchers have argued that Dalmatian Impressed Ware pottery also shows
local characteristics including impression motifs and techniques that are restricted in distribution,
and several competing stylistic sequences of Impressed Ware pottery have been proposed (Batović
1966, 1979; Benac 1957; Müller 1994; see also Spataro 2002). The relationship between Dalmatian
Impressed Ware and the broader Mediterranean Impressed Ware phenomenon remains unclear.
There is greater regionalization during the Middle Neolithic in the Mediterranean region, where
distinctive farming and technological characteristics indicate more dened economic and cultural
groups (e.g. Chapman 1988; McClure 2011). In the Dalmatian Middle Neolithic, Danilo pottery
shifts stylistically in form, with the introduction of new shapes such as cups and plates, and in deco-
ration. Danilo pottery consists largely of undecorated ne and coarse wares, but among the ne ware
is a distinctive subgroup of smudged and burnished wares with incised or carved ornamentation.
The typical motif is the spiral, usually encircling the entire vessel (Figure 2B). Other motifs include
meanders, triangles, and nets, and often these incisions are lled with red or white incrustations,
contrasting with the dark burnished vessels. Danilo assemblages furthermore include peculiar wares
known as gulina (also sometimes referred to as various types of polychrome, such as Danilo poly-
chrome or Southern Dalmatian Polychrome; see Korošec 1958:40–53; Forenbaher et al. 2013:600).
Figulina is a pink, white, or light orange buff ware made from ne, inclusion-free clays and painted
before ring (Figure 2C). It presents a very different manufacturing technology to the more common
smudged wares and has clear links to similar pottery styles in Italy. Current chemical evidence sup-
ports the historic interpretation that this ware was manufactured locally within Dalmatia (Korošec
1964:56–9, 66; Batović 1979:563–70; Spataro 2002; Teoh et al. 2014). Another distinctive type of
Middle Neolithic pottery found in Dalmatian sites is the rhyton. These footed “vessels” are unusual
in shape and often display zoomorphic features (Figure 2D), and have been interpreted as cult or
ritual paraphernalia (e.g. Perić 1996; Biagi 2003; Mlekuž 2007; Marijanović 2009; Rak 2011). No
consensus about their function yet exists and rhyta remain an easily identiable yet enigmatic fea-
ture of the Danilo period ceramic assemblages. Despite this ceramic diversity, there has been little
discussion of the nature, roles, and chronology of gulina and rhyta within the context of pottery
production and use at Middle Neolithic sites.
Finally, Hvar wares are similar to Danilo wares in that the ne vessels are also red in dark hues
and are highly polished or burnished (Figure 2E). Incisions continue as the most frequent decorative
technique, but motifs are more simplied and geometric than in Danilo, and tend to be “messier”
or less constrained. The main difference to Danilo wares, however, is the application of red paint to
ne wares after ring.
The corpus of 14C-dated archaeological deposits in the region has grown during the past 10 yr and
this study summarizes the available AMS 14C dates for Neolithic sites in central Dalmatia (Table 1;
1023
AMS 14C Chronology of Early Farmers in Eastern Adriatic
Figure 3; see also Miracle and Forenbaher 2006; Moore et al. 2007a,b; Forenbaher and Kaiser 2008;
Marijanović 2009; Podrug 2010; Forenbaher et al. 2013). Our research has focused on sites around
the Krka River valley and in particular on the sites of Danilo Bitinj and Pokrovnik (excavated during
the “Early Farming in Dalmatia Project”; NSF #0422195, 2004–2006; Moore et al. 2007a,b), Čista
Mala-Velištak (2007–present; Podrug 2010), and most recently the sites of Rašinovac and Krivače
(project “Neolithic Landscapes of Central Dalmatia”; NGS#9146-12; directed by McClure and Po-
drug). In addition, bone samples for AMS 14C dating were obtained from earlier excavations at
Krivače and Konjevrate, two key sites in the region. In the following, we outline sample selection,
methods, and results of AMS 14C dating, and then turn to the methods and results of stylistic analy-
sis of pottery from Pokrovnik and more detailed Bayesian chronology for these stratied deposits.
8
Figure3.RadiocarbondatesforsitesincentralDalmatia(calibratedwithOxCal
v4.2.3BronkRamsey(2013);r:5IntCal13atmosphericcurve(Reimeretal.2013)).
6400
6000
5500
5000
4800
LATE
Hvar
MIDDLE
Danilo
EARLY
Impresso
Seed samples
Faunal samples
Human samples
PK D-11a
PK A-33
PK D-23
PK D-21
RAS 1-3
PK D-22
PK C-7
DA A-31
KRI 3
DA E-5
DA E-14
DA A-17
PK C-23
KON 2
PK D-11b
PK A-8
DA C-15
DA B-24b
DA B-24a
KRI 2
DA A-42
DA A-36
KON 4
PK D-10
DA C-7
PK D-3
DA A-46
KRI 1-22
PK D-9
DA A-14
KRI 1-24
KRI 1-20
DA B-6
CMV F-73-1
CMV A-3a
CMV F-35
CMV A-3b
CMV A-23-1
CMV F-74-1
DA B-21
Figure 3 Radiocarbon dates for sites in cen-
tral Dalmatia, calibrated with OxCal v 4.2.3
Bronk Ramsey (2009); r:5 and IntCal13 atmo-
spheric curve (Reimer et al. 2013).
1024 S B McClure et al.
METHODS: SAMPLE SELECTION AND AMS 14C DATING
Two suites of AMS 14C dates are presented here. The rst group consists of dates generated at the
Oxford Radiocarbon Accelerator Unit (ORAU) as part of the “Early Farming in Dalmatia Project”
(Table 1). Samples consisted of 12 charred seeds and 1 bone (Ovis musimon) from Danilo Bitinj and
7 charred seeds from Pokrovnik (Moore et al. 2007a; Legge and Moore 2011). Samples were select-
ed from lower and upper levels of each unit to characterize the duration of occupation and sample
preparation followed conventions and procedures at ORAU (Brock et al. 2010).
A second suite of samples was selected from faunal remains from key sites in the region (Table 1)
for AMS 14C dating and stable isotope analysis (Zavodny et al. 2014). Bone samples from Konje-
vrate and the older excavations at Krivače were chosen from available holdings at the Šibenik City
Museum, while new excavations at Čista Mala-Velištak, Krivače, and Rašinovac provided material
with excellent stratigraphic control. Additional bone samples were chosen from Pokrovnik based
on changes in the ceramic assemblage identied during pottery analysis (see below). Bone collagen
for 14C and stable isotope analyses was extracted and puried at Penn State (Human Paleoecology
and Isotope Geochemistry Laboratory) using the modied Longin method (Brown et al. 1988) and
following protocols of the UC Irvine Keck Carbon Cycle AMS Facility (Beaumont et al. 2010).
Bone samples were initially cleaned of adhering sediment and the exposed surfaces were removed
by drilling or scraping with an X-ACTO® blade. Samples (200–400 mg) were demineralized for
24–36 hr in 0.5N HCl at 5°C followed by a brief (<1 hr) alkali bath in 0.1N NaOH at room tempera-
ture to remove humates. Collagen was rinsed to neutrality in multiple changes of Nanopure H2O,
and then gelatinized for 12 hr at 60°C in 0.01N HCl. Gelatin solution was pipetted into precleaned
Centriprep® 30 ultralters (retaining >30kD molecular weight collagen) and centrifuged three times
for 30 min, diluted with Nanopure H2O, and centrifuged three more times for 30 min to desalt the
solution. Ultraltered collagen was lyophilized and weighed to determine percent yield.
The recognition that foreign carbon could be introduced to samples during ultraltration by hu-
mectants (e.g. glycerol, glycerin) or lter material (e.g. reconstituted cellulose, polyethersulfone)
has spurred much recent methodological work to determine effective precleaning protocols (Bronk
Ramsey et al. 2004; Higham et al. 2006; Brock et al. 2007; Hüls et al. 2007). To remove the glycerin
coating from the Centriprep lters, the inner and outer portions of the lters were lled with 0.01N
HCl and sonicated at ~60°C for 1 hr and rinsed with Nanopure H2O. Nanopure H2O was centrifuged
through the lters 3 times for 30 min each, and the inner and outer portions were relled with Na-
nopure H2O and sonicated for 1 hr at ~60°C. After three further centrifuge runs with Nanopure H2O,
the lters were kept wet until use, no more than 48 hr after precleaning. Results on Pleistocene and
historic age bone standards processed along with the unknowns are used to detect contamination
from either modern or ancient carbon.
14C samples (~2.5 mg) were combusted for 3 hr at 900°C in vacuum-sealed quartz tubes with CuO
wire and Ag wire. At KCCAMS, sample CO2 was reduced to graphite at 550°C using H2 and a Fe
catalyst, with reaction water drawn off with Mg(ClO4)2 (Santos et al. 2004). Graphite samples were
pressed into targets in Al boats and loaded on the target wheel for AMS analysis. 14C ages were cor-
rected for mass dependent fractionation with measured δ13C values (Stuiver and Polach 1977), and
compared with samples of Pleistocene whale bone (background, >48k 14C BP), middle Holocene
pinniped bone, late AD 1800s cow bone, and OX-1 oxalic acid standards for calibration. Carbon
and nitrogen concentrations and stable isotope ratios were measured at the Penn State University
Light Isotope Laboratory with a Costech EA (ECS 4010), Thermo Finnigan Cono IV gas handling
device, and a Thermo Finnigan Delta V analyzer.
1025
AMS 14C Chronology of Early Farmers in Eastern Adriatic
Table 1 AMS 14C dates from open-air village sites in central Dalmatia; calibrated with OxCal
v 4.2.3 Bronk Ramsey (2009); r:5 and IntCal13 atmospheric curve (Reimer et al. 2013).
Site;
Sample #
Trench/
Level Material Lab # 14C BP 2σ cal BC Reference
Rašinovac
RAS-1
1/SJ3 Bos taurus (cow)
>30kDa gelatin
PSU-5612/UCI-
AMS-127394
7060 ± 25 6005–5895
Pokrovnik A/8 Triticum monococcum
(einkorn) charred grain
OxA-17195 6626 ± 39 5625–5490
Pokrovnik A/33 Triticum dicoccum
(emmer) charred grain
OxA-17328 6810 ± 40 5755–5630
Pokrovnik C/7 Triticum dicoccum
(emmer) charred grain
OxA-17124 6197 ± 39 5295–5240
(7.9%)
5235–5040
(87.5%)
Pokrovnik C/23 Triticum dicoccum
(emmer) charred grain
OxA-17125 6568 ± 36 5615–5585
(9.1%)
5570–5475
(86.3%)
Pokrovnik D/3 Triticum dicoccum
(emmer) charred grain
OxA-17223 6170 ± 35 5220–5015 Legge and
Moore
2011
Pokrovnik
PK-44
D/9 Ovis aries (sheep)
>30kDa gelatin
PSU-4960/UCI-
AMS-106477
6280 ± 20 5310–5215
Pokrovnik
PK-39
D/10 Bos taurus (cow)
>30kDa gelatin
PSU-5294/UCI-
AMS-116206
6190 ± 25 5220–5055
Pokrovnik D/11 Triticum dicoccum
(emmer) charred grain
OxA-17193 6625 ± 36 5625–5490
Pokrovnik
PK-45
D/11 Ovis aries (sheep)
>30kDa gelatin
PSU-4961/UCI-
AMS-106478
6840 ± 25 5765–5660
Pokrovnik D/21 Triticum monococcum
(einkorn) charred grain
OxA-17194 6999 ± 37 5985–5785 Legge and
Moore
2011
Pokrovnik
PK-7
D/22 Bos taurus (cow)
>30kDa gelatin
PSU-5293/UCI-
AMS-116205
7090 ± 25 6025–5965
(56.3%)
5960–5905
(39.1%)
Pokrovnik
PK-15
D/23 Ovis aries (sheep) PSU-5556/UCI-
AMS-119837
6975 ± 30 5980–5945
(8.3%)
5920–5760
(87.1%)
Konjevrate
KON-2
Ovis aries (sheep)
>30kDa gelatin
PSU-5291/UCI-
AMS-116203
6655 ± 25 5630–5535
Konjevrate
KON-4
Ovis aries (sheep)
>30kDa gelatin
PSU-5557/UCI-
AMS-119838
6175 ± 30 5220–5035
Danilo-
Bitinj
A/14 Triticum monococcum
(einkorn) charred grain
OxA-17196 6212 ± 35 5300–5190
(34.4%)
5185–5055
(61%)
1026 S B McClure et al.
Table 1 AMS 14C dates from open-air village sites in central Dalmatia; calibrated with OxCal
v 4.2.3 Bronk Ramsey (2009); r:5 and IntCal13 atmospheric curve (Reimer et al. 2013).
Site;
Sample #
Trench/
Level Material Lab # 14C BP 2σ cal BC Reference
Danilo-
Bitinj
A/17 Ovis musimon (sheep) OxA-14449 6284 ± 40 5365–5205
(94.2%)
5160–5150
(0.3%)
5145–5135
(0.3%)
5095–5080
(0.7%)
Moore et
al. 2007a
Danilo-
Bitinj
A/31 Triticum dicoccum
(emmer) charred grain
OxA-15764 6226 ± 37 5305–5195
(50%)
5180–5060
(45.4%)
Moore et
al. 2007a
Danilo-
Bitinj
A/36 Triticum monococcum
(einkorn) charred grain
OxA-17197 6121 ± 37 5210–4955
Danilo-
Bitinj DA-6
A/42 Ovis aries (sheep)
>30kDa gelatin
PSU-5290/UCI-
AMS-116202
6155 ± 25 5215–5025
Danilo
Bitinj
A/46 Triticum dicoccum
(emmer) charred grain
OxA-15681 6180 ± 34 5225–5020 Moore et
al. 2007a
Danilo-
Bitinj
B/6 Rosa sp. (wild rose)
charred seed
OxA-17329 6204 ± 38 5295–5050
Danilo-
Bitinj
B/21 Triticum monococcum
(einkorn) charred grain
OxA-15680 5987 ± 35 4985–4785 Legge and
Moore
2011
Danilo-
Bitinj
B/24 Rosa sp. (wild rose)
charred seed
OxA-17198 6093 ± 36 5210–5145
(10.2%)
5140–5095
(2.7%)
5085–4905
(82%)
4865–4855
(0.5%)
Moore et
al. 2007a
Danilo-
Bitinj
B/24 Rosa sp. (wild rose)
charred seed
OxA-17199 6103 ± 37 5210–5090
(21.9%)
5085–4935
(73.5%)
Danilo-
Bitinj
C/7 Rosa sp. (wild rose)
charred seed
OxA-17200 6161 ± 36 5215–5005
Danilo-
Bitinj
C/15 Rosa sp. (wild rose)
charred seed
OxA-17224 6083 ± 35 5210–5165
(5.6%)
5080–4895
(88.1%)
4870–4850
(1.8%)
Danilo-
Bitinj
E/5 Triticum monococcum
(einkorn) charred grain
OxA-17126 6237 ± 37 5310–5200
(63.2%)
5180–5065
(32.2%)
1027
AMS 14C Chronology of Early Farmers in Eastern Adriatic
Table 1 AMS 14C dates from open-air village sites in central Dalmatia; calibrated with OxCal
v 4.2.3 Bronk Ramsey (2009); r:5 and IntCal13 atmospheric curve (Reimer et al. 2013).
Site;
Sample #
Trench/
Level Material Lab # 14C BP 2σ cal BC Reference
Danilo-
Bitinj
E/14 Triticum monococcum
(einkorn) charred grain
OxA-15765 6245 ± 39 5315–5200
(70.1%)
5175–5070
(25.3%)
Moore et
al. 2007a
Krivače
KRI-2
III/A2 Sus scrofa (pig)
>30kDa gelatin
PSU-5558/UCI-
AMS-119839
6115 ± 30 5210–5145
(19.5%)
5140–5090
(6.2%)
5085–4945
(69.7%)
Krivače
KRI-3
III/A1 Bos taurus (cow)
>30kDa gelatin
PSU-5292/UCI-
AMS-116204
6300 ± 25 5320–5220
Krivače
KRI-6
1/SJ22 Homo sapiens
(human) >30kDa
gelatin
PSU-5613/UCI-
AMS-127395
6270 ± 20 5305–5215
Krivače
KRI-7
1/SJ24 Homo sapiens
(human) >30kDa
gelatin
PSU-5614/UCI-
AMS-127396
6285 ± 20 5310–5220
Krivače
KRI-8
1/SJ20 Homo sapiens
(human) >30kDa
gelatin
PSU-5615/UCI-
AMS-127397
6290 ± 20 5315–5220
Čista Mala-
Velištak
CMV-5
F/3 Ovis aries (sheep)
>30kDa gelatin
PSU-5289/UCI-
AMS-116201
5935 ± 20 4875–4870
(0.2%)
4850–4725
(95.2%)
Čista Mala-
Velištak
CMV-2
F/74-1 Bos taurus (cow)
>30kDa gelatin
PSU-5288/UCI-
AMS-116200
6045 ± 25 5020–4845
Čista Mala-
Velištak
CMV-38
A/23-1 Ovicaprid
>30kDa gelatin
PSU-3701/UCI-
AMS-78155
5975 ± 15 4935–4920
(2.7%)
4915–4795
(92.7%)
Podrug
2010
Čista Mala-
Velištak
CMV-28
A/3 Ovicaprid
>30kDa gelatin
PSU-3702/UCI-
AMS-78156
5920 ± 15 4840–4725 Podrug
2010
Čista Mala-
Velištak
CMV-12
A/SJ3 Homo sapiens
(human) >30kDa
gelatin
PSU-5616/UCI-
AMS-127398
5945 ± 20 4900–4860
(8.7%)
4855–4765
(84.7%)
4760–4740
(2%)
Čista Mala-
Velištak
CMV-3
F/73-1 Bos taurus (cow) PSU-5563/5564/
UCIAMS-125829/
125830
5903 ± 11 4800–4720
1028 S B McClure et al.
RESULTS: AMS 14C DATES
Early Neolithic Sites
Three sites—Pokrovnik, Konjevrate, and Rašinovac—attest to the development of the Impressed
Ware Neolithic and 14C dates from Pokrovnik and Rašinovac document the beginning of food pro-
duction in central Dalmatia at ~6000 cal BC. Pokrovnik is an open-air village that was occupied
continuously during the Early (Impressed Ware) and Middle (Danilo) Neolithic and that is well
documented (Table 1, Figure 3). Surface and occasional nds also indicate it was likely inhabited
during the early Hvar period. The site has been excavated twice, rst by Brusić in 1979 and more re-
cently by Moore and Menđušić in 2006 (Brusić 2008; Legge and Moore 2011; Moore et al. 2007b).
The 2006 excavation consisted of four large trenches (A–D) ranging in size from 5 × 5 to 8 × 5 m.
These trenches were excavated to the subsoil in a strip eld that transected the site, uncovering a
cross-section of the inhabited area of the village. Trenches A and D had multiple layers of habita-
tion debris, hearths, pits, and house remains. Large stone walls were documented in Trenches A,
C, and D and likely served as terraces and boundaries. Faunal remains, polished and chipped stone
tools, and large quantities of pottery were unearthed (Moore et al. 2007b; Legge and Moore 2011).
The data suggest that farming arrived in coastal Dalmatia as a full package at 6000 cal BC—with
domesticated plants and animals, as well as pottery—and early farmers did not change their basic
economic activities for 800 yr (Moore et al. 2007a,b; Legge and Moore 2011).
Analysis of Pokrovnik continues, but a suite of 12 AMS 14C dates has been generated for different
parts of the site (Table 1). In particular, we concentrated on Trench D for the pottery analysis (see
below) and obtained a number of 14C dates specically for this unit. Our goal was to constrain the
stylistic shifts we documented in the pottery analysis in concert with ongoing research on animal
management practices and stable isotope analyses (Zavodny et al. 2014). The resulting chrono-
logical framework consists of 8 AMS 14C dates (Table 1). More detail on the pottery analysis is
presented below.
Konjevrate is located under a modern churchyard and was test excavated in the late 1980s and
1990s. Although excavation was not extensive, material remains in the form of pottery, stone tools,
and animal bones were recovered. Stylistically, the pottery is Impressed Ware, dating to the Early
Neolithic. Two bone samples were 14C dated from this site and one, KON-2, fell into the expected
Early Neolithic chronology (Table 1, Figure 3). Surprisingly, however, a second sample (KON-4)
is Middle Neolithic in date. Although only general provenience data are available for these sam-
ples and not specic information on the stratigraphy, no Middle Neolithic pottery or other material
culture was unearthed. The signicance of the AMS 14C date is that there was some type of Middle
Neolithic occupation at or around the Impressed Ware site that led to a Middle Neolithic bone being
deposited there. Alternatively, it may reect a continuity of an Impressed Ware pottery tradition into
the Middle Neolithic. This issue will be discussed in more detail below. This location is currently a
cemetery, so it is unlikely that additional excavations can take place to clarify the settlement history
of this site.
Rašinovac is a newly discovered Early Neolithic site in the Piramatovci Valley (Figure 1) that was
test excavated (2 × 2 m unit) by Podrug and McClure in May 2013. A substantial Impressed Ware
cultural horizon (40 cm thick) was documented during these excavations. Analysis is ongoing, but
the assemblage is largely comprised of Impressed Ware pottery and Early Neolithic stone tools.
Future work will clarify the extent and intensity of occupation at this Neolithic settlement, but the
AMS 14C date presented here is the rst date available for this site (Table 1).
1029
AMS 14C Chronology of Early Farmers in Eastern Adriatic
The earliest Impressed Ware dates center at 6000 cal BC at both Pokrovnik and Rašinovac. This
is contemporary with published dates from other sites in the eastern Adriatic including Nakovana
Cave (OxA-18120: 7050 ± 37, 6008–5846 cal BC; Forenbaher et al. 2013: Table 1) and the open-
air site SU-002 (ETH-22912: 6925 ± 65 BP, 5877–5736 cal BC) on Sušac Island (Radić 2009:17;
Forenbaher et al. 2013:598).
Middle Neolithic Sites
Danilo Bitinj is located in the fertile Danilo Valley, ~18 km east of Šibenik, and was occupied in the
Middle Neolithic (Korošec 1964; Moore et al. 2007a; Legge and Moore 2011). It is the type-site for
the Middle Neolithic and gives the chronological phase its name. Early excavations in the 1950s,
rescue excavations in 1992, and a series of ve 5 × 5 m trenches (A–E) excavated in 2004/2005
uncovered a total 2700 m2 of the site. Based on surface distributions, the site is estimated to span
8–9 ha. Excavations unearthed several areas of habitation with remains of pits, house oors, walls,
and large quantities of faunal and oral remains, pottery, and stone tools. A suite of 14 AMS 14C
dates has been generated for four trenches, spanning 5300–4900 cal BC (Table 1, Figure 3).
Krivače is located in the Bribir-Ostrovica Valley and was surface collected in 1963 (Korošec and
Korošec 1974) and excavated in the early 2000s and in 2013. Pottery from Impressed Ware, Danilo,
and Hvar periods were recovered in surface collections, suggesting this site was an open-air village
occupied throughout the Neolithic. However, limited test excavations in the 2000s only unearthed a
Middle Neolithic occupation. Recent excavation in May 2013 by Podrug and McClure focused on a
2 × 2 m test trench. Over 60 cm of cultural deposits contained large quantities of pottery, including
gulina and rhyton fragments, stone tools made of chert and obsidian, and faunal remains. Further-
more, ditches, hearths, pits, and house oors were uncovered. Material is still under analysis, but
ve AMS 14C dates are available (Table 1, Figure 3). Two of these dates, KRI-2 and KRI-3, are on
animal bones from the 2000s excavations. Three additional AMS 14C dates are on human remains
found embedded in the clay oors at the base of the unit, just above the sterile subsoil. KRI-3 is
statistically identical to the human remain 14C dates and suggests that occupation of this part of the
village began around 5300 cal BC. Further dating of other parts of the cultural horizon will help us
identify the duration of settlement at Krivače.
Based on these data, typical Middle Neolithic villages are documented in central Dalmatia by
5300 cal BC and remained occupied for up to 400 yr. People living at multicomponent sites like
Pokrovnik began to create the typical suite of Middle Neolithic pottery around the same time.
Late Neolithic Sites
Čista Mala-Velištak is a Hvar period open-air village currently under excavation by Podrug. It
was discovered in 2007 and in seven excavation seasons an area of 200 m2 has been examined
(Podrug 2010). The site has a vertical stratigraphy with several distinct contexts, including a series
of pits excavated into the subsoil and remains of aboveground house oors and hearths. A large
quantity of Hvar pottery and other materials has been collected. Čista Mala-Velištak is particularly
signicant because the Hvar Neolithic is largely known from cave sites, and this is currently the
only Hvar open-air village to be excavated in Dalmatia. In the case of the valley surrounding Čista
Mala-Velištak, museum collections contain Danilo and Impressed Ware pottery, indicating the pres-
ence of sites from these periods in the valley, though precise locations are unknown. Six AMS 14C
dates are currently available spanning ~5000–4700 cal BC (Table 1, Figure 3).
1030 S B McClure et al.
METHODS: POTTERY ANALYSIS
The large ceramic assemblage from Pokrovnik is ideal to address questions of the timing, tempo,
and nature of stylistic shifts during the Early and Middle Neolithic. This study is based on material
from the excavation unit, Trench D, with the deepest (up to 2 m) intact deposit and includes data
from over 26,000 sherds. Here, we present data on pottery style; future work will include technolog-
ical studies, petrography, sourcing analyses, and residue analyses.
Pottery was cleaned in the eld and is curated by the Drniš City Museum. Analysis was conducted
at the Šibenik City Museum and consisted of several phases. First, undiagnostic sherds (i.e. wall
fragments) were sorted by level and decorative type and counted. Diagnostic sherds were sorted by
level into rims, bases, and handles, and recorded individually with decorative information. All rim
sherds were drawn. Fragments with data on vessel shape were further studied and recorded.
RESULTS: POTTERY ANALYSIS
In total, we analyzed 23,327 undiagnostic and 2919 diagnostic sherds from Pokrovnik Trench D
(Tables 2, 3). The stratigraphy of this excavation unit spanned 2 m in its deepest area. Figure 4A
illustrates the western prole wall (C–D) with the succession of strata, while Figure 4B shows the
Harris matrix for the unit as a whole. As can be seen in Figure 5, pottery becomes less decorated
through time. Early Neolithic Impressed Ware vessels are often completely decorated, with recur-
ring impressions on the entire external surface of the pot. A steady decline in decoration is visible
through time. This shift is due to an increase in more zonal decoration on the pottery, resulting in
a greater percentage of the vessel surfaces being simply smoothed or burnished. Although zonal
decoration is well known from the Danilo wares with their bands of meandering spirals and oth-
er motifs (Figure 2), these data indicate that this trend begins in the Early Neolithic, particularly
represented in Levels 14 and 11. A shift towards zonal decoration in Impressed Wares has been
documented elsewhere, and in some cases, distinct phases have been suggested to capture the shift
to zonal decorations (Impressed A vs. Impressed B; Batović 1979; Müller 1994; Čečuk and Radić
2005). However, the signicance of this shift as a clear chronological marker has been questioned
(Forenbaher et al. 2013:598) and our data suggest it was more of a gradual change. This trend in
the Pokrovnik pottery specically and the Adriatic more generally echoes Impressed Ware pottery
styles found in the western Mediterranean, where zonal motifs are common (Bernabeu et al. 2012;
McClure and Bernabeu 2012).
Table 2 Summary of pottery analysis for Pokrovnik Trench D. In columns with multiple numbers,
the rst refers to number of diagnostic sherds and the second to undiagnostic sherds (diag/undiag).
Level Total
Diag.
total Undec. Decorated Impressed Danilo Figulina Rhyton
1, 2, 3 6115 726 644/5259 82/130 0/6 64/142 9/64 9/0
5, 6, 7, 8 2087 293 246/1767 47/27 0/6 31/37 11/69 5/0
9 3397 404 381/2951 23/42 2/22 14/27 6/38 1/0
10 4430 419 406/3985 13/26 6/21 2/10 5/24 0/0
11 2670 242 231/2321 11/107 11/107 0/0 0/0 0/0
14 2236 241 231/1794 10/201 9/202 0/0 1/0 0/0
15, 16, 18,
20, 21
3308 368 309/2016 59/924 59/924 0/0 0/14 0/0
19 719 61 47/393 14/265 13/266 0/0 1/0 0/0
22, 23 1284 165 114/516 51/603 51/603 0/0 0/0 0/0
Total 26,246 2919 2609/21,002 310/2325 151/2157 111/216 33/209 15/0
1031
AMS 14C Chronology of Early Farmers in Eastern Adriatic
Table 3 Summary of Impressed Ware decorative techniques in number of diagnostic
pottery sherds.
Level Cardial Roulette Tremolo Other Impressed
9 1 0 0 1
10 3 3 1 0
11 3 2 3 6
14 0 2 5 9
15, 16, 18, 20, 21 18 1 9 56
19 2 2 0 11
22, 23 15 2 0 44
Total 42 12 18 127
A summary of the distribution of decorative types (Figure 6A, Table 2) indicates clear shifts in
wares and decoration types during the Early and Middle Neolithic. For the Early Neolithic, Fig-
ure 6B shows changes in techniques and demonstrates that a variety of styles were in use during the
earliest phases of pottery production at Pokrovnik (see also Table 3). Stylistic diversity increased in
later phases of the Early Neolithic assemblage with the addition of tremolo decorations. This partic-
Figure 4 Pokrovnik Trench D: A) Western prole (C–D). Stars indicate levels with 14C dates (see Table 1). B) Harris matrix.
B)
A)
1032 S B McClure et al.
ular decorative technique uses a sharp object to make very small, regular impressions in a zig-zag
motif. The data presented here support prior claims (e.g. Müller 1994; Spataro 2002; Brusić 2008)
that tremolo was a later addition to an Early Neolithic decorative repertoire.
Levels 9 and 10 show a more interesting pattern with a mix of impressed, typical Danilo incised, and
gulina wares (Figure 6A). We were unable to identify transitional pottery in techniques or motifs
between Impressed Ware and Danilo, and these levels in particular had the highest concentrations
of gulina pottery of all levels analyzed. We interpret these levels as a transitional phase between
Impressed Ware and Danilo since there are no data to suggest that they were mixed with underlying
Impressed Ware levels (see also Brusić 2008 for similar observations; Figures 3 and 4).
Finally, Figure 6A highlights questions regarding the timing and signicance of gulina and rhyta.
Although the relative proportion of gulina is very small throughout the Neolithic, it rst appears at
Pokrovnik in the Early Neolithic. The numbers presented in Figure 6A are very small, but an addi-
tional 14 undiagnostic sherds were recovered from Early Neolithic levels that are not captured in the
bar chart (see Table 2). It is striking that only gulina appears in the earlier levels and none of the
more ubiquitous Danilo pottery is present. We discuss issues of taphonomy in greater detail below.
Based on published research, gulina wares have typically been analyzed in Dalmatia without any
emphasis on chronology. Our results suggest that the timing and contexts of gulina in Dalmatian
Neolithic sites may be more complex and should be examined more closely.
Bayesian Analysis of the Pokrovnik Sequence
The basic distributions of stylistic groups through time at Pokrovnik in conjunction with the AMS
14C chronology reported above provide an opportunity to delve more deeply into the chronologi-
cal patterning of stylistic change in the region. We established a stratigraphic model for Trench D
using OxCal (Bronk Ramsey 2009) that combines 8 AMS 14C dates with stratigraphic information
to understand better the pace and timing of changes seen archaeologically. Figure 7A presents the
available AMS 14C dates for Pokrovnik in a phased stratigraphic sequence from the earliest (bottom)
Figure 5 Relative proportion of decorated and undecorated pottery through time (n = 26,246). Box indicates Early
Neolithic levels.
1033
AMS 14C Chronology of Early Farmers in Eastern Adriatic
to latest (top) Neolithic horizons. Phases were constructed based on the Harris matrix (Figure 4B) for
the site, and events and strata that were not directly dated were modeled as boundaries (e.g. Str. 14,
the Early to Middle Neolithic transition, end of occupation). In Figure 7B, the light gray probability
distributions represent calibrated date ranges (2σ) without model constraints. The dark gray distri-
bution shows these probability estimates constrained by the stratigraphic model. Agreement indices
(A) provide a measure of t between the data and model with values higher than the critical threshold
(A′ = 60%) indicating good concordance. This measure is provided for each date distribution (e.g.
R_Date D11 (b) Ovis, [A:95]) and for the model as a whole (Sequence [Amodel]). Overall, there is
good agreement between the available data and the stratigraphic model (Amodel 75).
Boundaries dening the ages of ceramic-bearing strata are reproduced in Figure 7B and shown
relative to the abundance of diagnostic pottery styles through time. Pottery abundances are shown
vertically for each level and add up to 100%. The age of each level on the timescale uses the mean
of each distribution for a point estimate. These data indicate that the use of specic ceramic styles
as chronological markers needs to be re-evaluated. Specically, the co-occurrence of gulina wares
Figure 6 A) Comparison of relative (%) distributions of diagnostic decorated pottery: Impressed Ware, typical Danilo
ware, rhyta and gulina (does not include undecorated diagnostics; n = 314; Table 2). Box indicates Early Neolithic
levels. B) Comparison of relative (%) distributions of Early Neolithic pottery decorative types (n = 204; Table 3).
The category “Other Impressed” includes a variety of impressed decorative types (e.g. non-denticulated, ngernail).
B)
A)
1034 S B McClure et al.
with Impressed Ware pottery as well as the persistence of Impressed Wares into otherwise typical
Middle Neolithic contexts indicates longer chronologies for both wares and the need for a more nu-
anced approach to pottery styles in early farming communities. This is echoed by recent research on
sequences largely from caves and rockshelters elsewhere in the eastern Adriatic (Forenbaher et al.
2013), where the utility of traditional divisions of the Neolithic into Early, Middle, and Late phases
has been questioned.
Figure 7 A) Bayesian chronology of ceramic sequences at Pokrovnik, Trench D. B) Radiocarbon phases
paired with the relative distribution of pottery styles highlighting the overlap of ceramic typologies.
B)
A)
1035
AMS 14C Chronology of Early Farmers in Eastern Adriatic
In central Dalmatia, Impressed Wares appear by 6000 cal BC (e.g. Pokrovnik, Rašinovac) and
continue to be produced at Pokrovnik until ~5300 cal BC, when they were replaced by more typical
Danilo wares. The persistence of Impressed Ware is 3 centuries longer than documented in other
parts of the eastern Adriatic. In the northern part of the coast, Danilo-Vlaška wares, a variant of the
Dalmatian Danilo wares, appear by 5600 cal BC and overlap with the end of the Impressed Ware
production for only around 100 yr, although they do not occur in the same sites and there is no mix-
ing within any of the stratied cave sites (Forenbaher et al. 2013). In Dalmatia, the co-occurrence of
these styles does not appear to be taphonomically structured.
Instead, we argue that the assemblage is distinct from cave sites due to differences in site use, pottery
production areas, and vessel function. This observation is consistent with pottery from the Italian
side of the Adriatic, where Impressed Wares co-occur with stylistically “later” pottery at a number of
open-air villages until around 5250 cal BC (Malone 2003:243; Robb 2007:163–72). The production
of Impressed Ware into what has traditionally been termed the Middle Neolithic in Dalmatia may
help explain the later date from Konjevrate (KON-4; Table 1) mentioned earlier. This date (KON-4:
6175 ± 30 BP; 5220–5035 cal BC) is statistically identical to the date for Pokrovnik Trench D,
level 10 (PK-39: 6190 ± 25 BP; 5220–5055 cal BC) that had 45% Impressed Wares among the
diagnostic pottery and 69% of all decorated sherds along with gulina and typical Danilo wares.
The similarities to the Early Neolithic in the Italian Adriatic do not end there. At Pokrovnik, gulina
appears archaeologically in levels dated to 5700–5500 cal BC (Table 1, Figure 7), but becomes more
common after ~5200 cal BC. Only a small number of gulina sherds were recovered from the earlier
levels 14 through 19, begging the question if their presence is due to taphonomic issues.
The levels in Trench D were well stratied and seemingly suffered very little from disturbance or
mixing. For example, the terrace wall (level 19) was well grounded and largely in situ. Higher up the
sequence, level 7 consisted of an intact clay oor with impacted limestone fragments that covered the
entire trench. This clay oor was also found in adjacent Trench A. The relatively undisturbed nature
of the stratication in Trench D is conrmed by the locations and frequencies of the artifacts. The
main categories other than pottery, including chipped stone, ground and other stone tools, and bone
artifacts, were all much more abundant in the Danilo levels than lower down. Very few of them seem
to have been displaced. Notably, there was no obsidian in the lower, Impressed Ware levels. It should
further be noted that the AMS dates from the trench are in stratigraphic order (Table 1, Figure 3).
Most striking, however, is that gulina is the only “typical” Middle Neolithic ware to be recovered
from these lower levels, despite its minor role in the Middle Neolithic ceramic repertoire. If its pres-
ence in the earlier levels were due to mixing or postdepositional effects, one would expect to nd at
least some Danilo wares in the same contexts. This was not the case and supports our interpretations
that these few fragments do indeed suggest an earlier presence of gulina at the site.
Figulina is present in all levels analyzed at Danilo Bitinj, indicating a clear presence of this type in
central Dalmatia by 5300 cal BC. This stands in contrast to elsewhere on the eastern Adriatic coast,
where gulina has a relatively late appearance, ~5200–4800 cal BC (Forenbaher et al. 2013:600–1).
Although the identication of gulina in the earlier levels at Pokrovnik was surprising given the
trends observed elsewhere in the region, gulina commonly overlaps with Impressed Wares in east-
ern Italy, beginning in the early 6th millennium BC (Robb 2007:170).
The Danilo-Vlaška wares on the northern coast mentioned earlier predate Danilo wares in central
Dalmatia by several centuries. At Pokrovnik, Danilo wares appear by 5300 BC, and are present at
Krivače and Danilo Bitinj, both with associated earliest dates of ~5300 cal BC (Table 1). The tem-
1036 S B McClure et al.
poral framework of Danilo pottery supports Forenbaher et al.’s (2013:604) inference that Dalmatian
Danilo wares may have originated in Istria and the Trieste karst, but developed independently into
regionally specic Hvar wares in the Late Neolithic. Although none of the sites analyzed in this
study span the entire Neolithic, the pottery assemblage from Čista Mala-Velištak is clearly a classic
“outline style” pottery type with dates spanning 5000–4700 cal BC. Later phases of the Late Neo-
lithic known from cave sites on the Dalmatian islands (Forenbaher and Kaiser 2008; Forenbaher et
al. 2013) have yet to be identied on the central Dalmatian mainland.
It appears, then, that Early Neolithic ceramic sequences in central Dalmatia are more comparable to
other Neolithic open-air villages in Italy than to caves and rockshelters on the northern end of the
eastern Adriatic. There is some evidence of contact among peoples in the Adriatic during this peri-
od, including seafaring and the establishment of settlements on islands (Forenbaher 2009). Howev-
er, we suggest that the similarities in ceramic assemblage composition and change through time on
either side of the Adriatic may have little to do with direct contacts between areas, but rather more
generally with the nature of ceramic production and consumption at village sites. Regardless of the
myriad taphonomic issues of the archaeological record, we can safely assume that the majority of
pottery found in caves and rockshelters was transported there for a reason, whereas we would expect
a greater diversity of ceramics to have been produced, used, and discarded in villages. Given these
issues, the disparities between data sets produced for the eastern Adriatic is not surprising. Why
should we expect the pottery found in caves/rockshelters and contemporary villages to have been
the same, when we know that the activities and duration of occupation at these locations were dif-
ferent? The key, instead, is to focus more energy on building independent chronological frameworks
for early farming sites on the one hand, and understanding the mechanisms of pottery production,
consumption, and discard on the other. As demonstrated here, the link between pottery style and
chronology is more regional, tenuous, and nuanced than previously appreciated.
CONCLUSION
The eastern Adriatic is a key area for understanding the mechanisms and effects of the spread of
agriculture into Europe. New AMS 14C dates from Neolithic village sites in central Dalmatia are
creating a temporal framework to assess these developments. Focused AMS 14C chronologies are
a relatively new line of research in the region; previous investigations used long-standing ceramic
typologies to date archaeological sites and features. Our data shed light on the limitations of tradi-
tional ceramic chronologies for addressing ne-grained questions in subregions and archaeological
contexts in the eastern Adriatic. Specically, we are redening the “suites” of ceramic wares that
co-occur and their relative placement in time and space. With these more ne-grained approaches
to chronology and pottery, we can begin to address issues of inter-regional contacts and similarities,
the roles of caves, rockshelters, and villages to farming populations, and the social and economic
relevance of shifts in ceramic technologies through time.
ACKNOWLEDGMENTS
Our gratitude is due to Thomas Higham, Marko Menđušić, and Joško Zaninović for their efforts and
expertise. Special thanks to Thomas Harper and Emily Zavodny for their help with maps and g-
ures, and to two anonymous reviewers for their insightful comments and revision suggestions. This
research was funded by the National Science Foundation (NSF #0422195), National Geographic
Society (NGS#9146-12), Rochester Institute of Technology, The Pennsylvania State University,
and the Šibenik City Museum.
1037
AMS 14C Chronology of Early Farmers in Eastern Adriatic
REFERENCES
Batović Š. 1966. Stariji neolit u Dalmaciji. Beograd: Ar-
heološko društvo Jugoslavije.
Batović Š. 1979. Jadranska zona. In: Benac E, editor.
Praistorija jugoslavenskih zemalja II. Neolit. Saraje-
vo: Akademija nauka i umjetnosti Bosne i Hercegov-
ine, Centar za balkanološka ispitivanja. p 473–634.
Beaumont W, Beverly R, Southon JR, Taylor RE. 2010.
Bone preparation at the KCCAMS Laboratory. Nu-
clear Instruments and Methods in Physics Research
B 268(7–8):906–9.
Benac A. 1957. Zelena pećina. Glasnik Zemaljskog
muzeja 12:61–92.
Bernabeu J, Rojo MA, Molina Ll, editors. 2012. Las Pri-
meras Producciones Cerámicas: El VI milenio AC
en la Península Ibérica. Saguntum-Extra 12, Uni-
versitat de València.
Biagi P. 2003. The rhyton of the Balkan Peninsula:
chronology, origin, dispersion and function of a
Neolithic “cult” vessel. Journal of Prehistoric Reli-
gion XVI–XVII:16–26.
Brock F, Bronk Ramsey C, Higham T. 2007. Quality
assurance of ultraltered bone dating. Radiocarbon
49(2):187–92.
Brock F, Higham TFG, Ditcheld P, Bronk Ramsey C.
2010. Current pretreatment methods for AMS radio-
carbon dating at the Oxford Radiocarbon Accelera-
tor Unit (ORAU). Radiocarbon 52(1):103–12.
Bronk Ramsey C. 2009. Bayesian analysis of radiocar-
bon dates. Radiocarbon 51(1):337–60.
Bronk Ramsey C, Higham T, Bowles A, Hedges R.
2004. Improvements to the pretreatment of bone at
Oxford. Radiocarbon 46(1):155–63.
Brown TA, Nelson DE, Vogel JS, Southon JR. 1988.
Improved collagen extraction by modied Longin
method. Radiocarbon 30(2):171–7.
Brusić Z. 2008. Pokrovnik - naselje iz neolitika. Šibenik:
Muzej grada Šibenika.
Čečuk B, Radić D. 2005. Vela spila. Višeslojno pretpo-
vijesno nalazište - Vela Luka, otok Korčula. Vela
Luka: Centar za kulturu “Vela Luka.”
Chapman J. 1988. Ceramic production and social differ-
entiation: the Dalmatian Neolithic and the Western
Mediterranean. Journal of Mediterranean Archaeol-
ogy 1/2:3–25.
Forenbaher S. 2009. Archaeological record of the Adri-
atic offshore islands as an indicator of long-distance
interaction in prehistory. European Journal of Ar-
chaeology 11(2–3):223–44.
Forenbaher S, Kaiser T, editors. 2008. Grapčeva Špilja.
Pretpovijesni stan, tor i obredno mjesto. Split: Kn-
jiževni krug.
Forenbaher S, Miracle PT. 2005. The spread of farming
in the Eastern Adriatic. Antiquity 79(305):514–28.
Forenbaher S, Kaiser T, Miracle PT. 2004. Pupićina
Cave pottery and the Neolithic sequence in North-
eastern Adriatic. Attli della Società per la preeisto-
ria e protostoria della regione Friuli-Venezia Giulia
14:61–102.
Forenbaher S, Kaiser T, Miracle PT. 2013. Dating the
East Adriatic Neolithic. European Journal of Ar-
chaeology 16(4):589–609.
Gheorghiu D. 2008. The emergence of pottery. In: Jones
A, editor. Prehistoric Europe: Theory and Practice.
Oxford: Blackwell Publishing. p 164–94.
Higham T, Bronk C, Karavanić, Smith F, Trinkaus E.
2006. Revised direct radiocarbon dating of the
Vindija G1 Upper Paleolithic Neandertals. Proceed-
ings of the National Academy of Sciences of the USA
103(3):553–7.
Hüls M, Grootes P, Nadeau M-J. 2007. How clean is ul-
traltration cleaning of bone collagen? Radiocarbon
49(2):193–200.
Korošec J. 1958. Neolitska naseobina u Danilu Bitinju. Za-
greb: Jugoslavenska akademija znanosti I umjetnosti.
Korošec J. 1964. Danilo in Danilska kultura. Ljubljana:
Univerza v Ljubljana.
Korošec J, Korošec P. 1974. Bribir I njegova okolica u
prapovijesno doba. Diadora 7:5–33.
Legge AJ, Moore AMT. 2011. Clutching at straw: the
Early Neolithic of Croatia and the dispersal of ag-
riculture. In: Hadjikoumis A, Robinson E, Viner S,
editors. The Dynamics of Neolithisation in Europe:
Studies in Honour of Andrew Sherratt. Oxford: Ox-
bow Books. p 176–95.
Malone C. 2003. The Italian Neolithic: a synthesis of re-
search. Journal of World Prehistory 17(3):235–312.
Marijanović B. 2005. Gudnja. Višeslojno prapovijesno
nalazište. Dubrovnik: Dubrovački muzeji - Arhe-
ološki Muzej.
Marijanović B. 2007. Neka pitanja ranog neolitika is-
točnog Jadrana. Archaeologia Adriatica I:7–54.
Marijanović B. 2009. Crno Vrilo 1. Zadar: Sveučilište
u Zadru.
McClure SB. 2011. Learning Technology: Neolithic Pot-
tery Production in Valencia, Spain. British Archae-
ological Reports, International Series 2300. Oxford:
Archaeopress.
McClure SB, Bernabeu J. 2012. Technological style,
chaîne opératoire, and labor investment of early
Neolithic pottery. In: Bernabeu J, Rojo MA, Molina
Ll, editors. Las Primeras Producciones Cerámicas:
El VI milenio AC en la Península Ibérica. Universi-
tat de València: Saguntum Extra 12. p 53–60.
Miracle PT, Forenbaher S, editors. 2006. Prehistor-
ic Herders of Northern Istria. The Archaeology of
Pupićina Cave. Volume 1. Pula: Arheološki Muzej
Istre.
Mlekuž D. 2003. Early herders of the Eastern Adriatic.
Documenta Praehistorica XXX:139–51.
Mlekuž D. 2005. The ethnography of the Cyclops: Neo-
lithic pastoralists in the eastern Adriatic. Documenta
Praehistorica XXXII:15–51.
Mlekuž D. 2007. ‘Sheep are your mother’: rhyta and the
interspecies politics in the Neolithic of the eastern
Adriatic. Documenta Praehistorica XXXIV:330–42.
Moore AMT, Menđušić M, Smith J, Podrug E. 2007a.
1038 S B McClure et al.
Project “Early Farming in Dalmatia”: Danilo Bitinj
2004–2005. Vjesnik Arheološkog muzeja u Zagrebu
40:15–24.
Moore AMT, Menđušić M, Smith J, Zaninović J, Pod-
rug E. 2007b. Project “Early Farming in Dalmatia:
Pokrovnik 2006.” Vjesnik Arheološkog muzeja u Za-
grebu 40:25–34.
Müller J. 1994. Das Ostadriatische Frühneolithikum.
Prähistorische Archäologie in Südosteuropa, Band
9. Frankfurt: Phillip von Zabern.
Özdoğan M. 2011. Archaeological evidence on the west-
ward expansion of farming communities from East-
ern Anatolia to the Aegean and the Balkans. Current
Anthropology 52(Supplement 4):S415–S430.
Perić S. 1996. Kult-Rhytone der neolitischen Viehzüch-
ter der Balkanhalbinsel. Starinar 47:21–66.
Podrug E. 2010. Čista Mala-Velištak: prve tri is-
traživačke kampanje na nalazištu hvarske kulture
[Čista Mala-Velištak: the rst three excavation cam-
paigns at a Hvar culture site]. Diadora 24:7–25.
Price TD, editor. 2000. Europe’s First Farmers. Cam-
bridge: Cambridge University Press.
Radić D. 2009. The beginnings of trans-Adriatic naviga-
tion: a view from Vela Spila Cave (Korčula Island).
In Forenbaher S, editor. A Connecting Sea: Mari-
time Interaction in Adriatic Prehistory. Oxford: Ar-
chaeopress. p 13–24.
Radović S. 2009. Analiza ostataka faune sisavaca [Anal-
ysis of faunal remains of mammals]. In: Marijanović
B, editor. Crno Vrilo 2. Zadar: Sveučilište u Zadru.
p 53–66.
Rak O. 2011. The Rhyton from Danilo: Structure and
Symbolism of a Middle Neolithic Cult-Vessel. Ox-
ford: Oxbow Books.
Reimer PJ, Baillie MGL, Bard E, Bayliss A, Beck JW,
Blackwell PJ, Bronk Ramsey C, Buck CE, Cheng
H, Edwards RL, Friedrich M, Grootes PM, Guilder-
son TP, Haidason H, Hajdas I, Hatté C, Heaton TJ,
Hoffman DL, Hogg AG, Hughen KA, Kaiser KF,
Kromer B, Manning SW, Niu M, Reimer RW, Rich-
ards DA, Scott EM, Southon JR, Staff RA, Turney
CSM, van der Plicht J. 2013. IntCal13 and Marine13
radiocarbon age calibration curves, 0–50,000 years
cal BP. Radiocarbon 55(4):1869–87.
Robb J. 2007. The Early Mediterranean Village. Agency,
Material Culture, and Social Change in Neolithic It-
aly. Cambridge: Cambridge University Press.
Rowley-Conwy P. 2011. Westward Ho! The spread of
agriculture from central Europe to the Atlantic. Cur-
rent Anthropology 52(Supplement 4):S431–S451.
Santos GM, Southon JR, Druffel-Rodriguez KC, Grif-
n S, Mazon M. 2004. Magnesium perchlorate as
an alternative water trap in AMS graphite sample
preparation: a report on sample preparation at the
KCCAMS Facility at the University of California,
Irvine. Radiocarbon 46(1):165–73.
Spataro M. 2002. The First Farming Communities of the
Adriatic: Pottery Production and Circulation in the
Early and Middle Neolithic. Società per la preistoria
e protostoria della Regione Friuli-Venezia Giulia,
Quaderno 9. Trieste: Edizioni Svevo.
Stuiver M, Polach HA. 1977. Discussion: reporting of
14C data. Radiocarbon 19(3):355–63.
Teoh M, McClure SB, Podrug E. 2014. Macroscopic,
petrographic and XRD analysis of Middle Neolith-
ic gulina pottery from central Dalmatia. Journal
of Archaeological Science (in press). http://dx.doi.
org/10.1016/j.jas.2014.07.007.
Zavodny E, McClure SB, Culleton BJ, Podrug E, Ken-
nett DJ. 2014. Neolithic animal management prac-
tices and stable isotope studies in the Adriatic. Envi-
ronmental Archaeology (in press). http://dx.doi.org/
10.1179/1749631414Y.0000000021.
