ChapterPDF Available

Archaeometric characterization of Middle Bronze Age pottery from the settlement at Borg in-Nadur

4. Archaeometric characterization of Middle Bronze Age pottery
from the settlement at Borġ in-Nadur
1. Introduction
A major weakness in the research about Maltese prehistoric
pottery has been represented by the paucity of archaeometric
data for the characterisation of fabrics. Besides preliminary
research on technological issues of Maltese prehistoric
pottery production attempted by Martha Molitor (1992),
the analysis carried out on a group of Late Neolithic, Early
and Middle Bronze Age specimens sampled from the
University of Malta excavations at Tas-Silġ (Mommsen et
alii 2006), remains the only study that has appeared in this
field. Without petrographic and chemical reference points the
classification of pottery classes and the distinction of fabrics
have been invariably based on the macroscopic observation
of sherd surface treatment and the composition of core and
body as seen through fractures. This traditional approach has
significant limitations (Maniatis 2009).
With regard to the study of pottery from the site of Borġ in-
Nadur which represents the principal ceramic evidence for the
Middle Bronze Age of Malta, all attempts at classification, and
subsequent chronological reconstructions, have been based on
a visual examination.
In the first typological study of Borġ in-Nadur pottery, David
Trump distinguished two ware types, fine and coarse (Trump
1961). On the basis of surface fabric and decoration of the
fine ware alone, he distinguished between three pottery classes
which he labelled II B1, II B2, and II B3. These he considered
as evidence for three separate and successive chronological
phases. In particular, he identified a bright red to scarlet
burnished slip as peculiar of II B1 and a dark red unburnished
slip with black blotches as characteristic of II B2. In his
typology, the II B3 group comprised three different surface
treatments: dark unburnished slip with brownish or deep red
blotches; dark slip used for drawing simple patterns over the
clay body, also labelled as ‘Dribbled ware’ or the total absence
of slip (Trump 1961, 259). No significant distinguishing
features were defined for the coarse ware and its evolution
across the three phases of Trump’s periodization scheme. It was
only later that John D. Evans referred to ‘unslipped fabrics’ for
the Borġ in-Nadur coarse ware (Evans 1971, 226).
The revision of Trump’s theories about the Borġ in-Nadur
culture was at the basis of recent research by Claudia
Sagona (2008; 2011). In particular, on the account of the
evidence from the University of Malta’s excavations at the
southern enclosure of Tas-Silġ, she redefined the known
principal pottery classes, discussed and provided additional
data concerning fabrics identified through visual analysis.
The traditional II B1 class is as a first phase of the culture,
labelled as Early Borġ in-Nadur, and described as a Red
Slipped Ware:
‘[…] red-slipped and burnished inside and out, sometimes the slip
can be blackened presumably from conditions in the kiln, from use
wear or from burning. It is invariably hand-made, probably using
the coil technique given the slightly undulating surfaces of many
examples. In section, the vessels reveal dark grey to black medium
coarse clays with mixed gritty inclusions that often include angular
white, quartz-like particles.’ (Sagona 2011, 400)
Associated with the II B1 ware, Sagona presented a less
common ‘White Gritty Ware’, interpreted as the domestic
cooking ware: ‘Hitherto, this fabric has not been isolated as
a ware type in its own right, but it is a useful grouping for
categorising drab (slipped or unslipped) wares that are found
along with burnished red slipped wares.’ (Sagona 2011, 410).
In the subsequent phase, labelled by Sagona Middle Borġ
in-Nadur, the emergence of a ‘Thin Walled Variant’ of the
previous red slipped class is identified, characterised by hard-
fired fabrics and often haphazardly incised with lines and rows
of zig-zag designs (Sagona 2011, 415).
In the third and last phase of Sagonas classification, Late
Borġ in-Nadur, she distinguished the main ware as a ‘Chalky
Reddish Yellow Ware’:
‘[…] soft and its eroding surfaces shed reddish yellow (pale orange)
chalky dust when handled. Fragments reveal dark day at the core
in brownish grey hues (in the range of 10YR 7/2-4) and surfaces
fire to reddish colour (around 7.5YR 6/6-8) suggesting poor firing
techniques. Vessels are invariably hand-made from day that carries
a moderate to large amounts of mixed gritty particles. Sometimes
slipped in reddish day that is mottled or blackened from conditions
in the kiln or perhaps from use ware. Unlike Borġ in-Nadur Red
Slipped Wares, the surfaces are usually matt and decaying rather
than burnished”. Besides this class, considered by the author an
imitation of Levantine prototypes, she even presented the Hand-
Made Pink Buff Ware, just attested on large thick-walled domestic
pottery.’ (Sagona 2011, 417)
About Trump’s II B3 pottery, Sagona re-interpreted the
stratigraphic sequence of the settlement at Borġ in-Nadur and
argued that the pottery attributed by Trump to the II B3 phase
ought to be related instead to the Tarxien Cemetery period of
the Early Bronze Age. In her view, the Borġ in-Nadur II B3
pottery never existed (Sagona 2008, 503).
More recently Trump’s classification has been questioned
on the basis of the new evidence coming from the Italian
excavations carried out in the northern enclosure at Tas-Silġ
(Copat et alii 2012). Based on the stratigraphic sequence
G B, P M, S R,
D T, A G
G B, P M, S R, D T, A G
(Tanasi 2013) (Table 2). Finally, Fabric 5 was recognised as
corresponding to Sagona’s White Gritty Ware whilst Fabric 3
was found to be equivalent to the ‘unslipped coarse ware with
smoothed surfaces of a beige hue’ identified by the Italian
archaeologists at Tas-Silġ (Copat et alii 2012).
In the foregoing attempts at pottery classification, it
transpires that a distinctive trait of the fine ware is the
presence of a slip while the lack of a slip is a recurrent feature
of the coarse ware. In these studies, the characteristics of the
coarse wares have never been defined and an identification of
a class of cooking vessels is entirely missing.
2. Archaeometric study: methodological approach
It is against this state of play that the decision was taken to
carry out an archaeometric study of samples of Middle Bronze
Age pottery from Malta with the intent of establishing a
system of reference that distinguishes properly between ware
types, Once permission was granted by the Superintendence
of Cultural Heritage (Malta) and Heritage Malta, in
December 2012, 28 samples were taken from a diagnostic
group of vessels representing all the ware types identifiable
in the ceramic assemblage from the settlement at Borġ in-
Nadur (Trump’s excavations) held at the National Museum
of Archaeology, Valletta (Figs. 1-2). Characterisation of the
samples was carried out by petrographic analysis and X-Ray
Fluorescence (XRF) in order to provide the first technological
study and fabric characterisation of the ceramic production
from Borġ in-Nadur. In addition, two samples of clay (BN
33, BN 34) belonging to the Blue Clay Formation were
collected from Ġnejna Bay in north-west Malta for further
comparative analyses. The sampled pottery came the most
important contexts, including Hut 1 (Trench G, layer 2)
and Hut 2 (Trench H, layer 2). These have been selected in
order to cover all the possible fabrics and surface treatments,
classified according to the aforementioned scheme (Tanasi
2011, 2013) (Table 3).
Unfortunately, the occurrence of Fabric 4 turned out to be
very scanty and it was not possible to sample any specimen
while sample BN7 was taken from the fragmentary cup
and relying on shape, surface treatment, and decoration as
distinguishing criteria, a new chronological arrangement
of the pottery production of the Borġ in-Nadur phase has
been presented. Three new chronological phases, each
characterised by different classes of pottery, are presented:
Early, Classic, and Later Borġ in-Nadur. In the first phase
(Early Borġ in-Nadur) burnished orange surfaces of Tarxien
Cemetery tradition are attested, together with black and
orange or reddish surfaces; a low percentage of samples shows
orange, grey, and beige surfaces and an even lower number
have a red slip. In the second phase (Classic Borġ in-Nadur),
corresponding to Trump’s II B1 and II B2, the most common
treatment is the red slip with or without black blotches; the
surfaces are sometimes burnished or just smoothed. In the last
phase (Late Borġ in-Nadur), there is the concurrent presence
of two pottery classes: a black/brown or red-black slip
ware and an unslipped coarse ware with smoothed surfaces
of a beige hue (unslipped). A third class, with partial slip
decoration or painted dots and bands is doubtfully situated
between Late Borġ in-Nadur and Baħrija phases.
In the overall re-assessment of the material evidence from
the excavations carried out in 1920s at the temple of Borġ
in-Nadur, all pottery from the Middle Bronze Age layers
was examined in an attempt of creating a chrono-typological
sequence (Tanasi 2011). That exercise was supplemented and
revised through the study of additional assemblages of Maltese
Bronze Age pottery (Tanasi 2013). After a meticulous study
(Tanasi 2011) it was concluded that no clear correspondence
was found with Sagona’s classification; instead convergence
with Trump’s descriptions was apparent. The visual study of
the ceramics allowed three fine fabrics to be distinguished:
Fabric 1 (Reddish yellow fabric with thick red slip), Fabric
2 (Pink fabric with red mottled slip), and Fabric 4 (Reddish
yellow fabric with dark red to black mottled slip). These can
be attributed respectively to Trump’s groups II B1, II B2,
II B3 (Tanasi 2011, 89-91; 2013) (Table 1). Moreover, an
additional two fabrics were distinguished not comparable
with anything described by Trump: a semi-fine Fabric 3 and
a coarse Fabric 5. The class traditionally known as ‘Dribbled
Ware’ was also identified and later observed in other contexts
associated with ceramics of all of Trump’s three phases
T 1. B -N      T (2011, 2013).
A   M B A 
T 2. B -N -       
  T (2011, 2013).
T 3. L      , N M  A O
I , ,   , ,     .
G B, P M, S R, D T, A G
37267 with a fabric and surface composition not referable
to the same classification. In fact, the sample was closer to
the ‘burnished orange’ related to an Early Borġ in-Nadur,
as a transitional phase between Early and Middle Bronze
Age (Copat et alii, 2012). Furthermore, two specimens of
‘Painted Ware’, all in all comparable to the ‘Dribbled Ware’
(Tanasi, this volume, Chapter 3), were sampled due to the
absence of significant dribbled examples. The samples cover
the most common shapes of the entire repertoire besides one
clay anchor and one of a rough clay render.
The results obtained were also used to ascertain the
provenance of the Borġ in-Nadur type pottery recovered from
Middle Bronze Age archaeological contexts in Syracuse and
Agrigento in Sicily (Tanasi 2008; 2011). In particular, data
have been compared with results of petrographic and chemical
analyses carried out on a group of sherds from Ognina
(OG12/128, OG12/131, OG12/140, OG12/151) and
Cannatello (CAN/71) considered as Borġ in-Nadur ceramic
imports (Tanasi 2014, fig. 3). Middle Bronze Age Thapsos
pottery samples from Ognina (Raneri et alii, in press) and
raw materials hailing from eastern Sicily (Barone et alii, 2012)
have been used in order to verify eventual distinctions between
Sicilian and Maltese productions on the basis of petrographic
and chemical analyses.
With regards to Borġ in-Nadur samples (hereafter, BN#),
petrographic analyses on thin sections were performed with
the aim to characterise texture and mineralogical composition
(Table 4). Two fabrics were recognised following the scheme
proposed by Whitbread (1995), which facilitates a detailed
F 1. M B A      B -N 
        ID   ID.
A   M B A 
F. 2. M B A     
     ID   ID.
characterisation of pottery in terms of texture, groundmass,
and inclusions (Table 4):
Fabric A (specimens BN1, BN3, BN5, BN6, BN7, BN11,
BN13, BN16, BN18, BN19, BN26, BN28), characterised by
coarse chamotte and fine quartz inclusions, and groundmass
with abundant fossil and high-medium birefringence;
Fabric B (specimens BN2, BN4, BN9, BN15, BN20, BN22,
BN23, BN27, BN30, BN31), with abundant chamotte, spatic
calcite, fine quartz inclusions, and fossiliferous groundmass
with high birefringence.
Furthermore, Fabric A samples are characterized by a sub-
millimetric red slip (Fig. 4b), absent in those of Fabric B. In
G B, P M, S R, D T, A G
particular, specimens belonging to Fabric A are related to a
pottery class characterised by a particular repertoire of shapes
and a distinctive red lustrous slip, while specimens of Fabric B
are mainly represented by cooking vessels.
The microstructure of Fabric A shows channels and planar
voids with remains of carbonaceous material suggesting
the use of straw as temper. Less abundant are vughy and
slightly preferential oriented vescicles (Fig. 4a). The fossil-
rich groundmass is heterogeneous and it is characterised
by medium to high optical activity and a brownish-greyish
colour. The inclusions have an open space distribution and
bimodal grain size: the coarser inclusions are represented by
chamotte (Fig. 4b) with prevalently sub-angular shape and
millimetric dimensions while the finer ones are mainly quartz
grains. Finally, brownish-black amorphous phases are present,
measuring about 0.5 mm in diameter.
The voids microstructure of Fabric B is similar to that
observed in Fabric A; noteworthy is the presence of
carbonaceous materials inside planar voids (Fig. 4c). The
groundmass rich in fossils is very heterogeneous (Fig. 4d)
with high optical activity and yellowish-greyish color. The
grain size distribution of inclusions is polymodal with
chamotte and spatic calcite (Fig. 4e) in the coarse fractions
and quartz in the finer. Volcanic inclusions, whose type
cannot be explained at this step of the research, are present
only in sample BN27 (Fig. 4f ). The chamotte is sub-angular
in shape with millimetric dimensions and it has the same
paste as the samples in which it occurs.
Finally, mainly reddish
amorphous phases are also present in samples of this fabric.
The study of the thin sections produced relevant information
on the technology of pottery production. Over all, the pottery
studied is petrographically quite similar for its groundmass
characteristics. The most peculiar discriminating factor is the
presence of spatic calcite inclusions in the Fabric B samples. As
has been clearly ascertained, this type of inclusion represents
a technical expedient introduced to increase the resistance to
fire of artefacts devoted to cooking food, making them suitable
to hold out against thermal shock (Fabbri et alii 1997).
With regard to optical activity, data may be used for
estimating the firing temperature of the pottery. In particular,
medium to high birefringence is in many cases indicative
of the attainment of medium to low firing temperatures
respectively without reaction and modifications of the
original mineralogical association of the raw materials. In the
case of the Borġ in-Nadur samples, it is possible to suggest
a medium firing temperature for the samples belonging
to Fabric A (about 800°C) and lower T
°C for samples
belonging to Fabric B. This assumption is supported by the
presence of microfossils that only in the Fabric A samples
show evidence of calcite decarbonatation.
In order to confirm these suggestions, X-Ray Diffraction
analysis (XRD) has been carried out on a sample of Fabric
A (BN7) and a sample of Fabric B (BN9). Although other
samples are being subjected to the same type of analysis, the
results obtained reveal that in the two samples new formed
Ca-silicates are absent, confirming that the pottery was fired
at a temperature below 850°C.
The mineralogical and petrographic data, therefore, suggest
that the technological level is quite low. As far as the choice of
the raw materials is concerned, the clay used to produce all the
pottery samples studied is probably the same. On the contrary,
the use of spatic calcite in pottery of Fabric B could suggest
that the artisans were knowledgeable of the properties of this
temper and used it for a specific purpose.
For the specimens from Ognina and Cannatello sampled for
the same analyses, it is worth noting that these present the
same visual and petrographic features as those of Fabric A,
strongly suggesting for them a Maltese provenance.
F. 3. M B A   O (OG12/131, OG12/128, OG12/140)  C
(CAN/17)    . (OG: O; CAN: C).
A   M B A 
F 4. M   .
F A: () BN1; () BN5. F B: () BN 28; () BN4; () BN2;  () BN27.
The chemical composition of the Maltese samples belonging
to the two fabrics determined petrographically and referential
Maltese clay (labelled as BN33 and BN34) are listed in Table 5.
With regard to major elements, the data obtained show a linear
trend in the SiO2/CaO vs Al2O3/CaO diagram (Fig. 5a) and a
parabolic trend in the diagrams representing SiO2/CaO vs Al2O3/
Fe2O3 and Al2O3/CaO vs SiO2/MgO (Fig. 5b, c), suggesting a
mixing process of the clay sediments used as raw materials in terms
of relative proportion of clay and silt/sand fractions. In fact, the
samples of the two fabrics are both distributed along the trends,
with the clay specimens falling at the starting point of the trends.
In order to confirm the hypothesis that the Borġ in-Nadur-type
potteries from Ognina and Cannatello has a Maltese provenance,
their chemical composition was compared to the results
achieved on the pottery samples from Malta. Furthermore, in
order to verify a possible distinction on chemical base between
Maltese and Sicilian raw materials, a selection of diagnostic
Thapsos pottery from Ognina and clays from Syracuse (Barone
et alii, 2012; Tanasi, in press) were also plotted. In particular,
a useful discriminant diagram is represented by Zr vs Rb (Fig.
5d), in which it is possible to distinguish different clusters
related to BN# samples and to Sicilian pottery and clays. In this
context, it is noteworthy that Ognina (OG#) and Cannatello
(CAN#) are plotted with the Borġ in-Nadur Maltese pottery
trend, confirming the hypothesis about their provenance.
Furthermore, chemical analysis carried out on samples
for which the low number of specimens did not allow
G B, P M, S R, D T, A G
F. 5. B         .
()  () Al3O3/CaO  Al2O3/Fe2O3 . SiO2/CaO , ;
() SiO2/MgO . AL3O3/CaO ; () Rb . Zr  .
T 4. S    F A  B,  
      .
A   M B A 
T 5. C     . (M     %,    ).
G B, P M, S R, D T, A G
F 6. H       
     .
petrographic analysis to be carried out (namely, Unknown
fabric: BN8, BN10, BN12, BN14, BN17, BN21) confirmed
the homogeneity of the BN# reference local group, suggesting
a possible correlation with Fabric A samples due to their
comparable chemical trends.
In order to highlight differences in the chemical composition
of Maltese and Sicilian samples, a multivariate discriminant
analysis (Huberty 1994) has been performed on chemical data
of trace elements. As shown in Fig. 6, all the samples studied
are correctly classified by the discriminant function in which
the elements with higher load are Zr and Cr with standardized
canonical coefficient -3.97 and 3.08 respectively.
3. Remarks on the archaeometric data
Petrographic and chemical analyses carried out for the first time
on specimens sampled from the site of Borġ in-Nadur in Malta
have allowed the definition of a homogeneous reference local
group, representing an important benchmark in archaeometric
provenance studies of Middle Bronze Age pottery from Malta.
Furthermore, analyses allows us to distinguish the samples
analysed into two main groups, despite previous classifications
based on decoration and surface treatments. Data related
to micromass optical activity and the inclusions testify to
a rather mediocre technological level of local production
going by the low-medium firing temperature estimated. The
choice of inclusions, moreover, may be attributed to a good
specialization in the manufacturing processes. The chemical
data suggest a local provenance of raw materials, identified in
the Blue Clays sampled from Ġnejna Bay and characterised by
a mixing evolution trend (Carbone et alii, 1987).
Moreover, the presence of volcanic fragments as inclusions in
sample BN27 – shown to be made of Maltese clay – opens
new perspectives in defining the importation into Malta of
materials in the Middle Bronze Age, a phenomenon that is
known to have occurred even in the Roman era (Bruno 2004,
92). Further analyses are required to establish the source of
such volcanic materials.
Finally, the comparison between Sicilian and Maltese samples
allow us to distinguish the Middle Bronze Age pottery
production of the two islands on the basis of petrography and
geochemistry. In particular, the discriminant analysis appears
to be a powerful tool that allows the separation of the two
ceramic productions even if further data are necessary in order
to validate the procedure.
4. Archaeological implications
Results of this first in-depth archaeometric study of Borġ
in-Nadur pottery have several archaeological implications
of note. As summarised in Table 6, all the pottery classes
identified during the reappraisal of the ceramic assemblage
from the settlement, and their equivalents in the classification
of other scholars, can now be divided in the two groups with
Fabric A and B (Table 6).
A   M B A 
T 6. C         
      .
The ramifications of the analyses presented above are several
and further study is required to understand their significance:
r the fact that Fabric A includes fabrics 1 and 2, representing
the production of pottery over a long period of time
covering phases II B1 and II B2, points to a certain degree
of technical conservatism in pottery production;
r the conservative trend can also be adduced by the fact
that ‘burnished orange ware’, considered a forerunner of
Borġ in-Nadur pottery, is chemically and petrographically
identifical to other classes in the Fabric A group;
r the petrographic and chemical characteristics that the
‘Painted Ware’ shares with other classes in the Fabric A
group suggests that it can be distinguished only on the
basis of different manufacturing skills even though it
appeared initially to have been more technically advanced;
r the identification of variants with red slip for fabrics 3
and 5 turned out to be misleading, as suggested by their
diverse petrographic and chemical compositions, and
ought to be classified differently;
r the distinction between Fabric A and Fabric B corresponds
to the binomial fine ware/coarse ware, where the fine ware
is generally tableware, invariably slipped and decorated,
and the coarse ware is represented by domestic or cooking
r the analysis of the two samples taken from the conical foot
37331 (BN9) and the bell-shaped foot 37338 (BN 15)
respectively showed that they belong to Fabric B, rich in patic
calcite, suggesting a use of pedestal vessels possibly related to
the exposure of the vessels to a heat source;
r alternatively, Fabric B might not have been always singled
out for the use of objects that required resistance to
thermal shock, as suggested by the analysis of samples
BN30 and BN31, respectively taken from a clay anchor
(37458) and a portion of clay render (37461).
In conclusion, this research exercise on a selection of Middle
Bronze Age pottery from Malta has been able to show that the
more traditional typological study of pottery stands to gain
when it is supported by archaeometric data.
Permission to sample the pottery held at the National Museum
of Archaeology was g ranted by Dr Anthony Pace (ref. no. SCH
422/2012), Superintendent of Cultural Heritage (Malta)
who we thank for the interest shown in this work. We are
particularly grateful to Sharon Sultana, Senior Curator of the
National Museum of Archaeology for supporting our request
and for facilitating all the work in the museum despite a heavy
working load.
B, G., M, P., A, A., B, G.,
2014. The Hellenistic and Roman Syracuse (Sicily)
fine pottery production explored by chemical and
petrographic analysis, Archaeometry 56: 70-87.
G B, P M, S R, D T, A G
B, B. 2004. L’arcipelago maltese in età romana e bizantina.
Bari: Edipuglia.
C, S., G, M., L, F., P, M.
1987. The distribution and palaeoenvironment of
Early Miocene phosphorites of southeast Sicily and
their relationship with the Maltese phosphorites,
Palaeogeography, Palaeoclimatology, Palaeocology 58: 35-
C, V., D, M., R, C. 2012. Late Neolithic
and Bronze Age pottery from Tas-Silġ sanctuary:
new research perspectives for the Maltese prehistoric
sequence, Rivista di Scienze Preistoriche 18: 39-63.
E, J.D. 1971. The Prehistoric Antiquities of the Maltese
Islands: A Survey. Athlone Press, London.
F, B., G, S., S, S. 1997. L’alternativa
chamotte/calcite nella ceramica grezza/prove tecniche,
in S B, S., F, B. (eds), Atti del
convegno Il Contributo della Analisi Archeometriche
allo Studio delle Ceramiche Grezze e Comuni: Forma/
Funzione/ Impasto, Bologna 28 febbraio 1997: 183-190.
Bologna: University Press.
H C.J. 1994. Applied Discriminant Analysis. New
York: John Wiley & Sons.
M, Y. 2009. The Emergence of Ceramic Technology
and its Evolution as Revealed with the use of Scientific
Techniques, in S, A.J., F, I.C.,
R, T. (eds), From Mine to Microscope: Advances in
the Study of Ancient Technology: 1-18. Oxford: Oxbow
M, M.A. 1992. Pots and potters of prehistoric Malta:
a dissertation submitted in partial satisfaction of the
requirements for the degree of Doctor of Philosophy in
archaeology, UCLA 1988. University Microfilms Inc.
M, H., B, A., C B, K.,
K, I., M, M., S, C., S,
A., V, N.C., Z, N. 2006. Characterization
of Maltese pottery of the Late Neolithic, Bronze Age
and Punic Period by neutron activation analysis, in
M, M., M, B., Geomaterials in Cultural
Heritage (Geological Society special publications 257): 81-
89. London: Geological Society.
R, S., B, G., M, P., T, D., C,
E., in press. Mobility of men versus mobility of goods:
archaeometric characterization of Middle Bronze Age pottery
in Malta and Sicily (15th-13th century BC), Periodico di
S, C. 2008. Malta: between a rock and a hard place,
in S, C. (ed.) Beyond the Homeland: Markers in
Phoenician Chronology (Ancient Near Eastern Studies
Supplement 28): 487-536. Leuven: Peeters.
S, C. 2011. Observations on the Late Bronze Age and
Phoenician-Punic pottery in Malta, in S, C. (ed.)
Ceramics of the Phoenician-Punic world: collected essays
(Ancient Near Eastern Studies Supplement 36): 397-432.
Leuven: Peeters.
T, D.H. 1961
. The Later Prehistory of Malta, Proceedings
of the Prehistoric Society 27: 253-262.
T, D. 2011. The Prehistoric Pottery, in T, D., V,
N.C. (eds), Site, Artefacts, Landscape: Prehistoric Borġ in-
Nadur, Malta, (Praehistorica Mediterranea 3): 71-158.
Monza: Polimetrica.
T, D. 2013. ‘Prehistoric Painted Pottery in Malta’: One
century later, Malta Archaeological Review 8 (2008-2009):
T, D. 2014. Uniti e divisi dal mare. Mobilità di
uomini e merci tra la Sicilia e l’arcipelago maltese nel
II millennio a.C., in C, M., M, C.,
M S. (eds), Viaggio in Sicilia. R acconti, segni e città
ritrovate, 99-116. Caltanissetta: Sciascia.
W, I.K. 1995. Greek Transport Amphorae. A
Petrological and Archaeological Study. Athens: The
British School at Athens.
P M is Associate Professor at the Department
of Biological, Geological and Environmental Sciences of the
University of Catania where he has lectured in petrography
and applied petrography. He teaches numerous training and
specialization courses in the context of geo-mineralogical
disciplines applied to environmental studies and cultural
heritage. From the academic year 2010/11 he co-ordinates
the M.Sc. Course in Geology at the University of Catania.
He currently conducts research on the use of innovative
methodologies in mineralogy and microstructure of stone and
pottery analysis, also in collaboration with both Italian and
foreign researchers. He has authored 58 ISI publications in
national and international scientific journals.
G B, Researcher at the University of Catania
in mineral geo-resources and mineralogical and petrographic
applications for the environment and cultural heritage at the
Department of Biological, Geological and Environmental of
the University of Catania. Her research interests are focused
on the archaeometric characterisation of cultural heritage
materials, collaborating with national and international
research agencies, museums and superintendencies. The
main topics of her publicatons include innovative destructive
and non-destructive analytical methods for the minero-
petrographic and geochemical study of rocks and artificial
A   M B A 
stone materials. She has authored 54 ISI papers in national and
international scientific journals.
S R, Conservation Scientist for Cultural
Heritage and Geologist, is a doctoral candidate in the
Department of Biological, Geological and Environmental
Sciences at the University of Catania, where she works on a
research project devoted to the modelling of the inner structure
of natural building stone by applying fractal geometry to test
nanostructured products for the use in conservation, under the
supervision of Dr G. Barone and Prof. P. Mazzoleni. In addition
to her main research activity, she works also on archaeometric
characterisation of artificial stone materials through innovative
non-invasive and non-destructive methodologies. She has
delivered papers in several international conferences and
published papers in national and international journals.
D T holds a Ph.D. in Archaeolog y and is Professor
of Archaeology at Arcadia University, The College of Global
Studies – Arcadia Sicily Center. His research interests include
Mediterranean prehistory, island archaeology, archaeometry
of ancient ceramics, computer graphics in archaeology, and
digital communication of cultural heritage. He has authored
a hundred scientific papers in these fields and produced
3D documentaries about Sicilian archaeology and cultural
heritage. His publications include La Sicilia e l’arcipelago
maltese nell’età del Bronzo Medio
(Officina di Studi Medievali,
2008) and Site, artefacts and landscape. Prehistoric Borġ in-
Nadur, Malta (Polimetrica, 2011, co-edited with N.C. Vella).
He is the director of the Field School in Archaeology of
Arcadia University in Sicily and editor of the international
journal Open Archaeology (De Gruyter).
A G, M.Sc. student in Geology at the
University of Catania, works on chemical, mineralogical,
and spectroscopic characterisation of volcanic ashes. Since
2013, he collaborates with the research group of Applied
Petrography at the Department of Biological, Geological
and Environmental Sciences at the same university for the
archaeometric characterisation of archaeological materials. He
has co-authored several scientific papers.
... We propose here to look at the ceramic evidence in order to answer questions about raw materials use, the spatial relationships involved, and the sites of ceramic production and sites of ceramic use; and whether any technological knowledge movement across space and time might be seen. In this paper, we are not taking into consideration the extra-island movement of people and objects from the Aegean, which has been already widely investigated ( Barone et al., 2015;Day and Joyner, 2005;Jones et al., 2014;Tanasi et al., 2017). ...
This paper reviews the materials primarily studied by the two authors in the attempt to reconstruct the Bronze Age cultural landscape of northern, central and eastern Sicily and the Aeolian islands from the perspective of ceramic production. In particular, we aim at tracing a diachronic reconstruction of the choices made by potters and ancient communities in relation to the landscape resources present in Sicily and to the different wares produced at each site. We pursue the question of natural resources beyond the traditional concept of "raw materials" by also considering some more overlooked resources in ceramic studies. In doing so, we intend to shed light on the web of knowledge created by potters and on their role in shaping the cultural landscape of Bronze Age Sicily. This is a first step toward a more integrative study of ancient ceramic production and natural resources for Bronze Age Sicily.
... Conversely, shapes including tiles and jars are characterized by micaceous groundmass and fine quartz (Fabrics 3, 4 and 5), with the presence of metamorphic rock fragments only in samples DC02/195, IPM/INQ981/ 34 and IPM/INQ981/86 (Fabric 1). It is interesting that Fabric 2 includes samples archaeologically and typologically classified as Maltese materials (e.g. the kenur tal-fuħħar, see above); in this sense, for all of these samples a local provenance can be suggested on the basis of minero-petrographic and geochemical features (see for comparison Raneri et al., 2015;Barone et al., 2015). For samples characterized by metamorphic rock fragments (Fabric 1), similarity with well-known Sicilian ceramic workshops active in the area of Strait of Messina can be assessed (see for comparison Barone et al., 2005aBarone et al., , 2005b. ...
Abstract The development of post-medieval archaeology has been uneven in the Mediterranean. While major advances have been made in areas, such as the Aegean and Italy, others have garnered little sustained archaeological interest. In Malta, one of the principal barriers to the development of post-medieval archaeology has been the lack of knowledge and published findings regarding one of the most important find-types: pottery. The present study presents the results of the first archaeometric (chemical and minero-petrographic) analysis of post-medieval pottery in Malta, which sheds light on the changing ceramic relationships between Malta and Sicily during the 17th to early 20th centuries. The results demonstrate that, on the one hand, locally-made Maltese pottery remained important throughout the post-medieval period, as did pottery from the Messina Straits area. While, on the other hand, pottery from the Syracuse area can be shown to be linked to the activities of the navy of the Knights of St John.
An underwater survey off the southwest coast of the island of Gozo revealed a well-preserved shipwreck 110 m below the surface. The site belonged to a previously unknown wreck with a cargo of volcanic millstones and ceramic amphorae dating to the 7th century BC. This article presents the first results of thin-section analysis taken from the pottery objects, and concludes that the ship was carrying a heterogeneous cargo of amphora-borne goods from the Maltese islands, North Tunisia, and possibly Sicily, making it the earliest, known shipwreck in the central Mediterranean; and provides the earliest evidence for Maltese external trade in the central Mediterranean.
The end of prehistory in the Maltese archipelago is characterized by the production of a problematic class of pottery, until now attested just at the site of Baħrija, on the western coast of Malta. Such a production represents a break with the tradition in terms of repertoire of shapes, style and technology and it has been interpreted as the result of contact between locals and foreign immigrants. The recent overall reappraisal of the unpublished ceramic assemblage collected during the excavations carried out at Baħrija, represents a unique opportunity to focus on the technological aspects of the production, trying to shed light on the issue of the break with the tradition and the impact of external influxes. Petrographic analysis on thin sections and chemical analyses via X-ray fluorescence spectroscopy (XRF) and laser ablation-inductively coupled plasma mass spectroscopy (LA-ICP-MS) have been carried out to characterize the Baħrija pottery production in order to interpret from a different angle the issue of the possible arrival of newcomers and establishment of a foreign enclave in Malta, which until now has been hypothesized only on the basis of the sudden emergence of the Baħrija pottery.
Full-text available
The aim of this research was to determine the provenance of Maltese ceramics and to determine the role pottery played in Maltese prehistoric trade and interaction networks. This study involved 236 Maltese ceramic samples, 19 geological clay samples from Ġnejna Bay & Selmun along with 18 ceramic samples from Ognina, Sicily, and four Sicilian clay samples from the outskirts of Siracusa that were nondestructively analyzed using a portable X-ray fluorescence (pXRF) spectrometer in order to determine their trace elemental compositions (Th, Rb, Sr, Y, Zr and Nb). The results of this analysis were statistically analyzed using principal component analysis in order to ascertain relationships in the chemical compositions among the samples. The results of this analysis indicate that the majority of all the Maltese ceramic samples have a local Maltese provenance and that pottery played a more significant role in defining the nature of Malta’s trade relationships during the Bronze Age. The following study has provided new insights into Malta’s role in trade and interaction networks from the late Neolithic to the Bronze Age and has allowed for new ideas in explaining the cultural change observed from the Temple Period to the Bronze Age.
Full-text available
During the study of a pottery assemblage collected at the Sicilian site of Licata-Caduta (Agrigento), continuously occupied between the Neolithic and Early Bronze Age, some ceramics presented typologies, fabrics and surface treatments alien to Sicilian repertoires and closer to Maltese productions. Sampled for petrographic (OM) and chemical (XRF) analyses one artefact turned out to be produced in Malta and imported into Sicily. The search for its chronological and cultural placement has led to the discovery of the first certified Żebbuġ phase import in Sicily.
Full-text available
A set of 41 samples from Tas-Silg, Malta, has been analysed by neutron activation. It contained nine ware groups formed by visual examination covering the Late Neolithic, Bronze Age and Punic Periods (c. 3000-218 BC). Despite this diversity and long time range, seven of these ware groups, including the 'Thermi Ware', all have a similar chemical composition and, therefore, have been made from the same clay. This points most probably to a local origin. One group from the Punic Period, containing only Bricky Red cooking ware, is chemically separate and represents a second distinct pattern probably assignable to a local production. Five amphora sherds also from the Punic Period, and consisting of a micaceous fabric, all have different chemical characteristics and are probably imports from overseas production sites of unknown location.
For many years down to 1953, our knowledge of Maltese prehistory could be summed up succinctly if rather unkindly in the phrase—‘Neolithic 3,000 B.C., Bronze Age 2,000, Punic 1,000.’ In that year, J. D. Evans's researches were published in these Proceedings . These at last provided a framework for his Period I, no longer called ‘Neolithic’ because the overlap of its later phases with metal-using cultures in nearby Sicily made it unlikely that metal was quite unknown. It was assigned a duration from the mid-second to mid-first millennium B.C. The absolute chronology will need revision in the light of the C-14 dates not then available and the correlations with the Sicilian development have met with some criticism. In any case, the isolation of the different phases was an enormous advance on which all further work in Malta will have to be based, even if, as at the time of writing begins to seem likely, certain amendments to the sequence become necessary. The later prehistoric Period II, lasting down to the 9th century when the Phoenician settlement opened Period III, was described in much less detail. Enormous quantities of material of the first period, megalithic buildings as well as pottery and small finds, were available for study: the material remains of Period II were much more scanty, there being in effect at that time only a single site known of each of its three phases, which were correspondingly named after the Tarxien Cemetery, Borġ in-Nadur and Baħrija. In 1956, Evans published a more detailed study of the first of these phases bringing forward evidence for attributing to it the local dolmens. Phases II B and C were not ready for such treatment as the only excavations were at Borġ in-Nadur in 1881 (a sketch plan found in a Valletta photographer's shop twenty years later being the only record) and 1921-7 (disturbed levels overlying a Period I temple); and at Baħrija in 1909 (three days' work). These two phases therefore remained the most urgent problem in the prehistory of Malta. Accordingly, further excavation was undertaken on their type sites by the Museum Department of the Maltese Government in the spring and autumn of 1959.
The aim of this paper is to offer the first petrographic and chemical characterization of Middle Bronze Age pottery of Malta (Borġ in-Nadur, half of 15th - early 12th century BC) to discriminate, in the multitude of Borġ in-Nadur vessels found in coeval sites of southeastern and central-western Sicily, between imports and imitations and to precisely define their provenance. The present research may provide new data in order to shed light on the long standing issue about ‘mobility of men’ and ‘mobility of goods’ between the two insular contexts. A significant amount of Borġ in-Nadur pottery from Malta and Sicily have been sampled and subjected to petrographic exam on thin sections; moreover, mineralogical and chemical analyses of major and trace elements were performed by X-ray diffraction and X-ray fluorescence spectrometry, respectively. Finally, chemical data were treated with statistical method (Aitchison, 1986) in order to attempt to distinguish Maltese and Sicilian Middle Bronze Age pottery productions.
The chemical composition and petrographic features of 51 samples of fine pottery selected from the archaeological site of Syracuse have been established by means of optical microscopy and X-ray fluorescence spectrometry, with the aim of creating a reference group for the ceramics produced in this important Greek colony during the Hellenistic–Roman period. This reference group is constrained by the analysis of six kiln wasters and of raw materials outcropping in the studied area. Among the studied findings, the main part is assignable imputably to local production, while some specimens probably come from Greece and other colonies (Messina and Gela). The Syracuse fine pottery is characterized by inclusions formed mainly by quartz and micaceous and fossiliferous groundmass. The used pastes were more or less purified through the removal of the sandy-to-coarse silty granulometric fraction. The multivariate statistical analysis of chemical data permits the identification of production at Syracuse on the basis of comparison with the kiln wasters and the Plio-Pleistocene clays.
Several phosphatic limestone localites occur within the Irminio Member (ragusa Formation) of the western Hyblean of southeast Sicily. In areas of maximum development,such as around Sampieri and Modica, the phosphatized, intraslcast-dominated deposits attain a thickness of up to 0.6m and are associated with a phosphatized underlying hardground. Thin and possibly discontinous spreads of phosphatized intraclast form a broader zone beyond the central semi-autochthonous zone.Microfossil analysis of associated strata demonstrates that all occurences of phosphatized strata within the Irminio Member are restricted to a single level of probable Late Burdigalian (Early Miocene) age. The bed correlates well both biostratigraphically and lithologically with the “Upper Main Phosphorite Conglomerate Bed” of the Maltese Islands, which lie c.95 km south of the Sicilian province.Five more distinct phosphorite beds occur in the Maltese Islands, being best developed in the western and northern parts of the islands. They range in age from Aquitanian to Late Burdigalian. These are principally francolite replaced micrate conglomerates which may be associated with phosphatized underlying hardgrounds in places.A major central Mediterranean phosphatogenic province of Early Miocene age is postulated inorder to account for the simultaneous occurrence of phosphorites in Sicily and Malta.