Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

2008 Apulian ENX

Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

Journal of Archaeological Science 35 (2008) 1533e1541

http://www.elsevier.com/locate/jas

Technological features of Apulian red figured pottery


A. Mangone a, L.C. Giannossa a, A. Ciancio c, R. Laviano b, A. Traini a,*
b

a
Dipartimento di Chimica, Universita` di Bari, via Orabona 4, 70126 Bari, Italy
Dipartimento Geomineralogico, Universita` di Bari, via Orabona 4, 70126 Bari, Italy
c
Soprintendenza Archeologica della Puglia, 70122 Bari, Italy

Received 13 November 2006; received in revised form 6 October 2007; accepted 26 October 2007

Abstract
Apulian red figured pottery samples, dating back to the 5th and 4th centuries BC, from the archaeological site of Monte Sannace (Gioia del
Colle, Bari, Italy) have been characterized from the physicalechemical, mineralogical and morphological points of view. Scanning electron
microscopy, X-ray diffraction and atomic spectroscopy investigations have been carried out on the ceramic body, red decorated area and black
gloss of the fragments, with the aim of outlining the technological features and of defining the nature of coatings and decorations. All 5th century
fragments show the same features: fine texture of the ceramic body, red figures saved from the ceramic paste and black gloss painted directly on
the ceramic body. The statistical treatment of compositional data of ceramic bodies excludes the local production of these objects. As regards the
4th century fragments, some show similar features to the 5th century ones; however others are characterized by the coarse texture of their
ceramic body and an intermediate red layer of finer clay between the black gloss and the ceramic body. The analytical results make it possible
to distinguish two different production technologies of red figured Apulian vases in Monte Sannace during the 4th century BC. Certain vases
were produced using the classic Attic technology and others with a different technique, not previously known, which involved the application
of a red engobe layer on the clay paste, before the black gloss painting.
2007 Elsevier Ltd. All rights reserved.
Keywords: Apulian red figured pottery; Scanning electron microscopy; Atomic spectroscopy; Ceramic; Black gloss and red decoration; Production technology

1. Introduction
From the 3rd quarter of the 5th century BC to the end of the
next century, there was intense production of red figured
ceramic items in Apulia (South Italy). Known as Apulian
red figured pottery, it is characterized by excellent drawing
ability and by remarkable quality. Between 450 and 300 BC,
Apulian red figured pottery was the most important quantitative handcraft production group of figured pottery in Magna
Grecia, the most widespread and commercialized both within
and outside the region. The traditionally accepted start date for
this manufacture is 440 BC, with the so called proto-Apulian
production. Three phases of truly Apulian production followd
Ancient, Middle and Late (Robinson, 1990; Trendall, 1989;
Trendall and Cambitoglou, 1978, 1982)dthroughout the
* Corresponding author. Tel.: 39 080 544 2021; fax: 39 080 544 2129.
E-mail address: fabia@chimica.uniba.it (A. Traini).
0305-4403/$ - see front matter 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jas.2007.10.020

4th century, during which an increase in production was


observed. The increasing trend was quite gradual during the
Ancient and the Middle phases and became clearly sharper
in the Late period, when iconographically and formally excellent handcraft pieces, such as huge monumental vases with detailed drawings from mythology and theatre plays, and low
quality mass production pieces are found together.
From the proto-Apulian period, the production management
in the workshops was complex and well structured. Two different craftsmen could work on a single vase, apprentices could
co-work with the main painter for easy tasks and minor details,
the production could be diversified depending on the final use
of the item (e.g., for daily use, for a burial, etc.) and on the
customer target. This production management persisted
throughout the next century, but in the Late period, the relevant production increase and the sharper separation between
low-quality mass production and the excellent artistic items
by the most famous artists, offer us many questions still to

1534

A. Mangone et al. / Journal of Archaeological Science 35 (2008) 1533e1541

be answered on the system of production of Apulian red figured pottery.


The most interesting problem to be solved deals with the
identification of the workshops, especially for the Middle
and Late periods, when production increased. The most likely
theory accepted so far hypothesizes that a temporary transfer
of expert craftsmen and artists occurred from Taras (Taranto)
to the wealthiest Apulian villages, the most interested in gaining such expertise, at least for the most valuable pottery
pieces. Then these became branch centres of production outside the main polis (city). However, the production of items
of minor importance for local use was organized locally.
The problem of solving the complex questions raised by the
stylistic investigations on Apulian red figured pottery opened
this field to archaeometric research. The goal is to clarify,
on the basis of unambiguous elements, the technological processes characterizing this particular kind of pottery and, most
of all, to identify possible differences between objects that
could help to distinguish between the different production processes and workshops.
Archaeometric investigations on red figured pottery have so
far mainly dealt with Attic production, and were aimed at
understanding the specific characteristics of each step in the
overall production cycle, with particular attention devoted to
the technique utilized to realize the black gloss (Canosa
et al., 1996; Ingo et al., 2000; Jones, 1986; Kingery, 1991;
Maniatis et al., 1993; Mirti, 2000; Mirti et al., 1996, 2004,
2006; Noble, 1960; Tang et al., 2001; Tite et al., 1982). There
is as yet a lack of data about red figured vase production from
Magna Grecia (Canosa et al., 1996; Grave et al., 1997; Mirti,
2000; Mirti et al., 1996, 2004, 2006); only approximately ten
fragments of Apulian red figured pottery have been analysed
(Canosa et al., 1996; Grave et al., 1997), a small number
and from different sites, so the results cannot be considered
representative of such complex, structured and assorted
production.
This investigation concerned Apulian red figured pottery
production coming from Monte Sannace (Gioia del Colle,
Bari, Italy), a site among the most relevant of ancient Peucetia
(corresponding to central Apulia).
2. Experimental
2.1. Samples
The collection of the examined samples consists of 37
items, codified in Table 1, either fragments or unbroken and
reassembled objects differing in shape and dimension, recovered during archaeological excavations carried out in the last
century at the site of Monte Sannace. Only the pelike (sample
15, inv. no. 61262) came from another site of archaeological
interest in the same area, the archaic-classical necropolis of
Santo Mola.
Monte Sannace, one of the most important cities of ancient
Peucetia, underwent particularly significant economic and
cultural growth in the 4th and 5th centuries BC. In that period,
in that centre, the red figure ceramics, mainly from tomb

Table 1
Analysed samples
Code
1
2
3
4
5
6e14
15
16
17
18
19
20
21
23e38

Inventory numbera
505
504
339
343
333
61262
3862
871
20054
20055
892

Typology

Dating (century BC)

Bell crater
Skyphos
High-footed bowl
High-footed bowl
Vase wall
Fragments
Pelike
Oinochoe
Bell crater
Crater border
Mask volute crater
Pelike
Bell crater
Fragments

Mid 4th
Beginning of 4th
First half of 4th
First half of 4th
End of 5th
Mid 4th
4th
Mid 4th
2nd half of 4th
End of 4th
2nd Half of 4th
Mid 4th

a
Unbroken and reassembled items, as well as fragments kept at National
Archaeological Museum of Gioia del Colle (Bari).

contexts, constitute a well represented class, with both intact


examples in various forms (craters, oinochoi, amphoras,
plates, bowls, skyphoi), as well as fragmentary material.
Some of the findings sampled can be classified stylistically
and chronologically on the basis of comparison with other ceramics and thanks to discovery data (the context which they
belong). Sample 5 comes from a fragment of the wall of
a closed vase (inv. no. 333), which conserves a part of the accessory decoration (palms and scrolls), as well as the traces of
Greek letters over-painted in white. The stylistic examination
indicates an attic or colonial product (from Metapontum or
Taras) from the end of the 5th century BC. Samples 3 and 4
(inv. nos. 339 and 343 respectively) come from two rounded
bowls turned back at the rim, resting on a high circular base,
both fragmentary and incomplete. The external decoration
consists of two male cloaked figures walking towards a small
square column. The inside shows a naked athlete, with a strigil
in his hand and boots on his feet. The bowls belong to Apulian
production in the decade between 380 and 360 BC, which
recall those of an artist, the so-called Pittore di Tarporley,
who operated in the early 4th century BC under the influx
of coeval production from the neighbouring region (Lucania).
This latter recalls sample 2, a skyphos (inv. no. 504) with two
female figures with mirrors in their hands, one on each side of
the vase, reconstructed from a number of fragments and filled
in some parts, while the oinochoe (sample 16, inv. no. 3862),
very fragmentary and incomplete, with decorative motifs not
clearly identifiable, can only be generically collocated chronologically around the 4th century BC.
The three vases (samples 1, 17 and 21, inv. nos. 505, 871
and 892 respectively) belonging to the production of the Middle period, bell craters with a rather common scene (maenads
and satyrs, cloaked male figures) and the pelike, sample 15, are
datable to the mid 4th century BC.
Finally, the large mask crater (sample 19, inv. no. 20054), is
noteworthy, with a representation of the deceased eroticised
inside the funerary temple, an example belonging to the highly
valued Apulian production of the late phase. Stylistically the
vase belongs to a group of products realised by artists,

A. Mangone et al. / Journal of Archaeological Science 35 (2008) 1533e1541

followers of an important maestro of the period (the so-called


Pittore di Dario), which have been found in various archaeological sites of Peucetia and Daunia.
The characteristics of the collection of samples strongly
influenced the analyses and forced some limits. For example
black gloss and red decorations were only analysed on the
fragments, while chemical composition of ceramic bulks was
carried out both on fragments and on items in museum
exhibitions.
The visual examination of objects revealed that the ceramic
body colour of some fragments differs from that of the red
decorations, the ceramic bulk being light brown / greyish
and the decorations purplish red.
2.2. Techniques
The fragments were examined with different complementary techniques, namely: polarized-light optical microscopy
(OM); scanning electron microscopy (SEM) with energy dispersive spectrometry (EDS); X-ray diffraction (XRD); atomic
absorption spectroscopy (AAS) and inductively coupled
plasma emission spectroscopy (ICP-OES). Multivariate statistical techniques were used for treating the compositional data.
Orthoscopic observations of the mineralogical textures
were performed by means of optical microscope (Carl Zeiss)
on polished thin sections, whereas SEM investigations were
carried out on untreated samples and on polished thin sections,
after graphite sputter-coating of the samples. Two SEM instruments, a S360 (Cambridge Instruments) and a EVO-50XVP
(LEO) were used. Microanalyses were conducted using an
Oxford-Link EDS instrument equipped with a Ge detector
and with a 0.4-mm-thick Super Atmosphere Thin Window
(SATW).
X-ray diffraction analysis was performed by using a Philips
XPert Pro X-Ray diffractometer, with the following working
conditions: CuKa Ni-filtered radiation, 40 kV, 40 mA, divergence slit 1 , anti-scatter slit 0.5 , receiving slit 0.2 mm, speed
0.5 in 2q per minute.
The analyses were carried out with the specimen directly
inserted into the sample-rack of the instrument, with the
X-ray beam pointing directly on the specimen surface, as
described by Tang et al. (2001). Powder samples were not
used, due to the impossibility of: (1) removing matter exclusively from the surface decorations without contaminating it
with bulk material; (2) obtaining enough powdered material
to perform XRD investigation within the detection limits of
the technique.
XPS (X-ray photoelectron spectroscopy) analyses were
performed using a Kratos Axis ULTRA DLD X-ray photoelectron spectrometer upgraded with a Delay Line Detector.
Data were acquired using monochromated Al Ka X-rays. All
analyses were performed on the samples as received and after a 120 s ion gun sputter-etching with Ar ions at 2.0 KeV.
The automatic charge neutralizer system of the instrument
was used since the samples are insulators, and the C 1 s photoelectron peak (Binding Energy calibrated at 285.0 eV) was
used.

1535

The elemental chemical composition of the ceramic body


of the samples was investigated; bulk ceramic matter was
scraped off fractures already existing on the fragments and
small visible points of the object, inside or under the base,
on the items in museum exhibitions, after removing the outermost external contaminated layer.
Aliquots of about 30 mg of bulk ceramics (Hatcher et al.,
1980), a good agreement between archaeological microdestructivity and analytical data quality, were dissolved by
acid attack with a solution of 37% HCl, 70% HNO3 and
40% HF (Fluka trace selected for trace analysis reagents), in
a 2:1:1 (v/v/v) ratio (Bruno et al., 2000). Ten elements were
quantified by atomic absorption spectrometry (Shimadzu
AA-6701 spectrometer): Na, K, Fe, Ca, Mg, using an airacetylene flame, Cr, Sr and Ni by the graphite furnace method.
Al and Ti were determined by inductively coupled plasma
emission spectrometry (Varian Liberty 110 sequential plasma
emission spectrometer). External calibration with matrix
matching standards was employed for quantification and five
replicate readings were performed on both standards and samples. The entire analytical procedure was tested on standard
clay material (Brick clay SRM 679 (National Institute of
Standards and Technology, Gaithersburg, MD, USA)) and
the results of ten replicates are reported in Table 2.

3. Results
3.1. Black gloss
The SEM analysis of the black gloss on all samples revealed an average thickness of about 20 mm; a very compact
structure with no voids and a large degree of sintering with
no clay structure evident, as shown in Fig. 1. These data
clearly indicate that a finer clay was used in the production
of the black gloss than that utilized for the ceramic body, probably with a granulometry lower than 2 mm.
ED spectra of the black gloss layer revealed a homogeneous
composition through all samples, as well as larger quantities of
Al, Fe, K and sometimes also of Ti, and lower quantities of Ca,
with respect to the ceramic body. As reported by Kingery
(1991), the black gloss could have been obtained in three
ways: by selecting a specific clay; by using the finer fraction,
richer in iron and potassium oxides of a decanted claydthe
coarser fraction, rich in calcium oxide, could be used for the
ceramic bodydor by adjusting the composition of the clay
with specific compounds. In our case, the high Al/Si (y1)
ratio measured and the negligible amount of Ca suggest that
two different clays were used for the black gloss and ceramic body. The use of the so-called terre rosse is compatible
with our data. Terre rosse, very common all over Apulia, are
continental sedimentary layers characterized by a silty-clay
granulometry. Their mineralogical composition includes
mainly partially crystalline Fe and Al oxides and hydroxides,
clay minerals (illite and kaolinite) and traces of quartz, feldspars, micas, pyroxenes and other minerals (DellAnna,
1967; DellAnna and Laviano, 1987, 1991). The finest fraction

A. Mangone et al. / Journal of Archaeological Science 35 (2008) 1533e1541

1536

Table 2
Determined element concentrations by Atomic Spectrometry of ten replicates of SRM 679
Replicate

Content, mg/g

Content, wt%
Al

Fe

Mg

Ti

Na

Ca

Cr

Sr

1
2
3
4
5
6
7
8
9
10

10.25
10.35
10.15
10.26
10.41
10.82
10.26
10.47
10.38
10.30

8.92
8.93
8.76
9.04
9.04
9.25
9.02
9.19
8.97
8.90

2.38
2.36
2.28
2.24
2.13
2.35
2.14
2.25
2.20
2.28

0.73
0.73
0.72
0.69
0.72
0.76
0.71
0.74
0.73
0.72

0.58
0.59
0.57
0.58
0.59
0.61
0.58
0.59
0.59
0.58

0.13
0.13
0.12
0.14
0.11
0.13
0.10
0.13
0.13
0.13

0.141
0.188
0.128
0.149
0.201
0.153
0.169
0.133
0.173
0.180

87.3
91.9
91.1
103.3
87.8
87.0
88.9
89.4
86.3
87.6

68.6
70.4
70.3
73.1
73.5
71.1
70.7
71.3
69.1
69.2

Mean
SD
Reference value

10.37
0.18
11.010.34

9.00
0.14
9.050.21

2.26
0.09
2.4330.047

0.73
0.02
0.75520.0088

0.59
0.01
0.5770.033

0.13
0.01
0.13040.0038

0.161
0.024
0.16280.0013

90.1
5.0
109.74.9

70.7
1.6
73.42.6

of such terre rosse, easily separated by decantation, could have


been used to realize these black glosses. In fact, this fraction,
richer in kaolinite and illite as well as in Fe and Al oxides,
could give the gloss investigated its technological and chemical properties.
The mineralogical phases of black gloss, obtained from
XRD spectra, include hercynite, magnetite, haematite, quartz
and feldspar. The coexistence of Fe-spinels, i.e., hercynite
and magnetite, allows us to hypothesize that the reducing
phase of the firing took place at a temperature not higher
than 900e950  C (Maggetti et al., 1981). The presence of haematite could be due to incomplete reduction of Fe(III) compounds during the reducing phase of firing.
The black gloss of sample 13 is different from the others. It
is much thicker, about 50 mm, with a large amount of inclusions as well as voids, craters and channels (Fig. 2). Nevertheless, no evident compositional difference was shown between
this black gloss and the others by EDS analysis. These peculiar
morphological features allowed us to infer gas leakage during
firing and led to further surface analysis by XPS.
XPS spectra of the black gloss surface before and after Ar
etching were compared and a larger amount of carbon was
found after etching (about 20%, atom content), far too much
to be explained only by the surface contamination due to the
soil during burial; in fact, other samples, and the same sample
13 in the red area, exhibited a carbon content between 5% and
8% after Ar etching under the same conditions. Unless the
presence of carbon was caused by the decomposition of
carbonates present in the clay body, these results support the
hypothesis that a significant amount of some kind of organic
matter could have been added to the clay used for the black
gloss. The thicker layer of this sample would have led to
some sort of enhanced segregation of carbon at the uppermost
layer, originated by carbon oxides diffusing from the deeper
layers of the glaze. This would also explain the craters on
the surface and the channels seen in cross section, opened
by the leakage of carbon oxides during firing. Organic matter
was probably also added to the clay of the other black gloss

samples but their lower thickness allowed the complete removal of the gas developed during firing, so less carbon was
found there.

3.2. Ceramic body and red decorated areas


3.2.1. Chemical analysis and statistical treatment of data
The results of the chemical analysis by atomic spectroscopy
of the ceramic bodies are reported in Table 3, expressed as
percentage weight for major and minor element oxides in
the solid samples, and as ppm (mg/g) for the trace elements.
These compositional data were processed with Principal Components Analysis (PCA), using the SCAN software package
(Minitab), with the main aim of identifying groups of objects
distinguished on the basis of their compositional features.
The results of the multivariate statistical treatment are
shown in Fig. 3, illustrating the scores plotted on to the first
three principal components subspace, which accounts for
65% of the total variance, and the loading plot of the different
parameters. Three markedly distinct groups can be identified.
The same figure shows the 95% isoprobability ellipsoids,
whose surfaces define the boundary of three clusters, A, B
and C.
Clusters A and B differ from C along PC1 principally due
to the loadings relative to the Ca, Al, Fe, Sr and Ni parameters.
In particular, the A and B scores are characterized by negative
values of PC1 due to higher amounts of Al, Fe, Sr and Ni,
while C scores are spread along positive PC1 values due to
the higher Ca and Na content. The widening of cluster B along
negative PC3 is mainly due to the loadings of Mg, K and Cr.
More ancient objects, dating from the late 5th century BC
and the first half of 4th according to archaeological classification, are grouped inside cluster A, while clusters B and C
include objects dating back to the 4th century BC. To identify
the reasons that split the scores of 4th century fragments into
the two different clusters - B and C -, the fragments were
examined from the morpho-mineralogical point of view.

A. Mangone et al. / Journal of Archaeological Science 35 (2008) 1533e1541

1537

Fig. 1. SEMeBSE photomicrograph of the thin sections of the fragment 23 with highlighted characteristic differences between the black gloss (upper brighter
layer) and the ceramic body (lower layer). The ED spectra of the black gloss (upper) and of the ceramic body (lower) are shown on the right.

Fig. 2. SEMeBSE photomicrograph of the fragment 13. Left: black gloss surface, where a loose structure is evident, with many voids and inclusions. Right: thin
section of the same fragment, with the black gloss layer (above, thickness 50 mm; evident gas channels), and the ceramic body (below).

A. Mangone et al. / Journal of Archaeological Science 35 (2008) 1533e1541

1538

Table 3
Ceramic body composition by atomic spectroscopy
Sample

Element or element oxide concentration


a

Cluster

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
a

B
A
A
A
A
C
C
C
C
A
A
A
C
C
B
B
B
B
B
B
B
B
C
C
B
C
C
C
C
B
B
B
B
B
B
B
B

(% w/w)

(ppm)

CaO

MgO

Na2O

K2O

FeO

Al2O3

TiO2

Cr

Ni

Sr

8.25
8.94
8.51
8.27
7.60
11.22
8.80
14.47
13.93
9.53
8.55
9.92
13.78
10.97
7.39
7.33
12.02
9.83
10.81
8.74
8.20
8.56
14.10
8.72
8.00
8.98
8.65
14.47
14.73
5.78
8.11
7.32
8.16
8.72
8.52
9.23
9.39

2.29
2.06
2.09
2.14
1.97
1.31
1.59
1.16
1.49
2.37
2.29
2.30
1.39
1.51
1.49
2.24
3.65
3.98
3.75
3.42
3.68
1.81
1.46
1.18
1.69
1.19
0.80
1.49
1.19
2.47
2.37
2.34
2.50
2.57
2.54
2.70
2.64

0.48
0.40
0.40
0.40
0.39
0.50
0.96
0.66
0.92
0.40
0.43
0.43
0.48
0.44
0.78
0.74
0.74
0.73
0.71
0.63
0.66
0.77
0.70
0.84
0.84
0.79
0.82
0.67
0.65
0.62
0.98
0.90
0.89
0.88
0.94
1.06
1.23

3.72
1.43
2.45
2.23
2.19
2.36
2.57
2.37
2.14
2.75
2.76
2.43
2.10
2.37
2.54
2.59
2.59
2.39
2.90
2.30
2.28
2.58
2.25
2.42
2.77
2.31
2.46
2.28
2.00
3.00
2.24
2.01
2.95
2.82
2.77
2.99
2.89

5.25
6.01
6.39
6.76
6.06
5.48
5.08
5.89
5.18
7.10
6.82
6.62
4.72
5.62
8.58
4.63
4.84
4.49
4.45
3.64
3.64
4.57
3.18
3.33
4.49
3.05
3.29
1.95
2.15
2.64
7.07
6.65
6.82
6.61
6.10
6.37
5.70

10.32
13.96
12.77
12.91
12.13
11.03
10.17
12.11
10.96
16.34
15.42
15.23
10.88
11.26
8.75
9.69
13.57
12.30
13.91
11.87
11.41
13.98
10.13
9.83
12.70
10.68
10.51
8.94
7.92
12.36
15.29
14.51
14.36
15.46
13.87
13.74
12.60

0.90
1.02
1.05
1.08
0.91
0.79
0.75
0.76
0.76
1.24
1.20
1.15
0.75
0.81
0.96
1.23
1.48
1.30
1.60
0.88
1.33
1.83
0.89
0.83
1.54
1.19
0.89
1.01
0.99
1.34
1.20
1.15
1.21
1.20
1.04
0.88
0.96

108
200
149
159
137
103
117
128
107
125
134
118
97
102
165
111
114
112
140
118
121
130
111
99
153
100
89
110
96
130
120
108
105
133
101
116
126

60
757
516
652
640
407
355
213
178
468
1066
534
242
169
103
136
78
53
67
55
60
68
42
52
74
44
47
43
34
331
82
54
66
68
59
73
50

563
265
319
298
269
304
280
308
271
282
271
287
267
258
791
680
486
496
424
470
459
320
340
303
322
295
341
313
279
190
174
165
164
174
133
135
143

Cluster assignment of samples after multivariate statistical treatment of chemical data.

3.2.2. Morpho-mineralogical characterization by OM,


SEM-EDS and XRD analysis
All fragments from clusters A and B are characterized by
fine texture paste with remarkable amounts of iron oxides
(Fig. 4). The absence of large inclusions led us to assume
that refined clay was employed in their fabrication. In addition, an evident morphological and compositional continuity
between the red-coloured surface and the bulk was found
and black gloss is layered directly on the ceramic body.
The pastes of all fragments from cluster C, on the contrary,
are characterized by coarse textures with flakes, mainly of
muscovite, quartz, feldspar and pyroxene (Fig. 5). A complex
coating structure was revealed for all fragments. The SEM
images of thin sections of the fragments show a clay layer
with a texture finer than that of the ceramic body below and
comparable with that of the ceramic body of shards in clusters
A and B (Fig. 6). Where the black gloss is present at the

surface of the fragment, three different layers are visible in


section. The outermost layer exhibits the typical characteristics of the black gloss, including the thickness and the chemical composition. The intermediate layer shows transitional
characteristics between black gloss and ceramic body (texture,
sintering degree and chemistry), with a thickness of 40e
70 mm. The inner layer, corresponding to the ceramic body,
shows a non homogeneous structure and coarse texture.
ED spectra of the two layers (Fig. 7) show Al/Si and K/Ca
atomic ratios as well as the quantity of Fe increasing from the
ceramic body upward to the intermediate layer. The presence
of this intermediate layer, which we define as ingobbio rosso,
explains the differences in colours between the ceramic body
and the red decoration, observed in the fragments belonging
to cluster C.
The ingobbio rosso is only found on the external side in the
fragments of thinner wall objects, or on both sides in thicker

A. Mangone et al. / Journal of Archaeological Science 35 (2008) 1533e1541

1539

Fig. 3. Scores and loadings diagram for the first three principal components related to the objects examined. The accounted variance is 65% of the total variance.

ones. The thickness and the texture of the ingobbio rosso also
depend on the overall thickness of the walls, which are thinner
and less refined in smaller vases. This is very probably correlated to the firing technology; large vases, in fact, require large
furnaces where it is difficult to keep the temperature constant
for the proper amount of time. This could lead to surface lightening of thick ceramic bodies rich in calcium. To avoid this,
a thick, properly prepared engobe had to be utilized, to allow
the surface to keep its deep red colour after firing. In smaller
vases this problem is less evident, the quality of the engobe
plays a minor role, so its preparation could be less accurate.
As regards firing temperature, the presence of neo-formed
phases, such as pyroxenes and gehlenite, suggests that ceramic
bodies of all vases, irrespective of cluster, reached a temperature of about 950  C (Heimann and Maggetti, 1981). In addition, the presence of calcite in the ceramic body of the

monumental vases belonging to cluster C indicates either an


uneven temperature reached during firing or too short a firing
time.

Fig. 4. Crossed polars optical microscopy of the fabric of shard 10 showing


fine textured clay body.

Fig. 5. Crossed polars optical microscopy of the fabric of shard 6 showing


coarse textured clay body.

4. Conclusions
The results obtained by the different techniques on red figured pottery from Monte Sannace provide detailed information
about technological-productive aspects of Apulian red figured
pottery and lead to interesting results from both the archaeometric and archaeological points of view.
Our results show that the analysed findings differ in age
(5th and 4th centuries BC) and the later items (4th century
BC) differ in production technology. The marked compositional diversity of objects enclosed in cluster A, which justifies
the formation of a sharply distinct cluster from the other ones,

1540

A. Mangone et al. / Journal of Archaeological Science 35 (2008) 1533e1541

Fig. 6. SEMeBSE photomicrographs of thin sections of fragment 24 show overlapping areas characterised by different grey shades. Right: section corresponding
to the surface of the fragment where the black gloss is present. Three different layers are visible: black gloss layer (brighter), intermediate layer (light grey) and
ceramic body (dark grey). Left: section corresponding to red surface area. Two layers are visible: intermediate layer (light grey) and ceramic body (dark grey).

and furthermore the presence inside it of sample 5 led us to


exclude the local production of cluster A objects. They could
have been imported from Greek colonies on the coast of the
region or directly from Greece as hypothesized in previous archaeological studies (Scarf`, 1962). However, the comparison
between chemical compositional data in the literature (Jones,
1986; Mirti et al., 2006) and ours does not support an Attic
provenance.
As regards clusters B and C, the objects, stylistically classified by the archaeologists as local production, are characterized by two different ceramic bodies. Some are defined by
a fine texture, the same red colour as the figures with the black
gloss painted directly on it, others by a coarse texture and
a layer of ingobbio rosso, on the whole surface for large vases
and only on the external surface for smaller ones. These results
show that there were two different production technologies of
red figured vases at the Monte Sannace site during the 4th
century BC. Certain vases were produced with the classic

Attic technology, other vases with a different technology,


never previously reported in the literature, which used the
application of the engobe layer. The intentional red Attic
technique has been described (Farnsworth and Wisely, 1958;
Mirti et al., 2006; Tite et al., 1982); however, our results
(chemical composition, thickness, absence of birefringence)
are not consistent with this technique. They are consistent
with the use of fractions of the same clay for the ceramic
body and for the ingobbio rosso but with different granulometry. The ceramic body was very probably made with the
coarse clay fraction, usually discarded for Attic vases, while
the ingobbio rosso was prepared with the finer fraction, typically used for the ceramic body of the Attic vases. Since the
quality of the ceramic body is not visible from the outside,
this use of the clay can be considered smart, in that it allowed
the makers to obtain visual effects similar to those obtained
with the traditional technology, while saving relevant quantities of raw materials. This production evolution was very

Fig. 7. ED spectra of the two layers shown in Fig. 6 (intermediate layer and ceramic body) relative to the thin section of fragment 24. Left: ceramic body, right:
intermediate layer.

A. Mangone et al. / Journal of Archaeological Science 35 (2008) 1533e1541

probably started by the need for better material, from the


mechanical and structural point of view, to make larger vases
typical of the Apula production of the late 4th century BC.
These results add significant details to our knowledge on
the production technology of red figured Apulian pottery of
the 4th century BC in Monte Sannace. Furthermore, from
these promising results a research plan should soon start,
including samples from other sites. The extension of this investigation to other important archaeological sites in central
and northern Apulia will allow us to understand whether this
pottery technology evolved during the 4th century BC, as for
the findings at Monte Sannace, or if different technological
procedures were used at different sites. Such data could possibly indicate specific workshops related to certain sites or to
artists who were working at many sites, and offer solid pieces
of evaluation of one of the most important examples of pottery
handcraft production in Magna Grecia.
Acknowledgements
The authors would like to thank Dr C. Blomfield and
Dr S. Coultas (Kratos Analytical Limited, Manchester, UK),
for providing XPS analysis and fruitful discussion.
References
Bruno, P., Caselli, M., Curri, M.L., Genga, A., Striccoli, R., Traini, A., 2000.
Chemical characterisation of ancient pottery from south of Italy by Inductively Coupled Plasma Atomic Emission Spectroscopy (ICP-AES). Statistical multivariate analysis of data. Anal. Chim. Acta. 410, 193e202.
Canosa, M., Lacarbonara, F., Laviano, R., Lorenzoni, S., Acquafredda, P.,
1996. Archaeometric investigations of the ceramiche a figure rosse of
Darios painter (IV cent. BC) coming from Timmari. In: OrtegaHuertas, M., Lopez-Galindo, A., Palomo-Delgado, I. (Eds.), Advances in
Clay Minerals. Proceedings of the 1996 Spanish-Italian meeting on clay
minerals, pp. 278e280. Granada.
DellAnna, L., 1967. Ricerche su alcune terre rosse della regione pugliese.
Periodico di Mineralogia 36 (2), 539e592.
DellAnna, L., Laviano, R., 1987. Penisola Salentina: stato delle conoscenze
mineralogiche e geochimiche. In: Proceedings of the 1987 Convegno
Centro Studi Geotecnici dIngegneria. Quaderni Ricerche Centro Studi
Geotecnici dIngegneria, 11, Lecce, pp. 303e321.
DellAnna, L., Laviano, R., 1991. Mineralogical and chemical classification of
Pleistocene clays from the Lucanian basin (Southern Italy) for the use in
the Italian tile industry. Applied Clay Science 6, 233e243.
Farnsworth, M., Wisely, H., 1958. Fifth century intentional red glaze. Am.
J. Archaeol. 62, 165e173.

1541

Grave, P., Robinson, E., Barbetti, M., Yu, Z., Bailey, G., Bird, R., 1997. Analysis of South Italian pottery by PIXE-PIGME. Meditarch 9/10, 113e125.
Hatcher, H., Hedges, R.E.M., Pollard, A.M., 1980. Analysis of Hellenistic and
Roman fine pottery from Benghazi. Archaeometry 22 (2), 133e151.
Heimann, R.B., Maggetti, M., 1981. Experiments on simulated burial of calcareous terra sigillata (mineralogical change). Preliminary results. In:
Hughes, M.J. (Ed.), Scientific Studies in Ancient Ceramics, pp.
163e177. British Museum Occasional Paper 19, London.
Ingo, G.M., Bultrini, G., de Caro, T., Del Vais, C., 2000. Microchemical study
of the black gloss on red- and black-figured Attic vases. Surf. Interface
Anal. 30, 101e105.
Jones, R.E., 1986. Greek and Cypriot Pottery: A Review of Scientific Studies.
The British School at Athens, Athens.
Kingery, W.D., 1991. Attic pottery gloss technology. Archeomaterials 5,
47e54.
Maggetti, M., Galetti, G., Schwander, H., Picon, M., Wessicken, R., 1981.
Campanian pottery; the nature of the black coating. Archaeometry 23
(2), 199e207.
Maniatis, Y., Aloupi, E., Stalios, A.D., 1993. New evidence for the nature of
the Attic black gloss. Archaeometry 35 (1), 23e34.
Mirti, P., Casoli, A., Calzetti, L., 1996. Technology of production of fine pottery excavated on a western Greek site investigated by scanning electron
microscopy coupled with energy-dispersive X-ray detection. X-Ray Spectrom 25, 103e109.
Mirti, P., 2000. X-ray microanalysis discloses the secrets of ancient Greek and
Roman potters. X-Ray Spectrom 29, 63e72.
Mirti, P., Gulmini, M., Perardi, A., Davit, P., Elia, D., 2004. Technology of
production of red figure pottery from Attic and southern Italian workshops.
Anal. Bioanal. Chem. 380, 712e718.
Mirti, P., Perardi, A., Gulmini, M., Preacco, M.C., 2006. A scientific investigation on the provenance and technology of a black-figure amphora attributed to the Priam group. Archaeometry 48 (1), 31e43.
Noble, J.V., 1960. The technique of Attic vase-painting. Am. J. Archaeol. 64,
307e318.
Robinson, E.G.D., 1990. Workshops of Apulian red-figure outside Taranto. In:
Descoeudres, J.P. (Ed.), EUMOUSIA. Ceramic and Iconographic Studies
in Honour of Alexander Cambitoglou, vol. 1. Meditarch, Supplement,
pp. 181e196.
Scarf`, B.M., 1962. Notizie degli scavi di antichita`. In: Atti della Accademia
Nazionale dei Lincei, serie ottava, vol. 16, pp. 269e283. Roma.
Tang, C.C., MacLean, E.J., Roberts, M.A., Clarke, D.T., Pantos, E.,
Prag, A.J.N.W., 2001. The study of Attic black gloss sherds using synchrotron X-ray diffraction. J. Archaeol. Sci. 28, 1015e1024.
Tite, M.S., Bimson, M., Freestone, I.C., 1982. An examination of the high
gloss surface finishes on Greek Attic and Roman Samian wares. Archaeometry 24 (2), 117e126.
Trendall, A.D., 1989. Red Fig. Vases of South Italy and Sicily. A Handbook
(World of Art). Thames-Hudson, London.
Trendall, A.D., Cambitoglou, A., 1978. The Red-Figured Vases of Apulia. In:
Early and Middle Apulian, vol. 1. Oxford University Press, Oxford.
Trendall, A.D., Cambitoglou, A., 1982. The red-figured vases of Apulia, vol. 2.
Late Apulian, Oxford.

You might also like