Clay Analyses of Archaic Greek Pottery
Clay Analyses of Archaic Greek Pottery
Clay Analyses of Archaic Greek Pottery
http://journals.cambridge.org/ATH
The Annual of the British School at Athens / Volume 68 / November 1973, pp 267 - 283
DOI: 10.1017/S0068245400004470, Published online: 27 September 2013
T H E study of Greek pottery of the Archaic period has reached the point at which most groups
can be identified on stylistic grounds alone and confidently assigned to their place of manu-
facture, while in some groups works of individual studios, painters, and potters can be detected.
There remain, however, both many problem pieces whose origin is disputed or unknown, and
some major classes about which, for example, it is still possible to argue whether they were made
in Italy or Greece. This is a field in which the techniques of clay analysis might prove beneficent,
and this article presents the provisional results of the first of a series of tests designed to explore
this potential.
Analysis by optical emission spectrography has already been applied with some success to
problems of Bronze Age pottery in the Greek world. The work was undertaken by the Oxford
Research Laboratory for Archaeology and the History of Art, under the archaeological guidance
of Dr. Catling, and the results published in this journal and elsewhere.1 Further tests on Iron
Age Greek pottery in Italy were sponsored by Mr. Ridgway and Mr. Prag, the material being
chosen on the same general basis as that for the Bronze Age tests, by selection of series of sherds
from particular sites. The material for the tests described here has been selected according to
somewhat different principles, as will appear.
The process of analysis and its application have been conveniently described in BSA Iviii
(1963) 95—101.2 The presence of the nine trace elements (strictly speaking, their oxides) deemed
the most sensitive to local variations in clay composition is determined, and displayed graphi-
cally in diagrams of the type used to illustrate this article, which allow of easy comparisons.
Where a large group of similar composition is involved the results are presented in diagrams of
the same type as a series of range measurements for each element representing an 80 per cent
level of confidence. 'This means that in the long run 80 per cent, i.e. four out of five, sherds
belonging to the same population should have concentration within the calculated range' (ibid.
97). Otherwise individual tests are plotted. The results of earlier tests on Bronze Age and other
pottery in the Oxford laboratory have been used here, adjusted to correlate with the tests
conducted after the recalibration of instruments in 1971. The recalibration will be explained in
detail by Mr. Schweizer in a forthcoming number of Archaeometry.,3 Similar tests on other classes
of Archaic Greek pottery are being carried out in the laboratory of the Louvre Museum, under
the archaeological guidance of Mile Waiblinger. Mr. Noble reported briefly the results of some
tests in New York in a book published in 1965. Otherwise this is not a field which has been much
explored by these techniques, although there have been useful studies by others employing
other techniques of microscopic examination, which is likely to prove particularly effective for
coarser wares.4 Comparison of analyses by neutron activation and by optical emission spectro-
graphy have proved reassuring.5
Certain 'articles of faith' in the interpretation of these analyses deserve a moment's attention
in the light of the problems of Archaic wares. It might be thought that the clays from different
1
BSA Iviii (1963) 94 ff.; and articles in Archaeometry iv-xi Prag, F. Schweizer, L. R. Llewellyn, a n d j . Williams.
(1961-9). * J . V. Noble, The Techniques of Painted Attic Pottery (1965)
2
And see Archaeometry viii (1965) 16. 2 f., 90 f., 212; M. Farnsworth, AJA lxviii (1964) 221 ff.
3 5
'Hellenistic Glazed Ware from Athens and Southern I. Perlman and F. Asaro in Science and Archaeology (ed.
Italy: Analytical Techniques and Implication', by A. J.N.W. R. H. Brill, 1971), and Archaeometry xi (1969) 21 ff.
268 J. BOARDMAN AND F. SCHWEIZER
areas of one bed or from different beds in one area could have markedly different compositions.
So far, observation of material from Greece and Italy has shown that the variations are slight
and, within the elements chosen for inspection, negligible. Thus, quite extensive areas in south
Italy, in the Peloponnese, and in central Crete appear to use clays of consistent composition for
each area although readily distinguishable one from another. And in central Euboea tests on
material of four different periods from the Late Bronze Age to modern are consistent. It has
been demonstrated elsewhere, however, that a clay bed can yield significantly different compo-
sitions in different layers, and the observed consistency presumably indicates that in any one
period only one area was normally being worked, or that the beds happen to be fairly homo-
geneous. The problem becomes acute when, with a given sample compared with an established
range pattern, there is marked disparity in one element, or slight disparity in two or more. Only
a fuller study of 'permissible' variations or fuller knowledge of the composition of possible
alternative sources will resolve this.
There are, furthermore, clays from different areas which cannot as yet readily be distin-
guished at all (Euboea and south Italy is a sore example). It may be that the answer to such
impasses lies in the plotting of other trace elements. But it means that while almost total
confidence can be placed in negative identifications ('this is not Attic, Corinthian, etc.'),
positive identifications require the support of other criteria, archaeological or stylistic. This
brings us to the differences in selection and interpretation of the material presented here from
that for the Bronze Age tests. For the latter large numbers of sherds from chosen excavated sites
were submitted for analysis. From the results the dominant composition type was judged
probably local. The sites were generally those for which local production might be assumed, and
the pottery of a type for which no stylistic criteria had been discovered to determine place of
origin, or for which the criteria were in dispute. In the Archaic Greek period—we are thinking
principally now of the black-figure vases of the sixth century B.C.—the situation is very different
and the problems to be answered of a somewhat different character. They are either more
rigidly historical, or purely art-historical. Regardless of provenience it is usually possible to
identify an Athenian or Corinthian black-figure vase. The physical appearance of the clay, the
shape, the style of decoration often allowing even identification of painter, inscriptions—there
are many and sure criteria. The majority of surviving complete or near-complete Athenian
black-figure vases have not been found in Athens but in Italy. And on colonial foundations,
where the identity of pottery imported may determine important historical problems about
trade or the homeland of the colonists, it is common for most of the vases to be imported. To
take an example which has yielded some specimens analysed here: the Greek colony of Tocra
(Taucheira) in Cyrenaica has yielded hundreds of decorated vases and fragments belonging to
the first hundred years of its history—late seventh to late sixth century B.C.—from votive
deposits. Approximately 85 per cent was obviously imported, and the remainder, mainly plain
wares, presumed local.6 Of the imported, the majority could be identified by use of the standard
archaeological criteria of visual appearance of clay, shape, and decoration, but there are still
classes and individual pieces which are homeless. For these the only hope lies in analysis or in
the patient expectation that more finds in Greece will reveal their origin. Black spots in the
study of Archaic vases have been some major sites which, since they cannot be or have not yet
been extensively excavated, have offered no clear criteria for the identification of local pottery.
Notable examples are Megara, Sicyon, Miletus, Ephesus. Analysis of samples of local clay beds—
a study which has been underestimated hitherto, and can only be used in a limited fashion
here—could produce negative identifications for some homeless classes ('this is not Milesian');
6
For the statistics see J. Boardman and J. Hayes, Tocra ii (1973) 5.
CLAY ANALYSES OF ARCHAIC GREEK POTTERY 269
or plausible positive identifications ('this could be Milesian and stylistically it agrees with what
we would expect of Miletus'). The instance cited here, of Miletus, is one closely connected with
the problems of the identity of Greek pottery exported to Syria and Egypt in the Archaic
period (Al Mina and Naukratis), which is the subject of current tests to be reported at another
time.
The art-historical problems are only slightly different. To take a relevant example: during
the preparation of a Corpus Vasorum of Attic black-figure amphorae in Oxford one vase, which
had passed (albeit uneasily for some) as Attic, resisted attempts to place it exactly in the Attic
series. Analysis demonstrated that it was not Athenian (see below, p. 276). On the other hand
analysis has not yet demonstrated where it was made, and instead has presented an embarrass-
ing choice. Other classes of black figure have been in recent years suspected of being non-
Athenian, although hitherto classified as such. For these, or some examples of these, new homes
can now be assured. All this involves far more than a correction to a museum label, since it
makes it easier now to establish sure stylistic criteria for the identification of similar classes
found in historically valuable contexts (in colonies, etc.) and enriches our knowledge of the
iconography and art-history of 'non-metropolitan' cities in Archaic Greece.
While, therefore, the identification of the clay composition of vases made in specified areas
offers exactly the same problems and benefits for the Archaic period as it has for the Bronze
Age, in practice much more may depend on the analysis of and on the judgement of analysis of
individual pieces. Even within a major and apparently homogeneous group, study of individual
analyses and their grouping has proved rewarding, and might prove rewarding also in restudy
of Bronze Age wares. For the identity of'controls', too, stylistic criteria may be able to give us
greater confidence in the coherence of a small group being tested than identity of find place
could for a larger group.
What is urgently required is analysis of samples from known clay beds in Greece, Italy, and
other Greek colonial areas. Until this is done we shall have more negative identifications than
positive ones. The material presented here is a first stage, premature in some degree in the light
of what has just been said, and provisional in its conclusions, towards the establishing of in-
dependent criteria for the identification of Archaic wares. Further progress will depend on
concerted programmes of study in Greece and Italy, and collaboration between laboratories
conducting analyses with these techniques. Even at this early stage the prognosis is good.
The analyses published here have been made under the supervision of F. Schweizer at the
Oxford Laboratory with the assistance of Miss L. R. Llewellyn. Text and commentary are Mr.
Boardman's. We are grateful to the Trustees of the British Museum, the Keeper of the Depart-
ment of Antiquities in the Ashmolean Museum, the Curator of the Museum of Greek Archaeo-
logy, University of Reading (Mrs. Ure), and the Curator of Antiquities in the University
Museum, Manchester (Mr. Prag), for permitting us to use clay samples from their vases; and to
Dr. G. Buchner and Professor G. Vallet for providing sherds from Ischia and Megara Hyblaea.
Dr. Hayes brought specimens from the excavations at Tocra. Mr. Ridgway and Mr. Prag have
kindly allowed us to make use of the results of other analyses undertaken for them in Oxford. In
this account, where possible, an indication is given of the appearance of the wares to the eye or of
descriptions by others—a criterion, however unscientific, which has proved effective in a good
many cases and may remain so. The details of each analysis are listed at the end, and for one or
two vases which have now lost a home or hope to find a new one there is appended a separate
discussion in order not to interrupt the main account with iconography and art-history. The
subjects are tests on black figure in the Attic-Euboean-'Chalcidian' range, on Protocorinthian,
and on some suspected Cretan pottery from Tocra. Tests are continuing with especial reference
270 J. BOARDMAN AND F. SCHWEIZER
to the problems of 'Melian', the East Greek schools represented at Naukratis and well-known
groups of clay figures and figure vases.
NiO CrjO3 MnO TiO, MgO CaO Na,0 Fe2O3 AJjO NiO Cr2O3 MnO T i ^ MgO CaO Na2O Fe2O3 AI2O3 NiO Cr2O3 MnO TiOj MgO CaO Na2O Fe2O3 AI2O;
:1000 xlOO x100 xiO x10 X1000 x1OO xlOO X10 xiO xlOOO x100 xlOO x10 xlO
-
~i 1 1 1 1 1 1 1 r~ 1 i 1 1 1 1 1 r~
:
DIAGRAM I DIAGRAM II DIAGRAM
ATHENIAN B.F. \ ATHENIAN BLACK GLAZE
RANGE OF 12 \ (2c B.C.) •
. I RANG OF 20 -
f
w\
/
V •
1 1 1 1
inscriptions on the vases themselves. The following twelve pieces were chosen from the Oxford
collection to give a good range of sixth-century Athenian workshops:
(1) Oxford 509 (G. 268). Neck amphora by the Affecter. ABV 239, 5.
(2) Oxford 510 (G. 272). Neck amphora akin to the Group of Wiirzburg 199. ABV 290.
(3) Oxford 220 (1874.291: G. 210). Neck amphora, Light-make Class. Paralipomena 300.
(4) Oxford 214 (1885.656). Neck amphora of the Dot-band Class, from Caere. ABV 484, 10.
(5) Oxford 1913.164. Tyrrhenian amphora from Caere. ABV 100, 64.
(6) Oxford 1952.549. Panathenaic amphora fragment by the Eucharides Painter, from Al
Mina. ABV 396, 4.
(7) Oxford 1960.1290. Neck amphora. Early Lydan. Archaeological Reports for 1963-64, 53 f.
fig. 10.
(8) Oxford 1965.100. Belly amphora by the Lysippides Painter. ABV255, 10; Paralipomena 113.
(9) Oxford 1965.115. Neck amphora by the Antimenes Painter. ABV 269, 49; Paralipomena
118.
' J . V. Noble, The Technique of Painted Attic Pottery (1965) source by ancient writers; see G. M. A. Richter, The Craft
1-3. Cape Kolias (just SE of Phaleron) is mentioned as a of Athenian Pottery (1923) 97.
CLAY ANALYSES OF ARCHAIC GREEK POTTERY 271
(10) Oxford 1965.124. Belly amphora in the manner of the Princeton Painter. ABV 300, 6;
Paralipomena 130.
(11) Oxford 1965.125. Neck amphora near the Class of New York 96.9.9. Paralipomena 295.
(12) Oxford 1965.141. Belly amphora in the manner of the Princeton Painter. ABV 299, 1;
Paralipomena 130 f.
The composition range pattern of these analyses is shown in Diagram I. Plotting individual
vases produced two pairs with very close similarities: (3) and (12) which have very low figures
for Al, Fe, Ti, and Cr; and (5) and (6) with high figures for Fe, Mn, Cr, and Ni. There are no
date or workshop connections here, so this may indicate no more than local varieties in the clay
beds used.
Another series of Athenian pottery analyses, taking twenty samples of black-painted vases
from a second-century B.C. group found in the Athenian Agora, was undertaken for Mr. Prag
who kindly permits me to illustrate the composition range pattern in Diagram II. This corre-
sponds closely with Diagram I, but allows for higher readings in Fe, lower in Ca.
Diagrams I and II give a clear indication of the composition of clays used in the Athenian
potters' quarter in the Archaic and Hellenistic periods. It is always possible, however, that other
potteries in Attica used clay beds of different composition. The existence of 'countryside'
workshops may certainly be suspected in the sixth and fifth centuries, supplying vases, for
example, to Eleusis, Brauron, and Vari. This may be worth further investigation. For example,
Diagram III gives the corrected composition range pattern of eleven Bronze Age fragments
from the site at Perati in East Attica, which were studied some years ago.8 The high Ca readings
in particular show a marked difference from the Athenian of Diagrams I and II, but we cannot
be sure that the Perati pottery was made locally.
'GHALCIDIAN'
'Chalcidian' vases are the most distinguished sixth-century black-figure series after the
Athenian. Virtually all have been found in Italy and Sicily, and are named for the 'Chalcidian'
forms of the inscriptions upon them. The clay is orange to brown, less ruddy than Attic. Current
opinion favours an origin in the south Italian colony Rhegion (Reggio) founded by Chalcis.9
The style is distinctive, but owes much to Corinth, Athens, and Ionia, and in its later years is
strongly Atticizing. Six 'Chalcidian' vases in Oxford were analysed:
(13) Oxford 1950.6. Neck amphora by the Phineus Painter. Ashmolean Museum Visitors Reportfor
ig50 pi. 2.
(14) Oxford 191. Neck amphora from Caere (?), Group of the Vienna Amphora. A. Rumpf,
Chalkidische Vasen (1927) 18 no. 32, pis. 66, 84.
(15) Oxford 1929.29. Neck amphora. Group of the Vienna Amphora. H. R. W. Smith, The
Origin of Chalcidian Ware (1932) pi. 19; Ashmolean Museum, Beazley Gifts igi2-66pl. 9.103.
(16) Oxford 192 (1885.652). Neck amphora from Caere. Group of the Orvieto Hydria. Rumpf
21 no. 50 pis. 84, 91.
(17) Oxford 1929.403. Hydria. Group of the Orvieto Hydria. Smith pi. 17.
(18) Oxford 1965.132. Neck amphora. Group of the Orvieto Hydria. Rumpf 18 no. 26 pis. 52-4.
The composition range pattern for these 'Chalcidian' black-figure vases is shown in Diagram
8
BSA lviii (1963) 103ff.no. 5, Group C. a brief statement of the problem, admitting Etruria as a pos-
9
See R. M. Cook, Greek Painted Pottery (1972) 158 for sible alternative source.
272 J. BOARDMAN AND F. SCHWEIZER
IV. For Cr three readings were of < 0-026; the three others were 0-034, °'°38; a n d 0-052, so
a considerable range in this oxide must be admitted.
There are also a number of vases related to the 'Chalcidian' in style and shape which are
generally called pseudo-'Chalcidian'. Four examples in London and one in Reading were
examined:
(19) London B 16. Neck amphora. Memnon Group. Rumpf 156, c, pi. 198.
(20) London B 17. Neck amphora. Memnon Group. Rumpf 156, d, pi. 199.
(21) London B 154. Neck amphora from Vulci. Polyphemos Group. Rumpf 161, VI, pis. 202,
203.
(22) London 1949.2-17.1. Neck amphora. Polyphemos Group. BMQxvi (1951) pi. 27.
(23) Reading University Museum 130.51 RM. Oinochoe. Archaeological Reports for 1962-63,
57 f. fig. 67, where Professor Banti's observation of the resemblance to figures on a vase of
the Polyphemos Group is recorded (also in EAA s.v. 'Calcidesi, vasi' as an amphora). Its
shape also resembles oinochoai of this Group (as Rumpf pis. 221, 222) more closely than
the 'Chalcidian'.
The results are plotted individually in Diagrams V and VI and in VII all are superimposed
on the 'Chalcidian' range. Over all a comparison with the 'Chalcidian' results shows that
pseudo-'Chalcidian' gives uniformly lower readings for Mg and Ti and generally higher ones
for Na. (23) differs from the other pseudo-'Chalcidian' in respects which, for Cr, Ca, Na, and
Ni, are very close to the 'Chalcidian' (16), but its place with the pseudo-'Chalcidian' seems
assured. Although the analyses are few there seem grounds for believing that the two series used
different, though not very different, clay beds.
Once the clay beds from all the areas which might have produced 'Chalcidian' vases have
been analysed, the problem of their origin should be solved. Rumpf thought that they were
made in homeland Chalcis, in Euboea, and we shall have later to consider what analysis can
and cannot say on that score. Dr. Walter-Karydi has more recently suggested that they are
successors to the Archaic 'Melian' vases,10 but analysis of these vases (to be published later) does
not bear this out, nor does analysis of clay and vases from Melos itself, or of pottery from Paros,
one of the proposed sources for 'Melian'. (The reader unfamiliar with Archaic Greek vases will
have become aware already of important areas of uncertainty in their study!) The local clays of
Rhegion have yet to be examined, but analyses of fifth- and fourth-century local wares from
other areas of south Italy (Apulia), to be published by Mr. Prag, give results comparable with
those for 'Chalcidian' in all except Ca, where 'Chalcidian' is generally lower. Etruria is the
other popularly supported source and local Etruscan clays have also yet to be examined. It can
at least be said that 'Chalcidian' and Pontic (a sixth-century ware certainly made in Etruria)
seem dissimilar ('Chalcidian' has far more Mg, rather less Ca).
Rumpf had affirmed a visual fabric difference between 'Chalcidian' and pseudo-'Chalcidian',11
denied by Vallet.12 Analysis suggests a difference hardly greater than there seems to be in the
style, but the geographical significance of this is hard to assess as yet, as we shall see. The minor
respects in which pseudo-'Chalcidian' differs from 'Chalcidian' bring it close to the composition
of some eighth-century vases from Veii, and Etruria has been suggested as a source for pseudo-
'Chalcidian'. But the problems of the origin of the Veii pottery (analysed for Mr. Ridgway)
cannot be discussed here. This is another area for closer study.
10
CVA Munich vi. 23 f. " In Chalkidische Vasen (1927) 167 f.
12
REA 1956, 46 f. and, on the whole problem, in Rhegion et £amle (1958) 211 ff.
CLAY ANALYSES OF ARCHAIC GREEK POTTERY 273
EUBOEAN
An Atticizing school of black figure has been identified at Eretria, on the island of Euboea,
from a number of vases of distinctive shape and decoration found at the site.13 Their clay is
pink-red, slightly paler than Attica, occasionally even yellowish, without mica. Other vases,
from other sites or without known provenience, have been attributed to Eretria or more
generally Euboea,14 including a number which have long passed as Athenian, and which have
appeared in Beazley's lists as Attic. Mrs. Ure makes another important contribution to this
study in another article in this volume of BSA (pp. 25ff.),citing several of the pieces mentioned
below. There is a happy congruence here of results from stylistic and chemical analysis.
The black-figure vases from Eretria itself are not available for analysis, and it might seem
improper to rely on those attributed to Eretria on grounds of style for an identification of
Eretrian clay. There are, however, other sources of information for the clay used at Eretria and
at the other great Euboean city, Chalcis, which is only some 20 kilometres away to the west.
Between the cities lies the Lelantine plain and midway, on the coast, is the important Bronze
Age and Early Iron Age site of Lefkandi beside which there are rich clay beds. A Bronze Age
site at Amarynthos, east of Eretria, has also been explored.15 The following samples from the
sites named have already been examined:
(a) Lefkandi. Two samples from the modern clay beds.16 Diagram VIII (corrected).
(b) Amarynthos. Of sixteen Bronze Age sherds tested17 eleven are of the uniform composition
indicated in Diagram IX (corrected).
(c) Chalcis. Twelve Late Geometric sherds from the ancient site at Chalcis were tested for
Mr. Ridgway.18 Diagram X (corrected).
Diagrams V I I I - X can be seen to correspond closely except for the slightly higher readings in
Ti for the two samples from Lefkandi clay beds, which may not be significant. They can be
taken to indicate clearly the composition of Euboean clays in the Lelantine plain area.
Before the analyses of problem pieces are presented we should compare the three main groups
so far distinguished—the Athenian, 'Chalcidian', and Euboean. The Athenian is triumphantly
distinctive. Between the 'Chalcidian' and Euboean there is less to choose, however. The
'Chalcidian' has consistently higher readings for Mg, but we may recall that the lower Mg was
also a feature of the pseudo-'Chalcidian'. In fact the Euboean and pseudo-'Chalcidian' are
found to correspond closely except for the generally slightly lower readings in Ti for the latter.
From the analyses so far made it is not possible to say categorically that 'Chalcidian' vases
could not have been made in Euboea, but the detectable, though slight, differences, and the
apparently strong archaeological and stylistic arguments in favour of looking for the homeland
of both 'Chalcidian' and pseudo-'Chalcidian' in the west, make the proposition most unlikely.
We are faced, however, with the fact that on present evidence, which may be inadequate
either on grounds of numbers tested or of elements recorded, it would be unwise to attempt to
distinguish between Euboean black-figure products of east and west on grounds of analysis
alone.
Study of the composition of local wares of Apulia of rather later date suggests that the clays of
13
BSA xlvii (1952) 30if. (Lefkandi), 64-6 (Amarynthos); M. Popham and H.
•• Articles by D. A. Amyx, AJA xlv (1941) 64 ff.; A. D. Sackett, Excavations at Lefkandi (1968).
l6
Ure, JHS lxxx (i960) 160 ff., lxxxii (1962) 138 ff., lxxxviii Details in Archaeometry xi (1969) 8 no. 32.
(1968) 140 f., BICS vi (1959) 1 ff., xii (1965) 22 ff.; by the " Details in BSA lviii (1963) 103 ff. no. 7a, Amarynthos
writer in BSA Hi (1957) 18 ff.; D. von Bothmer, Metr. Mus. I, Group D.
l8
Journal ii (1959) 27 ff. The figures given in Archaeometry xi (1969) 8 are from
15
For the Bronze Age sites see BSA lxi (1966) 60 f. a wider choice of sherds.
D384 T
274 J. BOARDMAN AND F. SCHWEIZER
NiO Cr,O 3 MnO TiO 2 MgO CaO Na2O Fe2O3 AI2O3 NiO Cr 2 O 3 MnO TiO 2 MgO CaO Na2O Fe2O3 AI2O3 NiO Cr2O3 MnO TiO 2 MgO CaO Na2O Fe2O3 AI2O3
xiOOO x1OO xlOO x1O xiO xlOOO xlOO xlOO xlO xlO x1000 xlOO xlOO xlO xiO
/A ^ « \i //
i,\ \\ f\ if
-\
V t i
NiO Cr2O3 MnO TiO 2 MgO CaO Na2O Fe2O3 AI 2 O 3 NiO Cr2O3 MnO TQ MgO CaO Na2O Fe2O3 AI2O3 NiO Cr2O3 MnO TiO2 MgO CaO Na2O Fe2O3 AI2O3
1OOO xlOO x100 xlO xlO xlOOO XlOO x1OO xlO xTO xlOOO xlOO x100 xlO XlO
~I 1 1 1 1 1 1 1
'A
W
V—I- V—L
.NiO Cr 2 O 3 MnO TfO 2 MgO CaO Na2O Fe2O3 AI 2 O 3 NiO Cr 2 O 3 MnO TiO 2 MgO CaO Na2O Fe2O3 AI 2 O 3 NiO Cr2O3 MnO TiO 2 MgO CaO Na2O Fe2O3 AI2O3
xiQOO x1QO xlOO x1O x1O XlOOO xlOO xlOO x10 x10 xlOOO x100 xlOO xlO X10
K$ \/J :
:
V
-
^Ul 1 1 1
CLAY ANALYSES OF ARCHAIC GREEK POTTERY 275
NiO Cr2O3 MnO TiO 2 MgO CaO Na2O Fe2O3 AI2O3 NiO Cr 2 O 3 MnO TiO 2 MgO CaO Na2O Fe2O3 AI2O3 N:O Cr 2 O 3 MnO TiO, MgO CaO Na.O Fo,0, Al-O,
:1000 xlOO xlOO xlO _ XlO X1OOO x1OO xlOO x1O x1O xlOOO x100 xlOO xlO xib
D I A G R A M XIII DIAGRAM XV
f3l) EUBOEAN BF.
(32)
(33) R A N G E OF 7
\
\
A //
-\—
/ , \ //
—-//\V/--H
— . ' • -
\
; \
\
\ // V
/
w / //
/ /
\, /
NiO Cr2O3 MnO TiO 2 MgO CaO Na2O Fe2O3 AI2O3 NiO O 2 0 3 MnO TiO 2 MgO CaO Na2O Fe2O3 NiO Cr r 0j MnO TiO MgO CaO too Fe ; 0j AI;O
xiOOO xlOO xlOO xlO xlO x1000 xlOO xlOO xlO xlO iOOO r.:OQ xlOO xlO xlb
A\ /A V /
NiO Cr2O3 MnO TiO 2 MgO CaO Na2O Fe2O NiO Cr2O3 MnO TiO? MgO CaO Na2O F e ^ AI2OS NiO Cr t O, MnO TiO2 MgO CaO Na ; O FH^O. AI.O.
xiOOO xlOO xlOO xlO xlO X I O O O xlOO x1OO xlO x10^ »1OOO *100 xiQO xlO x10
TABLE OF ANALYSES
shields and spears. The fighters have swords slung from a white baldric and with white chapes.
The one on the left wears a bell corslet over a short chiton, the one on the right a short red
chiton. His shield device is a white lyre with plektron, with a border of dots, and that of the
fallen man a snake. The interior of the other warrior's shield is painted red, with a fine white
porpax and the antilabe attached very close to the rim.
On B three maenads dance. All wear chitons, but the sleeves of the third are tight and short,
the upper part of her dress is painted red, and the skirt covered with red blobs. While the others
have projecting fillets on their heads, she wears her hair bound up at the back. All have incised
necklets and triple-dot ear-rings. All play clappers—the central pair red. The outer two wear
animal skins, one red, over their chitons. Their eyes have black pupils.
The vase is not Attic—this much the eye can suspect and analysis has confirmed. But the
composition of its clay is not decisive between pseudo-'Chalcidian' and Euboean, so we have to
turn to stylistic criteria, with an eye also to the 'Chalcidian' and Attic features which may have
inspired the work, and, for good measure, to Boeotia.
First, its neck and handles. The palmette-lotus cross with spiral corners is quite unfamiliar in
this or any other position in 'Chalcidian' or pseudo-'Chalcidian'. It is commonplace in Attic,
notably on Tyrrhenian vases or the work of artists like the Camtar Painter, of the second quarter
and middle of the sixth century, on amphora necks or in animal friezes. In Euboea it is borrowed
for the Eretrian Wedding Amphora, although in a different, unincised technique, and for the
lekane, Reading 56.6.8, whose Euboean origin was suspected on grounds of style and confirmed
by analysis (27).30 On Boeotian black-figure lekanai the compositions are different again, and
spirals rare as corner pieces, but the general type is known.31 We cannot be sure that the upper
corners of the complex on the Oxford vase were not occupied by flowers rather than spirals.
The lotus has a central leaf or spike. This is normal in Attic down to about 550, already being
omitted often by that date. It is normal on the Boeotian lekanai, which are dated in the third
quarter of the century. It is never met in 'Chalcidian', but in pseudo-'Chalcidian' it can be
found twice in the Polyphemos Group.32
The handles are triple-reeded. This construction has a long history in east Greece but appears
for neck amphorae in Athens little before the mid century on some Tyrrhenian vases (double-
reeded) and soon becomes normal. So by Attic standards this combination of handle and neck
pattern suggests a date of about 550. Reeded handles for amphorae are unknown in 'Chalcidian'
or pseudo-'Chalcidian', where the neck amphora shape is common, but are found on an attri-
buted Euboean amphora (in the Louvre). The handles run down the sides of the neck on to the
shoulder. This is an unusual treatment, although standard Attic neck amphorae of the second
half of the century come close to this, especially the Affecter's.33 The handles of a Boston neck
amphora, taken for Eretrian, are also like this.34
We turn to the other florals. A floral cross or fork attached to the handle is not seen on
'Chalcidian' or pseudo-'Chalcidian' vases, where any floral, even unattached, is very rare in
this position." The handle scheme is commonplace in Attic, but with palmettes only, not flowers.
Independent of the handles the flower terminals are, of course, common in 'Chalcidian' and on
the Boeotian lekanai.
The flowers in the handle complex have the bulging calyx haft characteristic of 'Chalcidian'
30
Pictures of these features conveniently in BICS xii ii pis. 61.3, 4; 62.
33
(1965) pi. 3.1-3. e.g. CVA Munich vii pis. 332, 336.
31 3
Metr. Mus. Stud, iv (1932) 25fig.9, where the centre * Metr. Mus. Journal ii (1969) 30fig.5.
bottom member is a palmette, not a lotus, and the upper 35 Unattached floral cross on Rumpf, pi. 46 (22).
corners have flowers. Vertical pendent florals of a quite different type, ibid. pis.
32
Rumpf, pis. 204-5 (Vatican 223); and CVA Adolphseck 52 (35), 84 (50) and in pseudo-'Chalcidian', pi. 212 (XI).
282 J. BOARDMAN AND F. SCHWEIZER
but generally ignored in Attic, where there is often found no more than a pair of incised lines
beneath flowers and buds—exactly the scheme of the floral band lower down on the Oxford
vase.36 The double incision, without the bulge, is seen on the Boeotian lekanai;37 and with the
bulge (as under the handles of the Oxford vase), on the amphora in Boston, attributed to
Eretria and already mentioned for its handles.38 The colourful floral band on the belly of our
vase is a type current on Attic vases through the second quarter of the century, but no later; not
on 'Chalcidian' but common on pseudo-'Chalcidian' of the Polyphemos Group where, however,
the centrepiece is drop-shaped, not a wedge.39
The horizontal, concave top to the foot is odd. Most 'Chalcidian' amphorae have a steeper or
convex profile, but some poorer late examples approach this.40
Finally, the figure scenes. The pseudo-'Chalcidian' vases follow this scheme of a deep frieze
below tongues, as do very many Attic, and only the main 'Chalcidian' series prefers a separate
shoulder frieze with figures or florals. Our artist cannot be faulted for composition. On one side,
the precise symmetry of the fight, with the geometrical balance of the three shields.41 On the
other, the three dancers whose differing actions and glances neatly interlock to fill the field.
Beside them the usual Attic or 'Chalcidian'42 dancers look positively arthritic. On 'Chalcidian'
the dress for maenads is quite different and on pseudo-'Chalcidian' they wear peploi.43 For dress
and manner they are far closer to the Attic of the third quarter of the century, and given the
other comparanda with Attic already observed, it is difficult to date the vase much after the
Where was it painted? Clearly not in the Attic, Boeotian, 'Chalcidian', pseudo-'Chalcidian'
(the Polyphemos Group is apparently the work of one man), 44 or Eretrian workshops as we
know them. Our mainland comparisons are far more numerous and more telling than the
western 'Chalcidian' or pseudo-'Chalcidian'. Clay composition hints at Euboea. I would have
expected a vase painter at Chalcis in Euboea to display just this degree of deviation from Attic,
with echoes of styles practised in Eretria and Boeotia, and nothing of Corinth. One of the ones
who stayed at home—whose further work or influence was either cut short or has yet to be
discovered?
36 For this feature in 'Chalcidian' and Attic see H. R. W. It is generally agreed that the Polyphemos Group is more
S m i t h , The Origin of Chalcidian Ware g i . strongly Atticizing than the main 'Chalcidian' series, yet
37
Metr. Mus. Stud, iv (1932) 27 fig. 16, 31fig.2 1 . closely linked to it, old-fashioned, and comparatively late
38
Above, n. 34. (last third of the sixth century). If it is so late (and I am not
39
e.g. R u m p f pis. 211-19. wholly convinced—Rumpf suggested c. 530) I do not see
40
As R u m p f pis. 60 (30), 67 (33), 68 (34), 139 (151). how it could exhibit long-out-of-date Attic features (the
41
The fallen warrior on his hands and knees is a little centre leaf or spike and patterned bands on the lotuses)
unusual but can be matched in 'Chalcidian': Rumpf, pi. 7 unless it was either made in an area of mainland Greece
(3) for Eurytion, and cf. pis. 115 (65), 218 (XVII); and in where such features lingered, as they did in Boeotia and at
Attic, e.g., for another Eurytion, ABV 133, 8 (D. M. Eretria, or by an emigrant (presumably to Italy) from that
Buitron, Attic Vase Painting in New England Collections (1972) area (presumably Chalcis) at a date appreciably later than
28 f. n o . 10). that of the inception of the main 'Chalcidian' series. Either
42 circumstance would explain some similarities to the Oxford
As Rumpf, pis. 27-30 (Kunstwerke der Antike, Coll.
K a p p e l i , D 23) a n d K . Schefold, Meisterwerke n o . 153. vase. The earliest of the Polyphemos Group would be the
« Rumpf pis. 206 (III), 216 (XV). Vatican (Rumpf II) and Adolphseck (above) vases, but even
44 if they are brought nearer the mid century their connection
Some examples published since Rumpf:
P. E. Corbett, BMQ_x\r'\ (1951) 74-6 pi. 27. with the rest of the series is clear, since similar archaizing
CVA Adolphseck ii pis. 61.3, 4; 62. features persist (the colourful florals, the animals with the
D. von Bothmer, Metr. Mus. Bull, iv (1947) 131 ff. (New old Attic double shoulder line) and Rumpf's arguments for
York 46.11.5). a single hand remain unshaken. All this indicates an artist
Miinzen und Medaillen Auktion xi no. 315. who did not take much notice of the styles affected by the
G. Vallet, REA 1956, 42 ff. (Louvre Campana 10498, 'Chalcidian' series in the 530s; perhaps a man who learned
10532). something, but not enough, from the artist of the Melchett
Reading University, see above, analysis (23). amphora.
CLAY ANALYSES OF ARCHAIC GREEK POTTERY 283
45
M o s t fully b y G . v a n H o o r n i n Choes and Anthesteria Dionysos could impersonate satyrs in the Anthesteria,
(1951). But dissociated b y R u m p f in Bonner Jahrbiicher clxi accompanying the god on his ship car through the streets of
(1961) 208-14. Athens; see J. Boardman, JHS lxxviii (1958) 6 f.
46 50
J . R . Green, AA 1970. 475 ff. H . K e n n e r , Das Phdnomen der verkehrten Welt in der
47
M . Crosby, Hesperia xxiv (1955) 76 ff. griechisch-romischen Antike (1970) 112.
48 51
L. Deubner, Attische Feste 142-7 a n d in Abhandl. Preuss. Oxford 1966.1007, Ashmolean Museum, Beazley Gifts
Akad. Wiss. 1943.12, ' D a s attische Weinlesefest'. igi2-66p\. 8.95, as Boeotian.
49 s2
I n general see F . Brommer, Satyrspiele (1959) a n d Athens 1721; van Hoorn, op. cit. 68 no. 68; P. Wolters
A. W. Pickard-Cambridge, The Dramatic Festivals of Athens and G. Bruns, Das Kabironheiligtum bei Theben i (1940) 126
(1968). In the later sixth century it seems that priests of pi. 61.6; S. Karouzou, AJA 1 (1946) 130 n. 48.
td
C/2
ON
00
(*)
CLAY ANALYSES OF ARCHAIC GREEK POTTERY
Oxford 1960.741 >
H
w
Ui
PLATE 54
B.S.A.
(*)
CLAY ANALYSES OF ARCHAIC GREEK POTTERY
Oxford 1960.741
B.S.A. 68 PLATE 55