The Early Cycladic Period: Matters of Definition and Terminology Author(s): R. L. N. Barber and J. A.

MacGillivray Reviewed work(s): Source: American Journal of Archaeology, Vol. 84, No. 2 (Apr., 1980), pp. 141-157 Published by: Archaeological Institute of America Stable URL: http://www.jstor.org/stable/504262 . Accessed: 08/03/2012 09:57
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Archaeological Institute of America is collaborating with JSTOR to digitize, preserve and extend access to American Journal of Archaeology.

http://www.jstor.org

The

Early

Cycladic
and

Period':

Matters

of

Definition

Terminology

R.L.N. BARBERand J.A. MAcGILLIVRAY

Abstract Two essentially different terminological systems (as well as other variations upon these) are at present in use for description of the Early Cycladic period. These are discussed against the background of a survey of the material evidence for the EC period and of some general observations on terminology and approach. Since evidence for genuine cultural distinctions
1 This article is one result of researchto which the assistance and generosity of many scholars have greatly contributed. Among these, we would mention in particular Professor P. Amandry, Professor J.L. Caskey, Professor J.N. Coldstream, Mr. J. Mellaart,ProfessorA.C. Renfrew and Mrs. Ph. Zapheiropoulou. An earlier draft was read by Professor J.L. Caskey, Professor J.E. Coleman, Mr. M.S.F. Hood, Professorand Mrs. J. Overbeck and ProfessorP.M. Warren. We are extremely grateful for their comments and criticisms, although responsibilityfor the opinions expressed rests with the writers alone. The following abbreviationsare used in addition to those listed in AIA 82 (1978) 3-10: C. Doumas, "Notes on Early Cycladic ArchiArchitecture tecture,"AA 1972, 151-70. Art and Culture J. Thimme, ed., Art and Culture of the Cyclades (Karlsruhe 1977). C. Doumas, Early Bronze Age Burial Habits Burial Habits in the Cyclades (SIMA 48, G6teborg 1977). A. Plassart, Les Sanctuaires et les Cultes du Delos XI Mont Cynthe (Paris 1928). EH Argolid J.L. Caskey, "The Early Helladic Period in the Argolid," Hesperia 29 (1960) 285-303. A.C. Renfrew, The Emergence of Civilisation: Emergence The Cyclades and the Aegean in the Third Millennium B.C. (London 1972). Interconnections J.E. Coleman, "The Chronologyand Interconnections of the Cycladic Islands in the Neolithic and Early Bronze Age," AJA 78 (1974) 333-44. Kastri E.M. Bossert, "Kastri auf Syros," Deltion 22 (1967) Cols. 53-76. J.E. Coleman, Keos 1: Kephala (Princeton Kephala 1977). Keos Pt. I J.L. Caskey, "Investigationsin Keos. Part I: Excavations and Explorations, 1966-70," Hesperia 40 (1971) 359-96. Keos Pt. II J.L. Caskey, "Investigationsin Keos. Part II: A Conspectus of the Pottery," Hesperia 41 (1972) 357-401. C. Tsountas, "Kykladika," ArchEph 1898, KykladikaI
137-212.

within the EC period is lacking (with one possible exception), we see no justification for adopting the new "culture" terminology. The traditional system (EC I, II, III), with slight modification, remains the most appropriate and allows a much more constructive and illuminating historical approach to the material. The last twenty years have seen a degree of activity in the field of Cycladic prehistory unmatched
KykladikaII MB Cyclades C. Tsountas, "Kykladika,"ArchEph 1899, 73134. R.L.N. Barber, "The Middle Bronze Age in the Cyclades," in C. Doumas ed., Thera and the Aegean World I (London 1978) 367-79. R.J. Howell, "The Origins of the Middle Helladic Culture," in R.A. Crossland and A. Birchall eds., Bronze Age Migrations in the Aegean (London 1973) 73-1o6. O. Rubensohn, "Die priihistorischen und friihgeschichtlichenFunde auf dem Burghiigel von Paros,"AthMitt 42 (1917) 1-98. T.D. Atkinson and others, Excavations at Phylakopi in Melos (London 1904). J.L. Caskey, "Aegean Terminologies," Historia 27 (1978) 488-91. C. Zervos, L'art des Cyclades (Paris 1957).

Origins

Paros

Phylakopi Terminologies Zervos

Notes to Illustrations.A wide range of Early Cycladicmaterial has been illustratedin publicationsand is thus available for use in conjunction with this text. For this reason, we present only summary figures of EC pottery forms and decorative motifs (ills. 2 and 3). Among the various sources, the schematic drawings provided by Doumas in Burial Habits are the most useful, because the range of material shown is wide and the objects are presented in chronologicalgroups (figs. 3-7: EC I [6 and 7 are "Kampos group"]; figs. 8-io, 12: EC II; fig. II: EC IIIA; fig. 13: EC IIIB). Zervos has good photographsof many objects but they are not assigned to periods. Art and Cultureis cited below. Pottery: Doumas illustratesthe greatest range of shapes in Burial Habits (see above). Also useful are Emergence (figs. Io. 1-4: EC I; figs. II. I and 20. 4, no. 5: EC II; figs. iI. 2 and 20. 4, nos. 1-4: EC IIIA; figs. 12. 1-2: EC IIIB); Kastri (figs. 3-5: all EC IIIA); Keos Pt. II (figs. 3-4: EC II; figs. 6-7: EC IIIA). Coleman's discussion of the pottery in Art and Culture 109-I7 is helpful but illustrationsare confined to vases in the exhibition concerned. Decorative motifs are discussed and illustrated in Art and Culture 129-41 and figs. 113-32. The illustrationsare weak for EC III, for which see our text and Phylakopi pls. IV-V, VII-X (the status of the pottery on pls. XI-XIII is uncertain but some

142

BARBER AND MAcGILLIVRAY

[AJA 84

sincethe timeof Tsountas.2 Excavation surface and have produceda substantialbody of exploration new materialand a numberof importantstudies and reports have appeared.3 naturalconsequence A of these developments been a reconsideration has of the archaeological sequencein the Cycladesin the EarlyBronzeAge and of the termsused to describeit (Table I). Below, we summarize the essential material of characteristics the successive phases of the EC period.On this basis,we then discussour reasons for preferring retainthe standard to terminological system,with one modification-the subdivisionof the final stageinto EC IIIA and EC IIIB.4
PRELIMINARY REMARKS

Terminology has become something of a problem in Aegean Bronze Age studies in recent years. Difficulties and possible solutions have been discussed by Caskey5and Renfrew.' We offer the following observations:
is probablyEC IIIB). Marble vessels: Art and Culture 95-1o8 and figs. 83-88. Marble figurines: The typological series proposed by Renfrew is
illustrated in Emergence fig. 11. 8. See also infra ns. 128, 132 and Art and Culture 59-91.

i. The most important factor governing the choice of terminology is the degree of convenience with which it can be used. In practice, the convenience of students is usually best served by the retention of an established system, especially one which has been carefully worked out and continuously modified over a long period; 2. The introduction of a new terminology does not of itself solve outstanding archaeological problems. It may in fact make their resolution that much more difficult by rendering the whole body of literature less comprehensible; 3. It is not possible entirely to separatequestions of cultural characterfrom those of chronology. The main reason for this is that pottery is both the most satisfactory chronological indicator and, at the same time, a primary feature of cultural identity; 4. The identification of a "culture" is never so simple a process as Childe's often-quoted phrase "a constantly recurring assemblage of
(1974) fig. I, 65 (MMII8). ILL. 3 Early Cycladic decorativemotifs. The comments above on ill. 2 apply also here. The sources from which the drawings are adapted are as follows: I. Deltion 17 (1961-62), pl. 44; 2. Zervos fig. 62; 3. Zervos fig. 84; 4. E.J. Forsdyke, PrehistoricAegean Pottery (Catalogue of Vases in the British Museum, Vol. i, Part I, London 1925) 54, A 302; 5. Zervos fig. 225; 6. Art and Culture 353, no. 4o1; 7. Zervos fig. 224; 8. Zervos fig. 225; 9. Emergence pl. 5, no. 3; io. Zervos fig. 152; II. Zervos fig. 152; 12. Zervos fig. 234; 13. Zervos fig. 220; 14. Zervos fig. 220; 15. Zervos fig. 204; 16. Zervos fig. 211; 17. Zervos fig. 236; 18. Emergence pl. 9, no. 3; 19. Kastri fig. 5, no. I; 20. Zervos fig. 126; 21. Phylakopi pl. 4, no. 6; 22. Phylakopi pl. 4, no. 5; 23. Phylakopi pl. 9, no. 2; 24. Phylakopi pl. 4, no. 7; 25a. Phylakopi pl. 5, no. 15; b. Phylakopi pl. 5, no. 8b; c. Phylakopi pl. 5, no. 8c; 26a. Phylakopi pl. 9, no. 9; b. Phylakopi pl. 8, no. 9. 2 For a review of archaeologicalinvestigations of the EC period, see Art and Culture 185-91. 3 The most important works of substance are Art and Culture, Burial Habits and Emergence. Brief but very useful are CAH' (infra n. 84) and Interconnections.Zervos has a good range of photographs. Summaries of the EC sequence which supplement our own are to be found in Burial Habits (11-27) and Interconnections (340-43). Discussions of figurines (infra ns. 128, 132) and bronzes (infra n. 129) are cited below. 4 This is the system used in discussion throughout the present article, except where the context requires otherwise. The relationships of the different systems are shown in Table I. 5 Terminologies. 6 Emergence 53-55.

Bronzes etc.: See above ns. 129, 136 and Art and Culture
117-22. Jewellery etc.: Art and Culture 123-29 and figs. 95-1o8.

ILL. I Map showing the location of sites mentioned in the text. The periods of occupation representedat these sites are indicated in Table II. Readers should note that the sites have been selected within the context of this article and that many others are also known (see ns. 20, 46). A complete list would undoubtedly dilute the emphasis on continuity which our own shows, though without materially affecting the argument. ILL. 2 Early Cycladic vase forms. The picture presented thus in diagrammaticform is inevitably oversimplified,since it is impossible to illustrate a full range of forms or to indicate points of detail. The drawings are free adaptationsfrom the following sources:
I. Art and Culture 348, no. 392; 2. Burial Habits pl. 29 j; 3. Deltion 17 (1961-62) col. pl. 44; 4. Emergence 154, fig. Io. I,

no. 3; 5. Art and Culture354, no. 402; 6. Burial Habits pl. 37 a; 7. Emergence pl. 5, no. 3; 8. Burial Habits pl. 41 e, f; 9. Art

and Culture 356, no. 41o; 10o.Keos Pt. II fig. 5, B 29; ii. Zervos fig. 200; 12. Zervos fig. 223; 13. Art and Culture 347, no. 386; 14. Zervos fig. 231; 15. Zervos fig. 234; 16. Zervos fig. 19o; 17. Keos Pt. II fig. 5, C 13; 18. Delos XI 45, fig. 42 (bottom left); 19. Kastri fig. 5, no. I; 20. Zervos fig. 195; 21. Delos XI 44, fig. 41 (top center) (restored); 22. Kastri fig. 4, no. 4; 23. BSA 69 (1974) fig. 9,-(MM337); 24. Zervos fig. 92; 25. BSA 69 (1974) fig. 2, I (MM65); 26. Paros 42, fig. 42; 27. BSA 69

19801

THE EARLY CYCLADIC PERIOD TABLE I

143

SETTLEMENT STRATIGRAPHY

TRADITIONAL TERMINOLOGY (Modified)

RENFREW DOUMAS GROUPS CULTURES GROUPS CULTURES Lakkoudes Pelos >

4,

PHYLAKOPI A I

EC I

Plastiras

Kampos Kampos KEA B

PHYLAKOPI (A2) I-i

EC II

Syros 0

KEA C KASTRI

EC III A Kastri

0O

O Kastri

0 Amorgos
PHYLAKOPI I-ii I-iii EC III B Phylakopi I

Amorgosi

PHYLAKOPI II MC
PHYLAKOPI II MC

of the periods shown. See also J. Coleman's table in AJA 78 (1974) 344-

The Table demonstrates relationsonly and does not give a true indicationof the relativelength terminological

Doumasregardshis Amorgos,Kastriand PhylakopiI groups as approximately contemporary (Burial Habits with each other and with the 25-26). Renfrew'sKastriand Amorgosgroupsare regardedas partlycontemporary I "Phylakopi culture,"and the Amorgosgroup as havinglastedto the beginningof the MiddleBronzeAge. The culture,"as manifestedin the Amorgosgroup, can thus be seen as continuingas late as the MBA "Keros-Syros
(Emergence 195, 535).

144

BARBER AND MAcGILLIVRAY

[AJA 84

artefacts" might superficially suggest.7 First, THE EARLY CYCLADIC SEQUENCE there are other elements-settlement type and FINAL NEOLITHIC location, graves and burial customs-which must also be taken into account. Then, in No clear Neolithic sequence has yet been estabdealing with the artifacts,there is no firm rule lished for the Cyclades and there are only two excaas to how much similarity in how many vated sites. The position of the later of these, Keclasses is enough to set one assemblage clearly phala on Kea, in the Final Neolithic is suggested within a cultural group, or how much diver- by material parallels with sites in the Aegean area gence is necessaryto divorce it. Archaeological outside the Cyclades and by radiocarbondates.'2 The nature of the relationship between FN theory is not yet sufficientlysophisticated to be specific on such questions and there are no Kephala and EC I is unclear and likely to remain clear principles to follow; so until more sites have been excavated. Some simi5. There can only be a case for radical termino- larities are discernible in the cist graves, figurine logical alteration when the material evidence heads and marble vase types,"l but the characteris plentiful, from good contexts, well-studied istic pottery of EC I is not found at Kephala. and quite unambiguous-and when there is universal agreement on the need for it. EARLY CYCLADICI It is worth citing one striking instance of the EC I material has nowhere confusion that can be caused by speculation in Stratification.Although been found in clear stratigraphic relationship to this area. This may be found in the variety of that of the Final Neolithic, excavations at Phylaterms proposed for the final stage of the EH kopi on Melos have demonstrated its position in period. This period, traditional EH III, was the EC sequence. In the original excavations at the redefined by Caskey in I960.8Since then, it has which is recsite, the so-called "Earliest Pottery,""4 been entitled "Protominyan,"9 the "Tiryns ognizably EC I, was found in levels antedating the culture"'"and MH I.11 earliest phase of the First City.15 The pottery of The proponents of these various alterations all this primary First City phase (I-i)16 is clearly what had useful points to make about strictly archaeo- we would now call EC II and can be related to logical problems, but the end result of the introduc- finds from Period II (Ceramic phase B) on Kea."7 tion of this range of new terms is simply to retard These findings are confirmed by the recent excavations at Phylakopi"8where EC I material occurred progress towards their solutions. For reasons which we hope to make clear below, in the earliest levels (A i) and was succeeded (in we feel that there is insufficient evidence on which A 2) by EC II pottery, the two being separatedby to propose a variety of genuinely distinct cultural a transitional phase. The material is not yet fully groups in the EC period and that the traditional published. and essentially chronological divisions best reflect A stage which may now be describedas late EC I the facts as we know them and are still the most or transitional between EC I and EC II (the helpful to students. "Kampos group") was first isolated by Bossert'"
7 The use of the term "culture" itself is not without problems. See Terminologies 489. 8 EH Argolid. 9 Origins 94.

10 Emergence

11 H. and M. van Effenterre, "Comment croire ' 1'HellaKephala 98-

102-103.

dique Ancien III?," BCH 99 (1975) 35-50. 12 Interconnections 335-40; Emergence 75-77;
II0.

13 Kephala 107.
14 Phylakopi 82-85. 15 Phylakopi 239-43.

16 Phylakopi 85-87 ("Some Other Early Wares"), 248. EC II Glazed Ware (Urfirnis) and the characteristicsaucer shape can be recognized from the descriptionson p. 86.

The notation I-i etc. to indicate the divisions of the Phylakopi sequence into the Cities and their subphasesis directly derived from Phylakopi Ch. io and may be found in BSA 69 (I974) 4-5. It should be noted that, in the BSA article cited, the EC II characterof I-i was not clearly demonstrated. 17 We use the numbered period designationsfor the sequence at Ayia Irini (AJA 83 [19791 412), adding for convenience the lettered ceramic phases according to the system used in Keos Pt. II. For the EC II material from Kea, see Keos Pt. II 362-69. 18Art and Culture 20; A.C. Renfrew, "Phylakopi and the Late Bronze I Period in the Cyclades," in C. Doumas ed., Thera and the Aegean World I (London 1978) 405 (Table II); JHS-AR 1975-76, 2519Burial Habits 15, 18-20; Jdl 75 (1960) I-I6; Emergence 153, 527-28. Further references may be found in the works

1980]

THE EARLY CYCLADIC PERIOD

145

and furtherdefinedand discussed Renfrewand Type A, the earliest,is a cist which has all four by Doumas. sides lined with slabs.Type B, introduced slightly later, is slab-linedon only three sides, the fourth Settlements.About thirty-threeEC I sites were consistingof dry stonewalling.Both typesare first recordedup to 1970.20The cemeteriesare better found in EC I and persistinto EC IIIA. reknown than the settlementsand architectural Some fifteencemeteries Doumas'list"1 in appear were to have been used and cemeteries mains are sparse.Settlements in EC I (his Lakkoudes, only situatedclose togetherand normallybut not inevi- Pelos, Plastirasand Kamposgroups). Four others tablycoastal.Some of the buildingsseem likely to show some signs of further use later in the EC materials.At have been constructed perishable of periodbut no continuity. the EC I finds from the earlierexcavaPhylakopi, Doumas showsthat the five cemeteries Akroat with any structures but tiri, Kastrakiand tions were not associated on Naxos, at Livadhion Spedos some wall remainswere discoveredin 1975.21At Despotikon and at Akrotirakion Siphnos were Grottaon Naxos," remainsof rectangular houses used in both EC I and EC II. To these instances were associatedwith materialwhich may be re- should be added Ayioi Anargyroi,32 Phyrroyes" stage and Karvounolakkoi34 Naxos and Zoumbaria gardedas belongingto the final ("Kampos") on in of EC I.23Doumas24 EC II connections the on sees It Despotikon.35 should be noted that six of features of this thesenine siteswith bowls and architectural marble fromEC I to EC II continuity "Kampos"material.Traces, possiblyof a settle- yielded,amongthe EC I material, someof the late of ment,werefoundin pits nearthe cemetery Kato "Kampos" phase. Akrotirion Amorgos.The finds includedpottery may be madebetween to fromPhylakopi(EC 1)25 Pottery.Certaindistinctions comparable the earliest the potteryfrom the settlements that from the and A as well as EC II forms."2 single room at Panorcemeteries. The settlementmaterialhas been dismos on Naxos may belong to the final phase of The by EC 1.27It appears havebeenrectangular was cussedand illustrated Renfrew.36 fabricis to but semicoarse with a highly burnisheddark surface. damagedby laterbuilding. Pottery belonging to the end of EC I is also The most commonforms are the bowl, often with known from Paroikia Paros.28 on Furtherevidence rolled rim,"3and the collaredjar with vertically A of EC I settlements providedby surfacefinds of pierced lugs set on the shoulder.38 few other is were found at Phylakopi.39 Some of this settlement-typepottery near the cemeteries at shapes on PelosandAyiosPandeleimon Melos,Akrotiraki pottery is decoratedwith incised patterns(ill. 2, on Siphnos,Ayioi Anargyroiand Aila on Naxos no. 3). Fragmentsof "Kampos-type" frying pans (ill. 2, no. 5) have been found at Grottaand Kato and on Ano Kouphonisi.29 sometimesbearingstampeddecoration. Akrotiri,40 Gravesand Cemeteries. These have been compre- The cemeterypotteryhas been studiedand preHe hensivelystudiedby Doumas.He has shown that sentedby Doumas.41 dividesthe materialinto there were two main typesof tomb (A and B).30 four groups which he considers to representa
cited here.
20 Emergence I53-57, 509-25, fig. appendix I. 2. (The true

number is certainly greater since some of the sites of uncertain date must be EC I); ArchitectureI5I-56.
21 Phylakopi 82-85, 239-43; JHS-AR 1975-76, 25. 22 Praktika 1949, 112-22. 23 Interconnections 341. 24 Architecture 152. 25 Kykladika I pl. 9: 3, 9, i6, 17. 26 Kykladika I pl. 9: 19, 20.

27Architecture156, fig. 8.

28 Paros 20, fig. Io; 44, fig. 46. 29 Emergence 153.

Burial Habits 37-46. 31 Burial Habits 26 (Table).

30

32 Burial Habits

100-20.

The bulk of the material from the

graves belongs to latest EC I. Continuity of use into EC II is suggested by fragments of a glazed sauceboat (p. o104) and of hat-vaseswith stamped spiral decoration (e.g. p. 117 no. 29). 33G.A. Papathanasopoulos,Deltion I7 (1961-62) I38-39. Tomb 27 contained a cylindrical pyxis (EM 6122, pl. 71b) clearly related to an EC I example from Akrotiri on Naxos. Tomb 28 contained a jug (EM 6io9 [21, pl. 72b) with EC II-type dark-on-light decoration. 34 Emergence 518. 35Emergence 517. 36 Emergence 153-57, figs. Io. I, 10. 2. 37Emergence fig. Io.I. 38 Zervos pl. 71. 39 Phylaopi 82-85. 40 Zervos pls. 224-27; Emergence 527-28. 41 Burial Habits 15-20, figs. 3-7 (captions p. 134).

146

BARBER AND MAcGILLIVRAY

[AJA 84

44V3,

3,3

10

If

.21

o
1

to
_1

20
i

30

l/

5o
i

M.

ILL. I.

Mapshowingthe locationof sitesmentionedin the text (see n. I for key)

TABLE II Map no. Island/Site FN I

2

Period(s) of Occupation I ? K II X IIIA X IIIB

3 4

KEA Ayia Irini Kephala SYROS Chalandriani Kastri DELOS

X X

5
6 7 8 9 Io ii
12

MountKynthos
SIPHNOS Akrotiraki DESPOTIKON Livadhi Zoumbaria PAROS Kampos Paroikia Plastiras Pyrgos (Cem.) NAXOS Aila Akrotiri Ayioi Anargyroi Grotta Karvounolakkoi Kastraki Lakkoudhes Panormos X X X X X X X X X X X X X X X X

X
X X X

X
X

X X

13 14 15

16
17 18 19 20
21 22

X X ? X X ? X

Phyrroyes Spedos
ANO KOUPHONISI Agrilia KEROS Dhaskalio AMORGOS Arkesine Kato Akrotiri MELOS Aspro Chorio Ayios Pandeleimon Pelos Phylakopi THERA
Akrotiri

X X
X

X X
X

X X
X X

23

24

X X

25
26

X X X

27 28 29
30

X X X

X
X

31

32

CHRISTIANA

N.B.: FN =Final Neolithic; K = "Kampos group"

148
SAUCER SPHERICAL PYXIS SMALLBOWL

BARBER AND MAcGILLIVRAY

PAN FRYING VASE DUCK JAR FOOTED JUG

[AJA 84
PERIODS FORMS CONFINED SPECIFIC TO

K A M P
0
S

67

(I

13

14

lilA

2324211

1262

ILL. 2.

EarlyCycladicvaseforms (see n. I for key). Scale--27

1:12; 5,12,24

I:I0;

8,25,26 I:8; all others1:6

chronological sequence and names respectively after the sites at Lakkoudes, Pelos, Plastiras and Kampos. The most primitive forms are found in the early Lakkoudes group. The late "Kampos group" includes some forms (ill. 2, no. 6) which are developed from earlier types and others (ill. 2, no. 5; also 6) which are ancestral to those of EC II. It may accordingly be seen as transitional between EC I and EC II."2 In fabric and surface treatment, the cemetery material is similar to that from the settlements but there is a much greater incidence of incised wares. The course of development (ill. 2) of the most common forms (the collared jar and the spherical and cylindrical varieties of pyxis) may be traced from an early stage of EC I through to EC II.43
42 Burial Habits 24-25. 43 Supra n. 41.

Other forms include the footed collared jar (ill. 2, nos. 2, 6) and the hat-vase (ill. 2, no. 8). The bottle-shaped vase (ill. 2, no. 7), which belongs to final EC I, may be regarded as a variety of the collared jar. The patterns (ill. 3) are simple and rectilinear (hatched triangles, herringbone) in vertical or horizontal zones. Curvilinear ornament (circles, spirals) appearsfirst in the "Kampos group." EARLY CYCLADIC II Stratification. Extensive and well-stratified deposits of EC II material represent Period II (Ceramic phase B) of the sequence at Ayia Irini on Kea,44 where Period III (Ceramic phase C) is EC IIIA
44 Keos Pt. 11 362-69.

1980]
HERRING BONE CHEVRON TRIANGLE

THE EARLY CYCLADIC PERIOD

ZIG ZAG KERBSCHNITT CIRCLE SPIRAL STAR

149
OTHERS

IO

II

12
1 1111 1111111 i1 iii ii 1

13

14

1617
w V

ilIIAIIiI IIii1l IIIIII I

18 19 20

21

22

23

24

25

26

decorative motifs (see n. I for key). Not to scale ILL.3. EarlyCycladic (see below). The pottery of the preceding phase (Period I: Ceramic phase A) is Neolithic or earlier EBA in character45 and there was evidently a gap in occupation between A and B. Although the body of EC II material from Phylakopi is less substantial, it is clearly stratified between EC I (see above) and EC IIIB (see below). Settlements. Approximately thirty-one sites of this period were known in 1970.46Excavated settlements are Ayia Irini, Phylakopi, Pyrgos on Paros,47 Mount Kynthos on Delos48 and perhaps Grotta on Naxos,49 where the related tombs have EC II material. Surface finds suggest that Spedos on Naxos and Dhaskalio on Keros50may also have been occupied in this period but neither has been excavated.51 Most of the settlements are located on low hills or slopes near the sea and are undefended. Mount Kynthos seems an exception, since it was situated on a hilltop and apparently fortified,52though it is not clear to which phase of occupation the bastions there belong. Although there are more architectural remains than in EC I, it is difficult to isolate features characteristic of the period since these are strongly in5 Keos Pt. II 360-62. 176-78; 509-25 and appendix fig. I. 2. The true number is again almost certainly larger, supra n. 20. 47 Architecture151-52. 48Delos XI. Recent restudy of material from the Mount Kynthos excavations by J.A. MacGillivraypoints to phases of occupation contemporary with Kea Periods II and III (EC II and EC IIIA). 49 See above and Architecture 152. 50Architecture 163-64; Ph. Zapheiropoulou, "Ostraka ek
46Emergence

fluenced in each case by local conditions and local materials. On Kea, the houses are rectangular and laid out on a consistent axis53and the same may have been the case at Grotta.54At both these sites local conditions permitted such a layout, whereas at Mount Kynthos the lack of space and the uneven terrain resulted in rooms with rounded walls crowded together in an apparently disorganized fashion.55 Graves and Cemeteries.Tomb types A and B continue in use from the earlier periods as do some of the cemetery sites (see above). Corbelled graves are new but, within the Cyclades, are so far known only from Syros. On the coast of Attica they occur at Ayios Kosmas, a site with strongly Cycladic characteristics."Of the large number of cemeteries known from EC II,57only Spedos on Naxos, Akrotiraki on Siphnos and Chalandriani on Syros continued in use in EC IIIA. Pottery. There are three main classes of EC II pottery: i. Patterned ware. This has a fine light-colored fabric, with a smoothed surface. Decoration is applied in lustrous black, brown or reddish-brown paint. The patterns (ill. 3, nos. o0, II, 12, 17) are generally rectilinear and consist of hatched or crossKerou," AAA 8 (1975) 79-85. 51 Other settlements assigned by Renfrew to his "Keros-

Syros culture" may be either EC II or IIIA.

52 Architecture 162-63. 3 Keos Pt. 1 369, fig. 7. 54 Supra n. 22. 55 Delos XI pl. 3. See also n. 73. 56 Burial Habits 41-4957 Emergence Appendix I 509-25.

150

BARBER AND MAcGILLIVRAY

[AJA 84

hatched triangles, connected strings of lozenges, EARLY CYCLADIC IIIA zigzags, chevrons or, in one case, connected circles. A distinct phase following EC II is The most common forms are the sauceboat (ill. 2, Stratification. stratigraphically attested at Kea (Period III: Ceno. 16), the pyxis in its various forms (e.g. ill. 2, ramic phase C).64 It is marked in the pottery by no. o0)58 and the jug (ill. 2, no. 14). the appearance of classes of fine and semifine to Patterned ware is known from settlements and semicoarse highly burnished wares in forms not graves throughout the Cyclades. It appears for the found in the Cyclades before this time and apparfirst time in the Cyclades in EC 11;59 ently intrusive. Some fabrics and forms continue 2. Stamped and Incised ware. The fabric is coarse from the preceding period. Kastri on and micaceous with a brown to dark brown bur- unchanged Syros was probably founded and only occupied nished surface. The decoration is stamped or inthis period, to which all the material recovcised. The motifs (ill. 3, nos. 13-16) include cir- during ered there seems to belong. The pottery"5 closely is cles, spirals, tangents, stars and Kerbschnitt. The related to that of Kea Period III (Ceramic grooves are often filled with a white substance."6 The most common forms are the footed jar (ill. phase C). 2, no. 13), the pyxis (ill. 2, no. ii) and the Syros- Settlements. There appears to have been a marked reduction in the number of occupied sites between type frying pan (ill. 2, no. 12). Stamped and Incised pottery is known from sev- EC II and EC IIIA.6"In addition to Kea Period III eral sites throughout the Cyclades. The forms (ill. (Ceramic phase C) and Kastri, occupation in this 2, nos. I1-13) and decorative motifs (ill. 3, nos. 13period is attested at Panormos on Naxos67 and Mount Kynthos on Delos (phase 2)."8 Paroikia on 16) are clearly derived from EC I incised wares; 3. Glazed ware (Urfirnis). The fabric is usually Paros69and Dhaskalio on Keros70may also have fine and light colored, coated with a dark paint or been occupied at this time. In addition, a small glaze. It is closely similar to EH II Glazed or group of pottery from the islet of Christiana, near Urfirnis pottery61but it is not clear whether any Thera, may be from a settlement." This pottery of that found in the Cyclades was actually im- has no diagnostic EC II features and the jugs are ported: some is quite definitely local. The most of types known in EC IIIA on Kea and at Kastri. common forms are the saucer and sauceboat (ill. 2, Kastri, Mount Kynthos, Panormos and Dhaskanos. 9, 16). lio are situated on hilltops close to the sea. All have Other wares found in EC II are: (a) a type of only one convenient line of approach and appear brown or grayish-brown burnished ware in forms to have been fortified.72Excavations have revealed including the spherical pyxis (ill. 2, no. ii), the small houses with curvilinear walls, connected by saucer/bowl with inturned rim (ill. 2, no. io) and narrow passages. The well-ordered plans of the the pedestalled bowl;62 and (b) red-brown semi- houses on Kea and the location of Ayia Irini on a fine to semicoarse undecorated wares which make low promontory on the sea seem exceptional for up a significant proportionof the pottery from Kea the period.73 Period II (Ceramic phase B).63 With the exception of Paroikia, and perhaps
58 See also Zervos pls. 233-37.

Pottery with painted patterns occurs in Crete in EM I and there may be some connection between this and EC II developments in the Cyclades.
60

59 EH Argolid 292.

kykladike Kerameikeapo ta ChristianaTheras," ArchEph 1976,

8, n. 2. 68 Supra n. 48.

Art and Culture 113-14. 61 Keos Pt. II 363.

62

69 Paros fig. 45; also unpublishedfragment of a red burnished tankard in the Paros Museum.
70 AAA 8 (1975) fig. 4 a, b.

Zervos pl. 78; Keos Pt. IH365, fig. 3, B 27; Delos XI fig.

71 Supra n. 67.
72 Architecture I64.

41, bottom left. 63 Keos Pt. 11 365-66. 64 Keos Pt. IH 370-75. 65 Kastri 67-75. 66 MB Cyclades 374. 67 Deltion 19 (1964) Chronika 411-12;

73 Supra n. 53. We are grateful to Mr. M.S.F. Hood for pointing to a possible relationship between apsidal houses at Paroikia and Mount Kynthos and examples in Anatolia. These will be discussedby Mr. Hood in a forthcoming paper.
C. Doumas, "Proto-

1980]

THE EARLY CYCLADIC PERIOD

151

Dhaskalio, which may have been occupied into EC IIIB, all known settlements were abandoned within the EC IIIA period. Graves and Cemeteries.There appear to have been no significant changes. The few tombs assignable to the period are of Types A and B which originate early in EC I.7 These are found in cemeteries together with EC II graves. The graves at Chalandriani were not fully described but seem likely to have been corbelled, like the EC II examples from the same cemetery." Pottery. The most distinctive pottery of the period is of fine or semifine to semicoarse fabric, slipped red, black, reddish- or golden-brown and highly burnished to give a smooth lustrous finish. The most common forms are the tankard (ill. 2, no. 21), the bell-shapedcup,76the depas amphikypellon (ill. The use of 2, no. 22) and the straight-sidedplate.77 the wheel has been observed once or twice in Kea Period III (Ceramic phase C);78 at Mount Kynthos, all the pottery of the period is handmade. A wheelmade vase was noted by Tsountas among material from Kapsala (Amorgos)." Incised pottery, similar to the above in fabric and surface treatment, is found but is less common. The decoration (ill. 3, no. i8) may be in the form of zigzag patterns or series of vertical or horizontal lines with diagonal slashes in imitation of rope. The incisions, especially when on black slipped vases, are often filled with a white substance, as in EC II. Prominent forms are the pyxis (ill. 2, no. i8), the spouted pyxis,8s the narrownecked jug (ill. 2, no. 20) and the proto-duck vase (ill. 2, no. 19). Painted decoration is found on a few vases with burnished surfaces. The decoration consists of
7" Supra n. 26.

groups of double criss-crosslines (ill. 3, no. 20) or in cross-hatching"8 red or brown paint. The only form in painted ware is the pedestalled cup.82 The red-brown semifine to semicoarse wares show continuity from the preceding period (at least on Kea)83 and some EC IIIA shapes (the rounded saucer, spherical pyxis and narrow-necked jug) are probably derived from EC II forms (ill. 2). EARLY CYCLADIC IIIB Stratification. Material exemplified by finds from the later phases of the First City at Phylakopi (I-ii/iii) have traditionally been taken to represent EC III." I-iii is distinguished from I-ii by the absence of incised pottery, the appearanceof light-ondark painted decoration (probably as a substitute for white-filled incised) and the development of some curvilinearityin the designs on the Geometric painted pottery. It seems, however, that I-ii and I-iii are part of the same overall stage (EC IIIB), since "Geometric" pottery occurs in both. The presence of incised pottery in I-ii and its disappearance in I-iii appear to be of chronological significance. This sequence is to some extent confirmed by reanalysis of the original excavation record and by newly excavated finds, on the basis of which EC II material (in I-i) can be seen to have preceded that of EC IIIB in the sequence.85 Kea, however, the At EC II phase Period II (Ceramic phase B) is followed by one with material characteristicswhich we have already classed as EC IIIA. The relative position of EC IIIA and EC IIIB material is nowhere yet clearly established in a stratified sequence in the Cyclades and it is possible that they are to some degree contemporary."8 But for material of both these periods outside parallels
Keos Pt. IH 372. 84 CAH8, Vol. I, Part II, 795-99. 85 Supra n. 18.
86 Indicationsof contemporaneityare at present unconvincing. An unpublished tankard fragment from Paroikia (supra n. 69) is conspicuousamong predominantlyEC IIIB pottery. It is also possible to postulate some connection between EC IIIA pottery and that of Phylakopi I-ii (EC IIIB early) in the use of burnished ware and similarities in one or two forms. The relations do not, however, seem particularly close or intensive and the burden of the evidence at present clearly favors a sequence such as describedhere. The evidence for this will be fully presented in a forthcoming article by J.A. MacGillivray.
83

75 Kykladika II 82. 76 Keos Pt. II fig. 6, C2, C4I. For a restored example, see M.R. Popham and L.H. Sackett, Excavations at Lelkandi in Euboea 1964-1966. A PreliminaryReport (London 1968) 8, fig. 7, no. 6. 7 Keos Pt. II fig. 6, C5, C34. 78 Keos Pt. 11 373, C 36. 9 ArchEph 1898, 153: we are grateful to Prof. Coleman for drawing our attention to this piece.
80

81 Keos Pt. II pl. 8o, C 27.

Zervos pls. 192-94.

82 Zervos pls. 125-26.

152

BARBER AND MAcGILLIVRAY

[AJA 84

the Cyclades strongly suggest that the chronological disposition which we have adopted is essentially correct. For example, clear connections have been established between EC IIIB and Middle Helladic matt-painted pottery, and EC IIIB incised pottery is found in late EH III or early MH contexts." The connections of EC IIIA, on the other hand, are with EH II-III.88 At the Heraion on Samos, material of EC IIIA type belongs to periods II and III,89 while EC IIIB types come from period IV.9O Settlements. Phylakopi and Paroikia are the only two well-known sites of the period, though the existence of a few others is indicated by surface finds." Both the two excavated settlements are substantial and solidly constructed on regular lines. They appear to have been unfortified."2 Graves and Cemeteries. No cemetery areas are as yet closely defined, though there appears to have been a group of tombs close to the settlement at Phylakopi.3 Tomb types seem more various than previously. Cist graves probably continued in use since pottery in the "Geometric" style occurs in a double tomb of this type on Amorgos94and the cist is still found later in MC times."5It is not known to which of Doumas' types these cists should be assigned. Rock-cut tombs at Phylakopi96 were associated with "Geometric"-style pottery, as were others on Melos at Aspro Chorio."9Burials of children in urns within the settlements are known from Phylakopi, Paroikia and Thera.9 At Ayia Irini"9burials of this kind appear to belong to the MC period and, with one exception, are extramural.
87Matt-painted: Hesperia 33 (1964) e.g. 283-84, 301-303.

The diversity of types is interesting. The cist clearly continues a traditional form. Rock-cut tombs are known on Euboea at Manika,100 where the pottery includes EC IIIA types. It has recently been suggested that the Cycladic practice of urn burial may have derived from Anatolia.1o' Pottery. The full range of pottery found in this period will be better known when study of the newly excavated material from Phylakopi is complete. Meanwhile, it is clear that there are two classes of decorated pottery that are characteristic of the period, though the first (the Incised ware) belongs only to the earlier part of it (I-ii). The fabric of the Incised pottery is moderately coarse and gritty. The surface is dark and lustrous, an effect produced by slip and burnishing, though there is no high shine such as is often found on the pottery of EC IIIA. The class was first described in the original publication of the Phylakopi excavations'02but has not recently been studied in detail. The most common forms are the conical pyxis (ill. 2, no. 26), the duck vase (ill. 2, no. 24) and the beaked jug (ill. 2, no. 25). The incisions are often white-filled, as in earlier periods. Decorative motifs (ill. 3) include chevrons, zigzags, diamonds, triangles and circles, in various combinations. Pointillie filling (ill. 3, no. 25 b-c) and hatching are quite frequently used. There are some scenes with figures, for example, a boat and a fish from Phylakopi.'03 The fabric"'0and some of the forms (e.g. ill. 2, nos. 24-25)105 and motifs (ill. 3, nos. 21-25)106 seem to have local antecedents but mainland parallels for the decoration are EH III-MH,1' and incised matenatos, Excavations at Thera 7 (Athens 1976) 12, pl. 8b (not from the main site). 9 Keos Pt. 11 383-85. We are grateful to Mrs. G. Overbeck for informationon this point. 100G. Papavasileiou, Peri ton en Euboia Archaion Taphon
AJA 102Phylakopi 87-92, pls. IV-V ("An Early Group of Darkfaced Vases").
103 Phylakopi pl. 5, nos. 8a, 8c.

Incised: S.A. Immerwahr, The Athenian Agora 13: The Neolithic and Bronze Ages (Princeton 1971) 58-59 and references.
88 Supra n. 86.

98 Phylakopi 15; BSA 17 (I9I0-II)

6-9; Paros 12; S. Mari-

V. Milojcic, Samos I (Bonn 1961) 43-48, pl. 47 (e.g.). Milojci' (supra n. 89) 48-49, pl. 49 (e.g.); H.P. Isler, "An Early Bronze Age Settlement in Samos," Archaeology 26
89

90

(1973) 175 (duck vases and conical pyxides). 91 Emergence 509-25, fig. appendix I. 2 (510). 92 Architecture 168-69. 93 Phylakopi 234-36.

(Athens 1910) 1-19. 101 T.S. Wheeler,

78 (1974)

184.

94 The material is discussed and illustrated by E.M. Bossert, "Zur Datierung der Griber von Arkesine auf Amorgos," in
Festschrift fiir P. Goessler (Stuttgart 1954) 23-34.

104 Emergence
105

189. The beaked jug (ill. 2, no. 25) is found in Incised ware,

95 MB Cyclades378.

though not commonly.

96 Supra n. 93. 97 Emergence 511-12;

unpublished

vases in the Melos Mu-

seum.

106 See also Art and Culture figs. 123, 129, 130, 131; Phylakopi pls. 4-5. 107 PPS 22 (1956) 192.

1980]

THE EARLY CYCLADIC PERIOD

153

rial from Kea is MC.'08 As already indicated, this pottery is not found after I-ii at Phylakopi.1'9 The second class, the so-called "Geometric"pottery, was also described in the original Phylakopi report."'?Similar pottery is known from Paroikia and, to a lesser extent, Amorgos (Arkesine). The fabric is quite distinctive--hard fired and brittle, often with reddish surfaces and gray core. There is usually a thin white wash, at least over the decorated area of vases with dark-painted decoration. The decoration is in reddish-brown to blackish paint which now seems matt rather than lustrous, as has several times been observed, in spite of the description given in the primary publication. Use of a light-on-dark system is less common; this latter method has been thought to have directly replaced white-filled incision. A number of shapes are found with some frequency, among which may be mentioned the barrel jar (ill. 2, no. 27), beaked jug (ill. 2, no. 25), two varieties of cup"' and the so-called "Melian bowl.""11 The motifs are simple and rectilinear and, in some cases, close to those found on the incised pottery,"3 but there is a greater range, and the exploitation of circles and other curvilinear motifs (ill. 3, no. 26 a-b) is a conspicuous development (in I-iii). In a detailed study Buck has demonstrated the close connection of this class of pottery with MH matt-painted.14 Possible relationships with material from southern Anatolia and farther to the East are not yet fully understood. "5 MIDDLE CYCLADIC (EARLY) Some brief remarks about the character of the early Middle Cycladic period may help to define the final stage of Early Cycladic.
108Keos Pt. 11 381, 383, pl. 84 D56-57, 137-38. 109Phylakopi 250. 110Phylakopi 96-1o2, pls. 7-10 ("Vases with Geometric Designs in Lustrous Paint"). 111Zervos pls. 140-41. 112 Phylakopi pl. 33, 3-9. The shape is generally related to ill. 2, no. 23. 113 Comparealso Phylakopi pls. 4, 7 and 5, 4 with 7, 15 and 9, 2 respectively (pls. 4, 7 and 9, 2 ill. 3, nos. 23, 24). 114R.J. Buck, "Middle Helladic Mattpainted Pottery," Hesperia 33 (1964) 231-308. 115AnatSt 4 (1954) 207; BSA 51 (1956) 14 and n. 9; PPS 22 (1956) 192.

Stratification. The division between the First and Second Cities at Phylakopi, which is marked by a major destruction level, has normally been taken to define the change from Early to Middle Cycladic. The Kea sequence seems broadly to confirm this division, though some links between Kea Period IV (Ceramic phase D - early MC) and late Phylakopi I,116 chiefly in pottery and the practice of urn burial, indicate that careful analysis of Cycladic interrelationshipsat this time will be necessary before we are able fully to understand the chronology of the transition. Settlements. The three major excavated MC settlements (Kea, Paroikia and Phylakopi) were all occupied in at least two of the main phases of the EC period. The MC town sites are important local centers and relatively sophisticated in construction and design.11 Graves and Cemeteries. These are not very well known except on Kea. Grave types seem even more numerous than previously.l"8Cist graves were still used in the MC period and both simple and more elaborate forms are found. The use of rock-cut tombs may have continued from EC III. Urn burials are quite common on Kea, whereas at Phylakopi such burials seem to belong to EC IIIB. At Ayia Irini there are two extramural cemeteries of infant burials, chiefly in cists and urns, in Period IV (Ceramic phase D).119 Pottery. The typical classes of the mature MC period are Dark Burnished and Cycladic White (the "Early Mycenaean Style with Designs in Matt Black" of the original report).120 These and subsidiary MC fabrics have been discussed elsewhere.21' They occur at all the three sites mentioned above.122Minyan shapes are common in
116 Amplificationof remarksmade here and further references may be found in MB Cyclades.The situation will become clearer on final publication of the Kea material. Meanwhile we are grateful to J.L. Caskey and J.C. Overbeck for information and comments which are necessarilyprovisional. 117 Phylakopi 38-50, pl. I. 118 MB Cyclades 378.

119We are indebted to Mrs. G. Overbeck for pointing this out to us.
122 I am informed that the frequency of Cycladic White on Kea is considerablygreater than I mistakenly reported in MB

120Phylakopi Io8-I8, pls. 14-I9. 121 MB Cyclades 374-77.

154

BARBER AND MAcGILLIVRAY

[AJA 84

Dark Burnished ware and there are connections between Cycladic White and later MH Mattpainted. The characterof the very earliest MC pottery has become rather obscure with the recognition'23that the EC IIIB "Geometric" class described above probably did not last into the Second City and with the great difficulty posed by the distinction of a supposedly later category of "Geometric"pottery ("Geometric Pottery with Designs in Matt Black") made in the original report.'24 These problems compound the difficulty of determining the degree of cultural continuity between Phylakopi I and Phylakopi II. We should, however, note that this was not doubted by the and the clear derivation of original excavators"25 certain MC forms from EC predecessors'26supports their view.

NON-CERAMICFINDS
Non-ceramic finds have not been included in the above summary since the chronology of individual types is rarely firm enough to allow valid generalizations. Nevertheless, the importance of this material is self-evident since it enables us to paint a much more complete cultural picture than is possible solely with the use of the evidence so far described. A summary of the non-ceramicobjects of the EC period is included in Doumas' discussion of the EC sequence."'2Useful studies of the figurines'28 and bronzes'29have appeared. The body of material includes marble vases and figurines, bronze weapons, tools and ornaments, some objects of gold
Cyclades375 (R.L.N.B.). 123 MB Cyclades376-77.
124 Phylakopi 1o2-o16, pls. 11-13. 125Phylakopi 258. 126 Bowl, Phylakopi 259 and pl. 33, I-8; beaked jug, com-

and silver, a little lead and bone. Obsidian is common. One or two points seem to be of particular significance. Marble vessels do not appear in the earliest stages of EC I.130Figurines occur only in Doumas' second EC group and subsequently there is a development from schematic to relatively naturalistic and more complex pieces-a development which is surely clear evidence of strong cultural continuity.'3 It is doubtful if any were manufactured after EC IIIA.3a2 The range of marble and ceramic forms expanded greatly in EC II.133 Obsidian and bone objects are found throughout the EC period. Bronzes are not regularly found until late EC I-EC II. Many of those formerly dated generally to the "Keros-Syrosculture" or the "Early Bronze 2 Peseem to belong more specifically to EC riod""'34 In some cases,'36there are clear Anatolian IIIA.'35 connections. Reinvestigation of this material is now urgently required since the evidence already cited suggests that the chronology and character of EC metallurgical developments may have to be reassessed. It may be that these innovations in EC IIIA are connected with the new ceramic forms which have already been noted.
CONCLUSIONS

This summary picture of the Early Cycladic period seems to us to demonstrate clearly two fundamental points. The first is that generations of painstaking research have at last begun to produce a reasonably detailed pottery-basedchronological sequence which gives us some chance of charting the history of these times. The second point of crucial
77-79), it must be doubtful whether the figurine illustrated as Type VII in Emergencefig. 11. 8 (Burial Habits fig. 13 q) is in fact so early. The other pieces shown on Phylakopi pl. 39 seem likely (pace Doumas, Burial Habits 24) to be earlier than EC IIIB (late Phylakopi I). 133Burial Habits 2o-22. 134 AJA 71 (1967) 12 etc. 135E.g. those from Kastri, Kastri 63-67, fig. 2. 136 The material will be presented shortly by J.A. MacGillivray. See S. Dietz "Aegean and Near-EasternMetal Daggers in Early and Middle Bronze Age Greece. The Dating of the Byblite Bronze Hoards and Aegean Imports,"ActaA 42 (1971) 1-22 (especially I-6); Kastri 66-67; F. Schachermeyr, Die digaischeFriihzeit I (1976) I89; D.B. Stronach, "The Development and Diffusion of Metal Types in Early Bronze Age Anatolia," AnatSt 7 (I957) 89-125.

pare Phylakopi pl. 9, i-io (First City) with pl. 14, I-6.
127 Burial Habits 11-27.

128A.C. Renfrew, "The Development and Chronology of the Early CycladicFigurines,"AJA 73 (1969) 1-32. 129A.C. Renfrew, "Cycladic Metallurgy and the Aegean
Early Bronze Age," AJA 71 (1967) ford 1974).
130 Burial Habits 16. 131 Emergence fig. Ii. 132 In view of

i-2o;

also K. Branigan,

Aegean Metalwork of the Early and Middle Bronze Age (Ox8.

J.L. Caskey's discussion of the Kea pieces in

II13-26 (and further Hesperia 43 119741

Hesperia 40 (197)

1980]

THE EARLY CYCLADIC PERIOD

155

significance is the demonstration of the essential cultural continuity of Cycladic civilization throughout much of the EC period,'37certainly to the end of EC II and, further, though to an extent not yet clear, into the MC period. Perhaps the most striking evidence for this second point is the continuous tradition of burial in cist tombs from EC I (or, rather, from the Final Neolithic) into MC times. We may also note the continuity of use of some from EC I to IIIA. cemeteries138 Continuity of settlement occupation1l9is less easy to show, though one may suspect that this is at least partly due to the lack of excavated sites. At Phylakopi, there was continuity from EC I to EC II, also perhaps at Grotta on Naxos and Kato Akrotiri on Amorgos. Later, both Mount Kynthos on Delos and Ayia Irini on Kea were occupied in both EC II and EC IIIA. The two main EC IIIB sites, Phylakopi and Paroikia, have produced virtually no evidence of EC IIIA occupation, although they were both occupied earlier in the EC period. Some EC IIIA settlements are or seem to be new foundations (Kastri, Christiana and perhaps Panormos). Continuity in ceramic styles (ills. 2, 3) is clearly demonstrablebetween EC I and EC II, as has been illustrated above. The new features which can certainly be distinguished in EC II may be largely attributable to contacts with the Argolid. In EC IIIA the elements of change are the most evident but we have already pointed out certain features of ceramic continuity. Among other finds, we may note that marble vases and figurines are found in contexts from EC I until at least EC IIIA. Alterations in type surely represent natural development rather than radical innovation. So far we have stressed the indications of continuity. But certain changes are also, of course, highly significant. We should note, in particular, the greatly increased range of ceramic forms and other artifacts in EC II and the striking innovations in pottery and bronzes in EC IIIA. The former may be explained in terms of steady economic and technological progress, stimulated in part by
137 As observed recently by Doumas, Burial Habits 26-27 and Coleman, Interconnections340. 138 See Note to ill. i. 139See Note to ill. i. 140Compare Phylakopi pls. IV, ii and IX, o0with ArchEph

contact with adjacent areas, and do not warrant description in terms of basic cultural change. The latter almost certainly do indicate radical cultural innovation and the arrival of a new population element from Anatolia. In every case the elements of innovation must be balanced against those of continuity but, in our view, it is only at this point in the course of the Early Cycladic period that there seems real justification for postulating any genuine "cultural"distinction and even this requires validation by fresh and careful study of the relevant material. Whatever its "cultural"character,the chronological significance of this group of material is

clear.

EC IIIB material cannot at present be correlated in detail with that of the earlier EC sequence and this apparent discontinuity may eventually prove significant. We have already noted an example of cist burial in this period on Amorgos. Some EC IIIB pottery shapes, in particular the beaked jug (ill. 2, nos. 14, 20, 25),140 must be derived from EC II-IIIA types and the incised potteryt"' of earlier EC IIIB clearly belongs in the EC tradition. The alternative framework proposed*42 disfor cussion of EC material attempts to separatecultural character and chronology by discussing cultural groups in a very loose chronological framework (EB I, EB 2, etc.) which is applied to the whole Aegean area. This sweeping rearrangement causes other problems, beyond the scope of this article, since it requires the imposition of a new set of descriptive terms on the Mainland material as well. These have so far found no favor with scholars. Crete is excluded from this universal realignment1' on the ground that the culture of the island is a unified one. But, in our view, the Cyclades are in an essentially similar position and for the Mainland, in the words of J.L. Caskey'44: "The term Helladic remains satisfactoryas a general designation in the district to which it was applied originally." The cultural groups defined for the Cyclades cannot be truly distinct "cultures,"if the continuity of EC civilization, at least in its earlier stages, is as
1976, pls. 1-4. 141 Phylakopi pls. 4-5. 142 Emergence 53-55.

143Emergence 53-54. 144Terminologies 489.

156

BARBER AND MAcGILLIVRAY

[AJA 84

strong as, in our opinion, the material evidence suggests. Thus the main reason adduced for terminological change is seen to be invalid. Further, we find the loss of chronological exactitude involved in the alternative system most unsatisfactory,and a positive barrier to a developing historical understanding. It is true, as Renfrew has pointed out,145that there is a degree of illogicality in allowing the possibility of any overlap between periods which are labelled in numerical sequence (EC I, II, III, etc.). Provided, however, that the sequence is essentially correct, we do not feel that this is a serious difficulty. It has certainly not proved to be such to date. Within the traditional system we can continue to construct a truly historical framework, based on a pottery sequence which can be seen to have a general validity, while admitting the existence of minor regional variations which may be discussed in terms of sites or local groups,146such as those usefully formulated by Doumas. The "culture"system has resulted in a good deal of confusion. It was not intended that the "cultures" should be exact equivalents of the traditional periods EC I-III, and indeed it would seem that the middle phase of the new groupings encompasses aspects of all three traditional periods (viz. much of standard EC II, the "Amorgos Group"147 which includes pottery of EC IIIB type, and the "earlier and as yet undocumented phase of the Keros-Syros which presumably belongs in EC I). Culture"148 But, in fact, they have often and inevitably been taken as providing such equivalents.'49The problem is, of course, not susceptible of a definitive solution, since terminological systems are partly a matter of taste and prehistoric archaeology is of necessity such an inexact science that neither its rules nor the methods of their application can be definitively prescribed. At this point our reasoned preference for the retention of an established system seems worth reiterating. Our main objections to the proposed alterations rest on the two central points propounded above, but others may be worth noting. There is a standing risk in the choice of site names for "cultures" which is only partly acceptable. In the case of the
145 Emergence 54. 146 Burial Habits 15-27. 147 Emergence 534-35.

EC period the unsuitability of some of those chosen is a clear indication that the body of material and research necessaryto underwrite such a radical adjustment of terminology is not available. Problems arise at every point. The name "GrottaPelos" for the earliest EC "culture" was quickly questioned."s'It was pointed out that the material from Grotta had characteristicswhich seemed more appropriate to the following period and, in any case, was hardly typical of the earliest cultural stage. Thus there are now two terms current for this "culture."It may be observed here that the very fact of such a difficulty arising strongly suggests that the "Grotta-Pelos"and "Keros-Syros"cultures are not readily distinguishable. In the case of the subsequent group, the "KerosSyros culture," Keros is unexcavated, while the island of Syros has produced material of two quite distinct chronological (and possibly cultural!) stages, EC II and EC IIIA. If we are to talk in terms of "cultures,"it is between these two periods, as observed above, that we see more radical evidence of cultural difference than exists between any of the other "cultures," groups or periods so far defined. In this context it is interesting to note that the compelling new cultural features of EC IIIA do not correspond to any of the major cultural divisions of the alternative system, merely to a subgroup of one of them (i.e. the Kastri group of the "Keros-Syrosculture"). As for the third main division, the "Phylakopi I culture," inspection of the original Phylakopi report indicates that it is erroneous to regard all the material of the First City as belonging to the latest stage of the EC period. We have already pointed out that the pottery of the first phase (I-i) is recognizably EC II ("Keros-Syros"in "culture"terms). Further, no account is taken of the apparentchronological distinction that can be observed in the pottery sequence which shows that incised pottery is absent from the last phase of the First City at Phylakopi (I-iii). It will be evident therefore that we see few, if any, advantages in the adoption of a new terminological system for the EC period and several for retaining the establishedone.
Emergence 185. 149 Burial Habits I2; Origins fig. 9. 3; BSA 69 (1974) 48. Architecture 151 n. I. 150o
148

1980]

THE EARLY CYCLADIC PERIOD

157

We suggest that a detailed chronological sequence is necessaryif we are to investigate the past in historical terms. Further, we consider that "cultural" distinctions can only be reliably made on the basis of an assessment of a full and adequately documented range of material evidence-a state of affairs which does not at present obtain for any part of the EC period. Finally, we insist that the continuity of the earlier stages of EC culture is so striking as to forbid any subdivision other than a chronological one and that marked cultural change, when it does seem to occur in EC IIIA,
Is1 Terminologies 491.

appears to have an equal chronological significance so that its recognition within the traditional framework poses no problem. We would accordingly urge, with Caskey,lS'further careful collection and analysis of the evidence on the basis of the traditional system until such time as a broad change may universally be agreed to be desirable. ARCHAEOLOGY DEPARTMENT OFCLASSICAL
UNIVERSITY OF EDINBURGH

16-20 GEORGESQUARE
EDINBURGH EH8 9JZ, UNITED KINGDOM