Franchthi Neolithic Pottery, Volume 1: Classification and Ceramic Phases 1 and 2

By Karen D. Vitelli

The first of two systematic reports on the more than one million sherds of pottery recovered from the Franchthi Cave in Greece.

Over two and a quarter metric tons of pottery were recovered from Neolithic deposits at Franchthi and Paralia which will significantly increase our understanding of Neolithic pottery and Neolithic society in southern Greece. Through the development and application of a new system of ceramic classification, this fascile analyzes the pottery from the earlier Neolithic deposits as a direct reflection of the human behavior that produced it.

“A highly innovative study that foregrounds the decision-making and technological choices of Neolithic potters.” —Antiquity

“Imaginative, rigorous and admirably lucid study.” —Journal of Hellenic Studies

• Language: English
• Publisher: Open Road Integrated Media
• Release date: Oct 26, 2018
• ISBN: 9780253044402

Book preview

Franchthi Neolithic Pottery

Excavations at Franchthi Cave, Greece

T.W. Jacobsen, General Editor

Lisa Williams, Managing Editor

FASCICLE 8

Franchthi Neolithic Pottery

VOLUME 1

CLASSIFICATION AND

CERAMIC PHASES 1 AND 2

KAREN D. VITELLI

INDIANA UNIVERSITY PRESS

Bloomington & Indianapolis

Copyright © 1993 by Karen D. Vitelli

All rights reserved

No part of this book may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and recording, or by any information storage and retrieval system, without permission in writing from the publisher. The Association of American University Presses’ Resolution on Permissions constitutes the only exception to this prohibition.

This book has been supported by a grant from the National Endowment for the Humanities, an independent federal agency

Library of Congress Cataloging-in-Publication Data

Vitelli, Karen D.

Franchthi Neolithic pottery.

p. cm. — (Excavations at Franchthi Cave, Greece; fasc. 8)

Includes bibliographical references and index.

Contents: v. 1. Classification and ceramic phases 1 and 2.

1. Franchthi Cave (Greece) 2. Paralia Site (Greece) 3. Neolithic period—Greece. 4. Pottery, Prehistoric—Greece. I. Title. II. Series.

GN816.F73V58 1993 738.3’ 82’ 09388—dc20 92-43943

ISBN 0-253-31980-3 (pbk.: v. 1)

1 2 3 4 5 97 96 95 94 93

CONTENTS

TABLES I–XV

TABLES 1–13

1.  The ranked system of the Franchthi ceramic classification

2.  The Franchthi ceramic categories

3.  Ceramic sequence in FAS:SE corner (counts)

4.  Ceramic sequence in FAS:SE corner (weights)

5.  Ceramic sequence in FAS:SW corner (weights)

6.  Ceramic sequence in FAS:NE corner (weights)

7.  Ceramic sequence in FAS:NW corner (weights)

8.  Ceramic sequence in FAN:SE corner (weights)

9.  Ceramic sequence in FAN:SW corner (weights)

10.  Ceramic sequence in FAN:NE comer (weights)

11.  Ceramic sequence in FAN:NW corner (weights)

12.  Temporal occurrence of FCP 1 and FCP 2 varieties

13.  C-14 dates from Franchthi relevant to FCP 1 and FCP

PLANS

1.  Paralia: All walls

2.  Cave: Post-Neolithic disturbance

3.  Paralia: Terraces

4.  Paralia: Int 0/1

5.  Cave: Int 0/1, A-FA-F area

6.  Paralia: FCP 1

7.  Cave: FCP 1

8.  Paralia: Int 1/2

9.  Cave: Int 1/2

10.  Paralia: FCP 2.1

11.  Cave: FCP 2.1

12.  Paralia: FCP 2.2

13.  Cave: FCP 2.2

14.  Paralia: FCP 2.3

15.  Cave: FCP 2.3

16.  Paralia: FCP 2.4

17.  Cave: FCP 2.4

18.  Paralia: FCP 2.5

19.  Cave: FCP 2.5

FIGURES I–V

FIGURES 1–102

1.  FCP 1 Lime Monochrome Burnished variety

2.  FCP 1 Lime Monochrome Burnished variety

3.  FCP 1 Lime Monochrome Burnished variety

4.  FCP 1 Lime Monochrome Burnished variety

5.  FCP 1 Lime Monochrome Burnished variety

6.  FCP 1 Lime Monochrome Burnished variety

7.  FCP 1 Lime Monochrome Burnished variety

8.  FCP 1 Sandy Monochrome Burnished variety

9.  FCP 1 Sandy Monochrome Burnished variety

10.  FCP 1 Andesite Monochrome Dark Burnished and Ungritted varieties

11.  FCP 1 Lime Patterned and Serpentine Monochrome Burnished varieties

12.  FCP 1 Lime Patterned variety

13.  FCP 1 Sandy Patterned variety

14.  Int 1/2 Lime Monochrome Burnished variety

15.  Int 1/2 Lime Monochrome Burnished variety

16.  Int 1/2 Lime Monochrome Burnished variety

17.  Int 1/2 Sandy Patterned and Lime Patterned varieties

18.  Int 1/2 Sandy Monochrome Burnished variety

19.  Int 1/2 Sandy Monochrome Burnished variety

20.  Int 1/2 Andesite, Serpentine, Ungritted, Monochrome Urf, and Patterned Urf varieties

21.  FCP 2.1 Lime varieties

22.  FCP 2.1 Lime varieties

23.  FCP 2.1 Lime varieties

24.  FCP 2.1 Sandy, Ungritted, and Burnished-Over Urf varieties

25.  FCP 2.1 Andesite, White on Red, Serpentine, and Sandy Patterned varieties

26.  FCP 2.1 Monochrome Urf and Lime Patterned varieties

27.  FCP 2.1 Monochrome Urf variety

28.  FCP 2.1 Monochrome Urf variety

29.  FCP 2.1 Patterned Urf variety

30.  FCP 2.1 Patterned Urf variety

31.  FCP 2.1 Patterned Urf variety

32.  FCP 2.1 Patterned Urf variety

33.  FCP 2.1 Coarse Urf variety

34.  FCP 2.2 Lime, White, Sandy, and Andesite varieties

35.  FCP 2.2 Monochrome Urf variety

36.  FCP 2.2 Monochrome Urf variety

37.  FCP 2.2 Monochrome Urf variety

38.  FCP 2.2 Patterned Urf variety

39.  FCP 2.2 Scribbled Urf and Burnished-Over Urf varieties

40.  Early Urf: Monochrome Urf and Coarse Urf varieties

41.  Early Urf: Patterned Urf variety

42.  Early Urf: Patterned Urf variety

43.  FCP 2.3 Sandy and Lime varieties

44.  FCP 2.3 Andesite varieties

45.  FCP 2.3 Andesite varieties

46.  FCP 2.3 Scribbled Urf and Monochrome Urf varieties

47.  FCP 2.3 Monochrome Urf variety

48.  FCP 2.3 Monochrome Urf variety

49.  FCP 2.3 Patterned Urf variety

50.  FCP 2.3 Patterned Urf variety

51.  FCP 2.3 Patterned Urf variety

52.  FCP 2.3 Patterned Urf variety

53.  FCP 2.3 Patterned Urf variety

54.  FCP 2.3 Coarse Urf variety

55.  FCP 2.3 Burnished-Over Urf variety

56.  FCP 2.4 Lime varieties

57.  FCP 2.4 Sandy varieties

58.  FCP 2.4 Monochrome Urf variety

59.  FCP 2.4 Monochrome Urf variety

60.  FCP 2.4 Monochrome Urf variety

61.  FCP 2.4 Monochrome Urf and Burnished-Over Urf varieties

62.  FCP 2.4 Patterned Urf variety

63.  FCP 2.4 Patterned Urf variety

64.  FCP 2.4 Patterned Urf variety

65.  FCP 2.4 Patterned Urf variety

66.  FCP 2.4 Coarse Urf and Scribbled Urf varieties

67.  Late Urf: Monochrome Urf variety

68.  Late Urf: Patterned Urf variety

69.  Late Urf: Patterned Urf variety

70.  Late Urf: Patterned Urf variety

71.  Late Urf: Patterned Urf variety

72.  Late Urf: Patterned Urf variety

73.  Late Urf: Patterned Urf variety

74.  Late Urf: Patterned Urf variety

75.  Late Urf: Patterned Urf variety

76.  Late Urf: Patterned Urf variety

77.  Late Urf: Burnished-Over Urf, Patterned Urf, and Monochrome Urf varieties

78.  Late Urf: Coarse Urf variety

79.  Late Urf: Coarse Urf variety

80.  FCP 2.5 Lime, Sandy, Andesite, and Serpentine varieties

81.  FCP 2.5 Monochrome Urf variety

82.  FCP 2.5 Monochrome Urf and Scribbled Urf varieties

83.  FCP 2.5 Patterned Urf variety

84.  FCP 2.5 Patterned Urf variety

85.  FCP 2.5 Scribbled Urf variety

86.  FCP 2.5 Scribbled Urf variety

87.  FCP 2.5 Pattern-Burnished Urf and Burnished-Over Urf varieties

88.  FCP 2.5 Pattern-Burnished Urf variety

89.  FCP 2.5 Pattern-Burnished Urf variety

90.  FCP 2.5 Pattern-Burnished Urf variety

91.  FCP 2.5 Pattern-Burnished Urf variety

92.  FCP 2.5 Coarse Urf variety

93.  L5 East section

94.  FCP 1 vessel shapes, all wares and varieties

95.  Int 1/2 vessel shapes, all wares and varieties

96.  FCP 2.1 vessel shapes, all wares and varieties

97.  FCP 2.2 vessel shapes, all wares and varieties

98.  Early Urf (FCP 2.1 or 2.2) vessel shapes

99.  FCP 2.3 vessel shapes, all wares and varieties

100.  FCP 2.4 vessel shapes, all wares and varieties

101.  Late Urf (FCP 2.3 or 2.4) vessel shapes

102.  FCP 2.5 vessel shapes, all wares and varieties

PLATES

1.  Building procedures

2.  Painting and firing effects

3.  Vessel function

4.  Recycled sherds

FOREWORD

It is with great pleasure and pride that I introduce this impressive volume as the latest installment in the Franchthi final publication series. Fascicle 8 is the product of years of exhaustive study and analysis and represents the first in a two-part presentation of the Neolithic pottery from our site. The companion volume (Vitelli, forthcoming b) is currently in preparation.

This publication is of major importance for a number of reasons. The conclusions arrived at by Professor Vitelli are, of course, absolutely fundamental to an understanding of the earlier Neolithic chronological sequence at the site of Franchthi and to the correlation of areas of prehistoric human activity both inside and outside the cave. Accordingly, the results of this study provide valuable independent information that complements and/or clarifies stratigraphic successions generated from other bodies of evidence (see, e.g., Wilkinson and Duhon 1990).

The significance of this study, however, goes well beyond the chronological — as was the author’s intent. It is a pathfinding contribution to our understanding of prehistoric ceramic technology and classification, and, in my judgment, promises to be the most influential publication of prehistoric Aegean pottery since that of Arne Furumark more than fifty years ago. Indeed, I believe that it surpasses Furumark’s monumental work in the explicitness and back-to-basics practicality of its classification system — to say nothing of Vitelli’s aim to get at the people, especially the potters, behind the pots. Up to now, concerns of this kind have been rare in Aegean archaeology. In short, I recommend this volume to all Aegean archaeologists, as well as to all others interested in the production and use of prehistoric pottery.

The preparation of this volume has departed somewhat, by way of experiment, from the procedures employed in the production of other components of this series. The typesetting was done in England through the agency of Christopher Chippindale and Dora Kemp, and the bulk of the proofreading and copyediting was done there and elsewhere outside Bloomington. Most of the production costs of this and a subsequent fascicle (9) were provided by the Office of Research and the University Graduate School at Indiana University, to whom we are most grateful.

Those familiar with this series will note that, for the first time, a subject index has been added to this volume. We hope that the reader will find it useful.

The reader is reminded to consult the first fascicle in the series (Jacobsen and Farrand 1987) for section drawings illustrating the discussion of context and phasing in Part II of this volume. Please note, too, that the author has chosen to follow an uncalibrated radiocarbon chronology in Part I and a calibrated chronology in Part II.

Once again I would like to thank Lisa Williams for her valued help with the last-minute preparations of this volume for publication, and the National Endowment for the Humanities for support (via a grant to the Indiana University Press) in its printing.

T. W. JACOBSEN

PREFACE

When Tom Jacobsen invited me, in the Fall of 1974, to publish the Neolithic pottery from Franchthi, I remember hesitating slightly at the size of the responsibility, and, more seriously, at the three-year commitment we agreed would be necessary to complete the undertaking. Had I known how long the job would really take, I doubt I would have agreed to it. Fortunately, such things are unknowable, for the years with the Franchthi project have been, and continue to be, extremely rewarding. The ceramics still interest me and the Franchthi friendships and experiences have shaped my adult life.

My first study season was in the summer of 1975, in the Leonardo, the storeroom of the Archaeological Museum in Nauplion, Greece, where all the cultural materials from the Franchthi excavations are stored. I returned for four to six weeks of additional study each summer for most of the next ten years. Often I had one or more student assistants working with me: Betsy Anfield, Maureen Carrol-Spielecke, Heather Chepko, Tracey Cullen, Susan Cummings, Sharon DeHoff, Terri Schneider, Patricia Smith, and Lillie Vitelli. Steve Koob directed the conservation work and provided valuable consultation on related technical issues.

Finds from excavations in Greece, of course, must be stored in Greece, in a location approved by the Greek authorities, who also control access to them. We must go to Greece to work with the material — an increasingly expensive requirement — and once there, may work only at the times and in the space allotted to us. Artifacts may not be removed from the approved storeroom. One adjusts goals and procedures to those limitations of time and space.

During the busiest Franchthi study seasons in the Leonardo, the Franchthi staff shared the limited work space with staff from the Asine, Halieis, Mycenae, and Tiryns excavations, as well as from the local excavations of the Greek Archaeological Service. I thank, once again, all those colleagues who, under difficult circumstances, made room for me to spread pottery. I am grateful, too, for the courtesies extended to me by the officials and staff of the Nauplion Ephoreia.

I made the measured pencil drawings of the Franchthi pottery in Nauplion. Ayla Akin, Lorna Gentry, Fran Huber, Priscilla Murray, and Laurie Talalay helped prepare them for publication. Dave Schalliol, of the Mathers Museum at Indiana University, also provided good advice and useful equipment. Jan Abernathie, Leslie Hammond, Ada Kalogirou, and Dimitris Sagias assisted with tables and references. Michael Fotiadis, Cheryl Munson, Curtis Runnels, Jerry Rutter, and Tjeerd van Andel commented helpfully on early drafts.

Tony Wilkinson’s analyses of the Paralia sediments were completed in time for me to use them. He intentionally incorporated little of the cultural evidence in his analyses. By adding the ceramic evidence, I was able to take the interpretation of Paralia a few steps forward, sometimes finding it necessary to disagree with Tony’s preliminary suggestions. Without his analyses and many useful discussions with him and with Catherine Perlès, however, my study of Paralia would have been far more difficult and less rewarding.

Tom Jacobsen had the courage to entrust this important body of material to a raw beginner and gave me the freedom to develop it as I chose. He has been encouraging and supportive, even when some of my bright ideas have gone further out on the shaky limb of prehistory than he thought prudent or justified. Catherine Perlès has been sounding board, critic, booster, and the best of friends. Tracey Cullen, too, has lived through this study with me, from the early days in Nauplion, through all stages of the manuscript, editing and vastly improving it, providing encouragement and inspiration through her own thoughtful and good work on some of the same material. Fred Matson’s influence is evident. Reg Heron has been infinitely patient and supportive. Aspasia Armaou, Anne and Chris Chippindale, Dora Kemp, Lisa Williams — thank you. My gratitude to all of you, and to others too numerous to name, is heartfelt.

Financial support for work on the Franchthi ceramics is gratefully acknowledged from the University of Maryland Baltimore County Faculty Summer Fellowship Program, an NEH Faculty Research Fellowship, an American Philosophical Society travel grant, an Indiana University Graduate School Grant-in-Aid of Research, Prof. John Grimes, the NEH Franchthi Publications Grant (T.W. Jacobsen, PI), the E.A. Schrader Fund of the Program in Classical Archaeology at Indiana University, and the Office of the Vice-President of the Graduate School, Indiana University.

November 1991

INTRODUCTION

Franchthi Cave is located in a limestone headland on the north shore of Kiladha Bay, near the market-town of Kranidhi in the southern Argolid, Greece (see Jacobsen and Farrand 1987:2–3). Excavations were carried out inside the mouth of the cave (the rear of the cave is covered with huge boulders from one or more roof collapses), on the terrace directly in front of the cave (Plan 2), and beyond the terrace on the slopes of Paralia, the modern shoreline (Plan 1, Jacobsen and Farrand 1987:2–9). In Neolithic times the shoreline was about a kilometer to the northwest (van Andel and Sutton 1987:38–44), and a coastal plain traversed by a river spread out below the cave in what is now Kiladha Bay. Coring in the bay has suggested that Neolithic activities, at least in the earlier phases, extended into that now-submerged plain (Gifford 1983, 1990).

A full discussion of the natural setting, its resources, attractions, and demands may be found in van Andel and Sutton (1987) and in more summary form in Perlès (1987:3–17). That setting should not be viewed as a simple backdrop to the human activities of Neolithic times, but as an active participant, strongly influencing and guiding, if not actually determining, the rhythm of daily life. The reader is urged to consult those, as well as the other fascicles in the Franchthi series¹ for a more complete appreciation of prehistoric Franchthi than is provided in this fascicle. A close reading of the stratigraphic analysis of the earlier Neolithic deposits in the cave and on Paralia, presented here in Part II, requires reference to the sections published in Jacobsen and Farrand (1987).

The present fascicle, in addition to the presentation of the classification and phasing, reports on the pottery from the earlier Neolithic at Franchthi: Franchthi Ceramic Phases 1 and 2, equivalent to Early and Middle Neolithic in more general Greek terms.² A second fascicle (Vitelli, forthcoming b) will report on the later Neolithic pottery from Franchthi Ceramic Phases 3 through 5, or Late through Final Neolithic, and the ceramic objects other than pottery and figurines. The figurines are the subject of a separate fascicle by Lauren Talalay (forthcoming).

The decision to present the pottery in two fascicles was taken to make this portion of the study available without further delay. A clear break in the ceramic sequence between Ceramic Phases 2 and 3 (see Chapter 14) determined the end point of this fascicle and the beginning of the next.

This study has been a long time coming, and not solely because of the quantity of pottery recovered from the Franchthi excavations. Certainly there is an abundance of pottery: over two and a quarter metric tons, more than a million pieces, only three of which are complete vessels. But what has really taken time is the slow process of reinventing a way to study pottery from an excavation. I did not realize when I began the study that I would need to do this.

I joined the Franchthi team in 1968, the first season of full-scale excavation, and participated in most field seasons thereafter. The 1970s were an exhilarating time in archaeology generally, and the Franchthi team felt itself at the cutting edge. Team members were full of new questions and approaches, excited by the potential of whole new categories of material and ways of recovering and analyzing them. In the excitement over the new materials, and especially the potential of environmental studies, ceramics seemed rather old hat. As we talked after dinner one night, one of the proponents of the great unexploited potential of lithics said something to the effect of Look at all the people we have washing and numbering sherds, when pottery isn’t really good for anything except chronology, and it isn’t even very useful for that. If we dumped it all at sea, we’d hardly miss it. The dump-it-all-at-sea suggestion wasn’t serious, but I felt a sting of truth in the statement. It became a guiding challenge, along with a desire to share in the excitement of the times, to make pottery tell us more than chronology. But how?

Sorting sherds is an almost instinctive response for an archaeologist. Give one a collection of sherds and he or she will immediately begin sorting the pieces into categories. The ceramic categories in use at Franchthi during the excavation seasons were essentially those in use throughout southern Greece: Rainbow ware, Slipped and Burnished ware, Spongy ware, etc. While I sorted sherds into those categories, I pondered ways to make the Franchthi sherds speak to big important issues. I remember thinking about the origins of agriculture while moving a sherd back and forth among piles, unable to decide where it belonged. Rainbow ware was supposed to have mottled coloring. Slipped and Burnished ware had a smooth, usually shiny surface. Spongy ware was covered with little holes that made it look like a sponge. So where should I put a sherd that was smooth and shiny and mottled on the exterior and covered with holes on the interior?

Eventually, I realized I had a big question sitting on the table in front of me: what did ceramic, or any other archaeological categories, mean? What was responsible for the qualities I used to decide which category a sherd belonged in?

The mottled colors of Rainbow ware were caused by exposure to variations in the atmosphere of a direct-fuel fire. The smooth, shiny surface of Slipped and Burnished pieces was created when the potter burnished the vessel by rubbing the nearly dry, unfired surface with a hard, smooth tool. The holes in Spongy ware were created when bits of calcium carbonate present in the raw clay body were dissolved by acid at some point after the pot was fired, which was often after excavation when the archaeologists soaked the sherds in hydrochloric acid to clean them.

Each of those three ware categories is defined by a different primary variable from a different stage in the history of a vessel. So there can be, and are, sherds from vessels that were made with a clay body containing fragments of calcium carbonate, that were burnished until shiny, then fired in a direct-fuel fire that produced a mottled surface, and were later exposed to acid that dissolved the calcium carbonate, i.e., sherds that legitimately belong in all three ware categories. The decision to classify such a sherd as one ware rather than another is essentially arbitrary. What can the results of such a process of classification mean? What inferences can legitimately be drawn from them? The categories are apples and oranges. They are not comparable.

Reluctantly, I realized I could not just sort the sherds into standard categories first and decide later how I would use the results to address big questions. With considerable trepidation, I abandoned the standard categories and nomenclature and undertook to devise a new system that would be based on variables I could apply consistently to produce mutually exclusive, comparable categories with which I could address questions about human behavior (see Chapter 1).

The classification system that I developed, as well as the descriptions of the resultant categories, is based in and ordered by the materials and procedures used by the Neolithic potter, a term used here to describe anyone who made a pot, with no implication about frequency or exclusivity of the practice. The potter is at the forefront of the following analyses because it is through the potter that we may move most directly from the sherds to human behavior.

Each sherd is a fragmentary pot, each pot, the product of a series of choices and actions taken by a potter. The potter’s choices were governed, at least in part, by the needs and attitudes of the community of which she was a part. Seen in that light, each sherd is a product and reflection of human behavior.

Since 1974 my students and I have been digging clays, discovering and inventing tools as needed to build pots inspired by Neolithic models, and firing them, essentially without the aid of post-Neolithic technology. The practical experience is crucial in learning to recognize and diagnose the Neolithic potters’ choices and procedures. Our experiences as experimental potters are a significant part of the following analyses.

Throughout the text I use female pronouns to refer to the Neolithic potters. Some readers may find that jarring — which is my intent. That Neolithic potters were probably women is an idea that has been around for a long time, although the implications have never been fully considered or explored. Male-female relations are a vital part of every society. One of the few things we know with certainty about Neolithic society is that it was composed of males and females. It is foolish not to make use of good hard fact.

If pottery-making was one of the tasks assigned by sex, as is likely, then the chance that pottery was made by women is roughly 50–50. The ethnographic record improves those odds, since it documents women as the potters in most traditional societies where pottery is produced without the wheel (Arnold 1985:101–108).

Dean Arnold suggests that women are most often the potters because the seasonal, part-time craft is usually household based and women tend to be closely attached to the household by requirements of pregnancy, infant care and other household tasks (1985:101). I think a more likely reason that we should expect women to have made the first pots is that they, because of their role as cultivators and gatherers of plant foods and medicines, had expert knowledge about kinds and locations of soils, including clays; had frequent occasion to experience the properties of clay, especially its plasticity; and no doubt often witnessed and brought about the accidental transformation of clay by fire. Women were, by virtue of their other roles, likely to have need of containers that possessed the particular properties of ceramics. As Autumn Stanley notes, it makes sense that the workers invented their tools (1981:290).

Once I had developed a logic for establishing the ceramic sequence, as well as the system for classification, several study seasons of re-sorting the pottery according to the new system followed before I had sufficient data, ordered temporally, to indicate that the effort had been worthwhile, that patterns in the data were present, clear, consistent, and enlightening. The system of classification used here opens many paths for exploring human behavior in the Neolithic. Some of those complex issues are merely touched upon in the following pages, as time and data do not yet permit their elaboration. Others are explored, if not exhaustively, at least in some depth.

Some phases that represent half a millennium or more are here refined with the addition of subphases that in some cases represent fewer than a hundred years — a generation or two — so that we may think meaningfully about possible interaction among individuals or groups in those subphases. The analysis of where the vessels were made and on what scale raises doubts about the usually assumed household mode of production. A review of the evidence for vessel function concludes that cooking pots were a relatively late addition to the ceramic repertoire, and that most other pots were created primarily for ceremonial purposes.

The find contexts at Franchthi contribute little to considerations of vessel function, but the patterns of comings and goings they document suggest that the earlier Neolithic was not a time of uninterrupted calm and growth. Tensions evident within the Middle Neolithic community may have anticipated the major changes that took place in the Late Neolithic in the Argolid.

As I bring this portion of the Franchthi ceramic study to a conclusion, I sense we are just beginning to discover and exploit the potential wealth of information encoded in prehistoric ceramics.

NOTES

1. To date, the series includes Jacobsen and Farrand 1987 (fascicle 1), van Andel and Sutton 1987 (fascicle 2), Perlès 1987 (fascicle 3), Shackleton 1988 (fascicle 4), Perlès 1990 (fascicle 5), Wilkinson and Duhon 1990 (fascicle 6), and Hansen 1991 (fascicle 7).

2. Throughout the study, Early, Middle, Late, and Final Neolithic are employed in the conventional fashion, to refer to general phases throughout Greece. Franchthi Ceramic Phases are, of course, site specific.

PART I

Theory and Methods

CHAPTER ONE

The Ceramic Classification

INTRODUCTION

The historical moment and my own inclination contributed to setting goals for this study that relate to questions of social and economic organization in Neolithic Greece: how did people at Franchthi live and order their lives, and how did they interact with each other and with their environment? Initially, I expected to build on previous work, using Weinberg’s classification of southern Greek Neolithic pottery (Weinberg 1937), along with subsequent elaborations (Weinberg 1947, 1962, 1965; French 1972; Lavezzi 1973, 1978; Phelps 1975). I planned to sort the sherds from Franchthi quickly into the wares those scholars had defined, perhaps add a few new ones, and provide the sorely lacking stratigraphic basis for the ceramic sequence. Then I could concentrate on the challenges of deciphering people’s lives.

When I attempted to apply the existing scheme to the Franchthi pottery, however, I encountered difficulties and confusion. For example, in sorting a unit of sherds into those familiar wares, I sometimes found that fragments from a single vessel might be assigned to several categories. Often, the exterior of a sherd qualified as one ware, while the interior of the same sherd qualified as a different ware. Intuition told me that one vessel should be assigned to one ware, and to one only. The major problem with the existing ware categories is that different criteria have been used to define each. Thus, no single set of criteria nor any standard ranking can be used to determine the ware assignment when several sets of criteria are present in a single sherd.

After considerable thought about the process of classification and its relationship to goals, I decided that existing Aegean Neolithic ceramic categories were inadequate and inappropriate for the material and opportunities that Franchthi provided. I thus devised a system of classification with a set of variables relevant to the nature of the Franchthi ceramic assemblage and to the specific goals of my study, and that could be applied consistently (Hill and Evans 1972:252–255).

Three salient features of the Franchthi ceramic assemblage distinguish it from other site assemblages and provide unique opportunities for study:

1.  The assemblage from one area (FA) derives from stratified deposits that span most of the Neolithic sequence. Other trenches provide columns through portions of the same sequence and could be used to test and refine the ceramic sequence derived from FA.

2.  Most of the ceramic material has been saved and is stored by original unit of excavation,¹ in contrast with normal practice for excavations in Greece. Thus, virtually the entire recovered assemblage, with information on its contextual associations, is available for study.

3.  Because all excavated sediments were sieved, all sherds larger than 5 mm were recovered. An objective standard for sherd recovery makes possible some meaningful quantifications of the ceramics.

The Greek Neolithic period lasted from ca. 6000 be to ca. 3000 be. In southern Greece the period has been subdivided, largely on the basis of ceramic styles, into four phases, each of which represents a great many generations of human experience. Archaeologists are generally forced to understand all events that occurred within a given phase as contemporary (Cullen 1985a: 188–192). If we could subdivide even a few of those phases to encompass spans approaching a human lifetime, we might begin to talk meaningfully about human interactions. The Franchthi stratigraphic sequences raise the possibility of doing that. Because virtually all the pottery was saved with its contextual information intact, I could examine evidence for the range of ceramic production, use, and discard at one site. Because we recovered and saved all the pottery larger than 5 mm, quantification of the ceramic data, based on consistently applied criteria, reflects actual frequencies in the excavated areas, rather than archaeological recovery and discard practices. The reasonably objective record of sherd frequencies from a number of deposits should permit us, for example, to see if features and objects that have been impressionistically described in the literature as typical and characteristic are indeed typical and characteristic of a given phase, or whether they are simply the most readily recognized or best preserved.

Ultimately I am interested in human behavior. By conceiving of the data, from the first stage of analysis, as manifestations of human behavior, and selecting variables that reflect human choices, it should be possible to address social and economic behavior in the Greek Neolithic with the results of the classification.

The ceramic sample from Franchthi consists primarily of sherds, so the variables used for the classification must be measurable in sherds, as well as in complete pots. Moreover, the variables should be recognizable in every sherd, to take advantage of the level of recovery, produce statistically valid sample sizes, and encompass the full range of ceramic production. The variables should be interrelated and have the potential for a large number of combinations so that innovations and variations over time will be apparent. The variables should be related to human choices based in social and economic decisions so resulting categories can be appropriately used to address the reasons for those decisions. The resulting categories, within the single classification system, should be mutually exclusive.

Variables derived from the pot-making process meet all those requirements. Every sherd is a product of that process and retains some record of it. The potting process involves a large number of intricately interrelated variables that are manipulated by the potter. Each stage of the production process calls for human decisions.

Certain restrictions apply to any potter and have not changed since the first pot was made. A potter must begin with the acquisition and preparation of raw materials before proceeding to build the vessel. Finishing and elaborating the surface necessarily follow forming. Drying and firing complete the requisite stages. Every potter goes through those stages, in that sequence, with every pot. Every potter must also work within the limitations of the technology available for preparing clay, forming, finishing, and firing a pot. Some ceramic technology is, in turn, dependent upon other technology for making tools which the potter might use, such as grinders, mixers, wheels, and fires. Every potter is also limited by his or her own skill, and skill varies from individual to individual, as well as within an individual’s lifetime. Depending on the other variables of the potting process, it may impose greater or lesser limitations on the individual’s ability to produce satisfactory pots.

At the same time, the potting process requires a large number of choices from the potter: Which and how many raw materials shall I use and where and how shall I acquire them? How shall I prepare them? How shall I build what shape, with what tools? How shall I finish the surface of the pot, with what tools and further raw materials? How, where, and for how long shall I dry it? How shall I fire it, and with what fuels and other equipment? How long shall I fire, and at what temperature? And so on.

Each decision the potter makes affects the range of choices at the next point of decision-making. Whether made consciously or not, the choices are grounded in cultural tradition, by which I imply the broadest cultural experience of the potter that molded her sense of taste, style, and rightness or wrongness of actions or appearances.

I assume the Neolithic potter began with an idea of why she was making a pot, that she had some practical or symbolic functions in mind. She was limited in what she could make and how she could fulfil those functional requirements by available raw materials, technology, personal skill, and cultural tradition. A change in any of those limitations could produce a change in the resulting pot.

We can often identify the stage of the potting process during which the potter produced a specific effect or property in a vessel. We should also be able to trace differences among contemporary products or changes in products over time to choices made at specific stages of the potting process. We can then analyze the probable reasons for the choices. Some may be obviously dependent on raw materials, technology, or individual skill. Others may imply a rationale outside the ceramic process, in cultural and environmental factors.

Of the many possibilities within the pot-making process, I have based my classification on the raw materials — specifically, the nonplastic inclusions visible with a 10X hand lens, in the fired clay body, or fabric — and on the finishing of the surface, since those are easily measurable variables from the potting process that are present in every sherd. The clay bodies and surface finishes represented in the Franchthi assemblage proved sufficiently diverse to provide a useful number of categories.

I have also ranked the variables in hierarchical order, based on the sequence of the potter’s choices in making a pot. Raw materials are the potter’s first choice. They are given higher priority in the classification than the potter’s later, and dependent, choices among surface finishes (Tables 1, 2). Other aspects of production, i.e., the techniques and skills of construction, the choices and details of shape and size, and the techniques and execution of decoration, are not used in the classification because evidence for them is not preserved in every sherd. They enter the discussion after the classification is completed (Part III).

A hierarchical classification system provides mutually exclusive categories that support quantification. The number of variables included in this system, together with their ranking, proved sufficient when applied to the Franchthi pottery sequences to define more stages of temporal variation than the four phases defined in prior studies. The variables reflect the Neolithic potters’ choices, so are relevant to discussions of issues that affected those choices. Why potters made specific choices — i.e., the causes of ceramic variation — is one of the provocative questions we may address with the resulting patterns of variation.

RATIONALE AND PROCEDURES

Classes

The Franchthi ceramic assemblage includes both calcareous and non-calcareous ceramics. That distinction forms the first level of the Franchthi ceramic classification.

Rationale

Calcium carbonate derives from shell and various carbonate rocks and is a major ingredient of many Neolithic, and later, ceramic bodies. It occurs naturally in some clay beds. It was, at least sometimes, intentionally added to clay bodies by Neolithic potters (see Chapter 16). When present, it may occur as very fine particles invisible to the naked eye, as larger nonplastic inclusions, or both. Calcium carbonates undergo extensive physical and chemical changes when exposed to temperatures in the range of 650 to 900°C (Shepard 1968:30). Those changes can affect color, hardness, porosity, and the ability of a vessel to survive firing. The nature of the changes depends on, among other things, particle size, the rate of temperature increase, the duration of time at specific temperatures, the atmosphere of the firing, and the presence and behavior of other chemical compounds in the clay body or the fuel. The temperatures that initiate those changes fall well within the range of Neolithic firing temperatures (see below, Firing Procedures).

Few, if any, other minerals commonly present in Neolithic pottery have the potential to affect a ceramic body in so many profound ways at the temperatures achieved by Neolithic firings. Because of the potentially adverse effects of carbonate inclusions in high-fired ceramic bodies, most contemporary potters avoid them (Rhodes 1957:47). Perhaps as a consequence of this attitude, archaeologists are only now beginning to investigate the benefits that might accrue to potter or consumer from the use of calcareous clays and tempers for low-fired ceramics (Grimshaw 1980:353; Maniatis and Tite 1981:75; Rye 1981:127; Jones 1986:751–754; Letsch and Noll 1983:109; Vandiver and Koehler 1986: 205–211).

Procedures

A simple field method was used to discriminate among sherd fabrics. I determined the presence or absence of calcium carbonate inclusions by dipping each sherd in a dilute solution (10%) of hydrochloric acid (HCl) and observing the reaction, if any. An effervescence indicates the presence of carbonates.² I tried to distinguish carbonates that were redeposited during burial from those originally present in the clay body by determining where the reaction was coming from in each sherd. If I could see the particle that was reacting within the sherd, I assumed it was a part of the original clay body. If I could not determine the source of a general fizzing, especially for sherds from the Paralia trenches, where carbonate redeposition is a known and more serious problem than in the cave, I assumed that the carbonate was absorbed through redeposition. If I had difficulty in determining the source of the effervescence, I broke the sherd and tested a new break that had been less exposed to absorption.

Wares

Each ceramic class is subdivided into a number of wares, based on the kind, size, and quantity of nonplastic inclusions.

Rationale

Of the various raw materials in a clay body, nonplastics are relatively easy for the archaeologist to see, describe, and sometimes identify in the field.

Archaeologists imply an aesthetic judgment in their descriptions of fabrics as coarse or fine, but from the potter’s standpoint the inclusion of nonplastics in a clay body is largely a practical concern. A primary role of nonplastics in pottery production is the reduction of shrinkage during drying and firing.

Clay becomes plastic as water coats the surfaces of individual clay particles, lubricating them so they slide over one another. Very fine clays have a large surface area,³ absorb a large amount of water, and are usually highly plastic. Nonplastics are, by definition, larger in size than clay particles. Their presence reduces the total surface area of the clay body particles, which means the clay body absorbs less water and is generally less plastic (see also Rice 1987:54–60). Since a gritty clay body absorbs less water initially, it has less water to lose during drying and firing, and hence will shrink less than a finer clay body. Shrinkage itself, however, is not likely to have been a major concern of Neolithic potters. Rather, they would have worried about the side effects of shrinkage.

During air drying, water evaporates from the surface of a pot. As the surface dries, the particles at the surface move closer together, filling the spaces previously occupied by water, and closing the exit routes for water still present in the core. If the surface particles dry and pack together quickly while the core remains swollen with water, tensions are set up within the walls of the vessel. Those tensions may be sufficient to crack or break the pot. Nonplastics reduce overall shrinkage and, by preventing the tight packing of the smaller clay particles, open the clay body with passages for water to evaporate from within the vessel walls. Heavily gritted clays are less likely than cleaner clays to crack or break from shrinkage tension during drying and firing.

The nonplastics in a clay body also affect the way a potter can manipulate it. Since they reduce plasticity, they are poorly suited to procedures that require highly plastic clays (e.g., throwing on the potter’s wheel). Gritty clays tend to dry out more quickly than cleaner clays while they are being worked, so the relative amount of nonplastics influences the rhythm of the potter’s work. Heavily gritted clays may require special surface treatments to achieve an appearance similar to that of an ungritty clay. Some tempering materials reduce strength (Shepard 1968:27) and increase porosity in the fired vessel. Nonplastics may increase thermal shock resistance, making the fired vessel less likely to break when exposed to a sudden temperature change, such as occurs when a pot is placed directly on a fire. Nonplastics also contribute to chemical and physical reactions during firing. They act as fluxes, reducing the temperatures necessary to chemically and physically alter the clay beyond the slaking point.⁴

Most clays occur naturally with nonplastic impurities. Potters may remove some impurities by extracting the larger ones with their fingers as they are encountered; by sieving liquid or dry clay, perhaps through a basket; or by levigating clay, mixing it with sufficient water to turn it into a suspension, and allowing the heavier, larger particles to settle out. Grinding dry clay, using the same tools and motions that are used in grinding grain, separates out nonplastics that collect around the edge of the grinding stone.

Nonplastics may also be added to a clay (temper). Tempering materials, such as sand, may occur naturally in a form that needs only to be collected and mixed with the clay. Other tempers, such as calcite, may require crushing or grinding⁵ before adding to the clay. Temper may be added to clay at any stage of preparation. If added to a wet plastic clay, it needs to be mixed thoroughly, by kneading or wedging. It reduces plasticity less if left to age, or sour, i.e., to soak in water for at least several days (Grimshaw 1980:450, 505). Some of the pots from Franchthi were certainly made with tempered clay. Although the process of tempering clay is simple, it can be time-consuming. That it was practiced during the early Neolithic indicates that the potters were aware of at least some of the functions of nonplastics.

Procedures

The nonplastic inclusions in the Franchthi sherd material were identified on the basis of visual inspection with a 10X hand lens. Actual clay particles are microscopic in size, measured in microns (thousandths of a millimeter). Descriptions frequently encountered in the archaeological literature suggesting that a clay is fine or coarse are in fact referring to the visible structure of the fired fabric, including the nonplastic inclusions, and not directly to the clay particles themselves.

I used non-technical descriptive terms (for color, shape, size, relative frequency, and distribution throughout the fabric) to distinguish among the different inclusions observed in the Franchthi sherds. I consulted with colleagues, including mineralogists, when they were present in the storeroom and could examine a number of sherds with me. Wherever possible, I have included the mineralogical identification of the nonplastics. In recent years, determination of ceramic composition has become a form of analysis left largely to the materials scientists, whom the archaeologist provides with a selected sample of sherds.⁶ To use a full characterization of raw materials as the basis of my classification system would, ideally, require analyzing all sherds — clearly an unworkable procedure. To submit samples of sherds for characterization and then use the results to classify the remaining sherds is equally unworkable, since sampling implies that one already knows the range to be sampled. Additionally, it is difficult, if not impossible, to go from oxide percentages and other numerically coded results of chemical characterizations back to the field and identify other sherds with those characteristics.

The rough identification of nonplastics provided here is not a mineralogical characterization comparable to a petrographic analysis, nor does it claim to have identified all the raw materials in any specific clay body.⁷ In the Franchthi sample, the nonplastic inclusions that I could distinguish in the field proved to be sufficiently diverse to provide an archaeologically useful classification that reflects a choice of materials by the potters.

Varieties

Varieties are subsets of wares, and are determined by the kinds of surface finish given to each vessel. The term surface finish refers to the potter’s treatment of the surface of a vessel after the shape was formed, from scraping away surface irregularities to various techniques of decoration, and the color to which it was fired.

Rationale

Information about finishing procedures is available in every sherd that preserves at least part of an original surface. The potter’s finishing procedures affect the appearance of a vessel and may also determine appropriate and possible vessel functions. Of the variables included in this classification system, those related to surface finishing reflect most the potter’s skill, sense of style, traditional aesthetic, and any conscious choices related to intended vessel function. The procedures and options exercised by the potters are detailed below.

Procedures

I employed close visual examination of sherd surfaces and drew on my experience as an experimental potter to identify surface qualities and to relate them to the potters’ procedures. I have learned much from published descriptions of the techniques of non-wheel potters. Shepard’s Ceramics for the Archaeologist⁸ has been an invaluable guide. Because the procedures for finishing vessel surfaces, the terms I use to describe them, and the traces they left on the sherds are fundamental to this study, I review them below in some detail.

CERAMIC PRODUCTION AT FRANCHTHI

Building and Finishing Procedures

All the Franchthi pots were built by hand, without the use of a potter’s wheel. Specific evidence for coil building and pinching, the primary procedures used, is presented with the discussion of each variety.

Until the latest phase represented at Franchthi (FCP 5), all sherds reflect some surface treatment beyond that resulting from the building process itself. Most of the Franchthi vessels were scraped, on at least one surface, in the course of building and/or after completion, with any hard, sharp-edged object, such as a sherd, shell, stick, bone, or piece of stone. While scraping, the potter removes clay from high spots along the surface, fills in depressions, and generally removes the surface irregularities of the handbuilding process. While thinning the walls, scraping removes surface traces of previous procedures and leaves behind shallow gouges where nonplastics have been dragged along by the scraper. Irregularities in the edge of the scraping tool may also create shallow striations. As it is used, the scraping tool is scratched and worn down by the nonplastics in the clay body, so the marks it leaves on a surface will change as the tool itself is altered by use.