A Pottery of the Middle Roman Imperial Period in Augst: (Venusstrasse-Ost 1968/69) 3715100028, 9783715100029

Citation preview

Forschungen in Augst 2 A Pottery of the Middle Roman Imperial Period in Augst (Venusstrasse-Ost 1968/69) by William C. Alexander

Ausgrabungen in Augst bisher erschienen:

I.

R u d o l f L a u r - B e l a r t , Insula X X I I I , 1948; Gewerbehallen mit Ölen. 1949. 31 Seiten, 18 A b b . , 1 Planbeilage. vergriffen II. Oswald Lüdin, M a d e l e i n e Sitterding, R u t h Steiger, H u g o D o p p l e r , Insula X X I V , 1939-1959.1962. 88 Seiten, 30 A b b . III. L u d w i g B e r g e r , E i n römischerZiegclbrennofen bei Kaiseraugst. M i t einigen B e m e r k u n g e n z u r T y p o l o g i e römischer Ziegelbrennöfen. 1969. 43 Seiten, 29 A b b . IV. U r s u l a R e i n h a r d t , T e o d o r a T o m a s e v i c , R . M . S w o b o d a , Gräber an der Rheinstrasse und A u s g r a b u n g e n in Kaiseraugst 1968 und 1970. 1974. 101 Seiten; A b b . , T a f e l n und Planbeilagen.

Fortsetzung der R e i h e :

Forschungen in Augst i. R u t h Steiger, G . T h e o d o r S c h w a r z , R i c h a r d S t r o b e l , H u g o D o p p l e r , Insula X X X I , 1960/61.1975. ca. 200 Seiten und ca. 100 A b b . 2. W i l l i a m C . A l e x a n d e r , A Pottery of the M i d d l e R o m a n Imperial Period in Augst (Venusstrasse-Ost 1968/69). 1975. 88 Seiten, 18 A b b . , 12 T a f e l n . zu beziehen d u r c h :

Pro Augusta Raurica CM-4000 Basel, R h e i n s p r u n g 20, oder CH-4302 Augst

Forschungen in Augst 2 A Pottery of the Middle Roman Imperial Period in Augst (Venusstrasse-Ost 1968/69) by William C . Alexander

P o s i t i o n of the kilns (measurement and drawing Ines Matter).

R e d a k t i o n : Jürg E w a l d G r a u w i l l e r Offsetdruck L i e s t a l

Contents Dedication Acknowledgements

5 6

Chronology

7

The Categories of Wares Category I - T h e C l a y G l o s s Wares Category II - T h e C a r b o n S m u d g e d Wares Category III - T h e W h i t e E n g o b e d Wares Category I V - T h e M i c a G i l t Wares Category V - T h e P l a i n Wares Category V I - T h e C o o k i n g Wares

9 9 13 15 16 17 17

The Raw Materials The Clay The White Engobe The M i c a The Temper

18 18 21 22 22

The Forms The Flagon Forms The Jug Forms T h e Bottle F o r m s T h e F i n e Wares The Bowl Forms T h e C o o k i n g Pots T h e Plates T h e M ort aria

24 24 29 29 34 36 38 38 38

The Derivation of the Forms

41

The Kilns

45

The Potters Wheels

49

The Wasters

51

Footnotes Photographs a n d Plates Inventory o f the E x c a v a t i o n Bibliography Zusammenfassung Resumé

55 60 77 80 82 84

Dedication W h e n Prof. D r . R u d o l f L a u r - B e l a r t , the late Conservator o f A u g u s t a R a u r i c a , first wrote to me i n the fall of 1969 proposing that I come to S w i t z e r l a n d to work o n the recent finds i n A u g s t , I was stricken w i t h feelings of m i x e d emotions. F i r s t , I was elated at the possibilities that this u n i q u e opportunity presented; and second, I c o u l d only w o n d e r that m y b a c k g r o u n d and preparation w o u l d be equal to the task at h a n d . F r o m the first day at the m u s e u m , Prof. L a u r was m y advisor, m y m e n t o r and m y guide t h r o u g h the c o m p l e x w o r l d of E u r o p e a n archaeology. Just as his contributions to S w i t z e r l a n d and to the better understanding of her varied past cannot be measured i n finite terms, neither can I put into words the value o f his assistance to me i n this endeavor. It is to h i m that this modest v o l u m e is respectfully dedicated.

William C. Alexander A s s o c i a t e Professor C o l o r a d o State U n i v e r s i t y Fort Collins, Colorado ( U S A )

January 1973

Acknowledgements So m a n y people h e l p e d m e d u r i n g the year that I w o r k e d i n A u g s t researching this paper that a complete list of those to w h o m I a m indebted w o u l d be impractically l o n g . W h e r e v e r I t u r n e d , I was met w i t h the friendliness and spirit o f h e l p f u l cooperation w h i c h typifies the Swiss character. E v e n so, I w o u l d be seriously remiss were I to o m i t reference to a few without w h o m this entire venture w o u l d have b e e n impossible. F i r s t and foremost I w o u l d like to thank Prof. D r . E l i s a b e t h E t t l i n g e r for her k i n d and t h o u g h t f u l reading of the text and for the invaluable suggestions she offered. It was she w h o prepared the condensed translations of the text. M y colleagues, D r . T e o dora T o m a s e v i c , D r . E m i l y R i h a and D r . R u t h Steiger, of the professional staff of the E x c a v a t i o n s and the R ö m e r m u s e u m h e l p e d m e daily i n i n n u m e r a b l e ways and m y debt to t h e m cannot be measured i n words. M y gratitude to M r . W e r n e r Hürbin, T e c h n i c a l Assistant at the m u s e u m , for his expert advice and help i n restoring the finds f r o m the excavation and M r . E r n s t W ä l c h l i , M a n a g e r of the T e s t i n g L a b o r a t o r y at the D a c h z i e g e l w e r k , F r i c k , for supervising the testing o f the raw materials. I a m especially grateful to M r . A r n o l d E i s e n h u t for his careful and sensitive rendering of the P r o f i l e Plates and Illustrations i n the p u b l i c a t i o n . I w o u l d also l i k e to thank D r . J . E w a l d , C o n s e r v a t o r o f the Stiftung P r o A u g u s t a R a u r i c a and the K a n t o n s m u s e u m i n L i e s t a l , t h r o u g h whose efforts this p u b l i c a t i o n came to print.

Chronology O n e o f the first clues i n establishing the dating of the finds f r o m Venusstrasse Ost came i n the f o r m of negative i n f o r m a t i o n i n that only a few of our sherds were covered i n D r . E l i s a b e t h E t t l i n g e n s n o w classic w o r k o n the excavation o f the Frauenthermen i n A u g s t . Since these baths were i n use during the S e c o n d C e n t u r y , only a few sherds f r o m that p e r i o d were deposited and these were found i n m i x e d complexes s u c h as the water drainage canals where n o sequential chronology c o u l d be established. A few o f our f o r m types, particularly the beakers w h i c h were probably used as d r i n k i n g vessels i n the hot baths, were w e l l covered i n D r . E t t l i n g e n s text, but the flagon and bottle forms were entirely m i s s i n g . T h i s i m p l i e d that the sherds stem f r o m the S e c o n d C e n t u r y . Efforts to learn a more specific dating were greatly h a m p e r e d by the scarity of T e r r a Sigillata (TS) sherds and coins. O f the m o r e than 10 000 sherds t a k e n f r o m the excav a t i o n , less t h a n 50 were T S a n d only 8 o f those were dateable. T h r e e T S sherds dating f r o m the C l a u d i a n era were f o u n d i n the service area i n front of the kilns w h i c h was originally f l o o r e d w i t h brick. U n f o r t u n a t e l y , the strata i m m e d i a t e l y above and b e l o w the f l o o r were excavated as one unit m a k i n g it impossible to relocate the sherds vertically i n the excavation and thereby date the floor. I n view of other evidence, however, it n o w seems likely that the C l a u d i a n sherds were in situ at the t i m e that the kilns were b u i l t a n d that the floor was l a i d over t h e m . T h r e e other T S sherds h a d to be discounted because they came f r o m disturbed strata y i e l d i n g contradictory evidence. I n the f i n a l analysis, o n l y two o f the T S sherds were of value i n dating the finds. T h e first of these was a large R i m S h e r d ( R S ; Inv. N r . 68.13945) f r o m a D r a g e n dorff 45 m o r t a r i u m dated i n the second half o f the S e c o n d C e n t u r y . T h e r u b b i n g surface of the m o r t a r i u m shows no sign of wear w h i c h may indicate that it was b r o k e n and deposited shortly after i m p o r t a t i o n to A u g s t . T h e second dateable T S sherd was a D r a g e n d o r f f 38 R S (Inv. N r . 68.14219) dated as H a d r i a n i c to A n t o n i n e . O f the six coins f o u n d , only four were dateable and these had to be discounted since insufficient data were available to relocate t h e m i n the excavation. O u r best sources o f dating proved to be the l o c a l copies of M e d i t e r r a n e a n T S forms and the c o m p a r i s o n w i t h previously dated clay gloss wares. P r o f i l e 21 shows a l o c a l copy of a T S f o r m w h i c h was p r o d u c e d f r o m the reign o f H a d r i a n u n t i l the T h i r d C e n t u r y . P r o f i l e 22 was also p r o d u c e d i n A u g s t and is a copy of the D r . 38 piece m e n t i o n e d above. A s m a l l beaker i d e n t i c a l i n f o r m to P r o f i l e 1 5 A , but somewhat larger, was excavated i n Kaiseraugst i n 1963 (Inv. N r . 63.5835). C o i n s f r o m the same c o m p l e x date this piece i n the second h a l f of the S e c o n d C e n t u r y . R u t h Steiger shows an almost i d e n t i c a l piece i n her report o n the excavations i n A u g s t f r o m 1961; this pot too dates f r o m the late S e c o n d C e n t u r y . B i n s f e l d shows profiles d u p l i c a t i n g 14A a n d 16A a n d dated t h r o u g h coins i n the reign o f M a r c u s A u r e l i u s . P r o f i l e 16A finds other counterparts i n excavations i n A u g s t , the L i n d e n h o f , F a i m i n g e n a n d N i e d e r b i e b e r . E t t l i n g e r ' s examples f r o m A u g s t come f r o m a c o m p l e x not more precisely dateable t h a n S e c o n d C e n t u r y ; D r e x e l places his ware f r o m F a i m i n g e n w e l l after 150 A . D . ; and Vogt reporting o n the excavations of the L i n d e n h o f i n Zürich dates his pieces i n accord w i t h D r e x e l a n d o n the basis of H e l v e t i c sigillata sherds f r o m the same c o m p l e x as not before the last t h i r d o f the S e c o n d C e n t u r y . T h e N i e derbieber sherds c o u l d hardly have been deposited before the early years o f the T h i r d Century. 1

2

3

4

5

8

9

6

7

A study o f the p o s i t i o n i n g of the sherds f r o m the major F o r m T y p e s s h o w n i n the P r o f i l e Plates showed that a l l o f these forms were represented i n b o t h upper and lower strata and suggested that a l l of the sherds were approximately contemporary. T h e references given above and the exact d u p l i c a t i o n of some of our clay gloss forms i n N i e d e r b i e b e r indicate that a l l of the major f o r m types f r o m this excavation were made and deposited i n an 50 year p e r i o d spanning the last three decades o f the S e c o n d C e n t u r y a n d the first two o f the T h i r d , i.e. 170-220 A . D .

T h e Categories of W a r e s T h e excavation p r o d u c e d a surprisingly w i d e variety o f pottery types not only i n terms o f forms but also i n regard to techniques used, a n d during the course o f the investigation it became clear that the techniques used a n d the wares w h i c h resulted f r o m t h e m fell into regular patterns w h i c h a l l o w e d for a positive categorization. T h e descriptions o f categories w h i c h f o l l o w apply specifically to the wares p r o d u c e d i n A u g s t and f o u n d i n the excavation reported o n , but there is considerable overlapping w i t h techniques used i n other areas. I n some cases this necessitated a general discussion to place o u r finds i n a proper context w i t h i n the picture o f R o m a n pottery as a w h o l e . M u c h o f it applies to a l l pottery regardless o f its time or place o f m a n u facture. Category I - T h e C l a y Gloss W a r e s T h e clay gloss coatings o f the G r e c o - R o m a n w o r l d have long intrigued potters, archaeologists a n d scientists alike a n d research into the p h e n o m e n o n , w h i c h began as early as the 1700's, has c o n t i n u e d up to the present day. M o s t o f the investigation has been carried out o n G r e e k pottery since its i r o n reds a n d i r o n blacks p r o d u c e d o n the same pot presented the most d i f f i c u l t t e c h n i c a l p r o b l e m s , ones w h i c h defied rediscovery u n t i l comparatively recent times. T h i s was t r u l y one o f the lost arts o f the ancients. M o s t o f what is presently k n o w n about the G r e e k black p o l i s h applies to T e r r a Sigillata and other clay gloss coatings as w e l l since they are n o w recognized as a f a m i l y o f finishes a l l resulting f r o m the same basic p h e n o m e n o n . T h i s body o f ware, being as it was, a complete t e c h n i c a l mystery f o r a l o n g p e r i o d of t i m e w h e n other areas o f archaeological research h a d achieved a higher plane o f understanding, p r o d u c e d a welter o f inaccuracies, m a n y o f w h i c h are unfortunately still w i t h us today. T h e t e r m »Firnis« is a g o o d example o f this a n d ranks along w i t h the M e s o - A m e r i c a n »Plumbate« as one o f the classic m i s n o m e r s o f a l l times. T h e G e r m a n w o r d Firnis translates into E n g l i s h as varnish. T h e F r e n c h duplicate the error by n a m i n g the ware vernis w h i c h carries the same m e a n i n g . T h e error no doubt arose due to the superficial similarity between the surface texture o f Firnis ware and a varn i s h e d surface. A d m i t t e d l y , most people w o r k i n g i n the f i e l d realize that an organic varnish plays n o part i n the p r o d u c t i o n o f the ware, but we are still harnessed to a t e r m i n o l o g y w h i c h is both technically inaccurate a n d hopelessly m i s l e a d i n g to the l a y m a n . D r e x e l recognized the error i n 1929 w h e n he refused to use the term Firnis. U n f o r t u n a t e l y , he substituted »glazed« (glasiertes Geschirr) w h i c h , w h i l e closer to the fact, is still l a c k i n g i n t e c h n i c a l precision. Since t h e n , others have increasingly resisted the c o n t i n u a t i o n o f inaccurate t e r m i n o l o g y a n d i n recent years suitable substitutes w i t h no m i s l e a d i n g connotations have been used i n scientific papers o n archaeology a n d ceramic research. R . J . C h a r l e s t o n is credited w i t h the first use o f the w o r d »gloss« i n Roman Pottery a n d M a v i s B i m s o n c o n t i n u e d its useage i n her paper o n » T h e T e c h n i q u e o f G r e e k B l a c k and T e r r a Sigillata R e d « . T h e accepted t e r m i nology n o w used i n G e r m a n is Glanztonüberzugwhich translates as »clay gloss coating«. I n point o f accuracy, the G e r m a n t e r m is the better o f the two i n that it states clearly and concisely the appearance o f the surface a n d - a n d this is quite important - the material used to produce it. F o r this reason, I have used the terms clay gloss a n d clay gloss coated throughout. T e r r a Sigillata also deserves some discussion since i t , t o o , is not perfectly accurate i f taken literally. Terra translates as »earth« a n d alludes n i c e l y to the earthy nature 11

18

of the materials f r o m w h i c h it is m a d e , but sigillata means »decorated w i t h s m a l l figures«. T h i s inaccuracy no doubt got its start because m u c h o f the ware n o w referred to as T e r r a Sigillata is decorated w i t h small figures. Sigillata is also used i n reference to the stamps or seals used to produce the figures, but t h r o u g h misinterpretation it is also sometimes extended to i n c l u d e the name stamps used by the R o m a n potters to identify their works. I have seen T e r r a Sigillata translated i n t o E n g l i s h as »sealed earth« w h i c h , due to the h o m o g r a p h , is usually understood to m e a n sealed i n the sense o f closed off or made i m p e r m i a b l e . Strangely e n o u g h , this misinterpretation is technically more accurate i n describing the ware than the correct one of an »earthy material decorated w i t h s m a l l figures«. T h e fact is that m u c h »sigillate« pottery is completely u n a d o r n e d . O v e r the years, the t e r m T e r r a Sigillata has gained such wide acceptance that there is no longer even the faintest hope of rectifying the error, nor is such necessary i n this case since it has evolved i n m e a n i n g to the point that the original requirement for figurative decoration has been entirely lost and it n o w clearly refers to the w h o l e body of ware whether decorated or plain. In ceramic t e r m i n o l o g y T e r r a Sigillata is usually understood to refer to the clay gloss itself w i t h the result that even such things as sewer pipes are sometimes p r o d u c e d i n »Terra Sigillata«. T o a v o i d c o n f u s i o n i n this w o r k , I shall restrict the use o f T e r r a Sigillata to references to the ware and shall use clay gloss w h e n discussing the surface finish. 19

T h e Technique C l a y gloss coatings are best defined as m i c r o s c o p i c a l l y t h i n , v i t r i f i e d or s e m i - v i t r i fied engobes w h i c h produce a decided sheen without the need of burnishing. O n l y the very finest clay particles w i l l produce the effect and the preparation of a clay gloss slip is a process of separating out sub-microscopic clay crystals t h r o u g h selective flotation. Joseph V . N o b l e i n his excellent w o r k o n The Techniques of Painted Attic Pottery p u b l i s h e d electron micrographs of A t t i c clay and the gloss p r o d u c i n g particles w h i c h he extracted f r o m it by flotation. T h e clay itself proved to be exceptionally fine w i t h the largest particles measuring o n l y t e n m i c r o n s i n diameter, but i n the gloss material the particles ranged f r o m two m i c r o n s d o w n to several hundredths o f a m i c r o n w i t h the average crystal size being a few tenths of a m i c r o n . T h i s dramatically illustrates the exceptional fineness needed to produce a satisfactory g l o s s . It is generally accepted that the ancient potters deflocculated their clay w i t h soluble salts extracted f r o m plant a s h . S o m e clays, particularly those h i g h i n i r o n , do not deflocculate easily and require a d d i t i o n a l treatment f r o m such materials as tannic or u r e m i c a c i d , b o t h of w h i c h w o u l d have been available to the R o m a n s f r o m nut galls and urine respectively. E a c h clay reacts somewhat differently and the deflocculant used o n one w i l l not always w o r k o n another. W h a t e v e r the materials and techniques u s e d , the basic concept of separating the fine particles f r o m the coarse remains the same. N o t a l l clays l e n d themselves to the p r o d u c t i o n o f a clay gloss. R e s i d u a l clays i n particular are poorly suited to the purpose since they often lack the fine particles w h i c h are needed. M a v i s B i m s o n d e t e r m i n e d that I l l i t i c clays usually produce the best clay glosses. M o n t m o r i l l o n i t i c clays, despite their exceptionally fine particle size w h i c h may range d o w n into the c o l l o i d a l scale, do not as a rule produce g o o d glosses. T h i s is probably e x p l a i n e d by the excessive shrinkage f o u n d i n most m o n t m o r i l l o n i t i c clays. N o r can a l l clays w h i c h w i l l y i e l d a g o o d gloss be expected to produce the same results w h e n fired. T h e clay glosses used o n the T e r r a Sigillata o f 20

21

22

A r e z z o and L e z o u x and the G r e e k black wares of A t t i c a are usually opaque unless very t h i n l y applied and show a u n i f o r m c o l o r despite m i n o r variations i n thickness. I n the body o f ware designated as »Firnis«, however, the clay gloss is m o r e a semitransparent coating and t h i c k and t h i n areas are often revealed by m o t t l e d c o l o r i n g . Despite the fact that iron is the principle colorant i n a l l of the clay gloss coated wares of the R o m a n provinces, ferric oxide i n the case of the red wares and ferrous oxide i n the black and grey wares, there are easily recognizeable differences i n b o t h hue and c h r o m a f r o m type to type. These variations i n color depend u p o n four factors: the amount of i r o n present, its fineness (and by extension its dispersal t h r o u g h the gloss), the presence of other oxides w h i c h m a y alter its c o l o r , and the firing t e m perature. T h e general rule of t h u m b is, the more i r o n , the deeper or m o r e intense the c o l o r , but two clays w i t h i d e n t i c a l i r o n contents may show decided differences i n hue and c h r o m a due to finer dispersal of the i r o n i n one. M o s t i r o n bearing clays deepen i n c o l o r as the temperature o f the firing increases. T h e oxides o f c a l c i u m , m a g n e s i u m and t i t a n i u m w i l l t e n d to b l e a c h ferric oxide and give a final c o l o r lighter than what might otherwise have been expected. T h e effect of separating out the finest fraction of the clay is that the larger f u s i o n resistant particles and those w h i c h are still f l o c k e d together are e l i m i n a t e d . T h e gloss material then consists almost entirely of flat, plate-like clay crystals w h i c h are excedingly fine. W h e n applied to the piece, they tend to orient themselves i n layers parallel to the surface o f the pot w i t h their flat sides out. E a c h crystal acts as a m i r r o r p r o d u cing a sheen. W h e n f i r e d , usually not higher than about 950° C , the gloss layer sinters f o r m i n g a relatively watertight abrasion resistant surface over the still porous paste. I n most cases, b o t h paste and gloss are made f r o m the same raw material and vary but little one f r o m the other i n their c h e m i c a l make u p . T h e gloss may have a slightly higher alkaline content, due to the soluble salts used as peptising agents, and these are f l u x e s , but this is hardly enough to account for the wide disparity i n sintering points. It is a well-demonstrated fact that finely g r o u n d materials fuse m o r e readily than coarser materials of the same c h e m i c a l c o m p o s i t i o n , and it is w i d e l y accepted that the sintering of clay gloss coatings is p r i m a r i l y attributable to the fineness of the particles. I n reduced wares, the i r o n content is also a contributing factor i n that ferrous i r o n ( F e O ) is a p o w e r f u l f l u x . T h a t the i r o n content plays a significantly smaller role t h a n the fineness of the material is demonstrated by the w o r k of F a r n s w o r t h and W i s e l y . I n their research, they were able to reproduce the »intentional red« w i t h the » G r e e k B l a c k « o n the same pot i n one firing by simple a d d i t i o n o f ochre to the black gloss before firing. T h e a d d i t i o n of ochre a c c o m p l i s h e d two things: first, it i n t r o d u c e d a non-sintering particle of larger size, and second, it increased the i r o n content f r o m 8.5 to ca. 11%. A t the same firing temperature (ca. 850° C.) and i n a red u c i n g atmosphere, the untreated gloss sintered to an i m p e r v i o u s coating whereas the same material w i t h the added ochre r e m a i n e d porous and r e o x i d i z e d w h e n o x y g e n was i n t r o d u c e d into the k i l n . T h i s demonstrates clearly that the particle size of the gloss is of considerably greater importance than the i r o n content. It also seems likely that the natural i r o n content of the clay w h i c h has been c o m b i n e d over periods of geologic t i m e serves better as a f l u x than that w h i c h is subsequently added. H e r e , t o o , the i r o n w o u l d be m o r e finely d i v i d e d and i n more intimate contact w i t h the clay platelets than additions o f free i r o n . Setting aside for the m o m e n t the factor of temperature, we can state that the color, the characteristics and to a degree the quality of the gloss are p r i m a r i l y dependant 23

24

u p o n the raw material used and cannot be completely c o n t r o l l e d by the potter. T h i s leads to differences w h i c h can be of invaluable aid to the archaeologist i n locating wares geographically and by extension i n dating t h e m . A g o o d example of this is the differences w h i c h can be noted between the clay gloss wares f r o m A u g s t and those w h i c h were made i n other areas. A s noted above, the clay gloss o n the body o f ware designated as T e r r a Sigillata (usually l i m i t e d to A r r e tine pottery and most o f the clay gloss wares p r o d u c e d i n G a u l ) is opaque or nearly so. B y contrast, the clay gloss p r o d u c e d i n A u g s t is semi-transparent and shows a definite change i n value depending u p o n the thickness of the gloss. N o r were the potters of A u g s t ever able to produce the h i g h glance and durability w h i c h was achieved i n other pottery centers even t h o u g h the techniques they used were probably very similar i f not i d e n t i c a l . F o r this reason, some of the pots produced i n A u g s t and other areas where the clay d i d not y i e l d an opaque gloss have a m o t t l e d appearance as opposed to the u n i f o r m color of T e r r a S i g i l l a t a . T h i s same effect is responsible for the variation i n value i n the clay gloss wares made i n A u g s t . I n the reduced wares the values range f r o m a m e d i u m grey to b l a c k , and o n o x i d i z e d sherds t h i n areas of the clay gloss are i n distinguishable i n color and value f r o m the paste and t h i c k e r applications are darker. It seems u n l i k e l y that the potters i n A u g s t consciously regulated the thickness of the clay gloss to produce lighter or darker pots, but our sherds do not present a complete picture and this cannot be definitely stated. It is certain, however, that they regulated their firing methods to produce b o t h o x i d i z e d and reduced wares. I n spite o f the fact that a good m a n y of the sherds bearing a clay gloss must be wasters j u d g i n g f r o m the percentage o f wasters i n the other f o r m types, there is a pattern of regularity i n that some forms are always reduced and others are always o x i d i z e d . F o r instance, a l l of our sherds f r o m indented beakers are reduced. T h i s c o u l d hardly be chance. D e s p i t e the fact that our sherds were b u r i e d for nearly two thousand years, and i n some cases are somewhat damaged due to l o n g exposure to h u m i c a c i d , it seems very u n l i k e l y that the potters of A u g s t were ever able to produce a better sheen than that w h i c h remains o n some o f our best preserved sherds. T w o factors point strongly to this c o n c l u s i o n . F i r s t , our best preserved specimens were f o u n d i n the relatively protected c o m p l e x under the f l o o r o f K i l n II i n direct association w i t h a large n u m b e r o f white engobed sherds. T h e engobed sherds f r o m this c o m p l e x show but little deterioration although laboratory tests showed c o n c l u s i v e l y that even weak acids over a short p e r i o d of t i m e seriously weakened the b o n d between paste and engobe. F r o m this we must assume that there was little or no contact between the sherds f r o m this c o m p l e x and the h u m u s a c i d f r o m the o v e r l y i n g strata. S e c o n d , we ran a series of tests o n clay samples f r o m A u g s t i n an effort to extract a suitable clay gloss. E v e n t h o u g h we used m o d e r n methods and c h e m i c a l elutrients w h i c h are highly effective and were unavailable to the R o m a n s , we were unable to produce a greater degree of sheen than that o n the w e l l preserved sherds f r o m the e x c a v a t i o n . T h e paste of the clay gloss wares also showed some variations i n that some o f the pieces were made f r o m a clay w h i c h had been b l u n g e d and settled and some showed the characteristic unslaked particles f o u n d i n the clay of A u g s t w h e n it is not treated (see section o n R a w M a t e r i a l s ) . W i t h invariable regularity, the larger forms w i t h clay gloss surfaces such as profiles 10B, 21 and 22 were made f r o m untreated clay whereas the finer wares - the indented beakers and the decorated pieces of profile 14A - were exclusively made f r o m a f i n e , even textured paste. R e m a r k a b l y , the fine textured 25

26

27

paste proved to have a l o w e r specific gravity and higher absolute porosity than the coarse paste. A s a rule this w o u l d indicate that the more porous ware was fired at a lower temperature, but i n this case that c o u l d hardly be true i n v i e w of the fact that the clay gloss requires approximately the same heat treatment whether applied to a coarse or s m o o t h paste (see section on R a w M a t e r i a l s ) . Categories IA and IB differ f r o m each other i n the m a n n e r of firing the ware. I A pots were f i r e d i n an p r e d o m i n a n t l y o x i d i z i n g atmosphere so that the color is i n the reddish orange range ( M u n s e l l 10 R 5/6) and the I B sherds range f r o m a m e d i u m olive grey ( M u n s e l l 5 Y 5/1) to black w i t h a l l gradations between. T h e paste is very fine textured and lacks the l u m p s w h i c h are charactistic of the clay of A u g s t suggesting that the potters processed this clay i n a special way to insure the quality of the ware. S p a l l i n g due to l i m e chips is not as c o m m o n as i n the coarser wares, but may occur. M a n y o f the sherds f r o m Category I A were rather poorly preserved w i t h o n l y a few traces o f the gloss r e m a i n i n g , but the I B sherds were for the most part i n a fairly good state of preservation. A s a general r u l e , it is possible to say that Category I A sherds are not as resistant to c h e m i c a l attack and weathering as those of Category I B despite the fact that the o n l y difference between t h e m is the m a n n e r of firing. Shepard noted the same p h e n o m e n o n w i t h regard to M e s o - A m e r i c a n P l u m b a t e ware and attributes the difference i n resistance to a difference i n density caused by the greater f l u x i n g action of i r o n i n the ferrous state. T h e same materials were no doubt used for both types o f wares, but there is a regularity i n the way the colors appear o n the pots w h i c h insists that the potters regulated their firings to produce either grey or red wares. T h e forms w h i c h occurred i n Category I A were: 14 A (all sherds except the p r o f i l e d piece), 15 A , 16 A - C , 13B and 1 4 B - E . The forms appearing i n Category I B w e r e : 1 2 , 1 3 A , 14A (profiled piece o n l y ) , 1 4 D , 1 4 E , 1 5 A , 15B and 17. 28

29

Category IC includes all of the rougher clay gloss wares. I n our examples the pieces are always o x i d i z e d and the clay gloss is o f the same color and texture as i n Category I A , but the paste differs i n that it is r o u g h and u n r e f i n e d and shows the characteristic lumpiness o n u n t r i m m e d , u n r i b b e d surfaces. It is the same paste used for the flagon and b o w l forms. A l l of the larger and coarser clay gloss wares were executed i n this t e c h n i q u e ; they were: 1 0 A , 10B, 13B, 1 4 B - E , 2 1 , 2 2 , 2 4 , 2 7 , 2 8 , and 29. Category II - The C a r b o n Smudged W a r e s C a r b o n smudging is an ancient technique c o m m o n l y used for the p r o d u c t i o n of black wares. It has been used the w o r l d over i n p r i m i t i v e and pre-industrial cultures, and is still i n use today i n parts of A f r i c a , A s i a , S o u t h A m e r i c a and M e x i c o . S u c h w e l l k n o w n types as terra nigra, bucchero nero and »Belgian« ware were done i n this technique. T h e color o f carbon smudged ware ranges f r o m grey to a true black and is achieved by impregnating the ware w i t h carbon during the last stages of the firing. T h e r e are a n u m b e r o f methods u s e d , but a l l depend u p o n p r o d u c i n g a thick c l o u d of carbonbearing smoke at the end of the firing cycle w h i c h enters the pores of the clay depositing a residue of carbon effectively t u r n i n g the ware b l a c k . T h e c o l o r of the clay used is, t h e n , relatively unimportant and the potters rely o n the b l a c k e n i n g power of carbon to produce the c o l o r . T h e r e are three c o m m o n l y seen variations of carbon smudged wares and they are the basis for o u r categories. T h e first is b u r n i s h e d ware w h i c h uses an engobe to pro3 0

31

32

duce the sheen - Category IIA. T h e second is also b u r n i s h e d but has n o engobe o n the surface - Category IIB. Category IIC is not b u r n i s h e d . T h e Techniques T h e pottery o f this type made during the G r e c o - R o m a n E r a was largely w h e e l t h r o w n , but the technique used to produce the surface sheen is m u c h older and is a c o n t i n u a t i o n of one o f the oldest surface treatments i n the l o n g history o f w o r l d ceramics. I n surface appearance they m a y closely resemble the black clay gloss wares o f the G r e e k s , but the techniques used were quite different. I n order to f u l l y e x p l a i n the p h e n o m e n o n it is necessary to begin w i t h a b r i e f discussion of the nature o f clay. C l a y is a cryptocrystalline material w h i c h fact was not fully realized u n t i l the advent o f the electron microscope. C l a y crystals are so s m a l l that observation t h r o u g h a n o r m a l optical microscope reveals no pattern or structure. W h i l e there are a n u m b e r o f clay-related m i n e r a l s , the three major minerals associated w i t h the material designated as clay, K a o l i n i t e , Illite and M o n t m o r i l l o n i t e , a l l have a platel i k e crystal structure. T h e peculiar properties o f clay are dependant u p o n this crystal form. In the dry state clay is n o m o r e plastic than sand or any other fine particled material. O n l y w h e n m i x e d w i t h water does it realize its special characteristics. W a t e r m i x e d w i t h clay i n the proper proportions, usually 25-35% by weight, serves t w o purposes. F i r s t it encapsulates each clay platelet a n d acts as a lubricant so that w h e n pressure is applied the crystals c a n slide one against the other w i t h a m i n i m u m o f f r i c t i o n , and w h e n the pressure is released, the surface tension o f the water holds the platelets i n place. H e n c e , the plasticity o f clay. T h e plate-like structure o f the clay crystal also bears directly o n the p h e n o m e n o n o f the sheen p r o d u c e d w h e n clay is b u r n i s h e d w i t h a hard instrument. B u r n i s h i n g causes n o t h i n g more than a physical alignement o f the clay crystals so that each crystal is forced to lie h o r i z o n t a l to the surface o f the pot w i t h its flat side out. E a c h crystal t h e n acts as a tiny m i r r o r reflecting light and causing the piece to shine. T h e water content of the surface at the time o f b u r n i s h i n g is critical. T h e piece must contain enough water to allow the clay crystals to move and align themselves, but not enough to cause any appreciable shrinkage as the piece dries completely. Shrinkage subsequent to the b u r n i s h i n g disarranges the alignment o f the crystals and destroys the s h e e n . W i t h these parameters i n m i n d , it is easy to see that not a l l clays w i l l l e n d themselves to this treatment. I n some clays the platelets are not w e l l separated, but are stacked up i n bundles k n o w n as »flocks« and cannot be aligned sufficiently to produce a sheen. S u c h clays are as a rule relatively non-plastic. I n other cases, the water content needed to a l i g n the crystals by b u r n i s h i n g is sufficient to cause a sheen destroying shrinkage o n drying. D u e to the rarity o f clays w h i c h take a n d h o l d a g o o d sheen w h e n b u r n i s h e d , it has been c o m m o n practice throughout history to use one clay for the body o f a pot and cover it w i t h a very fine grained clay as a p o l i s h coat. These were sometimes treasured clays, sparingly used a n d perhaps i m p o r t e d . T w o characteristics are h e l p f u l i n distinguishing b u r n i s h e d wares. F i r s t , the surface w i l l usually show lines o f striation caused by the p o l i s h i n g instrument. T h e s e are sometimes hardly noticeable a n d depend i n large measure o n the s k i l l o f the potter. If the piece has been p o l i s h e d o n the w h e e l , w h i c h is c o m m o n i n pottery f r o m the R o m a n E r a , the p o l i s h i n g lines w i l l r u n parallel to the t h r o w i n g a n d t r i m m i n g lines and m a y be d i f f i c u l t to distinguish f r o m t h e m . O n h a n d p o l i s h e d pots, the p o l i s h i n g lines are usually somewhat helter-skelter a n d do not f o l l o w the axis o f the p o t , b u t 33

they may be so skillfully done that the entire surface appears to be s m o o t h and u n striated. Identification is further c o m p l i c a t e d by the fact that some pieces w h i c h were finished i n a clay gloss were b u r n i s h e d or t r i m m e d prior to the application of the gloss. T h e b u r n i s h e d or t r i m m e d areas may have a better sheen due to the smoother surface, and this can be m i s t a k e n for b u r n i s h i n g o n top of a p o l i s h engobe. S e c o n d , the necessity of b u r n i s h i n g to produce the sheen negates the possibility of relief or depressed decoration such as rouletting. F i n e relief cannot withstand the pressures needed to a l i g n the clay particles. H e n c e the pots w i l l be largely unadorned and the forms simple w i t h a m i n i m u m of c o m p l e x handles and the like. W h e n fired these polished surfaces retain the sheen i f the firing temperature is not h i g h enough to cause more t h a n a m i n i m u m of shrinkage. T h i s dictates that most wares of this type w i l l be rather soft and porous. T h e b u r n i s h i n g hardens the surface somewhat due to c o n s o l i d a t i o n of the clay particles, but the paste is usually soft and easily scratched. W h e n the clay used for the paste both accepts and retains the sheen f r o m b u r n i s h i n g , a p o l i s h engobe may be o m m i t e d or a slip made f r o m the same clay may be used. T h i s is apparently the case w i t h the piece used for P r o f i l e I I A (Inv. N r . 68.13866). A s m a l l sherd f r o m this pot was refired i n an o x i d i z i n g atmosphere and there was no difference i n color to be seen between paste and surface nor does it seem possible that an engobe was u s e d . F e w sherds of either category I I A or I I B were f o u n d i n the excavation of kilns I a n d I I . 35

36

Category HC pots are u n b u r n i s h e d and have a matt black surface. T h e only pieces adhering to this technique i n this excavation were bowls of Profiles 23 A & B . Category III - T h e W h i t e Engobed Wares T h i s category is reserved almost exclusively for container forms such as bottles a n d flagons although it is also occasionally seen o n such things as incense cups. T h e color ranges f r o m an almost pure white (rare) into a creamy ivory tone ( M u n s e l l 10 Y R 8/4) w h i c h is the n o r m . W h e n it is very t h i n l y applied the engobe has a p i n k i s h cast caused by the red clay of the paste w h i c h shows through. T h e h i g h c a l c i u m content makes the engobe very susceptible to soil acids and the state of preservation is frequently very poor. I n some examples particularly those f r o m the h u m u s layer where the a c i d action is the strongest, the engobe may have been entirely eaten away. Category HI A. W i t h the e x c e p t i o n of plastic considerations such as the p r o f i l i n g o f the l i p and the ridging of handles, these pieces are u n a d o r n e d . T h e engobe here was used i n the classic sense w h i c h is to say that its sole purpose was to change the c o l o r of the ware. It is neither denser nor harder t h a n the paste so it c o u l d not have been expected to waterproof the pot or give it a more abrasion resistant surface. A s w i t h other techniques w h i c h are k n o w n to have been used i n A u g s t , this is a onefire process. T h e engobe was applied to the leather hard pot. T h i s is clearly s h o w n on a n u m b e r of sherds of w h i c h one of the best examples is Inv. N r . 68.13465. T h i s is a sherd f r o m a T y p e 2 A flagon and has four deep scratches o n the under belly. T h e y penetrate the engobe layer and cut into the paste. T h e sharp edges of the cuts positively show that they occurred w h i l e the pot was still damp. O n closed f o r m s , only the outside surface is engobed and this was probably a c c o m p l i s h e d by t u r n i n g the stiff leather hard pot i n a vat of the engobe slip. D r i p s and runs on the inside are frequent.

Category HIB. T h e engobe a n d techniques o f Category H I B are exactly the same as those of Category I I I A w i t h the e x c e p t i o n that the pots have been decorated w i t h a red clay gloss. T h e red pigment used was the same as that used o n the other clay gloss wares and was brushed o n over the raw, u n f i r e d engobe. T h i s excavation revealed only three patterns, but it seems l i k e l y that others m a y be f o u n d i n the course of future excavations. (See P r o f i l e 3 A and illustrations 1 and 2). T h i s technique seems to have been reserved for bottle f o r m s . T h i s sort of f o r m w i t h its w i d e belly and no handles w o u l d have been difficult to use for any practical purpose. M o r e than l i k e l y , they were purely decorative. 37

Category I V - T h e M i c a G i l t W a r e s T h e m i c a gilt wares form a unified group o n l y i n so far that they have a l l been f i n i s h e d w i t h some f o r m of m i c a as a surface decoration. T h e r e are various techniques of firing and at least three different methods for applying the m i c a . G e r m a n language p u b l i cations make a d i s t i n c t i o n between Goldglimmerüberzüge and Silberglimmerüberzüge (gold m i c a gilt and silver m i c a gilt, i n rough translation). T h e d i s t i n c t i o n here is solely the result of the firing t e c h n i q u e , o x i d i z i n g or r e d u c i n g , w h i c h was used. O u r excav a t i o n d i d not f u r n i s h enough sherds of reduced fabric to determine whether there was a regularity or pattern by w h i c h some forms w i t h m i c a gilt were always reduced and others always o x i d i z e d . F o r this reason, I have based the sub-categories o n the m a n n e r of application of the m i c a platelets, and it may be assumed that the three techniques listed c o u l d be used for either o x i d i z e d or reduced wares. Category IVA. T h i s is the simplest and by far the most prevalent t e c h n i q u e of g i l d i n g o n sherds f r o m this excavation. It was f o u n d exclusively o n the insides and rims of b o w l f o r m s , particularly those of Types 18,19, 20 and the plates of T y p e 24. T h e outsides and t r i m m e d surfaces of these forms occasionally show a few scattered flecks of m i c a , usually o n the foot, but so few and so irregular that one must assume t h e m to be u n i n t e n t i o n a l . T h e explanation of this peculiarity is that the m i c a was applied before the pots were t r i m m e d . A n u m b e r of b o w l sherds, Inv. N r . 68.13746 w h i c h was used for P r o f i l e 18E, for e x a m p l e , show f i n e , linear striations w h i c h do not r u n parallel to the t h r o w i n g rings. O n e also occasionally finds imprints where hairs have been i m p l a n t e d i n the surface or the piece and later b u r n e d out i n the firing process. T h e s e two factors dislose that the m i c a was applied w i t h a brush of some sort w h i l e the pot was very soft, presumably just after the throwing process and before it was r e m o v e d f r o m the w h e e l . L a b o r a t o r y tests c o n f i r m e d that this w o u l d have been the opportune m o m e n t . Just after t h r o w i n g , the surface of the pot is fresh and quite soft due to the water used i n the f o r m i n g process. M i c a platelets brushed o n i n water suspension automatically lie flat against the f o r m and partially i m b e d themselves i n the soft clay where they r e m a i n f i r m l y seated after f i r i n g . T h i s technique was reserved for forms o n w h i c h the treated surface c o u l d r e m a i n u n t r i m m e d and needed no further treatment. Category IVB. T h e materials are the same as used i n I V A , but i n this case the m i c a has been applied w h e n the pots were i n the leather hard state. B y this t i m e , the surface o f the pots had hardened to the point that the m i c a platelets c o u l d no longer i m b e d themselves and w o u l d have flaked off unless further treated. T o prevent this they were b u r n i s h e d to adhere t h e m f i r m l y to the surface of the pot. T h i s technique

was needed only o n pots where the surface to be g i l d e d had first to be t r i m m e d . P r o files IOC and D were f i n i s h e d by this m e t h o d . Category IVC Pieces i n this category differ f r o m the others i n that the m i c a particles are carried i n a n engobe layer. A l l o f the sherds o f this type that w e were able to observe were b o w l forms a n d were g i l d e d b o t h inside a n d out. T h e y were n o doubt slipped i n the leather hard state just after t r i m m i n g . T h e scarcity o f sherds o f this type i n this a n d other excavations i n A u g s t indicates that this was not a c o m m o n technique.

Category V - The Plain Wares T h e wares o f Category V were made f r o m the same paste as the other wares o f A u g s t , but their surfaces were left untreated; they carry neither engobes n o r clay gloss n o r have they been b u r n i s h e d . D e c o r a t i o n , i f any, is l i m i t e d to the use o f plastic clay such as the pressed coils seen o n P r o f i l e 8 or the stamps used o n the m o r t a r i a o f P r o f i l e 30. T h e y are as a r u l e , f i r e d under p r e d o m i n a n t l y o x i d i z i n g conditions.

Category VI - The Cooking Wares T h e paste for the c o o k i n g wares differs sharply f r o m that used i n the other wares o f A u g s t i n that it contains a large amount o f tempering materials. T h e temper, w h i c h p r o b a b l y a c c o u n t e d for 20-30% o f the v o l u m e o f the paste, served to o p e n the clay and give it greater resistance to the extremes o f t h e r m a l shock to w h i c h it was subj e c t e d by the u n e v e n heat o f a n open c o o k i n g fire. T h e materials used as tempering agents are discussed i n the section o n M a t e r i a l s and f r o m the basis for the subcategories. Category VIA sherds w h i c h accounted for more than 90% o f all the c o o k i n g pot sherds f r o m t h i s e x c a v a t i o n were g r i t t e d w i t h the sharp q u a r t z sand f o u n d i n Eupererde. Category VIB sherds were tempered w i t h a r o u n d e d , water-washed sand o f m i x e d c o m p o s i t i o n , a n d Category VIC sherds, o f w h i c h only a very few were f o u n d , used a c a l c i u m based grit derived f r o m the fossil remains o f early marine animals. Category VIA and B sherds were made exclusively o n the potters w h e e l , but a l l examples o f Category VIC were h a n d built. O n l y very few sherds o f this last category were f o u n d i n this e x c a v a t i o n , and their p o s i t i o n i n g i n the dig leads one to speculate that they were of an early type not frequently seen as late as the closing decades o f the S e c o n d Century.

The Raw Materials The Clay T h e ruins of A u g s t lie o n one of the terraces w h i c h step d o w n f r o m the J u r a Plateau to the bed of the R h i n e i n the » V « f o r m e d by the confluence of the E r g o l z river a n d the F i e l e n b a c h . E x c a v a t i o n s i n A u g s t always reach »gewachsenen Boden« or o r i g i n a l , undisturbed strata i n a clay bed w h i c h appears to u n d e r l i e the entire area. F r a u Prof. D r . E l i s a b e t h S c h m i d , D i r e c t o r of the Laboratorium für Ur- und Frühgeschichte of the U n i v e r s i t y of B a s e l , made a study of this material and identifies it as »intensiv verlehmter Lösslehm« (Loess sand heavily infiltrated w i t h c l a y ) . E . S c h m i d ' s samples were taken f r o m Insula X X I V i n A u g s t w h i c h is less than one-half k i l o m e t e r away f r o m the potteries o f Venusstrasse Ost where the same f o r m a t i o n crops out i n the lower strata of the excavations. O u r tests were r u n o n clay samples f r o m Insula X X I I where an excavation was i n progress giving easy access to an u n c o n t a m i n a t e d clay deposit. Samples of the clay and sherds f r o m the excavation were sent to the Eidgenössische Materialprüfungsanstalt (the Swiss national materials testing laboratory) where they were tested spectrographically. T h e results of those tests proved conclusively that this was the clay used by the R o m a n potters. T a b l e I shows the results of the tests and lists the elements present i n each of the samples and their relative quantities i n declining order. Sample A was a clay sample f r o m Insula X X I I and it proved to be virtually identical i n c o m p o s i t i o n to the three R o m a n sherds o f Samples B - D . Sample E was an u n f i r e d R o m a n sherd w h i c h , i n e x p l i c a b l y , was preserved i n the excavation. It showed a considerably higher C a l c i u m content due to the presence of s m a l l chips of limestone. A s a rule the clay i n A u g s t w h i c h comes f r o m areas where the overb u r d e n has not been previously r e m o v e d is practically free o f l i m e , but i n open clay pits where the protective overburden has been carried away by erosion or other forces, l i m e chips are a c o m m o n contaminant. A great m a n y o f the sherds f r o m the excavation have spitouts due to l i m e particles, so it is very likely that the potters m i n e d their clay f r o m an area where the clay was already exposed. T h e most likely l o c a t i o n is somewhere along the banks o f the F i e l e n b a c h w h i c h lies o n l y 100 m to the east o f the potteries. C l a y f r o m A u g s t does not slake completely i n water. S m a l l i n c l u s i o n s o f i r o n bearing m i n e r a l s , p r e d o m i n a n t l y l i m o n i t e , have f o r m e d s m a l l concretions w i t h the clay and sand particles and these r e m a i n after soaking as small l u m p s 1-3 m m i n diameter. O n e frequently sees these o n the surface of the pottery particularly o n the surfaces w h i c h were neither r i b b e d nor t r i m m e d . T h e finer wares such as the beakers frequently have a m o r e homogeneous paste w i t h no l u m p s . I n this case the potters probably processed their clay more carefully prior to use and achieved a smoother paste by b l u n g i n g the clay w i t h water and a l l o w i n g the coarser portions to settle. T h e most notable physical characteristics o f the clay are its exceptional plasticity and its astounding dry strength. B o t h of these hint at its fineness. A sieve test showed that o n l y 4.25% of the material was larger than 60 m i c r o n s i n diameter. T h e s e particles were largely quartz sand m i x e d w i t h i r o n a n d some m i c a and none were larger than 0.5 m m i n diameter. T h r o w i n g tests showed the clay to be very plastic; it can be t h r o w n very t h i n l y and shows little tendancy to s l u m p . T h e dry strength is excellent and the greenware can be h a n d l e d without risk o f breakage. 10

1 1

12

13

T a b l e I - Spectrographic A n a l y s e s

Large

Amounts:

Considerable

A mounts:

A

B

C

D

E

Si Al

Si Al

Si Al

Si Al

Si Al

Fe Mg

Fe Mg

Fe Mg

Fe Mg

Fe Mg Ca K Ti Na B

Small

Amounts:

K Ti Na B Ca

K Ti Na B Ca

K Ti Na B Ca

K Ti Na B Ca

Trace

Amounts:

Sr Ba Cr Mn Ni Zr Cu V

Sr Ba Cr Mn Ni Zr Cu V Zn

Sr Ba Cr Mn Ni Zr Cu V Zn

Sr Ba Cr Mn Ni Zr Cu V Zn

Legend:

A = B = C = D = E =

clay test f r o m Insula X X I I Category I C sherd Category I V A sherd Category I B sherd u n f i r e d R o m a n sherd w i t h l i m e chips

T a b l e II - C l a y Gloss Tests

Water Clay NaOH Calgon Volume yield Weight yield percent yield

A

B

C

1 1 360 g 1.5 g

1 1 360 g 1.2 g

1 1 360 g

500 m l 1.0 g 0.278%

360 m l 0.7 g 0.195%

3.5 g 190 m l 2.35 g 0.653%

Sr Ba Cr Mn Ni Zr Cu V

F i r i n g tests were r u n o n the clay f r o m Insula X X I I i n an effort to determine the t e m perature range used by the potters of A u g s t . A l t h o u g h the test kilns were equipped w i t h pyrometers, we used p y r o m e t r i c cones (Seger K e g e l ) f r o m the Seger C o r p o r a t i o n since they m o r e accurately measure the effect of the heat energy o n the ware. T h e test bars were saturated after each firing and the effective porosity calculated and these figures were c o m p a r e d w i t h the results of saturation tests previously r u n o n representative R o m a n sherds. Tests were fired at every cone l e v e l f r o m S K 014 to S K 010. E x a c t d u p l i c a t i o n of results f r o m the saturation tests was not a c c o m p l i s h e d , nor was this expected due to the changes w h i c h c o u l d have o c c u r r e d i n the R o m a n sherds during nearly two m i l l e n i a of b u r i a l and the fact that the R o m a n potters probably fired their wares by eye measurement of the color i n the k i l n or draw tiles and c o u l d not duplicate their firing as accurately as we using the Seger cones. O u r tests d i d , however, show that the R o m a n sherds were fired i n a temperature range of 840° to 880° C or ca. S K O l l a . N o differences c o u l d be detected w h i c h i n d i c a t e d that the various wares fired at differing temperatures. R a t h e r , indications were that wares f r o m all Categories were fired at approximately the same temperatures. D u e to the c o m p l e x i t i e s of the process and the short t i m e available for testing and study, no attempt was made to extract a clay gloss f r o m the clay f r o m Insula X X I I using p r i m i t i v e methods such as might have been used by the R o m a n potters. W e d i d , however, test the clay w i t h m o d e r n m e t h o d o l o g y and were able to extract a clay gloss w h i c h was virtually i d e n t i c a l to the gloss seen o n the best preserved sherds f r o m the excavation. Several different deflocculants and combinations of materials were t r i e d , but the best results came f r o m the recipe given by N o b l e w h i c h calls for V21 water, 2 /2 g C a l g o n and 115 g of clay. I n order to test this material further, we used the clay f r o m Insula X X I I to throw some copies of the R o m a n forms and applied the gloss material to t h e m . A f t e r firing to S K O l l a our forgeries were indistinguishable i n c o l o r , texture and sheen f r o m their R o m a n counterparts. I n an effort to extract a finer gloss f r o m the clay, we ran a series of tests using the recipes s h o w n i n T a b l e II and a l l o w e d the m i x t u r e s to stand undisturbed for one week. O n e liter of each m i x t u r e was used for the test. A t the end of the settling p e r i o d the upper layer of each test varied i n amount. T h e two N a O H tests p r o d u c e d the deepest layers of gloss w i t h 500 and 370 m l respectively and i n b o t h cases the top layer was only slightly m u r k y . T h e top layer of the C a l g o n test was considerably smaller w i t h only 190 m l but was m u c h thicker. T h e gloss material was evaporated to dryness to determine the amount of gloss it contained and the y i e l d of each recipe. I n each case the amount was very s m a l l , as s h o w n o n the chart. A l l three p r o d u c e d a decided sheen, but not perceptibly better than our previous tests and still a far cry f r o m f i n ishes seen at T r i e r and A r e z z o . T h e s e tests i n w h i c h only the very finest particles were separated out for use demonstrate that the clay used as a source determines the character of the clay gloss material (See S e c t i o n o n Category I wares for a fuller discussion of this point). T h e water of plasticity (that amount of water needed to bring the dry clay to the plastic state) is very h i g h and ranges up to 30% and m o r e . T h i s and the plasticity of the clay w o u l d lead one to expect a h i g h shrinkage w h i c h was not the case i n the tests we ran. T h e greatest total shrinkage f o u n d i n our tests was 11% w h i c h is entirely acceptable and m u c h l o w e r than the n o r m for h i g h l y plastic clays. T h e h i g h fraction o f loess sand i n the clay is probably a c o n t r i b u t i n g factor here since loess is an inert, 14

1 5

1

n o n - s h r i n k i n g material and w o u l d tend to retard the overall shrinkage of the clay. I n short, the clay is very plastic, throws w e l l without s l u m p i n g , is hard enough to m i n i m i z e breakage i n the greenware and fires to a suitable density at l o w temperature. It is an almost perfect natural clay body and needs no additions. Saturation tests r u n o n the sherds y i e l d e d some surprising results. T h e o x i d i z e d sherds, i.e. those f r o m Categories I A , III, I V A , and V A , proved to be w i t h i n fractional percentage points of each other i n the amount of water they c o u l d absorb. T h e reduced clay gloss sherds invariably absorbed 2 to 2Vi percent more water i n d i cating a more porous structure. T h i s was an unexpected result since the paste i n the reduced wares is usually finer and these sherds give every appearance of being denser and harder than the o x i d i z e d sherds. T h e explanation probably lies i n the greater f l u x i n g power of i r o n i n the reduced state. It is a w e l l k n o w n p h e n o m e n o n that clay sometimes becomes more porous as the firing temperature is increased f r o m 850° to 900° C . T h i s is caused by a partial sintering of the clay particles during w h i c h they p u l l back f r o m the pores i n the paste increasing pore size. A s the temperature increases even further, sintering continues closing off some o f the pores and the porosity begins to decrease again. I n these sherds, the extra sintering c o u l d easily have been affected by the p o w e r f u l f l u x i n g a c t i o n of reduced i r o n . E i t h e r a higher or lower firing temperature c o u l d also e x p l a i n the difference, but other factors indicated that there was no temperature differential i n this case. A n o t h e r possibility is that the elutriation of the clay for the fine wares w o u l d have e l i m i n a t e d some of the coarser particles of L o e s s sand. T h e s e particles have a 0% absorption and effectively reduce the porosity and increase the specific gravity o f the clay w h e n present. T h e W h i t e Engobe T h e white engobe used i n A u g s t is not a pure white but ranges into a pale cream color. Some sherds are lighter or darker than others, but the great majority fit into the M u n s e l l c o l o r chart at 10 Y R 8/4. A quantitative analysis of the engobe gave positive reactions o n C a O , S i 0 , A 1 0 , F e 0 , and showed M g O to be present as an i m portant trace element. O n e of the strongest reactions came f r o m C a O and indicated that the u n f i r e d engobe contained a m i n i m u m of 25% C a C 0 3 or other c a l c i u m m i n erals. T h e i r o n content gives the engobe its creamy c o l o r and it is prevented f r o m giving an even stronger color reaction by the bleaching action of the C a O i n the fired engobe. T h e r e are two possibilities for the c o m p o s i t i o n o f such an engobe. S o m e natural m a r l s are sufficiently h i g h i n l i m e to produce a creamy white color w h e n fired despite a relatively h i g h i r o n content, or mixtures of clay and C a C 0 , crushed l i m e s t o n e , for instance, can produce m u c h the same effect. O u r attempts to reproduce the color and texture of the engobe were not entirely successful even t h o u g h we tried b o t h possibilities i n a n u m b e r of c o m b i n a t i o n s . M i x t u r e s of A u g s t clay and w h i t i n g ( C a C 0 ) beginning w i t h 80% clay and 20% w h i t i n g and c o n t i n u i n g at 10% integrals up to a m a x i m u m of 50% w h i t i n g were made and fired to S K O i l . A l l o f these were too dark and had a decided reddish hue. A n o t h e r factor w h i c h makes it u n l i k e l y that this sort of c o m p o s i t i o n was used for the engobe material is the difficulties the potters w o u l d have had i n getting a very f i n e l y powdered limestone or other c a l c i u m material. T h e r e are a n u m b e r o f ways that this c o u l d have been a c c o m p l i s h e d , c a l c i n i n g the l i m e s t o n e , for instance, but all are rather complicated not to mention expensive and time consuming. 2

2

3

2

3

16

3

3

U n t i l recent t i m e s , the vast majority of effects p r o d u c e d o n pottery were the results of very simple procedures. T h e fact that the engobe used i n A u g s t was used i n large quantities indicates that it was a readily available material and cheaply w o n . T h e r e is a deposit of m a r l near the east gate of A u g s t w h i c h was m i n e d for c o m m e r c i a l use i n m o d e r n times. It has not been established that the R o m a n s k n e w about this deposit, but it is entirely possible that they d i d . U n f o r t u n a t e l y the pit has been partially f i l l e d i n n o w and it was not possible to get a sample o f the material for testing. W e d i d however test other marls and it was f r o m t h e m that we got o u r nearest a p p r o x i mations of the R o m a n engobe. T h e brick factory i n F r i c k ( K t . A a r g a u ) where we d i d our firing tests uses m a r l f r o m a deposit located adjacent to the plant. T h e best results were p r o d u c e d by simply slaking the m a r l i n an excess of water, b l u n g i n g briefly and a l l o w i n g the unslaked particles to settle out. T h e finer fraction was t h e n decanted, t h i c k e n e d and used as an engobe. A f t e r firing to S K O i l , the color was the same as that of the R o m a n sherds, but a shade or two darker i n value ( M u n s e l l 10 Y R 8/3). W e can be certain that the potters of A u g s t d i d not use the m a r l f r o m F r i c k , but we can be almost equally certain that they d i d use a m a r l , either f r o m the pit near the east gate or f r o m some other l o c a t i o n i n the i m m e d i a t e v i c i n i t y of A u g s t , for the white engobe. The M i c a T h e m i c a used o n the m i c a gilt wares was i d e n t i f i e d as muscovite. T h i s is a very c o m m o n m i n e r a l and w o u l d have been available to the potters i n A u g s t f r o m a n u m b e r of sources. T h e most l i k e l y source of supply w o u l d be one of the deposits of decaying granite i n the area. T h e r e are a n u m b e r of these particularly o n the right side o f the R h i n e not far f r o m A u g s t . I n such a deposit, the o r i g i n a l stone has c r u m b l e d due to the forces of nature to a coarse sand. T h e m i c a is easily separated out by stirring the material i n water. T h e sand, largely quartz a n d feldspar, settles q u i c k l y , but the platelike structure of the m i c a keeps it i n suspension longer so that it can be decanted off w i t h the water. F o r c o m p a r i s o n , l e p i d o l i t e , a l i t h i u m m i c a , was also tested, but its f u s i b i l i t y even at l o w temperatures caused it to melt rather than retain the platelike crystal w h i c h is necessary to the gilded effect. T h e Temper O n l y the c o o k i n g wares showed definite signs that a tempering material had been added to the paste. M i n e r a l determinations for each of the three types o f t e m p e r i n g materials were made by E l i s a b e t h S c h m i d . T h e c o m m o n e s t type of temper used i n c o o k i n g pots is a sharp quartz sand. T h e most l i k e l y source of this k i n d of sand i n this region is a material k n o w n as »Hupererde«. Hupererde is a fine plasic clay w h i c h naturally contains a h i g h percentage of quartz sand. T h i s material may have b e e n used unaltered for the c o o k i n g pots, or it may have been b l e n d e d w i t h A u g s t clay to obtain the proper w o r k i n g , firing and c o o k i n g properties. T h e second type of temper is a r o u n d e d sand c o m p o s e d p r i m a r i l y of quartz and f e l d spar, but also containing a n u m b e r of other m i n e r a l fragments. T h e r o u n d i n g of the sand particles inicates that it is a c o m m o n river sand and it c o u l d have been collected i n the F i e l e n b a c h , the E r g o l z or i n the bed o f the R h i n e itself.

W e also f o u n d temper. Some S c h m i d as the aquatic life i n available.

a few sherds f r o m h a n d m a d e c o o k i n g pots w i t h a c a l c i u m m i n e r a l s m a l l s h e l l fragments were still v i s i b l e and were i d e n t i f i e d by E . fossil remains of Cocinopora infundibuliformis w h i c h was a f o r m of the J u r a s s i c P e r i o d . T h i s m a t e r i a l , t o o , w o u l d have b e e n l o c a l l y

The Forms T h e Flagon Forms T h e m a i n body of flagon fragments falls into two closely related types. T h e smaller o f these appears i n six graduated sizes and is represented by more t h a n fifty neck and r i m fragments. O n l y three of the pieces were reconstructable f r o m the o r i g i n a l sherds to a degree w h i c h made exact profiles possible (see Profiles 2 A , 2 B and 2 C ) . E v e n so, the other sherds were sufficiently complete that there can be no doubt that the forms were m o r e or less i d e n t i c a l i n their proportions w i t h the possible e x c e p t i o n of the largest size of this type (2C) where the w i d e r m o u t h r i m may have necessitated some change i n f o r m to achieve a pleasing balance of proportions. T h e s e pieces are typically w i d e r t h a n they are tall giving the body a distinctly o v o i d f o r m . T h e neck is quite constricted and is s u r m o u n t e d by a collar w h i c h is slightly c o n i c a l . T h e majority of the pots have two handles each p r o f i l e d w i t h three ridges, although there were three examples w h i c h had three h a n d l e s and one w h i c h had o n l y o n e . O n the largest pieces o f this type, the handles have o n l y two ridges a n d are s i m p l y smaller version of the handles seen o n Profiles 1 A , I B and 1 C . A l l sherds f r o m F l a g o n Types 1 and 2 showed at least some traces o f the white engobe used o n Category III wares. 38

39

40

T h e best preserved s p e c i m e n of this type is s h o w n i n P r o f i l e 2 A . A l l but a p p r o x i mately one-third of the b e l l y was reconstructed f r o m original sherds. C o n s i d e r i n g the extreme swelling of the b e l l y and the exceptional thinness of the walls, w h i c h average o n l y 3,5 m m i n most sections, it is remarkable that it was made i n one section. It remains as a testament to the t h r o w i n g abilities of the potters of A u g s t . T h e f i n i s h i n g of the neck is very w e l l c o n t r o l l e d and the collar at the m o u t h was a c c o m p l i s h e d by a m i n i m a l t h i c k e n i n g at that point plus a slight, but noticeable f o r m change o n the inside. A s the P r o f i l e shows, the basic f o r m of the pot is that of a rather flattened oval. T h e line o f the shoulder breaks over almost to the h o r i z o n t a l before reversing itself and sweeping up into the c y l i n d r i c a l neck. T h e curve of the under-belly does not flatten to the degree seen o n the shoulder. T h i s gives an uplift to the f o r m a n d , c o m b i n e d w i t h a slight curve reversal at the foot, gives the piece an elegant stance and avoids the dumpiness so often seen i n forms of this sort. T h e handles are set o n i n the approximate m i d d l e o f the neck and rise slightly before curving over to the shoulder insertion. T h e t r i m m i n g o f the foot, w h i c h is very expertly d o n e , shows a peculiarity w h i c h was apparently one of the trade marks of the potters o f A u g s t . T h e central p l a i n of the foot is not s i m p l y recessed, but is set off by a m o r e deeply depressed groove just inside the foot ring. T h i s same treatment was observed o n a l l of the base sherds f r o m this type of flagon and o n those o f Profiles 1 A , 1 B , and 1 C . R e c o n s t r u c t i o n of this piece was considerably hampered by sherd warpage and because of this some of the sherds w h i c h were reassembled c o u l d not be attached to the reconstructed pot. T h e neck of the piece no longer sits upright a n d one shoulder is visibly lower than the other. T h i s c o u l d not have been the case w h e n the pot was t r i m m e d else the t r i m m i n g c o u l d not have been so expertly done. S u c h severe warpage after t r i m m i n g w h e n the pot was i n a stiff leather hard state w o u l d very l i k e l y have caused cracking i n the raw pot such that it w o u l d have been discarded at that point. N o r can we assume warpage due to the pressures of o v e r b u r d e n , since the sherds were not deeply b u r i e d and were i n a relatively protected area u n d e r the f l o o r o f K i l n II. M o r e l o g i c a l l y , the piece was b r o k e n during the firing of the k i l n and is a firing failure. 41

P r o f i l e 2 B shows a smaller v e r s i o n of this f o r m type. It is u n i q u e i n that it is the only piece of this type that shows any signs o f having been decorated. T h e white engobe, and hence the »decoration«, is very poorly preserved, but i n those areas where the reddish pigment is still v i s i b l e , there does not appear to be any regular pattern. R a t h e r , there is a r a n d o m quality w h i c h suggests that this was the result of a shop accident i n w h i c h some o f the red clay gloss was spilled over a finished piece of greenware. P r o f i l e 2 C shows the largest f o r m of this type f r o m the excavation, and the o n l y size w h i c h was t h r o w n i n two sections. T h e handles are broad and straplike and have only two ridges. T h e second major flagon type was represented by fragments of at least eight large pieces, three of w h i c h are depicted i n P r o f i l e 1 A , I B and 1 C . These flagons were made i n two sections. T h e body of the pot was t h r o w n first and allowed to stiffen somewhat before m o r e clay was added and f o r m e d into the neck and m o u t h r i m . H e r e again, the basic f o r m of the pot is that of an oval, showing a strong relationship to the smaller flagons of P r o f i l e 2. T h e major difference between the two forms other than size lies i n the f i n i s h i n g of the m o u t h r i m w h i c h i n P r o f i l e 1 flares rather than constricting as i n P r o f i l e 2. C l o s e e x a m i n a t i o n of the sherds showed that four of these pieces were made by one potter and four by another. B o t h potters were o b v i o u s l y w o r k i n g toward the same f o r m i d e a l , but their different w o r k habits and particularly the disparity i n their techn i c a l abilities made it possible to distinguish between t h e m . F o r ease i n reference I have designated t h e m as Potter I (see P r o f i l e 1 A ) and Potter II (see Profiles I B and 1 C ) . Seven factors can be used as determinants i n distinguishing their w o r k .

1. The Necks. I n pots made by Potter I, w h o was by far the more s k i l l f u l of the t w o , the added necks are set o n so skillfully that it is almost impossible to t e l l where the j o i n t was made. E v e n the inside surface is carefully smoothed over and the j o i n e r y is so expertly done that none of our f o u r examples cracked at this point. Potter II, by contrast, was not at the least at pains to effect a s m o o t h transition between shoulder and m o u t h r i m and his added necks rise angular and f u n n e l - l i k e f r o m the almost h o r i z o n t a l shoulders. T h e j o i n t is quite apparent f r o m the inside and may show a deep ridge as indicated i n P r o f i l e 1 C . C r a c k i n g at this point is a c o m m o n flaw i n his w o r k . 2. The Handles. H a n d l e s made by Potter I show an irregularity (read n o n - u n i f o r m i t y ) w h i c h is a result of their having been p u l l e d i n d i v i d u a l l y . T h e edges are r o u n d e d i n a way that makes Potter F s two ridged handles l o o k almost l i k e two oval straps j o i n e d at the edge. T h e handles broaden somewhat at the shoulder insertion and to a lesser degree where they j o i n the neck. T h e central depressions d i v i d i n g the two ridges are weaker toward the shoulder insertion sometimes playing out altogether, whereas they m a i n t a i n f u l l depth toward the top and disappear u n d i m i n i s h e d under the neck r i m . These two factors indicate that they were attached upside d o w n , i.e. as Potter I stroked t h e m f r o m a l u m p of plastic clay, the b o t t o m t a i l , w h i c h is usually t h i n n e r , became the neck insertion and the t h i c k e r top section was attached at the shoulder. A n o t h e r characteristic of these handles is a r o l l of clay going d o w n the underneath side of the handle o n the right. T h i s is not always seen, nor is it always just o n the right side - it may be o n both. T h i s aberation is a result o f the p u l l i n g process and dovetails neatly w i t h the postulation that the handles were attached upside d o w n . A s s u m i n g that

Potter I was right h a n d e d - percentages favor the a a u m p t i o n - he w o u l d have h e l d the clay i n his left hand and stroked it d o w n w a r d w i t h his right, elongating the clay into a handle f o r m w h i c h was slightly t h i c k e r at the top. T h e right edge o f the handle (later to become the left) w o u l d t h e n be c o n t r o l l e d by the flap of s k i n between t h u m b and forefinger o n the right h a n d . T h i s edge is usually easier to c o n t r o l . T h e left edge (right edge w h e n inverted and attached to the pot) w o u l d have been f o r m e d by the crook of the forefinger. T h i s surface is more d i f f i c u l t to c o n t r o l and the result i n the case of Potter I is the ridge so often seen o n his handles. O n e last factor is the central depression of the h a n d l e , w h i c h is usually not just a neat dip between the two oval ridges; rather it has a s m a l l sharply raised ridge r u n n i n g just off center the length of the depression. T h i s ridge was caused by the potter's t h u m b n a i l as he stroked d o w n ward cutting the central l i n e . Potter IPs handles are just as distinctive i n an entirely different way. T h e y are m u c h flatter and more rectilinear i n profile and lack the subtle r o u n d i n g of handles by Potter I. O n e side o f the h a n d l e , always the right i n these examples, is squared off and may even have rather sharp edges. One's first i m p r e s s i o n is that these handles, t o o , flare out as they approach the shoulder i n s e r t i o n , but a more careful e x a m i n a t i o n shows that this is the result of b e n d i n g the outer edges u p w a r d rather t h a n an actual change i n w i d t h . T h e central depressions m a i n t a i n f u l l strength f r o m top to b o t t o m ending only where they are smeared out i n the process of attaching. T h e s e factors show an entirely different technique of manufacture. Potter II almost certainly threw his handles o n the potters w h e e l . H e w o u l d first have t h r o w n a simple c y l i n d e r of the appropriate thickness. T h e n w i t h the w h e e l t u r n i n g s l o w l y , he drew a h o r i z o n t a l line just under the r i m u s i n g a stick or some other t o o l . T h i s became the central depression. T h e n using a sharp p o i n t e d i n s t r u m e n t , he cut a strap f r o m the top of the cylinder. T h e upper r i m of the c y l i n d e r ( w h i c h became the left side of the handle) c o u l d be smoothed and r o u d e d , but the b o t t o m or right side of the handle w o u l d be flat and angular as seen o n his pots. 3. The Mouth Rim. A t least four factors make it possible to distinguish between Potter I and Potter IPs m o u t h rims. F i r s t , the general character of Potter I's t h r o w i n g is precise and w e l l c o n t r o l l e d whereas Potter IPs w o r k is always a bit inexact at best. S e c o n d , the profiles differ consistently i n one factor. Potter IPs necks are simple spindle forms s u r m o u n t e d by a collar ( m o u t h r i m ) . Potter I, o n the other h a n d , always adds a ridge about 1 c m b e l o w the m o u t h r i m and seems to use this as a guide i n attaching his handles. T h i r d , the m o u t h rims themselves are different. Potter I's rims are a bit taller and flare slightly at the top. T h e y are decorated w i t h two depressed l i n e s , one just below the top o f the r i m and one just above the b o t t o m , w i t h flat band between. I n Potter IPs w o r k , the rims are not as t a l l , the lines are not as accurately placed and m a y be o m i t t e d entirely as o n P r o f i l e I B . 4. The Foot. B o t h potters use the same style of foot o n their f l a g o n , i.e. the same style as described o n P r o f i l e 2 A , and the differences are mostly matters of w o r k m a n s h i p . A s one w o u l d expect, Potter I's feet are precisely t r i m m e d and any roughened edges w i l l usually be seen to have occurred after t r i m m i n g and as a result of the slipping operation. T h e depressed groove setting off the outer ring of the foot is semicircular or nearly so. A Potter II foot may show ragged edges as a result of t r i m m i n g and the depressed groove is more angular.

5. The Finger Ridges. T h e finger ridges inside a Potter II pot are usually quite deep cover the entire surface. Potter P s w o r k is m u c h smoother inside and the finger ridges are m u c h less p r o n o u n c e d . T h e deeper finger ridges indicate that Potter II struggled w i t h the clay and used a great deal of pressure to raise and t h i n the pot. T h i s is not u n c o m m o n w i t h potters whose t h r o w i n g technique is a trifle faulty. B o t h potters probably used a rib to c o n t r o l the outer surface, so distinctions here w o u l d be difficult to find.42

6. Craftsmanship. A s repeatedly suggested i n the preceding n u m e r a l s , Potter F s craftsmanship is excellent. H i s w o r k shows a mastery of technique w h i c h is not approached i n that of Potter II. 7. The Forms. A l l of the preceding analyses are based o n one type of pot, a wide b e l l i e d flagon w i t h a decorative lip and two or m o r e handles and some variations are to be expected i n h a n d work. Craftsmanship plays a large role i n this and one's ability to c o n t r o l the material is decisive i n d e t e r m i n i n g whether or not a desired f o r m w i l l be achieved. B u t there is also that intangible factor of aesthetic sensitivity. T h e most s k i l l f u l of potters m a y create works w h i c h are technically perfect but decidedly lacking i n aesthetic merit. T h i s is not the case w i t h Potter I. H i s w o r k is not o n l y remarkable f r o m the t e c h n i c a l aspect, but his forms are posessed of a grace and elegance w h i c h bespeak the best o f pottery m a k i n g whatever the t i m e or place. B y c o m p a r i s o n , Potter II is a m e d i o c r e potter at best. T h e forms are somewhat d u m p y and lack luster. T h e r e is little or no d i s t i n c t i o n to be made between the materials (clay and engobe) used by the two potters. T h e y are for a l l intents and purposes the same. T h i s and the obvious similarities i n f o r m made it l i k e l y that all of these pieces are the products of the same w o r k s h o p . It is l o g i c a l to assume that Potter I was the master and Potter II was an apprentice or perhaps a slave. A l l three of the pots of this type f r o m w h i c h we f o u n d sufficient sherds for a partial reconstruction are wasters and a l l failed i n the same way as discussed i n the section o n Wasters and F i r i n g F a i l u r e s . P r o f i l e 1 A defines the type for this f o r m as executed by Potter I and shows a l l the characteristics discussed above. A s i n P r o f i l e 2 A , the curve of the shoulder flattens considerably more than that of the under b e l l y , and the general character of the w o r k manship strongly suggests that b o t h pieces were made by Potter I. T h e added neck section is perfectly executed and the line is u n b r o k e n f r o m foot to r i m displaying great s k i l l i n t h r o w i n g and a prior concept o f the f i n i s h e d f o r m w h i l e t h r o w i n g the body section. E v e n i n s i d e , the j o i n e r y of the neck is scarcely discernable and is betrayed by o n l y a slightly t h i c k e n e d ridge at the point of j u n c t u r e . T h e smoothness o f the inside walls and their thinness are remarkable i n a piece of this size. T h e average thickness is ca. 4 m m w h i c h gradually increases toward the b o t t o m to a m a x i m u m of 7 m m . T h e t r i m m i n g of the foot maintains the excellence of the rest of the piece and shows that the pot was perfectly centered w h e n this operation took place. T h e two handles do not f o r m i d e n t i c a l profiles and part of this difference is probably e x p l a i n e d by the fact that one side of the pot was dented during the firing process. T h i s dent, caused by pressure o n the shoulder and one h a n d l e , is evidence that the piece broke i n the k i l n . Because of sherd warpage s u r r o u n d i n g this dent, a complete reconstruction of the pot was not possible. Profiles I B and 1 C show examples of the f o r m as executed by Potter II. N e i t h e r piece

is as w e l l done as P r o f i l e 1 A and both exhibit the differences i n technique discussed above. Sherd warpage shows I B to be a firing failure. P r o f i l e I D , I E and I F show other variations of the f o r m . I E is an unicum, but there were three examples of P r o f i l e I D and four of I F . P r o f i l e 8 is unique i n that it is the only non-intrusive neck fragment i n Category V A (all others are coated w i t h the white engobe of Category III A ) and it is decorated w i t h finger pressed coils o f clay. T w o base sherds t r i m m e d i n the same manner as the foot of P r o f i l e 1 A , but also l a c k i n g the engobe surface were f o u n d i n the same l o c a t i o n as were several w a l l sherds w i t h c o i l decoration, but too few pieces fitted together to a l l o w for a complete reconstruction. T h e general character of the workm a n s h i p , particularly i n the f i n i s h i n g o f the m o u t h r i m w i t h its carefully p r o f i l e d ridging and the presence of its extra ridge m a r k i n g the level where the handles are attached, suggest that this may be the work of Potter I, although the handles are less precisely made than those seen o n P r o f i l e 1 A . T h e paste, however, proved to be i d e n t i c a l w h e n e x a m i n e d under h i g h m a g n i f i c a t i o n . S o m e of the wall sherds show occasional flecks of m i c a identical to those seen o n the bowls of Category I V A wares, but they are too few to be intentional and too large to be natural occurances i n the clay. T h i s w o u l d suggest that the same pottery produced wares i n Category III A , I V A and V A simultaneously. V o g t shows drawings of two w a l l sherds with similar dec o r a t i o n . H e describes his first example as having come f r o m a b r o w n storage jar (braunes Vorratsgefäss), but it seems equally likely that the original piece may have been a flagon, but not recognizable as such because the neck was missing. P r o f i l e 9 was pieced together f r o m eight sherds all of w h i c h were badly warped i d e n t i fying this piece as a firing failure. T h e strong relationship to Profile 1 A , I B and 1C is readily apparent and we may consider this piece to be a variation thereof. T h e r e are two raised ridges of clay a r o u n d the neck and upper shoulder, both of w h i c h lie inside the arch made by the handles. These added coils o f clay are further decorated w i t h s m a l l diagonal indentations. T h e s e are so regular that they c o u l d hardly have been done by h a n d and were probably pressed o n w i t h a roulette s i m i l a r to that used to decorate the clay gloss wares. D r e x e l discusses this type of decoration as having been used d u r i n g the first half of the Second C e n t u r y . O u r piece comes f r o m an undateable c o m p l e x , but its f o r m relationship and the similarity i n w o r k m a n s h i p make it l i k e l y that it was made d u r i n g the second half of the century. Its white engobe and the fact that it is clearly a firing failure label it as a product of A u g s t and belie Schoppa's supposition that this style of decor was l i m i t e d to the area a r o u n d M a i n z . T h e workm a n s h i p , the technique used i n adding the neck and the style of the handles make it highly likely that this piece was the work of Potter II. T h e interior groove m a r k i n g the j o i n t between shoulder and added neck w h i c h characterizes his work is readily apparent here as is the flat side of the handle showing it to have been t h r o w n rather t h a n p u l l e d . T h i s raises an interesting speculation w i t h regard to the dating of the various strata i n the excavation. F o u r clearly identifiable examples of Potter IPs w o r k were f o u n d i n the lowest levels of the dig just south of K i l n II i n quadrants A 3 and 4 and B 3 and 4 whereas this fragment comes f r o m the h u m u s layer. If this piece is the w o r k of Potter II, w h i c h cannot be stated without some e q u i v o c a t i o n , the i n d i c a t i o n is that the entire depth of the dig cannot represent a time span of more than the »working« life of one m a n , w h i c h is to say 20 to 30 years i n this case. M y feeling is that the major p o r t i o n o f the sherds were deposited over a relatively short time p e r i o d . A c c e p t a n c e of this supposition must be tempered by the knowledge that it is u n 43

44

4 5

supported by any data other than the fact that we were unable to establish a sequential c h r o n o l o g y for any of the major f o r m types f r o m the excavation. T h e Jug Forms Fragments of at least 11 j u g forms representing two distinct neck profiles were f o u n d . T h e one u n i f y i n g factor is that i n every case the paste is hard-fired a n d contains a surprizingly large percentage of sand temper. T h e tempering sand is visually the same as that used i n the c o o k i n g pot paste. S o m e of the other forms show s m a l l amounts o f the same sand, but i n no case is the paste as heavily gritted as i n the c o o k i n g pots and these jugs. T h i s raises the question o f the purpose intended for these pots. T h e y w o u l d hardly be suitable for c o o k i n g vessels w i t h their narrow necks. T h e added or p i n c h e d spouts indicate that they were designed to dispense a l i q u i d o f some sort, but offer no explanation for the excess a m o u n t of temper. A l l but one of the sherds f r o m P r o f i l e T y p e s 6 and 7 were f o u n d i n the h u m u s layer and probably date f r o m the T h i r d C e n t u r y . T h i s is supported by the fact that s i m i l a r forms i n bronze are c o m m o n l y f o u n d i n deposits of the T h i r d C e n t u r y . S o m e o f the b o t t o m sherds show string marks and none are t r i m m e d although some have been s m o o t h e d w i t h the fingers. M o s t o f the sherds are rather light c o l o r e d ( M u t i seli 10 Y R 5.5/3), but a few are smudged black i n the firing process. N o n e show any sign o f use. Profile 6 is a composite reconstructed f r o m the sherds f r o m three different pots. N o n e of our fragments were sufficiently complete for a w h o l e p r o f i l e , but there is no doubt that this reconstruction gives an accurate picture of the o r i g i n a l f o r m . C o u n t i n g handle fragments alone, there are at least six pots o f this type represented. T w o handle fragments, one b l a c k , the other as above, have s m a l l c o n i c a l t h u m b lugs just where they curve to descend to the shoulder attachment; they may or may not have come f r o m similar forms, but the paste is m u c h the same. P r o f i l e 7 is u n i q u e i n the d i g , and the p u l l e d spout and the two-ridged handle distinguish it f r o m T y p e 6. H e r e again, the paste is very heavily gritted, so m u c h so that it caused some cracking i n the wet clay w h e n the potter p u l l e d up the edges of the lip to f o r m the spout. The Bottle Forms B y far the most distinctive and i n t r i g u i n g finds of the entire excavation are the decorated bottle forms s h o w n i n Profiles 3 A and 3 B . P r i o r to the excavations o n V e n u s strasse Ost i n 1968, only two other examples of this style of decoration had been f o u n d i n A u g s t . T h e flamboyant i f somewhat haphazard decoration w h i c h girdles the neck and upper shoulder and distinguishs these pieces is done w i t h the red clay gloss seen o n the finer wares of Category I A , and these pots are the definitive examples o f Category H I B . P r o f i l e 3 A , w h i c h is identifiable as a firing failure because of sherd warpage of the same k i n d as seen i n the flagon f o r m s , is the most complete of our examples. T h e decoration is c o m p o s e d of a series of rings beginning o n the neck and progressing d o w n onto the upper shoulder. T h e t h i r d of these rings is accentuated by diagonal hash marks o n either side transforming it into an abstract wreath. T w o m o r e bands end the decoration just below the widest point of the belly. T h e s e two areas of b a n d i n g serve to delineate a wide strip o n the lower shoulder w h i c h carries a striking pattern o f alternating trees and »snakes«, seven o f each. T h e trees are carried out i n simple

0 I

L

j

5 CM I

Illustration 1. Schematic drawing of the decorative pattern o n P r o f i l e 3 A .

straight l i n e b r u s h strokes and some o f the trees are further decorated w i t h s m a l l hash marks giving t h e m both the f o r m and appearance of pine trees w h i c h they were most l i k e l y intended to represent. T h e wavy vertical lines between the trees seem to represent snakes, but they are very abstract and positive i d e n t i f i c a t i o n is uncertain. P r o f i l e 3 B duplicates 3 A i n f o r m and the placement of the design but varies f r o m it in that the tree patterns are formed from a straight vertical line embellished with opposing volutes. T h e r e is no suggestion o f the pine tree i n the f o r m of this pattern and it was more l i k e l y derived f r o m one o f the deciduous tree. P r o f i l e 4 shows a neck fragment w h i c h duplicates the small flagon o f Profile T y p e 2 i n the treatment o f the l i p and is decorated w i t h five h o r i z o n t a l bands. T h e complete pattern of this pot remains an enigma since no other sherds c o u l d be f o u n d . Illustration 3 shows the reconstruct i o n of the pattern of another decorated bottle somewhat smaller than Profiles 3 A and B . H e r e the pattern is l i m i t e d to broad cross-hatching i n the area set o f f by the h o r i z o n t a l banding. P r o f i l e 5 is an unicum i n the excavation and its p o o r l y preserved white engobe showed no traces of red pigment. T h e r e can be little doubt that these bottle forms derive f r o m the spherical decorated jars o f the L a T e n e culture, but the question w h i c h remains unanswered is their purpose or use i n the R o m a n environment o f A u g s t . T h e L a T e n e tendancy to decorate even c o m m o n h o u s e h o l d pottery had largely died out by the time these pieces were made as is borne out by the paucity o f decorated sherds o f l o c a l manufacture w h i c h have been f o u n d i n the residential sections o f A u g s t . It seems l i k e l y t h e n that these pieces had some special f u n c t i o n or purpose, perhaps p r i m a r i l y decorative since they lack the handles of the f u n c t i o n oriented flagons. O n e possibility is that they were i n some way connected w i t h religious practice since b o t h trees and snakes figure heavily i n the symbology of religions throughout history. I n the G a l l o - R o m a n w o r l d the most likely point of reference is the C y b e l e - A t t i s cult. A G a l l o - R o m a n temple dedicated to C y b e l e was excavated i n A u g s t i n 193 3 . Just w h e n the worship o f C y b e l e arrived i n A u g s t is not k n o w n , but sherds f r o m the temple excavation show it to have b e e n b u i l t sometime i n the S e c o n d C e n t u r y . O n e o f these sherds (Inv. N r . 33.1023) is an exact duplicate o f our P r o f i l e I F and establishes a c o n n e c t i o n i n point of time between the sherds of o u r excavation and the temple. D u e to the G a l l i c f o r m of the t e m p l e , it is assumed that C y b e l e b l e n d e d w i t h an aboriginal mother-earth-goddess i n A u g s t . T h i s assumption is supported by other factors as w e l l . F i r s t , it is k n o w n that C y b e l e is only one of the forms of the architypal mother-goddess w h i c h was very widespread i n early times. A s p o i n t e d out by F r a z e r , the cults o f C y b e l e - A t t i s , Isis-Osiris and V e n u s - A d o n i s are simply different expressions o f the same more basic concept of an original mother-earth-goddess and her l o v e r . T h e C e l t i c P a n t h e o n no doubt had its o w n version of this goddess. S e c o n d , the patterns of trees and »snakes« o n our sherds seem certain to be direct descendants of almost i d e n t i c a l patterns o n late L a T e n e sherds f r o m B a d e n w h i c h are dated as early as the F i r s t C e n t u r y and the trees are seen again i n b o t h volute and straight line forms o n sherds o f certain C e l t i c o r i g i n f r o m N a g y v e n y i m a n d A q u i n c u m i n H u n g a r y . If we assume that b o t h types, the L a T e n e a n d our decorated bottles, had some religious signifiance, it follows that the early sherds f r o m B a d e n were used by a cult w h i c h was i n existance i n the first t i m e of the R o m a n occupation. T h e pine tree, w h i c h the pattern o n P r o f i l e 3 A most nearly resembles, is closely c o n nected w i t h the worship of C y b e l e , or more specifically w i t h her unfortunate ward and lover A t t i s . Legends s u r r o u n d i n g A t t i s frequently l i n k h i m w i t h pine trees a n d 46

47

48

49

5 0

forests. O n e account o f his death has it that he emasculated h i m s e l f u n d e r a pine t r e e after a lover's quarrel w i t h C y b e l e thus i n i t i a t i n g a series o f annual deaths and resurrections s y m b o l i z i n g the a u t u m n death and spring rebirth of the plant w o r l d . H e n c e , pine trees and pine cones, also symbols of eternal life and fruitfullness, became associated w i t h A t t i s and his cult of vegetation. T h e c o n n e c t i o n of the snake is less clear, but documentable to a degree. Snakes are at one and the same t i m e the most hated and the most revered of a l l animals. A l m o s t every r e l i g i o n f r o m the J u d e o - C h r i s t i a n w i t h its »snake i n the garden« to the p r i m i t i v e cultures of aboriginal A u s t r a l i a have had snake dieties, semi-dieties and anti-dieties of one sort or a n o t h e r . N o r is the snake as a s y m b o l foreign to A u g s t . T h e snakecult u r n f o u n d i n A u g s t i n 1898 has been associated w i t h the Sabazius c u l t , w h i c h is i n t u r n related to the cult o f C y b e l e - A t t i s . U n f o r t u n a t e l y , this piece came f r o m an undateable d i g , but its exact counterpart was f o u n d i n V i n d o n i s s a w h i c h establishes an e n d date of 101 A . D . T h e votive hands f o u n d i n A v e n t i c u m and the Great St. B e r n a r d , b o t h attributed to the Sabazius c u l t , show snakes and the f o r m e r also features a bust of C y b e l e and a pine c o n e . R o m a n reverence for snakes is d o c u m e n t e d i n Men and Snakes by R a m o n a and D e s m o n d M o r r i s w h o state that snakes were kept i n R o m a n households and regarded as personal g u a r d i a n s . T h e A e s c u l a p i a n snake (Elaphe longissima) was apparently not native to the region a r o u n d A u g s t i n early t i m e s , but is n o w considered to be an indigenous species ( N a t u r a l H i s t o r y M u s e u m o f Basel - E x h i b i t of Indigenous R e p t i l e s and A m p h i b i a n s - D e c e m b e r 1970). Because this snake was a favored R o m a n house pet, it is assumed that they were i m p o r t e d to the region by R o m a n settlers. A n d , t o o , Ishtar, the B a b y l o n i a n equivalent of C y b e l e , was charcterized by a snake e m b l e m . T h e r e are n u m e r o u s references to the snake as a s y m b o l o f i m m o r t a l i t y because it sheds its skin. T h e shedding of s k i n has also been associated w i t h c i r c u m c i s i o n and may i n the case of A t t i s have been confused w i t h total emasculation. Snakes were also regarded as messengers to the u n d e r w o r l d where A t t i s awaited C y b e l e ' s a n n u a l call to come and rejuvenate the plant w o r l d . O f the other sherds o f this type f o u n d i n other excavations i n A u g s t , one is an almost complete bottle w i t h a d e c o r a t i o n , w h i c h although p o o r l y preserved, is i n every way comparable to the trees and snakes of P r o f i l e 3 A . T h e f o r m of the piece varies somewhat i n that the m o u t h r i m constricts as i n P r o f i l e 2 A and is decorated w i t h two i n dented rings as i n P r o f i l e 1 A . T h e other is a s m a l l fragment o f the shoulder o f a decorated piece (Inv. N r . 1954.237) w h i c h was excavated o n the site of the present m u s e u m b u i l d i n g . T h i s piece carries the »snake« design, but replacing the tree is what appears to be a w h e e l . A l l of the i n d i v i d u a l patterns used to compose the decorations o n these pieces w i t h the e x c e p t i o n of the cross-hatching o n Illustration 3 offer r i c h possibilities for s y m b o l i c interpretation. 51

53

52

54

55

5 6

57

58

59

60

61

6 2

The Fine Wares Profiles 1 0 A - D and 1 2 A and B comprise a f o r m f a m i l y w h i c h finds direct counterparts i n the beakers excavated i n N i e d e r b i e b e r . 1 0 A and B are executed i n the r e d clay gloss o f Category I C , whereas 10C and D were f i n i s h e d i n the g o l d m i c a gilt of Category I V B . T h i s difference i n surface treatment accounts for the m i n o r , but noticeable, variations i n the forms. O n none o f the examples of 10C and D d i d we find the decorative external foot ring seen o n 10A and B . T h i s was no doubt o m i t t e d i n order to facilitate the b u r n i s h i n g needed to adhere the m i c a platelets to the t r i m m e d surface (see S e c t i o n o n Category I V B ) . 63

12 A and B are indented beakers and a l l of our examples are executed i n the reduced fabric of Category I B . T h e color ranges f r o m a m e d i u m olive grey ( M u n s e l l 5 Y 5/1) to black w i t h a l l shades between. T h e black sherds are, however, not comparable to the black o f O e l m a n n ' s T e c h n i q u e » d « w h i c h he describes as having a red paste w i t h an excellent black gloss w i t h a m e t a l l i c g l a n c e . A s w i t h the vast majority of the intentionally reduced wares f r o m A u g s t , the paste of these beakers is grey. N o r is the quality o f the gloss comparable to O e l m a n n ' s examples. T h e black sherds come for the most part f r o m the h u m u s layer and are p o o r l y preserved, but it seems very u n l i k e l y that they ever had the beautiful sheen o f the black clay gloss o f N i e d e r b i e b e r , T r i e r or K ö l n w h i c h are comparable i n quality to the black gloss of A t t i c pottery. T h e indented beakers also differ f r o m the p l a i n wares i n the treatment of the foot. I n every case, the feet of the indented forms are t r i m m e d w i t h the recessed groove s u r r o u n d i n g a raised circular plane such as seen o n the feet of the flagons of Profiles 1 & 2. T h e p l a i n forms have o n l y a slightly concave foot. O e l m a n n ' s pieces also show b o t h variations, but apparently not as regularly as ours i n that his small indented beaker (Type 33 A ) is pictured w i t h a simple f o o t . 13 A is an exceptionally f i n e , beautifully crafted beaker w h i c h is one o f the h i g h points of the excavation i n terms of mastery of the materials. T h e paste and gloss are i d e n t i c a l to those of the indented beakers and there can be n o doubt that they came f r o m the same w o r k s h o p . E t t l i n g e r f o u n d similar forms i n the FrauenthermenP but none i n the reduced grey of our examples. H e r descriptions fit better to our P r o f i l e 13 B w h i c h is a larger f o r m w i t h a light b r o w n ( M u n s e l l 5 Y R 5/6) clay gloss. 1 4 A is a decorated beaker f r o m the most c o m m o n f o r m type a m o n g the clay gloss sherds. N o t a l l o f the sherds of this type are decorated and the r i m profiles vary as s h o w n i n 1 4 D - F . T h e decoration o n 14 A consists of two bands o f rouletting w h i c h set o f f the raised patterns o f »willow leaves«, »horse shoes« and circles. T h e horse shoes have the s m a l l spur at the top w h i c h D r e x e l noted i n c o n n e c t i o n w i t h Swiss clay gloss w a r e s . T h e gloss o n 1 4 A is a dark olive grey ( M u n s e l l 5 Y 4/1), but a l l other sherds s h o w i n g this decoration are red ( M u n s e l l 10 R 5/6) a l t h o u g h they have exactly the same lip profile. P r o f i l e 14B is a m u c h cruder type and here the decor is l i m i t e d to three bands o f rouletting. 14C shows another variation of the f o r m w i t h a sharper break i n the f o r m as the curve reverses itself at the belly. N o sherds of type 14C showed a raised decoration. 14F is decorated i n the same style as 14A but is very poorly executed. T h e paste is red and coarse and the black clay gloss is p o o r l y preserved. E t t l i n g e r f o u n d examples of this f o r m throughout the S e c o n d C e n t u r y strata of the Thermen.^ She relates the type to O e l m a n n ' s T y p e 30 f r o m N i e d e r b i e b e r , but I have some reservations i n assuming a f o r m c o n n e c t i o n between the t w o . O e l m a n n ' s T y p e 30a is m u c h smaller, quite t h i c k , apparently has no decoration and has a lip profile dissimilar to any of Ettlinger's or our examples. H i s T y p e 30b relates better i n terms of f o r m , but again lacks the p r o f i l i n g of the lip and the figurative decoration is a far cry f r o m the reserved patterns o f our sherds. A more l i k e l y point o f reference lies i n D r e x e l ' s c y l i n d r i c a l beaker f o r m w h i c h he describes as having bands of rouletting and a weak l i p f o r m a t i o n . A s E t t l i n g e r n o t e d , this f o r m was not reported i n any of the excavations o f the Obergermanische-Rätische L i m e s nor does V o g t show any i d e n tical profiles f r o m the L i n d e n h o f i n Zürich. H i s profile A b b . 44,43 shows some s i m i larities, but lacks the inner t h i c k e n i n g o f the lip w h i c h is characteristic of our pieces. T h e fact that the f o r m first appeared i n the S e c o n d C e n t u r y explains its absence i n V i n d o n i s s a . C o n s i d e r i n g the differences f r o m s i m i l a r forms f r o m other areas, it is possible that this f o r m type may have originated i n this f o r m i n A u g s t . 64

65

66

68

7 0

7 1

15 A and B are smaller versions of T y p e 14. N o n e of the sherds of this type were decorated b e y o n d a simple i n d e n t e d line ca. 1 c m b e l o w the r i m . C o l o r s of the paste and gloss varied f r o m red to grey to black and i n no case was the quality as fine or the workmanship as careful as i n the indented beakers or the decorated forms of Profile 14. I n contrast to the limited distribution of T y p e 14, Profile 16 A is duplicated by finds i n the L i n d e n h o f , N i e d e r b i e b e r , F a i m i n g e n and the Frauenthermen in A u g s t . A l l but one of our sherds are done i n a reduced fabric and range i n color f r o m a light grey ( M u t i seli 5 Y 5.5/1) to dark grey ( M u n s e l l N 3 ) . T h e one exception has a p i n k i s h paste ( M u n s e l l 5 Y R 8/6) and a dark red gloss ( M u n s e l l 10 R 3/4; these colors were not observed i n any of the other sherds f r o m the excavation and it seems likely that this piece was imported). 16B varies only slightly f r o m 16 A i n the h a n d l i n g of the shoulder w h i c h breaks sharply d o w n w a r d f r o m the concave collar of the neck. T h e band of rouletting is set off at the top by a fine indented line. I n a l l but one e x a m p l e , this decorative band was placed i n such a way that the height o f the neck and the w i d t h of the undecorated band o n the shoulder are approximately equal. A larger v e r s i o n of this f o r m is shown i n P r o f i l e 16C. T h e b o w l depicted i n P r o f i l e 17 is an unicum i n that it is the only b o w l fragment f r o m the excavation executed i n an intentionally reduced clay gloss and the o n l y n o n beaker fragment showing a raised decoration. T h e fragment is too s m a l l to show the complete pattern, but the i n d i v i d u a l elements show some variations f r o m the other decorated sherds. T h e horseshoe lacks the spur and the raised parallel lines close at the end like hairpins rather t h a n r e m a i n i n g o p e n as do the w i l l o w leaves of P r o f i l e 14 A . T h e only other r e m a i n i n g design element is a row of raised dots. C l a y gloss and paste are identical to those of the indented beakers, so it is certain that the piece was made i n A u g s t . 72

73

74

75

T h e B o w l Forms T h e most prélevant f o r m f r o m the excavation, represented by nearly 600 i n d i v i d u a l r i m sherds, is an o p e n , carinated b o w l w i t h a h o r i z o n t a l r i m and g o l d gilt f i n i s h . T h e s e are closely paralleled i n the writings of E . E t t l i n g e r where she refers to t h e m as the »einfache, grobe Schüsseln« - s i m p l e , coarse bowls - characteristic of the S e c o n d Century. B o t h r o u n d e d and carinated forms were f o u n d , but the latter by far o u t n u m b e r the former. T h e usual f o r m (see P r o f i l e 18A) has a flat b o t t o m o n the i n s i d e , usually not m o r e than 6-8 c m . i n diameter; f r o m t h i s , the walls slope sharply outward to nearly the f u l l w i d t h of the b o w l before angling upward i n an almost vertical line w h i c h rises 4 - 7 c m . where the b o w l ends i n an ornate r i m o n the outer perimeter. T h i s gives the forms a distinctly angular, somewhat m e t a l l i c character w h i c h is heightened by the g o l d gilt f i n i s h . A l l of these pieces have been t r i m m e d and the bottoms are s i m p l y struck flat w i t h no foot rings. T h e profiles vary, but do not fall into neat categories; rather, they f o r m general groupings each w i t h a core of v i r t u a l l y identical sherds and others w h i c h show deviations i n some details and are related to one or more groupings. It was not possible to establish a positive c h r o n o l o g i c a l order since profiles w h i c h p r e d o m i n a t e d i n the lower strata were f o u n d i n duplicate i n the upper layers as w e l l , but the t h i c k e r , clumsier r i m profiles increase i n numbers i n the upper strata w h i c h does seem to bear out E t t l i n g e n s postulation that this type of b o w l became heavier i n r i m profile as the S e c o n d C e n t u r y progressed. Profiles 18 A , B and C show some of the finer r i m types. T h e y were chosen f r o m a 76

77

group o f some 70 r i m sherds w h i c h showed very little variation i n either profile or competency o f e x e c u t i o n . T h e r i m s were w e l l articulated and are characterized by the sharp under edge. T h e bowls range i n diameter f r o m 22 to 32 c m . Profiles 1 8 D , E and F represent a smaller group of sherds and are closely related to Profiles 1 8 A , B and C by b o t h association of sherds i n the excavation and the fact that the external appearance of the original bowls w o u l d have been very similar whether the lip d r o o p e d as i n D , E and F or was m o r e s o l i d as i n A , B and C . I h o l d this k i n d o f variation to be the result of different w o r k habits and the techniques used by the potters rather that signs of progressive changes i n the forms. T h e similarity between ProFiles 18 D and E w o u l d seem to indicate that the f o r m of the b o w l , whether r o u n d e d or carinated, d i d not necessarily dictate the type of r i m to be used. I n Profiles 18 G and H the mass of the lip is placed somewhat higher, a tendancy w h i c h is carried even further i n Profiles 181 and J . T h e y stem almost e x c l u s i v e l y f r o m the upper strata and are m o r e crudely made and usually not as w e l l fired as examples of Profiles 18 A - F . P r o f i l e 18 J is the o n l y sherd of this f o r m type w h i c h shows traces of a red clay gloss; a l l other sherds have gold m i c a g i l d i n g . Profiles 18K and L are again of a m o r e compact type and were f o u n d closely associated w i t h sherds f r o m Profiles 1 8 A - F . T h e c o m m o n e s t type of r i m sherd f r o m the excavation w i t h more than 250 i n d i v i d u a l examples is s h o w n i n P r o f i l e s 1 8 M and N . H e r e , the heaviness o f the r i m is quite apparent and is a c c o m p a n i e d i n most cases by a c l u m s i e r treatment of the b o w l i n general. T h e sharper r i m seen o n N and the rounder profile of M were f o u n d side by side and i n almost equal numbers and represent, i n m y v i e w , the k i n d of variation caused by h a n d w o r k as m e n t i o n e d above. Profiles O and P show the crudest r i m type f o u n d and c o m e exclusively f r o m the h u m u s layer. Profiles 1 9 A - C show a triple ridged r i m profile w h i c h appears to have been reserved exclusively for r o u n d e d bowls. E t t l i n g e r shows s i m i l a r forms i n the Thermen * and V o g t shows a profile almost i d e n t i c a l to o u r s but describes it as having a red-coating (roten Überzug) whereas our examples are done exclusively i n the g o l d m i c a g i l d i n g o f Category I V A . Profiles 20 A and B were taken f r o m the o n l y examples o f this type i n the excavation. H e r e the f o r m is quite l o w a n d open w i t h a slightly constricted collar and carefully executed p r o f i l i n g around the body of the pot and at the r i m . T h e s e sherds were f o u n d i n direct association w i t h the sherds used for Profiles 1 8 A - F and were also executed i n g o l d m i c a gilt. Profiles 21 and 22 show two b o w l forms w h i c h were p r o d u c e d i n A u g s t i n direct i n i tation of T e r r a Sigillata forms (Dragendorff 37 and 38 respectively). B o t h profiles were t a k e n f r o m wasters and b o t h were executed i n the red clay gloss of Category I C . A n o t h e r sherd of the D r . 38 type (Inv. N r . 68.14225), although somewhat larger than Profile 22, was done i n the g o l d m i c a gilt of Category I V A . Profiles 23 A and B show two r i m profiles f r o m an enclosed b o w l f o r m w h i c h enjoyed an e n d u r i n g popularity i n A u g s t . T h e y are one o f the most c o m m o n b o w l types f o u n d i n excavatins throughout the city despite w h i c h only a few of these sherds were f o u n d i n this excavation. E t t l i n g e r reports very s i m i l a r i f not i d e n t i c a l forms f r o m the Thermen* ; she makes reference to the longevity of the f o r m and dates her sherds throughout the F i r s t C e n t u r y . O u r sherds were f o u n d scattered t h r o u g h a l l strata o f the excavation i n c l u d i n g positions superior to dateable sherds f r o m the A n t o n i n e p e r i o d , so we must assume that they were c o n t i n u e d u n t i l w e l l into the S e c o n d C e n tury. D u e to their scarcity, the w o r k m a n s h i p ( w h i c h is usually somewhat imprecise) 1

79

0

and the absence of any recognizeable firing failures, it is d o u b t f u l that they were made i n the same w o r k s h o p w h i c h manufactured the m i c a gilt bowls o f Profiles 18-20. P r o f i l e 26B shows a somewhat u n u s u a l variation o n the c o m m o n l y f o u n d b o w l f o r m w i t h externally flanged l i p . I n each of the six R S of this type f o u n d i n the excavation, the external flange rises as it leaves the w a l l and reaches the level of the vertical lip before dropping sharply d o w n w a r d . A n o t h e r distinctive feature of this type is the carination of the w a l l w h i c h supplants the usual h e m i s p h e r i c a l shape i n other variations of this f o r m . A l l examples of this f o r m c o m e f r o m the h u m u s or upper strata o f the excavation and it seems i m p r o b a b l e that they c o u l d date f r o m earlier t h a n the early decades of the T h i r d C e n t u r y . T h e r e m a i n i n g R S , w h i c h is an unicum i n the e x c a v a t i o n , is s h o w n i n P r o f i l e 2 6 A . I n this case the sharp i n c l i n e of the w a l l shows it to have come f r o m a deeper f o r m than is usual. T h i s fragment came f r o m a stratum u n d e r l y i n g the major depositions of fragments f r o m b o w l forms 18A t h r o u g h F . Patterns of o x i d a t i o n and r e d u c t i o n o n its b r o k e n edges prove it to have been a firing failure. T h e C o o k i n g Pots P r o f i l e 2 5 A shows the c o m m o n f o r m of the c o o k i n g pots f r o m A u g s t . T h e sizes range f r o m 11 to 24 c m . i n diameter at the m o u t h , but the proportions of the body r e m a i n unchanged regardless of size. S i x t e e n of the r i m sherds have a ridge o n the r i m as s h o w n o n P r o f i l e 25B w h i c h was probably intended to accept a l i d . T h e s e pieces were executed almost exclusively (ten exceptions i n more than 500 sherds) i n a heavily gritted grey paste (Category V I A ) . A few o x i d i z e d sherds were also f o u n d , but these may w e l l be accidents of the fire since grey or black was o b v i o u s l y the prefered color. T h e reasons for this c o l o r preference probably stem i n this case largely f r o m aesthetic grounds. I n use, these pots w o u l d naturally have been smudged f r o m the c o o k i n g fires, and this d i s c o l o r a t i o n w o u l d have been m u c h less noticeable o n a grey pot than a red one. A n d , t o o , the grey paste w o u l d have been m u c h closer i n color to the m e t a l c o o k i n g pots w h i c h were no doubt more desired. O n l y a very few of these sherds show s m u d g i n g w h i c h c o u l d have been caused by use o n an open fire and most are probably wasters. T h e Plates T h e plate sherds f r o m the excavation showed a remarkable u n i f o r m i t y . O f the nearly 1200 plate sherds f o u n d , o n l y three d i d not c o n f o r m to P r o f i l e 24. T h e only diversity seen i n this group was i n the manner of f i n i s h i n g the surfaces. F o u r t e e n sherds were smudged black (but not f r o m c o o k i n g fires), 159 h a d a red clay gloss and 1005 were f i n i s h e d i n g o l d m i c a gilt. T h e s e sherds were f o u n d throughout the excavation and ranged i n diameter f r o m 18 to 32 c m . The M o r t a r i a P r o f i l e 27 is an unicum i n the excavation and a firing failure. T h e paste shows a grey k e r n e l , but some of the b r o k e n edges are r e o x i d i z e d i n a way w h i c h c o u l d only have resulted f r o m breakage d u r i n g the firing process. T h i s piece is a p r o v i n c i a l copy o f C u r i e T y p e 21 w h i c h was p r o d u c e d i n T e r r a Sigillata d u r i n g the A n t o n i n e E r a . T h e r i m and a band o n the i n n e r surface extending d o w n to the sanded surface are covered w i t h a red clay gloss ( M u n s e l l 10 R 4/6) ; the outside was left p l a i n . T h e p o u r i n g spout was made by cutting away a section o f the vertical ridge o n the r i m and adding t w o

fillets of clay to f u n n e l l i q u i d over the outer edge. T h e b o t t o m has been t r i m m e d and may have had a recessed foot ring as o n C u r i e T y p e 21. P r o f i l e 28 is very s i m i l a r i n treatment although the profile of the r i m is quite different. It, t o o , has a red clay gloss coating ( M u n s e l l 10 R 5/4) identical i n placement to that o n P r o f i l e 27 and the spout is f o r m e d i n the same manner. T h e t r i m m i n g was l i k e l y the u n d o i n g o f the pot since the walls near the base t h i n d o w n to a scant 2/2 m m as opposed to the 1 c m and m o r e usually seen i n mortaria. F r o m this evidence alone it is i m p o s s i b l e to state that the piece is a certified t e c h n i c a l f a i l u r e , but this seems very probable. I n any case, the materials identify it as a product of A u g s t . P r o f i l e 29B is also a firing failure a n d duplicates the previous two profiles i n the placement o f the red clay gloss ( M u n s e l l 10 R 5/6), the m a n n e r o f m a k i n g the spout and the fact that it is t r i m m e d . O n l y a s m a l l segment of the base r e m a i n s , but there is no hint of a recessed b o t t o m , and we can assume that the entire base was struck flat as i n the bowls of P r o f i l e 18. T h e major change here is that the unsanded p o r t i o n is distinctly concave d e m a r k i n g it f r o m the r u b b i n g surface below. E t t l i n g e r f o u n d identical forms i n the Augster Thermen^ and agrees w i t h D r e x e l w h o dated t h e m as having begun i n the S e c o n d C e n t u r y A . D . P r o f i l e 2 9 A belongs to the same type as 2 9 B , but is f r o m a larger version. 1

8 0 b

P r o f i l e 30 is a larger and m o r e open f o r m and i n some of our examples the external flange curls back o n itself at the outer edge. H e r e the entire inner surface of the b o w l is sanded and clay has been added to f o r m the spout giving it a sculptured l o o k . I n this case the sand has apparently been applied to the surface i n a t h i c k engobel i k e layer rather t h a n just s p r i n k l e d o n the raw clay and pressed i n , w h i c h is the m o r e usual technique. T h e engobe is quite soft w i t h the result that the r u b b i n g surface is not very resistant to abrasion. T h e stamped pattern a r o u n d the r i m was probably made w i t h a fired clay stamp and was applied w h i l e the freshly t h r o w n pot was still quite soft, perhaps before it was r e m o v e d f r o m the w h e e l . Just u n d e r the r i m u n d e r each stamp, there is a finger mark where the potter supported the r i m w h i l e stamping to prevent d e f o r m i n g the pot. Sherds f r o m at least three mortaria o f this type were f o u n d and the stamp marks o n each were apparently made w i t h the same t o o l i n d i c a t i n g that they were a l l made i n the same shop. Sizes vary somewhat and o n the larger f o r m s , two stamps are placed side by side o n either side of the spout. T h i s may have been a means o f i n d i c a t i n g size or v o l u m e . A s the size increases, the cross-section o f the r i m becomes heavier, but the general character of the profile remains the same. T h e bases were apparently left u n t r i m m e d . Profiles 3 1 A and B are very closely related types and differ f r o m P r o f i l e 30 only i n the m o d e l i n g of the spout. T h e y are not quite the same i n that the inner r i m o n 3 1 A is the highest point of the pot whereas o n 31B the outer r i m is higher. N o n e of our examples of 31B has a complete spout, but one sherd shows the beginnings of the m o d e l i n g and they are the same as o n 31 A . P r o f i l e 31C is probably a variant of the two above types a n d is an unicum i n terms of size a n d color. It is the smallest mortari u m o f this type and has a very coarse sand o n the interior. T h e c o l o r of the paste is quite light ( M u n s e l l 10 Y R 7/4) i n c o m p a r i s o n w i t h a l l other m o r t a r i a sherds, a n d is not u n i f o r m i n that it shows striations of a deeper reddish hue. T h e clay beds i n A u g s t often have an uppermost layer w h i c h is of a lighter color t h a n the deeper strata and this pot m a y have been made f r o m that clay or it m a y be an i m p o r t . O n l y two sherds of the type s h o w n i n P r o f i l e 31D were f o u n d and they may belong

to the same pot a l t h o u g h they c o m e f r o m different parts of the dig. I n b o t h the sanded surface is very fine. T h e r e does not appear to be a pattern i n the size of the sand used o n various sizes of mortaria. Some o f the largest examples have fine sand a n d vice versa. B o t h sherds of T y p e 31D came f r o m m i x e d strata y i e l d i n g c o n f l i c t i n g data although there was some evidence to suggest that they are some of the youngest sherds i n the excavation. I n any case, they post date Types 3 1 A - C . N o tests were r u n to determine that these two sherds were made i n A u g s t a n d they m a y be imports. E a c h of these types was f o u n d i n c o n j u n c t i o n w i t h sherds f r o m one of more other profiles and no c h r o n o l o g i c a l order c o u l d be established. R a t h e r , it seems likely that all forms were p r o d u c e d m o r e or less contemporaneously. T h e only T . S . m o r t a r i u m f r o m the excavation is sufficiently complete to be identified as a D r . 45 and is dated i n O s w a l d and Pryce as second h a l f of the S e c o n d C e n t u r y . 81

The Derivation of the Forms T h e c o m p l e x u n d e r the collapsed floor o f K i l n II y i e l d e d a large n u m b e r of sherds w h i c h proved i n the course of study to be the most important a n d informative area i n the excavation. T h e i r p o s i t i o n i n the dig was such that they were deep enough to have largely escaped the damaging acids of the h u m u s layer; hence they are for the most part but little changed i n appearance despite the l o n g b u r i a l . T h i s is i n direct contrast to the sherds w h i c h were f o u n d i n the upper strata where soil acids altered and sometimes destroyed the surface characteristics to the point that even general determinations as to type were impossible to state w i t h any assurance of accuracy. T h e enclosure o f the k i l n also afforded the advantage that the sherds were less scattered and m o r e nearly complete pieces c o u l d be reconstructed i n the tedious task o f sorting and reassembling the thousands o f sherds f o u n d . M o s t o f the pieces w h i c h were sufficiently complete to allow for f u l l profiles came f r o m this c o m p l e x . F r o m the very b e g i n n i n g it was apparent that some o f our f o r m s , the flagons for instance, were new i n that no counterparts c o u l d be f o u n d i n previously p u b l i s h e d literature o n p r o v i n c i a l R o m a n pottery. T h e fact that this c o m p l e x was relatively closed off f r o m the rest of the excavation also presented its problems since it contained no T e r r a Sigillata sherds, no coins, nor any other items w h i c h were easily dateable (see section o n C h r o n o l o g y ) . A s was the case w i t h the entire excavation, m a n y of the sherds f o u n d i n this c o m p l e x proved to be f r o m wasters clearly l a b e l i n g t h e m as products of A u g s t . B e i n g a s m a l l part o f a s m a l l excavation (ca. 55 m i n total), the c o m p l e x h e l d a surprisingly large n u m b e r of f o r m types w h i c h were sufficiently w e l l represented to suggest that they were standard p r o d u c t i o n items of the potteries of A u g s t . T h e study of these forms and its c o n c o m i t a n t effort to search out their derivations p r o d u c e d an u n r e l e n t i n g ly recurrent pattern w h i c h i n the final analysis brands t h e m as being closely allied to the forms of indigenous C e l t i c pottery. T h e r e can be little doubt that the potters w h o p r o d u c e d these wares were native to the region rather than i m m i g r a n t s f r o m the Italian p e n i n s u l a . M a n y of the best preserved flagon fragments came f r o m this c o m p l e x . A l l sherds f r o m P r o f i l e 1 A - C and the best preserved specimens o f P r o f i l e 2 were f o u n d there and none of these c o u l d be related to the flagons f r o m the p u b l i s h e d literature f r o m other sites or even f r o m other excavations i n A u g s t . T h e flagon is, i n and of itself, a distinctly R o m a n species. T h i s f o r m w i t h its c o n stricted neck a n d one or m o r e handles was apparently not used i n the p r e - R o m a n C e l t i c c u l t u r e ; yet our examples f i n d their roots i n purely C e l t i c antecendents. T h e y are notably plastic as opposed to the harder, u n y i e l d i n g forms seen i n O s w a l d and P r y c e , N i e d e r b i e b e r and particularly V i n d o n i s s a where the m i l i t a r y post brought a m u c h stronger a n d m o r e direct M e d i t e r r a n e a n i n f l u e n c e . O n l y i n the m o u t h rims of P r o f i l e T y p e 2 do we f i n d a severity w h i c h suggests a metallic or non-clay i n f l u e n c e . T h i s m o u t h treatment is atypical w h e n c o m p a r e d w i t h the finds f r o m other areas and was apparently not c o m m o n i n p r o v i n c i a l pottery. T h e only s i m i l a r profiles we were able to locate were f r o m pots excavated i n A l z e y and W o r m s . Since those pieces date between 350 and 355 A . D . , the relationship is tenuous at best. T h a t the spherical pot of the G a l l o - R o m a n w o r l d is a c o n t i n u a t i o n o f the L a T e n e tradition is w i d e l y a c c e p t e d . T h e most obvious antecedents of our pieces are the large spherical jars reported by M a j o r . A piece w h i c h shows strong relationships 2

82

83

8 4

85

86

to b o t h our pots and the L a Tene pieces was reported by R i t t e r l i n g and m a y be a developmental l i n k between the two. I n our pots, the development has progressed to the adaptation o f the f o r m as a flagon w i t h o u t , however, having completely lost its C e l t i c identity. I n the treatment o f the feet we f i n d clear indications o f the ties to the L a T e n e prototypes. T h i s style o f t r i m m i n g w i t h its depressed groove just i n side the foot ring relates directly to the base treatment o n C e l t i c pots f r o m B a s e l , H o f h e i m , V i n d o n i s s a , Z ü r i c h and others. B y far the more c o m m o n flagon base is a simple stand ring w i t h a recessed interior. R e s e a r c h into the development of this style o f foot r i n g w o u l d probably lead far back into history o f the h a n d f o r m e d pottery o f the region. I n its o r i g i n a l f o r m , the u n usual m o d e l i n g was most l i k e l y an outgrowth of the f o r m i n g process. D e e p l y recessed feet o n c o i l b u i l t jars and some kinds o f t h r o w n pieces as w e l l are m o r e easily acc o m p l i s h e d by adding t h e m i n a second process w h e n the freshly f o r m e d pot has stiffened to the leather hard state. T h i s technique is strongly suggested i n the sharp interior angle at the base o f many p r o f i l e d C e l t i c p i e c e s . I n the flagons o f A u g s t the peculiar foot treatment has been c o n t i n u e d as a stylistic device. T h e r e are also obvious differences i n the forms. T h e bottles o f the L a Tene c u l t u r e , w h i l e quite f u l l and r o u n d e d , never approach the exaggeration o f f o r m seen i n o u r examples. T h i s may be accounted for by the i n t r o d u c t i o n o f a superior potters w h e e l of R o m a n design. M o s t o f the p r e - R o m a n w h e e l t h r o w n pottery o f G a u l has a reserved quality w h i c h is t y p i c a l o f w o r k done o n a rather p r i m i t i v e , slow t u r n i n g w h e e l . L a t e r works f r o m the same areas executed under R o m a n d o m i n a t i o n most usually have a more flamboyant character and were certainly made o n a s m o o t h - r u n n i n g w h e e l o f sophisticated design. C o n s i d e r i n g R o m a n achievements i n other areas o f technology, it seems more than just l i k e l y that their potters wheels were also o f good quality and that they i n t r o d u c e d t h e m to this area. A n d , t o o , the G r e c o - R o m a n practice o f b l u n g i n g clay to a fine slip and a l l o w i n g the coarse particles t o settle o u t , thereby r e a l i z i n g its m a x i m u m plasticity, may have been a c o n t r i b u t i n g factor. It seems very l i k e l y that the practice o f b l u n g i n g clay was brought to this area by the R o m a n s especially i n view o f the fact that clay gloss coatings w h i c h are p r o d u c e d by a variation o f this technique f o u n d their beginnings i n the F i r s t C e n t u r y , i.e. after the beginnings o f R o m a n i n f l u e n c e . I n c o m b i n a t i o n or a l o n e , these changes w o u l d have a l l o w e d f o r more extravagant forms and led to the development o f the type of pots seen i n A u g s t . T h e bottles o f P r o f i l e T y p e 3 show further evidence o f their d e r i v a t i o n f r o m the L a T e n e i n their decoration and there exist between the two types some interesting parallels as w e l l as some important differences. A s a point o f c o m p a r i s o n , see E m i l M a j o r ' s Gallische Ansiedelung mit Gräberfeld bei Basel, T a f e l X X I I , 15. B o t h M a j o r ' s pieces, w h i c h we w i l l take to be t y p i c a l o f the sort, and ours display their painted decorations o n the upper shoulder and have their patterned areas set o f f above a n d b e l o w by h o r i z o n t a l banding. These factors alone are not sufficient to assume a derivitive relationship since work i n this m o d e is f o u n d i n m a n y cultures f r o m such w i d e ly scattered areas as A n c i e n t C h i n a and the p r e - C o l u m b i a n A m e r i c a n Southwest. N o r c a n we m a k e a favorable c o m p a r i s o n between the patterns i n terms o f either design or e x e c u t i o n . T h e L a T è n e pot has a tight geometric pattern w i t h an ingenious design o f asymetrical balance, carefully p l a n n e d and painstakingly executed. P r o f i l e 3 A , by contrast, was decorated i n a m a n n e r o f minutes w i t h a s i m p l e , u n s o p h i s t i cated pattern o f trees a n d »snakes«. T h e r e c o u l d have been little i f any pre-planning 87

88

89

90

91

92

93

since the trees are not equally spaced a r o u n d the pot. T h e fact that there are seven trees seems i l l o g i c a l since a circle cannot be divided into seven equal parts by n o r m a l methods of geometry, but this may have some m y s t i c a l signifiance since the n u m b e r seven is sometimes held to have magical properties. Despite this seemingly unbridgeable gap, our trees f i n d their roots f i r m l y planted i n the painted ware of the C e l t s as s h o w n by a virtually i d e n t i c a l tree pattern o n a L a T e n e sherd f o u n d i n V i n d o n i s s a . E v e n the »snakes« are foreboded i n the zig-zag line f r o m the same sherd. N e i t h e r the white engobes nor the red pigment used o n L a T e n e pottery is comparable to the materials o f similar color used o n o u r wares. T h e white pigment used by the C e l t i c potters was a white firing clay, perhaps a k a o l i n or pipe clay, w h i c h may have been i m p o r t e d . It was, i n any case, used rather sparingly. T h e engobe used i n A u g s t , o n the other h a n d , is neither pure white nor was there any reservation i n using it lavishly. T h e materials were l o c a l and cheaply w o n (see section o n M a t e r i a l s ) . T h e red p i g m e n t , t o o , is a sharp departure f r o m C e l t i c t r a d i t i o n . W h e r e a s the red used o n L a T e n e pottery was a simple iron-bearing clay, carefully applied and perhaps later p o l i s h e d (almost any red-firing clay w o u l d serve the purpose), the red decor o n our wares is carried out i n a clay gloss material. It has a sheen w i t h o u t the need o f burnishing and the nature of the material dictated that the pattern be executed i n a one-stroke process. A s discussed i n the section o n C l a y G l o s s C o a t i n g , the clay of A u g s t p r o d u c e d a somewhat transparent gloss. Over-strokes show as such. T h i s , i n and of itself, dictated the abandonment of the tight geometric patterns of the L a T e n e . It was no longer possible to m a i n t a i n u n i f o r m i t y of color w i t h these n e w , but available, materials so the o l d style was discarded i n favor of an expedient, but acceptable substitute. T h e t i m e lapse between our sherds (the last one-third of the S e c o n d C e n t u r y at the earliest) and the last b l o o m i n g s of the L a T e n e culture is m i n i m a l i n view of the w o r k of E . E t t l i n g e r w h o proved the c o n t i n u a t i o n of the L a T e n e into the S e c o n d C e n t u r y . I n any case, some intervening p e r i o d was necessary for the change i n style and the acceptance of new methods. T h e differences discussed above stem almost e x c l u s i vely f r o m the techniques used and the painted pots of A u g s t demonstrate the a m a l gamation of the C e l t i c and R o m a n cultures through the use of factors c o m m o n to each. T h i s b l e n d i n g of cultures is not u n i q u e to A u g s t and has been noted i n other areas as w e l l , most notably i n recent years i n new finds f r o m H u n g a r y . T h e striped pottery f r o m B r i g e t i o is contemporary w i t h the painted pots f r o m A u g s t and also c o m b i n e s the clay gloss and rouletting of T e r r a Sigillata technique w i t h forms directly derived from Celtic prototypes. E v e n m o r e striking parallels to our pots are seen i n the painted pots f r o m P a n n o n i a . A s noted earlier, b o t h the volute and straight line trees seen o n our sherds are also represented o n sherds f r o m N a g y v e n y i m a n d A q u i n c u m . P r o f i l e 5 shows a lip f o r m a t i o n c o m m o n l y seen o n bottle forms of the G a l l o - R o m a n provinces. W a l k e and V o g t show i d e n t i c a l profiles o n decorated bottles f r o m Straubing and the L i n d e n h o f . T h e bottles f r o m Straubing also display the peculiar foot treatment discussed above. I n the bowls f o r m s , t o o , we f i n d positive evidence of C e l t i c influence. T h e bowls of Profile Types 18 and 19 fall into the same category as those shown i n D r e x e l o n Taf. X I I , 3 and 7. D r e x e l is positive i n his discussion of the type that it developed i n R a e t i a f r o m aboriginal prototypes. Schörgendorfer also attributes the development of these carinated bowls to a b o r i g i n a l , i.e. C e l t i c , G a u l i s h prototypes. O u r P r o f i l e T y p e 20 b o w l shows a definite relationship to the L a T e n e forms reported by A l f r e d H a f f n e r . 94

95

953

9 5 b

95c

T h e r e are n u m e r o u s literary references to the n o n - R o m a n origins o f »Rätische Firnisware« a n d o u r clay gloss coated wares are g o o d examples. V i r t u a l l y every f o r m type i n this category of ware finds its f o r m antecedents i n C e l t i c pottery. O n e o f the clearest examples lies i n the f o r m f a m i l y of pots o f o u r P r o f i l e T y p e 10 and their i n dented counterparts o f P r o f i l e T y p e 12. T h i s f o r m w i t h its short neck a n d wide belly clearly shows its C e l t i c origins. E a r l y antecedents of the f o r m are s h o w n i n M a j o r and S c h o p p a . O u r P r o f i l e I I A is an intermediary f o r m and the most likely i m m e d i a t e precursor to Types 10 a n d 12. P r o f i l e I I B shows a s m a l l T e r r a N i g r a beaker w h i c h was excavated i n A u g s t i n 1970 under the d i r e c t i o n o f T e o d o r a T o m a s e v i c a n d dated t h r o u g h T e r r a Sigillata sherds f r o m the same c o m p l e x as having c o m e f r o m the second half of the F i r s t C e n t u r y . E t t l i n g e r f o u n d s i m i l a r forms i n the Frauenthermen of A u g s t a n d noted their relationship to the later clay gloss w a r e s . I n discussing a T e r r a N i g r a pot o f the same type, D r e x e l points to L a T e n e pottery f o r the precursory f o r m s , a n d R i t t e r l i n g expressed similar v i e w s . J u d g i n g f r o m the nature o f the paste, I feel certain that I I B was made i n A u g s t . A n interesting feature o f this piece is the t r i m m i n g of the foot w h i c h relates directly to the foot treatment o n P r o f i l e Types 1 a n d 2 a n d forms a bridge between flagons a n d the i n d e n t e d beakers o f T y p e 12 w h i c h show an i d e n t i c a l foot treatment. T h e chief f o r m difference between T y p e s 10 a n d 11 lies i n the treatment of the l i p , and this can be e x p l a i n e d as a r a m i f i c a t i o n o f the different techniques used i n f i n i s h ing the surface. I n the T e r r a N i g r a piece at least some b u r n i s h i n g was needed to achieve the sheen; this p r e c l u d e d a c o m p l e x l i p profile since this w o u l d have h a m pered the p o l i s h i n g operation. W i t h the a d o p t i o n o f clay gloss coatings the necessity for s i m p l i c i t y fell away since the gloss was applied i n l i q u i d f o r m a n d c o u l d be used over even very c o m p l i c a t e d surface features. P r o f i l e 1 1 A is a T e r r a N i g r a sherd f r o m the excavations o n Venusstrasse Ost a n d shows the beginnings o f a l i p . Its p o s i t i o n i n the d i g indicates that it is one of the oldest sherds f r o m that excavation, but it is probably m o r e recent than T o m a s e v i c ' s piece. T h e C e l t i c origins o f the r e m a i n i n g f o r m types f r o m the c o m p l e x , P r o f i l e Types 12, 13,14 a n d 16, are w e l l d o c u m e n t e d a n d need not be recapitulated h e r e . C o n s i d e r i n g the i n f o r m a t i o n given above, we can t h e n state that C e l t i c influences were very m u c h alive i n A u g s t u n t i l very late i n the S e c o n d C e n t u r y a n d perhaps longer as evidenced by their impact o n the l o c a l l y p r o d u c e d pottery o f the colony. E v e n so, it must be realized that a l l o f these pots have undergone substantial transformations not only i n f o r m but particularly i n t e c h n i q u e , a n d these are of R o m a n origin. T h e pots of Venusstrasse O s t , t h e n , represent a b l e n d o f C e l t i c a n d R o m a n cultures. 9 6

97

98

99

100

101

The Kilns K i l n I a n d K i l n II rank a m o n g the best preserved k i l n s o f R o m a n o r i g i n ever excavated. I n both cases damage and deterioration is slight and very little restoration was needed to m a k e t h e m suitable for display. F o r this reason, they have been enclosed under a protective shelter where they can be preserved f o r p u b l i c view and study. B o t h o f these k i l n s were built o n a basic design principle w h i c h was c o m m o n t h r o u g h out the R o m a n E m p i r e as evidenced by n u m e r o u s finds i n such w i d e l y scattered areas as E n g l a n d , F r a n c e and T u r k e y . E x a m p l e s o f this k i n d o f k i l n were still i n use i n isolated regions o f southern Italy as late as I960. T h e y are simple updraft k i l n s w i t h a c o m b u s t i o n chamber b e l o w and a ware chamber above separated by a pierced f l o o r (Lochtenne) o n w h i c h the pots were stacked. I n o u r examples, the pierced f l o o r is supported by a w a l l w h i c h extends forward f r o m the back w a l l o f the c o m b u s t i o n chamber a n d by ledges w h i c h corbel out f r o m the side walls f o r m i n g simple supportive arches. N o keystone arch construction was used i n either k i l n . I n areas where bridging was necessary, such as the f o r m a t i o n o f the f l o o r o f the ware chamber a n d over the fire m o u t h s , either c o r b e l l i n g or post and l i n t e l systems were used. T h e chambers are oval i n shape and are slightly elongated f r o m front to rear. A s is c o m m o n for kilns o f this type, the major p o r t i o n o f both k i l n s is s u n k e n into a bank o f clay w h i c h buttresses the walls a n d provides i n s u l a t i o n . T h e walls o f b o t h c o m b u s t i o n a n d firing chambers were b u i l t o f b r o k e n r o o f i n g tiles and the like a n d vary between 15 a n d 30 c m i n thickness. 103

T h e potters o f A u g s t apparently d i d not m a k e special bricks or tiles for their k i l n s but s i m p l y availed themselves o f suitable materials f r o m the city's trash pile. T h e upper r i m o f K i l n I I , w h i c h extends above g r o u n d l e v e l , is surrounded by r o u g h blocks a n d chunks o f limestone w h i c h support a n d buttress the walls at that point. T h e c o n s t r u c t i o n is rough a n d somewhat haphazard a n d sherds o f pots are sometimes f o u n d f i l l i n g gaps between the larger pieces o f b r o k e n roofing tile. T h e o n l y mortar used was clay and this was probably dug at the k i l n site. T h e k i l n s sit i n the upper strata o f the clay o f A u g s t , w h i c h as discussed i n the section o n M a t e r i a l s , has a very h i g h sand content. T h e character a n d appearance o f the mortar suggest that this was the material used. T h i s same clay was used to line the inner surface o f the ware chamber a n d to s m o o t h the pierced floor. A s shown by the finger marks o n the f l o o r o f K i l n I I , the clay was smeared o n i n a soft m u s h state to a depth of 1-3 c m . T h e kilns sit at right angles to each other and their fire mouths open onto a c o m m o n service area. T h i s suggests that they were built at about the same time and probably belonged to the same potter or pottery. A b o v e the f i r e m o u t h o f each k i l n is a narrow walkway and a door leading into the ware chamber. T h e ware chamber o f K i l n I measures 135 c m f r o m front to rear, 105 c m f r o m side to side a n d is 85 c m deep. T h e same measurements f r o m K i l n II are 210 c m , 180 c m and 40 c m respectively. A s s h o w n by these figures, K i l n I is proportionately m u c h deeper than K i l n I I , but there is no r e m a i n i n g evidence to suggest that the walls o f K i l n II were ever m u c h higher than they are at present. A more l i k e l y explanation is that they were used to fire different kinds o f ware. T h e size a n d c o n f i g u r a t i o n o f K i l n I w o u l d have made it ideally suited for firing smaller pieces such as the clay gloss coated ware. T h e s e s m a l l f o r m s , for the most part beakers, are such that they c o u l d be stacked f r o m the floor to the r i m o f the k i l n and fired w i t h n o great danger of breakage. T h e flagon f o r m s , o n the other h a n d , cannot be stacked one o n the

Illustration 11. Cross-section of a contemporary k i l n f r o m Santa L u c i a i n southern Italy w h i c h continues ancient traditions and illustrates the basic plan of the kilns f r o m A u g s t ( A f t e r H a m p e and W i n t e r , p. 35, A b b . 25).

Illustration 12. Stamp of profile 30, plate X I I (1:1).

Illustration 13. T h e limestone cup (1:1).

other to any great depth w i t h o u t the real danger of breakage i n the fire and a significant loss i n each k i l n load. K i l n II w i t h its large f l o o r area and depth sufficient f o r two or three layers of flagons w o u l d have been appropriate. I n any case, there c a n be little doubt that the majority o f sherds f o u n d i n the excavation were not fired i n either of the kilns. M a n y of the sherds were f o u n d i n the service area at levels above the top of the fire mouths o f the k i l n s , a n d one must assume that the kilns were n o longer used at the t i m e these were deposited. T h e r e is no evidence to suggest that either o f the k i l n s ever h a d a d o m e . I n either k i l n it w o u l d have been necessary to raise the vertical walls of the ware chamber by 70 to 100 c m before arching c o u l d b e g i n i n order to a l l o w for a d o o r t a l l enough to permit free access to the ware chamber. T h i s w o u l d have made the chambers so large that it is very d o u b t f u l that the c o m b u s t i o n chambers w o u l d have been large e n o u g h to accomodate sufficient f u e l to fire the kilns to the needed temperature. I n b o t h k i l n s , the walls of the ware chamber slope slightly outward as they rise f r o m the pierced f l o o r . N o r were domes necessary. H a m p e a n d W i n t e r report a n u m b e r of kilns of this d e s i g n w h i c h were closed by simply covering the stacked pots w i t h two or three layers of sherds. T h e sherds provide little i n the way o f i n s u l a t i o n , but serve to d a m the hot gases a n d slow their passage t h r o u g h the pots. T h i s w o u l d have been sufficient for the relatively l o w firing temperatures used for the ware o f A u g s t and also helps e x p l a i n some o f the effects the potters were able to achieve i n their wares. K i l n s w h i c h have n o stack (Schornsteinreihe) do not i n d u c e a strong draft i n the early stages of firing a n d tend to b u r n w i t h a rather s m o k y flame w h i c h produces a pred o m i n a n t l y r e d u c i n g atmosphere. A s the firing progresses, the draft m a y b e c o m e m o r e lively a n d w i t h careful management of the fires o x i d i z i n g conditions c a n be m a i n t a i n e d . A t the e n d o f the firing as the fires are a l l o w e d to die d o w n , the draft continues drawing 0 into the k i l n o x i d i z i n g the still r e d hot wares giving the f i n a l effect o f o x i d a t i o n . T h i s is the technique by w h i c h wares f r o m Categories I A a n d C , III, I V a n d V - A were fired a n d accounts for the reduced kernel sometimes f o u n d i n these sherds. I n order to m a i n t a i n r e d u c i n g conditions u n t i l the wares h a d c o o l e d to the point where no r e o x i d a t i o n was possible, the potters c o u l d have simply covered over the top o f the k i l n w i t h a m i x t u r e o f soft clay a n d chopped straw prior to a l l o w i n g the fires to die a n d closing the fire m o u t h . T h i s c o u l d prevent the entry o f free o x y g e n and m a i n t a i n the wares i n a reduced state. I n order to smudge the pots as i n C a t e gory II, they c o u l d have f o l l o w e d the same procedure as above, being careful to stoke a large quantity of extra f u e l just before c l o s i n g the fire m o u t h . T h e heavy carbon smoke thus i n d u c e d w o u l d penetrate the ware s m u d g i n g it black. T h u s it was possible for the potters of A u g s t to fire o x i d i z e d , reduced or smudged wares i n these kilns w i t h a s i m p l e sherd cover. T h e area of the excavation was not large enough to ascertain whether the k i l n s sat i n a b u i l d i n g or were i n a n open court yard. D u e to the intense heat that they w o u l d have generated, it seems very u n l i k e l y that they w o u l d have been under a roof. K i l n I sits i n a corner f o r m e d by t w o masonry walls a n d the foundations o f the k i l n sit o n the foundations o f the walls i n d i c a t i n g that the walls predate the kilns. 104

105

106

2

The Potters Wheels T w o bits of evidence made it possible to reconstruct i n prototype at least the type of w h e e l used by the potters of Venusstrasse Ost. T h e first was a s m a l l rough-hewn limestone cup (Illustration 13) w i t h h o r i z o n t a l striations i n its central depression caused by rotation o f a shaft and the other is to be f o u n d i n the physical characteristics of the pottery f r o m the excavation. B o t h of these point unmistakeably to the sort o f basic w h e e l design s h o w n i n Illustration 4. T h e outside surfaces o f most o f the pots are s m o o t h and offer no clue as to the type o f equipment u s e d , but the finger ridges o n the insides invariably spiral upward to the right (viewed f r o m the inside) and show t h e m to have b e e n t h r o w n o n a w h e e l w h i c h rotated i n a counterclockwise d i r e c t i o n . T h e rule-of-thumb i n w h e e l construction is that hand powered wheels t u r n clockwise and foot powered wheels t u r n counter-clockwise. T h e r e are exceptions, but i n the m a i n the rule holds true due to that peculiarity of nature by w h i c h most people are right-handed a n d , to c o i n the t e r m , also right-footed. T h e n o r m a l m o t i o n w h e n setting a h a n d powered w h e e l , such as the one i n Illustration 5, into rotation is to grasp the outer perimeter w i t h the right hand and p u l l it sharply to oneself causing a clockwise rotation. V a r i a t i o n s o n this m e t h o d , such as using a stick i n a n o t c h o n the outer edge of the wheel usually produce the same d i r e c t i o n of spin. F o o t powered wheels, by contrast, are n o r m a l l y d r i v e n by k i c k i n g away f r o m oneself w i t h the right foot o n a m o m e n t u m w h e e l causing counter-clockwise rotation. T h e extreme forms of the flagons and the delicacy of the clay gloss coated wares indicate a very s m o o t h r u n n i n g w h e e l w i t h considerable m o m e n t u m and a l o n g d u r a t i o n of s p i n after each a p p l i c a t i o n of power. H a n d powered wheels usually have a short duration of spin and l o w m o m e n t u m due to the relatively s m a l l mass of the w o r k i n g head w h i c h also serves as m o m e n t u m w h e e l . T h e limestone cup c o n f i r m e d these indications and narrowed the choice of possible designs. Since the i n c e p t i o n o f the foot powered w h e e l d u r i n g H e l l e n i s t i c times, two basic designs have b e e n developed. T h e older o f the two was a simple adaptation of the even older h a n d w h e e l (see Illustration 6). T h e pivot point r e m a i n e d just under the center of the t h r o w i n g surface and a m o m e n t u m power w h e e l was dropped d o w n over the supporting f i x e d stake. I n the other v a r i a t i o n , s h o w n i n Illustration 4, the pivot point moves to the base of the shaft w h i c h connects the w o r k i n g head w i t h the m o m e n t u m w h e e l . Since the gyroscopic action o f the two wheels is not sufficient to m a i n t a i n the w h e e l i n a stable upright p o s i t i o n , one other point o f contact is n e e d e d ; this is usually located just u n d e r the t h r o w i n g head. T h e p r o b l e m is to secure the shaft and deny it lateral m o t i o n w h i l e at the same t i m e a l l o w i n g it to t u r n freely. I n m o d e r n w h e e l construction a variety o f bearings and bushings are used for this purpose; early potters, l a c k i n g these sophisticated devices, must have used m u c h simpler means to achieve the same purpose. T h e exact design of the m e c h a n i s m used by the potters o f A u g s t w i l l probably never be k n o w n w i t h certainty since the wheels were constructed of w o o d and have not survived the l o n g b u r i a l . T h e author knows o f two wheels o f traditional design w h i c h are still i n use near A u g s t . O n e is i n the shop of M r . G i s l e r i n Dällikon, K a n t o n Zürich, and the other i n one of the potteries of S o u f f l e n h e i m , A l s a c e ; b o t h are w e l l over h u n d r e d years o l d and b o t h use the same means of securing the shaft. T h e shaft is simply c l a m p e d between two pieces o f w o o d w h i c h have been carved out to the h a l f - r o u n d ; a piece of b a c o n r i n d wedged between the two sides of the clamp and the shaft serves as lubricant 107

and allows the w h e e l to t u r n freely (see Illustration 7b). A similar device was probably used i n the R o m a n wheels at A u g s t . A reconstruction of the wheel i n its most likely f o r m is s h o w n i n Illustration 7.

Illustration 4

Illustration 6

Illustration 5 ( A f t e r R i e t h , pp. 74)

a b c

Detail of b

1 Illustration 7. a) w h e e l head b) upper pivot point c) shaft d) m o n u m e n t u m w h e e l e) lower pivot point f) limestone cup g) o i l e d leather or something similar.

The Wasters A percentile b r e a k d o w n o f the types o f sherds f o u n d i n the excavation shows clearly that the area excavated was not a »city dump« as was the case w i t h the Schutthügel i n V i n d o n i s s a ; the m i n i m a l numbers o f T . S . sherds a n d sherds o f other k n o w n i m port types clearly indicate this. R a t h e r , the excavation brought to light the rather specialized trash pile o f one or m o r e o f the R o m a n potteries o n Venusstrasse Ost as demonstrated by the large n u m b e r of wasters w h i c h were f o u n d . T h e wasters f r o m the excavation break d o w n into firing failures w h i c h were caused by either the m e t h o d o f stacking or the fire itself, technical failures w h i c h resulted f r o m h u m a n error or poor techniques and failures caused by problems w i t h the clay. T h e flagon and bottle forms are g o o d examples o f firing failures caused by the m a n ner o f stacking the pots i n the k i l n to be fired. Profiles 1 A - C , 2 A and B , 3 A a n d 9 were a l l taken f r o m demonstrable wasters a n d a l l seem to have failed i n exactly the same way, i.e. f r o m pressure o n a few points o n the upper shoulder during the f i r i n g . T h e shelving used i n m o d e r n potteries is a relatively recent i n n o v a t i o n i n E u r o p e . A d a m W i n t e r and R o l a n d H a m p e i n their investigations into the pottery o f p r i m i t i v e areas o f G r e e c e , Italy and S i c i l y where k i l n s very s i m i l a r to those used i n A u g s t are still i n use do not m e n t i o n a single instance o f shelving being used to support the ware for firing. W i t h unglazed pottery, shelving is rarely needed even today since raw clay surfaces do not stick together w h e n fired. Illustrations 8 a n d 9 show two artifacts f o u n d i n the excavation w h i c h were probably used as stacking aids. T h e y w o u l d have been o f little value i n stacking closed forms such as the flagons and were probably used to stabilize stacks o f large bowls or mortaria. 108

A s the potters o f A u g s t were no doubt aware, w i d e - b e l l i e d flagon a n d bottle forms are d i f f i c u l t to stack without the aid o f shelving since they cannot be p i l e d i n a way w h i c h w i l l distribute the weight equally. T h e only practical m a n n e r o f stacking these forms w o u l d have been to pile the pots i n the k i l n m u c h as c a n n o n balls are stacked w i t h the first layer set out o n the pierced f l o o r (Lochtenne) side by side and t o u c h i n g and w i t h subsequent layers offset and resting o n the shoulders o f the pots below. T h e clay o f A u g s t is quite strong i n the dry state a n d w o u l d have permitted several layers o f flagons to be stacked without serious danger o f breakage, but the entire weight o f a l l the pots i n the k i l n w o u l d have been l o c a l i z e d to a few points o n the upper shoulders o f the b o t t o m layer o f flagons. A s the temperature i n the k i l n i n creased, these pressure points w o u l d be the first to warp or even collapse. If one pot o n the b o t t o m layer s h o u l d be c r u s h e d , the w h o l e l o a d c o u l d shift causing a disastrous c h a i n reaction o f breakage. S u c h a happening is the bane o f a potter's very existance a n d m a y have been the case w i t h the large flagons s h o w n i n P r o f i l e 1 A - C since a l l eight examples o f this f o r m type were f o u n d i n the same c o m p l e x a n d a l l failed because o f pressures o n the shoulder as described above. T h e piece used f o r P r o f i l e I B was d o o m e d to failure regardless o f the m a n n e r o f stacking a n d is a good example o f the sort o f failure attributable to h u m a n error. I n this case, the w a l l o f the pot as it adjoined the base ring was t r i m m e d so t h i n that it c o u l d not support the weight o f the rest o f the pot, and it broke at that point leaving the flagon w i t h o u t a b o t t o m . T h e break no doubt o c c u r r e d i n the k i l n , but cannot be b l a m e d o n the stacking or the f i r i n g ; it i s , rather, the fault o f the potter. T h i s k i n d o f error is frequently seen i n the w o r k of beginning potters or those whose technique is poor. A s discussed i n the section o n the flagons, I B was executed by

Potter II whose mastery of pottery techniques was far below the standard set by Potter I. T h e r e were also a n u m b e r of wasters a m o n g the b o w l s , one of the clearest of w h i c h is s h o w n i n the pot used for P r o f i l e 18E. A l l of the r i m , the w h o l e base and a n u m b e r of w a l l sherds to this piece were f o u n d , but it soon became apparent that reconstruct i o n was not possible due to severe sherd warpage. I n this case the failure does not appear to stem f r o m the m a n n e r of stacking the b o w l s , but f r o m their p o s i t i o n i n g i n the k i l n . D e e p bowls are usually stacked r i m - t o - r i m and foot-to-foot as s h o w n i n Illustration 10. T h i s style of stacking is usually preferred to the other alternative of nested (i.e. one i n the other) stacks since it distributes the weight m o r e evenly and avoids pressure o n the inside of the r i m s w h i c h often leads to r i m cracks. S m a l l bowls can be set i n large ones to conserve space. T h e entire weight of the stack is supported by the b o t t o m b o w l , but this is usually not a p r o b l e m since the pressure is evenly distributed around the r i m and comes directly f r o m above. T h e degree of sherd warpage and particularly the warpage of the base sherds suggests that 18E was the b o t t o m b o w l i n such a stack and that its base rested o n an u n e v e n surface i n the k i l n . Perhaps it was placed i n such a way that one-half of its foot extended over one o f the holes i n the pierced floor. T h e sherds do not appear to have been overf i r e d , so it is u n l i k e l y that heat alone c o u l d have caused the base to w a r p ; considerable pressure must also have played a role. S o m e times a single sherd offers enough evidence to state that it came f r o m a firing failure, such is the case w i t h 18D. O n this large r i m sherd the patterns of o x i d a t i o n and r e d u c t i o n indicate that the pot broke during the firing. O n e b r o k e n edge shows a clearly defined kernel of reduced grey, w h i l e the other is fully o x i d i z e d . T h e explanation lies i n the change i n c o m p o s i t i o n of k i l n gases during the firing cycle. I n the early stages of the fire, the atmosphere i n the k i l n was reducing w h i c h is to say that u n b u r n e d carbonaceous gases pervaded the k i l n . S u c h gases are c h e m i c a l l y very active and are capable of r o b b i n g other materials of their o x y g e n content. I r o n o x i d e w h i c h is present i n nearly a l l kinds of clay is particularly susceptible and is easily »reduced« f r o m red ferric oxide ( F e 0 ) to black ferrous oxide ( F e O ) . If oxygen is a l l o w e d to enter the k i l n w h i l e the pots are still hot enough for a reaction to take place, i.e. red heat or m o r e , the i r o n w i l l revert to its ferric state on exposed surfaces. If the t e r m i n a l o x i d a t i o n period continues l o n g enough or i f the ware is quite porous the paste may be r e o x i d i z e d throughout its entire thickness, leaving no traces o f residual r e d u c t i o n . I n the case of 1 8 D , one b r o k e n surface still shows the reduced core whereas the other does not. T h e o x i d i z e d surface m u s t , t h e n , have been exposed during the f i n a l o x i d a t i o n p e r i o d clearly i n d i c a t i n g that the pot broke i n the k i l n . O n e other point of interest o n 18D is that the nature of the r e o x i d a t i o n supports the theory that the potters of A u g s t stacked their bowls r i m - t o - r i m . O n the r i m section of the sherd, the r e o x i d a t i o n has penetrated equally f r o m a l l directions leaving the reduced core i n a central position. I n the w a l l , the o x y g e n penetrated to a depth of 2 /2 m m f r o m the outside, but there is no penetration f r o m the inside. Since there c o u l d be no difference i n density, the lack of penetration f r o m the inside must be accounted for by the fact that o x y g e n was denied access to the interior of the pot. R i m - t o - r i m stacking accounts for this. That the potters of A u g s t also had problems w i t h the clay is evidenced by the base sherds o f P r o f i l e 10B. I n this large beaker f o r m , the base has cracked i n a peculiar 2

1

3

» S « shaped pattern. H e r e , again, the patterns of o x i d a t i o n and r e d u c t i o n o n the b r o k e n edges of the sherds show that the cracking occurred prior to the f i n a l o x i d a t i o n of the ware. T h i s k i n d of cracking is caused by pressures set up i n the clay d u r i n g the t h r o w i n g process w h i c h are not released u n t i l the pot is heated i n the k i l n . V e r y fine grained clays such as that used i n A u g s t exhibit this d i f f i c u l t y m o r e frequently than coarse clay.

Illustration 10. B o w l forms stacked r i m - t o - r i m and foot-to-foot.

54

Footnotes 1 2 3 4 5 6 7 8 9 10 11 12 13 14

15 16 17 18 19 20 21

22 23 24 25

26 27

O s w a l d and Pryce - Plate L X X I V and pp. 216/217. O s w a l d a n d Pryce - Plate L X X V , 1, and pp. 223. Steiger - A b b . 42,68. B i n s f e l d - pp. 23. E t t l i n g e r - Taf. 2 2 , 6 and 14. V o g t - A b b . 4 4 , 4 2 , and A b b . 4 2 , 3 and 5. D r e x e l - A b b . 6,2. Oelmann - F o r m Type 32C. D r e x e l - pp. 81. L o e s s is a n exceptionally fine aeolean sand w h i c h takes its name f r o m the enormous deposits i n C h i n a . T h e one s m a l l difference is that R o m a n sherds contained a negligible a m o u n t of Z i n c w h i c h was not f o u n d i n the clay sample f r o m Insula X X I I . S l a k i n g is the p h e n o m e n o n by w h i c h any clay disintegrates w h e n dropped i n water. T h i s process is k n o w n to have been used by G r e c o - R o m a n potters and is one of the first steps i n the p r o d u c t i o n o f clay gloss coatings. F i r i n g a n d saturation tests were conducted i n the testing laboratory o f the Dachziegelwerk Frick, Kt. Aargau, w i t h the k i n d p e r m i s s i o n o f H e r r D i r e k t o r R o t h , the G e n e r a l D i r e c t o r of the factory. N o b l e II - pp. 309. M a r l is a heavily calcareous clay. D r e x e l - p. 80. B i m s o n - p. 200. Georges - p . 2658 - »sigillatus, -a, -um, mit kleinen Figuren (in erhabener Arbeit) verziert...« N o b l e - pp. 38-39 and figures 178-179. D e f l o c c u l a t i o n is the dispersal of the s o l i d matter i n a water suspension t h r o u g h the a d d i t i o n o f an electrolyte w h i c h effects the positive and negative charges o n the surface of the particles causing t h e m to repel rather than attract and affecting a drop i n the specific gravity o f the slip without the a d d i t i o n o f more water. B i m s o n - pp. 200-204. Parmelee - p. 88. F a r n s w o r t h a n d W i s e l y - pp. 165-173. L a c k o f a clear understanding o f this p h e n o m e n o n has contributed greatly to some o f the c o n f u s i o n f o u n d i n the artificial classification o f some wares as T e r r a Sigillata and some as T e r r a Sigillata imitations. T h i s entire area is sorely i n need o f c l a r i f i c a t i o n and réévaluation. T h i s is not to be confused w i t h »flamed« or »sponge decorated« ware where the m o t t l i n g was i n t e n t i o n a l . D u r i n g the course o f our efforts to reproduce the clay gloss used by the potters of A u g s t , we happened to fire one o f the pieces twice. T h e first firing was done i n a reducing atmosphere i n the hope that we c o u l d achieve the grey to black o f Category I B . D u e to the fact that we were firing i n a n electric k i l n , we were forced to delay the i n t r o d u c t i o n o f the r e d u c i n g gases u n t i l the e n d

28 29

30

31 32

33

35

36

37

38

o f the f i r i n g cycle (carbon m o n o x i d e is h a r m f u l to the electric heating elements i f it reaches t h e m w h i l e they are being heated by a n electric current). Since the ware was already matured at that p o i n t , the clay gloss had already sintered a n d we a c c o m p l i s h e d only a partial r e d u c t i o n o f the ware. W e t h e n applied another coat o f clay gloss a n d refired the piece i n a purely o x i d i z i n g atmosphere. T h e second firing was also a failure i n that the second layer o f gloss, w h i c h was very d i f f i c u l t to apply due to the sintering o f the first coat and the l o w absorbency o f the surface, peeled badly. T h i s clearly demonstrates that D r a c k (p. 28) was m i s t a k e n i n his assumption that the »flamed« or »sponged« wares were made i n a two-fire process. Shepard - pp. 220-221. Sherds f r o m types 14 B - E appeared i n a l l three categories o f clay gloss wares w i t h n o apparent regularity except that the finest examples were done i n C a t e gory I B and the roughest i n Category I C . T h e most notable m o d e r n exponents o f the technique are M a r i a M a r t i n e z and her s o n , P o p o v i D a , the famed A m e r i c a n I n d i a n potters o f S a n Ildefonso P u e b l o i n N e w M e x i c o . M a r i a ' s »black-on-black« pottery has brought her international fame a n d is i n c l u d e d i n m u s e u m s a n d pottery collections i n a l most every country i n the w o r l d . T h e procedure most l i k e l y used i n the k i l n s o f A u g s t is discussed briefly i n the section o n K i l n s . I n their report o n their research o n terra nigra (»Versuch über die Rekonstruktion der Terra Nigra« - G e r m a n i a 31, 1953, pp. 67-68) E d g a r D e n n i n g e r a n d H e i n z E b i n g e r agreed that even clays c o n t a i n i n g a considerable a m o u n t o f i r o n need the a d d i t i o n a l c o l o r i n g action o f c a r b o n to produce a good black color. U n f o r t u n a t e l y , their w o r k was marred by their assumption that it w o u l d have been necessary to pack the pots i n charcoal for firing to facilitate the carbon i m p r e g n a t i o n . P o l i s h i n g the ware after firing as suggested by D r a c k - p. 27 - is not possible as e x p l a i n e d i n the description o f the process above. I n the f i r e d state, it is not possible to rearrange the clay particles; they are permanently f i x e d . It is a virtual certainity that this piece was made i n A u g s t . T h e interior surface shows the characteristic l u m p s o f u n s l a k e d clay a n d A u g s t clay takes a n excellent p o l i s h w h e n b u r n i s h e d , n o r , considering the t h r i v i n g pottery industry i n A u g s t , w o u l d it have been necessary to i m p o r t ware o f this type. A good m a n y forms carried out i n this t e c h n i q u e , particularly b o w l f o r m s , are indiscriminately classified as Terra Sigillata imitations. T o be sure some of them do m i m i c traditional sigillate f o r m s , but others o f w h i c h D r a c k T y p e 20 is a good example are purely aboriginal i n b o t h f o r m a n d technique. D r a c k h i m self notes this. W h y t h e n the designation as a »Terra Sigillata imitation« w h e n neither the f o r m n o r the technique are imitative ? I h o l d the one e x c e p t i o n to this, the piece used for P r o f i l e 2 B , to have been a shop accident i n w h i c h some o f red clay gloss was inadvertently spilled over a n already engobed flagon. T h e r e is no discernable pattern a n d the rem a i n i n g areas o f red pigment c o u l d scarely have been applied w i t h a brush. T h e r e c a n be no doubt that these size graduations were i n t e n t i o n a l o n the part o f the potter i n that w h e n the r i m fragments were sorted into their appropriate size groups, they varied less than 1 m m one f r o m the others. T h e m e d i a n

39 40 41 42

43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

63 64 65 66

67 68 69 70 71 72

diameters of the six sizes of the m o u t h rims are : 30 m m , 33 m m , 37 m m , 41 m m , 49 m m and 68 m m . Inv. N r s . 68.13359,68.13096 and 68.13452. Inv. N r . 68.13727. F o r a fuller discussion o f the i m p l i c a t i o n s of this treatment, see the section o n D e r i v a t i o n s of the F o r m s . A rib is a t o o l usually made of w o o d or metal w h i c h is used instead of the fingers i n t h r o w i n g pottery. M o s t ribs are rather flat and have at least one long straight or slightly curved surface. T h e advantages to using a rib are that the surface of the piece can be s m o o t h e d m o r e completely than w i t h the fingers and less water can be used i n the t h r o w i n g process. T h i s is particularly h e l p f u l i n t h r o w i n g large pieces since excessive water softens the clay and often leads to s l u m p i n g . V o g t - A b b . 43,12 and 48,20. D r e x e l - p. 98. Schoppa - p. 50. L a u r - p. 158. L a u r - p. 160. L a u r - p. 160 and Stähelin - p. 557. F r a z e r - C h a p t e r I. E t t l i n g e r and Simonett - Taf. 28,7 and p. 9. V e r m a s e r e n - p. 6. F r a z e r - p. 164. V e r m a s e r e n - p. 31. M o r r i s and M o r r i s - pp. 11-49. Stähelin - p. 551. Stähelin - p. 552 and E t t l i n g e r and Simonett - pp. 108 and 128. Stähelin - pp. 554,556. M o r r i s and M o r r i s - p. 33. M o r r i s and M o r r i s - p. 3. M o r r i s and M o r r i s - p. 48. G r a i l l o t - p. 3. I a m indebted to D r . R u t h Steiger, M u s e u m s a s s i s t e n t i n at the R ö m e r m u s e u m , A u g s t , for a l l o w i n g me to m e n t i o n this piece prior to the p u b l i c a t i o n of her o w n w o r k o n the same subject w h i c h was i n progress at the t i m e of this writing. O e l m a n n - F o r m T y p e 33. O e l m a n n - p. 35. »Roter Ton mit ausgezeichnetem schwarzem oder schwarzbraunem Überzug von metallischem Hochglanz«. E t t l i n g e r and Simonett show a foot treatment s i m i l a r to ours o n A b b . 14,11, but their piece is i n »rough cast« or »sand faced« ware - a technique w h i c h is totally absent f r o m this excavation. E t t l i n g e r - Taf. 23,15 and 16. D r e x e l - p. 82. E t t l i n g e r - Taf. 22,19; Taf. 23,1-4 and p. 87. D r e x e l - Taf. X I , 2 and p. 97. E t t l i n g e r - p . 86. V o g t - A b b . 44,42 and A b b . 42,3 and 5.

73 74 75 76 77 78 79 80 80a 80b 81

82 83 84 85 86 87 88 89 90 91 92

93 94 95 95a

O e l m a n n - T y p e 32c. D r e x e l - A b b . 6,2. E t t l i n g e r - Taf. 22,6 and 14. E t t l i n g e r - Taf. 20,19-43 and pp. 103-104. E t t l i n g e r - p. 103. E t t l i n g e r - Taf. 20,40. V o g t - A b b . 44,34. E t t l i n g e r - Taf. 19,4 and 5. E t t l i n g e r - Taf. 21,12 and 13. See also V o g t - A b b . 40,22. O s w a l d and Pryce - Plate L X X I V , 2. A m o n g our mortaria sherds there are three i d e n t i c a l firing failures and none of the sherds show any sign of use. T h i s leads one to speculate that a l l of the m o r t a r i a f o u n d i n the excavation w i t h the e x c e p t i o n of the T.S. piece are wasters of one sort or another and that the sherds left to us are simply not recognizable as such. T h i s is a logical assumption i n v i e w of the nature of the other finds. It is p u z z l i n g that the one T . S . mortaria m e n t i o n e d above is also i n pristine c o n d i t i o n and the w o r k i n g surface is neither scratched nor w o r n i n any way. Since it is a k n o w n i m p o r t type it cannot be regarded as a waster, and another interpretation is o b v i o u s l y called for. I n this case, the lack of wear probably means that the pot was b r o k e n soon after manufacture and i m p o r t a t i o n and its value i n dating the strata becomes m o r e important, because we can be relatively certain that we are not dealing w i t h the »heir-loom« p h e n o m e n o n m e n t i o n e d by H a w k e s and H u l l . D r e x e l - p . 73. E t t l i n g e r and Simonett - Profiles 434, 440, 453, and 530; O e l m a n n - A b b . 27, N r s . 2b, 2c, 2k, 10a, et a l . ; O s w a l d and Pryce - Plates L X X X I I and L X X X I I I . U n v e r z a g t - A b b . 13,1 and 3 and A b b . 14,12. Prof. D r . E l i s a b e t h E t t l i n g e r - personal interview, Zürich, N o v . 1970. M a j o r - T a f . X X I I , 15. R i t t e r l i n g - Taf. X X X V I I , 120A. M a j o r - Taf. X X I I , 13 and 18-20. R i t t e r l i n g - Taf. X X X V I I , 120A and 121. E t t l i n g e r and Simonett - Taf. 1,12. V o g t - A b b . 31,1. A s a g o o d example o f this, see M a j o r - Taf. X X I I , 15. It is not u n u s u a l i n the history of ceramics to f i n d that techniques w h i c h were originally dictated by the nature of the materials c o n t i n u e d l o n g after the necessity had disappeared and became style factors and characteristics of the pottery of the culture. A good example of this is f o u n d i n Japan where some potters continue the trad i t i o n of m a k i n g even s m a l l pots by a m e t h o d w h i c h combines c o i l b u i l d i n g and t h r o w i n g rather t h a n the simpler process o f t h r o w i n g t h e m i n one section. I n earlier times i n that r e g i o n , the clay was so non-plastic that no other technique c o u l d be used. N o w , despite the availability of better clay, the potters continue this c o m p l i c a t e d process »because it has always been done so«. D r e x e l - p. 74. E t t l i n g e r and Simonett - Taf. 28,7. E t t l i n g e r and Simonett - p. 8. E . B o n i s i n her report o n the striped pots f r o m Brigetio ( F o l i a A r c h a e o l o -

95b 95c 96 97 98 99 100 101

103 104

105 106

gica X X I - Separatum pp. 89) states that the region surrounding Brigetio had no indigenous p o p u l a t i o n and that the forms f o u n d there were not developed locally. Nevertheless, one cannot assume that the V i n d o n i s s a L e g i o n , w h i c h was m o v e d to B r i g e t i o , brought this style of ware w i t h t h e m i n this f o r m since no similar pieces have ever been f o u n d i n V i n d o n i s s a . C o n s i d e r i n g the p r o x i m i t y of A u g s t to V i n d o n i s s a , one can hardly avoid the speculation that the V i n d o n i s s a L e g i o n may have carried the influences to Brigetio f r o m A u g s t . B o n i s I I - A b b . 2,1 and 2. W a l k e - Taf. 57 and V o g t - A b b . 42,2. M a j o r - T a f . X X I , 2. Schoppa - Taf. 13 and 14 and p. 54. E t t l i n g e r - Taf. 16,1 and p. 96. D r e x e l - Taf. X I , 2 6 and p. 76. R i t t e r l i n g - Taf. X X X V I , 1 2 1 and p. 349. F o r T y p e 13 see E t t l i n g e r - Taf. 23,16 and p. 87; for T y p e 14 see E t t l i n g e r Taf. 23,1-4 and p. 87; for T y p e 16 see V o g t - A b b . 42,3-5 and p. 186 and E t t l i n g e r and Simonett - Taf. 12,247 and p. 53; for T y p e 12 see above under P r o f i l e T y p e 10; for general i n f o r m a t i o n see E t t l i n g e r pp. 82-89, E t t l i n g e r and Simonett pp. 38,41 and D r e x e l pp. 80-84. H a m p e and W i n t e r - A b b . 25,72 and 83. T h e author is aware of a statement to the contrary i n an article by R . M . S w o b o d a , D e r Töpfereibezirk a m Südostrand v o n A u g u s t a R a u r i c a , H e l v e t i a A r c h a e o logica5(1971,2),pp.7-21. H a m p e and W i n t e r - A b b . 25,73 and 83 and Taf. 13. O x i d a t i o n and r e d u c t i o n are terms used i n ceramics to describe the character of the hot gases passing through a k i l n . I n o x i d a t i o n , the gases contain sufficient o x y g e n that a l l materials i n the k i l n can c o m b i n e w i t h 0 to their m a x i m u m possible, such as F e 0 i n the case of i r o n . U n d e r reducing conditions the gases contain carbon m o n o c i d e ( C O ) w h i c h reduces the oxide constituents of the materials i n the k i l n to their lowest o x i d e content, or F e O for i r o n . R i e t h (p. 52) reports that a foot powered w h e e l was f o u n d i n C i n e l l i near A r e z z o i n 1779 and that it was very similar i n design to wheels i n use i n Italy i n recent times. U n f o r t u n a t e l y this valuable f i n d d i d not survive the excavation. It, t o o , was constructed of w o o d and c o u l d very w e l l have b e e n an Italian counterpart o f our wheels. W o r k i n g i n G e r m a n and w r i t i n g i n E n g l i s h has l e d to some difficulties w h i c h merit discussion. I n German-language archaeology the t e r m Fehlbrand is a l l inclusive and refers to any pot w h i c h failed for any reason d u r i n g the course of its manufacture, whether it had to do w i t h the firing process or not. F o r this reason, the obvious translation into E n g l i s h of »firing failure« is inappropriate. »Waster«, however, carries the same c o n n o t a t i o n and w i l l be used here i n place of Fehlbrand as a catch-all t e r m , and f i r i n g failure w i l l be used for those pieces i n w h i c h the failure was a direct result of the f i r i n g or the stacking. 2

2

107

108

3

F o t o 2. K i l n I T o p view showing oval c h a m b e r and a p o r t i o n of the floor.

F o t o 3. K i l n II F r o n t view showing the stoke hole and the entrance to the ware chamber.

F o t o 5. K i l n II T h e pierced floor of the ware chamber. N o t e the finger marks left by the potters as they smeared the floor w i t h soft clay to s m o o t h it.

Plate

1-1:4

Î A - F and 2 A - C - F l a g o n s , reddish paste, white engobe. 3 A - B and 4 - Bottles, reddish paste, white engobe, red clay gloss decoration. 5 - B o t t l e , reddish paste, white engobe. 6 , 7 - Jugs, heavily gritted paste, grey to black.

8 - F l a g o n , reddish b r o w n paste, pressed c o i l decoration. 9 - F l a g o n , reddish paste, white engobe, rouletted c o i l decoration.

Plate 111-1:2 ÎOA-B - Beakers, reddish paste, red clay gloss, rouletted decoration. 1 0 C - D - Beakers, reddish paste, burnished gold m i c a gilt. I I A - Beaker, smudged black, burnished surface.

PlateIV-1:2 IIB - Beaker, smudged black, burnished surface. 1 2 A - B - Beakers, grey paste, grey to black clay gloss, rouletted and indented. 13A - Beaker, grey paste, grey clay gloss, rouletted. 13B - Beaker, reddish b r o w n paste, light b r o w n clay gloss, rouletted.

Plate V - 1 : 2 1 4 A - F - Beakers, red, grey or black clay gloss.

1 5 A - B - Beakers, r e d , grey or black clay gloss. 1 6 A - C - Beakers, reddish paste, red clay gloss, rouletted. 17 - B o w l , grey paste, grey clay gloss.

1 8 A - E - C a r i n a t e d bowls, reddish paste, g o l d m i c a gilt.

18 H

18 J

18 L

18 P

Plate V I I I - 1 : 2 1 8 G - I and 1 8 K - P - C a r i n a t e d b o w l s , reddish paste, gold m i c a gilt. 18J - C a r i n a t e d b o w l , reddish paste, red clay gloss.

20 B

PlateIX-1:2 1 9 A - E - B o w l s , reddish paste, g o l d - m i c a gilt. 2 0 A - B - B o w l s , reddish paste, gold m i c a gilt.

PlateX

-1:2

21 - B o w l , reddish paste, red clay gloss. 22 - B o w l , reddish paste, red clay gloss or g o l d m i c a gilt. 2 3 A - B - B o w l s , grey to black paste, unadorned. 24 - Plates, diverse finishes.

PlateXI-1:2 2 5 A - B - C o o k i n g vessels, heavily gritted paste, grey to black. 2 6 A - B - C a r i n a t e d b o w l s , grey to reddish paste, u n a d o r n e d .

Plate XII

-1:4

2 7 , 2 8 , 2 9 A - B - M o r t a r i a , reddish paste, red clay gloss. 30 - M o r t a r i u m , reddish paste, stamped r i m (stamp see 111. 12). 3 1 A - C - M o r t a r i a , reddish paste, unadorned. 31D - M o r t a r i u m , light reddish y e l l o w paste w i t h reddish striations.

Inventory of the E x c a v a t i o n Totals

Profile N u m b e r

Inventory N u m b e r of P r o f i l e d Piece

1A IB 1C ID IE IF

68.15024 68.15026 68.15025 68.14161 68.13856 68.14547

5 1 3 3 1 3

16

2 A r i m 0 41 m m 2B r i m 0 37 m m 2 C r i m 0 68 m m T y p e 2 r i m 0 30 m m T y p e 2 r i m 0 33 m m T y p e 2 r i m 0 49 m m T y p e 2 size undeterminable

68.14653 68.13503 68.13359

18 22 3 1 2 10 2

58

3A 3B

68.15022 68.15023

1 1

2

4

68.13796

4

5

68.14554

1 1432

T y p e s 1-5:1327 W S a n d 105 B S were discarded 6

68.13897,68.14066,68.14539

7

7

68.13731

5

8

68.13844

1

9

68.14545

1

10A 10B 10C 10D

68.13424 68.13426 68.14008 68.13770

3 7 13 8

IIA I I B (not f r o m this excavation)

68.13866 70.5724

1

12A 12B

68.13386,68.13400 68.13387,68.13401

15 28

43

13A 13B

68.13388,68.13402 68.13997

9 1

10

31 1

Profile N u m b e r

Inventory N u m b e r o f P r o f i l e d Piece

Totals

14A 14B 14C 14D 14E 14F

68.13385 68.14430 68.13860 68.13741 68.13859 68.13742

32

32

64

15A 15B

68.13484 68.14445

18 1

19

16A 16B 16C

68.13552 68.13906 68.14453

10 8 1

19

17

68.14449

18A 18B 18C 18D 18E 18F 18G 18H 181 18J 18K 18L 18M 18N 180 18P

68.13810 i 68.13075 68.13811 68.13823 . 68.13746 68.13748 68.14054 68.14272 68.13681 68.13237 68.13828 68.13815 68.13668 68.13667 68.13148 68.13666

1 70 29 48 86 16 263 7

T y p e 18: 446 R S , 397 W S a n d 105 B S (some m a y have belonged to type 19) were discarded

519

948

19A 19B 19C 19D 19E

68.13498 v 68.13231 68.13674 68.13153 1 68.14057 1

20A 20B

68.13830 68.13829

21

68.13806

4

22

68.14264,68.14265

6

18 o y

27

1 1

2

Profile N u m b e r

Inventory N u m b e r o f P r o f i l e d Piece

Totals

23A 23B

68.13071 68.14610

8 10

T y p e 23: 53 W S were discarded

53

24 68.13835 T y p e 24: 1191 W S B S were discarded 25 A 25B

18

68.13443 68.13066

4 1191

1195

93 12

105

T y p e 25: 278 W S a n d 146 B S were discarded

424

26 A

68.14025

6

26B

68.13876

1

27

68.13766

1

28

68.14629

1

29A

68.13696

9

29B 30

68.14319 68.13767

1

31A 31B 31C 31D

68.13167 68.14512 68.14028 68.14026

5 10 1 2

111. 13

68.7579

7

10 5

18

Bimson Binsfeld Bonis I B o n i s II Charleston Drack Drexel Ettlinger E t t l i n g e r and Simonett Farnsworth and Wisely Fischer Frazer Graillot Georges Haffner H a m p e and Winter Hawkes and Hull Laur Major Morris and Morris Munsell Noble I N o b l e II Oelmann Oswald and Pryce

M a v i s B i m s o n , T h e T e c h n i q u e o f G r e e k B l a c k a n d T e r r a Sigillata R e d , A n t i q u a r i e s J o u r n a l 36,1956. W o l f g a n g B i n s f e l d , Z u d e n römischen Töpfereien a m R u d o l f p l a t z i n K ö l n , K ö l n e r Jb. für V o r - u . Frühgeschichte 7,1964,19. E v a B o n i s , F o l i a A r c h a e o l o g i c a X X I , Sep., Budapest 1970. R e i Cretariae R o m a n a e F a u t o r u m A c t a I X , 1967. R . J . C h a r l e s t o n , R o m a n Pottery, L o n d o n 1955. W a l t e r D r a c k , D i e helvetische Terra-sigillata-Imitation, Schriften d . Inst. f. U r - u . Frühgesch. 2, Basel 1945. F r i e d r i c h D r e x e l , D a s K a s t e l l F a i m i n g e n , D e r obergermanischrätische L i m e s B d . 66 C , L e i p z i g 1929. E l i s a b e t h E t t l i n g e r , D i e K e r a m i k der A u g s t e r T h e r m e n , M o n o g r . z. U r - u . Frühgesch. d. S c h w e i z 6, Basel 1949. E l i s a b e t h E t t l i n g e r u n d C h r i s t o p h S i m o n e t t , R ö m i s c h e K e r a m i k aus d e m Schutthügel v o n V i n d o n i s s a , Veröff. d . G e s . P r o V i n d o n i s s a 3, B a s e l 1952. M a r i e Farnsworth and Harriet W i s e l y , F i f t h Century Intentional R e d , A m e r i c a n J o u r n a l of A r c h a e o l o g y 62,1958. U l r i c h F i s c h e r , C a m b o d u n u m f o r s c h u n g e n 1953 I I , Mat.hefte z. Bayer V o r g e s c h . 10, K a l l m ü n z 1957. J . G . F r a z e r , A d o n i s , A t t i s , O s i r i s , Studies i n the H i s t o r y of O r i e n t a l R e l i g i o n , L o n d o n 1906. H e n r i G r a i l l o t , L e culte de C y b è l e , mère des d i e u x à R o m e et dans l'empire r o m a i n , Paris 1912. K a r l E r n s t G e o r g e s , Ausführl. L a t . - D t . Handwörterbuch, Basel 1962. A . H a f f n e r , D a s Treverer-Gräberfeld m i t Wagenbestattungen v o n Hoppstädten-Weiersbach,Kreis B i r k e n f e l d , T r i e r e r Z s . 32,1969,71 ff. R o l a n d H a m p e u n d A d a m W i n t e r , B e i Töpfern u n d Z i e g l e r n i n Süditalien, S i z i l i e n u n d G r i e c h e n l a n d , M a i n z 1965. C . F . C . H a w k e s a n d M . R . H u l l , C a m u l o d u n u m , F i r s t R e p o r t o n the E x c a v a t i o n s at Colchester 1930-39, O x f o r d 1947. R u d o l f L a u r - B e l a r t , Führer d u r c h A u g u s t a R a u r i c a , Basel 1966. E m i l M a j o r , G a l l i s c h e A n s i e d e l u n g m i t Gräberfeld b e i B a s e l , Basel 1940. R a m o n a a n d D e s m o n d M o r r i s , M e n a n d Snakes, L o n d o n 1968. M u n s e l l S o i l - C o l o r C h a r t s , ed. by M u n s e l l C o l o r C o m p a n y , 2441 N . Calvert St., B a l t i m o r e 18, M a r y l a n d 1954. Joseph V e a c h N o b l e , T h e T e c h n i q u e s o f P a i n t e d A t t i c Pottery, N e w Y o r k 1965. Joseph V e a c h N o b l e , T h e Technique of A t t i c Vase Painting, A m e r i can J o u r n a l of A r c h a e o l o g y 64,1960. F r a n z O e l m a n n , D i e K e r a m i k des K a s t e l l s N i e d e r b i e b e r , M a t . z. röm.-germ. K e r a m i k 1, F r a n k f u r t 1914. F e l i x O s w a l d a n d T . D a v i e s P r y c e , A n I n t r o d u c t i o n to the Study o f T e r r a Sigillata, L o n d o n 1920.

Parmelee Rieth Ritterling Schonfield Schoppa

Schörgendorfer Shepard Stähelin Steiger

Tomasevic Unverzagt Vermaseren Vogt Walke

W . C . Parmelee, C e r a m i c G l a z e s , Industrial P u b l i c a t i o n s , C h i c a g o 1951. A d o l f R i e t h , 5000 Jahre Töpferscheibe, K o n s t a n z 1960. E . R i t t e r l i n g , D a s frührömische L a g e r bei H o f h e i m i m T a u n u s , A n n a l e n des V e r e i n s für Nassauische A l t e r t u m s k u n d e 40,1912. H . J . S c h o n f i e l d , T h o s e Incredible C h r i s t i a n s , N e w Y o r k 1968. H e l m u t Schoppa, D i e F u n d e aus d e m V i c u s des Steinkastells H o f h e i m - M a i n t a u n u s k r e i s , Veröff. d. Landesamtes f. kulturgesch. B o denaltertümer B d . II, T e i l I, W i e s b a d e n 1961. A u g u s t Schörgendorfer, D i e römerzeitliche K e r a m i k der Ostalpenländer, M ü n c h e n 1942. A n n a O . Shepard, C e r a m i c s for the A r c h a e o l o g i s t , W a s h i n g t o n D . C . 1968. F e l i x Stähelin, D i e S c h w e i z i n römischer Z e i t , Basel 1948. R u t h Steiger, G . T h e o d o r S c h w a r z , R i c h a r d Strobel, H . W . D o p p l e r , Insula X X X I 1960/61, F o r s c h u n g e n i n A u g s t 1, L i e s t a l 1975 (forthcoming). T e o d o r a T o m a s e v i c , D i e K e r a m i k der X I I I . L e g i o n aus V i n d o n i s s a , Veröff. d. G e s . Pro V i n d o n i s s a 7, B r u g g 1970. W i l h e l m U n v e r z a g t , D i e K e r a m i k des K a s t e l l s A l z e i , M a t . z. röm.germ. K e r a m i k 2, F r a n k f u r t 1916. M . J . V e r m a s e r e n , T h e L e g e n d of A t t i s i n G r e e k and R o m a n A r t , L e i d e n 1966. E m i l V o g t , D e r L i n d e n h o f i n Zürich, Zürich 1948. N o r b e r t W a l k e , D a s römische D o n a u k a s t e l l S t r a u b i n g - S o r v i o d o r u m , L i m e s f o r s c h u n g e n 3, B e r l i n 1965.

Zusammenfassung 1968/69 wurde i m Südosten der Stadt A u g u s t a R a u r i c a , a n der Venusstrasse-Ost e i n grosser Töpfereibezirk freigelegt (s. P l a n S. 2). H i e r w i r d über die F u n d e aus e i n e m T e i l dieses B e z i r k e s berichtet ( A b b . S. 3). D i e untersuchte Töpferei muss etwa v o n 170220 n . C h r . i n Betrieb gewesen sein. Besonders guten A u f s c h l u s s s o w o h l i n technischer H i n s i c h t w i e i n B e z u g a u f d e n F o r m e n s c h a t z lieferte ein K o m p l e x , der sich unter d e m z u s a m m e n g e b r o c h e n e n B o d e n des Töpferofens II fand. H i e r Hessen sich ganze P r o f i l e aus d e n Scherben wiederg e w i n n e n u n d die Oberflächen der Keramikstücke w a r e n n i c h t d u r c h Bodeneinflüsse angegriffen. D a m i t eignete s i c h dieser F u n d k o m p l e x i n besonderer W e i s e für die t e c h n i s c h e n B e o b a c h t u n g e n u n d U n t e r s u c h u n g e n , d e n e n diese Studie eines m o d e r n e n K e r a m i k f a c h m a n n e s gilt. D i e ausführliche B e s c h r e i b u n g der verschiedenen hier i n A u g s t gefundenen Töpferwaren dürfte weitgehend a u c h für andere Orte u n d Zeitabschnitte ihre Gültigkeit haben. E s w e r d e n als K a t e g o r i e n vorgelegt: / Glanztonwaren m i t eingehender A n a l y s e der Herstellungstechnik. I A u n d I B unterscheiden sich n u r d u r c h d e n o x y d i e r e n d e n oder r e d u z i e r e n d e n Feuerungsprozess, welcher i m einen F a l l e die K e r a m i k rot-orange, i m anderen olive-grau bis schwarz erscheinen lässt. Dass die unterschiedliche F a r b g e b u n g gewollt war, zeigt sich an d e n weitgehend verschiedenen F o r m e n i n b e i d e n K a t e g o r i e n . I A = P r o f i l e 1 4 A , 1 5 A , 1 6 A - C , 13B, 1 4 B - E . I B = 12, 1 3 A , 14A (nur das profilierte S t ü c k ) , 14D u n d E , 15A u n d B , 17. I C umfasst gröbere G l a n z t o n w a r e n m i t d e n P r o f i l e n 10A u n d B , 13B, 1 4 B - E , 2 1 , 2 2 , 2 4 , 2 7 , 2 8 u n d 29. // Geschmauchte Ware. Sie w i r d erzeugt, i n d e m i m letzten S t a d i u m des Brandes eine dicke R a u c h w o l k e d u r c h d e n O f e n gelassen w i r d , deren K o h l e p a r t i k e l sich i n d e n P o r e n des T o n e s festsetzen, der damit grau bis schwarz gefärbt w i r d . B e i I I A w i r d eine E n g o b e gebraucht u m d e n G l a n z z u erzeugen, die ausserdem n o c h geglättet w i r d . B e i IIB w i r d die Oberfläche nur geglättet u n d erhält einen Überzug. D e r G l a n z - E f f e k t d u r c h Glättung k o m m t b e i b e s t i m m t e n , sehr f e i n e n T o n e n zustande, w e n n sie n a c h der Glättung keiner starken S c h r u m p f u n g m e h r ausgesetzt s i n d , also auch nicht b e i h o h e n T e m p e r a t u r e n gebrannt werden. H C ist graue W a r e m i t matter, nicht geglätteter Oberfläche, n u r i n d e n P r o f i l e n 2 3 A undB. III Weisse Engobe, nur bei flaschenartigen Behältern v o r k o m m e n d . D e r matte Überzug ist selten r e i n weiss, meist elfenbeinfarbig. E r w i r d sehr leicht d u r c h Bodeneinflüsse angegriffen (Säuren !), was bis z u r fast gänzlichen Auflösung führen k a n n . I I I A : Gefässe, die n u r m i t weisser E n g o b e versehen sind. B e i H I B k o m m t n o c h eine B e m a l u n g m i t Glanztonfarbe hinzu. IV Ware mit Glimmerüberzug i n drei V a r i a n t e n . B e i I V A wurde der G l i m m e r zumeist i n n e n u n d a u f d e m R a n d v o n Schüsseln m i t e i n e m P i n s e l direkt nach d e m D r e h e n auf der Scheibe aufgetragen (Profile 18,19,20,24). B e i I V B wurde der G l i m m e r i m lederharten Z u s t a n d aufgetragen u n d d a n n eingeglättet (Profile IOC u n d D ) .

B e i I V C handelt es sich u m Schüsseln, die i n lederhartem Z u s t a n d u n d n a c h d e m die Oberfläche fertig überarbeitet w a r , i n n e n u n d aussen m i t e i n e m Glimmerüberzug versehen w u r d e n . Solche Stücke sind sehr selten. V Tongrundige Gefässe ohne spezielle B e h a n d l u n g der Oberfläche. VI Kochtöpfe, m i t 3 verschiedenen Sorten v o n M a g e r u n g . D a b e i war auffällig, dass die w e n i g e n Scherben handgeformter W a r e sich a u c h i n der M a g e r u n g strikte v o n den anderen unterschieden. D i e s e Stücke dürften älter sein als die sonstige hier besprochene K e r a m i k . E i n e U n t e r s u c h u n g über das Rohmaterial der Töpferei ergab, dass der i n A u g s t selbst r e i c h l i c h anstehende »verlehmte L ö s s l e h m « verwendet wurde. E s liess sich ferner feststellen, dass alle Sorten der besprochenen K e r a m i k b e i etwa 8 4 0 ° - 8 8 0 ° C gebrannt w o r d e n waren. E s w u r d e n a u c h B r e n n v e r s u c h e m i t d e m A u g s t e r L e h m angestellt, welche formgleiche K o p i e n der römischen Gefässe m i t Glanztonüberzug ergaben, die i n F a r b e , Struktur u n d G l a n z v o n d e n echten nicht z u unterscheiden waren. D i e weisse E n g o b e zeigte i n ihrer c h e m i s c h e n Z u s a m m e n s e t z u n g , dass hier wahrs c h e i n l i c h M e r g e l t o n verwendet w u r d e , der nahe b e i m A u g s t e r Osttor z u f i n d e n ist. D e r G l i m m e r besteht aus M u s k o v i t , d e n m a n leicht aus z e r f a l l e n d e m G r a n i t ausschwemmen kann. A l l e beschriebenen F o r m e n dürften g l e i c h z e i t i g hergestellt w o r d e n sein. D i e Häufigkeit der e i n z e l n e n T y p e n ist aus d e m Inventory ersichtlich. I n v i e l e n Fällen sind Fehlbrände erhalten. D i e auffallendsten F o r m e n sind zweifellos die z w e i - oder d r e i h e n k l i g e n Krüge des T y p u s 1 u n d die nahe verwandte F o r m 3, die fast denselben U m r i s s zeigt, j e d o c h keine H e n k e l besitzt u n d dafür m i t einer roten G l a n z t o n b e m a l u n g auf der Schulter dekoriert ist. E s liess s i c h n a c h w e i s e n , dass Krüge der F o r m I A v o n der H a n d eines anderen (bessern!) Töpfers s t a m m e n als die gleichartigen Krüge 1B u n d C . K r ü g e des T y p u s 2 sind i m m e r m i t weisser E n g o b e versehen. D i e K a n n e n der P r o f i l e 6 u n d 7 bestehen i m m e r aus gemagertem T o n w i e die K o c h t ö p f e u n d sind teils h e l l t o n i g , teils grau. V o n d e n z a h l r e i c h e n , gut bekannten B e c h e r f o r m e n 10-16 sei h e r v o r g e h o b e n , dass 10A u n d B m i t gut p r o f i l i e r t e m F u s s i m m e r roten Glanztonüberzug tragen, während IOC u n d D m i t nicht abgesetzter Fusspartie m i t Glimmerüberzug versehen sind. D e r weitaus grösste A n t e i l a n der G e s a m t m e n g e fällt auf die Schüsseln der P r o f i l e 18 u n d 19 sowie auf die einfachen T e l l e r T y p 24. B e i b e i d e n F o r m e n überwiegt der Glimmerüberzug weitaus d e n roten G l a n z t o n . D i e P r o f i l e 1-16 zeigen eine starke Affinität z u alten L a t è n e f o r m e n , was a n v i e l e r l e i E i n z e l h e i t e n aufgezeichnet w e r d e n k a n n . D i e b e i d e n Öfen sind sehr gut erhalten. Ihre A b d e c k u n g w i r d m a n sich a m besten d u r c h mehrere L a g e n v o n T o n s c h e r b e n z u d e n k e n h a b e n , die b e i m r e d u z i e r e n d e n B r a n d n o c h m i t w e i c h e m T o n u n d S t r o h überdeckt w u r d e n . E i n e n H i n w e i s auf die Töpferscheibe bildet eine kleine runde K a l k s t e i n p l a t t e m i t Mittelloch und Drehspuren. ( E l i s a b e t h Ettlinger)

Résumé U n vaste quartier de potiers a été dégagé en 1968/69 au sud-est de l a v i l l e d ' A u g u s t a R a u r i c a , Vénusstrasse-Est (cf p l a n , p.2). L e s trouvailles d ' u n secteur déterminé, une poterie en activité a p p r o x i m a t i v e m e n t de 170 à 220 de notre è r e , sont étudiées i c i (fig. P. 3). D e précieux renseignements concernant technique et typologie ont été donnés par u n ensemble trouvé sous le sol effondré d u four de potier I L C e s tessons, dont les surfaces n'ont pas été altérées par le séjour sous terre, ont permis la r e c o n s t i t u t i o n de profils c o m p l e t s , fournissant ainsi le matériel nécessaire a u x observations et recherches technologiques d ' u n expert-céramiste m o d e r n e . L a description détaillée des différents types de poterie découverts à A u g s t devrait v a l o i r , en général, p o u r d'autres endroits et d'autres périodes. L e s catégories suivantes nous sont présentées: I - Poterie lustrée, avec une analyse détaillée de l a t e c h n i q u e de fabrication. L a seule différence entre I A et I B réside dans le procédé de c u i s s o n , résultant soit en o x y d a t i o n , soit en réduction, provoquant une c o l o r a t i o n rouge-orangée dans u n cas, o u olivegrisâtre à noire dans l'autre. L e s poteries de ces d e u x catégories étant de formes dissemblables, i l est évident que la différence de teinte est voulue. I A = Profils 1 4 A , 1 5 A , 1 6 A - C , 13B, 1 4 B - E . I B = Profils 12, 1 3 A , 14A (profil seulement), 14 D et E , 15A et B , 17. I C = Poteries lustrées grossières, profils 10A et B , 13B, 1 4 B - E , 21, 22, 2 4 , 2 7 , 2 8 et 29. /7 - Poterie fumée. E l l e s'obtient e i n faisant passer u n épais nuage de fumée par le four au dernier stade de l a cuisson. L e s particules de c h a r b o n se déposent dans les pores de l'argile et l u i donnent u n e c o l o r a t i o n grise a noirâtre. I I A : pour obtenir le lustre, l a poterie est enduite d ' u n engobe ultérieurement lissé. I I B : poterie sans engobe, à surface lissée. L e lustre est obtenu par simple lissage de l a surface de certaines argiles très fines a c o n d i t i o n de ne plus exposer l a poterie à une réduction et d'éviter une c u i s s o n à trop haute température ultérieurement. I I C : pterie grise, à surface mate, sans lissage. P r o f i l s 2 3 A et B seulement. III - Engobe blanc. E m p l o y é u n i q u e m e n t p o u r des récipients à usage de bouteille. L ' e n d u i t , m a t , est rarement blanc p u r , plutôt ivoire. F a c i l e m e n t altéré sous les i n fluences acides d u s o l , i l peut disparaître presque entièrement. I I I A : récipients à engobe blanc. H I B : récipients a engobe blanc ultérieurement lustrés. IV - Poterie micacée, en trois variantes. I V A : e n général, le m i c a est appliqué avec u n p i n c e a u à l'intérieur et sur le rebord des plats immédiatement après tournage sur tour. Profils 18,19,20 24. I V B : le m i c a est appliqué sur l a poterie alors qu'elle offre l a consistance d u cuir. L a surface est ensuite lissée. Profils IOC et D . I V C : i l s'agit de plats enduits de m i c a à l'intérieur et à l'extérieur après t e r m i n a i s o n d u travail de l a surface, l'argile ayant encore l a consistance d u cuir. C e s pièces sont très rares. V - Récipients

en terre glaise, sans travail spécial de l a surface.

VI - Pots (marmites). E m p l o i de trois sortes de dégraissants. L e s quelques tessons de cette catégorie, façonnés à l a m a i n , se distinguent des autres poteries par le c h o i x des dégraissants. C e s pièces sont probablement plus anciennes que celles dont nous avons parlé j u s q u ' i c i . L ' e x a m e n des matières premières utilisées à l'atelier de poterie en question a prouvé l ' e m p l o i fréquent, à A u g s t , de l a terre glaise locale. L e s céramiques énumérées ont été soumise à l a cuisson à l a température approximative de 8 4 0 ° - 8 8 0 ° C . L e s essais de cuisson faits avec l'argile native d ' A u g s t ont donné de véritables copies des poteries romaines lustrées, impossibles à distinguer des o r i g i n a u x quant à l a c o l o r a t i o n , l a structure et le lustre. L a c o m p o s i t i o n c h i m i q u e de l'engobe blanc laisse supposer l ' e m p l o i de l'argile marneuse d ' A u g s t , dont u n gisement se trouve près de l a porte de l'Est. L e m i c a est d u »Muskovit« facile à extraire d u granite e n désagrégation. T o u t e s les formes décrites datent probablement de l a m ê m e époque. L e tableau e n f i n de texte i n d i q u e l a fréquence de chaque type. D a n s plusieurs cas, des rebuts de cuisson nous sont parvenus. L e s cruches à d e u x o u trois anses d u type 1, et celles, apparentées, d u type 3, de f o r m e semblable, mais sans anse, avec u n lustre rougeâtre sur l'épaule, sont, sans conteste, les formes les plus remarquables. I l est prouvé que les cruches f o r m e 1 A sont de l a m a i n d ' u n potier plus expert que celles semblables 1B et C . L e s cruches d u type 2 sont toujours enduites d ' u n engobe blanc. L e s pots des profils 6 - 7 , e n terre glaise enrichie de dégraissant c o m m e les pots ( V I ) , sont tantôt de teinte terreuse, tantôt grisâtres. A u sujet des formes de gobelets 10 à 16, n o m b r e u x et b i e n c o n n u s , remarquons que 10A et B , au pied nettement profilé, sont couverts d ' u n lustre rouge, alors que IOC et D , dont le pied n'est pas profilé, sont enduits de m i c a . L a plus grande partie de l a poterie consiste e n plats des profils 18 et 19, et en assiettes d u type 24. D a n s les d e u x cas, l ' e n d u i t m i c a c é prédomine sur le lustre rouge. L e s profils 1-16 dénotent u n e grande affinité avec d'anciennes formes de L a T è n e , reconnaissable à de n o m b r e u x détails. L e s deux fours sont très b i e n conservés. I l faut les imaginer recouverts de débris de poterie en plusieurs couches et d ' u n supplément e n argile m o l l e et paille lors de cuissons p o u r obtenir une réduction. L a petite plaque circulaire en calcaire, avec u n creux au m i l i e u , portant des traces de tournage, i n d i q u e le t o u r d u potier. ( T r a d u c t i o n par Y v o n n e Carjat)