Conservation issues related to Egyptian faience: A closer look at damaged shabti from the Dutch National Museum of Antiquities

(1)

Conservation issues related to Egyptian faience:

A closer look at damaged shabti from the

Dutch National Museum of Antiquities

Master’s thesis

Corinna de Regt, MPhil/MA (5935784) University of Amsterdam, Amsterdam Supervisor: Mandy Slager

Second assessor: Bas van Velzen July, 2017

(2)

De Regt, UvA, 2017

Foreword and acknowledgements

This thesis is written in the scope of the glass, ceramics and stone specialization within the MA trajectory ‘Conservation and Restoration of Cultural Heritage’ at the University of Amsterdam (UvA). The research subject concerns damage phenomena on Egyptian faience shabti from the Dutch National Museum of Antiquities (RMO).1 A better understanding of the causes of damage is crucial for the safeguarding of these objects in the future and may be of use to conservators and curators dealing with faience collections worldwide.

First and foremost, I would like to express my gratitude to my mentor and supervisor Mandy Slager, who has been of immense support to me, providing me with valuable insights and directions that were essential to complete my research and write this thesis.

I would also like to thank UvA supervisors Prof. dr. Maarten van Bommel, Prof. dr. Ella Hendriks and Dr. Rene Peschar, for their guidance on technical research and academic writing and Dr. Luc Megens and Rutger Morelissen (RCE) for their advice and support and for carrying out SEM-EDX analysis. Without their help and expertise, this research would not have been possible.

Enormous gratitude also goes out to Renske Dooijes, conservator at the RMO and tutor in this process, for providing me with this interesting and challenging case-study and for her never ceasing hospitality and guidance throughout the research process.

Many thanks go to those whose works, research and valuable insights have helped me to write this thesis: Kate van Lookeren Campagne (UvA), Nienke Besijn (UvA), Guus Verhaar (UvA), Prof. Dr. Maarten Raven (RMO), Robert Ritter (RMO), Mariëlle Bulsink (RMO), Alejandra Marmonde (RMO), Prof. dr. rer. nat. Gerhard Eggert (ABK Stuttgart), Willem van Haarlem (Allard Pierson Museum), René van Beek (Allard Pierson Museum), Chris Wilkins (Cardiff University), Denise Ling (British Museum), Nina Loschwitz (Ägyptisches Museum Berlin), Nikè Haverkamp (UvA) and my fellow students Anne Rupert, Caitlin Southwick and Roy van der Wielen.

Finally, I would like to dedicate this thesis to William, who has always believed in me and encouraged me to pursue my dreams.

Corinna de Regt Leiden, June 2017.

(5)

Dutch language summary

Twee shabti (funeraire beeldjes) van Egyptisch faience werden in 2013 opgesteld in een tentoonstelling over mummiekisten in het Rijksmuseum van Oudheden (RMO) in Leiden. Nog geen jaar na afloop van de tentoonstelling, werd opgemerkt dat er breuken en barsten in het glazuur waren ontstaan, terwijl andere faience shabti uit dezelfde vitrine onaangetast leken te zijn. Wel werden in het depot nog meer shabti met verschillende vormen van schade aangetroffen. Om deze objecten zo goed mogelijk te kunnen behouden voor de toekomst, was het van belang dat meer inzicht verkregen zou worden in het verband tussen de kwetsbaarheid en de huidige conditie van het materiaal. De kwetsbaarheid van faience is gerelateerd aan productiematerialen en –technieken, waaronder drie verschillende glazuurmethoden. Het doel van dit onderzoek is dan ook geweest om meer inzicht te krijgen in verbanden tussen de productietechnieken van de shabti en de conditie waarin de objecten zich bevonden, wat zou kunnen bijdragen aan kennis over gevoeligheid voor degradatie.

Omdat breuken en barsten in de twee tentoongestelde objecten doen denken aan schade door het uitkristalliseren van oplosbare zouten, werd verwacht dat dit onderzoek eveneens de aanwezigheid van zouten zou bevestigen. Zouten vormen een cruciale component in de productie van faience, wat gevormd wordt uit een pasta van silica, kalk, metaaloxiden en water. Daarnaast zouden ook bodemprocessen, vroegere conserveringsmethoden, uitgassende vitrinekasten en andere aspecten een rol kunnen hebben gespeeld in de aanwezigheid van zouten in het materiaal. Na het testen van samples op chloriden, sulfaten en nitraten en onderzoek van vijf verschillende samples met behulp van Scanning Electron Microscopy/Energy Dispersive X-ray Spectroscopy (SEM-EDX), werden echter geen onverwachte zouten aan het licht gebracht. De SEM-EDX-analyse leverde daarentegen bruikbare informatie op voor het onderscheiden van de verschillende glazuurmethoden, waardoor meer inzicht verkregen kon worden in verschillende productietechnieken. Hierbij werd gelet op de dikte van de glazuurlaag, de herkenbaarheid van de tussenlaag, aan- of afwezigheid van glasachtig materiaal in de kern en de verspreiding van koperoxiden in de microstructuur van faience.

De resultaten van het onderzoek suggereren het gebruik van de ‘applicatie methode’, of ‘cementatie methode’, met weinig tot geen glasvorming in de kern. Dit heeft geleid tot fragiel kernmateriaal, dat voornamelijk bij elkaar gehouden wordt door een relatief dikke glazuurlaag. Barsten in het glazuur, vermoedelijk reeds ontstaan tijdens het

(6)

productieproces, hebben deze fragiele kern toegankelijk gemaakt voor invloeden van buitenaf en daarmee waarschijnlijk nog gevoeliger voor degradatie. Andere objecten die werden aangetroffen in het depot, vertonen juist een vrij massieve kern met een relatief dunne, kwetsbare glazuurlaag. Verschillen in productiemethoden lijken wel een rol te hebben gespeeld bij verschillen in kwetsbaarheid. De opgedane kennis over de aan het productieproces gerelateerde schade aan de shabti kan van nut zijn voor restauratoren en conservatoren die te maken hebben met het conserveren en behandelen van faience objecten.

(7)

English language summary

Two Egyptian faience shabti (funerary figurines) were put on display in an exhibition on mummy coffins at the Dutch National Museum of Antiquities (RMO) in Leiden in 2013. Within a year after the exhibition had ended, breaks and cracks were noticed in the glazed surface, while other faience shabti kept in the same display case, did not appear to be affected. In the depot however, eleven other damaged shabti were encountered. In order to safeguard the condition of faience objects for the future, it was relevant to gain more insight into the vulnerability of the material and the associated reasons for degradation. The vulnerability of faience is related to production materials and –methods, including the three different glazing methods. The aim of this research has been to investigate how the applied materials and techniques for the production of faience may have influenced the degradation of the shabti. Moreover, as the breaks and cracks in the two objects that had been exhibited suggested the crystallization of soluble salts, the expectation was that this research would confirm the presence of soluble salts. Salts form a crucial component in the production of faience, which is shaped from a paste of silica, lime and alkali metal oxides.2 Burial conditions, past conservation treatments, off-gassing display cases and other aspects could also have contributed to the presence of salts in the material.

After testing samples for chlorides, sulfates and nitrates and investigating five different samples by means of Scanning Electron Microscopy/Energy Dispersive X-ray Spectroscopy (SEM-EDX), no salt crystallization was detected. SEM-EDX analysis however did deliver useful information for the identification of different glazing methods, by which more insight could be obtained into different production techniques and their relation to observed degradation. Distinguishing characteristics were studied, such as thickness of the glaze layer, the nature of the boundary between interaction layer and core material, the presence or absence of interparticle glass in the body and the copper weight concentration throughout the faience microstructure.3

The outcome of the research suggests that either the ‘application method’ or ‘cementation method’ was used, which results in limited to no interparticle glass formation. This has led to fragile core material, that is mainly held together by a relatively thick glaze layer. Cracks in the glaze, which appear to have formed during the production process, have made the fragile core vulnerable for outside influences and therefore more

2_{See glossary (appendix I).} 3_{See glossary (appendix I).}

(8)

susceptible to degradation. Other objects that were encountered in the depot, seem to have a firm core and a relatively thin, vulnerable glaze layer. Different production methods seem to have played a part in these differences in vulnerability. The knowledge acquired about damage phenomena related to the production process of the shabti, could be of relevance to conservators and curators dealing with the conservation and restoration of faience objects.

(9)

Introduction

After an exhibition on Egyptian mummy coffins in 2013, staff members from the Dutch National Museum of Antiquities (RMO) in Leiden, the Netherlands, noted that two shabti had been heavily damaged (see fig. 1 and fig. 2).4 These figurines, produced by the ancient Egyptians as grave goods, were made of Egyptian faience.5 This material may easily be mistaken for ceramics or glass by its appearance, but in fact has a distinctive composition of silica, metal oxides and lime.6_{It has been called the ‘first high-tech ceramic’, despite the}

absence of clay minerals, to stress the unique position this earliest and longest-made vitreous material has in history.7_{The term ‘faience’ is somewhat misleading as it can also}

refer to a particular type of blue glazed earthenware.8 Many different names have been proposed to describe the material, but as ‘faience’ is still the most commonly used and in order to avoid confusion, this is the term that will be used here.9

While the two shabti are claimed to have been in a seemingly stable condition, they showed large cracks and loss of material after display.10 Because of this alarming

Figure 1: After an exhibition at the RMO in 2013, shabti F93/10.29 showed severe damage, as a part of the object had broken away, exposing the powdery core. A significant part of this core material is missing (Image and color checking: De Regt, 2016).

4_{The exhibition ‘Mummiekisten van de Amon-priesters’ (‘Mummy coffins of the Amon priests’) concerned} funerary objects found in Bab el-Gasus, a tomb at Deir el-Bahari in the Theban Necropolis in Egypt.

5_{Ushabtis, shawabtis, ushebtis of shabtis were made from a wide range of materials and were produced as} grave goods. As every individual was supposed to work for the gods in the afterlife, shabti substitutes were invented to escape this destiny. Schneider, Shabtis, 2-5.

6_{Nicholson, Egyptian Faience and Glass, 9.} 7_{Nicholson, “Materials and Technology,” 50.}

8_{European archaeologists used the misleading term in the nineteenth century (Matin and Matin, “Egyptian} Faience 1,” 763.). The blue glazes of ceramics produced in Faenza, seemed to them to be closely related to the blue vitreous material found in Egypt (Nicholson, “Materials and Technology,” 50).

9_{Terms like ‘blue frit’ are only adding to the confusion, as they are also used for other materials (Nicholson,}

Egyptian Faience and Glass, 16). The term ‘glazed siliceous ware’ seems too broad (Davison, Conservation and Restoration, 3).

10_{No condition reports exist for these objects. The only information comes from a publication by Schneider} from 1977, in which some of the objects discussed here are described and photographed (Schneider, Shabtis). These descriptions suggest that the major damage on object F93/10.29 was not yet present at that time, although it did contain a crack in the leg area where it has now split open (see chapter one).

(10)

7 De Regt, UvA, 2017 Figure 2: After an exhibition at the RMO in 2013, shabti F93/10.34 had cracked open. A small piece of the glaze has chipped off (Images and color checking: De Regt, 2016).

(11)

discovery, conservator Renske Dooijes was asked to check the condition of the objects. Puzzled by the sudden degradation, she suspected damage by soluble salts, as visual examination of the objects showed a fragile core that looked as if it had burst open, causing the glaze to crack. This observation suggested internal stress, possibly due to the presence of salt crystals. As salts are a crucial component in the production of faience, this was mentioned as another argument for their presence. It did not explain, however, why and how the salts would have caused damage after the exhibition, or why other faience shabti presented in the same display case, had not cracked. When quickly browsing through the entire collection of shabti stored in the depot, curator Prof. dr. Maarten Raven noticed eleven other shabti that showed damage (see fig. 3-5). It was decided to study these objects as well, as they would enable comparison when studying degradation of faience material.

As it is highly relevant for the future conservation of faience objects to know what has affected the condition of these shabti and how the cracks may have occurred, a diagnostic analysis has been carried out. Many different factors could have been of influence on the degradation of the objects, such as climate changes, mechanical shock, handling, past conservation treatments, off-gassing of materials from display cases, exhibition circumstances and burial conditions. As it was not possible to cover all of these aspects within the given time-span, the focus of this research has been the impact of manufacturing techniques on the stability of faience material and its susceptibility to damage.11 These techniques, which are distinguished in the literature by three different glazing methods, contribute to the development of distinctive microstructures. As faience is known to vary in composition and to be very vulnerable to damage, it is relevant to carry out a detailed study of the construction and material properties of the shabti.12 Cracks and other issues may have already developed during production and may have led to degradation. How have different vulnerable properties, inherent to the material by the type of production methods applied, contributed to the degradation phenomena observed? Only when these properties are better understood, one can start investigating the influence of other aspects related to degradation. This research, therefore, lays the foundation for further research that falls outside of the scope of this thesis.

11_{The exhibition space has been refurbished and it is not known what the display case was made of, or what} its micro-climate was. Also, there is no record of conservation treatments. As a large amount of data is missing, studying the cause of degradation is not suitable as an MA thesis topic.

12_{Especially the composition of earlier faience varied considerably. Kaczmarczyk and Hedges, Ancient}

(12)

9 De Regt, UvA, 2017 Figure 3 (left): In the Egyptian depot of the RMO, more damaged faience shabti were encountered (Image and color checking: De Regt, 2016).

Figure 4 (bottom left): These objects (F93/10.41 and F93/10.42) were degraded to such an extent, that they were no longer recognizable as shabti. The powdery material has a brown color, which is uncommon for faience. On the old evelopes is written, that the figurines have fallen apart (Image: De Regt, 2016).

Figure 5 (bottom right): The provenance of these faience fragments (RBK 14272) is unknown. The material is very powdery and soft (Image: De Regt, 2016).

(13)

The main question proposed here, is: ‘How could the production materials and characteristics of different production techniques have influenced the degradation of Egyptian faience shabti from the Dutch National Museum of Antiquities?’

Differences in manufacturing techniques and vulnerability to damage from salt crystallization are believed to have a high impact on the stability of faience.13 The hypothesis proposed here is that the shabti have suffered from damage by soluble salts and that the manner in which they were produced has contributed to their susceptibility to damage. Different tools and methods have been used to investigate this hypothesis, which have provided information on the possible presence of salts and the different production materials and techniques. These tools included: visual examination, microscopy, salt test strips and SEM-EDX analysis of a number of samples.14_{Expert opinions, relevant}

literature to the subject and information from museum records have been of help too in answering various sub-questions that contributed to providing an answer to the main research question.

A concise description of the condition and history of the thirteen shabti investigated, will be presented in chapter two. An overview of current knowledge on the subject is provided, as well as a detailed account of the research that was undertaken to answer the main research question. Although mainly diagnostic in nature and limited to the study of a small number of objects, this research has yielded results that can contribute to our broader knowledge and understanding of Egyptian faience objects. Conservators as well as museum curators can benefit from the increased knowledge on the relationship between production technique and degradation of Egyptian faience, when making decisions about storing, handling and treating faience objects, so they can be preserved for the future.

13_{According to conservator Sandra Smith, in order to conserve the material properly, it is important to} understand its production process. Smith, “The Manufacture,” 849.

14_{As three different glazing techniques have been utilized for the production of faience, each resulting in} different characteristics, attempts were made at identifying these methods through microscopic and analytical research. The presence or absence of salts could also be detected in this manner.

(14)

1. Object biographies

Faience shabti

Thirteen damaged shabti from the collection of the Dutch National Museum of Antiquities (RMO), presumed to be made of faience, are the subject of this research (see fig. 3-5). Some resemble each other in appearance.15 Portrayed are both men and women, often provided with a name in hieroglyphic writing. All except one (F1985/3.2) are broken or have cracks, while some have disintegrated (see fig. 4-5).16 Damage phenomena are diverse and include breaks, cracks, chipping of the glaze layer and discoloration (see appendix II). There are traces of previous treatments and restorations, such as adhesives, fillings, retouching material and a mounting pin (F93/10.34).17

Provenance

Six shabti are recorded as coming from the ‘priest cachet’ at Deir el-Bahri in Thebes (see appendix II).18 The mummies and funerary gifts stored in the cachet, date to the Third Intermediate Period (21st Dynasty) and belonged to priests of Amen and their families. Originally, they had been located elsewhere, but when tomb robbing became an increasing problem at the end of the New Kingdom, they were hidden in cachets.19 In 1891, the director of the antiquities service Eugène Grébaut was informed about the existence of a tomb from which objects were being sold by a local resident, and the site was excavated (see fig. 6). Two burial chambers were found, containing 110 boxes, which held thousands of shabti.20 Most were made of faience with a light blue glaze.21 The clearing of the cachet took eight days, but further context is lacking, as no records were made.22 The shabti that entered the collection of the RMO in 1893 were a gift from the Egyptian government.23

15_{It seems that there are four ‘pairs’ of matching shabti; AF127a - AF127b, F93/10.97 - F93/10.98, H**12 -} H**13 and F93/10.41 - F93.10.42.

16_{This concerns objects F93/10.41, F93.10.42 and RBK14272.}

17_{Objects F1985/3.2 and F93/10.34 do not show any traces of treatment, but the latter has a metal mounting} pin inserted into its body.

18_{As these six objects entered the collection of the RMO in 1893, they are all registered by a number starting} with ‘F93’.

19_{Janes, Shabtis, vi.} 20_{Schneider, Shabtis, 336.} 21_{Janes, Shabtis, xi}

22_{The only exception is a list of 58 shabti names, made by George Daressy. Janes, Shabtis, xi-xii.} 23_{Schneider, Shabtis, 14.}

(15)

12 De Regt, UvA, 2017 Figure 6: These drawings by Marianne Brocklehurst show the excavations at Deir el-Bahri in 1891.24

(16)

All information available on the history of the other shabti, which is rather scarce, can be found in appendix II.25

Storage

There is not much information about storage conditions of the six shabti from Deir el-Bahri. Willem Pleyte (museum director from 1891-1903) however describedthe arrival of the boxes with Egyptian antiquities in 1893 (see fig. 7). Some of the objects were heavily damaged by transportation and required immediate restoration, which could have included the shabti as well, although they are not specifically mentioned. Soon after their arrival, the objects were put on display.26

Even less is known about the storage of the other shabti. The museum’s entire Egyptian collection is known to have moved to the current depot in the 1980s.27 Although it may be expected that the shabti were moved around several times, they have not been on loan since 1992.28 When put on display, objects are carried from the depot to the museum across the street, in closed boxes. The museum itself is climate controlled and there is an air filtration system, but the climate in the depot is unregulated.29

Display

The museum aims at displaying all objects in display cases that are made of materials that are recommended for safe storage.30 Before the exhibition in 2013, object F93/10.29 had been stored along with other shabti in one of the drawers of a storage unit in the Egypt depot. Object F93/10.34 had been on display in the main collection since 2001. From April 20 until September 15 in 2013, the RMO exhibited in a large display case with other shabti (see fig. 8).31Contrary to what was initially assumed, the damage to these objects was not immediately noticed after the exhibition.

25_{There could be more information from before 1992 that is not digitalized in The Museum System, but} limited time did not permit elaborate research.

26_{Source: archival records (1890-1905) describing the arrival of the objects at the RMO: pages 5 and 43.} 27_{In the early 1980s, the Egyptian depot was located elsewhere, in a room in the museum building itself.} Source: personal communication with Prof. dr. Maarten Raven (16-06-2017).

28_{If objects were on loan before that date, is unknown, as this information is not digitalized.}

29_{The museum aims at maintaining a temperature of 19°C and a relative humidity of 50% in the museum} itself and 18°C/50RH in the Egyptian depot. Source: Personal communication with collection manager Marianne Stauthamer (July 27, 2017).

30_{Off-gassing of wooden display or storage cases or paint can cause issues with soluble salts. As salts in} porous objects deliquesce at a specific relative humidity and then recrystallize during drier periods, they can cause damage to faience objects.

31_{Objects were mounted using existing pins in the objects, or with separate aluminum clamps. Source:} Personal communication with collection manager Robert Ritter (April 16, 2017).

(17)

14 De Regt, UvA, 2017 Figure 7: In his letter about the arrival of the objects from deir el-Bahri, Willem Pleyte stated that: “Na opening der kisten bleek het dat sommige voorwerpen vrij sterk beschadigd waren.” (“After opening the boxes, some of the objects were found to be heavily damaged.” Translation: De Regt, 2017) Source: handwritten letter by Willem Pleyte (RMO archive of correspondence).

(18)

15 De Regt, UvA, 2017 Figure 8: These images show a 3D map of the exhibition of 2013. The display case that contained the shabti was put against the wall (a) and contained numerous similar objects (b). Photos: Courtesy of RMO.

(19)

16 De Regt, UvA, 2017 Figure 9: Image to the left shows a photograph taken in the 1970s of object 1970/1.1, while image to the right shows a current photo of the same object. Image to the left by Schneider32_{and to the right by} De Regt, 2016.

Figure 10: Image to the left shows a photograph taken in the 1970s of object AF127a, while image to the right shows a current photo of the same object. Image to the left by Schneider33 and to the right by De Regt, 2016.

The collection manager recalled they had not been put back into their storage drawers in the depot, due to lack of time. The objects did not enter the conservation studio until August 2014. If this account of events is correct, the two shabti had been lying in an open plastic basket for almost a year, before damage was noticed.34_{Unfortunately, no detailed}

images exist of the objects before display.

Conservation

Information on conservation practices prior to the 1980’s, is rather scarce. A few objects have been mentioned by former director at the RMO Hans Schneider in a publication on shabti in 1977. Object F93/10.29 is described as not only broken at the hips, but also showing firing cracks in the legs.35 Object 1970.1.1 is mentioned to be ‘broken and repaired’, but it is unclear when this happened. It is also reported to be ‘damaged’ and looks very similar to how it does today in an old photograph (see fig. 9).36_{A photo of}

object AF127a was taken as well (see fig. 10).

32_{From: Schneider, Shabtis, 96.} 33_Ibid.

34_{Source: telephone call between Dooijes and Ritter. (April 16, 2017).} 35_{Schneider, Shabtis, 124.}

(20)

2. The production of Egyptian faience shabti

Production history

As knowledge of the fabrication and composition of faience is vital to gain an understanding of the vulnerable aspects of the shabti, it is relevant to reflect upon the history of faience in Egypt. Faience production developed alongside the older practice of glazing steatite in pre-dynastic times (5500-3050 B.C.), but later grew dominant (see fig. 11 and 12.).37 It is unclear whether the material was first created in Egypt or in Mesopotamia.38 Gradually, production centers were established throughout different parts of the world, including Asia and northern Europe.39 Faience, called tjehnet by the Egyptians, referred to the word ‘dazzling’ or ‘shining’ and was valued for its shiny glazed surface.40 Its blue or green color is often mentioned to have been a more affordable substitute for the mineral turquoise and lapis lazuli.41_{As faience could be shaped in molds,}

it enabled large scale production.42_{The notion that faience was considered an inferior}

substitute for precious stones has been debated however, as the material was placed in tombs of all socio-economic strata, and has been found alongside objects of gold and other costly materials.43 As faience is argued to have been associated with the sun god Re and the sacred blue of the skies, the reason for its popularity may be related more to its symbolism and its material properties.44

The first faience objects were small items such as amulets and beads, but gradually larger objects were made. 45 Most seem to have been produced specifically for use in burial.46 Although no specific depictions were made of faience making, one scene from a tomb may be a reference (see fig. 13).

37_{Nicholson, Egyptian Faience and Glass, 18. For the definition of steatite, see glossary (appendix I).} 38_{Various sources claim that faience was first produced in the Near East. Lucas, Ancient Egyptian Materials,} 464-465; Matin and Matin, “Egyptian Faience 1,” 763.

39_{Foster, Aegean Faience.}

40_{Nicholson, Egyptian Faience and Glass, 11}

41_{The goddess Hathor was given the title “Mistress of Turquoise” as well as “Mistress of Faience”. Dunn} Friedman, “Faience,” 15.

42_{Dunn Friedman, “Faience,” 15.} 43_{Patch, “By Necessity,” 43.}

44_{Bianchi argues how the word tjehnet seemed closely related to “luminosity” and may have been associated} with the sun god Re (Bianchi, “Symbols and Meanings,” 24). Faience is also believed to have been associated with skies and eternity (Schlick-Nolte, “Ägyptische Fayence,” 21).

45_{Vandiver and Kingery, “Manufacture,” 79-90.} 46_{Patch, “By Necessity,” 32.}

(21)

Chronology

Pre/Protodynastic Period Before 3050 BC n/a n/a

Early Dynastic Period 3050 – 2613 BC 3050 – 2682 Dynasties I and II

2686 – 2613 Dynasty III

Old Kingdom 2613 – 2181 BC 2613 – 2498 Dynasty IV

2498 – 2345 Dynasty V 2345 – 2181 Dynasty VI

First Intermediate Period 2181 – 2040 BC 2181 – 2040 Dynasties VII-X

2134 – 2060 Dynasty XI (Theban)

Middle Kingdom 2040 – 1782 BC 2060 – 1991 Dynasty XI

1991 – 1782 Dynasty XIII

Second Intermediate Period 1782 – 1570 BC 1782 – 1650 Dynasties XIII & XIV (Egyptian)

1663 – 1555 Dynasties XV & XVI (Hyksos) 1663 – 1570 Dynasty XVII (Theban)

New Kingdom 1570 – 1070 BC 1570 – 1293 Dynasty XVIII

1293 – 1185 Dynasty XIX

Third Intermediate Period 1070 – 713 BC 1070 – 945 Dynasty XXI

945 – 712 Dynasty XXII

828 – 712 Dynasty XXIII

724 – 713 Dynasty XXIV

Late Period 713 – 332 BC 713 – 656 Dynasty XXV (Nubian)

664 – 525 Dynasty XXVI

525 – 404 Dynasty XXVI (Persian)

404 – 399 Dynasty XXIII

399 – 380 Dynasty XXIX

380 – 343 Dynasty XXX (Egyptian/Persian)

Graeco-Roman Period 332 BC – AD 395 332 – 30 Ptolemies

30 – 395 Roman Emperors

Figure 11: Overview of different time periods in ancient Egypt47

(22)

19 De Regt, UvA, 2017 Figure 12: This map gives a general impression of the most well-known locations in ancient Egypt.48

Figure 13: A scene from a tomb dating to the Twenty-Sixth Dynasty, shows what might be faience making. Since there is no text associated with it, it is not exactly clear what is represented though.49

48_{From: Experience Ancient Egypt, “Ancient Egyptian.”} 49_{From: Nicholson, “Materials and Technology,” 56.}

(23)

Faience shabti first appeared in the Middle Kingdom and continued to be made until the Ptolemaic period.50 The shabti with the hardest cores are claimed to be from the 18th and 19th Dynasty, although no clear argumentation for this notion is given in the literature.51

During the Third Intermediate Period (1070-712 B.C.), shabti by then had become the most common faience objects after beads and inlays.52 The u-shabti, which translates as ‘answerers’, were meant to execute the tasks the deceased would be asked to carry out in the afterlife.53 During this period, one burial could contain as many as 401 shabti.54 Those from the late New Kingdom and Third Intermediate Period often have a very friable body, with details emphasized by the use of thick layers of glaze and black painted designs.55 The shabti from Deir el-Bahri in particular, are known for their ‘remarkable glossy appearance’.56 Although in some areas in the Near East production never seems to have ceased, it ended in Egypt during the Later Periods.57

Raw materials

When considering the composition of faience shabti, a distinction can be made between the inner core material, or the “body” of the object, and its surrounding glaze layer. Up to 99% of the body comprises of silica (SiO2), which was easily available in the Egyptian

desert (see fig. 14).58

Experiments with the production of faience replicates by La Delfa et al has shown that the use of normal quartz sand delivers a surface that is too rough, while grinded quartz (< 500-100 µm) could be used successfully.59_{Sand also has the disadvantage that it can}

contain impurities such as calcium, chalk and iron, which are able to discolor the glaze. Crushed pebbles are relatively pure in silica though, producing a white body material. 60

50_{Schneider, Shabtis, 235.}

51_{Schneider based his observations on the collection of shabti in the RMO, but did not explain how many} objects he had studied and how he arrived at his conclusion. Ibid.

52_{Ibid., 236.}

53_{Patch, “By Necessity,” 32.}

54_{One for each day of the year, with 36 overseers for each work week. Schlick-Nolte, “Ägyptische Faience,”} 42.

55_{Schneider, Shabtis, 235.} 56_{Ibid., 236.}

57_{In Egypt, it was during the Later Periods (1070 BC- 395 AD) that the production of faience ended} (Davison, Conservation and Restoration, 83). Faience was still produced in Qom and Shushtar in Iran in the 1960s and this practice continues today (Matin and Matin, “Egyptian Faience 1,” 763-764.)

58_{According to Vandiver, the typical faience body comprises 92-99% SiO}

2, 1-5% CaO, 0.5-3% Na2O with minor quantities of CuO, Al2O3, TiO2, MgO and K2O. Nicholson, “Materials and Technology,” 50.

59_{La Delfa, Formisano, and Ciliberto, “Laboratory Production,” e114.} 60_Ibid.

(24)

Composition of the typical Egyptian faience body

92-99% SiO2 (silicon dioxide)

1-5% CaO (calcium oxide)

0,5-3% Na2O (sodium oxide)

Minor quantities CuO (cuprous oxide)

Al2O3 (aluminium oxide)

TiO2 (titanium oxide)

MgO (magnesium oxide) K2O (potassium oxide)

Figure 14: Table showing the composition of the typical faience body.61

The theory that crushed pebbles were used, was also suggested by Sir Flinders Petrie, after excavations carried out at the site of Amarna between 1891 and 1892.62

Lime comprises 1-5% of the faience body.63 Lime mainly consists of calcium containing inorganic materials in which calcium carbonate, oxide and hydroxide predominate.64_{Its source could have been limestone or chalk, when added intentionally, or}

it may just have been present as an impurity if sand was used.65 Lime is known to act as a stabilizer in glazes and glass.66

The third component of the faience body is claimed to take up between 0.3 and 5% of the total components and consists of alkali, primarily sodium oxide (Na2O).67 This

relatively small amount of alkali was added to the mixture as a flux, lowering the melting point of silica.68 One source of sodium is natron, which occurs naturally in Egypt and is a mixture of sodium carbonate and sodium bicarbonate with sodium chloride and sodium sulfate.69 Another effective source is the ash of halophytic plants.70 From Predynastic to Roman times, alkali were mainly obtained from plant ashes.71 Silica, lime and alkali were mixed with water, and possibly with an organic binder, to hold the particles together and

61_{Based on: Vandiver and Kingery, “Manufacture,” 79-90.}

62_{Petrie found quartz pebbles, which he assumed had been part of a furnace floor, where repeated heating} and cooling would have facilitated cracking them. Ibid.

63_Ibid.

64_{Matin and Matin found that calcium carbonate did not give good results for cementation glazed replicates,} as the glaze became too dense. Calcium hydroxide did produce a successful glaze. Matin and Matin, “Egyptian Faience 1,” 767.

65_{Nicholson, “Materials and Technology,” 50.}

66_{Newton and Davison, Conservation of Glass, 135-164.}

67_{The ratio of different components is based on data obtained from fired objects, or replicates. As no recipes} have survived and compounds must have been lost during firing, it is difficult to establish exact amounts. 68_{For the definition of flux, see glossary (appendix I). The melting point of silica is 1710°C. Lime and alkali} acted as a flux and lowered it considerably. Kaczmarczyk and Hedges, Ancient Egyptian Faience,7.

69_{The most likely primary sources seem to have been Wadi Natrun and al-Barnuj. Shortland et al., “Natron,”} 528. For the definition of natron and wadi, see glossary (appendix I).

70_{Particularly those of the salicornia family. Nicholson, “Materials and Technology,”50. For the definition} of halophytic, see glossary (appendix I).

(25)

form a paste.72 Considering the low lime percentage in the core material, just enough of this component seems to have been mixed to obtain a paste, but not enough to properly fuse the particles together during firing.

The glaze of faience objects consists of a soda-rich flux, which is generally assumed to

have been obtained from natron or plant ashes.73 Another component of the glaze comprises of copper containing minerals, which produce a green or blue color.74 During the New Kingdom, antimony, lead and cobalt containing oxides were used for the production of shabti as well, so different colors could be obtained.75 Other components of faience occur only in small amounts and were added either accidentally as impurities, or deliberately to color the glaze.76

Modelling

Because of the thixotropic nature of faience paste before firing, the material is rather complicated to work with.77 It breaks quite easily when handled and shaped.78 The addition of lime and soda helps to cement the quartz grains together and improve mechanical strength.79

Faience paste was initially shaped by hand, but from about 2040 B.C., open molds were introduced. 80 The faience paste was pressed into the mold and then removed before firing, so it could dry and be re-worked with tools to add details.81 Most of the molds discovered

72_{Binders suggested in the literature are: lime, gum Arabic (Nicholson, “Materials and Technology,” 51;} Noble, “The Technique”), gum of tragacanth (Wulff, Wulff, and Koch, “Egyptian Faïence”), serish (Matin and Matin, “Egyptian Faience 1,” 775), and the alkali themselves, which had a binding effect during firing (Lucas, Ancient Egyptian Materials, 175). They must have helped in preventing material from cracking due to low firing temperatures and low flux content.

73_{Tite, Manti, and Shortland, “A Technological Study,” 1576.} 74_{Smith, “The Manufacture,” 845.}

75_{Nicholson, Egyptian Faience and Glass, 30-31.} 76_{Nicholson, “Materials and Technology,” 50.}

77_{When the right amount of water is added, the paste can be (re)modeled. It is thick at first and then soft and} flowing as it begins to be deformed. Nicholson, “Materials and Technology,” 51.

78_{Source: personal experience during a faience making workshop at UvA, under supervision of conservator} Nienke Besijn.

79_{Nicholson, “Materials and Technology,” 50.} 80_{Davison, Conservation and Restoration, 83.} 81_{Nicholson, “Materials and Technology,” 51.}

(26)

23 De Regt, UvA, 2017 Figure 15: This image shows a open-faced ceramic mold that was used for producing faience shabti. It is now part of the collection of the Petrie Museum.82

Figure 16: This schematic image gives an impression of the faience microstructure. The development of interparticle glass, fuses the quartz grains together 83

82_{From: Tajeddin, “Egyptian Faience,” 126.} 83_{From: Tite and Bimson, “Faience”.}

(27)

are single-sided.84 Shabti, too, were produced by means of molds (see fig. 15).85 When the mold was used, they could still differ in details, as they were worked with a sharp tool afterwards.86 When a faience object was constructed of different pieces, they were joined by means of a quartz slurry.87 There is no indication that shabti were also produced from different pieces however, and their size suggests this was not a necessity.

Glazing methods

Faience has many properties in common with glass, as it shares the same major components, such as soda, lime and silica. Silica is known as the ‘network former’ and calcium as the ‘network stabilizer’, while sodium helps to reduce the melting temperature of the silica and serves as ‘network modifier’ (flux). In glass production, an amorphous material is formed. In faience however, the proportions are different, the firing temperature is lower and the firing time is shorter, which results in a crystalline material. Cohesion is provided by the development of some degree of interparticle glass, which partly fuses the quartz grains together (see fig. 16).88 The glaze layer also keeps the material in place.

The glazed surface of a faience object could be obtained in different ways, each influencing its composition, construction and long-term stability in a specific manner. Initially, it was assumed that the glaze had been applied as a layer to the faience body and that the objects were fired twice, as in the production of ceramics.89 _{When Sir Flinders}

Petrie found faience production molds at the archaeological site of Amarna, he assumed the paste that had remained inside came from the quartz body, instead of the residue of efflorescence.90 Alfred Lucas recognized in 1926 that the basic process of faience production involved fusing a quartz core with sufficient alkali and impurities, to produce a rigid structure with an expansion coefficient similar to that of crystalline quartz. This enabled the application of low-melting alkali glazes.91 He also observed that the body of faience ‘…is generally so friable that it can readily be rubbed away with the fingers, except in those instances in which the glaze has penetrated below the surface.’92 His classification of different types of faience was mainly based on visual aspects of faience however and

84_Ibid.

85_{Lucas, Ancient Egyptian Materials, 159.} 86_{Schneider, Shabtis, 236.}

87_{Smith, “The Manufacture,” 845.}

88_{Nicholson, “Materials and Technology,” 51.} 89_{Nicholson, Egyptian Faience and Glass, 13.} 90_{Nicholson, “Materials and Technology,” 60.}

91_{Kaczmarczyk and Hedges, Ancient Egyptian Faience, 6.} 92_{Lucas, “Problems,” 437.}

(28)

therefore did not provide much information on its material properties.93 Stone and Thomas divided faience into true faience and glassy faience, which is still used today.94 The efflorescence glazing method was first identified by Binns, Klem and Mott in 1932 and further elaborated upon by Noble in 1969.95 The cementation method was first identified in the 1960s by Wulff et al.96

The distinction between the “efflorescence”, “cementation” and “application” techniques, as still used in present day faience research, can be seen in fig. 17.97 This chronology by Nicholson indicates generally when a certain glazing method or other technique was used.98 As the shabti from the RMO are documented to have been produced between 1187 and 343 B.C, each different glazing and manufacturing technique could have been applied (see appendix II).99_{It is relevant to have a closer look at all three}

methods, as each may have a different effect on the vulnerability of faience objects.

Efflorescence

For this ‘self-glazing’ method, the components that form the faience body (quartz, lime and soluble alkali salts) were mixed with copper compounds and water to form a paste.100 As the faience paste was left to dry, the alkali salts migrated to the surface and formed an efflorescent outer layer. During firing, this layer melted and fused with the quartz, copper oxide and lime, resulting in a glazed surface (see fig. 18A).101 The efflorescent layer contains sodium and to a lesser degree potassium carbonates, chlorides and sulfates from the natron or plant ash.102 Production of replicates showed that a higher rate of evaporation of the water leads to the greatest glaze thickness, resulting in a variable glaze thickness that is highest in protruding areas.103 Carving, cutting or incising the object was done after drying and therefore objects can show reduced efflorescence in worked areas.104

Another characteristic of efflorescence glazing is the amount of interparticle glass,

93_{Nicholson, Egyptian Faience and Glass, 14.}

94_{This research concerns only ‘true faience’, as ‘glassy faience’ is a homogeneous material that resembles} glass. Peltenburg, “Early Faience,” 9.

95_{Noble, “The Technique” 439.}

96_{Wulff, Wulff, and Koch, “Egyptian Faïence,” 98-107.}

97_{Vandiver and Kingery, “Manufacture,” 79-90; Nicholson, “Materials and Technology,” 52.} 98_{Nicholson, Egyptian Faience and Glass, 11.}

99_{Their age is determined by archeological data and decorations in hieroglyphic writing.} 100_Ibid.

101_{Matin and Matin, “Egyptian Faience 1,” 764.} 102_{Vandiver, “Appendix A”, A31.}

103_{Good results were only obtained in a very hot climate, with fast evaporation. La Delfa, Formisano, and} Ciliberto, “Laboratory Production,” e115.

(29)

Period Body Manufacture Glazing Process Factory Evidence

Predynastic (5500 – 2920 BC)

Modeling a core for grinding Surface grinding

Free-form modeling (rare)

Application (?) Cementation (?) Efflorescence (?) None Early Dynastic (2920 – 2649 BC) Modeling Surface grinding Efflorescence None Old Kingdom (2649 – 2134 BC)

Painting with slurry Layering (rare) First Intermediate

(2134 – 2040 BC)

Core forming (rare) Marbleizing (rare) Molding (?) Abydos Middle Kingdom (2040 – 1640 BC) Modeling Molding on a form Efflorescence Cementation Lisht Kerma Second Intermediate (1640 – 1532 BC) Core forming Marbleizing Layering

Painting with colored quartz slurry

Incising Inlaying Resisting

Painting with pigment wash

Application

New Kingdom (1570 – 1070 BC)

Molding on a form

Pressing into openface molds Forming over a core

Joining of molded parts with quartz slurry

Layering Incising

Inlaying with quartz slurry Painting with pigment wash Throwing (?)

Efflorescence Application

Finely powdered glass, added to body or inlay to extend color range (?)

Amarna Lisht

Late Period

(1070 BC – AD 395)

All New Kingdom

techniques, plus throwing

Application Efflorescence

Memphis Naukratis

Figure 17: This chronology by Nicholson, which provides a rough overview of when certain faience techniques were used, is an adapted version of an earlier chronology made by Vandiver. 105_{Nicholson based} his chronology mostly on archaeological data. As can be observed here, the cementation method seems not to have been used until the Middle Kingdom (2040-1570 B.C.). The application method was not used again until the Second Intermediate Period (1640-1532 B.C.) until the Later Periods (1070 B.C.- 395 A.D.), when it was used alongside with the efflorescence method.

(30)

developed as a result of incomplete migration of the alkali salts, which is the greatest in this glazing method and which hardens the fired body.106 In cross-sections, the interface between glaze and body would be the clearest to recognize of all three methods, as the transition from core to glaze is gradual and increasing amounts of interparticle glass towards the glaze can be observed.107 The glaze itself is reported to be thin and fragmentary, but also clearly distinguishable as a separate layer of fused material, with no distinguishable quartz grains (see fig. 18A). Objects may show drying marks in areas where they were put on stands while drying, unless they were put on supports that prevented this.108

Cementation

The cementation technique is another ‘self-glazing’ method.109 It was first recorded in the village of Qom in Iran in the 1960s.110 Cementation glazing involves burying the object in glazing powder with a high flux content, within a vessel.111 This powder was composed of lime, ash, silica, copper compounds and possibly charcoal.112

During firing, the glazing powder starts to melt. After firing up to twelve hours, the object slowly cools down and alkaline and copper compounds migrate from the glaze powder into the silica body, reacting with it and forming a glaze and interaction layers (see fig. 18B).113_{After firing, the unreacted excess powder is brushed away (see fig. 19).}114

Cementation glazing produces a thin, evenly distributed glaze layer, with little interparticle glass in the core.115 The glazed surface on the side of objects that are facing downwards may be thicker, because of gravity.116 Drying marks should not be present on

106_Ibid.

107_{Davison, Conservation and Restoration, 84.}

108_{If a non-wetting support was used, confusion with the cementation glazing method is possible. Vandiver,} “Appendix A”, A33.

109_{Matin and Matin, “Egyptian Faience 1,” 764.} 110_{Wulff, Wulff, and Koch, “Egyptian Faïence,” 99.}

111_{La Delfa, Formisano, and Ciliberto, “Laboratory Production,” e114.}

112_{Charcoal is used for modern faience production in Iran. (Wulff, Wulff, and Koch, “Egyptian Faïence”).} Matin and Matin however found no convincing arguments for its use in the past (Matin and Matin, “Egyptian Faience 1,” 768).

113_{Tite, Manti, and Shortland, “A Technological Study,” 1576.} 114_{Matin and Matin, “Egyptian Faience 1,” 764.}

115_{Nicholson, Egyptian Faience and Glass, 13; Rehren, “A review,” 1347.} 116_{Nicholson, “Materials and Technology,” 53.}

(31)

28 De Regt, UvA, 2017 Figure 18A: This image shows the process of efflorescence glazing and a schematic depiction of what a

cross section of a faience object would look like in this case. Figure 18B: According to Smith, the cementation method produces a thin and clearly defined transition layer between the body and the glaze. Figure 18C: This image shows the process of application glazing and a schematic depiction of what a cross section of a faience object would look like in this case. 117

Figure 19: In the process of cementation glazing, unreacted excess powder is brushed away from the glazed object. This figure shows a cementation glazed replicate.118

117_{All three images from: Smith, “The Manufacture,” 846.} 118_{From: Matin and Matin, “Egyptian Faience 1,” 764.}

(32)

small objects, as they were entirely surrounded by powder.119 Larger objects may show firing marks however.120 According to Smith, the cementation method produces a thin and distinguishable interaction layer (see fig. 18B).

Application

A third way to glaze faience was to apply a glaze to the unfired quartz core which had been produced by melting during firing, grinding and then refiring.121 The raw materials may have been crushed, or fritted together.122 By grinding, re-melting and regrinding the raw materials multiple times, a homogeneous glaze is produced.123 The porosity of the quartz body allowed the mixture to be absorbed so that it adhered to it and formed a coating upon drying (see fig. 18C).124_{Dipping, immersing or brushing of glaze may have been practiced,}

although Vandiver only found traces of brushing on polychrome faience of the New Kingdom (see fig. 20).125

Objects glazed by the application method show a distinctive, thick glaze layer, which often shows drips and flow lines.126 Considerable variations in the glaze thickness may be observed.127 Usually, very little interparticle glass is present in the core and the amount depends on the porosity of the quartz body and how much water has been added to the slurry. The interaction layer is not well defined.128_{Sometimes the objects have unglazed}

areas, to enable the object to be handled or prevent sticking to kiln supports.129

Even though three different glazing methods can be distinguished, one may find a combination of techniques in one single object. Vandiver and Kingery for instance, investigated a faience chalice produced by efflorescence glazing, that was finished with application glazing.130 Adding an extra layer by the application method may have compensated for cut out areas with very thin glazing.131

119_{Nicholson, Egyptian Faience and Glass, 13} 120_{Vandiver, “Appendix A”,}_A39_.

121_{Liang et al, “Optical Coherence,” 3683.} 122_{Nicholson, Egyptian Faience and Glass, 14}

123_{Tite, Freestone, and Bimson, “Egyptian Faience,” 17-27.} 124_{Nicholson, Egyptian Faience and Glass, 14}

125_{Nicholson, “Materials and Technology,” 54.} 126_{Nicholson, Egyptian Faience and Glass, 15}

127_{Tite, Shortland, and Angelini, Production Technology, 48-49; 202.} 128_{Davison, Conservation and Restoration, 85.}

129_{Nicholson, “Materials and Technology,” 54.} 130_Ibid.

(33)

30 De Regt, UvA, 2017 Figure 20: This image shows different methods that could have been used to apply a glaze layer.132

According to Vandiver, most shabti from the Late Period have the visual characteristics of an applied glaze, with splotchy, uneven surfaces with more glaze accumulated in cavities.133

Color and decoration

Turquoise colored glazes for faience, were obtained by adding metal parings or minerals containing copper, such as malachite.134 Many copper colored faience objects contain a small amount of tin oxide, probably resulting from the use of oxidized bronze as a colorant.135

Around 1500 B.C., a wide range of colors was produced with cobalt oxide (violet or purple), manganese (black or purple), light green (lead) and antimonite (yellow).136 In later periods, polychrome glazes were produced.137 The use of antimony and cobalt oxides shortly declined during the Intermediate Period, while lead and tin remained common.138

During the Twenty-Fifth and Twenty-Sixth Dynasties, antimony and lead were reintroduced.

132_{From: Nicholson, “Materials and Technology,” 53.} 133_{Vandiver, “Appendix A”, A-127.}

134_{Smith, “The Manufacture,” 845.}

135_{Bronzes contain between 5 and 10% tin. Tite, Manti, and Shortland, “A Technological Study,” 1580.} 136_{Nicholson, Egyptian Faience and Glass, 30-31; Tite, Manti, and Shortland, “A Technological} Study,”1568.

137_{Kaczmarczyk and Hedges, Ancient Egyptian Faience,}_140-184. 138_{Nicholson, “Materials and Technology,” 62.}

(34)

After ca. 1780 B.C., painted details such as hieroglyphic writing and facial features were applied to the shabti before firing, by application with manganese, or manganese and iron.139 After the Third Intermediate Period, incising became more popular.140 Faience became almost glass-like and a matte surface was prefered.141

Firing methods

Faience and glassy materials may have been produced in the same workshops, once glass was developed.142 Both techniques require similar raw materials.143 Faience is likely to have been fired betweeen 800 to 1000°C.144_{The higher temperatures can only be used for}

cementation glazing though, because of the tendency of the glaze to start flowing at 1000°C. As the cementation powder is rich in lime content, it remains solid at higher temperatures.145 It is unclear what material was used as fuel, but wood and cattle dung have been suggested.146

Few kilns directly associated with faience production have been found.147 Several production sites have been located however.148 Petrie excavated a faience workshop in Memphis, which contained misfired vessels, some with kiln supports still attached.149

Faience vessels were fired in saggars, to keep gasses out of the vessel when heated (see fig. 21). Petrie also found small ceramic cones and marl clay balls, possibly used as supports. These were also found at Lisht, Abydos and Amarna (see fig. 22).150 At Abydos, several circular features were excavated that are believed to have been kilns.151 They have no superstructure and remind of open-firing kilns that are known from ethnographic studies of

139_{Davison, Conservation and Restoration, 83, Smith, “The Manufacture,” 847; Nicholson, “Materials and} Technology,” 54.

140_{Vandiver, “Appendix A”, A-126.}

141_{Nicholson, Egyptian Faience and Glass, 39}

142_{The title given to faience makers, ìmy-r irw hsbd, contains the word hsbd, which also refers to lapis lazuli,} faience and glass. Nicholson, “Materials and Technology,” 55.

143_{Vandiver and Kingery, “Manufacture,” 79-90.}

144_{Although exact temperatures are unknown, this range was successful in producing replicates. Nicholson} and Peltenburg, “Egyptian Faience,” 191.

145_{Nicholson, “Materials and Technology,” 55.} 146_{Ibid.; Matin and Matin, “Egyptian Faience 2,” 133.}

147_{Vandiver, “Appendix A”, A-30; Nicholson and Peltenburg, “Egyptian Faience,” 192.}

148_{Production sites include: Abydos, Kerma, Lischt, Malqata, Tell Amarna, Gurob, Qantir, Tell} el-Jahudija and Naukratis. Schlick-Nolte, “Ägyptische Faience,” 14-16.

149_{Nicholson, “Materials and Technology,” 62.} 150_{Ibid., 56.}

(35)

32 De Regt, UvA, 2017 Figure 21: This image shows a faience bowl with a kiln support that has adhered to its glazed surface. Faience vessels were fired in saggars.152

Figure 22: These images show different materials used in the manufacture of faience beads during the Middle Kingdom, including beads and marl clay balls. The latter may have served to support saggars in the kiln, or to prevent objects from adhering to the kiln.153

152_{From: Dunn Friedman, “Faience,” 170.} 153_{Ibid., 172.}

(36)

ceramic production. The many fragments of jars that have been found here, possibly indicate faience objects were placed in jars to protect them from ash and fuel. In 2013, a faience kiln from the early Roman Period was excavated at Kom Helul, Memphis.154

Thousands of open face molds have been found at various sites, such as Malkata in Thebes (see fig. 23).155 As no ‘master’ archetypes for the construction of molds have been found, these were probably made of a perishable material such as wax.156 Many workshops seem to have been associated with temples and the production and design of shabti seems to have been carefully controlled.157 The distribution of molds at Amarna suggests a large-scale industry.158

Smaller scale workshops have also been found that indicate cooperation with glass, metal and ceramics industries.159_{Studies from the 1960s by Wulff et al, described modern}

small scale production of cementation glazed faience in Qom, Iran, where each employee of the workshop had its own task (see fig. 24).160 More recent ethno-archaeological fieldwork by Zahed Tajeddin confirmed that faience is still being produced there, largely following the same production style.161 Experimental research by Mark Eccleston has demonstrated that faience can be produced in simple ‘bread ovens’ and that there are strong indications that faience was also made in regular homes, possibly as an extra source of income.162

154_{Nicholson, Working in Memphis.}

155_{Excavations by the Metropolitan Museum of Art in 1910-1921 at Malkata, Thebes. Nicholson, “Materials} and Technology,” 60.

156_{Quirke and Tajeddin, “Mechanical Reproduction,” 356.} 157_{Bianchi, “Symbols and Meanings,” 28.}

158_{Quirke and Tajeddin, “Mechanical Reproduction,” 351.} 159_{Ibid., 358.}

160_{Wulff, Wulff, and Koch, “Egyptian Faïence,” 101.} 161_{Tajeddin, “Egyptian Faience,” 112.}

(37)

34 De Regt, UvA, 2017 Figure 23: These molds were found to contain unfired faience paste.163

Figure 24: These images show a faience workshop in Qom, Iran, where beads are made. An employee is seen crushing quartz pebbles with a sledge hammer. Afterwards, they were to be grinded with a hand mill to a fine powder. Others are seen packing beads in glazing powder, which are put in large earthenware pots. The final image shows the downdraft cylindrical kiln, into which about fifty pots could be stacked. While in past times wood was used as fuel, this workshop used black oil in the 1960s, a residue from the oil refineries. The objects were fired near 1000°C for twelve hours, with a cooling time of twelve hours as well.164

163_{From: Dunn Friedman, “Faience,” 172.}

(38)

3. Current scientific knowledge

After pioneering research in the late nineteenth and early twentieth century into the composition of faience and initial attempts at distinguishing different glazing methods, most research in the second half of the twentieth century was influenced by the application of new analytical techniques. Many archaeological and chemical studies were carried out and the first technical typology of faience was constructed by Kacmarczyk and Hedges in the 1980s, adapting earlier classifications by Lucas and Harris.165 The chemical composition of more than a thousand faience samples were investigated by means of X-Ray Fluorescence.166 The publication included a large technological appendix by Vandiver, who tried to determine the three different glazing methods by means of optical microscopy and visual examination.

SEM-EDX analysis has been used by many scholars in studying faience, as it provides information on chemical composition as well as the morphology of a material.167 On the subject of archaeological reports related to production sites of faience, important work has been published by Shortland and by Nicholson.168 Although quite a few scholars have carried out experiments with faience replicates, not much has been written on the shaping of faience objects, or on conservation issues and treatment.169

Identification of the glazing method

As it is difficult to discern with the naked eye which glazing method has been used, technical analysis has been of great importance to the study of faience objects. Tite et al. and Tite and Bimson suggested microstructural criteria for distinguishing these glazing methods and stressed that macroscopic evidence should be supplemented by microstructural and analytical evidence obtained from SEM-EDX.170 When comparing information from these different research methods, an attempt can be made to distinguish different aspects that are typical for a specific production technique (see appendix III). It is possible to distinguish three layers in faience: the first layer (the glaze layer where no quartz particles can be distinguished, or GLZ), the second interaction layer (the glaze-body

165_{Rehren, “A review,” 1347.}

166_{Kaczmarczyk and Hedges, Ancient Egyptian Faience.}

167_{Tite, Freestone, and Bimson, “Egyptian Faience”; Tite and Bimson, “Faience”; Nicholson, Egyptian}

Faience and Glass; Nicholson and Peltenburg, “Egyptian Faience”; Rehren, “A review”; Matin and Matin,

“Egyptian Faience Part 1”.

168_{Shortland, Vitreous Materials; Nicholson, Brilliant Things.}

169_{Several authors made faience replicates. Noble, “The Technique”; Tite, Freestone, and Bimson, “Egyptian} Faience”; Matin and Matin, “Egyptian Faience Part 1”.

Conservation issues related to Egyptian faience: A closer look at damaged shabti from the Dutch National Museum of Antiquities