Ancient repairs in archaeological research:
a Near Eastern perspective
Renske Dooijes and Olivier Peter Nieuwenhuyse
ABSTRACT Archaeological research has long neglected the phenomenon of ancient repairs. Only recently have archaeologists
and restorers begun to study past practices of repair and reuse in closer detail, recognising the need for careful registration
of ancient repairs in artefact descriptions. A restoration represents an important element in the object’s cultural biography,
the study of which may highlight shifts in function, value and symbolic meaning. This paper discusses case studies selected
from three successive periods in the archaeology of the ancient Near East: the Neolithic, the Late Uruk and the Late Bronze
Age. We argue that the study of ancient repairs may contribute to insights into social organisation and symbolic meanings
in the past.
KEYWORDS archaeology, ancient repairs, restoration, cultural biography, bitumen, pottery
Introduction
Archaeologists and restorers alike often come across ancient
repairs. Various clamps and patches, perforations, alien pieces
of pottery inserted into gaps and traces of glues show how
people in the past prolonged the use life of their vessels. Yet,
in spite of their frequency, ancient repairs have generally been
neglected in archaeological research, both from a descriptive
and from a theoretical perspective. Only recently have archaeologists
and restorers begun to study past practices of repair
and reuse in closer detail (Bentz and Kästner 2007; Chapman
and Gaydarska 2007). We increasingly acknowledge that restorations
represent an important element in the object’s cultural
biography (Kopytoff 1986). We have begun to explore the
technological aspects of ancient repairs more systematically,
and the context in which they occur. The study of ancient
repairs may help us to understand changes in function, value
and symbolic meaning of archaeological objects (Dooijes and
Nieuwenhuyse 2007).
In this paper, we discuss some case studies from three successive
periods in the archaeology of the ancient Near East:
the Neolithic, the Uruk and the Late Bronze Age.
The Neolithic
In the Neolithic (8000–5000 cal. BC) people used various
organic materials to repair broken vessels. One of the main
techniques involved drilling holes along the breakages, probably
using a flint or obsidian tool. The sherds were then tied
together by stringing leather, rope or another organic material
through the holes. Alternatively, bitumen was used as a glue.
Bitumen occurs naturally at various locations across the Near
East (Connan and Deschesne 1996) and studies show that this
material was exchanged over very large distances throughout
the Neolithic (Connan et al. 2004). Often these two techniques
were combined. In many instances these repairs may have
been waterproof, especially when the vessel was made of a
compact material and the breaks were fresh. The earliest examples,
dating from before the invention of pottery, have been
found on stone vessels. After the introduction of ceramics
(c.7000 cal. BC), the same techniques were applied to pottery
vessels as well. Interestingly, Neolithic repairs do not seem to
have been applied randomly to just any type of broken vessel.
Instead repairs were associated with high quality serving vessels
(Nieuwenhuyse 2007).
Late Neolithic Tell Sabi Abyad, situated in northern Syria,
was a four-hectare site inhabited between c.7000 and 5500 cal.
BC (Akkermans 1993; Akkermans et al. 2006). Excavations by
theNationalMuseumofAntiquities,Leiden(TheNetherlands),
have yielded numerous examples of ancient repairs using this
technique. Pottery and stone vessels are thought to have been
locally produced at the site, but some of the best quality items
were probably imported from elsewhere. A stone vessel from
this site shows the two main techniques combined (Figure 1).
This object was found in situ, with the sherds still in position.
No strings have been preserved; presumably these were made
of organic materials. A thin layer of bitumen covers the breaks,
but does not extend over the surface. Two Hassuna/Samarra

ANCIENT REPAIRS IN ARCHAEOLOGICAL RESEARCH: A NEAR EASTERN PERSPEC TIVE
pottery vessels continued. A famous example is the Warka
vase, a ritual vessel showing scenes in relief decoration of food
offering processions to the city deity. The vase was broken in
antiquity and repaired with copper wire or clamps set in perforations
along the breaks. A fragment belonging to another
vase was inserted to replace a missing part at the rim.
In addition, however, a more practical, functional approach
to pottery repairs emerged during the Uruk period. The
National Museum of Antiquities, Leiden owns a large collection
of Uruk pottery from Jebel Aruda, a ritual site containing
several large, high status buildings adjacent to two large temples.
The settlement was situated on a high spur overlooking
the central Syrian Euphrates Valley, looking down onto the
contemporaneous urban settlement of Habuba Kabira (van
pottery vessels show the same two main techniques (Figures
2 and 3). Here, the strings were not preserved either. A thin
layer of bitumen covers the breaks and some part of the surface
along the breaks. Painted Halaf vessels also occasionally
show the same technique (Figure 4). Here bitumen was used
as a glue, covering the breaks and part of the surface. In some
cases perforations along the breaks can be observed.
The Uruk period
With the rise of city states in the Uruk period (c.3200–3000
cal. BC), the practice of repairing high quality stone and
Figure 1 A Late Neolithic stone vessel from Tell Sabi Abyad,northern
Syria.EarlyPotteryNeolithic,c.6500BC.Reddishstone,diameterofvessel
c.18 cm.(© National Museum of Antiquities,Leiden.)
Figure 2 A Hassuna/Samarra pottery vessel from Tell Sabi Abyad,
northernSyria.Transitional(Proto-Halaf)period,c.5900BC.StandardFine
Ware,diameter 14 cm.(© National Museum of Antiquities,Leiden.)
Figure 3 A Hassuna/Samarra pottery vessel from Tell Sabi Abyad,
northernSyria.Transitional(Proto-Halaf)period,c.5900BC.StandardFine
Ware,diameter 15 cm.(© National Museum of Antiquities,Leiden.)
Figure 4 Painted Late Halaf vessel repaired in antiquity with bitumen,
northern Syria,c.5500 BC.(© National Museum of Antiquities,Leiden.)
10
RENSKE DOOIJES AND OLIVIER PE TER NIEUWENHUYSE
Driel and Murray 1979, 1983; van Driel 2002). Habuba Kabira
and Jebel Aruda were key points in the extensive exchange
network that connected societies from Anatolia to southern
Mesopotamia (Algaze 1993). During a recent restoration
project on several Late Uruk vessels, a number of Uruk period
restorations using bitumen as a glue were discovered (Dooijes
et al. in press). Bitumen was traded in bulk during the Uruk
period (Schwartz et al. 2000). A preliminary inspection of the
spatial context of the bitumen-treated ceramic vessels at Jebel
Aruda suggests an even distribution across the excavated settlement.
Quite probably the occasional pottery repair could
easily be executed locally, whenever the need arose, by melting
the bituminous material and pouring or brushing it in place.
One large pottery jar from the Jebel Aruda collection in
Leiden shows a significant amount of bitumen covering the
exterior rim and large parts of the shoulders. The bitumen
has also been applied in broad, vertical bands on the body
(Figure 5). These vertical bands may have been applied to the
surface with brushes. They may mark the locations where supportive
or protective netting was added after the vessel had
been repaired, the bitumen serving to keep it in place or to
protect the surface of the pot from the abrasive effects of the
rope. The break edges of a large loose sherd, just below the
rim, had been covered with bitumen. Clearly this is the place
where the vessel had already been broken in the past, possibly
during firing. A second example (Figure 6), is a mediumsized
jar with red slipped and engraved decoration, which has
a massive splash of bitumen on its interior, covering a number
of breaks and seeping through these breaks onto the exterior
surfaces (Dooijes et al. in press).
The Late Bronze Age
In the Late Bronze Age people restored pottery vessels using
techniques that were essentially the same as those used in
prehistoric periods. However, a broader, more diverse range
of techniques can be observed. A Middle Assyrian fortified
rural settlement dated to c.1200 BC was excavated at Tell Sabi
Abyad, northern Syria. The settlement, which consisted of a
series of administrative buildings surrounded by a wall and a
ditch, contained a number of pottery workshops (Akkermans
et al. 1993; Akkermans and Wiggermann 1999; Akkermans
and Duistermaat 2001). The Middle Assyrian ceramic output
was very similar across the empire, and consisted of massproduced,
wheel-made plain vessels. Damage sometimes
occurred during production, and afterwards vessels often
broke during use. Ways of repairing broken vessels included
drilling perforations along the cracks, using bitumen as a glue
or as a filling between the cracks and filling the cracks or damaged
parts with a gypsum/lime paste. Occasionally the potters
repaired damage before the vessel was fired, by closing the
cracks with clay (Duistermaat 2008).
Repairs, the biography of objects and archaeology
Ancient repairs are not just distortions of the authentic appearance
of an object – they contain valuable information concerning
the cultural biography of an artefact. Pottery vessels and
other objects can be said to have ‘social lives’ (Kopytoff 1986;
Chapman and Gaydarska 2007). They are not just ‘tools’ with
a particular function, but they are also invested with social
and symbolic meaning.Importantly, their role in society was
often dynamic: it changed as the object moved from one use
Figure 5 A large jar from the excavations at Jebel Aruda (JA 1642).
Late Uruk period,c.3200–3000 BC.Height 90 cm.(© National Museum
of Antiquities,Leiden.)
Figure 6 A medium-sized jar from the excavations at Jebel Aruda
(JA 1801). Late Uruk period, c.3200–3000 BC. (© National Museum
of Antiquities,Leiden.)
11
ANCIENT REPAIRS IN ARCHAEOLOGICAL RESEARCH: A NEAR EASTERN PERSPEC TIVE
history. The same considerations that apply to recent restoration
practices also apply to the more ancient ones: they provide
insight into aesthetic choices and changing cultural values.
Not so long ago, restorers habitually chose to remove ancient
repairs or to render them invisible. By doing so they made
a cultural choice, ironically by removing part of the cultural
biography of the object. Today ancient repairs are increasingly
left untouched and shown to the public (Bentz and Kästner
2007).
The potential of ancient repairs for reconstructing the cultural
biography of objects has been neglected by archaeologists
and museums. We are poorly informed about the frequency
of repairs in archaeological contexts, or in museum collections.
Pottery restorations should no longer be treated as a less
important by-product of archaeological find processing and
publication. Interdisciplinary cooperation between museum
staff, restorers and archaeological scholars is essential.
Acknowledgements
The authors would like to thank the National Museum of Antiquities,
Leiden, Professor Peter Akkermans, director of the excavations
at Tell Sabi Abyad, Dr Kim Duistermaat, Erica Dooijes-Hoogwijk
and Ans Bulles.
References
Akkermans, P.M.M.G. 1993. Villages in the Steppe: Later Neolithic
Settlement and Subsistence in the Balikh Valley, Northern Syria.
Ann Arbor, MI: International Monographs in Prehistory 5.
Akkermans, P.M.M.G. and Duistermaat, K. 2001. ‘A Middle Assyrian
pottery kiln at Tell Sabi Abyad, Syria’. In Beiträge zur Vorderasiatischen
Archäologie, Winfried Orthmann gewidmet, Meyer,
J.-W., Novák, M. and Pruss, A. (eds), 12–19. Frankfurt am Main:
Archäologisches Institut.
Akkermans, P.M.M.G. and Wiggermann, F. 1999. ‘La forteresse de
Tell Sabi Abyad: Sentinelle de l’empire assyrien’. Archéologia 358:
56–65.
Akkermans, P.M.M.G., Limpens, J. and Spoor, R.H. 1993. ‘On the
frontier of Assyria: excavations at Tell Sabi Abyad, 1991’. Akkadica
84/85: 1–52.
Akkermans, P.M.M.G., Cappers, R., Cavallo, C., Nieuwenhuyse,
O.P., Nilhamn, B. and Otte, I. 2006. ‘Investigating the Early Pottery
Neolithic of northern Syria: new evidence from Tell Sabi Abyad’.
American Journal of Archaeology 110(1): 123–56.
Algaze, G. 1993. The Uruk World System. Chicago: Chicago University
Press.
Bentz, M. and Kästner, U. (eds). 2007. Konservieren Oder Restaurieren,
die Restaurierung Griechischer Vasen von der Antike bis
Heute (3rd suppl. to the CVA Germany). Munich: Beck.
Chapman, J. and Gaydarska, B. 2007. Parts and Wholes: Fragmentation
in Prehistoric Context. Oxford: Oxbow.
Connan, J. and Deschesne, O. 1996. Le Bitume à Suse: Collection du
Musée du Louvre. Paris: Réunion des Musées Nationaux and Elf
Aquitaine Production.
Connan, J., Nieuwenhuyse O.P., van As, A. and Jacobs, L. 2004.
‘Bitumen in early ceramic art: bitumen-painted ceramics from
Late Neolithic Tell Sabi Abyad (Syria)’. Archaeometry 46(1):
115–24.
Dooijes, R. and Nieuwenhuyse, O.P. 2007. ‘Ancient repairs: techniques
and social meaning’. InKonservieren Oder Restaurieren, die Restaucontext
to another in the course of its life. Each time people
established new relationships with the object, it gained a new
meaning.
Archaeologists can often reconstruct different stages within
an object’s cultural biography from the distribution, consumption,
circulation and exchange to the discarding of the object
once damaged. However, an object’s biography did not always
end when it broke. Often it was repaired or curated, and given
a new lease of life. Archaeological excavation itself initiates
a whole new lifecycle of the broken and discarded artefacts.
Repairs, both modern and ancient, ‘mark’ an object. They offer
valuable information about changes in context and use of the
artefact. They deserve careful study alongside other physical
transformations such as traces left by the production process,
erosion, use traces, graffiti, breakage, and so on.
In the Near East, repairs during the Late Neolithic were
associated with high quality serving vessels. The repairs were
highly visible however it should be borne in mind that technological
constraints prevented the restorations from being
less conspicuous. It is also possible that people simply did not
bother to cover up the repairs. Perhaps the visible presence of a
repair even contributed to the prestige value of the object. We
know that some of these vessels were exchanged, sometimes
across considerable distances, from village to village. It may
have signalled that the object was old and valuable enough to
merit restoration (Nieuwenhuyse 2007).
Although high status objects were repaired with similar
methods in the Late Uruk period and in the Neolithic, cost and
practical considerations seem to have become more relevant.
At Jebel Aruda, the uniquely isolated location of the settlement
apparently made it worthwhile occasionally to restore a large,
heavy storage vessel, rather than replacing it with something
new. Pottery was not produced locally at the site: each vessel
had to be carried up the mountain from the city located far
below on the river bank. Given the visible character of the
coarse repairs, it may be argued that aesthetic considerations
were of minor concern, and that the repairs were purely functional.
Interestingly, the technique of drilling holes along break
edges and using clamps has thus far not been attested at this
site. In the case of the large, heavy storage jars at Jebel Aruda,
using metal would have been more efficient than bitumen glue.
Considerations of cost in this case perhaps overruled those of
efficiency (Dooijes et al. in press). The application of additional
supporting netting around the vessel was perhaps a practical
alternative to using costly metal clamps.
In the Late Bronze Age restoration of mass-produced plain
ware pottery appears to have been an essentially pragmatic
affair. The restorations were carried out as part of the pottery
production process. It appears that in some cases these pottery
restorations were executed by specialists (ie the potters)
in workshops. This is in contrast to the repairs dated to the
prehistoric and Uruk periods. Generally, these repairs seem
to have been made locally, within the private domestic surroundings
of an individual household.
Conclusions
Past stages in an object’s cultural biography are just as important
as the biography we can reconstruct for its more recent
12
RENSKE DOOIJES AND OLIVIER PE TER NIEUWENHUYSE
rierung Griechischer Vasen von der Antike bis Heute (3rd suppl.
to the CVA Germany), Bentz, M. and Kästner, U. (eds), 15–20.
Munich: Beck.
Dooijes, R., Burghout, F., Düring, M.H. and Nieuwenhuyse, O.P. In
press. ‘Restorations on the Late Uruk pottery of Jebel Aruda – old
and new’. Leiden Journal of Pottery Studies 23.
Duistermaat, K. 2008. The Pots and Potters of Assyria: Technology
and Organization of Production, Ceramic Sequence and Vessel
Function at Late Bronze Age Tell Sabi Abyad, Syria. Turnhout:
Brepols, Publications on Archaeology of the Leiden Museum of
Antiquities.
Kopytoff, I. 1986. ‘The cultural biography of things: commodization
as process’. In The Social Life of Things. Commodities in Cultural
Perspective, Appadurai, A. (ed.), 64–94. Cambridge: Cambridge
University Press.
Nieuwenhuyse, O.P. 2007. Plain and Painted Pottery: The Rise of
Late Neolithic Ceramic Styles on the Syrian Plains. Turnhout:
Brepols, Publications on Archaeology of the Leiden Museum of
Antiquities.
Schwartz, M., Hollander, D. and Stein, G. 2000. ‘Reconstructing
Mesopotamian exchange networks in the 4th millennium BC:
geochemical and archaeological analyses of bitumen artifacts from
Hacinebi Tepe, Turkey’. Paléorient 25(1): 67–82.
van Driel, G. 2002. ‘Jebel Aruda: variations on a Late Uruk domestic
theme’. In Artefacts of Complexity: Tracking the Uruk in the Near
East, British School of Archaeology in Iraq reports 5, Postgate,
J.N. (ed.), 191–205. Warminster: Aris and Phillips Ltd.
van Driel, G. and van Driel-Murray, C. 1979. ‘Jebel Aruda 1977–1978’.
Akkadica 12: 2–31.
van Driel, G. and van Driel-Murray, C. 1983. ‘Jebel Aruda, the 1982
season of excavation, interim report’. Akkadica 33: 1–29.
Authors’addresses
Renske Dooijes, Conservation Department, National Museum of
Antiquities Leiden, PO Box 11114, 2301 EC Leiden, The Netherlands
(r.dooijes@rmo.nl).
Olivier Peter Nieuwenhuyse, Faculty of Archaeology, PO Box 9515,
2300 RA Leiden, The Netherlands (onieuw@xs4all.nl).