Pottery styles context synchronisms and the dating of the Attic Early Iron Age

July 2nd, 2017

The following information comes from the authors PhD thesis on Attic Geometric and Orientalising pottery. This is an isolated article that discusses how pottery styles and context synchronisms were initially used for the dating of the Attic Early Iron Age. This article discusses dates that have been established from the beginning of the the 20th century with the use of 'traditional' methods. Scientific techniques in the dating of the Greek (and more specifically Attic) Early Iron Age, such as radiocarbon and dendrochronology, will be discussed in another article.

Pottery styles and context synchronisms in dating the Attic Early Iron Age
Chronologies for the Greek Early Iron Age have been traditionally established with three methods: stratigraphy or sequencing, stylistic analysis or attribution, and context comparisons or synchronisms (Cook & Dupont 1998, 8-9; Whitley 2001, 63). For Athens, the best know sequences come from Kerameikos, as the excavations conducted by Karl Kübler and the German Archaeological Institute in 1926 were to “set new standards in the stratigraphical recording of finds and deposits” (Whitley 2001, 35; also see Knigge 1991, 166-7).

Before the excavations at Kerameikos, early scholars followed a rough chronological system suggested by Schweitzer (1917; 1918), in which the 11th and 10th centuries BC were characterised by the Protogeometric style, and the 9th and 8th centuries BC by the Geometric style. The Protoattic style diversified from the Geometric towards the end of the 8th century BC (Böhlau 1887). This rough dating system was primarily stylistic.

The chronologies for the Protoattic period became clear from the beginning of the 20th century and the Protoattic style never attracted different arguments in relation to its duration. In 1935, J.M. Cook studied the evolution of the style and defined the chronological span among different groups of painters based on attribution techniques. His comparisons allowed the construction of a relative dating sequence for the Protoattic period, divided in three phases (Cook 1935, 205):

The dating of the Protogeometric was more challenging due to the presence of two pottery styles which seemed to overlap during the 11th century BC: the first one was Schweitzer’s (1917; 1918) Attic Protogeometric and the second one was Skeat’s (1934, 28) Submycenaean (1). Kraiker et al. (1939) saw that the early phases of the necropolis at Kerameikos belonged to the Submycenaean period, followed by the Protogeometric. Both phases belonged the 11th and 10th centuries BC, even though their exact duration was unclear. Kraiker et al. (1939, 162-4) produced synchronisms with Palestine and estimated that the passing from the Protogeometric to the Geometric in Athens was sometime after the middle of the 10th century BC, and more specifically between c.950 and c.930 BC. His estimation was based on Athenian flat-based cups and a fragment of a skyphos recovered at Tell Abu Hawam in the levels immediately preceding the destruction of the settlement in 926 BC. The internal development of the Attic Geometric style was mapped a year later in a short stylistic study by Peter Kahane (1940), who did not consider any context comparisons for dating.

Based on stylistic observations and by considering the broader seriation from Kerameikos, Desborough (1952, 294-5) placed the beginning of the Attic Protogeometric around 1050 BC. He rejected the previous dates offered by German scholars and suggested that the passing from the Protogeometric to the Geometric needed to be placed half a century later, between 900 and 875 BC. He also argued that due to differences in styles between different regions, the Early Iron Age cannot be divided in the same way for the whole of Greece. Desborough’s (1952, 295) dates for the Attic Protogeometric sequence were:

In 1954, Kübler suggested a terminus post quem for the Geometric era sometime in the first quarter of the 8th century BC based on a bronze bowl from Cyprus that was excavated in Grave 42 at Kerameikos (Kübler 1954, 202, fig.5; Schweitzer 1969, 16-19). However, his conclusions regarding the dating of the Geometric sequence were seriously questioned later by Hachmann (1963) not only for high dating, but also for their whole stylistic basis.

In 1957, Desborough produced new synchronisms between Attica, Cyprus and three contexts from the Levant: Megiddo, Tell Abu Hawam and Tell Qasile. Absolute chronologies for these sites had been previously established on known dates of Israelite kings in relation to the foundation of Samaria, and also according to the destruction layers after the invasion of Shishak I in c.918 BC (Desborough 1957, 216). Desborough noted that the Levantine contexts produced two different dating systems, a higher and a lower one. If both were applied in the Attic Early Iron Age, then the results were controversial: firstly, the high (‘biblical’) dates suggested that the Attic Geometric would have lasted about 300 years, while Late Helladic IIIC, Submycenaean and Protogeometric would have all been between c.1150 and c.1025 BC. Secondly, according to the lower dates, the Attic Protogeometric would have ended a little after c.900 BC and the Late Geometric a little before c.650 BC (Desborough 1957, 218). Even though the low dates from the Levantine contexts were lower than the conventional dates established by Kraiker et al. (1939) and Kübler (1954), they seemed to lie within the limits of probability for the Attic Protogeometric and Geometric (Desborough 1957, 218). In later years, Desborough (1964) produced synchronisms between Philistine and Mycenaean contexts correlated with changes that followed the invasion of the ‘Sea Peoples’, and verified that the Attic Protogeometric stood between c.1050 and c.900 BC.

Following Desborough (1952; 1957) and by revising the German chronological system developed by Kahane (1940) and Kübler (1954), R.M. Cook (1960; 1997) produced a new chronological chart, where the Attic Geometric style began at about 900 BC. The style appeared not much later in Argos, Corinth and Boeotia, while in Euboea, the Cyclades and the East Greek cities it appeared at about 850 BC; in Thessaly and Crete in the beginning of the 8th century; and, in Laconia and Western Greece quite later. As for the end of the style, the ‘Orientalising’ became established in Corinth at about 720 BC, while in Athens, the Cyclades and Crete at about 700 BC. Cook’s analysis was based on stylistic comparisons among different Greek regions and his synopsis of the Attic sequence suggested:

Following R.M. Cook (1960), Coldstream (1968, 302-31) examined the internal developments of the Attic Geometric style through attribution techniques (also in relation to Davison 1961), and expanded the internal subdivisions of the three Attic Geometric groups:

Furthermore, Coldstream (1968, 302-10) re-examined Attic and Atticising imports at three sites in Palestine (Tell Abu Hawam, Megiddo and Samaria) and concluded that the end of the Attic MGII was to be placed no later than c.750 BC based the dating of Period V at Samaria. In relation to the imports at Megiddo, he placed the beginning of MGI towards the middle of the 9th century BC. The duration of both sub-phases of the Attic MG were cross-referenced with regard to Attic and Atticising pendent-semicircle skyphoi and kraters recovered at Hama and Al-Mina in Syria (Coldstream 1968, 310-6). During comparisons of Corinthian and Euboean imports found at destruction layers associated with the campaigns of Sargon II at Hama, a terminus ante quem was established for the Corinthian Geometric period at 720 BC (Johansen 1957, 106-8; Coldstream 1968, 313), which verified the observations by Cook (1960); therefore; the end of the Geometric era for Attica and Euboea was placed shortly after that time (Coldstream 1968, 316), at 700 BC.

In relation to Egypt and the Western Greek colonies, Coldstream (1968, 316-7) noted that a child inhumation at Pithekoussai (in Grave 102) contained Corinthian skyphoi and globular aryballoi together with an Egyptian scarab from the time of Pharaoh Boccoris (718-712 BC) (de Salvia 1993, 777-80). The scarab provided a terminus ante quem for the earliest occupation phases of the colony, and also a date for Euboean and Corinthian Late Geometric, and Protocorinthian (2) pottery found at the site (Ridgway 1992; Buchner & Ridgway 1993, 378-82; Coldstream 1995, 251-67). Similar Protocorintian imports found in Late Geometric, Subgeometric and Protoattic wells at the Athenian Agora (Brann 1961a; 1961b) verify the Attic sequence based on its connections to the Corinthian.

Schweitzer (1969, 16-20) re-examined the dating of finds from Kerameikos by Kraiker and Kübler (Kraiker et al. 1939, Kübler 1943; 1954) and considered it in relation to the arguments by Desborough (1952; 1957) and Cook (1960). He refined the existing German chronological system for the Attic Geometric and divided it according to five ceramic styles:

Schweitzer’s chronologies bridged the gap between the traditional ‘German’ dates from Kerameikos and the ‘British’ dating system for the Greek Early Iron Age, even though his phases were slightly different compared to those suggested by Coldstream (1968). Again, both dating sequences were primarily established through stylistic observations.

Coldstreams’s chronological system is widely accepted nowadays, despite the fair amount of criticism that has received from Francis & Vickers (1985). Its main problem is that despite the cross-referencing of his Attic MG and LG divisions with contexts of know dates from the Palestine and North Syria, the identification of his EG divisions and LG sub-divisions are solely stylistic. With regard to the Attic Early Geometric, he accepts the dates suggested by Desborough (1952; 1957) and Cook (1960) for the end of the Protogeometric era and continues his discussion on purely stylistic ground. Lemos (2002, 24-5) notes the same problem with regard to the dating of the Protogeometric in general: absolute dates cannot be cross-referenced with Eastern Mediterranean sites and PG dating depends highly on fixed dates assigned to Late Helladic IIIC. Evidence of trade and connections between Attica and Euboea (3) have helped further in verifying Coldstream’s EG divisions. For example, Athenian Early Geometric II ceramic imports have been excavated at Subprotogeometric graves at Lefkandi (Popham et al. 1980, 350-4; Coldstream 1977, 63-5), pointing to a terminus post quem for both ceramic styles c.850 BC.

Coldstream’s chronological system is followed in this thesis for two reasons: firstly, it provides sub-divisions for the Late Geometric period, which may be stylistic, but they have been used in the discussion regarding Geometric ceramic workshops through connoisseurship. Secondly, his dates have been verified through latest radiocarbon studies, which will be explained in detail in the following section. Morris (1996, 58) concludes that according to the existing synchronisms, particularly in relation to the Western Greek colonies, the absolute chronologies for Greek pottery between the 8th and 7th centuries BC are fixed securely.

In the latest volume on Early Iron Age chronology, Whitley (2001, 61) suggests a slightly revised version of Coldstream’s (1968) chronological system for Attica:

Furthermore, Lemos (2002, 3-26) offers a full discussion on relative chronology for the Protogeometric period across Greece based on the comparison of grave contexts. It must be clarified that according to Snodgrass (1971, 1-25), Coldstream (1977, 25-106) and Whitley (2001, 61) such pot styles are not necessarily chronological periods, as for example, the Geometric style was not universal in ‘Geometric Greece’ during the 9th century BC (4); therefore, any discussion on chronology based solely on pottery must be treated with caution.


  1. The Submycenaean style was no other than the ‘Salamis style’, previously excavated by Kavvadias (1893) and studied by Wide (1910). For the dating of the Submycenaean period see Furumark (1941; 1972); Desborough (1964; 1972); Mountjoy (1986; 1988); Iakovidis (1979, 462) and Ruppenstein (2003; 2007). For arguments regarding the distinctiveness of the Submycenaean style see Whitley (1991, 83-4; 2001, 79); Snodgrass (1971; 2003, 34); Rutter (1977; 1978); and Osborne (1996a, 24).
  2. Additional fixed points for the end of the Corinthian Geometric and Early Protocorinthian were established on evidence from other Western Greek Colonies. Thucydides provided dates for the foundation of Naxos at 734 BC, Syracuse at 733 BC, Leontini at 729 BC, Megara Hyblaea at 728 BC and Gela at 688 BC (Coldstream 1968, 323; also see Dunbabin 1948, 435-71; Graham 1982, 89-91; Osborne 1996, 119-27; Morris 1996, 52). Such dates functioned as a terminus post quem for Corinthian Late Geometric and Protocorintian pottery recovered in the above sites (Coldstream 168, 322-7; also see Vallet & Villard 1952, 329-40).
  3. In a latest publication, Charalambidou (2011) also notes the connections in pottery production between Euboea and Oropos towards the beginning of the 7th century BC.
  4. Similar problems in establishing comparative chronologies for the Geometric and Archaic periods through ceramic styles and typologies have also been discussed by Rückert & Kolb (1993) with regard to Asia Minor.