The Aegean from Bronze Age to Iron Age

Continuity and Change Between the Twelfth and Eighth Centuries BC

Oliver Dickinson

• Publisher: Routledge
• Year: 2006
• ISBN:
  9781134778713
• Edition: 1
• Topic: History
• Subtopic: Greek Ancient History
• Index: History

Following Oliver Dickinson's successful The Aegean Bronze Age, this textbook is a synthesis of the period between the collapse of the Bronze Age civilization in the thirteenth and twelfth centuries BC, and the rise of the Greek civilization in the eighth century BC.

With chapter bibliographies, distribution maps and illustrations, Dickinson's detailed examination of material and archaeological evidence argues that many characteristics of Ancient Greece developed in the Dark Ages. He also includes up-to-date coverage of the 'Homeric question'.

This highly informative text focuses on:

• the reasons for the Bronze Age collapse which brought about the Dark Ages
• the processes that enabled Greece to emerge from the Dark Ages
• the degree of continuity from the Dark Ages to later times.

Dickinson has provided an invaluable survey of this period that will not only be useful to specialists and undergraduates in the field, but that will also prove highly popular with the interested general reader.

1
TERMINOLOGY AND CHRONOLOGY

In this chapter, for clarity, classificatory terms will be spelled out in full; in later chapters the abbreviations listed on p. xv will be used. As set out in the Introduction, the period to be covered will be divided between the Postpalatial Period, equivalent to Late Helladic/Minoan IIIC and Submycenaean/earlier Subminoan, and the Early Iron Age, equivalent to Protogeometric and Geometric.

Establishing a terminology

The outer limits of the period covered in this book take in a half-millennium, the twelfth to eighth centuries (the proposal in P. James et al., Centuries of Darkness (London: Jonathan Cape, 1991) to reduce this period to a century at most will not be discussed here, since it has been universally rejected; see Dickinson 1994a: 17 for references, also Snodgrass [1971] 2000: xxvi). If only because of its situation between periods for which some kind of historical chronology can be established, this long period cannot be treated as a unit. In fact, as noted in the Preface, it is becoming customary to separate the eighth century from the rest as a period on the threshold of true Greek history. But the severe problems of establishing a reliable absolute chronology that can be applied throughout the Aegean region make it impossible to discuss developments for the whole period purely in terms of centuries or fractions of them, let alone to date individual events within it.

The period has a reasonably clear-cut beginning in the series of destructions that marks the end of what, following the usage of Dickinson (1994a), will be termed the Third Palace Period (which in Dickinson 1994a incorporates what is often called the Mycenaean Palace Period of the fourteenth and thirteenth centuries). Although such destructions cannot be identified at every significant Aegean site, they must reflect an important series of events that effectively represents the collapse of the Aegean palace societies within a relatively short period of time (see further, p. 44). But thereafter all is uncertainty, for further ‘destruction horizons’ of the kind which so conveniently divide important stages of the Aegean Bronze Age are lacking, and it is quite uncertain whether the site destructions that have been identified within the Postpalatial Period can be grouped to form a comparable horizon. It is even less clear whether any destructions can be related to the severe climatic event represented in tree-ring evidence found in Turkey, now sited between c. 1174 and 1162 (Kuniholm, cited by M.H. Wiener in BSA 98 (2003) 244) and often attributed to an eruption of Mt Hekla in Iceland which fell in the first half of the twelfth century (Kuniholm 1990: 653–4; but Buckland et al. 1997: 588 question whether this eruption had widespread climatic effects). After the Postpalatial Period, destructions of major sites are hard to identify at all, except in Crete, until the eighth century, when they may sometimes be given a historical setting (e.g. the destruction of Asine in the Late Geometric pottery phase have been linked to the tradition of conquest by Argos, supposedly datable near 700).

Thus, the archaeological record offers no natural breaks to facilitate a subdivision of the period that is not based on pottery phases. Nevertheless, it is the only source that can provide a framework for such subdivision. Sometimes, the ‘Dorian invasion’, ‘Ionian migration’ and related population movements reported in the Greek traditions have been treated as historical events that can be approximately dated and used as chronological signposts in the earlier part of the period. But, even if these traditions could be accepted as containing genuine information, the basis for dating them is shaky indeed. As noted in the Introduction, such movements were essentially dated by reckoning downwards from the Trojan War, but this was not fixed; very varied dates were calculated for it by ancient scholars, all of whom must have based their work on varying interpretations of the genealogies that linked historical persons with famous heroes. But it has long been recognised that these genealogies, among which those of the Spartan royal families are the best known, are too short to fit any possible chronology, if the ‘age of heroes’ is assumed – though this is a very questionable assumption – to have a historical basis in the world of the Mycenaean palaces (see Snodgrass 1971: 10–13; Desborough 1972: 323–5).

Snodgrass has nevertheless argued that several sources suggest a comparable number of generations going back to an ancestor figure or god, and if one allows some thirty years to a generation and reckons backward from the time that these genealogies were recorded, in the fifth century, all seem to begin around the tenth century, which could be considered to represent a horizon of historical significance, even perhaps reflecting the date of the ‘Ionian migration’. But such genealogies cannot be assumed to be trustworthy, even in their ‘human’ generations, without external checks. As is demonstrated in Thomas (1989: ch. 3, especially 180–6), there is good reason to suppose that such ‘full’ genealogies were effectively created from family traditions by the first systematic genealogists in the fifth century, and that this process of creation involved many misinterpretations and distortions, in addition to anything that may have happened during the transmission of the material previously (cf. Davies 1984: 90–1). It would be unwise to believe that even as accounts of descent such genealogies represent information transmitted intact from the past, and this must discredit any dating system based on them.

The archaeological material, then, must be the only basis for establishing some kind of framework for the period. This has principally been done on the basis of pottery phases (cf. most recently Whitley 1991: 83–6, primarily concerned with Athens, and Morris I. 1997, where slightly differing sequences are proposed for central, western and northern Greece and for Crete). But it is undeniable that historical processes do not necessarily fit themselves neatly into phases defined stylistically, and other attempts have been made to suggest a sequence which, although taking its chronology from dates estimated for the pottery phases, is based on more general perceptions of observable processes. But such perceptions are subject to change as new material is discovered. Thus, Snodgrass argued that after the end of the Bronze Age there was continuing decline, which reached a nadir in a phase of ‘bronze shortage’, for which he suggested outer limits of c. 1025 and 950, and which he saw as the time of maximum isolation and poverty. This was followed by recovery, with the revival of communication, in the later tenth century, and culminating in a final ‘renaissance’ in the eighth century (1971: ch. 7; see also Snodgrass 1987: ch. 6). However attractive this might once have appeared as a model of development, more recent finds have called it into question (see Muhly 2003: 23). In particular, the rich burials that the Lefkandi ‘Heroön’ contained are agreed to date to the local Middle Proto-geometric phase, which has close links with Attic Middle Protogeometric and should fall chronologically within Snodgrass’s period of maximum isolation and poverty (see now Snodgrass [1971] 2000: xxvii–xxix, in which many of Snodgrass’s older views are substantially qualified if not totally withdrawn).

Similarly, Coldstream has withdrawn the original subdivisions of ‘Isolation’, ‘Awakening’ and ‘Consolidation’ which he saw in the period c. 900–770, termed ‘the passing of the Dark Ages’, accepting that this reflected a largely Athenocentric viewpoint (1977: ch. 1; see now 2003: 371).

The lesson to be drawn from these cases is that the evidence of one or a few sites cannot provide a universal pattern of historical development for the whole Aegean region. The local histories of regions and sites may have been very different for much of the period, so that no general system of historical phases can be devised for the whole Aegean until the eighth century, when it does become possible to speak of processes that seem to have been at work over much of Greece. For want of anything better, the material will have to be discussed in terms of pottery sequences, of which something must now be said.

Relative chronology

The difficulty with using decorated pottery as the primary basis for relative chronology in the period is that, with certain notable exceptions, the pottery produced for the greater part of the Early Iron Age in most parts of Greece has little distinctive character, being extremely limited in its range of shapes, motifs, and styles of decoration (see further, pp. 128–9). There are enough variations in details between regions for it to be clear that at first there were no centres whose stylistic lead was closely followed over a very wide area. Even when Athens achieved something like a leading position in this respect, its example was never followed completely faithfully, and sometimes seems to have been positively resisted, even in neighbouring regions. Much of the time, the simplicity of the motifs and of the manner in which they were deployed makes it difficult to identify connections between the local styles with any certainty: occurrences of the same motifs, deployed in similar ways, could often be attributed as plausibly to inheritance of a common stock derived from the Late Helladic IIIC styles, as to the exertion of influence by one region on another. Similarly, a generally limited range of shapes makes resemblances between those of one region and another unsurprising.

Also, the deficiencies in the database cannot be stressed too often. Although the situation is improving, large stratified deposits of settlement material are still rare and most of our evidence comes from graves. These generally contain complete vases, in contrast with settlement deposits which consist mostly of sherds, but such vases are seldom numerous and tend to differ considerably from those used in settlements, both in preferences for particular shapes and in the quality of their decoration. Hence it is difficult to compare styles defined from grave-associated material and those defined from settlement deposits with complete confidence.

Enough common features can nevertheless be identified in the archaeological material of the Postpalatial Period to make it possible to suggest a very general sequence of Late Helladic IIIC phases, that applies through much of the Aegean, although precisely how the phases of Cretan Late Minoan IIIC, which are less well defined, should be correlated with this sequence remains a subject of debate. There was a quite substantial Late Helladic IIIC Early phase, marked by competent but unexciting pottery that has many similarities over a wide region, to which at least two building levels can be attributed at Mycenae, Tiryns and Lefkandi. This was followed by a comparably substantial Middle phase, during which several sites and regions produced some extremely individual fine wares that can be seen to interact in complex patterns of cross-influences. Finally, there was a Late phase, generally reckoned to be shorter, of abrupt-seeming decline in quality and range. Even at the beginning of this sequence, although there is a good deal of homogeneity over a wide area, local features can be observed. In the course of it, the local sequences tend to diverge increasingly, retaining a family resemblance but becoming more and more distant towards the end, so that placing individual pieces within the sequence is often a matter for prolonged discussion of parallels.

A Submycenaean stage following Late Helladic IIIC Late has often been identified, but the use of this term has become fraught with difficulties. Morgan has commented (1990: 235) that the term’s originators intended it to indicate whatever falls between Late Helladic IIIC and Protogeometric in any region, without any implication of a strictly defined style, culture, or chronological period. The tendency has nevertheless been to use it as a pottery term, but, as is pointed out in accounts of its use (most recently Whitley 1991a: 81–2; Papadopoulos 1993: 176–81; Morgan 1999: 254–6; Mountjoy 1999: 56), it has been given very varied meanings. Sometimes it has been interpreted as simply the local west Attic variant of Late Helladic IIIC Late (Desborough’s original view) or as the style found on vases buried in graves contemporary with Late Helladic IIIC Late settlements (Rutter 1978), sometimes as the style that succeeded Late Helladic IIIC over a substantial area of the central mainland (Desborough’s later view, but see Mountjoy 1999: 56–7).

The difficulty of producing an agreed definition arises partly because the increasingly marked regionalism that appeared in Late Helladic IIIC pottery entails that whatever followed is hardly likely to show any marked stylistic coherence between different regions. More fundamentally, despite attempts at definition by Desborough and Mountjoy, Submycenaean does not have a very distinctive character; as Desborough comments, ‘even the word “style” is hardly appropriate’ (1972: 41). It is often unclear what distinguishes it from Late Helladic IIIC Late, let alone what is so distinctive about it that it deserves a separate term (Frizell 1986 uses ‘Final Mycenaean’). It is symptomatic that Mountjoy has felt able to reclassify as Late Helladic IIIC Late some of the material from Salamis and the Kerameikos that was originally used by Furumark to define Submycenaean (Myc. IIIC:2 in his terminology, 1972: 77–8), and that levels 13–23 at Kalapodi, originally defined as covering the end of Late Helladic IIIC and Submycenaean, have now been reassigned to Submycenaean, Transition to Protogeometric and Early Protogeometric (compare Felsch 1987: 3 n. 8 with 1996: xvi). All this emphasises how the material of this obscure period, which is known much better from grave-groups than settlement deposits, is typified by gradual transitions rather than marked changes of direction, allowing different scholars to interpret the material in different ways (cf. Lemos 2002: 7–8 on Submycenaean, and Catling in NorthCem: 295–6 on Subminoan).

In fact, the most distinctive types are those of the stage which has been defined as late Submycenaean in the Athens and Lefkandi cemeteries, in which the lekythos has become very prominent, largely replacing the stirrup jar, and Cypriot-derived forms (the bird or duck vase, ring vase, bottle, and flask) are found (Desborough 1972: 43–4, 54; Lemos 2002: 79–80, 81–3).

The bird vase has a long and complex history, and probably originated earlier in the Aegean and came back there from Cyprus in two forms, one found in various parts of the Greek mainland, especially Achaea, and on Skyros, the other on Crete and Cos (Lemos 1994, 2002: 82–3). Examples of the flask and bottle also come from Early Protogeometric tombs at Lefkandi and settlement contexts at Asine, and from a tomb at Karphi (Desborough 1972: 61) (see Figure 7.1 for distribution). Although the closest parallels for these types are Late Cypriot IIIB (but they continue later in Cyprus), occurrences in the Aegean need not represent a very tight chronological link with this phase. Some certainly occur in Protogeometric contexts, while the earliest bird vases of the later forms may well be of later Late Helladic IIIC/Late Minoan IIIC date (all examples from Achaea and Palaiokastro in Arcadia are classified as Late Helladic IIIC Late in Mountjoy 1999: 299, 441). Only at Athens and Lefkandi does it seem legitimate to suggest a close chronological link, although a more generalised overlap between Submycenaean, earlier Subminoan, and the latest Achaean Late Helladic IIIC may be supposed (see Desborough 1972: 61–2, 93). This ‘late Submycenaean’ stage merges with the beginnings of Protogeometric, although the Cypriot types largely disappear from the ceramic repertoire. In fact, Attic Early Protogeometric really seems a transitional phase, in which types that might separately be classified as Submycenaean and Protogeometric are found in the same grave.

The difficulties of definition must lead to the question whether Sub-mycenaean is a useful term. At best, it can be taken to identify the last, not in itself especially significant, stage of the Mycenaean pottery style in some central regions of the mainland, which was probably not very long-lived and can hardly be taken to represent a major phase of historical development (contra Lemos 2002: 26, which allows it two generations, with a third covering the transition to Protogeometric). There seem to have been comparable but different survivals of Mycenaean types elsewhere, as at Kalapodi, already mentioned, and in western Greece, especially the ‘Submycenaean’ material of the graves at Elis (Morgan 1990: 235–7, considers this probably later than Attic Submycenaean) and the Dark Age I of Nichoria. But whether the chamber tomb cemeteries of Achaea and Cephallenia continued in use into the eleventh or even tenth century (see most recently Morris, I. 1997: 549) is quite uncertain; Mountjoy (1999) classifies nothing from them as Submycenaean, although she identified Submycenaean and a transition to Protogeometric on Ithaca (1999: 475–8). In Crete, the current styles (the classification of Subminoan, well defined in the Knossos region, may not be appropriate everywhere; cf. the discussion of the Kavousi sequence in Hallager and Hallager 1997: 366–9) clearly derived from Late Minoan IIIC traditions, and certainly continued well past the point when Protogeometric styles, had become established in central mainland regions. In other regions there is either a still unbridgeable gap between the latest Mycenaean and the next well-represented style (as in Laconia), or a sequence that has no very obvious links with anything Aegean, as in Thasos, Epirus and to a great extent Macedonia, although here a painted style inspired by Mycenaean was popular for a considerable time.

With the establishment of the full Attic Protogeometric style the position becomes much more clear-cut. Knowledge of the Attic sequence is based almost entirely on the evidence of graves at present, particularly those of the Kerameikos and Agora, although considerable deposits, mainly from wells, have been found in the Agora region and some details have been published concerning them (Papadopoulos 2003: 5 and ch. 2). But despite the resulting rather limited amount of material for detailed analysis, the phases of the Attic Protogeometric and Geometric styles as defined in Desborough (1952) and Coldstream (1968) (Early, Middle and Late Protogeometric, Early Geometric I and II, Middle Geometric I and II, Late Geometric Ia, Ib, IIa and IIb) have been generally accepted as valid. Desborough argued that Early and Middle Protogeometric were short phases, no more than a generation each, but Late Protogeometric was long (Desborough 1972: 134–5), a view followed by Lemos (2002: 26). Smithson developed a four-part classification on the basis of the Agora material (Papadopoulos 2003: 5 n. 11). But this material remains unpublished at present, and it seems best to continue with the conventional terminology for now.

Once fully established, examples of the Attic Protogeometric style were evidently exported to, and had some influence over, an increasingly wide region of Greece, while the sequence of changes in the Attic Geometric style has been shown in Coldstream (1968) to be followed with varying degrees of closeness in many other regions. Examples of imported Attic pottery and close local imitations are sufficiently numerous to make establishing the relative position of most material a fairly easy matter in neighbouring parts of Greece from Late Protogeometric onwards, but of these only the Argolid and Euboea have produced good evidence for earlier stages in the sequence.

The Argolid sequence has largely been established on the basis of graves, with some support from settlement deposits not yet fully published, from Argos, Tiryns (Papadimitriou 1998) and Asine. The Asine deposits run without break from ‘Final Mycenaean’ into Protogeometric and are overlaid by Middle and Late Geometric deposits. They include numerous Attic Middle and Late Protogeometric imports, though nothing that looks like the very latest Attic Protogeometric, but unfortunately their value is limited. Although there is a stratigraphical and architectural sequence, there are Attic Late Protogeometric pieces even in the first Protogeometric phase distinguished by Wells (I am grateful to Dr I. Lemos for confirming this; see Lemos 2002: 5–6 for critical comments on Wells’s analysis of the Asine material). The deposits must, therefore, be mixed, which explains why there is little discernible development. Nevertheless, there is local Early Proto-geometric at Asine, which can be paralleled in other Argive material, and although Middle Protogeometric is poorly represented, Late Protogeometric is well documented here and elsewhere (Lemos 2002: 13–14, 17, 21–2). Later, the Argive potters can be seen to follow the changes in Attic Geometric style so closely that they even imitated Early Geometric I, uniquely (on present evidence) in Greece. There are indications that Argive Geometric pottery was influential in other areas of the Peloponnese, but earlier links are very few and often speculative.

The Euboean style is best represented by the material from Lefkandi, though an increasing amount is being published f...