Collapse and Transformation: The Late Bronze Age to Early Iron Age in the Aegean

By Oxbow Books

The years c. 1250 to 1150 BC in Greece and the Aegean are often characterised as a time of crisis and collapse. A critical period in the long history of the region and its people and culture, they witnessed the end of the Mycenaean kingdoms, with their palaces and Linear B records, and, through the Postpalatial period, the transition into the Early Iron Age. But, on closer examination, it has become increasingly clear that the period as a whole, across the region, defies simple characterisation – there was success and splendour, resilience and continuity, and novelty and innovation, actively driven by the people of these lands through this transformative century.

The story of the Aegean at this time has frequently been incorporated into narratives focused on the wider eastern Mediterranean, and most infamously the ‘Sea Peoples’ of the Egyptian texts. In twenty-five chapters written by 25 specialists, Collapse and Transformation instead offers a tight focus on the Aegean itself, providing an up-to date picture of the archaeology ‘before’ and ‘after’ ‘the collapse’ of c. 1200 BC. It will be essential reading for students and scholars of the Aegean and eastern Mediterranean regions, as well as providing data and a range of interpretations to those studying collapse and resilience more widely and engaging in comparative studies.

Introductory chapters discuss notions of collapse, and provide overviews of the Minoan and Mycenaean collapses. These are followed by twelve chapters, which review the evidence from the major regions of the Aegean, including the Argolid, Messenia, and Boeotia, Crete, and the Aegean islands. Six chapters then address key themes: the economy, funerary practices, the Mycenaean pottery of the mainland and the wider Aegean and eastern Mediterranean region, religion, and the extent to which later Greek myth can be drawn upon as evidence or taken to reflect any historical reality. The final four chapters provide a wider context for the Aegean story, surveying the eastern Mediterranean, including Cyprus and the Levant, and the themes of subsistence and warfare.

• Language: English
• Publisher: Oxbow Books
• Release date: Apr 9, 2020
• ISBN: 9781789254266

Book preview

1

Introducing collapse

Guy D. Middleton

Readers of this volume should be aware that there are many different ideas of what ‘collapse’ means, as well as theories about what caused any given collapse (see e.g. Middleton 2017a; Storey and Storey 2018). This short review aims to provide an introduction to some of these ideas and related concepts, so that readers may weigh up the evidence presented in the volume and place it in what they feel to be an appropriate context.

What is collapse?

It has been suggested that arguments about collapse, its nature and causes, often stem from different understandings of what is meant by the term and what it is applied to (Demarest 2001; Middleton 2017b; Tainter 2006). Schwartz (2006, 5–6) gives a useful summary of what many archaeologists have in mind: ‘the fragmentation of states into smaller political entities; the partial abandonment or complete desertion of urban centres, along with the loss or depletion of their centralizing functions; the breakdown of regional economic systems; and the failure of civilizational ideologies.’ These are all potentially detectable by archaeologists through changes in material culture and without historical evidence. Inevitably, though, a lack of historical evidence means that narratives of prehistoric collapses are difficult to construct and will tend to lack securely identifiable actors and events. This can increase the tendency for overly neat, simplistic or deterministic explanations (such as ‘megadrought’) to be proposed and found plausible.

Some authors on collapse have focussed on population or the environment – in combination, overpopulation and environmental damage, or climate change, caused apocalyptic collapse, which was characterised in particular by population loss – a neo-Malthusian ‘overshoot’ model. Diamond thus states that collapse is ‘a drastic decrease in human population size and/or political/economic/social complexity, over a considerable area, for an extended time’. Post-collapse periods are then seen as times of environmental regeneration and low population (Chew 2005; Redman and Kinzig 2003).

However, this overshoot model and Diamond’s approach to collapse have been found wanting by Tainter (2006) and others (McAnany and Yoffee 2014; Middleton 2012). The apocalyptic view of collapse – hundreds or thousands dying in a nightmare scenario of chaos and violence – is not what most archaeologists have in mind when they discuss the subject, even though in a number of examples declining population is thought to have accompanied or resulted from collapse. Poverty too can explain the decreased visibility of parts of the population, or why people ‘disappear’ (in some cases, the poorer sections of the population may never have been visible in the first place). Relevant to note here is that the massive population loss caused by the Black Death in Europe did not cause any states to collapse, nor did either the Athenian plague of the late fifth century BC or the Justinianic plague of the sixth century AD bring about Athenian or eastern Roman collapse.

A bogey word in collapse studies is ‘civilisation’, on which Yoffee and Cowgill, and the other authors in their 1988 volume, provide useful guidance. Both Yoffee (1988) and Cowgill (1988) suggest separating consideration of states, or political units, which can and do collapse, from civilisations or ‘great traditions’, which do not. Civilisations, if the word must be used, are distinct constellations of material and non-material ‘phenomena’, constantly transforming, and the collapse of political states embedded in them usually – and unsurprisingly – involves some visible changes to the civilisation, especially in elite material culture and practices. Cowgill (1988, 256) also argues that neither the term collapse or fall should be used to describe the political fragmentations of states or empires into smaller parts, because there might be no reduction in complexity in those parts. In practice this is what the term is usually used for, though, but Cowgill’s point about complexity in the remaining parts should be held in mind.

We can consider the changing material, and political and social culture of the Roman Republic and Empire over centuries as a clear example of ‘civilisational’ transformation; when the last western emperor died, the western Empire fell and was succeeded by ‘barbarian’ kingdoms and post-Roman enclaves; it was the collapse of a political unit. In this instance, material culture and established traditions did not come to an abrupt end, despite significant economic and material repercussions from the events around the collapse (compare Brown 1971 and Ward-Perkins 2005). The Hittite collapse too was political, but much of established elite Hittite culture continued in the Neo-Hittite kingdoms, if not in the old Hittite heartland (the capital Hattusa having been gradually abandoned – what happened to the king and inhabitants is unknown). The Classic Maya collapse involved the end of many independent city-states at different times (the end of royal lines and the abandonment (total or partial) of urban areas) within the three-century Terminal Classic period, with a long transformation in the totality of Maya material culture observable between, say AD 800 and 1100. Neither the Maya nor their culture disappeared. A political collapse, the collapse or failure of a state, whilst certainly having knock-on effects, need not result in an allout loss of culture and traditions, or a sudden mass die-off of people (Middleton 2017b).

More widely referred to in the archaeological literature, though not always followed, is the definition of Tainter in his classic work on collapse (Tainter 1988). He suggested that collapse is when a society ‘displays a rapid, significant loss of an established level of socio-political complexity’ (Tainter 1988, 4). In this clear and helpful view collapse is a political process in which a society becomes less complex and its hierarchy and parts are reduced. This simplification may also affect other aspects of life, such as the ability of a power to organise groups, a reduction in the level of investment in monumental architecture and art, a reduction in socio-economic differentiation, stratification, and general stability, and may result in the appearance of smaller post-collapse polities. As for causes, Tainter (1988, 38) adopts an economic perspective; he argued that ‘declining marginal returns’ were key – an increase in complexity to solve problems at some point stopped paying off, leaving structures to decomplexify over perhaps two or three decades. Issues that arises here concern the role of people and events in collapse – how did a process of simplification actually play out ‘on the ground’?

Tainter’s views owe something to Renfrew (1984, 367– 369), who provided an outline of the features of ‘system collapse’. These are worth reproducing in full (in Table 1.1), not least because they are in great part influenced by the circumstances of the Mycenaean collapse and Early Iron Age Greece (Renfrew 1984, 367). In Renfrew’s view, collapse could take around a century to fully complete, and would affect the political, social, and economic worlds as well as population and settlement. It would be followed by a dark age that displayed elements of continuity and that was at least partly constructed by post-collapse origin myths and the tastes of modern archaeologists. The collapse would not have an obvious cause.

Table 1.1 Features of collapse identified by Renfrew (1984)

More recent contributions have produced somewhat different or extended definitions, owing something to the development of the archaeological and wider discourses of collapse (Middleton 2017a). Johnson (2017), for example, dislikes the term ‘collapse’:

What archaeologists see as a collapse is usually just a transition to a different way of life … the idea of a rapid failure of the systems on which a population depends is intriguing but not an accurate way to describe what happens to most complex societies … ‘Transition’ is a neutral term that better conveys what happens … I use the term ‘collapse’ in a general way, and in most cases I will avoid ambiguity by qualifying what type of breakdown occurred.

This view contrasts strongly with the ‘traditional’ – or at least ‘popular’ – idea of collapse as a negative change or a terrible catastrophe, though ‘just a transition’ perhaps underplays the nature and significance of some of the changes we call collapse, and their impacts on the people who lived through those periods.

Storey and Storey (2017, 11–12) in their informative comparative study of the collapse of the western Roman Empire and the Classic Maya give both a more and less formal explanation, which attempt to draw together the variety of terms centring around collapse. Thus: decline = ‘things going to hell’; political fall = ‘when things go to hell to the extent that major political institutions cease to function’; collapse = ‘if things go so completely to hell that the culture loses coherence and the major defining elements and dimensions of that culture disappear’; and resilience = ‘after collapse, there is a giving way to a new cultural entity’. More formally they see collapse as:

a major disjuncture in the trajectory of a complex culture (those commonly called ‘civilizations’); the political integration completely fails, and the Great Tradition (the assemblage of material culture and reflected ideologies unique to that culture) similarly comes to an end.

Unlike some of the authors mentioned above, this definition suggests that ‘fall’ applies to political entities and ‘collapse’ should be seen as an end of cultural entities – civilisations.

In terms of timescale, there are also some differences of opinion. For Tainter (1988), collapse is rapid, taking two or three decades, but for Renfrew (1984) and Storey and Storey (2017) it can take a century. Partly the problem with this area is where to sketch in the beginning and end of any collapse ‘event’. The western Roman collapse can be seen in hindsight to have had its origins in the late fourth century AD and taken around a century to work out to its final end (Heather 2005), although collapse was not inevitable and the Roman Empire could have collapsed at many points much earlier than it did (there were many instances of civil war and successful and attempted secessions). The Classic Maya collapse, so-called, took place across the course of the Terminal Classic Period – three centuries. But in the latter case, the collapse of individual Maya states could be very rapid indeed – marked by conflict and the execution of royal families. Arguably the collapse of individual Mycenaean states could have been very rapid too, even if the collapse of the palace states as a whole took place over several decades.

Given the difficulty in identifying neat boundaries to collapse, it can be hard to detect clear beginnings of processes of collapse, and so to isolate causes on multiple levels. While tempting to try to recognise signs of ‘anxiety’, ‘crisis’, or ‘decline’, or at least ‘problems’ prior to collapse, it is possible to fall foul of hindsight and teleology and then to overlook the possibility of sudden and unexpected collapse. Even the collapse of the Soviet Union at the time it happened, came as a surprise, though was predicted in a general sense (Aron 2011). Many ancient states were frequently in a state of crisis, for one reason or another, but clearly these crises did not always lead to collapse; this suggests a key role for chance historical factors in collapse (Bronson 1988; Kaufman 1988). There is an inescapable ‘fuzziness’ here, which makes the study of collapse and causality particularly difficult.

Collapse – good or bad?

Collapse is often envisioned as ‘a bad thing’ – after all, when businesses, ecosystems, or populations collapse, it can well be. Scott (2017, 186) points out that collapse is ‘often understood to be a deplorable regression away from a more civilized culture’. The apocalyptic turn in collapse studies has been noted above; while not dominant, it lurks ever present in the minds of many. It is also formalised in some popular and authoritative views of collapse. Diamond thus painted a gloomy and violent picture of the end of the Norse settlements in Greenland, which he says was:

sudden rather than gentle, like the sudden collapse of the Soviet Union [the Eastern settlement was] like an overcrowded lifeboat … famine and disease would have caused a breakdown of respect for authority … starving people would have poured into Gardar, and the outnumbered chiefs and church officials could no longer prevent them from slaughtering the last cattle and sheep … I picture the scene as … like that in my home city of Los Angeles in 1992, at the time of the so-called Rodney King riots … thousands of outraged people from poor neighbourhoods … spread out to loot businesses and rich neighbourhoods (Diamond 2005, 271–273).

From a Dahlem workshop on sustainability and collapse comes an affirmation of this view, in which Young and Leemans (2007, 450), define collapse as a rate of change to a system that ‘has negative effects on human welfare, which, in the short term, are socially intolerable’. In the archaeological literature, Storey and Storey (2017, 11–12) add that the ‘process also entails human suffering on a large scale, largely through diminution of population, which almost never means total disappearance of a population; but there is significant loss of life and a smaller population left behind’.

Yet to characterise such ‘big’ events as the collapse of a state as good or bad, or a state as a success or failure, is too simplistic – and thus any very general portrayal is bound to be partisan and to break down on close examination. Young and Leemans’ statement is immediately problematic, given that many people could and did live in socially intolerable conditions in the heyday of a given state or empire (even in the contemporary UK, nearly four million people in poverty have used foodbanks; also consider in the ancient world those at the bottom of the heap – slaves, prostitutes, and the rural and urban poor). Scott (2017, 186) observes that collapse does ‘not necessarily mean a decline in regional population … in human health, well-being, or nutrition [it] may represent an improvement …’

Numerous authors have pointed out that past collapses appear usually to affect a society’s elite much more than its majority population of peasant or subsistence farmers, and that at a lower level both populations and traditions continue (e.g. Renfrew 1984; Tainter 1999; 2016; Scott 2017). Elites that were disempowered or on the losing side would have experienced events as ‘bad’, such as the executed rulers of Maya cities mentioned above (see Demarest 2014), yet for elites on the winning side, assuming conflict of some kind to be a key part in many collapses, the outcome may have been positive. For the majority there may have been little change – or possibly a positive reduction in taxes and duties owed to the rulers (although increased instability and conflict could have negatively affected rural life). Storey and Storey (2017, 54) suggest that the fact that peasants survive is ‘unremarkable’, but this seems to privilege the place of the elite in society and to downplay the resilience of the majority – this resilience would seem to suggest an important conclusion for modern complex societies looking for lessons from history. With imperial collapse, formerly ‘attached’ regions may have achieved a cherished – and perhaps more practical – independence, though in these cases it may still have been elites that benefitted more than those still at the lower end of the socio-economic spectrum (though if independence brought about less conflict and exploitation, the benefits would have been wider).

Collapse can be seen as a ‘reset’ of sorts, a situation in which there were both winners and losers before, during and after. For some change was a disaster, for some just change, and for some things stayed largely the same. Given that so many perspectives are possible and that multiple ‘accurate’ narratives can be constructed for any historical event and the people living through them, those seeking to characterise collapse or its consequences, or to add some kind of qualitative dimension to their descriptions, must proceed carefully.

Collapse, resilience and fragility

The concepts of resilience and the regeneration of complex societies after collapse have both come to the fore in collapse studies (e.g. Blanton 2010; Faulseit 2016; Redman 2005; Schwartz and Nichols 2006). Resilience has a specific meaning drawn from ecology, which Walker and Salt (2006, xiii) explain as ‘the capacity of a system to absorb disturbance and still retain its basic function and structure’. Resilience conceptualised as an adaptive cycle, expressed as a figure ∞. There are four phases in the cycle: exploitation (r), conservation (K), release (Ω), and reorganisation (a). Collapse is seen as the release or Ω phase, which is followed by reorganisation (a). When multiple adaptive cycles are linked, the term ‘panarchy’ is applied, represented as connected ‘figure eights’ (Gunderson and Holling 2002). Thus repeated and cyclical change – growth and collapse is imagined, in something resembling a biological metaphor.

Resilience thinking is applied to population groups, to the environment, and to cultures and societies as a whole and in terms of parts (e.g. Redman and Kinzig, 2003). When using the concept, it is helpful to be specific about what exactly it is being applied to. Collapse would imply that a system (using this word in a broad sense) as a whole was not resilient, though some constituent parts of it and visible expressions of it may have been. A population may be resilient through collapse (i.e. a village, or region, or broader defined group) in terms of continued biological survival, whilst abandoning or rejecting elements of the former culture or system – in which case that system (economic/ideological etc.) could be said to be not resilient. To be less abstract, resilience can refer, in an archaeological sense, to the themes of continuity and change in all areas and a society or system as a whole.

However, it might be wondered how useful resilience thinking is as a tool for understanding past collapse – or non-collapse. Iannone argues that it can provide a common language and set of concepts for communicating ideas across disciplines, but this assumes many people will be familiar with the terms and concepts (Iannone 2016, 130). He also observes that ‘not all systems pass through the various phases of the adaptive cycle in the anticipated order’ and ‘not all of the ideal characteristics of a particular phase will be exhibited by a specific archaeological example’, thus fitting an example into a resilience framework can be ‘a matter of taste’ (Iannone 2016, 181, 204–205). Resilience seems able to be used for both continuity through and reorganisation (‘adaptation’) after collapse. Thus, for this author at least, the concept of resilience seems – beyond a common sense understanding of the idea – somewhat problematic.

Fragility is another concept that has attracted recent attention in relation to the formation, survival and collapse of ancient states (Scott 2017; Yoffee 2019). Scott suggests that early states can be seen as like multi-storey pyramids built by children – the higher the rows of blocks rose, the less stable and more prone to collapse the growing pyramid was; ‘that it soon falls apart is hardly surprising’ (Scott 2017, 183–184). The blocks themselves were more resilient than the larger structures they were incorporated into (see also Kaufman 1988; Yoffee 2005, 135–136; Simon 1965). Thus it is clear to see how smaller units could survive the collapse of an overarching unit. States seen this way are fragile, even those that endured for centuries, and here the role of chance and the historically specific comes into play (Kaufman 1988). Longevity is not necessarily the same as stability – or strength, and, to borrow a phrase from the financial sector, ‘past performance is no guarantee of future results’.

Reorganisation and regeneration

The resilience cycle has a reorganisation phase following the release phase, and a number of collapse scholars have similarly identified both collapse and post-collapse periods as periods of reorganisation and even regeneration. Renfrew (1984) noted that collapse could result in the emergence of simpler and smaller polities, which sometimes echoed those found in earlier times and even followed earlier territorial arrangements. He also identified the sometimes rapid regeneration, after a gap, of collapsed regions into chiefdoms or even states. This view is followed by Storey and Storey (2017, 11), who focused on the western Roman and Classic Maya collapses:

The people (in most cases) survive, persist, and regenerate into another complex society; usually there is a gap (large or small) before re-establishment or re-integration of complexity into a new political system with a new Great Tradition, which is partly derivative of the old but also distinct.

Scott (2017, 186) agrees with Renfrew that ‘a collapse’ at the centre is less likely to mean a dissolution of a culture than its reformulation and decentralization’. Far from slipping necessarily into savage, culturally barren and sparsely populated dark ages, then, there is always continuity in some form through collapse – often at a folk level, sometimes, in part, at a cultural or even organisational level.

None of this is intended to downplay the severity of events or obscure the magnitude of change in many identified collapses. Changes could be sudden and profound, but in the big picture change can be seen as a normal part of the ebb and flow of dynamic human socio-political organisations and dominant ideologies. It must be remembered that people were always, even when environmental factors may have been key, the agents of collapse and transformation, even though these could be the unintended, rather than planned, consequences of actions and policies on many levels.

At an individual level, this means that there was no automatic erasure of the past in the minds of those who lived through collapse events – though certainly the memory and representation of the past could be manipulated (the erasure of the Bastille and symbols of the old regime in revolutionary France springs to mind). People made deliberate choices to shape their worlds before, through and after collapse, though naturally there were also constraints on action that may have made the perpetuation or reconstruction of a system difficult or impossible. Change over time meant that each generation grew up in different circumstances, with different realities, expectations, and desires, and so we should not automatically expect or look for total continuity. In illiterate societies, the ‘real’ historical past may have been easily lost. It is an open question as to why the re-emergence of states in Greece after the Late Bronze Age took so long to occur – some four-to-five centuries – but it is wrong to devalue the intervening period as in some way a failure and indeed to expect new states as some kind of natural or inevitable outcome. If collapse involved a rejection of what came before, state regeneration may have been deliberately avoided.

Eisenstadt (1988, 242) has commented succinctly that ‘the investigation of collapse in ancient states and civilizations really entails identifying the various kinds of social reorganization in these types of societies and so viewing collapse as part of the continuous process of boundary reconstruction’. States were and are not ‘natural’ and were not and are not necessarily the ‘best’ organisation for human communities – they are certainly not the only forms. By seeing states as entities that sometimes appear in the flow of human history, we remove the necessity to consider collapse as either an aberration or something with a moral value. Collapse can itself be reorganisation and adaptation, and post-collapse periods too can see different forms of organisation appear.

Conclusion

This short discussion has highlighted some of the key areas of thinking on collapse and associated terms and concepts. It has considered what collapse is – to different scholars, whether collapse is ‘good’ or ‘bad’, and has introduced the ideas of resilience, fragility, reorganisation, and regeneration. It is necessarily brief and many of the issues raised could be explored in much more depth and nuance. Nevertheless, it should serve as something of an introduction to collapse and provide some context to the evidence presented in the rest of the volume.

Acknowledgments

I thank Norman Yoffee for alerting me to the publication of his new and important edited volume on fragility (Yoffee 2019), which clearly bears on the issue of the collapse of ancient states.

Further reading

For readers interested in following up this discussion, works to consult include Yoffee (2019), Storey and Storey (2018; 2017), Middleton (2017a and b; 2012), Faulseit (2016), McAnany and Yoffee (2014), Schwartz and Nichols (2006), and the foundational works of Tainter (1988; also 2016; 1999), Yoffee and Cowgill (1988), and Renfrew (1984). An exploration of the various discourses of environmental collapse in modern society can be found in Vogelaar, Hale and Peat (2018).

References

Aron, L. (2011) Everything you think you know about the collapse of the Soviet Union is wrong. Foreign Policy. https://foreign-policy.com/2011/06/20/everything-you-think-you-know-about-the-collapse-of-the-soviet-union-is-wrong/

Bronson, B. (1988) The role of barbarians in the fall of states. In N. Yoffee and G.L. Cowgill (eds) The Collapse of Ancient States and Civilizations, 196–218. Tucson, The University of Arizona Press.

Brown, P. (1971) The World of Late Antiquity. London, Thames and Hudson.

Chew, S.C. (2005) From Harappa to Mesopotamia and Egypt to Mycenae: Dark Ages, political/economic declines, and environ-mental/climatic changes. In C. Chase-Dunn and E.N. Anderson (eds) The Historical Evolution of World-Systems, 52–74. New York, Palgrave.

Cowgill, G.L. (1988) Onward and upward with collapse. In N. Yoffee and G.L. Cowgill (eds) The Collapse of Ancient States and Civilizations, 244–276. Tucson, The University of Arizona Press.

Demarest, A.A. (2001) Climatic change and the Maya collapse: the return of catastrophism. Latin American Antiquity 12(1), 105–107.

Demarest, A.A. (2014) The Classic Maya collapse, water, and economic change in Mesoamerica: critique and alternatives from the ‘Wet Zone’. In G. Iannone (ed.) The Great Maya Droughts in Cultural Context: Case Studies in Resilience and Vulnerability, 177–206. Denver, University Press of Colorado.

Diamond, J. (2005) Collapse: How Societies Choose to Fail or Survive. London, Penguin.

Eisenstadt, S.N. (1988) Beyond collapse. In N. Yoffee and G.L. Cowgill (eds) Collapse of Ancient States and Civilizations, 236–243. Tucson, University of Arizona Press.

Faulseit, R.K. (2016) Beyond Collapse: Archaeological Perspectives on Resilience, Revitalization, and Transformation in Complex Societies. Carbondale, Southern Illinois University Press.

Gunderson, L.H. and Holling, C.S. (eds) (2002) Panarchy: Understanding Transformations in Human and Natural Systems. Washington, DC, Island Press.

Heather, P. (2005) The Fall of the Roman Empire. London, Pan.

Iannone, G. (2016) Release and reorganization in the tropics: a comparative perspective from southeast Asia. In R.K. Faulseit (ed.) Beyond Collapse: Archaeological Perspectives on Resilience, Revitalization, and Transformation in Complex Societies, 179–212. Carbondale, Southern Illinois University Press.

Kaufman, H. (1988) The collapse of ancient states and civilizations as an organizational problem. In N. Yoffee and G.L. Cowgill (eds) The Collapse of Ancient States and Civilizations, 219–235. Tucson, The University of Arizona Press.

McAnany, P.A. and Yoffee N. (2014) Questioning Collapse: Human Resilience, Ecological Vulnerability, and the Aftermath of Empire. Cambridge, Cambridge University Press.

Middleton, G.D. (2012) Nothing lasts forever: environmental discourses on the causes of past societal collapse. Journal of Archaeological Research 20(3), 257–307.

Middleton, G.D. (2017a) The show must go on: collapse, resilience, and transformation in twenty-first century archaeology. Reviews in Anthropology 46 (2–3), 78–105.

Middleton, G.D. (2017b) Understanding Collapse: Ancient History and Modern Myths. Cambridge, Cambridge University Press.

Redman, C.L. and Kinzig, A.P. (2003) Resilience of past landscapes: resilience theory, society, and the longue durée. Conservation Ecology 7(1), 14.

Renfrew, C. (1984) Systems collapse as social transformation. In C. Renfrew Approaches to Social Archaeology, 366–389. Cambridge, Harvard University Press.

Schwartz, G.M. (2006) From collapse to regeneration. In G.M. Schwartz and J.J. Nichols (eds) After Collapse: The Regeneration of Complex Societies, 3–17. Tucson, Arizona University Press.

Schwartz, G.M. and Nichols, J.J. (eds) (2006) After Collapse: The Regeneration of Complex Societies. Tucson, Arizona University Press.

Scott, J.C. (2017) Against the Grain: A Deep History of the Earliest States. New Haven, Yale University Press.

Simon, H. (1965) The architecture of complexity. Yearbook of the Society for General Systems Research 10, 63–76.

Storey, R. and Storey, G. (2017) Rome and the Classic Maya: Comparing the Slow Collapse of Civilizations. New York, Routledge.

Storey R. and Storey G.R. (2018) The collapse of complex societies. In C. Smith (ed.) Encyclopedia of Global Archaeology. Springer, Cham.

Tainter, J.A. (1988) The Collapse of Complex Societies. Cambridge, Cambridge University Press.

Tainter, J.A. (1999) Post-collapse societies. In G. Barker (ed.) Companion Encyclopedia of Archaeology, 988–1039. Routledge, London,

Tainter, J.A. (2006) Archaeology of overshoot and collapse. Annual Review of Anthropology 35, 59–74.

Tainter, J.A. (2016) Why collapse is so difficult to understand. In R.K. Faulseit (ed.) Beyond Collapse: Archaeological Perspectives on Resilience, Revitalization, and Transformation in Complex Societies, 27–39. Carbondale, Southern Illinois University Press.

Vogelaar, A., Hale, B., and Peat, A. (eds) (2018) The Discourses of Environmental Collapse. London, Routledge.

Walker, B. and Salt, D. (2006) Resilience Thinking: Sustaining Ecosystems and People in a Changing World. Washington DC, Island Press.

Ward-Perkins, B. (2005) The Fall of Rome and the End of Civilization. Oxford, Oxford University Press.

Yoffee, N. (1988) Orienting collapse. In N. Yoffee and G.L. Cowgill (eds) The Collapse of Ancient States and Civilizations, 1–19. Tucson, The University of Arizona Press.

Yoffee, N. (2005) Myths of the Archaic State: Evolution of the Earliest Cities, States, and Civilizatons. Cambridge, Cambridge University Press.

Yoffee, N. (ed.) (2019) The Evolution of Fragility: Setting the Terms. Cambridge, McDonald Institute for Archaeological Research. https://doi.org/10.17863/CAM.40692

Yoffee, N. and Cowgill, G.L. (eds) (1988) The Collapse of Ancient States and Civilizations. Tucson, The University of Arizona Press.

Young, M.N. and Leemans, R. (2007) Group report: future scenarios of human-environment systems. In R. Costanza, L.J. Graumlich and W. Steffen (eds) Sustainability or Collapse? An Integrated History and Future of People on Earth, 447–470. Cambridge, Dahlem University Press and Massachusetts Institute of Technology.

2

Mycenaean collapse(s) c. 1200 BC

Guy D. Middleton

Introduction

This chapter provides some context and orientation to the Mycenaean palace period of the Late Bronze Age (LBA), the collapse c. 1200 BC, and the postpalatial period, and critically discusses several suggested explanations. It proposes that, rather than focusing on ‘external’ causes, the existence of which is dubious, we see the collapse primarily at a human level as an outcome of the dynamic political situation that must have existed within and between the states and regions of LBA Greece. Following this line of reasoning, it suggests that expansion, fragility and hegemonic failure might provide at least some explanation of the collapse of the palace states and the archaeological record, as it appears to do in several other cases of collapse, e.g. the Akkadian and Hittite empires, and Classic Maya state collapses (see Middleton 2017, 2018c and forthcoming). It also emphasises that collapse was only partial; that is, only the palace states collapsed and much of thirteenth-century BC Greece was ‘non-palatial’. This means that there was clear uninterrupted continuity in many regions, in terms of population, culture and traditions. It was specifically state-level political units that collapsed c. 1200 BC and some aspects of elite-sponsored material culture and traditions that disappeared, though this was accompanied by a reduction in visible sites that began in the thirteenth century.

The Mycenaean world in the thirteenth century

By 1200 BC, there had been fairly simple state-level societies in mainland Greece for perhaps two centuries (Bennet 2013; Dickinson 2014a). It is generally reckoned that six or seven mainland states developed there, each with small (by Near Eastern standards) urban centres with royal palaces as capitals for ruling kings (wanax = king). There was an Argolid state, dominated by the twin palace sites of Mycenae and Tiryns, a Messenian state, with its capital at Pylos, a Lakonian state, with its palace at Ayios Vasileios, and palaces at Thebes in southern Boeotia, Orchomenos and Gla in northern Boeotia, and Iolkos (and possibly Kastro-Palaia) in Thessaly; there may have been a Mycenaean state based at Athens in the later thirteenth century too. Tablets with Linear B writing in Greek have been found at several but not all of these palaces (including recently at Kastro-Palaia in Thessaly (Archaeological Reports 57 (2011), 77–78)), and these demonstrate the existence of a clear social hierarchy with a king, ‘companions’, priests and priestesses, dependent workers, and the existence of a civic body called the damos (later Greek demos – the people) (Bennet and Shelmerdine 2008, 292–295). However, not all palaces or palace states were the same in all aspects (Olivier 1984). Various differences in history and society have been deduced from the Linear B records of Pylos and Knossos (see e.g. Olsen 2014 and Shelmerdine 1999a). By way of comparison, in Classical times, there were more than a thousand city-states in Greece, of various sizes, political and social organisation, and clout.

How powerful palace centres were, or came to be, in ‘their’ own territories is unclear, and again there may have been differences. Bennet (2007; see also Davis and Bennet 1999) has described the development and expansion of the Pylos state from small local beginnings to a large territorial state. This no doubt involved the Pylian elite employing techniques such as alliance and co-operation, strategic intermarriage, as well as coercion and conflict, over time. We should not assume a simple linear ‘progress’, expansion, or even consolidation – policy no doubt changed across the LBA and according to circumstance and the personal ambitions and successes of rulers and factions. The Pylos Linear B documents indicate, at the end of the thirteenth century, an administrative organisation of two provinces, sixteen districts and around 240 sites; Cosmopoulos (2006) proposes a four-tier hierarchy of sites, with Pylos at the top.

The recent excavation of major buildings and the recovery of fresco decoration (Smith 2013, 30) and (the earliest mainland) Linear B from Iklaina, only 10 km from Pylos, add new detail to this story. The evidence suggests that Iklaina may once have been the centre of an independent state close to Pylos (Cosmopoulos 2019). The site was destroyed around the mid-thirteenth century BC, at the same time as Pylos itself was expanding; probably the Iklaina state was conquered and absorbed by the Pylos state. Iklaina seems to have remained important as a local ‘capital’ that was recorded as *a-pu2 on the later Pylian Linear B (Cosmopoulos 2019, 370).

Messenia in the late thirteenth century can therefore be considered a large territorial state, or it was imagined as such in the Linear B, but even so the direct influence (or interference) of the palace may have been primarily in its own environs, and at key strategic locations and a ‘light touch’ in peripheral areas. Palaima (2004, 269–270) suggests that their direct influence of the palaces may have been felt ‘by most inhabitants of the territory only intermittently and indirectly’. Palaces and palace states are no longer seen as totalitarian-type entities that regulated all aspects of life and redistributed most produce. However, elements of the population of palace states may periodically have been ‘recruited’ by the palaces for military service, building, or other work, and some, perhaps of the more elite sort, took part in palace-sponsored events such as feasting.

The Mycenae state’s expansion in the fourteenth century may be visible in the destruction and rebuilding of a new and different type of palace at Tiryns (Maran 2015, 280) and in the thirteenth century may be visible with the founding and management of the port town of Kalamianos on the Saronic Gulf (Pullen 2019). This port, emphasising Mycenae’s clout, was presumably used for trade and military purposes, and may have helped Mycenae to eclipse the formerly important site of Kolonna, on nearby Aegina (Tartaron et al. 2011, 630–631). This ties in with the spread of goods originating in the Argolid across the Aegean (Voutsaki 2001). Unfortunately, Linear B texts are not as plentiful as at Pylos. How far Mycenae’s power stretched is unclear – and we return to this below.

The relationship between the mainland centres and Crete in Mycenaean palatial times is a matter of considerable debate (Driessen and Langohr 2007). Knossos had a Linear B-using administration after the Neopalatial destruction of the Minoan palace, but this final palace was also destroyed before 1200 BC (in LM IIIA2-B early, according to Rehak and Younger 1998, 92, 149, 160); Chania, also with Linear B, survived until later on, perhaps to c. 1200 BC or after, but its significance, whether it was the centre of an independent palace state, as it seems to have been in Minoan times, or was attached to a mainland centre, is not clear. There was movement of goods from Crete to the mainland, as demonstrated by the presence of inscribed stirrup jars (e.g. Hallager 1987).

Sherratt (2001, 238) has characterised the palaces and their administrative culture as ‘clumsily grafted’ onto a client-based warrior society, but this underestimates their importance and longevity in the history of Mycenaean Greece. The long and ongoing development of the palaces and states over a significant time reflected political, social, cultural and economic changes in the regions and within a Mycenaean culture zone and more widely within the Aegean and eastern Mediterranean. As they cohered and expanded, and became more deeply embedded in the fabric of Greek society, they in turn acted as influencers, creating in some areas across Greece a ‘palatial habitus’ that enculturated the population around in a certain ‘palace-centric’ world view. The palace societies of the late thirteenth century BC thus had a not inconsiderable body of history and tradition behind them, and the palaces undoubtedly served as focal points in the political, economic, and social landscape of their own regions and the wider Aegean. For the inhabitants of these parts of Greece in the later thirteenth century, their mental world would have incorporated a long palatial past, full of characters and events, and a normative world of kings, palaces, armies and states.

Many regions of Greece, such as Achaea, Corinthia, Elis, Arcadia, Phokis and Lokris, Aetolia, Acarnania, have not, at least as yet, revealed any palace centres or Linear B, though there were some notable sites, such as Delphi, Krisa, and Teikhos Dymaion (e.g. Arena 2015; Pullen and Tartaron 2007).¹ These regions were no less ‘Mycenaean’ than the palatial areas. The Aegean islands too have revealed no palaces, though sites such as Kanakia on Salamis were important in the thirteenth century (Middleton 2010, 6, 76–77). Rhodes is also considered an important ‘Mycenaean’ region, although no major LBA urban centre or palace is known; it has even been proposed as the centre of Ahhiyawa, an important kingdom located to the west of the Anatolian Hittites, generally now accepted to be a Mycenaean/Greek (‘Achaean’) state (Mountjoy 1998). But there was certainly a flow of culture – material and otherwise – and presumably people between palatial and the equally ‘Mycenaean’ non-palatial areas; Eder (2007) has argued for palatial control of some of these, based on the distribution of seal stones.

These areas need not be seen as backward, or as having ‘failed’ to develop, or indeed as subordinate; non-state societies have flourished throughout history alongside states (see e.g. Scott 2011). Contemporary with the Mycenaean period, the Hittites in Anatolia lived alongside the non-state Kaska peoples. These non-palatial regions too would have had their own histories and traditions, which scaffolded a different but still ‘Mycenaean’ habitus for their inhabitants.

We should imagine, then, that there were various political and social arrangements in Mycenaean Greece and the Aegean towards the end of the thirteenth century (Dickinson 2014a, 158), and that people’s views of their own domains and their expectations and place in the world, were varied and the product of a deep, dynamic, complex and entangled history.²

Finally, it should be pointed out that the term ‘Mycenaean’ strictly designates a material culture, and only secondarily the people who used that culture. Whilst convenient, referring to the people of Greece as ‘Mycenaeans’ (or as Greeks) may obscure some real diversity felt by the populations in question – we have no idea whether there was any panhellenic feeling amongst (all or some of) the bearers of Mycenaean culture, let alone across Greece, Crete and the Aegean, such as existed much later. It is perhaps most likely that if such a feeling did exist to any extent, it was most apparent amongst the palace rulers and elites, who would have shared more frequent and regular contacts and sets of royal and aristocratic values.

Not everyone in the Aegean or even mainland Greece necessarily spoke Greek – even if it was, with apparent local variations, the language of the Linear B tablets (Janko 2018). Diversity is probable, given that Eteocretan and perhaps other languages (e.g. ‘Pelasgian’) survived in the Aegean into historical times (Middleton 2018b, 119–120). The famous quote about the peoples and languages of Crete from the Odyssey may be indicative. Thus, in Crete:

… there are many men, past counting, and ninety cities. They have not all the same speech, but their tongues are mixed. There dwell Achaeans, there great-hearted native Cretans, there Cydonians, and Dorians of waving plumes, and goodly Pelasgians (Odyssey 19, 172–177).

Some have thought of a Greek-speaking Mycenaean elite and an oppressed non-Greek underclass, imagined from the notion of Greeks as a distinct group of Indo-European invaders, but there is little support for this view (Dickinson 2016; Janko 2018). In the Late Bronze Age, different dialects of Greek were presumably present, as they were later on, and probably different languages were spoken too.

But just as a common language or culture does not automatically equate with shared identity, or fellow-feeling and peaceful relations, differences and diversity need not inevitably result in conflict and fault lines. We need only look at the historical Greeks to see that a shared identity above the level of polis identity in no way reduced competition or conflict between them – and many were happy to fight on the side of ‘barbarians’ against their fellow Greeks. Whilst they may not have been unusually ‘warlike’, as they have been portrayed (in contrast to supposedly ‘peaceful Minoans’ – see Dickinson 2014b), the Mycenaeans will have regularly engaged in conflicts throughout Greece and the Aegean, and perhaps even further afield, in the Late Bronze Age, whether as state actors, other groups, or on an individual level (i.e. as mercenaries).

Destructions, endings and dates

The collapse is marked by fiery destructions at major palace sites, including in the Argolid at Mycenae, Tiryns (and Midea), Pylos in Messenia, and elsewhere in northern (Iolkos) and central (Orchomenos, Thebes and Gla) Greece, and probably Athens (a palace?), as well as at the non-palatial defended site of Teikhos Dymaion in Achaea and other smaller sites (Middleton 2010, 12–15). The palaces, built around megaron throne rooms, each with a large hearth surrounded by four columns and a throne on the right wall, were destroyed. The Linear B-style writing and by implication the administrative system used, by at least some of the palaces, was abandoned. The building of ‘cyclopean’ constructions (bridges, citadels, hydrological engineering, roads, tombs), and numerous fine arts, including seal carving and fresco painting, ceased. Many sites appear abandoned, which may indicate population decline, dispersal, and nucleation, a trend perhaps beginning some time before c. 1200 BC; this is particularly noticeable in Messenia. Funerary traditions generally began to shift slowly away, over time, from multiple burials to single burials and cremation and Mycenaean pottery eventually merged into and was superseded by the Protogeometric style in the eleventh century (Lemos 2002, 8).

The date of collapse in Greece is generally given, conveniently but almost certainly misleadingly, as a point date – c. 1200 BC, at the end of the LH IIIB2 ceramic phase (Bennet 2013, 252–254; Deger-Jalkotzy 2008; Dickinson 2010; Drews 1993; Knapp and Manning 2016; Middleton 2010). Manning (2007) has discussed the difficulty of providing a secure absolute dating for the end of the LBA, but suggests that:

a date range ca. 1200 BCE can still be used as a suitable ‘textbook’ round number approximation, so long as we are mindful that the relevant time period might in fact have been a few decades earlier or later (and need not have been contemporary across the relevant cultures/areas), and that the processes involved covered periods of time rather than point events.

This last comment is certainly important and is generally agreed. Jung (2003, 254), for example, has also pointed out that ‘to assume that all Mycenaean palaces of the Aegean and even other sites in remote regions … were destroyed within a short period of time, i.e. more or less contemporarily … is surely an oversimplification’; Popham (1994, 281) too suggested a period of twenty-five years or more.³

If we consider ‘the collapse’ as a process that took some time, punctuated by historically specific events, rather than a point, it remains unclear when we should place its beginnings and endings. It has often been suggested that the later thirteenth century was a period of crisis, but this seems largely based on hindsight and a ‘business as usual’ scenario for this period is also supported by the evidence (Maran 2009; Middleton 2017). The construction of wells in citadels and the continued construction of cyclopean defences at Mycenae, Tiryns and Athens were not rushed and need not have been carried out in response to imminent specific threats rather than being continued strategies of display and simple improvements to sites by rulers. The ‘crisis’ interpretations of Linear B texts as evidence for heightened defensive measures, such as coastal lookouts and the requisition of ‘temple bronze’ to make extra weapons, are questionable; there is no indication in terms of the supply and demand of goods recorded, or the activities being carried out, that suggest anything out of the ordinary. Surely any major or unusual threat would have resulted in the creation of some more obviously military records at the time (see comment by Shelmerdine 1999b, 405). Based on his work at the citadel of Tiryns, which has involved close examination of the architectural history, and a consideration of the other evidence more generally, Maran (2009, 255) doubts ‘that on the eve of the catastrophe the political dignitaries felt they were living under the shadow of a crisis’. We should be wary of giving a single clear characterisation of the ‘mood’ of the mainland, or the Aegean as a whole, in the later thirteenth century.

Postpalatial Greece

Despite the very visible collapse of the Mycenaean palaces, and the abandonment of a variety of elements of high elite or royal culture (e.g. the megarons, Linear B, major building works etc.), the Mycenaean period did not end. The postpalatial Late Helladic IIIC period was a long phase of about a century and a half, some five or more generations; it cannot simply be relegated to the status of an irrelevant adjunct to the heyday of the Mycenaean palaces or as a bleak forerunner to a Dark Age. More positive views of the period in its own right are possible. Rutter (1992, 70), for example, states that ‘despite some major shifts in how people were distributed in the landscape and what was probably a substantial decline in the overall population, the Aegean world weathered the actual palatial collapse of c. 1200 BC well enough’. Maran (2016) and Thomatos (2006) have suggested something of an LH IIIC Middle resurgence. Continuing work at Tiryns and the publications of Maran especially also show interesting things happening after the collapse, with a new building, Building T, making selective use of the earlier palatial megaron plan and throne emplacement, which suggests the presence of an organising ‘authority’ (see Maran 2002; 2006; 2011; 2016). Of LH IIIC Middle Mycenae, French notes (2002, 140) that ‘it is clear that there must have been some kind of governance’. Even in Messenia, which seems the most affected by apparent population decline (Harrison and Spencer 2008, 148–149), there was activity on the site of the palace, though its precise nature is unclear (Griebel and Nelson 2008; see also Popham 1991 and Mountjoy 1997); a tholos tomb at Tragana was reused from the twelfth to tenth/ninth centuries BC and a chamber tomb built at Pisaskion, which contained a bronze bowl and LH IIIC Middle pictorial Mycenaean pottery (Eder 2006, 550–554; Lemos 2002, 194). Janko (2018) has suggested linguistic continuity of the ‘Mycenaean’ Arcado-Cypriot dialect, which he argues remained widespread until c. 1100 BC, after which there was some change and development in language and the distribution of dialects.

In non-palatial areas continuity is assured, since there were no palace states to collapse. Morgan (1999, 365–366), for example, writing about Corinthia, suggests that there is ‘no reason to assume anything other than a very gradual and peaceful transformation of community life’ and also that Corinth ‘gained in size and importance during LH IIIC’. Achaea and the Euboean Gulf regions