Skip to navigation – Site map

Solid items made to break, or breakable items made to last? The case of Minoan peak sanctuary figurines

Céline Murphy


The thousands of figurines found at peak sanctuaries across Crete have been widely understood as votive offerings. Owing to their fragmentary condition, it also has been suggested that they were broken as part of the rituals performed at these sites by being thrown into bonfires or to the ground. This hypothesis was further strengthened by the perception of some archaeologists of the figurines as poorly made. Upon a closer material examination of these artefacts, however, it appears that they were modelled and assembled with techniques ensuring their solidity. I therefore propose here that the figurines may not have been broken as immediately as is usually assumed, and that they may have been placed on display at peak sanctuaries for some time, before fragmentation took place. Keywords

Top of page

Full text


1The island of Crete is replete with small Middle Bronze Age mountain-top sites typically termed ‘peak sanctuaries.’ To date, approximately fifty such sites have been identified and all present a particular assemblage consisting of drinking, food containing, serving, cooking and low-scale storage vessels. Alongside these ceramics are always clay male and female anthropomorphic and zoomorphic figurines. It is commonly accepted that Minoan peak sanctuary ceramics – both vessels and figurines – were broken as part of the rituals believed to have been carried out at these sites. Indeed, be they scattered randomly across the sanctuaries’ precincts, clustered in some form of pattern, crammed into rocky nooks and crannies, lodged in crevices or simply strewn amongst ashy soil layers, these artefacts — with only very rare exceptions — have all been retrieved in a fragmentary state. What is more, alongside their broken condition, the somewhat simplistic appearance of most of these items has led scholars to believe that they were not of much quality, were thus not made to last or to be reused, and consequently, that they were effectively made to be smashed.

  • 1 Murphy 2016.

2I revisit these perspectives here with the argument that peak sanctuary figurines were, in fact, in most cases, well-made objects. Materially-oriented, experimental examinations of such artefacts1 revealed that they were made with techniques ensuring solidity. While I do not counter the suggestion that peak sanctuary figurines may have been intended to eventually come apart in some way or another, or, in any case, that they were certainly not designed to be stored for reuse, I argue that these objects were not of such poor quality that they had to be, or would have, broken immediately. Rather, I suggest that they underwent a period of display in between the moment of their arrival on site and the time of their fragmentation.

3In offering a fresh perspective on peak sanctuary figurines, this paper also demonstrates the benefits of studying these artefacts from a materially-orientated and experimental perspective, rather than simply relying upon their appearances. While few such items have been published, I draw information gathered from the few available publications, museum displays, and the results of my analysis of the figurines from Philioremos. I show how this approach indeed opens doors for new understandings of what the objects were initially designed for and helps dispel biases and assumptions constructed over decades. Consisting of both very ubiquitous and drastically understudied artefacts, maybe precisely because of their broken condition, peak sanctuary figurines now more than ever require close analyses of their modelling, assemblage and breakage.

Earlier views on Minoan peak sanctuary figurines

  • 2 Peatfield 1992, 67.
  • 3 Myres 1902/3.
  • 4 Dussaud 1910; Davaras 1981; Peatfield 1987, 1990, 1992; Pilali-Papasteriou 1992; Rethemiotakis 1997 (...)
  • 5 Myres 1902/3, 356.
  • 6 Supra, n. 5.
  • 7 Rutkowski 1991.

4It has been widely understood that peak sanctuary figurines, as Peatfield, puts it, were broken during the rituals in which they were used, as part of an “act of deposition.”2 This idea was, in fact, first advanced by Myres, in his excavation report on the peak sanctuary of Petsophas, but with some inconsistencies.3 This idea was then widely accepted and followed.4 The inconsistencies in Myres’ account concern the method by which the figurines were fragmented. Based on his discovery of figurines in ashy layers, which he identified as the remains of a bonfire, Myres argued that the artefacts were ritually thrown into the fire.5 He further argued that, because of the presence of “whole”6 figurines in these layers, the fragmentary figurines were broken due to exposure to heat, or due to the bonfire’s cleaning following the ritual. However, it seems very unlikely that any clay figurine would have survived a fall covering the distance from an individual’s hand to the ground. What is more, nearly 90 years after their excavation Rutkowski7 composed a catalogue of the Petsophas figurines that, interestingly, does not contain a single ‘whole’ figurine, in spite of Myres’ claim (Fig. 1). Every item is in a fragmentary condition, although some have been reassembled or reconstructed. While some are not as fragmentary as other examples, they do nevertheless have missing parts.

Fig. 1. Figurines from the peak sanctuary of Petsophas

Photo from Dimopoulou-Rethemiotaki 2005.

  • 8 Rethemiotakis 1997, 117.
  • 9 Supra, n. 8.
  • 10 Pilali-Papasteriou 1992.
  • 11 Morris 2009.
  • 12 Dussaud 1910, 59; Rutkowski 1991, passim.
  • 13 Morris 2009.
  • 14 Murphy 2016.
  • 15 Myres 1902/3, e.g., 362, 367.
  • 16 Pilali-Papasteriou 1992.
  • 17 Zeimbeki 2004.

5Despite Myres’ inconsistencies, it nevertheless has been widely accepted that peak sanctuary figurines were not meant to “outliv[e] the action of their dedication.”8 They were not items intended to last. Rethemiotakis’ reasoning is based on his perception of the manufacturing techniques of the majority of peak sanctuary figurines as poor9, which thus did not allow the objects to survive for lengthy periods of time, and which also justified their generally less attractive appearance. What is more, the very little information published about the firing conditions of these items has also led to assumptions that they were either all poorly fired, as is the case of some of the Metaxas Collection’s items.10 These perspectives are, however, first of all, as Morris notes, hugely generalising.11 Indeed, they lump the thousands of figurines into a single category without considering the existence of a range of different manufacturing techniques whose uses might have required different forms of knowledge. Second, these perspectives are visually biased. A “simple”12 appearance does not necessarily equate with poor structural quality. Morris,13 in fact, notes that anthropomorphic figurines were generally well assembled; she argues that their manufacture followed a clear structure, which is a perspective I also have identified for the Philioremos figurines.14 It is interesting that Myres himself noted that the joins made between the different body parts of the figurines were “secured”15 through the use of smoothing and pressing gestures, or by overlapping layers of clay, implying that they were solidly assembled. The observation of the use of cores,16 or pegs17 in the composition of figurines from other sites also points to the use of good manufacturing techniques. These observations consequently imply that the items were well constructed.

  • 18 Murphy 2016; see also Kyriakidis forthcoming.

6Contradicting information, therefore, results from these previous perspectives on the manufacture and the breakage of peak sanctuary figurines. Although variations do, of course, occur from site to site, hence allowing for different statements to be advanced about the nature of the manufacture techniques and breakage methods of peak sanctuary figurines, it is conspicuous that most of the cited views were either directly based on Myres’ account of the Petsophas material, or drawn from exclusively visual examinations of such artefacts. This is how some large assumptions about these artefacts have prevailed. As a result, following Rutkowski’s initiative to more attentively examine the items individually, I chose to take a closer look at — and undertook a meticulous experimental program with — the figurines from the peak sanctuary of Philioremos18 (Fig. 2). I focused especially on their material aspects, in other words, on how they were made, and how and when they might have been broken. As is outlined below, it became clear that the figurines’ techniques of manufacture, including their firing, were sound, thus casting doubt on previous sweeping assumptions.

Fig. 2. Figurine fragment from Philioremos

Photo courtesy of E. Kyriakidis

Experimental research and observations

  • 19 See Nanoglou 2015.
  • 20 See Murphy 2016.

7In order to correctly address the material quality and breakability of peak sanctuary figurines, it is necessary first to examine the way in which the artefacts were made. This is, of course, required for the nature of this study, but also because I do not believe that breakage and manufacture can be so easily separated. Bracketing the end of the figurines’ production process and the beginning of their consumption process is difficult. These separations correspond to an academic archetype, rather than clear visible milestones in the life of an object.19 Thus, it is important to study the figurines’ assemblage and breakage alongside each other.20

  • 21 See Myres 1902/3; Rutkowski 1991; Pilali-Papasteriou 1992.
  • 22 Myres 1902/3, 362.
  • 23 Rutkowski 1991, 23.

8Here I cite but one manufacturing technique employed in the production of male figurines, both for reasons of space and because this technique has been attested at other peak sanctuaries21. This technique, which was summarily described my Myres22, and later detailed by Rutkowski23, consists of rolling out two cylinders of clay. The first is used for the head and the torso. The cylinder is modelled into a sphere at one end to form the head then shaped appropriately to form the torso. The facial features are marked. The underside of the abdomen is then pressed in to form a hollow into which the lower body can fit. The second cylinder is used for the lower body. It is folded over in two. The extremities of the cylinder are brought close to each other and inserted into a clay base, thus placing the figurine in a straight vertical position. The joint is fortified by smoothing clay over it, from which the feet are formed. Some very moist clay is then placed in the hollow of the torso’s lower part and then placed onto the lower body (Fig. 3). The join is smoothed over with additional clay, and sometimes further strengthened with a strip of clay marking a kilt. The arms, which are modelled separately, are then attached to the torso through smoothing.

Fig. 3. Example of an assemblage technique used for peak sanctuary anthropomorphic male figurines

Drawing: the author

9Following experimentation with the aforementioned modelling and assemblage technique, it became clear that through its use of as few pieces as possible, while allowing for speed and ease, it undeniably demonstrated a concern for solidity. By making the head and the torso, as well as the legs, in one piece, the figurine makers decreased the chances of the head or a leg coming off. It can thus be stated that the use of this technique ensured the items’ stability and firmness. Indeed, the experimental breakage of figurines I modelled with this technique demonstrated that the figurines broke at their most vulnerable and exposed points, such as their legs, narrow midriff or protruding arms, rather than at their joins. It is, moreover, noteworthy that a third of the retrieved Philioremos fragments consist of the upper and lower body join, or of the solid join between the arm and the shoulder or the base and the foot. The assemblage techniques were therefore so efficient that the pieces did not come apart. Consequently, the figurines cannot be described as having been being poorly made. It would appear that they were made to stay in one piece for as long as possible.

  • 24 Lewis-Robinson forthcoming.
  • 25 Nodarou forthcoming.
  • 26 See Murphy 2016.

10What is more, macroscopic24 and microscopic25 analyses of the clay composition of the Philioremos fragments’ revealed that all figurines were fired, and deliberately so, and that different temperatures were employed. Figurines presenting dusty surfaces were generally fired at temperatures around 650˚C. Figurines of harder texture were generally fired at temperatures rising to 950˚C, a heat spectrum considered as preferable. Experimental engagement with ceramic firing enabled me to note that my experimental figurines, which were made with clay mixes closely replicating the pastes of the original items’,26 could easily be fired at temperatures ranging from high (around 1000°C) to low (around 650°C) without experiencing damage during, or straight after, the firing process. Moreover, they could not be snapped by hand, at least by individuals of standard strength. Of course, it is probable that the items fired at lower temperatures came apart sooner than those fired at higher temperatures, yet no evidence points to the immediate breakage of the lower fired examples. In the absence of conclusive evidence, I am inclined to believe that the differences in firing temperatures demonstrated by the Philioremos figurines result from the work of different groups of people, the use of different firing structures, the use of different forms of fuels, or the climatic conditions at the time of the firing, rather than from a deliberate production of easily damageable objects.

  • 27 See, e.g., Myres 1902/3; Rutkowski 1991; Pilali-Papasteriou 1992; Morris and Peatfield 1995; Chryss (...)
  • 28 Myres 1902/3, 362.

11Furthermore, the presence of bases and supports under the lower body fragments of the Philioremos figurines, as is the case for most other peak sanctuary figurines,27 suggests that they were intended to be stood upright. Were these objects to be immediately thrown to the ground, into fires or down crevices, bases would not have been required. Also, the fact that peak sanctuary figurines are always decorated in the round, as is noted by Myres28 in his comment about their backsides, implies that they were to be seen from all angles, and thus probably to be seen for some time. While, aesthetically, these artefacts maybe not have corresponded to the tastes of earlier scholars, it can certainly not be denied that care and thought was put into their production.

12These observations, therefore, cast doubt on the previous suggestion that peak sanctuary figurines were immediately broken during rituals because they were not designed to last. While, of course, they may have been designed to be broken — and it is probable that they were, as is discussed below — my observations point to the fact that they might also have played a role of a longer duration than is usually assumed, and that they may have, in fact, survived for longer than just for the duration of the ritual. This view, in turn, also questions when, how, and by whom the figurines were broken. But the main question that arises is whether peak sanctuary figurines were solid items made to be broken, or breakable items made to last. These points are explored in the following discussion.


13In light of the above, I suggest that peak sanctuary figurines were placed statically on a surface for some time. What their exact role during this period was, however, remains unknown. Nevertheless, I propose here that they may have been set up in a form of display, left as a group for the site visitors to see or look at while other activities took place, or to just remain on site alone after the rituals were completed. I insist upon the notion of the group, because it could be an explanation for the objects’ somewhat ‘crude’ or ‘simple’ appearances, if one is to judge them aesthetically. Indeed, given the sheer quantity of figurines found at peak sanctuaries and the sites’ generally limited sizes, it is evident that the objects would have been left standing in close proximity to each other. Intricate details would not have been visible, thus not requiring their depiction on the figurines’ surfaces, but the use of trichrome decoration would have ensured their contrast with the rocky landscape surrounding them.

  • 29 See Gosden 2004.

14Displaying objects is a way of abstracting them from the general flow of life and is a way of creating a form of fiction.29 Indeed, it is conceivable that at peak sanctuaries, which are non-domestic sites associated with ritual activities, the group display of figurines embodied precisely what peak sanctuary visitors experienced in accessing the sites. In climbing up the mountains in order to attend the rituals, visitors in a sense abstracted themselves from their daily lives. I further suggest that the figurines were reintegrated into the flow of life once they were left to decay, in other words, left to the mercy of the weather and animal and later human interference, after all visitors left the peak and reinserted themselves into their routines. The figurines were, therefore, like the visitors going back to their homes, reintegrated into relationships with their surroundings. This is when the figurines might have begun to undergo fragmentation.

  • 30 Kyriakidis forthcoming.

15Without entirely rejecting the suggestion that figurines might have been broken through human interference at certain sites, pushed into rocky cracks, or that they could have been broken at the very end of the rituals, I propose that these artefacts might also have come apart over time, after the rituals, due to exposure to the wind, to rain, and snow, to goats, and sheep crossing the open sites, and visitors stepping over them during later visits. Incidentally, it is important to note that, at Philioremos, the stratigraphy is deeply jumbled due to construction works carried out during the Late Bronze Age.30 The soil’s disturbance could indeed have also contributed to the figurines’ breakage. There is thus certainly some truth in earlier statements that peak sanctuary figurines were intended to be broken. They are today all in a fragmentary condition; not a single one remains in a complete state, as was noted above. In this light, it is most likely that they were indeed meant to break eventually. And indeed it is possible that they were eventually swept aside, or placed in between rocks or crevices. Yet, the above observations alter the dynamics of previous understandings of their breakage; it is indeed also very possible that they were not instantly broken, nor necessarily fragmented by human hands alone. Peak sanctuary figurines therefore sit on a very narrow fence. Ought they to be described as solid items made to be broken, or breakable objects made to last? They are effectively both.


16In this paper, I have demonstrated that a different approach to examinations of the manufacture and fragmentation of peak sanctuary figurines can offer quite different perspectives on their nature and the processes through which they went following their deposition. By using a closer material investigation of, and experimental engagement with, the items from Philioremos, I was able to better understand the material dynamics that led the artefacts to the condition they are in today, and consequently to question some assumptions and biases frequently associated with this category of figurines more generally. Having argued that peak sanctuary figurines are well-made objects, I suggest here that they may not have been broken as quickly as is usually assumed, and that these items may have been placed on display for some time before fragmentation took place naturally, due to weather conditions, or due to animal, and later human, interference on the sites.

Top of page


Chapman, J. 2015. “Bits and pieces. Fragmentation in Aegean Bronze Age context.” In K. THAVSMA. Contextualising the Intentional Destruction of Objects in the Bronze Age Aegean and Cyprus, edited by K. Harrell and J. Driessen, 25–48. Louvain-La-Neuve: Presses Universitaires de Louvain.

Chryssoulaki, S. 2001. “The Traostalos peak sanctuary: aspects of spatial organisation.” In Potnia: Deities and Religion in the Aegean Bronze Age, edited by R. Laffineur and R. Hägg, 57–66. Proceedings of the 8th International Aegean Conference/8e Rencontre égéenne internationale. Göteborg, Göteborg University, 12–15 April 2000. Aegaeum 22. Liège: Université de Liège.

Davaras, C. 1981. “Three new Linear A libation vessel fragments from Petsophas.” Kadmos 20: 16.

Dimopoulou-Rethemiotaki, N. 2005. Το Αρχαιολογικό Μουσείο Ηρακλείου. Athens: EFG Eurobank Ergasias S.A. and John S. Latsis Public Benefit Foundation.

Dussaud, R. 1910. Les Civilisations Préhelléniques dans le Bassin de la Mer Égée: Études de Protohistoire Orientale. Paris: Librairie Orientaliste Paul Geuthner.

Gosden, C. 2004. “Making and display: our aesthetic appreciation of things and objects.” In Substance, Memory, Display: Archaeology and Art, edited by C. Renfrew, C. Gosden and E. DeMarrais, 35–45. Cambridge: McDonald Institute Monographs.

Kyriakidis, E. forthcoming. The Peak Sanctuary of Philioremos: A Minoan Open Air Ritual Site.

Lewis-Robinson, H. Forthcoming. “The Philioremos pottery fabrics.” In The Peak Sanctuary of Philioremos: A Minoan Open Air Ritual Site, edited by E. Kyriakidis.

Morris, C. 2009. “Configuring the individual: bodies of figurines in Minoan Crete.” In Archaeologies of Cult. Essays on Ritual and Cult in Crete in Honor of Geraldine C. Gesell, edited by A.-L. D’Agata and A. Van de Moortel, 179–187. Hesperia Supplement 42. Princeton: The American School of Classical Studies.

Morris, C. and Peatfield, A. 1995. “The pottery from the peak sanctuary of Atsipadhes Korakias.” In Proceedings of the 7th Cretological Congress, Rethymnon 1991, 643–648. Rethymnon: Society of History and Folklore.

Murphy, C. 2016. “Reconciling Materials, Artefacts and Images. An Examination of the Material Transformations Undergone by the Philioremos Anthropomorphic Figurines.” PhD thesis: University of Kent.

Myres, J. 1902/3. “The sanctuary site of Petsofa.” ABSA 9: 356–387.

Nanoglou, S. 2015. “Situated intentions: providing a framework for the destruction of objects in Aegean prehistory.” In THAVSMA. Contextualising the Intentional Destruction of Objects in the Bronze Age Aegean and Cyprus, edited by K. Harrell and J. Driessen, 49–60. Louvain-La-Neuve: Presses Universitaires de Louvain.

Nodarou, E. Forthcoming. “Petrographic analysis of selected samples from the peak sanctuary at Philioremos-Gonies.” In The Peak Sanctuary of Philioremos: A Minoan Open Air Ritual Site, edited by E. Kyriakidis.

Peatfield, A. 1987. “Palace and peak: the political and religious relationship between palaces and peak sanctuaries.” In The Function of the Minoan Palaces, edited by R. Hägg and N. Marinatos, 89–93. Proceedings of the Fourth International Symposium at the Swedish Institute in Athens, 10–16 June, 1984. Uppsala: Paul Åströms förlag

Peatfield, A. 1990. “Minoan peak sanctuaries: history and society.” OpAth 18: 117–"132.

Peatfield, A. 1992. “Rural ritual in Bronze Age Crete: the peak sanctuary at Atsipadhes.” CAJ 2-1: 59–87.

Pilali-Papasteriou, A. 1992. Μινωικά Πήλινα Ανθρωπόμορφα Ειδώλια της Συλλογής Μεταξά. Συμβολή στη Μελέτη της Μεσομινωικής Πηλοπλαστικής. Θεσσαλονίκη: Θεσσαλονίκη Εκδόσεις Βάνιας.

Rethemiotakis, G. 1997. “Minoan clay figures and figurines: manufacturing techniques.” In TEXNI: Craftsmen, Craftswomen and Craftsmanship in the Aegean Bronze Age, edited by R. Laffineur and P. Betancourt, 117–121. Proceedings of the 6th International Aegean Conference. Aegaeum 16. Liège: Annales d’Archéologie de l’Université de Liège.

Rethemiotakis, G. 2001. Minoan Clay Figures and Figurines: From the Neopalatial to the Subminoan Period. Athens: Archaeological Society at Athens.

Rutkowski, B. 1991. Petsophas: A Cretan Peak Sanctuary. Warsaw: Art and Archaeology.

Sphakianakis, D. 2012. “The Vrysinas Ephebe: the lower torso of a clay figurine in contrapposto.” In Philistor: Studies in Honor of Costis Davaras, edited by E. Mantzourani and P. Betancourt, 201–212. Philadelphia: INSTAP Academic Press.

Zeimbeki, M. 2004. The organisation of votive production and distribution in the peak sanctuaries of state society Crete: a perspective offered by the Juktas clay animal figures.” In Knossos: Palace, City, State, edited by G. Cadogan, E. Hatzaki, and A. Vasilakis, 351–361. London: The British School at Athens.

Top of page


1 Murphy 2016.

2 Peatfield 1992, 67.

3 Myres 1902/3.

4 Dussaud 1910; Davaras 1981; Peatfield 1987, 1990, 1992; Pilali-Papasteriou 1992; Rethemiotakis 1997; Chapman 2015.

5 Myres 1902/3, 356.

6 Supra, n. 5.

7 Rutkowski 1991.

8 Rethemiotakis 1997, 117.

9 Supra, n. 8.

10 Pilali-Papasteriou 1992.

11 Morris 2009.

12 Dussaud 1910, 59; Rutkowski 1991, passim.

13 Morris 2009.

14 Murphy 2016.

15 Myres 1902/3, e.g., 362, 367.

16 Pilali-Papasteriou 1992.

17 Zeimbeki 2004.

18 Murphy 2016; see also Kyriakidis forthcoming.

19 See Nanoglou 2015.

20 See Murphy 2016.

21 See Myres 1902/3; Rutkowski 1991; Pilali-Papasteriou 1992.

22 Myres 1902/3, 362.

23 Rutkowski 1991, 23.

24 Lewis-Robinson forthcoming.

25 Nodarou forthcoming.

26 See Murphy 2016.

27 See, e.g., Myres 1902/3; Rutkowski 1991; Pilali-Papasteriou 1992; Morris and Peatfield 1995; Chryssoulaki 2001; Rethemiotakis 2001; Sphakianakis 2012.

28 Myres 1902/3, 362.

29 See Gosden 2004.

30 Kyriakidis forthcoming.

Top of page

List of illustrations

Credits Photo from Dimopoulou-Rethemiotaki 2005.
File image/jpeg, 120k
Caption Photo courtesy of E. Kyriakidis
File image/jpeg, 112k
Credits Drawing: the author
File image/jpeg, 40k
Top of page


Electronic reference

Céline Murphy, « Solid items made to break, or breakable items made to last? The case of Minoan peak sanctuary figurines », Les Carnets de l’ACoSt [Online], 17 | 2018, Online since 10 April 2018, connection on 18 January 2019. URL : ; DOI : 10.4000/acost.1089

Top of page

About the author

Céline Murphy

Post-doctoral researcher
The Heritage Management Organization, Athens

By this author

Top of page


Licence Creative Commons
Les Carnets de l'ACoSt est mis à disposition selon les termes de la licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Top of page
  • Logo Association for Coroplastic Studies
  • OpenEdition Journals