Swords and Swordsmanship In the Aegean Bronze Age

AMERICAN JOURNAL OF ARCHAEOLOGY THE JOURNAL OF THE ARCHAEOLOGICAL INSTITUTE OF AMERICA This article is © The Archaeological Institute of America and was originally published in AJA 114(3):403–28. This reprint is supplied to the primary author for personal, non-commercial use only, following the terms stipulated by the AJA Editor-in-Chief. The definitive electronic version of the article can be found at http://www.atypon-link.com/AIA/doi/abs/10.3764/ aja.114.3.403. Volume 114 ●฀฀No. 3 July 2010 ARCHAEOLOGICAL INSTITUTE OF AMERICA 2010 OFFICERS C. Brian Rose, President Elizabeth Bartman, First Vice President Sebastian Heath, Vice President for Professional Responsibilities Jenifer Neils, Vice President for Publications Mat Saunders, Vice President for Education and Outreach Alexandra Cleworth, Vice President for Societies Brian J. Heidtke, Treasurer Teresa M. Keller, Executive Director HONORARY PRESIDENTS Robert H. Dyson, Jr., James R. Wiseman, Martha Sharp Joukowsky, James Russell, Stephen L. Dyson, Nancy C. Wilkie GOVERNING BOARD Susan Alcock Michael Ambler Carla Antonacchio Cathleen Asch Robert Atwater Barbara Barletta David R. Boochever Eugene Borza Laura Childs Lawrence Coben Mitchell Eitel William Fitzhugh Harrison Ford Peter Herdrich Lillian Joyce William A. Lindsay Donald W. Morrison Robert E. Murowchick Helen Nagy Eleanor Powers Lynn Quigley Ann Santen William Saturno Glenn Schwartz Ava Seave David C. Seigle Charles Stanish Charles Steinmetz Douglas A. Tilden John J. Yarmick TRUSTEES EMERITI Norma Kershaw Charles S. LaFollette PAST PRESIDENT Jane C. Waldbaum Mitchell Eitel, Sullivan & Cromwell, General Counsel MEMBERSHIP IN THE ARCHAEOLOGICAL INSTITUTE OF AMERICA AND SUBSCRIPTION TO THE AMERICAN JOURNAL OF ARCHAEOLOGY The American Journal of Archaeology is published by the Archaeological Institute of America in January, April, July, and October. An annual print or electronic subscription is $80 (international, $110); the institutional rate is $280 (international, $310). A combination (print and electronic) subscription is an additional $10 (individual), $30 (institution). The AJA is also available with membership in the Institute. For more informaion, contact membership@ aia.bu.edu. All communication regarding membership, subscriptions, and back issues should be directed to membership@aia.bu.edu or addressed to Membership Department, Archaeological Institute of America, located at Boston University, 656 Beacon Street, Boston, Massachusetts 02215-2006, tel. 617-353-9361, fax 617-353-6550. Swords and Swordsmanship in the Aegean Bronze Age BARRY MOLLOY Weapons, by definition, have a primary purpose of fulfilling the needs of warriors in combat. By charting changes in the design and functional attributes of swords in particular, this paper explores the changing character and social influence of combat and warfare. In turn, swords are shown to be more than fossils of war in itself, as they are also products of the social mechanisms that afforded violence a cultural value. Experimental archaeology, along with taxonomic, usewear, iconographic, mortuary, and literary evidence, are all brought together—in an approach I call combat archaeology2—to illuminate dynamics of war and society in the Aegean Bronze Age. The primary material has been well published, and a brief history of research and introduction of sword typologies follows. Abstract Warfare and combat are often considered to have played central roles in the characterization of elite identities and the social evolution of Aegean Bronze Age polities of Crete and the Greek mainland. Iconography and mortuary practice provide insights into how warrior identity and violence were materially celebrated. To understand better the systemic and reflexive impact of warfare on social structures, consideration of the technical aspects of combat practice is important. The actual weapons, particularly swords and spears, that warriors used to participate in wars survive in large quantities. Practical functions of these weapons can be determined with the help of metric and use-wear analysis and experimental archaeology; these efforts provide insights into the manner in which combat was conducted, a picture that is enhanced by iconographic, literary, and mortuary evidence. By focusing on swords and swordsmanship, this article seeks to use the material culture of war to illuminate the dynamic relationship between war and society in the Aegean Bronze Age.* aegean swords Aegean prehistory was born in Schliemann’s hunt for the heroes of Troy. At the site of Mycenae, in Greece, he unearthed a wealth of archaeological evidence for warfare and violence that included weaponry, fortifications, human remains, and artifacts bearing depictions of war and violence.1 While Schliemann was seeking evidence for one specific war, we now recognize that prehistoric wars in general have left few archaeological traces. Our evidence, primarily weapons and images of violence, directs research toward explorations of interpersonal combat and the social reception of violence. In 1930, Karo studied the early forms of Aegean swords from Grave Circle A at Mycenae, dubbing them Types A and B.3 Some decades later, a growing interest in the study of prehistoric swords from Europe4 led to the first systematic study of Aegean sword types by Sandars, who established the classification system still widely used today (fig. 1, table 1).5 Driessen and Macdonald made minor modifications to this system while retaining the broad groups of Type A–G swords.6 The intrusive Central European Naue ii sword tradition was studied by Catling;7 this work completed the sequence of Aegean bronze swords. Kilian-Dirlmeier8 more recently reset many of these typological categories (table 2) in a detailed catalogue of swords from the Aegean and southern Balkans (excluding the Peloponnese).9 * I wish to thank Alan Peatfield for his many years of support in this research. I am grateful to Editor-in-Chief Naomi J. Norman and the two anonymous reviewers for the AJA for their many helpful comments. I also wish to thank Neil Burridge and all at Sword Forum International (http://www.swordforum.com) for advice over the years. The pig carcasses were kindly provided by the Ashtown Food Centre and had been slaughtered as part of their abattoire work program. This research was conducted under funding from the Irish Research Council for Humanities and Social Sciences. 1 Schliemann 1878. 2 Molloy 2006; see http://www.combat-archaeology.org. 3 Karo 1930. 4 Cowen 1951, 1955, 1966; Catling 1956, 1961; Trump 1962; Eogan 1965. 5 Sandars 1961, 1963. 6 Driessen and Macdonald 1984. 7 Catling 1956, 1961. 8 Kilian-Dirlmeier 1993. 9 The nomenclature in this study follows the older system of sword typologies. While the results of Kilian-Dirlmeier’s (1993) comprehensive study are taken fully into account, Sandars’ (1961, 1963) broader, sequential series is considered more accessible to the general reader and so is used in this paper for the sake of clarity. Correlations between types in both systems are illustrated in table 1. introduction 403 American Journal of Archaeology 114 (2010) 403–28 © 2010 Archaeological Institute of America 404 BARRY MOLLOY [AJA 114 lochori in Crete, and Mycenae and around Pylos on mainland Greece (fig. 3).11 These swords are typified by long, tapering blades supported by a pointed midrib and were once attached by rivets to an organic handle that had decayed prior to recovery. The sheet gold that once covered an organic core on a sword from Shaft Grave Delta at Mycenae indicates that, when hilted, some had squared shoulders similar to the Type B weapons,12 whereas the gold hilt furniture from Shaft Grave V suggests that some were akin to Type C.13 The examples about 1,000 mm in length are the most visually striking, though the 20% of examples that measure 550–700 mm remind us of the considerable variability in lengths, weights, and balances of these swords (fig. 4).14 Fig. 1. Major categories of Aegean swords: 1, Type A; 2, Type B; 3, single-edged; 4, Type C; 5, Type Di; 6, Type Dii; 7, Type Fii; 8, Type Gi; 9, Type Gii; 10, Type Naue ii. I include here only a brief overview of forms and chronologies (fig. 2) as a foundation for the discussion of sword functionality that follows. Type A Sword The Type A sword appears virtually fully formed in Middle Minoan (MM)/Middle Helladic (MH) II, with a blade typically 700–1,000 mm in length. There are very few examples that are earlier than MM/MH III,10 when numerous examples are known from Arka- 10 Examples from Thebes, Aegina, and Malia are likely to be of this date (Chapouthier 1938; Kasimi-Soutou 1980, 92; Walter 1981, 182; Kilian-Dirlmeier 1993, 26–8). 11 Karo 1930 (Mycenae); Marinatos 1935 (Crete); Blegen et al. 1973 (Pylos). At least 92 examples are known. The fragmentary state of many pieces is the foremost problem in ascertaining exact quantities. 12 Athens, National Archaeological Museum, inv. no. 8710, from Grave Circle B, Grave Delta (Mylonas 1973, 85–6). 13 Karo 1930; Sandars 1963, 117. 14 Karo 1930; Blegen et al. 1973; Tripathi 1988; Fortenberry 1990. 15 Fortenberry 1990, 185–86. 16 Karo 1930; Fortenberry 1990. © 2010 Archaeological Institute of America Single-Edged Sword Contemporary with the Type A swords, a robust, single-edged type developed, sometimes called a Schlachtmesser.15 Of these, 26 are known from the Greek mainland, primarily from the shaft graves at Mycenae.16 They could reach lengths of 700 mm (typically 350–550 mm) and weigh as much as 1 kg.17 The earliest type (Type I) has a curved, single-edged blade terminating in a sharp point. The handle consists of a full tang with organic hilt plates set into deep flanges that are hammered closed to secure the plates in position. A second variant (Type II) has a solid cast handle and a loop at the terminal end of the handle. These weapons have neither predecessor nor successor, in typological and functional terms. Type B Sword Most surviving examples of Type B weapons18 come from the shaft graves at Mycenae and are contemporary with Type A swords. They are characterized primarily by a tongue of metal, or tang, that projects from the butt of the blade to form the core of the handle. This ensured that all Type B weapons had 17 Karo 1930; Molloy 2006. The dividing line between daggers and swords has been variously set by scholars in the past (Tripathi 1988, 72–3; Fortenberry 1990, 143), though the generic term “weapon” is sometimes preferable for general discussion, as other components of blade design (e.g., wdth., thickness, balance) can affect preferred manner of use and push some weapons to either side of a dividing line based on length. In broad terms, weapons shorter than 300 mm from point to pommel are called daggers, those between 300 and 600 mm are called shortswords, and those longer than 600 mm are longswords. The term “rapier” is avoided entirely because of its lack of applicability (Molloy 2006, 2007). 18 Cross-Section Modal Lgth. Range (mm) Main Sites A MH II– LH IIIA1 double-edged; short tang with 1–3 rivets and 2 in shoulders flat blade with midrib 550–1,000 Arkalochori, Zakros, Mycenae, Pylos B MH III– LH IIIA1 double-edged; long, broad tang with 1–3 rivets and 2–3 in shoulders graduations between flat blade with midrib to pointed oval 350–830 Mycenae, Orchomenos, Dendra Singleedged MH III– LH IIIA1 single-edged; flanged or solid cast handle extended triangular 350–700 Pylos, Mycenae C LH I– IIIB double-edged; full-flanged tang with 0–3 rivets and 0–2 in shoulders; projecting quillons on shoulders and flanged ricasso flat blade with midrib 500–900 Athens, Dendra, Mycenae, Knossos, Chania Di LH II– IIIB double-edged; full-flanged tang with 0–3 rivets and 2 in shoulders; projecting rounded knobs and flanged ricasso typically flat with midrib; later variants low-profile, flat midrib to pointed-oval cross-section 350–650 Dendra, Mycenae, Knossos, Chania Dii LH IIIA2–C double-edged; full-flanged tang with T-shaped pommel extension; 0–3 rivets in tang and 0–2 in shoulders; projecting rounded knobs and flanged ricasso low-profile; flat midrib to pointed-oval cross-section 350–450 wide distribution F LH IIIA2–C double-edged; full-flanged tang (typically) with T-shaped pommel extension; 2–4 rivets in tang and 0–2 in squared shoulders pointed-oval cross-section 350–500 wide distribution, esp. central and east Crete, Mycenae, Epirus Gi LH IIIA2–B double-edged; full-flanged tang (sometimes) with T-shaped pommel extension; 0–3 rivets in tang and 0–2 in shoulders; projecting quillons on shoulders and flanged ricasso flat blade with narrow midrib 400–600; 800–1,005 Mycenae, Syme, Knossos Gii LH IIIC double-edged; full-flanged tang with T-shaped pommel extension; 1–4 rivets in tang with 0–2 in shoulders; projecting quillons on shoulders and flanged ricasso pointed-oval cross-section 350–575 wide distribution Naue ii LH IIIB2–C double-edged; full-flanged tang (sometimes) with pommel spur, 2–5 rivets in tang and 2–4 in shoulders pointed-oval cross-section 500–850 wide distribution 405 Description SWORDS AND SWORDSMANSHIP IN THE AEGEAN BRONZE AGE Maximum Date Range Type 2010] © 2010 Archaeological Institute of America Table 1. Main Aegean Sword Types. 406 BARRY MOLLOY [AJA 114 Table 2. Cross-Referencing of Aegean Bronze Age Sword Types. Sandarsa Kilian-Dirlmeierb Type A Typ A (Variante 1–3) Type B Typ B Type Ci Hörnerschwert Typ 1a, 1b Type Cii Hörnerschwert Typ 1a, 1b Type Ciii Hörnerschwert Typ 3b Type Di Kreuzschwerter Typ 1 (Variante 1a–g) Type Dii Kreuzschwerter Typ 2 (Variante 2a–b) Type Fi Typ F 1 Type Fii Typ F 2 (Variante 2a–c) Type Gi Hörnerschwert Typ 2a Type Gii Hörnerschwert Typ 2a, 2b a Sanders 1961, 1963; amended by Driessen and MacDonald 1984 b Kilian-Dirlmeier 1993 squared shoulders, though none of their organic handles survives. Type B weapons have great variability in length,19 with the majority being long daggers/ shortswords ranging from 300 to 550 mm; only eight pieces exceed this. Type C Sword Daggers from Shaft Grave VI20 and Shaft Grave Delta21 at Mycenae incorporate the integral/full tang of the Type B with the “horned” hilt of some Type A swords. This hilt form came to characterize the Type C swords22 of Late Helladic (LH)/Late Minoan (LM) II–IIIA, marked also by flanges running along the tang, shoulders, quillons, and ricasso (fig. 5). These significantly strengthen the hilt-blade attachment beyond that of their predecessors.23 Blades were also narrower, thinner, lighter, and shorter than typical sword length varieties of Type A,24 though they retained the 19 Twenty-six examples are known overall (Karo 1930; Kilian-Dirlmeier 1993). 20 Athens, National Archaeological Museum, inv. nos. 904, 905 (Karo 1930). 21 Athens, National Archaeological Museum, inv. no. 8712 (Mylonas 1973). 22 In typological terms, Type C can be broken down into at least three or more subtypes that have more to do with rivet layouts and subtleties of the hilt configuration than with functional martial attributes. Therefore, for this broader discussion, only the main categories are significant. 23 There are three major variations of this type (Ci, Cii, Ciii), © 2010 Archaeological Institute of America Fig. 2. Schematic chronological range of Aegean Bronze Age swords. distinctive tapering profile and midrib of their predecessors. As with the contemporary Type D, these but for the purposes of the functional analysis of this paper, all are treated together. There is significant variation in blade forms and lengths within, as well as among, each group, so discussion is kept more general for brevity and clarity. 24 Modal range is 600–700 mm, with minimum length of dagger proportions, though rare examples occur between 80 and 104 mm. They were also usually light weapons with a typical weight range of just 300–500 g (Molloy 2006). Weights for all swords are given in relation to their current state of preservation and are irrespective of hilt plate and pommel weights, which could add up to 200 g more (Molloy 2006). 2010] SWORDS AND SWORDSMANSHIP IN THE AEGEAN BRONZE AGE 407 Fig. 3. Major sites with concentrations of bronze swords. are found widely throughout the Aegean,25 though significant concentrations occur in cemeteries in the Argolid26 and Crete.27 Type Di and Dii Swords Type Di swords differed from the contemporary Type C primarily in terms of hilt design.28 They substitute the backswept quillons with lower-profile, rounded knobs. Blade forms were similar to Type C, though a greater proportion of shorter examples exist.29 Generally, there is an increasing tendency later in 25 At least 39 pieces are known (Fortenberry 1990; KilianDirlmeier 1993). 26 Karo 1930; Persson 1931; Mylonas 1973. 27 Evans 1906; Hood and De Jong 1952; Hood 1956; Popham and Catling 1974; Vlasaki 2005. 28 Thirty-eight pieces are known (Fortenberry 1990; KilianDirlmeier 1993). These have similar proportions and thus a similar weight range to the Type C swords, though they start at 250 g (Molloy 2006). They were also relatively thin, measuring just 3.2 mm thick halfway down the blade (from the hilt). 29 Fortenberry 1990; Kilian-Dirlmeier 1993. © 2010 Archaeological Institute of America the series toward shorter, more robustly proportioned weapons, following the same multivariate tradition of Type B weapons. The Type Dii30 diverges from Type Di in two important ways: it incorporates a T-shaped plate at the base of the handle for attaching a pommel, and the midrib profile is significantly reduced and sometimes abandoned entirely in favor of a pointed-oval blade cross-section.31 The new cross-section design allowed for deeper cuts to inflict more dangerous injuries than was possible with midribbed blades, as discussed be- 30 Fourteen pieces are known (Fortenberry 1990; KilianDirlmeier 1993). These typically weighed 250–400 g. 31 This blade section is also found on some later Type Di swords. One example comes from Hagios Sylas in Crete (Heraklion, Heraklion Archaeological Museum, inv. no. 4467 [Ioannidou-Karetsou 1985; Kilian-Dirlmeier 1993, no. 136]). Another example comes from Mavro Spilio, Chamber Tomb XVIII (Heraklion, Heraklion Archaeological Museum, inv. no. 2141 [Forsdyke 1926–1927; Kilian-Dirlmeier 1993, no. 142]). 408 BARRY MOLLOY Fig. 4. Type A swords from Shaft Grave V, Grave Circle A at Mycenae, showing size variation in this type. Athens, National Archaeological Museum, inv. nos. 767 (left), 751 (right). [AJA 114 Fig. 5. Components of bronze swords. Type Fi and Fii Swords Type Fi weapons are rare and virtually identical to Type Fii swords in functional terms. The earliest example of Type Fii belongs to LM IIIA233 and comes from Tomb 95 at Knossos.34 These became the Aegean sword par excellence of LH IIIB and IIIC.35 The pointed-oval cross-section of the Type E double-edged knives36 and some Type Dii weapons is adopted and typifies Type Fii weapons.37 The blade edges are parallel, and the point has a broader angle than preceding sword types. These are stoutly proportioned with squared shoulders 32 They were typically 450–500 mm, with only one unprovenanced example (Oxford, Ashmolean Museum, inv. no. 1967.1273) measuring 700 mm (Kilian-Dirlmeier 1993, no. 148; Molloy 2006, no. 244). 33 Heraklion, Heraklion Archaeological Museum, inv. no. 1103 (Evans 1906; see also Kilian-Dirlmeier 1993, no. 195). 34 This sword had been heavily resharpened, as evidenced by the slightly convex blade edges, indicating that it was likely to have been used, possibly in combat. 35 At least 40 pieces are known (Fortenberry 1990; Kilian- Dirlmeier 1993). Typically, these were ca. 3.5 mm thick and weighed 300–400 g (e.g., from Dodona: London, British Museum, inv. no. 1975.7–12.67 [Kilian-Dirlmeier 1993, no. 208; Molloy 2006, no. 250]). The larger Cretan varieties could weigh up to 500 g and were 4–5 mm thick (e.g., from “Siteia”: Oxford, Ashmolean Museum, inv. no. 1966.542 [KilianDirlmeier 1993, 203; Molloy 2006, no. 251]). 36 Sandars 1963. 37 A shortsword from Shaft Grave VI (Athens, National Archaeological Museum, inv. no. 906 [Karo 1930]) is the earli- low. These were usually shorter than typical Type Di and C weapons.32 © 2010 Archaeological Institute of America 2010] SWORDS AND SWORDSMANSHIP IN THE AEGEAN BRONZE AGE and a T-shaped pommel plate. In LH IIIB, Type Fii swords are usually 400–450 mm in length, though in LH IIIC, examples from modern Epirus and Albania, as well as Crete, occasionally exceed 500 mm.38 Type Gi and Gii Swords The Type Gi sword39 is the successor to the Type C, and the two forms dovetail typologically.40 The blades retain the midrib, though it is considerably narrower, and the blades themselves become wider.41 They diverge from the Type C series in that they possess horns that are at virtually 90° to the hilt and have characteristically deeper cutting edges.42 Many examples have T-shaped pommel extensions,43 though not exclusively so. They die out early in LH IIIB and only lend superficial traits to Type Gii. Type Gii swords appear primarily in LH IIIC contexts. This type is rare,44 and the variety of blade proportions is such that there is little to compare among examples.45 In general, they are defined by a hilt (functionally) similar to the Type Fii weapons, though thin quillons hook out from the shoulders and curve toward the point of the blade. The blades are thin and pointed-oval in cross-section, sometimes with multiple low-profile ribs. A sword from Këlcyrë, in Albania,46 has a hilt typical of a Type Fii, shoulders of a Type Dii, and a blade typical of a Type Gii, demonstrating considerable fluidity of design by LH IIIC and underscoring the love of variability that characterized the Aegean sword tradition. 409 parallel cutting edges and a pointed-oval cross-section, as well as a fully flanged tang. They were typically longer than contemporary Aegean-type swords, being 580–680 mm in length,48 though their blades were narrower. In most respects, they had similar design components to local swords, being closer to a Type Fii than Type Fii was to a Type C, for example. This was one of the more common forms in use in LH IIIC, and apart from an exclusive concentration of Naue ii swords in Achaea,49 they have a very similar distribution pattern and frequency to Aegean Type Fii swords. Central and east Crete possessed a very high density of both Naue ii and Type Fii swords.50 sources The three categories of evidence best suited to an archaeological analysis of combat are iconography, mortuary/osteological remains, and weaponry. A brief analysis of the first two categories comes before a discussion of weaponry to provide the framework for understanding its roles in society. Naue ii Sword In the 13th century, the Naue ii developed between the Alpine region and the Balkans and was adopted in the Aegean by LH IIIB2.47 These swords had two Iconography Iconographic scenes were not intended to constitute a deliberate or cohesive narrative on war. Images occur on stone, metal and pottery vessels, sealstones, inlaid daggers, frescoes, stone stelae, and larnakes— objects with very different contexts of consumption and with equally marked chronological heterogeneity.51 While such scenes indicate that martial symbolism played a role in the display and construction of warrior identity, it cannot reveal how such identity was manifested in daily life or even during wars. In this sense, iconography is informative about perceptions or idealizations of war in Aegean societies; thus, est example of a sword with a pointed-oval cross-section. The midrib on several Type B shortswords blends out to the blade edges, though there is poor continuity between these and the weapons of similar cross-section in LM IIIA. 38 Kilian-Dirlmeier 1993, 76–86. 39 Driessen and Macdonald 1984. 40 Thirteen pieces are known (Driessen and Macdonald 1984; Fortenberry 1990; Kilian-Dirlmeier 1993). 41 Daggers and swords exist in this form. They fall into a group measuring ca. 40–60 mm and a longer group measuring ca. 80–105 mm. 42 Kilian-Dirlmeier (1993) conflates later examples of Sandars’ Type C and Gi swords into a single Hörnerschwerter 2a group. This lead is followed in this article, and all swords with forged closed flanges on quillons that are at ca. 90° to the hilt are moved from Type C to Type Gi. Differentiation is not made between examples with pommel spurs (Hörnerschwerter 2a1) and those with T-shaped pommel extensions (Hörnerschwerter 2a2), as focus is placed on weapon functionality. The reader is referred to Kilian-Dirlmeier (1993) for all typological discussions. 43 In Driessen and Macdonald (1984), the T-shaped pommel extension is a characteristic of this type, though this group is here amended to follow the observations made by Kilian-Dirlmeier (1993) in her Hörnerschwerter 2a category. 44 Five examples are currently known (Fortenberry 1990; Kilian-Dirlmeier 1993; Eder 1999). 45 They range in length from 400 to 600 mm, following a broadly similar pattern to Type Fii and Dii weapons. The example from “Ithaca” (London, British Museum, inv. no. 1938.1–10 342/2753) is 404 mm long, 4 mm thick (across the raised ribs), and weighs 275 g (Kilian-Dirlmeier 1993, no. 102; Molloy 2006, no. 249). 46 Bodinaku 1988, 34–8; Kilian-Dirlmeier 1993, no. 147. 47 Catling 1956; Harding 1984, 163–65; 1995, 20–3; KilianDirlmeier 1993, 94. 48 This is the modal length range (Fortenberry 1990; KilianDirlmeier 1993; Molloy 2006). 49 Papadopoulos 1978, 1999; Papadopoulos and KontorliPapadopoulou 1984; Papazoglou-Manioudaki 1994. 50 Kilian-Dirlmeier 1993. 51 Hiller 1999. © 2010 Archaeological Institute of America 410 BARRY MOLLOY there are limits to what it can tell us about swordsmanship or combat.52 In a study of combat, iconography is most helpful in revealing the different combinations of weapons used and providing information about the form and dimensions of lost organic components, such as shields and spear shafts. Iconographic scenes relating to warfare can be loosely categorized as images of group combat, images of single combat, and images with a general martial theme.53 From the first group, the image on the “Battle Krater” from Mycenae depicts two opposing lines of spearmen and is the single surviving scene of groups of warriors engaging one another in combat.54 The Lion Hunt dagger55 from the same grave depicts a similarly accoutered line of warriors facing a pack of lions. The similarity of the warriors in these scenes indicates that this may be a stock image of a “line of battle” used on different media. There are three primary observations about these group combats. First, no swords are depicted. Second, two different forms of large shields are used in the same line of battle, and spears are the main offensive weapon. Third, an archer is working amid the line of spearmen, combining missile with shock combat weapons.56 The “Miniature Fresco” from Akrotiri57 depicts a similar line of spearmen, though in this case all shields are of the same form, and swords are carried at the waist. Irrespective of artistic conventions, it appears likely that spears and shields were the preferred weapons at the outset of a group combat engagement and that cooperative/close-order formations were used. The use of swords was thus contingent on combat developments, such as the loss of a spear or a break in a line of battle, leading to a rout or even single combats before or during a wider engagement. The Silver Siege rhyton58 depicts archers and slingers working alongside spearmen, again combining elite warriors with spear and shield and missile troops, in this case fighting naked. When swords are depicted in use,59 it is exclusively on seal images between 1700 and 1400 B.C.E. on the mainland and Crete.60 The tiny size of seals (typically 52 Papadopoulos (2006) has described the methodological pitfalls faced when using these disparate sources, and they are considered accordingly in this research. 53 This includes warriors marching, warriors displaying weapons in noncombat contexts, warriors in ambiguous settings, nonwarriors bearing weapons, agonistic (nonlethal) contests such as boxing, and weapons depicted in isolation (including shields and helmets). 54 Sakelleriou 1974. 55 Karo 1930; Dickinson 1997. 56 Shock combat is essentially infantry combat using handto-hand weaponry. 57 Warren 1979; Morgan 1988. © 2010 Archaeological Institute of America [AJA 114 10–20 mm wdth.) rarely allows more than two individuals to be depicted, so combat scenes selected by the artists usually involve two opposing warriors. These images are not to be seen as excerpts from a combat manual,61 as they are vehicles intended to portray symbolic and historically contingent meanings. LM/ LH I–II seals depicting interpersonal violence come from Knossos, Hagia Triada, Zakros, and unspecified locations in Crete, Mycenae, Athens, and Pylos.62 Of these, eight depict opposed pairs of combatants and one depicts four people, two actively engaged in fighting each other. The three-time repeated scene63 of the swordsman defeating a spearman may reflect heroic symbolism of a lightly armed warrior using a personal sidearm to defeat a heavily armored opponent—the hero vs. the horde.64 The sword-cut to the neck may be as much an echo of scenes of sacrifice as it is of an attack on a vulnerable area of soft tissue,65 and we are reminded again of the “bloodletting” design of some contemporary swords.66 On the “Battle in the Glen” ring (fig. 6) from Shaft Grave IV at Mycenae,67 four combatants are depicted; the larger size of the bezel allows the depiction of more characters. In this scene, two swordsmen are fighting: Fig. 6. The Battle in the Glen ring from Shaft Grave IV at Mycenae. Athens, National Archaeological Museum. 58 Evans 1906; Karo 1930; Hooker 1967, 270; Hiller 1984, 29; 1999. 59 The scenes on the shaft grave stelae depict swords but not their use in combat. 60 Hiller 1999; Peatfield 1999. 61 As we find for medieval combat (e.g., dei Liberi 1410; Talhoffer 2000). 62 Younger 1988; Hiller 1999; Papadopoulos 2006. 63 CMS 1, no. 11; CMS 7, nos. 100, 130. 64 Younger 1988; Peatfield 1999; Papadopoulos 2006. 65 Peatfield 1999. 66 Molloy 2008. 67 Schliemann 1878. 2010] SWORDS AND SWORDSMANSHIP IN THE AEGEAN BRONZE AGE the central character uses a shortsword,68 while the falling figure to the right uses a longsword. A spearman is about to enter the fray but is not actively engaged, and a fourth character is on the ground and not fighting. The only other two scenes with opposing swordsmen depict combatants using the short varieties of sword.69 Depictions of swords in combat contexts only occur on these small media, where we cannot tell if the image was chosen to fit the small medium or if the medium dictated the character of the image or, indeed, if both factors governed the design. In Aegean art, when the field becomes large enough to contain more than two figures, more figures are inevitably depicted.70 This has important implications for the use of swords, and dueling in particular, as discussed below. In LH IIIA and IIIB images, warriors are depicted in ceremonial/processional contexts, particularly on ceramics,71 and virtually no scenes of opposing military forces survive that would enable us to reconstruct elements of combat. A solitary example from Hall 64 at Pylos depicts Mycenaean72 warriors (identified by boar’s tusk helmets, greaves, and kilts) fighting barbarians (identified by animal-skin clothing, either exaggerated or symbolic ) using swords and spears. The lack of formations or shields allows us to derive little of the realities of combat, and the animal skins of the “barbarians” underscore the symbolic/hyperbolic nature of the scene. However, it can be noted that two different types of shortsword are being used, and that the alleged barbarians are depicted wielding similar swords to the Mycenaeans.73 By the time art returns to a more overt martial flavor in LH IIIC, actual scenes of combat remain rare and inconclusive,74 while swords are still not clearly shown in fights. The replacement of the traditional Aegean tower and figure-eight shields with smaller round shields marks a significant change in the manner of combat possible. The shields of the Warrior Vase from Mycenae (figs. 7, 8) were probably about 700 mm in diameter, based on relative proportions in the image, though the more commonly depicted shields are about 400–500 mm in diameter.75 These shields were adopted during LH IIIB2,76 and it is noteworthy that their shape and range of sizes have ready 68 This may be a dagger, but for the purpose of this argument, this distinction is not pertinent. 69 The size of the medium may dictate the length of blades depicted, but the examples of swordsman vs. spearman are evidence that longswords could be depicted when desired. 70 Preziosi and Hitchcock 1999. The later Tanagra larnax (LH IIIA) contains the only scene of two swordsmen fighting that ignores the potential to add further figures. © 2010 Archaeological Institute of America 411 Fig. 7. The Warrior Vase from Mycenae, depicting LH IIIC warriors. Athens, National Archaeological Museum, inv. no. 1426. Fig. 8. Model of a shield, manufactured by the author, after those illustrated on the Warrior Vase. 71 Vermeule and Karageorghis 1982; Hiller 1999; Papadopoulos 2006. 72 Lang 1969, no. 22 H 64; Kontorli-Papadopoulou 1999. 73 The blade outlines are distinctly different and are similar to those of Type Gi and Type Fii. 74 Vermeule and Karageorghis 1982. 75 Vermeule and Karageorghis 1982; Hiller 1999. 76 Lang 1969, 69, pls. 13, 116, 122, B. 412 BARRY MOLLOY [AJA 114 parallels throughout Europe (fig. 9).77 This is also the time that Naue ii swords first enter the Aegean. Thus, art reveals a new defensive weapon arriving with this archaeologically attested new type of offensive weapon. These shields, as I discuss below, had a dramatic impact on combat systems.78 Mortuary Evidence Alongside images of warriors, the remains of actual warriors survive in the mortuary record. Burial as a warrior79 provides evidence for warlike elements in the construction of identity after death. The assumption that the presence of weapons in burials indicated a warrior identity in life was questioned by Whitley80 in relation to the so-called Warrior Graves at Knossos. Smith subsequently demonstrated that burials with weapons in the Athenian Agora had little evidence of combat trauma, while other individuals possessing no weapons in death had blade-inflicted trauma consistent with combat injuries in life.81 This suggests that warriors were not buried with weapons as a matter of course and that the use of weapons to construct identity after death reflects the importance of the image of warrior identity, though its cultural purpose was contingent on social factors now lost. This bias in burial practice also calls into question whether the swords interred with the dead were the tools of their trade in life or symbolic representations of them. As most of our swords come from graves, we might ask if our actual data set is representative of real weaponry, or were they nonfunctional, symbolic weapons? This line of questioning may create unnecessary polarities in meaning and cultural processes, as some swords in burials may have been functional and workaday,82 while others may have been finely crafted yet unserviceable.83 There is no need for such an either/or scenario; a single sword may have elements of 77 Coles 1962; Molloy 2009. It is particularly noteworthy that the votive shields from the Idaean cave in Crete, of later Geometric date, had Vnotches unknown in Greek shield designs but common to European designs, as discussed in Coles 1962; Osgood 1998; Molloy 2009. 79 This term is preferred to “warrior burials,” as it reflects the nature of representation rather than suggesting a reality of identity in life. 80 Whitley 2002; see also Alberti 2004. 81 Smith 2009. 82 E.g., the Type Fii sword from Tomb 95 at Zapher Papoura (Heraklion, Heraklion Archaeological Museum, inv. no. 1103) was functional in proportions and had evidence for resharpening. 83 E.g., the Type A–related sword from Zapher Papoura, Tomb 44 (Heraklion, Heraklion Archaeological Museum, 78 © 2010 Archaeological Institute of America Fig. 9. Leather shield of common European form, from Clonbrin, Ireland (courtesy National Museum of Ireland). both poles in varying degrees. Funerary assemblages were brought together during specific events relating to different people, and so, as with the person, the biography of the artifacts need not fit a uniform pattern. We can, however, note that the physical characteristics of swords from burials and examples from other contexts are mechanically consistent, indicating functional similarity.84 Literary Evidence The Ra Linear B tablets from Knossos include Ra 1540, which lists 50 swords, and Ra 7498, which lists 18 and 99 swords; 20 fragmentary tablets list an un- inv. no. 1456A [Kilian-Dirlmeier 1993, no. 45]) or the welldecorated, excessively long and thin Type C from Tomb 36 (Heraklion, Heraklion Archaeological Museum, inv. no. 1097 [Evans 1906; Kilian-Dirlmeier 1993, no. 52]). 84 Similarity can be noted between the Type A swords from graves at Mycenae (Karo 1930) and the palace at Zakros (Platon 1966). The same can be said for the Type C swords from graves at Zapher Papoura (Evans 1906) and the shrines at Kato Syme (Lembesi 1975); and Dodona (Athens, National Archaeological Museum, Karapanos Collection, inv. no. 140 [Sandars 1963, 145]). The Type Gi sword from Zapher Papoura, Grave Tomb 14 (Heraklion, Heraklion Archaeological Museum, inv. no. 1102 [Evans 1906]), equates well with an example from the acropolis hoard (Tsountas 1891, 25–6) from Mycenae (Athens, National Archaeological Museum, inv. no. 2537). 2010] SWORDS AND SWORDSMANSHIP IN THE AEGEAN BRONZE AGE certain number of swords.85 These numbers contrast sharply with the 24 bronze swords of all periods known archaeologically from this area, demonstrating that a tiny fraction of weapons once in circulation have been discovered. To mitigate these discrepancies, I give the technology of sword design greater weight than numerical superiority or finer points of typologies in my analysis below.86 interpersonal combat and the birth of the sword The taxonomic evolution of the swords provided above spans nearly a millennium of social evolution, and we now must turn to the social context of developments in sword design and use. This process began in the Early Bronze Age with the widespread use of daggers. Before swords, daggers were the only bladed weapons produced that were suited to combat,87 particularly because of their two-edged design.88 They have a wide distribution throughout the Aegean in the third millennium and earlier second millennium B.C.E., particularly in Crete and the Cyclades.89 The images of daggers that appear on Cycladic and Cretan peak sanctuary figurines and the daggers themselves that are found in Early Bronze Age to early Middle Bronze Age burials demonstrate a long tradition that associates daggers with masculine identity.90 Throughout this time, the diffuse and low-density settlement systems91 may indicate that to provide a fighting force of a realistic size, most males in a community could be required to fight in larger-scale conflicts. This argument follows the assumption that each community would have to be capable of matching the fighting potential force of neighboring communities to sustain balance. This does not rule out battles of champions 85 Driessen and Macdonald 1984, 64. This, of course, relates to relatively common sword designs that fit the pattern and not the occasional exceptional pieces such as the unprovenanced Type Dii in the Ashmolean Museum in Oxford (Kilian-Dirlmeier 1993, no. 148; Molloy 2006, no. 244). 87 A bladed weapon is taken to mean one whereby the blade projects from the hand. Spears are characterized as shafted weapons. 88 Peatfield 1999. Experiments carried out by the author with similar daggers revealed that they were also suitable for cutting and serving joints of meat, another role that may have been associated with masculine identity, particularly with reference to game meat. 89 Renfrew 1972; Branigan 1974. 90 Peatfield 1999; Broodbank 2002, 253–54. 91 Renfrew 1972; Broodbank 2002. 92 Nakou 1995, 9–13; Broodbank 2002, 253. 93 Grossman (1996), Grossman and Christensen (2004), and Molloy and Grossman (2007) discuss interesting moral 86 © 2010 Archaeological Institute of America 413 or intercommunity coalitions. Mortuary evidence suggests that dagger possession was widespread,92 and as an emblem of male identity, it may be seen to symbolize the right to engage in violence. By nature, daggers are close-quarter weapons, and they make use of both the point and the edges for making attacks, particularly to the peripheral areas of the body, such as the hands, arms, legs, and head. When used in single combats or in a small-scale fracas, the dagger closely mimics the trajectories of unarmed punches, and indeed the arms and legs are both a means of attack and targets for attack in such close-quarter engagements. In group combats where proximity is decreased, it may be better to consider them secondary weapons to spears and bows, perhaps to be used for a coup de grâce or during a rout. In any case, the dagger is an up-close-and-personal weapon mainly suited to individualist rather than cooperative fighting styles.93 Despite superficial similarities, the sword is not simply a longer dagger; it is a new weapon that required an entirely different mode of use and the development of novel skill sets. The first swords appear late in the Protopalatial period at Mallia, in Crete, in a technologically well-developed form.94 Our experimental work has shown that casting very long, thin artifacts such as these requires a complexity in moldmaking and pyrotechnology that was not needed for daggers.95 Swords thus represent a leap forward in metallurgical technology that may imply that military requirements played an important role in driving bronzesmiths to push their craft in new directions. Acquiring the skill level to use these long weapons would represent a quantum leap in martial arts equal to the leap in technological know-how required to pro- and psychological implications of this sensory proximity. 94 They were the product of a mature smithing tradition with refinements in design that suggest they were past any experimental phases of design technology. 95 The casting challenges of producing a Type A sword relate to the difficulty of pouring molten bronze into a long, thin mold while ensuring that it does not freeze before it fills evenly throughout. The mold is far cooler than the molten bronze, so the liquid will begin to cool as it travels through the mold, causing it to begin to solidify. The farther it has to travel, the more it cools. The midrib provides a conduit for the bronze from butt to tip, though ensuring the bronze fills the thin webs of the blade edge would have presented a technological challenge. The other challenge is producing a ceramic mold of this length and preventing warping and cracking during drying, though it remains possible that stone molds were used. Having worked with modern bronzesmiths producing replica weapons, it is clear to me that the smiths who produced Type A weapons were very highly skilled. See O’Faolain (2004) for further details on manufacturing bronze weapons. 414 BARRY MOLLOY duce the swords. Their gracile proportions brought a far greater risk of mechanical failure through incorrect use than had existed with daggers. This paradigm shift in martial arts required commensurately greater investment of resources by communities and combatants themselves to allow participation in the requisite training.96 These swords were the first-ever purposemade tools of interpersonal combat, as they could serve no other practical function.97 While the users of the multipurpose dagger could hold pluralistic identities (e.g., warrior, hunter, trader, butcher, farmer, headman), the sword points specifically to the advent of a more unique identity, that of the warrior. combat archaeology and the use of bronze swords Peatfield initiated an important new direction in the study of Aegean swords by undertaking a functional study of their martial capacities.98 I further developed this approach by integrating artifact studies, use-wear analysis, experimental archaeology, and iconographic study to create a combat-archaeology methodology.99 Such an approach is intended to bridge the gaps between specialist artifact analysis and general narratives on warfare and society. To investigate functionality, a representative sample of ancient swords was measured for length, weight, cross-sectional thickness, edge preparation (checking for evidence of cold-working and states of sharpness), distal taper (the rate of decreasing thickness of the blade from the hilt to the tip), center of percussion,100 blade geometry,101 use-wear evidence, and (surviving) point of balance.102 These factors dictate how a given sword can be used most effectively,103 as discussed in detail in original and modern studies on medieval and Renaissance swords and swordsmanship.104 Accurate replica swords were used for a program of experimental archaeological research to complement the analysis of ancient weapons.105 96 Peatfield 2007; Molloy 2008. Their symbolic and ritual value derived from a recognition of their potential functions as a weapon. Without this martial element, they would be objects devoid of meaning or identity (Molloy 2009). 98 Peatfield 1999. 99 Molloy 2004, 2005, 2006, 2007, 2008, 2009; see also http://www.combat-archaeology.org. 100 The center of percussion is the point on a blade— typically about two-thirds the distance from the hilt to the tip—that will vibrate the least when hitting a target, thus representing the “sweet spot” on a sword where the maximum force is transferred to a target with minimal shock to the weapon or wrist of the user. 101 Blade geometry is the angles of the cutting edge and 97 © 2010 Archaeological Institute of America [AJA 114 This practical work focused on addressing the cutting and stabbing capabilities of various Aegean swords. Test cutting with swords used test pieces called tameshigiri mats (fig. 10),106 the forelegs and torso of recently slaughtered pigs, and samples of body armor. The mats are rolled into a cylindrical test piece and struck with the blade edge during cutting strikes (they are not useful in assessing the effectiveness of thrusting attacks), allowing the number of layers penetrated to broadly quantify the efficacy of each sword examined. Test cutting on the pigs also approximated human flesh and allowed a quantitative and qualitative assessment of the ways in which particular swords functioned best. This indicated the range of attacks best suited to particular models. Choreographed sparring was conducted to ascertain the nature and extent of edge damage inflicted from blade-on-blade impacts. This was not pursued to destruction because of the need to conduct repeated cutting tests and the difficulty in procuring accurate replica weapons for this research. Copper alloyed with approximately 7–12% tin produces a bronze that is hard and tough enough to manufacture sword-length castings that will serve as effective weapons. The inflexibility of the metal, however, differentiates it significantly from later iron swords. Long and thin forms of bronze sword would bend, or more likely break, if heavy force is applied in an attempt to cleave a target. Shortswords, however, usually had insufficient blade length or potential sharpness to create a long, drawing motion to slice the skin deeply. The edges of bronze swords could not be sharpened as much as those of (technologically mature) iron swords, as they would roll, chip, or dull too quickly, so edge angles were broad and edge sharpness moderate, not high. Bronze Age warriors and weaponsmiths were aware of these mechanical issues, so the forms of bronze swords would be expected to fit well with their intended functions. By isolating the processes that allow them to cut effectively, we can general shape of the blade (e.g., straight or tapering edge). 102 Examination of combat damage through use-wear analysis could not be conducted on Aegean swords, as too high a proportion of the edges are lost from corrosion. 103 The subjectivity of criteria for “effectiveness” is mitigated by the fact that incorrect use will not inflict injury and is likely to break the sword itself. 104 Burton 1987; Clements 1998, 1999, 2007; Talhoffer 2000; Wagner and Hand 2003; Forgeng and Kiermayer 2007. 105 Molloy 2006. The replica swords were manufactured by the author and Neil Burridge of Bronze Age Craft. 106 Tameshigiri mats are straw mats prepared in a traditional Japanese fashion that makes them demonstrably analogous to human flesh in their resistance to cutting. 2010] SWORDS AND SWORDSMANSHIP IN THE AEGEAN BRONZE AGE 415 Fig. 10. Testing a replica Type C sword against tameshigiri mats. thus determine aspects of how they were designed to be used. Amberger has isolated three major ways in which a sword may cut: (1) penetration by incision and percussion, (2) penetration by incision and laceration, and (3) percussion.107 The first of these requires that the blade strikes with downward force and that the sharp cutting edge split the flesh on impact so that the blade cleaves into the target; edge sharpness and impact force combine to inflict injury. The second type of cut requires the blade to be drawn along the target in a linear motion so that it slices using the sharpness of the edge rather than force of impact. The third type of cut rarely applies to bronze weapons, as it does not require a sharp edge but cleaves by force of impact alone, transferring high energy to a small surface area. For swords of bronze, the long, thin examples use the second form of cutting while shorter, stouter examples use a combination of both the first and second form of cut (occasionally 107 Amberger 1998, 94; see also McCarthy et al. 2007. Karo 1930, 204–6; Sandars 1961, 17; Fortenberry 1990, 148. It can also be noted that many of the broken tangs on these weapons are the result of corrosion of this comparative108 © 2010 Archaeological Institute of America the third), depending on specifics of design and the immediate combat environment. Type A Swords The first swords of the Aegean were also the longest, and their apparent fragility, and in particular their short tangs, has sometimes called their functionality into question.108 The earliest design with one rivet in the tang and two in the shoulder had wide currency and longevity. The development of longer, thicker tangs accommodating three rivets (fig. 11) suggests that the technological choice to strengthen the hilting system was not always practiced.109 When the handle is attached around the shoulders of the grip plate and the tang, the hand actually grasps part of the shoulders of the blade so that the tang is located between the middle two fingers. Weapons that exhibit broken tangs, rather than demonstrating unworthiness in combat, are direct evidence for their use in this context. ly thin part of the weapon. 109 For Mycenae, see Karo 1930. For Zakros, see Platon 1966. For Pylos, see Blegen et al. 1973. 416 BARRY MOLLOY Fig. 11. Tang on Type A sword from Zakros, illustrating thicker tang than on earlier variants. Heraklion, Heraklion Archaeological Museum, inv. no. 2590. Thrusting attacks using Type A swords are possible, but rapid changes of direction are difficult because of their forward-weighted design. For this reason of balance, the longer examples would have maneuvered much more slowly than historic fencing swords or rapiers. The thin design of the blade edges indicates that they were once sharp and suited to making cutting attacks. Percussive force would potentially damage these lengthy swords, though their morphology was ideally suited to making long draw cuts that could penetrate by incising and lacerating the flesh (fig. 12). The midrib was a perpendicular obstacle to the line of a cut (fig. 13), indicating that cutting strikes were intended to slice muscle rather than cleave bone.110 These attacks could be effectively made to the arms, shoulders, legs, neck, and face from a frontal stance. When moving past or sidestepping an opponent, the blade could slice along other exposed areas as the attacker’s whole body moves past the target, even when allowing for relative motion between opponents. It also may have been possible to grasp the blade of the weapon with the free hand, wearing a glove, as in historical swordsmanship.111 This would allow for tighter blade control and the ability to move in to close quarters with greater ease.112 The superficial similarity of the Type A sword to Renaissance rapiers113 and the occasional appearance of longswords in single combat on Aegean seal images might suggest that these were tools designed for duels, single combats, or “battles of champions.” The Renaissance concept of aristocratic dueling also accords well 110 The significance of this has been discussed in detail in Molloy (2008) and is addressed briefly below in relation to the Type C and D swords. 111 The snug-fitting gold bands on the swords from Shaft Grave V at Mycenae could be marking these areas rather than being remnants of scabbards. 112 Talhoffer 2000. 113 Type A, B, C, and D swords are often inappropriately termed “rapiers.” The term is borrowed from civilian weapons used for dueling in Renaissance times, but, as we do not know © 2010 Archaeological Institute of America [AJA 114 with the elite status of the Aegean burials containing swords (exemplified by Shaft Graves IV and V at Mycenae). Unfortunately, our basis of comparison stops there because, as noted above in the discussion on seal images, we have no tangible evidence for longswords being used against each other in combat.114 While they would undoubtedly have served well in this context, as with virtually any sword, the depiction of longswords on the hips of spearmen on the Miniature Fresco and the charioteer from Stele V (fig. 14) from Grave Circle A at Mycenae reminds us that they could have served in many different contexts. Type B Swords The short proportions of most Type B weapons represent a deliberate martial and technological design choice that contrasts with the Type A tradition.115 Shortswords were more common than long Type B Fig. 12. Cutting technique with early long swords of Types A, B, C, and D. Cutting force used is penetration by incision and laceration. Note midrib inhibiting cut depth. the certain context of use of Type A swords, the term is misleading. Type A swords were long, forward weighted, broad, and midribbed, and possessed two cutting edges, whereas rapiers were long, well-balanced, narrow, and flat or triangularsectioned with blunt edges. 114 The later Tanagra larnax depicts such a scene, flanked by chariots, at the time when Type C and Di swords were in use. 115 There are also weapons of similar function that are not strictly Type B in form, notably in Shaft Grave VI of Grave Circle A at Mycenae. 2010] SWORDS AND SWORDSMANSHIP IN THE AEGEAN BRONZE AGE variants and are more frequently depicted in iconography.116 The continuing use of daggers alongside swords indicates that changes in their symbolic value did not negate the need for them as weapons; they likely continued in use because of their multifunctionality. Kilian-Dirlmeier has suggested that the martial effectiveness of Type B swords relegated the Type A to a ceremonial weapon or status symbol.117 However, Type A swords served different combat functions than Type B weapons, apparent from their markedly different size range and the longevity of both forms alongside each other, and Type B is rarely found outside Mycenae itself. Furthermore, most long Type B swords would have looked remarkably similar to Type A swords when both had their original hilts. Single-Edged Swords Alongside these two varieties of double-edged sword, there are more than 26 examples of singleedged swords known from the Greek mainland.118 Imposing and robust weapons, they were capable of making powerful percussive cuts as well as thrusting attacks. They appear in the hands of warriors on foot on two of the Shaft Grave stelae (see fig. 14).119 Tests with a replica from Shaft Grave IV demonstrated that it could easily cut through flesh and into the bone when tested against the pig forelegs. When cutting against the tameshigiri mats, it could cut through 14 layers with ease. Thrusting attacks penetrated the ribcage perpendicular to the line of the ribs of the pig (i.e., it did not slide between them). The wooden plates on Type I single-edged swords provided good grip on the round-sectioned handle. The solid-cast hilts can prove slippery in the hand, though a cord passed through the loop at the end of the handle and secured around the wrist prevented the hand from slipping forward onto the blade edge, causing injury to the user, particularly during a thrusting attack. This device also inhibited rotation of the handle on cutting attacks, ensuring that the edge struck a target square-on. While experimental work has revealed that these swords are effective weapons, they virtually disappear by the end of LH I and are a typological dead end. 116 E.g., CMS 1, nos. 11, 16; CMS 11, no. 34. “Typ A-Schwerter nicht wegen ihrer Brauchbarkeit als Kampfwaffe weiter herestellt wurden, sondern wegen ihrer Bedeutung als Zeremonialschwert und Stattussymbol” (Kilian-Dirlmeier 1993, 41). 118 Karo 1930; Blegen et al. 1973; Tripathi 1988; Fortenberry 1990. 119 Stele 14 from Grave Gamma in Grave Circle B and Stele 117 © 2010 Archaeological Institute of America 417 Fig. 13. Cross-section of a Type C sword, illustrating the blade edges and obtrusive midrib. Oxford, Ashmolean Museum, inv. no. AE 482. Fig. 14. Stele V from Grave Circle A at Mycenae, depicting chariot rider with Type A or B sword and man on foot with single-edged sword. Athens, National Archaeological Museum, inv. no. 1428. Type C Swords The Type C sword possessed a midrib and had long and thin proportions (see fig. 13). The full tang with flanges provides evidence for the actual form of the handle, which does not survive on earlier swords; thus, it is possible to determine how it could have been held. The user cannot comfortably hold120 a Type C sword in a full fist grip, or hammer grip,121 and is forced to V from Grave Circle A (Karo 1930; Mylonas 1973; Younger 1997). 120 It can be noted that the latest variants had an increased angle to the quillons, making the grip more flexible. 121 In a hammer grip, the four fingers of the hand form a fist around the weapon, literally holding it as one would a hammer, so that the weapon is held perpendicular to the forearm. 418 BARRY MOLLOY adopt a saber grip,122 with the index finger typically curled around the quillon123 and gripping the ricasso.124 The main purpose of this grip is to align the edge of the blade to the natural trajectory of a strike with the arm so that the blade edge strikes square-on to a target to inflict injury in a cutting attack. This grip helpfully limits the ability to make percussive strikes, as this force would easily cause damage to the blade itself rather than the target. On the few examples where edges are comparatively well preserved, edge damage of combat causation has not yet been macroscopically detected. That it could potentially be visible is indicated by two examples of Type C swords with deliberately “killed” blades from Olympia and Dodona.125 This suggests that most swords that survive in burials had typically not been used in combat, and may indeed have been manufactured for deposition in the event of the funeral. The “Type A” sword from Zapher Papoura is likely to have been either a miscast example or a partially tanged sword deliberately manufactured to look like a Type C or Di sword when hilted.126 This small tang would have been hidden within an organic hilt, though the attachment appears to have been too flimsy for this to have been a serviceable weapon. Replicas of a Type C and a Type Di sword (fig. 15) from burials at Knossos were manufactured, and testing revealed new insights into their modes of use. Experiments with the replica Type C sword demonstrated that cutting attacks would have been capable of slicing or lacerating the flesh down to the depth of the midrib, typically 10–20 mm from the blade edge. This blade setup was similar to the Type A and Type B swords, though with an even shallower maximum cut depth. The blade needs to be drawn in a whiplike manner along the target, combining edge sharpness with length of draw to cut using incision and laceration; percussive force is not used. The light blade weight in relation to length, the presence of the midrib, and the limited sharpness achievable on bronze meant that test cuts typically penetrated four to six layers of mats. On the forelegs of the pig, while it could slice through the skin with ease, it inflicted minimal damage on muscle tissue, largely because of the inhibiting midrib. The 122 A saber grip stretches the fist at an angle so that the blade is aligned with the forearm and flexibility in the wrist is significantly reduced. 123 This finger superficially appears more exposed, as it is isolated and ahead of the shoulders, but it must be noted that there is no protection for the entire sword hand, so this grip does little to increase risk of injury to any particular digit. 124 Peatfield 1999, 2007; Molloy 2008. 125 Sandars 1963, 122, 145; Kilian-Dirlmeier 1993, 45, 47; © 2010 Archaeological Institute of America [AJA 114 Fig. 15. Replicas, manufactured by N. Burridge and author, of Type C (right) and Di (left) swords from Knossos, Crete. Note the curve on the Type C sword caused by test cutting. grip layout and nature of the edges, or indeed their existence in the first place, indicate that cutting strikes were an intended mode of attack, despite their apparent inefficacy. These attacks could only inflict injury on an unarmored opponent, and the sword was incapable of cutting through linen, leather, or copper armor test pieces.127 Attacks that landed on exposed Molloy 2006. 126 This weapon only awkwardly fits into the Type A category and is best seen as a singularity. 127 Molloy (2006) provides details of tests against these armors. Ten layers of medium-duty linen held together with animal-hide glue exploited the toughness of the fibers and the hardness of the glue, much like modern fiberglass. The leather armor was baked and waxed, creating a hard and tough test piece. The copper was 0.9 mm thick. 2010] SWORDS AND SWORDSMANSHIP IN THE AEGEAN BRONZE AGE flesh would make bloody and very visible wounds but would not typically result in incapacitation from the first few cuts. Cuts to most areas of the body would not be sufficiently deep to reach tendons or arteries and would cause largely superficial injury. The points of the swords could be effectively used for thrusting attacks, primarily to soft tissue areas, such as the abdomen or throat, but also to exposed areas such as the sword hand, forearm, face, legs, and feet. It was relatively easy to penetrate the ribcage of the pig, perpendicular to the ribs, and it is clear that thrusts with these swords could be lethal. It is noted, however, that the longer varieties risked damage because of poor balance between length and strength if attacking “deep” targets such as the chest, somewhat compromising the usefulness of these swords in multiopponent environments. They could stab through the leather and linen armor with relative ease, but the copper armor proved impenetrable.128 All bronze swords tested replicated this same pattern: they could not cut through any of the corselet sections, and they could stab through the leather and linen but not the copper. 419 the Type C sword, though later varieties of Type D were usually of broader proportions with a less acute point angle. Type F Swords It is significant that the Type D sword follows a markedly different trajectory from the Type C sword. The fist/hammer grip of the Type D sword allowed for closer-quarter use, and the blades evolved to take advantage of this enhanced efficacy for cutting attacks. The Type F sword abandoned the shoulder knobs of the Type D, optimizing it for a hammer grip while making a saber grip difficult. The shorter, more robust blade design was also suited to the rougher handling of a hammer grip, and, important to note, all vestige of the midrib was abandoned. A replica of a short LM IIIA Type F sword (based on the Zapher Papoura Tomb 95 example) was considerably better at cutting attacks than might have been expected, given its overall length of less than 400 mm. It was capable of cutting through 10 to 12 layers of a test mat with little difficulty and slicing to the bone on the foreleg of the pig. A longer Type F Type Di Swords The Type D sword in its earliest incarnations is very similar to the Type C sword in its blade form. The two are differentiated primarily by the form of the protrusions on their shoulders. A significant feature of the Type D hilt configuration is that the absence of quillons allows it to be held in a fist or hammer grip, as well as the saber grip common to Type C swords. The hammer grip of the Type D sword allows it to be held with the blade perpendicular to the forearm so that the wrist can be broken,129 enabling the user to execute more rapid changes of direction and trajectories of attack. As a result, the user can move in closer to an opponent while presenting the edge of the blade offensively. Another feature of this grip is the ability to engage in more percussive strikes using a combination of force and finesse. Test cutting with the Type Di cut through 10 layers of mat, and on the pig forelegs it could cut deeply, on occasion down to the bone. Late Type Di and Type Dii Swords It is noteworthy that within the Type D category, we get a functional evolution in forms toward weapons with proportions that are well suited to close-quarter percussive attacks. The reduction or abandonment of the midrib further enhanced their cutting potential (fig. 16). Thrusting attacks were similar in efficacy to Fig. 16. Type Dii, Fii, Gii, and Naue ii sword cutting processes. Cuts are by incision and percussion. 128 See Molloy (2009) on shields of leather, wood, and copper alloy. 129 Broken in this sense refers to the ability to be flexible, as it is not locked in position by the posture of a saber grip. © 2010 Archaeological Institute of America 420 BARRY MOLLOY [AJA 114 sword (fig. 17) based on the piece of LH IIIC date from Siteia, Crete,130 was also manufactured and, allowing for subjectivity, was the best balanced and most wellproportioned weapon of all those tested. Its efficacy was similar in testing to the shorter variety, cutting up to 14 layers of mat, and its broad point angle ensured that it would not penetrate so deeply on a thrusting attack to make it difficult to retrieve.131 These weapons were capable of inflicting serious injuries at close quarters using short, powerful strokes. The balance of the blade facilitated easy maneuverability, and its ability both to cut and to thrust would have made it a versatile weapon suited to close-quarter fighting. In the case of the shorter varieties, they differ from daggers largely on the basis of blade geometry and strength. The ability of these sword designs to cut with percussive force, as well as make incisions, made them very versatile weapons (see fig. 16). Even the longest varieties can comfortably be termed shortswords, and their longevity from LH IIIA to LH IIIC in the face of martial and social changes is testament to their perceived efficacy. It should be remarked that the Type F sword in itself was not a novel concept, as weapons of this length and roughly these proportions are found in the shaft graves at Mycenae. Its prevalence by LH IIIC best represents an increasing preference for shorter weapons and significant advances in the technology of their design. Type Gi Swords The angle of the quillons on the Type Gi sword allows for the saber grip associated with the Type C sword but also for the hammer grip of its contemporaries. Examples of this type at Zapher Papoura and Syme indicate that they were in use for many decades alongside the Type C, D, and F swords, marking LM/LH IIIA2 as a time with a rich weaponsmithing tradition.132 The Type Gi sword was the last in a long tradition of Aegean sword technology. While it retained the characteristic midrib of Aegean swords, the increased distance between it and the edge is a highly important development.133 Cutting attacks to soft tissue could slice deep into the target, separating the flesh so that by the time the low midrib was encountered, little energy was dissipated by it following into the line of the cut. Experimental cutting with these weapons showed them to be capable of cutting 14 layers of mat. During a demonstration at the World Archaeological Congress in Dublin, cutting strikes against the test mats and 130 Kilian-Dirlmeier 1993, no. 203 (Oxford, Ashmolean Museum, inv. no. 1966.542). 131 For Roman swords’ broad point angle, see Bishop and Coulston 1993; Connolly 1998, 232. © 2010 Archaeological Institute of America Fig. 17. Replica, manufactured by author, of Type Fii sword from Mouliana, Crete. watermelons (for safe display purposes) quite rapidly led to the bending of these blades across the flat of the blade. This demonstrated that while they stood up to relatively rough use in the short term and were effective for both cutting and thrusting attacks, they were not as robust as the Type F swords. During LH IIIB, the pointed-oval cross-sectioned designs with full fist/hammer grips came to entirely dominate bladed weaponry forms. Type Gii Swords The rare Type Gii sword was only superficially related to the Type Gi sword, as it was virtually identical to the 132 Evans 1906 (Zapher Papoura); Lembesi 1975 (Syme). This was typically double, and sometimes triple, that of the classic Type C and D swords (cf., e.g., Kilian-Dirlmeier 1993, nos. 88, 90–3 [with nos. 52–9]). 133 2010] SWORDS AND SWORDSMANSHIP IN THE AEGEAN BRONZE AGE Type F sword in functional terms. One particular difference was that the vestigial quillons with accompanying ricasso allowed for a slightly cumbersome version of a saber grip, a throwback to the earlier traditions. Naue ii Swords A new type in LH IIIB2 was the Naue ii sword, a weapon sometimes seen as revolutionizing combat at the end of the Bronze Age.134 They are often referred to as “cut-and-thrust” swords, a term once again borrowed from a specific, yet entirely different, form of Renaissance sword.135 The true definition of this term effectively means that the sword could cut and it could thrust—features that do little to differentiate it from a long tradition of Aegean swords that could do the same tasks. Yet the term “cut-and-thrust” has somehow generated a myth that the Naue ii sword brought something decisively new to combat. It did not. Naue ii swords were typically 100–200 mm longer than contemporary swords of the Aegean tradition. It is noteworthy that there is virtually no attempt at direct synthesis between the two traditions, as no hybrid forms are known, despite nearly two centuries of contemporaneous use.136 The greater length typical of the new sword forms need not imply martial superiority, as swords contemporary to the Naue ii were rarely manufactured in matching lengths. The evolution of the Type Gii sword within this milieu is important, as the variability in their lengths is a clear sign that the Aegean propensity for breadth of choice in sword design remained common. The Naue ii sword simply offered a new option in a martial tradition that had always been characterized by diversity. The evolution of the Naue ii sword in Crete is quite different from the rest of Greece (fig. 18). Most examples in Crete are only superficially related to those on the Greek mainland. The Cretan examples are thinner, narrower, and shorter, usually measuring <550 mm, compared with the 600 mm> examples common throughout the Balkan Peninsula.137 This makes 134 Drews 1993; Eder 1999; Toms 2000, 125. Clements 1999; Molloy 2006, 2007. 136 Kilian-Dirlmeier’s (1993) Gruppe C, Variante 3 may reflect a degree of hybridization. 137 Peroni 1970; Kilian-Dirlmeier 1993; Harding 1995. 138 Molloy 2006. 139 The swords in this area are quite distinctly different from those of Crete, but the pattern of functional similarity between Naue ii and Fii swords holds true. 140 These are Kilian-Dirlmeier’s Gruppe C, Variante 3. In Greece, two examples come from Graditsa (Oxford, Ashmolean Museum, inv. no. 1927.1383 [Kilian-Dirlmeier 1993, no. 245]; Oxford, Ashmolean Museum, inv. no. 1927.1384 [Kilian-Dirlmeier 1993, no. 246]), one from Steni (Athens, 135 © 2010 Archaeological Institute of America 421 them both lighter and faster than the earliest Naue ii swords to arrive in the Aegean. The local Type Fii swords of Crete are most often about 500–550 mm long and are of similar weight to the Naue ii swords from the island.138 The Cretan Type Fii swords have a good native pedigree in typological and functional terms, but the Naue ii swords diverge from their functional tradition in order to match the martial characteristics of local weapons. They do not, however, change their typological traits to any great degree, indicating that it was important to their owners and manufacturers to retain morphological traits of the “foreign” tradition. A similar pattern occurs139 in the area of modern-day southern Albania and Epirus in Greece, with markedly shorter-than-average Naue ii swords occurring alongside Type Fii swords of the same general size range. Another largely localized phenomenon is the occurrence of long, thin Naue ii swords with midribs in northern Boeotia and Thessaly.140 In this case, it is likely that the intrusive sword tradition was modified to suit existing traditions of swordsmanship, as weapons of these proportions do not occur farther north.141 Another example of this hybrid form also occurs at Mouliana in Crete,142 though its full length is not preserved. Replica swords that were tested included a mainland Naue ii sword of Kilian-Dirlmeier’s143 Class A and a Cretan Naue ii sword of Class C. All swords performed well against the tameshigiri mats, and there was no major discernible performance difference, despite variations in length and weight for the longer weapons. Both were capable of cutting 12–14 layers of mat, but not even these robust designs could cut through the radius and ulna of the pigs. The long bones in human arms are not load-bearing and so have less dense cortical bone than pig bones. As human bones are also longer, it may be possible to cleave through them with one of these swords, though this obviously cannot be investigated. Testing showed that none of the bronze swords was intended to contend with bone. This is not en- National Archaeological Museum, inv. no. 8017 [KilianDirlmeier 1993, no. 248]), and one from Vranezi (Chaeronia, Chaeronia Museum, inv. no. 750 [Kilian-Dirlmeier 1993, no. 249]). 141 The closest parallels are two examples from south of Belgrade in modern-day Serbia, both of which are close to the main riverine artery, via the Sava River, to the Adriatic Sea (Belgrade, National Museum of Serbia, inv. no. 1741 [Harding 1995, no. 100]; Šabac, National Museum of Šabac, inv. no. P/73 [Harding 1995, no. 101]). 142 Heraklion, Heraklion Archaeological Museum, inv. no. 999 (Kilian-Dirlmeier 1993, no. 247). 143 Kilian-Dirlmeier 1993, 95–8. 422 BARRY MOLLOY [AJA 114 risk of gradation in injury discovery, so that we only identify certain specific forms of attack, such as parrying injuries, where an individual uses an arm to (desperately) deflect a blow. It is important to state that bronze swords may have been able to cleave bone, and in some cases they quite probably did so. In terms of swordsmanship, however, the level of force required in relation to the blade design would make this a highrisk attack in terms of potential damage to the weapon and compromising one’s own guard. swordsmanship and society Fig. 18. Type Fii (left) and Naue ii (right) swords from Crete. Oxford, Ashmolean Museum, inv. nos. 1966.542, 1966.543 (respectively). tirely surprising, as it is only necessary to sever tendons, muscles, and arteries to disable a limb or even to kill a person. In fact, most effective attacks would have been against soft-tissue targets, such as the flesh of the forearms, neck, abdomen, or legs, but rarely seeking to injure bones. In relation to osteological analysis, the repercussions of this are that most forms of sword attack result in soft-tissue traumas that may be difficult to recognize or differentiate from slight excavation damage on bones. There is therefore a 144 As explained in detail for Maya society by Webster 2000, 89. 145 Halsall 2003, 22–4. When a modern state is not formally “at war,” it is considered “at peace,” whereby warlike action is an internationally recognized legal infringement. Lacking such legal support, 146 © 2010 Archaeological Institute of America Territorial boundaries of Aegean polities were more likely to have been marked by identities and loyalties than by lines on maps, and conflict at political interfaces may have been perpetual.144 A formal state of war, as we conceive it today, would have had little meaning in prehistory. Intercommunity tensions could be manifested as raiding, territorial incursions, kidnappings, or any number of lower-level conflicts that could either dissipate or escalate, contingent on circumstances and motivations.145 Open war was thus at an extreme end of a wide spectrum, with truce more likely at the other end than any formal state of “peace.”146 The character of weapons demonstrates that war was not a random social malfunction based on diminution of political order; rather, it was a structured element of Aegean societies. Violence was expected, and technology created tools suited to the multifarious ways it occurred while at the same time seeking to circumscribe its character. The creation of increasingly specialized weapons led to increasingly specialized martial arts, making competitive male fitness, rather than warlikeness, a characteristic of Aegean elites. Such simmering competition within and between factions and polities can also be seen in competitive feasting, control of ritual practice, and territoriality.147 Public and competitive displays of virility, from bull leaping to boxing, were common in art (and society by extension), particularly in Crete.148 In this context, it can be noted that the longer varieties of Type A, B, C, and (early) Di swords were well suited to agonistic (nonlethal) combats where severe injury or death were not the primary objectives. The limited cutting potential imposed by the presence of midribs, coupled with their need for nonpercussive draw cuts, meant that they could create bloody injuries that were not neces- prehistoric polities could not be “at peace,” as there was a perpetual risk of war deterred by threat of reciprocal action. 147 Knappet and Schoep 2000; Hamilakis 2002; Schoep 2006; Haggis 2007. 148 Hiller 1999; Preziosi and Hitchcock 1999. 2010] SWORDS AND SWORDSMANSHIP IN THE AEGEAN BRONZE AGE sarily debilitating. In single or small-group combats, this afforded ample room for movement and showmanship. The use of swords in killing strokes against spearmen on seals, symbolic value aside, reminds us that they were likely to have had more than one context of use and that they could be used to lethal effect if required. The old maxim that changes in armor wrought changes in weaponry is demonstrably simplistic,149 though changes gradually took place throughout the century following the introduction of bronze plate armor ca. 1450 B.C.E. Experiments have shown that Type C and D swords were not capable of contending with such armor; indeed, the only contemporary weapons that could penetrate it were spears of Type F and Type H.150 It is worth noting, however, that the neck guard, or gorget, on the Dendra cuirass would protect the wearer against a spear or sword thrust to the throat, a blow that is depicted frequently on earlier glyptic art. In Medieval times, plate armor was also virtually impregnable to contemporary edged weapons, yet knights in this armor wielded swords in combat.151 This was because not all combatants could afford such armor, leaving many men susceptible to sword attack. We may therefore posit that a Bronze Age battlefield had a range of categories of combatants, only some of whom possessed sophisticated protective equipment. The question of who owned swords and had the right to use them must be considered. In Crete during the Protopalatial and Neopalatial periods, graves of any form were rare. In contrast, the shaft graves at Mycenae represent the rare value of the sealed contexts of royal burials. Evely has warned about the dangers of comparing such unequal data sets from these two regions, though it is noteworthy that even though it lacks many grave finds, Crete has produced so many swords and the majority of martial images.152 In both regions, it appears that elites actively engaged in combat, and weapon forms were tailored to their needs. This need not exclude people of lower status from being called upon in larger-scale battles (esp. as seen with the archers and slingers on imagery from Shaft Grave IV), but it indicated that military hardware was orientated toward small-group153 or individual combats. Figure 19 shows broadly the quantities of swords known from various periods, including and excluding 149 Clements 2007. Höckmann 1980. 151 Clements 1998. 152 Evely 1996; Molloy 2006; Papadopoulos 2006. 153 A maximum estimate of individual forces numbering in the tens or hundreds rather than thousands fits the material 150 © 2010 Archaeological Institute of America 423 the shaft graves, and highlights the impact that depositional practice and the accident of survival play in framing our data sets. As discussed above, an individual’s possession of a sword in death tells little about the framework of military organization. The technology for mass production evident in sword design and the Linear B records of large quantities of swords154 hints at a much wider circulation than is attested through the artifactual evidence. We can infer that any sizable polity had tens or hundreds of individuals in possession of swords from MM/MH II–III to LM/LH IIIA. This appears to change in LH IIIB, with an increase in workaday designs of sword, and by LH IIIC, the wider distribution of burials as warriors throughout Greece indicates an increased enfranchisement of persons characterizing themselves as warriors.155 The fresco from Hall 64 at Pylos suggests that some Mycenaean states may have been involved in conflicts with peoples they characterized crudely as barbarians. These foreigners are depicted using shortswords, perhaps representative of a new combat milieu. The adoption of both the Naue ii sword from Italy156 and/ or the western-central Balkans and the European-style shield late in LH IIIB complements this scenario. The increasing preference for pointed-oval crosssectioned blades over midribbed styles began after the introduction of armor, though it became more pronounced by later LH IIIB. The long-term trajectory of developments demonstrates that the Naue ii sword complemented rather than catalyzed changes in military equipment in LH IIIB. The new shields of late LH IIIB/LH IIIC were perhaps more important in transforming swordsmanship and combat as a whole, as they dictate spatial relations and group performance patterns more than any other weapon. There is a standardization of military equipment apparent in the artwork of LH IIIC, particularly visible on the Warrior Vase from Mycenae (see fig. 7), though there were distinct regional and chronological trends (fig. 20). In most areas of Greece, there was a preference for swords 400–500 mm (Type Fii or Gii) and 550–700 mm (Naue ii) in length, accompanied by shields ranging perhaps 400–700 mm in diameter. Spears were more consistent, as virtually all examples were designed for single-handed thrusting attacks.157 Art indicates that swords were the secondary weapon and iconographic record best. 154 Driessen and Macdonald 1984, 64. 155 Kilian-Dirlmeier 1993. 156 Jung and Mehofer 2008. 157 Smaller examples were also suited to throwing (Höckmann 1980). 424 BARRY MOLLOY [AJA 114 Fig. 19. Schematic illustration of the varying quantities of swords in the Aegean: left, by time period; right, by time period without the finds from shaft graves from Mycenae. on the battlefield, as was the case much later in hoplite warfare; swords were weapons designed first and foremost for the unpredictable environment of a compact and restricted space.158 The forms of sword, spear, and shield common to this period suggest that the hoplite analogy is not entirely misplaced in combat terms, as these weapon combinations encouraged cooperative fighting styles and organized lines of battle. A military element in the destructions of settlements at the end of LH IIIB and the broader enfranchisement of warriors in LH IIIC may indicate that armed forces increased to number in the low thousands in the wars throughout the decades of collapse in the Aegean world. The social processes that promoted urbanism and centralization of resources in Protopalatial Crete bred new forms of social conflict and competition.159 The Type A sword that emerged in this milieu was an entirely new type of object. It was a deadly weapon that instigated new social practices and a new forum for elite interaction and competition through a specific warrior identity. Contemporary changes in shield and spear design would have complemented these changes, though they were less revolutionary in terms of patterns of action and interaction. This can be seen to reflect a restriction in those permitted to undertake legitimate violence on behalf of a community to those equipped with, and trained to use, specialist weaponry. The sword might therefore be seen not to have simply replaced the dagger in combat—it supplanted an ancient tradition of exercising legitimate, socially sanctioned violence on behalf of a community. The widespread adoption of swords may therefore have re- 158 In terms of length, they very closely approximate classical Greek swords (Snodgrass 1999) and the Roman gladius (Bishop and Coulston 1993); see also Peatfield 1999. 159 Knappet and Schoep 2000; Schoep 2006; Haggis 2007. conclusion © 2010 Archaeological Institute of America 2010] SWORDS AND SWORDSMANSHIP IN THE AEGEAN BRONZE AGE 425 Fig. 20. Regions with distinctive functional traditions of sword design in LH IIIB–C. lated as much to political, as to military, expediency. This process began in the established political powers of Crete and spread from there to the Greek mainland, as seen in the material from the shaft graves at Mycenae.160 Elites in both areas characterized themselves as participants in this new martial tradition, and the osteological evidence indicates that some had recourse in life to violent engagements with these weapons.161 The first swords were exclusively Type A longswords. These were followed within a century by Type B weapons, a group that included varieties of many lengths, demonstrating an early complexity in the development of swordsmanship. Mainland single-edged swords had brief popularity, though they did not survive long beside the Type C and D swords in LH/LM II–IIIA. During this period, the ratio between long, elaborate swords to short, robust varieties shifted toward 160 O’Brien 2009, 83. © 2010 Archaeological Institute of America the latter, though both forms remained in use. This increased preference for workaday forms dominated military material culture by LH IIIB and LH IIIC and represented a growing preference for more close-knit fighting formations. A simple three-tiered division of swords into early (Type A, B, and single-edged), middle (Type C, Di, and Gi), and late (Type Dii, F, Gii, and Naue ii) works in broad terms, though a division into midribbed and non-midribbed weapons has been shown to be more pertinent for understanding the evolution of swordsmanship. By the end of the Bronze Age, the final refinements of bronze military hardware in the Aegean saw the elimination of any ostentation in sword design in favor of maximizing the mechanical properties of the metal. The occurrence of weapons, images, textual references, burial practices, and fortifications all concur 161 Angel 1972; Arnott 1999; Smith 2009. 426 BARRY MOLLOY with Homer’s warlike vision of Bronze Age Greece. This was a pattern, however, consistent across the entire continent of Europe, and the explosion of military technology was a defining characteristic of many Bronze Age societies.162 Despite the wealth of evidence laid out, characterization of warfare remains elusive because of our dearth of evidence on demographics, scale, frequency, strategies, logistics, intensity, and so on. By focusing on methods of combat, and swordsmanship in particular, we can best apply our wealth of evidence by exploring the social systems that were behind war and the technologies that made it possible. Bronze Age Aegeans lived with these components of war in their daily lives, and thus exploration of this theme is more salient than seeking glorious battles for understanding their civilization. school of archaeology university college dublin belfield dublin 4 republic of ireland barrymolloy@gmail.com Works Cited Alberti, L. 2004. “The Late Minoan II–IIIA Warrior Graves at Knossos: The Burial Assemblages.” In Knossos: Palace, City, State, edited by G. Cadogan, E. Hatzaki, and A. Vasilakis, 127–37. British School at Athens Studies 12. Athens: British School at Athens. Amberger, J.C. 1998. The Secret History of the Sword: Adventures in Ancient Martial Arts. Burbank, Calif.: MultiMedia Books. Angel, J.L. 1972. “Human Skeletons from Grave Circles at Mycenae.” In Taphikos Kyklos B ton Mykinon, edited by G. Mylonas, 379–97. Athens: Archaiologike Hetaireia. Arnott, R. 1999. “War Wounds and Their Treatment in the Aegean Bronze Age.” In Polemos: Polemos: Le contexte guerrier en Égée á l’âge du Bronze. Actes de la 7e Rencontre égéenne internationale Université de Liège, 14–17 avril 1998, edited by R. Laffineur, 499–506. Aegaeum 19. Liège and Austin: Université de Liège and University of Texas at Austin. Bishop, M.C., and J.C.N. Coulston. 1993. Roman Military Equipment from the Punic Wars to the Fall of Rome. London: Batsford. Blegen, C., M. Rawson, W. Taylour, and W. Donovan. 1973. The Palace of Nestor at Pylos in Western Messenia. Vol. 3, Acropolis and Lower Town: Tholoi, Grave Circle and Chamber Tombs. Discoveries Outside the Citadel. Princeton: Princeton University Press. Bodinaku, N. 1988. “Dy armë bronzi nga Këlcyra Deux armes de bronze provenant de Këlcyra.” Iliria 18(1):34–49. 162 Harding 1999, 2007. © 2010 Archaeological Institute of America [AJA 114 Branigan, K. 1974. Aegean Metalwork of the Early and Middle Bronze Age. Oxford: Clarendon Press. Broodbank, C. 2002. An Island Archaeology of the Early Cyclades. Cambridge: Cambridge University Press. Burton, R.F. 1987. Reprint. The Book of the Sword. Ontario: Dover Publications. Original edition, London: Chatto and Windus, 1884. Catling, H. 1956. “Bronze Cut-and-Thrust Swords in the Eastern Mediterranean.” PPS 22:102–25. ———. 1961. “A New Sword from Cyprus.” Antiquity 35: 115–23. Chapouthier, V.F. 1938. “Deux epées d’apparant découvert en 1936 au de palais de Mallia.” ÉtCrét 5:1–36. Clements, J. 1998. Medieval Swordsmanship: Illustrated Methods and Techniques. Boulder, Colo.: Paladin Press. ———. 1999. Renaissance Swordsmanship: The Illustrated Use of Rapiers and Cut-and-Thrust Swords. Boulder, Colo.: Paladin Press. ———. 2007. “The Myth of Thrusting Versus Cutting Swords.” In The Cutting Edge: Studies in Ancient and Medieval Combat, edited B.P.C. Molloy, 168–76. Stroud: Tempus. Coles, J.M. 1962. “European Bronze Age Shields.” PPS 28:156–90. Connolly, P. 1998. Greece and Rome at War. London: Greenhill Books. Cowen, J.D. 1951. “The Earliest Swords in Britain and Their Origins on the Continent of Europe.” PPS 17:195– 213. ———. 1955. “Eine einführung in die Geschicte der bronzenen Griffzungenschwerter in Suddeutschland und den angrenzenden Gebieten.” BerRGK 36:53–155. ———. 1966. “The Origins of the Flange Hilted Sword of Bronze in Continental Europe.” PPS 32:262–312. dei Liberi, F. 1410. Flos Duellatorum, or Fior di Battaglia. Florence: Privately printed. Dickinson, O.P.T.K. 1997. “Arts and Artefacts in the Shaft Graves: Some Observations.” In Texnh: Craftsmen, Craftswomen and Craftsmanship in the Aegean Bronze Age, edited by R. Laffineur and P.P. Betancourt, 47–9. Aegaeum 16. Liège and Austin: Université de Liège and University of Texas at Austin. Drews, R. 1993. The End of the Bronze Age. Princeton: Princeton University Press. Driessen, J., and C. Macdonald. 1984. “Some Military Aspects of the Aegean in the Late Fifteenth and Early Fourteenth Centuries BC.” BSA 79:49–75. Eder, B. 1999. “Late Bronze Age Swords from Ancient Elis.” In Polemos: Le contexte guerrier en Égée á l’âge du Bronze. Actes de la 7e Rencontre égéenne internationale Université de Liège, 14–17 avril 1998, edited by R. Laffineur, 443–49. Aegaeum 19. Liège and Austin: Université de Liège and University of Texas at Austin. Eogan, G. 1965. Catalogue of Irish Bronze Swords. Dublin: National Museum. Evans, A. 1906. “The Prehistoric Tombs at Knossos.” Archaeologia 59(2):391–562. Evely, D. 1996. “The Neo-Palatial Warrior: Fact or Fiction.” In Minotaur and Centaur: Studies in the Archaeology of Crete and Euboea Presented to Mervyn Popham, edited by D. Evely, I. Lemos, and S. Sherratt, 59–69. BAR-IS 638. Oxford: Tempus. 2010] SWORDS AND SWORDSMANSHIP IN THE AEGEAN BRONZE AGE Forgeng, J., and A. Kiermayer. 2007. “‘The Chivalric Art’: German Martial Arts Treatises of the Middle Ages and Renaissance.” In The Cutting Edge: Studies in Ancient and Medieval Combat, edited B.P.C. Molloy, 153–67. Stroud: Tempus. Forsdyke, E. 1926–1927. “The Mavro Spelio Cemetery at Knossos.” BSA 28:282. Fortenberry, C.D. 1990. “Elements of Mycenaean Warfare.” Ph.D. diss., University of Cincinnati. Grossman, D. 1996. On Killing. New York: Little, Brown. Grossman, D., and L.W. Christensen. 2004. On Combat. New York: PPCT Research Publications. Haggis, D.C. 2007. “Stylistic Diversity and Diacritical Feasting at Protopalatial Petras: A Preliminary Analysis of the Lakkos Deposit.” AJA 111(4):715–75. Halsall, G. 2003. Warfare and Society in the Barbarian West 450–900. London: Routledge. Hamilakis, Y. 2002. “Too Many Chiefs? Factional Competition in Neopalatial Crete.” In Monuments of Minos: Rethinking the Minoan Palaces. Proceedings of the International Workshop “Crete of the Hundred Palaces?” Held at the Université Catholique de Louvain, Louvain-la-Neuve, 14–15 December 2001, edited by J. Driessen, I. Schoep, and R. Laffineur, 179–99. Aegaeum 23. Liège and Austin: Université de Liège and University of Texas at Austin. Harding, A. 1984. The Mycenaeans and Europe. London: Academic Press. ———. 1995. Die schwerter im ehemaligen Jugoslawien. Prähistorische Bronzefunde 4(14). Munich: C.H. Beck. ———. 1999. “Warfare: A Defining Characteristic of Bronze Age Europe?” In Ancient Warfare, edited by J. Carmen and A. Harding, 157–75. Stroud: Sutton. ———. 2007. Warriors and Weapons in Bronze Age Europe. Budapest: Archaeolingua Alapítvány. Hiller, S. 1984. “Pax Minoica Versus Minoan Thalassocracy: Military Aspects of Minoan Culture.” In The Minoan Thalassocracy: Myth and Reality, edited by R. Hagg and N. Marinatos, 27–30. SkrAth 4°, 32. Göteborg: Paul Åströms Förlag. ———. 1999. “Scenes of Warfare and Combat in the Arts of the Aegean Late Bronze Age: Reflections on Typology and Development.” In Polemos: Le contexte guerrier en Égée á l’âge du Bronze. Actes de la 7e Rencontre égéenne internationale Université de Liège, 14–17 avril 1998, edited by R. Laffineur, 319–31. Aegaeum 19. Liège and Austin: Université de Liège and University of Texas at Austin. Höckmann, O. 1980. “Lanze und Speer im spatminoischen und mykenischen Griechenland.” JRGZM 27:13–158. Hood, S. 1956. “Another Warrior Grave at Ayios Ioannis near Knossos.” BSA 51:81–99. Hood, S., and P. De Jong. 1952. “Late Minoan WarriorGraves from Ayios Ioannis and the New Hospital Site at Knossos.” BSA 47:243–77. Hooker, J.T. 1967. “The Mycenae Siege Rhyton and the Question of Egyptian Influence.” AJA 71(3):269–83. Ioannidou-Karetsou, A. 1985. “Hagios Sylas.” ArchDelt 33: 352–59. Jung, R., and M. Mehofer. 2008. “A Sword of Naue II Type from Ugarit and the Historical Significance of ItalianType Weaponry in the Eastern Mediterranean.” Aegean Archaeology 8:111–35. Karo, G. 1930. Die Schachtgräber von Mykenai. Munich: F. Bruckmann. Kasimi-Soutou, M. 1980. “Mesoelladikos taphos polemiste apo te Theba.” ArchDelt 35:88–101. © 2010 Archaeological Institute of America 427 Kilian-Dirlmeier, I. 1993. Die Schwerter in Griechenland (Ausserhalb der Peloponnes), Bulgarien und Albanien. Prähistorische Bronzefunde 4(12). Munich: C.H. Beck. Knappet, C., and I. Schoep. 2000. “Continuity and Change in Minoan Political Power.” Antiquity 74:365–71. Kontorli-Papadopoulou, L. 1999. “Fresco Fighting-Scenes as Evidence for Warlike Activities in the Late Bronze Age Aegean.” In Polemos: Le contexte guerrier en Égée á l’âge du Bronze. Actes de la 7e Rencontre égéenne internationale Université de Liège, 14–17 avril 1998, edited by R. Laffineur, 331–40. Aegaeum 19. Liège and Austin: Université de Liège and University of Texas at Austin. Lang, M.L. 1969. The Palace of Nestor at Pylos in Western Messenia. Vol. 2, The Frescoes. Princeton: Princeton University Press. Lembesi, A. 1975. “Ieron Ermou kai Afroditis eis Eimin Viannou.” Prakt:322–29. Marinatos, S. 1935. “Ausgrabungen und Funde auf Kreta, 1934–1935.” AA 50:159–244. McCarthy, C.T., M. Hussey, and M.D. Gilchrist. 2007. “On the Sharpness of Straight Edge Blades in Cutting Soft Solids: Part I—Indentation Experiments.” Journal of Engineering Fracture Mechanics 74:2205–24. Molloy, B.P.C. 2004. “Experimental Combat with Bronze Age Weapons.” Archaeology Ireland 17(4):32–4. ———. 2005. “The Adoption of the Naue ii Sword in the Aegean.” In Proceedings of SOMA 2003: Symposium on Mediterranean Archaeology, edited by C. Briault, J. Green, A. Kaldelis, and A. Stellatou, 115–17. BAR-IS 1391. Oxford: British Archaeological Reports. ———. 2006. “The Role of Combat Weaponry in Bronze Age Societies: The Cases of the Aegean and Ireland in the Middle and Late Bronze Age.” Ph.D. diss., University College Dublin. ———. 2007. “What’s the Bloody Point? Fighting with Irish Bronze Age Weapons.” In The Cutting Edge: Studies in Ancient and Medieval Combat, edited B.P.C. Molloy, 90–112. Stroud: Tempus. ———. 2008. “Martial Arts and Materiality: A Combat Archaeology Perspective on Aegean Swords of the Fifteenth and Fourteenth Centuries BC.” WorldArch 40(1): 116–34. ———. 2009. “For Gods or Men? A Reappraisal of the Function of European Bronze Age Shields.” Antiquity 83: 1052–64. Molloy, B.P.C., and D. Grossman. 2007. “Why Can’t Johnny Kill? The Psychology and Physiology of Interpersonal Combat.” In The Cutting Edge: Studies in Ancient and Medieval Combat, edited B.P.C. Molloy, 188–202. Stroud: Tempus. Morgan, L. 1988. The Miniature Wall Paintings from Thera: A Study in Aegean Culture and Iconography. Cambridge: Cambridge University Press. Mylonas, G. 1973. O Taphikos Kyklos B ton Mykinon. Athens: Archaiologike Hetaireia. Nakou, G. 1995. “The Cutting Edge: A New Look at Early Aegean Metallurgy.” JMA 8(2):1–31. O’Brien, S. 2009. “Beyond the Sharp Bronze: Warfare and Society in Mycenaean Greece.” Ph.D. diss., University of Liverpool. O’Faolain, S. 2004. Bronze Artefact Production in Late Bronze Age Ireland. BAR-IS 382. Oxford: Archaeopress. Osgood, R. 1998. Warfare in the Late Bronze Age of North Europe. BAR-IS 694. Oxford: British Archaeological Reports. Papadopoulos, A. 2006. “The Iconography of Warfare 428 B. MOLLOY, SWORDS AND SWORDSMANSHIP IN THE AEGEAN BRONZE AGE in the Bronze Age Aegean.” Ph.D. diss., University of Liverpool. Papadopoulos, T.J. 1978. Mycenaean Achaea. SIMA 55. Göteborg: Paul Åströms Förlag. ———. 1999. “Warrior Graves in Achaean Mycenaean Cemeteries.” In Polemos: Le contexte guerrier en Égée á l’âge du Bronze. Actes de la 7e Rencontre égéenne internationale Université de Liège, 14–17 avril 1998, edited by R. Laffineur, 267–77. Aegaeum 19. Liège and Austin: Université de Liège and University of Texas at Austin. Papadopoulos, T.J., and L. Kontorli-Papadopoulou. 1984. “Notes from Achaea.” BSA 79:221–27. Papazoglou-Manioudaki, L. 1994. “A Mycenaean Warrior’s Tomb at Krini near Patras.” BSA 89:171–200. Peatfield, A.D. 1999. “The Paradox of Violence: Weaponry and Martial Art in Minoan Crete.” In Polemos: Le contexte guerrier en Égée á l’âge du Bronze. Actes de la 7e Rencontre égéenne internationale Université de Liège, 14–17 avril 1998, edited by R. Laffineur, 67–74. Aegaeum 19. Liège and Austin: Université de Liège and University of Texas at Austin. ———. 2007. “Reliving Greek Personal Combat: Boxing and Pankration.” In The Cutting Edge: Studies in Ancient and Medieval Combat, edited B.P.C. Molloy, 20–33. Stroud: Tempus. Peroni, V.B. 1970. Die Schwerter in Italien. Prähistorische Bronzefunde 4(1). Munich: C.H. Beck. Persson, A. 1931. The Royal Tombs at Dendra near Midea. Lund: Gleerop. Platon, L. 1966. “Anaskafi Zakrou.” Prakt:142–68. Popham, M.R., and H.W. Catling. 1974. “Sellopoulo Tombs 3 and 4, Two Late Minoan Graves near Knossos.” BSA 69:195–258. Preziosi, D., and L.A. Hitchcock. 1999. Aegean Art and Architecture. Oxford: Oxford University Press. Renfrew, C. 1972. The Emergence of Civilisation. London: Methuen. Sakellariou, A. 1974. “Un cratère d’argent avec scène de bataille provenant de la IVe tombe de l’acropole de Mycènes.” AntK 17:195–208. Sandars, N.K. 1961. “The First Aegean Swords and Their Ancestry.” AJA 65(1):17–29. ———. 1963. “Later Aegean Bronze Swords.” AJA 67(2): 117–53. Schliemann, H. 1878. Mycenae. London: Murray. Schoep, I. 2006. “Looking Beyond the First Palaces: Elites and the Agency of Power in EM III–MM II Crete.” AJA 110(1):37–64. © 2010 Archaeological Institute of America Smith, S.K. 2009. “Skeletal Evidence for Militarism in Mycenaean Athens.” In New Directions in the Skeletal Biology of Greece, edited by L.A. Schepartz, S.C. Fox, and C. Bourbou, 99–111. Hesperia Suppl. 43. Princeton: American School of Classical Studies at Athens. Snodgrass, A. 1999. Arms and Armour of the Greeks. Baltimore: The Johns Hopkins University Press. Talhoffer, H. 2000. Reprint. Medieval Combat: A FifteenthCentury Illustrated Manual of Swordfighting and CloseQuarter Combat. Translated and edited by M. Rector. London: Greenhill Books. Originally published as Fechtbuch. Privately printed, 1467. Toms, J. 2000. “The Aegean.” In Bronze Age Warfare, edited by R. Osgood, S. Monks, and J. Toms, 115–37. Stroud: Sutton. Tripathi, D.N. 1988. Bronzework of Mainland Greece from c. 2600 BC to c. 1450 BC. SIMA-PB 69. Göteborg: Paul Åströms Vörlag. Trump, B.A.V. 1962. “The Origin and Development of British Middle Bronze Age Rapiers.” PPS 28:80–102. Tsountas, C. 1891. “Ek Mykinon.” ArchEph:1–43. Vermeule, E., and V. Karageorghis. 1982. Mycenaean Pictorial Vase Painting. London: Harvard University Press. Vlasaki, M. 2005. “Chania.” AR 51:117. Wagner, P., and S. Hand. 2003. Medieval Sword and Shield: The Combat System of Royal Armouries MS I.33. Highland Village, Tex.: The Chivalry Bookshelf. Walter, H. 1981. “Anaskaphi sto Logo Kolona, Aegina 1981– 1982.” Archaiologika Analekta Athenon 14:182. Warren, P. 1979. “The Miniature Fresco from the West House at Akrotiri, Thera, and Its Aegean Setting.” JHS 99:115– 29. Webster, D. 2000. “The Not So Peaceful Civilization: A Review of Maya War.” Journal of World Prehistory 14(1):68– 119. Whitley, J. 2002. “Objects with Attitude: Biographical Facts and Fallacies in the Study of Late Bronze Age and Early Iron Age Warrior Graves.” Cambridge Journal of Archaeology, 12(2):217–32. Younger, J.G. 1988. The Iconography of Late Minoan and Mycenaean Sealstones and Finger Rings. Bristol: Bristol University Press. ———. 1997. “The Stelae of Mycenae Grave Circles A and B.” In Texnh: Craftsmen, Craftswomen and Craftsmanship in the Aegean Bronze Age, edited by R. Laffineur and P. Betancourt, 229–42. Aegaeum 16. Liège and Austin: Université de Liège and University of Texas at Austin.