Contents lists available at ScienceDirect Journal of Anthropological Archaeology journal homepage: www.elsevier.com/locate/jaa Landscape, typologies, and the social meaning of castles Scott D. Kirk a, ⁎ , Evan S. Sternberg b , Paulina F. Przystupa a a University of New Mexico, Department of Anthropology, Albuquerque, NM, USA b University of New Mexico, Office of Contract Archeology, Albuquerque, NM, USA 1. Introduction Castles are fortifed elite residences, built and maintained by members of militarized ruling classes in preindustrial societies (cf. Anderson, 1970; Brown, 2012; Duby, 1982; Morris, 2017; Prior, 2006; Wickham, 2016). Defned in this way, they appear in a wide variety of temporal and spatial settings, not just medieval Europe. 1 Commonly cited non-traditional examples include the castles of feudal Japan and colonial fortresses in places like Havana, Cuba. Regardless of spatio- temporal setting, castles are often the permanent residences of elites (i.e. Noble Houses, the Houses of a military commanders, and/or the residences of holy orders), equipped with provisions for their extended households and a garrison. However, they have also been built as summer retreats and temporary refugia. Despite obvious diferences in form and specifc functions, there are cross-cultural features of castles that make them recognizable as part of the same human behavior. In this paper we propose an anthropological explanation for why castles are universally recognizable and a typology for exploring variability in their development based on landscape. Anthropological models in castle studies are rare, as most scholar- ship on the subject currently utilizes historically particular approaches. Contrasting the militaristic explanations for castle development pub- lished in the early to mid 20th century (e.g. Hogg, 1981; Lawrence, 1936; Pettengill, 1979), which often saw ancient and non-Western 2 fortresses as similar to European castles, contemporary research largely rejects functional, behavioral, and evolutionary approaches (see Creighton, 2012; Johnson, 2002; Morris, 2017; Prior, 2006). Since the 1970s, many scholars, particularly in England, have understood castles as unique expressions of elite life, regionally distinct and steeped in local symbolism (e.g. Coulson, 2004, 1979; Johnson, 2002; O’Keefe, 2007). Newer defnitions for these structures have increasingly linked the term “castle” to social change and the rise of Christian States in Western Europe (e.g. Molinari, 1998, 1997; see also Creighton, 2012). However, this focus generally ignores real cross-cultural, chronological, and functional similarities between castles across socio-political and geographic divides (cf. Boone, 2009; also see Quirós Castillo, 2014). In contrast to the current trend explaining castles as a feudal or Christian phenomenon, we propose that the term “castle” has become too narrow. In this paper we argue that these monumental structures should be studied in light of the cross-cultural behavioral processes that led to their rise across disparate parts of the globe. With most castles functioning as elite residences, we suggest that the theoretical frame- work of the House Societies Model (HSM, see: Beck, 2007; Chesson, 2003; Gillespie, 2000; Joyce and Gillespie, 2000; Lévi-Strauss, 1982)is uniquely suited for exploring parallels in the development and use of castles globally. Originally devised as a cross-cultural, comparative approach to understanding social organization (Lévi-Strauss, 1982), the HSM sees the House as the basic unit for human sociality. Within this model, the House is defned as a corporate residential group (cf. Hayden and Cannon, 1982; Honigmann, 1959) that acts as an individual with respect to land, material goods, immaterial wealth, and namesake (Lévi-Strauss, 1982). Its material manifestation, as well as the landscape it inhabits, are thus as important to its protection, prosperity, and the maintenance of its traditions as the people who inhabit it (cf. Beck, 2007; Chesson, 2003; Joyce and Gillespie, 2000; Lévi-Strauss, 1982). Landscape (cf. Sánchez, 2015) and placement are particularly im- portant for understanding why castles appear as manifestations of the built environment because of their ability to protect individuals and resources, as well as generate new wealth, within the House. Assuming a functional relationship between the environment, and continuity in major landscape features over time, we suggest that categorizing castles based on landscape helps to illuminate why these structures developed in such culturally diverse locations by largely ignoring the aesthetic diferences in castle design that are often the focus of castellology. Using a sample of 459 medieval and Early Modern castles from around the world (Fig. 1), we collected data on common landscape features around each castle to conduct a series of k-means cluster analyses after the algorithm of Hartigan and Wong (1979) in (R Core Team, 2019). Analyzed regionally, supra-regionally, and as one sample, we sorted castles into eight types that broadly defne these structures across https://doi.org/10.1016/j.jaa.2020.101224 Received 20 April 2020; Received in revised form 25 July 2020 ⁎ Corresponding author. E-mail address: kirks@unm.edu (S.D. Kirk). 1 We defne the medieval period, or the Middle Ages, in Europe as the time from the collapse of the Western Roman Empire in the late 5th Century through to the Reconquista of Granada at the end of the 15th Century. This is subsequently followed by the Early Modern Period in Europe stretching from the 16th through the 19th centuries. 2 By non-Western we are referring to elements of cultures that are not European or European colonial. Journal of Anthropological Archaeology 60 (2020) 101224 0278-4165/ © 2020 Elsevier Inc. All rights reserved. T