Journal of Ancient near Eastern Religions 18 (2018) 86–110 Journal of Ancient Near Eastern Religions brill.com/jane Counterintuitive Demons: Pazuzu and Lamaštu in Iconography, Text, and Cognition Brett Maiden Emory University brett.maiden@emory.edu Abstract This paper examines the demons Pazuzu and Lamaštu from a cognitive science perspective. As hybrid creatures, the iconography of these demons combines an array of anthropomorphic and zoomorphic properties, and is therefore marked by a high degree of conceptual complexity. In a technical sense, they are what cognitive researchers refer to as radically “counterintuitive” representations. However, highly complex religious concepts are difficult in terms of cognitive processing, memory, and transmission, and, as a result, are prone to being spontaneously simplified in structure. Accordingly, there is reason to expect that the material images of Pazuzu and Lamaštu differed from the corresponding mental images of these demons. Specifically, it is argued here that in ancient cognition and memory, the demons would have been represented in a more cognitively optimal manner. This hypothesis is further supported by a detailed consideration of the full repertoire of iconographic and textual sources. Keywords demons – Pazuzu – Lamaštu – monsters – iconography – incantations – cognitive science of religion * I am grateful to Brent Strawn for his generous support and detailed comments on earlier versions of this article, and to Joel LeMon, Izaak de Hulster, and Byron McCane for their helpful feedback. © koninklijke brill nv, leiden, 2018 | doi 10.1163/15692124-12341295 Counterintuitive Demons Figure 1 1 87 Bronze Pazuzu-Lamaštu plaque, Neo-Assyrian Period. Source: Othmar Keel, Symbolism of the Biblical World: Ancient Near Eastern Iconography and the Book of Psalms (Reprint ed.; Winona Lake: Eisenbrauns, 1997 [1978]): Fig. 91. Cf. Henrike Frey-Anthes, Unheilsmächte und Schutzgenien, Antiwesen und Grenzgänger: Vorstellungen von “Dämonen” im alten Israel (OBO 227; Fribourg: Academic Press, 2007): Fig. 29. Introduction The demons Pazuzu and Lamaštu are among the most fantastic and frightening creatures from ancient Mesopotamia. From a conceptual point of view, they are also among the most complex. As hybrid creatures, the appearance of each figure combines an impressive array of anthropomorphic and zoomorphic properties—from Lamaštu’s bird talons and raging leonine face to Pazuzu’s gigantic wings and serpent-headed phallus (Fig. 1). It would be an understatement to say that the iconographic representation of these demons is marked by a high degree of anatomical complexity. Journal of ancient near eastern religions 18 (2018) 86–110 88 Maiden We start with Pazuzu. In his role as the king of demons, Pazuzu was associated with pestilence and destructive winds, but was also enlisted as a source of protection against evil demons.1 The early history of Pazuzu is somewhat obscure, but he achieves a “sudden, iconographically fully developed appearance” during the Iron Age.2 The earliest attested visual evidence appears in the royal tombs at Nimrud near the end of the 8th century BCE, although most objects date to the 7th or 6th centuries.3 On a well-known bronze statuette, Pazuzu is identified by his canine face, jaws, and teeth, bulging eyes, caprid horns, scaly body, bird talons, massive wings, scorpion tail, and erect serpentine penis (Fig. 2).4 Involving a complex transfer of numerous human and animal body parts, this is an example of what we might call the demon’s full-blown iconographic profile. We encounter a similar full-blown iconography with Lamaštu (Sumerian: dimme), a demonic figure whom Walter Farber describes as “an almost satanic force, a personification of evil and aggressiveness.”5 In material 1 On the demon Pazuzu, see Nils P. Heeßel, Pazuzu: Archäologische und philologische Studien zu einem altorientalischen Dämon (Leiden: Brill, 2002); id., “Evil Against Evil: The Demon Pazuzu,” SMSR 77 (2011): 357–68; F. A. M. Wiggermann, “Pazuzu,” RlA, 10:373–81; idem, “The Four Winds and the Origins of Pazuzu,” in Das Geistige Erfassen der Welt im Alten Orient: Beiträge zu Sprache, Religion, Kultur und Gesellschaft (ed. Claus Wilcke; Weisbaden: Harrassowitz, 2007), 125–66; R. Borger, “Pazuzu,” in Language, Literature and History: Philological and Historical Studies Presented to Erica Reiner (ed. F. Rochberg-Halton; AOS 67; New Haven: American Oriental Society, 1987), 15–32. On the demonic in ancient Mesopotamia and Israel in general, see Anthony Green, “Beneficent Spirits and Malevolent Demons: The Iconography of Good and Evil in Ancient Assyria and Babylonia,” in Visible Religion: Annual for Religious Iconography Volume III: Popular Religion (ed. H. G. Kippenberg et al.; Leiden: Brill, 1984), 80– 105; and Henrike Frey-Anthes, Unheilsmächte und Schutzgenien, Antiwesen und Grenzgänger: Vorstellungen von “Dämonen” im alten Israel (OBO 227; Fribourg: Academic Press, 2007). 2 Heeßel, “Evil Against Evil,” 358. Cf. Wiggermann, “Four Winds,” 125. 3 Heeßel, “Evil Against Evil,” 359. While most prevalent in the Mesopotamian heartland, Pazuzu artifacts have also been found in the Levant. For example, Heeßel (“Evil Against Evil,” 360 n. 9) mentions Pazuzu items from Horvat Qitmit, Tel Beth Shean, and Beth Shemesh (with references). 4 For similar statuettes, see Heeßel, Pazuzu, figs. 2–13; and P. R. S. Moorey, “A Bronze ‘Pazuzu’ Statuette from Egypt,” Iraq 27 (1965): 33–41. 5 Walter Farber, Lamaštu: An Edition of the Canonical Series of Lamaštu Incantations and Rituals and Related Texts from the Second and First Millennia B.C.E. (MC 17; Winona Lake: Eisenbrauns, 2014), 3. On Lamaštu, see also Farber, “Lamaštu,” RlA, 6:439–46; and F. A. M. Wiggermann, “Lamaštu, Daughter of Anu: A Profile,” apud Marten Stol, Birth and Babylonia and the Bible: Its Mediterranean Setting (Groningen: Styx, 2000), 217–52; Mordechai Cogan, “A Lamashtu Plaque from the Judaean Shephelah,” IEJ 45 (1995): 155–61. Both demons are depicted alongside a variety of other Mischwesen on a plaque discovered in Northern Syria, for which see ANEP, 658. Cf. Frey-Anthes, Unheilsmächte und Schutzgenien, fig. 29; Othmar Journal of ancient near eastern religions 18 (2018) 86–110 89 Counterintuitive Demons Figure 2 Bronze Pazuzu statuette, Assyrian or Babylonian, early first millennium BCE. Source: Keel, Symbolism of the Biblical World: Fig. 93. Cf. Frey-Anthes, Unheilsmächte und Schutzgenien: Fig. 31. art, she has the face of a ferocious lion; a naked, spotted fur-covered body; donkey ears and teeth; bird talons; and long, sharp claws (Fig. 3). She is depicted both with and without wings, and occasionally appears with a tail. Finally, she is often shown grasping serpents in her claws and with a piglet and puppy suckling at her breasts. Despite the potent nature of this visual imagery, there are reasons to expect that the material images of Pazuzu and Lamaštu were not identical to the corresponding mental images of these demons. Recent empirical and theoretical insights from the cognitive science of religion (CSR), for example, reveal that highly complex supernatural concepts are difficult to process, remember, and Keel, Symbolism of the Biblical World: Ancient Near Eastern Iconography and the Book of Psalms (trans. J. Hallett; repr. ed.; Winona Lake: Eisenbrauns, 1997), figs. 91–92. Journal of ancient near eastern religions 18 (2018) 86–110 90 Maiden Figure 3 Stone Lamaštu amulet, Assyrian, ca. 800–550 BCE. After Henrike Frey-Anthes, Unheilsmächte und Schutzgenien: Fig. 28. transmit faithfully.6 As a result of this cognitive cost, highly complex concepts are often either forgotten and lost, or else spontaneously simplified into more manageable forms. The present article builds on these findings and proposes that most ancient mental understandings of Pazuzu and Lamaštu likely 6 For overviews of the field, see Ilkka Pyysiäinen, “Cognitive Science of Religion: State-of-theArt,” Journal for the Cognitive Science of Religion 1 (2012): 5–28; Justin L. Barrett, “Cognitive Science of Religion: Looking Back, Looking Forward,” Journal for the Scientific Study of Religion 50 (2011): 229–39; idem, “Exploring the Natural Foundations of Religion,” Trends in Cognitive Sciences 4 (2000): 29–34; Jesper Sørensen, “Religion in Mind: A Review Article of the Cognitive Science of Religion,” Numen 52 (2005): 465–94; Pascal Boyer, “Religious Thought and Behavior as By-Products of Brain Function,” Trends in Cognitive Sciences 7 (2003): 119–24; Pascal Boyer and Brian Bergstrom, “Evolutionary Perspectives on Religion,” Annual Review of Anthropology 37 (2008): 111–30. Journal of ancient near eastern religions 18 (2018) 86–110 Counterintuitive Demons 91 differed in important ways from their more elaborate visual iconographies. Specifically, it is argued that in ancient cognition and memory, these demons would have been represented in a more cognitively “optimal” manner—one that adheres more closely to psychological domain-specific knowledge and expectations.7 Moreover, as I will show, this hypothesis is further supported by a detailed consideration of the full repertoire of iconographic and textual sources, the bulk of which tends to portray Pazuzu and Lamaštu in less complex forms that reflect this cognitive optimum. In brief, then, this paper brings together iconographic, textual, and cognitive perspectives in order to shed new light on the ways people conceptualized these striking demonic figures. Naturally, we cannot access the mental lives of ancient Mesopotamians by scanning their brains with fMRI machines or by inviting them into the lab for psychological experimentation. Nevertheless, in combination with contemporary textual and iconographic data, a cognitively informed methodology affords a glimpse into religious cognition, and therefore offers a fruitful avenue for penetrating the ancient religious imagination.8 2 Cognition and Hybrids In the transmission of culture, not all concepts are created equal. A so-called epidemiological approach to culture and cultural transmission aims to explain why certain concepts, or representations, are more successful than others— that is, why some concepts manage to catch on, stick, and lodge themselves in our minds and memory. Along these lines, the anthropologist Dan Sperber 7 A concise discussion of cognitively optimal religious concepts is found in Harvey Whitehouse, Modes of Religiosity: A Cognitive Theory of Religious Transmission (Walnut Creek, CA: AltaMira, 2004), 29–47. 8 Research at the intersection of cognition and ancient culture is growing, evinced by recent scholarship under the rubric of “cognitive historiography.” See Luther H. Martin, “The Future of the Past: The History of Religions and Cognitive Historiography,” Religio 20 (2012): 155–71; Luther H. Martin and Jesper Sørensen, eds., Past Minds: Studies in Cognitive Historiography (London: Equinox, 2011); Harvey Whitehouse and Luther H. Martin, eds., Theorizing Religions Past: Archaeology, History, and Cognition (Walnut Creek, CA: AltaMira, 2004); as well as the Journal of Cognitive Historiography. Within the field of biblical studies, see István Czachesz and Risto Uro, eds., Mind, Morality and Magic: Cognitive Science Approaches in Biblical Studies (Durham: Acumen, 2013). For an excellent discussion of supernatural concepts, cognition, and iconography, see Beate Pongratz-Leisten and Karen Sonik, “Between Cognition and Culture: Theorizing the Materiality of Divine Agency in Cross-Cultural Perspective,” in The Materiality of Divine Agency (ed. B. Pongratz-Leisten and K. Sonik; Berlin: de Gruyter, 2015), 3–69. Journal of ancient near eastern religions 18 (2018) 86–110 92 Maiden has long suggested that the spread of culture can be studied in much the same way as diseases are studied in the field of epidemiology.9 This is because, as Sperber writes, “The human mind is susceptible to cultural representations in the same way as the human organism is susceptible to diseases.”10 When a disease spreads among a population, the result is an epidemic. But when representations spread among a community of human minds, the result is what is commonly referred to as “culture.”11 Or, as Peter Richerson and Robert Boyd succinctly put it, “Culture is (mostly) information stored in human brains.”12 In this view, the package of culture comprises two types of representations. On the one hand, there are mental representations that are located in individual minds as thoughts, beliefs, desires, memories, and so forth; on the other hand are public representations that take the form of utterances, texts, and artifacts.13 Mental beliefs often give rise to public representations, but 9 10 11 12 13 Dan Sperber, Explaining Culture: A Naturalistic Approach (Cambridge: Blackwell, 1996); idem, “Anthropology and Psychology: Towards an Epidemiology of Representations,” Man 20 (1985): 73–89; idem, “The Modularity of Thought and the Epidemiology of Representations,” in Mapping the Mind: Domain Specificity in Cognition and Culture (ed. Lawrence A. Hirschfeld and Susan A. Gelman; New York: Cambridge University Press, 1994), 39–67. Sperber, Explaining Culture, 57. It is worth clarifying that the epidemiology metaphor is conceptual and does not imply negative connotations. Epidemiology seeks to explain macro-scale phenomena at population levels (e.g., an influenza epidemic) as the aggregation of micro-processes at the individual biological level (e.g., individuals contracting and transmitting the flu virus). Whereas this research requires knowledge of human physiology, by analogy the transmission and spread of cultural phenomena entails knowledge about human psychology. See Sperber, Explaining Culture, 2, 56–61. Employing a similar metaphor, the biologist Richard Dawkins coined the term “meme” to denote the cultural equivalent of genes, describing them as “viruses of the mind.” See Richard Dawkins, The Selfish Gene (Oxford: Oxford University Press, 1976), 189–201. This idea was later popularized in Susan Blackmore’s The Meme Machine (Oxford: Oxford University Press, 1999). However, so-called memetic approaches have been trenchantly critiqued; on which see Daniel C. Dennett, Darwin’s Dangerous Idea: Evolution and the Meaning of Life (New York: Touchstone, 1995), 352–60; and the collected essays in Robert Aunger, ed., Darwinizing Culture: The Status of Memetics as a Science (Oxford: Oxford University Press, 2000). See also the informative discussion by Christophe Heintz, “Cognitive History and Cultural Epidemiology,” in Past Minds: Studies in Cognitive Historiography (ed. L. H. Martin and J. Sørensen; London: Equinox, 2011), 11–28. Peter J. Richerson and Robert Boyd, Not By Genes Alone: How Culture Transformed Human Evolution (Chicago: University of Chicago Press, 2005), 61. See also Robert Boyd and Peter J. Richerson, Culture and the Evolutionary Process (Chicago: University of Chicago Press, 1982). Sperber, Explaining Culture, 61–63. Journal of ancient near eastern religions 18 (2018) 86–110 Counterintuitive Demons 93 physical representations are also capable of influencing and modifying mental representations as well. Ultimately, the two forms are inextricably linked in what Sperber describes as “complex causal chains.” As he writes, “Mental representations caused by public productions can in turn cause further public productions, that can cause further mental representations, and so forth.”14 The distinction is crucial, however, especially when comparing public representations, such as ancient iconography and material art, with their corresponding mental representations in human minds. One’s memory of a Pazuzu artifact, for instance, may be influenced by, though not identical to, the actual artifact itself. At present, then, we may simply note that mental representations are shaped not only by interaction with the physical environment and communication with other individuals, but are also constrained by speciesspecific features of our cognitive architecture.15 The latter produce cognitive biases that steer concepts in particular directions, towards what Sperber calls cultural “attractor” positions.16 From an epidemiological perspective, supernatural concepts are especially contagious and effective attractors. The cognitive anthropologist Pascal Boyer argues that, given the way human minds perceive the world and process 14 15 16 Sperber, Explaining Culture, 99. Scholars often draw a distinction between dispositions and susceptibilities: dispositions emerge as adaptive products of natural selection, while susceptibilities emerge as byproducts of the evolutionary process. See Sperber, Explaining Culture, 66–67, and also Dan Sperber and Lawrence Hirschfeld, “The Cognitive Foundations of Cultural Stability and Diversity,” Trends in Cognitive Science 8 (2004): 40–46. On the role of evolved psychology in shaping culture, see further John Tooby and Leda Cosmides, “The Psychological Foundations of Culture,” in The Adapted Mind: Evolutionary Psychology and the Generation of Culture (ed. Jerome H. Barkow, Leda Cosmides, and John Tooby; New York: Oxford University Press, 1992), 19–136. Sperber, Explaining Culture, 98–118. See also Nicolas Claidière, Thomas C. Scott-Phillips, and Dan Sperber, “How Darwinian is Cultural Evolution?” Philosophical Transactions of the Royal Society B 369 (2014): 1–8. Richerson and Boyd (Not By Genes Alone) speak similarly of “biased cultural transmission,” according to which some cultural variants are preferentially adopted over others. It should be noted that epidemiological models of culture primarily concern transmission in oral contexts, where the focus centers on memory and communication. This picture is complicated, however, by the introduction of material modes of representation such as writing and iconography, which may function as external mnemonic aids, or what has been called a form of “external symbolic storage.” On this idea, see Colin Renfrew and Chris Scarre, eds., Cognition and Material Culture: The Archaeology of Symbolic Storage (Cambridge: McDonald Institute for Archaeological Research, 1998). Journal of ancient near eastern religions 18 (2018) 86–110 94 Maiden information, religious concepts are easy to acquire, remember, and transmit.17 As a result, they achieve what is called a “cognitive optimum”—a kind of conceptual Goldilocks zone around which religious concepts tend to cluster. Boyer suggests that optimal supernatural concepts are “minimally counterintuitive” (MCI), in the sense that they only violate in a limited way intuitive ontological expectations about objects in the world.18 Although reflectively articulated culturally specific ontologies may vary dramatically, intuitive ontological understandings are similar in human beings the world over.19 People across cultures, for example, reliably classify their environment into basic categories—such as humans, animals, plants, artifacts, and natural objects—and thereafter develop strong expectations about these ontological domains.20 Yet unlike ordinary objects, supernatural concepts violate the 17 18 19 20 See Pascal Boyer, Religion Explained: The Evolutionary Origins of Religious Thought (New York: Basic Books, 2001); idem, “Religious Thought and Behavior as By-products of Brain Function,” Trends in Cognitive Sciences 7 (2003): 119–24; idem, The Naturalness of Religious Ideas: A Cognitive Theory or Religion (Berkeley: University of California Press, 1994). On the idea that religion is “natural,” see further Robert N. McCauley, Why Religion is Natural and Science is Not (Oxford: Oxford University Press, 2012), 145–221; and Paul Bloom, “Religion is Natural,” Developmental Science 10 (2007): 147–51. The term “counterintuitive” is used in a technical sense to describe, “information contradicting some information provided by ontological categories” (Boyer, Religion Explained, 65). Therefore, representations that are counterintuitive in this limited sense should not be confused with what is either unfamiliar or unreal. Because the ordinary connotations of the word “counterintuitive” may be misleading, Boyer admits that “counterontological” would be a more precise (though cumbersome) alternative. See further Justin L. Barrett, “Coding and Quantifying Counterintuitiveness in Religious Concepts: Theoretical and Methodological Reflections,” Method and Theory in the Study of Religion 20 (2008): 308–38. See Boyer, “Why Evolved Cognition Matters to Understanding Cultural Cognitive Variations,” Interdisciplinary Science Reviews 35 (2010): 376–86. For a discussion of native ancient Mesopotamian views about ontology and the natural world, see Francesca Rochberg, Before Nature: Cuneiform Knowledge and the History of Science (Chicago: University of Chicago Press, 2016), esp. 85–102. These conceptual categories and expectations appear at an early age in development and arise independent of social learning. See Elizabeth S. Spelke and Katherine D. Kinzler, “Core Knowledge,” Developmental Science 10 (2007): 89–96; and the essays in Lawrence A. Hirschfeld and Susan A. Gelman, eds., Mapping the Mind: Domain Specificity in Cognition and Culture (New York: Cambridge University Press, 1994). Barrett (“Coding and Quantifying Counterintuitiveness”) proposes the following five ontological categories: Spatial Entities, Solid Objects, Living Things, Animates, and Persons. These categories correspond, in turn, to five different intuitive “expectation sets” that Barrett labels, respectively: Spatiality, Physicality, Biology, Animacy, and Mentality. Journal of ancient near eastern religions 18 (2018) 86–110 Counterintuitive Demons 95 implicit rules of our intuitive ontology.21 Consider, for example, that a ghost is a person but without a material body, a virgin birth involves a human with special biology, and a talking snake is a normal animal with special psychology.22 Crucially, these concepts have two important cognitive effects. On the one hand, they confound our intuitive expectations and, as a result, become salient and attention grabbing. At the same time, they adhere to the default ontological domain and thereby remain coherent and intelligible enough to be remembered and communicated in their original form. As Ilkka Pyysiäinen aptly states, “It is the intuitive aspects of religious representations that makes them understandable and learnable, but it is the counter-intuitive aspect that makes them religious.”23 Successful religious concepts thus strike a careful balance between being attention grabbing and memorable, which in turn boosts their success in cultural transmission.24 21 22 23 24 See also Pascal Boyer, “Cognitive Tracks of Cultural Inheritance: How Evolved Intuitive Ontology Governs Cultural Transmission,” American Anthropologist 100 (1998): 876–89; idem, “Cognitive Constraints on Cultural Representations: Natural Ontologies and Religious Ideas,” in Mapping the Mind: Domain Specificity in Cognition and Culture (New York: Cambridge University Press, 1994), 391–411. Conceptual domain violations occur in one of two ways: either by a breach of ontological categories, or through the transfer of features across domains. See Pascal Boyer and Charles Ramble, “Cognitive Templates for Religious Concepts: Cross-Cultural Evidence for Recall of Counter-Intuitive Representations,” Cognitive Science 25 (2001): 535–64; Boyer, Religion Explained, 65–74. Ilkka Pyysiäinen, How Religion Works: Towards a New Cognitive Science of Religion (Leiden: Brill, 2001), 21. There is robust empirical support from experimental studies demonstrating enhanced recall for MCI concepts. See Boyer and Ramble, “Cognitive Templates”; Justin L. Barrett and Melanie A. Nyhof, “Spreading Non-natural Concepts: The Role of Intuitive Conceptual Structures in Memory and Transmission of Cultural Materials,” Journal of Cognition and Culture 1 (2001): 69–100; Justin L. Barrett, “Cognitive Constraints on Hindu Concepts of the Divine,” Journal for the Scientific Study of Religion 37 (1998): 608–19; M. Afzal Upal et al., “Contextualizing Counterintuitiveness: How Context Affects Comprehension and Memorability of Counterintuitive Concepts,” Cognitive Science 31 (2007): 1–25; Benjamin Grant Purzycki, “Cognitive Architecture, Humor and Counterintuitiveness: Retention and Recall of MCIs,” Journal of Cognition and Culture 10 (2010): 189–204; Konika Banerjee, Omar S. Haque, and Elizabeth S. Spelke, “Melting Lizards and Crying Mailboxes: Children’s Preferential Recall of Minimally Counterintuitive Concepts,” Cognitive Science 37 (2013): 1251–89. Experimental work by Atran and Norenzayan examine MCI narratives and “belief sets,” rather than discrete concepts. They argue that an optimal recipe consists of a small proportion of MCI concepts sprinkled among a majority of intuitive ones. The former activate interest, while the latter provide a framework to ensure long-term recall. See Ara Norenzayan, Scott Atran, Jason Faulkner, and Mark Schaller, “Memory and Mystery: Journal of ancient near eastern religions 18 (2018) 86–110 96 Maiden In contrast to the cognitive optimum, concepts that include too many domain violations have the opposite effect, resulting instead in a kind of “conceptual overload.”25 Thus, a talking snake is minimally counterintuitive, but a talking snake that is invisible, made of cashmere, exists in all places at once, and gives birth to zebras, would be highly difficult to recall in toto, or to communicate faithfully. For this reason Boyer writes that, “a combination of one violation with preserved expectations is probably a cognitive optimum.”26 3 Visualizing the Demonic From an epidemiological perspective, Pazuzu and Lamaštu are certainly counterintuitive. As hybrid creatures, they violate intuitive biological expectations by mixing species and combining human and animal properties.27 In one sense, then, hybrid animals are, to borrow from Claude Lévi-Strauss’ dictum, concepts that are “good to think.”28 However, as we have seen, several 25 26 27 28 The Cultural Selection of Minimally Counterintuitive Narratives,” Cognitive Science 30 (2006): 531–53; and Scott Atran, In Gods We Trust: The Evolutionary Landscape of Religion (Oxford: Oxford University Press, 2002), 100–107. Boyer and Ramble, “Cognitive Templates,” 546–50. See also Lauren O. Gonce et al., “Role of Context in the Recall of Counterintuitive Concepts,” Journal of Cognition and Culture 6 (2006): 521–47. Boyer, Religion Explained, 86 (emphasis original). In general, this is how most subsequent researchers have understood and applied the theory. For further discussion, see Barrett, “Coding and Quantifying Counterintuitiveness.” On the development of intuitive biology, see Susan Carey, Conceptual Change in Childhood (Cambridge: MIT Press, 1985); Susan A. Gelman, “The Development of Induction Within Natural Kind and Artifact Categories,” Cognitive Psychology 20 (1988): 65–95; Frank C. Keil, “The Birth and Nurturance of Concepts by Domains: The Origins of Concepts of Living Things,” in Mapping the Mind: Domain Specificity in Cognition and Culture (ed. Lawrence A. Hirschfeld and Susan A. Gelman; New York: Cambridge University Press, 1994), 234–54. Claude Lévi-Strauss, Totemism (Boston: Beacon, 1963). On symbolism and hybrids, see also Dan Sperber, “Why Are Perfect Animals, Hybrids, and Monsters Food for Symbolic Thought?” Method and Theory in the Study of Religion 8 (1996): 143–69. Note also the fascinating study by David Wengrow, The Origins of Monsters: Image and Cognition in the First Age of Mechanical Reproduction (Princeton: Princeton University Press, 2014). Although utilizing the lens of cultural epidemiology, Wengrow somewhat minimizes the role of cognitive factors in the spread of ancient hybrid representations, arguing instead that the mass dissemination of hybrid imagery occurs only with the appearance of technological and political supports. This conclusion is unfortunate, in my view, since it depends partly on an unnecessary dichotomy between “cognition” and “culture.” Despite these Journal of ancient near eastern religions 18 (2018) 86–110 Counterintuitive Demons 97 images of these demons involve multiple domain violations beyond the attractor position of the cognitive optimum, and therefore should be regarded as more than just minimally counterintuitive. Based on physical appearance alone, then, these artifacts of Pazuzu and Lamaštu involve multiple violations of intuitive ontology and therefore can be regarded as maximally or radically counterintuitive. From a cognitive perspective, however, there are some initial reasons for thinking that not all conceptions of Pazuzu and Lamaštu were identical to these full-blown, radically counterintuitive iconographies. First, it is important to recall the distinction between public and mental representations, and to reiterate that the two are not necessarily identical. In short, material images are not the same as mental images.29 They are different phenomena that involve different cognitive processes. This was recognized already by the psychologist Rudolf Arnheim, who problematized the misguided idea that “mental images are faithful replicas of the physical objects they replace.”30 Given the conceptual complexity of some Pazuzu and Lamaštu iconography, there is prima facie reason to think that these public, material representations differed from their mental representations in ancient minds and memory. Second, there is a related distinction between what cognitive scientists refer to as intuitive and reflective beliefs and processes.31 So-called dualprocessing theories contrast two modes of thinking: (1) explicit, reflective, and (mostly) conscious cognition on the one hand, and (2) implicit, intuitive, (mostly) non-conscious cognition on the other.32 Reflective cognitive activity involves 29 30 31 32 reservations, I share Wengrow’s interest in the nexus of material culture and ancient cognition. Yet, unlike his study, a focus of the present paper is to explore how the material mechanisms discussed by Wengrow give rise to radically counterintuitive ancient imagery. In English, the term “image” is commonly used for both mental and visual objects. However, this slippage in terminology only obscures important cognitive differences between public and mental representations, in terms of processing and modes of transmission. In general, it is tricky to infer mental representations based on explicit public representations alone, whether in the form of texts or material artifacts. See W. J. T. Mitchell, “What is an Image?” New Literary History 15 (1984): 503–37. Rudolf Arnheim, Visual Thinking (Berkeley: University of California Press, 1969), 102. See Nicolas Baumard and Pascal Boyer, “Religious Beliefs as Reflective Elaborations on Intuitions: A Modified Dual-Process Model,” Current Directions in Psychological Science 22 (2013): 295–300. The literature on this topic is vast, but see J. St. B. T. Evans, “Dual-Processing Accounts of Reasoning, Judgment and Social Cognition,” Annual Review of Psychology 59 (2008): 255– 78; idem, “Intuition and Reasoning: A Dual-Process Perspective,” Psychological Inquiry Journal of ancient near eastern religions 18 (2018) 86–110 98 Maiden slow, deliberate thinking, while intuitive cognition is fast and automatic— a distinction captured in the title of Daniel Kahneman’s recent book, Thinking, Fast and Slow.33 Artifacts of Pazuzu and Lamaštu are the products of deliberate, reflective artistic activity. As material art, they likely functioned as a valuable medium for creating, stabilizing, and transmitting complex concepts, which in turn paved the way for the development of elaborate artistic traditions and radically counterintuitive imagery.34 I would even suggest that, because optimal concepts must strike a balance between being salient, memorable, and communicable, material art (in much the same way as written texts) can function as a mnemonic aid that helps bypass the ordinary demands of cognition and memory. As such, careful consideration of the role of ancient material media would be vital to developing and refining current cognitive theorizing regarding the shaping of mental representations and the elastic constraints placed on religious cognition and imagination. Nevertheless, at present the important thing to bear in mind is that there is a difference, in terms of cognitive processing, between this reflective imagery on the one hand, and concepts that arise through more intuitive cognition on the other. As we have seen, the downside of highly counterintuitive representations is that they exert a greater cognitive burden, and consequently, are difficult to remember and communicate in their current form. Thus, for instance, even if a Pazuzu artifact is rendered in a complex manner, it may not correspond precisely to how the Pazuzu 33 34 21 (2010): 313–26; idem, “In Two Minds: Dual Process Accounts of Reasoning,” Trends in Cognitive Sciences 7 (2003): 454–59; G. Keren and Y. Schul, “Two is Not Always Better than One: A Critical Evaluation of Two-Systems Theories,” Perspectives on Psychological Science 4 (2009): 533–50; A. W. Kruglanski, G. Gigerenzer, “Intuitive and Deliberate Judgments Are Based on Common Principles,” Psychological Review 118 (2011): 97–109. With regard to religion, in particular, see also Nicolas Baumard and Pascal Boyer, “Religious Beliefs as Reflective Elaborations on Intuitions: A Modified Dual-Process Model,” Current Directions in Psychological Science 22 (2013): 295–300. Daniel Kahneman, Thinking, Fast and Slow (New York: Farrar, Straus and Giroux, 2011). The different modes of thought are sometimes termed System 1 (intuitive) and System 2 (reflective). Intuitive capacities appear early in childhood development, do not require explicit instruction or cultural support, and involve a set of skills common to all humans, but which operate outside of conscious inspection. A modified dual-processing model is offered by McCauley, who further differentiates between two types of natural cognition: (1) “maturationally natural cognition,” which emerges in the course of normal cognitive development, and (2) “practiced naturalness,” a cognitive mastery achieved through practice, experience, and expertise in a particular domain. See McCauley, Why Religion is Natural, 11–30. See also Renfrew and Scarre, eds., Cognition and Material Culture. Journal of ancient near eastern religions 18 (2018) 86–110 Counterintuitive Demons 99 concept is mentally stored and recalled within an individual’s mind. Indeed, research indicates that different levels of representation coexist within individual minds, and that the further a concept strays from intuitive expectations, the more likely it is to be transformed into a cognitively optimal form during real-time thinking.35 Far from being merely theoretical, the distinctions between mental/public representations and intuitive/reflective beliefs, are relevant when we take into consideration the full corpus of ancient Pazuzu and Lamaštu artifacts, many of which illustrate the tendency to favor more optimal representations. Although the full-blown, radically counterintuitive images discussed above are striking, they are also exceptional and therefore potentially misleading. They are a kind of prototype that exhibits the maximum number of exotic features—what Nils Heeßel calls, with respect to Pazuzu, the demon’s “Idealikonographie.”36 However, the visual record is far from uniform, and in fact is dominated by images that are less idealized and less complex. The best example in this regard is the large assemblage of apotropaic Pazuzu heads (Figs. 4–7).37 These amulets are found in a variety of materials, including bronze, iron, gold, glass, and bone, but most are made of terracotta. As Wiggermann notes, “Pazuzu’s apotropaic power resides in his head,” and many of these objects have holes and suspension loops for attaching a chain or string, allowing women to wear them around their necks during pregnancy 35 36 37 This occurs during the phenomenon known as “theological incorrectness,” wherein people abandon complex theological conceptions of God and spontaneously default to a more intuitive person-like understanding. See Justin L. Barrett and Frank C. Keil, “Conceptualizing a Non-Natural Entity: Anthropomorphism in God Concepts,” Cognitive Psychology 31 (1996): 219–47; idem, “How Ordinary Cognition Informs Petitionary Prayer,” Journal of Cognition and Culture 1 (2001): 259–69; idem, “Theological Correctness: Cognitive Constraint and the Study of Religion,” Method and Theory in the Study of Religion 11 (1999): 325–39; Jason Slone, Theological Incorrectness: Why Religious People Believe What They Shouldn’t (New York: Oxford University Press, 2004). There is nothing strictly “theological” about the phenomenon; the larger issue centers on the difference between intuitive and reflective beliefs and their differential effects on cognition and memory. Heeßel, Pazuzu, 11. See, Heeßel, Pazuzu, 41–50, 115–69, 180–93 and figs. 61–163; Eva Andrea Braun-Holzinger, Figürliche Bronzen aus Mesopotamien (Prähistorische Bronzefunde I/4; Munich: Beck, 1984), figs. 277–279; E. Klengel-Brandt, “Pazuzu-Köpfe aus Babylon,” Forschungen und Berichte 12 (1970): 37–40, plates. 3–4; Agnes Spycket, From the Collections of Elie Borowski: The Human Form Divine (Jerusalem: Bible Lands Museum), figs. 145a-c; Uri Gabbay, “A Collection of Pazuzu Objects in Jersualem,” RA 94 (2001/2): 149–54. Wilfred G. Lambert, “Inscribed Pazuzu Heads from Babylon,” FuB 12 (1970): 41–47. Journal of ancient near eastern religions 18 (2018) 86–110 100 Maiden Figures 4–5 Pazuzu heads, carnelian (left) and terracotta (right). Source: Uri Gabbay, “A Collection of Pazuzu Objects in Jerusalem,” Revue d’Assyriologie et d’Archéologie Orientale 94 (2001/2): 149–154: Figs. 2, 4. in order to ward off evil spirits that might harm the unborn child (Figs. 8–10).38 Clay molds of Pazuzu heads indicate that they were mass-produced and, overall, the number of heads far exceeds the number of full-body figurines.39 These objects indicate that Pazuzu was identified above all by the head and face. This was the demon’s most essential and iconic feature. Indeed, the prevalence of Pazuzu heads shows that, as Heeßel notes, the head was used pars pro toto for the demon as a whole.40 By omitting the composite body and isolating the head alone, these material representations of Pazuzu become considerably less counterintuitive. That is, rather than incorporating an array of discrete animal features, they are marked instead by a more generic monstrous countenance. Indeed, it is worth noting that despite theriomorphic features 38 39 40 Wiggermann, “Four Winds,” 125. Heeßel, “Evil Against Evil,” 358; Wiggermann, “Four Winds,” 125. Heeßel, “Evil Against Evil,” 358. The head of the leonine-faced Humbaba is likewise used as an iconic symbol for the figure itself, both in text and iconography. For discussion see Sarah B. Graff, “The Head of Humbaba,” Archiv för Religionsgeschichte 14 (2013): 129–42. Journal of ancient near eastern religions 18 (2018) 86–110 Counterintuitive Demons Figures 6–7 101 Pazuzu heads, marble (left) and clay (right). Source: Nils P. Hee ß el, Pazuzu: Archäologische und philologische Studien zu einem altorientalischen Dämon (Leiden: Brill, 2002): Figs. 94, 156. such as canine mouth and teeth, the facial configuration and basic shape of the head remain markedly humanlike.41 For example, Pazuzu heads tend to be vertically oblong in shape, sitting in superior vertebral position atop the neck, with frontal facing eyes, minimal nasal protuberance, and distinctively human ears. Thus, many Pazuzu heads, while frightening and grotesque, maintain an arguably human appearance. The Lamaštu iconography is more limited, but while there is variation in the demon’s head and face, the distinctive identifying feature is always her claws 41 Moreover, rather than having clearly discernable, discrete human and animal features, the purpose of Pazuzu’s head and face is instead to showcase more generally the demon’s “malformed inhuman ugliness” (Wiggermann, “Four Winds,” 125). Journal of ancient near eastern religions 18 (2018) 86–110 102 Maiden Figures 8–10 Bronze Pazuzu heads. Source: Hee ß el, Pazuzu: Figs. 61, 74, 81. Journal of ancient near eastern religions 18 (2018) 86–110 Counterintuitive Demons 103 Figures 11–12 Lamaštu amulets. Sources: Fig. 11 (left): Frey-Anthes, Unheilsmächte und Schutzgenien: Fig. 27; Fig. 12 (right): Walter Farber, Lamaštu: An Edition of the Canonical Series of Lamaštu Incantations and Rituals and Related Texts from the Second and First Millennia B.C.E. (Mesopotamian Civilizations 17; Winona Lake: Eisenbrauns, 2014): Fig. 3. (Figs. 11–12). Beyond that, however, the rest of Lamaštu’s body remains quite humanlike.42 As with Pazuzu, she is always depicted in an upright anthropomorphic posture, indicating that the form of both demons was modeled on a human foundation characterized by bipedalism (e.g., Figs. 13–14). In sum, when the full array of images of Pazuzu and Lamaštu is taken into account, it becomes clear that, notwithstanding a few exceptional artifacts, most material representations of the demons are limited in their conceptual complexity. The majority of images display at best only a modest level of counterintuitiveness. Pazuzu is mostly embodied with a largely anthropomorphic head and face, while Lamaštu is characterized above all by her sharp claws. And most importantly, both demons have a predominantly person-like form. These simplified images, despite being the result of deliberate artistic activity, therefore reflect a more intuitive and optimal conceptualization of the demons. Since ancient artists tended to craft less idealized, more optimal 42 Wiggermann, “Lamaštu,” 232; cf. Farber, Lamaštu, 5. Journal of ancient near eastern religions 18 (2018) 86–110 104 Maiden Figure 13 Bronze Pazuzu statuette, Iraq, Neo-Assyrian Period. Source: Joel LeMon, Yahweh’s Winged Form in the Psalms: Exploring Congruent Iconography and Texts (OBO 242; Fribourg: Academic Press, 2010): Fig. 2.18. Cf. Hee ß el, Pazuzu: Fig. 5. Figure 14 Lamaštu amulet, Assyria. After Frey-Anthes, Unheilsmächte und Schutzgenien: Fig. 26. Cf. Farber, Lamaštu: Fig. 14. Journal of ancient near eastern religions 18 (2018) 86–110 Counterintuitive Demons 105 material renderings of the demons, we may infer that most ancient minds likewise gravitated towards more optimal mental representations as well. The visual evidence therefore coincides with cognitive theorizing and, together, brings us closer to how the demons would have been mentally visualized and remembered, particularly in the absence of complex, maximally counterintuitive artistic forms. 4 The Textual Evidence We can now compare the iconographic imagery with textual descriptions of Pazuzu and Lamaštu attested in ancient literary sources, which contribute to a thicker picture of how ancient people imagined these demons. As we shall see, although the textual and visual evidence match up in some ways, they also diverge in important respects. Pazuzu texts mostly consist of short, standardized spells inscribed on small apotropaic objects.43 For example, the following amulet incantation is presented as the demon’s first person speech: I am Pazuzu, son of Hanbu, king of the evil phantoms. I ascended the mighty mountain that quaked. The winds that I went against were headed toward the west. One by one I broke their wings.44 Other incantations include a long list of verbal formulations detailing Pazuzu’s destructive power: he parches the land, withers trees, frosts the mountains, and attacks young men and women.45 One text concludes by appearing to address the demon in the second person as follows: 43 44 45 There are two standard incantations: (1) standard incantation A, written in the first-person perspective of Pazuzu, and (2) standard incantation B, addressed to Pazuzu in the second-person. See Heeßel, Pazuzu, 55–80. Benjamin R. Foster, Before the Muses: An Anthology of Akkadian Literature (3rd. ed.; Bethesda, MD: CDL, 2005), 978. Cf. Heeßel, Pazuzu, 62. Wiggermann (“Four Winds,” 126) suggests that the quaking mountains in this passage act as a metaphor for the belly of a pregnant woman, while the hostile winds represent Lamaštu. See Foster, Before the Muses, 977; Heeßel, Pazuzu, 61. Journal of ancient near eastern religions 18 (2018) 86–110 106 Maiden Agony of mankind, disease of mankind, suffering of mankind. Do not enter the house I enter, do not come near the house I come near, do not approach the house I approach!46 These incantations clearly reflect Pazuzu’s destructive and malevolent character, but what is most striking about the textual evidence is the complete absence of any description regarding the demon’s physical appearance. There is no mention of his wings, tail, body, or face. Instead, he is described simply as “fierce” or “furious,” while corporeal details are conspicuously lacking. It is possible that the above texts implicitly allude to the demon’s wings and power of flight, but even this is unclear since wings are never explicitly mentioned. In general, references to the demon’s physical form are either vague or nonexistent and, in this regard, the literary evidence therefore shares little in common with the idealized iconography. Lamaštu texts are more plentiful and include various spells from the second and first millennia BCE, the most important of which is the canonical series of Lamaštu incantations and rituals from the first millennium.47 As an evil figure, the demon Lamaštu was said to attack women in pregnancy and to snatch and torture newborn babies.48 Unlike Pazuzu, the texts are more forthcoming about her appearance. For example, an Old Babylonian incantation describes her origins by stating, “Anu begot her, Ea reared her. Enlil doomed her the face of a lioness. She is furious.”49 A first millennium text provides a more detailed anatomical description: Her head is a lion’s head, donkey’s teeth are [her] teeth?. Her lips are a gale and spread death. From deep in the mountains she came down, roaring like a lion, whimpering all the time like a bitch.50 This text explicitly refers to Lamaštu’s animal features and hybrid form. Throughout the corpus, the most common feature mentioned is her leonine 46 47 48 49 50 Heeßel, “Evil Against Evil,” 365. See now Farber, Lamaštu. Although primarily a connoisseur of infants and children, she is also said to prey on grown men, the elderly, and domesticated animals. See further Wiggermann, “Lamaštu”; and Konrad Volk, “Kinderkrankheiten nach der Darstellung babylonisch-assyrischer Keilschrifttexte,” Or 68 (1999): 1–30. Foster, Before the Muses, 173. Cf. Farber, Lamaštu, 281. Farber, Lamaštu, 179. Cf. Foster, Before the Muses, 982. Journal of ancient near eastern religions 18 (2018) 86–110 Counterintuitive Demons 107 face.51 However, other features—such as her bird talons, wings, and tail—are attested either rarely or not at all, and are found scattered across different texts rather than consolidated into a single extended physical description. Moreover, as with Pazuzu, references to her appearance are often vague and have nothing to do with her hybrid body; these include references to her disheveled hair, wild face, or exposed breasts.52 These physical traits, while bizarre, are not counterintuitive in the technical sense, much less radically counterintuitive. Although Mesopotamian authors considered animal traits integral to Lamaštu’s identity, the incantations spend far more time reciting the demon’s malicious deeds and how to avert them. Indeed, when the textual corpus is viewed as a whole, very little of this literature is devoted to Lamaštu’s physical appearance. Rather, we encounter frequent pithy reports of her doings—for example, “she lives among dung,” “she blocks oxen and donkeys,” “she kills young men and girls,” “she yanks out the pregnant woman’s baby,” and so on. Passages in the canonical Lamaštu series are often quite detailed about her activities while at the same time remaining silent about her physical appearance: [She is] clad in scorching heat, fever, cold, frost, (and) ice. The root of the licorice tree, the seed(s) of the chaste tree, the fruit of the poplar, pride of the river meadow, she spoiled. By crossing a river, she makes it murky. By leaning against a wall, she smears (it) with mud. When she has seized an old man, they call her “The Annihilator.” When she has seized a young man, they call her “The Scorcher.” When she has seized a young woman, they call her “Lamaštu.” When she has seized a baby, they call her “Dimme.” Because you (= Lamaštu) came here and attacked his face, took hold of the joints, destroyed the limbs, (are now) consuming the muscles, twisting the sinews, make faces turn green, turn features the way they should not be, cause depression, burn bodies like fire. (Lam. I, lines 62–75)53 51 52 53 See also W. Fauth, “Ištar als Löwengöttin und die löwenköpfige Lamaštu,” WO 12 (1981): 21–36. Farber, Lamaštu, 299. Farber, Lamaštu, 153. Compare also the lengthy incantation of Lam. I, Incantation 5, of which only two of the 118 total lines make any reference to her physical features: “Teeth of a dog are her teeth, talons of an eagle are her talons.” Journal of ancient near eastern religions 18 (2018) 86–110 108 Maiden Here, again, Lamaštu is described with vivid verbal expressions: she crosses rivers, seizes and attacks her victims, destroys their limbs, consumes their muscles, twists their sinews, and burns their bodies. Elsewhere, she is said to slither like a serpent through doorways and windows in search of prey.54 Overall, as with the Pazuzu texts, Lamaštu is primarily depicted using verbs that describe what she does rather than with adjectives describing how she appears.55 Unlike the full-blown “ideal” iconography, the textual evidence does not highlight the radically counterintuitive anatomy of the demons, but opts instead for a more cognitively optimal description. It is worth noting that the ritual instructions accompanying the Lamaštu incantations also hint that the demon’s physical appearance was only of secondary importance. In these ritual prescriptions, individuals are instructed to fashion clay figurines of Lamaštu to be used during the ritual procedures. Yet interestingly, no specific details are provided about how the figurine should be made or what the product should look like—one is simply told to “make a figurine of Lamaštu” (ṣalam Lamaštu teppuš) or to “make a daughter of Anu from clay” (mārat Anu ša ṭīdi teppuš).56 The terse and elliptical instructions either assume that people already knew how to make a Lamaštu figurine, or else that the figurine’s material details were irrelevant to the efficacy of the ritual. Overall, then, despite a few areas of convergence with the visual images of Pazuzu and Lamaštu, the literary evidence offers its own distinct presentation of these demons. Whereas the iconography accentuates the demons’ unnatural and fantastic physical form, the literary sources focus on action over appearance, emphasizing the demons’ destructive deeds at the expense of their anatomy. Thus, like the majority of Pazuzu and Lamaštu iconography, the textual sources reflect a far more cognitively optimal representation of the demons. Now, of course, texts and artifacts are different modes of material expression, and each has its own peculiar constraints. Material artifacts, for instance, are limited in their ability to convey a figure’s movement and action. Therefore, 54 55 56 “She comes in by the window, slithers in by the door pivot. She enters the house (and) leaves the house (as she pleases, saying:) ‘Bring me your sons: I want to suckle (them), and for your daughters I want to be their nanny. In the mouth of your daughters I want to place my breast!’” See Farber, Lamaštu, 299. Moreover, as with Pazuzu, she is most often described with generic adjectives that are silent on her appearance. For example, see Lam. II, lines 154–5: “Cruel, raging, malicious, rapacious, violent, destructive, aggressive is the Daughter-of-Anu” (Farber, Lamaštu, 179). See, e.g., Farber, Lamaštu, Lam. I, lines 47, 94, Lam. III, line 119. The most detailed instruction merely states, “You [cover] the head [of the molded figurine] with hair” (qaqqassa perta t[ukattam?]). Note also that the few ritual instructions for making Pazuzu heads are equally vague, which likely explains the idiosyncratic variety encountered in the morphology of these artifacts. Journal of ancient near eastern religions 18 (2018) 86–110 Counterintuitive Demons 109 the dual portrayals of Pazuzu and Lamaštu in text and iconography might be explained in part by the limitations and emphases of each medium. However, the differences still reflect and communicate divergent conceptualizations of the monsters.57 In particular, our comparison of text and image has shown that full-blown, radically counterintuitive versions of Pazuzu and Lamaštu do not appear in the textual record, and therefore did not govern each and every understanding of these creatures. Rather, as we saw with much of the iconography, literary descriptions steer towards less counterintuitive, more optimal portrayals. What is more, literary sources represent the demons in largely ordinary, anthropomorphic terms as intentional human-like agents with capacities such as seeing, hearing, and thinking.58 That is, they are represented as beings with minds, social agents with whom one can interact.59 As agents with minds, Pazuzu and Lamaštu were treated as members of the human social system, based upon the principles of exchange and reciprocity. This is reflected in the collection of amulets, spells, and incantations addressed to the demons, which show that people attempted to appease, repel, supplicate, or win favor from them. People offered gifts to Lamaštu, including cosmetics, shoes, and pots, and she is depicted travelling on humanlike means of transportation, such as boats or donkeys. Lastly and most notably, ancient texts impute to these demons the unique capacity of human language and speech. All of this activity is predicated on the representation of the demons as agents with intelligible person-like minds and cognition.60 Ancient Mesopotamian ritual specialists and laypeople alike could therefore expect, automatically and at no extra cognitive cost, Pazuzu and Lamaštu to have beliefs, desires, and preferences, all of which would crucially influence the potential actions and decisions of these demonic figures. 57 58 59 60 This holds true even if the textual and visual media were intended to supplement and inform one another. For even when it comes to Pazuzu or Lamaštu texts inscribed upon objects of the demons, both representations (image and text) would only be available to literate individuals who could read the inscriptions. For this reason, it is important not to interpret one type of evidence in light of the other, or to harmonize the different media. On the notion of intentional agency, see Ilkka Pyysiäinen, Supernatural Agents: Why We Believe in Souls, Gods, and Buddhas (Oxford: Oxford University Press, 2009). See Todd Tremlin, Minds and Gods: The Cognitive Foundations of Religion (New York: Oxford University Press, 2006), 112–13. Supernatural agency in relation to the demonic is apparent, too, in the textual representation of disease and illness. As Geller notes with respect to Lamaštu, the explanation for illness befalling (literally “seizing”) a child was that a “Lamaštu-demon has chosen him.” The demon Lamaštu “chooses” (ḫiāru) a baby as a victim. See Markham J. Geller, Ancient Babylonian Medicine: Theory and Practice (Malden, MA: Wiley-Blackwell, 2010), 196 n. 206. Journal of ancient near eastern religions 18 (2018) 86–110 110 5 Maiden Conclusion By way of conclusion, we may emphasize that there was not simply one way of imagining the ancient demons Pazuzu and Lamaštu. Some artifacts combine a surplus of human and animal properties in their representation of the demons, in what might be called full-blown iconographic profiles. In cognitive terms, these images are maximally counterintuitive and as a result, they are expected to be cognitively costly when it comes to cognitive processing and memory. However, this was not the only possible presentation and many objects are instead far less complex and more cognitively optimal in form. Moreover, and importantly, the bulk of the literary sources also depict Pazuzu and Lamaštu in largely anthropomorphic terms. In short, the majority of ancient iconographic and textual evidence, each in their own way, opt for a cognitively optimal conception of the demons. The iconographic and textual records therefore converge with cognitive theorizing and reflect a more intuitive, optimal understanding of the demons, one that more closely approximates how they might have been mentally represented and remembered most of the time in actual religious reasoning, intuitive cognitive thinking, and everyday life.61 61 The distinction between intuitive and reflective modes of thinking is again relevant, since different contexts and situations will activate or require one mode or the other. Barrett (“Coding and Quantifying,” 330) illustrates how a single reflective concept can have different degrees of relevance depending on the situation, writing that, “It may be that an idea with great inferential potential in off-line processing (e.g., relativity theory for a physicist in the lab) may have rather poor on-line inferential potential (e.g., relativity theory for a physicist driving in rush hour traffic).” We can repackage this idea with reference to the conceptions of Pazuzu or Lamaštu by stating, “An idea with great inferential potential in off-line processing (e.g., the iconographic representation of Pazuzu for an artist designing an amulet) may have rather poor on-line inferential potential (e.g., the mental representation of Pazuzu when performing an incantation, treating an illness, or attributing misfortune).” Ancient situations that necessitated fast, intuitive thinking would have been more likely to activate the optimal mental versions of Pazuzu or Lamaštu. Journal of ancient near eastern religions 18 (2018) 86–110