Paul’s Iconic Christ among Mediterranean Cult Statues: A Comparison of Divine Images

Abstract

While scholars continue to debate whether Paul ever explicitly called the messiah a god, there can be no question that he called the messiah an image of a god (2 Cor. 4.4). Even so, there has been no shortage of comparative studies that situate Paul’s messiah among ancient Mediterranean gods, including the Jewish god himself, whereas what is arguably more proximate comparanda for Paul’s iconic claims tend not to be taken into consideration, namely, ancient Mediterranean images of gods. Taking for granted that Paul’s god—the Jewish god—was a Mediterranean god among others, in this article I redescribe Paul’s messiah as a Mediterranean cult image among others. By exploring the relation between gods and their images, between divinity and materiality, with images of gods as material media that make present the gods they image, new light is cast on the debated relation between the Jewish god and his christological image in Paul’s letters.

Introduction: The Status Quaestionis of the Iconic Christ

In the letter that has come to be known as 2 Corinthians, Paul tells his gentile followers that the Jewish messiah is an image of the Jewish god. ¹ ‘The god of this age has blinded the minds of the unfaithful, to keep them from seeing the light of the good news of the glory of the messiah, who is an image of god’ (ὅς ἐστιν εἰκὼν τοῦ θεοῦ, 2 Cor. 4.4). As such, it is in the materiality of this messiah that the luminous presence of the Jewish deity is seen and known. ‘For it is the god who said, “Let light shine out of darkness”, who has shone in our hearts to give the light of knowledge of the glory of god (τῆς δόξης τοῦ θεοῦ) in the face (προσώπῳ) of the messiah’ (2 Cor. 4.6 NRSV, slightly modified). While this is the only instance in the undisputed letters of Paul where Jesus is explicitly dubbed a divine image (implicitly, 1 Cor. 15.49; 2 Cor. 3.18; Rom. 8.29; cf. Col. 1.15; 3.10), the basic concept has had a disproportionate reception history, informing centuries of christological debate about the humanity of Jesus and its relation to divinity. ² In this article, I bring new comparanda to an old debate by drawing from a hitherto neglected body of relevant scholarship, namely, classical and art-historical studies on images of gods in the visuality of everyday Greeks and Romans in ancient Mediterranean material culture. ³ The result is to open up fresh avenues for understanding Paul’s iconic claims about the Jewish messiah in light of wider ritual perceptions of divine images and their material entanglement with the gods they image. ⁴

In recent decades, a strong majority of scholars have appealed to Paul’s ancestral Jewish heritage to elucidate the alleged ‘origins’ of his so-called ‘image christology’. Some have pointed to ‘merkabah mysticism’ and visionary texts about divine ‘glory’ (כבוד‎, δόξα) in the Hebrew Bible, Septuagint, or hekhalot literature (most famously, Kim 1981: 136–268; 2002: 165–213; cf. Newman 1992). ⁵ Others have pointed to a ‘wisdom’ or ‘word’ tradition attested in Pseudo-Solomon and Philo of Alexandria (most recently, Kugler 2020: 111–48). ⁶ And still others have pointed to a wider constellation of early Jewish texts about humanity writ large, involving traditions about Adam, Moses, Israel, the high priest, the son of man, or some combination of any or all the above (most fully, Fletcher-Louis 1999, 2000, 2002, 2004a, 2004b, 2007, 2015). ⁷ Whatever one makes of such claims, it is clear that, with only a few notable exceptions, ⁸ the older agenda of the religionsgeschichtliche Schule to pinpoint the Greek and Roman philosophical ‘roots’ of Paul’s claims—be they gnostic redeemer myths (Bousset 1913; Jervell 1960; Schwanz 1970) or Middle Platonic intermediaries (Eltester 1958; Colpe 1961; Larsson 1962)—has been outpaced by that of the so-called ‘new’ Schule, which, in the words of Wayne Meeks (2001: 21), is ‘to demonstrate the distinctiveness of Christianity against its pagan environment’ by appealing to ‘the distinctiveness it shared’ with ancient Israel and/or early Judaism. ⁹ Indeed, even Chris Kugler (2020: 24), who frames his recent analysis as developing insights from the ‘old’ Schule by highlighting Middle Platonic elements in Jewish wisdom traditions, nevertheless insists on ‘the crucial ways in which Paul’s imago Dei theology differs radically from the philosophical imago Dei traditions’.

In this article, by contrast, I am not concerned with the questions that so preoccupied the religionsgeschichtliche Schule in both its older and newer varieties, that is, questions fixated on origins, borrowings, backgrounds, or influences (whether Greek, Roman, Jewish, or otherwise). ¹⁰ Rather, the present article takes its point of departure with what Jennifer Eyl (2019: 2) has argued should ‘go without saying’: Paul was intelligible and recognizable to his own gentile followers within the wider world of ancient Mediterranean religion. ¹¹ In other words, I treat Paul not as a singular, isolated figure set against any one particular ‘background’ but as ‘one among many’, ¹² whose letters provide ‘evidence for ancient Mediterranean religion’ more broadly as one peculiar case relative to widely recognizable modes of social practices and cognitive processes involving the gods and their presence among mortals (Wasserman 2022: 217, 219). ¹³ So, while I take for granted that Paul lived and died a Jew ‘within Judaism’, and that his letters should be treated as primary evidence for Jewish religious expertise promoting a foreign god—the Jewish god—among gentiles (Wendt 2016: 146–89), I further insist that Paul, like other early Jews, actively participated in and creatively contributed to widely shared social practices, ritual perspectives, visual resources, and material technologies that were otherwise common to Greeks and Romans, no less than Jews. ¹⁴

By introducing comparanda from the study of Greek and Roman cult images, then, I intend no categorical dichotomy with Jewish sources. On the contrary, it is my contention that the latter (Paul’s letters included) are themselves better understood as operating within the same set of variously linked social practices and cognitive processes involving divine images as the former—deliberately cutting across traditionally disparate categorizations to treat the whole as ethnically coded examples of ancient Mediterranean religion. ¹⁵ So, while I begin by focusing on images in the visuality of everyday Greeks and Romans in the pragmatic flow of my argument, I do so with the aim of defining the redescriptive terms that I go on to apply to Jews, too. ¹⁶ Far from reinscribing an implicit divide between them, this offers one way of demonstrating that Greeks, Romans, and Jews alike can be understood within the porous and permeable boundaries of ancient Mediterranean religion, working towards what Eyl (2020) has called ‘a kind of integrated Greco-Romano-Judeanness’.

This article thus proceeds as a comparative experiment of redescription and contextualization. ¹⁷ In what follows, I construe Paul’s claim that the messiah is an image of the Jewish god in terms of concepts and categories derived from wider ritual perspectives involving images of other Mediterranean gods among everyday Greeks and Romans. I do so in the attempt to pose historical boundaries for a disciplined imagination of how to understand the relation between the Jewish god and his christological cult image within the wider world of ancient Mediterranean religion. In the first half of this article, I draw from recent classical and art-historical scholarship on ancient Mediterranean cult statues in order to probe the relation between gods and their images. This then supplies the concepts and categories for a redescription of Paul’s iconic Christ in similar terms in the second half of this article, in which I offer an ‘iconographic’ or ‘materialist’ approach to Paul’s claims about the messiah as an image of god. In so doing, I argue that the materiality of the messiah can be viewed as a technology of divine epiphany comparable to the statuary of cult images, both of which serve as material media of presencing gods among mortals. In this way, I hope to show that, notwithstanding Paul’s polemics against images of gentile gods elsewhere in his letters (e.g., 1 Thess. 1.9–10; 1 Cor. 8.1–11.1; 12.2; 2 Cor. 6.14–7.1; Rom. 1.18–32), his iconic claims about the messiah as an image of the Jewish god can be understood to operate within and innovate upon widespread ritual perceptions of the visuality of the gods and epiphanic strategies involving the presence of those gods vis-à-vis their images.

Ancient Mediterranean Cult Statues: Deity, Statuary, and Material Epiphany

In recent decades, the so-called ‘material turn’ has continued to gain prominence across the humanities, not least in the study of Mediterranean antiquity. Following increased attention to the materiality of ‘things’, the production of presence, and the co-constitutive assemblage of nonhuman objects and human/superhuman agents, there has been a resurgence of interest among classicists and art historians in images of gods in ancient Mediterranean cult and material culture. As scholars now regularly acknowledge, previous scholarship as far back as the father of classical art history and archaeology himself—Johann Joachim Winckelmann (1764)—has long been prejudiced by the legacy of Jewish and Christian polemics against ‘idolatry’ (Donohue 1988: 85–103, 175–231; Rives 2007: 32–34; Squire 2009: 15–89). Alice Donohue (1997: 42) put it bluntly: ‘On the whole, the scholarly imagination has been iconoclastic, and few analyses have moved far enough away from the iconoclastic position to entertain seriously the possibility of approaching the images as the images that the Greeks insisted they were’. ¹⁸

Over the past twenty years, however, scholars have increasingly wrestled with ‘questions of agency and the degree to which Graeco-Roman viewers ascribed to material artefacts the capacity for presence, sentience, and the exertion of power’, precisely in the attempt to interrogate ‘emic conceptions of materialist thought’ about images among Greeks and Romans themselves (Gaifman and Platt 2018: 406–7). ¹⁹ This has necessarily led to a critical re-examination of a dizzying array of literary, artistic, and archaeological evidence, from the archaic period well into the so-called Second Sophistic, informed especially by theoretical advancements in visual anthropology, cognitive archaeology, and materiality studies. ²⁰ Such studies take seriously the material agency of ‘things’, their active participation in public life, and their dynamic entanglement in networks of human, nonhuman, and superhuman actors. As a result, it was once common for classicists—and is still relatively common for biblical scholars ²¹ —to interpret such evidence with reference to a dualistic model of ‘representation’, which firmly maintains that Greeks and Romans did not conflate or equivocate between images (as ‘signifiers’) and gods (as ‘signified’). Many have since come to conclude that this model is largely indebted to later Christian semiotics, which do not adequately reflect the complex material entanglement and continual cognitive renegotiation between gods and their images in the ritual-orientated visuality of everyday Greeks and Romans. ²² In this section, I survey a selection of the evidence reassessed by scholars to probe the relation between deity and statuary, and then outline the now dominant category by which classicists and art historians have come to interpret such evidence and its role in the production of divine presence.

Between Deity and Statuary

The complexity of Greek and Roman ‘visual theologies’ of images is aptly portrayed in ancient art, in which the relation between gods and images is visually depicted in a wide variety of ways (see esp. Gaifman 2016; cf. Scheer 2015). But it is also reflected in ancient ekphrasis, which conveniently interprets the images depicted in visual art. ²³ In a now famous example, Richard Gordon (1979) observes that Pausanias could often write of the ‘statue of x’ (usually, ἄγαλμα + the name of a god in the genitive), but that he just as readily refers to the deity directly. ‘A statue of Aphrodites (Ἀϕροδίτης ἄγαλμα) stood under the first tripod’, Pausanias writes at one place, ‘and Artemis (Ἄρτεμις) under the second’ (Description of Greece 3.18.8). As Gordon (1979: 7) points out, scholars subtly correct the seemingly ‘false’ equivocation between the statue of Artemis and the goddess Artemis herself by glossing the text to read: ‘The first tripod is supported by an image of Aphrodites, the second by one of Artemis’ (trans. Jones 1895: 26, emphasis added)—despite the fact that Ἄρτεμις is in the nominative, not the genitive. In other words, scholars insert a statue of Artemis into Pausanias’s ekphrasis when, strictly speaking, there is ‘a statue of Aphrodites’ (Ἀϕροδίτης ἄγαλμα), and then there is simply ‘Artemis’ (Ἄρτεμις).

Similarly, when Pausanias transcribes an inscription from a wooden horse at Delphi, listing ‘images’ (εἰκόνας) of gods and heroes dedicated from the spoils of war, he unequivocally states: ‘These are Athena and Apollos’ (εἰσὶ δὲ Ἀθηνᾶ τε καὶ Ἀπόλλων, 10.10.1). But once again Gordon (1979: 7) cites scholars who ‘correct’ the presumed ‘failure’ to distinguish between gods and images by glossing the Greek to read: ‘These represent Athena and Apollos’ (trans. Jones 1895: 96–7, emphasis added). Thus, Gordon (1979: 8) concludes: ‘It is we who introduce the concept of “representation”, where Pausanias says “they are …”’. This fundamental observation about the ‘ambiguous status of the art object’ (Gordon 1979: 10–11) has since been followed and further developed by scholars with respect to a range of other Greek and Latin texts, to the point where it is now largely taken for granted. ²⁴ Clifford Ando (2008: 22–27) thus concludes that by glossing over the semantic tension between deity and statuary in Greek and Roman writings about images, and by projecting a semiotic theory of ‘representation’ onto evidence where there is none, we mask an important feature of ancient ekphrasis, namely, the ability to depict images of the gods both as images and as gods.

This artistic and ekphrastic phenomenon is further reinforced by the ways in which statues are described in inscriptions. Joseph Day (2010) has identified two common patterns whereby epigrams textualize the ritual performance of dedication and frame visual encounters with the image. Most often, epigrams follow a ‘narrative sequence’ involving the dedicant (subject), the act of dedicating (verb), the dedication itself (direct object), and the divine recipient (indirect object): ‘person x set up dedication y for god z’ (Day 2010: 6, slightly modified; citing DVA 111–28 and ThesCRA 1.269–78). In these cases, the statue of the god is offered as a gift to the god, who is, as a result, presumably distinguished from the votive statue as its recipient. Less often, but not uncommonly, epigrams are written from the perspective of the dedicated statue, rather than the human dedicant, which addresses readers directly and identifies itself either as the god (i.e., a divine name in the nominative, sometimes with εἰμί), or as belonging to the god (i.e., a divine name in the genitive, sometimes with εἰμί, sometimes with the substantive ἱερόν): ‘(I am) god x’, or ‘(I am) (the consecrated object) of god x’ (Day 2010: 6–7; citing CEG 251, 302, 354, 400; cf. 407). Blending these two common patterns together, other inscriptions follow the sequence of the first but narrate it from the perspective of the statue as in the second: ‘person x set me up, dedication y, for god z’ (CEG 190, 207, 259, 281, 363). In these latter two cases, the distinction between the image and the god is blurred nearly to the point of erasure, with the deity ostensibly present as a public interlocutor with its viewers in the form of its statue (see further Day 2010: 130–80). To look at a cult statue is to see ‘a being who looked back’ (Platt 2011: 78, emphasis original), and to read its accompanying inscription is to hear that being speak.

Finally, as with art, ekphrasis, and epigrams, so also with dreams and their interpretation. ²⁵ The fluidity and instability of dreams—or what Patricia Cox Miller (1994: 30) calls ‘the oneiric imagination’—provide a medium of visuality capable of bending conventional ontological categories, making them well suited to probe the already fuzzy relation between gods and their images. In dream visions, gods and other nonobvious divine beings often appear in the form of their statues. For instance, when three nymphs appear to Daphnis in a dream, they are described as ‘tall and beautiful ladies, semi-clothed and barefoot, looking like their statues’ (τοῖς ἀγάλμασιν ὅμοιαι, Longus, Daphnis and Chloe 2.23). In other cases, gods do not merely look like their statues; they quite simply are animated statues (Cox Miller 1994: 28–31). According to Artemidorus, an ancient expert on dreams and divination, this should not be wholly surprising: ‘statues of the gods have the same meaning as the gods themselves’ (Interpretation of Dreams 2.39.9). When Artemis appears in a dream vision, Artemidorus claims that ‘it makes no difference whether we see the goddess herself, as we have imagined her to be, or whether we see her statue (ἄγαλμα αὐτῆς). For whether gods appear in the flesh (σάρκινοι) or as statues (ἀγάλματα) fashioned out of some material, they have the same meaning’ (2.35). While there is a material difference between gods (‘in the flesh’) and statues (‘fashioned out of some material’), Artemidorus insists that there is ‘no difference’ at the level of visual encounter (Platt 2011: 285). He goes on to apply this same logic to (statues of) Zeus (2.35); Heracles (2.37); Poseidon, Amphitrite, and Nereus (2.38); Serapis, Isis, Anubis, and Harpocrates (2.39). So, it would seem, in response to the question posed in the title of H. S. Versnel’s famous essay, ‘What Did Ancient Man See When He Saw a God?’ (1987), one answer would be, quite simply, its cult statue.

Material Media of Divine Epiphany

In light of recent reevaluations of the sort of evidence discussed above, the view that has come to dominate the fields of classical and art-historical scholarship in recent decades is that a cult image constitutes the material epiphany of a god. ²⁶ In other words, an image of a god is a material ‘mode of presencing’ the god among mortals: it makes seen and accessible normally nonobvious beings, who are, for all intents and purposes, out of sight and out of touch to most mortals most of the time. ²⁷ In fact, this view was anticipated already by Donohue and others, who suggested that ‘the phenomenon of epiphany is central’, lying at ‘the heart of Greek iconism’ and providing the ‘key to understanding the Greek images of the gods’ (Donohue 1997: 44–45; cf. Gladigow 1985/1986, 1990, 1994; Versnel 1987). But it has since been developed most fully by Verity Platt (2011: 77–123) and Georgia Petridou (2016: 49–64), who built on the earlier work of Jean-Pierre Vernant (1983, 1991) and the so-called École de Paris that pooled around him, which pioneered the academic study of the physiology of the gods and its manifestation among mortals. ²⁸

On this epiphanic approach, the relation between deity and statuary is not one of dualistic representation; nor is it one of straightforward identification. It is, in a sense, both and neither. In the famous coinage of Vernant, a cult statue is a technology of presentification, providing one epiphanic strategy of many in the ancient Mediterranean, which facilitates the material production and visual perception of divine presence for ritual interaction and social exchange at local sites. Vernant (1991: 153) explains: ‘the idea is to establish real contact with the world beyond, to actualize it, to make it present, and thereby to participate intimately in the divine; yet by the same move, it must also emphasize what is inaccessible and mysterious in divinity, its alien quality, its otherness’. ²⁹ In this way, the category of presentification both collapses and reinforces the cognitive dissonance between deity and statuary in a way that generates ‘a play of presence and absence, reality and representation, in which godhead can be apprehended not only in multiple images, but also on multiple ontological planes’ (Platt 2011: 199–23). The image both is and is not the god, and the complex entanglement between deity and statuary is such that gods simultaneously can and cannot be neatly distinguished from their images. ³⁰ Michael Squire (2011: 159) put it thus:

images did not simply re-present the divine, then, but could also render deities literally present before a viewer’s eyes. Greek and Roman art, in short, made a nonsense of twentieth century semiotic jargon, predicated on rationalist Christian theology: visual ‘signifiers’ did not ‘signify’ the divine ‘signified’; the divine ‘signified’ was in some sense the ‘signifier’, and so too was the ‘signifier’ the ‘signified’. (emphasis original)

Unsurprisingly, perhaps, the phenomenon of material epiphany could be conceptualized in a variety of different ways in the visuality of different viewers (so Scheer 2015: 175–76). As a result, we should resist the impulse to identify ‘a single, consistent ancient explanation’ for conceptualizing the precise epiphanic relation between a god and its image (Johnston 2008: 449–50). ³¹ Scholars have convincingly shown, for instance, that some could view cult images as empty ‘vessels’ or ‘receptacles’, which housed the presence of divinity within them, like ‘containers’ filled with πνεῦμα or numen (Faraone 1992: 18–35, 94–223; Steiner 2001: 120–34; Neer 2010a: 109–21). But others could see them as ‘seats’ temporarily facilitating divine presence upon them (Scheer 2000: 21–23, 143–45; Bettinetti 2001: 52–54), or as statuary ‘disguises’ akin to bodily metamorphosis, concealing divine presence for safe interaction with unsuspecting mortals (Petridou 2016: 62–64), or as living ‘doubles’ or ‘second like objects’, standing in for the deity as an avatar (Vernant 1983: 305–20, 1991: 164–85, 186–92). Still others could intellectualize images with a clearer distinction between the stuff of divinity (e.g., πνεῦμα/numen, power, immortality) and that of statuary (e.g., wood, stone, gold, ivory), while nevertheless conceding their epiphanic utility in making distant deities visually present for mundane interaction with mortals (cf. Rives 2007: 36–37; Ando 2008: 31–34). What is important for our purposes is that the ritual perspective of cult images as material media of presencing the gods remains consistent (however construed) in the ritual-orientated visuality of everyday Greeks and Romans.

There were, however, different kinds of divine image, each of which, precisely as technologies of material epiphany, represent different strategies of ‘presentification’ (so Platt 2013). This is an important final point to consider when exploring the relation between deity and statuary. Depending on their artistic style and formal architecture—ranging from the figural (or semi-figural) to the non-figural (or ‘aniconic’)—the materiality of images could be manipulated in various ways to evoke the presence of gods through various means. The Athenian Acropolis provides an illuminating example, which housed not one but a few celebrated statues of the goddess Athena. Most famously, this includes Phidias’s colossal, naturalistic, chryselephantine statue of the Athena ‘Parthenos’, and a smaller, cruder, archaic, olive-wood xoanon of the Athena ‘Polias’. ³² Both were variously dubbed ‘the statue’ (τὸ ἄγαλμα), ‘the statue of Athena’ (τὸ ἄγαλμα τῆς Ἀθηνᾶς), ‘the Athena’ (ἡ Ἀθήνη), or simply ‘the goddess’ (ἡ θεός) (see Ridgway 1992). And both facilitated a visual encounter with the deity as ‘a locus of presence’. ³³ But, as Platt (2011: 83–123) has argued, the material quality, ritual function, and artistic technique (τέχνη) deployed in the production and perception of the image exemplify different epiphanic strategies to facilitate such an encounter. For chryselephantine statues, such as the Parthenos, not only were gold and ivory widely considered to have magical properties well suited to conjure divine presence, but they were also among the purest, brightest, finest, smoothest, rarest, and priciest materials known to antiquity—best able to convey the sublime materiality that constituted the physiology of ‘godsbodies’ (Lapatin 2001; 2010: 139–42, 150). ³⁴ Furthermore, the naturalistic forms, colossal proportions, and reflective surfaces of chryselephantine statues impressed upon viewers the anthropomorphic, supersized, and luminous characteristics of divine bodies, visually eliding the god and the image through naturalistic mimesis. But for materially crude, archaic, and semi- or non-figural xoana, such as the Polias, the presence of the divine was visually evoked not through mimetic approximation but through ritual performances that reified the detection of divine agency, such as bathing, feeding, drinking, or clothing (Gladigow 1985/1986, 1990, 1994), combined with aetiological myths about their celestial origins (e.g., Pausanias, Description of Greece 1.26.6). Thus, Platt (2011: 119) concludes that archaic images (including the Polias) acted as ‘material markers of divine presence’ when activated as such through ritualization and mythmaking, whereas naturalistic images (including the Parthenos) might have ‘no specific ritual function’ but visually elided the god and the image to varying degrees (and to varying levels of success) by approximating the likeness of godsbodies through mimesis.

To summarize: images of the gods served as material media of divine epiphany; but different kinds of images deployed different epiphanic strategies in the production and perception of divine presence. Such strategies manipulated materiality as a technology of ‘presentification’, which could be conceptualized in a variety of ways and achieved through a variety of means, including visual approximation, ritualization, and mythmaking. So, to return to Paul, we now ask: How might the materiality of the messiah as an image of the Jewish god fit into this wider world of divine images as technologies in the production of divine presence? And what might it say about the relation between Paul’s iconic Christ and the Jewish god he images?

The Iconic Christ in the Letters of Paul: Images of the Jewish God and the Materiality of the Messiah

Images of the Jewish God

In the history and anthropology of religion, the use or non-use of divine images has long served as a bright line dividing Jews from non-Jews throughout the ancient Mediterranean basin and beyond. On this view, ritual aniconism (the de facto absence of figural divine images) and rhetorical anti-iconism (the programmatic antagonism towards figural divine images) together constitute a veritable poster child for early Jewish (and, subsequently, Christian) uniqueness in the ‘sculptural environment’ of the Graeco-Roman Mediterranean. ³⁵ As Jaś Elsner once put it: ‘Images were at the centre of traditional religion in the Graeco-Roman world’ (1998: 119), whereas early Judaism, in stark contrast to the wider world it inhabited, was ‘the most un-iconic (indeed anti-iconic) of religions’ (1998: 215). Setting aside, for the moment, the historiographical issues with such claims and the material evidence for early Jewish art and iconography, ³⁶ as well as widespread traditions of aniconism and anti-iconism among Greeks, Romans, and other non-Jews throughout West Asian and Mediterranean antiquity, ³⁷ a growing number of scholars have come to argue that the Jewish god, no less than his ‘pagan’ contemporaries, in fact always had divine images, by which he, too, was made present among his people. ³⁸ In other words, non-Jews could be aniconic, even anti-iconic, just as Jews could be iconic. As Milette Gaifman and other art historians have been at pains to point out, aniconism, anti-iconism, and iconism were not mutually exclusive: they can and did coexist, even in the same ancestral tradition—whether Greek (Gaifman 2012), Roman (Kiernan 2020), Persian (Shenkar 2014), Phoenician (Doak 2015), Nabatean (Patrich 1990; Healey 2001), or, I suggest, Jewish. ³⁹

If, as Paula Fredriksen (2006: 241) boldly put it, ‘in antiquity, all gods exist’, I would want to add that, in antiquity, all gods have images. Paul’s god—the Jewish god—was no great exception. Like those of Greek and Roman gods, images of the Jewish god took a number of different material forms in the service of a number of different epiphanic strategies. They ranged from standing stones, vacant spaces, and empty thrones (Mettinger 1995: 135–97; Levtow 2008: 132–43; Doak 2015: 125–32; McClellan 2022: 133–46) to law tablets, torah scrolls, and ‘iconic books’ (van der Toorn 1997; Parmenter 2011; Stavrakopoulou 2013; Watts 2017, 2018; 2020; McClellan 2022: 175–93; cf. Watts 2013; Watts and Yu 2021). ⁴⁰ Each of these images effectively serve as material media that made the Jewish god present among his people for human-divine interaction, whether for ritualized warfare, reciprocal exchange, or communication. Daniel McClellan (2022: 199) has gone so far as to suggest (rightly, in my view) that it is within this diverse spectrum of Jewish iconicity that literary depictions of god’s ‘wisdom’ (σοϕία), god’s ‘word’ (λόγος), and perhaps also god’s ‘spirit’ (πνεῦμα) can also be understood as ‘presencing media’ in a textual mode, promoted as images of the Jewish god in the mythmaking of religious experts. ⁴¹

Perhaps the most famous image of the Jewish god, however, still well known as such to most scholars today, is more specifically anthropomorphic, even anthropological: not merely human-like statuary but humanity itself. ⁴² ‘God made humankind (τὸν ἄνθρωπον); according to an image of god (κατ᾽ εἰκόνα θεοῦ), he made him; male and female, he made them’ (LXX Gen. 1.27). In what is popularly dubbed the imago Dei tradition, humans have been traced throughout the Hebrew Bible and early Jewish literature—from the Septuagint to the Mishnah—as de facto cult statues, animated by the reception of divine pneuma and thereby facilitating the production of divine presence. ⁴³ In this sense, humanity is to the Jewish god what statuary is to gentile gods. ⁴⁴ And like statuary, I suggest, the materiality of humans could be manipulated as a technology of ‘presentification’ in the service of different epiphanic strategies, with the result that different kinds of humans effectively functioned as different kinds of divine images, which produced divine presence in different kinds of ways. So, as Stephen Herring (2013) has argued, humanity in general is one kind of image of the Jewish god (87–127), but Israel in particular is another kind of divine image (164–208), Moses still another (127–63), and so on. The messiah is, for Paul, one such kind of image.

But what do we make, then, of early Jewish polemics against so-called ‘idols’ (e.g., Jer. 10.1–16; Isa. 44.9–20; Pss. 115.1–8; 135.15–18; Wis. 13–15; Bel; Let. Jer.), Paul’s not least among them (e.g., 1 Thess. 1.9–10; 1 Cor. 8.1–11.1; 12.2; 2 Cor. 6.14–7.1; Rom. 1.18–32)? After all, many scholars still tend to appeal to ‘anti-idolism’ as a point of ‘wholesale incompatibility’ that demarcates early Judaism from so-called ‘paganism’ (Fine 2005: 79; cf. Halbertal and Margalit 1992: 236). ⁴⁵ One way scholars have sought to avert this seeming dichotomy is by arguing that the discourse of ‘idolatry’ is exactly that: a discourse (Anderson 2016; Barbu 2016). More specifically, as I have argued elsewhere, it is a discourse designed by ‘ethnically coded experts in foreign religion’ (or ‘ethnopolitical entrepreneurs’) to create and categorize ethnic difference, not necessarily to reflect it or describe it (Chantziantoniou 2023: 69–79). ⁴⁶ As such, Jewish polemics against images of gentile gods proceed as an ethnic group-making strategy of social formation (what Brubaker 2004 calls ethnic ‘groupism’), which produces the perception of alterity and glosses over possible points of similarity by creating, categorizing, and differentiating ethnic groups through totalized stereotypes of their ancestral images. ⁴⁷ In fact, this strategy was common among Greeks and Romans, no less than Jews, who similarly stereotyped the images of others to facilitate self-definition and differentiation from ethnic foreigners (e.g., Persians, Scythians, Nabateans, Phoenicians, and other so-called ‘barbarians’). ⁴⁸ As a result, it would be a mistake to treat discursive claims to difference as sociological evidence of difference, reifying ancient ethnic stereotypes as modern historiographical facts, and, in the process, dismissing evidence of ‘a spectrum of iconicity’ that exists alongside that of ‘a spectrum of the aniconic’ in early Judaism itself, due to performative rhetoric about the images of others. ⁴⁹ Put simply, Paul’s negative polemicizing against images of gentile gods need not detract from his positive mythmaking about images of the Jewish god, which coexisted in his mission to turn gentiles from the former to the latter.

The Materiality of the Messiah

It is in this sense that Paul, too, can be understood as ‘deeply iconic’ (Rowe 2005: 301). The messiah, he says in no uncertain terms, ‘is an image of god’ (ἐστιν εἰκὼν τοῦ θεοῦ, 2 Cor. 4.4; cf. Col. 1.15). And, as such, ‘the glory of god’ (τῆς δόξης τοῦ θεοῦ) is seen and known ‘in the face of the messiah’ (ἐν προσώπῳ Χριστοῦ, 4.6). In these texts, δόξα functions as an epiphanic category referring to the luminous presence of the Jewish god, and πρόσωπον functions as a somatic category referring to the physical body of the messiah (specifically his face), quite precisely as a figural, three-dimensional, anthropomorphic image (εἰκών) of god (Lorenzen 2008: 139–256). ⁵⁰ The relationship between the use of δόξα in early Jewish literature (including Paul’s letters) and the wider history of the term in postclassical Greek is a lexicographical puzzle in need of further interrogation, especially the ‘semantic shift’ it apparently underwent through Greek translations of כבוד‎ in the Septuagint (Aitken 2014: 193; cf. Frey 2014). ⁵¹ But Carey Newman (1992) and others (Gaventa 2014; Grindheim 2017; Jackson 2018a, 2018b) have convincingly shown that the term appears to refer to the radiant presence of the Jewish god. ⁵²

Newman (1992: 134–53) goes so far as to suggest that the phrase δόξα θεοῦ (or δόξα κυρίου) is a ‘technical term’ denoting the visible, mobile, manifest presence of Israel’s god, closely corresponding to כבוד יהוה‎ in the Hebrew Bible. M. David Litwa (2012: 123–26) goes further still, emphasizing rather more strongly the pneumatic physiology and celestial illumination that characterizes the manifest presence of divinity, leading him to argue that τῆς δόξα τοῦ θεοῦ in 2 Cor. 4.6 evokes ‘a visible, luminous, and divine corporality’ made present among mortals in the human face of the messiah. Either way, this is quite precisely an instance of material epiphany: a divine image facilitates the presence of a deity, and it does so as a property of its materiality. As an image of the Jewish god, it is the human physiology of the messiah that constitutes the material medium of divine epiphany. So, while Paul never refers to divine ‘epiphany’ (ἐπιϕάνεια) in his undisputed letters, strictly speaking, he does refer to divine ‘glory’ (δόξα); and, not insignificantly, he often does so in close connection with εἰκών (1 Cor. 11.7; 15.39–49; 2 Cor. 4.4–6; Rom. 1.23; cf. 8.29–30, δοξάζω). ⁵³ This does not suggest an alleged synonymity between δόξα and εἰκών (pace Kim 1989: 230), but it does seem to elide a firm distinction between divine presence and divine image.

Indeed, in what Margaret Mitchell has called the ‘epiphanic logic’ of Paul’s argument throughout 2 Cor. 2.14–7.4—a main concern of which is how the messiah is perceived through the senses (so Heath 2013: 193–97)—Paul portrays himself as an ‘epiphanic envoy’ who carries around the messiah like a cult statue paraded in iconic ritual procession during epiphanic festivals (Mitchell 2004: 187–91; cf. Duff 1991a, 1991b, 2008, 2015: 93–98, 117–20). ⁵⁴ Immediately after Paul refers to the messiah as an image of god (4.4–6), he refers to himself in no uncertain terms as carrying around (περιϕέρω) and making visible (φανερόω) that image in his own mortal body (4.7–11) as ‘an aural-visual icon of Christ crucified’ (Mitchell 2004: 189). ‘We have this treasure in earthen vessels’, Paul says, ‘always carrying around (περιϕέροντες) the dying of Jesus in the body, so that the life of Jesus might also be made visible (ϕανερωθῇ) in the body’ (4.7, 10). As Paul Duff (1987, 1991a, 2008: 776–79) has demonstrated at length and in detail, Paul here uses language and imagery that commonly refers to parading the presence of gods by carrying around their images in iconic ritual procession, which was a practice designed to evoke their presence and power to the public (e.g., Dionysius of Halicarnassus, Roman Antiquities 7.72.13; Plutarch, Isis and Osiris 357f, 365b; Athenaeus, Learned Banqueters 5.197c–203b; Pausanias, Description of Greece 9.22.1–2). ⁵⁵ So, it would seem, in ‘Paul’s visual piety’ (Heath 2013’s apt phrase, adopted from Morgan 1998), the Jewish god indeed has a cult image, and that image is the dead and deified messiah whom Paul himself proclaims and parades.

To be sure, elsewhere in his Corinthian correspondence, Paul affirms that ‘a man’ (ἀνήρ)—that is, a gendered male mortal—is also ‘the image and glory of god’ (εἰκὼν καὶ δόξα θεοῦ, 1 Cor. 11.7). ⁵⁶ Some scholars view this text as ‘untypical’ of Paul’s claims about divine images and divine presence (e.g., Dunn 1989: 105), due to his insistence elsewhere that ‘all … lack the glory of god’ (πάντες … ὑστεροῦνται τῆς δόξης τοῦ θεοῦ, Rom. 3.23), having exchanging it for an image (εἰκόνος) of humans or animals (1.23) and only recovering it through conformity to the image (εἰκόνος) of the messiah (8.29–30; cf. 2 Cor. 3.18). ⁵⁷ But I see no good reason not to assume that Paul took for granted that gods had many different kinds of images, each of which served different kinds of epiphanic strategies. Hence, in the same letter where Paul denies divine presence among immoral humans (Rom 3.23) and chronicles the gentile exchange of god’s presence for images of their own (1.23), he nevertheless affirms that ‘the glory’ (ἡ δόξα) continues to belong to his Israelite kinsfolk (9.4), presumably referring to the divine presence residing in the innermost sanctuary of the Jerusalem temple (so Stowers 1994: 130–31; Fredriksen 2017: 35), where the Jewish deity was made present through a form of iconography that Tryggve Mettinger (1995: 18–27) famously dubbed ‘empty-space aniconism’. ⁵⁸

Even so, it is clear that, at least so far as human images of the Jewish god are concerned, there is another step involved: humans, for Paul, bear the image of an image. ⁵⁹ Either they bear the image of Adam, a living but mortal being with a terrestrial materiality (i.e., ‘the image of the terrestrial’, τὴν εἰκόνα τοῦ χοϊκοῦ), or they bear the image of the last Adam, Jesus, who at the resurrection became a life-giving and immortal πνεῦμα with a celestial materiality (i.e., ‘the image of the celestial’, τὴν εἰκόνα τοῦ ἐπουρανίου, 1 Cor. 15.42–49). ⁶⁰ It is in this latter sense that the resurrected messiah, as a pneumatic being (cf. 2 Cor. 3.17), constitutes a divine image of a different kind with different material qualities than the terrestrial body of ordinary mortals—one to which the latter ultimately ought to conform (2 Cor. 3.18; Phil. 3.20–21; Rom. 8.29; cf. Col. 3.10). As such, the materiality of mortal humans and the materiality of the immortal messiah facilitate the presence of god in different ways and to different degrees. As Paul himself makes clear: ‘There are both celestial bodies (σώματα ἐπουράνια) and terrestrial bodies (σώματα ἐπίγεια), but the glory of the celestial (ἡ τῶν ἐπουρανίων δόξα) is one thing, and that of the terrestrial (ἡ τῶν ἐπιγείων) is another’ (1 Cor. 15.40). As a result, for bearers of the image of the messiah to be conformed to that same image can be understood as moving ‘from glory to glory’ (ἀπὸ δόξης εἰς δόξαν, 1 Cor. 3.18). A change in divine image necessarily constitutes a change in material medium of divine presence.

For Paul, then, both mortal humans and the immortal messiah are divine images. And, as such, both are technologies of divine epiphany. But they represent different kinds of epiphanic strategy, which proceed on the basis of different kinds of materiality. As argued in the previous section, the more naturalistic an image in its visual approximation of a deity, the more mimesis and material quality play into the production of divine presence (Platt 2011: 113–23; cf. Neer 2010a). So, too, in the case of the messiah, the immortality, imperishability, and pneumatic physiology of his celestial body blurs the visual and material distinctions between the image and the deity—so much so that it is possible to view Paul’s imagemaking as a kind of deification, both as it applies to the messiah (Litwa 2014) and as it applies to those who conform to the image of that messiah (Litwa 2012). As a result, Paul’s iconic Christ both is and is not divine, and the complex entanglement between the presence of god and the materiality of the messiah is such that Paul’s god simultaneously can and cannot be neatly distinguished from his christological image.

Conclusion

Approaching longstanding christological debates about the humanity of Jesus and its relation to divinity from the perspective offered here has several benefits. Scholars still debate whether Paul ever explicitly calls the messiah a god (Rom. 9.5). ⁶¹ But there can be no question that he calls the messiah an image of a god (2 Cor. 4.4). While scholarly comparisons between Jesus and other Mediterranean gods may be fruitful in many respects, I suggest that a productive way forward would be to compare Jesus with what is arguably more proximate comparanda for Paul’s iconic claims, namely, images of other Mediterranean gods. This is by no means to suggest that Greek and Roman sources are somehow more (or less) proximate to Paul’s iconic claims than Jewish ones. Rather, it is more simply to suggest that literary, artistic, and archaeological evidence involving images of gods (whether Greek, Roman, Jewish, or otherwise) may prove more illuminating than that involving gods as such (again, whether Greek, Roman, Jewish, or otherwise). This is a practice-orientated approach that focuses on the iconic content of the comparanda rather than their ethno-religious classification, cutting across disparate categories towards ‘a kind of integrated Greco-Romano-Judeanness’ (Eyl 2020). In so doing, it demonstrates that social practices and cognitive processes that inform the visuality of everyday Greeks and Romans can also be understood to inform that of Jews, including Paul.

When Paul’s iconic claims about Christ are redescribed and contextualized in terms derived from such comparanda, it becomes possible to conclude that the messiah is no mere representation of the Jewish god, indexing a human ‘signifier’ of the divine ‘signified’ with a hard and fast distinction between them. Nor is there a straightforward identification of the messiah with or as the Jewish god himself, contained within ‘the unique identity of the one God’ (Bauckham 2020: 146), as somehow one and the same by virtue of shared acts, attributes, or prerogatives. Rather, Paul’s iconic Christ can be understood more precisely as a presentification of the Jewish god, making visually present and locally accessible a deity who is normally nonobvious to human sense perception. Put simply, the messiah is not the Jewish god; he is an image of that god. But, precisely as such, he is a living ‘proxy’ or ‘cult manifestation’ of divine presence, power, agency, and action, serving as a site for ritualized acts such as invocation, petition, meals, and so on (Novenson 2019: 13), while nevertheless remaining entangled with but distinguished from the god he images. ⁶²

The point is well illustrated by Plato’s dialogue between Socrates and Cratylus.

If there were two things, such as Cratylus and an image (εἰκών) of Cratylus, if someone of the gods not only were to make it with regard to your color and shape just as painters do, but also were to make all the internal qualities like yours … and were to place inside it the movement and life and thought such as you have … and were to stand this other thing close to you, would there then be Cratylus and an image (εἰκών) of Cratylus, or two Cratyluses? (Cratylus 432b-c; trans. Steiner 2001: 69, slightly modified)

Cratylus, of course, concedes the point. The image, while always remaining nothing less than an image, has effectively become something more: a ‘double’ or a ‘second like object’. The same logic applies to cult statues. In a guidebook to the seven wonders of the ancient world, attributed to Philon of Byzantium, the author claims that, by fashioning an image of the god Helios out of a quality of material that corresponds to the splendour of the god himself, the imagemaker ‘made the god equal to the god’ (τῷ θεῷ τὸν θεὸν ἴσον ἐποίησεν), and ‘in this way he achieved a great work … since he established for the world a second Helios’ (On the Seven Wonders of the World 4.6). ⁶³ As it happens, the Greek word commonly used to connote this kind of image (i.e., a ‘double’ or ‘second like object’) is εἴδωλον, popularly known by its transliteration into English as ‘idol’. ⁶⁴ So, I conclude, provocatively perhaps, that Paul’s iconic Christ may well be redescribed in other terms as a true and living idol. Or, in descriptive terms closer to Paul’s own, the messiah is a cult image of the Jewish god.